Peritoneal dialysis

Peritoneal
dialysis
edited by
KARL D. NOLPH, M.D.
Division of Nephrology,
University of Missouri Health Sciences Center,
VA Hospital, and Dalton Research Center,
Columbia, Missouri, USA

Second, revised and enlarged edition

1985 Springer-Science+Business Media, B.V.

ISBN 978-94-017-2562-0 ISBN 978-94-017-2560-6 (eBook)
DOI 10.1007/978-94-017-2560-6

Library of Congress Cataloging in Publication Data

Main entry under title:

Peritoneal dialysis.

Includes bibliographies and index.

1. Peritoneal dialysis. I. Nolph, Karl D.
[DNLM: 1. Peritoneal Dialysis. WJ 378 P446]
RC901.7.P47P47 1984 617'.461059 84-14789

ISBN 978-94-017-2562-0
Copyright

© 1985 Springer Science+Business Media Dordrecht

Originally published by Martinus NijhoffPublishers, Boston in 1985
Softcover reprint of the hardcover 2nd edition 1985

All rights reserved. No part of this publication may be reproduced. stored in a retrieval system, or
transmitted in any form or by any means. mechanical. photocopying, recording. or otherwise, without
the prior written permission of the publishers. Springer Science+Business Media Dordrecht.
Foreword to first edition

A year or so after Dr. Robert Popovich arrived in Seattle in 1965 to begin working
on his doctoral thesis under Dr. A.L. Babb, we had just begun work to try to
prove the prediction that the peritoneum had a higher permeability to 'middle
molecules' than hemodialysis membranes [1]. Several years later, when Dr.
Popovich accepted a position at the University of Texas in Austin, he decided to
concentrate his research efforts in the area of peritoneal dialysis and everyone
knows how successful that effort has become [2]. Indeed, because of continuous
ambulatory peritoneal dialysis (CAPD), long-term peritoneal dialysis after a
two-decade incubation period is finally becoming an equal option to hemodialysis
and transplantation in the management of chronic renal failure.
For me this development represents final vindication of a twenty-year effort to
help promote peritoneal dialysis, often in the face of enormous opposition. I
particularly remember a policy meeting at the NIH a few years back in which it
was decided by my colleagues on the committee that long term peritoneal dialysis
had no future and therefore no funds for projects in this area would be forthcom-
ing. Based on the excellent results that Boen and later Tenckhoff had been
getting in our Seattle program, I knew the committee was wrong and tried to
convince them otherwise. Naturally, being the only favorable vote, I failed. I
often wonder how many years this decision and others like it set back peritoneal
dialysis.
Long term peritoneal dialysis was born out of necessity. After starting the first
three patients on long term hemodialysis in early 1960, the program was com-
pletely shut down because the hospital administration decreed that due to lack of
funding, no additional patients could be accepted until one of the first three died.
Since that did not occur until 11 years later, it would have been a long wait.
Later the administrators relented - mainly on the strength of a small research
grant from Dean George Aagaard that permitted us to add two more patients.
The second of these, J. R., proved to be the first failure on chronic hemodialysis,
who would have died had not Fred Boen arrived in Seattle about that time. J.R.
was dying simply because he immediately clotted the same A V shunt that was
VI

working so well in the other 4 patients. (Imagine what might have happened if
J.R. had been patient number 1 instead of number 5). The reasons for this
accelerated clotting were never identified. However, recently a small subgroup of
dialysis patients with an accelerated tendency to clot has been described [3] and
J.R. may have belonged to that group. In any event, Dr. Boen determined to try
to save J.R. by means of long term peritoneal dialysis. Figure 1 shows this patient
on the cycler cleverly fashioned out of equipment that had been developed five
years earlier by Dr. Thomas Marr for use in gastrodialysis [4].
In 1962, when Dr. Boen started working with J.R., long term peritoneal
dialysis had been abandoned because of the high incidence of peritonitis. Boen
decided to eliminate the bottle change as one source of infection and developed a

Figure 1. Patient JR on a 20-1 carboy dialysis system in 1962. From 1963 onwards 40-1 units were used.
 VII

closed sterile system using first twenty-liter and later forty-liter carboys of dialysis
fluid . This remarkable system required that a 'fluid factory' be built which could
manufacture and sterilize forty-liter bottles. A remote corner in the sub-base-
ment of the University of Washington Hospital was donated to the project and
this factory, Figure 2, operated successfully until 1979, when it was finally
refurbished and moved upstairs to more respectable quarters. To this day all
in-hospital peritoneal dialysis still is done with forty-liter carboys. Being com-
pletely closed and sterile, it is the safest system ever devised and permitted Boen
to keep J.R. going for many months until he finally became infected repeatedly
through the access device and eventually died. As a result Boen decided to
abandon attempts to develop a peritoneal access device and in January 1963,
started a second patient, J.D ., on peritoneal dialysis using a repeated puncture
technique which involved inserting a catheter for each dialysis. J. D. did very well
on peritoneal dialysis remaining virtually free of peritonitis for three years until
she was switched to hemodialysis. Eventually she received a transplant from her
sister and is alive and well today , seventeen years after starting peritoneal
dialysis. Boen regards his experience with J. D. as the crucial first step in finally
proving the potential feasibility of long-term peritoneal dialysis in the manage-
ment of end-stage kidney disease. Boen's research fellow, Henry Tenckhoff,
went one step further. He started a patient, M. 0., on peritoneal dialysis and
immediately moved her into the home, Figure 3. This approach required Dr.

Figure 2. Dr. Fred Boen and his chief technician, Mr. George Shilipetar in the fluid factory , 1962.
VIII

Figure 3. Patient M.O. on peritoneal dialysis at home , using a 40-1 carboy cycler which remained
completely closed throughout the procedure .

Tenckhoff to visit M.O. at home 3 times weekly to insert the peritoneal catheter.
Whereas the repeated puncture technique worked splendidly as a peritoneal
access technique with very little risk of infection and thereby proved to Boen and
Tenckhoff that long term peritoneal dialysis was feasible, it frustrated any wide-
spread application of the technique because cannula insertion for each dialysis
was so unpleasant and demanding. Some form of long term access simply had to
be devised. Henry Tenckhoff finally came up with the answer after overcoming
the bias against seeking a device which he inherited from his mentor, Fred Boen .
Dr. Tenckhoff received his initial encouragement from the success of the
Palmer-Quinton silicone catheter which had a long subcutaneous tunnel [5] .
 IX

About that time my colleague, Jack Cole, was experimenting with the bonding of
dacron felt to silicone arteriovenous shunts as an anchor and infection barrier.
Although this technique did not work for A. V. shunts, it permitted Tenckhoff to
re-design the Palmer-Quinton catheter into a shorter device that could be inser-
ted through a trochar and be held firmly in place by the dacron felt cuffs [6] . It is of
some interest that we later modified Tenckhoff's design for use as a right atrial
catheter for home parenteral nutrition [7]. This device saved that project from
utter failure. Originally we had proposed to infuse through a side-arm in an A-V
shunt [8]. This latter technique works well in uremia, but not in patients with
chronic bowel disease who readily clot both A-V shunts and A-V fistulas.
With long term peritoneal access finally assured , there still remained the
problem of a safe and practical source of peritoneal dialysis fluid . The forty-liter

Figure 4. The Cobe portable autoclave system for home peritoneal dialysis.
x
bottle system worked well enough, provided one had a 'fluid factory' and was
willing to deliver forty-liter carboys to homes in the area. It simply was not a
practical system.
Tenckhoff's first attempt to solve this problem is shown in Figure 4. This
apparatus was nothing more nor less than a miniature fluid factory with a
miniature autoclave to sterilize the fluid. Reverse osmosis replaced the still as a
source of pure water. Several of these machines were built by Cobe Laboratories
and used with great success locally. However, the heat of sterilization, the huge
size, the weight and the complicated operational sequence precluded widespread
application. The next generation of machines, developed by Curtis at the Seattle
V.A. and Tenckhoff, used reverse osmosis as the method of sterilization. These
machines have evolved into today's highly usuable units.
Despite the success of these reverse osmosis machines, the technique of long
term peritoneal dialysis never really took hold except in a few centers. It re-
mained for CAPD to really start things moving. I daresay that this book would
not have been planned had it not been for CAPD. I believe that both CAPD and
this volume will prove to be landmarks in the history of the therapy of chronic
renal failure. I sincerely hope I am right - time surely will tell.

January 1981 BELDING H. SCRIBNER

References

1. Babb AL, Johansen PJ, Strand MJ, Tenckhoff H, Scribner BH: Bidirectional permeability of the
human peritoneum to middle molecules. Proc Europ Dialysis Transplant Assoc 10: 247-262, 1973.
2. Popovich RP, Moncrief JW, Nolph KD, Ghods AJ, Twardowski ZJ, Pyle WK: Continuous
ambulatory peritoneal dialysis. Ann Intern Med 88: 449-456, 1978.
3. Kauffman HM, Edbom GA, Adams MB, Hussey CV: Hypercoagulability: A cause of vascular
access failure. Proc Dialysis Transplant Forum 9: 28--30, 1979.
4. Marr TA, Burnell JM, Scribner BH: Gastrodialysis in the treatment of acute renal failure. J Clin
Invest 39: 653-661, 1960.
5. Palmer RA, Newell JE, Gray JE, Quinton WE: Treatment of chronic renal failure by prolonged
peritoneal dialysis. N Eng J Med 274: 248--254, 1966.
6. Tenckhoff H, Schechter H: A bacteriologically safe peritoneal access device. Trans Am Soc Artif
Intern Organs 14: 181-186, 1968.
7. Atkins RC, Vizzo JE, Cole JJ, Blagg CR, Scribner BH: The artificial gut in hospital and home.
Technical improvements. Trans Am Soc Artif Intern Organs 16: 260-266, 1970.
8. Scribner BH, Cole JJ, Christopher TG, Vizzo JE, Atkins RC, Blagg CR: Long-term total
parenteral nutrition. The concept of an artificial gut. JAMA 212: 457-463, 1970.
Foreword to second edition

When the foreword to the first edition was written the big question was whither
CAPD? Many of the answers to that question will be found in the pages of this
second edition.
In 1979 I predicted that a new technique of combined ~utomated and ambula-
tory peritoneal dialysis might emerge as a sort of compromise between IPD and
CAPD [1]. What was then rank speculation has turned out to be correct in part, at
least, as CCPD has emerged as a new and apparently satisfactory approach. Now
I will go one step further and predict that some form of nightly peritoneal dialysis,
NPD, may evolve as the best compromise of all. It seems to me that NPD might
end up as the method of choice for most patients because it has the potential to
maximize the advantages of both IPD and CAPD while minimizing the disadvan-
tages. For example, the greatest advantage of IPD was its low incidence of
peritonitis, i.e. one episode every 41/2 patient years in the Montpellier experience.
With a properly designed reverse osmosis fluid supply system for NPD, it should
be possible to reduce the number of connect-disconnect procedures to seven per
week. In addition, improved connectology should make it possible to again
approach this low incidence of peritonitis.
NPD would eliminate all of the mechanical disadvantages of carrying 1-2 liters
of fluid in the abdomen including hernias, chronic back pain and the adverse
cosmetic effect of a swollen abdomen. Also, the daily tedium and inconvenience
of repeating the connect-disconnect procedure would be eliminated.
NPD probably requires lower dextrose concentrations to achieve the same
amount of weekly ultrafiltration. In addition, the elimination during the day of
both the dextrose load and the abdominal distension might improve protein
intake and help reverse the documented trend of CAPD patients to store fat on
the one hand and develop protein malnutrition on the other [2].
Finally, it should be possible to design a peritoneal fluid supply system based on
modern reverse osmosis and ultrafiltration technology that would be easy to
XII

operate, virtually maintenance-free, and always ready for nightly use by the
patient. Such a system would be very cost effective - the watchword of the future.

Seattle, January 1984 BELDING H. SCRIBNER

References

1. Scribner BH: A current perspective on the role of intermittent vs continuous ambulatory peritoneal
dialysis. Proc NE Regional Meeting of Renal Physicians Assoc 3: 76-81, 1979.
2. Heide B, Prenatos A et al.: Nutritional status of patients undergoing CAPO. Perit Dial Bulletin 1:
138-141, 1983.
Contents

Foreword to first edition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V

BELDING H. SCRIBNER
Foreword to second edition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XI
BELDING H. SCRIBNER
Preface.. .. . ..... .... . .. ... .. ... .. .. . .. .. . ... .. .. ... .. .. . .. .. . .. XV
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. XVII
1. History of peritoneal dialysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
S.T. BOEN
2. The peritoneal dialysis system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
KARL D. NOLPH and ZBYLUT J. TWARDOWSKI
3. The peritoneal microcirculation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
FREDERICK N. MILLER
4. Peritoneal ultrastructure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
CHRISTIAN VERGER
5. Transport kinetics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
ROBERT P. POPOVICH and JACK W. MONCRIEF
6. Ultrafiltration with peritoneal dialysis. . . . . . . . . . . . . . . . . . . . . . . . . . . 159
LEE HENDERSON
7. Intermittent peritoneal dialysis as renal replacement therapy. . . . . . . 179
SUHAIL AHMAD, Fu-HSIUNG SHEN and CHRISTOPHER R. BLAGG
8. Continuous ambulatory peritoneal dialysis. . . . . . . . . . . . . . . . . . . . . .. 209
JACK W. MONCRIEF and ROBERT P. POPOVICH
9. Continuous cyclic peritoneal dialysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
JOSE A. DIAZ-Buxo
XIV

10. Pharmacologic manipulation of peritoneal transport. . . . . . . . . . . . . . 267

JOHN F. MAHER and PRZEMYSLAW HIRSZEL
11. Comments on dialysis solution, antibiotic transport, poisoning, and
novel uses of peritoneal dialysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
JACK RUBIN
12. Managing the nutritional concerns of the patient undergoing peri-
toneal dialysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
MICHAEL J. BLUMENKRANTZ, ISIDRO B. SALUSKY
and R. WILLIAM SCHMIDT
13. Peritonitis................................................... 411
STEPHEN I. VAS
14. Complications of peritoneal dialysis other than peritonitis. . . . . . . . . 441
RAMESH KHANNA and DIMITRIOS G. OREOPOULOS
15. Peritoneal dialysis in children. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525
STEVEN R. ALEXANDER
16. Peritoneal dialysis in diabetics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
MARCEL LEGRAIN and JACQUES ROTTEMBOURG
17. Peritoneal membrane stability and the kinetics of peritoneal mass
transfer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 581
PETER C. SPENCER and PETER C. FARRELL
18. The USA CAPD Registry.. ... .. ... . .... . .. .. ... .. ... .. . .. . . . . 597
SETH M. STEINBERG, SIDNEY J. CUTLER, JOEL W. NOVAK
and KARL D. NOLPH
19. Peritoneal dialysis results in the EDTA Registry. . . . . . . . . . . . . . . . . 637
A.J. WING, R. MOORE, F.P. BRUNNER, C. JACOBS, P. KRAMER
and N.H. SELWOOD
20. Quality of life and psychosocial aspects of chronic peritoneal dialysis 667
R.M. LINDSAY, H.J. BURTON and S.A. KlINE
Index of subjects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 685
Preface

Peritoneal dialysis represents an internal technique for blood purification. In this

dialyzer the blood path, the membrane, and the dialysate compartment are
provided by nature. The developments of chronic peritoneal catheters, auto-
mated cycling equipment, solution preparation by reversed osmosis, manipula-
tions of transport with drugs, and the experiences with continuous ambulatory
peritoneal dialysis and continuous cycling peritoneal dialysis have increased the
interest in peritoneal dialysis. Publications related to peritoneal dialysis exceed
400 annually. The Peritoneal Dialysis Bulletin represents a new journal devoted
to peritoneal dialysis developments. The Third International Symposium on Peri-
toneal Dialysis is to be held in Washington, D.C. in 1984. From this meeting it is
likely that an International Society for Peritoneal Dialysis will emerge.
This book is meant to provide an overview of the state of the art of peritoneal
dialysis. Many clinicians are making extensive commitments to peritoneal dialysis
for the first time. Nephrologists, physiologists, pharmacologists, biomedical
engineers, and even physicists are involved in studies to better understand
peritoneal dialysis. The complexities of peritoneal dialysis and the peritoneal
membrane are becoming apparent. Studies of peritoneal dialysis increase under-
standing of the anatomy and physiology of biological membranes and of factors
influencing the passive movement of solutes across the microcirculation and
related structures. Peritoneal dialysis provides a 'window' to the visceral micro-
circulation in animals and humans.
Peritoneal dialysis may be useful to treat problems other than renal failure.
Beneficial effects in the treatment of hemorrhagic pancreatitis, dysproteinemias,
psoriasis, hypothermia, and many metabolic problems have been reported. The
intraperitoneal administration of chemotherapeutic agents is under investi-
giltion.
As I stated in the preface to the first edition, I feel fort~mate to have been
involved in peritoneal dialysis research during the past fifteen years. New ideas
and new developments have been an almost daily occurrence. Yet our under-
standing of this dialysis system is still in its infancy. The authors of the chapters in
XVI

this book have been actively investigating and writing about their respective
topics for many years. They are all individuals with whom I have had the good
fortune to have frequent contact.
As in the first edition, each chapter is an extensive review of a given topic. I
have not edited out all overlap between chapters since I feel the reader benefits by
exposure to slightly different perspectives of complex material and by allowing
each author to deal with all issues that relate to their respective topics.
The last sentence of the preface of the first edition summarized the major
purpose of this book. 'It is hoped that this book will serve as a reference text for all
those with more than a casual interest in peritoneal dialysis.' This remains my
hope for the second edition.

February 1984 KARL D. NOLPH, M.D.

Contributors

SUHAIL AHMAD, M.D., Northwest Kidney Center, 700 Broadway, Seattle, WA

98122, USA
Chapter 7
STEVEN R. ALEXANDER, M.D., The Oregon Health Sciences University Hospital,
3181 SW Sam Jackson Park Rd., Portland, OR 97201, USA
Chapter 15
CHRISTOPHER R. BLAGG, MD., Executive Director, Northwest Kidney Center,
700 Broadway, Seattle, WA 98122, USA; and Professor of Medicine, Uni-
versity of Seattle, WA, USA
Chapter 7
MICHAEL J. BLUMENKRANTZ, M.D., Adjunct Associate Professor of Medicine,
University of California, Los Angeles School of Medicine, Los Angeles, CA
90073, USA
Chapter 12
S.T. BOEN, M.D., Head, Department of Nephrology and Dialysis, Sint Lucas
Ziekenhuis, Jan Tooropstraat 164, Amsterdam-West, The Netherlands
Chapter 1
F.P. BRUNNER, M.D., EDTA Registry, St. Thomas' Hospital, London, England
SE17EH
Chapter 19
H.J. BURTON, Dr P.H., [Link]. (Hyg) , M.S.W., Associate Professor of Psychiatry,
The University of Toronto, Toronto, Canada
Chapter 20
SIDNEY J. CUTLER, Sc.D., The EMMES Corporation, Data Coordinating Center
for the USA CAPO Registry, 11325 Seven Locks Rd., Suite 214, Potomac,
MD 20854, USA
Chapter 18
JOSE A. DIAZ-Buxo, M.D., FACP, Director, Nalle Clinic Kidney Center, P.A.,
928 Baxter Street, Charlotte, NC 28204, USA
Chapter 9
XVIII

PETER C. FARRELL, Ph.D., Director, Centre for Biomedical Engineering, P.O.

Box 1, University of New South Wales, Kensington, Australia 2033
Chapter 17
LEE HENDERSON, M.D., Professor of Medicine, Associate Chief of Staff, V.A.
Hospital, 3350 LaJolla Village Drive, San Diego, CA 92161, USA
Chapter 6
PRZEMYSLAW HIRSZEL, M.D., Professor of Medicine, Division of Nephrology,
Uniformed Services University of Health Sciences, Bethesda, MD 20814,
USA
Chapter 10
C. JACOBS, M.D., EDTA Registry, St. Thomas' Hospital, London, England
SE17EH
Chapter 19
RAMESH KHANNA, M.D., Associate Professor of Medicine, Division of Nephrol-
ogy - M472, University of Missouri Health Sciences Center, One Hospital
Drive, Columbia, MO 65212, USA
Chapter 14
S.A. KLINE, M.D., FRCP(C), Clinical Director, Department of Psychiatry,
Toronto Western Hospital, 399 Bathurst, Toronto, Canada M5T 2S8
Chapter 20
P. KRAMER, M.D., EDTA Registry, St. Thomas' Hospital, London, England
SE17EH
Chapter 19
MARCEL LEGRAIN, M.D., Service de Nephrologie, Groupe Hospitalier de la Pitie-
Salpetriere, 83, bd de I'H6pital, 75654 Paris, Cedex 13, France
Chapter 16
BOB LINDSAY, M.D., FRCP(E), FRCP(C), FACP, Professor of Medicine, The
University of Western Ontario, Director, Renal Unit, Victoria Hospital
Corp., London, Ontario, Canada
Chapter 20
JOHN F. MAHER, M. D. , Professor of Medicine, Director, Division of Nephrology,
Uniformed Services University of Health Sciences, Bethesda, MD 20814,
USA
Chapter 10
FREDERICK N. MILLER, Ph. D. , Department of Physiology and Biophysics, Health
Sciences Center, Building A, University of Louisville, Louisville, KY 40292,
USA
Chapter 3
JACK W. MONCRIEF, M.D., Acorn Research Laboratory and Austin Diagnostic
Clinic, Austin, TX 78765, USA
Chapter 8
R. MOORE, M.D., EDTA Registry, St. Thomas' Hospital, London, England
SE17EH
Chapter 19
 XIX

KARL D. NOLPH, M.D., Director, Division of Nephrology - M472, Professor

of Medicine, University of Missouri Health Sciences Center , VA Hospital and
Dalton Research Center, Columbia, MO 65212, USA
Chapters 2 and 18
JOEL W. NOVAK, M. S., The EMMES Corporation, Data Coordinating Center for
the USA CAPD Registry, 11325 Seven Locks Rd., Suite 214, Potomac,
MD 20854, USA
Chapter 18
DIMITRI G. OREOPOULOS, M.D., Division of Nephrology, Toronto Western Hos-
pital, 399 Bathurst, Toronto, Canada M5T 2S8
Chapter 14
ROBERT P. POPOVICH, Ph.D., University of Texas, Department of Chemical and
Biomedical Engineering, Austin, TX 78712, USA
Chapter 5
JACQUES ROTTEMBOURG, M.D., Service de Nephrologie, Groupe Hospitalier de la
Pitie-Salpetriere, 83, bd de I'Hopital, 75654 Paris, Cedex 13, France
Chapter 16
JACK RUBIN, M.D., Division of Nephrology, University of Mississippi Medical
Center, 2500 North State Street, Jackson, MS 30216, USA
Chapter 11
ISIDRO B. SALUSKY, M.D., Assistant Professor of Pediatrics, University of Cali-
fornia, Los Angeles School of Medicine, Los Angeles, CA 90073, USA
Chapter 12
R. WILLIAM SCHMIDT, M.D., Assistant Professor of Clinical Medicine, Medical
Director, Kidney Disease Center, St. Francis Memorial Hospital, San Fran-
cisco, CA 90073, USA
Chapter 12
BELDING H. SCRIBNER, M.D., Professor of Medicine, University of Washington,
Division of Kidney Diseases, Box Room 11, Seattle, WA 98105, USA
Forewords
N.H. SELWOOD, M.D., EDTA Registry, St. Thomas' Hospital, London, England
SE17EH
Chapter 19
Fu-HsIUNG SHEN, M.D., Associate Professor of Medicine, Northwest Kidney
Center, 700 Broadway, Seattle, WA 98122, USA
Chapter 7
PETER C. SPENCER, M.D., Centre for Biomedical Engineering, P.O. Box 1,
University of New South Wales, Kensington, Australia 2033
Chapter 17
SETH M. STEINBERG, Ph.D., The EMMES Corporation, Data Coordinating
Center for the USA CAPD Registry, 11325 Seven Locks Rd., Suite 214,
Potomac, MD 20854, USA
Chapter 18
xx
ZBYLUT J. TWARDOWSKI, M.D., Associate Professor of Medicine, University of
Missouri Health Sciences Center, One Hospital Drive, Columbia, MO 65212,
USA
Chapters 2 and 18
STEPHEN I. VAS, M.D., Ph.D., Professor of Medical Microbiology, Toronto
Western Hospital, 399 Bathurst, Toronto, Canada M5T 2S8
Chapter 13
CHRISTIAN VERGER, M.D., Service de Medicine, Interne et Nephrologie, Centre
Hospitalier De Pontoise, 95301 Pontoise, France
Chapter 4
TONY WING, M.D., EDTA Registry, St. Thomas' Hospital, London, England
SE17EH
Chapter 19
1. History of peritoneal dialysis

S.T. BOEN

1. First period: 1923-1962

1.1. Early clinical experience

In 1923 an article by Ganter [1] was published in which he described intermittent

infusion and removal of saline solution into and from the peritoneal cavity of a
guinea pig made uremic by ureteral ligation. The urea-N concentration in the
fluid was close to the bloodconcentration after a dwell time of 1 h. After several
instillations and removals of fluid the animal improved.
Furthermore he stated briefly that 1.51 of saline solution was infused into the
peritoneal cavity of a woman with uremia due to ureter blockage by a uterus
carcinoma, and that he had the impression that there was an improvement in her
condition.
Ganter was of the opinion that the peritoneal membrane would also allow
passage of other kinds of toxic substances, as he observed a reversal of the
unconsciousness of a patient with diabetic coma after intraperitoneal infusion of 3
1 of saline solution.
Heusser and Werder [2] mentioned in 1927 that they performed peritoneal
dialysis in 3 patients; they noted that there was no clinical improvement because
the amount of fluid used was too small.
Prior to 1940 additional uremic patients were treated with peritoneal dialysis by
Balazs and Rosenak [3] in 3 cases, by Wear etal. [4] in 5 cases and by Rhoads [5] in
2 cases. One of Wear et al. 's patients recovered from the uremic state, and
tolerated an operation for bladder stones. Rhoads used peritoneal dialysis 3 times
in one patient, each session lasted 21/2h. Although a substantial amount of urea
nitrogen was removed the decline in blood urea nitrogen was small or absent.
From the many publications after 1946, we will select a few to illustrate the
developments in its clinical use.
Fine et al. [6] in 1946 used peritoneal dialysis in 4 patients, in one of them for 12
2

days. Pulmonary edema developed in 3 cases because of too much fluid admin-
istration by intravenous route; at the other hand using hypertonic dialysate, water
could be removed by peritoneal dialysis. They mentioned also the importance of
adjusting the dialysate composition in order to improve acidosis in the uremic
patient. Furthermore it was calculated that using 351 of dialysate with 2% glucose
concentration in 24 h, 200 to 300 g of glucose was absorbed in this period.
Derot et al. [7] reported in 1949 their first successful experience in acute renal
failure: 9 out of 10 patients survived compared to no survivors in 1947.
The duration of dialysis was between 5 and 240 h, using 2.25 to 1501 of fluid.
After their most recent experience it was advocated to dialyze for 24 to 36 h,
adding penicilline and sometimes streptomycine to the dialysate.
Legrain and Merrill [8] used peritoneal dialysis in 3 patients, in one of them 3
procedures were performed in a 2-week period after a renal transplant with
oliguria and hyperpotassemia. In another patient a practically sodium-free hyper-
tonic irrigating fluid with glucose was used for 7 h, removing about 1000 mEq of
sodium from a patient with marked edema due to nephrotic syndrome.
Odel et al. [9] collected 101 patients from the literature between 1923 through
1948: 63 had reversible lesions, 32 irreversible lesions and in 2 the diagnosis was
undetermined. Of the 63 patients with reversible renal diseases 32 recovered. The
cause of death was reported in 40 cases. Three complications accounted for the
death in 88% of the cases; in 13 cases (33%) death was caused by uremia, in 16
cases (40%) it was due to pulmonary edema and in 6 cases (15%) peritonitis was
the primary cause of death. It could be assumed that pulmonary edema was
brought about by use of an unbalanced perfusing fluid or by injudicious and
excessive use of parenteral fluid.
Grollman et al. [10] demonstrated that peritoneal dialysis can keep bilaterally
nephrectomized dogs alive for periods of 30 to 70 days. Furthermore 5 patients
with uremia were treated with intermittent peritoneal dialysis.
In the Netherlands peritoneal dialysis was first used by Formijne in 1946 in 2
patients with acute uremia [11].

1.2. Method and technique

1.2.1. Catheters
Usually the catheters used for peritoneal dialysis were improvised and adapted
from tubings available on the ward. Wear et al. [4] used a regular gallbladder
trocar for the inflow, and a trocar with numerous small holes in the distal third for
the outflow. Fine et al. [6] employed a rubber catheter or a perforated small
stainless steel tube as inlet tube, and a whistle-tip catheter or a large bore
mushroom-tip catheter as outlet tube. Because of frequent plugging, the outlet
tube later on was changed to a stainless steel sump-drain, which was similar to the
metal perforated suction tube commonly used in operating rooms.
 3

Derot et al. [7] and Legrain and Merrill [8] used polyvinyl chloride tubes with
small holes in the distal part of the catheter; this tube was inserted through a
trocar.
Grollman et al. [10] used polyethylene plastic tubes. Bassett et al. [12] used a
brass fenestrated tube as an outlet channel. Boen [13, 14] used rubber gastric
tubing with side holes as a peritoneal catheter. Doolan et al. [15] initially used
plastic gastric or nasal oxygen tubes in which additional holes were made. These
proved unsatisfactory, and they developed a polyvinyl chloride catheter with
transverse ridges to prevent kinking as well as to prevent blockage by the
omentum.
Maxwell et al. described a nylon catheter, 27.5 cm in length with multiple small
perforations at its distal 7.5 cm. It was slightly curved at the distal end with a
rounded solid tip. It fit into a 17 F. metal trocar [16]. This catheter became
commercially available and was widely used in the following years.

1.2.2. Technique
There are two techniques for peritoneal dialysis. With the continuous flow
technique two catheters are used, one in the upper abdomen and a second one in
the lower pelvis. Fluid is continuously infused through the upper catheter and is
drained out through the lower one. This technique was used by Heusser and
Werder [2] and by others [3, 6, 7, 8, 11].
With the intermittent technique only one catheter is used, which is placed with
its end in the small pelvis (lowest part of the abdominal cavity) to ensure good
removal of the fluid. The dialysate is run in and after a dwell time run out again
through the same tube, whereafter the cycle is repeated.
Abbott and Shea [17], Grollman et al. [10], Boen [13, 14], Doolan et al. [15],
Maxwell et al. [16] and others used this technique. It has the advantage of lesser
chance for leakage, for infection and of avoiding short-cut fluid channels.
Since 1950 nearly all clinicians have been using the intermittent technique.

1.3. The dialysate

1.3.1. Composition of the fluid

In the first years of peritoneal dialysis either normal saline solution or 5% glucose
solution was used as dialysate. Heusser and Werder [2] advised the use of saline
solution with the addition of 2 to 5% glucose to make the dialysate hypertonic.
Large shifts of water and minerals occurred during dialysis and the acidosis of
uremic patients was not corrected. Later lactate was added to the fluid as a source
of bicarbonate [5], and bicarbonate was part of the mineral composition when
Ringer's solution (2.4 mEq bicarbonate/I) or Tyrode's solution (12 mEq bicarbo-
nate/I) was used as dialysate. With higher bicarbonate concentrations a correction
of the patient's acidosis could be achieved. Abbott and Shea [17] added 26 mEq
4

bicarbonate per liter, Odel etal. [9] 24 to 36mEq/l, Grollman et al. [10] 35.8 mEq/
1, and Boen 35 to 40mEq/l [13, 14].
Commercial solution became available in 1959; it contains 35 to 40 mEq lactate/
l. In 1962 we started to use acetate as a source of bicarbonate in a concentration of
35 mEq/1 [18]; this is still being used in the Seattle area and some manufacturers
are also using acetate in the dialysate.
The sodium concentration in the fluid varies from 130 to 140 mEq/l, potassium 0
to 5 mEq/l, calcium 2 to 4 mEq/1 and magnesium 0 to 2 mEq/1.
Glucose concentration of the fluid is between 1.5 to 5 gm %.
Instead of glucose, 5% gelatin was added by Fine et al. [6] to make the fluid
hypertonic; this is not being used anymore.
In Table 1 the composition of different dialysates are seen.

1.3.2. Preparation of the fluid

Factory made fluid became available in 1959. Up till then the clinicians had to
make their own dialysate. Usually mineral solutions and glucose solutions were
mixed prior to dialysis, making a dialysate composition which can be adjusted to
the patient's need. More glucose was added when the patient was overhydrated,
and the potassium concentration was varied dependent upon the serum po-
tassium level.
In the early nineteen-fifties we used sterile distilled water in calibrated bottles
to which measured amounts of concentrated solutions of NaCl, NaHC0 3 , CaCI2 ,
MgCI2 , KCl and glucose were added immediately before use [13, 14].
Kop [19] made dialysate by sterilizing solutions in a container consisting of 2
compartments; a smaller one of 51 for NaCl, NaHC0 3 and KCl, and a larger one
of 281 for glucose, CaCl2 and HCl. The solutions were boiled for 30 min; after
cooling the solutions from the two compartments were mixed.
Caramelization of glucose had to be avoided; glucose could not be sterilized
together with all the minerals. Furthermore solutions containing both calcium
and bicarbonate could not be stored because of precipitation of calcium carbo-
nate. Once calcium was added to the dialysate, the fluid had to be used soon.
Maxwell et al. [16] in 1959 introduced the use of factory made fluid in one liter
bottles. Using a Y-connection, 2 bottles of fluid were infused simultaneously; the
empty bottles were used to receive the drained out dialysate. Instead of bicarbo-
nate, lactate was used as a source of bicarbonate. The availability of commercial
fluid has enhanced the wide-spread use of peritoneal dialysis.

1.4. Quantitative data of peritoneal dialysis

In earlier publications peritoneal dialysis seemed unable to improve the blood

chemistry of the uremic patient. Frequently the duration of dialysis was too short
(a few hours). Furthermore although sometimes the dialysis duration was long,
Table 1. Composition of irrigation fluid

NaCI Rmger Locke Tyrode Rhoads Hart- Abott Ferris Odel Groll- Boen Boen Tenck- Commercial
0,9% mann and man Amster- Seattle hoff RO
Shea dam machme 1 2 3 4
(CAPO) (CAPO)

Na mEqll 154 156.4 156.4 1664 276 130 131 139 143 134.5 140 135 130 130 140 132 134
K mEq/1 4 3.2 27 5 4 5 3 3 2.7 0-3 0-2
Ca mEq/1 4.5 7.6 3.6 4 4 4 2 2 3.6 3 3 3.5 40 40 3.0 3.5
Mg mEq/1 2.1 1 2 2 1.1 0-1.5 1.5 1.0 1.5 15 1.5 1.0
HC03 mEq/1 2.4 2.4 12 26 36 24 35.8 40
Cl mEq/1 154 162.5 164.8 162.4 257 110 114 109 109 106.1 103 107.5 96.5 101 101 101.5 105.5
Lactate mEq/1 28 28 35.0 35.0
Acetate mEq/1 35.0 38.0 35.0 45.0
Citrate mEq/1 16 3
Glucose Gmll 10-20 20 20 10-30 20 and 20 and 15 and 70 15 15,42.5 5,15
higher higher higher 42.5

Ul
6

the amount of fluid used was too small to produce significant removal of waste
products. The difference in outcome between hemodialysis and peritoneal di-
alysis lead Hamburger and Richet to state that only the artificial kidney can
correct the abnormalities in calcium, chloride and phosphate levels, and that this
can not be achieved with peritoneal dialysis [20].
Contrary to this statement, we demonstrated in 1959 that the same improve-
ment in bloodchemistry could be achieved provided a large amount of dialysate
was used and the dialysis duration is prolonged [21, 13]. Some data are seen in
Figure 1.

PRESENT STUDIES (1958) HAMBURGER and RICHET (1956)

C"ealinlne~ 246
UREA"~4.53
~134
~2.05Gm/L
mg/L
~[Link] 1ZZZZ12.04
150
mEeK

100
No CI

50

o
BEFORE AFTER NORMAL BEFORE AFTER NORMAL
PERITONEAL DIALYSIS (10) tlAEMODIALYSIS(60)

AIIERAGE DURATION: 36 H 4-6 H

01'" DIALYSIS
mEq/L
Na 130.5 136.3 137.3 136 138 142
Ca 3.8 4.5 5 3.9 5.6 5
Mg 3.3 2.2 1.6 2.6 1.6 15
K 6.9 4.3 4.1 6 5.1 5
CI 89.3 96.2 101.1 86 95.9 103
PROTEIN 14.5 15.2 17 16.8 15 17
HC03 15.7 24.7 26.4 19.5 22.1 27
P04 7.2 4.5 2 5.3 3 2
R- 17.8 6.8 1.5
504 12.2 15
5.S
ORGANIC ACIDS
8.7 8.5 3
(R- = 504 + [Link])

Figure 1. Comparison between peritoneal dialysis and hemodialysis.

 7

The clearance obtained with peritoneal dialysis is far lower than the hemo-
dialysis clearance. For instance, the peritoneal urea clearance on the average is
12 ml/min when 11 dialysate is cycled per hour and around 20 mllmin with 21/h,
25 ml/min with 3l1h and around 30 ml/min, with 41/h. There is a clear relationship
between flow rate and clearance value. Later studies by Tenckhoff et al. showed a
further increase of the urea clearance to 40 ml/min at a dialysate flow rate of 10 IIh
[22]. This value is about the limit which can be achieved by peritoneal dialysis.
Bomar et al. [23] and Villaroel [24] found good agreement of our data with their
mathematical models.
The diffusion curves for creatinine, uric acid and phosphate are lower than the
urea curves. Accordingly the peritoneal clearances of these substances are lower
than the urea clearance. These differences are similar to the artificial kidney.
However, the peritoneal membrane does also have areas with large pore sizes,
because all fractions of serum proteins do appear in the dialysate [13]. The inverse
relationship between the molecular weights of the plasma proteins and its per-
itoneal clearance was demonstrated in later years [25]. The permeability to
molecules with a molecular weight between 1500 and 5000 daltons was investig-
ated too [26].
Enhancement of peritoneal clearance by drugs and by osmolarity changes was
demonstrated in later years, but has not been integrated in clinical practice [27-
31].

2. The period of chronic intermittent peritoneal dialysis

2.1. Devices for access into the peritoneal cavity

When chronic peritoneal dialysis started to be used in 1962, it was felt that
frequent access into the peritoneal cavity should be made easier. In Seattle we
developed access devices made of teflon and silicone rubber tubes which were
implanted in the abdominal wall [18]. The catheter was inserted through this tube,
and after dialysis the tube was closed by a cap. Others also developed conduits
[32-34].
All efforts ended with failure because of peritonitis and formation of adhesions
which blocked the pathway of the catheter. After a few years this approach was
abandoned. An indwelling rod to provide a permanent tract for a catheter has
also been developed [35]; in our experience this kind of tract between the skin and
the peritoneal cavity gave too easy entry of bacteria although others have used it
for some period of time [36].
Subcutaneous peritoneal devices were designed and used in patients [37-39],
although not extensively. More investigation seems justified as this kind of device
may decrease the risk of peritonitis.
8

2.2. The repeated puncture technique

To prove that chronic peritoneal dialysis is possible, we had to eradicate per-

itonitis which was the limiting factor for long-term use of this method. In 1963 we
elected to abandon indwelling devices and to use repeated puncture for each
dialysis [40]. Initially a small trocar was used for insertion of a commercially
available small bore nylon catheter, but by the end of 1964 we used a stylet-
catheter [41]. The catheter was removed after each dialysis. Furthermore a closed
sterile system during an entire dialysis run was achieved using fluid in 40 1carboys
and an automatic machine.
The patient had complete freedom of movement in between dialyses, and it
was the first time that a patient could go on vacation-trips without any risk.
The pre-dialysis BUN concentrations varied between 100 and 150 mg%, the
serum creatinine level was around 15 mg% and the uric acid concentration was
9 mg%. The post dialysis values were: BUN usually around 50 mg% (sometimes
90mg%), serum creatinine around 9mg% and uric acid around 5mg%. The
hematocrit was about 30% without transfusions. The serum albumin levels were
in the low normal range. The protein loss was between 20 and 50 g/dialysis. The
blood pressure was well controlled without any hypotensive medication. Over a
2-year period the peritoneal urea clearance was measured periodically, the values
did not show a decrease. This was the first long-term successful experience with
peritoneal dialysis.
Tenckhoff demonstrated, that the repeated puncture technique with the use of
an automatic machine was also possible in the home-setting [42]. This method
was carried out for 31/ 2 yr at home, during which period 380 catheter punctures
were performed. Later on this patient was dialyzed using the silicone rubber
catheter devised by Tenckhoff. The repeated puncture technique was also suc-
cessfully used in 5 patients by Lasker et al. [43].

2.3. Automatic machines

One of the causes of peritonitis is contamination of the dialysate when changing

the bottles. Using fluid in one liter bottles during a 10-h run and a cycle volume of
41/h, 40 bottles had to be changed with as many chances for bacterial invasion. To
minimize this risk, we produced dialysate in 40-1 bottles and used a closed sterile
system throughout the entire dialysis. In order to minimize attendance by a nurse
for clamping and opening fluid lines, timers and clamps were incorporated in the
system to make the procedure automatic. The first kind of peritoneal dialysis
machine was constructed in Seattle in 1962 [18]. This machine is still being used at
the University Hospital in Seattle. A schematic drawing of the machine is seen in
Figure 2. From the 40-1 carboys the fluid is pumped into a head-tank from which it
goes into the patient by gravity flow. The outflow is collected in a sterile 40-1
 9

bottle. Timers control the inflow-time, dwelltime and outflow time. In later years
a camcycler was used instead of the clock-like timers [44].
To sterilize fluid in 40-1 bottles special equipment was required which could not
easily be installed in other places. Later commercial fluid in 2-1 bottles became
available; by connecting 4 bottles, a reservoir of 81 each time was obtained and a
cycler could be used for automatic dialysis [43]. Independently, we developed a
system with over 40 I of dialysate by connecting 3-1 bottles in series and using
timers for automated dialysis [45]. Mion connected 4 to 8 plastic containers with
10 I of fluid each in series for closed circuit peritoneal dialysis [46].
Shipping large amounts of fluid to the patient's home was cumbersome, and
there was a need for an apparatus which could make sterile dialysate in the home
of the patient. Although cold sterilization of water has been tried using a 0.22JL

2 IIttr capaclfy·· "

10 overf low

.. HOI plo ..
Vo!yltS VI. Vz Off'
Ivb. clomponq lyP'
·· SI •• I f,om.

SEOUENCE OF OPERATION
TIME
o VI [Link]. fluod 01 :n° C
pones Inlo pot l tnt
OrO In bOll ll'
:;, ,n v, cios.s
Roll., pump Slo,,,
30 mon [Link]
40 L
Vacuum pump Slorts
~5 m in Rollrr pump SlOPS
4'5 rTl1n V2 closes
Va cuum pump stops
VI optns

Figure 2. Schematic drawing with sequence of operation of the first automatic peritoneal dialysis
machine.
10

filter [47, 48], small viruses and bacterial breakdown products can still pass the
filter. Pyrogenic reactions and sterile peritonitis can follow (unpublished personal
data and ref. [49]). In 1969 Tenckhoff et al. [49] described a pressure boiler tank
with a capacity of 160 I which was used for on the spot sterilization of either mixed
dialysate or water. In the latter case the water was mixed in a 20 to 1 proportion
with mineral concentrate using a proportioning roller pump. In this way the cost
of dialysate could be reduced. A much refined version was used successfully both
in the home and in the hospital. The weight and bulkiness of the machine
represented major disadvantages. Advances in water treatment technology per-
mitted the development of a new system, which Tenckhoff et al. published in 1972
[50]. It incorporated a reverse osmosis filter to produce sterile, pyrogen free
water from tap-water which is mixed by a 20:1 proportioning (roller) pump with
mineral concentrate to make sterile dialysate. The machine was compact and
easily movable for home or hospital use. This reverse osmosis automatic machine
did increase the number of patients treated at home with peritoneal dialysis.

2.4. Indwelling catheters

Although the repeated puncture technique made long-term peritoneal dialysis a

success (by preventing peritonitis), the procedure was too time-consuming for the
physicians and could not be used on a larger scale. Furthermore, occasionally
bleeding occurred during puncture of the abdominal wall. The stiff nylon catheter
occasionally produced pain during dialysis.
Palmer et al. [51, 52] in 1964 devised an indwelling silicone rubber peritoneal
catheter; it was 84 cm long, with a lumen of 2 mm. The intraperitoneal end of the
tube was coiled and had many perforations extending 23 cm from the tip. Halfway
along the tube was a triflanged step for seating the tube in the deep fascia and
peritoneum. The rest of the tube was placed in a long spiral tunnel in the deep
subcutaneous tissue emerging from the skin surface in the left upper quadrant of
the abdomen. The catheter was sealed by a small cap if the patient was off
dialysis. The long subcutaneous part was meant to prevent extension of surface
infection.
Straight silicone rubber catheters were used by Gutch [53], and by McDonald et
al. [54], the latter incorporated a teflon velour skirt in the subcutaneous tissue and
a dacron-weaveknit sleeve from the skirt down to the peritoneum.
Tenckhoff's design of indwelling silicone rubber catheter [55] was accepted
widely since its publication in 1968 and became the most important factor in
promoting chronic peritoneal dialysis in other centers. Figure 3 shows the cathe-
ter. The silicone rubber tube had an internal diameter of 2.6 mm and an outer
diameter of 4.6 mm. The intra-abdominal section of the catheter was 20 cm long
and had 60 spaced perforations of 0.5 mm diameter in its terminal 15 cm; the end
of the tube remained open. One dacron cuff was bonded to the catheter just
 11

outside the peritoneum; the second dacron felt cuff was immediately beneath the
skin in the subcutaneous tissue. The distance between the 2 cuffs was 1Ocm, and
this part is placed in a curve in the subcutis. The external part of the catheter was
10 cm long. The dacron felt cuffs were designed to close the sinus tract around the
catheter against bacterial intrusion. The shorter subcutaneous part compared to
Palmer's catheter made it possible to implant the catheter through a specially
designed trocar. Furthermore if a catheter had to be removed because of infec-
tion, there was still space left for repeated puncture dialysis and reimplant at ion of
a new catheter.
Later modification included a balloon [56] and discs in the intra-abdominal part
to prevent easy dislocation of the catheter and omental wrapping [57]. A sub-
cutaneously implanted device with 2 tubes in the peritoneal cavity has also been
used [58]. With silastic catheters, the incidence of pain during dialysis is very low.
Although the Tenckhoff catheter is most advantageous from the practical point of
view, as an indwelling device it still carries the risk of peritoneal infection through
the lumen of the catheter or alongside the catheter. Good instructions for aseptic
technique remain imperative.

2.5. Clinical results

The use of the Tenckhoff catheter and automated machines enlarged the chronic
peritoneal dialysis program in the Seattle area. In 1973 Tenckhoff et al. reported
the experience of 12 000 peritoneal dialysis sessions in 69 patients, mostly at home
[59]. In 1977 in the Seattle area 161 patients had been on peritoneal dialysis, many
of them for over 4 yr and one patient for 8 yr [60].
The second largest population of intermittent peritoneal dialysis patients was in
Toronto, Canada. Oreopoulos reported on 150 patients in 1975 [61]. Other
centers in Europe and in the USA also reported satisfactory results (ref. [61]
through [81]) and peritoneal dialysis became an alternative method for treating
patients with end-stage renal disease.
However, real long-term treatment (over 4 yr) was not often achieved.
Ahmad et al. [82] calculated the cumulative technical survival rate for the
Seattle patients: 72% after 1 yr, 43% after 2 yr and 27% after 3 yr. Conversion to
hemodialysis because of complications and inadequate dialysis was common.
This reflects the usual experience in other centers. In contrast Diaz-Buxo re-
ported a survival rate of 86% after 1 yr, 83% after 2 yr and 80% after 3 yr; these
figures are comparable to hemodialysis survival rates [66].
Inadequate dialysis was one of the cases for conversion; it is therefore impor-
tant to adjust the duration of dialysis to the declining residual renal function.
Peritonitis did still occur, although the incidence was low when a closed sterile
system was used with automated dialysis.
An index for adequate peritoneal dialysis using the total creatinine clearance
12

Figure 3. Tenckhoff silicone rubber catheter with two dacron cuffs.

 13

(renal + peritoneal clearance) has been proposed [83, 84); we observed that
patients on peritoneal dialy~is did as well as patients on hemodialysis although the
small solutes removal was less than with hemodialysis [85]. The total creatinine
clearance in peritoneal dialysis is far less than in hemodialysis, but the clearance
for presumably toxic larger molecule weight substances ('middle molecules') is
higher.

3. Continuous ambulatory peritoneal dialysis (CAPD)

In 1976 Popovich, Moncrief et al. [86] submitted an abstract to the American

Society for Artificial Internal Organs describing what was called the equilibrium
peritoneal dialysis technique. Two liters of dialysate were infused intraperi-
toneally and allowed to equilibrate 5 h while the patient conducted his normal
activities. The dialysate was then drained and fresh fluid was run in again. Five
exchanges per day, 7 days per week were carried out. In 1978 experience in more
patients was reported by Popovich, Moncrief, Nolph et aI., and the name of the
method was changed to continuous ambulatory peritoneal dialysis or CAPD [87].
The dialysate was then only available in bottles in the USA, whereas in Canada
dialysate was delivered in plastic bags. Oreopoulos's modification of CAPD using
bags made the technique easier to perform and decreased (but did not eliminate)
the rather high incidence of peritonitis [88]. After inflow, the empty bag is folded
and carried on the body allowing free movement of the patient.
The peritoneal urea clearance with 5 exchanges of 21 each per day and 21 of
ultrafiltration per day is 8.4 mllmin. The clearance for larger molecules like Vit.
Bn and inulin, however, are higher than with hemodialysis or intermittent
peritoneal dialysis. For instance Vito B12 clearance values are: CAPD 5 exchanges
per day around 5 mllmin, intermittent peritoneal dialysis 40 h/week 1.6 mlimin,
hemodialysis 15h/week 3.0mllmin [89. 90].
In contrast to hemodialysis and intermittent peritoneal dialysis, the blood
chemistries of patients on CAPD is steady after a few weeks of treatment because
there is a constant removal of waste products from the body. This makes CAPD
the most physiologic way of dialysis. Because of sufficient fluid removal each day,
the fluid allowance of the patients is more liberal. The blood pressure is usually
well controlled. CAPD patients show a rise in hematocrit which is not seen with
other dialysis methods. The clinical condition of the patient is good provided
complications like peritonitis do not occur and the patient can eat sufficient
amount of protein to compensate for the protein loss with dialysis. Many more
reports about CAPD have been published [91-162].
Special indications are: dialysis of diabetic patients because of easier blood
sugar control (usually insulin is administered intraperitoneally) and dialysis of
children.
The interest for CAPD has grown tremendously in the last few years. Many
14

centers have started CAPD and many have expanded their program. By summer
1983 it was estimated that 8000 patients were on CAPD in the USA [136],
compared to 20 patients in summer 1977 and 1700 patients in summer 1980 [109].
On December 31, 1982 there were 1232 patients on peritoneal dialysis in Canada,
which was 34% of the total dialysis population. Of the new patients accepted in
1982, 50% were on CAPD [137]. In Europe 1837 patients started CAPD in 1982
and 2141 patients in the years before 1982 [138]. In most countries a further
increase of the number of CAPD patients took place in 1983.
According to the EDTA Registry report over the year 1982, for patients aged
45-54 excluding diabetics, the patient survival was around 82% at 2 years [138].
For the USA patient survival after one year was 78.1% for diabetics and 87.8%
for all others [136].
In Toronto Western Hospital the cummulative survival of diabetics at 4 years
was around 50% compared to around 70% for non-diabetics. For type I diabetics
the survival at 4 years was 62% as opposed to 27% for type II [139]. At the
Vancouver General Hospital the overall patient survival at 42 months was 88.9%
and the overall treatment survival was 37.5%. In the diabetic patients survival
was 67.5% at 36 months compared to 92.9% for the non-diabetics [140]. In
Montpellier, using intermittent peritoneal dialysis, the actuarial survival rate for
type I diabetics was 83% at 4 years and around 25% for type II patients [141].
Although its real long-term success is not yet known, it is clear now that CAPD
became a first choice initial treatment method for many patients. Technique
survival rate is improving, but sclerosing peritonitis with loss of ultrafiltration
capacity is probably an irreversible complication. This serious problem has been
seen mostly in France [142, 143].
New technical developments include the use of a bacterial filter in the fluid line
[144,145], the use of a sterile splice for connecting new dialysate bags [146], the
use of heat sterilization [147] or ultra violet light sterilization [148] for connec-
tions, the abandonment of carrying empty bags during dwell time [140, 149, 150,
151, 152], the introduction of new catheters like the curled Tenckhoff catheter
[153], the column disc catheter [154], a new type of balloon catheter [155] and a
catheter with a transcutaneous segment with a cuff and a flange made from
silicone elastomer and expanded polytetrafluoroethylene which allow tissue
ingrowth [156].
A modification of CAPD commonly known as continuous cyclic peritoneal
dialysis, CCPD, but which also has been called prolonged dwell time peritoneal
dialysis, PDPD, requires a peritoneal dialysis cycling machine [157, 158, 159].
Upon retiring at night the peritoneal catheter is connected to the cycler. The
machine is programmed to deliver 3 exchanges of 2 liters each, lasting 3 hours per
exchange. After the last drainage, an additional 2 liters of dialysate is infused.
The equipment is disconnected and the catheter capped. The fluid remains in the
peritoneal cavity during daytime. The procedure is repeated nightly.
Another modification is the so-called semi-continuous semi-ambulatory per-
 15

itoneal dialysis which can be carried out according to two schedules: (a) rapid
exchange of 8 liters of fluid late in the evening plus equilibration of 2 liters of
dialysate during nightime and daytime; (b) rapid exchange of 4 liters of dialysate
both in the morning and in the evening plus two equilibrations of 2 liters of
dialysate at daytime and at nighttime each [160].
High-volume (3 liters) low-frequency CAPD gave satisfactory results in pa-
tients who can tolerate such a volume [161, 162].
Better instructions and better training facilities have improved the results of
CAPD. Even so, a back-up hemodialysis facility is necessary with a CAPD
program, either for temporary or permanent takeover in case of complications or
inability to continue CAPD.
The resurgence of peritoneal dialysis has been remarkable and has stimulated
many people to search for new ways.
Undoubtedly further improvements can be expected in the near future.

References

1. Ganter G: Uber die Beseitigung giftiger Stoffe aus dem Blute durch Dialyse. Muench Med
Wochenschr 70(2): 1478-1480, 1923.
2. Heusser H, Werder H: Untersuchungen iiber Peritonealdialyse. Bruns Beitr Klin Chir 141:
38-49, 1927.
3. Balazs J, Rosenak S: Zur Behandlung der Sublimatanurie durch peritoneal Dlalyse Wien Klin
Wochenschr 47(2): 851-854, 1934.
4. Wear JB, Sisk IR, Trinkle AJ: Peritoneal lavage in the treatment of uremia. J Urol :39: 53-62,
1938.
5. Rhoads JE: Peritoneal lavage in the treatment of renal insufficiency. Am J Med Sci 196: 642-647,
1938.
6. Fine J, Frank H, Seligman AM: The treatment of acute renal failure by peritoneal irrigation.
Ann Surg 124: 857-875, 1946.
7. Derot M, Tanzet P, Roussillon J, Bernier 11: La dialyse peritoneale dans Ie traitement de
I'uremie aigue. J Uro155: 113-121, 1949.
8. Legrain M, Merrill JP: Short term continuous transperitoneal dialysis. N Engl J Med 248: 125-
129, 1953.
9. Ode I HM, Ferris DO, Power MH: Peritoneal lavage as an effective means of extrarenal
excretion. Am J Med 9: 63-77, 1950.
10. GroHman A, Turner LB, Mc Lean JA: Intermittent peritoneal lavage in nephrectomized dogs
and its application to the human being. Arch Intern Med 87: 379-390,1951.
11. Formijne P: De behandeling van de acute uraemie. Ned Tijdschr voor Geneeskd 90: 1181, 1946.
12. Basset SH, Brown HR, Keutmann EH, Holler J. Van Alstine HE, Mocejunas 0, Schantz H:
Nitrogen and fluid balance in treatment of acute uremia by peritoneal lavage. Arch Intern Med
80: 616-636, 1947.
13. Boen ST: Kinetics of peritoneal dialYSIS. Medicine 40: 243-287, 1961.
14. Boen ST: Peritoneal Dialysis in Clinical Medicine. Charles C Thomas Publ, Springfield, Illinois,
1964.
15. Doolan PD, Murphy WP, Wiggins RA, Carter NW, Cooper WC, Watten RH, Alpen EL: An
evaluation of intermittent peritoneal lavage. Am J Med 26: 831-844, 1959.
16. Maxwell MH, Rockney RE, Kleeman CR, Twiss MR: Peritoneal dialysis. JAMA 170: 917-924,
1959.
16

17. Abbott WE, Shea P: The treatment of temporary renal insufficiency by peritoneal lavage. Am J
Med Sci 211: 312-319, 1946.
18. Boen ST, Mulinari AS, Dillard DH, Scribner BH: Periodic peritoneal dialysis in the manage-
ment of chronic uremia. Trans Am Soc Artif Intern Organs 8: 25&-262, 1962.
19. Kop PSM: Peritoneale Dialyse (Thesis), Groningen, 1948.
20. Hamburger J, Richet G: Enseignements tires de la pratique du rein artificiel pour l'interpreta-
tion electrolytiques de l'uremie aigue. Rev Fr Etud Clin Bioll: 39-54, 1956.
21. Boen ST: Peritoneal Dialysis (Thesis). Univ of Amsterdam, 1959.
22. Tenckhoff H, Ward G, Boen ST: The influence of dialysate volume and flow rate on peritoneal
clearance. Proc Eur Dial Transpl ~ssoc 2: 113-117,1965.
23. Bomar JB, Decherd JF, Hlavinka DJ, MoncriefJW, Popovich RP: The elucidation of maximum
efficiency - minimum cost peritoneal dialysis protocols. Trans Am Soc Art Intern Organs 20:
120-129, 1974.
24. Villaroel F: Kinetics of intermittent and continuous peritoneal dialysis. J Dial 1: 333-347, 1977.
25. Bonomini V, Zucchelli P, Mioli V: Selective and unselective protein loss in peritoneal dialysis.
Proc Eur Dial Transpl Assoc 4: 14&-149, 1967.
26. Babb AL, Johansen PJ, Strand MJ, Tenckhoff H, Scribner BH: By directional permeability of
the human peritoneum to middle molecules. Proc Eur Dial Transpl Assoc 10: 247-257, 1973.
27. Henderson LW, Nolph KD: Altered permeability of the peritoneal membrane after using
hypertonic peritoneal dialysis fluid. J Clin Invest 48: 992-1001, 1969.
28. Nolph KD, Rosenfeld PS, Powell JT, Danforth E: Peritoneal glucose transport and hyper-
glycemia during peritoneal dialysis. Am J Med Sci 259: 272-281, 1970.
29. Nolph KD, Ghods AJ, Van Stone J, Brown PA: The effects of intraperitoneal vasodilators on
peritoneal clearances. Trans Am Soc Artif Intern Organs 22: 586b-591, 1976.
30. Maher JF, Shea C, Cassetta M, Hohnadel DC: Isoproterenol enhancement of peritoneal
permeability. J Dial 1: 319-331, 1977.
31. Gutman RA, Nixon WP, McRae RL, Spencer HW: Effect of intraperitoneal and intravenous
vasoactive amines on peritoneal dialysis. Trans Am Soc Artif Intern Organs 12: 570-573, 1976.
32. Merrill JP, Sabbaga E, Henderson L, Welzant W, Crane C: The use of an inlying plastic conduit
for chronic peritoneal irrigation. Trans Am Soc Artif Intern Organs 8: 252-255, 1962.
33. Barry KG, Shambaugh GE, Goler D, Matthews EF: A new flexible cannula and seal to provide
prolonged access to the peritoneal cavity for dialysis. Trans Am Soc Artif Intern Organs 9: 105-
107,1963.
34. Henderson LW, Merrill JP, Crane C: Further experience with the inlying plastic conduit for
chronic peritoneal dialysis. Trans Am Soc Artif Intern Organs 9: 108-116,1963.
35. Jacob GB, Deane N: Repeated peritoneal dialysis by the catheter replacement method: Descrip-
tion of technique and a replaceable prosthesis for chronic access to the peritoneal cavity. Proc
Eur Dial Transpl Assoc 4: 13&-140,1967.
36. Bigelow P, Oreopoulos DG, De Veber GA: Use of Deane prosthesis in patients on long-term
peritoneal dialysis. Can Med J 109: 999-1001, 1973.
37. Malette WG, McPhaul 11, Bledsoe F, Mcintosh DA, Koegel E: A clinically successful sub-
cutaneous peritoneal access button for repeated peritoneal dialysis. Trans Am Soc Artif Intern
Organs 10: 39&-398, 1964.
38. GotIoib L, Nisencorn J, Garmizo AL, Galili N, Servadio C, Sudarsky M: Subcutaneous
intraperitoneal prosthesis for maintenance peritoneal dialysis. Lancet 1: 1318-1319,1975.
39. GotIoib L, Mines M, Garmizo AL, Rodoy Y: Peritoneal dialysis using the subcutaneous
intraperitoneal prosthesis. Dial Transpl 8: 217-220, 1979.
40. Boen ST, Mion CM, Curtis FK, Shilipetar G: Periodic peritoneal dialysis using the repeated
puncture technique and an automatic cycling machine. Trans Am Soc Artif Intern Organs 10:
409-413, 1964.
41. Weston RE, Roberts M: Clinical use of stylet-Catheter for peritoneal dialysis. Arch Intern Med
115: 659-662, 1965.
 17

42. Tenckhoff H, Shilipetar G, Boen ST: One year's experience with home peritoneal dialysis.
Trans Am Soc Artif Intern Organs 11: 11-14,1965.
43. Lasker N, McCauley EP, Passarotti CT: Chronic peritoneal dialysis. Trans Am Soc ArtifIntern
Organs 12: 94-97, 1966.
44. Curtis FK, Boen ST: Automatic peritoneal dialysis with a simple cycling machine. Lancet 2: 620,
1965.
45. Bosch E, De Vries LA, Boen ST: A simplified automatic peritoneal dialysis system. Proc Eur
Dial Transpl Assoc 3: 362-365, 1966.
46. Mion C: A peritoneal dialysis program. Proc Eur Dial Transpl Assoc 12: 140-145, 1975.
47. McDonald HP: An automatic peritoneal dialysis machine: preliminary report. Trans Am Soc
Artif Intern Organs 11: 83-85, 1965.
48. Vercellone A, Piccoli G, Cavalli PL, Ragni R, Alloati S: A new automatic peritoneal dialysis
system. Proc Eur Dial Transpl Assoc 5: 344-347, 1968.
49. Tenckhoff H, Shilipetar G, Van Paasschen WH, Swanson E: A home peritoneal dialysate
delivery system. Trans Am Soc Artif Intern Organs 15: 103-107, 1969.
50. Tenckhoff H, Meston B, Shilipetar G: A simplified automatic peritoneal dialysis system. Trans
Am Soc Artif Intern Organs 18: 436-439, 1972.
51. Palmer RA, Quinton WE, Gray JF: Prolonged peritoneal dialysis for chronic renal failure.
Lancet 1: 700-702, 1964.
52. Palmer RA, Newell JE, Gray EJ, Quinton WE: Treatment of chronic renal failure by prolonged
peritoneal dialysis. N Engl J Med 274: 248-254, 1966.
53. Gutch CF: Peritoneal dialysis. Trans Am Soc Artif Intern Organs 10: 406-407, 1964.
54. McDonald HP, Gerber N, Mishra D, Wolin L, Peng B, Waterhouse K: Subcutaneous dacron
and teflon cloth adjuncts for silastic arteriovenous shunts and peritoneal dialysis catheters. Trans
Am Soc Artif Intern Organs 14: 176b-180, 1968.
55. Tenckhoff H, Schechter H: A bacteriologically safe peritoneal access device. Trans Am Soc
Artif Intern Organs 14: 181-186,1968.
56. Goldberg EM, Hill W, Kabins S, Levin B: Peritoneal dialysis. Dial Transpl4: 50-56,1975.
57. Oreopoulos DG: Overall experience with peritoneal dialysis. Dial Transpl 7: 783-787, 1978.
58. Stephen RL Atkin-Thor E, Kolff Wl: Recirculating peritoneal dialysis with subcutaneous
catheter. Trans Am Soc Artif Intern Organs 22: 575-584, 1976.
59. Tenckhoff H, Blagg CR, Curtis KF, Hickman RO: Chronic peritoneal dialysis. Proc Eur Dial
Transpl Assoc 10: 363-370, 1973.
60. Tenckhoff H: Advantages and shortcomings of peritoneal dialysis in the management of chronic
renal failure. Seminar Uro-Nephrologie Hopital Pitie, p 107-118, Paris, 1977.
61. Oreopoulos DG: Home peritoneal dialysis. Proc Eur Dial Transpl Assoc 12: 139, 1975.
62. Counts S, Hickman R, Garbaccio A, TenckhoffH: Chronic home peritoneal dialysis in children.
Trans Am Soc Artif Intern Organs 19: 157-163,1973.
63. Black HR, Finkelstein FO, Lee RV: The treatment of peritonitis in patients with chronic
indwelling catheters. Trans Am Soc Artif Intern Organs 20: 115-119, 1974.
64. Von Hartitzsch B, Medlock TR: Chronic peritoneal dialysis':' a regime comparable to con-
ventional hemodialysis. Trans Am Soc Artif Intern Organs 22: 595-597, 1976.
65. Diaz-Buxo lA, Chandler JT, Farmer CD, Smith DL: Chronic peritoneal dialysis at home - a
comparison with hemodialysis. Trans Am Soc Artif Intern Organs 23: 191-193,1977.
66. Diaz-Buxo JA, Haas VF: The influence of automated peritoneal dialysis in an established
dialysis program. Dial Transpl 8: 531-533, 1979.
67. Fenton SSA, Cattran DC, Barnes NM, Waugh KJ: Home peritoneal dialysis. A major advance
in promoting home dialysis. Trans Am Soc Artif Intern Organs 23: 194-200,1977.
68. Brouhard BH, Berger M, Cunningham RJ, Petrusick T, Allen W, Lynch RE, Travis LB: Home
peritoneal dialysis in children. Trans Am Soc Artif Intern Organs 25: 90-93,1979.
69. Roxe DM, Del Greco F, Krumlowsky F, Ghantous W, Hughes J, Ivanovich P, Quintanilla A,
18

Salkin M, Stone N: A comparison of maintenance hemodialysis to maintenance peritoneal

dialysis in the maintenance of end-stage renal disease. Trans Am Soc Artif Intern Organs 25:
81-85, 1979.
70. Dawids SG, Christensen E: Chronic home peritoneal dialysis with a simple dialysis system. Proc
Eur Dial Transpl Assoc 12: 149-152, 1975.
71. Giordano C, De Santo NG, Cirillo D, Capodicasa, Rinaldi S, Cicchetti T, Di Maio F: Short daily
peritoneal dialysis and protein restriction. Proc Eur Dial Transpl Assoc 12: 132-138, 1975.
72. Buoncristiani V: Clinical results of long-term peritoneal dialysis. Proc Eur Dial Transpl Assoc
12: 145-148, 1975.
73. Heal MR, England AG, Goldsmith HJ: Four year's experience with indwelling silastic cannulae
for long-term peritoneal dialysis. Brit Med J 2: 596-600, 1973.
74. Lankish PG, Tonnis HJ, Fernandez-Redo E, GirndtJ, KramerP, Quellhorst E, SchelerF: Use
of Tenckhoff catheter for peritoneal dialysis in terminal renal failure. Brit Med J 2: 712-713,
1973.
75. Brewer TE, Caldwell FT, Patterson RM, Flanigan WJ: Indwelling peritoneal (Tenckhoff)
dialysis catheter. JAMA 219: 1011-1015,1972.
76. Rae A, Pendray M: Advantages of peritoneal dialysis in chronic renal failure. JAMA 225: 937-
941,1973.
77. Blagg CR: Peritoneal dialysis and the Medicare ESRD program. Dial Transpl 8: 1081-1085,
1979.
78. Blumenkrantz MJ: Controlled evaluation of maintenance peritoneal dialysis. Dial Transpl 7:
797-799, 1978.
79. Mion C, Slingeneyer A, Oules R, Selam JL, Delors J, Mirouze J: Home peritoneal dialysis in
diabetics with end-stage renal failure. Contrib Nephroll: 120-130, 1979.
80. Hood SA, Frohnert PP, Mitchell JC, Kurtz SB: Home peritoneal dialysis: dialysis therapy of
choice in chronic renal failure of juvenile-onset diabetes mellitus. Dial Transpl9: 843--844, 1980.
81. Boen ST, Mion C, Slingeneyer A: The past 15 years: the role of peritoneal dialysis in the
treatment of end-stage renal disease. Proc Third Capri Uremia Coufe. Publ, Wichtig Ed, Milan,
1980.
82. Ahmad S, Gallagher N, Shen F: Intermittent peritoneal dialysis: status re-assessed. Trans Am
Soc Artif Intern Organs 25: 86-88, 1979.
83. Boen ST: Overview and history of peritoneal dialysis. Dial Transpl6: 12-18, 1977.
84. Boen ST, Haagsma-Schouten WAG, Birnie RJ: Long-term peritoneal dialysis and a dialysis
index. Dial Transpl 7: 377-381, 1978.
85. Scribner BH, Fergus EB, Boen ST, Thomas ED: Some therapeutic approaches to chronic renal
insufficiencies. Ann Rev Med 16: 285-300, 1965.
86. Popovich RP, Moncrief JW, Decherd JF, Bomar JB, Pyle WK: The definition of a novel
portable-wearable equilibrium peritoneal technique. Abst Am Soc ArtifIntern Organs 64,1976.
87. Popovich RP, Moncrief JW, Nolph KD, Ghods AJ, Twardowski ZJ, Pyle WK: Continuous
ambulatory peritoneal dialysis. Ann Intern Med 88: 449, 1978.
88. Oreopoulos DG, Robson M, Izatt G, Clayton S, De Veber GA: A simple and safe technique for
continuous ambulatory peritoneal dialysis (CAPD). Trans Am Soc Artif Intern Organs 24: 484-
489, 1978.
89. Popovich RP, MoncriefJW: Kinetic modeling of peritoneal transport. Contr Nephrol17: 58-72,
1979.
90. Popovich RP: Physiological transport parameters in patients. Dial Transpl 7: 823-824, 842,
1978.
91. Moncrief J, Nolph KD, Rubin J, Popovich RP: Additional experience with continuous ambula-
tory peritoneal dialysis (CAPD). Trans Am Soc Artif Intern Organs 24: 476-483, 1978.
92. Hiatt MP, Pyle WK, Moncrief JW, Popovich RP: A comparison of the relative efficacy of CAPD
and hemodialysis in the control of solute concentration. Artif Organs 4: 37-43, 1980.
 19

93. Gill D, Morgan J , Ryan B, Gault MH, Churchill D N: Role of continuous ambulatory peritoneal
dialysis in a rural population. Dial Transpl8: 1182-1183, 1979.
94. Oreopoulos DG, Clayton S, Dombros N, Zellerman G, Katirtzoglou A: Experience with
continuous ambulatory peritoneal dialysis (CAPD). Trans Am Soc Artif Intern Organs 25:
96-97, 1979.
95. Oreopoulos DG: The coming age of continuous ambulatory peritoneal dialysis. Dial Transpl8:
460-517,1979.
96. RubinJ, Arfania D, Nolph KD, Prowant B, Fruto L, Brown P, Moore H: Peritoneal clearances
after 6-12 months on continuous ambulatory peritoneal dialysis. Trans Am Soc Artif Intern
Organs 25: 104-108, 1979.
97. Fenton SSA, McCready W, Cattran DC, Oreopoulos DG, Whiteside C: Selected clinical aspects
of continuous ambulatory peritoneal dialysis. Proc Symp On CAPD, Paris. Excerpta Medica,
Amsterdam, 1979, pp 107-112.
98. Price JDE, Moriarty MV: Continuous ambulatory peritoneal dialysis: selection criteria - failures
and causes - deaths in diabetes mellitus. Proc Symp on CAPD, Paris. Excerpta Medica,
Amsterdam, 1979, pp 113-119.
99. Gahl GM: Medical management of continuous ambulatory peritoneal dialysis. Proc Symp on
CAPD, Paris. Excerpta Medica, Amsterdam, 1979, pp 181-186.
100. Lindholm B, Ahberg M, Alvestrand A, Furst P, Karlander SG, Bergstrom J: Nutritional aspects
of continuous ambulatory peritoneal dialysis. Proc Symp on CAPD, Paris. Excerpta Medica,
Amsterdam, 1979, pp 199-206.
101. Lameire N, Ringoir S: An overview of peritonitis and other complications of continuous
ambulatory peritoneal dialysis. Proc Symp on CAPD, Paris. Excerpta Medica, Amsterdam,
1979, pp 229-237.
102. Gokal R, Freyer R, McHugh M, Ward MK, Kerr DNS: Calcium and phosphate control in
patients on CAPD. Proc Symp on CAPD, Paris. Excerpta Medica, Amsterdam, 1979, pp 283-
291.
103. Mion C: Maintenance hemodialysis versus intermittent peritoneal dialysis versus continuous
ambulatory peritoneal dialysis. Proc Symp on CAPD, Paris. Excerpta Medica, Amsterdam,
1979, pp 317-327.
104. Baillod RA: Continuous ambulatory peritoneal dialysis versus intermittent peritoneal dialysis at
the Royal Free Hospital. Proc Symp on CAPD, Paris. Excerpta Medica, Amsterdam, 1979, pp
328--334.
105. Legniin M, Jacob C: Place of chronic ambulatory peritoneal dialysis in the treatment of end-
stage renal failure. Proc Symp on CAPD, Paris. Excerpta Medica, Amsterdam, 1979, pp 347-
353.
106. Nolph KD, Parker A: The composition of dialysis solutions for continuous ambulatory per-
itoneal dialysis. Proc Symp on CAPD, Paris. Excerpta Medica, Amsterdam, 1979, pp 341-346.
107. Thomson NM, Walker RG, Whiteside G, Scott DF, Atkins RC: Continuous ambulatory
peritoneal dialysis (CAPD) in the treatment of end-stage renal failure. Proc Eur Dial Transpl
Assoc 16: 171-177,1979.
108. Farrell PC, Randerson DH: Mass transfer kinetics in continuous ambulatory peritoneal dialysis.
Proc Symp on CAPD, Paris. Excerpta Medica, Amsterdam, 1979, pp 34-41.
109. Moncrief J: CAPD Experience in USA (1980). Proc Panpacific Symp on Peritoneal Dialysis.
Churchill, Melbourne, 1981, pp 165-170.
110. Lindholm B, Bergstrom J, Karlander SG: Glucose metabolisms in patients on CAPD. Trans
ASAIO 27: 58--60, 1981.
111. Potter DE, McDaid K, McHenry K, Mar H: Continuous ambulatory peritoneal dialysis in
children. Trans ASAIO 27: 64-17, 1981.
112. Evans DH, Sorkin MI, Nolph KD, Whittier FC: Continuous ambulatory peritoneal dialysis and
transplantation. Trans ASAIO 27: 320-324, 1981.
20

113. Chan MK, Chuah P, Raferty MJ, Baillod RM, Sweny P, Varghese Z, Moorhead JF: Three
year's experience of CAPD. Lancet 1: 1409,1981.
114. Gokal R, Ramos JM, Vertch P, Proud G, Taylor RMR, Ward MK, Wilkinson R, Kerr DNS:
Renal transplantation in patients on continuous ambulatory peritoneal dialysis. Proc Eur Dial
Transpl Assoc 18: 222-227, 1981.
115. Oreopoulos DG, Khanna R, Williams P et al.: Continuous ambulatory peritoneal dialysis.
Nephron 30: 293-303, 1981.
116. Giangrande A, Cantie P, Limido A, de Francesco D, Malacrida V: Continuous ambulatory
peritoneal dialysis and cellular immunity. Proc Eur Dial Transpl Assoc 19: 372-379, 1982.
117. Luce E, Nakagawa D, Lovell J, Davis J, Stonebaugh BJ, Suki WN: Improvement in the
bacteriologic diagnosis of peritonitis with the use of blood culture media. Trans ASAIO 28: 259-
262,1982.
118. Lindholm B, Tegner R, Tranaeus A, Bergstrom J: Progress of peripheral uremic neuropathy
during CAPD. Trans ASAIO 28: 263-269, 1982 ..
119. Fabris A, Biasioli S, Chiaramonte C, Feriani M, Prisani E, Ronco C, Cantarella G, La Greca G:
Buffer metabolism in CAPD: relationship with respiratory dynamics. Trans ASAIO 28: 270-
275,1982.
120. Rottembourgh J, Issad B, Gallego JL, Degoulet P, Aime F, Gueffaf B, Legrain M: Evolution of
residual renal function in patients undergoing hemodialyses or CAPD. Proc Eur Dial Transpl
Assoc 19: 397-403, 1982.
121. Kurtz SB, Wong VH, Anderson CF, Vogel JP, McCarthy JT, Mitchell III JC: Continuous
ambulatory peritoneal dialysis. Three year's experience at the Mayo Clinic. Mayo Clin Proc 58:
633-639, 1983.
122. Prowant B, Ryan L, Nolph KD: Six years experience with peritonitis in a CAPD program. Perit
Dial Bull 3: 199-200, 1983.
123. Broyer M, Niaudet P, Champion G, Jean G, Chopin N, Czernichow P: Nutritional and
metabolic studies in children on continuous ambulatory peritoneal dialysis. Kidney Int 24
(Suppl) 15: 106--110, 1983.
124. Solusky JB, Kopple JD, Fine RN: Continuous ambulatory peritoneal dialysis in paediatric
patients. Kidney Int 24 (Suppl) 15: 101-105, 1983.
125. Garred U, Canaud B, Farrell PC: A simple kinetic modelling for assessing peritoneal mass
transfer in chronic ambulatory peritoneal dialysis. ASAIO 6: 131-137, 1983.
126. Blumberg A, Hanck A, Sander G: Vitamin Nutrition in patients on CAPD. Clin Nephrol20:
244-250, 1983.
127. Lameire N, Dhaene M, Matthys E, De Paepe M, Vereerstraeten P, Dratwa M, Ringoir S:
Experience with CAPD in diabetic patients. Proc Symp on Prevention and Treatment of
Diabetic Nephropathy, Paris. MTP, Lancaster, 1983, pp 289-297.
128. Nolph KD, Boen ST, Farrell PC, Pyle KW: Continuous ambulatory peritoneal dialysis in
Australia, Europe and the United States: 1981. Kidney Int 23: 3-8, 1983.
129. Thomson NM, Stevens BJ, Humphery TJ, Atkins RC: Comparison of trace elements in
peritoneal dialyses, hemodialyses and uremia. Kidney Int 23: 9-14, 1983.
130. Gokal R, Ramos JM, Ellis HA, Parkinson I, Sweetman V, Dewar J, Ward MK, Kerr DNS:
Histological renal osteodystrophy and 25 hydroxycholecalciferol and aluminium levels in pa-
tients on CAPD. Kidney Int 23: 15-21, 1983.
131. Widerbe E, Smeby LC, Berg KJ, Jorstad S, Svartas TM: Intraperitoneal insulin absorption
during intermittent and continuous peritoneal dialysis. Kidney Int 23: 22-28, 1983.
132. Von Baeyer H, Gahl GM, Riedinges H, Borowzak R, Averdunk R. Schurig R, Kessel M:
Adaptation of CAPD patients to continuous peritoneal energy uptake. Kidney Int 23: 29-34,
1983.
133. Rottembourg, J, El Shahat Y, Agrafiotis A, Thuillier Y, De Groc F, Jacobs C, Legrain M:
Continuous ambulatory peritoneal dialysis in insuline-dependent diabetic patients: a 40 month
experience. Kidney Int 23: 40-45, 1983.
 21

134. Zucchelli P, Chiarini C, Esposito ED, Fabbri L. Santoro A, Sturani A, Zuccalia A: Influence of
CAPD on the autonomic nervous system. Kidney Int 23: 46--50. 1983.
135. Vas SI: Microbiologic aspects of chronic ambulatory peritoneal dialysis. Kidney Int 23: 83-92,
1983.
136. Nolph KD: Results of the USA CAPD Registry. Abstract, 4th ISAO Official Satellite Symp on
CAPD. Kyoto, Japan, November 1983.
137. Posen G, Lam E, Rapaport A: CAPD in Canada 1982. Abstract, 4th ISAO Official Satellite
Symp on CAPD. Kyoto, Japan, November 1983.
138. Wing AJ, Broyer M, Brunner FP, Brynger H, Challah S, Donckerwolcke RA, Gretz N, Jacobs
C, Kramer P, Selwood NH: Combined report on regular dialysis and transplantation in Europe,
XIII, 1982. Proc Eur Dial Transpl Assoc 20: 5-67, 1983.
139. Khanna R, Wu G, Chisholm L, Oreopoulos DG: Further experience with CAPD in diabetics
with end-stage renal disease. Proc Symp on Prevention and Treatment of Diabetic Nephropathy,
Paris. MTP Press, Lancaster, 1983, pp 279-288.
140. Mavichak V, Moriarty MV, Cameron EC, Reeve CE, Bailon HS, LauenerRW, PriceDJ: Three
and a half years experience with CAPD using the beta cap technique. Trans ASAIO 28: 253-258,
1982.
141. Mion C, Slingeneyer A, Canaud B, Oules R, Branger B, Chong G, Mourad G: Home intermit-
tent peritoneal dialysis in the treatment of end-stage diabetic nephropathy: 1982 update. Proc
Symp on Prevention and Treatment of Diabetic Nephropathy. Paris. MTP Press, Lancaster,
1983, pp 263-277.
142. Slingeneyer A, Mion C, Mourad G, Canaud B, Faller B, Beraud 11: Progressive sclerosing
peritonitis: a late and severe complication of maintenance peritoneal dialysis. Trans ASAIO 29:
633-640, 1983.
143. Rottembourg J, Gahl GM, Poignet JL, Mertani E, Strippoli P, Langlois P, Tranbaloc P, Legrain
M: Severe abdominal complications in patients undergoing continuous peritoneal dialysis. Proc
Eur Dial Transpl Assoc 20: 236--241, 1983.
144. Slingeneyer A, Mion C: Peritonitis prevention in continuous ambulatory peritoneal dialysis:
long-term efficacy of a bacteriological filter. Proc Eur Dial Transpl Assoc 19: 388-395, 1982.
145. Winchester JF, Ash SR, Bousquet G, Rakowski TA, Barnard WR, Heeter E, Haley S:
Successful peritonitis reduction with a unidirectional bacteriological CAPD filter. Trans ASAIO
29: 611-616, 1983.
146. Hamilton R, Adams P, Burkart J, Disher B, Dillingham E, Crater C: Feasibility of a sterile
splice for connection in CAPD. Trans ASAIO 29: 623-628, 1983.
147. Di Paolo N, Buoncristiani V: Automatic peritoneal dialysis. Nephron 35: 248-252, 1983.
148. Moncrief JW, Mullins-Blackson C, Le Bourglois J, Popovich RP, Pyle K: Development and
testing of an ultraviolet light resterilizing procedure for CAPD. Abstract 4th ISAO Official
Satellite Symposium on CAPD. Kyoto, Japan, November 1983.
149. Bazzato G, Coli U, Landini S et al.: CAPD without wearing a bag: complete freedom of patient
and significant reduction on peritonitis. Proc Eur Dial Transpl Assoc 17: 266--275, 1980.
150. Buoncristiani V, Bianchi P, Cozzari M et al.: A new safe simple connection system for CAPD.
Int J Nephrol Urol Androll: 50-53, 1980.
151. Buoncristiani U, Di Paolo N: Autosterilizing CAPD connection systems. Nephron 35: 244-247,
1983.
152. Maiorca R, Cancarini GC, Brocolli R, Brasa S, Cantaluppi A, Scalamogna A, Graziani G,
Ponticelli C: Prospective controlled trial of a Y-connector and disinfectant to prevent peritonitis
in CAPD. Lancet 2: 642-644,1983.
153. Rottembourg J, De Groc F: Peritoneal access using the curled Tenckhoff catheter. Perspectives
in Peritoneal Dialysis 1: 7-8, 1983.
154. Ash SR, Slingenery A, Schardin KE: Peritoneal access using the column-disc catheter. Perspec-
tives in Peritoneal Dialysis 1: 9-11,1983.
22

155. Valli A, Comotti C, Torelli D, Crescimanno U, Valentini A, Riegler P, Huber W, Borghi M,

Gruttadauria C, Scarovanat P, Pecchini F: A new catheter for peritoneal dialysis. Trans ASAIO
29: 629--632, 1983.
156. Ehrlich LF, Powell SL: Care of the patient with a Gore-Tex peritoneal dialysis catheter. Dial
Transpl12: 572, 1983.
157. Diaz-Buxo JA, Walker PJ, Farmer DF, Chandler JT, Holt KL, Cox P: Continuous cyclic
peritoneal dialysis. Trans ASAIO 27: 51-53, 1981.
158. Price CG, Suki WN: Newer modification of peritoneal dialysis. Am J Nephroll: 97-104, 1981.
159. Nakagawa D, Price C, Stinebaugh B, Suki W: Continuous cycling peritoneal dialysis: a viable
option in the treatment of chronic renal failure. Trans ASAIO 27: 55-57, 1981.
160. Buoncristiani V, Cozarri M, Carobi C, Quintaliani G, Barbarossa D, Di Paolo N: Semicon-
tinuous semiambulatory peritoneal dialysis. Proc Eur Dial Transpl Assoc 17: 328-332, 1980.
161. Twardowski ZJ, Prowant BF, Nolph KD, Martinez AJ, Lampton LM: High volume low
frequency CAPD. Kidney Int 23: 64-70, 1983.
162. Twardowski ZJ, Nolph KD, Prowant BF, Moore HL: Efficiency of high volume low frequency
continuous ambulatory peritoneal dialysis. Trans ASAIO 29: 53-57, 1983.
2. The peritoneal dialysis system

KARL D. NOLPH and ZBYLUT J. TWARDOWSKI

1. Introduction

The peritoneal dialysis system can be considered as nature's version of a capillary

kidney [1]. Peritoneal dialysis probably represents solute and fluid exchange
mainly between peritoneal capillary blood and dialysis solution in the peritoneal
cavity [2]. The dialysis membrane consists of the vascular wall, the interstitium,
the mesothelium, and adjacent fluid films [1-5]. In this chapter, we will review the
anatomy of the peritoneum and the physiology of peritoneal transport. We will
also compare the peritoneal dialysis system to man-made hollow fiber dialyzers.
The features of the latter have been well characterized and are very familiar to
most nephrologists. Comparison of peritoneal and hollow fiber dialysis should
help the reader appreciate some of the unique characteristics of the peritoneal
dialysis system.
The peritoneal membrane covers visceral organs, forms the visceral mesentery
that connects loops of bowel, and reflects over and covers the inner surface of the
abdominal wall. This membrane is continuous and the closed space within
contains small amounts of fluid (probably less than 100 ml) under normal condi-
tions. This space can be enlarged by the instillation of fluid. Most normal-sized
adults can tolerate 2 or more liters of fluid without discomfort. A thin layer of
mesothelial cells covers the surface of the membrane lining the cavity. Beneath
the mesothelial layer there is interstitium containing extracellular fluid, con-
nective tissue fibers, blood vessels and lymphatics.
Visceral peritoneum courses over the surface of visceral organs. Visceral
mesentery between adjacent loops of bowel is formed as the visceral peritoneum
reflects over loops of bowel and consists of two layers of mesothelial cells with
interstitium interspersed between these layers.
The inner surface of the abdominal wall is covered by the parietal peritoneum.
The parietal peritoneum receives its blood supply from the arteries of the abdomi-
nal wall.
Visceral mesentery contains mainly large blood vessels on their way to visceral
24

Table 1. Indirect evidence that peritoneal capillary blood is a major source of solutes, cells, and water
removal during peritoneal dialysis

1. Hypotension with rep~ated hypertonic exchanges

2. Decreased clearances with hypotension
3. Decreased clearances with vasoconstrictors
4. Increased clearances with vasodilators
5. Drugs known to increase protein leaking from venules increase protein losses during peritoneal
dialysis
6. Decreased clearances with vasculitis
7. Decreased clearances with diabetic vascular disease
8. Dialysate potassium concentrations approach Gibbs-Donnan equilibrium with serum, not with
intracellular fluid
9. Convective removal of potassium per liter of uItrafiltrate does not exceed extracellular
concentrations
10. Limited pools of fluid and solutes in peritoneal mesothelium and interstitium-quickly exhausted
without rapid replacement
11. Lymphatic flow presumably quite low-drainage not chylous
12. Dialysate leukocyte counts and fibrin increase rapidly with inflammation

organs. At the point where blood vessels reflect over the loops of bowel and
divide into smaller vessels, arteriolar or capillary beds capable of participating in
peritoneal dialysis exchange have been noted on the bowel surface. Many
lymphatics are present in visceral mesentery but the extent of their participation
in peritoneal dialysis solute and water exchange is unknown. Characteristics of
the peritoneal microcirculation will be reviewed in the chapter by Miller.
The total gross surface area of the combined parietal and visceral peritoneal
mesothelium is thought to approximate the surface area of the skin (1-2 square
meters in most adults). The exact ratio of parietal to visceral peritoneal surface
area is unknown. Visceral mesentery represents a larger fraction of total per-
itoneal surface area because of its many folds. However, portions of the parietal
peritoneum may be more vascular than some of the nearly avascular sections of
mesentery. Thus, true vascular contributions of parietal and visceral peritoneum
to solute transport are unknown.

2. Evidence that peritoneal dialysis is primarily hemodialysis

Table 1 summarizes the indirect evidence that peritoneal capillary blood is the
major source of solutes, cells, and water removed during peritoneal dialysis. It
should be stressed that most of the evidence is indirect.
Hypertonic peritoneal dialysis solution containing 4.25% dextrose can gener-
ate net ultrafiltration in excess of 500 ml/h [6]. Many liters per day of ultrafiltra:"
tion can be tolerated with the rather rapid resolution of edema. If hypertonic
exchanges are interspersed so as to avoid severe hyperglycemia or hypotension,
 25

net ultrafiltration per hypertonic exchange can remain quite consistent [6]. It
seems unlikely that mesothelial cells, interstitium or lymphatics could yield net
ultrafiltration of this magnitude over short periods of time. It seems more
reasonable to assume that ultrafiltrate is primarily a capillary ultrafiltrate. Also,
dramatic reductions in blood pressure following one or two hypertonic exchanges
can sometimes be observed suggesting that net ultrafiltration without adequate
mobilization of edema fluid can jeopardize blood volume.
Hypotension can result in decreased peritoneal clearances [18]. Although such
reductions in clearances are often modest even in severe shock (for reasons to be
discussed below) the findings do suggest that solute clearances are affected by
peritoneal capillary blood flow [7-9].
Intraperitoneal or systemic vasoconstrictors have been shown to reduce per-
itoneal clearances [10-11]. Vasoconstrictors are known to decrease the number of
peritoneal capillaries perfused as well as peritoneal capillary blood flow [12].
Decreased clearances with vasoconstrictors supports the conjecture that the
status of the microcirculation influences peritoneal clearances.
Peritoneal clearances increase with intraperitoneal vasodilators [12-18]. These
agents increase peritoneal capillary blood flow as well as the number of capillaries
perfused [14-16]. Vasodilatation and/or direct effects of the agents used may
increase capillary permeability [19]. The point to be made here, however, is that
vasoactive agents affect peritoneal clearances in the expected direction if cleara-
nces relate directly to blood flow, numbers of capillaries, and vascular permeabil-
ity.
Wayland has shown that histamine applied topically to the rat peritoneum
widens the intercellular gaps in small venules [19]. His techniques have included
serial section studies with electron microscopy, computerized reconstruction of
venular intercellular gaps, and direct observations of the movement of fluores-
cent tagged albumin across the walls of small vessels in the rat mesentery. The
latter technique utilizes a laser beam microscope developed by Dr Wayland. With
this device an outpouring of albumin from small venules can be seen following a
topical application of histamine to rat mesentery. Miller and co-workers have
reported similar findings with nitroprusside [20]. In clinical studies, the addition
of nitroprusside to peritoneal dialysis solution markedly increases protein losses
[13-18].
Patients with severe systemic vasculitis, presumably involving the peritoneal
microcirculation have been reported to have significantly reduced peritoneal
clearances [21, 23]. To date this includes reports of reduced clearances with
systemic lupus erythematosus, scleroderma, and malignant hypertension.
Some patients with wide spread diabetic vascular disease have been found to
have significantly reduced clearances [21]. This is not a universal finding in all
diabetics but may relate to the basement membrane thickening and vascular
disease as it exists in the peritoneum.
The concentration of potassium in the intracellular fluid of mesothelial cells is
26

near 140 mEq/l [24-26]. Nevertheless, dialysis solution in the peritoneal cavity
approaches Gibbs-Donnan equilibrium with the potassium concentration in
serum water [27-28]. Since dialysis solution is similar in composition to extra-
cellular fluid it is not surprising that the mesothelial cells would maintain their
normal internal milieu even though bathed with dialysis solution. However, the
fact that intracellular electrolytes do not participate in peritoneal dialysis ex-
change to any great extent does not rule out the possibility that some creatinine
and urea are removed from intracellular fluid.
Using hypertonic solutions and thus generating net ultrafiltration, solutes can
be removed by convection in the absence of a concentration gradient for net
diffusion [29-32]. The net removal of potassium by convection per liter of
ultrafiltrate does not appear to exceed amounts of potassium in extracellular fluid
[27, 33]. Thus even hypertonic exchanges do not appear to mobilize much, if any,
intracellular potassium.
Certainly it is likely that diffusible solutes are removed from peritoneal inter-
stitium and perhaps to some extent from mesothelial intracellular fluid [1-4, 19,
34]. Ultrafiltrate would, of course, involve water movement through the inter-
stitium and perhaps to some extent through or from mesothelial cells [31].
However, mesothelial cells could only tolerate a modest degree of dehydration
and interstitial pools of water and solute would quickly be exhausted without
rapid replacement from peritoneal capillaries. The point is that most of the water
and solutes removed during peritoneal dialysis must represent water and solute
movement from peritoneal capillaries into the peritoneal cavity by way of path-
ways through the interstitium and the mesothelial layer .
Solutes and water could move into the peritoneal interstitium from peritoneal
lymphatics [34]. It is unknown what portion of net removal of solutes or water
comes from peritoneal lymphatics. This has been assumed to be of minor import-
ance since lymphatic flow rates are presumably low and drainage is not usually
chylous. We have followed one patient on continuous ambulatory peritoneal
dialysis whose drainage after an episode of streptococcal peritonitis obviously
contained lymph for nearly three years. Drainage was milky, particularly after
meals, and contained high triglyceride concentrations. There was no evidence of
inflammation (dialysate white counts were low and there were no symptoms).
This particular finding however is extremely unusual.
Additional evidence that peritoneal capillary blood can contribute rapidly and
significantly to what is removed in peritoneal dialysis solution is the finding that
with infection dialysate leukocyte counts can increase over several hours from
less than 100 to many thousands of white cells/mm 3 [35-36]. Also it would appear
that an outpouring of fibrinogen and the formation of fibrin in dialysate can occur
quite quickly in the presence of inflammation.
Thus, indirect evidence supports the hypothesis that peritoneal dialysis repre-
sents fluid and solute exchange between peritoneal capillary blood and dialysis
solution in the peritoneal cavity. The capillary, endothelium, peritoneal inter-
 27

stitium, and mesothelium represent the resistance sites which must be crossed by
fluid and solutes to result in a net exchange.

3. Peritoneal capillary blood flow

The absolute peritoneal capillary blood flow which participates in peritoneal

dialysis exchange is unknown. In adult humans, total splanchnic blood flow may
exceed 1200 mllmin [37]. However, most of this blood flow is on its way to visceral
organs and not to the small vessels of the peritoneum. In fact our observations of
the rat peritoneum would suggest that the mesentery is not particularly vascular
and that most of the small vesseJs capable of participating in exchange may be
located at those sites where the peritoneum reflects over loops of bowt:-l [38-39].
In adult humans maximum urea clearances usually do not exceed 40 mllmin
even with the most rapid cycling [40-42]. One possible explanation has been that
effective peritoneal capillary blood flow may not exceed 30 to 40 mllmin and
maximum urea clearances are approaching effective peritoneal capillary flow.
There is abundant indirect evidence to suggest that this is not the case. Table 2
summarizes indirect evidence that maximum peritoneal urea clearances are not
primarily blood flow limited.
First in animal studies it has been shown that urea clearances remain above
70% of control even with severe shock and 38% reductions in splanchnic blood
flow [7]. Although the magnitude of change in the effective peritoneal capillary
flow from control to shock conditions is unknown, this observation would suggest
that effective peritoneal capillary flow is well above urea clearance in the control
state and only falls into a modest flow limiting range with severe hypotension.
Secondly, urea clearances increase only modestly (usually less than 20% ) with
intraperitoneal vasodilators [13-18]. Vasodilators are also known to influence the
number of capillaries perfused, induce venodilation and in some instances di-
rectly alter vascular permeability [12, 38-39]. Since these latter effects might also
account for the modest increases in small solute clearances these findings suggest
that any increases in effective capillary flow with vasodilators have little or no
effects on urea clearances and that there is not a major blood flow limitation on
urea clearances.
In fact vasodilators primarily increase clearances of larger solutes [14-15]. Such
increases may exceed 100% for solutes of molecular weight 5200 daltons or
above. These observations would support the contention that vasodilator effects
may be more related to venodilation and alterations in permeability rather than
effects on blood flow per se [19, 39, 43]. This is not to deny the possibility that
vasodilators may increase effective peritoneal capillary flow; it is very likely they
do. However, if this were the main effect of the drugs and if urea clearances were
flow limited, proportionally greater increases in urea clearances should occur as
compared to increases in clearances of larger solutes.
28

Table 2. Indirect evidence that maximum peritoneal urea clearances' are not primarily blood flow
limited

1. Urea clearances remain 70% of control even in shock

2. Urea clearances increase <20% with vasodilators
3. Vasodilators increase clearances of larger solutes more than urea clearances
4. Clearances of CO 2 and H2 gases nearly three times maximum urea clearances
5. Urea clearances, if only l/J of effective capillary blood flow (as gas clearances suggest), would be
minimally flow limited according to kinetic modeling analysis and in vitro simulations
6. Increasing evidence that fluid films can readily explain low urea clearances

'Very high dialysate flow, >411h yields urea clearances in man <40mllmin.

Peritoneal clearances of CO 2 gas in humans and hydrogen gas in rabbits are two
•
to three times maximum urea clearances [2, 8]. Gas clearances should also be
limited by effective peritoneal capillary flow and should not exceed urea cleara-
nces to any great extent if effective capillary flow were the main determinant of
urea clearance. On the other hand gases should be able to diffuse across all
membrane resistances more rapidly. Gases might also utilize transcellular routes
while evidence to be discussed later suggests that nongaseous solutes pass pri-
marily through intercellular gaps and extracellular pathways [45-47]. The fact
that gas clearances can be two to three times urea clearances suggests that urea
clearances are limited by total membrane resistances including fluid films rather
than effective peritoneal capillary flow.
If it is true that urea clearances are only one third to one half of the effective
capillary blood flow, kinetic modeling analyses and in vitro simulations of per-
itoneal dialysis suggest that under such conditions effective capillary flow would
exert only modest limitations on peritoneal urea clearances and that the relation-
ship of urea clearance to effective capillary flow would be in the 'plateau' portion
of the clearance to blood flow relationship [2]. Figure 1 shows a hypothetical
relationship of urea clearance to blood flow at high dialysis solution flow rates for
peritoneal dialysis compared to typical findings with a hollow fiber artificial
kidney. For the hollow fiber, dialysate flow would be near 500 mllmin and for
peritoneal dialysis flow rates would exceed 411h. If effective peritoneal capillary
flow rate is near 70 ml/min (as CO 2 gas diffusion studies suggest) than urea
clearances would show a nearly 'plateau' relationship with effective blood flow.
This plateau presumably represents effects of membrane and fluid film resis-
tances. On the other hand, the hollow fiber dialyzer fluid film and membrane
resistances to urea transport are so low that the system is primarily blood flow
limited. Urea clearances remain at a very high fraction of blood flow up to very
high blood flows.
Finally, there is increasing evidence that fluid films can readily account for the
low urea clearances [2, 48]. Figure 2 shows the hypothetical resistance sites which
solutes must cross in moving from peritoneal capillaries into the peritoneal
dialysis solution. R J would represent stagnant fluid films in the peritoneal capil-
 29

Clearance
Urea 200

100

100 200 300

Blood Flow (ml/m.n)

Figure 1. Urea clearance is related to blood flow in hollow fiber and peritoneal dialysis. Dialysate flow
is assumed non-limiting .

lary and R2 the endothelium. The work of Karnovsky and others suggest that the
major path for solute movement across the endothelium may be through inter-
cellular gaps [45,47]. R3 would represent the basement membrane of the capillary
endothelium, R4 the peritoneal interstitium , and Rs the mesothelial layer. Inter-
cellular gaps are probably the major pathways for solute and water movement
across the mesothelium during peritoneal dialysis [45 , 47] . Finally , R6 represents
the stagnant fluid films in the peritoneal cavity. There is increasing evidence that
the interstitium and the fluid films in the peritoneal cavity may account for the
major limitations on urea clearance [19, 48]. The number of capillaries involved
may also limit total pore area even though mean pore size may be relatively high.

Interstitium

.,{'basement membrane

mesothel ial cell-"

Figure 2. Resistance to solute movement during peritoneal dialysis . See text for definitions of Rl thru
R 6 · Reprinted with permission from Nolph and Sorkin: CAPO in Chronic Renal Failure edited by
Brenner and Stein. Churchill Livingstone, New York , 1981 , p. 197.
30

4. Indirect evidence that fluid films and/or interstitial resistance limit urea cleara-
nces during peritoneal dialysis

Table 3 summarizes the indirect evidence that fluid films and/or the interstitial
resistance are important in limiting urea clearances during peritoneal dialysis.
First, as mentioned above, maximum urea clearances do not appear to exceed
40 ml/min in most humans even with rapid cycling or the use of intraperitoneal
vasodilators [40-42]. The rapid cycling should minimize limitations due to dialysis
solution flow rate. Vasodilators which presumably increase the number of capil-
laries perfused and alter capillary permeability should minimize endothelial
resistance. Studies in isolated mesentery suggest that mesothelial intercellular
gaps are greater than 500 A in width and offer very little resistance [49-51]. Thus,
under the conditions of rapid cycling and the use of intraperitoneal vasodilators
(and assuming that effective peritoneal capillary flow is not limiting) major
resistance sites that explain the limits on urea clearances should be the inter-
stitium and the stagnant fluid films in the peritoneal cavity.
Secondly, dialysate is always relatively stagnant in the peritoneal cavity within
the many folds of the mesentery even with rapid cycling [40-42].
Thirdly, dialysate channels are probably relatively wide [42]. Figure 3 shows a
comparison of dialysate channel dimensions during peritoneal dialysis with those
in a man-made hollow fiber kidney. Note that in the hollow fiber kidney much of
the cross section of the dialyzer represents blood path. In small dialysate channels
there is rapid counter-current flow with minimal fluid film resistance [52-53]. In
contrast, in the peritoneal system the interstitium and stagnant pools of fluid
between folds of mesentery represent substantial fluid film resistances.
The interstitial solute path probably represents a usually relatively long dis-
tance [4]. In Figure 2 we see that it may represent 100 IL or more. Figure 4 shows
that the situation may be even more complex. The work of Wayland suggests that
the interstitium may represent a network of aqueous channels through mucopoly-
saccharide and collagenous gels [19]. Hypertonic peritoneal dialysis solutions may
dehydrate the interstitium and, although the total distance may be shortened, the
aqueous network of channels could become more torturous and the resistance to
solute movement actually increase [39,54].

Table 3. Indirect evidence that fluid films and/or interstitial resistance limit urea clearances during
peritoneal dialysis

1. Maximum urea clearances near 40 mllmin even with rapid cycling or vasodilators
2. Dialysate relatively stagnant
3. Probably very wiqe dialysate channels
4. Interstitial solute path-potentially long distance
5. In vitro simulations of peritoneal dialysis demonstrate high fluid film resistances
6. Little evidence to support blood flow limitation
7. Vascular resistance appears low for small solutes
 31

Peritoneal Hollow Fiber Dialyzer

\ .
i :?---..
solule pathway., fiber

Relat ively I-
/0 ~-~
.- 0 ----

8~ _0 =_.: --.:
~
interst itium 8cf&
oO~8o
stagnant
dialysate ~
,/ 0 -""
-=- ~11)lt"-w-ay--
• "">
-:~\y~ ,:::.~=::
n - -. -
ij. capillary .
?,ooo
small dialysate channel
Adjacent folds (countercurrent fast flow)

01 mesentery

(Large Iluid film resistances) (Small fluid film resistances)

Figure 3. Dialysate channel features for peritoneal and hollow fiber dialysis are compared.

Peritoneal Interstitial Hydration and Resistance (R) to Solute Transport

A. Simple Fluid Film B. Aqueous Ne~k

interstitium Dehydrated
States

mesothelium

Figure 4. The interstitium is hypothetically shown as a simple film (A) or network of water channels
(8) dehydrated by hypertonic solutions in the peritoneal cavity . If it were a simple fluid film (A), going
from dehydration to hydration should increase resistance (R) by increasing distance . In B, hydration
decreases R by widening channels.
32

In vitro simulations of peritoneal dialysis, using hollow fiber dialyzers placed in

stagnant pools of fluid with the outer shell of the dialyzer removed, demonstrate
rapid deterioration in urea clearances attributable to high fluid film resistances [2,
48]. Even with the most rapid cycling techniques in and out of the simulated
peritoneal cavity, clearances cannot be restored. Vigorous shaking of the cavity
and improved mixing will diminish the effects of fluid film resistances to some
extent but never approach the performance that can be achieved with rapid
counter current flow of dialysate in the usual manner [48].
As mentioned above, there is little evidence to support a blood flow limitation
and, therefore, the importance of the fluid films is implied.
Finally, studies by Wayland suggest that endothelium offers very little resis-
tance to small solute movement from peritoneal capillary blood into peritoneal
dialysis solution [19]. If rats are injected with fluorescent tagged small solutes,
extensive migration of the solute into the interstitium can be observed. This is in
contrast to what is seen following injection of fluorescent tagged albumin; move-
ment of albumin across vascular walls is not obvious over many minutes unless
agents which increase vascular permeability are administered in solutions bathing
the peritoneum [19].

5. Evidence that vascular permeability is a major resistance for larger solutes

In contrast to the situation for small solutes where interstitial and fluid films
appear to be major determinents of removal efficiency, the permeability of the
microcirculation appears to be a major influence of the clearances of larger
solutes [13-20, 43]. The evidence for this is summarized in Table 4.
First, we have already mentioned that'increased protein losses occur with the
topical application to the mesentery of agents known to increase venular per-
meability [19]. Intraperitoneal nitroprusside, for example, markedly increases
protein losses.
Secondly, there are proportionally larger increases in inulin clearances as
compared to urea clearances with vasoactive drugs [13-18]. There is evidence to
suggest that vasoactive drugs alter vascular permeability [19-20]. This would
explain the greater proportional effects on larger solutes where vascular per-
meability has a major effect on clearances.

Table 4. Evidence that vascular permeability is a major resistance for large solutes

1. Increased protein losses with agents known to increase venular permeability

2. Proportionately larger increases in inulin clearances (as compared to urea clearances) with
vasoactive drugs
3. Increased protein losses with peritoneal inflammation
4. Laser studies with fluorescent tagged albumin in the rat microcirculation
 33

Thirdly, it is well known that protein losses increase with peritoneal inflamma-
tion [5,55-57] . Peritoneal inflammation from any cause stimulates an outpouring
of white cells into the peritoneal dialysis solution [36]. Inflammation in other
tissues of the body is usually associated with vasodilation and it seems reasonable
to assume that this would also occur in the peritoneum. Thus the protein losses
with inflammation may simply reflect endogenous mechanisms that induce vas-
odilation. Vasodilation per se may result in the perfusion of more permeable
capillaries [58]. Local release of histamine may increase vascular permeability.
Finally, studies with fluorescent tagged albumin already have been mentioned
[19]. Following an injection of this material into the rat, the albumin remains
within the microcirculation over many minutes of observation without obvious
leaking into the interstitium. With topical peritoneal application of agents that
alter vascular permeability, there is an almost explosive outpouring of albumin
from the microcirculation into the interstitium over a matter of seconds.

6. A comparison of capillary dimensions of the peritoneal dialysis system with the

hollow fiber dialyzer

Figure 5 shows cross sections of a peritoneal capillary and a synthetic fiber in a

hollow fiber dialyzer. The dimensions are drawn to scale as indicated. Although
the synthetic fiber wall is much thicker, ahigh fraction of the wall luminal surface
may represent 'pore' area. The fiber wall is a mesh of synthetic material with
many spaces between interstices. In contrast, the peritoneal capillary may not
only have a very small relative total luminal surface but only a small fraction (less
than 0.2% according to Pappenheimer [59]) of that luminal surface may represent

Peritoneal Hollow Fiber Dialyzer

'umen diameter 5 -IO..u

{

I
/ wall th ickness I -2.u

IOO..u (interstitium )

---------
mesothelium

--------
~~~~ ---~~-~~

~--------
-====::::: Dialysate -------

---
- Dialysate

Figure 5. Capillary diameters and wall thicknesses are compared for peritoneal and hollow fiber
dialyzers. Note that hollow fiber dialyzers are in direct contact with dialysate while peritoneal
capillaries are not.
34

'pore' area . This is only true, of course, if indeed intercellular gaps are the major
pathways for solute and water movement from the capillary.
Figure 6 shows lateral views of the fiber walls. This demonstrates even more
readily the great distance that may be between intercellular slits in capillary
endothelium and, in contrast, the high fraction of synthetic fiber walls represent-
ing space available for solute exchange between the molecules composing the
wall.
Figure 7 shows lateral views of the course of capillaries in the peritoneal
membrane and synthetic fibers in a hollow fiber dialyzer. Notice that the capillary
network in the peritoneal system would be quite complex with many interconnec-
tions. The total number of capillaries participating in exchange is unknown. In
contrast, in the hollow fiber dialyzers each fiber is a separate entity. There are no
interconnections and the numbers are well known depending on the brand of
hollow fiber dialyzer. In the peritoneal system only a portion of capillaries may be
perfused at anyone time as others may be essentially closed down by pre-capillary
sphincter tone [4] . In contrast, in the hollow fiber dialyzer most of the fibers are
perfused at anyone time in the absence of fiber plugging [60] .

7. Dialysate flow

Figure 8 compares typical flow rates in ml per minutes and in liters per week for'
dialysis with a hollow fiber dialyzer (12 hours per week), intermittent peritoneal

Pathways for Solute Movement

Peritoneal Capillary Wall Hollow Fiber Wall

. Endothelial ceil .:

Endothel ial ceil

. '.
.
", ;,
",
"
, .

(Caps <0. 2'10 sui1ace area 01 lummal O.OO2g averil4Je diameter of "pores"
endothelia l sui1acel (High "pore density" l

Figure 6. The intercellular gaps of the peritoneal endothelium are relatively wide but few. In hollow
fibers, 'pores' are smaller but many.
 35

Peri toneal Hollow FI ber Dialyzer

I 2 j!

. ~ --

Intereon nectlon s
Complex networ 7000 - 20,000 fibe;s
umber unknown Sepa rate, fixed length
Figure 7. Capillary arrangements in peritoneal and hollow fiber dialysis.

dialysis (40 hours per week), and continuous ambulatory peritoneal dialysis (four
2-1 exchanges per day). The figure does not include ultrafiltration rates which
have little impact on the figures for hollow fiber dialysis and intermittent per-
itoneal dialysis but add substantially to total flow for continuous ambulatory
peritoneal dialysis.
For hdllow fiber dialysis, urea clearances are primarily blood flow limited. For
intermittent peritoneal dialysis urea clearances during treatment are probably

FLOW RATE (ml/min) FLOW RATE

DURING TREATMENT IN L/WK
500
500 500

400 400
360

ml L
300 300
min WK

200 200
160

100 100
67 56

Hollow I PO CAPO Hollow IPO CAPO

Fiber (4 ex /day) Fiber (4 ex/day)
Dialyzer Dialyzer

Figure 8. Comparison of typical dialysate flow rates in hollow fiber, intermittent peritoneal (IPD) , and
continuous ambulatory peritoneal dialysis (CAPO); mllmin and IIweek.
36

6 .9-
6]
1.0.

0..8

0.6

0..4

0..2

Hollow
Fiber .Includes
I PO CAPO
Dialyzer
UF

Figure 9. Ratios of typical urea clearances to dialysate flow rates in different dialysis techniques as in
Figure 8.

limited primarily by fluid film resistances for the reasons discussed above. For
continuous ambulatory peritoneal dialysis, the urea clearances are limited by
dialysis solution flow rate and are nearly identical to dialysis solution flow rate.
Figure 9 shows that urea clearances typically are nearly one third of dialysis
solution rate for both hollow fiber dialysis and intermittent peritoneal dialysis
during the treatment sessions. For continuous ambulatory peritoneal dialysis, the
ratio of urea clearance to flow rate of dialysis solution approaches 1.0 [61-66]. .

8. A summary of the importance of different resistances for large and small solutes
during hollow fiber dialysis and peritoneal dialysis

Figure 10 shows a summary of the important resistance sites during peritoneal

dialysis and during hollow fiber dialysis. The height of each bar is a hypothetical
value since the actual numbers are for the most part unknown.
The upper portions of the figure show the importance of different resistance
sites on an absolute scale. In peritoneal dialysis, the vascular wall probably offers
substantial resistance only for the much larger solutes above 1000 daltons in
molecular weight [19]. For smaller solutes this is a short distance to traverse with a
relatively large mean pore size (perhaps greater than 40 A in width at the venular
 37

PERITONEAL

Absolute Absolute
Sca le Sca le

100
100

Per Cent Per Cent

of of
Total Total

Endothelium Interstllum Dia lysate Fiber Dialysate

Film Wall Fi lm

Figure 10. Hypothetical absolute and relative resistance values in peritoneal and hollow fiber dialysis.

end of the capillary) [45-47]. The interstitial resistance is substantial for both
small and large solutes. This would again be an even greater resistance for large
solutes since they must diffuse across this distance and greater hindrance by the
dimensions of the aqueous channels (if such truly exists) would be expected. The
fluid films in the peritoneal cavity offer substantial resistance to both small and
large solutes. The resistance to the latter would again be greater because of the
distances involved and their poorer diffusibility. Mesothelial resistance is not
shown since intercellular gaps may be 500 A or more in diameter and there is little
evidence to suggest that mesothelium is a major resistance site [49-51]. Very
recent findings, to be reviewed below, are challenging the concept of a very
'open' mesothelium, however, and argue for greater mesothelial resistance in
some areas of the peritoneum. Intracapillary stagnant fluid films and capillary
basement membranes are not shown since they are also not known to be major
resistance sites.
Below these hypothetical absolute resistance values for peritoneal dialysis is a
figure showing relative resistances. In the case of urea clearance, the interstitium
and the fluid films are proportionately greater resistances. Because the vascular
wall is a major resistance site for large solutes , interstitial and dialysis solution
fluid films are proportionally shown as less important.
In the hollow fiber dialyzer are only two major resistance sites, the fiber wall
and fluid films. The mean pore size of the fiber wall may be 20 Aor less [67]. For
very large solutes, like albumin, synthetic fiber resistance approaches infinity
since fibers are impermeable to albumin. The thickness of the fiber wall makes
the fiber an important resistance site for urea, perhaps primarily because of the
distance involved. Dialysis solution fluid film resistances are much smaller than in
38

peritoneal dialysis for reasons discussed. Thus on a relative scale the fiber wall
would offer a high proportion of the total resistance to the movement of both
small and large solutes.

9. Possible mechanisms for the net electrolyte sieving effects with peritoneal
ultrafiltration

There are numerous studies to suggest that ultrafiltration induced with osmotic
pressure removes water primarily from peritoneal capillaries. The net removal of
sodium and potassium by convection per liter of ultrafiltrate are usually well
below respective extracellular fluid concentrations [31-33}. The net convective
component of sodium and potassium removal can be calculated by subtracting net
removal accountable to diffusion from the net total removal [33]. Another way to
estimate convective transport is to instill solutions with sodium and potassium
concentrations in Gibbs-Donnan equilibrium with serum water [31-33, 68-69].
Although a net sieving effect creates a concentration gradient for some net
diffusion, net electrolyte removal per liter of ultrafiltrate still remains far below
that in extracellular fluid. Severe hypernatremia has been observed as a result of
overzealous peritoneal ultrafiltration and removal of extracellular water without
amounts of sodium equal to extracellular fluid concentrations [32].
In contrast, it is well known that ultrafiltration by hydrostatic pressure in a
hollow fiber dialyzer results in the net removal of an ultrafiltrate with electrolytes
in proportions comparable to those in extracellular fluid [70-71]. Isolated ultra-
filtration in hollow fibers yields an essentially protein free ultrafiltrate of serum
containing electrolytes in amounts near those predicted by Gibbs-Donnan equi-
librium.
How is it possible that a membrane as permeable as the peritoneum (more
permeable than hollow fibers in terms of protein losses) can hinder the convective
movement of electrolytes with ultrafiltration? The answer to this question is not
available. Table 5 summarizes possible mechanisms for the net electrolyte sieving
effect with peritoneal ultrafiltration.
First, there is substantial evidence to suggest that the width of intercellular gaps

Table 5. Possible mechanisms for the net electrolyte sieving effect with peritoneal ultrafiltration

1. Ultrafiltration through narrow intercellular gaps in proximal capillaries

2. Endothelial cell surface charges in intercellular gaps
3. Trans-endothelial cell water movement
4. Interstitial gel surface charges along aqueous channels
5. Mesothelial cell surface charges in intercellular gaps
6. Trans-mesothelial cell water movement
7. Glucose interaction with cations in intercellular gaps or interstitial channels
 39

progressively increases from proximal to distal portions of the capillaries with the
most permeable portions being in the small venules [19]. The capillaries of the
peritoneum may differ from man-made fibers in having a progressive increase in
pore width along the capillary while man-made fibers are more homogenous. At
the proximal end of the capillaries hydrostatic pressure should be higher [58].
Glucose should be more osmotically effective across this tighter portion of the
capillary whereas in the distal portion of the capillary glucose may be readily
absorbed and exert little osmotic pressure. Thus combined hydrostatic and
osmotic pressure could induce maximum ultrafiltration rates across portions of
the capillary that are least permeable.
If most of the water flows through the intercellular gap where junctions are
rather narrow, then endothelial cell surfaces in close proximity and their respec-
tive charges could impede the movement of electrolytes through the gap.
If transmembrane hydrostatic and osmotic pressures are high enough, perhaps
some transendothelial cell water movement does occur [31]. Such net movement
of water across the cell may occur without proportional movement of electrolytes
through the very complex internal cell milieu.
Surface charges on the capillary basement membrane or on the surfaces of
interstitial gels may impede the movement of charged solutes. This could be akin
to the charge interference offered by polar molecules in the glomerulus [72]. The
work of Glassock and others suggest that albumin is held back more by charge
than by pore dimensions [72].
Mesothelial cell surface charges in intercellular gaps could influence electrolyte
movements. However, if it is true that the permeability of the mesothelium is
much greater than that of the endothelium, then this could be less important than
the same phenomenon in the endothelium.
The movement of ultrafiltrate from the interstitium into the peritoneal cavity
could occur by hydrostatic pressure with the build up of fluid in the interstitium
and with some osmotic pressure induced by glucose gradients across the meso-
thelium. If it is true that the mesothelium is more permeable than the endo-
thelium then the major glucose gradient could be across the vessel wall with only a
modest glucose gradient maintained across the mesothelium. Nevertheless, if
water did move through mesothelial cells this again could interfere with the
convective transport of electrolytes.
Finally, we have published studies to show that even neutral molecules may not
accompany ultrafiltration induced by glucose osmotic pressure in the same pro-
portions as when the ultrafiltration is induced across the same membrane with
hydraulic pressure [73]. This is not an effect of osmotic pressure per se as perhaps
due to the use of a solute such as glucose which can enter the membrane and move
up stream against the flow of ultrafiltrate [74]. We have proposed a hypothesis
that molecular interaction within the membrane may alter the net sieving effects
[74].
40

Rubin et ai. have measured net sieving coefficients in clinical studies where
solutes were added to dialysis solutions prior to instillation at concentrations near
those of body fluid [75]. A concentration gradient for net diffusion was thus
absent at the beginning of the exchange. Net sieving coefficients of multiple
solutes (for 4.25% dextrose exchanges with a 30-minute dwell time) were calcu-
lated as mass transfer/ultrafiltration volume/plasma water concentration. Pro-
gressively lower net sieving coefficients were found as molecular weight increased
(see Table 6). Charged ions such as sodium and potassium yielded net sieving
coefficients lower than that of neutral solutes at comparable molecular weight.

10. Comparisons in animals and humans

Techniques to assess dialysis efficiency will be reviewed in the chapter by Popo-

vich. The mass transfer area coefficient represents the maximum clearance
possible by diffusion at rapid dialysis solution flow rates. It is limited by the
permeability and effective pore area of the membrane. Dedrick has compared
peritoneal mass transfer area coefficients for urea and inulin in several species as a
function of body weight (Fig. 11). Some of the values represent clearance mea-
surements at high flow rates but presumably approach the mass transfer area
coefficients. Several values for gas clearance (hydrogen in rabbit and carbon
dioxide in humans) are shown. Gas clearances provide minimum estimates of
effective capillary blood flow. Lower urea and inulin mass transfer area coeffi-
cients support the contention that mass transfer area coefficients for urea and
inulin are not limited by blood flow but by meinbrane resistances and area.
Nearly linear relationships with body weight suggest that the peritoneal dialysis
system behaves quite similarly from species to species in proportion to the
dimensions represented therein. These studies suggest that there are not striking
differences in membrane properties between species.

Table 6. Net sieving coefficients per exchange a

Solute Mean + SEM net sieving Number of

coefficient exchanges

Sodium 0.56± 0.04 14

Potassium 0.40 ± 0.04 11
Urea 0.63 ± 0.06 10
Creatinine 0.57 ± 0.09 8
Inulin 0.41 ± 0.08 6

a Masstransfer/ultrafiltration volume/plasma water concentration for 4.25% dextrose exchanges with

a 30-minute dwell (from Rubin et al. [75]; reproduced with permission).
 41
100

~
~
::!
......
...J
10
::!
U
z
~
(%:
~
W
...J
U
a:
0
~
0
n.

01~L- ________ ~ __________ ~ __________ ~

o
BODY WEIGHT 9

Figure 11. Peritoneal mass transfer area coefficients are compared for urea and inulin in several species
as a function of body weight. Some values are clearances at high flow rates thought to be approaching
the mass transfer area coefficients, clearances of hydrogen gas in the rat and rabbit and of carbon
dioxide in humans are also shown (from Dedrick et al. [76); published with permission) .

U. Vesicles vs intercellular gaps

In recent years interest has focused on the relative importance of vesicles and
intercellular gaps in regard to the passive movement of solutes across peritoneal
endothelium and mesothelium [77-85].
Wide intercellular gaps have been noted in certain species particularly in the
sub-diaphragmatic area [83-84]. In addition, earlier reports suggested that the
movement of tracers up to 30 000 daltons molecular weight occurs mainly
between cells rather than through cells via vesicles [45-47]. Conversely, recent
studies in rabbits have noted very tight intercellular mesothelial cell junctions and
numerous intra-cytoplasmic vesicles [85] . Capillaries were of the continuous type
and endothelial cells were also noted to contain many vesicles. Following intra-
venous injection of iron dextran, an electron dense tracer, the iron dextran was
found mainly in vesicles in the peritoneum of the rabbit. Scanning electron
microscopy of rat mesentery has also revealed very tight intercellular junctions
between mesothelial cells [86].
More work is needed to determine the relative roles of intercellular gaps and
vesicles in passive solute transport across the peritoneum. This may vary from
species to species, in different areas of the peritoneum and for solutes of different
42

molecular weight. It is not clear as to how peritoneal morphology and solute

transport in animals relate to humans. Also, it is not clear how tissue processing
may artifactually misrepresent the number of vesicles and/or the dimensions of
intercellular gaps in vivo. Finally, in vivo , inflammation such as with peritonitis,
may widen intercellular gaps resulting in misrepresentation of the situation under
non-infected conditions [86].

12. Changes with peritonitis

In humans and animals peritonitis has been associated with changes in peritoneal
transport [86-88). Clearances, glucose absorption rate and protein losses tend to
increase suggesting increases in permeability. Rapid glucose absorption is associ~
ated with an earlier osmotic equilibrium, early loss of the osmotic pressure
gradient for ultrafiltration and less net ultrafiltration with long dwell exchanges
(greater than 3 hours of dwell).
Verger and his colleagues have shown a marked widening of intercellular gaps
in visceral mesentery in rats with peritonitis (see Fig. 12-14) [86]. There is also loss
of mesothelial microvilli and the appearance of round forms (presumably white

Figure 12. Normal rat mesentery by SEM. There are many microvilli ; x2000 (from Verger et al. [86);
reproduced with permission).
 43

Figure 13. Scanning electron microscopy of mesentery with peritonitis. Note intercellular spaces
bridged by cytoplasmic processes (from Verger et al. [86J; reproduced with permission).

Figure 14. Scanning electron microscopy of infected mesentery. Round forms appear to be moving
through the intercellular spaces. Arrows point to several round cells; x1500 (from Verger et al. [86] ;
reproduced with permission).
44

cells) that seem to be moving through the wide intercellular gaps. With peritonitis
there are also changes in the interstitium and microcirculation; interstitial cellular
infiltration and vasodilatation are seen. Thus, it is impossible to determine
whether transport changes are primarily the result of mesothelial interstitial or
vascular alterations.

13. The importance of mesothelial resistance

Verger and colleagues produced morphological changes in the mesothelium of

the rat peritoneum with a laparotomy, heat drying technique and subsequent
closure. One to two days post-operatively, increases in clearances, protein loss
and glucose absorption were similar to increases with peritonitis [86]. Histologi-
cal studies at the same time showed mesothelial changes similar to peritonitis in
visceral mesentery except for the absence of inflammatory cells. However, there
were no interstitial or microcirculatory alterations.
These studies suggest that mesothelial resistance may be much greater than
previously believed, at least in some species and some areas of the peritoneum.
Mesothelial injury with or without vasodilatation may alter the passive transport
characteristics of the membrane. Further work is necessary to clarify these issues.
The mesothelial microvilli are 1.5-3.0 microns long and 400-900 A wide in mice
[79]. Microvilli markedly increase gross surface area. Surface charges may trap
water between microvilli and prevent friction of adjacent surfaces [80]. At the
base of microvilli, the open ends of surface vesicles can be seen [81]. There is
much to be learned about the effects of cellular structures on mesothelial passive
transport properties during peritoneal dialysis.

14. Capacities and pressures in the dialysis solution chamber of the peritoneal
dialysis system

Twardowski and co-workers [89] have related intra-abdominal pressure to intra-

peritoneal volume in normal sized adults. The mean results from his studies are
summarized in Figure 15. Note an almost linear increase in intra-abdominal
pressure in each position studied. The hydrostatic pressures shown relate to the
umbilicus in the supine position and the xiphoid process in the upright positions.
Filling the peritoneal cavity is somewhat like inflating a collapsed balloon. In
upright positions the muscle tone of the abdominal wall is greater and higher
pressures are required to inflate the less compliant wall. Many normal-sized
adults tolerated up to 4 liters of intraperitoneal volume without discomfort.
Those who develop dyspnea at intraperitoneal volumes between 2.5 to 4 liters
have had an associated decrease in forced vital capacity. This occured even
though their intra-abdominal pressure at the higher volumes was the same as
 45
lAP
em H2 0
18

16

14

12

10

2 - Supine 0 1.5 %
.4.25%
-- Upri;ht A I. 5 %
0
.4.25%
- Sittin; D 1.5 %
-2
t .4.25 %
0 2 3 4
IPV L

Figure 15. Intra-abdominal pressure (lAP) related to calculated intraperitoneal volume (IPV) of 1.5
and 4.25% glucose solutions. Means ± SEM and linear regressions are shown (from Twardowski et al.
[89]; reproduced with permission).

those who did not develop dyspnea or a decrease in forced vital capacity.
Therefore, it would appear that the major difference between patients who
tolerate and patients who do not tolerate large volumes has to do with diaphrag-
matic strength rather than the workload. Obviously at the extremes of patient size
variation, the relationship of intra-abdominal pressure to intraperitoneal volume
may be different.
In patients who tolerate larger volumes, daily clearances may be increased by
the use of larger volumes at the same frequency or maintained with larger
volumes and fewer daily exchanges. Clearance increases with 3-liter volumes as
compared to 2-liter volumes at fixed cycle times are mainly a function of the
increased dialysis solution flow rate [90]. Small solute clearances increase most;
larger solute clearances and protein losses remain stable, as they tend to do with
increases in dialysis solution flow rate by any technique. Studies in adult patients
by Goldschmidt and co-workers [91] showed similar clearances of small solutes
with 1L and 2L volumes provided the dialysis solution flow rates remained
unchanged. This suggests that solution membrane contact is near maximum with
l-liter volumes and that larger volumes mainly increase the width of the dialysis
46

solution channels with little effect on membrane contact. Thus, in most adults all
or most finger-like projections of the peritoneal cavity must be filled with fluid
with I-liter intraperitoneal volumes. Three-liter volumes may stretch them more
but do not open new projections. There must be some point below I-liter
volumes, however, where solution membrane contact would decrease and area
losses would reduce large solute clearances.

Acknowledgement

Supported in part by Veterans Administration Merit Review Funds, NIH Divi-

sion of Research Resources, General Clinical Research Center, Grant No
5MOIRR00287 and NIH contracts USPH Nol AM5-2216, USPH No 1 AM7-2217,
and USPH No 1 AM9-2208.

References

1. Nolph KD: CAPD - A logical approach to peritoneal dialysis limitations (A comparison of the
peritoneal dialysis system and hollow fiber kidneys). NUA 1: 5-8, 1980.
2. Nolph KD, Popovich RP, Ghods AJ, Twardowski Z: Determinants of low clearances of small
solutes during peritoneal dialysis. Kidney Int 13: 117-123, 1978.
3. Nolph KD: The peritoneal dialysis system. In: Today's Art of Peritoneal Dialysis. Contrib
Nephrol17: 44-49, 1979.
4. Nolph KD, Miller FN, Rubin J, Popovich R: New directions in peritoneal dialysis concepts and
applications. Kidney Int 18 (Suppll0): S111, 1980.
5. Nolph KD: Peritoneal dialysis. In: W Drukker, FM Parsons, JF Maher (eds). Replacement of
Renal Function by Dialysis. Martinus Nijhoff Medical Division, The Hague, 1978, pp 277-321.
6. Rubin J, Nolph KD, Popovich RP, Moncrief J, Prowant B: Drainage volumes during CAPD.
ASAIO 2: 2, 1979.
7. Erbe RW, Greene JA Jr, Weller JM: Peritoneal dialysis during hemorrhagic shock. J Appl
Physiol22: 131-135, 1967.
8. Aune S: Transperitoneal exchange 2. Peritoneal blood flow estimated by hydrogen gas clearance.
Scand J Gastroenterol5: 99, 1970.
9. Texter E, Clinton JR: Small intestinal blood flow. Am J Digest Dis 8: 587, 1963.
10. Hare HG, Valtin H, Gosselin RE: Effects of drugs on peritoneal dialysis in the dog. J Pharmacol
Exp Ther 145: 122-129, 1964.
11. Henderson LW, Kintzel JE: Influence of antidiuretic hormone on peritoneal membrane area and
permeability. J Clin Invest 50: 2437-2443, 1971.
12. Nolph KD, Ghods AJ, Brown P, Van Stone JC, Miller FN, Wiegmann DL, Harris PD: Factors
affecting peritoneal dialysis efficiency. Dial Transpl6: 52-90, 1977.
13. Nolph KD, Ghods AJ, Van Stone J, Brown PA: The effects of intraperitoneal vasodilators on
peritoneal clearances. Trans Am Soc Artif Intern Organs 22: 586-594, 1976.
14. MillerFN, NolphKD, Harris PD, RubinJ, WiegmanDL, JoshuaIG, TwardowskiZJ, GhodsAJ:
Microvascular and clinical effects of altered peritoneal dialysis solutions. Kidney Int 15: 630-639,
1979.
15. Miller FN, Nolph KD, Harris PD, Rubin J, Wiegman DL, Joshua IG: Effects of peritoneal
dialysis solutions on human clearances and rat arterioles. Trans Am Soc Artif Intern Organs 24:
131-132, 1978.
 47

16. Nolph KD: Effects of intraperitoneal vasodilators on peritoneal clearances. Dial Transpl7: 812-
817, 1978.
17. Nolph KD, Rubin J, Wiegman DL, Harris PD, Miller FN: Peritoneal clearances with three types
of commercially available peritoneal dialysis solutions: Effects of pH adjustment and intra-
peritoneal nitroprusside. Nephron 24: 35-40, 1979.
18. Nolph KD, Ghods AJ, Brown PA, Twardowski ZJ: Effects of intraperitoneal nitroprusside on
peritoneal clearances with variations in dose, frequency of administration, and dwell times.
Nephron 24: 114, 1979.
19. Wayland H: Action of histamine on the microvasculature. Proc 1st CAPD Int Symp. Excerpta
Medica, Amsterdam, 1980, 18-27.
20. Miller FN, Joshua IG, Harris PD et at.: Peritoneal dialysis solutions and the microcirculation.
(Vol 17 of Contributions to Nephrology). A Trevino-Becerra, F Boen (eds). In: Today's Art of
Peritoneal Dialysis. S Karger, Basel, 1979, pp 51-58.
21. Nolph KD, Stoltz M, Maher JF: Altered peritoneal permeability in patients with systemic
vasculitis. Ann Intern Med 78: 891-894, 1973.
22. Nolph KD, Miller L, Husted FC, Hirszel P: Effects of intraperitoneal isoproterenol on reduced
peritoneal clearances in patients with systemic vascular disease. J Int Urol Nephrol 8: 161-169,
1976.
23. Brown ST, Ahearn DJ, Nolph KD: Reduced peritoneal clearances in scleroderma increased by
intraperitoneal isoproterenol. Ann Intern Med 78: 891-894, 1973.
24. Manery JF: Water and electrolyte metabolism. Physiol Rev 34: 334-417, 1954.
25. Tarail R, Hacker ES, Tavmor R: The ultrafiltrability of potassium and sodium in human serum. J
Clin Invest 31: 23-26, 1952.
26. Folk BP, Zierler KL, Lilienthal JL: Distribution of potassium and sodium between serum and
certain extracellular fluids in man. Am J Physiol153: 381-385, 1948.
27. Brown ST, Ahearn DJ, Nolph KD: Potassium removal with peritoneal dialysis. Kidney Int 4:
67-69, 1973.
28. Kelton JG, Vlan R, Stiller C, Holmes E: Comparison of chemical composition of peritoneal fluid
and serum. Ann Intern Med 89: 67-70,1978.
29. Henderson LW: Peritoneal ultrafiltration dialysis: Enhanced urea transfer using hypertonic
peritoneal dialysis fluid. J Clin Invest 45: 950, 1966.
30. Henderson LW, Nolph KD: Altered permeability of the peritoneal membrane after using
hypertonic peritoneal dialysis fluid. J Clin Invest 48: 992-1001, 1969.
31. Ahearn DJ, Nolph KD: Controlled sodium removal with peritoneal dialysis. Trans Am Soc Artif
Intern Organs 28: 423-428, 1972.
32. Nolph KD, Hano JE, Teschan PE: Peritoneal sodium transport during hypertonic peritoneal
dialysis: Physiologic mechanisms and clinical implications. Ann Intern Med 70: 931-941,1969.
33. Nolph KD, Sorkin MI, Moore H: Autoregulation of sodium and potassium removal during
continuous ambulatory peritoneal dialysis. Trans Am Soc Artif Intern Organs 26, 1980.
34. Wayland H, Silberberg A: Blood to lymph transport. Microvasc Res 15: 367, 1978.
35. Nolph KD, Prowant B: Complications during continuous ambulatory peritoneal dialysis. Proc 1st
Int Symp on CAPD. Excerpta Medica, Amsterdam, 1980, pp 258-262.
36. Rubin J, Roger WA, Taylor HM, Everett ED, Prowant BP, Fruto LV, Nolph KD: Peritonitis
during continuous ambulatory peritoneal dialysis. Ann Intern Med 92: 7-13, 1980.
37. Wade OL, Combes B, Childs A W, Wheeler HO, Dournand A, Bradley SE: The effect of exercise
on the splanchnic blood flow and splanchnic blood volume in normal man. Clin Sci 15: 457-463,
1956.
38. Miller FN, Nolph KD, Joshua IG: The osmotic component of peritoneal dialysis solutions. Proc
1st Int Symp on CAPD. Excerpta Medica, Amsterdam, pp 12-17.
39. Nolph KD: Anatomy, physiology and kinetics of peritoneal transport during peritoneal dialysis.
Proc 1st Int Symp on CAPD. Excerpta Medica, Amsterdam, 1980, pp 7-11.
48

40. Tenckhoff H, Ward G, Boen ST: The influence of dialysate volume and flow rate on peritoneal
clearance. Proc Eur Dial Transpl Assoc 2:113-117,1965.
41. Stephen RL, Atkin-Thor E, Kolff WJ: Recirculating peritoneal dialysis with subcutaneous
catheter. Trans Am Soc Artif Intern Organs 22: 575-585,1976.
42. Goldschmidt ZH, Pote HH, Katz MA, Shear L: Effect of dialysate volume on peritoneal dialysis
kinetics. Kidney Int 5: 240--245, 1975.
43. Miller FN, Wiegman DL, Joshua IG, Nolph KD, Rubin J: Effects of vasodilators and peritoneal
dialysis solution on the microcirculation of the rat cecum. Proc Soc Exp Bioi Med 161: 605-608,
1979.
44. Nolph KD, Hirszel P: Recent developments in the understanding of peritoneal dialysis. Editorial
by invitation to Editorial Board of Przeglad Lekarski 5: 433-436, 1975.
45. Karnovsky MJ: The ultrastructural basis of capillary permeability studies with peroxides as a
tracer. J Cell Bioi 35: 213-235, 1967.
46. Cotran RS: The fine structure of the microvasculature in relation to normal and altered per-
meability. In: EB Reeve, AC Guyton (eds). Physical Bases of Circulatory Transport: Regulation
and Exchange. WB Saunders, Philadelphia, 1967, pp 249-275.
47. Karnovsky MJ: The ultrastructural basis of transcapillary exchanges. In: Biological Interfaces:
Flows and Exchanges. Little Brown, Boston, 1968, pp 64-95.
48. McGary TJ, Nolph KD, Rubin J: In vitro simulations of peritoneal dialysis: A technique for
demonstrating limitations on solute clearances due to stagnant fluid films and poor mixing. J Lab
Clin Med 96: 1, 148--157, 1980.
49. Nagel W, Kuschinsky W: Study of the permeability of isolated dog mesentery. Eur J Clin Invest 1:
149-154, 1970.
50. Gosselin RE, Berndt WO: Diffusional transport of solutes through mesentery and peritoneum. J
Theor Bioi 3: 487-495, 1962.
51. Rasio EA: Metabolic control of permeability in isolated mesentery. Am J Physiol276: 962-968,
1974.
52. Maher JF, Nolph KD: Factors effecting optimal performance of coil dialyzers. Proc Int Congr
Nephrol, Florence Italy, 1975, p 657.
53. Maher JF, Nolph KD: Resistance to diffusion in dialyzers. Clin Nephrol1: 333-335.
54. Rubin J, Nolph KD, Arfania D, MiIlerFM, Wiegman DL, Josua IG, Harris PD: Studies on non-
vasoactive peritoneal dialysis solutions. J Lab Clin Med 93: 910--915,1979.
55. Blumenkrantz MJ, Roberts CE, Card Bet al.: Nutritional management of the adult patient
undergoing peritoneal dialysis. J Am Diet Assoc 73(3): 251-256, 1978.
56. Giordano C, De Santo NG: Dietary management of patients on peritoneal dialysis. (Vol 17 of
Contrib Nephrol). In: A Trevino-Becerra, F Boen (eds). Today's Art of Peritoneal Dialysis. S
Karger, Basel, 1979, pp 77-92.
57. Kobayashi K, Manji T, Hiramatsu S et al.: Nitrogen metabolism in patients on peritoneal dialysis.
(Vol 17 of Contrib Nephrol). In: A Trevino-Becerra, F Boen (eds). Today's Art of Peritoneal
Dialysis. S Karger, Basel, 1979, pp 93-100.
58. Renkin EM: Exchange of substances through capillary walls: circulatory and respiratory mass
transport. In: GEW Wolstenholme (eds). Ciba Foundation Symp. Little Brown & Company,
Boston, 1969, pp 50--66.
59. Pappenheimer JR: Passage of molecules through capillary walls. Physiol Rev 33: 387, 1953.
60. Nolph KD, Ahearn DJ, Esterly JA, Maher JF: Irreversible morphological and functional changes
in hollow fiber kidneys with a single dialysis. Trans Am Soc ArtifIntern Organs 20: 604-612, 1974.
61. Nolph KD: Peritoneal clearances. (Invited Editorial). J Lab C1in Med 94: 519-525, 1979.
62. Nolph KD, Popovich RP, Moncrief JW: Theoretical and practical implications of continuous
ambulatory peritoneal dialysis. (Invited Editorial). Nephron 21: 117-122, 1978.
63. Popovich RP, Moncrief JW: Kinetic modeling of peritoneal transport. In: A Trevino-Becerra, F
Boen (eds). Today's Art of Peritoneal Dialysis. S Karger, Basel, p 59.
 49

64. Popovich RP, Pyle WK, Moncrief JW et al.: Peritoneal dialysis. AIChE Symp Series 75:31.
65. Popovich RP: Metabolic transport, kinetics in peritoneal dialysis. Proc 1st Int Symp on CAPD.
Excerpta Medica, Amsterdam, 1980, pp 28--33.
66. Popovich R, Moncrief JW, Nolph KD, Ghods AJ, Twardowski ZJ, Pyle WK: Continuous
ambulatory peritoneal dialysis. Ann Intern Med 88: 449-456, 1978.
67. Green DM, Antwiler GD, Moncrief JW, Decherd JF, Popovich RP: Measurement of the
transmittance coefficient spectrum of cuprophan. Trans Am Soc Artif Intern Organs 22: 627-636,
1976.
68. Donnan FG: The theory of membrane equilibrium. Chem Rev 1: 73,1924-1925.
69. Loeb J: Donnan equilibrium and physical properties of proteins. J Gen Physiol3: 691, 1920-1921.
70. Nolph KD, New DL: Effects of ultrafiltration on solute clearances in hollow fiber artificial
kidneys. J Lab Clin Med 88: 593-600, 1976.
71. Nolph KD, Stoltz ML, Maher JF: Electrolyte transport during ultrafiltration of protein solutions.
Nephron 8: 473-487, 1971.
72. Glassock RJ: The nephrotic syndrome. Hosp Prac 14: 105-129, 1979.
73. Nolph KD, Hopkins CA, New D, Antwiler GD, Popovich RP: Differences in solute sieving with
osmotic vs. hydrostatic ultrafiltration. Trans Am Soc Artif Intern Organs 22: 618--626, 1976.
74. Twardowski AZ, Nolph KD, Popovich RP, Hopkins CA: Comparison of polymer, glucose and
hydrostatic pressure induced ultrafiltration in a hollow fiber dialyzer: Effects on convective solute
transport. J Lab Clin Med 92: 619-633, 1978.
75. Rubin J, Klein E, Bower JD: Investigation of the net sieving coefficient of the peritoneal
membrane during peritoneal dialysis. ASAIO J 5: 9-15, 1982.
76. Dedrick RL, Flessner MF, Collins JM et al.: Is the peritoneum a membrane? ASAIO J 5: 1-8,
1982.
77. Feriani M, Biasioli 5'; Chiaramonte S et al.: Anatomical bases of peritoneal permeability: A
reappraisal. Anatomy of peritoneum. Int J Artif Organs 5: 345, 1982.
78. Odor DL: Observations of the rat mesothelium with the electron and phase microscopes. Am J
Anat 95: 433, 1954.
79. Baradi AF, Rao SN: A scanning electron microscope study of mouse peritoneal mesothelium.
Tissue Cell 8: 159, 1976.
80. Andrews PM, Porter KR: The ultrastructure morphology and possible functional significance of
mesothelial microvilli. Anat Res 177: 409,1973.
81. Baradi AF, Rayns DJ: Mesothelial intercellular junctions and pathways. Cell Tissue Res 173: 133,
1976.
82. Simionescu M, Simionescu N: Organization of cell junctions in the peritoneal mesothelium. J Cell
Bioi 74: 98, 1977.
83. Tsilibray EC, Wissig SL: Absorption from the peritoneal cavity; SEM study of the mesothelium
covering the peritoneal surface of the muscular portion of the diaphragm. Am J Anat 199: 127,
1977.
84. Dumont AE, Robbins E, Martelli A, I1iescu H: Platelet blockade of particle absorption from the
peritoneal surface of the diaphragm (41138). Proc Soc Exp Bioi Med 167: 137,1981.
85. Goitloib L, Digenis GE, Rabinovich S, Medline A, Oreopoulos DG: Ultrastructure of normal
rabbit mesentery. Nephron 34: 248,1983.
86. Verger C, Luger A, Moore HL, Nolph KD: Acute changes in peritoneal morphology and
transport properties with infectious peritonitis and mechanical injury. Kidney Int 23: 823,1983.
87. Rubin J, McFarland S, Hellems EW et al.: Peritoneal dialysis during peritonitis. Kidney Int 19:
460,1981.
88. Smeby LC, Wideroe TE, Svartas TM et al.: Changes in water removal due to peritonitis during
continuous peritoneal dialysis. In: KD Nolph (eds). Advances in Peritoneal Dialysis. GM Gahl,
M Kessel. Excerpta Medica, Amsterdam, 1981, p 287.
50

89. Twardowski ZJ, Prowant BF, Nolph KD etal.: High volume, low frequency continuous ambula-
tory peritoneal dialysis. Kidney Int 23: 64-70, 1983.
90. Twardowski ZJ, Nolph KD, Prowant B, Moore HL: Efficiency of high volume, low frequency
CAPD. Trans ASAIO 29: 53-57, 1983.
91. Goldschmidt ZH, Pote HH, Katz MD, Shear L: Effects of dialysate volume on peritoneal dialysis
kinetics. Kidney Int 5: 240-245, 1974.
3. The peritoneal microcirculation

FREDERICK N. MILLER

The peritoneal membrane which surrounds the peritoneal cavity covers a maze of
complex organs and tissues (e.g., stomach, small intestine, large intestine, liver,
abdominal wall muscles, and mesentery). These tissues have independent and
often dissimilar blood flow regulatory mechanisms. Thus, during peritoneal
dialysis there are a multitude of very different vascular areas that are available for
transfer of solutes from the blood, through the interstitial tissue and the per-
itoneal membrane, to the peritoneal cavity. The combined circulatory functions
of these tissues will ultimately regulate the type and quantity of solutes eliminated
by peritoneal dialysis. Aside from basic physiological differences among tissues,
there are, in addition, numerous factors which can independently influence the
local control of blood flow in these regions. Neural inputs, drugs, metabolic
products, local hormones, excitement, exercise, and intestinal motility all may
have a profound effect on the delivery of solutes (via peritoneal blood flow) to the
areas of solute transport and on the permeability of the vasculature. In addition,
diseases both systemic (e.g. hypertension) and local (e.g. peritonitis, cancer)
have been shown to alter blood flow patterns or vascular permeability.
Assuming constant cardiac output, these factors will participate in the regu-
lation of blood flow to the peritoneum primarily by affecting peritoneal vascular
resistance which in turn is predominantly a function of small arteriolar diameter.
The smooth muscle tone of the arterioles in the microvasculature of the per-
itoneal cavity will determine the number of capillaries and small venules per-
fused, and thus the total effective endothelial surface area available for solute
exchange. This review will concentrate on factors and mechanisms operative in
the peritoneal microcirculation and the peritoneal interstitial tissue which serve
to control tissue blood flow, vascular permeability, solute transfer and ultimately
the efficiency of peritoneal dialysis.
52

1. The circulation of blood in the peritoneal cavity

The blood supply to the organs of the peritoneal cavity and to the peritoneal
membrane , aserous membrane composed of connective tissue and mesothelium,
can be divided into two main categories. The first category includes that portion
of the membrane lining the peritoneal cavity along with the skeletal muscle
vasculature which it covers (parietal peritoneum). The second is that portion
which covers and courses through the viscera (visceral peritoneum). The inter-
relationships of the blood supply of the peritoneal membrane and the underlying
organs provide for a complex solute delivery and removal system that has the
potential for dramatically altering the efficiency of peritoneal dialysis. There are
essentially three interdependent but separate systems which function simul-
taneously to influence the delivery and exchange of solutes, the absorption of the
components of peritoneal dialysis solution (primarily glucose) and the effect of
drug additives on the efficiency of peritoneal dialysis. These three systems are
diagrammatically portrayed in Figure 1. They include: 1) the blood circulation of
the visceral peritoneum, 2) the blood circulation of the parietal peritoneum, and
3) the lymphatic circulation for both parietal and visceral peritoneum.

//
AORTA
~ VENA CAVA

~
~," /
t

-
/
, ":
/

'J
L IVER
..,
I/L/////,,/; //. '
/,%
.'
.'
'/'

PORTAL
~ .------ -, PAR I ETAL

~ ti.
VEIN
///// PER ITONEUM :-: .: ....

t /j
?
r-----"

VISCER AL
I ~Z% ...
//////~ "
PER ITO NEUM
J ~
//,.
,.
~. " .:. ',: ':', • ~ :' ,.. ::.: ~', .: "~ ,~,:'~ _: :-" ...:, .:. . ..... : ... " . . :.~~ _~ .. ".: .:: ,,: ,4".., ... ',: :.' : .. ". ,, 4: : '; :::
LYMPHAT IC VESSELS

Figure 1. Diagram of circulatory pathways in the peritoneal cavity.

 53

The arterial blood supply to the visceral peritoneum and underlying structures
comes from either the mesenteric or celiac arteries [1]. The mesenteric arteries
supply blood to the jejunum, ileum, cecum, and large intestine as well as the
mesentery. Organs supplied via the celiac artery include the liver, stomach,
pancreas, spleen, duodenum, gall bladder, the greater omentum and the lesser
omentum. Visceral peritoneum covers both sides of the mesentery, a thin com-
posite of connective tissue, fat, blood vessels, lymph glands and ducts. In general,
vessels in the mesentery act as conduits to supply blood to structures covered by
visceral peritoneum. Although the mesenteric blood flow may also participate in
fat deposition and utilization, a definite role, other than for transport of blood to
the viscera, has not been established [2].
The venous system draining the visceral organs and peritoneum converge to
enter the portal vein (Fig. 1). Consequently, drugs and solutes in peritoneal
dialysis solutions and those absorbed from the gastrointestinal tract may be
subject to immediate treatment by the liver [3]. The portal venous system is
essentially separate from the rest of the venous system, however, the portal and
systemic venous systems do form anastomoses in two places: around the rectum
and the esophagus. In the presence of portal hypertension or obstruction of the
portal system as may occur in cirrhosis or carcinoma, blood from the portal
system may back up and engorge these areas of anastomoses forming esophogeal
varices or hemorrhoids. The additional venous pressure would alter the delicate
balance of Starling forces in the visceral vasculature causing ascites and altering
solute clearances during peritoneal dialysis.
In contrast to the visceral peritoneum, the arterial blood supply to the parietal
peritoneum [1] and to the underlying skeletal muscles of the abdominal wall (i.e.
the circumflex, iliac, lumbar, intercostal and epigastric arteries) drains into veins
which enter the systemic rather than the hepatic portal system (Fig. 1). Thus the
local regulation of the parietal vs the visceral peritoneal vascular systems and the
relative proportion of blood in each of these systems, can be important when
considering drug metabolism and solute absorption from dialysis solutions. Phar-
macokinetic studies have indicated that compounds such as atropine, caffeine,
glucose, glycine and progesterone are primarily absorbed by the visceral per-
itoneum after intraperitoneal injection [3]. One might expect therefore, a de-
creased or minimal systemic effect of drugs which are metabolized by the liver
when these drugs are administered with dialysis solutions.
There is a very extensive lymphatic system in the abdominal cavity which
functions to maintain a balance of solutes and fluids in the interstitial tissue,
probably functioning to prevent edema [4, 5, 6]. Generally the visceral lymphatics
arise in the peritoneal membranes (omenta and mesentery), and drain into a
complex network of visceral lymph nodes. These in turn drain into parietal lymph
nodes and the lymph is then returned to the venous circulation (Fig. 1) primarily
via the thoracic duct. Consequently, unlike the venous drainage which is divided
into the systemic and portal systems, the lymphatic drainage of the parietal and
54

visceral peritoneum use common pathways. In light of the proposed function of

the lymphatics to prevent edema, the initial lymphatic vessels have been reported
to concentrate protein [7] relative to the interstitial connective tissue and thus
would be expected to draw water out of the interstitial tissue. Also protein laden
peritoneal fluid may slide along the peritoneal membrane within the peritoneal
cavity, reenter the interstitium and then enter the lymphatics. Thus, the high
dialysate protein concentrations found in the dialysate during human peritoneal
dialysis [8, 9, 10] may reflect a short circuiting of the flow of protein from the
vasculature to the lymph either by a 'washing' effect on the interstitium or on the
peritoneal membrane.

2. Architecture of the peritoneal vasculature

Since the 'textbook' pictures of the vascular tree are certainly not prevalent in the
peritoneal cavity, the anatomical arrangements of the peritoneal vasculature
present a fascinating basis for beginning our understanding of the role of the
peritoneal vasculature in the efficiency of peritoneal dialysis. In the cremaster
muscle, which is an extension of the abdominal wall muscles and is representative
of the vasculature underlying parietal peritoneum (Fig. 2), a major artery and
vein pair (approximately 100 and 150 /Lm in diameter respectively) enter the tissue
from the main feeder vessel. This pair has been designated as the first order artery

Figure 2. The rat cremaster muscle: a. first order artery and vein pair; b. second order arteriole and
venule pair; c. third order arteriole; d. third order venule; e. venule arcade; f. terminal arteriole; g.
post capillary venule; h. metarteriole.
 55

and vein [11]. Subsequent branchings are called 2nd, 3rd and 4th order vessels
respectively. The paired branching pattern continues through the 3rd and some-
times 4th order vessels. In skeletal muscle such as the cremaster, short, direct,
artery-vein anastamoses have not been observed by Baez [12] and Smaje et al.
[13]. However, there are arteriole to arteriole and venule to venule connections
sometimes called arcades, from which capillaries may arise (Fig. 2). Capillaries
may also branch from terminal arterioles (the smallest arteriole with smooth
muscle) and from metarterioles (arterioles which become venules) [14] before
draining into post-capillary venules. The capillaries themselves may exhibit a
complex branching pattern in the parietal as well as the visceral and mesenteric
circulations.
The visceral circulation is represented in Figure 3 by the vasculature on the
mesothelial surface of the rat cecum. This tissue has the paired arrangement of
arterioles and venules all the way down to, in some cases, the terminal arteriolar
level (3rd or 4th order arterioles) which just preceded the capillary branches.
Numerous artery to artery and vein to vein anastomoses or arcades are present in
the cecum. This system of arcades may function to equalize flow and provide
adequate perfusion in times of bowel compression of localized vascular areas.
Frasher and Wayland [2] have identified in the cat, discrete units of mesentery
bounded on all sides by pairs of artery to artery and vein to vein anastomoses or
arcades. The arcades in the mesentery increase in numbers as the intestinal wall is

Figure 3. The mesothelial surface of the rat cecum : a. first order artery and vein pair; b. second order
arteriole and venule pair; c. third order arteriole and venule pair; d. venule arcades; e. arteriole
arcades.
56

approached [15]. In contrast to the cat mesentery [2], the rat mesentery is
relatively avascular especially in young or male rats [16]. Generally larger vessels
which provide the blood supply to the intestine are found coursing through the rat
mesentery. These vessels are surrounded by fat with no or few arteriole and
venule branches to the mesentery itself (Fig. 4). These first order vessels give rise
to smaller, second order vessels which generally run parallel to the intestine.
Third order arterioles course over and perpendicular to the intestine and branch
to form the vessels which penetrate and nourish the muscular layers of the
intestine or enter the mucosal layer of the intestine for supply of blood to the villus
[15, 17]. Numerous arcades are also present in both the arteriole and venule
systems as they spread over the intestinal wall.
Throughout the mesentery are a widespread network of lymph vessels . The
smallest lymphatic vessels which perform the function of collecting fluid and
material from the blood and interstitial tissue are called the initial or terminal
lymphatics [7]. These vessels merge to form the collecting lymphatics which
transport the lymph fluid back to the blood. Collecting lymphatics in the mesen-
tery are often found near or between an artery - vein pair. An example of such a
lymphatic vessel, filled with fluorescent labeled albumin, is shown in Figure 5.

Figure 4. The rat mesentery and small intestine: a. avascular section of the mesentery; b. small
intestine; c. first order artery and vein pair; d. fat cells surrounding the larger vessels; [Link] order
arteriole and venule pair; f. third order arteriole and venule pairs : these are identified on both the near
and far side of the intestine ; g . numerous arcades are present but the low magnification precludes a
more precise identification.
 57

This figure also demonstrates that vascularity in the mesentery dramatically

decreases in the area just peripheral to the fat cells.
The role of the lymphatics during peritoneal dialysis has not been illucidated
although a role of the lymph vessels in removing normally accumulated peritoneal
fluid is well established [18]. Diaphragmatic contractions close the openings
(stomata) of the terminal lymphatics (lacunae) that are located between meso-
thelial cells [19]. Contractions also increase the protein concentration in the
lymphatics [20]. Thus alterations in movements, position, or contractility of the
peritoneal organs, induced by peritoneal dialysis solutions or activity of the
patient, may impact on the ability of the lymph to carry protein and fluid from the
peritoneal cavity. In addition, pressure in the peritoneal cavity, as may be
produced by the large volume of fluid used in peritoneal dialysis, has been shown
to open stomata [19] and could allow for an enhanced lymphatic drainage.
Tsilibary and Wissig [19] have also concluded that opening and closing the
stomata may be under physiological control with contractile elements of the
mesothelium or lymphatic endothelium actually responsible for the patency of
these openings. Furthermore the contractions of the lymph vessels themselves are
extremely sensitive to external calcium concentration [21]. Thus the compostion
of the dialysis solution could easily affect lymph formation or lymph vessel
function thus influencing macromolecular clearance during peritoneal dialysis.

Figure 5. The rat mesentery: the rat is given an intravascular injection of fluorescein isothiocynate
tagged rat serum albumin (FITC-RSA) and the in vivo fluorescent image of the mesentery is
observed: a. section of avascular mesentery (no FITC-RSA is present); b. venule; c. arteriole; d. fat
cells; e. lymph vessel with FITC-RSA in the lumen ; f. arteriole - venule anastomases .
58

2.1. Altered vasculature and macromolecular leakage

Various disease states may produce anatomical changes in the microcirculation

which could lead to changes in clearance during peritoneal dialysis. For instance
hypertension is associated with a reduction of diameter of small arterioies [22]
(Fig. 6). As hypertension progresses the degree of vessel involvement may
increase thus producing a continually changing hemodynamic picture. This could
contribute to the alterations in macromolecular permeability in hypertension that
has been observed in both rats [23] and humans [24] . The altered vasculature may
also restrict peritoneal blood flow and thus the delivery of solutes to the sites of
exchange. Also, peritoneal tumors are associated with ascites formation [25, 26]
which may in part be due to pronounced alterations in the tumor venous system.
Alternatively, there may also be a specific permeability factor emitted from the
tumor cells [27] which would increase macromolecular leakage. This obviously
could have a significant impact on macromolecular loss during peritoneal dialysis
used to treat chronic uremia or peritoneal cancers.

o NORMOTENSIVE
~¥m HYPERTENSIVE
2 WEEKS
100
E
::i..
a:: 60
w
I-
W
~
<t 20
0
a::
<t
-l
0 4 WEEKS
ir
~IOO
a::
<t
-l
0
a:: 60
I-
z
0
u
20

IA 2A 3A
ARTERIOLES

Figure 6. First. second and third order arteriolar diameters for normal and renovascular hypertensive
rats. Rats were made hypertensive by removal of one kidney and restricting the blood flow to the
contralateral kidney (from Joshua et al. [22]; reprinted with permission) .
 59

3. Barriers of solute exchange between the vasculature and the peritoneal cavity

The transport of solutes from the blood to the peritoneal cavity is potentially a
very complex event which may involve a number of anatomical resistance barri-
ers. The basic barriers to solute transport are the vascular wall (including the
endothelium and basement membrane) the interstitium and the mesothelium
[28]. Potential endothelial transport pathways include various degrees of fenes-
trations, vesicles and cell junctions [29]. The relative contributions of each
pathway to total solute (large or small) movement have not been established.
Similarly, the role of the interstitium in the normal transport of solutes is
undefined. The following discussion will focus on some of these potential path-
ways for solute movement and how they may interact to alter dialysate solute
concentrations.

3.1. Arterioles

Although the arterioles are not immediately thought of as having a major role in
solute exchange during peritoneal dialysis, there is anatomical and experimental
evidence to indicate otherwise [14]. Permeability of these vessels is important in
the maintenance of interstitial solute concentrations and fluid balance. In ad-
dition, pathways of solute transfer to the smooth muscle cells of the arterioles
playa major supportive role for blood, ionic, and hormonal information to reach
the smooth muscle and affect contractility.
Probably most solute transfer from arterioles involved in interstitial regulation
and ultimately in peritoneal dialysis clearances occurs in the terminal arterioles or
in the metarterioles. Chambers and Zweifach [30] have described these vessels as
having a discontinuous muscular layer. The absence of a complete smooth muscle
layer would leave the endothelial cells and basement membrane as the last
vascular barriers to diffusion in this high pressure side of the microcirculation
(Fig.7).
Solute transfer through the endothelial cells of the arterioles may occur as
transcellular movement or may occur via micropinocytotic vesicles which are
especially important for the transfer of large solutes [29, 31]. The smooth muscle
cells of the arterioles are in close proximity to the endothelial cells at discrete
areas known as myoendothelial junctions [14, 29]. These junctions, along with the
endothelial gap junctions (Fig. 7) provide an access to the smooth muscle cells for
blood nutrients and hormones as well as a waste pathway for metabolic products
from the smooth muscle cells.
60

Figure 7. Cross-sectional diagram of a terminal arteriole (from Rhodin, 1974) showing: a. arteriole
lumen; b. endothelial cell nucleus ; c. endothelial vesicles; d. endothelial gap junction; e. an area of
discontinuous muscle layer which characterizes these vessels; f. smooth muscle cell nucleus ; g.
myoendothelial junction; h. basal lamina.

3.2. Capillaries and endothelium

As the arteries repeatedly branch to form smaller vessels, a continuous smooth

muscle coat is replaced by a discontinuous coat and finally we are left with the
capillary vessels that have only an endothelial cell lining and a basal lamina. The
seemingly simple structure of the capillary, however, has many variations which
depend on the morphology of the endothelium [29]. The three most important
types in the peritoneal cavity are the continuous endothelium, the fenestrated
endothelium and the discontinuous endothelium (found in the liver sinusoids and
in the spleen and of unknown importance for peritoneal dialysis) . It has been
shown however [32] that in the liver of cirrhotic patients there is a macromolecu-
lar barrier to protein movement for large (immunoglobulin-M) but not medium
and small (immunoglobulin-G and albumin) proteins. This type of differentiation
could affect the composition of protein lost during peritoneal dialysis. Others
have shown with gel electrophoresis in rats, that indeed, most of the protein lost is
albumin [33].
Most capillaries in the peritoneum have the continuous type of endothelial
cells, similar in structure to arterial endothelium (Fig. 7). Transfer of molecules in
these cells probably occurs either via the intercellular gap junction or via micro-
 61

pinocytotic vesicles [34]. These vesicles are found on the luminal and basal lamina
surfaces as well as in the cytoplasm of the endothelial cell. The capillaries of the
smooth muscle, visceral peritoneum, parietal peritoneum and skeletal muscle
contain the continuous endothelium with many micropinocytotic vesicles. Mea-
surements in the mouse diaphragm [35] show that capillaries have an average
thickness of 0.25 ± 0.01 to 0.17 ± 0.07 JLm (arteriole capillary to venule capillary)
and an increasing number of vesicles per cubic micrometer of endothelium at the
venular end of the capillary (956 ± 114 at the arteriole end to 1171 ± 98 at the
venular end). In contrast, muscular arteries and venules had 187 ± 14 and 297 ± 31
vesicles, respectively. Vesicles may form an apparent chain to give rise to transen-
dothelial channels which are more prevalent on the venous end of the capillary
[35] and may function for macromolecular transport. However, current evidence
suggests that at least for agonist induced effects, endothelial gaps in the venular
walls are the most important structures for movement of protein [16].
The second major type of endothelium is found in the intestinal villus and is
characterized by the presence of numerous fenestrae. The fenestrae are pore like
structures, -50 mu in diameter [36] which are often covered by a thin membrane
composed largely of mucopolysaccharides [37]. In the intestinal villus of the
mouse the fenestrae cover approximately 10% of the total capillary surface area
[36]. Venular capillaries have far more fenestrae (-12 times) than arteriole
capillaries. Also, a greater proportion of fenestrae in the venular capillaries
(-73%) have a membrane (vs 54% in the arterial end of the capillary). Endo-
thelial junctional gaps may be less important in these fenestrated capillaries for
solute transfer since the ratio of fenestrae area to gap area is 16 in the arterial
capillary and 240 in the venous capillary. In the low hydrostatic pressure venous
system where the fenestrae are more prevalent, the movement of interstitial
protein back to the blood may be favored. The vehicles for solute transport
(vesicles, junctional gaps, fenestrae, etc.) are not static. They may be altered by
changes in hydrostatic pressure, tissue pressure, hormones, injury, etc. What
effect peritoneal dialysis has on these normal transport pathways is not known.
Although one might guess that the high osmolality, low pH and acetate or lactate
buffer of peritoneal dialysis solutions would affect these structures.
Besides the structural elements involved in capillary transport, there are also
differences in the physical dimensions of capillaries which may influence water
and solute transcapillary movement. Smaje et al. [13] has made extensive mea-
surements of capillary length, density, and diameter. Interestingly, he found that
the venular end of the capillary had approximately a 10% increase in diameter as
compared to the arteriolar end of the capillary and there was a wide variation in
capillary length (615 ± 194 JLm). Also, there is a very distinctive stop-go blood
flow pattern of the capillary bed which was recognized many years ago by Krogh
[38]. This pattern is the result of vasomotion (cycles of relaxation and con-
traction) in the terminal arterioles [39]. Such a flow pattern will provide a
continuously changing small vessel hydrostatic pressure and thus be instrumental
62

in the moment to moment regulation of the forces which govern solute and water
exchange with the interstitial tissue. Thus we could expect that altering vasomo-
tion and changing capillary flow patterns by the additions of vasodilators to
dialysis solution [8] would have a significant effect on water and solute movement
in the interstitium and on peritoneal dialysis efficiency.

3.3. The post capillary and collecting venules

The post capillary and collecting venules are anatomically similar to the capil-
laries. They have an endothelial layer, an occasional smooth muscle cell, and a
basal lamina. In addition they may have surrounding the endothelial cells,
pericytes [40] which appear to be precursors of smooth muscle cells. The role of
pericytes is not established but recent studies propose that they may contribute to
the regulation of lymph formation [41].
The small veins appear to be most important in protein transfer [42] with
fenestrae, vesicles and endothelial gaps that could provide routes for protein
transfer. The venular endothelium contains not only small micropinocytotic
(400-800 A) vesicles such as those found in the arterial endothelium (Fig. 7) but
also many large vesicles (2000-3000 A). These large vesicles may fuse to form a
vesicular chain or channel that provides a direct pathway from the intervascular
to extravascular space [35, 43] for protein transport. Endothelial gaps may also
playa significant role in the movement of protein [16] into the interstitium.

3.4. The basement membrane

The basement membrane or basal lamina appears to have collagen as an integral

part of its structure [44] and it is apparently permeable to most molecules. It
appears that only larger molecules may be retarded by the basement membrane
from movement into the interstitium [45]. These observations are supported by
theoretical calculations [4] and by the studies of Granger and Taylor [46] in the cat
ileum which imply that the basement membrane may playa significant role in the
movement of macromolecules from the blood through the interstitium to the
lymph. They have shown that the passage of endogenous proteins from the blood
to the lymph is dependent on molecular size. However, as venous pressure is
increased, molecular size makes progressively less of an impact on the fraction of
protein appearing in the lymph. Since the intestinal capillaries were no more
restrictive to macromolecular movement than skin capillaries, they concluded
that the actual barrier for macromolecular movement may be beyond the endo-
thelial wall, perhaps at the basement membrane.
In a series of interesting experiments, Johansson [47] injected horseradish
peroxidase (MW 50-60 A)into the interstitial tissue and found this tracer in
capillary vesicles but not in intercellular endothelial spaces. This demonstrated
 63

that vesicles may transport large molecules to the blood from the interstitium and
that these molecules cannot move through the intercellular spaces. However,
with a very large molecule, ferritin, (MW 750 000, diameter 115 A) Johansson [48]
found that the basement membrane apparently restricted the movement of
ferritin and there were only minimal amounts found in endothelial vesicles. Since
others [34] have found the basement membrane to be permeable to ferritin when
it is injected intravascularly, the basement membrane may function as a one way
valve for the passage of very large molecules. If the kinetics and factors governing
solute movement are different from the peritoneum to the blood than from the
blood to the peritoneum [49] this would have special interest in the use of
peritoneal dialysis for the administration of anti cancer or other drugs.

3.5. The interstitium

The interstitial tissue may play a unique role in regulating solute clearances
during peritoneal dialysis. This tissue is largely composed of mucopolysaccha-
rides [50] such as hyaluronate and chondroitin sulfate. These have been thought
to function as a barrier for diffusion of large molecular weight solutes [51, 52]. A
model for this type of hindrance has been developed by Wiederhielm [53] and
refined and supported by Watson and Grodins [54]. The interstitial space in these
models is viewed as a gel like matrix composed of mucopolysaccharides inter-
spersed with a free fluid phase. Solutes move by restricted diffusion in the gel
phase and by convection and free diffusion in the fluid phase (Fig. 8). Small
solutes travel easily in both the gel and fluid phases while the large solutes would
be dependent on the integrity of the free fluid pathways for transfer through the
interstitium. As such, it has been hypothesized [55] that dehydration of the free
fluid phase during peritoneal dialysis with hypertonic solutions would decrease
protein transport. Ordinarily, solute and water in the interstitum that does not
return to the capillaries or venules would flow into the lymphatic system. Thus
peritoneal dialysis may interrupt the normal interstitial flow of fluid that forms
the lymph.

3.6. The mesothelium

The last barrier to solute transfer from the peritoneal vasculature to the per-
itoneal cavity is the mesothelium. If the surface of the mesothelium normally
functions as a low resistance transport mechanism for movement of fluid and
solutes from the blood to the lymph as has been proposed by Wayland and
Silberberg [56], peritoneal dialysis may simply 'wash' the surface of the meso-
thelium. Gosselin and Berndt [57] concluded that passive diffusion was probably
the primary mechanism for transport of substances across the mesothelium.
64

Figure 8. A hypothetical model for protein transfer through the interstitium to the peritoneal cavity.
With a normal osmolality dialysis solution the interstitium is hydrated and protein passages are wide
open. Lymphatics collect the fluid and the solutes that leak out of the vasculature or slide across the
mesothelial surface . Solutes that reach the peritoneal cavity are washed away by the peritoneal
dialysis solution instead of returning to the blood or lymph system.

These investigators also concluded that mesentery mesothelium may be more

permeable than parietal mesothelium. This could account for the observations
that intraperitoneally injected drugs are largely absorbed via visceral peritoneal
blood vessels [3]. However, the permeability of the mesothelium linearly de-
creases with increasing molecular weight [58] and may not present a significant
barrier up until a molecular weight of 500000 [59]. Mesothelium diffusion can be
altered by temperature, pH, or drugs [59, 60, 61] and the changes in transport
differ depending upon the molecular species. These data therefore suggest that a
simple 'non-restrictive membrane' model can not account for the experimental
data and that the mesothelium may provide some significant barriers to diffusion
or transfer which can be altered by changes in dialysate volume or composition.
This mesothelial barrier may function to control the generation and flow of
lymph.

3.7. Endothelial lining

Recently, an increasing body of evidence suggests that the traditional resistance

pathways to fluid and protein movement which have been discussed in the above
sections may play less of a role than previously thought. A fibrous glycoprotein
 65

coat over the endothelium and even in the 'pores' may have a significant impact
on permeability. This has been suggested from the results of studies by Mason et
al. [62] who showed that perfusion of a capillary bed with different proteins,
reduced the filtration coefficients of the capillary wall and thus the net fluid flow
as compared to perfusion with a protein free fluid. Others have shown that
cationized ferritin (but not native ferritin) binds to the endothelium and decreases
the filtration coefficient of perfused capillaries [63]. These authors have proposed
that a change in binding properties of the endothelial coat could produce irregular
binding patterns thus altering macromolecular permeability with little or no effect
on small solute permeability. Differential anionic sites in the endothelium have
been observed in the lung [64]. Similar differences in anionic binding have been
identified in basement membranes [65]. An extension of this idea (an influence of
differential anionic binding) to peritoneal dialysis would suggest the potential
importance of changes in the composition of dialysis solutions or in the amount
and composition of plasma proteins present during chronic uremia.

4. Drug and autacoid effects in the peritoneal vasculature

There are many cardiovascular factors such as peritoneal blood flow, peritoneal
vascular permeability, and total venular and capillary surface area available for
exchange which may influence peritoneal dialysis efficiency. These factors can be
affected by normal regulatory mechanisms (i.e. neurotransmitters, hormones,
ions, etc.) or by drugs to alter peritoneal blood flow, or vascular permeability.
Thus they may have significant effects on the efficacy and efficiency of peritoneal
dialysis.

4.1. Vasoactive effects of autacoids

A group of endogenous compounds, the autacoids, have important regulatory

effects on the peritoneal microcirculation. These compounds have direct effects
on vascular smooth muscle or indirect effects via modulation of adrenergic
transmission or synthesis of other autacoids. For instance, Blumberg et al. [66]
have shown that prostaglandin synthesis can be stimulated by angiotensin and
bradykinin. Prostaglandins [67], histamine [68], adenosine [69, 70], serotonin
[71], angiotensin [72] and bradykinin [73] have all been implicated in altering the
release of norepinephrine from the adrenergic nerve terminal. In addition, at
least some of these agents such as histamine [74] and prostaglandins [75] may in
themselves be regulated by norepinephrine release from the adrenergic nerve
terminal. The direct effects of the autacoids may result in vasoconstriction,
vasodilation and changes in vascular permeability. The response to a particular
autacoid, is the result of many complex factors such as the level of sympathetic
66

tone, the level of contractile tone, the presence of other autacoids, the particular
tissue and the vessel size. The additional effects of peritoneal dialysis on the
regulatory role of autacoids on vascular tone or permeability has not been
investigated.

4.2. Vasoactive effects of drugs

Addition of vasodilators to the dialysate prior to instillation should allow for high
local concentrations of drug in the peritoneum while body concentrations remain
low due to slow absorption and subsequent metabolism after passage through the
portal venous system into the liver (Fig. 1). Indeed, the results of studies in dogs
[76], rats [77], and man [S, 7S] demonstrate that solute clearances may be
augmented by the addition of drugs to dialysis solutions without apparent sys-
temic effects. Furthermore, large changes in peritoneal blood flow can be pro-
duced by the injection of vasodilators in the perfused canine intestinal circulation
[79]. The implication of these studies is that the contractile state of the vascular
smooth muscle can regulate blood flow in the peritoneum to alter the efficiency of
peritoneal dialysis.
Changes in peritoneal vascular resistance however, do not appear to occur
uniformly in the visceral and parietal peritoneal vasculatures. In a rat intestinal
wall preparation [SO] and in gastric submucosa [Sl] dilation of arterioles has been
demonstrated but results were erratic without preconstriction by norepinephrine.
Miller et al. [S2] have developed a preparation which uses the mesothelial surface
of the rat cecum (Fig. 3) in a controlled tissue environment (pH, P0 2 , PCO l and
temperature). In this preparation they have demonstrated substantial dilation
(50-100% respectively) of arterioles (-70 JLm in diameter) and venules (-140 JLm
in diameter) with different pharmacological agonists. In vivo concentration re-
sponse curves are shown in Figure 9 for nitroprusside in this preparation. Parallel
small arteries and veins in the cecum circulation exhibit different sensitivities
(-log EDso or pDl values), to isoproterenol and different maximal responses to
isoproterenol, nitroprusside and papaverine (Table 1). Thus the response of the
vasculature may depend not only on the drug but on the vessel type and size.
Messina et al. [S3] demonstrated significant but minimal dilation (-13% in-
crease in diameter of 22 JLm arterioles) in the superfused rat mesentery in re-
sponse to prostaglandin E with no effect of PGA. However, in the rat cremaster
muscle which has been used for a model for parietal peritoneum, others [S4] have
found that blockade of prostaglandin synthesis produced a 40-60% decrease in
diameters of small arterioles. These data suggest that prostaglandins provide a
substantial local dilatory input to vascular smooth muscle. Altura [S5] and Altura
and Zweifach [S6] showed that isoproterenol and other vasodilators produce
some dilation (usually 10-30%) at various vascular levels of the mesocecum, but
they suggested that in general, vessels in their mesocecum preparation are usually
 67
CECUM
200

a:
W 175
1--
We;
::!!!:
«C
08150
LAio
C/l:.!!
~!...125
>

100, , , , ,
168 167 106 105 104
NITROPRUSSIDE [M]

Figure 9. The mean (± SEM) concentration-response curves (N = 4) for nitroprusside on small

arteries (x = 52 ± 6.4 Mm diameter) and on small veins (x = 84 ± 6.9 Mm in diameter) on the meso-
thelial surface of the rat cecum. The cecum was places in a tissue bath. The blood vessels and nerves
from the animal to the cecum remained intact. In vivo television microscopy was used to measure
vascular diameters. A five-minute control period with Krebs solution bathing the cecum was followed
by a five-minute exposure to a low concentration of nitroprusside. The bath was drained and fresh
Krebs solution was added along with the next higher concentration of nitroprusside. This sequence
was repeated until a maximal effect (no further increase in diameter) was achieved. Maximal diameter
during each drug period is plotted as a percent of the mean diameter during the control period.

dilated to near maximal levels by circulating epinephrine. These data in the

cecum [82] cremaster [84] and mesentery [83, 85, 86] suggest that primary blood
flow control mechanisms are operative in the peritoneal vasculature but may
differ depending upon the tissue. In addition it appears that the mesentery
subserves a delivery rather than a regulatory role for blood flow in the visceral
peritoneum since the mesentery vasculature generally appears to be in a dilated
state. The relatively avascular nature ofthe mesentery, except around the major
vessels and fat (Fig. 4 and 5) further supports this hypothesis.

Table 1. Effect of vasodilators on small arteries and veins in the rat cecum.

Control diameter Maximum dilation PD 2

(Mm, x±SEM) ('Yo, x±SEM) (x±SEM)

Artery Vein Artery Vein Artery Vein

Isoproterenol 67±3 141 ± 11 21 ± 6.3 48 ± 2.7 6.3 ±0.28 • 7.1±0.17

(7)
Nitroprusside 64±3 135 ± 11 52 ± 6.4 • 84 ± 6.9 5.6 ± 0.27 NS5.6±0.16
(11)
NS
Papaverine (7) 66 ± 6 141 ± 13 20 ± 4.6 • 81 ± 14.1 3.9 ± 0.31 NS3.9±0.11

• p<0.05 analysis of variance with Duncans multiple range test for unequal N size.
68

Particular segments of the microcirculation may also respond differently to

vasoconstrictors and vasodilators. Maximal vasoconstrictor effects induced by a
variety of agonists showed large differences in the rabbit superior mesenteric
artery [87]. Also in the rat mesentery [88] the precapillary sphincters are about
lOO-fold more sensitive than the arterioles to the vasoconstrictor effects of cate-
cholamines. Differential effects of vasoconstriction or vasodilation may explain
why two drugs (norepinephrine and dopamine) both of which increased blood
pressure in rabbits, had opposite effects on solute clearances during peritoneal
dialysis [89]. Norepinephrine decreased and dopamine increased urea and crea-
tinine clearances in this study. In the rat cremaster muscle there is a difference in
arteriolar sensitivity to prostaglandins [84] and in the rat cecum, there is a
difference in arteriole sensitivity to peritoneal dialysis solutions [82]. In both
studies, vessel size or branching order appeared to be the important determinants
of these differences.
The route of drug administration is also important since peritoneal barriers to
diffusion and/or liver metabolism may limit systemic drug effects or limit per-
itoneal vascular effects. For instance, Maher et al. [90] reported no increase in
peritoneal solute clearances with intraperitoneal glucagon in New Zealand white
rabbits. However, with intravenous glucagon, clearances of creatinine and urea
were increased significantly. They hypothesized that glucagon is too large (3484
daltons) to readily cross the peritoneal mesothelium, and thus glucagon could not
gain access to the vessel walls when it was added to peritoneal dialysis solutions.
This hypothesis is supported by data from Torres et al. [61]. These investigators
demonstrated that other drugs with an equivalent molecular weight to glucagon
are largely retained in the peritoneum. From a clinical viewpoint, intravenous
application of drugs to enhance peritoneal dialysis clearances does not seem
desirable since systemic drug effects or induced hormonal effects may negate the
desired action of the drug on the peritoneal vasculature. As one example,
Longnecker et al. [91] showed that intravenous injection of nitroprusside will give
only transient dilation or even constriction in small arterioles of the cremaster
while direct topical application will give a prolonged dilation of these same
vessels.

4.3. Effects of endogenous mediators and drugs on permeability

During peritoneal dialysis autacoid or drug effects on vascular permeability may

be even more important than effects on regional blood flow patterns. Fox et al.
[16] demonstrated a concentration-dependent histamine-induced leakage of pro-
tein across vessel walls in the rat and cat mesentery. Leakage of protein was
correlated with electron microscopic identification of endothelial gaps up to llLm
in diameter which developed after histamine application. These gaps appeared in
the area of the endothelial junctions of the non-muscular venules. This data
 69

supports the observations of Nakamura and Wayland [42] who have shown that
small molecules (MW <34(0) are able to pass across membranes of all micro-
vascular components (arterioles, capillaries and venules) in the rat mesentery but
larger molecules (MW <190(0) diffuse primarily across venules or the venular
end of capillaries with virtually no passage across arterioles.
The interaction of autacoids and vasodilators on protein leakage has been
examined in perfused dog forelimbs where a bradykinin induced increase in
lymph flow rate and total lymph protein concentration was inhibited by vas-
opressin, serotonin and methylprednisolone [92]. In guinea pig lung [93] the
selective ~2-adrenergic agonist, terbutaline, inhibited the histamine induced ex-
travasation of Evans blue dye. Others have also found an inhibitory effect of
catecholamines on histamine induced edema [94] that could not be blocked by
alpha-adrenergic receptor antagonism. There thus appears to be a general antag-
onistic effect of ~-adrenergic receptor stimulation on a histamine-induced change
in vascular permeability. However, during peritoneal dialysis, the ~-adrenergic
agonist, isoproterenol, increases the clearances of inulin, creatinine, and urea in
man [78] and in rats [77].
These data imply a complex relationship between drug induced changes in
vasodilation and in solute transfer. Since terbutaline (a vasodilator) decreases
macromolecular transport induced by PGE1, bradykinin and histamine [95] the
endothelial cell is not simply responding to altered hydrostatic pressure. Others
have also found by direct observations' of the microcirculation that agents which
decrease vascular resistance such as acetycholine, papaverine and isoproterenol,
do not increase macromolecular leakage [96]. Baker [97] has perfused the dog
gracilis muscle under maximally dilated conditions. He found that the vasodilator
papavarine did not cause fluid and protein movement out of the vasculature
whereas histamine, in the presence of papaverine did. He concluded that protein
leakage by histamine is not dependent on changes in hydraulic forces due to
vasodilation. On the other hand, with peritoneal dialysis in the hypertensive rat,
there is a significant increase in dialysate protein (Fig. 10). This leads to a
hypothesis that macromolecular leakage induced by histamine is a specific altera-
tion in vascular permeability and the regulation of the involved mechanism then,
may hold a key for alteration in dialysate protein during peritoneal dialysis. In
support of this hypothesis (Fig. 11), three vasodilators histamine, isoproterenol
and nitroprusside were all efficacious in production of vasodilation but vastly
different in their abilities to induce protein leakage (histamine > nitroprusside
> > > isoproterenol [98]. Furthermore histamine-induced protein leakage but
not histamine-induced vasodilation is dependent on prostaglandin synthesis [99].
Both indomethacin and mefenamic acid blocked protein leakage but not dilation
produced by histamine. We can only conclude then, that there is a complex
relationship between various endogenous agents and the regulation of protein
leakage in the peritoneal cavity.
70

DIANEALR, 355 mOsm

KREBS, 285 mOsm

NORMOTENSIVE HYPERTENSIVE
(N=5) (N=4)

10
III
!;i * NS
en 8
~
<t
o~
<t E 6
en'
a: OJ c
1-
4
(,)
l-
lL.
2

1-8 17-20 1-8 17-20

EXCHANGE NUMBER

Figure 10. Peritoneal dialysis in rats. Normotensive and renovascular hypertensive rats had eight 25 ml
exchanges (12-minute dwell, 3-minute drainage) with 1.5% dextrose Dianeal® peritoneal dialysis
solution (-350mOsm) followed by eight exchanges with Krebs solution (-285mOsm) and then
another four exchanges with the Dianeal® solution. Prior to dialysis, rats were given an intravascular
injection of fluorescein isothio cyanate tagged rat serum albumin (FITC-RSA) and the concentration
of FITC-RSA in the dialysate was measured for each exchange. An asterisk indicates a significant
difference (P <0.05) between the group mean values for the different solutions.

4.4. Permeability effects of drugs

In contrast to the many studies demonstrating autacoid effects on permeability,

there are only a few studies which have suggested that drugs may induce changes
in permeability. Hirszel et al. [100] have investigated hormonal and drug-induced
effects on clearances during peritoneal dialysis in New Zealand white rabbits.
Intraperitoneal nitroprusside and other drugs increased clearances of urea and
creatinine to over 145% of control values. Since the magnitude of changes for
small and large solutes were the same, and since different drugs gave different
effects on clearances, they suggested that vasodilation by itself may not be an
adequate explanation for the effect of nitroprusside on peritoneal dialysis solute
clearances.
Miller et al. [8] studied the effect of nitroprusside on the peritoneal clearances
of different molecular weight solutes during peritoneal dialysis. Nitroprusside
produced relatively greater effects on the clearances of the higher molecular
weight solutes. The combined effects for nitroprusside with and without pH
adjustment of the dialysis solution before instillation were a 21 % increase for urea
 71

- ARTERIOLE DIAMETER
350
----FITC-RSA LEAKAGE

...J 300
o
0::
I-
Z
8 250

u...
o 200
I-
Z
W
~ 150
w
a.

100

109 10-8 Ici 106 105 106 10-5 104 10- 7 166 105 10-4
ISOPROTERENOL (M) HISTAMINE (M) NITROPRUSSIDE (M)

Figure 11. Concentration-response curves for isoproterenol (N = 6) histamine (N = 8) and nitro-

prusside (N = 7). The experimental preparation is the rat cremaster muscle but the set-up and the
protocol are similar to that in Figure 9. Arteriolar diameter (solid line) and fluorescent intensity in the
interstitium (dashed line) of fluorescein isothiocyanate tagged rat serum albumin (FITC-RSA) were
measured during a control period in Krebs solution and during each exposure to a vasodilator. The
mean diameter and f1urorescent intensity during each drug period is plotted as a percent of the mean
value during the control period.

(MW = 60), a 23% increase for creatinine (MW = 113), a 36% increase in inulin
(MW = 5000) and 111 % increase for protein (albumin MW = 68000). Since there
were no differences in the dialysate drainage volumes with nitroprusside, these
differences in solute clearances could not be attributed to differences in ultra-
filtration. Miller et al. [8] hypothesized that, in addition to an effect on vascular
tone, nitroprusside also increases large solute clearance through some other
effect on the peritoneal transport process, perhaps through an effect on total
vascular pore area.
This hypothesis is supported by direct microcirculatory evidence (Fig. 11) in the
rat cremaster muscle [98] and mesentery [101] that nitroprusside induces the
leakage of protein out of the vasculature. Whether this drug effect is the result of
histamine release or a direct effect of nitroprusside remains to be investigated.
Our initial experiments using direct in vivo microscopy have shown (Fig. 11) that
both histamine and nitroprusside will produce substantial leakage of albumin and
that leakage induced by isoproterenol is minimal. Since the arteriole dilation
produced by isoproterenol was similar to that produced by histamine and nitro-
prusside, leakage does not appear to be dependent on changes in vascular
diameter. It is also of interest that leakage ioducedby histamine occurs at
72

concentrations which produce minimal changes in diameter while leakage in-

duced by nitroprusside occurs only at concentrations that produce maximal
vasodilation. Recent data from our laboratory [33] demonstrate however that
during peritoneal dialysis in normal rats, nitroprusside does not cause protein
leakage. Although there are a number of experimental differences between these
seemingly conflicting data, the answer to this dichotomy remains unresolved.
Since drugs can produce vascular permeability changes as well as vasoactive
effects, their release or activity during peritoneal dialysis could vastly alter
patterns of solute clearance through either direct effects on permeability and
blood flow, or by blockade of the normal effects of endogenous substances. The
interaction of drugs used to treat various disease states can thus further compli-
cate the final effect on protein loss during peritoneal dialysis. Beta-adrenergic
blocking agents for instance may prevent the normal inhibitory effect that beta-
adrenergic stimulation has on autacoid induced protein leakage [95, 102, 103]. In
addition, prostaglandin synthesis inhibitors [~9] significantly attenuate histamine
induced protein leakage in the microcirculation although prostaglandins them-
selves do not produce protein leakage [104]. These results are consistent with
other experimental findings. PGE effects have been shown [105] to be synergistic
to the action of histamine in the dog forelimb causing edema formation; also
human endothelial cells in culture release prostaglandins when incubated with
histamine [106]. In addition, Juan and Sametz [107] have shown that histamine
induces PGE z synthesis in the isolated, perfused rabbit ear preparation and that
PGE1 is necessary for the ensuing protein leakage. The importance of prostaglan-
dins in protein leakage could explain enhanced protein permeability in disease
states such as renovascular hypertension [22, 23, 108] where prostaglandin syn-
thesis appears to be increased [109, 110, 111].
In contrast to the apparent dependence of histamine on prostaglandins for the
production of protein leakage, the effect of histamine on vascular smooth muscle
does not appear to be dependent on prostaglandins. Prostaglandins are vasodila-
tors in the cremaster preparation [112] but prostaglandin synthesis blockade with
either indomethacin or mefenamic acid had no effect on the arteriole dilation
produced by histamine in the rat skeletal muscle [99]. A similar finding has been
reported by Guth and Moler [113] for gastric arterioles. This suggests to us an
action of histamine to cause leakage which is independent of arteriole dilation or
venous constriction and dependent on prostaglandin synthesis.
It is, however, particularly interesting to note that Van de Voorde and Leusen
[114] demonstrated that while indomethacin does not block the dilatory effect of
histamine in the rat hindquarter, ETYA and quinacrine do inhibit this dilation.
This suggests that histamine stimulated metabolites of arachidonic acid other
than prostaglandins (possibly leukotrienes or chemotactic lipids) mediate the
dilation caused by histamine while the prostaglandins mediate histamine induced
protein leakage. In addition to prostaglandin synthesis, calcium mobilization also
appears to be very important for autacoid effects. Juan [115] reported that the
 73

action of bradykinin is via an effect on calcium influx which results in activation of

phopholipase A2 leading to prostaglandin synthesis. Reports by Mayhan et al.
[116] in the hamster cheek pouch and Miller et al. [117[ in the rat cremaster muscle
demonstrated that calcium is necessary for histamine-induced leakage since
leakage may be blocked by the calcium channel blocker verapamil. Besides
altering prostaglandin synthesis, calcium influx could also contribute to endo-
thelial cell contraction and formation of the endothelial gaps [16, 35, 118]. This
gap formation is potentiated by prostaglandins [119] and it is possible that his-
tamine induces venular gap formation via a mechanism dependent, at least in
part, on prostaglandins and calcium influx. Thus the tissue environment or
perturbations during peritoneal dialysis which effect autacoid release may inter-
act with drugs used to treat common diseases to alter macromolecular loss. The
effects of peritoneal dialysis solutions with their high osmolality, low pH, and
non-bicarbonate buffer anion to release autacoids or alter drug activity should be
a subject for future investigations.

5. Effects of peritoneal dialysis solutions on the microcirculation and on solute

clearances

It is now well established that the commercial solutions used for peritoneal
dialysis in humans have a vasoactive effect. Topical application to arterioles of the
rat cremaster muscle [8] or arterioles and venules on the mesothelial surface of
the rat cecum [82] primarily produces vasodilation. In addition, it has recently
been demonstrated in the cat using radioactive microspheres [120] that this
vasodilation is associated with large increases in blood flow to major exchange
areas (mesentery, parietal peritoneum, omentum, and intestinal serosa). How-
ever, blood flow to the large organs such as the liver, stomach, intestine, pancreas
and spleen were unaffected by commercial dialysis solutions. Alterations of the
vasoactive effects of peritoneal dialysis solutions can lead to changes in the
patterns of solute clearances during clinical peritoneal dialysis [9]. Thus, it is
important to determine which constituents of these commercial solutions are
vasoactive and ultimately, how these individual constituents affect solute cleara-
nces during peritoneal dialysis. These commercial peritoneal dialysis solutions
are acidic (pH 5.2-5.8), hypocapnic, contain either acetate (McGaw®) or lactate
(Inpersol®, Dianeal®) as a buffer anion, and these solutions are hyperosmolar
(340-360 mOsm) relative to plasma due to high dextrose concentrations. Acidity,
hypocapnia, high osmolality, dextrose or the buffer anion could be responsible
therefore for the vasoactive effects of peritoneal dialysis solutions.
74

5.1. Osmolality

It has been shown that high osmolality will affect solute clearances. The use of a
high osmolality dialysis solution decreases dialysate protein concentrations in
comparison to those obtained with a lower osmolality dialysis solution [9, 121].
Zelman et al. [122] have increased peritoneal dialysis efficiency by alternating the
use of two levels of hyperosmotic peritoneal dialysis solutions in the goat. Brown
et al. [77] found that many intraperitoneal vasodilators would give some increase
in peritoneal solute clearances in the rat; yet, larger increases in solute clearances
could be achieved with a 4.25% dextrose dialysis solution. Since these larger
clearances persisted throughout subsequent exchanges with 1.5% dextrose di-
alysis solution, ultrafiltration alone could not have been the sole mechanism for
the increased solute clearances with the 4.25% dextrose dialysis solutions. In
humans, Henderson and Nolph [123] found that the use of a 7% dextrose
peritoneal dialysis solution increased the clearances of urea and inulin, and that
this increase also persisted during subsequent exchanges with a 1.5% dextrose
dialysis solution. Since high osmolality apparently has a persistent effect on solute
clearances after the end of the high osmolality exchange period [77, 123], high
osmolality solutions may induce cellular or interstitial changes that are not readily
reversible. Regardless of the mechanism, these data indicate that there are
complex effects of osmolality on solute clearances during peritoneal dialysis.
Since 1.5% Dianeal solution has been shown to increase mesenteric blood flow
and 4.5% Dianeal solution produces an even greater increase [120], one of these
effects appears to be vasodilation.
In most experimental animal preparations hyperosmolality produces vasodila-
tion. Hyperosmolar solutions abolish the electrical and mechanical activity of rat
portal veins [124,125] and will dilate small arteries of the hamster cheek pouch and
cremaster muscle [101, 126]. Hyperosmolar solutions of urea, dextrose and so-
dium chloride decrease perfusion pressure of the dog forelimb preparation [127,
128] and the cat ileum [129].
Other studies have also suggested that there are multiple effects of hyperosmo-
lar solutions. In the in vitro mesenteric ring preparation [130] hyperosmolar
solutions of dextrose, sucrose, NaCl and acetate all produced a transient (2-4
minutes) relaxation of the norepinephrine contracted tissue. Ten minutes after
the initial application of the hyperosmolar solution, tension had increased to
149 ± 7% of the norepinephrine-induced tension. A hyperosmolar mannitol solu-
tion [131] produces vasodilation and also inhibits both vasoconstrictor and vas-
odilatory responses in the perfused rabbit ear artery. Contractile responses to
angiotensin, norepinephrine and potassium in spontaneously active venous
smooth muscle of the rat portal vein [132, 133] were first depressed and then, after
continued exposure were augmented by solutions made hypertonic by the ad-
dition of sucrose or NaCl. In the canine renal and forelimb vascular beds, an
infusion of hyperosmotic NaCl, urea, or dextrose decreases vascular resistance;
 75

however with time, resistance gradually increases to eventually reach levels

above control values [127, 134]. These experiments all suggest that hyper-
osmolality not only produces dilation of resistance vessels but also induces a
secondary change in the active state of vascular smooth muscle to give a greater
contraction for a given level of smooth muscle contractile inputs.
It does not appear that the high dextrose concentration in dialysis solutions are
responsible for the vasodilatory effects of hyperosmolality. Both Mellander et al.
[124] and Miller et al. [101] failed to find an effect of a high dextrose, normal
osmolality solution. It is also important to note that although the qualitative effect
of hyperosmolality is vasodilation, different constituents (dextrose, sucrose,
sodium chloride or urea) will produce quantitatively different rates of dilation
and maximal dilations at the same osmolality [127, 129, 138]. Table 2 demon-
strates this point by giving the rate constants for dilation and the maximal effect
produced by topical application of different hyperosmolar solutions on arterioles
of the rat cremaster muscle. In addition these data show that a high osmolality
Krebs-acetate solution produces the same maximal effect and rate of dilation as
does McGaw® peritoneal dialysis solution which also contains acetate and is at a
high osmolality. Interestingly, the vasodilator nitroprusside, while producing
approximately the same maximal dilation as the other solutions, has a much faster
rate of dilation than the other solutions. This is probably the difference between a
receptor mediated and non-receptor mediated event.
Besides vasodilatory effects, hyperosmolar solutions could also affect solute
transport in the interstitial tissue. Electron microscopy has been used to identify
and clarify the role of intervascular and intravascular pathways for macromolecu-
lar transport which exist for various microvascular segments [35]. Others [135,
136] have studied macromolecular transport using limb perfusion and lymph

Table 2. Effect of experimental solutions on arterioles of the rat cremaster muscle.

Control diameter Maximal diameter Rate constant

(/1-, i±SEM) (/1-, i±SEM) of dilation
(sec-I, i ± SEM)

Krebs-dextrose 6.1±.43 IS.1 ± 1.S .21 ± .06

(355mOsm)
Krebs-sodium chloride 5.6 ± .40 14.0 ± 1.6 1.05 ± .27
(355mOsm)
Krebs-sucrose 5.6± .61 15.3 ± 1.2 43 ±.13
(355mOsm)
Krebs-acetate (37 mM) 7.2± .70 24.S ± 1.S .52± .OS
(355mOsm)
McGaw® 9.7 ± .96 25.1 ± 2.0 .54± .11
(355mOsm)
Krebs-nitroprusside 1O.7±0.S 23.2 ± 2.4 2.17 ± .63
(290 mOsm)
76

collection techniques. The results of these lymph collection studies have led to a
current hypothesis that the transport of macromolecules from the blood to the
lymph involves transport through free-fluid channels of a gel-like matrix in the
interstitium [54]. Since peritoneal dialysis with a normal osmolality solution may
increase loss of protein into the peritoneum in man [8, 9] and in both normoten-
sive and hypertensive rats (Fig. 10) we hypothesized that the dialysis procedure
could involve an alteration of macromolecular transport through this inter-
stitium. This hypothesis assumed that during peritoneal dialysis with a normal
osmolality solution the interstitium would be relatively hydrated as opposed to its
normally dehydrated state [5] with relative exclusion of macromolecules [137].
Thus, the interstitial resistance would be low with large, open, free fluid channels
(Fig. 8). This situation would be conducive to transport of high molecular weight
solutes such as protein. We suggested that exchanges with a hypertonic dialysis
solution would be expected to create a dehydrated state of the interstitium to give
free fluid interstitial channels which would be relatively narrowed and tortuous
and thus interstitial transport of large solutes like protein would decrease.
However, in a test of this hypotheses, Miller et al. [138] found that Krebs
solution which differed only in osmolality, gave similar protein concentration
curves for successive dialysis exchanges regardless of the initial osmolality (Fig.
12). Our conclusion was that if the interstitium does perform a regulatory role in
the loss of protein, the interaction of at least two factors, osmolality and pH
(charge) are probably important. In both studies that have found an increased
macromolecular loss with a Krebs dialysis (Fig. 10 and Ref. 9) both pH and
osmolality have been changed simultaneously. This may reflect a role of the low
pH of dialysis solutions on mechanisms of macromolecular transport. Others [62,
139] have suggested that a protein coat which lines the endothelium may influ-
ences the passage of proteins through the vascular endothelium. In addition,
Simionescu et al. [64] have identified a negative charge on the endothelium which
is not homogeneous in nature. There are areas of the endothelium without the
negative charge through which they feel proteins may be funneled into the
interstitium. If this protein coat and the negative charges associated with it may
also line the intercellular in the interstitium (Fig. 8), we might expect an effect
that would inhibit protein movement. Since most plasma proteins are negatively
charged they would be repelled by the negative charges on the proteins which
coat the channels. It would also explain the observations [137] that there is a
restricted intercellular space for macromolecules. Furthermore, pH may influ-
ence the elasticity of the surface layer protein [139]. Thus in the use of peritoneal
dialysis solutions with a very low pH component, the rigidity of these surface
layer proteins could increase therefore increasing the resistance to protein move-
ment.
 77

150

I
x- ---l( HYPEROSMOLAR KREBS (355 mOsm) followed by
~ NORMAL KREBS (290 mOsm). N=6

"'0
...... \ 0----0 NORMAL KREBS (290 mOsm) followed by
C> 100 \ 0----0 HYPEROSMOLAR KREBS (355 mOsm). N=6
E

z
.-0
w
0:
a.
50
w
.-<I:
C/)
>-
-J
<I:
a
0
2 3 4 5 6 7 8 9 10 II 12
EXCHANGE NUMBER

Figure 12. Each patient underwent twelve consecutive dialysis exchanges (10 min instillation, 30 min
dwell, 20 min drainage): six exchanges with one solution were followed by six exchanges with the
second solution. On another day the same procedure was followed but the order of the solutions was
reversed. Data are plotted as the X ± SEM (from Miller et al. [138]; reprinted with permission).

5.2. pH and peo2 : general effects

Peritoneal dialysis solutions are normally acidic and extremely hypocapnic. It has
been hypothesized [140] that the acidity (pH -5.5) could alter peritoneal dialysis
efficiency through an effect on vascular smooth muscle tone. Although most
studies on the effects of pH have suggested that small vessels tend to dilate as pH
is decreased [141, 142, 143] the microcirculatory effects of a pH as low as 5.5 have
not been extensively investigated. On the other hand, since hypocapnia results in
vasoconstriction, the combination of hypocapnia and low pH could result in
vasodilation, no change, or vasoconstriction.
A number of investigators [142, 144, 145] have demonstrated that hypercapnia
results in dilation of resistance vessels. Duckles [146] observed depression of
serotonin contractions in an in vitro cat bronchi preparation after addition of
either carbon dioxide or acid to the in vitro bath and other investigators [147]
found that a combination of high peo 2 and low pH decreased the sensitivity of
small arteries in the rat cremaster to the constrictor effects of norepinephrine.
Both intracellular and extracellular pH may be involved since in both muscle and
brain, intracellular and extracellular pH decreased in response to breathing an
78

inspired concentration of 10% CO 2 for one hour [148]. Duling [144] suffused the
hamster cheek pouch with O 2 and CO 2 equilibrated solutions and measured
microvascular diameters. Carbon dioxide produced vasodilation which he at-
tributed to an effect of CO 2 on the oxyhemoglobin dissociation curve to alter P02
and produce smooth muscle contraction. Alternatively, CO 2 may affect adre-
nergic receptor activity in arteries. Atkinson and Rand [149] observed decreased
blood pressure responses to both alpha- and beta-adrenergic agonists in rats
breathing 10% CO 2 • They suggested that hypercapnia, possibly through a de-
crease in pH, alters the ionization of receptor proteins or has a direct depressant
effect on vascular smooth muscle. It is unlikely, however, that the ionization
fraction of cathecholamines in the pH range of 6.8-7.4 is altered enough to effect
the response [150] although at the pH of dialysis solution (5.2-5.8) this might
occur.
It has been hypothesized [147, 151, 152] that these effects of carbon dioxide and
pH could lend a functional significance to the anatomical arrangement of the
microcirculation in the parietal and visceral vasculatures. The close proximity and
parallel arrangement of small arteries and veins (Figs. 2, 3 and 4) and the high
diffusibility of carbon dioxide [153] suggested that [147,151] there is a countercur-
rent exchange of carbon dioxide from the small veins to the small arteries. This
exchange of carbon dioxide provides a negative feedback control of the micro-
vasculature by altering the sensitivity of the small artery to alpha- and/or beta-
adrenergic receptor stimulation. Such a local control mechanism of tissue blood
flow might be abolished by the pH effects of dialysis solution since others have
shown profound effects of acidity on vasoactivity of drugs and on microcircula-
tory control mechanisms [154]. However, local pH changes failed to alter the
microvascular response to systemic hypoxia in the rat cremaster [155].

5.3. pH and PCO[adjusted peritoneal dialysis solutions

Despite the experimental evidence that a decrease in pH from 7.4 to 6.9 produces
vasodilation, it has been difficult to demonstrate an effect of the very low pH (5.5)
of peritoneal dialysis solutions on the microcirculation or on clearances during
human peritoneal dialysis [8]. The microvascular effects of Inpersol® (pH ~5.8)
on arterioles of the rat cremaster are shown in Figure 13. Inpersol®, (control
solution A) caused an initial constriction followed by a prolonged dilation. pH
adjustment to 7.4 (solution B) had no effect on this response.
In the human peritoneal cavity [28] the pH and PC0 2 of the peritoneal dialysis
solution increases rapidly during the first fifteen minutes of a one-hour exchange
(Fig. 14). Thus, in another group of animals the dialysis drainage solutions from
one patient were used instead of fresh solutions to determine if the microvascular
effects of the dialysis solutions were changed by dwell in the human peritoneum.
The effects of solutions A and B after one hour dwell in the peritoneum are shown
 79
30

20
A B
E
<- 10
0:
W
I- KREBS DIALYSIS
w
::2! 0 I SOLUTION SOLUT,ION , ,
~ 0 10 15 0 5 15
0

w
-1 30
0
0:
W
I-
0: 20
«
A -DRAINAGE B -DRAINAGE

10

o 5 10 15
TI ME (minutes)

Figure 13. Effects of pH adjustment of dialysis solutions on the microvascular response of arterioles in
the rat cremaster. The cremaster muscle was placed in a tissue bath with the blood vessels and nerves
from the animal to the cremaster still intact. In vivo television microscopy was used to measure
vascular diameters. A five-minute control period with Krebs solution bathing the cremaster was
followed by a ten-minute exposure to a peritoneal dialysis solution. Solution 'A' is the commercial
solution Inpersol® (pH -5.8). Solution 'B' is Inpersol® adjusted with sodium hydroxide to pH = 7.4.
'A-drainage' is solution A after one hour dwell in the human peritoneum and solution 'B-drainage' is
solution B after one hour dwell in the human peritoneum.

in the lower left panel (A - drainage) and lower right panel (B - drainage)
respectively of Figure 13. During the hour equilibration in the human peritoneal
cavity the pH of solution A increased from 5.8 to 7.2 and the pH of solution B
slightly decreased from 7.4 to 7.2.
Since Inpersol® at pH == 5.8 (solution A) produced the same responses as
Inpersol® adjusted to pH == 7.4 (solution B) it is unlikely that the relatively low
pH (5.3-5.8) of commercial peritoneal dialysis solutions has a significant effect on
the responses to these solutions. This conclusion was further supported by the
microcirculatory responses to the drainage solutions in which pH had partial
adjustment by equilibration with body fluids. Responses to these solutions
(A -drainage and B-drainage) were not distinguishable from the results obtained
using the fresh dialysis solution (A and B).
In additional experiments with the cremaster arterioles bicarbonate was added
80

7.5
7
6.5
DIALYSATE
pH 6D
(MEAN:!: SEM, no 3-6)

, , , ! ! I I , , ! , ,

-3 63 9 15 21 27 33 39 45 51 57 63

DIALYSATE
pC0 2
(mmHg)

-9 -3" 3 9 15 21 27 33 39 45 51 57 63
24

20

DIALYSATE I6
[ HC0 3] 12
(mEq!L) 8

-9 -3 0 3 9 15 21 26 33 39 45 51 57 63
TIME (min)

Figure 14. Equilibration of pH and PC02 during peritoneal dialysis in humans. Dialyate was sampled
every three minutes for the first half hour of a one-hour dwell in the human peritoneum using
McGaW® peritoneal dialysis solution. The mean values (± SEM) are plotted vs time of the exchange
(published with permission).

to McGaw® dialysis solution in five animals to give a solution with a low pH and
normal peo2 in comparison to McGaw® at a low pH and low peo2 • These data
are seen in Table 3. The normal peo2 (44 AF2mmHg) did not effect control
arteriolar diameters or the mean dilation produced by unaltered McGaw per-
itoneal dialysis solution. Thus in the cremaster of parietal peritoneal vasculature
the low pH or low peo2 of peritoneal dialysis solutions does not appear to effect
the microvascular response to these solutions. pH adjustment of McGaw® per-
itoneal dialysis solution with sodium hydroxide also has no effect on the microcir-
culatory response to small arteries and veins in the cecum or visceral peritoneum
(Table 4). The microcirculatory data in the cremaster and cecum correlated with
clinical data in which solute clearances do not change with pH adjustment of
 81

Table 3. Effect of PC0 2 on the response of rat cremaster arterioles to dialysis solution.

PCO z pH Arteriole diameter (t-tm)

(mmHg)
Control Response
(i±SEM) (i± SEM)

McGaW® (N = 5) 9± 1 5.7 ± .05 11.3±1.13 * 19.7±1.53

NS NS
McGaW® bicarbonate (N = 5) 44 ± 2 6.1 ± .07 10.0 ± 0.76 * 19.7 ± 1.50

• p<0.05.

Table 4. Effect of pH on the response of rat cecum venules and arterioles to dialysis solutions.

pH Venule diameter Arteriole diameter

(t-tm, i± SEM) (t-tm, i±SEM)

Control Response Control Response

McGaw® (N = 6) 5.8 157 ± 6.1 • 245 ± 5.4 n± 3.9 * 105 ± 5.5

NS NS NS NS
McGaw® (N = 6) 7.4 171±13.1 249 ± 10.8 75 ± 4.3 106 ± 3.2
(adjusted with NaOH)

• p<0.05.

Inpersol® [8]. Similar clinical findings have also been reported with other brands
of dialysis solutions [10]. Granger et al. [120] have confirmed these conclusions
since tyro des solution and tyrodes lactate with a pH5.3 did not produce different
effects of blood flow in the peritoneal cavity of the cat. Although low pH by itself
may not directly affect solute clearances, it may influence drug absorption from
the peritoneal cavity [61] or alternatively drug clearance in the dialysate. Cer-
tainly the pH may affect the proportion of drug bound to protein and this would
be influential in the clearance of drugs with peritoneal dialysis.

5.4. Acetate and lactate

The buffer anions, acetate and lactate, are important constituents of peritoneal
dialysis solutions and the vasoactive effects of these anions could influence the
efficiency of peritoneal dialysis. Liang and Lowenstein [156] showed that acetate
infusion will increase skeletal muscle and intestinal blood flow suggesting an
acetate induced relaxation of the arterial vasculature. However, a similar effect
oflactate has not been reported. A buffer anion plus hyperosmolality is necessary
to give quantitatively the same dilation as peritoneal dialysis solutions (Fig. 15).
82

30

0::
W
I- 20
w
!i:
,,
<[
is
,
I

wE .,
I

...J.3
0
0::
w 10
I-
0::
<[

DEXTROSE (15g/I),
290 mOsm

KREBS _ -+-___ EXPERIMENTAL SOLUTION

o SOLUTION

o 5 10 15
TIME (minutes)

Figure 15. Effects of Krebs-acetate, Krebs-dextrose and McGaw peritoneal dialysis solution on
diameters of small arterioles in the rat cremaster muscle. Experimental set-up is the same as in Figure
13. Each curve represents the group data (X ± SEM, N = 6) during a 5-min control period with Krebs
solution (285 mOsm, 2.1 gil dextrose, no acetate) bathing the cremaster and a lO-minute experimental
period with Krebs-acetate, Krebs-dextrose or McGaw peritoneal dialysis solution bathing the cremas-
ter.

In an in vitro experiment, with mesenteric artery rings [130] both acetate and
lactate produced a transient dose-dependent contraction of the norepinephrine
contracted ring (Fig. 16) followed by a more prolonged dose dependent relaxa-
tion . The 45 mM acetate found in McGaw dialysis solution was a supramaximal
concentration for acetate induced dilation [130] and at 30 mM acetate and high
concentrations, acetate produced significantly more relaxation than lactate. Thus
one would expect acetate containing peritoneal dialysis solution to have greater
vasoactive effects. This was not observed in the cremaster arterioles where three
different solutions gave the same amount of vasodilation [8]. However, in the
visceral peritoneum (Fig . 17), the McGaw®solution which contains acetate as the
buffer anion did give far more dilation than the Dianeal® solution which con-
tained lactate.
In a set of clinical experiments [10] the peritoneal clearances of four different
solutes were not different among three commercial brands of dialysis solutions.
Thus , chemical differences in the composition of these solutions, for instance
lactate in Inpersol® and Dianeal® vs acetate in McGaw®, may account for some
differences in the microcirculatory responses to these solutions; however, these
 83

1.0
NE No ACETATE,45mM (285mOsm)
~ ~
0.5
0>

0
Z
0

,
(f)
z 1.0
W
t-
NE No LACTATE,45mM (285mOsm)

0.5

0
, , i i
0 6 12 18
TIME (minutes)
Figure 16. Effect of acetate and lactate on norepinephrine contracted in vitro, mesenteric rings. An
EDsu concentration of norepinephrine was added to the in vitro tissue bath containing Krebs solution.
When the tissue had maximally contracted, Na acetate (top panel) or Na lactate (bottom panel) was
added to the bath such that the osmolality of the solution remained at the osmolality of the Krebs
solution (285 mOsm).

1.5% McGAW
1.5% DIANEAL
x-x-x 0.5 % DIANEAL

ARTERIOLE (65,um) VENULE ([Link])

lL.-.J
00
a:
t
---
t-t-125 ,...1_
Zz
Wo y-~

Uu
a:
_/'- ~lV"~
xrlV"lI-.:z:
lLJ
a.. 75 i i
0 10 20 30 0 10 20 30

TIME (minutes)

Figure 17. Effect of peritoneal dialysis solutions on the vascular diameter of arterioles and venules on
the mesothelial surface of the rat cecum. Experimental set-up is the same as given in Figure 9. After a
five-minute control period with Krebs solution bathing the cecum, the bath was drained and refilled
with peritoneal dialysis solution for a twenty-minute experimental period. This sequence was re-
peated for each solution. Arteriole and venule diametes are plotted as a percent of the mean value
during the control period which preceded the application of each solution.
84

chemical differences do not appear to give similar differences for peritoneal

clearances in humans.
During peritoneal dialysis, the peritoneal blood flow may be at a level where
hyperosmolality has produced enough dilation of peritoneal arterioles such that
differences in buffer anions among the commercial solutions did not further alter
peritoneal blood flow, solute delivery or peritoneal clearances. If this is true,
vasodilators which alter peritoneal clearances in humans when added to dialysis
solutions [78] must do so by a mechanism other than vasodilation of arterioles.
One hypothesis is that at least some vasodilators increase vascular permeability
(see Section 4).
The mechanisms by which acetate and lactate may cause arterial dilation could
be related to chelation of ionized calcium [157, 158]. Alternatively, these anions
may affect calcium release which is involved in contraction. Lactate is rapidly
taken up into coronary vascular smooth muscle [159] and increases the binding of
calcium to high affinity sites. These are apparently the same sites that are
activated by PGF2a to induce contraction and lactate blocks contraction of this
prostaglandin. An affect of lactate on calcium binding may be especially impor-
tant since it has been recently shown [117] that calcium is necessary for histamine
induced protein leakage. Calcium probably also contributes to endothelial cell
contraction and the formation of endothelial gaps [16, 35,118]. This gap formation
may be important for protein to leak out of the vasculature into the interstitium.
There are other possible effects of lactate such as a direct effect to increase plasma
insulin [160] but the vascular effects of acetate and lactate in peritoneal dialysis
solutions are more likely associated with metabolism of these anions.
Various tricarboxylic acid cycle metabolites produce dilation [158] and the
hemodynamic effects of acetate were abolished by pretreatment with fluoroace-
tate [156] which inhibits acetate utilization [161]. These data suggest that the
effects of acetate are not caused directly by acetate itself, but rather are produced
by a metabolite of acetate.
The effects of acetate on the circulation resulting in arterial dilation may be
associated with the formation of high-energy phosphates that accompanies ace-
tate activation [156, 158]. The activation of acetate to acetyl CoA by acetate
thiokinase requires A TP and yields AMP which is a highly potent vasodilator
[162]. In addition, adenosine, which causes vasodilation of the mesenteric artery
[163] and increases blood flow in dog intestine [164] can be formed from AMP. In
contrast, the conversion of lactate to acetyl CoA via pyruvate does not yield
AMP. This could explain why lactate containing solutions caused less relaxation
than acetate containing solutions in some preparations (Fig. 17).
The effects of adenosine and the adenine nucleotides in general, may be an
important ingredient in the vasoactive effects of peritoneal dialysis solutions
especially since the purinergic nervous system may be so important in the regu-
lation of visceral peritoneal blood flow [165]. The production of adenosine by
activation of the buffer anions of dialysis solutions or the alteration of the
 85

physiological role of the purinergic nervous system by the addition of drugs to

dialysis solutions is suggested by other investigators in this area. For instance,
aminophylline has been shown to decrease, and dipyridamole to increase urea
clearances with peritoneal dialysis in rabbits [166]. Aminophylline is a similar
compound to theophylline and has been used by various investigators as a
competitive inhibitor of adenosine to evaluate the physiological role of the
adenine nucleotides and their breakdown products [70, 165, 167, 168]. On the
other hand, dipyridamole, has been used to potentiate the effects of adenosine by
inhibiting adenosine uptake into nerve terminals or inhibiting adenosine de-
aminase activity [165, 169]. Either mechanims would result in increased levels of
adenosine near the adenosine active sites. Thus one might expect that if ade-
nosine is involved in the mechanisms regulating solute clearances during per-
itoneal dialysis, inhibition of adenosine (by aminophylline) would decrease clear-
ances and potentiation of adenosine (by dipyridamole) would increase clearances
as Maher et al. [166] have indeed found.

6. Summary

The regulatory mechanisms of the peritoneal vasculature are responsible for the
control of blood flow to the peritoneum and therefore for the delivery of solutes
to the areas for solute transfer across the endothelium. Some of these same
mechanisms may also modify the permeability of the vasculature and thus the first
resistance pathway (endothelial transfer) for solute movement to the peritoneal
cavity. Regulation of peritoneal blood flow and vascular permeabillity becomes a
function then of hormonal and neural inputs as well as the environment of the
peritoneal microcirculation (pH, paz, PCO z osmolality, ions, etc.). This environ-
ment may be strongly influenced by the composition and physio-chemical charac-
teristics of the peritoneal dialysis solutions. In addition, the anatomical arrange-
ment of the vasculature, vascular smooth muscle tone, endothelial protein coat,
endothelial charge, transfer processes in the endothelium and the mesothelium,
and the structure and charge of the interstitial tissue are important factors which
may participate in the eventual regulation of dialysate composition. Various
attempts have been made to alter control mechanisms of the peritoneal microcir-
culation by altering the composition of periton~al dialysis solutions. Many of
these attempts which include altered pH, osmolality, buffer anions, PCO z and the
addition of various drugs have altered peritoneal dialysis efficiency. Currently,
however, too little is known about the interactions of peritoneal dialysis solutions
with these hormonal, neural, or drug inputs, to predict how the composition of
peritoneal dialysis solutions can be altered to produce specific increases or
decreases in specific solutes. These interactions are the focus of much of the
research which is currently in progress on control of the peritoneal microcircula-
tion and the efficiency and solute specificity of peritoneal dialysis.
86

References

1. Crafts RC: A Textbook of Human Anatomy. The Ronald Press Company, New York, 1966.
2. Frasher WG Jr, Wayland H: A repeating modular organization of the microcirculation of cat
mesentery. Microvasc Res 4: 62-76,1972.
3. Lukas G, Brindle SD, Greengard P: The route of absorption of intraperitoneally administered
compounds. J Pharmacol Exp Ther 178: 562-566, 1971.
4. Casley-Smith JR: Calculations relating to the passage of fluid and protein out of arterial-limb
fenestrae through basement membranes and connective tissue channels, and into venous-limb
fenestrae and lymphatics. Microvasc Res 12: 13-34, 1976.
5. Granger HJ: Role of the interstitial matrix and lymphatic pump in regulation of transcapillary
fluid balance. Microvasc Res 18: 209-216, 1979.
6. Fraser PA, Smaje LH, Verrindes A: Microvascular pressures and filtration coefficients in the cat
mesentery. J Physiol (London) 283: 439-456, 1978.
7. Casley-Smith JR, Sims MA: Protein concentrations in regions with fenestrated and continuous
blood capillaries and in initial and collecting lymphatics. Microvasc Res 12: 245-257, 1976.
8. Miller FN, Nolph KD, Harris PD, Rubin J, Wiegman DL, Joshua IG, Twardowski ZJ, Ghods
AJ: Microvascular and clinical effects of altered peritoneal dialysis solutions. Kidney Int 15: 630-
639,1979.
9. Rubin J, Nolph KD, Arfania D, Joshua IG, Miller FN, Wiegman DL, Harris PD: Clinical
studies with a nonvasoactive peritoneal dialysis solution. J Lab C1in Med 93: 910-925, 1979.
10. Nolph KD, Rubin J, Wiegman DL, Harris PD, MillerFN: Peritoneal clearances with three types
of commercially available peritoneal dialysis solutions. Effects of pH adjustment and intra-
peritoneal nitroprusside. Nephron 24: 35-40,1979.
11. Bohlen HG, Gore RW: Comparison of microvascular pressures and diameters in the innervated
and denervated rat intestine. Microvasc Res 14: 251-264, 1977.
12. Baez S: An open cremaster muscle preparation for the study of blood vessels by in vivo
microscopy. Microvasc Res 5: 384-394, 1973.
13. Smaje L, Zweifach BW, Intaglietta M: Micropressures and capillary filtration coefficients in
single vessels of the cremaster muscle in the rat. Microvasc Res 2: 96-110, 1970.
14. Rhodin JAG: Histology: A Text and Atlas. Oxford University Press, New York, 1974.
15. Noer RJ: The blood vessels of the jejenum and ileum: A comparative study of man and certain
laboratory animals. Am J Anat 73: 293-334, 1943.
16. Fox J, Galey F, Wayland H: Action of histamine on the mesenteric microvasculature. Microvasc
Res 19: 108-126,1980.
17. Jacobson LF, Noer RJ: The vascular pattern of the intestinal villi in various laboratory animals
and man. Anat Res 114: 85-101, 1952.
18. Allen L, Vogt E: A mechanism of lymphatic absorption from serous cavities. Am J Physiol119:
776-782,1937.
19. Tsilibary EC, Wissig SL: Lymphatic absorption from the peritoneal cavity: regulation of patency
of mesothelial stomata. Microvasc Res 25: 22-39, 1983.
20. Casley-Smith JR: Varying total tissue pressures and the concentration of initial lymphatic
lymph. Microvasc Res 25: 369-379, 1983.
21. McHale NG, Allen JM: The effect of external Ca 2+ concentration on the contractility of bovine
mesenteric lymphatics. Microvasc Res 26: 182-192, 1983.
22. Joshua IG, Wiegman DL, Harris PD, Miller FN: Progressive microvascular alterations with the
development of renovascular hypertension. Hypertension 6: 61-67, 1984.
23. Miller FN, Anderson GL: Alterations in histamine-induced protein leakage during renovascular
hypertension. Microvasc Res 25: 249, 1983.
24. Parving HH, Rossing N, Jensen HAE: Increased metabolic turnover rate and transcapillary
escape rate of albumin in essential hypertension. Circ Res 35: 544-552, 1974.
 87

25. Van Den Brenk H, Crowe M, Kelly H, Stone M: The significance offree blood in liquid and solid
tumors. Br J Exp Pathol58: 147-159, 1977.
26. Katano M, Torisu M: New approach to management of malignant ascites with a streptococcal
preparation OK-432. II Intraperitoneal inflammatory cell-mediated tumor cell destruction.
Surgery 93: 365-373, 1983.
27. Heuser LS, Taylor SH, Folkman JF: Prevention of carcinomatosis in the rat. J Surg Res (in
press).
28. Nolph KD, Miller FN, Rubin J, Popovich R: New directions in peritoneal dialysis concepts and
applications. Kidney Int 18: S111-S116, 1980.
29. Renkin EM: Relation of capillary morphology to transport of fluid and large molecules: A
review. Acta Physiol Scand (Suppl) 463: 81-91, 1979.
30. Chambers R, Zweifach BW: Functional activity of the blood capillary bed, with special refer-
ence to visceral tissue. Ann NY Acad Sci 46: 683-695, 1946.
31. Westergaard E, Brightman MW: Transport of proteins across normal cerebral arterioles. J
Comp Neuro1152: 17-44, 1973.
32. Henriksen JH: Permselectivity ofthe liver blood-lymph (ascitic fluid) barrier to macromolecules
in decompensated cirrhosis: relation to caculated pore-size. Clin Physiol3: 163-171, 1983.
33. Miller FN, Hammerschmidt DE, Anderson GL, Moore IN: Protein loss induced by complement
activation during peritoneal dialysis. Kidney Int 25: 480-485, 1984.
34. Bruns RR, Palade GE: Studies on blood capillaries. II Transport of ferritin molecules across the
wall of muscle capillaries. J Cell Bioi 37: 277-299, 1968.
35. Simionescu N, Simionescu M, Palade GE: Structural basis of permeability in sequential seg-
ments of the microvasculature of the diaphram. II Pathways followed by microperoxidase across
the endothelium. Microvasc Res 15: 17-36, 1978.
36. Casley-Smith JR: Endothelial fenestrae in intestinal villi. Difference between the arterial and
venous ends of the capillaries. Microvasc Res 3: 49-68, 1971.
37. Karnovsky MJ: The ultrastructural basis of transcapillary exchanges. J Gen Physiol52: 645-95S,
1968.
38. Krogh A: The Anatomy and Physiology of Capillaries (rev. and 2nd ed.). Yale University Press,
New Haven, Connecticut, 1929.
39. Chambers R, Zweifach BW: Topography and function of the mesenteric capillary microcircula-
tion. Am J Anat 75: 173-205, 1944.
40. Rhodin JAG: Ultrastructure of mammalian venous capillaries, venules, and small collecting
veins. J Ultrastruct Res 25: 452-500, 1968.
41. Sims DE, Westfall JA: Analysis of relationships between pericytes and gas exchange capillaries
in neonatal and mature bovine lungs. Microvasc Res 25: 333-342, 1983.
42. Nakamura Y, Wayland H: Macromolecular transport in the cat mesentery. Microvasc Res 9:
1-21,1975.
43. Wolff JR: Ultrastructure of the terminal vascular bed as related to function in microcirculation.
In: G Kaley, BM Altura (eds). University Park Press, Baltimore, 1977.
44. Majno G: Ultrastructure of the Vascular Membrane. In: WF Hamilton, P Dow (eds). Handbook
ofPsysiology, Section 2: Circulation Vol. III. Am Physiol Soc, Washington, DC, 1965, pp 2293-
2375.
45. Clementi F, Palade GE: Intestinal capillaries. I Permeability to peroxidase and ferritin. J Cell
Bioi 41: 33-58, 1969.
46. Granger DN, Taylor AE: Permeability of intestinal capillaries to endogenous macromolecules.
Am J Physiol 238: H457-H464, 1980.
47. Johansson BR: Permeability of muscle capillaries to interstitially microinjected horseradish
peroxidase. Microvasc Res 16: 340-353, 1978.
48. Johansson BR: Permeability of muscle capillaries to interstitially microinjected ferritin. Micro-
vasc Res 16: 362-368, 1978.
88

49. Dedrick RL, Flessner MF, Collins JM, Schultz JS: Is the peritoneum a membrane? Am Soc Artif
Intern Organs 5: 1-8, 1982.
50. Laurent TC: II The ultrastructure and physical-chemical properties of interstitial connective
tissue. Pfluegers Arch (Suppl) 336: S21-S42, 1972.
51. Laurent TC: Interaction between proteins and glycosamineoglycans. Fed Proc 36: 24-27, 1977.
52. Fox JR, Wayland H: Interstitial diffusion of macromolecules in the rat mesentery. Microvasc
Res 18: 255-276, 1979.
53. Wiederhielm CA: The interstitial space. In: YC Fung, N Perrone, M Anliker (eds). Bio-
mechanics, its foundations and objectives. Prentice Hall, Englewood Cliffs, New Jersey, 1972,
pp 273-286.
54. Watson PD, Grodins FS: An analysis of the effects of the interstitial matrix on plasma-lymph
transport. Microvasc Res 16: 19-41, 1978.
55. Miller FN, Nolph KD, Joshua IG, Wiegman DL, Harris PD: Hyperosmolality, acetate and
lactate: Dilatory factors during peritoneal dialysis. Kidney Int 20: 397-402, 1981.
56. Wayland H, Silberberg A: Blood to lymph transport. Microvasc Res 15: 367-374, 1978.
57. Gosselin RE, Berndt WO: Diffusional transport of solutes through mesentery and peritoneum. J
Theor Bioi 3: 487-495, 1962.
58. Rasio EA: Metabolic control of permeability in isolated mesentery. Am J Physiol226: 962-968,
1974.
59. Nagel W, Kuschinsky W: Study of the permeability of the isolated dog mesentery. Eur J Clin
Invest 1: 149-154, 1970.
60. Berndt WO, Gosselin RE: Differential changes in permeability of mesentery to rubidium and
phosphate. Am J Physiol202: 761-767, 1962.
61. Torres IJ, Litterst CL, Guarine AM: Transport of model compounds across the peritoneal
membrane in the rat. Pharmacology 17:330-340, 1978.
62. Mason JC, Curry FE, Michel CC: The effects of proteins upon the filtration coefficient of
individually perfused frog mesenteric capillaries. Microvasc Res 13: 185-202,1977.
63. Turner MR, Clough G, Michel CC: The effects of cationized ferritin and native ferritin upon the
filtration coefficient of single frog capillaries. Evidence that proteins in the endothelial cell coat
influence permeability. Microvasc Res 25: 205-222,1983.
64. Simionescu D, Simionescu M: Differentiated distribution of the cell surface charge on the
alveolar-capillary unit. Characteristic paucity of anionic sites on the air-blood barrier. Microvasc
Res 25: 85-100, 1983.
65. Charonis AS, Wissig SL: Anionic sites in basement membranes. Differences in their electrosta-
tic properties in continuous and fenestrated capillaries. Microvasc Res 25: 265-285, 1983.
66. Blumberg AL, Denny SE, Marshall GR, Needleman P: Blood vessel-hormone interactions:
angiotensin, bradykinin and prostaglandins. Am J PhysioI232(3): H305-H31O, 1977.
67. Hedqvist P: Studies on the effect of prostaglandins E1 and E2 on the sympathetic neuromuscular
transmission in some animal tissues. Acta Physiol Scand (Suppl) 345: 1-40, 1970.
68. McGrath MA, Shepherd JT: Inhibition of adrenergic neurotransmission in canine vascular
smooth muscle by histamine: mediation by Hz-receptors. Circ Res 39: 566-573,1976.
69. Hedqvist P, Fredholm BB: Effect of adenosine on adrenergic neurotransmission. Prejunctional
inhibition and post junctional enhancement. Naunyn-Schmiedeberg's Arch Pharmacol293: 217-
223,1976.
70. Wakade AR, Wakade TD: Inhibition of noradrenaline release by adenosine. J Physiol (London)
282: 35-49,1978.
71. Shepherd JT, Vanhoutte PM: Veins and their control. Saunders WB, Philadelphia, 1975.
72. Starke K, Schumann HJ: Interactions of angiotensin, phenoxybenzamine, and propranolol on
noradrenaline release during sympathetic nerve stimulation. Eur J Pharmacol18: 27-30, 1972.
73. Malik KU, Nasjletti A: Attenuation by bradykinin of adrenergically-induced vasoconstriction in
the isolated perfused kidney of the rabbit: Relationship to prostaglandin synthesis. Br J Phar-
macol 67: 269-275,1979.
 89

74. Heitz DC, Brody MJ: Possible mechanisms of histamine release during active vasodilation. Am J
PhysioI28:1351-1357,1975.
75. Davies BN, Horton EW, Withrington PG: The occurence of prostaglandin E2 in splenic venous
blood of the dog following splenic nerve stimulation. Br J Pharmacol Chemother 32: 127-135,
1968.
76. Hare HG, Valtin H, Gosselin RE: Effect of drugs on peritoneal dialysis in the dog. J Pharmacol
Exp Ther 145: 122-129, 1964.
77. Brown EA, Kliger AS, Goffinet J, Finkelstein FO: Effect of hypertonic dialysate and vasodila-
tors on peritoneal dialysis clearances in the rat. Kidney Int 13: 271-277, 1978.
78. Nolph KD, Ghods AJ, Van Stone J, Brown PA: The effects of intraperitoneal vasodilators on
peritoneal clearances. Trans Am Soc Artif Intern Organs 22: 586-594, 1976.
79. Powlik W, Tague LL, Tepperman BL, Miller TA, Jacobson ED: Histamine Hr and H 2-receptor
vasodilation of canine intestinal circulation. Am J Physiol233: E219-E224, 1977.
80. Guth PH, Smith E: Histamine receptors in mesenteric circulaton of the cat and rat. Am J Physiol
234: E370-E374, 1978.
81. Guth PH, Moler TL, Smith E: H j and H2 histamine receptors in rat gastric submucosal
arterioles. Microvasc Res 19: 320-328, 1980.
82. Miller FN, Wiegman DL, Joshua IG, Nolph KD, Rubin J: Effects of vasodilators and peritoneal
dialysis solution on the microcirculation of the rat cecum. Proc Soc Exptl Bioi Med 161: 605-608,
1979.
83. Messina EJ, Weiner R, Kaley G: Microcirculatory effects of prostaglandins Ej, E2 and Aj in the
rat mesentery and cremaster muscle. Microvasc Res 8: 77-89, 1974.
84. Faber JE, Harris PD, Joshua IG: Microvascular response to blockade of prostaglandin synthesis
in rat skeletal muscle. Am J Physiol243: H51-H6O, 1982.
85. Altura BM: Pharmacology of venular smooth muscle: New Insights. Microvasc Res 16: 91-117,
1978.
86. Altura BM, Zweifach BW: Evidence for beta receptors in the rat mesenteric mirovasculature. J
Pharmacol Exp Ther 150: 23--25, 1965.
87. Gillespie JS, Rae RM: Constrictor and complicance responses of some arteries to nerve or drug
stimulation. J Physiol (London) 223: 109-130, 1972.
88. Altura BM: Chemica and humoral regulation of blood flow through the precapillary sphincter.
Microvasc Res 3: 361-384, 1971.
89. Hirszel P, Lasrich M, Maher JF: Augmentation of peritoneal mass transport by dopamine. J Lab
Clin Med 94: 747-754, 1979.
90. Maher JF, Hirszel P, Lasrich M: Effects of gastrointestinal hormones on transport by peritoneal
dialysis. Kidney Int 16: 130-136, 1979.
91. Longnecker DE, Creasy RA, Ross DC: A microvascular site of action of sodium nitroprusside in
striated muscle of the rat. Anesthesiology 50: 111-117, 1979.
92. Raymond RM, Jandhyala BS, Grega GJ: The interrelationship among bradykinin, various
vasoactive substances, and macromolecular permeability in the canine forelimb. Microvasc Res
19: 329-337,1980.
93. Persson CGA, Erjefiilt I: Terbutaline and adrenaline inhibit leakage of fluid and protein in
guinea-pig lung. Eur J Pharmacol55: 199-201, 1979.
94. Marciniak DL, Dobbins DE, Maciejko JJ, Scott JB, Haddy FJ, Grega GJ: Antagonism of
histamine edema formation by catecholamines. Am J Physiol 234(2): H180-H185, 1978.
95. Joyner WL, Svensjo E, Arfors KE: Simultaneous measurements of macromolecular leakage and
arteriolar blood flow as altered by PGEj and ~rreceptor stimulant in the hamster cheek pouch.
Microvasc Res 18: 301-310, 1979.
96. Svensjo E, Arfors KE, Raymond RM, Grega GJ: Morphological and physiological correlation
of bradykinin-induced macromolecular efflux. Am J Physiol236: H600-H606, 1979.
97. Baker CH: Nonhemodynamic effects of histamine on gracilis muscle capillary permeability. J
Pharmacol Exp Ther 211: 672-677, 1979.
90

98. Miller FN, Joshua IG, Anderson GL: Quantation of vasodilator-induced macromolecular
leakage by in vivo fluorescent microscopy. Microvasc Res 24: 56-67, 1982.
99. Anderson GL, Miller FN, Xiu RJ: Inhibition of histamine-induced protein leakage in rat skeletal
muscle by blockade of prostaglandin synthesis. Microvasc Res 28: 51-61, 1984.
100. Hirszel P, Maher JF, Chamberlin M: Augmented peritoneal mass transport with intraperitoneal
nitroprusside. J Dial 2: 131-142, 1978.
101. Miller FN, Joshua IG, Harris PD, Wiegman DL, Jauchem JR: Peritoneal dialysis solutions and
the microcirculation. In: A Trevino-Becerra, FST Boen (eds). Today's Art of Peritoneal
Dialysis. S Karger, Basel, Contrib Nephrol17: 51-58, 1979.
102. Svensj6 E, Foltz R, Grega G: The morphological basis of the antagonism of the histamine-
induced efflex of macromolecules produced by catecholamines in the cheek pouch. of hamsters.
Microvasc Res 21: 259,1981.
103. Svensj6 E, Adamski SW, Su K, Grega GJ: Quantitative physiological and morphological aspects
of microvascular permeability changes induced by histamine and inhibited by terbutaline. Acta
Physiol. Scand. 116: 265-273, 1982.
104. Joyner WL: Effect of prostaglandins on macromolecular transport from blood to lymph in the
dog. Am J Physiol 232: H690-H696, 1977.
105. Amelang E, Prasad CM, Raymond RM, Grega GJ: Interactions among inflammatory mediators
on edema formation on the canine forelimb. Circ Res 49: 298-306, 1981.
106. Alhenc-Gelas F, Tsai SJ, Callahan KS, Campbell WB, Johnson AR: Stimulation of pros tag lan-
din formation by vasoactive mediators in cultured human endothelial cells. Prostaglandins 24:
723-742, 1982.
107. Juan H, Sametz W: Histamine-induced release of arachidonic acid and of prostaglandins in the
peripheral vascular bed. Naunyn-Schmiedeberg's Arch Pharmacol 314: 183-190, 1980.
108. Huttner I, More RH, Rona G: Fine structural evidence of specific mechanism for increased
endothelial permeability in experimental hypertension. Am J Pathol61: 395-417, 1970.
109. Rioux F, Quinon R, Regoli D: The role of prostaglandins in hypertension. I The release of
prostaglandins by aorta strips of renal, doca-salt and spontaneously hypertensive rats. Can J
Physiol Pharmacol 53: 673-677,1977.
110. Pugsley DJ, Beilin L, Peto R: Renal prostaglandin synthesis in the Goldblatt hypertensive rat.
Circ Res (Suppll) 36-37: 81-88, 1975.
111. Romero JC, Strong CG: The effect of indomethacin blockade of prostaglandin synthesis on
blood pressure of normal rabbits and rabbits with renovascular hypertension. Circ Res 40: 35-41,
1977.
112. Faber JE, Harris PD, Miller FN: Microvascular sensitivity to PGE 2 and PGI 2in skeletal muscle
of decerebrate rat. Am J Physiol243: H844-H851, 1982.
113. Guth PH, Moler TL: The role of endogenous prostanoids in the response of the rat gastric
microcirculation to vasoactive agents. Microvasc Res 23: 336-346, 1982.
114. Van de Voorde J, Leusen I: Influence of prostaglandin-synthesis inhibitors on carbachol- and
histamine-induced vasodilation in perfused rat hindquarters. Pfluegers Arch 397: 290-294, 1983.
115. Juan H: Role of calcium in prostaglandin E release induced by bradykinin and the ionophore A
23187. Naunyn-Schmiedeberg's Arch Pharmacol307: 177-183, 1979.
116. Mayhan WG, Gilmore JP, Joyner WL: The effect of a calcium blocker (Verapamil) on the
characterization of microvascular leaky sites in the hamster cheek pouch. Microvasc Res 3: 247,
1983.
117. Miller FN, Joshua IG, Harris PD: Inhibition of histamine-induced protein leak during reno-
vascular hypertension by the calcium entry blocker verapamil. The Pharmacologist 25: 239,
1983.
118. Majno GS, Shea SM, Leventhal M: Endothelial contraction induced by histamine-type media-
tors: an electron microscopic study. J Cell Bioi 42: 647-672, 1969.
119. Svensj6 E: Bradykinin and prostaglandin E" E2 and F2-induced macromolecular leakage in the
hamster cheek pouch. Prostaglandins and Med 1: 397-410. 1978.
 91

120. Granger DN, Ulrich M, Perry MA, Kvietys PR: Peritoneal dialysis solutions and feline
splanchnic blood flow. Clin Exp Pharmacol Physiol (in press).
121. Nolph KD, Twardowski ZJ, Popovich RP, Rubin J: Equilibration of peritoneal dialysis solu-
tions during long-dwell exchanges. J Lab Clin Med 93: 246-256, 1979.
122. Zelman A, Parsons R, Anzola E, Whittman P: Augmentation of peritoneal dialysis efficiency
with programmed hyper-hypoosmotic dialysates. In: BB Mackey (ed). Eleventh Annual Con-
tractor's Conference. Washington, DC, 1978, pp 47-50.
123. Henderson LW, Nolph KD: Altered permeability of the peritoneal membrane after using
hypertonic peritoneal dialysis fluid. J Clin Invest 48: 992-1001,1969.
124. Mellander S, Johansson B, Gray S, Jonsson 0, Lundvall J, Ljung B: The effects of hyper-
osomolarity on intact and isolated vascular smooth muscle. Possible role in exercise hyperemia.
Angiologica 4:310-322,1967.
125. Arvill A, Johansson B, Jonsson 0: Effects of hyperosmolarity on the volume of vascular smooth
muscle cells and the relation between cell volume and muscle activity. Acta Physiol Scand 75:
484-495, 1969.
126. Duling BR, Staples E: Microvascular effects of hypertonic solutions in the hamster. Microvasc
Res 11: 51-56, 1976.
127. Gazitua S, Scott JB, Swindall B, Haddy FJ: Resistance responses to local changes in plasma
osmolality in three vascular beds. Am J Physiol220: 384-391, 1971.
128. Overbeck HW, Molnar JI, Haddy FJ: Resistance to blood flow through the vascular bed of the
dog forelimb: Local effects of sodium, potassium, calcium, magnesium, acetate, hypertonicity
and hypotonicity. Am J Cardiol8: 533-541, 1961.
129. Levine SE, Granger DN, Brace RA, Taylor AE: Effect of hyperosmolality on vascular resis-
tance and lymph flow in the cat ileum. Am J Physiol 234: HI4-H20, 1978.
130. Jauchem JR, Miller FN, Wiegman DL, 10shua IG, Harris PD: Interaction of acetate, lactate and
osmolality on contraction of mesenteric arteries. Microcirculation 1: 37-54, 1981.
131. Krishnamurty VSR, Adams HR, Willerson IT: Paradoxical inhibition of vasoconstrictor and
vasodilator responses by hypertonic mannitol in isolated arterial smooth muscle. Eur 1 Phar-
macol58: 379-388, 1979.
132. Blair-WestJR, McKenzie JS, McKinley Ml: The actions of angiotension II on the isolated portal
vein of the rat. Eur 1 Pharmacol15: 221-230, 1971.
133. McKinley Ml, McKenzie IS, Blair-West lR: Effects of maintained osmolarity changes on rat
portal vein spontaneous contractions. Am 1 Physiol226: 718-723, 1974.
134. Gazitua S, Scott IB, Chou CC, Haddy Fl: Effect of osomolarity on canine renal vascular
resistance. Am 1 Physiol217: 1216-1223,1969.
135. Renkin EM, Joyner WL, Sloop CH, Watson PD: Influence of venous pressure on plasma-lymph
transport in the dog's paw: Convective and dissipative mechanisms. Microvasc Res 14: 191-204,
1977.
136. Renkin EM: Transport pathways through capillary endothelium. Microvasc Res 15: 123-136,
1978.
137. Bell DR, Watson PD, Renkin EM: Exclusion of plasma proteins in interstitium of tissues from
the dog hind paw. Am J Physiol239: H532-H538, 1980.
138. Miller FN, Nolph KD, Sorkin MI, Gloor HI: The influence of solution composition on protein
loss during peritoneal dialysis. Kidney Int 23: 35-39, 1983.
139. Copley AL: The physiological significance of the endoendothelial fibrin lining (EEFL) as the
critical interface in the 'vessel-blood organ' and the importance of in vivo fibrinogenin formation
in health and disease. Thromb Res Suppl V: 105-145, 1983.
140. Nolph KD, Ghods AI, Brown P, Van StoneJ, MillerF, Wiegman D, HarrisP: Factors affecting
peritoneal dialysis efficiency. Dial TranspI6:52-63, 1977.
141. Harris PD, Longnecker DE, Miller FN, Wiegman DL: Sensitivity of small subcutaneous vessels
to altered gases and local pH. Am J PhysioI231(1): 244-251, 1976.
92

142. Kontos HA: Role of hypercapnic acidosis in the local regulation of blood flow in skeletal muscle.
Circ Res (SuppI1-28 and 29): 98--105, 1971.
143. Betz D, Enzenross HG, Vlahov V: Interaction of H+ and Ca++ in the regulation of local pial
vascular resistance. Pfluegers Arch 343: 79-88, 1973.
144. Duling BR: Changes in microvascular diameter and oxygen tension induced by carbon dioxide.
Circ Res 32: 370--376, 1973.
145. Radwaski D, Dabney JM, Daugherty RM Jr, Haddy FJ, Scott JB: Local effects on CO 2 on
vascular resistances and weight of the dog forelimb. Am J Physol222: 439-443, 1972.
146. Duckles SP, Rayner MD, Nadel JA: Effects of CO 2 and pH on drug-induced contractions of
airway smooth muscle. J Pharmacol Exp Ther 190: 472--481, 1974.
147. Wiegman DL, Miller FN, Harris PD: Modification of a-adrenergic responses of small arteries by
altered PC0 2 and pH. Eur J Pharmacol57: 307-315, 1979.
148. Lai YL, Attebery BA, Brown EB Jr: Intracellular adjustments of skeletal muscle, heart, and
brain to prolonged hypercapnia. Respir Physiol19: 115-122, 1973.
149. Atkinson JM, Rand MJ: Reduction of cardiovascular responses to some sympathomimetic
amines during hypercapnia. Eur J Pharmacol18: 166-173, 1972.
150. Reynolds RC, Hardman HF: The effect of pH changes and ionization on the action of
epinephrine upon the isolated rabbit ileum. Eur J Pharmacol1O: 249-255, 1972.
151. Miller FN, Wiegman DL, Devaney MJ, Harris PD: Effect of CO 2 on the a-adrenergic receptor:
a control mechanism for tissue blood flow. Microcirculation 1: 259-261, 1976.
152. Smiley M, Lawson M, Garlick D: Countercurrent exchange of highly diffusible tracers in
skeletal muscle: its absence in the lung. Clinical Expt Pharmacol Physiol4: 183--196, 1977.
153. Landis EM, Pappenheimer JR: Exchange of substances through the capillary walls. In: WF
Hamilton, P Dows (eds). Handbook of Physiology: Section 2: Circulation, Vol II. Am Physiol
Soc, Washington, DC, 1963, pp 1018--1019.
154. Wei EP, Thames MD, Kontos HA, Patterns JL Jr: Inhibition of the vasodilator effect of
hypercapnic acidosis by hypercalcemia in dogs and rats. Circ Res 35: 890--895,1974.
155. Morff RJ, Harris PD, Wiegman DL, Miller FN: Muscle microcirculation: effects of tissue pH,
PC02 , and P0 2 during systemic hypoxia. Am J Physiol 240: H746-H754, 1981.
156. Liang CS, Lowenstein JM: Metabolic control of the circulation. Effects of acetate and pyruvate.
J Clin Invest 62: 1029-1038, 1978.
157. Christoffersen GRJ, Skibsted LH: Calcium ion activity in physiological salt solutions: Influence
of anions substituted for chloride. Comp Biochem Physiol 52A: 317-322, 1975.
158. Frohlich ED: Vascular effects ofthe Krebs intermediate metabolites. Am J Physiol208: 149-153,
1965.
159. Hester RK, Weiss GB, Willerson IT: Basis of pH-independent inhibitory effects of lactate in
45Ca movements and responses to KC1 and PGF2 in canine coronary arteries. Circ Res 46: 771-
779, 1980.
160. Federspil G, Zaccaria M, Pedrazzoli S, Zago E, De Palo C, Scandellari C: Effects of sodium
dl-lactate on insulin secretion in anesthetized dogs. Diabetes 29: 33--36, 1980.
161. Peters RA: Mechanism of the toxicity of the active constituent of Dichapetalum cymosum and
related compounds. Adv Enzymol Relat Subj Biochem 18: 113--159, 1957.
162. Frohlich ED: Local effect of adenosine mono-, di- and triphosphate on vessel resistance. Am J
Physiol 204: 28--30, 1963.
163. Haddy FJ, Chou CC, Scott JB, Dabney JM: Intestinal vascular responses to naturally occurring
vasoactive substances. Gastroenterology 52: 444--451, 1967.
164. Granger HJ, Norris CP: Role of adenosine in local control of intestinal circulation in the dog.
Circ Res 46: 764-770, 1980.
165. Burnstock G: Purinergic nerves. Pharmacol Rev 24: 509-581, 1972.
166. Maher JF, Hirszel P, Lasrich M: An experimental model for study of pharmacologic and
 93

hormonal influences of peritoneal dialysis. In: A Trevino-Becerra, FST Boen (eds). Today's Art
of Peritoneal Dialysis. S Krager, Basel, 1979, Contrib Nephrol17: 131-138, 1979.
167. Honig CR, Frierson JL: Role of adenosine in exercise vasodilation in dog gracilis membrane.
Am J Physiol238: H703-H715, 1980.
168. Tabaie HMA, Scott JB, Haddy FJ: Reduction of exercise dilation by theophylline. Proc Soc
Exptl Bioi Med 154: 93-97, 1977.
169. Stafford A: Potentiation of adenosine and the adenine nucleotides by dipyridamole. Br J
Pharmacol Chemother 28: 218-227, 1966.
4. Peritoneal ultrastructure

Christian Verger

1. Introduction

Peritoneal dialysis was first attempted clinically by Ganter in 1923 [1]. Tenckhoff
[2], Boen [3] and Mion [4] made peritoneal dialysis possible in chronic renal
failure patients. U ntil1976 there was little concern about the fate of the peritoneal
membrane with chronic dialysis, since many patients treated with this method
were at high risk for a short life span. In 1976, with the appearance of continuous
ambulatory peritoneal dialysis [5] (CAPD), it became clear that many more
patients could be treated with this method and for longer periods; therefore, the
longevity of the peritoneal membrane became an important question. Changes in
the peritoneum exposed to long-term peritoneal dialysis will be explored herein.

2. Gross anatomy

The peritoneum is a smooth, translucent membrane lining the abdominal cavity

and reflecting onto the abdominal viscera.
For the purpose of peritoneal dialysis, the mesothelial surface and the microcir-
culation of the peritoneum both contribute to membrane resistance as discussed
in Chapter 2.
Since the original work of Wegner [6] cited by Boen [7] it is generally agreed
that the peritoneal membrane mesothelial surface area approximates skin surface
area. Visceral peritoneum predominates as the parietal peritoneum is estimated
to be only 10% of total peritoneal membrane area [8]. Functional studies suggest
that effective peritoneal membrane area is well below 1 m2 [9], probably because
only capillary 'effective pores' and nearby mesothelial 'pores' participate signifi-
cantly in exchange [10].
The visceral peritoneum is supplied by arteries coming from either the mesen-
teric or coeliac arteries [11]. After the blood has been delivered to the visceral
organs it is collected by the venous system to the portal vein. In contrast, the
96

parietal peritoneum is supplied by circumflex, iliac, lumbar, intercostal and

epigastric arteries [11].
Visceral lymphatics arise in the peritoneal membrane of omenta and mesentery
and join a network of visceral lymph nodes; these drain into the parietal lymph
nodes and the lymph is then returned to the venous circulation primarily via the
thoracic duct. For more details about the peritoneal microcirculation, we invite
the reader to refer to work by Miller [12] and Chapter 3. Whatever the importance
of the peritoneal microcirculation, it is worthy to note that, under some condi-
tions, only a fraction of capillaries may be perfused; closed capillary sphincters
[13] may prevent blood flow. It is possible that only 0.52% of the mesenteric
mesothelium would be adjacent to perfused capillaries [14]. Peritoneal structures
participating in exchange consist of a single layer of mesothelial cells resting on a
basement membrane, a subjacent loose connective tissue and blood and lympha-
tic vessels. These different structures will be described.

3. Technics of examination

Most investigators use standard techniques of light and electron microscopy [15,
16, 17]. I would like to emphasize the importance of careful handling of the
specimen when samples are taken from the peritoneum. The mesothelium is a
very fragile structure and a few seconds of drying or the touch of a glove or a
compress can induce artifacts. When a peritoneal biopsy is performed, either in
animals or humans, the surgeon must be aware of these problems and the sample
of the peritoneum must be kept wet by dropping it immediately into a 37° C
solution. The tissue must be fixed immediately in the appropriate fixative.
For electron microscopy, the choice of fixative concentrations and osmolality is
of great importance [18].
For cross-sections, either for light or electron microscopy, the preparation
must be oriented for cutting so that the mesothelium is first cut, then the tissue
underneath; not taking this precaution can induce detachment of the mesothelial
cells and lead to a false impression of the absence of mesothelium.

3.1. Normal peritoneum

3.1.1. Light microscopy

The surface of the mesothelium can be observed in animals using Hautchen
preparations as described by Poole [19] and modified by more recent authors [20,
21] (Fig. 1). In these preparations the mesothelium appears as a continuous
mosaic pattern of polygonal cells resting side by side without space between
them; contact lines are sinuous; each cell presents 5 to 7 sides.
The different structures on light microscopy are more readily observed in the
 97

Figure 1. Normal rat mesentery by light microscopy using an 'en face' silver stain. Magnification x72
Note 'pleated sides' .

mesentery; here the presence of adipose cells provides a clear separation between
the superficial and deeper structures. In the parietal peritoneum, the boundary
between interstitium and muscle fascia is less precise.
On cross-section (Fig. 2), the most superficial structure is the single and
continuous layer of flat mesothelial cells containing a dark nucleus. With tri-
chrome staining the tissue immediately beneath mesothelium is usually poorly
vascularized . The most superficial capillaries are mainly at the junction between
the adipose cells and the connective tissue which separates the adipose cells from
the mesothelium. We have previously measured the average distance between
capillaries and mesothelial surface in normal subjects [22]: it varies from 8 to 20
microns.

3.2. Ultrastructure of the peritoneum

3.2.1. Scanning electron microscopy (SEM)

On SEM the mesothelium usually appears as a continuous surface covered by
numerous microvilli (Figs. 3a and 3b); Odor [23] seems to have been the first to
confirm by transmission electron microscopy the presence of microvilli covering
the free surface of mesothelial cells. In 1974 [24] Andrews and Porter, using SEM,
described lengths, diameters and relative densities of microvilli on mesothelial
surfaces covering different internal organs. The microvilli vary considerably.
Lengths ranged from 0.04 to 2.5 mjL, diameters from 0.04 to 0.08 mjL and densi-
98

o.
'\
,. - -
.~ ...
•

_ J

Figure 2. Normal human mesentery (xlOO)-Trichromic staining. Connective tissue between meso-
thelium and adipocytes is extremely thin and not visible here .

ties from 200 per 10 mIL 2 on the bladder to 600 per 10 mIL2 on the membrane
covering the diaphragm and the heart. On the liver, intestine and stomach the
density is near 360 per 10 m1L2. However, some areas may appear completely
denuded of microvilli (Fig. 4). We frequently observed denuded areas in subjects
with end-stage renal failure before CAPD and during CAPD with or without
peritonitis. Although patchy denudation might result from fixation techniques,
Andrews and Porter feel they may be present in vivo [24]. In some areas microvilli
are coated with an amorphous material termed glycocalyx.
The functional significance of microvilli is not clear. Odor [23] proposed that
these structures may increase surface area and facilitate exchange between cells
and the body cavity. Another common hypothesis is that they act along with a
serous exudate to protect the surface ofthe mesothelium against friction [24]. The
observed higher density of microvilli on peritoneum covering very mobile organs
supports this hypothesis.
At higher magnification (Fig. 5) small invaginations are observed between
microvilli. These represent pinocytotic vesicles opening into the peritoneal cavity
and I will describe them in the section devoted to transmission electron micro-
scopy.
Baradi [25] and Motta [26] described in some areas on the mesothelium surface
a disjunction of mesothelial cells with round cells between them. They called
 99

Figure 3. SEM micrograph of the mesentery from a uremic non-dialyzed patient. Magnification:
3a = XISOO , 3b = xSOOO. No space is visible between mesothelial cells which are covered by numerous
microvilli.
100

Figure 4. SEM micrograph of the mesentery from a CAPO patient. Magnification x1S 000. Cell
surface is denuded of mirovilli; this aspect is also observed in normal subjects and might be due to
preparatory procedure (see text).

Figure 5. Normal mesothelial cells by SEM (rat mesentery) at magnification x30000 showing
cytoplasmic invaginations.
 101

them 'milky spots' and suggested they might be zones of mesothelial regenera-
tion. I personally seldom observed these 'milky spots' in normal rat or in normal
human mesothelium, but saw them frequently with the presence of peritonitis
(Fig. 11) as reported later in this paper.
The mesothelial surface on the muscular portion of the diaphragm has some
unique features. Tsilbary and colleagues [27] described mesothelial cells overly-
ing lymphatic lacunae (which are the blind endings of diaphragmatic lymphatics)
as different in form from the other mesothelial cells. Their boundaries were
prominent and the central region of each cell, containing the nucleus, protruded
towards the peritoneal cavity. However, we also observed identical findings in
human mesentery (unpublished personal data). Probably more specific to the
mesothelium over muscular portions of the diaphragm are the gaps they de-
scribed as follows: 'At some sites, the processes from adjacent cells diverged to
leave a round gap, 4-12 microns in diameter, between neighboring cells.' The
authors suggest that these stomata might represent the portal of a direct pathway
into the lumen of the lymphatic lacunae. For Hau et al. [28] these stomatae and
lymphatic lacunae may contribute to the removal of particulate material and
bacteria from the peritoneal cavity via the diaphragmatic lymphatics.

3.2.2. Transmission Electron Microscopy (TEM)

The mesothelium, interstitium, lymphatics and blood vessels will be desribed.
(a) The mesothelial layer. The mesothelium is composed of flattened cells lying
on a continuous and homogeneous basement membrane (Fig. 6). Microvilli,
previously described on SEM, appear as cytoplasmic prolongations. Andrews
and Porter observed small filaments (40 A, average diameter) within the matrix
of microvilli running parallel to the long axis. They also described, with SEM and
TEM in rats, other projections which have the characteristic structure of cilia with
a nine plus two microtubule arrangement. These cilia were not reported in the
recent work of Gotloib in the rabbit mesentery [14]. Neither were they observed
by me in rats or humans. The cilia described by Andrews were nearly three times
greater in diameter (approximately 0.2 microns) than the surrounding microvilli.
Between microvilli the surface of cells is randomly interrupted by openings of
pinocytotic vesicles (corresponding to cytoplasmic invaginations seen with
SEM). Vesicles are also frequently seen lining the opposed cell membranes of a
gap junction. The intracytoplasmic vesicle diameter (mean ± SD) has been
estimated by Gotloib and colleagues to be 71.7 +/- 46.4nm [14].
The nucleus, oval or reniform, is usually located in the central region. The
Golgi apparatus is well developed. Rough endoplasmic reticulum and ribosomes
are present and mitochondria can be observed in different sections of the cyto-
plasm.
Mesothelial intercellular junctions have been widely investigated [29-32].
Although general agreement is not established, the role of junctions might be of
importance in solute transport between the peritoneal cavity and capillaries [29,
102

Figure 6. TEM photomicrograph of the mesentery from a CAPD patient treated for 2 years: the
mesothelium is normal. A long and tortuous intercellular channel can be seen on the right side
(magnification x5600).

33 , 34, 35]. Probably many types of junctions are present in mesothelium . I

observed, in human tissue (Fig. 7) , the type of clefts described by Cotran and
Karnovsky. These appear tortuous, owing to overlappings and interdigitations of
contiguous mesothelial cells. The width of these clefts has been estimated to be
near 200 A with narrowings at one or more foci . In some places pinocytotic
vesicles open into the intercellular clefts as mentioned above. Large openings or
fenestrations are also observed [36]. Besides these gaps, a number of tight
junctions are observed comparable to those observed in vascular endothelium by
Karnovsky who suggested that these junctions consisted of maculae rather than
zonulae occludentes [37].
M. and N. Simionescu have studied the ultrastructure of junctions between cell
bodies or between the processes of one cell and the body of its neighbor, using a
freeze-fracture procedure and SEM [32]. They have distinguished tight junctions,
communicating junctions and macular junctions. Junctional complexes have
been described in human tissue examined by TEM by Dobbie [38]. He dis-
tinguished:
1. a fusion of adjoining cell membranes (the zonula occludentes or tight junc-
tion) ,
2. a deeper dense osmophilic zone beneath the plasmalemma , containing tono
filaments (the zonula adhaerans) , and
 103

J
Figure 7. TEM aspects of multiple intercellular junctions between several mesothelial cells. Same
patient as in Figure 6. Magnification : x10 000. Note mitochondria and the numerous vesicles opening
on both sides of cells.

3. the macula adhaerans or desmosome in the deepest part of the junctional

complex.
It is difficult to evaluate the impact of these structures on peritoneal membrane
permeability. For Forrest and co-workers [39] tight junctions define an anatomi-
cal barrier capable of restricting the passage of relatively small molecules such as
urea.
(b) The basement membrane of mesothelium. The basal mesothelial cell surface
is separated from the adjacent connective tissue by a membrane consisting of an
inner and outer dense line separated by a lighter area, the three structures
measuring about 500 A in width [23]. In some cases small vesicles or granules are
associated with this membrane. Often small prolongations of the cytoplasm of the
mesothelial cells dip deeply into the connective tissue. Beneath the basal lamina
is a layer of very fine connective tissue fibrils.
(c) The interstitium. The interstitium consists of a loose tissue made up of
bundles of collagenous fibrils separated by large structure less areas. Fibroblasts
can be observed. The interstitium also contains lymphatic and blood vessels.
The permeability of the interstitium is relatively high. Diffusion coefficients for
movement within interstitial space calculated for dextran are the same as in water
for solutes with a molecular weight below 150000 daltons [40] .
(d) Blood vessels. Capillaries in mesentery are sparse . The minimum distance
separating vessels from the surface in the human peritoneum is 5-20 microns [22].
104

The mesenteric microvessels have been subdivided into 5 types on the basis of the
flow pattern [41]. The segment between a diverging and converging dichotomy is
defined as midcapillary. Upstream are arterial capillaries and then arterioles.
Downstream are venous capillaries and venules. There are gradual structural
variations throughout the microvasculature with the presence of fenestrae (40-
50-nm in diameter) in venous capillaries and venules. Gotloib [14] distinguishes
true capillaries (without perithelial cells) and post capillary venules. Capillaries
and post capillary venules are made up of a single layer of endothelial cells lying
on a continuous basement membrane. Their cytoplasm contain mitochondria, a
Golgi apparatus, numerous ribosomes and rough endoplasmic reticulum. Many
vesicles are also present. The nucleus is oval. Pericytes are present on the other
face of the basal membrane of postcapillary venules. They are more numerous as
the luminal diameter of the venules increases. The capillaries of the mesentery
seem to be of the continuous type described by Majno [42] with cells sealed by
tight junctions, but with gaps between them [42-44]. These junctions do not seem
to be an important barrier to small solutes as suggested by the studies of Wayland
who demonstrated that endothelium offers very little resistance to small solute
movements [45]. On the other hand, capillaries probably play an important role
in the regulation of fluid movement between the peritoneal cavity and blood
stream, depending on the state of pre and post capillary sphincters [46].
(e) Lymphatic vessels. Gotloib and colleagues observed an important lympha-
tic network in the mesentery of the rabbit and they estimated that about 4.0% of
the mesothelial surface covered adjacent lymphatic vessels [14]. They did not find
perithelial cells associated with these lymphatics. The endothelial lining of lym-
phatics had the same structure as in blood capillaries, but tight junctions were not
seen. Furthermore, no continuous basal membrane was observed. These obser-
vations suggest that, at least in the rabbit mesentery, the lymphatics may be very
permeable and could play an important role in solute transport.

3.3. The peritoneum of uremic patients

3.3.1. Differences between uremic and non-uremic patients

Few data are available about the variations of peritoneal structure that could be
induced by uremia. In the work of Dobbie [38] comparing normal rodents,
normal man and uremic patients the ultrastructural findings were similar. I have
the same experience; however, most of my patients who had a laparotomy during
the first placement of the peritoneal catheter presented some degree of ascites. In
these cases, the interstitium separating the mesothelial layer from the adipose
cells looked thicker than usual and measured about 30 microns (Fig. 8). Further-
more, on light microscopy the mesothelial cells often appeared more turgescent
with a round instead of an elongated nucleus.
 105

Figure 8. Normal mesentery from a non-dialyzed uremic patient. Light microscopy with trichromic
stain-magnification xlOO. Note the thick submesothelial tissue. Red cells resting on the mesothelium
are due to a slight bleeding during the sampling.

3.3.2. The peritoneum during CAPD

After several months on CAPD, an increased distance between the mesothelial
surface and the capillaries can be seen. This distance is about 20 to 40 microns.
The change is due to both interstitial edema and interwoven bundles of collagen
fibers (Fig. 9). On SEM, the superficial structure of the mesothelium is usually
well preserved , even in patients who have a history of several previous episodes of
peritonitis . This normal appearance has also been shown by Sorkin and co-work-
ers in a diabetic patient after 3 years on CAPD with many episodes of peritonitis
[41] . In some circumstances, however, we observed the same alterations de-
scribed by Dobbie [38]; groups of individual cells showing intracellular edema
with disruption and degeneration of cytoplasmic structures. In these cases, large
gaps were present between the mesothelial cells on SEM .
In a few patients, who had severe or prolonged peritonitis we observed both a
total disappearance of the mesothelial cells and a fibrous reaction of the sub-
mesothelial connective tissue. These alterations have been shown to be associ-
ated with increased permeability and decreased ultrafiltration on CAPD .
During acute peritonitis in rats [48], besides the inflammatory response and
hypervascularization (Fig. 10), the first change in mesothelium seems to be a
rupture of intercellular junctions. Soon round cells (probably macrophages)
move from underneath the mesothelium toward the peritoneal cavity (Fig. 11) .
106

Figure 9. Mesentery of CAPD patient after 2 years. The submesothelial fibrosis is identical to what is
observed in uremic patients with slight ascitis . Without SEM it is not possible here to conclude if the
patchy disappearance of the mesothelium is real or due to the preparatory procedure . Magnficiation
x80.

Figure 10. H&E stained from a rat with peritonitis . Note hypervascularization and band of fibrosis
beneath the mesothelium. Magnification xlOO.
 107

Figure 11. SEM of infected mesentery. Round cells appear to be moving through the intercellular
spaces . Magnification x1500.

The mesothelial cell separation may allow rapid access of macrophages and other
leukocytes to the peritoneal surface. After severe injury, the entire mesothelium
can degenerate and desquamate. The peritoneum becomes very permeable;
rapid glucose absorption causes early dissipation of osmotic pressure and loss of
net ultrafiltration after a long dwell exchange.
Gentle peritoneal drying induces only mesothelial desquamation without evi-
dent alteration of underlying structures [48]. However, permeability changes as
during peritonitis are demonstrable [48] suggesting that mesothelium might
influence peritoneal permeability; transport changes associated with intraperi-
toneal protamine which causes mesothelial changes support this hypothesis [49].
Protamine induces separation of mesothelial cells associated with an increased
peritoneal clearance for urea and inulin. These alterations were reversible at
lower doses of protamine.
In rats, if the animal is sacrificed one or two days after peritoneal drying, two
populations of cells are observed on denuded surfaces. These are round cells
which are probably macrophages and flattened cells (Fig. 12) which are new
mesothelial cells adhering to the denuded peritoneum. Complete regeneration of
the mesothelium is observed in non-dialyzed animals 7 days following drying. In
humans on CAPD it is not known whether the presence of peritoneal dialysate
would delay or promote mesothelial regeneration. Rapid recovery of peritoneal
108

Figure 12. SEM of mesentery 2 days after total mesothelial desquamation by peritoneal drying .
Numerous cells, round and flattened now adhere to the denuded surface . Magnification x600,

transport properties with CAPD following clinical peritonitis , often in 7 to 10

days, suggests healing is not prolonged to any great extent.
The regeneration of mesothelium raises questions as to its origin, which is not
clear. Many authors have studied regeneration of the mesothelial cells [19, 20, 21,
50-52]. Cameron [53], Johnson and Whitting [54] and Bridges and Whitting [55]
believe that mature mesothelial cells become detached and float freely in per-
itoneal seroUS fluid before adhering to a denuded area and proliferating in situ to
form a new mesothelial layer . However, recent observations that free mesothelial
cells in the peritoneal fluid are usually altered introduces some doubts about their
ability to participate in mesothelial reconstruction. Eskeland [57], Eskeland and
Kjaerheim [57] and Ryan [21] believed that peritoneal macrophages are ubiq-
uitous and in some instances transform into mesothelial cells following grafting
onto the denuded surface. This last hypothesis has not been confirmed by the
studies of Raftery using labelled peritoneal macrophages; he suggested that
mesothelial cells were coming from the underlying tissue. According to this
hypothesis, the original mesothelial cells could result from the transformation of
either primitive mesenchymal cells or fibroblasts [20, 58].
Whatever the origin of mesothelium and the process of its repair, the above
works suggest that the loss of ultrafiltration in some patients on CAPD [59, 60,
22] may be related to reversible alterations of the mesothelium. The reap-
 109

pearance of normal peritoneal permeability after temporary interruption of

CAPD in some patients (personal unpublished data) supports this hypothesis.
Unfortunately, besides these probably benign and reversible alterations, more
severe and irreversible damages have been reported either on intermittent per-
itoneal dialysis or on CAPD. Denis and colleagues described massive adhesions
between intestinal loops bound together in a dense, opaque casing [61]. The same
kind of observation was published by Gandhi et al. [62] and recently reviewed
[63]. These observations are probably the equivalent of the so-called 'encapsulat-
ing and sclerosing peritonitis' published by Slingeneyer et al. [64] and Rottem-
bourg et al. [65]. In their cases, the peritoneal membrane appeared thicker than
usual, and was sclerotic with complete disappearance of the mesothelium. The
causes of these extremely severe alterations are not completely elucidated.
However, recent French studies suggest that acetate, sometimes used as buffer in
the dialysate, could be involved in the development of these complications.
However, other factors could be responsible as such lesions have also been
described in patients who had always been dialysed with solutions buffered with
lactate [66]. It is therefore possible that medications such as beta blocking agents
[67], the release of particulate material [68] from the bags containing the dialysis
solutions, the prescription of intraperitoneal medications and, of course, the rate
and the duration of peritonitis could be involved to different degrees.
The decreased exchange surface and the very sclerotic peritoneum explain that
the loss of ultrafiltration observed in these cases is associated with low clearances
and decreased permeability. Therefore, a decreased filtration with hypoper-
me ability must always lead the physician to consider irreversible sclerosing
peritonitis.
The fact that sclerosing peritonitis and the loss of ultrafiltration by hyperper-
meability have been frequent in the same centers suggests the two phenomena
could be linked. The protective role of the mesothelium for the subjacent
structures and its role in the prevention of adhesions [69] suggest the following
hypothetical sequence:
Initially there could be rupture of mesothelial junctions and progressive de-
squamation of the mesothelial cells. At this stage the peritoneal membrane would
be very permeable and any loss of ultrafiltration due to rapid glucose absorption.
These changes are reversible.
If peritoneal dialysis is not stopped, the mesothelium is inadequately regener-
ated and the underlying tissue is continuously exposed to chemical and mechan-
ical aggression. The result is a progressive development of sclerosis and/or
adhesions which result in decreased pore area, decreased clearances and loss of
filtration by hypopermeability. These lesions are irreversible.
Early diagnosis of mesothelial alterations on the basis of a careful examination
of peritoneal equilibration curves might help prevent definitive damage. How-
ever, the fact t~at some sclerosing peritonitis seems to develop spontaneously
even several months or years after the interruption of CAPD [64] raises the
110

hypothesis that, once started, the sclerosing process could be irreversible. Immu-
nological modifications could be responsible although this remains to be proven.
Fortunately, this last complication is probably rare and it has never been encoun-
tered in many centers.

4. Summary

Unlike the artificial kidney, CAPD is based on the use of a living membrane. This
means that its properties may vary with individuals and circumstances. Besides
the local variations of the peritoneal microvasculature, peritoneal permeability
characteristics are dependent on the mesothelium, the interstitium and probably
lymphatic vessels. The junctions between mesothelial cells are usually of the tight
type. Mesothelium probably is involved in membrane resistance to solute trans-
port and in the protection of the underlying tissue. Fibrotic reactions are fre-
quently encountered with severe alterations or total disappearance of the meso-
thelial layer. Even with an intact mesothelium, a mild thickening of the inter-
stitium may be observed in uremic and/or CAPD patients. Lymphatic vessels are
well developed, particularly over the muscular portion of the diaphragm where
they seem to play a role in the absorption of particles and in defenses against
peritoneal infection. Capillaries have the usual structure and do not seem to be
altered by uremia or by CAPD; their size and number vary considerably during
peritonitis. Other investigations are needed to understand and prevent peritoneal
membrane alterations, particularly in CAPD patients. Peritoneal biopsies when
possible may be helpful in this regard.

Acknowledgements

We are obliged to Professor Karl D. Nolph for his assistance, Dr Velu (Lab
d'anatomo-pathologie, Pontoise, France) for his advice in light microscopy and to
Professor P. Galle (Lab Biophysique, Creteil, France) who gave us access to his
laboratory; we warmly thank technicians Mrs M. Orrico and Mr P. Siry for their
extremely efficient help.

References

1. Ganter G: Uber die Beseitigung giftiger Stoffe aus dem Blute durch Dialyse. Muench Med
Wochenschr 70: 1478-1480, 1923.
2. Tenckhoff H, Meston B, Shilipetar G: A simplified automatic peritoneal dialysis system. Trans
Am Soc Artif Intern Organs 18: 436, 1972.
3. Boen ST: Overview and history of peritoneal dialysis. Dial Transpl6: 12, 1977.
4. Mion C, Slingeneyer A: La dialyse peritoneale adomicile, traitement au long cours de I'uremie au
 111

stade ultime. Bilan de 22 mois d'experience. In: Seminaire d'Uro-Nephrologie, 3e serie. R Kuss,
M Legrain (eds), Masson, Paris, p 119.
5. Popovich RP, Moncrief JW, Decherd JB et al.: The definition of a novel portable/wearable
equilibrium peritoneal dialysis technique abstract. Abstr Am Soc ArtifIntem Organs 5: 64,1976.
6. Wegner G: Chirurgische Bemerkungen iiber die Peritonealh6le, mit besonderer Benick-
sichtigung der Ovariotomie. Arch F Chir 20: 51-55, 1877.
7. Boen ST: Peritoneal Dialysis in Clinical Medicine. Charles Thomas, Springfield, Illinois, 1964.
8. Knapowski J, Feder E, Simon M, Zabel M: Evaluation of participation of parietal peritoneum in
dialysis: physiological, morphological and pharmacological data. Kidney Int 16: 227 (Abstract),
1979.
9. Henderson LW: The problem of peritoneal membrane area and permeability. Kidney Int 3: 409-
410,1973.
10. Henderson LW, Nolph KD: Altered permeability of the peritoneal membrane after using
hypertonic peritoneal dialysis fluid. J Clin Invest 48: 992-1001,1969.
11. Crafts RC: A textbook of Human Anatomy. The Ronald Press Company, New York, 1966.
12. Miller FN: The peritoneal microcirculation. In: KD Nolph (ed). Peritoneal Dialysis. Martinus
Nijhoff Publishers, Boston, 1981, pp 42-78.
13. Nolph KD: The peritoneal dialysis system. Contr Nephrol17: 44-50, 1979.
14. Gotloib L, Digenis GE, Rabinovich S, Medline A, Oreopoulos DM: Ultrastructure of Normal
Rabbit Mesentery. Nephron 34: 248--255, 1983.
15. Toner PG, Carr KE: The digestive system. In Biomedical research applications of scanning
electron microscopy Vol. 1, Ed GM Hodges, RC Hallowes. Academic Press, London, New York,
San Francisco, 1979, pp 203-209.
16. Arro E, Collins VP, Brunk UT: High resolution SEM of cultured cells: preparatory procedures.
Scanning Electron Microsc 11: 159-168,1981.
17. Whitaker D, Papadimitriou JM, Walters MN-I: The mesothelium; techniques for investigating
the origin, nature and behaviour of mesothelial cells. J Pathol132: 263-271, 1980.
18. Pentilla A, McDowell EM, Trump BF: Effects of fixation and postfixation treatments on volume
of injured cells. J Histochem Cytochem 23: 251-270, 1975.
19. Poole JCF, Sanders AG, Florey HW: The regeneration of aortic endothelium. J Pathol Bacterial
75: 133-142, 1958.
20. Raftery AT: Regeneration of parietal and visceral peritoneum, a light microscopial study. Br J
Surg 60(4): 292-299, 1973.
21. Ryan GB, Grobety J, Majno G: Mesothelial Injury and Recovery. Am J Pathol71: 93-102,1973.
22. Verger C, Brunschvicg 0, Le Charpentier Y, Lavergne A, Vantelon J: Structural and ultrastruc-
tural peritoneal membrane changes and permeability alterations during continuous ambulatory
peritoneal dialysis. Proc Eur Dial Transpl Assoc 18: 199-203, 1981.
23. Odor L: Observations of the rat mesothelium with the electron and phase microscopes. Am J
Anat 95: 433-465, 1954.
24. Andrews PM, Porter KR: The ultrastructural morphology and possible functional significance of
mesothelial microvilli. Anat Rec 177: 409-426, 1973.
25. Baradi AF, Rao SN: A scanning electron microscope study of mouse peritoneal mesothelium.
Tissue Cell 8: 159-162, 1976.
26. Motta P, Andrews PM, Porter KR: Microanatomy of cells and tissue surfaces. In: An Atlas of
Scanning Electron Microscopy. Lea and Febiger, Philadelphia, 1977, pp 76-79.
27. Tsilbary EC, Wissig SL: Absorption from the peritoneal cavity sem study of the mesothelium
covering the peritoneal surface of the muscular portion of the diaphragm. Am J Anat 149: 127-133,
1977.
28. Hau T, Ahrenholz David H, Simmons Richard L: Secondary bacterial peritonitis: the biologic
basis of treatment. Curr Probl Surg 16: 5-17, 1979.
29. Cotran RS, Kamovsky MJ: Ultrastructural studies on the permeability of the mesothelium to
horseradish peroxidase. J Cell Bioi 37: 123-137, 1968.
112

30. Fedorko ME, Hirsch JG: Studies on transport of macromolecules and small particles across
mesothelial cells ofthe mouse omentum. I Morphological aspects. Exp Cell Res 69: 113-127,1971.
31. Hama K: The fine structure of the desmosomes in frog mesothelium. J Biophys Biochem Cytol7:
575-579, 1960.
32. Simionescu M, Simionescu N: Organization of cell junctions in the peritoneal mesothelium. Cell
Bioi 74: 98--110, 1977.
33. Leak LV, Just EE: Permeability of peritoneal mesothelium: a TEM & SEM study. J Cell Bioi 70:
423 a (Abstract), 1976.
34. Nagel W, Kuschinsky W: Permeability of Isolated Mesentery. Europ J Clin Invest 1: 149-154,
1970.
35. Gosselin RE, Berndt WO: Diffusional transport of solutes through mesentery and peritoneum. J
Theor Bioi 3: 487-495, 1962.
36. Cotran RS, Majno G: Studies on inter cellular junctions of mesothelium and endothelium.
Protoplasma 63: 45-51, 1967.
37. Karnovsky MJ: Ultrastructural basis of capillary permeability studied with peroxidase as a tracer.
J Cell Bioi 35: 213, 1967.
38. Dobbie JW, Zaki M, Wilson L: Ultrastructural studies on the peritoneum with special reference
to chronic ambulatory peritoneal dialysis. Scott Med J 26: 213-223, 1981.
39. Forrest IN, Boyer JL, Ardito TA, Murdaugh HV, Wade JB: Structure of tight junctions during
Cl secretion in the perfused rectal gland of the dogfish shark. Am J Physiol242: c388--c392, 1982.
40. Nakamura Y, Wayland H: Macromolecular Transport in the Cat Mesentery. Microvasc Res 9:
1-21,1975.
41. Bundgaard M, Fr0kjaer-Jensen J: Functional Aspects of the ultrastructure of Terminal Blood
Vessels: A Quantitative Study on Consecutive Segments of the Frog Mesenteric Microvascula-
ture. Microvasc Res 23: 1-30,1982.
42. Majno G: Ultrastructure of the vascular membrane. In Hamilton, Down, Handbook of Physiol-
ogy, sect 2, iv: 2295 (Am Physiol Soc Washington).
43. Simionescu N, Simionescu M, Palade GE: Structural basis of permeability in sequential segments
of the microvascular of the diaphragm. II Pathways followed by microperoxidase across the
endothelium. Microvasc Res 15: 17-36, 1978.
44. Karnovsky MJ: The ultrastructural basis of transcapillary exchanges. J Gen Physiol52: S64-S95,
1968.
45. Wayland H: Action of histamine on the microvasculature. In: Proc 1st CAPD Int Symp. Excerpta
Medica, Amsterdam, 1980, pp 18--27.
46. Nolph KD, Miller FN, Pyle WK, Popovich RP, Sorkin MI: An hypothesis to explain the
ultrafiltration characteristics of peritoneal dialysis. Kidney Int 20: 543-548, 1981.
47. Sorkin MI, Luger AM, Prowant B, Kennedy J, Moore H, Nolph KD: Histological and functional
characteristics of the peritoneal membrane of a diabetic patient after 34 months of CAPD. Perit
Dial Bull 2: 24-27, 1982.
48. Verger C, Luger A, Moore HL, Nolph KD: Acute changes in peritoneal morphology and
transport properties with infectious peritonitis and mechanical injury. Kidney Int 23: 823-831,
1983.
49. Alavi N, Lianos E, Andres G, Bentzel CJ: Effect of protamine on the permeability and structure
of rat peritoneum. Kidney Int 21: 44-53, 1982.
50. Watters WB, Buck RC: Scanning Electron Microscopy of Mesothelial Regeneration in the Rat.
Lab Invest 26: 604-609, 1972.
51. Raftery AT: Regeneration of parietal and visceral peritoneum. An electron microscopical study.
J Anat 115: 375-392, 1973.
52. Reidy MA, Schwartz SM: Endothelial Regeneration. Lab Invest 44: 301-308, 1981.
53. Cameron G R, Hassan SM, de SN: Repair of Glisson's capsule after tangental wounds of the liver.
J Pathol Bacterial 73: 1-10, 1957.
 113

54. Johnson FR, Whitting HW: Repair of parietal peritoneum. Br J Surg 49: 653-660, 1962.
55. Bridges JB, Whitting HW: Parietal peritoneum healing in the rat. J Pathol Bacterial 87: 123-130,
1964.
56. Eskeland G: Regeneration of parietal peritoneum in rats. I A light microscopical study. Acta
Pathol 68: 353-378, 1966.
57. Eskeland G, Kjaerheim A: Regeneration of parietal peritoneum in rats. II An electron micro-
scopical study. Acta Pathol 68: 379-395, 1966.
58. Haar JL, Ackerman GA: A phase and electron microscopic study of vasculogenesis and
erythropoiesis in the yolk sac of the mouse. Anat Rec 170: 199-224, 1971.
59. Faller B, Marichal JF: Loss of ultrafiltration in continuous ambulatory peritoneal dialysis: Clinical
data. In: Proc 11th Int Symp Peritoneal Dialysis. Excerpta Medica Berlin, 1981, pp 227-232.
60. Slingeneyer A, Canaud B, Mion C: Permanent loss of ultrafiltration capacity ofthe peritoneum in
long-term peritoneal dialysis: an epidemiological study. Nephron 33: 133-138, 1983.
61. Denis J, Paineau J, Potel G, Fontenaille J, Guenel J: Continuous Ambulatory Peritoneal
Dialysis. Ann Intern Med 93: 508, 1980.
62. Ghandi VC, Humayun HM, Ing TS, Daugirdas JT, Jablokow VR, Iwatsuki S, Geis WP, Hano
JE: Sclerotic Thickening of the Peritoneal Membrane in Maintenance Peritoneal Dialysis Pa-
tients. Arch Intern Med 140: 1201-1203, 1980.
63. Gandhi VC, Ing TS, Daugirdas JT et al.: Failure of peritoneal dialysis due to peritoneal sclerosis.
Int J Artif Organs 6: 97, 1983.
64. Slingeneyer A, Faller B, Beraud JT: Progressive Sclerosing Peritonitis: late and severe complica-
tion of maintenance peritoneal dialysis. Trans Am Soc Artif Intern Organs 29: 633-638,1983.
65. Rottembourg J, Gahl GM, Poignet JL, Mertani E, Strippoli P, Langlois P, Crambaloc P, Legrain
M: Severe abdominal complication in patients undergoing Continuous Ambulatory Peritoneal
Dialysis. Proc EDTA 20: 236-241, 1983.
66. Bradley JA, Hamilton DNH, McWhinnie DL, Briggs JD, Junor BJR: Sclerosing Peritonitis after
CAPD. Lancet 11: 572-573,1983.
67. Ahmad S: Sclerosing peritonitis and propranolol. Chest 79: 361-362, 1981.
68. Verger C, Berry JP, Galle P, Lavergne A, Hoang C, Lecharpentier Y: Foreign material inclusions
in the peritoneum of CAPD patients: a study with X Ray microanalysis. Perit Dial Bu112: 138-139,
1982.
69. Ryan GR, Grobety J, Majno G: Postoperative peritoneal adhesions. Am J Pathol65: 117-140,
1971.
5. Transport kinetics

ROBERT P. POPOVICH and JACK W. MONCRIEF

1. Introduction

Comprehensive mathematical models of the patient-peritoneal dialysis system

are fundamental to the analysis and understanding of metabolite and fluid
transport in peritoneal dialysis. Theoretical models serve to: (1) illustrate the
system parameters which are most significant, (2) define how these parameters
relate to each other, (3) predict the behavior of the system, allowing manipula-
tion of the variables to produce optimized clinical results, (4) aid in the design of
clinical protocols to measure the parameters, and (5) suggest areas requiring
additional investigation. In short, a great deal of information can be acquired
from modeling of the peritoneal dialysis system in general with multiple appli-
cations to the diagnosis and treatment of individual patients.

2. Historical perspective

Ganter [1] is credited as being the first to use peritoneal dialysis in the treatment of
uremia in 1923. Following experimentation with animals, he dialyzed a patient
who demonstrated some clinical improvement. Other investigators, reviewed by
Cunningham [2], studied the absorption of a wide range of particulate and soluble
substances from the peritoneal cavity. As early as 1921, Clark [3] demonstrated
that intraperitoneal fluid tended to equilibrate both chemically and osmotically
with blood. This was later confirmed by Schechter et al. [4] who further demon-
strated that hypertonic glucose solutions infused into the peritoneal cavity in-
creased in volume before being absorbed. The effectiveness of peritoneal lavage
in resolving uremic symptoms in nephrectomized dogs was demonstrated by Bliss
[5] in 1931. In 1950, Odel et al. [6] published an extensive literature review in
which they noted over 100 reported cases of peritoneal dialysis. They concluded
that peritoneal dialysis had earned a definite place in the treatment of acute renal
failure.
116

Grollman et al. [7] resolved many of the technical problems and complications
of chronic peritoneal dialysis. The development of the Tenckhoff catheter [8]
coupled to subsequent refinements in techniques and the availability of commer-
cial dialysis supplies and solutions has made peritoneal dialysis a simple, safe, and
commonly employed technique in the treatment of renal failure [9-17].

3. Physiological principles

The peritoneum is the largest serous membrane of the body. It lines the inside of
the abdominal wall (the parietal peritoneum) and is reflected over the viscera (the
visceral peritoneum). The space between the parietal and visceral portions of the
membrane is called the peritoneal cavity. This is normally a potential space
lubricated by serous fluid secreted by mesothelial cells which cover its free surface
[18-21].
The detailed microscopic anatomy of the peritoneum has been described by
Baron [21]. It consists of five layers of fibrous and elastic connective tissue
covered by mesothelial cells. Blood and lymphatic capillaries are located in the
deepest layers in adults. Thus, for a substance to pass from the bloodstream into
the peritoneal cavity, it must pass the capillary endothelium, the interstitium, the
mesothelium and any fluid film resistances. This is illustrated schematically in
Figure 2 of Chapter 2. Six resistances to the diffusion of metabolites from blood
are shown. R 1 represents the mass transfer resistance of the blood fluid film where
the diffusion distance in the capillary is illustrated between 1 and 3 f.J., (the capillary
radius), R2 and R3 represent the resistances to diffusing past the endothelial
intercellular channel and the basement membrane, respectively. R4 represents
the resistance of the interstitial fluid. The interstitium may be a dense matrix of
collagenous fibers and mucopolysaccharide gels with channels or pores through
which the metabolites must diffuse. The nature of the pores may depend on the
degree of hydration [22] of th~ matrix. Note that the interstitial diffusion path can
approach 100 f.J., which would represent an appreciable resistance to diffusion. Rs
represents the resistance of the mesothelial cells and R6 is the dialysate fluid film
resistance. The total resistance is the sum of all the individual resistances.
The precise nature of absorption from the peritoneal cavity varies in relation to
the physical and chemical properties of the substance in question. Neutral, non-
colloidal substances such as urea will be transported primarily by passive diffu-
sion. On the other hand, it is postulated that plasma proteins, red blood cells, and
other particulate matter may be absorbed through the lymphatics because of the
greater permeability of the lymph capillaries to large molecules or particles [2,
23-25]. Allen and Weatherford [23] have demonstrated that spherical particles up
to 24 f.J., are absorbed through the peritoneum in the cat and rat. From this they
have postulated that there is a porous basement membrane through which
particles pass into the lymphatics.
 117

The movement of fluid and solutes across the peritoneum in response to

osmotic agents is well-documented [3,26-31]. Henderson [28] demonstrated the
transfer of inulin and urea across the peritoneal membrane in the absence of a
concentration gradient. This led Nolph [26] to conclude that solute transport
across the peritoneum occurs to some extent by convection or ultrafiltration.
The area of the peritoneal membrane available for mass transfer is uncertain.
The total peritoneal area was directly measured and found to be 20718 cm2 with
an area-to-body weight ratio of 284 cm2/kg in an adult male [32, 33]. The authors
calculated an area-to-body weight ratio of 522 cm2/kg in infants [33]. Esperanca
and Collins [33] also obtained a ratio of approximately 2 for infant peritoneal area
to adult peritoneal area per kilogram of body weight. Kallen [32] has suggested a
scaling approximation of:

S = lOOOWD7, (1)

where S is the peritoneal surface area in square centimeters and W is the boqy
weight in kilograms.
These measurements may provide an estimate of the total anatomical per-
itoneal surface area. However, the precise pore area of the adult peritoneum is
unknown. Gosselin and Berndt [34] computed a 0.6% open pore area for the
mesentery and a 0.2% open pore area for the intestinal peritoneum in rabbits.
The average thickness of the rabbit visceral peritoneum was 38.1 ± 1.7 f.L.
The area of the peritoneum available for transfer may also depend on the
volume of dialysate infused. The precise peritoneal surface area-to-dialysate
volume relationship is also unknown. The data of Miller et al. [15] indicate that
both a linear volume/area relationship and a constant area relationship are
unsatisfactory to explain observed equilibration data. Goldschmidt et al. [35]
have demonstrated that the relationship does not conform to that expected for a
sphere unless the dialysate volume is large (approximately 31).

4. Clinical protocols

Modeling of the patient-peritoneal dialysis system will to some extent be depen-

dent upon the details of the clinical protocol employed. Miller et al. [15] have
classified six intermittent methods:

Technique I - standard peritoneal dialysis (PD)

Technique II - recirculation PD
Technique III - continuous flow PD
Technique IV - continuous flow recirculation PD
Technique V - rapid intermittent PD
Technique VI - continuous flow compound dialysis
118

Several recent additions since the publication of Miller's paper are reciprocating
PD and Continuous Ambulatory Peritoneal Dialysis (CAPD) and Continuous
Cyclic Peritoneal Dialysis (CCPD).
Technique I, standard intermittent peritoneal dialysis was the usual clinical
procedure prior to the introduction of CAPD. In Technique I, two liters of
dialysate solution at 37° C are rapidly infused into the peritoneal cavity and
allowed to equilibrate for approximately 30 min. The dialysate fluid is then
drained by gravity flow and two liters of fresh dialysate are infused. The ex-
changes are repeated until the desired clinical result is obtained. The bulk of the
theoretical models in the literature are aimed at understanding and optimizing
this standard clinical procedure.
Intermittent recirculation PD (Technique II) is similar to Technique I except
that the dialysate is recirculated through the abdomen during the equilibration
phase. It was thought that this would greatly improve solute transport by prevent-
ing stagnation of the dialysate. Unfortunately, Miller et al. [15] demonstrated a
slight decrease in clearance using Technique II relative to Technique I controls.
Continuous flow PD (Technique III) employs two widely spaced catheters with
continuous infusion and drainage of dialysate. It has been used successfully by
early clinicians [6, 17,36,37] but again, Miller et at. [15] have demonstrated that
no significant improvements in clearance are obtained. Indeed, the addition of a
recirculation loop to achieve higher flow rates through the peritoneal cavity
(Technique IV - continuous recirculation PD) by Miller and colleagues [15] did
not yield results up to expectations. They suggested that this may have resulted
from channeling of dialysate at the higher flow rates. As will be demonstrated,
this may instead be a result of the peritoneal transport being limited by membrane
resistances (Rl to Rs)'
Rapid intermittent peritoneal dialysis (Technique V) has also been applied by
many investigators [15, 17, 38-40]. Drainage is started as soon as an infusion is
completed in this technique, in an attempt to achieve maximum dialysate flow
rates (4 to 51/h).
Continuous compound dialysis (Technique VI) is a continuous flow procedure
in which the peritoneal dialysis fluid is recirculated through a hemodialyzer.
Shinaberger et al. [41, 42] reported reduced protein losses using this method.
Rosenbalm and Mandanas [43] used this technique and recirculated the dialysate
through ion exchange columns to obtain enhanced phenobarbital removal.
A slight modification of this basic procedure termed reciprocating peritoneal
dialysis [7,18-21,23-24] has also been attempted. In this case, a preset volume of
dialysate is removed from the peritoneal cavity and pumped through a hollow
fiber artificial kidney. This is schematically illustrated in Figure 1. Recent ad-
vances involve the addition of a dialysate regeneration circuit to the basic re-
ciprocating system [44].
A novel approach to continuous peritoneal dialysis (developed by Popovich
and Moncrief) has recently been described [45-47] and termed Continuous
 119

:PD N,5

,
Gl ,

I 201
i
I,N5UL NEEDLE
TANK SKIN
PERITONE~
~- -I
CATHETER Iz DRAIN !
[Link]
..-
BAG
.i
I
I~- ._ .J

Figure 1, Schematic of reciprocating peritoneal dialysis (reprinted with permission of Dialysis &
Transplantation (Ref. 44»_

Ambulatory Peritoneal Dialysis. This technique and the analogous CCPD pro-
cedure employs the continuous presence (24 hours a day, 7 days a week) of
peritoneal dialysis solution in the peritoneal cavity. Dialysate fluid is drained and
immediately replaced 3 to 5 times per day. Additional procedural details of these
methods of dialysis are available in Chapters 7-9.

5. Theoretical analysis of the patient-peritoneal dialysis system

A complete exposition of the kinetic models associated with each of the various
peritoneal dialysis techniques is beyond the scope of this publication. Therefore,
only those models associated with techniques currently utilized in general clinical
120

practice will be examined. These include the various forms of standard intermit-
tent dialysis, CAPD and CCPD. For models of the less frequently employed
procedures, the literature should be consulted. For example, Villarroel [48] has
presented an excellent characterization of continuous models of peritoneal di-
alysis. Kablitz, et al. [44] have described a model of reciprocating peritoneal
dialysis. Naturally, the general principles of peritoneal transport apply to all the
techniques.

5.1. Diffuse mass transfer models

In 1966, Kallen [32] suggested the use of a simple exponential model for approx-
imating the decrease in blood urea concentration as a function of total dialysis
time. No provisions were included for the effects of infusion and drainage, or for
fluid transfer. This model tended to over-predict the urea removal in clinical
trials. The basic parameter of his model was the mean urea clearance which, as
will be subsequently demonstrated, is highly dependent upon the dialysis condi-
tions.
Miller et al. [15] also suggested a simple exponential model for the decay ,of the
concentration gradient between the body fluid and the dialysate during a single
exchange. On the basis of this model, they concluded that there are three primary
factors affecting mass transport. These are: (1) the general permeability charac-
teristics of the peritoneum, including its underlying interstitial space and the
capillary walls (resistances R4 and R 2 , respectively); (2) the volumes of distribu-
tion of the metabolite present on either side of the peritoneum; and (3) the
transperitoneal metabolite concentration gradient. Clearance measurements
using the techniques (I-III) outlined previously suggested that the dialysate side
of the dialysate fluid film diffusion resistance (R6) is of less importance.
Henderson and Nolph [30] employed the classic two-pool (body fluid and
dialysate) compartmental model utilized in pharmacodynamic studies [49] to
characterize diffusive mass transfer during a single exchange. They clearly dem-
onstrated that the dialysis clearance changes during the residence period and that
the mass transfer rate is relatively insensitive to large errors in estimating the
metabolite space. Mattocks [50] employed a similar two-pool model to character-
ize salicylate transfer in the presence of transfer accelerators. Popovich et al. [51]
expanded the two-pool model to include the effects of metabolite generation (an
important parameter), protein binding, and nonequilibrium distribution. They
simplified the mathematical analysis considerably by replacing the infusion and
drainage times with an equivalent residence time. Many important relationships
between the mass transfer parameters were demonstrated. They defined an
overall peritoneum mass transfer coefficient by

(2)
 121

m
where is the mass transfer rate, K is the overall mass transfer coefficient, A is
the effective peritoneal transfer area, and CBand CD are instantaneous blood and
dialysate metabolite concentration levels. They demonstrated the important fact
that unlike the dialysis clearance, the overall mass transfer coefficient is indepen-
dent of the dialysis schedule.
Equation (2) can be compared to the more familiar form of mass transfer -
Ficks' Law of Diffusion:

m_D dCA
(3)
A - AM dx

where DAM is the diffusivity of solute A in the peritoneum and CA is the solute
concentration of A as a function of distance across the peritoneum, X. Assuming
steady-state transport across a well-defined membrane film of thickness 6, this
can be simplified to:

m= D (CB-CD) = (CB-CD)
(4)
A AM 6 RM
where RM is the resistance to transport across the membrane. A comparison of
equations (2) and (4) illustrates that for a single well-defined film resistance the
mass transfer coefficient (K) is equal to the membrane diffusivity (DAM) divided
by the membrane thickness (6). It is also equal to the reciprocal of the membrane
resistance.
In actual practice metabolites diffusing from the blood to the dialysate encoun-
ter an entire series of resistances, not a particularly well-defined one such as
would be described by equation (4).
A total of six potential resistances have been identified as noted in Figure 2,
Chapter 2. Under these more complex circumstances the mass transport coeffi-
cient is related to these resistances by the reciprocal relationship:

(5)

The relative magnitudes of the individual resistances and the effective mass
transfer area have not been clearly defined. Therefore, Popovich et at. [51]
suggested that the product of the mass transfer coefficient (K) times the area (A),
known as the mass transfer-area coefficient (KA) ,is the transport parameter to be
utilized to characterize mass transfer in peritoneal dialysis. They also demon-
strated that the metabolite generation rate must be included in the analysis.

5.2. Peritoneal dialysis optimizations

Bomar et al. [52] presented the first comprehensive analytical optimizations of

standard intermittent peritoneal dialysis. They also employed a two-pool model
122

and accounted for metabolite transfer during the infusion and drainage periods.
In addition, metabolite generation and residual renal clearance were included to
allow simulation of the interdialytic period. The model was employed to deter-
mine optimum peritoneal dialysis protocols for urea removal.
As would be expected, this model predicts that the maximum urea clearance is
obtained if the residence time is reduced to near zero, as in continuous flow
intermittent peritoneal dialysis (Technique III). Villarroel [48] confirmed these
findings in a comparison of intermittent and continuous peritoneal dialysis. He
concluded that 'continuous flow peritoneal dialysis is more efficient than the
intermittent mode, particularly at higher flow rates'. However, Bomar found that
only a very slight decrease in urea removal rates occurs if the residence time is
increased from 0 to 10 min. Slightly increasing the residence time greatly reduces
the total volume of dialysate used. Thus, it was concluded that short-residence-
time standard intermittent peritoneal dialysis was economically superior to con-
tinuous flow dialysis with minimal effect on efficiency.
Bomar et al. [52] also determined the residence time which yielded the max-
imum urea removal per liter of dialysate (the 'dialysate utilization optimum') for
various degrees of reduction of serum blood urea nitrogen. They found the
optimum residence time to range from 40 to 50 min for reductions in serum BUN
ranging from 20% to 60%.
Finally, a cost function was generated to reflect both the fixed and variable
costs associated with peritoneal dialysis [52]. The model was used in conjunction
with this cost function to determine the dialysate residence time which yielded the
minimum cost. For the cost parameters selected, it was determined that a
minimum in cost would be obtained using a 29 min residence phase for a 60%
reduction in blood urea nitrogen.

5.3. Combined diffusive and convective mass transfer models

The pure diffusive models outlined above all assume no significant ultrafiltration
occurs during an exchange. However, osmotic agents are routinely added to
dialysate [12] to remove excess fluid. As will subsequently be demonstrated, this
gives rise to significant ultrafiltration rates (up to 26 mllmin) [53]. This fluid
movement has been shown to carry solute with it as it traverses the peritoneum
[26, 30]. Neglecting this important mass transfer mechanism results in the com-
putation of higher mass transfer-area coefficients than if it was included in the
models because of the model's attempt to 'best fit' the clinical data without the
convective term. For small molecular weight, rapidly-transferring solutes like
urea, ultrafiltration effects are small compared to diffusion [54] such that forcing a
fit with a pure diffusion model yields reasonable agreement between model
predictions and urea clearance. This will not hold true if high ultrafiltration rates
are obtained. Pyle [54] and Popovich et al. [53] have also demonstrated that the
 123

convective mass transfer mechanism becomes increasingly important as the mo-

lecular weight of a solute increases. Thus, models which account for the transport
effects of ultrafiltration are required to properly analyze most metabolites in
peritoneal dialysis.
In 1973, Babb et al. [56] utilized a simplified transport model which included
convective transport. The authors assumed that convective mass transport oper-
ated as a parallel mechanism with diffusion and included a term to describe its
effect. This model was also employed by Popovich et al. [45,57] who included
provisions for variable blood concentration and metabolite generation. These
models utilized a sieving coefficient, S, as measured by Henderson and Nolph [30]
in the ultrafiltration term. In the mathematical terms, the mass transfer rate is
expressed in the form (compare with Equation (2)):

(6)

where Q u is the ultrafiltration rate.

Bomar [58] presented a comprehensive mathematical model based upon a non-
equilibrium thermodynamics approach [59, 60]. It is a generalized, open, three-
pool model in which the dialysis volume and metabolite distribution volumes vary
as a function of time. Provision is made for solute generation, G, in either pool
and for residual renal clearance. The model is extremely complex, requiring the
simultaneous solution of over eight non-linear differential equations for fluid and
mass transfer. Excellent fits were obtained for blood and dialysate concentration
levels for urea, creatinine and vitamin B-12 over multiple exchanges. The results
are summarized by Popovich et al. [61].

6. Definition of a model of peritoneal dialysis

The most desirable mathematical model is one which closely duplicates the
natural events which occur during peritoneal dialysis but is an acceptable compro-
mise of utility and theoretical accuracy. Numerous models have been proposed as
outlined above with varying degrees of complexity. We believe that in order for a
model to accurately represent the patient-peritoneal dialysis system, it must at a
minimum account for fluid transport with resulting variable ultrafiltration rates
and dialysate volumes, both diffusive and convective transport, metabolite gener-
ation, and residual renal and/or metabolic clearance. While early models ne-
glected some of these factors, later models, such as Bomar'S, were extremely
complex and difficult to apply to the clinical situation.
Figure 2 presents a schematic of a model which conforms to all the criteria
outline above [53]. The peritoneum separates a dialysate compartment from a
body compartment. Both pools are assumed to be well-mixed with concentra-
tions designated by CB and CD' and volumes by VB and V D for blood and
124

Peritoneum
(KA, a)

Figure 2. Schematic of the patient undergoing peritoneal dialysis.

dialysate, respectively. Metabolite generation is assumed to occur at a constant

rate, G. Removal by remaining renal function or metabolism is at the constant
clearance, K R • Ultrafiltration occurs under the influence of hydrostatic and
osmotic forces at a variable rate denoted by Ou, The peritoneal membrane is
characterized by two membrane phenomenological coefficients: the mass trans-
fer area coefficient for diffusion and the reflection coefficient for convective
transport. The reflection coefficient is a constant employed to determine the
fraction of metabolite transferred by entrainment in ultrafiltration from one
compartment to another. These coefficients are assumed to be time invariant and
independent of osmotic and metabolite concentration gradients for a particular
solute.
Mathematically, the metabolite mass balance for the dialysate compartment
and peritoneal membrane is given by:

+ (I-a) Ou C (7)
Rate of accumulation Transfer by diffusion Transfer by convection

where:

(8)

KA = mass transfer-area coefficient

a = reflection coefficient
Ou = ultrafiltration rate
f =!_ 1 (9)
fJ exp (fJ) - 1

fJ = Ou (1- a) (10)
KA
 125

The form of the convective term conforms to homogeneous membrane theory

[62] and pore theory [63, 64]. It is equal to the product of the complement of the
reflection coefficient, the instantaneous ultrafiltration rate, and the weighted
average transmembrane concentratiol1. The function, f, is dependent on the ratio
of convective to diffusive transport, characterized by the Peelet Number, fi[65].
An overall metabolite balance for the system is:

(11)

where the superscript, 0, refers to an initial state. Pyle et al. [53, 54] have
g~monstrated that the ultrafiltration rate can be characterized by the exponential
function:

(12)

The dialysate volume is determined by integrating this equation over time to

yield:

(13)

Combining Equations (5)-(11) and solving yields characteristic expressions for

the dialysate and blood concentrations as functions of the model parameters and
time. Details of the solution are presented elsewhere [54].

7. Clinical measurement of model parameters

In order to measure the membrane parameters using the model outlined above,
concurrent evaluations of fluid and solute transport are required [53, 54]. The
rate of fluid transfer was determined by the degree of dilution obtained with a
large molecular weight substance added to the dialysate. Prior to dialysate
infusion, a known quantity of sterile, pyrogen-free radioisotopically tagged dex-
tran (70000 mol. wt.) was infused into the dialysate. In order to investigate a wide
range of osmotic effects, three dialysate solutions with varying osmolalities were
employed. These were Ringer's lactate (244 mOsm/I), and two commercially-
used Dianeal® dialysis fluids with 1.5% dextrose (332 mOsm/I), and 4.25%
dextrose (477 mOsm/I). Following the completion of infusion, dialysate samples
were obtained at frequent time intervals. Blood samples were obtained immedi-
ately at post-infusion and at the mid-point and end of dwell to determine the
degree of transport of the tagged dextran. By employing the dilution principle,
corrected for such dextran transfer, intraperitoneal dialysate volumes were calcu-
lated at each sample time. A computer program [54] was employed to fit Equa-
126

tion 11 to the calculated volumes using Gauss' non-linear least squares algorithm
[66, 67]. The resulting coefficients were then utilized to compute the instanta-
neous ultrafiltration rate using equation (10).
Transport parameters can be determined for any desired solute once the
intraperitoneal volume profile is known. Concentration-time data have been
generated for a variety of solutes by appropriate analyses of samples selected
from simultaneous fluid transfer evaluations [53, 54]. Since the solution of
equation (7) specifies the dialysate concentration profile for a given set of
transport parameters, a least squares fit to the solute data gives the mass transfer-
area and reflection coefficients characteristic of the patient's peritoneal.

8. Fluid transfer parameters

Typical intraperitoneal volume profiles for Ringer's lactate, 1.5 and 4.25 dextrose
dialysis solution (Dianeal, Baxter-Travenol) are illustrated in Figures 3-5 [53,
54]. In each figure, the asterisks indicate the actual volume data while the solid
line is the least squares fit of Equation (13) to this data. The coefficients which
gave these fits are shown in Table 1. Due to the initial hypotonicity of Ringer's
lactate, the curve in Figure 3 shows an initial rapid drop in volume. As the
osmolality of the dialysate rises due to solute transfer, the rate of volume decline
asymptotically approaches a constant rate. That is, ultrafiltration is initially high
from dialysate to blood (negative ultrafiltration), decreasing to a small negative
value. When a hypertonic solution such as 1.5 gldl dextrose dialysis solution is
2600

2400

..
-
2200

UJ 2000
:E
:J
...J
0
> 1800

.. .
1600

400
C 40 80 120 180 200 240 28C 320
TIME (min)

Figure 3. Intraperitoneal volume profile - Ringer's lactate (reprinted with permission of Pyle WK
(Ref. 54».
 127

Table 1. Volume profile equation coefficients

Solution VD t = 0 al a2 a3

Isotonic 2841.0 -21.4 -0.0195 -0.231

1.5 2130.0 7.48 -0.0207 -0.452
4.25 1988.0 28.4 -0.0170 -1.40

employed, the intraperitoneal volume will increase before reabsorption begins,

as shown in Figure 4. In this case, ultrafiltration is positive (from blood to
dialysate) for the first 136 min resulting in a maximum volume of 2408 ml. Since
the volume at the end of infusion was 2130 ml (t = 0), approximately 278 ml were
ultrafiltered during the residence phase. After the maximum was reached, the
volume begins to decrease and subsequently approaches a constant rate of
reabsorption. Since the 4.25 dialysis solution is also hypertonic but to a greater
degree than the 1.5 gldl dialysis solution, the 4.25 intraperitoneal volume profile
shown in Figure 5 is similar to that of Figure 4 but with a higher ultrafiltered
volume and greater initial slope. The peak volume for this exchange was 3325 ml
at 177 min. The 1337 ml of ultrafiltration represents a 67% increase over the post-
infusion volume of 1988 ml. As in each of the other cases, fluid reabsorption
eventually begins and approaches a constant rate.
The volume profile coefficients of Table 1 also specify the ultrafiltration profile
through equation (12). The ultrafiltration profile corresponding to the 1.5 gldl
dialysis solution volume profile of Figure 4 is shown in Figure 6 [53, 54]. The
2600

--
2500

*
2400
* *
.-. *
E

w
2300 . It
::2:
~
...J
0
>

2100

2000
0 50 100 150 200 250 300 350 400
TIME (min)
Figure 4. Intraperitoneal volume profile -1.5 g/dl dextrose dialysis solution (reprinted with permission
of Pyle WK (Ref. 54».
128

ultrafiltration rate is approximately 7 ml/min immediately post-infusion. The

ultrafiltration rate decreases to 0 at 136 min, and levels off at about -0.45 ml/min.
As indicated above, the higher tonicity of the 4.25 dialysis solution increases the
magnitude of the initial ultrafiltration. The profile shown in Figure 7 is derived
from the volume curve of Figure 5. The initial rate exceeds 26 ml/min dropping to
-1.4 mllmin after 4 to 5 h.
Pyle et al. [53] have reported the fluid transfer results in four studies each with
1.5 gldl and 4.25 g/dl dextrose dialysate. The average results are summarized in
Table 2. From these studies, the 1028 ml ultrafiltration induced by the 4.25
solution is approximately three times that removed with the average 1.5 solution.
Since the concentration of the osmotic agent is higher in the 4.25 dialysis solution,
the osmotic gradient will be maintained longer. Thus, ultrafiltration is positive for
a longer period when the 4.25 solution is employed. On the average, ultrafiltra-
tion is positive for 140 and 247 min with the 1.5 and 4.25 solutions, respectively.
Since fluid reabsorption begins after these respective times, these times mark the
point of maximum fluid removal and maximum intraperitoneal volume. The
initial or maximum ultrafiltration rate which will occur is also dependent on the
magnitude of the initial osmotic gradient. These studies have shown the average
post-infusion ultrafiltration rate to be 11. 7 mllmin for the 1.5 and 16.6 mllmin for
the 4.25 solution. The reabsorption rates reported by Pyle et al. [53] were -0.68
and -0.87mllmin for 1.5 amd 4.25 dialysis solution, respectively. Since both of
these values, shown in Table 2, were subject to relatively large variations, they
may represent a single value (p >0.65). Reabsorption may be utilized clinically to
3600
r .. .
3300
*
3000
...E
lJ.J 2700
~
:J
...J
0
> 2400

2100

1800
0 80 160 240 320 400 480 560 640
TIME (min)

Figure 5. Intraperitoneal volume profile - 4.25 g/dl dextrose dialysis solution (reprinted with per-
mission of Pyle WK (Ref. 54».
 129

10

·ec: 8

-
"-
E
6
lJ.J
I-
el
a::
4
z
o
~
a:: 2
~
Li:
el
a::
I-
...J 0 - - - - - - - - -_-=-=-::...::::.:-=--==~-=-=-==.::-=-::..=
:::>

-2 ~--~----------~--~----~----~--~-----
o 50 100 150 200 250 300 350 400
TIME (min)

Figure 6. Transperitoneal ultrafiltration profile -1.5 gldl dextrose solution (reprinted with permission
of Pyle WK (Ref. 54)).

40

·ec: 32
:::
E
24
lJ.J
I-
el
a::
16
z
o
~
a:: 8
~
Li:
el
a:
I- 0 --------------

-9 ~--~--~----~--~----~--~----~----
C 80 160 240 320 400 480 560 640
TIME (min)

Figure 7. Transperitoneal ultrafiltration profile - 4.25 gldl Dextrose solution (reprinted with per-
mission of Pyle WK (Ref. 54».
130

introduce fluids into the vascular space of a hypovolemic patient by leaving the
fluid in the peritoneal cavity.
Considerable variations in ultrafiltration characteristics between patients have
been noted [53, 68]. Some of the variations may be attributed to differences in
mass transfer characteristics and the relative degree of drainage. Dilution of fresh
solutions by residual dialysate can reduce fluid transfer. Nolph has hypothesized
that ultrafiltration is actually the result of two different fluid transfer mecha-
nisms: high proximal capillary ultrafiltration and lower negative distal ultrafiltra-
tion [69]. This suggests that a patient's ultrafiltration characteristics can be
affected by permeability changes at either the proximal or distal portions of the
capillary or both. Undoubtedly, fluid transfer is influenced by hydrostatic and
osmotic pressures in both the capillaries and in the dialysate, the interstitial water
path dimensions, and molecular surface charges in the interstitium [69]. Detailed
studies will be required to identify the importance of the various factors and the
sources of the reported interpatient variations.

9. Solute transfer parameters

Popovich et al. [70] have reported peritoneal mass transfer area coefficients
(MTAC) resulting from 34 studies in 8 patients on Intermittent Peritoneal
Dialysis and CAPD. These early studies predated the concurrent evaluations of
fluid and solute transfer. As a consequence, they assumed constant ultrafiltration
rates. Also, the convective term was assumed to be equal for all solutes in the
studies using hypertonic solutions after the results of Henderson and Nolph [30].
From the mean mass-transfer area coefficients, they produced the first empirical
correlation of MTAC to solute molecular weight, shown in Figure 8. While the
MTAC's for the different solutes exhibited considerable interpatient variations, a
trend given by:

KA = 333.6 (MW)-O 561 (14)

was evident over six orders of magnitude of molecular weight.

Table 2. Average fluid transfer results

mean ± S.D. (n = 4)

1.5 Dianeal® 4.25 Dianeal®

Maximum ultrafiltered volume (ml) 331 ± 187 1028 ±258

Time of maximum volume (min) 140±48 247 ±61
Maximum ultrafiltration rate (mllmin) 11.7 ± 13.0 16.6 ± 7.7
Reabsorption rate (mllmin) -0.68± 0.61 -0.87± 0.55
 131

Peritoneum
-0561
KA • 333.6 (M W I

r ·097
<I 10
Vl

- o
'-
c:
E I
'-
E Peritoneum
o
<I
o
'" o
o
o

MOLECULAR WE IGHT

Figure 8. Peritoneal mass transfer - area coefficient as a function of solute molecular weight (reprinted
with permission of the American Institute of Chemical Engineers (Ref. 61».

Pyle et al. [53] later performed sixteen comprehensive transport studies in five
CAPD patients, measuring fluid transfer rates and diffusive and convective solute
transfer parameters. Figures 9-12 present representative dialysate concentration
profiles for BUN, creatinine, inulin, and total protein, respectively, from these
studies. Dialysate concentrations are shown by the asterisks with the solid line
indicating their model's predictions. Blood levels are shown by circles and the
dashed lines. From this sequence of figures, the effect of molecular size on
transport rates becomes apparent. Due to its small molecular size (60 daltons),
urea has a relatively high MTAC of 15.3 ml/min and low reflection coefficient of
0.126. These membrane properties lead to a rapid rise in the dialysate urea level
and a slight drop in the blood concentration during the first 90 min of the
exchange as seen in Figure 9. Creatinine (113 daltons) is larger than urea and
transfers more slowly, as shown by Figure 10. In this case, the parameters are an
MTAC of 9.0mllmin and a reflection coefficient of 0.200. Since the transfer rate
is slower, the creatinine dialysate concentration profile does not show the high
initial slope or strong curvature of the BUN profile. Also, the difference between
blood and dialysate levels is greater at any given residence time for the larger
metabolite. This effect is even more prominent for inulin (5500 daltons) and total
protein (340000 daltons, weighted mean) shown in Figures 11 and 12, respec-
tively. In Figure 12, the plasma protein level of7900 mg/dl is so much greater than
in the dialysate that it is not practical to show it on the graph.
132

100

80 _ _ _ J>
--------~-----~

60
"0
"-
0'
E
z 40
:::J
CD
PARAMETERS

20 MTAC 1531
SIGMA 0126
CORRELATION 0999

o * - DIALYSATE
o -- BLOOD

__ ____ ____ __ ____- L____ __- J_____

o 80 160 240 320 400 480 560

TIME (min)

Figure 9. Blood urea nitrogen concentration profiles in CAPD (reprinted with permission of Pyle WK
(Ref. 54)).

24

20
~---------------~------------------~

~
0' 16
£
w
z
z 12
f-
<[
W PARAMETERS
a:: ..HAC 9041
u 8
SIGMA 0200
CORREL:'TION 0999

4 - * - DIALYS-"TE
o - - BLOOD

8 L -__________________ ~ ____ ~ ____ ~ ______ ~ ____ ~ _____________

4':; '30 120 160 200 240 280 320

Figure 10. Creatinine concentration profiles in CAPD (reprinted with permission of Pyle WK (Ref.
54)).
 133

The protein concentration profile of Figure 12 also illustrates the effect of high
fluid transfer rates on solute transfer. For about the first 80 min of the residence
phase of this exchange, the ultrafiltration was positive and relatively high, similar
to the situation shown in Figure 7. The high fluid transfer rate resulted in the
convection of protein across the peritoneum by entrainment and a rapid initial
rise in the dialysate protein level of about 30 mg/dl. As the ultrafiltration rate
declines, solute convection becomes relatively less important than diffusion and
the concentration profile, assumes a nearly linear trend. At the lower diffusion
induced transfer rate, about 320 min is required to obtain another 30 mg/dl rise in
the protein level. Thus, even a small period of convective transport can be quite
important in the dialysis of large solutes.
An important aspect of the studies by Pyle et al. [53] was the first determination
of peritoneal membrane reflection coefficients for a variety of solutes. Mean
reflection coefficients for urea (60 daltons), creatinine (113), uric acid (158),
glucose (180), inulin (5500), dextran (70000 mean) and total protein (340000
weighted mean) are presented in Figure 13 with an empirical correlation function
to molecular size. Prior to these studies, the most frequently referenced work [30]
on convective coefficients indicated that both urea and inulin had peritoneal
reflection coefficients of about 0.2. While these values fall within the range of
variation reported by Pyle, the mistaken conclusion that peritoneal sieving is
equivalent for all solutes could be drawn from the earlier work. Figure 13

18000
PARAM ETERS
MTAC 5! 75
5~OO
SIGMA 0243
C:JRRELATIO', 0999

E 12CCC .. - [Link]
"E o -- BLOOD
Co
'C

Z 900e
..J
:::l
Z
6000

3000

----- ------- -------------~

o~----~-~-~-~-~-~-~----~--~--~--------------
40 80 120 160 200 240
TIME ,min)

Figure 11, Inulin concentration profiles in CAPD (reprinted with permission of Pyle WK (Ref. 54)).
134

120

100

~
"-
0' 80
E

z
60
C PARAMETERS
?=
4C MTAC 0037
SIGMA. 0984
-<:::: CORRELATION 0990

c.~
* - - :lIALYSATE

::
8e '60 240 320 400 480 560 640

Figure 12. Total protein concentration profile in CAPD (reprinted with permission of Pyle WK (Ref.
54)).

'~" ~
~

Cl

C)
Cl
" 12
~
10 ill T
b ~

~
~ ~S:!
z
w
U
o 8 ~"'"
j::;:"q<!'l
~<t,,8
~ '"
~"'
ClCl
LL 't~~ ::l ~~
I..i... :::.,,:::.'"
w 06 -JCl
0 ~::!,
u E?~

z 04
0
f-
U
w 02 (J I - exp (- 0461 d )
-..J
I..i...
W 1
I I - exp (- 0609 Tmw )
0::
0
0 10 20 30 40 50 60 70 80 90 100
MOLECULAR DIAMETER,d (Al

Figure 13. Peritoneal membrane reflection coefficients as a function of molecular size (reprinted with
permission of Pyle WK (Ref. 54».
 135

indicates that this is clearly not the case. In these studies, the reflection coefficient
ranged from a mean of 0.18 for urea to 0.992 for total protein and was well-
correlated to the expression

a = 1 - exp[ -0.0609(MW)(1/3l ]. (15)

This function indicates that the majority of smaller solutes, such as urea,
entrained in the ultrafiltrate are convected through the membrane. Conversely,
less than one percent of the protein in the same ultrafiltrate passes through due to
the higher degree of reflection. As indicated above, the convective transfer rate
of large solutes can still be appreciable if the fluid transfer rate is significant.

10. Continuous ambulatory peritoneal dialysis

Prior to 1975, chronic peritoneal dialysis techniques were generally performed

intermittently, two or three times per week. The high efficiency of hemodialysis
clearances allowed for short treatment times; a total of approximately 15 h/week.
The less efficient standard intermittent peritoneal dialysis required 30-40 h of
treatment per week. This difference in treatment times is one factor resulting in
the vast majority of patients being treated with hemodialysis at that time.
During a typical 5 h hemodialysis treatment, the concentration of blood urea
nitrogen is reduced from approximately 100 mg/dl to about 40 mg/dl concomitant
with the removal of approximately 2-31 of water. For the next 43 to 67 h, no
dialysis is performed, and urea, other metabolites, and fluids accumulate. The
cycle is repeated by subsequent periods of dialysis and reaccumulation.
Kjellstrand et al. [71] have hypothesized that these fluctuations may be detrimen-
tal to the patients' health. In fact, the application of very high efficiency hemo-
dialysis results in a 'disequilibrium syndrome' characterized by headache, nausea,
vomiting and severe blood pressure declines [72]. Arieff et al. [72] have suggested
that this syndrome is a result of large intracellular-to-extracellular concentration
gradients with concomitant fluid movement under the resulting osmotic gra-
dients, particularly across the blood-brain barrier. This hypothesis is supported
by the results of Popovich et al. [73] who have shown that resistance to mass
transfer across the cellular membranes can result in significant transcellular
concentration disequilibrium during hemodialysis.
The clinical symptoms associated with short-term intensive dialysis treatments
can be eliminated by performing the treatment continuously at a modest effi-
ciency analogous to normal kidney function but at a reduced clearance. A clinical
technique which accomplishes this using peritoneal dialysis was developed by
Popovich and Moncrief in 1975 [46]. This procedure is termed Continuous
Ambulatory Peritoneal Dialysis (CAPD).
136

11. Theoretical basis of CAPD

The clinical protocol which will yield desired blood metabolite concentrations
levels employing CAPD can be specified by the use of a simple mathematical
model for the patient-peritoneal dialysis system. Popovich et al. [73] have demon-
strated that a single body pool assumption, such as the one schematically outlined
in Figure 2, is valid for low dialysis clearance systems such as CAPD. The verbal
statement which applies to metabolites in the body is that accumulation is equal to
the rate of generation, G, minus the rate by residual renal clearance, K R , or
dialysis clearance, K D • This verbal statement is illustrated in Equation (16).

d
at (VBCB) = G KRCB KDCB· (16)
Rate of Rate of Renal Dialysis
accumulation generation removal rate removal rate

As noted, the accumulation term is the time derivative of the total mass of
metabolite in the system (the blood concentration level, C B , times the volume of
distribution, VB). This term will equal zero if the dialysis treatment is continuous,
i.e. the concentrations and volumes are constant. Setting the accumulation term
equal to zero and solving the resulting algebraic equation for the steady state body
metabolite concentration level in term of the system parameters yields:

C _ G G (17)
B - KR +KD - K '

where K is the total metabolite clearance or the sum of dialysis, renal and, if
present, metabolic clearance. Note that these clearances are additive.
A typical blood area nitrogen generation rate for dialysis patients is 5.7 mg/min
[74]. If one desires to maintain a patient at a continuous BUN level of 80 mg/dl,
equation (15) predicts that a total clearance of 7.1mllmin is required as
illustrated:

K= G = 5. 7 mg/min = 7 1 1/ . (18)
CB 0.8 mg/ml . m mill.

Multiplying this by 1440, the number of minutes per day, yields the daily clearance
requirement:

K = (7.1mllmin) (1440 min/day) = 10200ml/day. (19)

Thus, a total clearance of about 10 l/day is required to maintain a patient with a

BUN of about 80 mg/dl.
This clearance can be obtained using any dialysis technique. Peritoneal dialysis
 137

clearance is defined by:

K = VDCD (20)
D TCB '

where V D is the drained dialysate volume at a mean concentration of CD over a

total time T.
As noted in Figure 9, the dialysate BUN concentration asymptotically ap-
proaches the blood concentration level following infusion of fresh dialysis fluid. If
the residence phase is sufficiently long, equilibration will occur such that CB =
CD' Under these circumstances equation (18) reduces to:

KD = ~D = QD = dialysate flow rate. (21)

This model leads directly to the clinical protocol of CAPD. The theory predicts
that a patient will maintain a steady BUN level of 80 mg/dl if 10 1 dialysate are
allowed to equilibrate with body fluids on a daily basis. Since the normal infusion
volume is 2.01, four infusions per day will result in a total of 8.0 1. Approximately
2l/day are ultrafiltered yielding a total drained volume of 10 l/day as predicted.
Four exchanges per day will require a mean residence time of approximately 6 h
per exchange for continuous treatment. This is more than adequate to achieve
equilibration for BUN. Equilibration will not be achieved for higher molecular
weight substances such as inulin (5500 MW) and proteins.

12. CAPD exchange rate criteria

Table 3 presents a list of the major factors which affect CAPD exchange rate
criteria. Under ordinary conditions, two liter infusions are employed. Mean
peritoneum transport characteristics and ultrafiltration volumes have been deter-
mined as outlined above. Using this data. Popovich et ai. [75] have computed the
number of two liter exchanges per day which will be required to maintain a BUN
of less than 80 mg/dl in a typical CAPD patient. The results are presented in
Figure 14. The results demonstrate that the typical patient with low residual renal

Table 3. Factors affecting exchange rate

Major factors Influenced by

Generation rate Diet, weight, metabolic rate

Residual renal clearancs
Ultrafiltration rate Fluid intake, dialysate tonicity
Mass transfer-area coefficient Peritonitis, vascular disease, vasodilation
138

4
>.
o
o G=
...
41 8.4 mg, .
0.. 'min
on
3
41
01
c:
o

.. 7. 0
.J::.
U

W 2

5 .6

4 .2

o 2 3 4 5
Res i dua l Renal Clearance (mi Imin)

Figure 14. CAPD exchange rate determined by residual renal clearance and BUN generation rate
(reprinted with permission of Excerpta Medica (Ref. 75» .

clearances (0 to 1.0 ml/min) will generally require 4 exchanges per day. Patients
with zero residual renal clearance will have a steady state BUN near 80 mg/dl.
Patients with renal clearances near 1.0 ml/min will obtain an additionall.4l/day
clearance resulting in a BUN level of about 67 mg/dl. At a renal clearance of
approximately 1.5 mllmin , only three exchanges per day are predicted to maintain
the patients at a BUN of 80 mg/dl. Two exchanges per day are predicted for a
residual renal clearance function of approximately 3.0 mllmin. Of course, the
generation rate is also an important parameter which can be related to protein
intake [74]. Figure 15 presents identical information but is based on maintaining a
steady creatinine level below 12 mg/dl.

13. Membrane transport limitations

The authors caution that the above results are based on patients with average
ultrafiltration rates and peritoneal transport characteristics [75] . Indeed,
Popovich et al. [76] have recently presented a documented case of a patient who
 139

3
I. 2 "'%,in

2
/.0

0.8

o 2 3 4 5
Residual Renal Clearance (mllm in)

Figure 15. CAPO exchange rate determined by residual renal clearance and creatinine generation
rate.

developed uremic symptoms on 4 exchanges per day as a consequence of dras-

tically reduced peritoneum mass transfer ability. The patient developed nausea,
vomiting anorexia, and lethargy while undergoing CAPD with four exchanges
per day. The patient became asymptomatic when placed on five exchanges per
day. Detailed transport studies were obtained at this time . The exchange sched-
ule was again altered from 5 to 4 to 5 exchanges per day with the development and
abatement of symptoms. Symptoms did not reappear following a final shift to 4
exchanges per day. Again, transport studies were obtained. The resulting mass
transfer-area coefficients are presented in Table 4.

The authors found that the patient had a creatinine mass transfer-area coeffi-
cient of only 5.1 ml/min (22% of normal) during the period he developed uremic
symptoms on four exchanges per day. His BUN and creatinine levels were 97 and
21.5 mg/dl, respectively. The creatinine MTAC had increased to 12.8 mllmin
(54% of normal) when he was asymptomatic on four exchanges per day. This led
the authors to conclude that decreased mass transport can result in insufficient
metabolite removal and the development of uremic symptoms on CAPD [76].
140

Also, because of high MTAC's for middle molecules in the patient and the known
large removal rates of middle molecules for CAPD, relative to other dialysis
techniques [75], the authors suggested that it is the smaller metabolites which
were responsible for the uremic symptoms noted.
A simple clinical protocol to evaluate the mass transport capabilities of a
patient based on the degree of equilibration of creatinine during a test exchange
has been presented [76]. The clinical protocol is outlined in Table 5. This can be
performed using a needle for dialysate infusion if a nephrologist wishes to test a
patient's ability to perform CAPD prior to insertion of an indwelling catheter.
The degree of creatinine equilibration in a four-hour period is determined
using 2.0 liters of standard 1.5% dextrose Dianeal®. The four-hour period indi-
cated refers to the period between the end of infusion and sampling of the blood
and dialysate. The MTAC can be estimated from the degree of equilibration
using Figure 16 which is based on the authors' theoretical correlations [76].
The average creatinine MTAC is about 23.5 ml/min [70] corresponding to 92%
of equilibration for creatinine after a four-hour dwell period. The authors suggest
that patients with MTAC's less than one-fourth of average (less than 50%
equilibration) may develop symptoms of under-dialysis and require additional
exchanges [76]. Patients with greater than one-half average MTAC should be
considered suitable candidates for CAPD from the standpoint of mass transport
ability.

Table 4. Mass transfer-area coefficients (ml/min)

Solute (mol. wt) Study 1 5 ex/day Study 2 4 exlday Controls (averaged

results)

BUN (60.1) 12.6 17.0 33.5

Creatinine (113) 5.1 12.8 23.5
Glucose (180) 5.2 6.6 18.1
Inulin (5500) 3.7 4.0 2.7
Dextra,n (70 (00) 1.2 1.9 0.6

Table 5. Transfer evaluation protocol

1. Infuse 2.011.5 gm% dextrose dialysis solution intraperitoneally.

2. Let solution dwell for 4 hours post-infusion.
3. Sample dialysate and plasma for creatinine concentration.
4. Calculate DIP creatinine concentration ratio.
5. Find DIP ration on 4 hour creatinine equilibration curve.
6. Read corresponding MTAC from abscissa.
 141

100
z
0
~
Cl::
75
ID
::::i
:::J
a 50
w
I-
Z
W
u 25
Cl::
w
Cl

0 10 20 30 40
MTAC

Figure 16. Percent creatinine equilibration (dialysate/plasma ratio) as a function of mass transfer
coefficient and dwell time (reprinted with permission of Artificial Organs (Ref. 93)).

Another factor which can influence CAPD exchange rate criteria is the neces-
sity to maintain proper fluid balance. Generally, CAPD provides excellent
control of fluids [68]. However, there may be difficulties in achieving adequate
ultrafiltration during episodes of peritonitis [77]. This may result from increased
glucose transport with resulting rapid loss of osmotic gradients; Randerson and
Farrell [78] have demonstrated substantial increases in the MTAC in two patients
during bouts of peritonitis. This may necessitate more frequent exchanges em-
ploying more hypertonic solutions.
Occasionally, patients under routine treatment exhibit hypotension [77]. This
can be alleviated by using less of the higher tonicity dialysate. During hypotensive
episodes, it may also be helpful to skip the next scheduled exchange. By leaving
the fluid in for a period of two exchanges, the body reabsorbs dialysate fluid at
approximately 1 ml/min during the second four to six hour period. This will help
any acute situations with little noticeable increase in BUN levels.

14. CAPD in children

CAPD is currently enjoying rapid expansion in the treatment of adult patients

with end stage renal disease. This is due in part to the advantages that CAPD
exhibits over hemodialysis [79-81]. These include: (1) continuous biochemical
and fluid control, (2) patient mobility and freedom from a machine, (3) greatly
reduced dietary restrictions, (4) simplicity of operation, and (5) elimination of the
need for routine blood access. Because of these advantages, preliminary clinical
studies suggest that CAPD may be an attractive alternative to hemodialysis in the
treatment of infants and children. Promising clinical results have been reported in
limited clinical trials [82, 83].
142

It is known that metabolite transport in children differs from that in adults [32,
33,84]. However, only recently have detailed transport studies been conducted
by Popovich et al. [85] to characterize their differences. Peritoneal mass transfer
characteristics were measured in seven studies in four children with an age range
of 17 months to 6 years. The clinical protocol of Pyle [54] outlined above was
employed using radioiodated human serum albumin (RISA) in place of the
tagged dextran [85]. The volume of the dialysate fluid present in the peritoneal
cavity was computed using the degree of dilution of the RISA. Mass transfer-area
coefficients were computed from an analysis of the concentration-time profiles
and known ultrafiltration rates.
Typical concentration-time results are presented in Figure 17. The BUN dialy-
sate to plasma concentration ratio versus time in three patients is illustrated. Note
that results similar to the adult reference are obtained. The same held true for
creatinine, uric acid and glucose [85].

The actual intraperitoneal dialysate volumes as a function of dwell time for

1.5% and 4.25% dextrose solutions in a typical patient are illustrated in Figure 18.
Again, the results are similar to that in adults. A rapid increase in dialysate
volume is obtained shortly after infusion. This occurs as a result of high glucose
concentration gradients with concomitant osmotic driving forces. The glucose is
rapidly absorbed causing the volume to reach a maximum and then decline as
dialysate is reabsorbed under the influence of the oncotic force of the blood.
Dwell times corresponding to maximum ultrafiltration were approximately 3 to
4h.

10 --------
o
~
II:
z 08
::>
(Il

<!
~ 06
<!
.J
a.
I'? 04 ~ 17 MO - 6 9 KG
x 6 YR • II 7 KG
W
I- • 6 YR -147 KG
<!
Ul
;>j 02
<!
o

o 100 200 300 400 500

DWELL TIME (min)

Figure 17. BUN dialysate/plasma ratio in children on CAPD.

 143

750

•
-E 700
425 G/DL
DEXTROSE

.., •
~
::J
..J
0
>
..,
~
<I
(J)
>
..J
<I
15

..,
..J
<I
Z
0 I 5 G/DL
~ DEXTROSE
ii:
LU
a. 450
<I
II:
~

~
400 INFUSION VOLUME
(400ml)

350
0 100 200 300 400 500
DWELL TIME ( ml n)

Figure 18. Intraperitoneal volume profiles for a 14.7 kg child on CAPD.

Mean mass transfer-area coefficients, determined by Popovich et al. [85], are

presented in Table 6 with an adult reference. Clearly, the MTAC's are quite
different from the adults. Some relationship between the MTAC and child size is
evident. The reader will recall the MTAC is a measure of the maximum possible
dialysis clearance which would occur if the dialysate metabolite concentration
level was somehow maintained at zero at all times. It is also equal to the area
available for mass transport divided by the sum of all resistances to transport (see
Equation (5)). If the anatomical and physiological characteristics of the per-
itoneum in the child per unit area are similar to that in adults, the sum of the
resistances would also be similar. This suggests that the decrease in the MT AC's
might be related to a decrease in the area for transport which could be correlated
for adult values by some scaling factor.
Table 7 shows how the mean solute MTAC's scaled by body surface area and
weight. Scaling by weight gives a much better correlation with adult values in the
children studied [85].
Scaling by weight rather than body surface area is also supported by an
ultrafiltration comparison. The mean rate of ultrafiltration immediately follow-
ing dialysate infusion (the maximum ultrafiltration rate) was determined to be 2.5
to 3.0mllmin compared to 11.7 to 16.6mllmin in adults [85]. Table 8 illustrates
that the mean maximum ultrafiltration rates give better correlation to adult
values if scaled by weight.
144

Table 6. Mean mass transfer area coefficients

Patient/wt (kg) in mllmin

1 2 3 4 Adult
7.0 8.2 11.7 14.7 reference

Solute
Urea 2.69 6.14 7.50 7.71 33.6
Creatinine 1.49 1.51 3.36 4.15 23.5
Uric acid 0.88 2.59 4.34 4.40 19.5
Glucose 1.67 3.35 4.15 3.52 18.1

Table 7. Mean BUN mass transfer-area coefficients scaled by body surface area and weight

Patient KA Surface area Wt KA scaled KA scaled by wt Adult

mllmin M2 Kg by surface area reference

2.69 0.34 7.0 13.7 26.9 33.6

2 6.14 0.38 8.2 28.0 52.4 33.6
3 7.50 0.51 11.7 25.4 44.9 33.6
4 7.71 0.65 14.7 20.5 36.7 33.6
Mean 21.9 ± 5.4 40.2± 9.5 33.6

Popovich et al. [85] noted the apparent contradiction of metabolite concentra-

tion equilibration data similar to adults in the face of drastically reduced mass
transfer-area coefficients. They resolved this by rearranging the basic mass
transport relationship (equation (1» for the change of dialysate concentration
versus time to obtain:

dCD _ KA (C
dt - VB-CD) QU[I-a
D
() ]
C-CD · +"D
(22)

The parameters which determine the rate of dialysate equilibration are KAIVD
and QuIVD. The authors had demonstrated that both KA and Qu appear to
correlate in direct proportion to body weight. Therefore, if the infused dialysate

Table 8. Maximum ultrafiltration rates scaled by body surface area and weight

% Measured Scaled by Scaled by Adult

Dextrose mllmin surface area weight mean

1.5 2.5 5.3 9.5 11.7

4.25 3.0 9.6 17.8 16.6
 145

volume, VD' is also decreased proportionally to decreasing weight, the parameter

will have constant values. For this case, mass transfer solutions of equation (22)
will yield similar dialysate concentration profiles regardless of patient size. In
fact, clinicians treating children have typically scaled infusion volumes by body
weight, which explains the rather consistent DIP curves obtained for the different
children.
Finally, the authors predicted daily infusion volumes necessary to obtain
specified creatinine concentration levels for adults and children using four CAPD
exchanges per day. Creatinine was selected because of the wealth of generation
rate data for children and adults [87-89]. The mass transfer-area coefficients and
ultrafiltration rates were assumed to scale by weight. Residual renal clearance
was assumed to be negligible as a worst case basis.
The results of the computations are illustrated in Figures 19 and 20. Also shown
are data points corresponding to actual infused volumes and weights of children
treated with CAPD. Excellent agreement was obtained between the predicted
and clinical results. However, the authors caution that 'their curves are based on
average data and should only be utilized to obtain an initial estimate. Considera-
ble individual variations in peritoneal transport characteristics are common.
Generation rates also vary' [85]. Nevertheless, the curves provide a basis for a
first estimate of infusion volume which can be adjusted depending in the particu-
lar circumstances involved.

15. Kinetic studies of dialysis

15.1. Effects of convective transport

It has long been recognized that fluid transfer occurs in peritoneal dialysis and
that some entrainment or convection of solutes must result. Until the studies of
Pyle et al. [53, 54], however, definitive data on solute convection was not
available and the contribution of convection to total solute transport could only
be estimated [90]. From comprehensive transport studies in CAPD patients, Pyle
[54] has determined the individual diffusive and convective mass transfer rates as
a function of time. Figure 21 illustrates the fraction of total solute transfer
resulting from convection for the solute concentration profiles shown in Figures 9
to 12 [54]. This data shows that convection is a small but significant mechanism in
urea transport. At the maximum, the contribution is about 12%. As solute size
increases, the importance of convective transport increases dramatically. Im-
mediately post-infusion (t = 0), convection accounts for about 86% of the total
protein transfer. Thus, it is apparent that convection cannot be neglected as a
transport mechanism, especially for large solutes. If this mechanism is not in-
cluded in models employed in the evaluation of transport parameters, the remain-
ing parameters may be seriously over-estimated.
146
1200
• PATI ENTS ON 5 EXCHANGES I DAY
• PATIENTS ON 4 EXCHANGESI DAY
1000 Plasma
C1«J1mmefmg/dJ}

E BOO
w
;:;;
:::>
...J
~ 600
z
0
iii 400
:::> 12
LL
;;:
200

0
0 3 6 9 12 15
WEIGHT (KG)

Figure 19. Predicted infusion volume required for CAPD patients as a function of weight (3 to 14 kg).

3500
" PATIENTS ON 4 EXCHANGES I DAY

3000

6
I
PlasmQ .....(>.
8 10 12 14
CrHlfll,,"e
2500
(mg/(1I)
w
;:;;
:::>
...J
~ 2000

z
o
~ 1500
LL
z
1000

500

o 15 30 45 60 75 90
VvEIGHT (KG)

Figure 20. Predicted infusion volume required for CAPD patients as a function of weight (greater than
14 kg).

15.2. Technique comparisons

The two classic treatments of uremia via dialysis are hemodialysis and peritoneal
dialysis, with many technical variations on each. The evaluation of these various
procedures is complicated by the lack of a clear understanding of uremia itself,
i.e. what the toxins involved are, and which are the most important. However,
more recent comparative studies have employed mathematical models of the
dialysis techniques to identify the important variables and their effects. Through
 147

1.0

0.8
z
\~
0
i=
u
<t
a::
u.. 0.6
\\
w
>
i= \\ --- -----
--- -----
u

\\
w 0.4 Z
>
Z
.c. ..... / '

~ ......-----
,/
0
..... .--
,/

.< ,//
.-- .--
0.2 ./. .-- .--
..... /

.-- .....
..... /
X:;' ;;,..---
:::;;---
0
0 100 200 300 400 500
TIME (min)
Figure 21. Fraction of solute transport due to convection in peritoneal dialysis (reprinted with
permission of Pyle WK (Ref. 54)).

the ability of these models to simulate a variety of therapeutic conditions, it is

possible to draw comparisons of techniques as dissimilar as hemodialysis and
peritoneal dialysis.
Popovich et ai. [75,92,93] have employed a two-pool patient model to compare
these techniques and CAPD for the ability to remove urea, creatinine, vitamin
B-12 (1355 daltons) and inulin (5500 daltons). Their model generated concentra-
tion profiles for intra- and extracellular body compartments as shown for urea
and the 5500 dalton solute in Figures 22 and 23, respectively. This data confirms
that 40 hour/week intermittent peritoneal dialysis is a less efficient method of
dialyzing urea than 15 hour/week hemodialysis. However, CAPD with 4 ex-
changes/day results in lower metabolite levels than IPD. In this case, the mini-
mum IPD levels and the CAPD levels (which are nearly constant) are both
approximately equal. The predialysis IPD levels are approximately 40% higher
than the CAPD levels. Also note the nearly constant BUN levels with CAPD
patients relative to the saw-tooth concentration pattern for both of the intermit-
tent procedures.
As outlined in Figure 23, the results are significantly different for a 5500 dalton
'middle' molecule. While CAPD is still more efficient than IPD (i.e. CAPD
results in lower concentration levels), both are substantially better than hemo-
dialysis. This is because of the rapid decrease in hemodialysis clearance with
 148

120

100

;,!?
CAPO 4 ex/day
0
Cf'
80
E

z
:::> 60
en
- - Intracellular
40 --- Extracellular

20
o 10 20 30 40 50 60
Tima (hrs.l
Figure 22. BUN concentration profiles for stable, average patient on IPD, CAPD, and hemodialysis.

30

HEMO
I') i', _-- ------
0 25 - '-------

o:;,!?
(j\ 20 -
E
- - Intracellular
z - - - Extracellular
o 15 _
I-
<!
0::
I-
z
w 10 IPD
---- ---
~
o
u
'--
------- ---
5 CAPD 4 ex/ day
f..-- - - - - - - - - - - - -----

o 10 20 30 40 50 60
Time (hrs.)

Figure 23. Middle molecule (5500 dalton) concentration profiles for stable, average patient on IPD,
CAPD, and hemodialysis.
 149

mcreasing molecular weight relative to peritoneal dialysis [76]. Again, the CAPD
levels are relatively constant, due to the continuous application ofthe procedure.
It is interesting to note that in CAPD there is very little transcellular difference
in the urea levels compared to the conditions during hemodialysis. As solute size
increases, the transcellular concentration gradient increases in magnitude, but
the gradient is never as great as with hemodialysis. Hiatt et al. [93] have shown
that the CAPD transcellular disequilibrium is virtually the same as would be
found in individuals with normal renal function. They have illustrated this by
performing mathematical simulations of concentration levels in subjects with
normal renal function and in those undergoing dialysis. The maximum dis-
equilibrium was then determined and a ratio of the dialysis to normal values was
calculated. The ratios, shown in Table 9, indicate that the disequilibrium experi-
enced by CAPD is never more than four percent greater than that in normals
while it may exceed 900% in hemodialysis patients. It has been hypothesized [72,
94] that this abnormal degree of disequilibrium is detrimental to the patients'
health.

It is apparent from the concentration profiles in Figures 22 and 23, that there is
some solute size for which hemodialysis and CAPD are equivalent at steady state.
Figure 24 shows the ratio of the predialysis hemodialysis to CAPD concentration
levels as a function of solute molecular weight. For urea, this ratio indicates that
four CAPD exchanges per day are roughly equivalent to three IS-hour hemo-
dialyses per week. As solute size increases, the hemodialyzer is relatively less
efficient and the concentration ratio increases. The increase is largely due to three
factors: the rapid decrease in membrane permeability with increasing molecular
weight for hemodialyzer membranes, the relatively high permeability of the
peritoneum to large solutes, and the longer treatment time with CAPD. Thus, for
metabolites which behave similar to inulin, four daily CAPD exchanges results in
a metabolite level one-seventh of that found on hemodialysis.
Another method with which to compare the various treatment modalities is to
quantitate them on the basis of solute clearances [76]. Representative clearance
values on a weekly basis are presented in Table 10 along with normal kidney
clearances. From this data, the superiority of hemodialysis in the removal of small
solutes is apparent. For urea or a similar-sized solute, hemodialysis can clear 107
liters of body fluids in a week while CAPD (4 exchanges per day) and IPD clear

Table 9. Maximum intracellular to extracellular metabolite disequilibrium ratio

Solute Mol. wt. CAPD/normal IPD/normal HD/normal

Urea 60 1.04 3.5 9.6

Vit. B-12 1.355 1.02 2.7 4.6
Inulin 5.500 1.00 1.3 1.5
150
0
0-
<:(
CAPO
0
::!:! 7 5 ex/day
::;
l<J 4 "
:I:
6 3 II

0 2 II

t
0::
5

U 4
z
0
u 3
(f)
en 2
~
<t:
0
w a
0::
a..
60.06 100 200 300 500 1000 2000 8000
M.W.

Figure 24. Hemodialysis to CAPO predialysis concentration ratio as a function of molecular weight
(reprinted with permission of Artificial Organs (Ref. 93)).

Table 10. Weekly solute clearances

Solute Normal Hemodialysis IPO CAPO exchanges/day

kidneys (15 hr/wk) (40hr/wk) 3 4 5

Urea 756 107 60 57 70 83

1355 dalton 1260 15 19 39 42 45
5500 dalton 1260 4 12 25 26 27

only 70 and 601/week, respectively. However, none of these procedures can

duplicate the removal rate of two normal kidneys, 7561/week. As solute size
increases, there is a drastic drop in the hemodialyzer clearance. The peritoneal
dialysis procedures are less strongly affected for the reasons noted above. For a
solute of 5500 daltons, the hemodialyzer clearance declines to 51/week, only
4.6% of the urea clearance. In contrast, the 5500 dalton clearances for CAPD (4
exchanges per day) and IPD of 26 and 121/week represent 37 and 20%, respec-
tively, of the urea clearance. Thus, either of the peritoneal dialysis procedures
will provide better large solute removal ability with CAPD being the clearly
superior technique. There can be little doubt that CAPD would be the treatment
of choice if the object of the treatment is the removal of large molecular weight
toxins.
 151

16. Transport limitations: the difference between CAPD and IPD

The BUN clearance of peritoneal dialysis is a complex function of three dominant

parameters [76]: the peritoneal blood flow rate, OB' the dialysate flow rate, 00'
and the mass transfer-area coefficient, MTAC. The clearance can never exceed
the lowest value of these three parameters. Also, the clearance will be approx-
imately equal to the value of one of the parameters if it is very much less than the
other two. For example, the clearance will be approximately equal to the MTAC
if the value of the blood and dialysate flow rates are very much larger than the
value of the MTAC. In this case the clearance is said to be mass transfer limited,
and increasing either the blood or dialysate flow rates will not significantly
increase the clearance. The clearance of a solute is said to be dialysate flow rate
limited if the value of the dialysate flow rate is much lower than the blood flow
rate and the MTAC. In this case the clearance will essentially equal the value of
dialysate flow rate.
These concepts are illustrated in Figure 25. The urea clearance is presented as a
function of dialysate flow rate for peritoneal dialysis. Intermittent peritoneal
dialysis is generally performed with dialysate flow rates of 30-40 mllmin, with a
urea clearance near 20 mllmin. Note that the urea clearance levels off at approx-
imately 35 mllmin (a value equal to the average MTAC) at high dialysate flow
rates. This demonstrates that the clearance of intermittent peritoneal dialysis is
limited to a considerable extent by a low mass transferarea coefficient. This
assumes that under ordinary circumstances the peritoneum capillary blood flow

60
- HEMODIALYSIS
(RegIon of Flow Rate LImItatIon)
50
c
E
~ 40
E o

w
u o
z 30 MASS TRANSFER

~
<[
a: LIMITED REGION
<[
w
--' PERITONEAL DIALYSIS
u 20 o
<[
w
a:
:J

FLOW LIMITED REGION

o 40 80 120 160 200 240 280 320

DIALYSATE FLOW RATE (ml/mln)

Figure 25. Urea clearances as a function of dialysate flow rate.

152

rate is sufficiently high to preclude it becoming a limiting parameter. This

assumption is supported by inferential evidence presented by Nolph et al. [96].
With the typical IPD dialysate flow rates of 30 to 40 ml/min, the urea clearance
is approximately 55% of the maximum possible clearance for a patient with
average peritoneal mass transfer ability. If a patient has impaired transport
ability, the maximum clearance will be decreased proportionately. This effect can
have serious consequences in an intermittent technique such as IPD where
efficiency must be maximized to provide adequate therapy. As shown in Figure
22, IPD is the least effective technique in terms of urea removal for the average
patient. Any loss or intrinsic lack of peritoneal transport capability will make IPD
less efficient, and will result in higher metabolite concentration.
There is known to be substantial interpatient variability in peritoneal mass
transfer-area coefficient. Some patients' MTAC's have been found to be about
20% of the average value [92]. Due to the low clearances found even under the
best circumstances, these patients are poor candidates for IPD. If treated with
IPD, they may well exhibit symptoms of under-dialysis of develop other serious
complications. Indeed, this may explain the poor survivability of some IPD
patients [95]. These problems can generally be overcome by performing longer
treatments to compensate for any transport limitations. However, longer IPD
treatments present practical difficulties. CAPD overcomes the problem of low
efficiency inherent in peritoneal dialysis by making continuous operation an
integral part of the procedure.
Since CAPD operates at much lower dialysate flow rates (7 to 9 mllmin) than
IPD, CAPD becomes a flow rate limited system with a maximum clearance of no
more than 9 ml/min. Thus, on first examination, it would seem that CAPD might
be an even poorer therapy than IPD. However, CAPD is performed continuously
(168h/week) as opposed to only about 40h/week for IPD. CAPD provides a
much higher weekly solute clearance for all solutes because of this longer treat-
ment time, as shown in Table 10. This shows that CAPD is, in fact, a much more
efficacious means of maintaining solute concentration control than IPD.
There is also a hidden benefit in the low clearance of CAPD. Since the dialysate
flow rate is much less than the average mass transfer-area coefficient, the flow
rate is the dominant factor in determining solute clearances. That is, the mass
transfer area coefficient would have to drop to less than half of the average value
before membrane properties could exert a significant effect on clearances. Thus,
only those patients with servere loss or lack of transport ability will have de-
creased CAPD clearances with the concomitant problems. In many of these
cases, adding another CAPD exchange each day can compensate for the decrease
in clearance in these patients. With IPD, even a slight decrease in transport ability
may significantly affect clearances.
These results suggest that IPD should be employed with caution due to the
possible consequences of under-dialysis. Probably those patients with known
membrane-transport problems should not be considered for chronic IPD unless
 153

there is a significant amount of residual renal function to assist in solute clearance.

Most patients thus exluded from IPD can be successfully maintained by CAPO
with or without residual renal function.
Patients being considered as peritoneal dialysis candidates may be tested for
peritoneal membrane transport properties before being placed on either IPD or
CAPO by employing the simple clinical protocol of Popovich [76] outlined above.
Those patients with average or greater than average transport ability will proba-
bly be acceptable candidates for either IPD or CAPO. For those in which any
significant impairment is noted, CAPO may prove generally preferable to IPD.
Finally, these results demonstrate that IPD and CAPO are fundamentally
different procedures in terms of transport limitations with resulting dissimilarities
in their abilities to maintain biochemical control. These differences suggest that
IPD clinical results should not be employed in attempting to make comparison
between CAPO and hemodialysis.
Numerous recent publications concerning the kinetics of peritoneal transport
may be consulted for additional information [97-116].

References

1. Ganter G: Uber die Beseitigung giftiger Stoffe aus dem Blut durch Dialyse. Munch Med
Wochschr 70: 1478-1480, 1923. (from Boen ST) Kinetics of Peritoneal Dialysis. Medicine 40:
243-287, 1961.
2. Cunningham RS: The physiology of the Serous Membranes. Physiol Rev 6: 242-280, 1926.
3. Clark AJ: Absorption from the Peritoneal Cavity. J Pharm Exp Ther 16: 415-433, 1921.
4. Schechter AJ, Cary MK, Carpentier AL, Darrow DC: Changes in Composition of Fluids
Injected into the Peritoneal Cavity. Am J Dis Child, 46: 1015-1026,1933.
5. Bliss S, Kastler AO, Nadler SB: Peritoneal Lavage. Effective Elimination of Nitrogenous
Wastes in the Absence of Kidney Function. Proc Soc Exp BioI Med 29: 1078-1079, 1932.
6. Odel HM, Ferris DO, Power MH: Peritoneal Lavage as an Effective Means of Extrarenal
excretion. Am J Med 9: 63-77, 1950.
7. Grollman A, Turner LB, McLean JA: Intermittent Peritoneal Lavage in Nephrectomized Dogs
and Its Application to the Human Being. Arch Intern Med 87: 379-389, 1951.
8. Tenckhoff HA, Schechter H: A Bacteriologically Safe Peritoneal Access Device. Trans Am Soc
Artif Intern Organs 14: 181-187, 1968.
9. Feldman W, Beliah T, Drummond KN: Intermittent Peritoneal Dialysis in the Management of
Chronic Renal Failure in Children. Am J Dis Child 116: 30-36, 1968.
10. Tenckhoff HA, Blagg CR, Curtis KF, Hickman RO: Chronic Peritoneal Dialysis. Eur Dial
Transpl Assoc 10: 363-370, 1973.
11. Flanigan WJ, Henderson LW, Merrill JP: The Clinical Application and Technique of Peritoneal
Dialysis. Gen Pract 28: 98-109.
12. Mattocks AM, El-Bassiouni EA: Peritoneal Dialysis. A Review J Ph arm Sci 60: 1767-1782,
1971.
13. Jones JH: Peritoneal Dialysis. Br Med Bull 27: 165-169, 1971.
14. Vidt DG: Intermittent Peritoneal Dialysis. Ohio State Med J 64: 1149-1153, 1968.
15. Miller JH, Gipstein R, Margules R, Swartz M, Rubin ME: Automated Peritoneal Dialysis:
Analysis of Several Methods of Peritoneal Dialysis. Trans Am Soc Artif Intern Organs 12:
98-105, 1966.
154

16. Maxwell MW, Rockney RE, Kleeman CR, Twiss MR: Peritoneal Dialysis. 1. Technique and
Application, JAMA 170: 917-924,1959.
17. Boen ST: Kinetics of Peritoneal Dialysis: A Comparison with the Artificial Kidney Medicine 40:
243-287,1961.
18. Gray H: Anatomy ofthe Human Body, 27th ed. C Mayo (ed), Lea and Febiger, Philadelphia, pp
1253-1272, 1959.
19. Copenhauer WM, Johnson DD (1958) Bailey's Textbook of Histology, 14th ed. Williams and
Wilkins, Baltimore, 1958.
20. Grant JCB: An Atlas of Anatomy, 4th eds. Williams and Wilkins, Baltimore, 1956.
21. Baron MA: Structure of the Intestinal Peritoneum in Man. Am J Anat 69: 439-496, 1941.
22. Wayland H: Transmural and Interstitial Molecular Transport. Proc Int'!. CAPD Symp. Paris,
1979. Excerpta Medica, pp 18--27, 1980.
23. Allen L, Weatherford T: Role of Fenestrated Basement Membrane in Lymphatic Absorption
from Peritoneal Cavity. Am J Physiol197: 551-554, 1959.
24. Karnovsky MJ: The Ultrastructural Basis of Capillary Permeability Studied with Perioxides as a
Tracer. J Cell Bioi 35: 213-235, 1967.
25. Lieb WR, Stein WD: Biological Membranes Behave as Non-Porous Polymeric Sheets with
Respect to the Diffusion of Non-Electrolytes. Nature 224:240-243, 1969.
26. Nolph KD, Hano JE, Teschan PE: Peritoneal Sodium Transport During Hypertonic Peritoneal
Dialysis: Physiologic Mechanisms and Clinical Implications. Ann Intern Med 70: 931-941,1969.
27. Stolz ML, Nolph KD, Maher JF: Factors affecting Calcium Removal With Calcium Free
Peritoneal Dialysis. J Lab Clin Med 78: 389-398, 1971.
28. Henderson LW: Peritoneal Ultrafiltration Dialysis: Enhanced Urea Transfer Using Hypertonic
Peritoneal Dialysis Fluid. J Clin Invest 45: 950-955, 1966.
29. Nolph KD, Rosenfeld PS, Powell JT, Danforth E: Peritoneal Glucose Transport and Hyper-
glycemia During Peritoneal Dialysis. Am J Med Sci 259: 272-281, 1970.
30. Henderson LW, Nolph KD: Altered Permeability of Peritoneal Membrane After Using Hyper-
tonic Peritoneal Dialysis Fluid. J Clin Invest 48: 992-1001, 1976.
31. Raja RM, Cantor RE, Boreyko C, Bushehri H, Kramer MS, RosenbaumJL: Sodium Transport
During Ultrafiltration Peritoneal Dialysis. Trans Am Soc Artif Intern Organs 18: 429-435, 1972.
32. Kallen RJ: A Method for Approximating the Efficacy of Peritoneal Dialysis for Uremia. Am J
Dis Child III: 156-160, 1966.
33. Esperanca MJ, Collins DL: Peritoneal Dialysis Efficiency in Relation to Body Weight. J Pediat
Surg 0: 162-169, 1966.
34. Gosselin RE, Berndt WO: Diffusional Transport of Solutes Through Mesentery and Per-
itoneum. J Theor Bioi 3: 487-495, 1962.
35. Goldschmidt ZH, Pote HH, Katz MA, Shear L: Effect of Dialysate Volume on Peritoneal
Dialysis Kinetics. Kidney Int, 5: 240-245,1974.
36. Frank HA, Seligman AM, Fine JJ: Treatment of Uremia After Acute Renal Failure by
Peritoneal Irrigation. JAm Med Assoc 130: 703--705.
37. Seligman AM, Frank HA, Fine JJ: Treatment of Experimental Uremia by Means of Peritoneal
Irrigation. J Clin Invest 25: 211-2??, 1946.
38. Penzotti SC, Mattocks AM: Effects of Dwell Time, Volume of Dialysis Fluid, and Added
Accelerators on Peritoneal Dialysis of Urea. J Pharm Sci 60: 1520-1522,1971.
39. Pirpasopoulos M, Lindsay RM, Rahman M, Kennedy AC: A Cost-Effectiveness Study of Dwell
Times in Peritoneal Dialysis. Lancet 2: 1135-1136, 1972.
40. Gross M, McDonald HP: Effect of Dialysate Temperature and Flow Rate on Peritoneal
Clearance. JAM A 202: 363-365, 1967.
41. Shinaberger JH, Shear L, Clayton LE, Barry FG, Knowlton M, Goldbaum LR: Dialysis for
Intoxications with Lipid Soluble Drugs: Enhancement of Glutathimide Extraction with Lipid
Dialysate. Trans Am Soc Artif Intern Organs 11: 173-177,1965.
 155

42. Shinaberger JH, Shear L, Barry KG: Pertioneal-Extracorporeal Recirculation Dialysis: A

Technique for Improving Efficiency of Peritoneal Dialysis. Invest Uro12: 555-560, 1965.
43. Rosenbaum JI, Mandanas R: Treatment of Phenobarbitol Intoxication in Dogs with Anion-
Recirculation Peritoneal Dialysis Technique. Trans Soc Artif Intern Organs 13: 183-189, 1967.
44. Kablitz C, Stephen RL, Duffy DP, Jacobsen SC, Zelman A, Kolff WJ: Technological Augman-
tation of Peritoneal Urea Clearance: Past, Present, and Future. Dial & TranspI9(8): 741-744,
1980.
45. Popovich RP, Pyle WK, Moncrief JW, Decherd JF, Brooks S: Preliminary Verification of the
Low Dialysis Clearance Hypothesis Via a Novel Equilibrium Peritoneal Dialysis Technique.
Proc 2nd Austral Conf Heat Mass Transfer 2: 217-223,1977.
46. Popovich RP, Moncrief JW, Decherd JF, Bomar JB, Pyle WK: The Definition of a Novel
PortablelWearable Equilibrium Peritoneal Dialysis Technique. Abst Am Soc Artif Intern
Organs 5: 64, 1976.
47. Popovich RP, Moncrief JW, Nolph KD, Ghods AJ, Twardowski ZJ, Pyle WK: Continuous
Ambulatory Peritoneal Dialysis. Ann Intern Med 88: 449-456, 1978.
48. Villarroel F: Kinetics of Intermittent and Continuous Peritoneal Dialysis. J Dial (4): 333-347,
1977.
49. Rescigno A, Segre G: Drug and Tracer Kinetics, Blaisdell, Waltham, Mass, 1966.
50. Mattocks AM: Accelerated Removal of Salicylate by Additive in Peritoneal Dialysis Fluid. J
Pharm Sci 58: 595-598, 1969.
51. Popovich RP, Moncrief JW, Okutan M, Decherd JF: A Model ofthe Peritoneal Dialysis System.
Proc 25th Ann Conf on Engr in Med And Bioi 14: 172, 1966.
52. Bomar JB, Decherd JF, [Link] DJ, Moncrief JW, Popovich RP: The Elucidation of Max-
imum Efficiency Minimum Cost Peritoneal Dialysis Protocols. Trans Am Soc Artif Intern
Organs 20: 120-129,1974.
53. Pyle WK, Popovich RP, Moncrief JW: Mass Transfer Evaluation in Peritoneal Dialysis. In
CAPD UPDATE. JW Moncrief, R.P. Popovich eds. Masson, N.Y., 1981. pp 35-52.
54. Pyle WK: Mass Transfer in Peritoneal Dialysis. Ph.D. Dissertation, Univ. of Texas, 1981.
55. Popovich RP, Cristopher TG, Babb AL: The Effects of Membrane Diffusion and Ultrafiltration
Properties on Hemodialyzer Design and Performance. Chern Eng Prog Symp Ser 67(114): 105-
115, 1971.
56. Babb AL, Johansen PJ, Strand MJ, Tenckhoff H, Schribner BH: Bi-directional Permeability of
the Human Peritoneum to Middle Molecules. Proc 10th Cong Europ Dial Transpl Assoc, Vienna
10: 247-262, 1973.
57. Popovich RP, Moncrief JW: Clinical Development of the Low Dialysis Clearance Hypothesis
Via Equilibrium Peritoneal Dialysis. 1st Ann Rep No. NOl-AM-6-2211, AK-CUP, NIAMDD,
NIH, Bethesda, MD, 1977.
58. Bomar JB: The Transport of Uremic Metabolites in Peritoneal Dialysis. Ph.D. Dissertation,
Univ of Texas.
59. Kedem 0, Katchalsky A: Thermodynamic Analysis of the Permeability of Biological Mem-
branes to Non-Electrolytes. Biochem. Biophys Acta 27: 229-246, 1958.
60. Katchalsky A, Curran PF: Nonequilibrium Thermodynamics in Biophysics. Harvard University
Press, Cambridge, Mass, 1967.
61. Popovich RP, Pyle WK, Moncrief JW, Bomar JB: Peritoneal Dialysis. Chronic Replacement of
Kidney Function. A1ChE Symp. Series, 75(187): 31-45, 1979.
62. Villarroel F, Klein E, Holland F: Solute Flux in Hemodialysis and Hemofiltration Membranes.
Trans Am Soc Artif Intern Organs 23: 225-233, 1977.
63. Anderson JL, Quinn J A: Restricted Transport in Small Pores: A Model for Steric Exclusion and
Hindered Particle Motion. Biophys J 14: 130-150, 1974.
64. Brenner J, Gaydos LJ: The Constrained Brownian Movement of Spherical Particles in Cylindri-
cal Pores of Comparable Radius: Models of the Diffusive and Convective Transport of Solute
Molecules in Membranes and Porous Media. J Colloid Interface Sci 58: 312-356,1977.
156

65. Bird RB, Stewart WE, Lightfoot EN: Transport Phenomena, John Wiley and Sons, New York,
1960.
66. Conte SD, de Boor C: Elementary Numerical Analysis, 2nd ed., McGraw-Hill, New York,
1965.
67. Brown KM: Computer-Oriented Methods for Fitting Tabular Data in the Least Squares Sense.
Fall Joint Computer Conf., Nat!. Center for Atmospheric Res., Boulder, Colo, 1972.
68. Rubin J, Nolph KD, Popovich RP, Moncrief JW, Prowart B: Drainage Volumes During
Continuous Ambulatory Peritoneal Dialysis. J Am Soc Artif Intern Organs 2(2): 54-60,1979.
69. Nolph KD, Sorkin MJ: A Hypothesis to Explain the Ultrafiltration Characteristics of Peritoneal
Dialysis. Kidney Int. (in press).
70. Popovich RP, Moncrief JW, Nolph KD, Pyle WK, Sawyer JW: Physiological Transport Param-
eters in Peritoneal and Hemodialysis. 3rdAnn Rep No. NOl-AM-3-2205, AK-CUP, N1AMDD,
NIH, Bethesda, Md, 1977.
71. Kjellstrand CM, Rosa AA, Shideman JR, Rodrigo F, Davin T, Lynch RE: Optimal Dialysis
Frequency and Duration: The 'Unphysiology Hypothesis'. Kidney Int 13 (supp!. 8): S-120-S-124,
1978.
72. Arieff AJ, Guisado R, Massry SG: Uremic Encephalopathy: Studies on Biochemical Altera-
tions in the Brian. Kidney Int 7 (Suppl): S-194-S-200, 1975.
73. Popovich RP, Hlavinka DJ, Bomar JB, Moncrief JW, Decherd JF: The Consequences of
Physiological Resistances on Metabolite Removal from the Patient-Artifical Kidney System.
Trans Am Soc Artif Intern Organs 21: 108-115, 1975.
74. Gotch FA, Sargent JA, Keen M, Lam M, Prowitt M, Grady M: Solute Kinetics in Intermittent
Dialysis Therapy. 9th Ann Rep - Contractors Conf, p 98-101,1976.
75. Popovich RP, Pyle WK, Hiatt MP, McCollough WS, Moncrief JW: Metabolite Transport
Kinetics in Peritoneal Dialysis. Proc Int CAPD Symp, Paris, Nov 2-3, 1979. Excerpta Medica,
pp 28-33, 1980.
76. Popovich RP, Hiatt MP, Moncrief JW, Pyle WK: Mathematical in Modeling and Minimum
Treatment Requirements Peritoneal Dialysis. Proc 3rd Capri Conf on chronic Uremia, 1980 (in
press).
77. Moncrief JW: Personal communication, Aug. 28, 1980.
78. Randerson DH, Farell PC: Assessment of Mass Transfer Properties ofthe Peritoneal Membrane
During Peritoneal Dialysis. J Am Soc Artif Intern Organs, 1980 (in press).
79. Moncrief JW, Popovich RP, Nolph KD, Rubin J, Robson M, Dombros N, de Veber G,
Oreopoulous DG: Clinical Experience with Continuous Ambulatory Peritoneal Dialysis. JAm
Soc Artif Intern Organs 2(3): 114-118, 1979.
80. Moncrief JW, Popovich RP: Peritoneal Dialysis for a Greater Number of Patients. Chap. In:
Controversies in Nephrology, Schreiner GE, ed, Georgetown University Press, Washington
D.C.,1979.
81. Moncrief JW, Popovich RP: Continuous Ambulatory Peritoneal Dialysis. Contrib Nephrol17:
139-145, 1979 (Karger, Basel).
82. Balfe JW, Irwin MA, Oreopoulos DG: An Assessment of Continuous Ambulatory Peritonean
Dialysis in Children. Proc CAPD Int Symp II, May 9-10,1980, Austin, Tx (in press).
83. Shmerling J, Kohaut E, Perry S: Cost and Social Benefits of CAPD in a Pediatric Population.
Proc. CAPD Int. Symp. II, May 9-10, 1980, Austin, Tx (in press).
84. Kohaut EC, Alexander S: Ultrafiltration in the Young Patient on CAPD. Proc. CAPD Int.
Symp. II, May 9-10, 1980, Austin, Tx (in press).
85. Popovich RP, Pyle WK, Rosenthal DA, Alexander S, Balfe JW, Moncrief JW: Kinetics of
Peritoneal Dialysis in Children. Proc CAPD Int Symp II, May 9-10,1980, Austin, Tx (in press).
86. Moncrief JW, Pyle WK, Simon P, Popovich RP: Hypertriglycerdemia, Diabetes Mellitus, and
Insulin Administration in Patients Undergoing Continuous Ambulatory Peritoneal Dialysis.
Proc. CAPD Int. Symp. II, May 9-10, Austin, Tx (in press).
 157

87. Arant BS, Edelman CM, Spitzer A: The Congruence of Creatinine and Inulin Clearances in
Children - Use of the Technicon Auto Analyzer. J pediat 81(3): 559-561,1972.
88. Clark LC, Thompson HL, Beek EI, Jacobson W: Excretion of Creatine and Creatinine by
Children. Am J Dis Child 81: 774-783, 1951.
89. Shull BC, Haughey D, Koup J, Baliah T, Li PK: A Useful Method for Predicting Creatinine
Clearance in Children. Clin Chern 24(7): 1167-1169, 1979.
90. Popovich RP, Moncrief JW: Kinetic Modeling of Peritoneal Transport. Contrib Nephrol17:
59-72. (Karger, Basel).
91. Babb AL, Popovich RP, Christopher TG, Scribner BH: The Genesis of the Square Meter-Hour
Hypothesis. Trans Am Soc Artif Intern Organs 17: 81-91,1971.
92. Popovich RP, Pyle WK, Hiatt MP, Moncrief JW: Comparative Kinetic Studies of Dialysis. Proc.
Northeastern Physicians Conf., New York, Oct. 20, 1979 (in press).
93. Hiatt MP, Pyle WK, Moncrief JW, Popovich RP: A Comparison of the Relative Efficacy of
CAPD and Hemodialysis in the Control of Solute Concentration. Artif Organs 4(1): 37-43,
1980.
94. Kjellstrand CM, Evans RL, Peterson RJ, Shideman JR, von Hartitzsch B, Buselman TJ: The
'Unphysiology' of Dialysis: A Major Cause of Dialysis Side Effects? Kidney Int 7: 530-534,
1975.
95. Kjellstrand CM et al.: Proc 3rd Capri Conf on Uremia, 1980 (in press).
96. Nolph KD, Miller F, Rubin J, Popovich RP, New Directions in Peritoneal Dialysis Concepts and
Applications Kidney Int 18(S-10): S-111-S-116.
97. Feriani M, Biasioli S, Chiaramonte A et al.: Anatomical bases of peritoneal permeability: a
reappraisal. Int J Artif Organs 5: 345-348,1982.
98. Nolph KD: Solute and water transport during peritoneal dialysis. Perspec Perit Dial 1: 4-8, 1983.
99. Garini G, Tagliavini D, Occhialini L: Mechanisms of transport and kinetics of the solutes in
peritoneal dialysis. Recent Prog Med 73: 569-580, 1982.
100. Rubin J, Klein E, Bower JD: Investigation of the net sieving coefficient of the peritoneal
membrane during peritoneal dialysis. ASAIO J 5: 9-15, 1982.
101. Indraprasit S, Namwongprom A, Sooksriwongse C et al.: Effect of dialysate temperature on
peritoneal clearances. Nephron 34: 45-47, 1983.
102. Maher J: Transport kinetics in peritoneal dialysis. Perit Dial Bull 3 (suppl): 4-6, 1983.
103. Nath IV, Sehgal S, Chugh KS et al. : Loss of immunoglobulins during peritoneal dialysis. J Assoc
Physicians India 29: 927-929,1982.
104. Ku K, Anderson R, Shoenfeld P: Kinetic modeling of urea in peritoneal dialysis. Dial Transpl12:
374-381, 1983.
105. Oreopoulos DG: Criteria for adequacy of peritoneal dialysis. Perit Dial Bull 3: 1-2, 1983.
106. Diaz-Buxo JA, Farmer CD, Walker PJ et al.: Effects of hyperparathyroidism on peritoneal
clearances. Trans Am Soc Artif Intern Organs 28: 276-279, 1982.
107. Grzegorzewska A, Baczyk K: Furosemide-induced increase in urinary and peritoneal excretion
of uric acid during peritoneal dialysis in patients with chronic uremia. Artif Organs 6: 220-224,
1982.
108. Hall K, Meatherall B, Krahn J et al.: Clearance of quinidine during peritoneal dialysis. Am
Heart J 104: 646-647, 1982.
109. Rubin J, Adair C, Barnes T et al.: Augmentation of peritoneal clearance by dipyridamole.
Kidney Int 22: 658-661, 1982.
110. Lang HL, Nolph KD, McGary TJ: Enhancement pf clearances by activated charcoal in an in
vitro model of peritoneal dialysis. Clin Exp Dial Apheresis 6: 85-95,1982.
111. Ratnu KS, Haldia KR, Panicker S et al.: A new technique - semicontinuous rapid flow, high
volume exchange - for effective peritoneal dialysis in shorter periods. Nephron 31: 159-164,
1982.
112. Twardowski ZJ, Prowant BF, Nolph KD et al.: High volume, low frequency continuous
ambulatory peritoneal dialysis. Kidney Int 23: 64-70,1983.
158

113. Lopot F: Ultrafiltration characteristics of peritoneal dialysis. Vnitr Lek 29: 230-237, 1983.
114. Manuel A: Failure of ultrafiltration in patients on CAPD. Perit Dial Bull 3 (Suppl): 38-41, 1983.
115. Slingeneyer A, Canaud B, Mion C: Permanent loss of ultrafiltration capacity of the peritoneum
in long-term peritoneal dialysis: an epidemiological study. Nephron 33: 133-138, 1983.
116. Kraus MA, Shasha SM, Nemas M et al.: Ultrafiltration peritoneal dialysis and recirculating
peritoneal dialysis with a portable kidney. Dial Transpl12: 385-388, 1983.
6. Ultrafiltration with peritoneal dialysis

LEE HENDERSON

1. Introduction

There is a clinical requirement to remove excess body water and its attendant
electrolytes on a regular basis from patients with end-stage renal failure. For
patients treated with peritoneal dialysis this is accomplished osmotically rather
than hydrostatically as is the common practice with hemodialysis [1-3]. To date,
glucose is the only agent that has been accepted clinically for regulating the
osmolality of peritoneal dialysis fluid for this purpose. The problem of overload-
ing the patient with carbohydrate calories is significant in patients under treat-
ment with CAPD. As a result other osmotically active agents will undoubtedly be
tested and applied clinically in the next few years. The present discussion will
focus on glucose and its use as a driving force for ultrafiltration across the
peritoneal membrane with the expectation that most of the principles developed
for this solute will apply equally to other solutes that may be utilized.

2. Description of the system

For water to move across a semipermeable membrane such as the peritoneum

there must be a driving force. This driving force is usually described in terms of a
transmembrane pressure gradient. For the present discussion units assigned to
this gradient will be either in millimeters of mercury of milliosmoles. An osmole
by definition is simply 6.02 x 1023 osmotically active solute particles present in one
kilogram of water. For a substance that does not dissociate in water solution, e.g.,
glucose, 1 osmole is equivalent to 1 mole of solute. One mOsm present on one side
of a membrane that is perfectly impermeable for that solute will exert enough
force to support a 19mm column of mercury. For a dissociable solute such as
sodium chloride there is the potential for 1 mole = 2 osmols as in dilute solution it
will dissociate into its constituent ions each of which is an osmotically active solute
particle. Dissociation may not be complete. For solutes that dissociate a factor
160

designated the osmotic coefficient may be calculated or (more usually) measured

to equate the ideal and the observed osmolality for a dissociable solute. For NaCI
the predominant electrolyte of peritoneal dialysis solution, we can assume rea-
sonably complete (93%) dissociation of the electrolyte present.
The 'anatomy' of the peritoneal membrane is described in detail in Chapter 2. I
would like to begin, however, by considering the peritoneal 'membrane' in a
'black box' manner, i.e., the domain that separates the bulk phase of plasma
water from the bulk phase of well-mixed dialysis fluid. This then comprises both
the anatomical structures such as the vascular endothelium, mesothelium, etc.,
and in addition, all of their attendant unstirred layers.
The peritoneal membrane is not perfectly semipermeable (i.e., is somewhat
leaky) for the solutes to be considered making our task of quantitating the driving
force for ultrafiltration more complicated. In particular, glucose moves across the
peritoneal membrane at a rate consonant with its rather small molecular weight
(180 daltons). Even protein molecules such as albumin (68,000 daltons) move
across the peritoneal membrane to some small extent. If the membrane was
impermeable for all solutes and easily permeable for water, then one would need
only to measure the 'blood to bath' concentration gradients across the membrane
for each of the solutes present on both sides of the membrane and then convert
each of these concentration gradient terms into mm Hg (L11T). Their sum would be
the computed osmotic transmembrane pressure gradient that would drive ultra-
filtration. Further, if the area (A) of the membrane was known and its hydraulic
permeability (Lp) it would be possible to compute the rate of ultrafiltration (Of) in
terms of mllmin for the clinical circumstance with which you were dealing.
As the peritoneal membrane is not ideally semipermeable for the solutes
present in dialysis fluid, a correction factor (0 = Staverman's reflection coeffi-
cient) for each solute concentration gradient (L11T) must be applied to adjust that
gradient for the degree of the membrane's 'semipermeability' for that solute.
Written more formally, these concepts could be set down as follows:

where L11Tj O j would equal the osmotic driving gradient for sodium, L11TP2 would
equal that for potassium, etc. until each of the solutes present on each side of the
membrane was represented. (Solutes that were nearly equal or equal in concen-
tration on both sides of the membrane would, of course, contribute little or no
driving force for ultrafiltration. Solutes such as glucose, urea and protein, for
example, which by therapeutic intent or biological circumstance will have signifi-
cant concentration gradients across the membrane may contribute significant
osmotic driving force.)
To understand just how much or even relatively how much force for ultrafiltra-
tion would be contributed, for example, when glucose is contrasted with protein
°
or urea requires further exploration of the term and an appreciation of the fact
 161

that the larger a solute is in terms of its molecular size the less likely it is to be
present in biologic solutions at a concentration that contributes much to the total
osmolality.
Albumin, for example, has a reasonably high a (a ranges from 0 to 1). That is to
say albumin moves through the membrane with difficulty. If it is present at a
concentration of 3.4 gldl. Taking the molecular weight of 68000 daltons and
assuming a = 1, the osmolar contribution of albumin to resisting movement of
plasma water into the peritoneal space is:

34g/kg H 20 _ _ _
68000 /k H 0 - 0.0005 osmoles - 0.5 mosmole - 9 mm Hg
g g 2

There is an additional 3-4 mmHg added to this figure that is contributed by the
unequal distribution of electrolytes in an ionized colloidal system such as plasma
(Gibbs-Donnan phenomenon). The total contribution of 12-13 mmHg must be
considered small.
Whereas glucose at a concentration of 1.5 or 4.25 g/dl in the dialysis fluid and
100 mg/dl in plasma would contribute a maximum potential osmolar driving
gradient for ultrafiltration of either 78 mOsm (1482 mm Hg) or 231 mOsm
(4381mm Hg). This maximum is never manifest, however, because of the rela-
tively low afor glucose across the peritoneal membrane. Furthermore, with time,
the concentration gradient deteriorates. There are at least two components to this
discharge of the gradient. The first is diffusion of glucose into the plasma and the
second is convective (also termed Poisseuillian, plug or bulk flow) water move-
ment countercurrent to the glucose. Clinically, the net movement of water is
obtained as the difference between inflow and outflow volumes. As such, it
represents an average value. Figure 1 shows the relationship of water flow rate
and dialysate volume with time for a 4.25% glucose-containing solution [1, 4].
Changes in flow rate with time as expected are exponential in character. Figures 2
and 3 show the relationship of osmolar gradient and glucose concentration with
time for 1.5 and 4.25% glucose containing dialysis fluid [1, 4].
The osmotic force contributed by urea (u) with a 60 dalton molecular weight
may be arrived at in a manner similar to that for glucose (g). For a plasma urea
concentration of 300 mg/dl (BUN of approximately 120 mg/dl) it offers a 50
mOsm (950 mm Hg) maximum potential osmolar driving gradient for water
movement. Clinical wisdom identifies that 'isotonic' (1.5% glucose-containing)
dialysis solutions used in patients with a blood urea nitrogen concentration of
150 ± 20 mg/dl usually result in little or no net removal of excess total body water.
In this circumstance for average values of 7T for a 60-min exchange time then
a u7Tu=ag7Tg where starting values for 7Tu = 950 and 7Tg = 1482. Therefore, au <ag and
ajag = about 0.6. 2 At present there is very limited information about a for the
peritoneal membrane and the various solutes that are present in the peritoneal
dialysis system (i.e. soluble constituents of uremic plasma water and of dialysis
162
3400

E 3200
W
:!!
=> 3000
...J
51
..."'
<l
2800
V>
>-
...J
<l
0

40 80 120 160 200 240 280 320 360 400 440 480

o 40 80 120 160 200 240 280 320 360 400 440 480
DWELL TIME (Monute,)

Figure 1. Dialysate volume and the rate of ultrafiltration are plotted vs dwell time for a 4.25% dextrose
containing solution.

500

450

~ 400
iii 1 5 % Dextrose
~ 350 SolutiOns
>-
>-
::; 300 i .....m
:30
:=; 250
'"0 200

Of,
120 180 240 300 360 420 480 540 600

'5
"-
.§ 3000

t20 t80 240 300 360 420 480 540 600

DWELL TI ME (MlOufes)

Figure 2. Change in glucose concentration and osmolality of 1.5% glucose containing solution plotted
vs. dwell time. Mean plasma osmolality of 300 mOsm/kg would be expected for a patient with
comparatively good control of BUN as occurs with CAPD. Note that osmolar equilibrium is achieved
prior to the glucose concentration falling to a value comparable to normal plasma glucose concentra-
tion. This reflects both dilution of other osmotically active constituents of dialysis fluid by ultrafiltered
plasma water as well as elevation of the plasma glucose and dialysis fluid urea concentrations.
 163

500

425 % Dextrose
Solutions

eN'LO~ 120 180 240 300 360 420 480 540 600

r
~
3500

~:~
~ 2000
8 1500
•

~ 1000 • 425% Dlolysote (n:4)

u
~ 500
°t'-'-0---'60-120~1~80-2~40-3~00~360~4~20-4-'-80---'54-0~600
INFLOW
DWELL TI ME (Mmules)

Figure 3. Change in glucose concentration and osmolality of 4.25% glucose containing dialysis fluid
plotted vs dwell time.

fluid) [4]. Direct measurement of the peritoneal membrane characteristic a for a

given solute is not presently possible. Even indirect or relative measurements
such as described above are not easy and subject to criticism. What is usually
measured clinically, however, is the solute sieving coefficient for the peritoneal
membrane. Table 1 lists these values for a variety of relevant solutes [4-7,8]. The
sieving coefficient (S) is the ratio of the concentration of a solute (e.g., urea) in
the plasma water to that in the ultrafiltrate.

S = Cf = urea concentration in the ultrafiltrate

Cw urea concentration in plasma water

As such it will have a maximum value of 1.0 (no membrane sieving) and a
minimum value of 0.0 (total sieving or complete restraint by the membrane). The
sieving coefficient for a solute (Su) then must be closely related to au'

Where E may be referred to as the rejection coefficient for urea. E is arrived at by

measuring S during clinical application of peritoneal dialysis whereas a is not.
There is a considerable body of theoretical information about a available for
synthetic and biological membranes in the literature of irreversible thermo-
dynamics that will not be included in the present chapter. However, an under-
164

Table 1. Sieving and reflection coefficients measured for the human peritoneal membrane

Solute Daltons Hydrated S±l SEM 1- a (range) Ref.

radius (A)

Urea 60 2.7 0.81 ± 0.03 5

0.63 ±0.06 8
0.82 (0.91 to 0.73) 4
Creatinine 113 0.57 ± 0.09 0.67 (0.83 to 0.51) 8,4
Uric acid 168 0.63 (0.83 to 0.54) 4
Chloride 35 3.86 0.78 ±0.Q2 6
Potassium 37 3.96 0.36±0.Q2 7
0.40 ± 0.04 8
Glucose 180 4.4 0.62 (0.82 to 0.57) 4
Sodium 23 5.12 0.54±0.2 6
0.56 ± 0.04 8
Inulin 5200 12.0 0.83 ±0.04 5
0.41 ± 0.08 8
0.63 (0.91 to 0.30) 4
Albumin 68000 35.5 <0.02
Plasma protein 340000 0.01 4

a See text for explanation.

standing of how E and a differ will provide a satisfactory theoretical background

with which the forces at work when ultrafiltration occurs may be understood.
In simple systems where a single solute in water is ultrafiltered by a membrane
of known dimensions and permeability (Pm) for the solute the following relation-
ship between E and a pertains [9, 10].
J
IfCf =j=(l-E)Cw
f

where Js = the solute flux rate in mols/min cm2

J f = the water flux rate in ml/min cm2
and
EP-1
E=--
EP-a

and

where J f is large E~a.

when J f is small E~Jp/Pm"
 165

Using a best guess estimate for the area of membrane, participating in the process
(we will take a small but defensible [3] area of 0.5 m2 in order to err if at all on the
high side when estimating the ultrafiltration flux rate for the peritoneal mem-
brane), one would judge the J f for peritoneal membrane to be 'small'. With
reference to Figure 1, the maximum ultrafiltrate rate measured at the onset of a
4.25% exchange when the glucose gradient is greatest ranges from 12-16ml/min
[1,4]. For an estimated membrane area of 0.5 m2 , J f computes to = 6 X 1O-5 ml/
min/cm2 • Contrast this with hemodialysis where a 1.0m2 hollow fiber
Cuprophan® dialyzer may be expected to remove 5 liters of excess total body
water over a 4-hour treatment period; J f = 2 X 10-3 ml/min/cm2 • From the work of
Villaroel this value for Cuprophan is considered 'small'. We may therefore expect
that for a given solute moving convectively across the peritoneum that e would
= Jp/Pm •
For the complicated system dealt with in peritoneal dialysis, however, (i.e.,
whole blood as the solution to be dealt with in peritoneal dialysis) this relation-
ship must be considered only an approximation. Note, however, that e will vary
with J f a phenomenon noted with ultrafiltration across the glomerular membrane
[11, 12].

3. Measuring the sieving coefficient

Previous experiments to measure S, in general, involve infusing peritoneal di-

alysis fluid that is made hypertonic to plasma water with glucose, but contains the
solute for which the measurement is to be made in a concentration equal to that in
plasma water. The average concentration of this solute in ultrafiltrate (Cf ) may be
computed from the concentration of the solute measured in inflowing and out-
flowing dialysis fluid. The number of milligrams of solute transported by con-
vection (solute entrained in the moving water stream that is carried across the
membrane) into the peritoneal space is computed by difference and factored for
the time of the exchange to obtain an average rate of solute removal (Q s =
mmoles/min). In like manner the volume of dialysis fluid removed in excess of
that infused provides the volume of ultrafiltrate which is also factored for the
exchange time (Q f = mllmi) as

C = Js (mmoles/min/membrane area)l
f J f (ml/min/membrane area)

The concentration of the solute in peripheral blood water is then measured. 4 This
is usually measured in plasma and a correction factor for the displaced volume
166

occupied by the plasma proteins ('protocrit?,) converts the measured value to

plasma water concentration (Cw)'

C
-f=S=1-e
Cw

Cw = C/1- e where e the volume fraction of hydrated proteins is taken as 0.0107

times the concentration of total plasma proteins (7.4 Gm/dl in normal plasma)
[13].

4. The convective solute mass transfer

The ability to move solutes across the peritoneal membrane in the absence of a
driving concentration gradient for that solute was demonstrated in 1966 for urea
[14]. Convective mass transfer across the peritoneal membrane is now well
accepted. Important to the understanding of this means for moving solutes into
the dialysis fluid is the role that the membrane plays in restraining or modulating
such movement. The concept of membrane sieving or rejection as explored above
is central to an understanding of how solute movement occurs. Solute mass
transfer and its quantitation has been dealt with in some detail by Dr Popovich in
Chapter 5. Therefore, I shall confine my comments to those addressing con-
vective mass transfer and the impact that this may have on simultaneously
occurring diffusive mass transfer. As previously mentioned, there have been only
a limited number of studies done to quantitate the sieving coefficients for solutes
of interest in peritoneal dialysis. Table 1 offers the values generated for urea,
inulin, glucose, sodium, potassium and chloride. Take the simplest case first, an
uncharged solute of small molecular weight for which S = 1. Let us assume for the
moment that urea fulfills these criteria. Furthermore, let us consider only the

Peritoneal
membrane

2
o
i= C2
~
I-
2
W
U
2
o
U

DISTANCE

Figure 4. Concentration profile for a solute with S = 1 during ultrafiltration across the peritoneal
membrane. The dialysis fluid has been made up to contain a concentration of the solute (C2) that is
equal to that in plasma water (C1).
 167

membrane separating blood and dialysis fluid. Figure 4 is a diagram which depicts
a concentration profile across the peritoneal membrane at a point where a
capillary runs just beneath the investing peritoneum. The concentration for the
solute of interest, e.g., urea, has been adjusted to be equal in plasma water and
dialysis fluid and, for the present, we will assume the sieving coefficient to be 1.0.
The concentration profile is drawn for the conditions existing a few moments
after the onset of ultrafiltration and is linear. That is, there is no change in the
concentration
"I
of the ultrafiltered plasma water as it crosses the membrane and its
attendant stagnant fluid films. There is convective movement of water and urea
and the average convective clearance of urea will be the product of the sieving
coefficient and the ultrafiltration rate.

c
S =--.-1JL
u CWu

- C[-
Su Q [ = C u Q[ = average clearance of urea from plasma water by
Wu convection

_ mass removed by ultrafiltration/exchange time

plasma water concentration

Qf would usually be obtained as the volume difference (inflow volume - outflow

volume) divided by the exchange time. It is apparent that the average convective
clearance for a solute with S = 1 is equal to the ultrafiltration rate.
The measurement of the ultrafiltration rate as performed clinically provides an
average value. As previously noted the osmotic gradient generated by glucose
will deteriorate with time as an exponential function. The convective clearance
described then is also an average value for the plasma cleared over the time
interval of the exchange rather than an instantaneous value. Had the driving
gradient for ultrafiltration been kept constant during the study period, then
clearance would be constant with time as, for example, is the case when measur-
ing the clearance of inulin by the native kidney in which the driving force for
ultrafiltration is hydrostatic and constant.
However, in the usual clinical situation where one may wish to quantitate the
convective clearance for a solute such as urea there is no urea present in the
dialysis fluid. Diffusion and convection then are simultaneously ongoing during
the exchange. While there are theoretical treatments for the comparative contri-
butions of diffusion and convection to overall clearance rate [15], there are only
limited experiments in vivo with peritoneal membrane to try and substantiate the
respective contributions of these forces to net mass transfer [4]. However,
reasoning from this body of work and from analogy with synthetic membranes in
vitro a qualitative appreciation of the interaction between these two modes of
solute transport may be developed. Figure 5 is different from Figure 4 in that
168

~ c,
~
a::
f-
zw
u
z
8

D,
DISTANCE

Figure 5. Concentration profile for a solute that diffuses across the peritoneal membrane in the
absence of any ultrafiltration. The 'steepness' of the driving gradient for diffusion is depicted by the
difference in concentrations between the bulk phase of the plasma water (C 1) and the of the dialysis
fluid (C 2) acting over the distance D1 to D2, i.e., (C 1-C2)/(D1-D2).

there is no ultrafiltration occurring and the more usual clinical circumstance is

presented where there would be no urea present in the dialysis fluid. The
concentration profile for Figure 5 would be expected for the circumstance where
diffusion alone is present. The fall in concentration from bulk phase plasma to the
surface of the membrane represents the 'blood side' unstirred layer that is
partially depleted of solute by diffusion across the membrane into the dialysis
fluid. In like manner the 'dialysate side' stagnant fluid film is depicted as a
continuing reduction in concentration in the dialysis fluid before achieving the
concentration in the dialysate bulk phase (which at the start will be zero). With
peritoneal dialysis where there is little mixing as compared with hemodialysis,
one may expect this component of resistance to transport to be significant.
Imposing a transmembrane driving force for ultrafiltration on the conditions
shown in Figure 5 results in the changes shown in Figure 6. The reduction to zero
(or toward zero for a more diffusively permeable solute) of the blood side fluid
film will enhance diffusion by shortening the length of the path over which the
gradient for diffusion is acting (D, to D 2) between bulk phase blood and bulk
phase dialysate. In addition the drop in concentration within the membrane has
been obliterated, again shortening the path over which the concentration gra-
dient is exercised. Finally, there is a shift of the dialysate side film away from the
membrane. In the event that mechanical mixing may be somewhat better away
from the membrane there will again be some enhancement of diffusion.
Figure 7 shows the circumstance where there is significant membrane restraint
(S<l) for the solute. The solute is convected to the membrane surface where the
concentration builds up (,concentration polarization' or 'solute polarization') [16,
17]. The concentration within the membrane falls with distance and the dialysate
side film is displaced in a manner analogous to the situation in Figure 6. Again the
overall effect is to steepen the concentration gradient enhancing the diffusive
component of transport. Precise quantitation of this enhancement for the solutes
 169

z
o
~
~
Z
lLI
U
Z C,
o
u
Of

0, 02
DISTANCE

Figure 6. Concentration profile for a solute with S = 1 moments after ultrafiltration has begun. Note
the steeper driving gradient for diffusion created by the obliteration of the unstirred layer on the blood
side, i.e., D I-D 2 for diffusion (Figure 5 is larger than that depicted here). The impact of protein
concentration polarization on solute concentration, i.e., displaced volume effect, at the membrane/
blood interface is neglected.

z
o
~
~ c,
~ - - - \...•.....
o
u
Of
5<1

0,

DISTANCE

Figure 7. Concentration profile for a solute that is partially restrained by the membrane (S<l)
moments after the onset of ultrafiltration. The concentration gradient is made steeper by the buildup
of solute on the membrane (Cl ) and acts over a shorter distance than was the case for simple diffusion
(Figure 5). As in Figure 6, the impact of protein concentration polarization is neglected.

routinely dealt with in uremic plasma is not presently possible for the peritoneal
membrane and may not be in view of data that indicate that exposure to hyper-
tonic solutions results not only in increased membrane permeability, but in-
creased membrane area as well (increase in number of capillary loops perfused?)
[5].
Finally, in considering the impact of convection on diffusion it should be
apparent that solute size is important. For larger solutes with poor diffusive
permeability for the peritoneal membrane (referred to now without its attendant
fluid films), convective mass transport if S is high enough can contribute the
major portion of the total mass transported. As a corollary, the fraction of overall
mass transfer resistance to diffusion contributed by unstirred fluid films becomes
less as the solute considered becomes larger due to the membrane dominating the
serial resistances to transport [18, 19].
170

Albumin, the smallest of the traditional plasma proteins, bears some special
comment. The presence of protein in the initiating exchange of peritoneal dialysis
(i.e., the 'ascites' present) and its subsequent decline over the next few exchanges
to a steady-state level point up the fact that under usual circumstances there is a
loss of protein from the plasma into the peritoneal space. The capillary membrane
is the likely source. Albumin is the predominant protein. Table 1 lists an S for
albumin of <0.02. Glomerular restraint of albumin in animals is somewhat more
complete with S=0.0003 [20]. The figure given in Table 1 for the peritoneal
membrane is not rigorously arrived at as there are no experiments to determine
this value so far reported. Rather, I have used data for protein (TCA precipita-
ble) loss in a series of isotonic exchanges (1.3 ± 0.1 gin = 13) taken after washout
and subtracted this from losses in hypertonic solution (1.7 ± 2 g) to which no
protein had been added to block diffusion. The ultrafiltrate concentration of 0.06
gms/dl divided by a normal plasma albumin of 3.5 gms/dl = 0.02. This is undoubt-
edly too high a figure as diffusive loss is not blocked and the assumption is made
that we are dealing only with albumin when, in point of fact, small amounts of
globulin are also present. It is of interest, however, that the 'peritoneal mem-
brane' value in man approaches that for the glomerular membrane in animals.
Further, the presence of protein in spent peritoneal dialysis fluid is frequently
cited as a point for the peritoneal membrane as being 'very permeable' when
contrasted with hemodialysis or hemofiltration membranes. It should be noted,
however, that more than 98% of the protein is held back by this 'very permeable'
membrane even under the stress of hypertonic solution. For perspective,
Cuprophan PT-150 has a transmittance (tr) value (Tr=S) for egg albumin (44000
daltons, 46.6 A) of 0.002 and the more permeable Rhone-Poulenc AN 69 of 0.04
[21]. In light of its slow diffusivity, it seems likely that the predominant mecha-
nism for protein loss across the peritoneal membrane is convective even in the
presence of isotonic dialysis solutions. The presence of dialysis fluid would, of
course, 'trap' by massive dilution, protein lost from the capillary that normally
would, under steady-state circumstances, be either convected back across the
venular end or returned to the vascular space via the lymphatics.

5. Critique of present concepts and some speCUlations

There are several potential problems with the methods that have been employed
to obtain the values reported in Table 1. First, exposure of the peritoneal
membrane to hypertonic solutions alters the permeability characteristics of the
peritoneal membrane. Enhanced diffusive transport is reported to occur after
such exposure. Data on the comparative overall mass transfer resistances for urea
and inulin indicate that there is both an increase in membrane area and per-
meability [5]. This complexity of having the study parameter altered by the
experimental condition used to test the parameter suggest that true estimates of
 171

'resting' peritoneal membrane £ and a will not be readily forthcoming. This will
disturb the physiologists more than the clinicians. Secondly, in computing the
sieving coefficient the use of peripheral plasma water concentration for the
denominator of that calculation assumes that we know that the ultrafiltrate that is
drawn across the peritoneal membrane originates directly from the vascular
space. This need not be the case and at present there is no clear experimental
evidence to guide this judgement. A corollary question to the origin of the
ultrafiltrate is 'what membrane participates in the ultrafiltration process and is it
the same membrane that participates in the diffusion process?' Again, there is no
direct experimental data to guide us.
A reasonable inferential judgement that approximately 20-30% or so of ultra-
filtrate is drawn from the extravascular space and, hence, utilizes a membrane
that is other than that described in Chapter 2 is possible. This other membrane
might be the cell wall, its investing peritoneal membrane, and the attendant
interstitial space andlor possibly the membrane dividing the interstitial space
from the peritoneal space. Evidence for these alternative membranes and sites
for derivation of ultrafiltrate stem from the following considerations. Inulin is
distributed in the extracellular space and urea in the total body water. This raises
a paradox when data from Table 1 are considered. Urea, a small uncharged solute
that is known to cross most biological membranes readily [22], should have a high
sieving coefficient, i.e., near 1.0 and yet the highest reported value is only 0.8.
Inulin, which we might expect to be restrained by biologic membranes, also has a
high value of 0.4 to 0.8. In order to reconcile these values we must speculate that
because the experimental design for measuring these values for S were obtained
after either 'washout' exchanges or one or more isotonic exchanges that urea and,
to a lesser extent, inulin were cleared from a certain fraction of cell or interstitial
space water and that with subsequent hypertonic dialysis fluid that this inulin and
urea-free water is delivered as ultrafiltrate in conjunction with ultrafiltrate from
intravascular sources. Approximately 20% of the ultrafiltrate then must have
been 'totally depleted' of its urea and inulin by diffusion prior to use of the
hypertonic exchange used to measure S for the two solutes. In order for this
mechanism to occur, the overall resistance to diffusion whether it be the compo-
nent resistance of the anatomical membranes or the length of the path for
diffusion between the plasma water (vascular space) and the intracellular or
interstitial spaces or both must be greater than that between these latter spaces
and the dialysis fluid for urea. In addition or alternatively, the greater depletion
that might occur for urea because of its faster diffusivity would have to be offset
by the fraction of inulin-free water drawn as ultrafiltrate from within the cell.
An alternative hypothesis for explaining the data from Table 1 would involve
the presence of a heterogeneous set of membranes for both diffusion and con-
vection. Specifically, examining what is known of capillary structure and func-
tion, it is apparent that the arteriolar end of the capillary and its attendant
arteriole have significantly narrower endothelial junctions than are present in the
172

venular end of the capillary [23, 24]. Mass transfer studies indicate that this
arteriolar end is 'less permeable' [25]. If these endothelial junctions are the much
speculated about 'pores' in the peritoneal membrane, then one may suggest that
we are dealing with a heterogeneous set of membrane pores. Also, there is
evidence that additional capillaries may be recruited when blood flow to the
peritoneum is enhanced by osmotic irritation or pharmacologic manipulation of
vascular resistances to flow with agents such as nitroprusside [26]. It has been
suggested that these 'dormant' capillaries noted in animal studies are more
permeable to the movement of water and solutes than capillaries perfused under
basal conditions [27] - again the potential for dealing with a membrane that is
heterogeneous with respect to permeability characteristics. Heterogeneity of
sieving properties may then well be assumed. With this model in order to explain
values of less than unity for urea, one must postulate an exceptionally 'tight'
portion of the membrane which then would effectively block passage of both urea
and inulin. Additionally, inulin and urea would then be passed with an S near
unity by the more open portion of the membrane. The difficulty I find with this
model lies in the degree of 'tightness' required of this biologic membrane to
achieve net S of 0.8 for urea. While the above considerations plus the present
uncertainty about the area of membrane involved and the required complexity of
the solutions used in delivering ultrafiltrate to hypertonic dialysis fluid, to say
nothing of the difficulties in adequately characterizing the concentration profile
in the unstirred layers on or in the anatomical peritoneal membrane, makes it
unlikely that during experiments with isotonic and hypertonic solutions valid
estimates of the absolute a for a given solute are likely to emerge for the
peritoneal membrane. These concerns, however, do not preclude the experimen-
tal development of relative overall reflection coefficients for non-electrolytes [4].
In addition, the spaces from which ultrafiltered water is drawn is well within
experimental reach.
It is harder to derive information from the electrolyte data given in Table 1 as
sodium and chloride (extracellular) and potassium (intracellular) are much gov-
erned by active transport and considerations relating to their charge. For exam-
pIe, it is apparent that cell water potassium does not achieve diffusion equilibrium
with isotonic dialysis fluid as equilibrium values for dialysis fluid approach plasma
water not cell water concentrations.
Dedrick et al. [28] have recently added a very interesting new dimension to the
above noted problems by raising the concept of a spacially distributed capillary
network rather than a porous membrane as the physical model by which to
understand solute diffusion between peritoneal fluid and blood. While the 'dis-
tributed' model shows reasonable agreement with the membrane model for
purely diffusive parameters it is unlikely that this similarity will hold for con-
vective mass transport. When good quality data are available on convective mass
transport for solutes of widely varying molecular weight and similar charge and
molecular configuration it will be possible to identify which of these two models is
in best agreement with these experimental measurements.
 173

6. Therapeutic applications of ultrafiltration

The removal of excess total body water and its constituent solutes remains the
single most important reason to use hypertonic peritoneal dialysis solutions.
Figure 1 is adapted from Rubin et al. to show the rate of osmotic equilibration for
the 4.25% glucose containing solutions. In addition, Figure 3 is offered to show
the changing rate of ultrafiltration during an exchange. It should be noted that at
20 min the vast majority of ultrafiltration has been completed. In order to
maximize fluid removal a short dwell time (20 min) is appropriate. In addition, a
nurse wishing to identify whether hypotension is imminent as excess total body
water is being removed during the course of a peritoneal dialysis would do well to
check the vital signs at the 20-minute mark with hypertonic exchanges.
Recent reports of impaired water transport in CAPD patients from France,
Norway and the USA are of concern [29, 30, 31, 32, 33]. The frequency of this
occurrence is higher in France than the United States. The predominant mecha-
nism for this event appears to be reduction in mass transport resistance (i.e.,
enhanced diffusive transport) for glucose such that the usual hypertonic exchange
swiftly loses its driving gradient for water removal through diffusive loss of
glucose from dialysate to blood [32, 34]. However, some patients manifest a
decrease in transport parameters for both water and small solute including
glucose [35]. The respective frequency of these two mechanisms remains to be
determined.
Review of a publication of the results of an International Cooperative Study
[36] measuring effluent exchange volume and dialysate glucose concentration
after a single four-hour hypertonic exchange adds the following points to our
understanding:
(a) while acetate as the alkali equivalent correlated with lower filtrate volume
and dialysate glucose concentration, lactate had no such correlation;
(b) there was no significant correlation with the number of episodes of per-
itonitis suffered nor the age of the patient; there was, however a significant
correlation with the months on treatment with CAPD;
(c) of the 185 patients studied most (162) were treated with lactate and the vast
majority of these (152) were supplied by a single manufacturer. This muddles the
interpretation from correlation (a) above raising a strong secondary correlation
with the manufacturer/supplier such that it is not appropriate at present to single
out acetate as the etiology for decreased filtrate volume. Further studies of
solution composition as well as a prospective study with time on treatment will be
important to have before the impact of this phenomenon on long-term CAPD can
be judged. Early reports of increased hydraulic permeability without changes in
small molecule clearance, i.e. urea, creatinine, in response to amphotericin Bare
of high interest [37].
The observed sieving coefficients for the electrolytes Na (0.5) and K (0.4) and
chloride (0.8) require the clinician to expect hypernatremia and hyper-
174

chloridemia if significant fluid removal is undertaken. As potassium is actively

being removed by diffusion in addition to convection (unlike sodium and chlo-
ride), it offers little problem and as previously noted the convective loss with S<1
will enhance the loss by diffusion. Treatment of these iatrogenic concentration
distrubances may be accomplished either by offering sodium chloride-free water
back in amounts calculated to offset the sodium and/or chloride-free water lost as
ultrafiltrate or by the use of alternating hypotonic or isotonic exchange that would
permit diffusion equilibrium or water reabsorption to readjust these supernormal
values back toward normality. At times plasma glucose levels rise considerably
when 4.25% glucose containing dialysis fluid is used resulting in osmolar shift of
water out of cells. (One may speculate that by analogy water movement out of the
cells bathed by the hypertonic dialysis fluid also occurs.) This would artifactually
ameliorate the presence of hypernatremia induced by serial hypertonic ex-
changes. With the return to isotonic exchanges or discontinuance of the dialysis,
glucose and water would move intracellularly revealing the true degree of hyper-
natremia. At present, these is no information on sieving coefficients for other
electrolytes.
Alternative solutes to glucose for generating hypertonicity with its attendant
caloric load have not been readily forthcoming. The biologic safety in a func-
tionally anephric population of compounds such as glycerol, sorbitol and man-
nitol is discussed in greater detail elsewhere. Suffice it to say that none has proved
to be a clear favorite.
Use of amino acids as osmotic agents has received a preliminary and positive
trial by the Toronto group [38]. Concentrations that provided 50% fewer calories
resulted in comparable water removal to hypertonic glucose solutions. Amino
acids, in addition to providing an osmolar gradient for water loss, would offset
amino acid losses across the peritoneal membrane. The cost of these solutions is a
potential drawback.
Animal work on the toxic charged polymer poly (sodium) acrylate demon-
strates the potential for such an approach using nontoxic agents [39]. In principle,
the large molecular size of the polymer would limit its movement out of the
peritoneal space, but would make it rather inactive osmotically. The polymer's
attendant ions would have enhanced osmotic activity as a result of being con-
strained to remain in the peritoneal cavity in order to maintain electro neutrality.
Simple uncharged polymers of glucose of appropriate size may also prove to be
useful.

Notes

1. This equation is usually written for ultrafiltering capillaries with a term to express the hydrostatic
diving force resulting from the blood to interstitial pressure gradient (L1P), i.e.,
n
Qf = ALp(L1P + L LllTpJ)
J~l
 175

Because the peritoneal space under normal circumstances is free of significant quantities of fluid, we
may reasonably assume that the hydrostatic and oncotic forces at play across the capillary membrane
from arteriolar to venular end are in balance with lymphatic run off. In order to effect net accumula-
tion of ultrafiltrate in the peritoneal space, these forces must be unbalanced by the osmotic force
contributed by the presence of glucose in the dialysis fluid. For a two-liter exchange added to the
peritoneal space, the hydrostatic force contributed would result either from the elastic recoil of the
abdominal wall and/or the hydrostatic head of pressure generated by the 'column' of water above the
dependent portion of the peritoneal membrane. As such it would be expected to be small and in a
direction favoring net uptake across the peritoneal membrane from 'bath to blood'. To be precise, this
force must be added as a component of LIP. Lastly, in considering this simplification there is the
unlikely possibility that hypertonic dialysis fluid by some mechanism may alter the afferent/efferent
resistances of the peritoneal capillary bed in such a manner as to enhance the hydrostatic pressure
gradient thereby causing ultrafiltration.
2. With the cellulose membranes for laboratory use (i.e. not Cuprophan) studied in vitro [2], the
values obtained were au = 0.024 and a g = 0.20, i.e. au/ag = about 0.1 suggesting that this synthetic
membrane is a good deal 'tighter' than peritoneal membrane - a conclusion arrived at by another line
of reasoning [3].
3. It is assumed that the area of the membrane for movement of the solute is the same as that for the
movement of water and, hence, may be canceled out of the ratio - a point that needs further
experimental proof.
4. Again there is the assumption that the concentration of the solute in the plasma water available
from a peripheral vessel (artery or vein) represents the concentration of that solute on the 'blood side'
of the peritonel membrane - again a point that needs further experimental proof.

References

1. Rubin J, Nolph KD, Popovich RP, MoncriefB: Drainage volume during continuous ambulatory
peritoneal dialysis. ASAIO J 2: 54,1979.
2. Durbin RP: Osmotic flow of water across permeable cellulose membranes. J Gen Physiol44: 315,
1960.
3. Henderson LW: The problem of peritoneal membrane area and permeability. Kidney Int 3: 409,
1973.
4. Pyle WK, Popovich RP, Moncrief JW: Mass Transfer Evaluation in Peritoneal Dialysis. In:
Moncrief JW, Popovich RP (eds), Proc. CAPD Int Symp II, May 9-10 1980, NY: Masson, pp
35-52, 1981.
5. Henderson LW, Nolph KD: Altered permeability of the peritoneal membrane after using
hypertonic peritoneal dialysis fluid. J Clin Invest 48: 992, 1969.
6. Nolph KD, Hano JE, Teschan PE: Peritoneal sodium transport during hypertonic peritoneal
dialysis. Ann Intern Med 70: 931, 1969.
7. Brown ST, Ahearn DJ, Nolph KD: Potassium removal with peritoneal dialysis. Kidney Int 4: 67,
1973.
8. Rubin J, Klein E, Bower JD: Investigation of the net sieving coefficient of the peritoneal
membrane during peritoneal dialysis. ASAlO J 5: 9, 1982.
9. Speigler KS, Kadem 0: Transport coefficients and salt rejection in uncharged hyperfiltration
membrane. Desalination 1: 311, 1966.
10. Villarroel F, Klein E, Holland F: Solution flux in hemodialysis and hemofiltration membranes.
Trans Am Soc Artif Intern Organs 23: 255, 1977.
11. Pappenheimer JR, Landis EM: Exchange of substances through the capillary walls. In: Hand-
book of Physiology, Vol 2, Section 2, p 961. American Physiological Society, Washington, DC,
1963.
176

12. Rosenbaum RW, Hruska KA, Anderson C, Robson AM, Slatopolsky E, Klahr S: Inulin: an
inadequate marker of glomerular filtration rate in kidney donors and transplant recipients?
Kidney Int 16: 999, 1979.
13. Colton CK, Smith KA, Merrill EW, Friedman S: Diffusion of urea in flowing blood. Am Inst
Chern Engin J 17: 800, 1971.
14. Henderson LW: Peritoneal ultrafiltration dialysis: Enhanced urea transfer using hypertonic
peritoneal dialysis fluid. J Clin Invest 45: 950, 1966.
15. Babb AL, Johansen PJ, Stand MJ, Tenckhoff H, Scribner BH: Bidirectional permeability of the
human peritoneum to middle molecules. Proc Eur Dial Transpl Assoc 10: 247, 1973.
16. Andreoli TE, Schafer JA, Troutman SL: Coupling of solute and solvent flows in porous lipid
bilayer membranes. J Gen Physiol 57: 479, 1971.
17. Colton CK, Henderson LW, Ford Ca, Lysaght MJ: Kinetics of hemodiafiltration I. In vitro
transport characteristics of a hollow-fiber blood ultrafilter. J Lab Clin Med 85: 355, 1975.
18. Henderson LW, Colton CK, Ford CA: Kinetics ofhemodiafiltration II. Clinical characterization
of a new blood cleansing modality. J Lab Clin Med 85: 372, 1975.
19. Colton CK: Permeability and transport studies in batch and flow dialyzers with application to
hemodialysis. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, 1969.
20. Carone FA, Banks DB, Post RS: Micropuncture study of albumin excretion in the normal rat.
Am J Physiol 55: 19A, 1969.
21. Green DM, Antwiller GD, Moncreif JW, Decherd JF, Popovich R: Measurement of the transmit-
tance coefficient spectrum of Cuprophan and RP 69 membranes: application to middle molecule
removal via ultrafiltration. Trans Am Soc Artif Intern Organs 22: 627, 1967.
22. Colton CK, Smith KA, Merrill and Reece JM: Diffusion of organic solutes in stagnant plasma and
red cell suspensions. Chern Eng Prog Symp Series 66: 85,1970.
23. Wayland H, Silberber A: Blood to lymph transport. Microvasc Res 15: 367, 1978.
24. Miller FN, Wiegman DL, Joshua IG, Nolph KD, Rubin J: Effects of vasodilators and peritoneal
dialysis solution on the microcirculation of the rat cecum. Proc Soc Exp Bioi 161: 605, 1979.
25. Wayland H: Transmural and interstitial molecular transport. Proc. Int. Sympt. CAPD. Paris,
1979, Excerpta Medica. Amsterdam, p 18,1980.
26. Nolph KD, Ghods A, Van Stone J, Brown PA: The effects of intraperitoneal vasodilators on
peritoneal clearance. Trans Am Soc Artif Intern Organs 22: 586, 1976.
27. Renkin EM: Exchange of substances through capillary walls. In: Circulatory and Respiratory
mass Transport, Ciba Foundation Symp. Little, Brown and Co, Boston, 1969.
28. Dedrick RL, Flessner MF, Collins JM, Schultz JS: Is the peritoneum a membrane? ASAIO J 5: 1,
1982.
29. Slingeneyer A, Canaud B, Mion C: Permanent loss of ultrafiltration capacity of the peritoneum in
long term peritoneal dialysis: An epidemiologic study. Nephron 33: 133, 1983.
30. Faloer B, Marichal JF: Loss of ultrafiltration in CAPD: clinical data; advances in peritoneal
dialysis. In: Gahl, Kessel, Nolph (eds), Int Congr Ser No 567, Excerpta Medica, Amsterdam,
1981, p 227.
31. Verger C, Brunschvicg 0, Le Charpentier Y et at: Structural and ultrastructural peritoneal
membrane changes in permeability alterations during CAPD. Proc Eur Dial Transpl Assoc
18: 199, 1981.
32. Smeby Lc, Wideroe T, Jorstad S: Individual differences in water transport during continuous
peritoneal dialysis. ASAIO J 4: 17, 1981.
33. Nolph KD, Pyle WK: NIH CAPD patient registry report: population demographcs and outcomes
for the period 1/1/82 through 12/31/82, 1983.
34. Rubin J, Ray R, Barnes T, Bower J: Peritoneal abnormalities during infectious episodes of
continuous ambulatory peritoneal dialysis. Nephron 29: 124, 1983.
35. Wu G, Khanna R, Oreopoulos DG, Vas SI: Incidence and pathogenesis of ultrafiltration failure
among CAPD patients. Abstract Am Soc Nephrol p. 125A, 1983.
 177

36. Nolph KD: An international cooperative study: factors effecting ultrafiltration in contino us
ambulatory peritoneal dialysis. Perit Dial Bull 4: 14-19, 1984.
37. Maher JF, Hirszel P, Bennett RR, Chakrabarti E: Amphotericin B selectively increases per-
itoneal ultrafiltration. Abstract Am Soc NephroJ p 121A, 1983.
38. Oren A, Wu G, Anderson GH, Marliss E, Khanna R, Petitt J, Mupas L, Rodella H, Brandes L,
Roncari DA, Kakis G, Harrison J, McNeil K, Oreopoulos DG: Effective use of amino acid
dialysate over four weeks in CAPD patients. Trans Am Soc Artif Intern Organs 29: 604,1983.
39. Nolph KD, Hopkins C, Rubin J, Twardowski Z, Popovich R, VanStone J: Polymer induced
ultrafiltration in dialysis: high osmotic pressure due to impermeant polymer sodium Trans Am
Soc Artif Intern Organs 24: 162, 1978.
7. Intermittent peritoneal dialysis as renal
replacement therapy

SUHAIL AHMAD, FU-HSIUNG SHEN and

CHRISTOPHER R. BLAGG

1. Historical perspective

In 1877 Wegner [1] described the effect of cold peritoneal lavage on body
temperature and noted that hyperosmolar solutions containing sugar or glycerine
resulted in increased outflow. Further studies on changes in outflow volume using
solutions of different concentrations were reported over the next 40 years, but the
ability of the peritoneum to absorb fluid was first used clinically in 1918 to treat
children with gastrointestinal problems that precluded oral intake.
Throughout the 1920s and 1930s studies were made on the diffusion of sub-
stances across the peritoneal membrane and the effect on blood levels, the
mechanism of increased outflow volume using hypertonic glucose, and the effect
of vasodilatation and vasoconstriction on diffusion. References to these and other
early reports are found elsewhere in this book.
Peritoneal dialysis was first used to remove uremic toxins by Ganter in 1923 [2]
in animals made uremic by ureteric ligation. After 2-4 h of dialysis the concentra-
tions of non-protein nitrogen in peritoneal fluid and blood were similar, and the
animals showed marked clinical improvement. Ganter also was the first to report
use of peritoneal dialysis in the human, studying the effects of physiological saline
solution introduced intraperitoneally in a patient with ureteral obstruction due to
uterine carcinoma, noting that her clinical condition showed slight improvement.
Peritoneal dialysis was used clinically in a few patients during the 1930s, but the
next major step was the report on the use of peritoneal dialysis in acute renal
failure by Fine and co-workers [3]. By 1950, Odel et al. were able to collect 101
patients from the literature who had been treated by either intermittent or
continuous peritoneal dialysis [4]. During the ensuing 10 yr peritoneal dialysis was
used more widely for treatment of acute renal failure in both adults and children,
and was also used successfully to treat hypercalcemia and various poisonings.
The modern era of peritoneal dialysis was ushered in by publication of a
monograph by Boen in the Netherlands in 1959. In this classic work, published in
the United States in 1964 [5], Boen described studies on the kinetics of peritoneal
180

dialysis and discussed the indications, technique, and complications of this treat-
ment. Togethtr with the report of Maxwell and co-workers from the United
States [6], this established the use of peritoneal dialysis in the treatment of acute
renal failure; and because hemodialysis was not widely available at this time,
peritoneal dialysis began to be used at a number of centers.
In 1960, with development of the teflon arteriovenous shunt for hemodialysis
[7], Scribner and co-workers at the University of Washington began studying the
practical problems of long-term treatment of patients with chronic renal failure.
Scribner invited Boen to Seattle in 1962 to continue his work on peritoneal
dialysis and to extend its application to the treatment of chronic renal failure.
Equipment used previously for the experimental treatment of chronic renal
failure by gastrodialysis was modified by Boen to make the first closed-system
peritoneal dialysis cycler [8] using dialysate sterilized in 40-L glass bottles (Fig. 1).
This closed system dramatically reduced the frequency of peritonitis because it
eliminated repeated connections to fresh dialysate containers during the course of
a dialysis [9]. In 1963 Boen started treatment of a 28-year old woman with chronic
renal failure by out-patient intermittent peritoneal dialysis (IPD), and this was
continued successfully for 4 yr until the patient was transferred to home hemo-
dialysis. This patient is alive today, having had a successful related donor trans-
plant in 1971.
Development of a closed system solved one of the two major problems posed
by long-term peritoneal dialysis. The other, repeated access to the peritoneal
cavity, was the subject of intensive study in the early 1960s, and various indwelling
'buttons' and other devices were developed. Generally, these were not successful
because of infection, flow problems, and lack of a closed system.
Tenckhoff, who came to Seattle in 1964 to work with Boen, addressed himself
particularly to the access problem. The need was to develop an indwelling access
device which would minimize the risk of infection. Palmer had described a
permanently implanted silastic catheter with a long subcutaneous tract [10].
Working from this concept, Tenckhoff attached dacron felt cuffs to the catheter at
levels just below the skin and immediately outside the peritoneum. Tissue
ingrowth into these cuffs fixed the catheter in place and provided an effective
barrier against bacterial invasion of the sinus tract around the catheter. This, the
Tenckhoff catheter [11, 12], originally implanted using a special trocar, became
the standard access device for long-term peritoneal dialysis. A modification with
a single cuff is widely used for patients with acute renal failure.
Tenckhoff also worked on development of peritoneal dialysis equipment which
would eliminate the need for 40-L containers of sterile fluid. Such equipment,
while useful in the hospital, could also be used for home IPD. The first such
equipment, using heat sterilization [13], became available in 1970 (Fig. 2). This
was large and heavy and, while used for home dialysis by a small number of
patients, was never a commercial success.
The early 1970s showed increased interest in IPD for chronic renal failure,
 181

Figure 1. Boen's closed-system peritoneal dialysis equipment, 1963.

182

Figure 2. Tenckhoff's peritoneal dialysis system using heat sterilization .

 183

mainly as a result of the continuing work of Tenckhoff, supported by the National

Institutes of Health. His development of a peritoneal dialysis fluid supply system
using reverse osmosis to produce sterile, pyrogen-free water was the significant
advance which made long-term IPD feasible (Fig. 3) [14]. As a result, both the
Northwest Kidney Center and the Veterans Administration Hospital in Seattle
developed programs offering home IPD as an alternative treatment for suitable
patients.

u u u u

Figure 3. Reverse osmosis peritoneal dialysis system.

184

In the late 1970s there was renewed interest in the use of pharmacologic agents
to improve peritoneal clearance and so reduce the long duration of a peritoneal
dialysis, and the possibility of peritoneal dialysis using dialysate regeneration also
was explored. These efforts are discussed elsewhere in this book. However, IPD
remained the stepchild of dialysis treatments, being used for less than 5% of
patients in the United States. In part, this related to lingering suspicions based on
the high infection rate associated with use of open peritoneal dialysis systems in
the 1960s. In addition, in the United States, this also reflected availability of
generous federal support of dialysis patients from 1973 onwards which encour-
aged establishment of out-patient hemodialysis units, discouraged IPD because
of its longer duration and resulting higher cost, and discouraged home dialysis
generally.
The most notable advance in peritoneal dialysis in the 1970s was the brilliant
conception by Popovich, a biomedical engineer who had previously worked on
hemodialysis kinetics in Seattle, of the principle of continuous ambulatory per-
itoneal dialysis (CAPD) [15]. As described elsewhere in this book, since then
there has been an astonishingly rapid increase in the use of CAPD worldwide. In
the United States this increase was aided by a major publicity campaign, was seen
by many patients who did not have access to home hemodialysis programs as a
simple form of home dialysis, and was welcomed by the Health Care Financing
Administration (HCFA) as a less expensive form of treatment despite a lack of
reliable cost data.
This increased interest in peritoneal dialysis also led to another approach. In
1979 Scribner proposed the use of nightly automated peritoneal dialysis together
with daytime ambulatory peritoneal dialysis [16], and this was developed by Diaz-
Buxo and co-workers [17] as continuous cycling peritoneal dialysis (CCPD).
As a result of these developments, more than 7 000 patients in the United States
are treated by one form of peritoneal dialysis or another (mainly CAPD); and
there has been interest by manufacturers in developing new tubing sets, connec-
tion devices, and fluid supply systems. However, an unfortunate side effect of the
boom in CAPD has been a decline in interest in the use of IPD, and so production
of reverse osmosis dialysis equipment has ceased in the United States because of
limited sales. Now, with introduction of CCPD, there is renewed interest in
development of automated cyclers, and it would appear to be only a matter of
time before a new generation of reverse osmosis equipment is developed.
Despite resolution of the major technical obstacles to IPD by the mid-1970s,
use of this modality increased slowly. In 1977 the European Dialysis and Trans-
plant Association reported 412 patients being treated with IPD in Europe [18]. By
1979 this had increased to 1696 patients, of whom 646 were treated by CAPD and
1050 by IPD [19]. Most recently, in 1982, 32 countries reported 1675 patients
treated by IPD, 4417 on CAPD, 240 treated by CCPD, and 155 using a combina-
tion of peritoneal and hemodialysis [20]. In the United States, National Dialysis
Registry data in 1974 showed only 72 of 5273 dialysis patients being treated by
 185

peritoneal dialysis. By 1983 the Health Care Financing Administration reported

on 65 765 dialysis patients, of whom 6 523 were treated by CAPD and 1701 by IPD
(CCPD was not counted separately) [21]. Thus, while IPD was slow in gaining
acceptance, CAPD, once established and with availability of fluid in bags rather
than bottles, had a meteoric increase. Some reasons for the slow growth in use of
IPD are discussed else~here in this chapter.

2. The role of peritoneal dialysis in acute renal failure

Peritoneal dialysis is an accepted alternative to hemodialysis for treatment of

patients with acute renal failure, and in particular for those with serious bleeding
problems and those with poor vascular access [22]. Firmat and Zucchini reviewed
reports on peritoneal dialysis and acute renal failure prior to 1979 [23]. Mortality
in various series ranged from 17% to 75%, averaging about 40%, similar to that
for patients with acute renal failure treated by hemodialysis. The reports included
patients treated by repeated abdominal puncture and those using a Tenckhoff
catheter, and patients treated using 1- or 2-L bags or bottles of dialysate, as well as
those treated with closed systems. The most common exchange rate was 2-L
dialysate twice every hour. Heparin, usually 1000 u in each 2-L exchange, was
used by some (although there have been no prospective studies to assess its
value), but antibiotics were not generally added to dialysate. There was little to
suggest that mortality was related to use of IPD; rather, this was associated with
the cause of the acute renal failure.

2.1. Complications

There have been several reviews of the complications of IPD in acute renal failure
[24, 25, 26]. Mechanical complications include pain, hemorrhage, leakage, and
inadequate drainage [27]. Pain usually responds to catheter adjustment and
analgesics. Bleeding into the dialysate only rarely requires specific treatment.
Dialysate leakage through the catheter site usually is not a problem and does not
result in infection. Poor drainage usually responds to catheter adjustment, al-
though occasionally requiring catheter replacement. Other mechanical complica-
tions, such as perforation of a hollow viscus, are most uncommon.
Bacterial peritonitis is the major complication, occurring in 31. 7% of patients
reported by Firmat and Zucchini [23] and 5% to 75% of patients in other series
[25,26,27,28]. One report notes that while 30% of repeated peritoneal cultures
were positive, only 6% of patients developed symptomatic peritonitis, usually
responding readily to appropriate antibiotic therapy [27].
Cardiovascular complications, principally tachyarrhythmias, are common, oc-
curring in 37% of patients. These are usually associated with underlying heart
186

disease or the poor general status of the patient, and only rarely are due to
fluctuations of serum potassium level [27]. Pulmonary complications include
pneumonitis, often associated with aspiration and pleural effusion [29, 30],
usually associated with fluid overload and readily managed by fluid removal
during dialysis. Neurologic complications include seizures and symptoms sugges-
tive of disequilibrium, probably related to osmotic shifti}, and are less common
than with hemodialysis. Metabolic complications include elevation of blood
glucose associated with the dextrose load, hyponatremia due to differential
absorption of sodium and water, and more rarely, hypokalemia, hypoglycemia,
and alkalosis.

2.2. Indications and contra-indications

The use of IPD in acute renal failure has been reviewed by Firmat and Zucchini
[23] and by Mathew [31], and there is also a report of the use of CAPD in acute
renal failure [32]. Peritoneal dialysis may be specifically indicated for patients
with intractable shock, a serious bleeding tendency, gastrointestinal hemorrhage,
severe electrolyte disturbances, or after severe head trauma [33]. Contra-indica-
tions to peritoneal dialysis in acute renal failure are few, since patients with recent
abdominal surgery or acute peritonitis may be helped by this treatment [34].
Serious contra-indications include hypercatabolism which may exceed the capac-
ity of IPD to remove uremic metabolites, undiagnosed intra-abdominal disease,
infection of the abdominal wall, previous peritonitis with adhesion formation,
ileus, perforation of a hollow viscus, a patent opening between peritoneal and
pleural cavities, or presence of severe respiratory insufficiency.
The advantages of IPD in acute renal failure include simplicity, short set-up
time, easy access, slower biochemical changes during dialysis, lack of need for
anticoagulation, and opportunity to use intra-abdominal antibiotics and lavage in
patients with abdominal sepsis. Peritoneal dialysis is commonly used for acute
renal failure in small children and is the treatment of choice in newborns and
infants [35].
In making the choice between hemodialysis and IPD for a patient with acute
renal failure, the tendency is to use IPD for patients with severe hemodynamic,
hematologic, or neurologic problems which might be aggravated by hemodia-
lysis. Thus one might anticipate that the mortality rate would be higher in patients
treated by IPD, but this does not appear to be the case [36, 37]. In the report of
Firmat and Zucchini [23], 40% of deaths were due to sepsis, 25% to irreversible
shock, and 14% to pulmonary embolism. No deaths were directly related to the
dialysis procedure.
Comparison of the effectiveness of hemodialysis and IPD in the treatment of
acute renal failure remains an open question [23, 34]. Some facilities use IPD for
most patients, reserving hemodialysis for those with specific contra-indications to
 187

IPD. In recent years, availability of femoral and subclavian catheters has made
hemodialysis much easier to perform. This, together with the large number of
hospitals in the United States with hemodialysis units, has resulted in less use of
IPD for acute renal failure. Even so, IPD is likely to continue to be used for at
least some patients with acute renal failure, and especially for small children.

3. Intermittent peritoneal dialysis in the treatment of chronic renal fairure

The use of IPD in patients with chronic renal failure has declined in recent years,
largely due to introduction of CAPD. For example, before 1979 (pre-CAPD)
about 32 new patients annually were started on IPD in the Northwest Kidney
Center program in Seattle; this has now declined to less than 10 patients per year.
However, developing interest in modifications of peritoneal dialysis may even-
tually lead to a recrudescence of some form of overnight IPD. In 1979 Scribner
proposed a combination of nightly IPD and daytime ambulatory peritoneal
dialysis, with less connect-disconnect procedures [16], and in 1980 Diaz-Buxo and
co-workers reported continuous cycling peritoneal dialysis (CCPD) utilizing such
a combination of nightly automated and daytime ambulatory peritoneal dialysis
[17]. This technique, using a cycler for nightly dialysis, is now being used in
selected patients. Thus, while use of three-times-a-week IPD has declined, a
variant of IPD is emerging as a new form of treatment [38].

3.1. Long-term survival

In the last decade there have been reports on more than 1000 patients treated by
IPD [39-48], but survival information remains scarce since patient selection
criteria in different programs may not be comparable. Unfortunately, the report
of a multicenter cooperative study performed by the Veterans Administration
[49] is still unpublished. Most of the data available suggest that survival with IPD
is not significantly different from that of comparable patients treated by hemo-
dialysis [19, 50, 51, 52]. However, a recent report from Israel comparing elderly
patients showed a similar first-year survival rate with hemodialysis and IPD, but
beyond 15 months survival with IPD was worse, and IPD patients required longer
periods of hospitalization [53].
In 1979 a retrospective study of adult non-diabetic patients treated in Seattle by
IPD was reported with a follow-up of at least one year [46]. Cumulative patient
survival calculated by the life-table m@thod [54] was 55% at three years (Fig. 4).
This compares with a three-year survival of 40% for IPD patients reported by
Price et al. [45]. The mean age of the Seattle IPD patients was 57.5 yr, and if
patients older than 60 were excluded, the three-year survival increased to 66%.
The lower survival reported by Price et al. could be due to patient selection or to a
188

100'wr~~-----------------------------'

80
~
(4)

(6)

:::::. ENTIRE GROUP

::::. VOUNGER GROUP

1 2 3 4
DURATION OF DIALYSIS, VEARS

Figure 4. Patient and technique survival for Seattle peritoneal dialysis patients starting dialysis 1975-
1977.

large number of drop-outs from treatment. These survival rates must be com-
pared with those of hemodialysis, but unfortunately patient selection precludes
reliable comparison. Only by use of techniques such as Cox's proportional
hazards model will it be possible to adjust for different patient populations [55].
However, for comparison, the three-year survival for all dialysis patients in the
50-59 age group reported by the Health Care Financing Administration is 57%
[56] and the figure for hemodialysis patients aged 55-64 in the Seattle program is
60%.
Despite patient survival which may be comparable to that with hemodialysis, a
large number of patients may convert from peritoneal to hemodialysis and so are
removed from the data analysis. Thus a better measure of the success of peri-
toneal dialysis as long term therapy is technique survival - comparable to graft
survival following transplantation. Technique survival reflects the rate at which
patients leave peritoneal dialysis because of death, transplantation, or conversion
to other forms of dialysis, and also is calculated by the life-table method. As
shown in Figure 4, technique survival steadily declined in the Seattle program,
and after 3 years only about one quarter of the patients remained on IPD. A
 189

similar high incidence of technique failure can be calculated from other reports.
For example, 15 of 36, 12 of 41, and 13 of 24 patients in various series [40, 57, 58]
were converted to hemodialysis.
The most frequent cause of technique failure in the Seattle series was inade-
quate dialysis, accounting for 19 of 28 (68%) conversions to hemodialysis. This
compares with 33% reported by others [40, 57]. Next most frequent reasons for
conversion were loss of support for home dialysis (14%) and frequent infections
(7%); pain during dialysis and patient preference accounted for less than 5% of
conversions. Gutman [40] noted malnutrition and hypoalbuminemia as causes for
conversion, but these are probably manifestations of inadequate dialysis, so
explaining the apparent discrepancy between his and the Seattle series.

3.2. Theoretical considerations regarding the cause of failure of intermittent per-

itoneal dialysis

Inadequate dialysis with IPD may result from several causes including peritonitis
with loss of membrane area, spontaneous loss of clearance, and decline of
residual renal function.
Although 40% of conversions from IPD to hemodialysis in the Seattle series
were preceded by at least one episode of culture-proven peritonitis, this does not
necessarily imply that the infection directly caused inadequate dialysis. Loss of
membrane surface area due to fibrosis and adhesion formation occurs rarely if
treatment of peritonitis is not unduly delayed [59]. Another possibility is spon-
taneous deterioration of peritoneal clearance as described by Finkelstein et al.
[60]. However, the main reason for conversion was failure of physician and/or
patient to increase dialysis time in proportion to the decline of residual renal
function.
To illustrate the mechanism underlying failure of IPD the conventional cleara-
nce approach is helpful:

minimum adequate clearance in ml/min C = Kdl~8td + K. (1)

where
Kd = dialyzer clearance in ml/min
td = dialysis time in h/wk
Kr = residual renal clearance in ml/min
168 = total hours in one week

Since discussion of the relationship of adequacy of dialysis to middle and small

molecular weight compounds is beyond the scope of this chapter, C has been
assumed to be 3 mVmin for middle molecules [61] and 5.5 ml/min for creatinine for
a 'standard' patient with body surface area of 1. 73 m2 as proposed by Boen et al.
[62].
190

Equation (1) can be restated as follows:

168
td = - P x Kr + P x C where p = K (2)
d

Equation (2) is expressed in graphic form in Figure 5. As Kr diminishes, td has to

be increased at a linear rate to maintain the same combined clearance C. The line
intercepts the abscissa (Kr) at C, indicating that no dialysis is needed (td = 0) when
renal function equals C. The line intersects the ordinate (td) at pC indicating that
pC hours of dialysis per week are required if Kr becomes zero. The slope of the
line (p) is determined by the membrane clearance. The lower the clearance, the
steeper the curve; p also represents the td increment needed for each ml/min
decline of K r.
Table 1 compares IPD and hemodialysis in terms of creatinine (Kd Cr) and
middle molecule (KdMM) clearances and their respective values for p. For each
1 mllmin decline in K" IPD will have to be increased by 9.3 h/week, whereas
hemodialysis will only have to be increased by 1.6 h/week. This is using creatinine
as the indicator of adequacy of dialysis.

--
-
"a

-
III
E
I II
III
>-
III
Q

Residual renal clearance (K r >, ml/min

Figure 5. Calculated relationship between residual renal clearance (Kr) and dialysis time (td) in
peritoneal dialysis patients.
 191

Table 1. Comparison of clearances with intermittent peritoneal dialysis (IPD) and hemodialysis (HD)

[Link] mllmin p Cr h/week' KdMM ml/min p MM hlweek'

IPD 18 9.3 7.2 23.3

HD 106 1.6 29.0 5.8
IPDIHD 0.17 5.9 0.25 40

• h/wk for each ml/min change.

The measured 1(. of 20 patients on IPD is plotted against their td in Figure 6.

This shows that td increased with declining Kr, but at a slow rate (p = 3.1). Tpis
increment of 3 h dialysis per week for each ml/min decline in Kr is only one third of
the calculated required change in td' Two important conclusions may be drawn
from this: 1) there is resistance by patients to dialyzing more than 40 h/week, and
2) when Kr reaches 2 mllmin or less, a patient becomes underdialyzed.
In a separate study the Kr of 5 patients treated by IPD was measured every 3
months from starting treatment until the time of conversion to hemodialysis. The
results are shown in Figure 7 and are based on the assumption that peritoneal
KdMM = 7.2 ml/min. The minimum adequate clearance (i.e., Dialysis Index = 1)
equals a total combined middle molecule clearance of 3 ml/min [40]. During the
first 9 months of IPD the clearance contributed by residual renal function (Kr) far
exceeded that from dialysis. Unfortunately, the Kr diminished rapidly from
4.7ml/min at the start of treatment to only 0.3 mllmin by the 15th month of IPD,
and the patients became underdialyzed within about a year of starting IPD. This
explains their conversion to hemodialysis shortly thereafter.
Consequently, we concluded that a reduction in Kr without sufficient adjust-
ment in td was the major cause for under dialysis and conversion of patients from
IPD to hemodialysis, and loss of membrane surface area probably is not a
60

50
•
-
~
• 30 •
-
•• • jf~,
.§ .,. 3")(1-
<II
20 " 3?~
'iii ,~
~
III 10 "c:....
c '~.t
0 2 3 4 5 6
Residual renal clearance (K r ), ml/min

Figure 6. Relationship between residual renal clearance (K,) and dialysis time (t d) in 20 patients
treated by peritoneal dialysis.
192

.Kd
CP
g
..
6
IV
IV
.!
u DKr
cp.5 4
- E
~::::: 3 ______ L~I~~ __
.!E
o
E 2

6 9 12 15 18
Duration on dialysis, month

Figure 7. The effect of duration of treatment on total middle molecular clearance (KdMM) in 5 Seattle
patients treated by peritoneal dialysis for 18 months.

significant factor for most patients. This experience provides a theoretical expla-
nation for the clinical impression that IPD is a satisfactory treatment for about
one year but usually is unsatisfactory for long-term use.

3.3. Possible reasons for past overoptimism about intermittent peritoneal dialysis

These disappointing results with long-term IPD are in striking contrast to some of
the early reports. There are several possible explanations for this.

3.3.1. Unsophisticated methodology

In most reports, follow-up of patients treated by IPD is short. Measures of
'success' such as 'average dialysis-months' are meaningless, but measures such as
cumulative survival rate, nutritional status, and cumulative drop-out rate have
not been used. Most importantly, clinical status of patients is not well defined,
and information is not usually available on residual renal function which pro-
foundly affects outcome. As a result, enthusiasm led to exaggerated claims, often
reflected in the titles and conclusions of papers, but not substantiated by the
reported results.

3.3.2. Biased patient selection

Selection for IPD usually favored elderly patients, often with multiple medical
problems. When such patients died it was taken for granted that death related to
factors other than dialysis, and potential problems with IPD were rarely con-
sidered. Death often occurred before significant changes could be detected in
indicators of underdialysis such as motor nerve conduction; and when patients
died within a year of starting IPD, the significant role of Kr was not appreciated.
 193

Also, many early IPD programs treated a predominance of females whose

smaller body size requires less td, so that underdialysis was less likely to occur.

3.3.3. Changes in dialysis regimens

In the early years patients accepted the need for prolonged peritoneal dialysis
treatment since hemodialysis with the Kiil dialyzer required 18 to 27 h of dialysis
weekly. Today, with hemodialysis time reduced to 9 to 15 h weekly, most physi-
cians and patients are reluctant to consider the 40 or more h per week required by
IPD. This may change, however, if dialysis is spread over most nights of the week
as in CCPD.

3.4. Important aspects of intermittent peritoneal dialysis

3.4.1. Adequacy of dialysis

Peritoneal dialysis has a lower clearance rate than hemodialysis, requiring longer
hours of treatment. This creates logistical and financial problems with out-patient
IPD, and resistance to undertake home IPD unless the patient is able to sleep
during dialysis. Efforts to reduce time by increasing peritoneal permeability have
not been of clinical value so far.
Nephrologists frequently have reassured themselves with the fallacious belief
that IPD clears middle and large molecules better than hemodialysis. This
erroneous assumption may lead to the conclusion that the serum creatinine level
is not a matter for concern provided the patient feels well. Belief that the
peritoneal membrane clears middle molecules better than a hemodialysis mem-
brane derived from the observation that IPD patients do not develop peripheral
neuropathy despite high serum creatinine levels. This, together with develop-
ments in hemodialysis, eventually led to formulation of the middle molecule
hypothesis [61]. In fact, IPD clears middle molecules better than hemodialysis
only when compared in terms of the clearance ratio of middle to small molecules.
As shown in Table 1, the ratio of KdCr for IPD compared to that of hemodialysis is
only 0.17, but the ratio for KdMM is 0.25. However, as far as patient well-being is
concerned, these ratios are not as important as absolute clearance. Although
CAPD with four exchanges daily results in a better clearance than hemodialysis
for molecules larger than 600 daltons, clearances with IPD for both middle and
small molecules will always be inferior to those of hemodialysis. Consequently,
when an IPD patient's residual renal function approaches zero underdialysis for
both small and middle molecules will result unless the patient is willing to dialyze
at least 50 h/week.
Unfortunately, manifestations of underdialysis in IPD patients are subtle and
often develop insidiously. The usual signs of uremia such as mucositis, pericar-
ditis, and neuropathy are encountered only infrequently. More commonly, un-
derdialysis results in anorexia leading to poor protein and calorie intake, muscle
194

wasting, generalized malaise, and cachexia. The serum creatinine level is high,
with a relatively low BUN-to-creatinine ratio, and the patient shrinks to a lesser
lean body mass compatible with the lower total clearance.
The phosphate clearance also becomes less than desirable with long-term IPD,
and patients often require higher doses of phosphate-binding agents, have higher
serum phosphate levels, and more frequently require parathyroidectomy. When
IPD patients are converted to CAPD, often the most impressive biochemical
improvements are the changes in serum creatinine and phosphate levels [63], and
there may also be improvement in serum chemistries and hematocrit [64].
Thus there are a number of potential obstacles to widespread use of IPD, but
inadequacy of dialysis is the major cause of serious problems and much more
important than peritonitis. In fact, the success of CAPD is achieved by obtaining
better clearances at the expense of an increase in the infection rate. Since the poor
clearance with IPD results from the nature of the peritoneal membrane itself,
manipulations such as use of vasodilators [65, 66] or hypo/hyperosmotic dialysate
cycles [67] would seem unlikely to be effective practical solutions. Because
residual renal function eventually will fall to less than 1 ml/min in almost all
patients, IPD is unlikely to be practical as a long-term treatment unless patients
are willing to accept the need for as much as 50 h dialysis weekly. This will require
systems for IPD while sleeping, and so CCPD may well prove to be the preferred
mode of peritoneal dialysis.

3.4.2. Exhaustion of vascular access sites

In 1974 Tenckhoff listed 'patients in whom reliable access for hemodialysis could
not be maintained' as an indication for IPD (68). Since then, improvements in
techniques have greatly reduced the number of patients with serious vascular
access problems (Fig. 8). Nine of 19 patients started on IPD in the Seattle program
[46] because of access problems subsequently were converted to hemodialysis.
Most of these started IPD in 1975, suggesting that either the initial impression
about their blood access was incorrect, or that subsequent improvements in
vascular surgery eventually enabled establishment of reliable blood access. Simi-
larly, 8 of 36 patients (22%) in Gutman's series were on IPD because of blood
access problems, but 3 of these subsequently converted to hemodialysis [40].
Vascular access now is a significant problem for only a very small number of
patients and, with ongoing advances, could almost cease to exist as an indication
for peritoneal dialysis.

3.4.3. Hemodynamic instability

Patients with an 'unstable' cardiovascular system characterized by severe coro-
nary artery disease, cardiomyopathy, serious arrhythmias or inadequate cardiac
output may have difficulty in tolerating hemodialysis, and have been reported to
do better with IPD [68-70]. Absence of rapid shifts of blood volume [71], a slower
ultrafiltration rate [72], lower acetate absorption and bicarbonate loss, slower
 195

EJ patients converted
to HemoDialysis

Years

Figure 8. The percentage of new patients started on peritoneal dialysis in Seattle because of access
problems, with the percentage later converted to hemodialysis.

osmotic changes [73], simultaneous glucose absorption [74], and relatively hypo-
tonic ultrafiltrate composition [75, 76] are all possible explanations for the smaller
swings of blood pressure during peritoneal dialysis.
However, only a relatively small number of patients has been reported who
have changed to IPD from hemodialysis because of vascular instability. Four of 91
patients (4 .3%) in Seattle and 1 of 36 (2.7%) over a 4-year period in Gutman's
series [40] were assigned to IPD because of cardiac problems. This suggests that
less than 1% of dialysis patients have severe cardiovascular .problems requiring
IPD. The availability of bicarbonate dialysis [77] , sequential ultrafiltration [78],
and other new techniques may make hemodialysis feasible even for these pa-
tients . Furthermore, the slower fluid removal during IPD may make extracellular
volume control more difficult in such patients, leading to poorer control of
hypertension , a factor with significant impact on patient survival [79, 80].

3.4.4. Thirst
Thirst is common in IPD patients, although often ignored, usually occurs late in
dialysis and/or immediately thereafter, and may lead to excessive water intake
and weight gain. Thirst is due to hypernatremia and hyperosmolarity late in
dialysis due to loss of hypoosmolar ultrafiltrate , and this problem was resolved by
lowering the dialysate sodium level to 118 mEq/L [81]. The dwell time recom-
mended by Tenckhoff in the early days of IPD was much longer than now used,
fluid removal was closer to iso-osmolar , and thirst was not usually a problem .
Disappearance of thirst may be one of the first differences noted by patients
following conversion from IPD to CAPO [82] .

3.4.5. Hematologic problems

Because systemic anticoagulation is unnecessary, IPD may be preferable to
196

hemodialysis for patients with active bleeding or potential bleeding complications

[68,83], and for those who refuse transfusion on religious grounds [68]. Fenton et
al. [84] found fewer bleeding complications in patients treated by IPD as com-
pared to those undergoing hemodialysis. Absence of obligatory blood loss may be
one reason for the relatively higher hematocrit and lower transfusion require-
ments of IPD patients [43, 83], but others have failed to find a significant
difference in hematocrit or hemoglobin level between hemodialysis and IPD
patients [44, 70]. In this context, although normal intestinal iron absorption has
been reported in IPD patients [85], those who do not receive iron supplementa-
tion have latent or overt iron deficiency, presumably due to increased iron losses
and insufficient dietary intake.
The greater peritoneal clearance for middle molecules has been used to explain
the more normal platelet aggregation function in IPD as compared to hemo-
dialysis patients [48], and Nenci et al. [86] have reported improved platelet
function with IPD or transplantation as compared with hemodialysis and untre-
ated uremia. However, improved platelet function has also been reported to
relate to more efficient dialysis, irrespective of modality [87]. Similarly, abnormal
lymphocyte function and a high incidence of circulating antinuclear antibodies
have been found in hemodialysis patients [88, 89] but not in IPD patients. More
attention to the underlying disease, residual renal function, and the details of
dialysis are needed to elucidate these issues further.

3.4.6. Home peritoneal dialysis

Home IPD with automated equipment or a cycler is simple to perform and does
not pose any immediate risks such as those associated with hemodialysis. Con-
sequently, with development of reverse osmosis systems, IPD became a practical
form of home dialysis [47, 84] which could be performed alone and which
encouraged patient independence. In contrast, it is preferable that a home
hemodialysis patient have someone with at least minimal knowledge of dialysis
available during treatment [90]. Because of the simplicity of IPD, small children
and elderly patients are able to perform their own dialysis.
With home IPD it is important the patient sleep during overnight dialysis
because of the prolonged time necessary. Automated equipment made this
possible, and many patients dialyzed three times weekly while sleeping during
most of the dialysis. However, when residual renal function declines the neces-
sary hours of dialysis may entail treatment more frequently than three times
weekly. This is perfectly feasible to do, although in the United states problems
with Medicare payment ensue because reimbursement is based on three-times-
weekly dialysis unless there is medical justification for more frequent dialysis.
Recently, with the rapid growth of CAPD and decline in the use of IPD,
manufacturers have ceased making reverse osmosis peritoneal dialysis equip-
ment. Consequently, IPD (and CCPD) now uses cyclers for which the tubing sets
and fluids are very expensive. Before CCPD and nightly IPD become more
 197

widely used, the cost of these supplies will have to be reduced significantly.
The cost of home IPD, home hemodialysis, and CAPD are similar in the
United States, ranging from $65 to $110 per dialysis (converting CAPD to the
equivalent of three-times-weekly dialysis) depending upon the equipment, di-
alyzer reuse in the case of hemodialysis, and whether a paid home dialysis helper
is provided. The cost of supplies alone for CCPD presently is approximately $ 400
per week as compared with $ 90 per week for home hemodialysis supplies.

3.4.7. Logistics of out-patient intermittent peritoneal dialysis

The long duration of IPD results in problems for an out-patient dialysis unit. An
out-patient peritoneal dialysis at the Northwest Kidney Center in Seattle costs
about 90% more than an outpatient hemodialysis because of the increased cost
for space, staff, equipment and supplies. Because of these problems, IPD in a
facility is best restricted to those patients for whom it is clearly the most appropri-
ate treatment. Such patients include those with blood access or severe car-
diovascular problems, those maintained on outpatient IPD while their catheter
becomes fixed prior to training for home IPD or CAPD, and also home IPD
patients with problems requiring backup dialysis.
It has been suggested that peritoneal dialysis is preferable to hemodialysis for
patients with diabetic retinopathy. So far, this has not been proved, and hemo-
dialysis using low-dose heparin is reported to be very successful [91]. However, if
it is shown that peritoneal dialysis is in fact better than hemodialysis for these
patients, there will be need for a significant increase in out-patient IPD capability
for diabetic patients unable to dialyze at home using IPD or CAPD. This could be
an expensive undertaking.

3.4.8. Intermittent peritoneal dialysis in the elderly

In addition to its use in elderly patients with serious cardiovascular problems, the
relative simplicity of IPD permits some elderly patients to dialyze at home who
might be incapable of safe home hemodialysis and who do not wish to undertake
CAPD. This has been a frequent reason for use of home IPD in Seattle. Patient
survival with all forms of dialysis is age-dependent [92]; to date, based on a small
number of patients, survival with IPD does not appear strikingly different from
that in comparable elderly patients treated by hemodialysis [19].

3.4.9. Growth, maturation, and intermittent peritoneal dialysis in pediatric

patients
Prior to CAPD, IPD proved particularly attractive to those caring for small
children with end-stage renal disease [93, 94]. This was because it did not require
vascular access and could be more readily carried out at home with less disruption
to family routine and school attendance. However, the few reported results have
been disappointing. Many pediatric patients were troubled by hypertension and
hyperphosphatemia with development of severe renal osteodystrophy [93].
198

Growth was poor, linear growth averaging only 37% of that expected [95]. In
addition, one report has shown that less than 10% of pediatric patients remain on
IPD long-term [95]. Such a high conversion rate can hardly be considered a
success. More recently CAPD has been successfully used for treatment of
children.

3.4.10. Peritonitis
Peritonitis is the most frequent complication of peritoneal dialysis, but its inci-
dence has been considerably reduced with improvements in aseptic technique and
development of indwelling catheters and closed-system automated equipment.
The incidence rate of infection in IPD is 0.2% to 1% [39], with skin and intestinal
flora and contaminated dialysis fluid as the commonest sources of infection [96,
97]. Occasional epidemics of peritonitis due to uncommon bacteria such as
mycobacterium chelonei have occurred, the usual source of such infection being
the water supply.
Peritonitis and its treatment are discussed elsewhere. One point worth stress-
ing is that the rate of peritonitis is related to the frequency of exposure - that is,
opening and closing of the system. This holds true for both bacterial and fungal
infections [98]. Consequently, the infection rate with CAPD is always several
times that with IPD; the infection rate with CCPD probably will fall between
these two rates.

3.4.11. Protein loss and malnutrition

The problems of nutrition with IPD were never adequately resolved. Data on
protein loss during peritoneal dialysis have been inconsistent [99], varying from
8.5 g/day with a short dialysis [96] to 17.9 g/day with CAPD [97]. With IPD,
reported protein loss has varied from 6 g/day [100] to 300 g/day [101]. Protein loss
depends on dialysate composition, technique, and dwell time [99]. During IPD,
using standard solutions and dwell time, Seattle experience was that daily protein
loss ranged from 4 to 22.3 g, averaging 16.4 g [102].
The adequately dialyzed peritoneal dialysis patient who is not infected and who
has a good appetite can compensate for protein loss with a diet containing 1 to
1.5 g protein/kg body weight/day. However, dietary adjustment becomes critical
with a catabolic event such as infection or administration of steroids, and the
serum protein level should then be monitored closely [103]. Similarly, as residual
renal function declines, a patient may require increased protein intake to avoid
malnutrition which may further increase the dialysis requirement.

3.4.12. Abormallipid metabolism

Whether there is a difference in serum triglyceride levels between patients on
peritoneal and hemodialysis remains uncertain. Hypertriglyceridemia is common
in uremic patients and has been reported to be a problem in IPD patients [104].
Cattran et al. [105] found a mean triglyceride level of 313 ± 57 mg% in IPD
 199

patients, a level which did not change significantly during dialysis. In contrast,
hemodialysis patients had a mean triglyceride level of 195 ± 19 mg% predialysis,
falling to 118 ± 10 mg% after dialysis. This might be due to heparin. Other
investigators have not found this disparity between IPD and hemodialysis pa-
tients [43]. Weisinger has reported data showing glucose intolerance and marked
insulin response to a glucose load in IPD patients as compared to hemodialysis
patients [106].

3.4.13. Osteodystrophy
A lower incidence of osteomalacia and absence of fractures and soft tissue
calcification in IPD patients compared to hemodialysis patients has been reported
[107, 108], but a subsequent review comparing 24 hemodialysis and 41 IPD
patients showed the latter to have more bone problems [83]. Renal osteodystro-
phy should be similar in well-dialyzed IPD and hemodialysis patients if serum
phosphate levels are well controlled and a dialysate containing sufficient calcium
is used.

3.4.14. Diabetic nephropathy and intermittent peritoneal dialysis

In recent years there has been a great increase in the number of patients with
diabetic nephropathy treated by dialysis. In the United States about 25% of all
new dialysis patients have diabetic end-stage renal disease [109]. Several papers
have discussed the supposed benefits of IPD for diabetic patients [110-112]' but
these reports generally included only a small number of patients studied over a
relatively short period of time. The supposed advantages and disadvantages of
peritoneal over hemodialysis for the diabetic patient are shown in Table 2. CAPD
is now used frequently for treatment of uremic diabetic patients and IPD is only
rarely used. Consequently, little recent information on the latter is available.

(a) Retinopathy. The most important benefit claimed for IPD was stabilization of
diabetic retinopathy and retention of existing vision [110-113]. Anticoagulation,
hemodynamic instability and retinal ischemia have been blamed for progression
of retinopathy and development of vitreous hemorrhages in hemodialyzed dia-

Table 2. Supposed advantages and disadvantages of peritoneal dialysis in diabetic end-stage renal
diasese

Advantages Disadvantages

Reduced risk of retinopathy Poor control of extracellular volume

Slower progression of neuropathy Difficulty with blood sugar control
No need for vascular access Loss of protein and malnutrition
Less hemodynamic stress Increased risk of infection
Poorer urea and creatinine clearance
200

betic patients [71]. However, doubt has been raised about the possible relation-
ship of these factors to progression of retinopathy [114, 115]. Mitchell et al. [113]
originally reported no progression of retinopathy in a small number of diabetic
patients treated by IPD, but with longer follow-up and a larger number of
patients, they noted progression in 20% of patients treated by IPD [116], under-
lining the importance of adequate sample size and sufficient follow-up. Together
with stabilization of vision, there have also been reports of improvements in
vision in a small number of diabetics treated by IPD [113, 116, 117].
Data on retinopathy in diabetic patients treated by hemodialysis is similarly
controversial. Rubin and Friedman [112], in a review of 340 diabetics treated by
hemodialysis, calculated that 33% showed deterioration of vision. However, this
has been less in diabetic patients treated by hemodialysis in recent years [115, 118],
probably as a result of lower doses of heparin. It has also been suggested that
retinopathy worsens during the second year on dialysis, regardless of the mode of
treatment [119]. There is need for a controlled study with long-term follow-up to
answer the question whether IPD (or CAPD) is a better treatment for diabetic
patients.

(b) Vascular access. Difficulty in maintaining vascular access because of wide-

spread vascular disease has also prompted the suggestion that IPD may be
preferable for diabetic patients. However, recent literature suggests that the
number of diabetics with serious vascular access problems is small. In 28 such
patients followed for 238 patient-months, Jacobs et al. [115] found that more than
60% required only one episode of access surgery, and only 2 patients had to be
converted to IPD because of blood access problems. Use of bovine and synthetic
grafts may minimize this problem, and Kassissie et al. [120] have reported that
bovine grafts have significantly longer life than primary arteriovenous fistulae in
diabetic patients.

(c) Cardiovascular problems and hemodynamic stress. Diabetes frequently is

complicated by cardiovascular involvement, and the hemodynamic stress of wide
swings of blood pressure during hemodialysis might prove a problem for some
patients. It has been suggested that diabetic patients may be more comfortable
and have less dialysis-related morbidity if treated by IPD .. However, problems of
volume control with IPD [117-121] may result in chronic extracellular volume
excess with left ventricular hypertrophy and hypertension.

(d) Peritonitis, blood sugar control, neuropathy, pericarditis, and peritoneal clear-
ance. Because diabetic patients are particularly susceptible to infections, it was
thought that IPD might prove impractical for diabetic patients. With improve-
ments in technique, it has become clear that the incidence of peritonitis in diabetic
and non-diabetic IPD patients is similar [98, 115, 116].
Adequate blood sugar control may be difficult to achieve during IPD in a
 201

diabetic patient. Blood glucose level is dependent on the dialysate dextrose

concentration and is quite variable from patient to patient. Blood sugar control
can be achieved either by addition of insulin to the dialysate [122] or by a
supplementary subcutaneous injection of insulin when using a dextrose concen-
tration of 2 g% or greater [123]. Stoddard and Gallagher [123] found 4-5 units of
subcutaneous regular insulin for 2 g% dextrose, 8-10 units for 3 g% dextrose, and
12-15 units for 4.25 g% dextrose dialysate to be effective. Oreopoulos found that
4-6 units of crystalline insulin per liter of 1.5 g% dextrose and 8-10 units per liter
of 4.25 g% dextrose dialysate also provided adequate blood sugar control [110]. It
has been general experience that blood glucose control has been much better with
use of intraperitoneal insulin.
Initial reports of slower progression of peripheral neuropathy in diabetic
patients treated by IPD have been questioned [116-121]' This needs further
evaluation with longer follow-up and consideration of residual renal function
[124]. Similarly, the incidence of pericarditis [118-121] and osteodystrophy needs
further evaluation. A report of poorer peritoneal urea and creatinine clearances
in diabetics [125] has been disputed [114].

(e) Survival. Information on survival of diabetic patients treated by IPD is

available only for a small number of patients with short follow-up, making
interpretation difficult. Nevertheless, overall survival of diabetic patients is poor
[118-121]. Twenty-five percent mortality after one year in 12 diabetic patients
treated by IPD was reported by Rubin and Friedman [112], and 5 of 10 patients
who died in the series of Blumenkrantz et al. [114] did so within 6 months of
starting treatment. A similar high mortality rate at 6 months was reported by
Mitchell et al. [113] with only 52% cumulative survival on IPD at 6 months
compared with 80% for hemodialyzed diabetics; however, survival for both IPD
and hemodialysis patients was equal at one year. Quelhorst et al. [121] found no
difference in survival rates of 16 IPD and 16 hemodialyzed diabetic patients over a
period of 21 months.
Analysis of 32 diabetic patients treated by IPD in Seattle showed poorer
survival at 2 years as compared with 54 diabetic patients treated by hemodialysis.
Two-year survival was 39% for IPD patients and 52% for hemodialysis patients.
The I-year survival of 65% for diabetic patients treated by IPD was comparable to
that reported by Mion et al. [117] and Quelhorst et al. [121], and better than that
reported by Mitchell [113]. However, considerable caution must be used in
comparing different series and in comparing IPD with hemodialysis patients at
the same center because of likely disparaties in age and severity of the diabetes
and its complications. Because of the poor 2-year survival of diabetics, whether
treated by hemodialysis or IPD, attention might better be focused on improving
survival generally rather than on any minor improvement in eyesight or question-
able differences in the incidence of osteodystrophy or neuropathy with the two
treatments.
202

4. The future of intermittent peritoneal dialysis

4.1. The impact of other forms of peritoneal dialysis on intermittent peritoneal

dialysis

The rapid increase in use of CAPD and recent introduction of CCPD have had
profound effects on the utilization of IPD. In some ways the situation is reminis-
cent of the 1960s when development of IPD was overshadowed and hampered by
the rapidly developing popularity of hemodialysis. Ironically, the concept and
practice of CAPD are exactly contrary to the objectives of developments in IPD
over the years - automation, sophistication, safety, and sterility. CAPD utilizes
manual methods rather than automation, and overcomes the inefficiency of IPD
by continuous dialysis. In addition to replacing IPD as the usual form of per-
itoneal dialysis, CAPD has also had a negative effect on the technology of IPD.
With automated reverse osmosis equipment no longer available, IPD has to be
performed using cyclers which require more patient time, entail more breaks in
the closed system, and are more expensive. Despite these problems, a small
number of patients continue to be treated by IPD.

4.2. Who should be treated by intermittent peritoneal dialysis?

IPD may be the only alternative available for patients who do not wish to be
treated by or who are medically unsuitable for hemodialysis, and who dislike the
daily rituals of CAPD and CCPD. However, in order to avoid early failure
because of the low clearance and long hours inherent in IPD, patients selected for
this treatment should fulfill the following criteria:
1. Body weight should be less than 55 kg or surface area less than 1.40 m2 • This
gives some reassurance that when Kr eventually approaches zero, clearance
may still be adequate, even if the patient refuses to dialyze more than 40 hi
week.
2. The patient should be able to dialyze at home. IPD is essentially a home
dialysis regimen, and to occupy an out-patient dialysis station for more than
40 hlweekly is too costly in terms of utilization of space, staff, time, and
ancillary services.
3. The patient should be able to dialyze at night while sleeping. Dialyzing while
asleep is a relatively unique benefit of IPD. It is also a necessity because dialysis
during the day means spending 50% of the patient's waking hours on dialysis,
clearly detrimental to both rehabilitation and other activities.
 203

4.3. Outlook for the future of IPD

IPD may still be the choice of a small number of patients, but the future of IPD is
likely to be related to further developments of its variant, CCPD. CCPD provides
good clearances while permitting nightly dialysis during sleep and may become
the preferred mode of peritoneal dialysis in the future if the cost can be reduced
sufficiently to make this practical.

Acknowledgements

The authors wish to express their continuing gratitude to Nancy Gallagher, R.N.,
and to Henry Tenckhoff, M.D., for their efforts in making peritoneal dialysis a
successful program in Seattle, and also to Marilynn Suthergreen for her assistance
in preparation of this manuscript.

References

1. Wegner G: Chirugische Bermerkungen iiber die Peritonealhohle mit besonderer Beriick-

sichtigung der Ovariotomie. Langenbecks Arch Chir 20: 51-145, 1877.
2. Ganter G: Uber die Beseitigung giftiger Stoffe aus dem B1ute durch Dialyse. Munch med Wschr
70: 1478-1480, 1923.
3. Fine J, Frank HA, Seligman AM: The treatment of acute renal failure by peritoneal irrigation.
Ann Surg 124: 857-878, 1946.
4. Odel HM, Ferris DO, Power MH: Peritoneal lavage as an effective means of extrarenal
excretion. Am J Med 9: 63-77,1950.
5. Boen ST: Peritoneal Dialysis in Clinical Medicine. Charles C Thomas, Springfield, Illinois,
1964.
6. Maxwell MH, Rockney RE, Kleeman CR, Twiss RL: Peritoneal dialysis. 1. Technique and
applications. JAMA 170: 917-923, 1959.
7. Quinton W, Dillard D, Scribner BH: Cannulation of blood vessels for prolonged hemodialysis.
Trans Am Soc Artif Intern Organs 6: 104-113, 1960.
8. Boen ST, Mulinari AS, Dillard DH, Scribner BH: Periodic peritoneal dialysis in the manage-
ment of chronic uremia. Trans Am Soc Artif Intern Organs 8: 256-262, 1962.
9. Boen ST, Mion CM, Curtis FK, Shilipetar G: Periodic peritoneal dialysis using the repeated
puncture technique and an automatic cycling machine. Trans Am Soc Artif Intern Organs 10:
409-414, 1964.
10. Palmer RA, Newell JE, Gray EJ, Quinton WE: Treatment of chronic renal failure by prolonged
peritoneal dialysis. N Engl J Med 274: 248-254, 1966.
11. Tenckhoff H, Schechter H: A bacteriologically safe peritoneal access device. Trans Am Soc
Artif Intern Organs 14: 181-187,1968.
12. Striker GE, Tenckhoff HA: A transcutaneous prosthesis for prolonged access to the peritoneal
cavity. Surgery 69: 70-74, 1971.
13. Tenckhoff H, Shilipetar G, van Paasschen WH, Swanson E: A home peritoneal dialysate
delivery system. Trans Am Soc Artif Intern Organs 15: 103-107, 1969.
14. Tenckhoff H, Meston B, Shilipetar G: A simplified automatic peritoneal dialysis system. Trans
Am Soc Artif Intern Organs 18: 436-440, 1972.
204

15. Popovich RP, Moncrief JW, Decherd JF, Bomar JB, Pyle WK: The definition of a novel
portable/wearable equilibrium peritoneal dialysis technique. Abstr Trans Am Soc Artif Intern
Organs 5: 64.
16. Scribner BH: A current perspective on the role of intermittent versus continuous ambulatory
peritoneal dialysis in the treatment of chronic renal failure. Renal Physicians Association
Northeastern Meeting Proceedings 3: 76-81,1979.
17. Diaz-Buxo JA, Farmer CD, Walker PJ, Chandler JT, Holt KL: Continuous cyclic peritoneal
dialysis: a preliminary report. Artif Organs 5: 157-161,1981.
18. Jacobs C, Brunner FP, Chantler C, Donckerwo1cke RA, Gurland HJ, Hathway RA, Selwood
NH, Wing AJ: Combined report on regular dialysis and transplantation in Europe, VII, 1976.
Proc Eur Dial Transplant Assoc 14: 3-69,1977.
19. Brunner FP, Brynger H, Chantler C, Donckerwo1cke RA, Hathway RA, Jacobs C, Selwood
NH, Wing AJ: Combined report on regular dialysis and transplantation in Europe, IX, 1978.
Proc Eur Dial Transplant Assoc 16: 2-73, 1979.
20. Brynger H, Brunner FP, Chantler C, Donckerwo1cke RA, Jacobs C, Kramer P, Selwood NH,
Wing AJ: Combined report on regular dialysis and transplantation in Europe, X, 1979. Proc Eur
Dial Transplant Assoc 17: 2-86, 1980.
21. Health Care Financing Administration: End-stage renal disease program highlights 1982. Health
Care Financing Administration, Baltimore, Maryland, 1983.
22. Vaamonde CA: Peritoneal dialysis. Current status. Postgrad Med 62: 148--156, 1977.
23. Firmat J, Zucchini A: Peritoneal dialysis in acute renal failure. Contrib Nephrol17: 33-38, 1979.
24. Vaamonde CA, Michael UF, Metzger RA, Carroll KE, Jr.: Complications of acute peritoneal
dialysis. J Chron Dis 28: 637--659,1975.
25. Roxe DM, Argy WP, Frost B, Kersin J, Schreiner GE: Complications of peritoneal dialysis.
South Med J 69: 584--587, 1976.
26. Kjellstrand CM, Pru CE, Jahnke WR, Davin TD: Acute renal failure. In: W Drukker, FM
Parsons, JF Maher (eds). Replacement of Renal Function by Dialysis (2nd ed). Martinus
Nijhoff, Boston, 1983, pp 536-568.
27. Valk TW, Swartz RD, Hsu CH: Peritoneal dialysis in acute renal failure: analysis of outcome
and complications. Dial Transpl 9: 48--54, 1980.
28. Kanter A, Nadler N, Verte! RM et al.: Peritoneal dialysis: indications and technique in the
surgical patient. Surg Clin N Am 48: 47-55, 1968.
29. Finn R, Jowett EW: Acute hydrothorax complicating peritoneal dialysis. Br Med J 2: 94,1970.
30. Lorentz WB (1979) Acute hydrothorax during peritoneal dialysis. J Pediatr 94: 417-419.
31. Mathew TH: Comparison of peritoneal and haemodialysis in acute renal failure. In: RC Atkins,
NM Thomson, PC Farrell (eds), Peritoneal Dialysis. Churchill Livingstone, Edinburgh, 1981, pp
80-86.
32. Posen GA, Luisello J: Continuous equilibration peritoneal dialysis in the treatment of acute
renal failure. Perit Dial Bull 1(2) : 6-7, 1980.
33. Sipkins JH, Kjellstrand CM: Severe head trauma and acute renal failure. Nephron 28: 36-41,
1981.
34. Tzamaloukas AH, Garella S, Chazan JA: Peritoneal dialysis for acute renal failure after major
abdominal surgery. Arch Surg 106: 639-643,1973.
35. Donckerwo1cke RA, Chankter C, Broyer MJC: Pediatric dialysis. In: Drukker W, Parsons FM,
Maher JF (eds), Replacement of Renal Function by Dialysis (2nd ed). Martinus Ni jhoff, Boston,
1983, pp 514--535.
36. Stott RB, Ogg CS, Cameron JS, Bewick M: Why the persistently high mortality in acute renal
failure? Lancet 2: 75-79,1972.
37. Kleinknecht D, Jungers P, Chanard J, Barbanel C, Ganeval D: Uremic and non-uremic
complication in acute renal failure: evaluation of early and frequent dialysis on prognosis.
Kidney Int 1: 190-196, 1972.
 205

38. Schmidt RW, Blumenkrantz MJ: IPD, CAPD, CCPD, CRPD - peritoneal dialysis: past,
present and future. Int J Artif Organs 4: 124-129, 1981.
39. Boen ST: Overview and history of peritoneal dialysis. Dial Transpl6: 12-18, 1977.
40. Gutman RA: Automated peritoneal dialysis for home use. Q J Med 47: 261-280, 1978.
41. Karanicolas S, Oreopoulus DG, Pylypchuk G, Fenton SSA, Cattran DC, Rapoport A, de Veber
GA: Home peritoneal dialysis: 3 years experience in Toronto. Can Med Assoc J 116: 266-269,
1977.
42. Sherrard DJ, Tenckhoff H, Curtis FK, Casaretto A: Advances in peritoneal dialysis. Kidney Int
(Suppl) 3: 422-424, 1975.
43. Roxe DM: Comparisons of maintenance peritoneal and hemodialysis. Dial Transpl 7: 800--802,
1978.
44. von Hartitzch B: Chronic peritoneal dialysis. A regime comparable to conventional dialysis.
Dial Transpl 7: 805-844, 1978.
45. Price JD, Ashby KM, Reeve CE: Results of 12 years' treatment of chronic renal failure by
dialysis and transplantation. Can Med Assoc J 118: 263-266, 1978.
46. Ahmad S, Gallagher N, Shen F: Intermittent peritoneal dialysis: status reassessed. Trans Am
Soc Artif Intern Organs 25: 86-89, 1979.
47. Diaz-Buxo JA, Chandler JT, Farmer CD, Smith DL: Chronic peritoneal dialysis at home - a
comparison with hemodialysis. Trans Am Soc Artif Intern Organs 23: 191-193, 1977.
48. Lindsay RM, Friesen M, Koens F, Linton AL, Oreopoulus D, deVeber G: Platelet function in
patients on long term peritoneal dialysis. Clin Nephrol 6: 335-339, 1976.
49. Blumenkrantz MJ, Curtis FK, Chan K, Gandhi VC, Smith BN, Barbour GL, Gutman RA,
Ivanovich PT, Goffinet JA, Murawski BG: Initiation of a cooperative study comparing the
clinical effectiveness of peritoneal dialysis and hemodialysis. Dial Transpl6: 64-67, 1977.
50. Roxe DM, delGreco F, Krumlovsky F, Gantous W, Hughes J, Ivanovich P, Quintanilla A,
Salkin M, Stone N: A comparison of maintenance hemodialysis to maintenance peritoneal
dialysis in the management of end-stage renal disease. Trans Am Soc Artif Int Organs 25: 81-85,
1979.
51. Diaz-Buxo JA, Chandler JT, Farmer DC, Smith DL: Chronic peritoneal dialysis at home - a
comparison with hemodialysis. Trans Am Soc Artif Organs 23: 191-193,1977.
52. Diaz-Buxo JA, Haas VF: The influence of automated peritoneal dialysis in an established
dialysis program. Dial Transpl 8: 531-533, 1979.
53. Marai A, Rathaus M, Gibor Y, Bernheim J: Chronic dialysis in the elderly: intermittent
peritoneal dialysis or hemodialysis? Perit Dial Bull 3: 183-186, 1983.
54. Cutler SJ, Ederer F: Maximum utilization of the life table method in analyzing survival. J
Chronic Dis 8: 699-712, 1958.
55. Cox DR: Regression models and life tables (with discussion). JR Stat Soc (B) 34: 187-220, 1972.
56. Eggers PW, Connerton R, McMullan M: The Medicare experience with end-stage renal disease:
trends and incidence, prevalence and survival. Health Care Financing Review 5: 69-88, 1984.
57. Ghantous WN, Salkin MS, Adelson BH, Ghantous S, McGinnis K, Valenziano A, Cronin M:
Limitations of peritoneal dialysis (PD) in the treatment of ESRD patients. Trans Am Soc Artif
Intern Organs 25: 100-103, 1979.
58. Gastaldi L, Baratelli L, Cassani B, Cinquepalmi M, Martegani M: Peritoneal dialysis for renal
failure. Lancet 2: 197-198,1979.
59. RubinJ, Arfania D, Nolph KD, Prowant B, Fruto L, Brown P, Moore H: Peritoneal clearances
after 6-12 months on continous ambulatory peritoneal dialysis. Trans Am Soc Artif Intern
Organs 25: 104-109, 1979.
60. Finkelstein FO, Kliger AS, Bastl C: Sequential clearance and dialysance measurements in
chronic peritoneal dialysis patients. Nephron 18: 342-347, 1977.
61. Babb AL, Strand MJ, Uvelli DA, Milutinovic J, Scribner BH: Quantitative description of
dialysis treatment: a dialysis index. Kidney Int (Suppl) 2, 23-29, 1975.
206

62. Boen ST, Haagsma-Schouten WAG, Birnie RJ: Long-term peritoneal dialysis and a peritoneal
dialysis-index. Dial Transpl 7: 377-380, 1978.
63. Popovich RP, Moncrief JW, Nolph KD, Ghods AJ, Twardowski EJ, Pyle WK: Continuous
ambulatory peritoneal dialysis. Ann Intern Med 88: 449-456, 1978.
64. Madden MA, Zimmerman SW, Simpson DP: Longitudinal comparison of intermittent versus
continuous ambulatory peritoneal dialysis in the same patients. Clin Nephrol16: 293-299, 1981.
65. Nolph KD, Ghods AJ, VanStone J, Brown PA: The effects of intraperitoneal vasodilators on
peritoneal clearances. Trans Am Soc Artif Intern Organs 22: 586-594, 1976.
66. Gutman RA: Toward enhancement of peritoneal clearance. Dial Transpl8: 1072-1099, 1979.
67. Zelman A, Gisser 0, Whittam PJ, Parson RH, Schuyler R: Augmentation of peritoneal dialysis
efficiency with programmed hyper/hypoosmotic dialysates. Trans Am Soc Artif Intern Organs
23: 203-209, 1977.
68. Tenckhoff H: Peritoneal dialysis today: a new look. Nephron 12: 420-436, 1974.
69. Oreopoulus DG: Peritoneal dialysis is reinstated. J Dial 2: 295-310, 1978.
70. Oreopoulos DG: Overall experience with peritoneal dialysis. Dial Transpl 7: 783-787, 1978.
71. Nolph KD: Chronic peritoneal dialysis in a patient with diabetes mellitus and heart disease.
Kidney Int 15: 698-708, 1979.
72. Nolph KD, Popovich RP, Ghods AJ, Twardowski Z: Determinants of low clearances of small
solutes during peritoneal dialysis. Kidney Int 13: 117-123, 1978.
73. Nolph KD: Peritoneal dialysis. In: W Drukker, FM Parsons, FJ Maher (eds), Replacement of
Renal Function by Dialysis (1st ed). Martinus Nijhoff, The Hague, 1978, pp 277-321.
74. Nolph KD, Rosenfeld PS, Powell JT, Danforth E Jr.: Peritoneal glucose transport and hyper-
glycemia during peritoneal dialysis. Am J Med Sci 259: 272-281, 1970.
75. Ahearn OJ, Nolph KD: Controlled sodium removal with peritoneal dialysis. Trans Am Soc Artif
Intern Organs 28: 423-428, 1972.
76. Nolph KD: Short dialysis, middle molecules, and uremia. Ann Intern Med 86: 93-97, 1977.
77. Graefe U, Milutinovich J, Follete WC, Vizzo JE, Babb AL, Scribner BH: Less dialysis induced
morbidity and vascular instability with bicarbonate in dialysate. Ann Intern Med 88: 332-336,
1978.
78. Bergstrom J, Asaba H, Furst P, Oules R: Dialysis, ultrafiltration, and blood pressure. Proc Eur
Dial Transpl Soc 13: 293-305, 1976.
79. Haire HM, Sherrard OJ, Scardapane AE, Curtis FK, Brunzell JD: Smoking, hypertension and
mortality in a maintenance dialysis population. Cardiovasc Med 3: 1163-1168, 1978.
80. Goodman WG, Haas LB, Sherrard DG: Left ventricular hypertophy and dialysis survival. Clin
Res 27: 61A (Abstract), 1969.
81. Shen FH, Sherrard OJ, Scollard D, Merritt A, Curtis FK: Thirst, relative hypernatremia, and
excessive weight gain in maintenance peritoneal dialysis. Trans Am Soc Artif Intern Organs 24:
142-145, 1978.
82. Oreopoulos DG, Khanna R, McCready W, Katirtzoglou A, Vas S: Continuous ambulatory
peritoneal dialysis in Canada: Dial Transpl 9: 224-226, 1980.
83. Blumenkrantz MJ, Lindsay RM: Co,?parisons of hemodialysis and peritoneal dialysis: a review
of the literature. Contrib Nephrol17: 20-29, 1979.
84. Fenton SSA, Cattran DC, Barnes NM, Waugh KJ: Home peritoneal dialysis: a major advance in
promoting home dialysis. Trans Am Soc Artif Intern Organs 23: 194-200,1977.
85. Milman N, Christensen T, Bartels U, Larsen L: Iron absorption and iron status in patients with
chronic uremia on regular peritoneal dialysis. Acta Med Scand 205: 629-635, 1979.
86. Nenci GG, Berrettini M, Agnelli G, Parise P, Buoncristiani U, Ballatori E: Effect of peritoneal
dialysis, haemodialysis and kidney transplantation on blood platelet function. I. Platelet ag-
gregation by ADP and epinephrine. Nephron 23: 287-292, 1979.
87. Lindsay RM, Friesen M, Aronstam A, Andrus F, Clark WF, Linton AL: Improvement of
platelet function by increased frequency of hemodialysis. Clin Nephrol1O: 67-70, 1978.
 207

88. Atkins RC, Holdsworth SR, Fitzgerald MG, Kraft NE, Hosking CS: The contrasting effect of
haemodialysis and peritoneal dialysis on lymphocyte function. Abstr Int Soc Nephrol, 1978.
89. Nolph KD, Sharp G, Ghods AJ: Antinuclear antibodies in chronic dialysis and renal transplanta-
tion. Proc 10th Ann Contactors' Conf, DHEW publ NIH 77-1442: 56-57, 1977.
90. Blagg CR: Seattle experience with peritoneal dialysis. Dial Transpl 7: 790-791, 1978.
91. Shapiro F, Umen A: Risk factors in hemodialysis patient survival. ASAIO J 6: 176-184, 1983.
92. Blagg CR, Wahl P: The Northwest Kidney Center: a statistical report, June 30th, 1977.
Northwest Kidney Center, Seattle, 1978.
93. Hickman R: Nine years' experience with chronic peritoneal dialysis in childhood. Dial Transpl7:
803,1978.
94. Potter DE: Comparison of peritoneal dialysis and hemodialysis in children. Dial Transpl7: 800-
802,1978.
95. Brouhard BH, Berger M, Cunningham RJ, Petrusick T, Allen W, Lynch RE, TravisLB: Home
peritoneal dialysis in children. Trans Am Soc Artif Intern Organs 25: 90-94,1979.
96. DeSanto NG, Cirillo D, Senatore R, Cicchetti T, Manzo M, Capasso G, Giodarno C: Compar-
ing schedules of daily peritoneal dialysis. J Dial 4: 311-324, 1978.
97. DeSanto NG, Cirillo D, Giodarno C: Peritoneal dialysis seven days a week. Int J Artif Organs 1:
154-156,1978.
98. Kraus ES, Spector DA: Characteristics and sequelae of peritonitis in diabetics and nondiabetics
receiving chronic intermittent peritoneal dialysis. Medicine 62: 52-57, 1983.
99. Giordano C, DeSanto NG: Dietary management of patients on peritoneal dialysis. Contrib
Nephrol17: 77-92,1979.
100. Pauli HG, Buttikofer E, Vorburger C: Clinical experience with peritoneal dialysis. Helv Med
Acta 33: 51-58, 1966.
101. Fretheim B, Selvaag 0: Peritoneal irrigation in uremia: report of 3 cases. Acta Chir Scand 96:
461-473, 1948.
102. Lindner A, Tenckhoff H: Nitrogen balance in patients on maintenance peritoneal dialysis. Trans
Am Soc Artif Intern Organs 16: 255-259, 1970.
103. Boen ST: Undesirable effects of chronic peritoneal dialysis. Contrib Nephrol17: 73-76, 1979.
104. Oreopoulos DG, Karanicolas S, Izatt S, deVeber GA: Dialysis and triglycerides. Ann Intern
Med 85: 679-680, 1976.
105. Cattran DC, Fenton SSA, Wilson DR, Steiner G: Defective triglyceride removal in lipemia
associated with peritoneal dialysis and hemodialysis. Ann Intern Med 85: 29-33, 1976.
106. Weisinger JR: Comparison of the effect of chronic hemodialysis and peritoneal dialysis on the
glucose intolerance of uremia. Abstr Int Soc Nephrol, 1978.
107. Cangiano JL, Ramirez-Gonzalez RE, Ramirez-Muxo 0: Bone disease and soft tissue calcifica-
tion in chronic peritoneal dialysis. Am J Med Sci 264: 301-306, 1972.
108. Oreopoulos DG, Rabinovich S, Memma HE, Lloyd GJ, Rapoport A, deVeber GA: Contrasting
bone changes in patients on chronic hemodialysis and chronic peritoneal dialysis. Int Symp Clin
Aspects Met Bone Disease 270: 628-631, 1972.
109. Kjellstrand CM: Treatment strategies for diabetic patients with end-stage renal failure. Mayo
Clin Proc 53: 819-820, 1978.
110. Oreopoulos DJ: Maintenance peritoneal dialysis. In: EA Friedman (ed), Strategy in Renal
Failure. John Wiley and Sons, New York, 1978, pp 393-414.
111. Blumenkrantz MJ, Kamdar A, Coburn JW: Peritoneal dialysis for diabetic patients with end-
stage nephropathy. Dial Transpl6: 47-52, 1977.
112. Rubin JE, Friedman EA: Dialysis and transplantation of diabetics in the United States. Nephron
18: 309-315, 1977.
113. Mitchell JC, Frohnert PP, Kurtz SB, Anderson CF: Chronic peritoneal dialysis in juvenile-onset
diabetes mellitus: a comparison with hemodialysis. Mayo Clin Proc 53: 775-781, 1978.
114. Blumenkrantz MJ, Shapiro DJ, Mimura N, Oreopoulos DG, Griedler RM, Levin S, Tenckhoff
208

H, Coburn JW: Maintenance peritoneal dialysis as an alternative in the patient with diabetes
mellitus and end-stage uremia. Kidney Int (Suppl) 1: 108-114, 1974.
115. Jacobs C, Rottembourg A, Frantz P, Slama G, Legrain M: Treatment of end-stage renal failure
in the insulin-dependent diabetic patient. Adv Nephrol 8: 101-126, 1979.
116. Hood SA, Frohnert PP, Mitchell JC, Kurtz SB: Home peritoneal dialysis: dialysis therapy of
choice in chronic renal failure of juvenile-onset diabetes mellitus. Dial Transpl9: 843-894,1980.
117. Mion C, Slingeneyer A, Oules R, Selam J-L, Delors J, Mirouze J: Home peritoneal dialysis in
diabetics with end-stage renal failure. Contrib Nephrol17: 120-130,1979.
118. Comty CM, Kjellsen D, Shapiro FL: A reassessment ofthe prognosis of diabetic patients treated
by chronic hemodialysis. Trans Am Soc Artif Intern Organs 22: 404-411, 1976.
119. Blagg CR: Visual and vascular problems in dialyzed diabetic patients. Kidney Int (Suppl) 1:
27-31, 1974.
120. Kaississieh SD, Yen MC, Lazarus JM, Lowrie EG, Goldstein HH, Takacs FJ, Hampers CL,
Merrill JP: Hemodialysis-related problems in patients with diabetes mellitus. Kidney Int (Suppl)
1: 100-107, 1974.
121. Quelhorst E, Schuenemann B, Mietzsch G, Jacob I: Haemo-and peritoneal dialysis treatment of
patients with diabetic nephropathy - a comparative study. Proc Eur Dial Transplant Assoc 15:
205-212, 1978.
122. Crossley K, Kjellstrand CM: Intraperitoneal insulin for control of blood sugar in diabetic
patients during peritoneal dialysis. Br Med J 1: 269-270.
123. Stoddard PD, Gallagher NM: Insulin regulation for the diabetic patient on peritoneal dialysis.
Nephrol Nurse 1: 16--19, 1979.
124. Ahmad S, Babb AL, Milutinovic J, Scribner BH: Effect of residual renal function on minimum
dialysis requirements. Proc Eur Dial Transpl Assoc 16: 107-114,1979.
125. Nolph KD, Stoltz ML, Maher JF: Altered peritoneal permeability in patients with systemic
vasculitis. Ann Intern Med 75: 753-755, 1971.
8. Continuous ambulatory peritoneal dialysis

JACK W. MONCRIEF and ROBERT P. POPOVICH

1. Introduction

There has been a rapid expansion in the number of dialysis units experienced with
continuous ambulatory peritoneal dialysis (CAPD). As of fall 1983, 561 dialysis
facilities in the United States and an approximate equal number throughout the
world have made CAPD available to most patients (Fig. 1). An increase in this
patient population is continuing and more than 9000 patients, nearly 13% of the
dialysis patients, in the USA are presently undergoing CAPD (Fig. 2). As of
November 1983, there are approximately 15000 CAPD patients worldwide. This
chapter will review the worldwide experience with this technique and attempt an
in-depth discussion of the present status of continuous ambulatory peritoneal
dialysis.
CAPD has become an increasingly utilized alternative method of treatment for
patients with end-stage renal disease (ESRD). The increased patient population
and renewed scientific interest in peritoneal dialysis is reflected in a tremendous
upsurge in scientific publications relating to peritoneal dialysis.

2. History

2.1. Theoretical and clinical

The mathematical and clinical parameters upon which CAPD are based were
developed at the Austin Diagnostic Clinic in 1975. The original concepts were
applied and materials, protocols, methods and results evaluated. Success was
achieved and the first published report was in the Abstracts of the American
Society of Artificial Internal Organs [1]. Subsequently, the National Institute of
Arthritis, Metabolism, and Digestive Diseases awarded a contract to evaluate
four more patients at the Austin Diagnostic Clinic. In 1977, the evaluation was
expanded to include patients under the direction of Dr Karl D. Nolph at the
210

Figure 1. Percent of patients by country presently on CAPD.

Figure 2. Number of CAPD patients in USA - rate to growth.

University of Missouri Medical Center. Successful control of uremia and an

improved patient sense of well-being was demonstrated [2, 3, 4]. The materials
available in the United States at that time required delivery ofthe dialysis solution
to the patient from glass bottles [3]. It soon became obvious that recurrent
infection made this technique impractical. Dr Dimitrios Oreopoulos, at the
Toronto Western Hospital, adapted the concepts of CAPD to improve the
delivery system by using collapsible plastic containers supplied as Dianeal (TM)
 211

by Baxter Travenol Laboratories in Canada [5]. This technique reduced the

number of times a patient was exposed to potential contamination and decreased
the incidence of peritonitis from one episode in every 7 patients weeks [3] to one
each 37 patients weeks [6]. Materials designed specifically for CAPD, refined
techniques and training methods have further reduced the incidence of infection
[7-14]. Publication of results from these original efforts stimulated others to
initiate CAPD programs [15]. The First International meeting relating specifically
to continuous ambulatory peritoneal dialysis was held in Paris, France, in No-
vember, 1979, and was chaired by Dr Marcel Legrain, then president of the
European Dialysis and Transplant Association. Scientific data was encouraging
and a monograph of the meeting was published in May, 1980 [16]. In that same
month, the Second International CAPD Symposium was held in Austin, Texas,
chaired by Dr Jack Moncrief. A seven-fold increase in attendance reflected the
increased interest in CAPD by the dialysis community.

2.2. Materials

The technique of CAPD has undergone modification with the development of the
specific materials designed for the procedure [7]. These include a catheter
adapter made of titanium with a luer-Iock configuration between the adapter and
the plastic connector of the extension tubing and an improved design of the spike
connection of the tubing to the bag.

3. Concepts of CAPO

3.1. Dialysate flow rate - urea clearance

Patients treated with intermittent dialytic therapies have; in the past, been
required to go to a specific place to receive treatment. The thrust of research in
the intermittent procedures has been to improve efficiency. The increase in
dialysate flow rate to 300-500 mllmin in hemodialysis has improved the small
molecule (urea) clearance to the range of 150 ml/min with a one-square meter
dialyzer [17]. This increased efficiency allows the patient to undergo adequate
dialysis (prevents uremia) if applied five hours, three times per week. To main-
tain this efficiency, however, a high blood flow is required, and some clinical signs
and symptoms are created with application of the procedure. These include
headache, muscle cramps, anorexia, nausea, vomiting, somnolence, and on
occasion shock and seizures [18]. As long as 24 hours may be required for the
patient to recover from the process itself.
Efforts have been made to increase the efficiency of intermittent peritoneal
dialysis (IPD) by increasing dialysate flow rate. Much ofthis effort, however, has
212

been unsuccessful [19-22]. In spite of techniques to improve dialysate flow rate,

maximum urea clearance achieved with IPD approximates 37 to 40 ml/min [23] at
a dialysate flow rate of70 to 80mllmin. Though the volume (flow rate) in IPD is
smaller, it is large enough to prevent portability.

3.2. Total drain volume and molecular size

3.2.1. Urea removal

The concept of continuous ambulatory peritoneal dialysis modeled by Popovich

and Moncrief [23-26] utilizes the smallest volume of dialysate, i.e. lowest dialy-
sate flow rate, to achieve the goal of preventing uremia. The remarkable decrease
in dialysate volume allows portability during the procedure (Table 1) [27-28].
If two liters of dialysis solution are placed in the peritoneal cavity and the dwell
times extended to allow equilibration (Fig. 3), the drained volume of dialysate
will then equal the urea clearance. If one assumes that adequate urea dialysis is
achieved when the blood urea nitrogen is maintained at 70 mg%, the dialysate
drained volumes which have equilibrated with the blood urea nitrogen will have
700 mg urea nitrogen/L. Urea nitrogen generation in the average 70 kg/man, on a
one gram of protein per kilogram body weight diet, will be 7000 mg/day. It is
possible, therefore, to remove this quantity of urea nitrogen with a total of ten
liters of drained volume if each liter is equilibrated with the blood urea nitrogen of
70 mg%. This can technically be accomplished by the following steps:
1. Infuse two liters of dialysis solution into the peritoneal cavity
2. Extend the dwell time to allow equilibration of urea nitrogen
3. Drain this volume and infuse two liters of fresh solution
4. Exchange the volume four times a day at convenient intervals
5. Adjust the solution tonicity to produce two liters of ultrafiltrate (two 1.5%
glucose, two 4.25% glucose), to achieve a total drained volume of ten liters/
day.
The procedure which will adequately remove small molecules (i.e. 60 MW urea)
is, therefore, defined.
Larger infusion volumes (2.5-3.0 liters) may allow less frequent exchanges and

Table 1. Calculation of minimal dialysate volume for adequate urea dialysis

Minimal dialysate volume = Dialysate volume equilibrated

with blood urea nitrogen
(Assume - acceptable blood urea nitrogen = 70mg%)
Dialysate urea nitrogen concentration = 700mg/1
(Measured - urea nitrogen generation rate = 7.0g/day)
Minimal dialysate volume = IOI/day
 213

Dialysate concentration
Plasma concentration

I0 - - - ---- -- - -- -- - ---- - --- -- ---1::--~-~"'..,.I.I",",",--~'O'"

UREA
09

08

07

06

05

04
INULIN
03

02

01
PROTEIN

00
20 60 90 120 160 200 240 300 360 420 480

Figure 3. Solute concentration-time profiles for a typical long dwell time exchange. (Published with
permission of Annals of Internal Medicine (Ref. 3».

achieve similar results [29]. Increased intra-abdominal pressure may produce

higher incidences of hernias, respiratory embarrassment and pain on infusion,
but many patients tolerate these volumes without complications [29].

3.2.2. Creatinine removal

When the procedure is evaluated for its efficacy in removing creatinine, it is seen
that the larger molecular size (113 daltons) increases resistance of creatinine
transfer from the blood to the dialysate compartment (Fig. 3). The prolonged
dwell time does not reduce the efficiency of creatinine removal as greatly as it
does that of urea. Therefore, the procedure is more efficient for the removal of
creatinine when compared to urea. Dialysis of creatinine will accrue throughout
the prolonged dwell time, and though the rate of creatinine movement from the
blood into the dialysate decreases progressively as the concentration gradient
between blood and dialysate is lost, dialysis of creatinine continues for seven to
eight hours as opposed to urea. In the latter case, dialysis virtually stops at the end
of two to three hours. CAPO is, therefore, not continuous ambulatory urea
dialysis but is continuous ambulatory creatinine dialysis.
214

3.2.3. 'Middle molecule' removal

As the molecules of interest increase in weight, higher efficiency is seen with the
prolonged dwell time. A molecule of 5200 daltons (inulin) will diffuse from the
blood compartment into the dialysate very slowly, and the concentration gradient
between the blood and the dialysate will be maintained for an extended period
(Fig. 3). A single exchange made every twelve hours will produce excellent
removal of this size molecule. CAPD is approximately six times better at remov-
ing inulin size molecules than is intermittent hemodialysis for 15 hours a week [2,
30].

3.3. Ultrafiltration rate

Although many factors effect ultrafiltration rates during peritoneal dialysis [31-
34] - osmotic gradient between blood and dialysate; peritoneal membrane thick-
ness; status of vasoconstriction; hydration of the peritoneal membrane; and total
infused volume [35, 36] - the present clinically controllable factor is the dialysis
solution tonicity [37]. As tonicity increases, ultrafiltration rate proportionately
increases. Glucose is the osmotic agent in commercially available solutions, and
upon instillation of a hypertonic solution of 4.25% glucose, an immediate ultra-
filtration rate of approximately 15 to 25 ml/min will occur. As glucose is simul-
taneously transferred from the dialysate to the blood side, tonicity will diminish
and the ultrafiltration rate will fall proportionately. This transfer will occur at a
gradually decreasing rate (related to the concentration gradient between blood
and dialysate) and total ultrafiltration will occur over approximately the first two
to four hours. A similar but less dramatic ultrafiltration will occur with instillation
of a 2.5% or 1.5% glucose solution, and isotonicity will occur at approximately
two to four hours. The total ultrafiltered volume with instillation of 2000 ml of
1.5% glucose solution will be approximately 300 ml. Five to seven hundred
milliliters of ultrafiltration will occur with use of 2000 ml of 2.5% glucose solu-
tion. As isotonicity between blood and dialysate is achieved, the reverse flow of
isotonic water from the peritoneal cavity into the blood vascular space will occur
at a rate of 1 +/- 0.5 mllmin [37]. This reverse flow is important in evaluating
total volume available for drainage and the effect of the prolonged dwell time. As
dwell time increases, drained volume will decrease by approximately 60 ml/h
after the first two to four hours of osmotically controlled ultrafiltration.
There are reports of progressive reversible and irreversible decline in ultra-
filtration rate and volume. Some of these appear to be related to peritonitis [38,
39]. A series of reports from Europe indicate that this problem is unrelated to
peritonitis [40-43]. There is survey evidence that this problem is limited to
patients using fluid manufactured in Europe [44]. Non-glucose osmotic agents are
being evaluated but none are clinically available in the United States [45]. Maher
has reported a selective increase in peritoneal ultrafiltration with the addition of
 215

Amphotericin B [46]. Diaz-Buxo has reported a decrease in peritoneal clearance

in patients with hyperparathyroidism [41]. Increase in solute transport in labora-
tory animals has been described with the addition of protamine to dialysis
solution [47-49] and with dipyridamole by mouth in humans [50].
Dialysis solution with a glucose concentration of 2.5% is now available and has
become the standard concentration for most exchanges. The advantage of this
solution concentration is that it avoids the intermittent symptoms of distention
(due to large ultrafiltration volume) frequently created by the 4.25% glucose
solution.

3.4. Clinical application of water flux data

3.4.1. Number of exchanges

How many exchanges are required to adequately perforO} CAPD [27, 28]? The
primary factor which controls the efficiency of the procedure relative to small
molecule removal is not the number of exchanges but the drained volume. The
number of exchanges per day is only tangentially related to the total drained
volume. If three exchanges of 4.25% glucose solution are performed each day,
the total drained volume will be approximately nine liters. If four exchanges of
1.5% glucose solution are performed per 24 hours, approximately eight liters of
drained volume will be achieved. As the drained volume is decreased from ten
liters per day to eight liters per day, there is some temporary sacrifice in total urea
removal (approximately 1.4 g the first day) and urea nitrogen will stabilize at a
slightly higher level before removal will again equal production. There is less
sacrifice in the removal of creatinine with a decrease in drained volume from ten
to eight liters per day. The serum creatinine will stabilize approximately 1.5 mg%
higher (12.5 +/-1.5mg% versus 14 +/-1.5mg%). A 'middle molecule' of 5500
daltons will be removed at approximately the same rate with a drained volume of
ten liters per day as compared to eight liters per day. As drained volume (flow
rate) requirements are clinically defined, the 'middle molecule' hypothesis should
be confirmed or disproved [30].

3.4.2. Dialysis solution tonicity

Vigorous removal of excess salt and water may be used to control edema and
hypertension. This can be accomplished rapidly by increasing the frequency of
exchanges or through the use of hypertonic solutions.
If intravascular volume depletion or orthostatic hypotension occur, the patient
may be instructed to skip a single exchange. This will prolong the dwell time to
approximately ten to twelve hours and produce an infusion of 60 ml/h from the
abdominal reservoir into the blood vascular space and re-expand the blood
volume. This can prevent the need for intravenous fluids. Only a minimal and
temporary sacrifice in the dialysis of small molecules will occur.
216

3.4.3. Glucose absorption

The total glucose absorbed during CAPD depends on the number, and glucose
concentration, of the exchanges [51, 52]. If two liters of 4.25% glucose solution
are used twice per day and two liters of 1.5% glucose solution used twice per day,
approximately 900 calories of glucose will be absorbed per 24 hours [51, 53].
Other types of osmotically active agents are being evaluated [54-56]. There is
suspicion that this large glucose load is etiologically related to the frequency of
obesity and hypertriglyceridemia in patients undergoing CAPD [57]. Reports of
increased ultrafiltration associated with the presence of insulin in the infused
solution have been unconfirmed [58].

3.5. Sodium and water balance

Rapid ultrafiltration associated with hypertonic glucose solutions will cause

movement of sodium-free water from the blood vascular space into the peritoneal
cavity [59]. Hypernatremia will occur if this ultrafiltration continues at a rapid
rate. The prolonged dwell time of continuous ambulatory peritoneal dialysis
allows near total equilibration between the concentration of sodium in the blood
and the dialysate. Thus, two liters of ultrafiltrate per day may produce a negative
sodium balance of 240 or more mEq. This may allow the patient an unrestricted
sodium chloride diet. Since this ultrafiltration volume is a function of the tonicity
of the solution used, and the procedure itself is not 'smart', the prescription for
the tonicity of the solution must be adjusted to accommodate for the salt and
water ingested by the patient. If rapid ultrafiltration with hypertonic glucose
solutions is carried out continuously (each 30 to 60 min), the increased serum
sodium will stimulate thirst. A water drinking - ultrafiltration cycle may occur.
This may be seen in the early training days during which rapid removal of salt and
water is initiated to control blood pressure. As gradual ultrafiltration produces
salt and water depletion, the intravascular volume will contract and blood pres-
sure will be controlled [60, 61]. The need for antihypertensive medication can
often be eliminated simply by continued removal of more salt and water than is
ingested until the blood pressure is normal. CAPD allows the body weight to be
maintained at a specific level throughout the day. The effects of angiotensin,
norepinephrine, and renin do not prevent blood pressure decline produced by
intravascular volume depletion, but only narrow the range of weight fluctuation
which may occur before hypertension or orthostatic hypotension will occur.
There is in every hypertensive patient a blood volume at which pressure is
normal.
 217

3.6. Potassium balance

CAPD will prevent most of the signs and symptoms of uremia and move a patient
from a state of ESRD to approximately eight to ten ml/min of urea clearance. It is
obvious, however, that the limits of the procedure will pose some constraints on
the patient's dietary discretion. These limitations are most significant for sodium,
water, potassium, and phosphate. Though it is unusual for any potassium intake
to exceed the ability of the procedure to remove potassium, hyperkalemia may
occur. On a clinical basis, however, hyperkalemia is uncommon, and a dietary
survey will demonstrate the etiology of the potassium load (i.e. two quarts of
orange juice per day). Access to potassium within certain limits is prescribed for
the patient undergoing the procedure, but frequent serum potassium evaluation
during the early training phase should be performed to detect indiscretion. The
amount of potassium removed by the dialysis procedure itself is not adequate to
explain the absence of hyperkalemia. In fact, the concentration of potassium in
the dialysate is almost invariably lower than the serum potassium [62]. There is,
however, an increased quantity of potassium found in the stool [62, 63], an
increased total body potassium, and an increased muscle mass associated with
anabolism [64, 65]. This combination may serve to explain the absence of the
hyerkalemia. A rapid increase in serum potassium does occur in some patients
who are temporarily taken off CAPD, and close attention to dietary restriction
and potassium evaluation should be paid if CAPD is even temporarily discon-
tinued (12 to 24 hours).

3.7. Phosphate balance

The concentration of phosphate in the drained volume following an eight hour,

overnight dwell time is near the serum level phosphate and can be used as a
measure of serum phosphate without resorting to blood sampling [66]. Phosphate
dialysis is better with CAPD than with intermittent peritoneal dialysis and many
patients require only modest doses (relative to hemodialysis patients) of alumi-
num hydroxide to maintain acceptable serum phosphate levels [66-68]. Since
calcium carbonate is itself a phosphate binding agent and some patients require
calcium supplement to maintain a calcium balance, calcium carbonate may be
considered for phosphate control when necessary [67, 69].

3.B. Protein loss

High protein diets (1.2 to 1.5 g protein/kg body weight/day) have been recom-
mended because of the protein losses in the dialysate [51, 70, 71]. These range
between three and twenty gram of protein loss/24 hours. Increased loss occurs in
patients with recurrent peritonitis [3]. In patients without peritonitis, three to six
218

gram protein 10ss/24 hours has been documented [72]. Positive nitrogen balance
has been demonstrated in patients on 1gm/kg body weight protein diets [73-75].
As little as 700 mg/kg body weight will allow protein anabolism in most patients
[73, 74, 76]. Therefore, protein intakes in excess of 1.0 to 1.5 g/kg do not appear
necessary and will increase the requirement for aluminum hydroxide or other
phosphate binding agents. Dietary education encouraging the ingestion of high
quality protein is essential but supplementation of the protein diet with milk, and
milk products will tend to produce hyperphosphatemia which may aggravate
itching, suppress serum calcium levels, and stimulate parathyroid hormone secre-
tion.

3.9. Calcium balance

Transport of calcium between blood and dialysate is a function of the concentra-

tion of diffusible calcium in the plasma and the calcium concentration in the
dialysis solution [77]. With 3.5 milliequivalents of calcium per liter, a diffusion of
calcium down the gradient from the solution into the plasma occurs, and a slightly
positive calcium balance will accrue [2]. However, ultrafiltration will cause a
transport of calcium from the patient into the dialysate and produce a negative
calcium balance [78]. The total balance between calcium diffusion and the
ultrafiltered calcium will determine the total net calcium flux. Some reports have
suggested a positive calcium balance [2]. Others report a negative calcium bal-
ance [78]. These differences may reflect a difference in the diffusible calcium in
the individual patient population and a variation in the dialysate calcium concen-
tration. (Canada, 2.5 mEq/L vs United States, 3.5 mEq/L). With five exchanges
per day, an ultrafiltration volume of two liters per day and a dialysate calcium
concentration of 3.5 mEq/L, a decrease in the serum parathyroid hormone levels
has been reported [2, 79]. This trend was reversed when the same patients were
placed on four exchanges a day using solution of the same calcium concentration.
Some authors have recommended supplementation of calcium with calcium
carbonate [67, 68]. Dialysate loss of 1.25 dehydroxy vitamin D3 has been reported
and small supplemental doses of this vitamin have been suggested [66, 68, 80].
Some patients tend to develop hypercalcemia and may have poor osteoblastic
activity when exposed to oral calcium and vitamin D supplement [79].

4. Dietary management

As previously stated, major dietary indiscretion can exceed the removal rate of
metabolic toxins, salt and water. However, because CAPD can ultrafilter large
volumes of water and remove large quantities of sodium, no major restriction is
required. Potassium restriction has been made unnecessary by the presence of
 219

large amounts of potassium in the stool [75]. However, adequate dietary instruc-
tion is essential to avoid dietary indiscretion. In patients who are malnourished or
protein depleted, a high protein intake of 1.5 g/kg body weight is recommended
[64,78]. When protein repletion has been achieved, free access to protein without
supplementation or restriction can be prescribed. The ingestion of very large
quantities of salt and water may require the utilization of increasing amounts of
hypertonic glucose solutions for removal. This cycle contributes to progressive
obesity and hypertriglyceridemia [59, 2]. Salt restriction followed by decreased
thirst and, thereby, less water intake may decrease the requirement for hyper-
tonic glucose solutions and alleviate this problem. It must be remembered,
however, that decreasing the ultrafiltration will decrease the total drained vol-
ume and, thereby, reduce small molecule dialysis [2]. The prescription for the
dialysis solution, therefore, must reach a balance between the need for maximum
total drained volumes for adequate dialysis and excessive hypertonic glucose
which may produce hypertriglyceridemia and obesity. The practice of using four
exchanges per day with one or two consisting of 4.25% glucose solution has been
demonstrated to produce adequate dialysis as evaluated by most parameters for
up to five years [81]. Dietary calcium and phosphate have been discussed
previously.

5. Clinical aspects of CAPD

5.1. Establishing a program

5.1.1. Personnel
Commitment of the nephrologist and nursing personnel to sterile technique,
chronic catheter management, and patient training methods is necessary for a
successful CAPD program [82, 83]. A well-trained dedicated staff thoroughly
familiar with the concepts of peritoneal dialysis and available for emergency and
routine outpatient follow-up should facilitate successful delivery of this system.
The technique is extremely simple. Problems may arise quickly, however, and be
demanding on the medical and nursing staff. Thus, adequate personnel, propor-
tionate to the patient population is essential. One nurse to each eight to ten out-
patients and a schedule for 24 hour per day coverage is also required. [84].
The first week of training should be carried out on a one-to-one basis, with a
nurse dedicated to education of each patient. By the end of the first week, the
patient is usually performing the technique independently. At that point, trou-
bleshooting and observation can be done in small groups, but no more than one
nurse to three patients is practical.

5.1.2. Site of training

In hospital CAPD training is practiced in some institutions. Out-patient training
220

is preferable, however, because it decreases the cost and prevents the 'ill patient'
syndrome that may occur with in hospital procedures [85, 86]. Little equipment is
required, and 100-120 square feet of training space per patient is adequate.

5.1.3. Hospital back-up

Although hospitalization has become less frequent [87], good hospital back-up
and expert management of the patient is required. Most centers have found that
specialized nursing personnel should manage both the out-patient training and
follow-up, as well as in hospital back-up and management [81, 84, 85].

5.1.4. Team work

Since CAPD replaces only a small amount of the function of the kidneys, a
physician who is trained and confident in the care of patients with renal failure
must necessarily participate in training and management of the patients. Retrain-
ing and support services must be easily accessible to every patient undergoing
continuous ambulatory peritoneal dialysis.

5.2. Patient selection

5.2.1. High-risk patient population

CAPD is a simple, easily taught technique which lends itself to management of
patients willing and able to undergo self-dialysis. the simplicity, however, tends
to make it an attractive technique to medical staff with patients who have major
difficulties undergoing other types of dialysis. The 'last ditch' effort, 'maybe you
ought to try CAPD' approach may produce statistical evidence of a high mortality
and an excessive drop out rate of patients who initiate training. Patients in whom
there is little left to offer certainly may be 'given a try' on CAPD, but a high
failure rate can be expected as well as a high mortality rate [88-90].

5.2.2. Selection
In those patients in whom there is a choice concerning which form of long-term
ESRD care is best, selection criteria will be required. Development of these
criteria will aid in finding the place that CAPD will best serve the end-stage renal
disease population.
CAPD is self-dialysis. It is best offered, therefore, to those patients who are
motivated to care for themselves and who can be expected to accept this responsi-
bility on a long-term basis because of the benefits. A good concept of the
relationship between end-stage renal disease and symptoms, a scientific approach
to the requirement of dialysis and a belief in the need for this therapy by the
patient is essential or failure is inevitable. Internal, rather than external, motiva-
tion factors are most beneficial, and the physical and mental capabilities of
performing the procedure are required [91]. Personality traits which lead to a
 221

desire for personal independence and freedom in a reasonably intelligent individ-

ual would make the very best CAPD patient.

5.2.3. Infants and children

Reports of success with continuous ambulatory peritoneal dialysis in the pediatric
population suggest that this modality may become the treatment of choice in the
dialysis of infants and small children [4, 34, 92, 93, 9S-98]. Adequate clearance to
prevent uremia has been reported [99, 100], and preliminary information suggests
accelerated linear growth [97, 101, 102]. The best results have been reported with
a high protein diet, and a forced feeding of 2.2-3.S gram protein/kg body weight
has produced 'catch up' growth in infants and small children [103]. Kohaut [104]
and Alexander [lOS, 106] have reported accelerated growth in children who have
normal parathyroid hormone levels. They recommend supplement with dihydro-
tachesterol or dihydroxycholecalciferol and close attention to calcium and phos-
phate control. The fact that ninety-four percent of patients in network six
(Southern California) below the age of 10 are reported to be on CAPD illustrates
the increased interest by the pediatric nephrology community [4].

5.3. Contra-indication

5.3.1. Blindness
Well-organized and successful programs designed to train blind patients have
allowed those patients to perform self-dialysis [107, 10]. Tools and devices are
now available to aid the blind patient in maintaining the sterility of the system [9,
10, 14].

5.3.2. Transfer surface area

There must be an adequate transfer surface area available for peritoneal dialysis,
and clinical evaluation of the adequacy of the surface and blood flow to the
peritoneum can be achieved, though it is rarely needed, prior to implantation of a
chronic peritoneal catheter.
This information can be obtained by passing a spinal needle through the
abdominal wall and infusing two liters of 1.S% glucose solution into the per-
itoneum. The needle is withdrawn and after a dwell time of 4 hours, a small
sample of the dialysate is removed and compared with the serum creatinine. Sixty
to 70% equilibration of creatinine in the dialysate as compared to plasma demon-
strates normal transfer. Less than 40% may cause concern and less then 30%
makes failure of CAPD likely. This technique may be used in those individuals
with multiple abdominal surgical procedures including aortic aneurysm repair or
previous serious abdominal inflammation, such as pancreatitis or peritonitis
secondary to ruptured viscus.
222

5.3.3. Triglycerides
Severe hypertriglyceridemia not associated with diabetes mellitus is also con-
sidered a relative contraindication to CAPD.

5.3.4. Immunosuppression
The requirement of chronic immunosuppressive drugs in patients with conditions
such as active systemic lupus erythematosis, have caused concern about the risk of
catastrophic results from an episode of peritonitis. Many patients undergoing
steroid therapy have now been freated with CAPD, and no increased mortality
has been reported. An improvement in both cellular and hemoral immunity has
recently been reported in patients transferred from hemodialysis to CAPD [88,
108, 109, 110].

5.3.5. Ostomies
Chronic abdominal wall infections including open colostomies, ileostomies, and
nephrostomies may increase the risk of recurrent peritonitis.

5.3.6. Hernia
Umbilical, inguinal and diaphragmatic hernias as well as hemorrhoids, may be
aggravated by increasing intra-abdominal pressure with CAPD and complicate
the long-term performance results [78]. The repair of 'large' hernias prior to
CAPD appears prudent. The complication appears more commonly in patients
using larger volumes of dialysis as suggested by Twardowski and Rubin [29, 111].

5.3.7. Neurological defects

Physical incoordination resulting from progressive neurological disease, cere-
brovascular accidents, movement disorders and severe arthritis, may make
CAPD impossible to perform.

5.3.B. Back pain

Low back pain may be aggravated by the presence of the fluid in the abdomen
with a shift in the center of gravity (78, 112].

5.3.9. Psychological and social problems

A cooperative and compliant patient is essential if CAPD is to be successful.
Patients who are psychotic, belligerant or uncooperative cannot be expected to
succeed with this form of self-dialysis.

5.4. Peritoneal catheters

5.4.1. Types of catheters

The chronic, indwelling catheter is the lifeline and access for the CAPD patient.
 223

Permanent access to the peritoneal cavity was made possible by the development
of the dacron felt cuff on the silastic catheter [113]. Various types of catheters,~
including the single- and double-cuffed Tenckhoff type, the Toronto Western
[114], the Ash (Life) catheter and the Valli catheter [115], lend evidence to the fact
that no perfect catheter is available. Past experience with the Tenckhoff catheter
has made it the most widely used and successfully employed catheter. The double
cuffed catheter may have the advantage of a better seal and make dislodging
unlikely and a tunnel infection less likely because the distal cuff is very close to the
skin (1 to 2cm). This, however, increases the risk of distal cuff erosion through
the skin which will then produce a chronic tunnel infection. Many investigators
feel that the frequency of this complication has made the double-cuffed catheter
less useful than one with a single cuff [116].
A new cuff material, PTFE, introduced by William Gore and Associates, Inc.,
may produce better skin healing. Such innovations are evidence of the great
interest in new access devices which should produce major advances in the near
future.

5.4.2. Methods of implantation

The Tenckhoff catheter can be inserted by a surgical technique or at the patient's
bedside. Many nephrologists have practiced the bedside technique in both hospi-
tal and office [117]. The majority of physicians prefer these catheters to be
implanted in the operating room, however, with the surgical approach allowing
direct vision. General anesthesia is preferred by many. Numerous variations have
been reported in the position of the catheter cuffs [114, 117, 118]. Kirksey has
developed and reported [119] a modification of the technique of cuff placement, in
which a single cuffed catheter is used. The cuff is placed superficial to the facia
and a non-absorbable suture sewn around the cuff and in the facia. This decreases
the tendency toward leakage or displacement in the early phase, prior to tissue
ingrowth. It allows early utilization and decreases early failure rate. All catheters
except the Tenckhoff require surgical implantation but the latter is much easier to
remove.
An evaluation of the integrity of the catheter seal may be performed by the
infusion of a large volume of dialysis solution (three liters in the average size
adult) while the patient is still on the operating table. If leakage does occur, repair
is then easily accomplished while the peritoneum is still accessible [120].

5.4.3. Catheter break-in

Catheter break-in is another controversial issue. Dr Tenckhoff prefers rapid
exchange [117], prolonged dialysis following implantation of the catheter. This
technique requires continued hospitalization, immobilization, and increases cost.
Moncrief has developed an out-patient irrigation technique [62] in which a single
container of 500 ml dialysis solution is attached to the transfer tubing. Fifteen
hundred units of heparin are added to the solution, and the fluid is infused into the
224

patient by gravity. It is immediately drained, then reinfused for a total of three

in-and-out irrigations. The last irrigation ends with the solution back in the
container, and the patient carries this bag in a pocket for the next six to eight
hours. At that time, the same solution is used to irrigate the catheter, and again,
the solution remains out of the abdomen. Eight hours later the procedure is
repeated a third time. Each 24 hours, the 500 ml solution container with 1500 units
of heparin is replaced by nursing personnel using aseptic technique, and the
patient irrigates the catheter. With this technique, three irrigations per day for
four days are performed, then CAPD training is initiated. This modification
allows the patient to be discharged from the hospital following catheter insertion,
does not require any understanding of sterile technique by the patient, and uses a
small volume of dialysate to perform the irrigation. If dialysis is required, either
hemodialysis or short-term (eight hours), low volume (1000ml), short dwell time
(20 to 30 min) peritoneal dialysis is performed with the patient flat in bed, and the
patient is then discharged to continue the irrigations at home.

5.4.4. Catheter position

Many units have preferred the catheter to be implanted in the midline because of
the simplicity and lack of muscle tissue in this area [114, 117, 118]. This area of
placement, however, may produce a body image problem, especially in young
adults, as this is an important area during sexual activity. Movement of the
catheter laterally, low in the abdomen, medial to the anterior superior iliac spine,
tends to reduce the body image problem and eliminates the catheter from the area
of sexual activity. A decrease in catheter site hernia has also occurred with this
lateral position [111, 121]. The single-cuffed Tenckhoff catheter implanted in the
operating room has been recommended in infants and small children [106, 122,
123].

5.4.5. Catheter exit site care

A knowledge of catheter exit site care is critical for both the teaching staff and
patient, and major efforts must be made to educate the patient about personal
hygiene. Some training units recommend that the catheter exit site be covered
with an airtight dressing [114, 117], but at the Austin Diagnostic Clinic, improved
catheter life and a decrease in the incidence of infection have resulted by leaving
the catheter exit site undressed. The patient is instructed to perform daily
catheter site care (more often if an infection is present) in the shower, scrubbing
the catheter site gently with soapy water and rinsing the site with tap water from
the shower. The catheter site is then painted with a povodone-iodine solution and
allowed to air dry. Removal of any encrustations or scabs is done with hydrogen
peroxide. No bandage is applied to the catheter site after the original insertion
wound has healed and drainage has stopped. Any evidence of moisture, tender-
ness or drainage from the catheter site is reported immediately to medical staff.
Cultures are obtained and antibiotics initiated. Early exit site drainage or tender-
 225

ness is frequently cleared within 24 to 48 hours with a single dose of 1 mg/kg body
weight of an aminoglycoside intramuscularly. Chronic catheter site drainage
suggesting cuff infections may require removal of the catheter and replacement
on the opposite side of the abdomen. No difficulty with secondary healing of this
old catheter site is to be expected if conscientious cleansing and attention are
practiced. Failure to remove an infected cuff will invariably lead to erosion and
peritonitis. Subclinical tunnel infection may be suspected when recurrent per-
itonitis caused by the same organism occurs.
An experimental technique using indium labeled granulocytes has been re-
ported and may allow earlier diagnosis of this important problem [145].

5.4.6. Swimming and bathing

Patients are allowed to swim and take a sitz bath, keeping the spike-container
connection dry, if possible. If the spike-container connection becomes wet,
however, a new povidone-iodine dressing is applied and allowed to soak the area
for at least one hour before an exchange is made.

5.5. Catheter failure

5.5.1. Early catheter failure

Causes of catheter failure include: obstruction secondary to improper placement,
folding and bending, occlusion by wrapping ofthe omentum, and migration ofthe
catheter from the pelvis upward in the peritoneal cavity. Postoperative bleeding
into the peritoneal cavity, with fibrin and clot formation and early infection, may
also cause obstruction. Early and persistent catheter leakage may require re-
placement.

5.5.2. Late catheter failure

Late failures (after two weeks' use) are almost invariably associated with per-
itonitis or catheter displacement, but may be associated with intra-abdominal
catastrophes unrelated to peritoneal dialysis such as ruptured diverticula, pan-
creatitis, peptic ulcer, bowel perforation, etc. Fibrin formation can be prevented
by the addition of heparin to the solution when peritonitis occurs. Heparin may
also be required intermittently in some patients who developed fibrin material in
the dialysate unrelated to infection. This latter problem can usually be docu-
mented by a decreasing rate of outflow or partial obstruction. Partial outflow
obstruction may be reversed by the early addition of 10,000 units of heparin to the
infusion solution and then allowing a prolonged drain time (four to twelve hours)
during which an increasing rate of flow may occur and the catheter obstruction
reversed. Several days of heparin use (2000 units per exchange) may then return
the flow to normal. Manipulation of the catheter with Fogerty catheters and
trocars yields little more than the potential risk of contamination and subsequent
infection with little hope of opening an obstructed catheter.
226

5.6. Training and follow-up

5.6.1. Day 1 to 4
Out-patient and in-patient training has been discussed previously (5.1.2). Ten to
fifteen training sessions are usually required to prepare a patient to manage
CAPD [84]. The training sessions are carried out on a daily basis. The sessions are
begun after catheter break-in and each session consists of six to eight hours as the
patient's physical, emotional, and mental status allows. During the first few days,
exchanges are made approximately each 11/2 to 2 hours. These rapid exchanges
allow better urea and small molecule control and avoid the need for back-up
dialysis. It also increases patient learning by repetition and may be used to more
rapidly control the expanded intravascular volumes so commonly present. This
will bring blood pressure under control as antihypertension medications are
tapered. During these early training days, the patient is taught to make the
exchanges using simulation models while the nurse makes the actual exchange.
At the end of the training day (4:30-5:00 p.m.), two liters of solution are left in
the abdomen overnight. If discomfort occurs during the night, the patient is
instructed to drain solution until relief is achieved. One hundred to 300 ml
drainage is usually adequate, but some patients may drain as much as one liter.

5.6.2. Day 4 to 6
After the first few days as technical skills improve, the nurse allows the patient to
make the exchanges with supervision. As this practice improves technique, the
patient then makes exchanges in the unit without supervision and eventually
makes his own exchange at night. When this is accomplished, the patient begins
the CAPD regime at a rate of four exchanges per day. This degree of expertise
can usually be reached within the first week of training, and the patient is ready to
make all four exchanges by Sunday in the absence of the staff.

5.6.3. Day 8 to 12
During the following week, exchange and theory training continue. Repeated
documention of successful adherence to the rigid technique is recorded by the
staff. Question and answer sessions also continue with a 'troubleshooting' format
and near the end of the second week, the patient is given a battery of examin-
ations to evaluate both the theoretical and practical aspects of accumulated
knowledge. These successfully completed, the patient is graduated from the
CAPD training program and signs a release stating that CAPD training has
included theoretical and practical aspects of the necessary information. The
patient is then given supplies and sent home to perform the technique for two
weeks. The nurse accompanies the patient, when feasible, evaluates the home
environment, and observes the first exchange performed after graduation.
 227

5.6.4. Follow-up
Within 2 to 3 days, contact is made by telephone. At the two week visit,
re-evaluation of the technique occurs, and suggestions for change are made as
clinically indicated. The patient is seen each two weeks for the first three months.
With each visit, a complete exchange is observed, and each month a transfer set
change is performed by nursing personnel. After three months, the visits are
reduced to one per month. During these visits, clinical evaluation by the physician
is made and the catheter site is inspected. Compliance and potential theoretical
problems are discussed and documented by the staff.

5.6.5. Telephone contact

Close contact with and encouragement of the patient is accomplished with a
weekly telephone call which includes specific questions about blood pressure, dry
weight, last scale weight, appetite, sense of well-being, catheter site evaluation,
and dialysate flow characteristics. The calls serve to detect incipient and slowly
developing problems, keep the patient aware of the concern of the staff and
reduce a reluctance to call if problems arise. The educational process also
continues with these calls.

5.6.6. Laboratory tests

During the training phase, laboratory evaluation is obtained on the first day and
every other day unless unusual parameters are reported. Hyperkalemia or hypo-
kalemia may require daily testing for serum potassium. If the BUN does not begin
to fall or if the blood sugar starts to rise, more frequent evaluation may be
necessary. A list of the routine laboratory tests is in Table 2. If the BUN and

Table 2. Routine laboratory tests

Every month:
BUN Total protein
Creatinine Albumin
Sodium Alk. phosphatase
Potassium LDH
CO 2 SGOT
Calcium Hct
Magnesium Hgb

Phosphate Dialysate protein

Every 3 months
WBC RBC Platelet count

Every 6 months:
Residual renal function Motor nerve conduction velocity
24 hour urine volume EKG
Chest X-ray Bone mineral density
228

creatinine are not stable at acceptable levels or slowly declining during the first
week, a clinical transport evaluation using the dialysate and serum creatinine
comparison as described earlier (5.3.2), is performed.

5.6.7. Medications
Medications taken by the patient before beginning dialysis are adjusted with the
initiation of CAPD. These include diuretics, phosphate binding agents, and
occasionally digitalis preparations. Patients are maintained on water-soluble
vitamins. Antihypertensive medications are generally tapered during the early
phase of training. During this time, intravascular volume is contracted by ultra-
filtration. Insulin dosages are adjusted to accommodate the increased glucose
load. If the blood sugar is difficult to control or serum triglyceride levels climb
over 500, initiation of insulin in the dialysis solution may be required [124, 59].
This latter route of administration has been shown to improve glucose control but
a safe technique to add insulin to the dialysis solution is essential [125].
Rarely, a patient who is very compliant and lives a great distance from the
center, will be instructed in the technique of adding antibiotics to the dialysis
solution and the patient given a starter dose of cephalosporin [126]. The patient is
instructed to contact the unit by telephone before using this medication.

6. Complications

These are discussed in greater detail in Chapters 13 and 14.

6.1. Complications during training

6.1.1. Distension and abdominal pain

Most patients experience some sense of distension and fullness as the CAPD
technique is initiated. This is most problematic in young, muscular individuals
and less in older multiparous females. These symptoms may be painful and
require smaller infusion volumes several days after initiation of training. This may
also occur during the overnight dwell time and occasionally cause the patient to
question the choice of CAPD as a dialysis technique. These symptoms invariably
disappear. The distension symptoms may be most severe in areas where previous
surgery has been performed and related to traction on adhesions.
Some patients complain of pain on infusion which may be aggravated by either
excessively cold or warm temperatures of solution. This pain is, however, most
frequently bicarbonate responsive. The pH of the dialysis solution has been found
to be 4.9 to 5.2. The addition of a small amount of 50% bicarbonate solution (5 to
10 ml) will raise the pH to 6.0 to 6.02 and relieve most of the infusion pain. Some
patients continue to have bicarbonate responsive pain for several weeks to
 229

months, and some require the instillation of small doses of bicarbonate for
prevention. With time, however, most patients have discontinued this procedure
as the pain is extremely shortlived and usually not worth the trouble required to
add the bicarbonate in a sterile manner. Decreasing the rate of infusion by
lowering the height of the bag during infusion may also be helpful. This latter
maneuver is most helpful in those patients who complain of rectal or peritoneal
pain.

6.1.2. Catheter leaks - early

Catheter site leaks are also common during the training period and increase in
frequency with early use. These may also develop in patients who experience
increased intra-abdominal pressure following catheter implantation. Examples
include straining due to constipation, respiratory distress secondary to asthma or
laryngeal edema post anesthesia, or a barium enema performed without draining
the abdomen. Most leakage will spontaneously disappear if the abdomen is
drained and left empty one to two days. Back-up dialysis may be required. Some
early leaks may last as long as three weeks before spontaneous closure. Evalua-
tion of continued leakage may be made each three to four days with a small
volume of dialysate. When the leakage has ceased, 48 hours should be allowed to
pass before CAPD is initiated, with only 1.5% glucose solutions utilized for
several days. The incidence of early leakage can be decreased by the three-liter
test as described earlier (5.4.2).
During this delay, training may continue. Exchange technique may be simul-
ated and the theory reinforced in order that the catheter site leakage need not
inordinately delay discharge to home. Most early catheter site leakage is not
associated with infection but early and empiric treatment of suspected infection
may decrease catheter loss and prevent peritonitis. The potential is present for
the transmission of organisms from the skin through the exit site leakage into the
peritoneal cavity.

6.1.3. Catheter leaks -late

Late catheter leakage is usually secondary to a tunnel infection and subsequent to
a cuff infection. Diagnosis is established by purulent drainage from the catheter
site which is unresponsive to treatment. When cuff infection is present, early
catheter removal before dialysate leakage occurs will avoid the peritonitis which
invariably follows communication from the peritoneum through an infected
catheter cuff.
Rarely, late catheter leakage will occur spontaneously or follow abdominal
trauma. Drainage of the dialysate and a dry abdomen for several days will usually
allow spontaneous closure and prevent catheter loss.
230

6.2. Blood pressure control

6.2.1. Hypertension - hypotension

Aggressive attempts to control the blood pressure through blood volume reduc-
tion by ultrafiltration will usually produce excellent results [127]. Rapid tapering
and discontinuation of antihypertensive medications may allow recrudescence of
severe hypertension, and reinstitution of medications for temporary control may
be required. Intravascular volume contraction increases the efficacy of any blood
pressure medication and even small doses may produce severe orthostatic hypo-
tension. As volume reduction is pursued, spontaneous orthostatic hypotension
unrelated to antihypertensive medications almost invariably occurs in those
patients where aggressive ultrafiltration is undertaken for blood pressure control
[127, 128]. Once this level has been reached, evaluation of body weight and a
slight increase in dry weight will produce normotension. If severe hypotension
occurs, a reclining posture, increased salt and water intake, and decreased
ultrafiltration by the use of only 1.5% glucose solutions will usually suffice (see
Section 3.4). The patient is then placed on a normal diet and the CAPD regimen
is prescribed to conform with the usual salt and water intake. Reports of early
contraction of extracellular water as a cause of increased hematocrit and ortho-
static hypotension have been disputed [129-132].

6.3. Skin edema

Even when catheter site leakage is not present there is occasional accumulation of
fluid in the subcutaneous tissue during the early training phase. This may occur in
the anterior abdominal wall at the catheter site or may dissect from the per-
itoneum into the scrotum.
A very strange phenomenon of massive scrotal edema in patients who had
previously undergone aortic aneurysm repair has been observed in three patients
at this institution. Surgical exploration of the scrotum in one of these patients
failed to reveal any evidence of hydrocele, hernia, or other correctible causes.
The edema subsides with drainage of the abdominal cavity and a reclining
posture. Edema recurred, however, over a three- to four-day period following
the reinstitution of CAPD. By intermittently leaving the abdomen empty during
the night, the rate of reaccumulation gradually decreased. After three weeks of
this conservative management, the scrotal edema completely disappeared, and
the patient remained edema free.
Two other patients were subsequently seen with aortic aneurysm repair who
developed severe scrotal edema. They were treated in a similar fashion without
the surgical exploration, and a similar result obtained. Changes in the tissue
planes following aortic aneurysm repair appear to be the cause of this phe-
nomenon, and it seems to be self-limiting and is not a reason to permanently
discontinue CAPD. A surgical 'approach' is not indicated in these patients unless
 231

a hernia can be demonstrated or if the swelling will not remit with draining of the
abdomen and assuming the reclining posture overnight.

6.4. Motivation

Patients placed on CAPD who are internally motivated and compliant, do

extremely well, learn quickly and achieve rehabilitation status. There are, how-
ever, personality traits and family situations which create a likelihood of failure
[133,134].
Many patients adjust their lives to in-center hemodialysis so completely that
their friends, parties and enjoyment are found in this setting. Such patients, who
are transferred from hemodialysis because of vascular access or other problems,
do poorly on CAPD.

6.4.1. Depression - fatigue

Once training is complete and the patient goes home to an empty house without
social contacts, the rapid onset of depression and loneliness may be manifested by
multiple somatic complaints. Recurring episodes of peritonitis due to lack of
compliance should be quickly taken as a signal to make further attempts to
establish another vascular access for institution of hemodialysis.
Some patients develop an enthusiasm for CAPD with a misunderstanding of
the daily requirements. The enthusiasm soon wanes with the constant demand of
a careful, compliant exchange technique, the patient becomes disappointed and
disillusioned. This occurs most frequently in patients who have come directly
onto CAPD without having undergone other forms of dialytic therapy and can
frequently be reversed by a tour through the hemodialysis unit.
In explaining CAPD to patients, a balance must be maintained between
enthusiasm and blunt honesty to fully inform the patient of the benefits and side
effects of the procedure. This must include the understanding of the possibility of
gradual progressive obesity, the need for restriction of carbohydrate intake, strict
compliance, tenacious adherence to the protocol and the requirements of long-
term self-management.

6.4.2. Non-compliance
The failure of patients to comply with the protocols of CAPD is easily detected.
This usually manifests itself as recurrent episodes of peritonitis. At this time, a
discussion with the patient to explain the risks of non-compliance and the symp-
toms produced by the infection either produces a turnabout in attitude or a
willingness to be removed from CAPD.
Some patients will, however, fail to comply by skipping exchanges or failing to
observe even minimal dietary restrictions. For example, an occasional patient
may ingest up to 5000 to 6000 ml water/day. These patients usually express
232

amazement that they are drinking excessive water and deny that they are skipping
exchanges. BUN and creatinine blood values, however, give evidence of the
failure of the patient to comply with the number of required exchanges. If the
patient does not perform the number of exchanges prescribed, one or two days of
compliance does not significantly change the blood urea nitrogen and creatinine
levels. The patient's metabolic product evaluation will reveal this on the routine
laboratory tests. The absence of change in the transfer characteristics of the
peritoneal membrance must be ascertained prior to confronting the patient with
the evidence of decreased number of exchanges (5.3.2). If the patient is unwilling
to comply with the number of exchanges required, another form of dialysis will be
necessary. Patients whose motivation and compliance on hemodialysis were
marginal as evaluated by sodium and water balance, phosphate control and
potassium control will not necessarily manifest these problems on CAPD. So-
dium and water load are rarely a problem, and phosphate balance is more easily
controlled.

6.5. Anorexia and vomiting

The distension symptoms frequently present with the initiation of CAPD may be
manifested by epigastric distress, anorexia and, on occasion, dyspepsia, reg-
urgitation and eructation. The symptoms frequently disappear within four to five
days and may be related to the lingering uremic toxicity prior to dialysis control.
The epigastric distress is most likely to persist and is frequently cimetadine
reponsive. If, after three to five days, the symptoms have not abated with
cimetadine, psychological support, decreasing the intra-abdominal volume, and
radiographic evaluation should be undertaken. These are usually negative but
may reveal gastrointestinal pathology such as esophageal reflux, duodenal ulcers
or may suggest erosive gastritis.

6.6. Serum potassium abnormality (see previous discussion, p. 217)

6.7. Hyperglycemia

Hyperglycemia may occur in both insulin-requiring and non-insulin-requiring

patients with diabetes mellitus, who were previously controlled by diet or sta-
bilized on insulin [59]. The increased load of glucose from the dialysis solution
may produce an increased insulin requirement [124]. In patients known to have
diabetes mellitus, daily blood sugar levels should be drawn during the early phase
until control is achieved. When insulin is added to the dialysis solution, the
amount added to the overnight exchange should be decreased because the rapid
 233

glucose absorption during the first 2 to 3 hours dwell time, will be followed by a
continued absorption of insulin after the glucose is metabolized. Nocturnal
hypoglycemia may occur. Many authors now feel that CAPD is the preferred
form of dialysis for the diabetic patient [10, 135-139]' and report better blood
sugar control, improved control of hypertension, steady-state control of uremia,
and improved hematocrit and cardiac function [140, 141].

6.8. Anemia

A rapid rise in the hemoglobin level with the institution of CAPD has been
reported by most investigators [129-132, 142]. The mean hematocrit of 32 volumes
percent is significantly higher than the levels reported in hemodialysis patients.
There is usually not, however, a complete correction of anemia in patients
undergoing CAPD. Blood transfusions are rarely required. Even patients requir-
ing two to four units of blood per month on hemodialysis have not required blood
while on CAPD. Frequent monitoring of the serum ferritin is indicated to prevent
secondary iron deficiency anemia since rapid hemopoiesis may produce a fall in
the ferritin level [143]. Increased protein intake and intravenous iron has caused a
secondary rise in hematocrit. Some question of a late fall in hemoglobin levels six
to twelve months after the initiation of CAPD has arisen but did not reach
statistical significance. There is uncertainty about the rapid rise in the hemoglobin
level in CAPD patients. Some authors report an increased red blood cell (RBC)
[129, 131, 132] mass while others suggest only a fall in extracellular water and
hemo concentration [130].

6.9. Hernias and hemorrhoids

Increased intra-abdominal volume and pressure may aggravate already present

hemorrhoids as well as abdominal and inguinal hernias [78]. Repair of these prior
to a commitment of CAPD should be discussed with the patient but neither are a
specific contra-indication to the initiation of CAPD.

6.10. Back pain

Lumbosacral spine pain in patients who have previously had low back pain may,
on occasion, require discontinuation of CAPD. A search for significant symp-
toms in this area prior to the initiation of CAPD is indicated and should be
considered a relative contra-indication when present [78, 112].
234

6.11. Bloody dialysate

Female patients consistently report blood tinged dialysate with each menstrual
period. This causes concern because of the turbidity which is created during this
time. Symptoms of peritonitis are not present; heparinization has not been
required; and spontaneous disappearance occurs.
Rarely a patient will present with spontaneous bleeding into the peritoneal
cavity. This appears to be associated with abdominal cramps and diarrhea, and
spontaneously disappears. One patient reported grossly bloody dialysate after
adding heparin to the dialysate because of fibrin formation. The bleeding into the
dialysate did not occur, however, until the patient ingested aspirin, and also
disappeared without further treatment.

6.12. Peritonitis (see Chapter 13 for a detailed review of peritonitis)

The primary complication which limits the applicability of CAPO to patients with
end-stage renal disease is recurrent peritonitis. There has been a rapid decline in
the incidence of this complication with the availability of collapsible plastic
containers (October, 1978) and materials specifically designed for CAPO (Sep-
tember, 1979).
Peritonitis is the most frequent cause of discontinuation of CAPO; however,
many patients have utilized CAPO for three to four years without infection. This
suggests that except for the small statistical risk of organisms floating onto the
spike at the time of spike exposure during exchanges, the incidence of peritonitis
in the patient is related to catheter compliance or an inadequate catheter seal.
Material defects may be more common than is currently reported, and recurrent
episodes of peritonitis indicate the need for careful examination of both materials
and the rigid sterile technique.
The description of continuous ambulatory peritoneal dialysis in 1976 by Popo-
vich and Moncrief also included a recognition of the potential for recurrent
peritonitis [1]. This early prediction was confirmed and further studied in a
cooperative effort between these authors and the University of Missouri Medical
Center under the direction of Dr Karl Nolph [3]. Selection of patients who are
compliant and motivated and removal of patients from the technique who have
recurrent episodes of peritonitis in several programs has decreased the incidence
to less than one episode per patient year.

6.12.1. New developments

1. Ultraviolet Sterilization Chamber (TM) [8, 149] - an ultraviolet light re-
sterilizing system, developed by Popovich and Moncrief and Baxter Travenol
Laboratories, was introduced to the market in 1983. In vitro studies demonstrated
successful resterilization while clinical trials produced a reduction in the incidence
 235

of peritonitis. It was noted, however, that the control group also expressed a
reduction in the rate.
2. Sterile Connection Device (TM) - DuPont laboratory is testing a knife
splicing technique which may also reduce the risk of connection induced contam-
ination.
3. Microbiological Filter - Mion and others have reported a reduced incidence
of peritonitis using a 0.22 micron pore size microbiological filter on the inflow line
[11-13]. Spent dialysate may obstruct flow through the filter necessitating a
bypass.
Variation in the design of the studies used to evaluate these new techniques
make comparison of the results difficult. Failure of the ultraviolet light system to
demonstrate a statistical difference in the test and the control groups suggest
many episodes of peritonitis are not due to the spike-bag connection contam-
ination. This information has stimulated a renewed interest in peritoneal access
design.

6.12.2. Diagnosis
Peritonitis is diagnosed by an increased number of white blood cells in the
dialysate. Rubin et al. [126] studied clear dialysate samples and demonstrated
3-25 white blood cells per cubic millimeter in noninfected dialysate. Cells are
predominantly mononuclear. A rapid increase in the dialysate white blood cell
count occurs with the onset of peritonitis. Frequently, within four to eight hours,
the white blood cell count may rise to 5000 or more per cubic millimeter. Other
symptoms associated with the development of peritonitis include epigastric dis-
tress, nausea, vomiting, vague abdominal discomfort, rebound tenderness and
fever. These may progress to ileus, lethargy, severe abdominal pain, hypotension
and changes in sensorium. None ofthese symptoms are essential for the diagnosis
of peritonitis but should lead to the examination of the dialysate for an increased
number of white blood cells. A comparison of the white blood cell count using the
hemocytometer as opposed to the Coulter-counter demonstrates that the Coul-
ter-counter fails to count all of the cells. Therefore, the hemocytometer white
blood cell count is preferred.

6.13. Treatment

6.13.1. Prevention
The single best way to treat peritonitis is by prevention. This is best achieved by
meticulous development of a technique which, if followed correctly, should
produce a patient population with near zero incidence of peritonitis. Once this
has been accomplished, peritonitis can then be considered either a materials or
patient compliance failure. Each episode of peritonitis should stimulate a search
for the etiological factors. Fifty percent of patients (ten) at this institution have
236

performed the technique for greater than one year without peritonitis. The other
50% of patients have had one or more episodes of peritonitis within the first year
and 10% of our patients (two) have had 75% of our episodes of peritonitis.

6.13.2. Early treatment

Aerobic cultures are obtained the first day prior to initiation of therapy. Table 3 is
a suggested clinical approach to decide when to start and adjust therapy.
When the onset of peritonitis is limited to turbid dialysate with mild clinical
symptoms, initiation of therapy through addition of antibiotics to the dialysis
solution on an out-patient basis is usually successful. The CAPD exchange
schedule is continued at the prescribed rate. Seventy-five percent of the per-
itonitis episodes contracted in the out-patient program are gram positive. Doses
of cephalosporin 250 mg per exchange for four days followed by five to seven days
of oral cephalosporin (i.e. cephalexin 500 mg four times per day) has produced
complete clearing in 85% of the cases. A starting dose of tobramycin 1.0 mg/kg in
the first medicated container, followed by 15 mg/2 liter bag until culture results
are available, will allow coverage for the less frequent gram negative organisms.

6.13.3. Unresponsive peritonitis

Occasionally, a patient will present one day after the initiation of therapy with
progressive symptoms. These include increasing dialysate turbidity and systemic
toxicity. This patient should be immediately hospitalized as should any patient
presenting initially with severe symptoms. In these patients, repeat aerobic,
anaerobic, tuberculus and fungal cultures should be obtained. A gram stain for
fungal hyphae and acid fast bacillus to rule out tuberculosis may occasionally be
diagnostic. If evidence of these latter organisms is found, immediate steps to
drain the abdomen and remove the peritoneal catheter should be undertaken.
Patients who do not rapidly respond to cephalosporin or those with major
systemic toxicity should receive aminoglycocide antibiotics. This protocol is
frequently documented by a decrease in the number of white blood cells within a
24-hour period and may be interpreted as responsiveness to the antibiotics.
Continued intraperitoneal aminoglycosides at 10 mg per exchange in a patient
who is clinically responding will usually produce clearing within three to four
days. The rapid lavage of the peritoneal cavity with exchanges every 30 minutes
may increase the rate of clearing, but has been unnecessary at this institution. If
the patient is unresponsive to rapid lavage within 24 to 48 hours or if systemic
toxicity persists, drainage of the abdomen and removal of the catheter is indi-
cated. The presence or absence of systemic toxicity, the rapidity of response to
empiric cephalosporins, the inclusion of intraperitoneal aminoglycocide antibio-
tics, and/or a seriously ill patient all influence the therapeutic approach. Drainage
of the peritoneal cavity and removal of the catheter in those patients with
progressive systemic toxicity or failure to respond may be life saving. The former
would be the case in fungal peritonitis, tuberculosis peritonitis and anaerobic
 237

Table 3. Treatment of peritonitis

White blood cell count Course of action

per cubic centimeter (unspun dialysate)

Day 1
0-50 White blood cells per mm 3 Re-examine dialysate visually each hour during
Abdominal signs or symptoms present. next exchange.
Repeat white blood cell on dialysate if
symptoms persist.
No antibiotics.

50-100 White blood cells per mm 3 Aerobic culture.

Repeat white blood cell on next exchange.

No symptoms.
100 or greater white blood cells mm 3 • Aerobic culture.
With or without symptoms. Cephalosporin 500 mg per exchange for two
exchanges.
250 mg per exchange for 4 days
500 mg cephalosporin by mouth for 5 to 7 days
thereafter.
If symptoms severe, consider hospitalization.

Day 2
Symptoms improved and/or decrease in Continue above - follow by phone.
white blood cells in dialysate.

Symptoms unchanged and mild. Culture aerobic, anaerobic, fungal, TBC.

White blood cells not decreased or if it is Add gentamicin or tobramycin 100 mg in 21 for
increased. one exchange.
15 mg per exchange for three exchanges
cephalosporin by mouth.
Consider hospitalization.

Day 3
Symptoms improved. Consider only change to oral cephalosporin.
White blood cells decrease to less than
150mm3
(Check cultures and sensitivity.)

Symptoms improved. Continue intraperitoneal

White blood cells decreased but greater cephalosporin, 250 mg with each exchange
than 150 (mm3).
(Check cultures and sensitivity.)

Symptoms improved after aminoglycocides Continue gentamicin or tobramycin at 15 mg

started 2nd day. per two liters for four exchanges.
White blood cells decreased (evaluate Continue cephalosporin if sensitive.
sensitivity of positive culture).

Symptoms unimproved or worsened. Hospitalize.

White blood cells unchanged or increased. Rapid 15 to 30 min dwell lavage with
gentamicin or tobramycin 10 mg per two liters.
238

peritonitis associated with a perforated bowel. Early surgical correction of the

bowel perforation may be essential. Though not always fulminent, anaerobic,
intestinal tract organisms are virtually diagnostic of bowel perforation or leak
[150].

6.13.4. Antibiotic therapy

Addition of antibiotics to the dialysis solution is performed via sterile technique.
The medication port of the solution container is not guaranteed to be sterile [151]
and must be soaked for at least five minutes with povidone-iodine solution prior
to the injection of antibiotics. If symptoms and white blood cell counts are
diminishing, solutions containing cephalosporin antibiotic are continued for 10
days, or therapy may be changed to oral cephalosporin at day four. If no
improvement is noted, aminoglycocide therapy may be added (after anaerobic,
fungal and tuberculosis cultures are obtained). If clearing has not begun after 24
hours, the patient is hospitalized. If clearing is present, cephalosporin is con-
tinued as above, and the patient is contacted by telephone during this interim for
evaluation. Most centers continue the treatment for 10 days total.
Any patient who develops peritonitis should be evaluated for materials or
compliance failure. After the peritonitis is cleared, the patient should be returned
to the training program and several exchanges closely observed. The patient
should be closely questioned for possible materials defects. It has been the policy
of this facility that a transfer tubing change is always performed during peritonitis
treatment. An invisible crack in the spike or a puncture hole in the tubing may
produce several episodes of peritonitis before discovery. Clinical and subclinical
catheter tunnel infections must also be sought (Indium WBC).

6.14. Removal of patielits from CAPD

In those patients who have more than three episodes of peritonitis within the first
six months of CAPD, counseling should determine if compliance is the problem.
After the third episode, discontinuation of CAPD and a return to hemodialysis is
suggested. Patients are frequently quite willing to discontinue CAPD because of
the symptoms and the risk.

6.15. Staff-induced peritonitis

In the last eighteen months, four episodes of peritonitis in this program have been
related to contamination by unit personnel. Perforation of the catheter by a clamp
during a tubing change was responsible for two episodes. Failure of the titanium
adapter to seal in the catheter, after replacement of the previous plastic adapter,
produced two episodes. Clamping of the catheter should always be achieved with
 239

a soft clamp as close to the catheter adapter as is practical since any perforation of
the catheter near the skin can cause catheter loss. The change from a plastic to a
titanium adapter should include soaking the original adapter for at least ten
minutes in povidone-iodine solution, cutting the catheter at the connection site
proximal to the portion of the adapter in the catheter, and re-soaking the cut end
for 10 min prior to insertion of the titanium adapter. Since adopting this tech-
nique, no episodes of peritonitis have been associated with catheter adapter
change at this institution [125].

6.16. Peritonitis summary

Initiation of therapy with any elevated dialysate white blood cell count ac-
companied by close clinical follow-up to demonstrate clearing is critical. Admis-
sion to the hospital when response is slow followed by early removal of the
catheter when systemic toxicity is present may be life saving. Anaerobic cultures
to rule out bowel perforation in patients not responding rapidly to therapy and
removal of the catheter in any patient who develops fungal or tuberculosis
peritonitis will usually produce clearing and disappearance of symptoms. These
complications can be reduced by patient selection and a meticulous teaching
program.

7. Conclusion

An expanding worldwide experience with continuous ambulatory peritoneal

dialysis has demonstrated that this new technique has been used to successfully
manage patients with end-stage renal disease. The use of this technique in the
ESRD patient population has demonstrated that CAPD may be offered as an
alternative form of dialysis to those patients capable of and interested in self-
dialysis. The declining incidence in the peritonitis rate associated with introduc-
tion of better techniques and materials is encouraging [150]. Caution must be
exercised and adequate personnel available to manage those patients trained on
CAPD.

References

1. Popovich RP, Moncrief JW, Decherd JP, Bomar JB, Pyle WK: The definition of a novel
portable/wearable equilibrium peritoneal dialysis technique. Trans Am Soc Artif Intern Organs
(Abstract) 5: 64, 1976.
2. Moncrief JW, Popovich RP, Nolph KO: AdditIonal experience with continuous ambulatory
peritoneal dialysis (CAPO). Trans Am Soc Artif Intern Organs 24: 476-483, 1978.
3. Popovich RP, Moncrief JW, Nolph KO, Ghods A1, Twardowski Z1, Pyle WK: Continuous
ambulatory peritoneal dialysis. Ann Intern Med 88-4: 449-456, 1978.
240

4. Nissenson AR, Gentile DE, Soderblum R, Brax C: Longterm outcome of CAPD - regional
experience, Dial Transpl13: 34-39, 1984.
5. Oreopoulos DG, Robson M, Izatt S, Clayton SL, DeVeber GA: A simple and safe technique for
continuous ambulatory peritoneal dialysis (CAPD). Trans Am Soc Artif Intern Organs 24: 484-
488,1979.
6. Oreopoulos DG: The coming of age of continuous ambulatory peritoneal dialysis (CAPD). Dial
Trans 8-5: 460-464, 1979.
7. Moncrief JW, Rutherford CE, Sorrels PAJ, Bailey A, Popovick RP: Technical aspects of
continuous ambulatory peritoneal dialysis new connection devices developed by Baxter Trav-
enol laboratories. In: continuous Ambulatory Peritoneal Dialysis, Ed Legrain M, Excerpta
Medica, Amsterdam, p 79-81, 1980.
8. Sorrels PAJ: New materials, In: Moncrief J, Popovich R (Ed) Proc CAPD Int Symp II, Masson,
Austin, Texas, p 167, 1980.
9. Moncrief JW, Mullins-Blackson C, LeBourgeois J, Popovich RP, Pyle WK: Development and
testing of an ultraviolet light resterilizing procedure for continuous ambulatory peritoneal
dialysis (CAPD), Abstract, 4th ISAD Offic Satel Symp on CAPD, Kyoto, p 56, 1983.
10. Flynn CT, Strosahl V: Self-Care by the blind diabetic on continuous ambulatory peritoneal
dialysis, Diab Neph 1: 22-24, 1982.
11. Mion C, Slingeneyer A, Liendo-Liendo C, Perez C; Despaux E: Reduction in incidence of
peritonitis associated with continuous ambulatory peritoneal dialysis, Proc Clin Dial Transplant
Forum 9: 9, 1979.
12. Winchester JF, Ash SR, Bousquet G, Rakowski TA, Barnard WF, HeeterE, Haley S: Success-
ful peritonitis reduction with a unidirectional bacteriologic CAPD filter, Trans Am Soc Artif
Intern Organs 29: 611-615, 1983.
13. Slingeneyer A, Mion C: Peritonitis prevention in continuous ambulatory peritoneal dialysis:
long-term efficacy of a bacteriological filter, Proc Eur Dial Transplant Assoc 19: 388, 1983.
14. Hamilton R, Adams P, Burkart J, Disher B, Dillingham G, Crater C: Feasibility of a sterile
splice for connection in continuous ambulatory peritoneal dialysis, Trans Am Soc Artif Intern
Organs 29: 623-627, 1983.
15. Legrain M: Introductory remarks. In: Continuous Ambulatory Peritoneal Dialysis, Ed Legrain
M, Excerpta Medica, Amsterdam, p 3--4, 1980.
16. Legrain M: Continuous Ambulatory Peritoneal Dialysis, Ed Legrain M, Excerpta Medica,
Amsterdam, 1980.
17. Bryan FA Jr, Evans RW: Hemodialysis parameters. Kidney Int 7-3: 350-356, 1980.
18. Enderson LW: Symptomatic hypotension during hemodialysis. Kidney Int 17-5: 571-576,1980.
19. Nolph KD: The peritoneal dialysis system. In: Today's Art of Peritoneal Dialysis, Eds Trevino-
Becerra A, Boen FST, Karger S. Basel, 17: 4-50,1979.
20. Villarroel F: Future of intracorporeal dialysis. In: Today's Art of Peritoneal Dialysis, Eds
Trevino-Becerra A, Boen FST, Karger S. Basel, 17: 111-119,1979.
21. Boen ST: Undesirable effects of chronic peritoneal dialysis. In: Today's Art of Peritoneal
Dialysis, Eds Trevino-Becerra A, Boen FST, Karger S, Basel, 17: 73-76, 1979.
22. Boen ST: Overview and history of peritoneal dialysis, Peritoneal Dialysis, Wichtig Editore,
Milano, pp 9-16, 1982.
23. Popovich RP, Moncrief JW: Kinetic modeling of peritoneal transport. In: Today's Art of
Peritoneal Dialysis, Eds Trevino-Becerra A, Boen FST, Karger S. Basel, 17: 59-72, 1975.
24. Popovich RP, Moncrief JW, Decherd JF: Physiological transport parameters in peritoneal and
hemodialysis in 2nd Annu. Report No. Am-2-3-2205. Artificial Kidney-Chronic Uremia Pro-
gram, national Institute of Arthritis, Metabolism and Digestive Diseases. HEW, 1977.
25. Popovich RP, Pyle WK, MoncriefJW, Decherd JF, Brooks S: Preliminary verification ofthe low
dialysis clearance hypothesis via a novel equilibrium peritoneal dialysis technique. Trans Austral
Conf Heat and Mass Transfer, Univ of Sidney Press, Sidney, Australia p 2: 217, 1977.
 241

26. Popovich RP: Kinetics of peritoneal transport. In: Peritoneal Dialysis. Wichting Editore,
Milano, pp 49-62, 1982.
27. Nolph KD: Concluding remarks. In: Today's Art of Peritoneal Dialysis. A. Trevino-Becerra,
FST Boen (eds). Karger S, Basel, 17: 146-148, 1979.
28. Cantarovich F, Perez Loredo J, Chena C, Wilberg R, Vernetti J. Correa C, Tizado J: C.A.P.D.
- 3 Daily exchanges. In: Moncrief J. Popovich R (Ed) Proc CAPD Int Symp II, Masson, Austin,
Texas, p 125, 1980.
29. Twardowski ZJ, Prowant BF, Nolph KD, Martinez AJ, Lampton LM: High volume, low
frequency continuous ambulatory peritoneal dialysis, Kidney Int 23: 64-70, 1983.
30. Babb AL, Popovich RP, Christopher TG, Scribner BH: The genesis of the square meter-hour
hypothesis. Trans Am Soc Artif Intern Organs 17: 81-91, 1971.
31. RubinJ, Nolph KD, Popovich RP, Moncrief JW, Prowant B: Drainage volumes during CAPD.
ASAIO J 2: 2, 1979.
32. Miller FN, Nolph KD, Joshua IG: The osmolality component of peritoneal dialysis solutions in
Continuous Ambulatory Peritoneal Dialysis, Ed Legrain M, Excerpta Medica. Amsterdam, p
12-17,1980.
33. Henderson LW, Nolph KD: Altered permeability of the peritoneal membrane after using
hypertonic peritoneal dialysis fluid. J Clin Invest 48: 992-1001, 1969.
34. Baum M, Powell D, McHenry K, Potter D: Comparison of continuous ambulatory peritoneal
dialysis (CAPD) and hemodialysis in children, Abstract, Am Soc Nephrol, p 50A, 1982.
35. Kohaut EC. Alexander SR: Ultrafiltration in the young patient on CAPD. In: Moncrief J,
Popovich R (ed) Proc CAPD Int Symp.n, Masson, Pub, Allstin, Texas, p. 125, 1980.
36. Simmons EE, Lockard DT, Moncrief JW, Hiatt MP, McCollough WS, Popovich RP: Experi-
ence with continuous ambulatory peritoneal dialysis in the maintenance of a surgically anephric
dog. Southwest Vet 33-2: 129-135,1980.
37. Pyle WK, Moncrief JW, Popovich RP: Peritoneal transport evaluation in CAPD. In: Moncrief J,
Popovich R (eds), Proc CAPD Int Symposium II, Masson, Pub, Austin, Texas, p. 35, 1980.
38. Boen ST, Haagsma-Schouten WAG, Birnie RJ: Change in ultrafiltration capacity during
peritonitis in CAPD patients, Abstract, 4th ISAO Offic Satel Symp on CAPD, Kyoto, p 30,
1983.
39. Hattori F, Nakamoto M, Miishima C, Fujimi S: Structural change of peritoneum in CAPD,
Abstract, 4th ISAO Offic Satel Symp on CAPD, Kyoto, p 38, 1983.
40. Slingeneyer A, Mion C, Mourad G, Canaud B, Faller B, Beraud 11: Progressive sclerosing
peritonitis: a late and severe complication of maintenance peritoneal dialysis, Trans Am Soc
Artif Intern Organs 29: 633-638, 1983.
41. Diaz-Buxo J A, Farmer CD, Walker PJ, Chandler JT, Holt KL: Effects of hyperparathyroidism
on peritoneal clearances, Trans Am Soc Artif Intern Organs 28: 276-278, 1982.
42. Manuel MA: Failure of ultrafiltration in patients on CAPD. Perit Dial Bull (Suppl) 3: 838-840,
1983.
43. Wu G, Khanna R, Oreopoulos DG, Vas SI: Incidence and pathogenesis of ultrafiltration failure
among CAPD patients. The Am Soc Nephrol, p 125A, 1983.
44. Nolph KD et ai.: A survey of ultrafiltration in continuous ambulatory peritoneal dialysis, An
International Cooperative Study, Second Report.
45. Bazzato G, Coli U, Landini S, Fracasso A, Morachiello P, Righetto F, Scanferla F: Xylitol and
low doseages of insulin: new perspectives for diabetic uremic patients on CAPD Perit Dial Bull
2: 161-164, 1982.
46. Maher JF, Hirszel P, Bennett RR, Chakrabarti E: Amphotericin B selectively increases per-
itoneal ultrafiltration, The Am Soc of Nephrol, p 121A, 1983.
47. Sacchi VA, Bentzel CJ, Mookerjee BK, Beam TB: Protamine Sulfate (PS) augments peritoneal
clearances in a rabbit model of CAPD and enhances antimicrobial activity In vitro, Abstracts,
Kidney Int 23: 159,1983.
242

48. Alavi N, Lianos E, Bentzel C: Enhanced peritoneal permeability of the rat, by intraperitoneal
use of protamine sulfate, Abstracts, Kidney Int 16: 880, 1979.
49. Alvai N, Lianos E, Andres G, Bentzel CJ: Effect of protamine on the permeability and structure
of rat peritoneum, Kidney Int 21: 44-53, 1982.
50. Rubin J, Adair C, Barnes T, Bower JD: Augmentation of peritoneal clearance by dipyridamole,
Kidney Int 22: 658-661, 1982.
51. Blumenkrantz MJ, Kopple JD, Moran JK, Grodstein GP, Coburn JW: Metabolic balance
studies in patients undergoing continuous ambulatory peritoneal dialysis. Am Soc Nephrol
(Abstract) 112A, 1979.
52. Nolph KD. Rosenfield PS, Powel JT. Danforth E: Peritoneal glucose transport and hyper-
glycemia during peritoneal dialysis. Am Med Sci 259: 272-281. 1970.
53. Grodstein GP. Blumenkrantz MJ, Kopple JD, Moran JK. Coburn JW: Glucose absorption
during continuous ambulatory peritoneal dialysis (CAPD). Abstracts, Kidney Int 22: 220, 1979.
54. Oreopoulos DG: Further experience with the use of amino acid containing dialysate (Amino-
Dianeal). In: Moncrief J. Popovich R (Ed). Proc CAPD Int Symp II. Masson, Pub. Austin,
Texas, p 109, 1980.
55. Cantarovich F, Perez Loredo J. Chena C, Reichart M, Tizado J. Castro L, Biana D, Juliamolli
V: Additives in CAPD. In: Moncrief J, Popovich R (Ed), Proc CAPD Int Symp II, Masson,
Austin. Texas, p 117, 1980.
56. Bazzato G. Coli U, Landini S. Fracasso A, Morachiello P, Righetto F, Scanferla F, Onesti G:
Xylitol as osmotic agent in CAPD: an alternative to glucose for uremic diabetic patients?, Trans
Am Soc ArtifIntern Organs 23: 280-285. 1982.
57. Heaton A, Ramos M, Johnston D. Gokal R. Ward MK, Kerr DNS: Glucose and lipid metabo-
lism in continuous ambulatory peritoneal dialysis (CAPD), Abstracts. Kidney Int 22: 220, 1982.
58. Moncrief JW, Pyle WK, Simon P, Popovich RP: Hypertriglyceridemia diabetes mellitus, and
insulin administration in patients undergoing continuous ambulatory peritoneal dialysis. In:
Moncrief J, Popovich R (Ed), Proc CAPD Int Symp II, Masson, Austin, Texas, p 143,1980.
59. Nolph KD. Hopkins CA, New D, Antwiler GD. Popovich RP: Differences in solute sieving with
osmotic vs hydrostatic ultrafiltration. Trans Am Soc Artif Intern Organs 22: 618-626, 1976.
60. Moncrief JW, Popovich RP: Continuous ambulatory peritoneal dialysis. In: Today's Art of
Peritoneal Dialysis, Eds Trevino-Becerra A. Boen FST, Karger S. Basel, 17: 139-145, 1979.
61. Moncrief JW: Round table discussion in Continuous Ambulatory Peritoneal Dialysis. Ed
Legrain M, Excerpta Medica, Amsterdam. P 174-178. 1980.
62. Nolph KD, Moncrief JW, Popovich RP: Multi-center evaluation of CAPD in Annu Report,
Artificial Kidney-Chronic Uremia Program, National Institute of Arthritis, Metabolism and
Digestive Disease, HEW, 1980.
63. Blumenkrantz MJ: Nutritional aspects of peritoneal aspects of peritoneal dialysis. In: Proc
Renal Physicians Assoc Symp Perit Dial.
64. Nolph KD, Twardowski ZJ, Popovich RP: Equilibration of peritoneal dialysis solutions during
long-dwell exchanges. J Lab Clin Med 93: 246-256, 1979.
65. Kobayaski K, Manji T, Hiramatsu S. Maeda K, Uemura J: Nitrogen metabolism in patients on
peritoneal dialysis. In: Today's Art of Peritoneal DialYSIS. A TreVIno-Becerra, FST Boen (eds),
Karger S, Basel, 17: 93-100, 1979.
66. Gokal R, Fryer R, McHugh M, Ward MK, Kerr DNS: CalcIUm and phosphate control in
patients on continuous ambulatory peritoneal dialysis. In: Continuous Ambulatory Peritoneal
Dialysis. Ed Legrain M. Excerpta Medica. Amsterdam. p 283-291. 1980.
67. Gokal R, Ellis HA, Ward MK, Kerr DNS: Histological renal bone disease in patients on
continuous ambulatory peritoneal dialysis. In: Moncrief J, Popovich R (Ed) Proc CAPD Int
Symp II, Masson, Austin, Texas. p 249. 1980.
68. Gokal R: Metabolic Effects (calcium. phosphate, vitamin D, PTH, bone disease and trace metal
metabolism) in CAPD. Abstract, 4th ISAO Offic Satel Symp on CAPD Kyoto, p 84,1983.
 243

69. Calderaro V, Oreopoulos DG, Meema HE, Khanna R, Quinton S, Carmichael D: Renal
ostodystrophy in patients on continuous ambulatory peritoneal dialysis (CAPD), a biochemical
and radiological study. In: Moncrief J, Popovich R (Ed), Proc CAPD Int Symp II, Masson,
Austin, Texas, p 243, 1980.
70. Blumenkrantz MJ: CAPD: Nutritional concerns - have we learned from the experience in the
1960's and 1970's? In: Moncrief J, Popovich R (ed) Proc CAPD Int Symp II, Masson, pub,
Austin, Texas, p 83, 1980.
71. Blumenkrantz MJ, Kopple JD, Moran JK, Grodstein GP, Coburn JW: Metabolic balance
studies in uremic patients undergoing continuous ambulatory peritoneal dialysis (CAPD),
Abstracts, Kidney Int 16: 882, 1979.
72. Moncrief JW, Popovich RP, Nolph KD, Rubin J, Robson M, Dombros N, DeVeber GA,
Oreopoulus DG: Clinical experience with continuous ambulatory peritoneal dialysis. ASAIO J
2-3: 114-119, 1979.
73. Lindholm B, Ahlberg M, Alvestrand A, Furst P, Larlander SG, Gergstrom J: Nutritional
aspects of continuous ambulatory peritoneal dialysis. In: Continuous Ambularoty Peritoneal
Dialysis, Ed Legrain M, Excerpta Medica, Amsterdam, p 199-206, 1980.
74. Gahl GM, Baeyer HV, Riedinger R, Borowzak B, Schuring R, Becker H, Kessell M: Caloric
intake, and nitrogen balance in patients undergoing CAPD. In: Moncrief J, Popovich R (ED),
Proc CAPD Int Symp II, Masson, Austin, Texas, p 87, 1980.
75. Lindholm B, Furst P, Alvestrand A, Bergstrom J (introduction by K Nolph): Metabolic effects
of continuous ambulatory peritoneal dialysis (CAPD). Abstracts Kidney Int 16: 892,1979.
76. von Baeyer H, Gahl GM, Riedinger H, Borowzak R, Averdunk R, Schurig R, Kessel M:
Adaptation of CAPD patients to the continuous peritoneal energy uptake, Kidney Int 23: 29-34,
1983.
77. Medical Letter, Home Peritoneal dialysis for end-stage renal disease, 21-17, Issue 538, 1979.
78. Oreopoulos DG, Vas S, Zellerman G: Continuous ambulatory peritoneal dialysis. In: Proc 12th
Ann Contractors ConL Artificial Kidney-Chronic Uremia Program, National Institute of Ar-
thritis, Metabolism and Digestive Diseases, January, 1979, p 58-65.
79. Miguel Alonso JL, Martinez ME, Selgas R, Carares M, Gomez P, Sanchez Sicilia L: Peritoneal
clearance of parathormone, Abstracts, Kidney Int 22: 216, 1982.
80. Gokal R: Round table discussion. In: M Legrain (ed) Continuous Ambulatory Peritoneal
Dialysis. Excerpta Medica, Amsterdam, 1980, pp 209-313.
81. Nolph KD, Sorkin M, Rubin J, Dariush A, Prowant B, Fruto L, Kennedy D: Continuous
ambulatory peritoneal dialysis: Threeyear experience at one center. Ann Intern Med 92-5: 609-
613,1980.
82. Moncrief JW, Sorrels PAJ, Druger VJ, Mullins-Blackson C, Pyle WK, Popovich RP: Develop-
ment of training programs for continuous ambulatory peritoneal dialysis - historical review. In:
M Legrain (ed) Excerpta Medica, Amsterdam, 1980, pp 149-151.
83. Teehan BP, Schleifer CR, Cupit M, Knapp J, Miles HS: Organized aspects of a continuous
ambulatory peritoneal dialysis program. In: M Legrain (ed), Continuous Ambulatory Peritoneal
Dialysis. Excerpta Medicia, Amsterdam, 1980, pp 152-157.
84. Sorrells PAJ, Kruger VJ, Moncrief JW, Popovich RP: Austin Diagnostic Clinic, continuous
ambulatory peritoneal dialysis training program. In: M Legrain (ed), Continuous Ambulatory
Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1980, p 167-170.
85. Clayton S, Finer C, Quinton C, Jabaz 0, Clark S, Lekman B, Oreopoulos DG: Training patients
for continuous ambulatory peritoneal dialysis at the Toronto Western Hospital. In: M Legrain
(ed), Continuous Ambulatory Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1980, pp 162-
166.
86. Prowant B, Fruto LV: Inpatient home training for continuous ambulatory peritoneal dialysis.
In: M Legrain (ed), Continuous Ambulatory Peritoneal Dialysis. Excerpta Medica, Amster-
dam, 1980, pp 158-161.
244

87. Khanna R, Wu G, Vas S, Oreopoulos DG: Mortality and morbidity on continuous ambulatory
peritoneal dialysis, ASAIO J 6: 197-204, 1983.
88. Oreopoulos DG: Selection criteria and clinical results of continuous ambulatory peritoneal
dialysis. In: M Legrain (ed), Continuous Ambulatory Peritoneal Dialysis. Excerpta Medica,
Amsterdam, 1980, pp 101-106.
89. Price JDE, Moriarty MV: Continuous ambulatory peritoneal dialysis: selection criteria - failures
and causes - deaths - diabetes mellitus. In: M Legrain (ed), Continuous Ambulatory Peritoneal
Dialysis. Excerpta Medica, Amsterdam, 1980, pp 113-119.
90. Shaldon S: A cynical critique of continuous ambulatory peritoneal dialysis. In: M Legrain (ed),
Continuous Ambulatory Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1980, pp 137-140.
91. Zapacosta A: Patient selection for CAPD. Seminar on CAPD, Philadelphia, 1980.
92. Alexander SR, Tseng CH, Maksym KA, Campbell RA, Talwalkar YB: Clinical parameters in
continuous ambulatory peritoneal dialysis for infants and children. In: J Moncrief, R Popovich
(ed) Proc CAPD Int Symp II, Masson, Austin, Texas, 1980, p 195.
93. Balfe JW, Irwin MA: Continuous ambulatory peritoneal dialysis pediatrics. In: M Legrain (ed),
Continuous Ambulatory Peritoneal Dialysis, Excerpta Medica, Amsterdam, 1980, pp 131-136.
94. Honda M, Ito H, Hasegawa 0, Nakahara C, Hayashi S, Kawamura T, Hasegawa A, Ogawa 0:
Nutritional status and growth in children on CAPD, Abstract, 4th ISAO Offic Satel Symp on
CAPD, Kyoto, 1983, P 86.
95. Baum M, Powell D, McHenry K, Potter D: Comparison of Continuous ambulatory peritoneal
dialysis (CAPD) and hemodialysis in children, Abstracts, Kidney Int, 23: 143,1983.
96. Salusky IB, Kopple JD, Fine RN: Continuous ambulatory peritoneal dialysis in pediatric
patients: a 20-month experience, Kidney Int 24: S101-S105, 1983.
97. Broyer M, Niaudet P, Champion G, Jean G, Chopin N, Czernichow P: Nutritional and
metabolic studies in children on continuous ambulatory peritoneal dialysis. Kidney Int 24: SlO6--
SlIO, 1983.
98. Salusky IB, Fine RN, Nelson P, Kopple J: Growth and nutritional status (NS) in children
receiving CAPD, Abstracts, Kidney Int 23: 159,1983.
99. Harmon WE: Continuous ambulatory peritoneal dialysis in children. (Letters) N Engl J Med
308: 968, 1983.
100. Baum M, Potter DE: Continuous ambulatory peritoneal dialysis in children. (Letters) N Engl J
Med: 308: 968, 1983.
101. Gruskin AB, Rosenblum H, Baluarte HJ, Morgenstern BZ, Polinsky MS, Perlman SA: Trans-
peritoneal solute movement in children, Kidney Int 24: S95-S100, 1983.
102. Morgenstern B, Pyle WK, Gruskin A, Baluarte HJ, Perlman S, Polinsky M, Kaiser B: Transport
characteristics of the pediatric peritoneal membrane, Am Soc Nephrol, p 122A, 1983.
103. Conley S: Pediatric renal nutrition, presented at the Symposium for Texans Active in Renal
Therapy (START), 1984.
104. Kohaut EC: Growth in children with end-stage renal disease treated with continuous ambula-
tory peritoneal dialysis for at least one year, Perit Dial Bull 2: 159-161, 1982.
105. Alexander SR, Lubischer JT: Continuous ambulatory peritoneal dialysis in pediatrics: three
years' experience at one center, Nefrologia 2: 53-62, 1982.
106. Alexander SR: Pediatric CAPD update - 1983, Perit Dial Bull (Supp!) 3: SI5-S22, 1983.
107. Flynn CT: Continuous ambulatory peritoneal dialysis in the diabetic patient. In: Legrain M (ed),
Continuous Ambulatory Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1980, p 187-193.
108. Lee HB, Whang SK, Ihm CG, Chun S: Cell-mediated immunity (CMI) in CAPD patients. Am
Soc Nephrol, p 121A, 1983.
109. Giacchino F, Alloatti S, Quarello F, Coppo R, Pellerey M, Piccoli G: The influence of peritoneal
dialysis on cellular immunity. Peri Dial Bull 2: 165, 1982.
110. Williams P, Key R, Harrison J et al.: Nutritional and anthropometric assessment of patients on
CAPD over one year: contrasting changes in total body nitrogen and potassium. Perit Dial Bull
6: 82-87, 1981.
 245

111. Rubin J, Raju S, Teal N et al.: Abdominal hernia in patients undergoing continuous ambulatory
peritoneal dialysis, Arch Intern Med 142: 1453-1455,1982.
112. Hamodraka-Mailis A: Pathogensis and treatment of back pain in peritoneal dialysis patients,
Perit Dial Bull (Suppl) 3: S41-S43, 1983.
113. Tenchkoff H, Schechter H: A bacteriologically safe peritoneal access device. Trans Am Soc
Artif Intern Organs 14: 181-186, 1968.
114. Oreopoulos DG, Zellerman G, Izatt S: The Toronto Western Hospital permanent peritoneal
catheter and continuous ambulatory peritoneal dialysis connector. In: M Legrain (ed), Continu-
ous Ambulatory Peritoneal Dyalysis. Excerpta Medica, Amsterdam, 1980, pp 73-78.
115. Valli A, Comotti C, Torelli, Crescimanno U, Valentini A, Riegler P, Huber W, Borghi M,
Gruttadauria C, Scarovanati P, Pecchini F: A new catheter for peritoneal dialysis (two years of
experience with Valli Catheter), Trans Am Soc Artif Intern Organs 29: 629-631, 1983.
116. Nolph KD: Personal communication, 1980.
117. Tenckhoff H: Chronic Peritoneal Dialysis: In: A Manual for patients, dialysis personnel and
Physicians, Univ. of Washington School of Medicine, Seattle, Washington, 1974.
118. Colombi A, Bianella C: Straight implantation of the Tenckhoff catheter for continuous ambula-
tory peritoneal dialysis. In: M Legrain (ed), Continuous Ambulatory Peritoneal Dialysis.
Excerpta Medica, Amsterdam, 1980, pp 69-72.
119. Kirksey T: Surgical implantation of the singlecuff (acute) Tenckhoff catheter. In: Proc CAPD
Int Symp II, Masson, 1980 (in press).
120. Blumenkrantz MJ: Personal communication, 1984.
121. Helfrich GB, Pechan BW, Alijani MR, Barnard WF, Rakowski TA, Winchester JF: Reduction
of catheter complications with lateral placement, Perit Dial Bull (Suppl) 3: S2-S5, 1983.
122. Alexander SR, Tank ES: Surgical aspects of continuous ambulatory peritoneal dialysis in
infants, children and adolescents, J Uro1127: 501-504, 1982.
123. Alexander SR, Tank ES: Technical considerations in the implantation of Tenckhoff catheters
for continuous ambulatory peritoneal dialysis in children, Nefrologia 2: 49-52, 1982.
124. Flynn CT, Nanson JA: Intraperitoneal insulin with CAPD - an artificial pancreas. Trans Am Soc
Artif Intern Organs 25: 114-116, 1979.
125. Going Home with Confidence, Procedure for Continuous Ambulatory Peritoneal Dialysis Staff
Program, Baxter Travenol Laboratories Inc., 1979.
126. Rubin J, Rogers WA, Taylor HM, Everett ED, Prowant DB, Fruto LV, Nolph KD: Peritonitis
during continuous ambulatory peritoneal dialysis. Ann Intern Med 92-1: 7-13, 1980.
127. Moncrief JW: Continuous ambulatory peritoneal dialysis concepts and clinical application. In:
M Marti, A Locatelli (ed), Int Symp Chronic Perit Dial, Buenos Aires, Argentina, 1980, p 153.
128. Osmond H, Loh Y, Dombros D, Oreopoulos DG, Paroport A: Effects of CAPD on renin-
angiotensin system. Am Soc Nephrol (Abstract) 67A, 1978.
129. DePaepe M, Schelstraete K, Ringoir S, Lameire NH: Influence of continuous ambulatory
peritoneal dilaysis on the anemia of endstage renal disease. Kidney Int 23: 744-748, 1983.
130. Mehta BR, Mogridge C, Bell JD: Changes in red cell mass, plasma volume and hematocrit in
patients in CAPD, Trans Am Soc Artif Intern Organs 29: 50--52, 1983.
131. De Paepe M, Lameire N, Schelstraete K, Ringoir S: Changes in red cell mass, plasma volume
and hemotocrit in patients on continuous ambulatory peritoneal dialysis. Proc Eur Dial Transpl
Assoc 18: 286, 1981.
132. Zappacosta Ar, Caro J, Erslev A: Normalization of hemotocrit in patients with end-stage renal
disease on continuous ambulatory peritoneal dialysis, Am J Med 72: 53, 1982.
133. Lindsay RM, Oreopoulos DG, Burton H, Conley J, Wells G: A comparison of CAPD and
hemodialysis in adaptation to home dialysis. In: J Moncrief, R Popovich (eds), Proc CAPD Int
Symp II. Masson, Austin, Texas, 1980, p 171.
134. Moncrief JW, Popovich RP: Peritoneal dialysis for a greater number of patients? In: Controver-
sies in Nephrology, Ed Schreiner CE, Georgetown University, Washington, D.C., p 31-44,
1979.
246

135. Berger PS, Alpert BE, Longnecker RE: Dialysis therapy for Diabetics, Diab Nephrol2: 22-25,
1983.
136. Sorkin MI, Luger AM, Prowant B, Kennedy J, Moore H, Nolph KD: Histological and func-
tional characteristics of the peritoneal membrane of a diabetic patient after 34 months of CAPD,
Perit Dial Bull 2: 24-27, 1982.
137. Wilhams C, Belvedere D, Cattran D, Clayton S, Cole E, Fenton S, Gutman K, Khanna R,
Knight S, Manuel A, Oreopoulos D, Pierratoss A, Roscoe J, Saiphoo C, Vas S: Experience with
CAPD in diabetic patients in Toronto, Perit Dial Bull 2: S12-S16, 1982.
138. Roscoe JM: Practices of insulin administration in CAPD, Perit Dial Bull 2: S27-S29, 1982.
139. Posen G, Lam E, Rappoport A: The management of end-stage renal disease (ESRD) in diabetes
mellitus (DM) in Canada in 1981, Trans Am Soc Artif Intern Organs 29: 116-118, 1983.
140. Amair P. Khanna R, Leibel B, Pierratos A, Vas S, Meema E, Blair G, Chisholm L, Vas M,
Zingg W, Digenis G, Oreopoulos DG: Continuous ambulatory peritoneal dialysis in diabetics
with end-stage renal disease. Perif Dial Bull (Suppl) 2: S6-Sll, 1982.
141. Leenen F, Smith DL, Khanna R, Oreopoulos DG: Changes in left ventricular anatomy and
function on CAPD. Perit Dial Bull (Suppl) 3: S26-S28, 1983.
142. Lamperi S. Carozzi S, Icardi A: In vitro and in VIVO studies of erythropoiesis during continuous
ambulatory peritoneal dialysis. Perit Dial Bull 3: 94-96, 1983.
143. Goldsmith HJ, Forbes A. Gyde OHD, Summerfield G: Hematological aspects of continuous
ambulatory peritoneal dialysis. In: M Legrain (ed). Continuous Ambulatory Peritoneal Di-
alysis. Excerpta Medica. Amsterdam, 1980. pp 302-308.
144. Nolph KD, Pyle WK, Hiatt M: Mortality and morbidity in continuous ambulatory peritoneal
dialysis. ASAIO J 6: 220-226. 1983.
145. Steiner RW, Kipper S, Savoia MC, Witztum KF: Identification by Scannin with indium 111
labeled lewkocytes, Ann Intern Med 99: 44-45, 1983.
146. Peritonitis in Chicago CAPD patients is traced to defects in catheters. Nephrol Rep 2: Jan. 1984.
147. Pisani E, Biasioli S, Borin D, Chiaramonte S, Fabns A, Feriani M, Ronce C, La Greca G:
Overview of peritoneal access. In: Peritoneal Dialysis. Wichtig Editore, Milano, 1982, pp 121-
130.
148. Slingeneyer A: Surgical implantation of the Tenckhoff catheter. In: Peritoneal Dialysis. Wichtig
Editore, Milano, 1982, pp 131-137.
149. Popovich RP, Moncrief JW, Sorrels-Akar P, Mullens-Blackson C, Pyle WK, Bellotti M, Taylor
L: An ultraviolet germicidal system to alleviate touch contamination in CAPD. Am Soc
Nephrol. 1983, p 123A.
150. Nolph KD, Sorkin MI: Diagnosis and treatment of peritonitis. In: J Moncrief, R Popovich (eds),
Proc CAPD Int Symp II. Masson, Austin, Texas, 1980, p 273.
151. Baxter Travenol Laboratories, personal communication.
9. Continuous cyclic peritoneal dialysis

JOSE A. DIAZ-BUXO

1. Introduction

Continuous cyclic peritoneal dialysis (CCPD) was introduced in 1980 [1], based
on the concept of continuous equilibration dialysis described by Popovich et al.
[2]. The primary objective was to provide automated continuous equilibration
peritoneal dialysis, with most exchanges taking place at night, and a long-dwell
diurnal exchange. The secondary goal was to reduce the rate of peritonitis.

2. Technique

Peritoneal access is provided by a permanent silastic catheter of the straight,

curled, or column-disc type. CCPD requires an automated cycler capable of
delivering variable volumes of dialysate for a prescribed dwell time. The per-
itoneal catheter is connected to the cycler line before the patient retires at night.
Three or four cycles are generally administered during the night, each lasting 2 to
3 hours, using 2 L of commercial dialysate per exchange. An additional exchange
is effected in the morning, prior to disconnection. The 2 L of dialysate infused in
the morning are allowed to dwell intraperitoneally for the next 14 to 15 hours with
the catheter capped. Hypertonic dialysate containing 2.50% to 4.25% dextrose is
recommended for the diurnal cycle in order to prevent significant absorption of
the solution. All connections and disconnections take place in the early morning
and at night, in the convenience of the patient's home. The average length of time
required to set up the equipment and connect the catheter to the cycler is 20
minutes.

2.1. Cyclers

The peritoneal cycler's main functions are to deliver a prescribed volume of

248

dialysate to the peritoneal cavity and to allow the dialysate to dwell in the
peritoneal cavity, followed by a period of drainage. In addition, many cyclers
incorporate safety and comfort features such as solution heaters and ultrafiltra-
tion monitors. Most cyclers are designed after Lasker's model [3] using gravity for
infusion and drainage of dialysate (Fig. IA). Commercial dialysate in bottles or
plastic containers can be utilized. The first step in intiating dialysis is to set the
cycler controls to determine the volume of exchanges, length of intraperitoneal
dwell, and outflow or drainage time. The dialysate flows from the containers to a
heating cabinet that will also determine the volume of inflow. Once a tempera-
ture of approximately 38° C is accomplished, the fluid is delivered to the per-
itoneal cavity. After the prescribed dwell time is completed, the cycler automati-
cally shifts into a drain cycle and the fluid is collected in a weight bag to monitor
adequate drainage. The spent dialysate finally flows into a disposable drain bag.
This basic system can be modified with the use of microcomputerized programs to
closely monitor net ultrafiltration.
Two important modifications to the original cycler concept have been intro-
duced. One uses a roller pump to deliver dialysate to a heating bag which is placed
at least 20 cm over the patient's abdomen, thereby preserving the ability to deliver
dialysate by gravity while using large dialysate containers (Fig. lB). The roller
pump has the dual function of delivering a prescribed volume of dialysate to the
heating bag and acting as an occluder to prevent the retrograde transit of bacteria
from the final drain container. A third type of cycler has been proposed which
utilizes roller pumps for both the active infusion and drainage of dialysate into
and out of the peritoneal cavity (Fig. IC). These systems have not found commer-
cial application mainly due to the fear of over-distending the abdominal cavity
through positive pressure in the event of malfunction and the potential risks
associated with application of negative pressure (suction) for drainage. The
advantages of this last system are the significant reduction in size and weight that
a simple set of pumps makes possible.

3. Solutions for CCPD

The recommended dialysate formulation for CCPD is essentially the <;ame as for
continuous ambulatory peritoneal dialysis (CAPD). However, for the occasional
patient who needs shorter and frequent exchanges in order to accomplish a high
rate of ultrafiltration and small solute removal (vide infra), lower sodium concen-
trations may be recommended. Where dwell times of short duration and hyper-
tonic glucose solutions are used, proportionately greater removal of extracellular
water than sodium often occurs with consequent hypernatremia [4]. This phe-
nomenon is most notable when the dwell times are extremely short, usually less
than 30 minutes, and applies mainly to intermittent peritoneal dialysis (IPD).
 Dlatysate

Dialysate

~
..-
Pallent ~

~
~F Y
0'd Patient
..--
----..
"'"., ---.. II -- fI' )
tit Pump ~ Pressure
Monitor

Drain
Bag
Drain
Bag

A B c

Figure 1. Diagrammatic representation of cycler systems. (A) Gravity infusion and drainage; (B) Gravity infusion and drainage, modified by active transfer of N
,J::..
solution; (C) active infusion and drainage by roller pumps. '-0
 250

4. Physiologic considerations

4.1. Ultrafiltration

The typical patient undergoing CCPD using 8 L of dialysate per day can accom-
plish a net ultrafiltration of 0.5 to 3.0 L with the use of dialysate containing 1.50%
to 4.25% dextrose. In the peritoneal dialysis system, ultrafiltration is mainly a
function of transperitoneal osmotic gradient; therefore, increments in the dex-
trose concentration of the solution can most conveniently increase net ultrafiltra-
tion. However, the peritoneal ultrafiltration rate curve exhibits an exponential
decay type of configuration. It follows that maximal net ultrafiltration can be
achieved by increasing the number of exchanges and reducing the dwell time.
Patients with extraordinary ultrafiltration requirements may benefit from ad-
ditional, shorter nocturnal exchanges.

4.2. Clearance

The clearance for small molecules, such as urea and creatinine, for the average
patient on four daily exchanges of CCPD are 67 Llweek for urea and 58 Llweek
for creatinine [1]. These clearances are slightly lower than for CAPD and superior
to IPD. Full equilibration between dialysis and plasma for small molecules is
attained within four hours of dialysate dwell (Fig. 2). Consequently, the long
diurnal cycle of CCPD becomes relatively inefficient in clearing small molecules.
However, this can be easily overcome by an additional nocturnal exchange where

/-.-,
1.0 UREA

;; _.- --~.:.
_ _ _ ._o-.-(I!--.:l-~ [Link]!~
A A
.9 /. 0 • _ _- • CREATININE
~.
.8
0 _ VIT. B12
.7 -rr-
0
.6

v,r:
c...
......
o /6 ~
v-<
Cl
.5
.4 /t
~.
-;,0'
.3
.2
.1

100 200 300 400 500 600 700 800 900

DWELL TIME (min)
Figure 2. Equilibration of urea, creatinine and vitamin BI1 using 1.50% (light symbols) and 4.25%
(dark symbols) dextrose dialysis solutions (from Diaz-Buxo et al. [1]; reprinted by permission).
 251

necessary [5]. In fact, for patients requiring high small molecular solute removal,
the addition of multiple short-dwell nocturnal exchanges allows enhancement of
small molecular clearances approaching those of hemodialysis.
The equilibration of middle molecules is significantly slower. Even at the end
of the long diurnal cycle, equilibration is incomplete for larger molecules such as
inulin and vitamin B 12 • For these larger molecules, the time of exposure of
dialysate to the peritoneal membrane becomes the determining factor in solute
removal, while manipulation of dialysate flow or length of intraperitoneal dwell
are of little consequence.
Twardowski et al. have suggested the use of high-volume dialysate exchanges
to increase the rate of solute removal for small molecules [6]. The concept is based
on the observation that the rate of transperitoneal equilibration for urea and
creatinine is similar for 2- and 3-L dialysate exchanges. Despite the practical
advantages offered by this protocol, many patients cannot tolerate high-volume
dialysate exchanges during the day due to the high intraperitoneal pressure
generated by 3 L of dialysate while the patient is in a sitting or standing position.
Convenient application of the high-volume dialysate exchange concept can be
made utilizing the automated nocturnal exchanges of CCPD, since the intra-
peritoneal pressure generated by the same volume of dialysate is lower when the
patient is in the supine position. Forced vital capacity diminishes when dialysate is
infused intra peritoneally and has been reported to deteriorate further with
increasing intraperitoneal volume in the supine than in the vertical position [6].
Therefore, caution must be exerted in prescribing volumes exceeding 2 L in
patients with chronic obstructive pulmonary disease and restricted vital capacity.

4.3. Relationship between dialysate volume and intra-abdominal pressure

Several investigators have recently demonstrated a linear relationship between

intra-abdominal pressure and intraperitoneal dialysate volume [7-9]. The linear
relationship is maintained regardless of the patient position, but the slope of the
curve shifts with changes in position (Fig. 3). Increments in pressure with a given
volume are higher in the upright than in the supine position and further increase
in the sitting position [8, 9]. Consequently, a patient is able to tolerate the same or
even larger volumes of dialysate at night while in the supine position than during
the active hours of the day.
In practice, we can take advantage of this better tolerance of high volumes of
fluid during the nocturnal cycles in the CCPD patient to enhance small solute
removal.
252

22
21
20
19

/~
18
_ 17
i 16
/A
! 15
o 14
E ~
.e
~ 12
13
I
/ r = 0.98
~/ y = 2.43 + 0.21x
<II
<II 11
/ .
...J····l
I!!
Q. 10
ii 9
c
!/?
'E /
0
'tI
.D
'lI
8
7
~
/ ..~····1
I! 6 ttl ••.••
.E 5 / •.'! r = 0.96
4 "" ••••• .L Y = -1.12 + 0.16x
3
2 f .' ..,·····! -Sitting
••'!... - - Upright
••••••• Supine
0
-1
•..'1'
o 10 20 30 40 50 60 70 80
Exchange volumes (ml/kg of body weight)

Figure 3. Comparative effect of dialysate volume and patient position on intra-abdominal pressure.
Means ± SEM [9].

4.4. Steady physiologic state

One of the desirable features of continuous peritoneal dialysis is the maintenance

of a steady physiologic state as reflected by minimal fluctuations in body chemis-
tries. Since most of the ultrafiltration and solute removal occurs at night in the
case of CCPD and the bulk of solute and fluid intake occurs during the day, we
could expect some fluctuation in the hydration status and body chemistries during
a 24-hour period. In practice, this should not be a significant limitation of the
therapy since most patients on three-times-weekly hemodialysis tolerate treat-
ment for many years without obvious dysequilibrium symptoms. Furthermore,
dysequilibrium is rare in patients undergoing intermittent peritoneal dialysis on
alternate nights. We have addressed the problem by studying selected biochemi-
cal parameters in a series of patients undergoing CCPD (Fig. 4). Minimal
variations in the blood composition were noted, the most apparent of which were
an increase in serum osmolality and sodium at night, and higher BUN levels in the
afternoon. These changes should not be of clinical significance.
 253

f]
:c
I I I I I 1 1 r

145

---

.
T
I

.
I
• T
T T
~ i ~ 1
• 1
135

80,-----------------------~--------

1 70
~ 804---~--~------------------------­
II)

504----------------------------------

J 1
•
280

I I i i I i I I

~ 8CD 0
0
0
0
0
0
on i
0
~ §
0 0 '"0 ~ ~ ~ N N

Time (hrs)
Figure 4. Variation in hematocrit, sodium, urea nitrogen and osmolality during a 24-hour period in
five patients undergoing CCPD.

5. Clinical experience with CCPD

5.1. Hematologic and biochemical parameters

The experience with the hematologic and biochemical profiles of patients under-
254

going CCPD has been previously reported [1, 10]. However, reference should be
made to a few selected parameters which have been studied in greater detail or
which have shown characteristic behavior in our patient population.
The hemoglobin and hematocrit concentrations have uniformly improved
among patients undergoing CCPD. The extent of the increments in hemoglobin
concentration are similar to those reported by other groups with CAPD [11-13].
The factors responsible for the improvement have not been fully delineated. The
most likely explanations for this phenomenon are a decrease in plasma volume,
an increase in red cell mass due to improved erythropoiesis or prolongation of red
cell life, or a combination of these factors. Figure 5 analyzes the hematocrit and
hemoglobin concentrations and correlates these changes with the reticulocyte
count, weight, and mean blood pressure in ten patients during the first twelve
months of CCPD. A characteristic pattern of rapid improvement in hematocrit
and hemoglobin concentrations can be appreciated for the first three months of
therapy followed by a very slow increase until the ninth month at which point the
values remain stable. The reticulocyte count significantly increases during the
third month and remains higher than during the predialysis period. There is a
significant drop in weight at initiation of therapy, probably reflecting contraction
of plasma volume and removal of edema fluid, and progressive weight gain
thereafter, possibly due to an accumulation of adipose tissue and/or an increase in
muscle mass. Improvement in blood pressure control is slowly attained during the
first four months and remains stable for the total observation period.
This pattern is consistent with the observations of De Paepe et al. who observed
an increase in hematocrit during the first six months of CAPD treatment con-
comitant with a decrease in plasma volume followed by an increase in red cell
mass during the subsequent months of observation [13]. Lamperi et al. have
shown an improvement in hematocrit, hemoglobin, and reticulocyte values, and a
strong correlation with the recovery of the erythroid cell proliferative activity in
patients undergoing CAPD [14]. They did not see a change in the level of serum
erythropoietin, suggesting that the improvement in bone marrow function ap-
pears to be due to better clearance of substances which inhibit the response of
bone marrow to erythropoietin. During the first six months of peritoneal dialysis,
it is likely that a concomitant decrease in plasma volume and a gradual increase in
red cell mass takes place, which accounts for the dramatic improvement in
hemoglobin and hematocrit values. An additional contributory factor to the
improvement in hemoglobin concentration may be the removal of toxic sub-
stances responsible for red blood cell lysis. The reduction in total body weight and
blood pressure further substantiates intravascular fluid removal. Between the
third and ninth months blood pressure remains stable and the weight increases
while the hematocrit concentration slightly improves and the reticulocyte count
remains high, suggesting an increase in red blood cell mass, consistent with
improved erythroid cell proliferative activity. In our experience the hemoglobin
 255
40
T T T

l .,.
•
J 1 1 T I J 1 I I
~ii 30
T
I
I
• ~
I
-
I 1 I
•
J. • ~ •
•E I
I
-
I

•
:t: T
I
20 •

~~~10;-----------~I--~--~----------------------------­
~~:=====;::==::::I=:I=:r~r~::=r~I=1~I-=:11-=:: ;1
f~ 9 +-----------~---------------­
• 8+-~T---l~-----------------------­
:t: 7~--~~L--------------------------------------

3.0
,. I
• •
I
l
-
L T
2.0 - I I I 1 I
•
T T
•
!
o
'S
1.0
I
•
J.

I
•
u T
i T J.
II:
•
0

90
a

IIIIIIII fI
~

I r
l:

I
80
f
III
"i
~
70

120
a
:t:
E 110
g
A-
ID
c 100
•
II

:E 90

I i i I i i I I I I I i
Pre 2 3 4 5 6 7 8 9 10 11 12
Time (Months)

Figure 5. Changes in hematocrit, reticulocyte count, weight and blood pressure in ten patients during
the first year of CCPD.

concentration has remained relatively stable after nine months of ccpn, but can
be drastically affected by severe and prolonged episodes of peritonitis and
concomitant malnutrition.
256

5.2. Calcium, phosphorus, and renal osteodystrophy

Most patients were dialyzed using a dialysate calcium concentration of 4 mEq/L.

Vitamin D supplementation (1,25 dihydrocholecalciferol) was used in 30% of the
patients with initial hypocalcemia. During the first six months of therapy, 65% of
the patients required calcium carbonate supplements in order to maintain their
serum calcium levels in the range of 8.5 to 9.5 mg/dl. After one year, however,
only 15% of the patients required oral calcium supplementation, while most were
able to maintain a normal serum calcium level. Eighty-five percent of patients
have required phosphate binders and only 15% were capable of maintaining a
normal phosphate level without binders. The serum alkaline phosphatase level
remained slightly elevated without apparent trends.
Twenty-two percent of patients showed radiologic evidence of subperiosteal
resorption at the beginning of dialysis, but did not show clinical deterioration in
their renal osteodystrophy. One patient suffered from severe secondary hyper-
parathyroidism prior to initiation of therapy, as evidenced by extensive sub-
periosteal resorption and metastatic calcifications on metabolic bone surveys,
elevated levels of intact parathyroid hormone, severe osteitis fibrosa on an
un decalcified bone biopsy, and negative aluminum stains. Despite good com-
pliance with therapy, excellent control of hyperphosphatemia, and improvement
in her nutritional status, the patient became hypercalcemic and developed more
metastatic calcifications. She underwent a partial parathyroidectomy after six
months on CCPD with significant improvement in her calcium and phosphorus
metabolism. A second patient entered the program with evidence of marked
vascular calcification on X-rays and generalized osteopenia. She also had minimal
subperiosteal resorption of bone. A parathyroid hormone (PTH) profile revealed
a modest elevation in intact and C-terminal PTH. Despite aggressive treatment
with vitamin D, supplemental calcium, and excellent control of serum phos-
phorus, her alkaline phosphatase has remained elevated, the radiographic ap-
pearance has shown signficant deterioration, and the bone biopsy showed severe
osteomalacia with negative aluminum stains.
It is premature to reach conclusions regarding the development of renal
osteodystrophy in patients undergoing CCPD due to the limited number of
patients studied and a maximum follow-up of only three years. In our experience
only two cases (2.5%) have developed progressive bone disease and both patients
had evidence of severe renal osteodystrophy at the beginning of therapy. Osteitis
fibrosa and osteomalacia may be progressive in patients undergoing CCPD
despite negative aluminum staining in bone and normal serum phosphorus and
calcium concentrations. Persistent hyperparathyroidism as well as severe os-
teomalacia with positive aluminum staining has been previously reported in
patients on long-term CAPD [15]. A dialysate calcium concentration of 4mEq/L
is probably adequate for most patients on CCPD. Although, it is true that the
majority of patients require vitamin D and calcium supplementation during the
 257

first six months of therapy, most patients are capable of maintaining adequate
serum calcium concentrations after six months without additional oral calcium
supplementation. Furthermore, we have only observed one case of hyper-
calcemia, which occurred in a patient with progressive hyperparathyroidism.

5.3. Nutritional status

Although all patients have been instructed to ingest 1.2 g of protein/kg of body
weight/day, nutritional histories and protein counts on a selective menu diet
revealed that the average patient only consumes 0.75 g protein/kg of body weight/
day. No specific caloric restrictions have been recommended unless the patient
develops obesity or is diabetic. On this program, the 40 patients observed for a
period of at least one year have shown a mean calculated dry weight increase in
body mass of 3.95 ± 1.21 kg. The serum total protein and albumin concentrations
have remained normal, or in the lower range of normal, despite a daily mean
dialysate protein loss of 10 g [1]. No correlation has been noted between total daily
peritoneal protein losses, number of dialysate exchanges, or total daily dialysate
volume. However, a slightly higher peritoneal protein loss of 12 g/day has been
observed in patients using predominantly 4.25% dextrose solution.
The serum cholesterol has not shown a significant rise in most patients. Six
patients developed hypertriglyceridemia. Of the four diabetic patients in this
group, three showed significant improwment in serum triglycerides with addition
of intraperitoneal insulin in doses of 5 to 15 units/L of dialysate (Fig. 6). The
correction of hyperlipidemia in the diabetic patients is consistent with the experi-
ence of Moncrief et at. [16]. The two nondiabetic patients failed to respond to
intraperitoneal insulin. Beardsworth et al. also concluded that there was no
significant change in any lipid parameters in nondiabetic hyperlipidemic patients
on CAPD despite a significant fall in fasting and postprandial glucose levels [17].

5.4. Blood pressure control

Excellent blood pressure control has been attained with manipulation of the
hydration state in most patients. While 80% of patients suffered from hyperten-
sion requiring multiple drug therapy prior to treatment with CCPD, only 10% of
the patients required antihypertensive agents after six months of therapy. Pos-
tural hypotension has been very infrequent among nondiabetic patients. How-
ever, five diabetic patients suffered from severe postural hypotension, despite
adequate or excessive hydration and normal or elevated blood pressure in the
supine position. Postural hypotension under these circumstances is characteristic
of autonomic neuropathy secondary to diabetes.
258

01 &
190~
150 Z A
s;:;e:: '-i
c,.-
~ii110 Sse;? ~ /-
m 70~-------~------------------------------------------~~---
....0.
900

..
0--0' ~
1/1
700 / ~--o..
"0
i::g / ~
:l", / "'\.

'"
500 a
~E
~
300
h
"
~

~ ...{
""-
100

.... 01
'2 -
.,-
-"0
:::---=1I~_.......--:
___.a_"",-_...-=-r~ _ _""""
___ ______ ·_~ --S;-----="""""".~___ :::o
"DE
.s:
o 100~-~--------------------------------------------------

84.01 ?~
82.0~-I----!Q.--.....l-=-_-a~Z~-------------------------------------­
e ~
~------~------------------------------
80.0 -1__________________________________________________

loP. Insulin

i i i i i I I I i i i i
Pre 2 3 4 5 6 7 8 9 10 11 12
Months on CCPD

Figure 6. Treatment of hypertriglyceridemia with intraperitoneal insulin in a Type II diabetic patient

undergoing CCPD. Open bar represents the addition of regular insulin, 5 units/2 L of 1.50% D and 10
units/2L of4.25% D dialysate; solid bar denotes addition of 10 units/2 L of1.50% D and 15 units/2 L of
4.25/D dialysate.

6. Peritonitis

Peritonitis remains the most frequent complication of chronic peritoneal dialysis

and the single most common cause of hospitalization for this population of
patients. Technical advances in connecting devices, improvement in the teaching
of aseptic techniques to patients, and incorporation of in-line filters in the
peritoneal dialysis system have significantly reduced the incidence of peritonitis
among patients undergoing CAPD. However, most centers still experience one
episode of peritonitis per patient per year. The incidence of peritonitis in our
CCPD experience has been one episode per 30 to 35 patient months for the total,
unselected population since 1980. Peritonitis is diagnosed whenever 100 cells or
more are detected in the peritoneal outflow, or by a positive dialysate culture
obtained through a 0.22 /.t filter. Figure 7 illustrates the probability of remaining
 259
(118)
100

! 90
~
C 80
~
ia. 70
'0
CD
~ 60
g>
c
C 50
iii
E
! 40 (6) (2)
'0
~ 30
:ii
III
.a
!:! 20
Do

10

6 12 18 24 30 36
Months on CCPD

Figure 7. Probability of remaining free of peritonitis for patients undergoing CCPD.

free of peritonitis as a function of time for all our patients. The likelihood that a
patient undergoing CCPD will remain free of peritonitis is 73% at one year and
51 % at two years. No significant difference in the rate of peritonitis has been
observed between the diabetic and nondiabetic populations. This low rate of
peritonitis has been corroborated and surpassed by others [18]. Of more signifi-
cance is the fact that at least two programs have reported twice as many episodes
of peritonitis with CAPD than with CCPD in simultaneously studied populations
[18,19]. Unfortunately, none of these studies have been controlled.
It is pertinent to analyze the potential factors responsible for the low incidence
of peritonitis in our experience (Table 1). The number of connections required
between the peritoneal catheter and the system is reduced to two per day.
Although there are multiple connections between the system's lines and the
bottles, or bags, the likelihood of contaminating two disposable sterile parts
seems minimal compared with a connection with the permanent catheter. All
connections in CCPD take place at a convenient time and location, which
probably improves the patient's concentration and minimizes the patient's fa-
tigue. Better aseptic control of the environment can also be accomplished in the
patient's home than in unfamiliar surroundings. Finally, we have proposed the
direction of dialysate flow following a connection in CCPD as a potential factor
influencing the incidence of peritonitis. In the cases of IPD and CAPD, dialysate
is immediately infused into the peritoneal cavity following a connection. In
contrast, with CCPD, the peritoneal fluid is drained into the collecting bag
260

Table 1. Potential factors responsible for low incidence of peritonitis in CCPD

1. Reduced number of connections between the peritoneal catheter and the cycler system
2. Connections and disconnections take place at convenient times and locations
3. Dialysate outflow following a connection

following a connection. If bacterial contamination were to occur during the

connecting procedure, it is likely that the transit of bacteria into the peritoneum
will be facilitated by inflowing dialysate in the case of IPD or CAPD, but bacteria
will be washed out by drainage of spent dialysate in the case of CCPD.

Hernias

Abdominal hernias have become a relatively common complication of continu-

ous peritoneal dialysis. Several centers have reported the de novo development of
hernias in 9% to 12% of patients undergoing CAPD [20-23]. Etiologic factors
that may contribute to abdominal wall herniation include the constant increase in
intra-abdominal pressure generated by the peritoneal dialysate and weaknesses
of the anterior abdominal wall, either constitutional or from multiple pregnan-
cies, previous surgery, and/or insertion of peritoneal catheters. Careful examin-
ation of the abdominal wall and inguinal areas is mandatory prior to insertion of a
peritoneal catheter in order to exclude the existence of hernias. Elective repair
prior to or at the time of insertion of the catheter is recommended in order to
prevent incarceration, inadequate drainage of dialysate, or expansion of the
hernial sac.
During our initial 30 months of experience with CCPD, 9% of patients de-
veloped hernias. All patients were multiparous, elderly females with weak ante-
rior abdominal walls. This is in marked contrast with our extensive experience
with IPD with only a 2% incidence of new hernias in patients undergoing
treatment. Following surgical repair and insertion of a Tenckhoff or column disc
catheter, all patients returned to CCPD, using reduced dialysate volumes for the
diurnal exchange (1,000 to 1,500 ml) without recurrence of hernias.
Six patients have presented with a hernia-like distention around or in proximity
to the catheter exit site. They were easily reducible and collapsed when the
abdominal cavity was emptied. Exploration of the hernial sac revealed a dilated
structure of fibrous tissue and muscle not lined by peritoneum; therefore, they
were not true hernias and we have elected to use the descriptive term
"pseudohernia". They are a simple accumulation of fluid that has dissected the
peritoneal space at the point where the catheter pierces the peritoneum. The
transit of peritoneal fluid around the catheter and into the hernial sac can be
demonstrated with the use of intraperitoneal contrast material injected while the
patient is in the hands-knees position [24). We have only seen this complication
 261

with the use of single-cuff peritoneal catheters. All patients have been treated by
simple removal of the catheter and replacement with a double-cuff device. After
a resting period of 10 to 14 days without peritoneal dialysis to allow healing,
CCPD has been reinstituted using volumes up to 2 L without recurrence.

8. Other complications

Tunnel and catheter exit -site infections have been more common with continuous
peritoneal dialysis. Many of these infections did not result in frank peritonitis, but
were responsible for chronic skin irritation and the eventual loss of the catheter.
The mean catheter life in our population has diminished significantly from 22
months for IPD to the present 14 months for CCPD.
Catheter outflow obstruction has been less frequent with CCPD than with IPD
and is usually corrected by conservative means. One-way obstruction is most
often due to omental wrapping around the intraperitoneal portion of the cathe-
ter, or to catheter migration. Conservative treatment consisting of simple exer-
cise and stimulation of the bowel by enemas corrects the problem in the majority
of patients.
Although recent reports suggest that some patients on CAPD lost effective
ultrafiltration even in the absence of peritonitis [25], we have not been able to
confirm a real loss of ultrafiltration in any of the patients undergoing CCPD. We
must admit that an apparent loss of ultrafiltration has been noted in several of our
patients; however, on further evaluation, the following factors have been charac-
terized as responsible for the inability to effectively remove fluid: 1) an increase in
sodium intake, 2) an increase in the sodium concentration of the dialysate, 3) a
significant decrease in residual renal function resulting in increased weight gain
and edema without an actual drop in peritoneal ultrafiltration, 4) mechanical
problems with the catheter reducing the surface area available for transperitoneal
exchange, 5) uncontrolled hyperglycemia in diabetics which may decrease the
transperitoneal osmotic gradient, and 6) severe hypoalbuminemia that may result
in edema and difficulty with mobilization of fluid from the interstitial tissue, while
maintaining a normal or supernormal net peritoneal ultrafiltration. It is impera-
tive to reevaluate the patient who experiences difficulty with fluid removal in
order to determine a real versus apparent ultrafiltration loss.

9. Patient and technique survival

Figure 8 illustrates the patient and technique survival for our first three years of
experience with CCPD in the nondiabetic population. The patient survival at
three years is 83%. The technique survival rates at one, two, and three years were
80%,62%, and 56%, respectively. The drop-out rate averaged 9% per year. The
262
(66)
100

90
(43) (25) (14) (10) (6)
- - -+------+---- ..... - - ...
80 Patient

70
,!!
~ 60
"ii
>
,.
.~
50 Technique
t/J

40

30

20

10

6 12 18 24 30 36
Months on CCPD

Figure 8. Patient and technique survival for non-diabetic patients undergoing CCPD.

patient and technique survival rates are similar to those of our home hemodialysis
population.

10. Treatment of diabetics with CCPD

Continuous peritoneal dialysis has provided a new and physiologic route for
insulin administration, sparing the patient multiple dialy injections. It is quite
feasible to obtain tight blood sugar control in the patient undergoing CAPD, due
to the multiple exchanges during the day which allow divided and variable doses
of insulin at the required times. It has been more difficult to design a uniform
method for intraperitoneal insulin administration in the patient undergoing
CCPD, due to the fact that most of the caloric load takes place during the day
while only one peritoneal dialysis exchange is administered. Nevertheless, excel-
lent glycemic control can be obtained in the majority of patients if the time is
spent to calculate the precise dose of insulin required, and if a regular and
predictable caloric intake is maintained with little day to day variation.
All insulin-dependent diabetic patients should be instructed in intraperitoneal
insulin administration and regular blood sugar determinations using the fin-
gerprick technique and a glucometer. Since the dialysate containers often do not
drain simultaneously during the nocturnal exchanges, it is recommended that the
insulin dose be divided among all containers in order to avoid a sudden and
massive infusion of insulin and consequent hypoglycemia. The average intra-
 263

peritoneal insulin dose required for good control of glycemia has been three times
the previous total subcutaneous dose. In most cases, 50% of the intraperitoneal
dose has been used for the long-dwell diurnal cycle with the remaining 50%
equally divided among the nocturnal exchanges. While the patient is still in the
hospital, blood sugars are determined four times daily, or more often if required.
All subcutaneous insulin injections are discontinued and an initial dose of regular
insulin, equivalent to two times the previous total 24-hour subcutaneous insulin
dose, is prescribed for intraperitoneal use. Fifty percent of the regular insulin is
added to the diurnal dialysate container (4.25% dextrose) and the other 50% is
equally divided among the three bottles used for the nocturnal exchanges. This
dose often requires adjustment with the final daily dose closer to three times the
previous total insulin dose required for good glycemic control. If additional
intraperitoneal insulin is needed, the guidelines suggested by the Toronto West-
ern Hospital protocol are followed [26].
The one-year patient survival for diabetic patients in our population is 74%.
This survival rate is definitely lower than for the non-diabetic population. How-
ever, it is significantly better than that reported for diabetic patients undergoing
IPD [27, 28]. Amair et al. have reported survival rates for diabetic patients
undergoing CAPD which are comparable or superior to the survival rates ob-
tained by other programs for non-diabetic patients undergoing CAPD or hemo-
dialysis [29]. A factor that may influence the lower survival rate for our diabetic
population is a high rate of transplantation, both living related and cadaveric, for
most young diabetic patients entering our program; thereby eliminating this
relatively healthy population from our statistics. We have also depended heavily
on CCPD for older, blind, and dependent diabetics who require a partner for
their treatment at home.
Because of the need for mUltiple divided doses of insulin during the day for
some diabetic patients, CAPD may offer an advantage in administering insulin
intraperitoneally in a more physiologic manner.

11. Indications for CCPD

There are no absolute contra-indications to CCPD and the advantages inherent

to continuous peritoneal dialysis generally apply to this therapy. However, there
are special circumstances where CCPD may prove to be the preferred mode of
therapy. Table 2 lists the potential indications according to the patient's prefer-
ence, partner's preference, and medical circumstances.
The convenience provided by CCPD in eliminating all diurnal interruptions in
the daily routine necessary for dialysate exchanges is probably responsible for the
preference expressed by most of the patients in our program. The majority of
these individuals are employed, and are unwilling or unable to perform ex-
changes during the day. Furthermore, some patients have chosen CCPD ex-
264

Table 2. Indications for CCPD

Patient preference
Employed, active patients
Unwilling or unable to perform exchanges
Psychological (self-image)

Partner preference
Employed partner
Partner fatigue

Medical circumstances
Inadequate dialysis
Inadequate small molecular clearances
Inadequate ultrafiltration
Non-compliance with number of exchanges
Recurrent peritonitis
Poor eye-hand coordination
Chronic low back pain
Recurrent catheter exit site leakage

elusively because of their ability to perform continuous peritoneal dialysis in an

inconspicuous way in the privacy of their homes.
When a partner is necessary, the freedom and convenience provided by CCPD
becomes particularly attractive. Partner's preference has been the major indica-
tion for CCPD in the case of working partners, partners dialyzing children,
relatives dialyzing elderly and debilitated patients, and those with poor eye-hand
coordination. Since the partners are only required twice daily for a short period of
time, at convenient hours, partner fatigue is reduced with CCPD.
Inadequate peritoneal dialysis can result from suboptimal solute removal and/
or insufficient ultrafiltration. This condition may be the product of noncompli-
ance with the number of exchanges, a defective peritoneal membrane, hyper-
catabolic states, or a combination of these circumstances. The prescription of
additional short nocturnal cyeles can significantly improve water and fluid re-
moval without the inconvenience imposed and the time required to perform
manual exchanges during the day. Furthermore, additional breaks in the sterile
system required for dialysate exchanges potentially increase the risk of per-
itonitis. CCPD allows flexibility in prescription of additional exchanges during
the nocturnal hours, in most patients, without placing an additional burden on
their time or increasing the risk of peritonitis.
CCPD is also better tolerated by patients with chronic low back pain, which is
often aggravated by large intraperitoneal dialysate volumes. The physiologic
lumbosacral angle is approximately 30 degrees. Increase in intra-abdominal
pressure accentuates the lordotic curve and increases the lumbosacral angle by 15
to 20 degrees [30]. This is usually associated with significant aggravation of back
 265

pain. Although exercises can markedly improve the symptomatology in these

patients, we have had gratifying results by reducing the diurnal intraperitoneal
volume of dialysate and compensating with equivalent increases in dialysate flow
during the noctutnal phase of CCPD.
As previously mentioned, patients with abdominal hernias and those with
recurrent catheter exit-site leaks may benefit from the reduced intra-abdominal
pressure provided by reduction in intra-abdominal volume during the day. These
patients can tolerate similar or even larger volumes of dialysate at night while in
the supine position without significant compromises in their total clearances.

U. Conclusions

CCPD provides an alternative for continuous peritoneal dialysis for the patient
who needs or desires automated treatment during the night without interruptions
in the daily routine for dialysate exchanges. The experience so far suggests that it
is particularly useful for children, individuals requiring a partner, and those
requiring more than four dialysate exchanges per day. Although the rate of
peritonitis has been impressively low in some programs, there are no objective
data from controlled studies to prove its superiority in reducing the rate of
peritonitis over CAPD. In the final choice of peritoneal therapy, the patient's
life-style, psychologic needs, and preferences should be considered above all
other factors in order to attain the highest level of rehabilitation.

References

1. Diaz-Buxo JA, Walker PJ, Farmer CD et al.: Continuous cyclic peritoneal dialysis- a preliminary
report. Artif Organs 5: 157-161, 1981.
2. Popovich RP, Moncrief JW, Decherd JF et al.: The definition of a novel portable/wearable
equilibrium peritoneal dialysis technique. (Abstract) Am Soc Artif Intern Organs 5: 64, 1976.
3. Lasker N, McCawley EP, Passarotti CT: Chronic peritoneal dialysis. Trans Am Soc Artif Intern
Organs 12: 94,1966.
4. Shen FH, Sherrard DJ, Scollard D et al.: Thirst, hyponatremia and excessive weight gain in
maintenance peritoneal dialysis. Trans Am Soc Artif Intern Organs 24: 142-145, 1978.
5. Blumenkrantz MJ: The need for individualization and'flexibility in treatment modalities. Con-
temp Dial 4: 12, 1983.
6. Twardowski ZJ, Nolph KD, Prowant B et al.: Efficiency of high volume, low frequency continu-
ous ambulatory peritoneal dialysis. (Abstract) Am Soc Artif Intern Organs 12: 69, 1983.
7. Gotloib L, Mines M, Garmizo L et al.: Hemodynamic effects of increasing intra-abdominal
pressure in peritoneal dialysis. Perit Dial Buill: 41-43, 1981.
8. Twardowski ZJ, Prowant BF, Nolph KD et al.: High volume, low frequency continuous ambula-
tory peritoneal dialysis. Kidney Int 23: 64--70,1983.
9. Diaz-Buxo JA: CCPD is even better than CAPD. Proc Fourth Int Conf Uremia, Capri, 1982 (in
press)
10. Diaz-Buxo .JA,Walker PJ, Chandler JT et al: Continuous cyclic peritoneal dialysis. In: GM Gahl,
266

M Kessel, KD Nolph (eds), Advances in Peritoneal Dialysis. Excerpta Medica, Amsterdam pp

126--130, 1981.
11. Moncrief JW, Popovich RP, Nolph KD et al.: Clinical experience with continuous ambulatory
peritoneal dialysis. ASAIO J 2: 114-118, 1979.
12. Lamperi S, Icardi A, Carozzi S et al.: Effect of CAPD on renal anemia. Int J Nephrol Urol Androl
1: 43-52, 1981.
13. De Paepe MBJ, Schelstraete KHG, Ringoir SMG et al.: Influence of continuous ambulatory
peritoneal dialysis on the anemia of end-stage renal disease. Kidney Int 23: 744-748, 1983.
14. Lamperi S, Carozzi S, Icardi A: In vitro and in vivo studies of erythropoiesis during continuous
ambulatory peritoneal dialysis. Perit Dial Bull 3: 94-96, 1983.
15. Digenis G, Khanna R, Pierratos A et al. : Renal osteodystrophy in patients maintained on CAPD
for more than three years. Perit Dial Bull 3: 81-86,1983.
16. Moncrief JW, Pyle WK, Simon P et al.: Hypertriglyceridemia, diabetes mellitus and insulin
administration in patients undergoing CAPD. In: JW Moncrief, RP Popovich (eds), CAPD
update. Proc 2nd Int Symp. Mason, New York, pp 143-165,1981.
17. Beardsworth SF, Goldsmith HJ, Stanbridge BR: Intraperitoneal insulin cannot correct the
hyperlipidemia of CAPD. Perit Dial Bull 3: 126--127, 1983.
18. Cavoretto CA, Jackson FE: A decrease in peritonitis with CCPD: One unit's experience.
Nephrol Nurse 5: 33,17,1983.
19. Walls J, Smith BA, Feehally Jet al.: CCPD - An improvement on CAPD. In: GM Gahl, M
Kessel, KD Nolph (eds), Advances in Peritonal Dialysis. Excerpta Medica, Amsterdam, pp 141-
143, 1981.
20. Chan MK, Baillod RA, Tanner A et al.: Abdominal hernias in patients receiving continuous
ambulatory peritoneal dialysis. Br Med J 283: 826, 1981.
21. Digenis GE, Khanna R, Oreopoulos DG: Abdominal hernias in patients undergoing continuous
ambulatory peritoneal dialysis. Perit Dial Bull 2: 115-117,1982.
22. Jorkasky D, Goldfarb S: Abdominal wall hernia complicating chronic ambulatory peritoneal
dialysis. Am J Nephrol 2: 323-324, 1982.
23. Rubin J, Raju S, Teal N et al.: Abdominal hernia in patients undergoing continuous ambulatory
peritoneal dialysis. Arch Intern Med 142: 1453-1455,1982.
24. Tucker CT, Cunningham JT, Nichols AM et al.: Cannulography with peritoneal air contrast
study. Contemp Dial 3: 9-13, 1982.
25. Slingeneyer A, Canaud B, Mion C: Permanent loss of ultrafiltration capacity of the peritoneum in
long-term peritoneal dialysis: An epidemiological study. Nephron 33: 133-138,1983.
26. Khanna R, Liebel B: The Toronto-Western protocol. Perit Dial Bull 6: 101-102, 1981.
27. Katirtzaglou A, Elzatt S, Oreopoulos D et al.: Chronic peritoneal dialysis in diabetics with end-
stage renal failure. In: EA Friedman, FA L'Esperance (eds), The Diabetic Renal-Retinal
Syndrome. Grune and Stratton, New York, pp 317-331, 1980.
28. Mitchell JC, Frohnert PP, Kurtz SB et al.: Chronic peritoneal dialysis in juvenile-onset diabetes
mellitus: a comparison with hemodialysis. Mayo Clin Proc 53: 775-781, 1978.
29. Amair P, Khanna R, Liebel Bet al.: Continuous ambulatory peritoneal dialysis in diabetics with
end-stage renal disease. N Engl J Med 306: 625-630,1982.
30. Goodman CD, Husserl FE: Etiology, prevention and treatment of back pain in patients undergo-
ing continuous ambulatory peritoneal dialysis. Perit Dial Buill: 119-122, 1981.
10. Pharmacologic manipulation of
peritoneal transport

JOHN F. MAHER and PRZEMYSLAW HIRSZEL*

1. Introduction

Recently, peritoneal dialysis has become an increasingly popular alternative to

maintenance hemodialysis for therapy of patients with chronic renal failure [1-3].
Concurrently, the effect of pharmacological and physiological manipulations on
peritoneal transport parameters have been explored seeking enhanced under-
standing of transport barriers and clinically useful methods to augment transport.

2. Rationale for augmenting transport rates

The mass transport rate of small solutes by peritoneal dialysis is rather slow when
compared to the rates during hemodialysis. Accordingly, peritoneal dialysis is
more time consuming when the therapeutic endpoint is to achieve a given degree
of control of the plasma concentration of low molecular weight solutes like urea.
The more often the fluid is exchanged the more likely is the occurrence of
peritonitis, the major complication of chronic peritoneal dialysis. Thus, inef-
ficient transport can contribute to the danger of peritonitis because more ex-
changes of dialysis solution are required. Once peritonitis occurs solute transport
may increase, while the rate of ultrafiltration simultaneously decreases because of
more rapid dissipation of the osmotic gradient. Thereafter, transport should be
returned to baseline rates unless treatment has been inadequate leading to loss of
peritoneal surface area or decreased permeability. Such patients may have margi-
nal transport rates, however, and in particular the ultrafiltration capacity may be
decreased to an unacceptable level [4]. Moreover, for hypercatabolic or hyper-
kalemic patients the transport inefficiency for small solutes may be quite signifi-
cant clinically, even when the peritoneal surface area and permeability have not
been reduced.
The efficiency of peritoneal mass transport may be particularly impaired in the
presence of systemic vascular disease [5]. Although the splanchnic blood vessels

• The opinions or assertions contained herein are the private views of the authors and should not be
construed as official or as necessarily reflecting the views of the Uniformed Services University of the
Health Sciences or Department of Defense. There is no objection to its presentation and/or publi-
cation.
268

may escape a generalized sclerosing or inflammatory process, usually diseases

such as diabetes mellitus, malignant hypertension, scleroderma and systemic
lupus become so widespread as to affect all the vasculature before causing
terminal renal failure.
Continuous ambulatory peritoneal dialysis does not really have the disadvan-
tage of a long duration of treatment because treatment time does not inhibit
rehabilitation [6]. Nevertheless, this technique requires an adequate level of
efficiency to be clinically satisfactory. With coexistent systemic vascular disease
or after multiple episodes of peritonitis, the efficiency of peritoneal mass trans-
port may be so borderline as to render the procedure inadequate unless more
frequent exchanges are used, with the attendant hazards of multiple tubing
disconnections. Moreover, some patients undergoing this treatment have low
rates of ultrafiltration across the peritoneum or acquire this abnormality. Under
other circumstances increased catabolism may increase the nitrogen load. De-
spite continuous peritoneal dialysis, augmented transport may be required when-
ever there is decreased transport efficiency or an increased catabolic rate.
When peritoneal dialysis is used for the removal of exogenous toxins it is
usually mandatory that the removal rates be maximal. On the other hand, when
protein loss is excessive it can be judicious to decrease the transport rates, at least
of larger solutes. It becomes obvious, therefore, that further understanding of the
mechanisms of mass transport and the influence of various pharmacologic and
physiologic manipulations on these mechanisms is important fundamental infor-
mation to provide the capability of accelerating or decreasing transport rates as
clinically indicated. Because recent evidence suggests that the major sites of
ultrafiltration and of diffusion across the peritoneum may differ [7] and it is
possible to affect these transport sites selectively [8, 9], it may be appropriate to
modulate these transport rates individually in certain circumstances. Frequently,
patients undergoing peritoneal dialysis also require a variety of drugs that have
specific vasoactive or membrane effects. Knowledge of the effects of such drugs
on transport parameters can influence the appropriate choice of a drug for a
particular indication.

3. Mechanisms of transport

When dialysis solution is placed in the peritoneal cavity it approaches concentra-

tion equilibrium with plasma by diffusion. Additionally, net osmotic and hydro-
static forces promote the movement of water, usually from plasma to dialysate.
Such ultrafiltration also augments the removal of solutes by convective transport.
Solutes also can be added to dialysate from adjacent tissue rather than from
plasma [10]. Finally, solutes that are absorbed from peritoneal dialysis solution
may undergo hepatic metabolism before reaching the systemic circulation, de-
creasing the absorbed concentration [11].
 269

4. Diffusion

Diffusion occurs by random kinetic movement of molecules, a process that tends

to spread any substance evenly throughout the space available to it. This process
is not affected by drugs directly, but the barriers to diffusion can be influenced
pharmacologically. Diffusion rates correlate directly with temperature, how-
ever.
The rate of linear diffusion of a solute in any direction throughout a cross-
sectional area, expressed as mass transport or quantity per unit time, is propor-
tional to the concentration gradient. Interposing a membrane with pores that are
large in relation to the diffusing molecules merely restricts the total area available
for free diffusion. Dividing the rate of mass transport by the electrochemical
gradient, or more simply by the plasma concentration, yields a clearance value
analogous to renal clearance. When intraperitoneal dwell is prolonged, the
concentration gradient dissipates decreasing the rate of mass transport. Hence,
unless clearances are calculated on short-term exchanges, e.g., hourly, they are
misleadingly low and the dialysance [12] or mass transfer coefficient must be
determined [13].
Free diffusion across capillary walls becomes progressively restricted as the
square root of the molecular mass of the solute increases in accord with the
Einstein equation for the diffusion constant [14]. Accordingly, peritoneal per-
meability area coefficients decrease as the square root of the molecular mass
increases, while clearances bear a slightly different relationship because concen-
tration equilibrium is approached as dwell time is prolonged [15]. It is difficult,
however, to know all the relations that exist between the many diffusion coeffi-
cients required to characterize multicomponent mass transfer across macroscopic
biologic membranes.
Water soluble solutes traverse intercellular channels, whereas lipid soluble
solutes dissolve in plasma membranes, thereby readily permeating cells. The rate
of diffusion of small water soluble solutes is so rapid that the observed peritoneal
transport rates can be accounted for by intercellular pores that total only 0.2% of
the surface area [16]. The total cross-sectional peritoneal surface area is not
precisely known but exceeds one square meter, i.e., approaches body surface
area. At the usual exchange rate of dialysis solution of two liters hourly and with a
peritoneal blood flow rate of 60-100ml/min [17] the clearances of small solutes
such as urea and creatinine are much lower than those achieved by hemodialysis,
whereas large solutes like inulin are removed relatively faster. These observa-
tions have been interpreted to indicate that the total pore area of the peritoneum
is less than that of synthetic dialysis membranes but that the pores are larger [18].
Preliminary studies of the transport of neutral dextrans are consistent with
heteroporosity of the peritoneum with some pores in the range of 35 Aor higher
[19]. Large solutes like polypeptides and small proteins appear to traverse the
capillary wall in vesicles adjoining or contiguous with intercellular clefts [20-22].
270

The effective size of the pores in capillaries can be influenced by the protein
concentration of the perfusate, by the capillary blood pressure and by drugs.
The thickest layer of the resistance barrier to transport is the dense interstitial
connective tissue between the capillary endothelium and the mesothelium. Since
this represents an unstirred layer of gelatinous fluid it contributes to the imped-
ence of transport. Dehydration surprisingly increases the resistance of this layer
to solute transport, a phenomenon explained by the resultant distortion of the
porous channels of this layer [23].
Studies of transport across isolated mesentery suggest that the mesothelial cells
do not contribute importantly to transport resistance. Such measurements are
open to question unless it is verified that the membrane remains both intact
(scanning electron microscopy) and viable (dye studies) throughout the study.
The mesothelial cells are flattened and overlapping with tight junctions between
them [22]. They lie on a continuous basement membrane and contain numerous
intracytoplasmic vesicles. Permeation of solutes into the isolated hemidiaphragm
is lower in areas covered by the mesothelium compared to bare areas. This
impedence is offset by the addition of a redox dye to the system and is restored by
adding malate or succinate but lost again when malonate is added [24]. These
results suggest that oxidative metabolism and A TP formation are intimately
linked in regulating the diffusion process through this cell layer , indicating that it
should be responsive to pharmacologic manipulation

5. Dialysate flow rate

The major determinant of the rate of transport Of any given solute by diffusion is
the electrochemical concentration gradient. This gradient dissipates as solute
leaves the plasma and accumulates in dialysate. Obviously impractical, infinitely
high blood and dialysate flow rates would maintain maximal concentration
gradients. For large poorly diffusible solutes the accumulation rate in dialysate is
so slow that increasing the mean rate of dialysis solution exchange above two
liters per hour cannot increase the clearance by more than the dialysate/plasma
concentration ratio ordinarily achieved, usually about 10%. With intermittent
peritoneal dialysis the usual drainage rate of dialysate is about 2.1l/hr or 35 mIl
min. Under this circumstance the clearance of a small, highly diffusible solute like
urea is about 20 ml/min indicating incomplete equilibration, i.e., a dialysate/
plasma concentration ratio of 20/35 or about 0.6. Since this concentration ratio
decreases with shorter dwell times, increasing the dialysate exchange rate can
only increase the clearance by about 30% which corresponds to the maximal
increment observed [25]. It should be recognized, however, that when dialysate
volume is insufficient to contact the entire peritoneal surface, clearances will be
suboptimal until the exchange volume is appropriately increased [26]. Accord-
ingly, clearance decreases as fluid is being exchanged and can be augmented by
 271

leaving a residual volume in the peritoneum as rapid exchange of the excess

volume proceeds [27]. But, greatly improved rates of mass transport must depend
on augmentation of blood flow or enhancement of peritoneal permeability or
area, just as improvements in hemodialyzer efficiency have accompanied the
development of larger surface area dialyzers with more permeable membranes
and higher blood flow rates with arteriovenous fistulae rather than external
arteriovenous shunts. Recognition of the limited value of high dialysate flow rates
prompted Popovich and colleagues [6] to develop the technique of continuous
ambulatory peritoneal dialysis which prolongs diffusion time rather than increas-
ing the volume or exchange rate of dialysis solution. The procedural variant,
continuous cyclic peritoneal dialysis, combines this concept with intermittent,
more rapid exchanges [28, 29]. The effect of dialysis solution flow rate and of
blood flow rate on peritoneal clearances are shown in Figure 1.

6. Mesenteric blood flow

Blood flow to the peritoneal dialysis membrane derives predominantly from the
mesenteric circulation since the visceral peritoneum is much larger than the
parietal peritoneum. Mesenteric blood flow rates, as determined by flow probes

30

C 20
ml/mln

10

10 20 30 40 50 60
QD ml/mln
Figure 1. As dialysate flow rate (or volume exchanged) increases, peritoneal clearance of a highly
diffisible solute (bars) increases rapidly until flow rate is high and thereafter gradually. At any given
dialysate flow rate, increases in blood flow induce a curvilinear increase in clearance (dots).
272

or by solute clearances, average about 10% of the cardiac output or 40 mllmin/

100 g [30, 31], while the effective blood flow rate to the human peritoneum
averages 60 to 100 ml/min [17]. When mesenteric blood flow is increased by 100%
the clearances of small solutes like urea increase by 30 to 50% [32], consistent
with a resting blood flow that normally exceeds the maximal rate at which the
capillary diffusion capacity can completely clear the perfusing blood [33].
The splanchnic vascular bed can sequester blood, excluding it from or releasing
it into the circulat~on as systemic volume changes. Thus, hemodynamic effects of
drugs can influence splanchnic blood volume and flow rate considerably. Because
drugs usually affect the splanchnic blood flow and volume pari passu, changes in
peritoneal transport that result from the increased volume can be misinterpreted
as resulting from the augmented flow rate. The mesenteric vasculature is ac-
companied by autonomic neuroelements from the celiac plexus. The sympathetic
innervation of the splanchnic vessels is the primary mechanism of neurocontrol.
Both alpha- and beta-adrenergic receptors are located in the mesenteric vessels
[34]. These vessels also contain dopaminergic receptors. Vasoconstriction and
vasodilation of the mesenteric vascular bed in response to pharmacologic manip-
ulations are well established. The vasoconstrictor response that normally occurs
with appropriate stimuli can be prevented by pretreatment of the mesenteric
vascular bed with the alpha receptor blocking agent, phenoxybenzamine. It has
also become increasingly clear that the prostaglandins are intimately involved in
the fine control of vascular dynamics by virtue of their capability of modifying
vasoconstrictor responses [35]. The opportunities for increasing peritoneal mass
transport by pharmacologic modulation of blood flow to the peritoneum are
numerous [36].

7. Convective transport

Solute movement also occurs by convection because of osmotically induced

ultrafiltration. If the convecting holes are large enough, a solution may be forced
from one compartment to another without a change in composition, e.g., by
hydrostatic pressure filtration. Since the pores, through capillary walls, restrict
the passage of protein, ultrafiltration occurs. A compositional change in smaller
solutes such as urea is minimal or does not occur with peritoneal ultrafiltration.
Solutes as large as inulin are sieved during peritoneal ultrafiltration and the
movement of proteins and dextrans is restricted considerably. The rate of ultra-
filtration is normally determined by the hydrostatic pressure of the blood, which
decreases from 32 to 15 mm Hg, from the arterial to the venous end of the
capillary. This ultrafiltration pressure is opposed by the plasma oncotic pressure,
normally 25 mm Hg, and by the interstitial hydrostatic pressure minus the intersti-
tial osmotic pressure. The ultrafiltration rate through mesenteric capillaries at
normal pressures should approximate 3.0 ml/min/m2 of surface area. Most of this
 273

ultrafiltrate returns to venules and lymphatics, the walls of which have lower
transport resistance than the mesothelium. Normally, peritoneal fluid resembles
lymph from the leg rather than from the hepatic or thoracic duct [37] and is
derived from mesenteric capillaries. With increased hepatic venous pressure the
surface of the liver contributes predominately to ascites formation.
When two liters of isotonic dialysis solution is infused intraperitoneally it
causes sufficient extravascular hydrostatic pressure to promote the absorption of
fluid at a rate of about 10% of residual dialysate volume per hour [38]. The
addition of dextrose raises the osmotic pressure of the dialysis solution suf-
ficiently to overcome the absorptive tendency and thereby induce net ultrafiltra-
tion in proportion to the dextrose concentration of the instilled fluid. Because of
the restricted diffusion coefficient of dextrose relative to the solvent, net ultra-
filtration occurs at the rate of about 3.0 mllminlm2 of surface area when 1.5%
dextrose dialysis solution is infused. Because ofthe inward diffusion of dextrose,
however, the osmotic pressure gradient dissipates rapidly despite metabolic
degradation of the absorbed glucose and the ultrafiltration rate decreases with
time. The rate of ultrafiltration can be increased by using a higher concentration
of dextrose or, if practical, a less permeant solute of comparable osmotic activity
or by drugs that increase the capillary filtration coefficient [39], or drugs that
increase capillary hydrostatic pressure because of inducing relatively more con-
striction of venules than of arterioles, which may be the case with dopamine [40].
In patients with renal failure the use of hypertonic dextrose dialysis solution is
accompanied by increased rates of solute loss, a change that has been attributed
to enhanced permeability [41]. Most of the increased solute removal with hyper-
tonic dextrose dialysis solution can be accounted for by increased convective
transport, however. Indeed, in the intact rabbit convective transport increases
but the rate of diffusion does not rise when hypertonic dextrose dialysis fluid is
used. Unlike diffusion which separates solutes according to molecular size,
removal by convection does not discriminate according to size until sieving occurs
as the dimensions of the effective pores are approached. Since convection adds
more to the transport rate of slowly diffusible solutes, the net effect mimics an
increase in permeability of the diffusion barrier.
The complex of lipid and protein forming biologic membranes has interstices
and discontinuities between lipid and protein so that pores of a sort exist for
diffusion and ultrafiltration. The Pappenheimer theory of restricted diffusion
across capillaries takes into account 1) the stearic hindrance at the entrance of a
pore, 2) friction between molecules moving within a pore and 3) molecular
friction with the stationary walls of a pore as factors impeding the passage of
molecules through pores of molecular dimensions. Such pores are lined by the
fixed ionic charge groups of protein (amino-, imino-, and carboxyl) and of lipid
(phosphate and choline) [42].
These ionic charges restrict the diffusive and convective passage of charged
solutes through the membrane. For example, the rate of absorption of acids and
274

bases from peritoneal fluid decreases to the extent of their ionization at phys-
iological pH [38]. Diffusive transport rates of potassium, lithium and phosphate
across the peritoneum are slower than the rates of uncharged solutes of similar
size, unlike the transport across synthetic hemodialysis membranes [15]. More-
over, in the absence of a diffusion gradient the sodium concentration of ultrafiltr-
ate induced by hypertonic peritoneal dialysis is much lower than that of plasma,
i.e., about 75 mEq/1 [43]. The addition of furosemide to the dialysis solution
decreases the inhibition of sodium transport during osmotic ultrafiltration across
the peritoneum [44], consistent with a pharmacological modification of mem-
brane physiology. The ionic charge of capillary walls has previously been shown
to be an important determinant of the composition of ultrafiltrate across the
glomerulus [45], but its importance in diffusion through biological membranes
such as the peritoneum has not been stressed. Recently, it has been demonstrated
that intraperitoneal but not intravenous administration of the fungicide, ampho-
tericin B, significantly increases peritoneal ultrafiltration without altering the
osmotic gradient [9]. Amphotericin B induces channel formation in biological
membranes promoting the passage of solutes and water.

8. Transport of lipids

The transport of fatty acids into peritoneal fluid does not proceed by simple
diffusion and convection from plasma. Because these lipids rapidly diffuse
through cell membranes, concentration equilibrium across biological membranes
is reached within a few minutes. The concentration ratio dialysate/plasma water,
however, is far above unity for several fatty acids. Diffusion equilibrium does not
occur from plasma to dialysate, or from gastrointestinal luminal contents to
dialysate and is uninfluenced by circulating concentrations of lipases [10]. Rather,
the flux of non-esterified fatty acids occurs from adjacent adipose tissue to
peritoneal fluid and thereafter into portal venous blood [46]. It has also been
shown that absorption of barbiturates of comparable size is a function of their
lipid partition coefficient [38]. Whether lipid soluble drugs can be removed from
fat stores in the mesentery by this process remains to be established and methods
to exploit this transport mechanism should be studied.

9. Mechanisms of accelerating peritoneal transport

An adequate rate of transport by peritoneal dialysis requires enough blood flow

to the dialyzing surface, sufficient area and permeability of the membrane to
allow rapid permeation of solutes and ultrafiltration of fluid, as well as rapid
diffusion throughout the dialysate which is periodically replaced, thereby main-
taining electrochemical gradients.
 275

Mechanisms whereby peritoneal transport might be augmented are outlined

below and conceptually presented in Figure 2. By increasing blood flow to the
peritoneum the rate of solute delivery to the membrane can be increased, which
accelerates the transport of small highly diffusible solutes, but only modestly.
Increased splanchnic perfusion, however, augments peritoneal clearances of
larger solutes to at least as great an extent as it does the transport of smaller
solutes. This may be explained by an increase in peritoneal surface area or
permeability resulting from vasodilation. This can be attributed to a combination
of dilation of the functional peritoneal capillaries and perfusion of an increased
number of capillaries. Such capillary dilation spreads the same wall mass over a
larger circumference, thereby decreasing the wall thickness and stretching pores.
Intercellular junctions widen allowing increased mass transport [23]. Evidence
has also been presented [47, 48] that augmentation of blood flow by local

CONTROL
o ~
t SOLUTE
DELIVERY
MORE SMALL SOLUTES
CAN PERMEATE

tPERMEABILITY LARGER SOLUTES

CAN PERMEATE

tAREA MORE SOLUTES

PERMEATE

t VOLUME HIGHER GRADIENT

SMALL SOLUTES

tREMOVAL ~ GRADIENT HIGH FOR

.ATE ~ SMALL SOLUTES

Figure 2. A schematic representation of solute removal by peritoneal dialysis. The circles represent
the peritoneum. The long arrow above the circle represents peritoneal blood flow.
276

application of vasodilators open previously closed capillaries, increasing the

surface area available for transport. In the resting state blood may circulate
predominately through metarterioles. Enhanced perfusion opens more capil-
laries, exposing blood to a more permeable surface. Furthermore, permeability
of the venular end of the capillary may be greater than that of the arteriolar end.
Vasodilators with a predominant venular site of action may cause greater in-
creases in diffusion rates but arteriolar dilators may increase the ultrafiltration
rate. By increasing blood flow, diffusion and ultrafiltration may occur throughout
a greater length of the capillaries than occurs under resting conditions.
Depending on the nature of the vasodilating agent there may be an increase
(arteriolar relaxation), decrease (lowered venular tone), or no change (balanced
effects) in capillary hydrostatic pressure. This hydrostatic pressure may affect
capillary diameter increasing capillary permeability and blood volume. It is also a
major determinant of the rate of filtration through the capillary. Certain drugs
can affect specifically the capillary filtration coefficient, which may be defined as
the volume filtered per unit of pressure per unit of time (mllmm Hglmin). The
rate of ultrafiltration under the artificial circumstances of intraperitoneal fluid
administration is largely determined, however, by the osmotic gradient across the
peritoneum, customarily induced by dextrose. Since dextrose is a diffusible
solute, the gradient dissipates rapidly and at any given time cannot be assumed.
Increased solute permeability of the peritoneum increases the rate of glucose
diffusion so that under this circumstance a constant fluid flux must be interpreted
as consistent with an increased flux/gradient.
Specific drugs may affect directly the permeability of the capillary or the
mesothelium. Drugs that influence membrane charge, cell volume, cell metabo-
lism or intercellular junctions may directly influence peritoneal permeability
without affecting flow rates.
Not only is the rate of flow of dialysis solution a determinant of the transport
rate, at least of small highly diffusible solutes by virtue of maintaining the
electrochemical gradient but also, dialysate contact with the membrane, which is
a function of dialysate volume, affects transport. It appears that most of the
membrane surface area is contacted by a dialysis fluid volume of about 1.51/m2
[26]. The rapid exchange of small volumes of dialysate by enhancing mixing
increases transport rates by decreasing the impedance due to unstirred layers.
When such techniques leave a residual volume of dialysate in the peritoneum they
maintain peritoneal surface contact throughout the interval of fluid exchange,
which also increases efficiency. In an in vitro model of peritoneal dialysis it has
also been shown that maintaining the chemical gradient by charcoal adsorption of
solutes in the dialysate increases clearances [49].
Finally, the transport rate of specific solutes may be accelerated, e.g., by
chelates or adsorbents, by changing pH, thereby influencing non ionic diffusion
or by adding protein to dialysate to bind toxins. It is likely that maximal transport
rates will only be achieved by combinations of maneuvers affecting different
 277

resistances to transport. It has been shown recently that the transport accelera-
tion effects of intraperitoneal nitroprusside, increased dialysis fluid flow rate,
temperature and dextrose concentration are additive [50].

10. Restoration of decreased transport rates toward normal

When peritoneal blood flow has been reduced by disease thereby lowering
clearances, transport rates may be restored toward normal by treating the specific
abnormality. For example, the mesenteric blood flow rate varies directly with
cardiac output. Treatment of congestive heart failure with digoxin and with
ancillary supportive measures should improve peritoneal clearances. In three
patients that we have evaluated before and after treatment of heart failure, mean
clearance of creatinine increased from 8.1 mVmin to 11.9 ml/min while urea cleara-
nce increased from 12.4 mllmin to 16.9 ml/min. Chronic congestive heart failure
with hepatic congestion may increase portal venous pressure, however, increas-
ing splanchnic volume and capillary diameter.
Depletion of effective circulating blood volume, as occurs in hemorrhagic
hypotension, reduces peritoneal transport rates of urea and potassium in the dog
[51]. When blood pressure and volume are restored toward normal by infusion of
blood or saline, clearances return to normal. After hemorrhagic hypotension,
clearances are not restored to normal by raising blood pressure by norepine-
phrine infusion, nor is transport affected adversely by lowering the blood pres-
sure further with phenoxybenzamine [52]. These studies suggest that blood
pressure, per se, does not influence importantly the efficiency of peritoneal
dialysis which does depend, however, on adequate splanchnic volume and perfu-
sion.
A variety of vascular diseases can impair the mesenteric arterial circulation, so
reducing peritoneal transport rates [5]. Vascular damage secondary to diabetes
mellitus is not considered reversible and the vasculitis of systemic lupus eryth-
ematosus does not readily respond to prednisone or other immunosuppressive
drugs [53]. On the other hand, some diseases that cause renal failure are charac-
terized by widespread vascular endothelial injury inducing platelet thrombi [54].
Examples include malignant hypertension and hemolytic uremic syndrome. Re-
duced peritoneal transport rates complicating these diseases improve with the use
of dipyridamole [55]. Since the augmentation of peritoneal transport rates per-
sists after the vasodilatory effects of dipyridamole have abated, it is attributed to
the antiplatelet aggregating effect of the drug. Peritoneal clearances of patients
with normal vasculature improve only modestly and transiently with di-
pyridamole administered orally or intraperitoneally and a modest increment in
solute transport across the peritoneum occurs in experimental animals given
dipyridamole intraperitoneally or intravenously [56].
The impaired peritoneal transport that complicates irreversible systemic vascu-
278

lar lesions also improves toward normal with the local application of vasodilators
such as isoproterenol [57, 58]. There is no evidence however, that increased
clearances result from improvement in the vascular disease. Rather, enhanced
transport may be attributed to vasodilation of diseased vessels as occurs when
such drugs are administered to patients with normal vasculature.

11. Increasing peritoneal transport above normal values

There is now ample evidence that peritoneal transport rates may be increased to
values exceeding normal, both in patients without vascular disease [59] and in
several animal models [60-62]. Evidence against a nonspecific effect associated
with an intraperitoneal inflammatory reaction includes the following. The effects
of mass transport can be separated from those on fluid flux and on solvent drag [8,
9, 44, 63]. Transport rates increase without the vasodilator inducing an intra-
peritoneal inflammatory exudate [59]. Certain drugs evoke acceleration of trans-
port when given either intravenously or intraperitoneally [56, 64]. The response
of peritoneal transport rates is in accord with the known properties of vasoactive
compounds, increasing with vasodilators and decreasing with vasoconstrictor
agents [64, 65]. Finally, inactive metabolites and drug vehicles do not affect
peritoneal transport rates [8, 65].
The study of a variety of vasodilator agents suggests that peritoneal clearances
will increase only if the compound selectively dilates the splanchnic vasculature
or is applied locally, e.g., by intraperitoneal instillation. Intravenous use of such
drugs may cause widespread vasodilation decreasing blood pressure, splanchnic
perfusion and splanchnic volume, thereby lowering peritoneal transport rates. To
date studies with membrane active agents have only demonstrated augmented
transport when the drugs are applied locally, i.e., instilled intraperitoneally.

12. Isoproterenol enhancement of peritoneal mass transport

In patients with reduced peritoneal clearances, Nolph and colleagues first demon-
strated improved rates of peritoneal transport by adding isoproterenol (0.06 mg/l)
to the dialysis solution instilled intraperitoneally [57, 58]. Although mean clear-
ances increased to the lower range of normal values, the effect was transient and
not all patients improved significantly [66]. No systemic effects of intraperitoneal
isoproterenol were detected even with cardiac monitoring. Such use of iso-
proterenol has been explored ir greater detail in experimental animals. In acute
experiments in anesthetized dogs, Gutman et al. [61] demonstrated mean in-
creases in peritoneal urea and creatinine clearances of 45% and 30%, respec-
tively, with intraperitoneal isoproterenol but subpressor doses of isoproterenol
given intravenously did not augment transport rates. To extend these studies we
 279

evaluated the effect of a comparable dose (0.04 ILmollkg) of intraperitoneal

isoproterenol in normal unanesthetized rabbits undergoing experimental per-
itoneal dialysis [67[. Mean urea and creatinine clearances increased to 150% of
baseline values, but osmotically induced water flux was unaffected. No systemic
effects were observed. On exposure to light and air isoproterenol rapidly became
ineffective.
It is of interest that Felt et al. [32] demonstrated that despite increasing
mesenteric blood flow to 188% of control values by intravenous isoproterenol,
clearances did not increase. With intraperitoneal isoproterenol a comparable
increase in blood flow improved peritoneal inulin and creatinine clearances by
27% and 18% respectively. When blood flow was reduced to normal by a clamp
the clearances returned to normal. It appears that we may be monitoring the
wrong parameter, i.e., blood flow. Vasodilators increase volume as well as blood
flow. A higher capillary volume requires perfusion of a greater number of the
capillaries or widening of the channels or both, changes that result in increased
area or permeability or both. The observed increases in the ratio clearance large/
clearance small solute are consistent with the interpretation that increased
splanchnic blood volume not splanchnic blood flow rate causes the clearances to
increase.
Isoproterenol is a known inhibitor of catecholamine responses in smooth
muscle, so it relaxes the mesenteric vascular bed. When used clinically to acceler-
ate peritoneal transport rates, no systemic toxicity has been demonstrated despite
continuous use of isoproterenol for 20 exchanges [68]. Nevertheless, a better
transport accelerator continues to be sought because of the potential cardiotox-
icity and the need to apply the drug topically.

13. Effects of theophylline on peritoneal fluxes

Seeking a stable, safe vasodilator for clinical use during peritoneal dialysis the
effects of aminophylline on transport parameters in rabbits have also been
studied. Changes in solute transport and in osmotic water flux were inconsistent,
both after intraperitoneal and intravenous aminophylline, despite doses (30 to
150 ILmollkg) sufficient to achieve blood theophylline concentrations exceeding
the normal therapeutic range [69]. Yet, theophylline is known to relax smooth
muscle and dilate the systemic vasculature. Peritoneal flux of theophylline was
almost as high as urea transport rates, suggesting that rapid absorption of this
xanthine induced widespread vasodilation, just as intravenous administration
did, with no net gain in splanchnic blood flow or volume. Indeed, the transport
rate of theophylline by peritoneal dialysis in the rat is high enough to remove 28%
of an administered dose, decreasing the biologic half-life significantly and sug-
gesting clinical application for treatment of overdosage of this drug [70].
280

14. Nitroprusside augmentation of peritoneal mass transfer

The observations by Nolph et al. [66] that intraperitoneal nitroprusside admin-

istration increases peritoneal mass transport values have been confirmed in
multiple laboratories in several species [71-74). Urea and creatinine clearances
increase as much as 50% above control values and the increments in inulin
clearances and protein loss are even greater, consistent with enhanced peritoneal
permeability or area or both rather than simply increased solute delivery. Osmo-
tic ultrafiltration increases slightly [72] or not at all because of rapid glucose
absorption dissipating the osmotic gradient. Nitroprusside-induced increases in
mass transport are dose dependent and can be seen with as little as 1.0 mg/kg [75].
Systemic effects of nitroprusside have not been detected in most studies and
intravenously the drug does not accelerate peritoneal mass transport. The trans-
port increment is sustained for several exchanges and on discontinuation may
persist somewhat for up to two hours [76]. Augmented transport can be explained
by an increase in permeance of the peritoneum (mass transfer coefficient x area)
resulting from capillary dilation, especially of the venous end and from opening of
previously non-perfused capillaries [75, 77]. Nitroprusside accelerates the onset
of vasodilation that usually follows the transient vasoconstriction of the mesen-
teric circulation which occurs after instillation of commercially available dialysis
solutions [77]. Although nitroprusside is metabolized to thiocyanate, this toxic
metabolite is removed rapidly by dialysis [78] and no evidence of accumulation
has been observed with repeated nitroprusside instillation.

15. Dipyridamole effects on peritoneal efficiency

As an accelerator of peritoneal transport dipyridamole has the advantage that it is

effective when given orally [56). Dipyridamole is a general smooth muscle
relaxant with pharmacologic properties similar to those of papaverine. It rapidly
but transiently induces vasodilation [79] and results in a sustained antiplatelet
aggregating effect. In addition to the restoration of clearances toward normal in
those patients with intravascular platelet aggregations, peritoneal transport of
urea and creatinine increase by 43 % and 70% respectively in patients with normal
vasculature, when given dipyridamole orally at a dose of 300mg daily [80, 81].
The clearance of radiolabeled EDTA and DPTA increase by 75% and 41%
respectively [81]. A modest increment in the clearances of uric acid and of inulin
has also been reported, but this response is delayed for a few days [82]. In normal
rabbits dipyridamole given intravenously (0.5 mg/kg) or intraperitoneally
(2.5 mg/kg) increased urea and creatinine clearances by 39 and 16% respectively
[56,83]. The limited effectiveness and the transient response make dipyridamole
a suboptimal accelerator of peritoneal transport for all patients, but it can be very
useful in selected circumstances when an oral drug is preferred or when systemic
disease involving platelet thrombi affects the mesenteric circulation.
 281

16. Influence of catecholamines on peritoneal transport kinetics

To explore further the influence of vasoactive drugs on peritoneal transport the

effects of catecholamines have been studied in experimental animals undergoing
peritoneal dialysis. In preliminary studies Gutman et at. [61] noted a decrease in
the increment in dialysate urea concentration when large doses of dopamine were
administered intraperitoneally to dogs, but did not measure dialysate volume.
Because blood pressure increased it was interpreted that splanchnic vasoconstric-
tion accounted for the lower accumulation of urea in the dialysate. To offset such
vasoconstrictor activity Parker et at. [84] added an alpha adrenergic blocker to the
dialysis solution. When phentolamine was added intraperitoneally while dopa-
mine was given intravenously they observed increased peritoneal clearances in
dogs. In patients, however, Chan et at. [85] observed no effect of a low dose (4 mg/
1) or high doses (20-160 mg/l) of dopamine added intraperitoneally on peritoneal
transfer rates of urea, creatinine or phosphate. They also did not measure
dialysate volume so the effect on ultrafiltration rate could not be discerned. Their
conclusions that dialysate flow rate «25 ml/min in their study) limits peritoneal
clearance of small solutes and that peritoneal blood flow rate is optimal are not in
accord with other studies. In rabbits intraperitoneal administration of dopamine
causes dose related (from 0.6 to 1.8 mg/kg) increases in peritoneal urea clearance
[64]. The increments occurred although the doses were lower than those used by
Gutman et at. [65] and drug concentrations (10 to 30mgll) were within the range
studies by Chan et at. [85]. Species differences may account for the discrepant
results.
The intravenous administration of I-norepinephrine significantly decreased
peritoneal clearances of urea and creatinine in unanesthetized rabbits [64, 85].
The decrement was dose dependent and correlated with the pressor response
[86]. Comparable pressor doses of dopamine given intravenously increased clear-
ances of urea and creatinine to 145% of control values, whereas low doses had
minimal and inconsistent effects. Osmotic water flux increased only slightly (from
0.18 to 0.24ml/kg (min) but significantly (p<0.02). Because dopamine vas-
oconstricts venules relatively more than arterioles as compared to norepine-
phrine [40], augmented water flux is interpreted to be mediated by increased
hydrostatic pressure rather than a change in the capillary filtration coefficient.
The augmented solute transport accompanying high doses of dopamine given
intravenously was unaffected by concurrent beta blockade with propranolol, was
decreased by alpha adrenergic receptor blockade by simultaneous administration
of phentolamine and was abolished by haloperidol, a drug that blocks
dopaminergic receptors [86]. Accordingly, the augmented transport is attributed
to the action of the drug on dopamine receptors causing mesenteric vasodilation
and in part on alpha adrenergic receptors simultaneously causing somatic vas-
oconstriction, increasing blood pressure while mesenteric blood flow is main-
tained. Although dopamine may not be a suitable drug for augmenting efficiency
282

of routine peritoneal dialysis, these data argue strongly that it should be prefera-
ble to I-norepinephrine when vasopressor therapy is required during peritoneal
dialysis.

17. Other vasodilators that affect peritoneal transport

No consistent change in peritoneal clearance of urea or creatinine was observed

in six studies of patients given 20 to 40 mg of hydralazine intraperitoneally [66].
Blood pressure decreased slightly with this dose of hydralazine by this route.
Because the molecular mass of hydralazine is 168 daltons it should be absorbed
from the peritoneal fluid, but its pharmacologic action may depend on bio-
transformation to an active compound. Accordingly, widespread vasodilation
may occur despite local application with no preferential effect on splanchnic
blood flow or volume.
Diazoxide caused a modest increase in peritoneal clearances of urea and
creatinine and a significant decrement in blood pressure when administered
intraperitoneally to patients at a dose of 100 to 300 mg [66]. An increase in
ultrafiltration rate approaching 50% of control values was found inconsistently.
In another study, a dose of 150 mg of diazo xi de increased peritoneal clearances of
urea, creatinine and phosphate by 58%, 48% and 39% respectively [81].
The intraperitoneal administration of 5 mg of phentolamine did not influence
peritoneal transport rates in five patients so investigated, nor did it affect osmotic
water flux [66].
In anesthetized rats, histamine (4 or 8 JLg) induced only modest increments (9%
to 16%) in the clearances of urea and inulin, whereas 3 JLg of bradykinin aug-
mented these clearances by 13% and 25% respectively [71]. Histamine causes
overt dilation and increased permeability, however, in the isolated rat mesentery
preparation viewed by television microscopy after fluorescein labelling [14].
Dilation is most prominent in the venous end of the capillary and similar changes
are noted with nitroprusside.
In preliminary studies of a few unanesthetized animals tolazoline increased
peritoneal urea clearances to more than 150% of control values [87]. Tolazoline,
which can be given by mouth, is a substituted imidazole that is a very potent
adrenergic blocker with pharmacologic properties somewhat similar to phento-
lamine.
It is thus apparent that a variety of vasoactive drugs can affect peritoneal
transport rates. For comparison, the effect of various drugs on peritoneal cleara-
nce of urea is shown in Figure 3. Since some of the drugs have been studied in a
limited number of species, only in a few animals or at only one dose, the estimates
may have considerable error. Isoproterenol, and especially nitroprusside, have
consistently augmented clearances in several studies from multiple laboratories.
Dipyridamole has the advantage that it can be given by mouth.
 283

CONTROL

ISOPROTERENOL (IP)

THEOPHYLLINE (IP, IV)

NITROPRUSSIDE (lP) ~
~==================~~~
DIPYRIDAMOLE (PO)

DOPAMINE (IP)

DOPAMINE (IV)

NOREPINEPHRINE (IV) ~
HYDRALAZINE (IP)

DIAZOXIDE (IP)

PHENTOLAMINE (IP)

HISTAMINE (IP)

BRADYKININ (IP)

TOLAZOLINE (lP)
I J J I I I

20% 40% 60% 80% 100% 120% 140% 160% 180%

Figure 3. The anticipated effect of vasoactive drugs on peritoneal clearance of urea is shown as a
percent of control by the bars. The lines indicate the maximal percentage increase reported.

Norepinephrine, as anticipated, consistently decreased clearances, whereas

dopamine not only did not but at certain doses appreciably increased transport
rates.
Because numerous other drugs are administered for various indications to
patients undergoing peritoneal dialysis the influences of a variety of other agents
on peritoneal mass transport have been explored.

18. Prostaglandin modulation of peritoneal transport

There is ample evidence that prostaglandins control regional blood flow by virtue
of their capability of modulating vasoconstrictor responses [35]. The prostaglan-
dins are a family of unsaturated 20 carbon lipids biosynthesized from a variety of
precursors, the most common of which is arachidonic acid, by prostaglandin
synthetase, an enzyme present in tissues [88]. The prostaglandin synthetic path-
way begins with the cyclooxygenation reaction wherein prostaglandin synthetase
converts arachidonic acid into the cycle endoperoxides PGG z and PGH z. These
are converted by specific terminal enzymes into pharmacologically active prod-
ucts including PGD z, PGA z, PGE z, PGF2a thromboxanes and prostacycline
(POI z). Depending on the local concentration of the specific terminal enzyme,
e.g., en do peroxide reductase leading to POF1a or endoperoxide isomerase lead-
284

ing to PGE2 , the pathway will favor a given product in a given tissue. Regional
blood flow is one determinant of enzyme activity. After release into circulation
the prostaglandins are degraded during a single passage through the pulmonary
circulation, thereby acting only locally with the exception of prostacycline and
thromboxanes which have half-lives of a few minutes. It is apparent that bio-
synthesis of these compounds can be inhibited by various drugs at varied sites in
the synthetic pathway. Prostaglandins of the PGA, PGE or PGI series vasodilate
whereas PGF2a and thromboxanes are potent vasoconstrictors [89, 90]. These
prostaglandins act locally in arterial walls to influence vascular tone and modulate
the response of vascular smooth muscle to other vasoactive agents [91].
Intraperitoneal administration of PGA1 or PGE1 increased peritoneal cleara-
nces of urea and creatinine modestly in the unanesthetized normal rabbit, where-
as 125 p.g/kg of PGE 2 significantly augmented creatinine clearance to 132% and
urea clearance to 180% of control values [60, 65, 92]. In contrast, intraperitoneal
administration of the vasoconstrictor PGF2a (125 p.g/kg) decreased peritoneal
clearances to 80% (urea) and 82% (creatinine) of control values [65]. No effect on
osmotic water flux occurred with these prostaglandins. Moreover, intravenous
administration of either type of prostaglandin did not influence peritoneal trans-
port significantly as anticipated because of rapid metabolic inactivation. Neither
intravenous nor intraperitoneal administration of prostacyclin affected per-
itoneal solute or water transport significantly over a dose range from 25 to 125 p.g/
kg [65]. It is interpreted that prostacyclin caused widespread vasodilation, so that
mesenteric blood flow and volume were not selectively increased. Hence, trans-
port remained unaffected.
Studies of prostaglandin synthetase stimulators and inhibitors have not shown
significant effects on peritoneal transport parameters. Oral pretreatment with 10
to 21 mg/kg of sulfinpyrazone, a potent stimulator of prostaglandin synthetase,
did not affect peritoneal clearances significantly [92]. When mefenamic acid, a
potent prostaglandin synthetase inhibitor, is administered intravenously or intra-
peritoneally to normal unanesthetized rabbits in doses sufficient to inhibit plate-
let function neither the peritoneal solute clearance (creatinine or urea) nor water
flux changes significantly [92]. Oral pretreatment of rabbits with indomethacin
blocked platelet aggregation but did not change clearances or ultrafiltration rates
significantly [92]. Intraperitoneal administration of indomethacin caused an in-
crease in the size of pinocytotic vesicles and a narrowing of intercellular spaces in
the rabbit, however [93]. But, such stimulation and blockade of prostaglandin
synthetase affects both vasoconstrictor and vasodilator prostaglandins, so there
may be no net change in regional blood flow. Imidazole, an agent that specifically
blocks the synthesis of vasoconstrictor thromboxanes, did not augment cleara-
nces, however, when given intraperitoneally in a dose of 5 mg/kg. Yet, when
vasodilator prostaglandin activity predominates because of homeostatic compen-
sation for increased renin'angiotensin activity or ischemic vascular disease, non-
specific aspirin or indomethacin decreases regional blood flow [94]. In selected
 285

patients blockade of prostaglandin synthetase by nonsteroidal anti-inflammatory

agents has caused acute renal failure [95] and could induce mesenteric ischemia
with decreased peritoneal clearances as well. However, the reduction of cleara-
nces induced by intravenous I-norepinephrine, which should be accompanied by
vasodilator prostaglandin stimulation, is exaggerated by pretreatment with in-
domethacin in only half of the animals so studied. These results suggest that
endogenous prostaglandins do not play a major role in regulating peritoneal
blood flow under ordinary circumstances. But, in patients that depend on vas-
odilator prostaglandins to maintain organ perfusion, blockade of prostaglandin
synthetase could reduce transport rates and a history of exposure to such drugs
should be sought if clearances are low. It is of interest that intraperitoneal
administration of the prostaglandin precursor, arachidonic acid (1.5 to 5.6mg!
kg), increased creatinine clearance by 36% and urea clearance by 24 % suggesting
an effect of endogenous prostaglandins, but systemic use of indomethacin did not
block this increase [92, 96].

19. Vasodilator gastrointestinal hormones

The gastrointestinal hormones are a group of polypeptides that specifically dilate

the splanchnic vasculature and induce certain physiologic effects [97]. Glycagon
and secretin are structurally similar and increase mesenteric blood flow by as
much as 100% above baseline when pharmacologic doses are given but augment
different gastrointestinal functions [98]. Secretin, like cholecystokinin, increases
predominantly hepatic blood flow, but glucagon has the more potent effect on the
mesenteric circulation. Gastrin, which is structurally similar to cholecystokinin,
also increases superior mesenteric arterial blood flow [98]. The augmented
mesenteric circulation can be induced after intraduodenal instillation of corn oil,
I-phenylalanine or acid, following a short latency consistent with the physiologic
release of secretin or cholecystokinin [99]. Similar changes can be induced by low
doses of these hormones given intravenously. The effects of secretin and cho-
lecystokinin on mesenteric blood flow are additive and are potentiated by the-
ophylline [100]. Mesenteric vasodilation caused by these hormones is unaffected
by adrenergic activity and is interpreted to represent direct relaxation of vascular
tone presumably mediated by cyclic AMP. In contrast to these hormones, soma-
tostatin and vasoactive intestinal peptide vasoconstrict the splanchnic bed, reduc-
ing mesenteric blood flow rates [97].
When administered intravenously immediately predialysis, 30 JLg/kg of glu-
cagon significantly increased peritoneal clearances of urea and creatinine in
normal unanesthetized rabbits [68, 101]' Intraperitoneal administration of the
same dose of glucagon did not augment clearances significantly. Since this large
peptide should traverse the peritoneum slowly it is interpreted that hormonal
activity occurs at the endothelial, rather than the mesothelial surface. Neverthe-
286

less, much higher doses given intraperitoneally do increase clearances, although

not to the extent that intravenous doses do.
Bolus administration of these hormones is known to be more effective than
continuous infusion despite their short half-lives [102]. Nevertheless, intravenous
infusion of about 30 JLglkg/hr increased superior mesenteric arterial blood flow
and peritoneal inulin clearance (but not creatinine clearance) in dogs unlike
intraperitoneal instillation [32]. Slight increments in urea and creatinine cleara-
nces occurred when secretin or cholecystokinin were administered [63]. Intra-
venously, but not intraperitoneally, secretin (10 U/Kg) increased osmotic water
flux significantly in rabbits from 0.19 to 0.29 mllkglmin [63]. The endogenous
release of cholecystokinin or secretin, or their intra-arterial infusion, relaxes
precapillary sphincters and increases the capillary filtration coefficient from 0.05
to 0.10 mllmin/mmHg/100 g [103]. It has also been shown that secretin increases
the capillary filtration coefficient in the mesenteric vasculature of the cat [39].
Neither glucagon nor cholecystokinin affected water transport during experimen-
tal peritoneal dialysis in rabbits, however [63]. These studies of peritoneal
kinetics after intravenous administration of gastrointestinal hormones, by dem-
onstrating a separation of the effects on diffusive and on convective transport,
suggest the possible use of separate pharmacologic agents acting additively.

20. Other hormones and drugs affecting peritoneal blood flow and diffusion

Parenteral administration of vasopressin to anesthetized dogs decreased per-

itoneal clearances of small solutes, consistent with a hormonally mediated reduc-
tion in mesenteric blood flow [104, 105]. Since inulin clearance increased slightly
under such circumstances, it has been interpreted that there is a concurrent
increase in membrane permeability [105], which is in accord with the accelerated
transport that occurs in isolated membrane preparations [106].
In preliminary studies in rabbits and patients large doses of methyl pred-
nisolone increased peritoneal clearances by as much as 69%, but such augmented
transport was observed inconsistently [107].
The peritoneal transport rates of potassium and 131iodide increased when
streptokinase or serotonin were administered systemically to anesthetized dogs
[104]. Whether these agents affect peritoneal permeability or simply augment
blood flow remains to be determined.
In sedated rabbits dialyzed with Hanks solution made hypertonic with man-
nitol, the intraperitoneal addition of procaine hydrochloride to achieve a final
concentration of 0.25% increased peritoneal urea and inulin clearances to more
than 160% of control values [108]. The effect persisted for at least one hour after
procaine was discontinued. Procaine vasodilates, which may contribute to the
augmented transport. However, in studies with isolated mesothelium the ad-
dition of procaine to either side of the membrane increased transport after a
 287

transient decrease. This effect may be explained by disruption of the microfila-

ments of tight junctions between cells.
Variations in electrolyte concentrations also influence peritoneal blood flow
and permeability. For example, topically applied low potassium solutions vas-
oconstrict several vascular beds including the mesenteric, interfering with normal
Na-K ATPase activity and should thereby decrease peritoneal transport effi-
ciency. Whether this contributes to the transient vasoconstriction that immedi-
ately follows instillation of peritoneal fluid remains to be established. Modest
increments in potassium concentration vasodilate, an effect that is inhibited by
oubain [109].
The dialysis solution itself affects the peritoneal vasculature. Miller and col-
leagues [47, 110] have documented carefully that the mesenteric vasculature
transiently vasoconstricts and then vasodilates when exposed to peritoneal di-
alysis solution. The prolonged vasodilation depends on the presence of hyper-
osmolar dialysis fluid containing either acetate or lactate. Bicarbonate buffered
isotonic Krebs solution did not induce vasodilation.

21. Membrane surface-active agents

Penzotti and Mattocks [62, 111, 112] measured accelerated rates of transport of
radiolabelled urea and creatinine across the peritoneum of sedated rabbits by the
addition of a variety of surface-acting agents including dioctyl sodium sulfosucci-
nate, cetyl trimethyl ammonium chloride and N-myristyl-j3-amino-proprionate.
Although none of these compounds have been employed clinically, they may
point the way to identifying agents in popular clinical use that could be combined
with appropriate vasodilators.
More recently, Dunham et al. [113] verified that a dose dependent increase in
creatinine and urea clearances occurred when docusate sodium was added to the
peritoneal dialysis solution in tranquilized rabbits. The effect persisted for up to
five hours after the agent was instilled.
The solubilizing agent, dimethyl sulfoxide, did not augment clearances of
potassium or urea in rabbits, however, except to the extent that large doses
intraperitoneally created an osmotic gradient and increased convective transport
[114].
Cytochalasin D disrupts intercellular junctions. When given intraperitoneally
this agent increases significantly the clearances of creatinine and urea in the
rabbit, consistent with augmented diffusion through intercellular gaps [8]. Con-
currently, the ultrafiltration rate decreases, which is attributed to accelerated
glucose transport diminishing the osmotic gradient more rapidly, mimicking the
transport alterations of acute peritonitis [115].
The addition of 1 mg/kg of furosemide to hypertonic peritoneal dialysis solution
increased the rate of sodium movement accompanying osmotically induced water
288

flux in rabbits [41]. Normally, electrolytes do not accompany water in the same
concentration as exists in plasma water, suggesting that membrane charge im-
pedes transport, a phenomenon that is interrupted by furosemide. A 27% in-
crease in peritoneal urea clearance also resulted from the intraperitoneal addition
of furosemide, but no demonstrable changes in transport rates occur in patients
undergoing intermittent peritoneal dialysis when treated systemically with this
diuretic. Moreover, oral use of furosemide did not affect sodium, potassium or
water transport into dialysate in patients undergoing continuous ambulatory
peritoneal dialysis [116]. On the other hand, furosemide does increase the trans-
port of uric acid and of barbiturates into peritoneal dialysate [117]. Intra-
peritoneal instillation of 1.25 mglkg of ethacrynic acid did not affect the move-
ment of sodium with the bulk flow of water across the peritoneum but augmented
urea clearance to about 165% of baseline values [44]. It has also been shown that
protamine sulfate, another agent that offsets membrane charge effects, increases
the mass transport of urea and inulin when added to the dialysis solution [118].

22. Increased ultrafiltration rates

Because the peritoneal ultrafiltration rate is normally low compared to the

clearance of low molecular weight solutes (10 to 20% ), even doubling this rate has
only a modest effect on removal rates of such substances. On the other hand,
ultrafiltration contributes importantly to the removal of large, poorly diffusible
solutes as has been demonstrated with hemodialysis [119].
The ultrafiltration rate can be increased by increasing the hydrostatic pressure
gradient across the capillary wall, by constriction of the venous end of the
capillary, or by arteriolar relaxation. Few drugs appear to affect capillary hydro-
static pressure but this mechanism can account for increased ultrafiltration by
dopamine [86].
Secretin exemplifies an agent that increases the hydraulic permeability of the
peritoneal membrane [63]. This action is selective for the splanchnic vascular bed
and occurs only when secretin is given intravenously. The aminonucleoside of
puromycin which is known to induce glomerular lesions characterized by in-
creased permeability to macromolecules also induces such an effect on peritoneal
capillaries [120]. A few days after this drug is administered a proteinaceous ascites
develops in rats. When radiolabelled test solutes are then administered intra-
venously the appearance rate in peritoneal fluid is faster than in control rats.
Although the prolonged effect of such a permeability augmenting drug could be
commendable, the likelihood that other nonglomerular capillary beds are simi-
larly affected would make it hazardous even in anephric patients.
Amphotericin B increases the rate of ultrafiltration in response to a constant
osmotic gradient, that is, raises the ultrafiltration coefficient [9]. Above the
therapeutic dose, 0.5 mg/kg, there is no dose effect and the drug is effective only
 289

on the serosal side of the membrane [9, 121]. Amphotericin B creates channels in
biological membranes through which solutes and water penetrate readily [122].
During peritoneal dialysis the increment in solute clearance when amphotericin B
is instilled intraperitoneally is only modest and can be accounted for by enhanced
convective transfer rather than diffusion [9]. However, peritoneal mass transport
of sodium also increases. Because osmotic ultrafiltrate during peritoneal dialysis
is hyponatric, the sodium gradient so established across the membrane is an
impediment to water transport that amphotericin B cancels [121]. Amphotericin
B or safer analogues may be useful in treating the reduced ultrafiltration capacity
that occasionally complicates long-term peritoneal dialysis. Chlorpromazine may
also have such usefulness.
The osmotic gradient across the peritoneum, which also determines the rate of
ultrafiltration, is increased when plasma protein concentrations are lowered or
more transiently when plasma sodium concentration is decreased. Therapeutic-
ally, however, a high osmotic gradient is achieved by adding dextrose to the
dialysis solution. Since excessive glucose absorption can be hazardous to patients
undergoing peritoneal dialysis, alternative osmotic agents have been evaluated
[123-126]. Amino acids added to the dialysate offer a nutritional advantage but
may be costly [123]. Although the addition of dextrans to peritoneal dialysate
[124, 125] does not have such nutritional advantages, they can be made large
enough to permeate the peritoneum poorly, thereby maintaining the osmotic
gradient throughout the exchange. Clearances of potassium and of creatinine
with dextran dialysis fluid are no different than with commercial fluid. When
xylitol is substituted for dextrose there is no effect on peritoneal transport
parameters, but plasma changes consistent with glucose deprivation occur [126].
By raising the osmotic gradient with higher concentrations of dextrose solute
removal is enhanced, at least in part because of increased convection [41].
Zelman et al. [127] have used hypertonic dialysis solution to increase solute
clearances alternating such exchanges with hypotonic solution to maintain fluid
balance, achieving higher rates of solute removal. With the appropriate osmotic
agent to enhance convective transport, clearances should be augmented while the
patient is simultaneously allowed and required to increase fluid intake. The role
of the hemodynamic effects of absorbed glucose on peritoneal transport rates
needs further evaluation.

23. Transport acceleration of specific solutes

Removal of barbiturates may be accelerated by increasing dialysate pH with tris

buffer, thereby influencing the rate of non-ionic diffusion [128, 129]. An increase
in uric acid transport with alkalinization of peritoneal dialysis fluid by THAM has
also been observed [130]. Peritoneal transport rates of barbiturates are also
increased by adding albumin to peritoneal dialysis solution since binding of short-
290

acting barbiturates to plasma proteins is a major impediment to net diffusion from

plasma to dialysate [131]. The observed increases in peritoneal clearances of
barbiturates, phenytoin and salicylates with intraperitoneal instillation of N-myr-
istyl alanine and anthranilic acid have been attributed to displacement of protein
binding of these drugs by the instilled chemicals [132]. The addition of albumin to
peritoneal dialysis solution also enhances removal of ethchlorvynol [133] and
salicylate [134] and predictably should augment the clearance of numerous other
drugs that circulate bound to plasma proteins, such as some aminoglycoside
antibiotics, quinine and phenytoin. The addition of albumin to peritoneal dialy-
sate also augments removal of trace metals, such as copper [135]. Such enhanced
removal of trace elements can also be accomplished with the use of a specific
chelate [136-139]. For drugs with a high lipid partition coefficient, such as
glutethimide and short-acting barbiturates, enhanced transport can be achieved
by adding lipid to the dialysate [140]. In general, the removal rate of drugs,
especially those with a high degree of protein or lipid affinity is so slow by
peritoneal dialysis that other methods should be used for severe overdosage. It is
advisable to be aware ofthese specific effects, such as chelation, however, as they
may influence therapeutic drug concentrations and certain uremic metabolites.

24. Conclusions

It is easy for clinicians as well as bioengineers to forget that the peritoneum,

unlike synthetic hemodialysis membranes, is alive. The mesenteric circulation is
remarkable for its size and complexity and, until recently, for the paucity of
knowledge about its physiology. The numerous drugs and hormones that affect
mesenteric blood flow and membrane physiology should have predictable effects
on peritoneal transport parameters [36, 141]. Patients undergoing chronic dialysis
characteristically receive several drugs regularly, many of which have hemo-
dynamic and membrane transport effects. The influence of these agents on the
peritoneum must be ascertained. Those treated for acute problems, for example
in an intensive care unit, are exposed to even a greater abundance of drugs,
potentially altering transport. Rational use of drugs and other physiologic manip-
ulations in patients maintained by peritoneal dialysis requires an understanding
of their effects on peritoneal blood flow and permeability. It is naive to consider
the peritoneum as an inert membrane with constant blood flow and transport
characteristics. Further investigation of the interactions of drugs and the per-
itoneum may identify optimal methods for augmenting transport efficiency.
 291

References

1. Broyer M, Brunner FP, Brynger H, Donckerwolcke RA, Jacobs C, Kramer P, Selwood NH,
Wing AJ: Combined report on regular dialysis and transplantation in Europe, XII, 1981. Proc
Eur Dial Transpl Assoc 19: 3, 1982.
2. Oreopoulos DG: Chronic peritoneal dialysis. Clin Nephrol 9: 165, 1978.
3. Nolph KD: Continuous ambulatory peritoneal dialysis. Am J Nephroll: 1, 1981.
4. Verger C, Brunschvicg 0, Le Charpentier Y, Lavergne A, Vantelon J: Structural and ultrastruc-
tural peritoneal membrane changes and permeability alterations during continuous ambulatory
peritoneal dialysis. Proc Eur Dial Transpl Assoc 18: 199, 1981.
5. Nolph KD, Stoltz ML, Maher JF: Altered peritoneal permeability in patients with systemic
vasculitis. Ann Intern Med 75: 753, 1973.
6. Popovich RP, Moncrief JW, Nolph KD, Ghods AJ, Twardowski ZJ, Pyle WK: Continuous
ambulatory peritoneal dialysis. Ann Intern Med 88: 449, 1978.
7. Nolph KD, Miller FN, Pyle WK, Popovich RP, Sorkin MI: An hypothesis to explain the
ultrafiltration characteristics of peritoneal dialysis. Kidney Int 20: 543, 1981.
8. Hirszel P, Dodge K: Peritoneal transport increase by changing intercellular junctions. Abstracts
Am Soc Artif Intern Organs 12: 50, 1983.
9. Maher JF, Hirszel P, Bennett RR, Chakrabarti E: Amphotericin B selectively increases per-
itoneal ultrafiltration. Abstracts Am Soc Nephrol16: 121A, 1983.
10. Maher JF. Hirszel P, Hohnadel DC, Abraham J, Lasrich M: Fatty acid removal during per-
itoneal dialysis. Mechanisms, rates and significance. ASAIO J 1: 8, 1978.
11. Dedrick RL, Myers CE, Bungay PM, De Vita VT Jr.: Pharmacokinetic rationale for peritoneal
drug administration in the treatment of ovarian cancer. Cancer Treat Rep 62: 1, 1978.
12. Henderson LW, Cheung AK, Chenowith DE: Choosing a membrane. Am J Kidney Dis 3: 5,
1983.
13. Garred LJ, Canaud B, Farrell PC: A simple kinetic model for assessing peritoneal mass transfer
in chronic ambulatory peritoneal dialysis. ASAIO J 6: 131,1983.
14. Pappenheimer JR, Renkin EM, Borrero LM: Filtration, diffusion and molecular sieving
through peripheral capillaries; a contribution to the pore theory of capillary permeability. Am J
Physiol167: 13,1951.
15. Lasrich M, Maher JM, Hirszel P, Maher JF: Correlation of peritoneal transport rates with
molecular weight: a method for predicting clearances. ASAIO J 2: 107,1979.
16. Gosselin RE, Berndt WO: Diffusional transport of solutes through mesentery and peritoneum. J
Theor Bioi 3: 487, 1962.
17. Aune S: Transperitoneal exchange II: Peritoneal blood flow estimated by hydrogen gas cleara-
nces. Scand J Gastroenterol5: 99,1970.
18. Nolph KD: The first hemodialyzer. ASAIO J 1: 2, 1978.
19 Hirszel P, Chabrabarti EK, Bennett RR, Maher JF: Permselectivity of peritoneum to neutral
dextrans. Abstracts Am Soc Artif Intern Organs 13: 44, 1984.
20. Casley-Smith JR: An electron microscopical study of the passage of ions through the endo-
thelium of lymphatic and blood capillaries and through the mesothelium. Q J Exp Physiol 52:
105, 1967.
21. Cotran RS, Karnovsky MJ: Ultrastructural studies of the permeability of the mesothelium to
horseradish peroxidase. J Cell Bioi 57: 123, 1968.
22. Gotloib L, Digenes GE, Rabinovich S, Medline A, Oreopoulos DG: Ultrastructure of normal
rabbit mesentery. Nephron 34: 248, 1983.
23. Wayland H: Transmural and interstitial molecular transport. In: M Legrain (ed), Proc Int
Symposium Continuous Ambulatory Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1980, p
18.
24. Cascarano J, Rubin AD, Chick WL, Zweifach BW: Metabolically induced permeability changes
across mesothelium and endothelium. Am J Physiol 206: 373, 1964.
292

25. Robson M, Oreopoulos DG, Izatt S, Ogilvie R, Rapaport A, deVeber GA: Influence of
exchange volume and dialysate flow rate on solute clearance in peritoneal dialysis. Kidney Int 14:
486,1978.
26. Twardowski Z, Nolph KD, Prowant BF, Moore HL: Efficiency of high volume low frequency
continuous ambulatory peritoneal dialysis (CAPD). Trans Am Soc Artif Intern Organs 29: 53,
1983.
27. Ratnu KS, Haldia KR, Panicker S, Mathur AS: A new technique - semicontinuous rapid flow,
high volume exchange for effective peritoneal dialysis in shorter periods. Nephron 31: 159, 1982.
28. Diaz-Buxo JA, Walker PJ, Farmer CD, Chandler JT, Holt KL, Cox P: Continuous cyclic
peritoneal dialysis. Trans Am Soc Artif Intern Organs 27: 51, 1981.
29. Nakagawa D, Price C, Steinbaugh B, Suki W: Continuous cycling peritoneal dialysis: a viable
option in the treatment of chronic renal failure. Trans Am Soc Artif Intern Organs 23: 55, 1981.
30. Grayson J, Mendel D: Physiology of the Splanchnic Circulation. Williams and Wilkins Co,
Baltimore, 1965.
31. Lanciault G, Jacobson ED: The gastrointestinal circulation. Gastroenterology 71: 851,1976.
32. Felt J, Richard C, McCaffrey C, Levy M: Peritoneal clearance of creatinine and inulin during
dialysis in dogs: effect of splanchnic vasodilators. Kidney Int 17: 459, 1979.
33. Renkin EM: Exchange of substances through capillary walls. In: GEW Wolstenholme (ed),
Ciba Foundation Symposium. Little, Brown and Co, Boston, 1969, p 50.
34. Swan KG, Reynolds DG: Adrenergic mechanisms in the canine mesenteric circulation. Am J
Physiol220: 1779, 1971.
35. Messina EJ, Weiner R, Kaley G: Prostaglandins and local circulatory control. Fed Proc 35: 2367,
1976.
36. Maher JF: Blood flow to the peritoneum: physiological and pharmacological influences. In: JW
Moncrief, RP Popovich (eds), CAPD Update Continuous Ambulatory Perioneal Dialysis.
Masson Publ USA, New York, 1981, p 53.
37. Courtice FC, Roberts DCK: Peritoneal fluid in the rabbit: permeability of the mesothelium to
proteins, lipoproteins and acid hydrolases. Lymphology 8:1,1975.
38. Torres IJ, Litterst CL, Guarino AM: Transport of model compounds across the peritoneal
membrane in the rat. Pharmacology 17: 330, 1978.
39. Richardson PDI: The actions of natural secretin on the small intestinal vasculature of the
anesthetized cat. Br J Pharmacol58: 127, 1976.
40. Goldberg LI: Cardiovascular and renal actions of dopamine: potential clinical applications.
Pharmacol Rev 24: 1, 1972.
41. Henderson LW, Nolph KD: Altered permeability of the peritoneal membrane after using
hypertonic peritoneal dialysis. J Clin Invest 48: 992, 1969.
42. Harris EJ: Transport and Accumulation in Biological Systems. Third Edition. Butterworths,
London, 1972.
43. Ahearn DJ, Nolph KD: Controlled sodium removal with peritoneal dialysis. Trans Am Soc Artif
Intern Organs 18: 423, 1972.
44. Maher JF, Hohndel DC, Shea C, DiSanzo F, Cassetta M: Effects of intraperitoneal diuretics on
solute transport during hypertonic dialysis. Clin Nephrol 7: 96, 1977.
45. Brenner BM, Hostetter TH, Humes HD: Molecular basis of proteinuria of glomerular origin. N
Engl J Med 298: 826, 1978.
46. Mermier P, Baker H: Flux of free fatty acids among host tissues, ascitic fluid and Ehrlich ascites
carcinoma cells. J Lipid Res 15: 339, 1971.
47. Miller FN, Joshua IG, Harris PD, Wiegman DL, Jauchem JR: Peritoneal dialysis solutions and
the microcirculation. Contrib Nephrol17: 51, 1979.
48. Nolph KD: Peritoneal anatomy and transport physiology. In: W Drukker, FM Parsons, JF
Maher (eds) , Replacement of Renal Function by Dialysis, 2nd ed. Martinus Nijhoff, The Hague,
1983, p 440.
 293

49. Lang HL, Nolph KD, McGary TJ: Enhancement of clearances by activated charcoal in an in
vitro model of peritoneal dialysis. Clin Exp Dial Apheresis 6: 85, 1982.
50. De Santo NG, Capodicasa G, Capasso G, Giordano C: Development of means to augment
peritoneal urea clearances: The synergistic effects of combining high dialysate temperature and
high dialysate flow rates with dextrose and nitroprusside. Artif Organs 5: 409, 1981.
51. Erbe RW, Greene JA Jr, Weller JM: Peritoneal dialysis during hemorrhagic shock. J Appl
Physiol22: 131, 1967.
52. Greene JA Jr., Lapco L, Weller JM: Effect of drug therapy of hemorrhagic hypotension on
kinetics of peritoneal dialysis in the dog. Nephron 7: 178,1970.
53. Baldwin DS, Gluck MC, Lowenstein J, Gallo GR: Lupus nephritis: clinical course as related to
morphologic forms and their transitions. Am J Med 62: 12, 1977.
54. Kincaid-Smith P: Participation of intravascular coagulation in the pathogenesis of glomerular
and vascular lesions. Kidney Int 7: 242, 1975.
55. Maher JF, Hirszel P: Augmentation of peritoneal clearances by drugs. In: M Legrain (ed), Proc
Int Symposium Continuous Ambulatory Peritoneal Dialysis. Exerpta Medica, Amsterdam,
1980, p 42.
56. Maher JF, Hirszel P, Abraham JE, Galen MA, Chamberlin M, Hohnadel DC: The effect of
dipyridamole on peritoneal mass transport. Trans Am Soc Artif Intern Organs 23: 219, 1977.
57. Brown ST, Ahearn DJ, Nolph KD: Reduced peritoneal clearances in scleroderma increased by
intraperitoneal isoproterenol. Ann Intern Med 78: 891, 1973.
58. Nolph KD, Miller L, Husted FC, Hirszel P: Peritoneal clearances in scleroderma and diabetes
mellitus: effects of intraperitoneal isoproterenol. Int Urol Nephrol 8: 161, 1976.
59. Nolph KD, Ghods AJ, Brown PA, Twardowski ZJ: Effects of intraperitoneal nitroprusside on
peritoneal clearances in man with variations of dose, frequency of administration and dwell
times. Nephron 24: 114,1979.
60. Maher JF, Hirszel P, Lasrich M: An experimental model for study of pharmacologic and
hormonal influences on peritoneal dialysis. Contrib Nephrol17: 131, 1979.
61. Gutman RA, Nixon WP, McRae RL, Spencer HW: Effect of intraperitoneal and intravenous
vasoactive amines on peritoneal dialysis: study in anephric dogs. Trans Am Soc Artif Intern
Organs 22: 570, 1976.
62. Penzotti SC, Mattocks AM: Effects of dwell time, volume of dialysis fluid and added accelera-
tors on peritoneal dialysis of urea. J Pharm Sci 60: 1520, 1971.
63. Maher JF, Hirszel P, Lasrich M: The effects of gastrointestinal hormones on transport by
peritoneal dialysis. Kidney Int 16: 130, 1979.
64. Hirszel P, Lasrich M, Maher JF: Divergent effects of catecholamines on peritoneal mass
transport. Trans Am Soc Artif Intern Organs 25: 110,1979.
65. Maher JF, Hirszel P, Lasrich M: Modulation of peritoneal transport rates by prostaglandins.
Adv Prostaglandin Thromboxane Res 7: 695,1980.
66. Nolph KD, Ghods AJ, Van Stone J, Brown PA: The effects of intraperitoneal vasodilators on
peritoneal clearances. Trans Am Soc Artif Intern Organs 22: 586, 1976.
67. Maher JF, Shea C, Cassetta M, Hohnadel DC: Isoproterenol enhancement of peritoneal
permeability. J Dial 1: 319,1977.
68. Vanichayakornkul S, Nimmanit S, Chirawong P, Nilwarangkur S: Accelerated peritoneal
dialysis with intraperitoneal isoproterenol. J Med Assoc Thailand 61 (Suppl1): 127-129, 1978.
69. Maher JF,Cassetta M, Shea C, Hohnadel DC: Peritoneal dialysis in rabbits. A study of
transperitoneal theophylline flux and peritoneal permeability. Nephron 20: 18, 1978.
70. Emonds AJG, Driessen OMJ: Treatment of theophylline intoxication: a model study ultilizing
peritoneal dialysis. Clin Toxicol13: 505, 1978.
71. Brown EA, Kliger AS, Goffinet J; Finkelstein FO: Effect of hypertonic dialysate and vasodila-
tors on peritoneal dialysis clearances in the rat. Kidney Int 13: 271, 1978.
294

72. Nolph K, Ghods A, Brown P, Miller F, Harris P, Pyle K, Popovich: Effects of nitroprusside on
peritoneal mass transfer coefficients and microvascular physiology. Trans Am Soc Artif Intern
Organs 23: 210, 1977.
73. Hirszel P, Maher JF, Chamberlin M: Augmented peritoneal mass transport with intraperitoneal
nitroprusside. J Dial 2: 131, 1978.
74. Raja RM, Kramer MS, Rosenbaum JL: Enhanced clearance with intraperitoneal nitroprusside
in high flow recirculation peritoneal dialysis. Trans Am Soc Artif Intern Organs 24: 133,1978.
75. Nolph KD, Ghods AJ, Brown PA, Twardowski ZJ: Effects of intraperitoneal nitroprusside on
peritoneal clearances in man with variations of dose frequency of administration and dwell
times. Nephron 24:114, 1979.
76. Riembau E, Lloveras J, Aubia J, Masramon J, Garcia C, Orfila MA, Llorach M: Sequential
intraperitoneal administration of nitroprusside in patients maintained on peritoneal dialysis. J
Dial 4: 203,1980.
77. Miller FN, Nolph KD, Harris PD, Rubin J, Wiegman DL, Joshua IG, Twardowski ZJ, Ghods
AJ: Microvascular and clinical effects of altered peritoneal dialysis solutions. Kidney Int 15: 630,
1979.
78. Danzig LE: Dynamics of thiocyanate dialysis. The artifical kidney in the therapy of thiocyanate
intoxication. N Engl J Med 252: 49, 1955.
79. Sano N, Satoh S, Hashimoto K: Differences among dipyridamole, carbochromen and Iidoflazine
in responses of the coronary and the renal arteries. Jap J Pharmacol22: 857, 1972.
80. Ryckelynck JP, Pierre D, DeMartin A, Rottembourg J: Amelioration des clairances perito-
neales par Ie dipyridamole. Nouv Presse Med 7: 472, 1978.
8l. Limido A, Cantu P, Allaria P, Colombo L, Giangrande G: VeIocita di flusso ed effetto dei
farmaci nella valutazione dell efficienza della dialisi peritoneale. Minerva Nephrol26: 161,1979.
82. Rubin J, Adair C, Barnes T, Bower JD: Augmentation of peritoneal clearance by dipyridamole.
Kidney Int 22: 658, 1982.
83. Maher JF, Hirszel P: Augmenting peritoneal mass transport. Int J Artif Organs 2: 59, 1979.
84. Parker HR, Schroeder JP, Henderson LW: Influence of dopamine and regitine on peritoneal
dialysis in unanesthetized dogs. Abstracts Am Soc Artif Intern Organs 7: 43, 1978.
85. Chan MK, Varghese Z, Baillod RA, MoorheadJF: Peritoneal dialysis: effect of intraperitoneal
dopamine. Dial Transpl 9:380, 1980.
86. Hirszel P, Lasrich M, Maher JF: Augmentation of peritoneal mass transport by dopamine.
Comparison with norepinephrine and evaluation of pharmacologic mechanisms. J Lab Clin Med
94: 747,1979.
87. Maher JF: Principles of dialysis and dialysis of drugs. Am J Med 62: 475, 1977.
88. Pace Asciak CR: Oxidative biotransformation of arachidonic acid. Prostaglandins 13: 811,1977.
89. Nakano J, McCurdy JR: Hemodynamic effects of prostaglandins E], AI and Fz in dogs. Proc Soc
Exp Bioi Med 128: 39, 1968.
90. Messina EJ, Kaley G: Microcirculatory responses to prostacyclin and PGEz in the rat cremaster
muscle. Adv Prostaglandin Thromboxane Res 7: 719, 1980.
91. Vane JR, McGiff JC: Possible contributions of endogenous prostaglandins to the control of
blood pressure. Circ Res 36, 37 (Suppll): 68, 1975.
92. Maher JF, Hirszel P, Lasrich M: Prostaglandin effects on peritoneal transport. Proc 2nd Int
Symposium on Peritoneal Dial 2: 65, 1981.
93. Mileti M, Bufano G, Scaravonati P, Pecchini F, Carnevale G, Lanzarini P: Effect of indom-
ethacin on the peritoneum of rabbits on peritoneal dialysis. Perit Dial Bull 3: 194, 1983.
94. Strong CG, Romero JC: Effects of indomethacin in rabbit renovascular hypertension. Clin Sci
Mol Med 51: 249s, 1976.
95. Kimberly RP, Bowden RE, Keiser HI, Plotz PH: Reduction of renal function by newer
nonsteroidal anti-inflammatory drugs. Am J Med 64: 804, 1978.
96. HirszeI P, Lasrich M, Maher JF: Arachidonic acid increases peritoneal clearances. Trans Am
Soc Artif Intern Organs 27: 61,1981.
 295

97. Rayford PL, Miller TA, Thompson JC: Secretin, cholecystokinin and newer gastrointestinal
hormones. N Engl J Med 294: 1093, 1976.
98. Thulin L, Samnegard H: Circulatory effects of gastrointestinal hormones and related peptides.
Acta Chir Scand (Suppl) 482: 73, 1978.
99. Fara JW, Rubenstein EH, Sonneschein RR: Intestinal hormones in mesenteric vasodilation
after intraduodenal agents. Am J Physiol223: 1058, 1972.
100. Farah JW: Effects of gastrointestinal hormones on vascular smooth muscle. Am J Dig Dis 20:
346,1975.
101. Hirszel P, Maher JF, LeGrow W: Increased peritoneal mass transport with glucagon acting at
the vascular surface. Trans Am Soc Artif Intern Organs 24: 136, 1978.
102. Farini R, Del Favero G, Adorati M, Perdrazzoli S, Fabris G, Giordano P, D'Angelo A, Zotti E,
Lise M, Chiaramonte M, Salvagnini M, Naccarato R: Comparison between bolus injection and
infusion of Secretin and Pancreozymin in the diagnosis of chronic pancreatic diasease (one-hour
test). Acta Hepatogastroenterol (Stuttg) 24: 462, 1977.
103. Biber B, Farah J, Lundgren 0: Vascular reactions in the small intestine during vasodilation.
Acta Physiol Scand 89: 499, 1973.
104. Hare HG, Valtin H, Gosselin RE: Effect of drugs on peritoneal dialysis in the dog. J Pharmacol
Exp Ther 145: 122, 1964.
105. Henderson LW, Kintzel JE: Influence of antidiuretic hormone on peritoneal area and per-
meability. J Clin Invest 50: 2437, 1971.
106. Shear L, Harvey JD, Barry KG: Peritoneal sodium transport: enhancement by pharmacologic
and physical agents. J Lab Clin Med 67: 181, 1966.
107. Maher JF, Hirszel P, LeGrow W: Enhanced peritoneal permeability with methyl prednisolone.
Clin Res 26: 64A, 1978.
108. Breborowicz A, Knapowski J: Augmentation of peritoneal dialysis clearance with procaine.
Kidney Int (in press).
109. Haddy F: The mechanism of potassium vasodilation. In: PM Vanhoutte, I Leusen (eds),
Mechanisms of Vasodilation. S Karger, Basel, 1978, p 200.
110. Miller FN, Nolph KD, Joshua IG, Wiegman DL, Harris PD, Anderson DB: Hyperosmolality,
acetate and lactage: Dilatory factors during peritoneal dialysis .. Kidney Int 20: 397, 1981.
111. Penzotti SC, Mattocks AM: Acceleration of peritoneal dialysis by surface-active agents. J Pharm
Sci 57: 1192, 1968.
112. El-Bassiouni EA, Mattocks AM: Acceleration of peritoneal dialysis with minimal N:myristyl-,B-
amino-proprionate. J Pharm Sci 62: 1314, 1973.
113. Dunham CB, Hak LJ, Hull JH, Mattocks AM: Enhancement of peritoneal permeability of the
rat by intraperitoneal use of docusate sodium. Kidney Int 20: 563, 1981.
114. Maher JF, Chakrabarti E: Ultrafiltration by hyperosmotic peritoneal dialysis fluid excludes
intracellular solutes. Am J Nephrol4: 169, 1984.
115. Raja RM, Kramer MS, Rosenbaum JL, Bolisay C, Krug M: Contrasting changes in solute
transport and ultrafiltration with peritonitis in CAPD patients. Trans Am Soc Artif Intern
Organs 27: 68, 1981.
116. Scarpioni L, Ballocchi S, Bergonzi G, Fontana F, Poisetti P, Zanazzi MA: High-dose diuretics in
CAPD. Perit Dial Bull 2: 177,1982.
117. Grzegorzewska A, Baczyk K: Furosemide-induced increase in urinary and peritoneal excretion
of uric acid during peritoneal dialysis in patients with chronic uremia. Artif Organs 6: 220, 1982.
118. Alavi H, Lianos E, Bentzel C: Enhanced peritoneal permeability of the rat by intraperitoneal
use of protamine sulfate. Kidney Int 16: 880,1979.
119. Nolph KD, Nothum RJ, Maher JF: Ultrafiltration: a mechanism for removal of intermediate
molecular weight substances in coil dialyzers. Kidney Int 6: 55,1974.
120. Avasthi PS: Effects of aminonucleoside on rat blood-peritoneal barrier permeability. J Lab Clin
Med 94: 295, 1979.
296

121. Hirszel P, Bennett RR, Chakrabarti E, Maher JF: Amphotericin B (A) augments peritoneal
water flux. Abstracts Am Soc Artif Intern Organs 13: 43, 1984.
122. Andreoli TE, Dennis VW, Weigl AM: The effect of amphotericin B on the water and non-
electrolyte permeability of thin lipid membranes. J Gen Physiol53: 133, 1969.
123. Kobayashi K, Manji T, Hiramatsu S, Maeda K, Uemura J: Nitrogen metabolism in patients on
peritoneal dialysis. Contrib Neprol17: 93, 1979.
124. Gjessing J: The use of dextran as a dialysing fluid in peritoneal dialysis. Acta Med Scand 185: 237,
1969.
125. Rubin J, Nolph K, McGary T: Osmotic ultrafiltration with dextran sodium sulfate. Potential for
use in peritoneal dialysis. J Dial 3: 251, 1979.
126. Bazzato G, Coli U, Landini S, Fracasso A, Morachiello P, Righetto F, Scanferla F, Onesti G:
Xylitol as osmotic agent in CAPD: An alternative to glucose for uremic diabetic patients? Trans
Am Soc Artif Intern Organs 28: 280, 1982.
127. Zelman A, Gisser D, Whittam PJ, Parsons DH, Schuyler R: Augmentation of peritoneal dialysis
efficiency with programmed hyper/hypoosmotic dialysates. Trans Am Soc Artif Intern Organs
23: 203, 1977.
128. Knochel JP, Clayton LE, Smith WL, Barry KG: Intraperitoneal THAM: an effective method to
enhance phenobarbital removal during peritoneal dialysis. J Lab Clin Med 64: 257, 1964.
129. Nahas GG, Giroux JJ, Gjessing J, Verosky M, Mark LC: The use of THAM in peritoneal
dialysis. Trans Am Soc Artif Intern Organs 10: 345, 1964.
130. Knochel JP, Masson AD: Effect of alkalinization on peritoneal diffusion of uric acid. Am J
Physiol 210: 1160, 1966.
131. Campion DS, North JDK: Effect of protein binding of barbiturates on their rate of removal
during peritoneal dialysis. J Lab Clin Med 66: 549, 1965.
132. Kudla RM, El-Bassiouni EA, Mattocks AM: Accelerated peritoneal dialysis of barbiturates,
diphenylhydantoin and salicylate. J Pharm Sci 60: 1065, 1971.
133. Schultz JC, Crouder DG, Medart WS: Excretion studies in ethchlorvynol (PI acidyl) intoxica-
tion. Arch Intern Med 117: 409, 1966.
134. Etteldorf IN, Dobbins WT, Summit RL, Rainwater WT, Fischer RL: Intermittent peritoneal
dialysis using 5% albumin in the treatment of salicylate intoxication in children. J Pediatr 58: 226,
1961.
135. Cole DEC, Lirenman OS: Role of albumin enriched peritoneal dialysate in acute copper
poisoning. J Pediatr 92: 955, 1978.
136. Covey TJ: Ferrous sulfate poisoning: a review, case summaries and therapeutic regimen. J
Pediatr 64: 218, 1964.
137. Mehbod H: Treatment of lead intoxication. Combined used of peritoneal dialysis and edetate
calcium disodium. JAMA 201: 972,1967.
138. Lowenthal DT, Chardo F, Reidenberg MM: Removal of mercury by peritoneal dialysis. Arch
Intern Med 134: 139, 1974.
139. Falk RJ, Mattern WD, Lamanna RW, Gitelman HJ, Parker NC, Cross RE, Rastall JR: Iron
removal during continuous ambulatory peritoneal dialysis using desferrioxamine. Kidney Int 24:
110,1983.
140. Shinaberger JH, Shear L, Clayton LE, Barry KG, Knowlton M, Goldbaum LR: Dialysis for
intoxication with lipid soluble drugs: enhancement of glutethimide extraction with lipid dialy-
sate. Trans Am Soc Artif Intern Organs 11: 173, 1965.
141. Maher JF: Characteristics of peritoneal transport: physiological and clinical implications. Min-
eral Electrolyte Metab 5: 201, 1981.
11. Comments on dialysis solution, antibiotic
transport, poisoning, and novel uses of
peritoneal dialysis

JACK RUBIN

1. Introduction

Peritoneal fluid absorption and removal has been a source of fascination for
physiologists and clinicians for years [1]. Cunningham's review of peritoneal
physiology in 1926 touched on still current issues [2].
Classic studies with phenosulfonpthalein administered into the peritoneal
cavity demonstrated that the drug appeared in peripheral blood in two to four
minutes. Uptake from the peritoneal cavity was primarily via the circulatory
system [3]. Larger particulate matter, was shown to be absorbed via the lym-
phatics.
Fluid instilled into the peritoneal cavity was found to be absorbed when it was
hypotonic and to induce ultrafiltration when hypertonic [4-12].
Studies of the rate of equilibration between plasma and dialysate concentration
have demonstrated the effects of altering contact area between dialysate and
peritoneum [13-15].
The peritoneal membranes have differing permeabilities in different areas
(visceral, parietal, mesentery) and are influenced by metabolic processes [16-21].
Peritoneal membranes surrounding the intestine and the omentum are likely the
most important sites for transfer [22].
Alterations in Starling forces influence peritoneal events [23-24]. It is known
that the formation of ascites in patients with liver disease is dependent on colloid
osmotic pressure [25]. Maher has suggested that hormonal manipulation, possi-
bly by influencing hydrostatic forces and/or vascular surface area, may alter water
flow into the peritoneal cavity [26] ..
Peritoneal permeability may be altered by alternating between hypo and
hypertonic dialysate, local temperature, vasoactive agents [27-32] and surface
active agents [33-39]. Whether this is due to changes in the peritoneal membrane
itself, metabolic processes within cells, peritoneal blood flow, stagnant fluid
films, or a combination of events, is unknown [38-41].
Systemic protein binding of a solute may retard transfer into the peritoneal
298

dialysate. Mattocks suggested that anthranilic acid (1%) and N-myristyl-beta-

aminopropionic acid (0.25%) accelerated salicylate removal by altering binding
of salicylate to plasma proteins [42, 43].
The peritoneum is more permeable to non-ionized species. Non-ionized or
lipid soluble compounds are absorbed to a greater extent than ionized or lipid
insoluble compounds. Neutral compounds are absorbed in relation to their
molecular weight [44, 45].
Dialysis solutions have active effects upon vascular beds and likely metabolic
effects upon peritoneal cells and should not be conceptualized as inert 'bathing
solutions' of the peritoneal membrane.
Miller et al. showed that unaltered dialysis solutions caused transient vas-
oconstriction followed by vasodilation of the parietal microvasculature [46-49].
Clinical studies with pH-adjusted dialysate failed to show any improvement in
peritoneal clearances [50]. In their model, osmolality, lactate and acetate contrib-
uted to vasodilation but glucose concentration did not. Study of the mesothelial
wall of the rat cecum showed that dialysis solutions produced maximal dilation of
small arterioles but less than half maximal dilation of the venules [50].
Venular dilation may be an important determinant of dialysate protein losses.
Studies by Miller using dialysis solutions which differed in osmolality, pH, and/or
buffer anion were carried out to determine whether any of these factors influ-
enced protein loss into dialysate. They showed that the protein found in dialysate
effluent was influenced by changes in two or more of these factors but that
changes in one variable only cannot produce these effects [51].
Other structure/function relationships likely will soon emerge. For example
Mileti, working with rabbits, has shown that indomethacin administered in
dialysate is associated with an increase in size of the pinocytic vesicles of the
endothelial cells and a decrease in intercellular spaces [52]. Hirszel, also working
with rabbits, has shown that arachidonic acid increases peritoneal clearances
[53].
Peritoneal absorption or removal is dependent upon an instilled dialysis solu-
tion. This chapter will discuss dialysis solutions, the use of antibiotics during
peritoneal dialysis, the role of peritoneal dialysis in poisoning and some novel
applications of peritoneal dialysis.
Reviews of drug metabolism and kinetics are available to the interested reader
and will not be discussed [54-60].

2. Dialysis solutions

Drs Ward, Klein, and Watham, submitted their report on The Investigation of
Risks and Hazards with Devices Associated with Peritoneal Dialysis. Table 1,
taken from their. report, summarizes the risks associated with peritoneal dialysis
solutions [61].
 299

Table 1. Summary of risks and hazards associated with peritoneal dialysis solutions

Event Risk (R) Source Underlying cause Clinical

or hazard of infor- consequences
(H) a mation

Bacterial H 62-63 Contaminated water bath Peritonitis

contamination of for heating bottles; Pin
dialysate hole leaks in dialysate
bags; contamination during
spiking of dialysate
container
2. Impurities in dialysate R,H 64-66 Inadequate quality Pain, peritonitis,
(particulate matter, assurance, incomplete sclerosis of
endotoxin rinsing of on-line peritoneum, loss of
formaldehyde) proportioning machines clearance
after disinfection
3. Trace chemicals in R 67-69 Inadequate water Clinical significance
water used for dialysate purification unknown
preparation
4. Hypertriglyceridemia R 70-72 Use of glucose as an Long-term effects
osmotic agent on patient
morbidity are
unknown
5. Hyperosmolarity and R,H 73-75 Use of glucose as an Hyperosmolar
hyperglycemia osmotic agent coma (with 70 gil
glucose) high
glucose loads in
diabetes requiring
additional insulin
therapy,
hypovolemia and
hypotension
6. Hypernatremia R 76-79 Formation of hypotonic Hypertension,
ultrafiltrate, high sodium cerebral edema
concentration in dialysate
7. Acidic dialysate R,H 80 Need to keep pH in range Pain, sclerosis of
5.0-5.5 to avoid glucose peritoneum, loss of
caramelization during heat clearance
sterilization
8. Hyperosmolarity H 81-82 Use of sorbitol as an Hyperosmolar
osmotic agent coma, death
9. Use of glass bottles for R, H 72,83,84 Weight, bulk and Increased
dialysate brittleness of bottle peritonitis in
continuous
ambulatory
peritoneal dialysis
from increased
connect -disconnect,
over balancing of
carousels, bottle
breakage
300

10. Use of incorrect R 83,85 Inadequate quality Poor control of

dialysate composition assurance, operator error fluid balance,
hypovolemia,
hypernatremia
11. Abdominal pain R,H 80,86 Instillation of dialysate Pain, discomfort,
under pressure, prolonged buffer loss by
use of hypertonic patient
dialysates, acidic dialysates
(see 7 above)
12. Hepatitis positive H 87-94 Treatment of hepatitis B Hepatitis outbreak
dialysate positive patients amongst staff

aA hazard is defined as an actual occurrence detrimental to patient safety, while a risk is an event
which has not occurred but which is considered likely to occur and to have detrimental clinical
consequences.

2.1. Electrolytes and glucose

Peritoneal dialysis solutions are formulated with electrolytes including: sodium,

chloride, calcium and magnesium, either acetate or lactate as potential bicarbo-
nate, and an osmotically active agent, usually glucose. Solutions are supplied
premixed or in concentrate of 30% and 50% glucose for use with reverse-osmosis
dialysis machines.
The ideal dialysis solution is not known and it is perhaps unrealistic to expect
one solution for all patients. Reviews by Gault and Tenckhoff on fluid and
electrolyte complications of peritoneal dialysis point out the potential complica-
tions with clinical use of these solutions; hypovolemia, hypernatremia, alkalosis,
hyperglycemia. Gault makes a plea for more formulations, whereas Tenckhoff
suggests that less choice may be safer [92-94].
Recently, attempts have been made to modify the peritoneal dialysis solutions
administered to patients. Modifications to the solutions are made by the manufac-
turers secondary to a perceived 'problem' by their consultants.

2.2. Magnesium

Perhaps the best example of this, is the marketing of a 'low magnesium' solution
(0.5 mEq/L) [95-98] instead of standard 1.5 mEq/L. The stimulus for altering the
solution, came from the observation that patients undergoing CAPD, commonly
evidenced serum magnesium concentrations that were above normal.
This author is not aware of any clinical sequelae to slightly elevated serum
magnesium concentrations in the peripheral blood. Magnesium is an important
cation involved in several enzymatic reactions. In the laboratory, it is almost
impossible to show abnormalities related to modestly elevated magnesium con-
centrations, but one can readily demonstrate abnormalities in the presence of
 301

lowered serum magnesium. Most patients on the lower magnesium solution for
CAPD show serum values in the normal range, but the potential for low values
seems greater in a poor eater.

2.3. Calcium

The ideal calcium concentration in dialysate is unknown. The aim of the manufac-
turers appears to be net absorption. Dr Oreopoulos has suggested that patients
maintained on continuous ambulatory peritoneal dialysis with an infused calcium
concentration of 3 mEq/L may be in negative dialysate calcium balance [99].
Studies with a dialysate calcium concentration of 3.5 mEq/L during continuous
ambulatory peritoneal dialysis have generally shown a positive calcium balance
[100].
Garrett showed that calcium absorption is dependent on at least three factors:
ultrafiltration, concentration of protein in the peritoneal cavity, and serum
ionized calcium [101]. Ultrafiltration will dilute dialysate and diminish the concen-
tration gradient for absorption, or even cause a reversal of the diffusion gradient.
Although protein concentrations are low in dialysate effluent, it is possible that
intraperitoneal protein binds a small amount of calcium. The level of ionized
calcium in dialysate and plasma are likely the major determinants of absorption.
The rate of ionized calcium absorption in general parallels the rate of total
calcium absorption [102-103].
In the post-operative care of hypocalcemic parathyroidectomized CAPD pa-
tients we have found it occasionally useful to add an ampule of calcium chloride to
the dialysate.
Calcium-free dialysis solutions have been used to remove calcium from patients
with hypercalcemia. The removal of calcium is greater with hypertonic ex-
changes. Peritoneal clearance has been reported to be 9.6 and 17 mllmin for total
and ionic calcium respectively for a 1.5% dextrose solution, and 12 and 24 ml/min
for a 7% dextrose solution. Clearances of calcium varied inversely with serum
protein concentration. Miach pointed out that a calcium-free dialysate of dex-
trose and saline was effective in removing calcium, but that the absence of
bicarbonate led to acidosis [104-107].

2.4. Sodium

Sodium has been added to dialysate in varying concentrations ranging from 120-
140 mEq/L. Hypernatremia may occasionally occur because the sodium concen-
tration in the ultrafiltrate approximates 70 mEq/L [77-79]. Raja suggested that
the increase in serum sodium concentrations could be avoided, and an isonatric
ultrafiltrate achieved with a dialysate sodium concentration of 115-120 mEq/L in
302

7% dextrose dialysate, and 125-130 mEq/L for a 4.25% dextrose dialysate [108-
109].
Delivered sodium concentration can vary, depending upon the manufacturing
process. If a 5% variation is possible, then a stated sodium content of 140 mEq/L
can vary between 135-148 mEq/L [110, 111]. Swales pointed out that it is possible
to induce net absorption of sodium in hyponatric patients.
Sodium transport across rabbit omentum can be modified by pharmacologic
and physical agents [21]. Nolph has suggested that sodium losses from patients
undergoing continuous ambulatory peritoneal dialysis are auto-regulated by
changes in serum sodium concentration [112].

2.5. Potassium

Potassium chloride is commonly added to dialysate to prevent potassium deple-

tion with peritoneal dialysis. Spital showed that 20 meq/L could be added to
dialysate [113]. At this dose one could expect a net absorption of 15 meq; 12 meq
may be absorbed during the first 15 minutes. If one added a greater concentration,
peritoneal irritation occurred [113].
Potassium is cleared via peritoneal dialysis but at maximum rates only near 116
and 113 of those possible with hemodialysis and resin exchange enemas [114].
Studies of potassium diffusion show that serum and dialysate do not reach
equilibration during long intraperitoneal dwells [12]. It has been suggested that,
since potassium may leach from red cells, serum potassium concentrations may be
spuriously elevated, thus accounting for this observation.

2.6. Lactate acetate and bicarbonate

There are occasional patients unable to maintain a serum bicarbonate concentra-

tion at acceptable levels. A formulation that delivers 45 mEq/L of acetate or
lactate to a patient undergoing intermittent peritoneal dialysis may be excessive
for some patients undergoing continuous ambulatory peritoneal dialysis, leading
to elevated serum bicarbonate concentrations, whereas 32 mEq/L may not be
sufficient for the patient undergoing intermittent dialysis.
Lactate and acetate are similar in their ability to repair acidosis. They are
similar in their effects on clinical clearances during peritoneal dialysis [115, 116].
It was originally felt that acetate solutions may offer some advantage because
of antibacterial activity of unused dialysate. Differences between lactate or
acetate solutions disappear when dialysate is studied following instillation into
the peritoneal cavity [117-120].
Dixon has shown that lactate is absorbed at a rate of 25 mEql hour throughout a
24-hour dialysis [121]. He points out that net absorption of base equivalent must
 303

include consideration of bicarbonate lost into dialysate. The endogenous cleara-

nce of lactate in this setting is approximately 9 ml/min [122-129].
In acutely ill patients, hyperlacticacidemia has been reported secondary to an
inability to metabolize the lactate ion. Lactate used in dialysate is a racemic
mixture of dextro and levo forms. The fate of the dextro isomer is not well
understood. The levo-isomer disappears from dialysate more slowly than the
dextro-isomer [130].
It has been suggested that patients with lactic acidosis requiring peritoneal
dialysis be treated with dialysate containing either acetate or bicarbonate [131,
132].
Bicarbonate dialysate would be ideal, and can readily be prepared from
available intravenous solutions. If one has a dialysate cycling device such as
manufactured by American Medical Products one can readily prepare a bicarbo-
nate solution. For example a solution containing 131 mEq/L sodium, 28 mEq/L
bicarbonate and 1.5% dextrose can be made by hanging 400 ml of 50% dextrose, 4
liters of normal saline and 3 liters of half normal saline. One would then add 45
meq sodium bicarbonate (available commercially in 50 ml containers) to five of
the liter containers. To make a 5% solution one could use 5% dextrose and saline.
The potential modifications are obvious. Calcium and magnesium supplementa-
tion intravenously may be necessary if such solutions are used over many ex-
changes.
Sterilization difficulties and the possibility of precipitation with calcium in
dialysate have precluded the routine use of bicarbonate.
Ing has described a method of preparing an 'on-line' bicarbonate solution
[133].
Dialysis with an iso-osmotic bicarbonate solution resulted in enhanced per-
itoneal protein losses and an increase in inulin clearance relative to urea clearance
[134]. Gjessing has suggested that an alkaline dialysis solution may have antibac-
terial properties [135, 136].
It is possible to induce severe metabolic alkalosis during peritoneal dialysis by
maintaining a patient on nasogastric suction and administering peritoneal di-
alysis. Vilbar has treated metabolic alkalosis with chloride-rich dialysate [137].

2.7. Osmotically active agents

Glucose has been the main agent used to generate ultrafiltrate. There has been a
lot of thought given to modifying the osmotically active agent in dialysate [138,
139].
The major objections to glucose have been:
1) In some patients glucose does not seem to induce ultrafiltration.
2) It provides unwanted calories and many patients on CAPD gain weight.
3) It may be associated with lipid abnormalities predisposing the patient to
cardiovascular disease [140].
304

4) The long-term effect of chronic glucose infusion are unknown [141].

5) There are occasional diabetic patients placed on CAPD who develop eleva-
tions in their blood sugar which were not evident on maintenance hemodialysis,
or intermittent peritoneal dialysis.
7) Dialysate solutions are formulated at an acid pH to prevent caramelization
of glucose. Numerous observations have related this acid pH to patient discom-
fort upon inflow of dialysate.
8) Acidic dialysate may impair the activity of peripheral blood leukocytes
'called' to the peritoneal cavity [142].
9) Acidic dialysate may inhibit aluminium removal from dialysate [143].
We submit that these objections are for the most part minor. For example
obesity may not be related to CAPD. Further history from patients who appear to
be getting fat often reveals that the patients are still returning to their non-uremic
weight [144]. Furthermore, the calories that the host receives from the glucose,
may be important for patients with inadequate food intake [145, 146].
Continuous glucose infusion may induce pancreatic insufficiency in those
predisposed to diabetes or contribute to hyperlipidemia and obesity. Studies
during hourly exchanges did not show any correlation between the rate of glucose
absorption and serum insulin levels, whereas, during long-dwell exchanges, the
pattern of insulin and glucagon secretion was found to be similar to normal
subjects [147, 148].
More basic work on the long-term effects of glucose containing dialysis solu-
tions is required.
Cunningham showed that intraperitoneal glucose induced morphologic
changes in the mesothelial lining cells of the peritoneal cavity of rat [4]. Peritoneal
membrane thickening occurs in dogs fed hypertonic solutions for a period of one
month. The peritoneal membrane returns to normal when the feedings are
stopped [149].
The range of glucose absorption from dialysate with 2L hourly exchanges for
1.5% and 4.25% glucose concentration is 5-10 g and 35-40 g/h, however, there is
wide variation [147, 150]. Studies of the change in intraperitoneal dialysate
glucose concentration over time have shown an initial rapid decrease followed by
a slower fall in concentration. The rapid decrement is due to absorption and
dilution. The ultrafiltrate generated in response to glucose is hypotonic [12, 151].
Hypertonic dialysis solutions are useful to remove excessive fluid from pa-
tients. They have been associated clinically with hyperglycemia, hyperosmolality
and excessive ultrafiltration leading to volume contraction with hypotension and
in one case report, delayed hypoglycemia [152, 153].
Other sugars have been used for peritoneal dialysis including fructose, which
may be useful for diabetics, xylitol, and sorbitol [82,154-164].
 305

2.8. Non-glucose osmotic agents

Sorbitol was initially considered because it did not caramelize during heat steril-
ization. In one study (not confirmed), it was suggested that ultrafiltration was
greater with sorbitol dialysate when compared to an equal concentration of
glucose dialysate. Unfortunately, hyperosmolar coma has occurred with concen-
trations of sorbitol greater than 20 giL dialysate. This may be due to persistence of
sorbitol leading to high concentrations in the blood.

2.9. Amino acids

The addition of amino acids to dialysate has been suggested as a mechanism to

prevent amino acid losses and as a nutritional supplement. It is not clear whether
this will be a useful approach to therapy.
Amino acid losses in dialysate are small (two grams daily); amino acid solutions
may be used as a dialysate solution and are capable of generating a useful
ultrafiltrate; amino acids are absorbed from the peritoneal cavity; and chronic
therapy with two liters of a 1% amino acid glucose solution alternating with a
glucose solution failed to improve total body potassium. It is possible that
incorrect amino acid combinations could lead to amino acid toxicity.
Jackson and Giordano, on the other hand felt that this therapy was able to
improve the nutritional health of patients. It is interesting to note that Giordano
failed to detect any significant lipid absorption from the peritoneal cavity [165-
176].

2.10. Polymers

Some work is being carried out on developing polymers as potential osmotically

active agents for use in peritoneal dialysis. These agents could be useful for
patients undergoing CAPD that fail to obtain adequate dialysate effluent returns.
They may potentially decrease the caloric intake to the extent that these polymers
are not absorbed during the period of intraperitoneal residence of dialysate. To
date there is no commercially available non toxic polymer [177,178].
Gjessing attempted to use 6% dextran in saline, but found it did not have
adequate ultrafiltration capabilities and dextran accumulated in the patient's
serum [179]. Jirka showed that protein losses in dialysate effluent were less with
dextran containing dialysate when compared to glucose solutions, although, this
could have been related to decreased dialysate volumes [180].
Fine, in early work, used gelatin to obtain ultrafiltration, and glycerine has
been used in animal work [1, 181, 182]. We were unable to obtain studies
characterizing the effects of gelatin or glycerine on the peritoneal dialysis system.
306

A useful modification of dialysate would be if the dialysate could be gifted with

antibacterial properties. Protamine administered with dialysate, may be such an
anti-bacterial agent. If it could be added only directly to the fluid in manufacture,
this would have the advantage of simply combatting one of the major problems of
peritoneal dialysis. It is impossible to know for sure whether this will help since
prophylactic antibiotics have generally not been found helpful in peritoneal
dialysis [183, 184].

2.11. Contaminants

Stewart showed that although dialysate packaged in plastic bags was sterile, the
potential space between the dialysate container and the plastic envelope may not
be sterile [136]. Moisture may accumulate between the bags, and, if there is a
defect in the manufacture of the plastic, contamination of dialysate could occur.
Abrutyn showed that an outbreak of Acinetobacter peritonitis using bottled
dialysate could be attributed to contaminated warming bath fluid spilling over the
administration ports [62].
Lasker and Verger showed that although particulate matter is administered
with peritoneal dialysis solutions, it did not produce peritoneal irritation [66,
185].
During CAPD solutions are commonly exchanged via the same set of plastic
tubing. Supposedly plasticizers are not leached by aqueous solutions.
Bisulfite is present in many dialysis solutions and is known to be a toxic
substance in high concentrations. It is used to prevent discoloration of dextrose
during autoclaving. It appears that autoclaving dialysis fluid reduces the bisulfite
concentration below toxic levels [81, 186, 187].
Aseptic peritonitis has occurred secondary to endotoxins released during steril-
ization of contaminated dialysate [188].

3. Antibiotics

Peritonitis occurring during peritoneal dialysis is clinically different from that

occurring following a surgical insult. Patients undergoing dialysis have suffered
some external contamination leading to infection. Dialysis therapy is used for
treatment of infection. The instillation and drainage of dialysis solution during
peritonitis removes organisms, provides a ready means of administering antibio-
tic, and may prevent adhesions secondary to infection [189]. It also removes host
defenses mobilized due to infection and spreads the organisms over the peritoneal
cavity. This may be beneficial to some extent, since removal of the mediators of
inflammation may prevent clinical toxicity, and the exposure to a larger per-
itoneal surface area may allow an increased area for phagocytosis [190]. Effective
antibiotic therapy must not only provide adequate serum and tissue levels, but
 307

must be able to eradicate organisms that may persist in stagnant residual pools of
dialysate within the peritoneal cavity.
Different regimens for antibiotic therapy are emerging. Some authors recom-
mend loading doses be given either intraperitoneally or by other parenteral
routes. Bunke has used modeling to predict plasma levels whereas Halstenson
has suggested an intermittent schedule for intraperitoneal administration of
tobramycin [191-192].

3.1. Compatability with dialysate solutions

It is possible to administer numerous substances to the patient within dialysate.

Compatability tables are available for dialysis solutions manufactured by Abbott
Laboratories, and these should be applicable to other solutions.
The clinical studies reported have usually added one antibiotic to dialysate just
prior to instillation. Although it is convenient to add antibiotics to the peritoneal
cavity via dialysate, there are few studies documenting whether antibiotics main-
tain efficacy within the dialysate.
When tested after proportioning, antibiotics added to 30% concentrate used in
reverse-osmosis dialysis delivery machines are inhibited, compared to their ac-
tivity in saline [193].
Guttman has shown that gentamicin, nafcillin, cephalothin and penicillin
added to commercially available concentrate for the reverse-osmosis dialysis
machine lose activity over time. Gentamicin retains 80% activity, nafcillin 70%
activity, cephalothin and penicillin, less than 50% activity after 24 hours [194].
Antibiotics are generally stable in dialysate containing 1.5% glucose [195, 196].
Sewell suggests that dialysate-drug solutions may be prepared everyone to two
days. In his studies the bioactivity of cefotaxime, naficillin, and vancomycin
declined 15-20% after 48 hours. Neither the addition of heparin nor insulin
altered antibiotic activity [197, 198].
Diskin suggests that dialysate effluent inhibits the antimicrobial activity of
antibiotics added to dialysate [120]. These studies await confirmation.
In general, study protocols related to antibiotics have been designed to inves-
tigate half-life of the antibiotics, either during dialysis, or off of dialysis, penetra-
tion of an antibiotic into dialysate after parenteral administration, or uptake from
the peritoneal cavity following local instillation.
Administration of antibiotics to the peritoneal cavity has the potential advan-
tages of high local concentration and the ability to maintain continuous antibiotic
levels without submitting the patient to intravenous or intramuscular administra-
tion. Most reports have documented therapeutic efficacy using this technique.
Failure to respond to therapy may occasionally be caused by the chronic dialysis
catheter or a small amount of fibrin that protects and harbors the bacteria [62,
199] acting as a foreign body.
308

Antibiotics are effective if they can be delivered to the site of infection in

appropriate concentrations. In treating peritonitis occurring during peritoneal
dialysis, it may not be sufficient to deliver high local drug concentrations. The
serum concentration, although an imperfect index of tissue penetration, may also
need to be adequate.
Study protocols by Atkins and Nielson are useful in that they emulate clinical
practice during intermittent peritoneal dialysis [200, 201].
Since patients undergoing CAPO frequently are able to treat themselves at
home treatment protocols for patients undergoing continuous ambulatory per-
itoneal dialysis will differ from those for intermittent peritoneal dialysis.
Clinical studies have demonstrated good results with many antibiotics without
adverse effects, and until more studies are carried out, these could act as useful
guidelines [202-206]. Tables 2 and 3 summarize reasonable therapeutic recom-
mendations.

3.2. Heparin

Heparin is commonly added to peritoneal dialysate to prevent catheter plugging

by proteinaceous deposits during routine dialysis and treatment of peritonitis.
Use of heparin has been felt to be important in preventing intraperitoneal
adhesions [189].
Heparin has been shown to have adverse effects on gentamicin activity, and
could be a potential problem for patients treated for peritonitis with intra-
peritoneal administration of antibiotics [207]. A concentration of less than 3 U/ml
of heparin should not adversely affect gentamicin.
Systemic blood coagulation was not found to be affected by 10 000 unit doses of
intraperitoneal heparin [208, 209].

3.3. Cephalosporins

The cephalosporin group of antibiotics are weil absorbed from the peritoneal
cavity. Cephalosporins and aminoglycosides maintain activity when both are
added to the same dialysate bag [210, 211]. The peritoneal clearance of this group
of antibiotics is approximately 10 ml/min. The half-life of this group of drugs
administered to a patient on dialysis ranges between 7-20 hours. An occasional
patient will experience abdominal pain when the drug is administered intra-
peritoneally at doses greater than 250 mg/L of dialysate.
Cephalothin has been used intraperitoneally. Intraperitoneal administration of
50-100 mcg/ml at a dialysate flow rate of 2Llhr led to equilibration of dialysate
and serum levels in 6-9 hours. Approximately half the intraperitoneal dose was
absorbed.
 309

Table 2. Antibiotics useful in treating peritonitis

Antibiotic Intraperitoneal Safe serum Serum level Hours to attain Drug stability
generic dose in mgIL or level attained with therapeutic in the
name JLg/m1 in JLg/ml routine serum levels dialysate
parenteral doses
(JLg/ml)

Cephalothin+ 20-50 50-100 10-30 4--6 36 h

Ampicillin 50 50-100 3-7 4--6 12 h
Methicillin 100 100-200 16-72 1-6 8-24 h
Penicillin G 1000-50.000 VII 80 Vlml 15 Vlml 6-12 Stable if acid
(lJLg = 1.6 V) pH
Carbenicillin 200 200 110-170 4-6 36h
Vancomycina 15 15-25 23-40 8-12 36h
Tobramycina. c 8 10-12 3-14 8-12 probably Stable
Gentamicina, c 8 10-12 5-7 8-12 36h
Amphotericin Bb 2-4 0.5-3.5 Stable
Clindamycind 10 20-40 10 Stable

Chloramphenicol: The only parenteral preparation that does not become active intraperitoneally. and
oral or parenteral administration does not lead to adequate intraperitoneal levels.
Tetracycline and erythromycin: Not recommended.
aAt least one parenteral loading dose recommended.
bLow dose parenteral therapy recommended also, 200-500 mg total; 5 mg i.v. day 1,10 mg i.v. day 2,
15 mg i.v. day 3,20 mg i.v. day 4, then 25 mg i.v. for 5-20 days.
c Aminoglycoside TI/2 lessened by carbenicillin and possibly other penicillins. Follow serum levels.

d Requires metabolic conversion intraperitoneally before effective and, therefore, should be given

parenterally as well.
+ Doses of 125 mg/l used in CAPD.

Intramuscular administration of one gram led to significant concentrations

(potentially therapeutic) in dialysate for approximately 12 hours. In dogs and
man, one gram given intraperitoneally, led to drug levels after two hours that
were similar to levels obtained with an intravenous dose. In a study in rabbits
using interstitial capsules, Gerding showed that systemic administration yielded
peak levels in omental interstitial fluid within 1-2 hours. The peak levels were
only 26% of the serum cephalothin level [212-217].
Cephalexin is removed via dialysate with the calculated Tl/2 ranging between
3-8 hours [218]. Buncke suggests this drug does not help in the management of
peritonitis[219]. It is not a useful agent for prophylaxis [184].
Studies have been carried out with Moxalacatam [220, 221], Ceftazidime [222],
Cefazolin [210, 219, 223, 226], Cefoxtin [227], Cephamandole [228-230], Cefu-
roxime [231-233], and Cephaloridine [234].
310

Table 3. Drugs recommended for postdialysis supplementation

(Additional 0.5 dose after hemodialysis, 0.2-0.3 dose after peritoneal dialysis)

Penicillins: penicillin, ampicillin, carbenicillin

Aminoglycosides: streptomycin, kanamycin, gentamicin, paromomycin, tobramycin
Cephalosporins: cephalothin, cephalexin, cephapirin, cephazolin
Other antibiotics: chloramphenicol, sulfonamides, trimethoprim, cycloserine, isoniazid, ethambutol,
5-fluorocytosine
Vasoactive drugs: aminophylline, methydopa, procainamide, quinidine
Immunosuppressive - chemotherapeutic agents: methotrexate, 5-fluorouracil, cyclophosphamide
Vitamins: ascorbic acid, thiamin, pyridoxine, folic acid
Miscellaneous: salicylates, phenobarbital, barbital, lithium

3.4. Pencillins

The penicillins are absorbed after peritoneal instillation, and penetrate the
peritoneal cavity from the serum [213].
Methicillin administered intraperitoneally in a dose of one gram, reached
therapeutic serum levels in approximately two hours [235]. When given intra-
venously with dialysis, the Tl/2 of oxacillin, cloxacillin and dicloxacillin ranged
between 1.5-2.5 hours.
These studies show the influence of protein binding. Oxacillin and dicloxacillin
are 80 to 90% protein bound. They penetrate poorly into dialysate from serum.
There is good absorption (50% of administered dose) of cloxacillin from the
peritoneal cavity. There is a relatively low protein concentration in the peritoneal
dialysate [234, 236-239].
Ampicillin is 20 to 60% absorbed from the peritoneal cavity over 1 hour. When
given intravenously, the T1I2 during dialysis is 10-14 hours [213, 234-239].
Ticarcillin and carbenicillin are cleared via peritoneal dialysis at a rate of 2-7
ml/min. The T1I2 of a one gram dose intravenously is five to seven hours [240-
242].
The peritoneal clearance of Mezlocillin is approximately 7 ml/min [243].
Azlocillin is removed during peritoneal dialysis. The plasma half life is short-
ened from 235 minutes when not on dialysis (60 minutes in normals) to 150
minutes when on peritoneal dialysis [244].

3.5. Lincomycin, clindamycin

Lincomycin and clindamycin are not significantly dialyzable. The T1I2 for lin-
comycin during peritoneal dialysis has been reported as 13 hours, and for clin-
damycin four hours [245, 246].
 311

3.6. Vancomycin

Vancomycin has enjoyed a resurgence in use. Studies during peritoneal dialysis,

suggest that vancomycin is moderately well dialyzed. The T1I2 has been reported
to range between 18-48 h and the clearance to be 6.1 mllmin. Intraperitoneal
administration of 25-50 mcg/mllead to effective serum levels in 5-10 h after the
start of dialysis. In one report, peritonitis did not appear to significantly alter the
half-life [196, 247-252].
We have found that a dose of 50-75 mg added to the overnight two liter bag of
dialysate among patients undergoing CAPD will maintain blood levels of 15-20
[Link]/ml.

3.7. Sulfas

The sulfa drugs were the first to be used intraperitoneally. They may have
potential use in Nocardia peritonitis occurring during peritoneal dialysis. Sul-
phamidine administered intraperitoneally, approached equilibration with plasma
after four days. In doses greater than 290 mg/ml, nausea occurred [253-255].
Fremont has used sulfamethoxazole and trimethoprim in doses of 80 mg/L and
16 mg/L respectively for the treatment of peritonitis [256].

3.B. Aminoglycosides

Gentamicin has been shown to have a peritoneal clearance between 7-19 ml/min,
and a half-life during peritoneal dialysis between 21-36 hours. When given
intraperitoneally at a dialysate flow rate of 2L1hr, it equilibrates with serum in
approximately 12 hours [257-266].
Tobramycin given intravenously, has a Tl/2 of 10-16 hours at a dialysate
exchange rate of 2L1hr. The clearance was calculated as 15 mllmin. Because of
low dialysate concentrations systemic antibiotic therapy for treatment of per-
itonitis usually requires intraperitoneal supplementation [267-269]. Tobramycin
can be considered almost identical to gentamicin.
Netilmicin behaves similarly to tobramycin and gentamicin [270].
Kanamycin has a Tl/2 of 12 hours during peritoneal dialysis, and a clearance of
8.3 ml/min. Intraperitoneal administration of 20 mg/2L1hr has led to therapeutic
serum levels in 10 h. There is one case report of apnea following intraperitoneal
administration in the post-anesthetic setting [200, 271-273].
Amikacin has been shown to be cleared peritoneally at a rate of 6.4-8 ml/min
with Tl/2 of 20-29 h during peritoneal dialysis [274-277].
312

3.9. Tetracycline, chloramphenicol, thiamphenicol

Tetracycline was once used extensively for treatment and prophylaxis during
peritoneal dialysis. It is absorbed following intraperitoneal instillation and is
cleared via the peritoneal cavity at a rate of 5.6 mllmin. At present its indications
are limited [235, 236, 271, 278, 279].
Peritoneal dialysis has been used to treat acute poisoning by tetracycline [280].
Chloramphenicol has a half-life of 2.5 to 7 hours during peritoneal dialysis at
dialysate flow rates of two liters every two hours. Peritoneal clearance is poor
[213,238,271].
Thiamphenicol is cleared at a rate of 7.7 ml/min, and has a Tl/2 of 13 112 h. It is
absorbed poorly from the peritoneal cavity, and little is removed with dialysis
[282,283].

3.10. Metronidazole

Metronidazole penetrates into dialysate in a sufficient concentration to treat

anerobic infections [284, 285].

3.11. Erythromycin

Erythromycin has been added to peritoneal dialysate in a dose of 125 mg/L of

dialysate. Apparently two liters of dialysate were exchanged every four hours.
The authors found that serum levels of 63-78 mg/L were associated with ototox-
icity, slurred speech, diplopia and confusion. The drug was given with sodium
fusidate [286].

3.12. Polymyxins

Colistin is highly protein-bound, and one report suggests removal of antibiotic is

more efficient with peritoneal dialysis than hemodialysis. The T1I2 is 10-14 hr
during peritoneal dialysis with a peritoneal clearance of 6-11 ml/min. One case of
apnea (associated with intraperitoneal administration) has been reported in the
post-operative setting [287-290].

3.13. Antifungal agents

Flucytosine has been used to treat peritonitis secondary to Candida species. It

readily penetrates dialysate from serum and is well absorbed from dialysate [291-
 313

295]. Jones has suggested a dose of 37.5 mg/kg followed by 10-15 mg/kg daily for
patients undergoing CAPD [294].
Amphotericin has been used intraperitoneally but causes pain to the patient.
Most treat fungal peritonitis by removing the peritoneal catheter and administer-
ing systemic therapy [296-299].
Miconazole has been used to treat selected episodes of peritonitis [300]. We
have used it in three patients at a dose of 30 .mg/2L of dialysate. In one instance
pain prompted us to discontinue therapy, one patient was cured and one patient
failed to respond. We have reserved this drug for use when hemodialysis can not
be achieved and the patient can not tolerate amphotericin.
Although Ketoconazole has been used to treat candida peritonitis we are not
aware of any studies on peritoneal transport of other antifungal agents [301, 302].

3.14. Antituberculous agents

Isoniazid is removed from the serum by peritoneal dialysis [303-305].

Ethambutol and cycloserine have been reported to be cleared using peritoneal
dialysis [306, 307].

3.15. Anti-viral agents

Amantadine and Acyclovir are poorly removed by peritoneal dialysis [308, 309].

3.16. Anti-malarials

Chloroquine is not effectively removed with dialysis [310].

Quinine intoxication has been treated using peritoneal dialysis [311-317]. Some
authors have reported effective removal whereas others have not. Addition of
tromethamine impaired removal. Quinine is a weak base and at physiologic pH is
in an ionic and non-ionic state. It has been suggested that the peritoneal mem-
brane is more permeable to the nonionized species. Dialysate is acidic when
instilled and rapidly reaches equilibration with plasma pH. Alkalinization of
dialysate above plasma pH enhances transport of nonionized quinine into the
peritoneal cavity. The ionized species is trapped when dialysate pH rises [311-
317].

3.17. Antiseptics

Antiseptics have been instilled intraperitoneally in attempts to treat peritonitis or

314

as a prophylactic measure. None of the agents described below are particularly

useful.
Povidone-Iodine has been used during surgery as an irrigating solution in
grossly contaminated abdomens. Experimental studies in animals have yielded
conflicting results as to efficacy. One author has incorporated it into the exchange
technique for continuous ambulatory peritoneal dialysis and reports a high inci-
dence of 'sterile' peritonitis, while others have advocated a saline-iodine flush of
the peritoneal cavity [318-322].
Taurolin (formaldehyde as the active agent) was used in infected surgical
abdomens with good results [323, 324]. It has not been instilled chronically.
Intraperitoneal chlorhexidine has been proposed as a useful prophylactic mea-
sure to prevent peritonitis. Sommerville found that this agent lead to severe
abdominal pain [325-328].

3.18. Poisoning

There are several reviews of treatment of poisoning by dialysis, the most exten-
sive of which is by Winchester [329]. In this section, we will discuss reports where
peritoneal dialysis was used to treat intoxications. Although peritoneal dialysis is
less efficient than hemodialysis, it may offer an advantage in particular situations;
the patient presenting with marked hypotension or hypothermia secondary to a
dialysable agent; intoxications in the very young where hemodialysis may be
technically impossible; or in patients too ill to transport and hemodialysis is not
available (Table 4).

Table 4. Drug intoxications clinically most responsive to dialysis

Drug Clearance (ml/min)

Peritoneal Hemodialysis

Salicylates 20 100
Bromide 14 >150
Thiocyanate >150
Barbiturates 3-10 65-110
Glutethimide 10 20-90
Ethchlorvynol 18 65
Meprobamate 20 60-100
Methyprylon 18 80
Methaqualone 7 23
Phenytoin 12
Lithium 14 75
Aminoglycosides 5 30-50
 315

3.19. Common analgesics

Salicylate intoxication may be treated by peritoneal dialysis. In children, per-

itoneal dialysis may be simpler to perform than hemodialysis. Addition of al-
bumin to dialysate improves the removal of salicylate. Salicylate binds to protein,
preventing back diffusion [330-335]. Alkalinization of instilled dialysate en-
hances removal but may be dangerous in patients with respiratory alkalosis.
Tromethamine has been added to dialysate in an attempt to raise dialysate pH
and maintain salicylate within the peritoneal cavity [336]. Tromethamine's effects
may be due to alterations in peritoneal permeability, since it has been shown that
infused dialysate approaches physiologic pH rapidly [12, 42, 43]. Hypokalemia,
occurring with salicylate intoxication, may be prevented by adding potassium to
dialysate.
Paracetamol is not readily removed with peritoneal dialysis. It is highly protein
bound [337]. It is possible that addition of albumin to dialysate may enhance
paracetamol removal, but this has not been shown.
Propoxyphene hydrochloride is dialysable, but drug removal by peritoneal
dialysis is small compared to the usual ingested dose during a suicide attempt
[338-340].

3.20. Hypnotics and sedatives

Barbituate intoxication has been treated with peritoneal dialysis. Henderson

calculated drug removal rates with hemodialysis, peritoneal dialysis, and forced
diuresis [341-344]. Clearance calculations in Ll24 hours showed peritoneal di-
alysis to be approximately 75% as efficient as hemodialysis and forced diuresis
[345]. Albumin and tromethamine has been shown to enhance removal [346-
349]. Exaire reported one case wherein intraperitoneal furosemide enhanced
peritoneal removal [350]. Rosenbaum showed that recirculating peritoneal di-
alysis through an anion exchange resin was effective in removing large quantities
ofbarbituate in dogs [351]. Occasionally, hypothermia may occur with barbituate
intoxication and peritoneal dialysis might be useful in warming the patient [352-
353].
Gluthethimide may be removed by peritoneal dialysis [354-355]. Von Har-
titzsch suggests that 24 h of peritoneal dialysis with albumin may be as effective as
six hours of hemodialysis [356]. McDonald's experience was less favorable [357].
In one clinical report, methylprylon overdose was treated successfully by
peritoneal dialysis [358, 359].
Meprobamate readily permeates the peritoneal cavity and peritoneal dialysis
has been used in therapy for intoxications. We calculated peritoneal clearance in
one reported case as 22.5 ml/min [360-363].
Ethochlorvynol peritoneal clearance has been reported at 18.5 mllmin. In one
316

study, removal of ethrochlorvynol was enhanced by albumisol and

thromethamine [364, 365].
Methaqualone is cleared by peritoneal dialysis at a rate of 4-8 ml/min [366,
367].

3.21. Antidepressants

Amphetamine poisoning occurring in a child has been treated by peritoneal

dialysis. One group has induced amphetamine poisoning in dogs and treated the
dogs by peritoneal dialysis. Their results are difficult to interpret, but the authors
felt there was some advantage [368-370].
Peritoneal dialysis is not effective therapy for removing tricyclic antidepres-
sants [371-374]'
Peritoneal dialysis has been used in pargyline hydrochloride intoxication but
the removal rate of the drug was not reported [375].

3.22. Alcohols

Methanol is removed with peritoneal dialysis. Clearance measurements in dogs

with 20-minute equilibration periods of dialysate within the peritoneal cavity
were 24 ml/min for 1.5% glucose exchange and 31 ml/min for a 7% exchange.
Peritoneal dialysis alone removed 30% of the administered dose of methanol
after six hours. Although less efficient than hemodialysis, peritoneal dialysis
could potentially be of benefit [376-381].
Ethylene glycol is removed by peritoneal dialysis. One of the metabolic con-
sequences of poisoning with ethylene glycol is lactic acidosis. If peritoneal dialysis
were to be used, non-lactate dialysis solutions might be considered. Oxalate, one
of the metabolites of ethylene glycol is cleared by peritoneal dialysis at 5ml/min
[382-384].
Isopropyl alcohol intoxication has been treated by peritoneal dialysis [385].
Severe ethanol poisoning in a child has been treated by peritoneal dialysis.
Clearance measurements were not reported but the authors felt that the rate of
drug elimination was increased by the use of peritoneal dialysis [386].

3.23. Heavy metals & fluoride

Perhaps the most important use for peritoneal dialysis is to treat the acute renal
failure that occurs with these intoxications.
Iron, mercury, chromium, thallium and fluoride are poorly removed by per-
itoneal dialysis [387-398].
 317

Deferoxamine may be considered in cases of iron poisoning [399, 400].

Systemic administration of dimercaprol enhanced peritoneal mercury re-
moval.
Significant removal with peritoneal dialysis occurs for intoxications with po-
tassium dichromate, lead, copper, gold, and cesium. Albumin enhanced removed
of copper, whereas intraperitoneal penicillamine did not [401-409].

3.24. Cardioactive agents

Digoxin is not removed effectively with peritoneal dialysis (nor hemodialysis)

[410-414]. Some have suggested that the dose be increased if a patient's residual
renal function is significant. The prudent approach is to measure drug levels and
titrate the patient.
Procainamide has been reported to be slowly removed via peritoneal dialysis in
a patient ingesting a toxic dose. The clearance via peritoneal dialysis contributed
little to this patient's recovery. Biotransformation was the major route of elimina-
tion [415].
There is no need to increase the dose of quinidine during peritoneal dialysis.
The clearance of the parent compound is around 1 ml/min and the clearance of
metabolites are low. The metabolites may interfere with the clinical assay leading
to spuriously high concentrations when the active compound is at a low concen-
tration [416, 417].
There are no data concerning the pharmacology of beta-blockers in peritoneal
dialysis. A method has recently been described for analysis in dialysate [418]. This
class of agents has been implicated in sclerosing obstructive peritonitis [419].
It is not necessary to increase the dose of Nifedipine in patients undergoing
peritoneal dialysis. It is not significantly cleared by peritoneal dialysis [420].
Amiodarone, an anti-arrythmic agent, administered to patients on dialysis is
not recovered in the dialysate [421].
Aldomet is removed by dialysis [422]. Approximately half the administered
oral dose of radioactivity was removed with dialysate effluent (n = 2 patients).
In one report, plasma levels of diazoxide were lowered in a patient undergoing
peritoneal dialysis, whereas others have suggested no significant effect [423,
424].
Theophylline alters peritoneal properties. It is dialysable and peritoneal di-
alysis could be potentially useful in intoxications [289]. Peritoneal clearance is
5-10 ml/min at dialysis flow rates of 2Llhr [425-429].

3.25. Anti-epileptics

Dilantin is poorly eliminated via peritoneal dialysis [430-432].

Valproic acid is poorly eliminated by peritoneal dialysis [433].
318

3.26. Chemotherapeutic agents

Peritoneal dialysis is being explored as a route for delivery of chemotherapeutic

agents [434]. In these trials the investigators use the low peritoneal clearance of
the anti-neoplastic agent to advantage. This low peritoneal clearance permits a
high local dose of chemotherapeutic agent at a lower serum concentration. The
goal of this therapy is direct delivery to the site of cancer while offering the
possibility of less systemic toxicity because of poor absorption [435].
Thiotepa [436-437], methotrexate (peritoneal clearance 6-7.5 ml/min 438,
439), cyclophosphamide [440, 441], 5-fluorouracil [442], adriamycin [443] and
cisplatin [444] have been administered intraperitoneally with varying results.
Intraperitoneal administration of immunotherapy is undergoing trial as well
[445-447].

3.27. Contrast agents

Iodine, administered with contrast agents, is cleared by peritoneal dialysis at

about the same rate as creatinine [448-451].
The peritoneal clearance of Diethylenetriaminepentacetate labeled with Tech-
netium-99m (Tc-99m DTPA) was reported as 14 mllmin [452].
Gallium-67 is not cleared efficiently by peritoneal dialysis [453].

3.28. Hyper/hyponatremia, lithium and bromide

Finberg and Kiley successfully treated sodium intoxication in neonates by per-

itoneal dialysis [454].
Since ultrafiltrate generated by dialysate is near half normal saline, peritoneal
dialysis may be used to correct hyponatremia. It is particularly helpful when
hypertonic saline cannot be infused for fear of inducing pulmonary edema [455].
Lithium and bromide are readily removed via peritoneal dialysis at a rate of
13-15 mllmin [456-460].

3.29. Miscellaneous agents

Cimetidine peritoneal clearance is approximately 4-10 mllmin. There is no need

to adjust the conventional renal failure dosing regime [461-465].
Oral hypoglycemic agents are not readily dialyzable. However, peritoneal
dialysis may be a convenient way to administer glucose to the patient [466-468].
Epsilon-aminocaproic acid is cleared at 13 ml/min. The dose of this drug should
be decreased to 25% of the usual recommended dose in renal failure [469].
 319

Potassium ferrocyanide poisoning has been treated using peritoneal dialysis

with correction of associated hyperkalemia [470].
Mannitol overdosage in the presence of renal impairment may lead to intra-
vascular fluid overload. Aviram reported three cases wherein peritoneal dialysis
corrected the fluid balance of the patient but failed to remove sufficient mannitol
leading to persistance of hyperosmolality and hyponatremia [471].
Peritoneal dialysis was used to reverse neuromuscular blockade by gallamine
[472].
Debendox® poisoning (antiemetic) which is a combination of dicyclomine
hydrochloride (effects similar to atropine) and doxylamine succinate (anti-
histamine) has been treated by peritoneal dialysis. Although the amount re-
moved was small, the authors felt it was clinically important [473].
Peritoneal dialysis in pheniramine male in ate (A VIL (R)) intoxication did not
remove significant amounts of the drug [474].
Mushroom poisoning has been treated by peritoneal dialysis. We were unable
to find removal rates of the toxic cyclopeptides (phalloidine and am ani tine ) by
peritoneal dialysis. The major importance of dialysis may be treatment for acute
[Link] that occurs with poisoning [475-477].
Eucalyptus oil intoxication has been successfully treated by a combination of
mannitol infusions, peritoneal dialysis and hemodialysis [478].
Peritoneal dialysis has been used in the treatment of organo-phosphorus poi-
soning. Removal in dialysate was demonstrated. The authors felt that peritoneal
dialysis shortened the period of illness [479].
Arsine exposure causes acute hemolysis. It is not effectively removed by
peritoneal dialysis [480].
Boric acid intoxication has been treated by peritoneal dialysis [481-484].
Lysol (Cresol 10%), chlorate, and hexachlorophene are not sufficiently re-
moved to warrant use of peritoneal dialysis during intoxications with these agents
[485-488].

4. Novel applications of peritoneal dialysis

4.1. Insulin

Patients with end stage renal disease secondary to diabetes mellitus have been
managed with peritoneal dialysis. Clinically, this requires a method of managing
blood sugar during dialysis. One approach is to supplement insulin by sub-
cutaneous doses as required [489-491]. Several authors have reported intra-
peritoneal administration of insulin to control blood sugar during intermittent
peritoneal dialysis and continuous ambulatory peritoneal dialysis [492, 493]. The
regimes are empiric and are derived by monitoring blood sugar during dialysis
and adjusting the dose of regular insulin added to dialysate until satisfactory
control of the patient's blood sugar is obtained.
320

Insulin is not significantly bound to the plastic bags used for peritoneal dialy-
sate [494-497].
We allow the patient their usual dose of insulin. When dialyzing with prepared
dialysate solutions 8-12 units of regular insulinlper 2 L of 4.25% dextrose solution
and 6-10 D/per 2L of 1.5% dextrose solution are usually required. Insulin binding
to the dialysis tubing does not appear to be significant. When the reverse-osmosis
machines are used, regular insulin can be added to the concentrate in a dose of 200
D/2L of 30% concentrate, 400 D/2L of 50% concentrate, and 50 units to each 500
cc bottle of 50% dextrose solution [498].
There is a risk of post-dialysis hypoglycemia [499]. During intermittent per-
itoneal dialysis with commercially packaged dialysate, we omit insulin from the
last four exchanges; whereas with the reverse-osmosis machines, 500 cc of 4.25%
dextrose dialysate is left in the peritoneal cavity at the end of dialysis.
We, and others, have found that diabetic patients undergoing continuous
ambulatory peritoneal dialysis can be managed with intraperitoneal insulin alone
[500, 501]. We have used regular insulin in a dose of 5-20 D/per 2L of 1.5%
dextrose solution, 10-30 D/per 2 L of 4.25% solution and in one patient, 24 D/per
2 L of 2.5% dextrose solution. Two-hour post-prandial blood sugars have ranged
between 200-300 mgllOO ml.
Nocturnal hypoglycemia may be prevented by lowering the overnight dose of
insulin or adjusting food intake. We supply our patients with a device to measure
blood sugars. Reliable patients are instructed to modify their insulin therapy to
control blood sugar.
The amount of insulin absorbed is a function of intraperitoneal residence of
dialysate. Its absorption rate from dialysate is similar to inulin. In studies using
relatively short periods of intraperitoneal residence of dialysate, Shapiro esti-
mated that between 6-25% of a dose of insulin administered intraperitoneally
was absorbed; whereas Crossley and Kjjellstrend estimated less than 5% of the
administered dose was absorbed [492, 493, 502-504].
Patients undergoing CAPD may be expected to absorb 50% of the admin-
istered dose of insulin over an 8 hour period [505].
It has been suggested that patients with markedly elevated triglycerides, with-
out elevated blood sugars, may obtain biochemical correction of this abnormality
with a low dose of intraperitoneal insulin. This observation has not been con-
firmed [506].
Intraperitoneal insulin may be a more physiologic route to administer insulin
since absorption is via the portal system. There is evidence that the liver can
increase the amount of insulin extracted from the portal circulation when an
increased load is presented. This may be a safety margin for patients using
intraperitoneal insulin in that a systemic circulation may be partially protected
from an excessive dose of intraperitoneal insulin.
A potential disadvantage to the use of intraperitoneal insulin for control of
blood sugar is the inability to readily increase the dose of insulin prior to meals.
 321

4.2. Edematous states

One of the early applications of peritoneal dialysis was congestive heart failure.
We have found it useful in the acute situation. It may be more efficient to remove
3-4 kg from a patient using peritoneal dialysis than to wait for a diuretic response
[507-512].
CAPD offers a method of controlling patients refractory to diuretics. We have
found CAPD particularly useful in the patients with heart failure, severe nephro-
tic syndrome in patients resistant to high dose diuretics and having moderate to
severe renal failure, and for dialysis related ascites [513). CAPD allows us to
control intravascular volume and liberalize the diet.

4.3. Oxygen

Intraperitoneal administration of oxygenated dialysate and 0.3% hydrogen per-

oxide have been attempted in models of respiratory insufficiency. Sufficient
benefit was not obtained to warrant clinical trials [514, 515).

4.4. Blood

Blood may be administered via the intraperitoneal route. Approximately two-

thirds of cells survive normally. Absorption may be prolonged over a week.
Absorption is felt to be primarily via lymphatics [516-521].

4.5. Thyrotoxic crisis

Thyroxine is removed with peritoneal dialysis and has been included in the
management of patients in crisis. Thyroxine was removed in protein-bound form
in a rat model of thyrotoxic crisis [522-525].

4.6. Bilirubin

Peritoneal dialysis has been found to remove significant quantities of indirect

bilirubin, particularly with albumin added to dialysate. Removal rates, however,
are not fast enough to check a rapidly rising bilirubin in neonates [526-528].
322

4.7. Pediatric enzyme deficiencies

Peritoneal dialysis has been used in the treatment of: leucinosis, hyperam-
monemia in urea cycle enzyme deficiencies, propionicacidemia, removal of
branched-chain amino acids and their keto-acids in Maple-Syrup-Vrine disease
[529-536].

References

1. Putnam TJ: The living peritoneum as a dialyzing membrane. Am J Physiol18: 548-565, 1923.
2. Cunningham RS: The physiology of the serous membranes. Physiol Rev 6: 242-280, 1926.
3. Dandy WE, Rowntree LG: Peritoneal and pleural absorption with reference to postural treat-
ment. Ann Surg 59: 587-596, 1914.
4. Cunningham RS: Studies on absorption from serous cavities. III. The effect of dextrose upon the
peritoneal mesothelium. Am J Physiol 53: 488-494, 1920-21.
5. Leathes JB, Starling EH: On the absorption of salt solutions from the pleural cavities. J Physiol
Cambridge 18: 106--116, 1920-21.
6. Clark AJ: Absorption from the peritoneal cavity. J Pharmacol Exp Ther 16: 415-433, 1920-21.
7. Shear L, Swartz C, Shinaberger JA, Barry KG: Kinetics of peritoneal fluid absorption in adult
man. N Engl J Med 272: 123-127, 1965.
8. Shear L, Castellot JJ, Shinaberger JH, Poole L, Barry KG: Enhancement of peritoneal fluid
absorption by dehydration, mercaptomerin and vasopresin. J Pharm Exp Ther 154: 289-297,
1966.
9. Shear L, Castellot JJ, Barry KG: Peritoneal fluid absorption. I. Effect of dehydration on
kinetics. J Lab Clin Med 66: 232-244, 1965.
10. Rubin J, Nolph K, Popovich R, Moncrief J, Prowant B: Drainage volumes during continuous
ambulatory peritonitis. ASAIO J 2: 54-60, 1979.
11. Gordon A, Bennett LR, Maxwell MH: Transperitoneal water transport during peritoneal
dialysis. Abstr J Clin Inv 45: 1015, 1966.
12. Boen ST: Peritoneal dialysis in clinical medicine. Charles C. Thomas, Springfield Illinois, 1964.
13. Robson M, Oreopoulos DG, Izatt S, Ogilvie R, Rapoport A, de Veber GA: Influence of
exchange volume and dilaysate flow rate on solute clearance in peritoneal dialysis. Kidney Int 14:
486-490, 1978.
14. Twardowski ZJ, Prowant BF, Nolph KD, Martinez AJ, Lampton LM: High volume, low
frequency continuous ambulatory peritoneal dialysis. Kidney Int 23: 64-70, 1983.
15. Goldschmidt ZH, Pote HH, Katz MA, Shear L: Effect of dialysate volume on peritoneal dialysis
kinetics. Kidney Int 5: 240-245, 1974.
16. Rasio EA: Metabolic control of permeability in isolated mesentery. Am J Physiol226: 962-968,
1974.
17. Berndt WO, Gosselin RE: Action of vasopressor on the permeability of mesentery. Science 34:
1987-1988, 1961.
18. Berndt WO, Gosselin RE: Rubidium and creatinine transport across isolated mesentery.
Biochem Pharm 8: 359-366, 1962.
19. Berndt WO, Gosselin RE: Differential changes in permeability of mesentery to rubidium and
phosphate. Am J Physiol202: 761-767, 1962.
20. Fleisher MS, Loeb L: Studies in edema. VI. The influence of adrenalin on absorption from the
peritoneal cavity, with some remarks on the influence of calcium chloride on absorption. J Exp
Lancaster, Pennsylvania 13: 288-310, 1910.
 323

21. Shear L, Harvey ID, Barry KG: Peritoneal sodium transport: enhancement by pharmacologic
and physical agents. 1 Lab Clin Med 67: 181-188, 1966.
22. Buxton BH, Torrey IC: Absorption from the peritoneal cavity, Part III. The function of the
diaphragm. 1 Med Res 15: 41-54, 1906.
23. Starling EH: On the absorption of fluids from the connective tissue spaces. 1 Physiol Cambridge
19: 312-326, 1895-1896.
24. Starling EH, Tubbyy AH: On absorption and secretion into the serous cavities. 1 Physiol
Cambridge 16: 140-155, 1894.
25. Atkinson M, Losowsky MS: Plasma colloid osmotic pressure in relation to the formation of
ascites and oedema in liver disease. Clin Sci 22: 383-389, 1962.
26. Maher IF, Hirszel P, Lasrich M: Effect of gastrointestinal hormones on transport by peritoneal
dialysis. Kidney Int 16: 130-136, 1979.
27. Henderson LW, Nolph KD: Altered permeability of the peritoneal membrane after using
hypertonic peritoneal dialysis fluid. 1 Clin Invest 48: 992-1001, 1969.
28. Zelman A, Gisser D, Whittam PI, Parsons RH, Schuyler R: Augmentation of peritoneal dialysis
efficiency with programmed hyperlhypo-osmotic dialysate. Trans ASAIO 23: 203-209, 1977.
29. Nolph KD, Ghods AI, Brown P, VanStone 1, Miller FN, Wiegmann DL, Harris PD: Factors
affecting peritoneal dialysis efficiency. Dial Transpl6: 52-64, 1977.
30. Gutman RA: Toward enhancement of peritoneal clearances. Dial Transpl 8: 1072-1078,1979.
31. Maher IF, Hirszel P, Lasrich M: An experimental model for study of pharmacologic and
hormonal influences on peritoneal dialysis. Contrib Nephrol17: 131-318, 1979.
32. Nolph KD: Effects of intraperitoneal vasodilators on peritoneal clearances. Dial Transpl7: 812-
817,1978.
33. Hare HG, Valtin H, Gosselin RE: Effect of drugs on peritoneal dialysis in the dog. 1 Pharm Exp
Ther 145: 122-129, 1964.
34. Penzotti SC, Mattocks AM: Effects of dwell time, volume of dialysis fluid, and added accelera-
tors on peritoneal dialysis of urea. 1 Pharm Sci 60: 1520-1522, 1971.
35. Mattocks AM, Penzotti SC: Acceleration of peritoneal dialysis with minimum amounts of
dioctyl sodium sulfosuccinate. 1 Pharm Sci 61: 475-476, 1972.
36. El-Bassiouni EA, Mattocks AM: Acceleration of peritoneal dialysis with minimal N-Myristyl-
R-aminopropionate. 1 Pharm Sci 62: 1314-1317, 1973.
37. Penzotti SC, Mattocks AM: Acceleration of peritoneal dialysis by surface active agents. 1 Pharm
Sci 57: 1192-1195, 1968.
38. Gross M, McDonald HP: Effect of dialysate temperature and flow rate on peritoneal clearance.
lAMA 202: 363, 1967.
39. Patton IF, Doolittle WH, Hamlet MP: Peritoneal clearance of urea and potassium following
experimental hypovolemia, 1 Appl Physiol 36: 403-406, 1974.
40. Greene lA, Lapco L, Weller 1M: Effect of drug therapy of hemorrhagic hypotension on kinetics
of peritoneal dialysis in the dog. Nephron 7: 178-183,1970.
41. Henderson LW, Kintzel IE: Influence of antidiuretic hormone on peritoneal membrane area
and permeability. 1 Clin Invest 50: 2437-2443, 1971.
42. Mattocks AM: Accelerated removal of salicylate by additives in peritoneal dialysis fluid. 1
Pharm Sci 58: 595-598, 1969.
43. Kudla RM, El-Bassiouni EA, Mattocks AM: Accelerated peritoneal dialysis of barbiturates,
diphenylhdantoin, and salicylate. 1 Pharm Sci 60: 1065-1067, 1971.
44. Knochel IP, Mason AD: Effect of alkalinization on peritoneal diffusion of uric acid. Am 1
Physiol 210: 1160-1164, 1966.
45. Torres IJ, Litterst CL, Guarion AM: Transport of model compounds across the peritoneal
membrane in the rat. Pharmacology 17: 330-340, 1978.
46. Miller FN: Effects of peritoneal dialysis on rat microcirculation and peritoneal clearances in
man. Dial Transpl 7: 818, 1978.
324

47. Miller FN, Joshua IG, Harris PD, Wiegman DL, Jauchem JR: Peritoneal dialysis solutions and
the microcirculation. Contrib Nephrol17: 51-58, 1979.
48. Miller FN, Nolph KD, Harris PD, Rubin J, Wiegman DL, Joshua IG: Effects of peritoneal
dialysis solutions on human clearances and rat arterioles. Trans ASAIO 24: 131-132, 1978.
49. Miller FN, Wiegman D L, Joshua I G, Nolph KD, Rubin J: Effects of vasodilators and peritoneal
dialysis solution on the microcirculation of the rate cecum. Proc Soc Exp Bioi Med 161: 605-608,
1979.
50. Nolph KD, Rubin J, Wiegman DL, Harris PD, Miller FN: Peritoneal clearances with three types
of commercially available peritoneal dialysis solutions: Effects of pH adjustment and intra-
peritoneal nitroprusside. Nephron 24: 35-40, 1979.
51. Miller FN, Nolph KD, Sorkin MI, Gloor HJ: The influence of solution composition on protein
loss during peritoneal dialysis. Kidney Int 23: 35-39, 1983.
52. Mileti M, Bufano G, Scaravonati P, Pecchini F, Carnevale G, Lanzarini P: Effect of indom-
ethacin on the peritoneum of rabbits on peritoneal dialysis. Perit Dial Bull 3: 4-194, 1983.
53. Hirszel P, Lasrich M, Maher JF: Arachidonic acid increases peritoneal clearances. Trans
ASAIO 27: 61-63, 1981.
54. Rubin AL, Stenzel KH, Reidenberg MM: Symposium on drug action and metabolism in renal
failure. Am J Med 62: 459-460, 1977.
55. Bennett WM: Symposium on clinical pharmacology and the kidney patient. Dial Transpl8: 7,
1979.
56. Lasrich M, Maher M, Hirszel P, Maher JF: Correlation of peritoneal transport rates with
molecular weight: A method for predicting clearances. J ASAIO 2: 107-113, 1979.
57. Anderson RJ, Bennett WM, Gambertoglio JG, Schrier RW: Fate of drugs in renal failure. In:
BM Brenner, FC Rector (eds), The Kidney. WB Saunders Co, 1981, pp 2659-2708.
58. Maher JF: Pharmacology of peritoneal dialysis and permeability of the membrane. Dial Transpl
9: 197-200, 1980.
59. Manuel MA, Paton TW, Cornish WR: Drugs and peritoneal dialysis. Perit Dial Bull 3: 117-125,
1983.
60. Bennett WM, George R, AronoffR, Morrison G, GolperTA, PulliamJ, Wolfson M, Singer I:
Drug prescribing in renal failure: Dosing guideline for adults. Am J Kidney Dis 3: 155-193, 1983.
61. Ward RA, Klein E, Wathen RL: Investigation of the risks and hazards with devices associated
with peritoneal dialysis and sorbent regenerated dialysate delivery systems. Perit Dial Bull
(Suppl) 3: S1-S52, 1983.
62. Abrutyn E, Goodhart GL, Ross K, Anderson R, Buxton A: Acinetobacter ca1coaceticus
outbreak associated with peritoneal dialysis. Am J Epidermiol107: 328-335, 1978.
63. Ashline V, Stevens A, Carter MJ: Nosocomial peritonitis related to contaminated dialysate
warming water. Am J Infect Control 9: 50-52, 1981.
64. Gaunter WC, Feldman HA, Puschatt JB: Peritonitis in chronic peritoneal dialysis patients. Clin
Nephrol13: 255-259, 1980.
65. McDonald HP: An automatic peritoneal dialysis machine for hospital or home peritoneal
dialysis: Preliminary report. Trans ASAIO 15: 108-113, 1969.
66. Lasker N, Burker JF, Patchefsky A, Haughey E: Peritoneal reactions to particulate matter in
peritoneal dialysis solutions. ASAIO 21: 342-345, 1975.
67. Thomson NM, Stevens BJ, Humphery TJ, Atkins RC: Trace elements in patients on continuous
ambulatory peritoneal dialysi~. In: GM Gahl, M Kessel, KD Nolph (eds), Advances in Per-
itoneal Dialysis. Excerpta Medica, Amsterdam, 1981, pp 473-477.
68. Gokal R, Parkinson IS, Ramos M, Ward MK, Kerr DNS: Elevated serum aluminum levels in
patients on continuous ambulatory peritoneal dialysis. In: GM Gahl, M Kessel, KD Nolph (eds),
Advances in Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1981, pp 478-482.
69. Sorkin MI, Nolph KD, Anderson HO, Morris JS, Kennedy J, Prowant B, Moore H: Aluminum
mass transfer during continuous ambulatory peritoneal dialysis. Peri Dial Buill: 91-93, 1981.
 325

70. Cumming AD, Simpson G, Bell D, Cowie J, Winney RJ: Acute aluminum intoxication in
patients on continuous ambulatory peritoneal dialysis. Lancet 1: 103-104, 1982.
71. Oreopoulos DG, Clayton S, Dombros N, Zellerman G, Katirtzoglow A: Nineteen months
experience with continuous ambulatory peritoneal dialysis (CAPD) Proc Europ Transpl Assoc
16: 178-183, 1979.
72. Nolph KD, Sorkin M, Rubin J, Arfania D, Prowant B, Fruto L, Kennedy J: Continuous
ambulatory peritoneal dialysis: three-year experience at one center. Ann Intern Med 92: 609-
613, 1980.
73. Triolo G, Segoloni G, Carta Q, Boggio-Bertinet D, Lloatti S, Stratta P, Messina M, Priasca G,
Vercellone A: CAPD with intrabag insulin administration (IBIA): An easy strategy to remove
the uremic diabetic patient (UDP) management. In: GM Gahl, M Kessel, KD Nolph (eds),
Advances in Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1981, pp 424-426.
74. Boyer J, Gill GN: Hyperosmolality complicating peritoneal dialysis. Ann Intern Med 67: 568-
572, 1967.
75. Handa SP, Cushner GB: Hyperosmolar hyperglycemic nonketotic coma during peritoneal
dialysis. Southern Med J 61: 700-702, 1968.
76. Food and Drug Administration, Device. Experience Network Reports #33000, 1978 and
#36269, 1980.
77. Nolph KD, Hano JE, Teschan PE: Peritoneal sodium transport during hypertonic peritoneal
dialysis. Ann Intern Med 70: 931-941, 1969.
78. Ahern DJ, Nolph KD: Controlled sodium removal with peritoneal dialysis. Trans ASAIO 18:
423-428, 1972.
79. Raja RM, Cantor RE, Boreyko C, Bushehri H, Kramer M, Rosenbaum J: Sodium transport
during ultrafiltration peritoneal dialysis. Trans ASAIO 18: 429-435, 1972.
80. Schmidt RW, Blumenkrantz M: Peritoneal sclerosis: A 'sword of democles' for peritoneal
dialysis? Arch Intern Med 141: 1265-1267, 1981.
81. Halaby SF, Mattocks AM: Absorption of sodium bisulfite from peritoneal dialysis solutions. J
Pharm Sci 54: 52-55, 1965.
82. Bischel MD, Barbour BH: Peritoneal dialysis with sorbitol versus dextrose dialysate. Nephron
12: 449-463, 1974.
83. Quellhorst UA, Mietzsch G, Doht B, Fernandez-Redo E, Kubosch J, Leititis U, Voiles E,
Thorwort V, Scheler F: Severe side effects during peritoneal dialysis caused by sorbitol contain-
ing dialysate. Dtsch Med Wochenschr 100: 1431-1435, 1975.
84. Food and Drug Administration: Device experience network reports #33000, 1978; #36269,
1979.
85. Bisson PG, Bailey KM: Sodium in peritoneal dialysis solutions. Br Med J 5: 1322-1324, 1979.
86. Kopp VH, Tonnis HJ: The effect of temperature on the effectiveness of peritoneal dialysis.
Munch Med Wochenschr 109: 2438-2443, 1967.
87. Oreopoulos DG: Chronic peritoneal dialysis. Clin Nephrol 9: 165-173, 1978.
88. Dahner E: Hohes hepatitis-infektionsrisiko fUr das personal bei der peritonealdialyse-be-
handlung Australiaantigen positiver patienten. Med Welt 25: 1705-1707, 1974.
89. Spector D: Hepatitis B mini epidemic in a peritoneal dialysis unit. Arch Intern Med 137: 1030-
1031, 1977.
90. Bond WW, Petersen NJ, Gravelle CR, Favero MS: Hepatitis B virus in peritoneal dialysis fluid:
A potential hazard. Dial Transplll: 592-600, 1982.
91. Goodman W, Gallagher N, Sherrard DJ: Peritoneal dialysis fluid as a source of hepatitis antigen.
Nephron 29: 107-109,1981.
92. Gault MH: Peritoneal dialysis solutions. J Can Med Assoc 108: 325-327, 1973.
93. Gault MH, Ferguson EL, Sidhu JS, Corbin RP: Fluid and electrolyte complications of peritoneal
dialysis. Ann Int Med 75: 253-262, 1971.
94. Tenckhoff H: Choice of peritoneal dialysis solutions. Ann Int Med 75: 313-314,1971.
326

95. Parker A, Nolph KD: Magnesium and calcium mass transfer during continuous ambulatory
peritoneal dialysis. Trans ASAIO 26: 194-196, 1980.
96. Nolph KD, Prowant B, Serkes KD, Morgan L, Baker B, Chary tan C, Gham K, Hamburger R,
Husserl F, Kleit S, McGuinness J, Moore H, Warren T: Multicenter evaluation of a new
peritoneal dialysis solution with a high lactate and a low magnesium concentration. Perit Dial
Bull 3: 63-65, 1983.
97. Kohaut EC, Balfe JW, Potter D, Alexander S, Lum G: Hypermagnesemia and mild hypocarbia
in pediatric patients on CAPD. Perit Dial Bull 3: 40, 1983.
98. Nolph KD, Prowant B, Serkes KD, Morgan L, Baker B, Chary tan C, Gham K, Hamburger R,
Husserl F, Kleit S, McGuinness J, Moore H, Warren T: Multicenter evaluation of a new
peritoneal dialysis solution with a high lactate and a low magnesium concentration. Perit Dial
Bull 3: 63-65, 1983.
99. Oreopoulos DG: Personal communication.
100. Blumenkrantz MJ, Kopple JD, Moran JK, Coburn JW: Metabolic balance studies and dietary
protein requirements in patients undergoing continuous ambulatory peritoneal dialysis. Kidney
Int 21: 849-861, 1982.
101. Garrett JJ, Cuddihee RE: Calcium absorption during peritoneal dialysis. Trans ASAIO 14: 372-
375, 1968.
102. Rubin J, Rust P, Brown P, Popovich RP, Nolph KD: A comparison of peritoneal transport in
patients with psoriasis and uremia. Nephron 29: 185-1189, 1981.
103. Anderson KEH: Calcium transfer during intermittent peritoneal dialysis. Nephron 29: 63-67,
1981.
104. Stoltz ML, Nolph KD, Maher JF: Factors affecting calcium removal with calcium-free peritoneal
dialysis. J Lab Clin Med 78: 389-398, 1971.
105. Nolph KD, Stoltz M, Maher J: Calcium free peritoneal dialysis, treatment of vitamin D
intoxication. Arch Intern Med 128: 809:814,1971.
106. Lemann J, Donatelli AA: Calcium intoxication due to primary hyperparathyroidism. Ann
Intern Med 60: 447-461, 1964.
107. Miach PJ, Dawborn JK, Martin TS, Moon WJ: Management of the hypercalcemia of malignancy
by peritoneal dialysis. Med J. 97: 782-784, 1975.
108. Raju RM, Kramer MS, Rosenbaum JL, Manchanda R, Lazaro N: Evaluation of hypertonic
peritoneal dialysis solutions with low sodium. Nephron 11: 342-353, 1973.
109. Quellhorst E, Lowitz HD, Scheler F: The influence of dialysate osmolarity on the effectiveness
of peritoneal dialysis. Klin. Wschr. (Berlin) 49: 583-587,1971.
110. Swales JD: Sodium in peritoneal dialysis solutions. Br Med J 19: 1322-1323, 1979.
111. Swales JD: Sodium uptake in peritoneal dialysis. Br Med J 5: 345-347, 1967.
112. Nolph KD, Sorkin M, Moore H: Autoregulation of sodium and potassium removal during
continuous ambulatory peritoneal dialysis. Trans ASAIO 26: 334-338, 1980.
113. Spital A, Sterns R: Potassium supplementation via the dialysate in continuous ambulatory
peritoneal dialysis. Am Soc Nephrol Abstr P124A, 1983. Kidney International 25(1): 261,1984.
114. Brown ST, Ahearn DJ, Nolph KD: Potassium removal with peritoneal dialysis. Kidney Int 4:
67-69, 1973.
115. Rubin J, Nolph KD, Arfaania D, Wiegman DL, Miller FN, Harris PD: Comparison of the
effects of lactate and acetate on clinical peritoneal clearances. Clin Nephrol12: 145-147, 1979.
116. Rossen B, Ladefoged J: A comparison between the effects of acetate and lactate in peritoneal
dialysis solutions. Scand J Urol Nephrol16: 279--281, 1982.
117. Richardson JA, Borchardt KA: Antibacterial effect of different dialysates. Br Med J 20: 468-
469, 1972.
118. Richardson JA, Borchardt KA: Adverse effect on bacteria of peritoneal dialysis solutions that
contain acetate. Br Med J 3: 749-750, 1969.
119. Binswanger U, Keusch G, Bammatter F, Heule H, Kiss D: Peritonitis during continuous
 327

ambulatory peritoneal dialysis: Improving patient defense by type of buffer of dialysate?

Nephron 28: 300-302, 1981.
120. Diskin CJ, Coplon N, Feldman C, Vosti K: Antimicrobial activity in continuous ambulatory
peritoneal dialysis. Perit Dial Bull 3: 150-154,1983.
121. Dixon SR, McKean WI, Pryor JE, Irvine ROH: Changes in acid-base balance during peritoneal
dialysis with fluid containing lactate ions. Clin Sci 39: 51-60, 1970.
122. Jurgessen JC: Dialysis for lactic acidosis. N Engl J Med 278: 1350, 1968.
123. Lee HA, Hill LF, Hewitt V, Ralston AJ, BerJyne GM: Lacticacidemia in peritoneal dialysis.
Proc Eur Dial Transpl Assoc 4: 150-155,1967.
124. Nash MA, Russo JC: Neonatal lactic acidosis and renal failure: The role of peritoneal dialysis. J.
Pediatr 91: 101-105, 1977.
125. Sheppard JM, Lawrence JR, Oon RCS, Thomas DW, Row PG, Wise PH: Lactic acidosis:
Recovery associated with use of peritoneal dialysis. Aug N Z J Med 4: 389-392, 1972.
126. Hayat JC: The treatment of lactic acidosis in the diabetic patient by peritoneal dialysis using
sodium acetate: A report of two cases. Diabetologia 10: 485-487, 1974.
127. Vaziri ND, Ness R, Wellikson L, Barton C, Grup N: Bicarbonate buffered peritoneal dialysis-
An effective adjunct in the treatment of lactic acidosis. Am J Med 67: 392-396, 1979.
128. Vaziri ND, Warner AS: Peritoneal dialysis clearance of endogenous lactate. J Dial 3: 107-113,
1979.
129. Breborowicz A, Szule R: Removal of endogenous lactates via the peritoneum in experimental
lactic acidosis. Intensive Care Med 7: 297-300, 1981.
130. Rubin J, Adair C, Johnson B, Bower J: Sterospecific lactate absorption during peritoneal
dialysis. Nephron 31: 224-228, 1982.
131. Naparstek Y, Rubinger D, Friedlaender MM, Popovtzer MM: Lactic acidosis and peritoneal
dialysis. Isr J Med Sci 18: 513-514, 1982.
132. Foulks Maj CJ, Wright Maj LF: Successful repletion of bicarbonate stores in ongoing lactic
acidosis: A role for bicarbonate-buffered peritoneal dialysis. South Med J 74: 1162-1163, 1981.
133. Ing TS, Quon MJ, Daugirdas JT, Liu P, Gandhi VC, Reid RR: On-line preparation of
bicarbonate-containing dialysate for use in peritoneal dialysis. Int J Artif Organs 4: 308, 1981.
134. Rubin J, Nolph KD, Arfaania D, Joshua IG, Miller FN, Wiegman DL, Harris PD: Studies on a
non-vasoactive peritoneal dialysis solution. J Lab Clin Med 93: 910-915,1979.
135. Gjessing J: Bacterial growth in the dialysate fluid and the reaction of peritoneum to peritoneal
dialysis. Acta Medica Scand 182: 509-512, 1967.
136. Stewart WK, Anderson DC, Wilson MI: Hazard of peritoneal dialysis: Contaminated fluid. Br
Med J 11: 606-607,1967.
137. Vilbar RM, Ing TS, Shin KD, Gandhi VC, Viol GW, Chen WT, Geis WP, Hano JE: Treatment
of metabolic alkalosis with peritonea dialysis in a patient with renal failure. Artif Organs 2: 421-
422,1978.
138. Wu G: Osmotic agents for peritoneal dialysis solutions. Perit Dial Bull 2: 151-153, 1982.
139. Giordano C, De Santo NG, Capodicasa : Adding glucose to dialysate: pros and cons. Int J of
artificial organs 5: 225-117, 1982.
140. Heaton A, Johnston DG, Burrin JM, Orskov H, Ward MK, Alberti KGMM, Kerr DNS:
Carbohydrate and lipid metabolism during continuous ambulatory peritoneal dialysis (CAPD):
the effect of a single dialysis cycle. Clin Sci 65: 539-545, 1983.
141. Breckenridge WC, Roncari DAK, Khanna R, Oreopoulos DG: The influence of continuous
ambulatory peritoneal dialysis on plasma lipoproteins. Atherosclerosis 45: 249-258, 1982.
142. Duwe AK, Vas SI, Weatherhead JW: Effects of the composition of peritoneal dialysis fluid on
chemiluminescence, phagocytosis, and bactericidal activity in vitro. Infect Immun 33: 130-135,
1981.
143. Gilli P, Bastiani P De, Fagioli F, Buoncristiani U, Carobi C, Stabellini N, Squerzanti R, Rosati
G, Farinelli A: Positive aluminium balance in patients on regular peritoneal treatment: An effect
of low dialysate pH? Proc EDTA 17: 219-225, 1980.
328

144. RubinJ, Kirchner K, Barnes T, Teal N, Ray R, Bower J: Evaluation of continuous ambulatory
peritoneal dialysis. Am J Kid Dis 3(3): 199, 1983.
145. Gahl GM, Baeyer V, Averdunk R, Riedinger H, Borowzak B, Schurig R, Becker H, Kessel M:
Outpatient evaluation of dietary intake and nitrogen removal in continuous ambulatory per-
itoneal dialysis. Ann Int Med 94: 643-646, 1981.
146. Baeyer HV, Gerhard MG, Riedinger H, Borowzak R, Averdunk R, Schurig R, Kessel M:
Adaptation ofCAPD patients to the continuous peritoneal energy uptake. Kidney Int 23: 29-34,
1983.
147. Nolph KD, Rosenfeld PS, Powell JT, Danforth E Jr: Peritoneal glucose transport and hyper-
glycemia during peritoneal dialysis. Am J Med Sci 259: 272-281, 1970.
148. De Santo NG, Capodicasa G, Senatore R, Cicchetti T, Cirrillo D, Damiano M, Torella R,
Giugliano D, Improta L, Giordano C: Glucose utilization from dialysate in patients on continu-
ous ambulatory peritoneal dialysis. Int J Artif Organs 2: 119-124, 1979.
149. Selby R, Gilsdorf R, Potter D, Schon D: Systemic appearance of nutrients used in peritoneal
nutritional support: J Parent Nutrit 8: 93, 1984.
150. Blumenkrantz MJ, Kopple JD, Moran, JK, Coburn JW: Glucose absorption during continuous
ambulatory peritoneal dialysis. Kidney Int 19: 564-567, 1981.
151. Andersson G, Bergquist-Poppen M, Bergstrom J, Collste LG, Hultman E: Glucose absorption
from the dialysis fluid during peritoneal dialysis. Scand J Urol & Nephrol5: 77-79, 1971.
152. Nolph KD, Twardowski ZJ, Popovich RP, Rubin J: Equilibration of peritoneal dialysis solu-
tions during long-dwell exchanges. J Lab Clin Med 93: 246-256,1979.
153. Greenblatt DJ: Fatal hypoglycaemia occurring after peritoneal dialysis. Br Med J 2: 270-271,
1972.
154. Brown DJ, Adam WR, Dawborn JK: Glucose absorption and hyperglycaemia during peritoneal
dialysis. N Z J Med 1: 1-5, 1973.
155. Raja RM, Kramer MS, Manchanda R, Lazaro N, Rosenbaum JL: Peritoneal dialysis with
fructose dialysate. Ann Intern Med 79: 511-517,1973.
156. Robson MD, Levi J, Rosenfeld JB: Hyperglycaemia and hyperosmolarity in peritoneal dialysis:
its prevention by the use of fructose. Proc Eur Dial Transpl Assoc 6: 300-306, 1969.
157. Yen TS: Experimental study of peritoneal dialysis using xylitol containing solution. J Formosan
Med Assoc 69: 292-303, 1970.
158. Nihei N, Yamanaka T, Kurokawa K, Kurosawa T, Nagase M, Shimizu K, Kaneko T, Okaniwa
H, Sanjo T, Oka H, Neito C, Maeda T, Yoshitosi Y: Study of relationships between penetration
and metabolism: Case IV. (xylitoi) Jap J NephrollD: 162, 1968.
159. Y otuc W, Ward G, Shilipeter G, Tenckhoff H: Substitution of sorbitol for dextrose in peritoneal
irrigation fluid: A preliminary report. Trans ASAIO 13: 168-171,1967.
160. Mailloux L, Allerhand J: Comparison of glucose and sorbitol peritoneal dialysis. Abstr Am Soc
Nephrol, p 52, 1970.
161. Raja RM, Maros JG, Kramer MS, Rosenbaum JL: Hyperosmotic coma complicating peritoneal
dialysate with sorbitol dialysate. Ann Int Med 73: 993-994, 1970.
162. Olmsted WH: The metabolism of mannitol and sorbitol. Diabetes 2: 132-137, 1953.
163. Bazzato G, Coli U, Landini S, Fracasso A, Morachiello P, Righetto F, Scanferla F: Xylitol and
low dosages of insulin: new perspectives for diabetic uremic patients on CAPD: Perit Dial Bull 2:
159-160, 1982.
164. Bazzato G, Coli U, Landini S, Fracasso P, Morachicello F, Righetto F, Scanferia, Onesti G:
Xylitol as osmotic agent in CAPD: An alternative to glucose for uremic diabetic patients? Trans
ASAIO 28: 280-286,1982.
165. Kopple JD, Blumenkrantz MJ, Jones MR, Moran JK, Coburn JW: Plasma amino acid levels and
amino acid losses during continuous ambulatory peritoneal dialysis. Amer J Clin Nutr 36: 395-
402,1982.
166. Armstrong VW, Buschmann U, Ebert R, Fuchs C, Rieger J, Scheler F: Biochemical investiga-
 329

tions of CAPD: Plasma levels of trace elements and amino acids and impaired glucose tolerance
during the course of treatment. Intern J Artif Organs 3: 237-241, 1980.
167. Gjessing J: Addition of amino acids to peritoneal-dialysis fluid. Lancet 12: 812,1968.
168. Gjessing J: Absorption of amino acids and fat from the peritoneum. Opuscula Medica 13: 251-
252,1968.
169. Dombros N, Oren A, Marliss EB, Anderson GH, Stein AN, Khanna R, Petit K, Brandes L,
Rodella H, Leibel BS, Oreopoulos DG: Plasma amino acid profiles and amino acid losses in
patients undergoing CAPD. Perit Dial Bull 2: 27, 1982.
170. Anderson GH, Marliss E, Khanna R, Pettit J, Mupas L, Rodella H, Brandes L, Roncari DA,
Kakis G, Harrison J, McNeil K, Oreopoulos DG: Effective use of amino acid dialysate over four
weeks in CAPD patients. Perit Dial Bull 3: 66, 1983.
171. Williams P, Marliss E, Anderson GH, Oren A, Stein A, Khanna R, PetittJ, Brandes L, Rodella
H, Mupas L, Dombros N, Oreopoulos DG: Amino acid absorption following intraperitoneal
administration in CAPD patients. Perit Dial Bull 2: 124, 1982.
172. Giordano C, De Santo NG, Capodicasa G, Di Leo VA, Di Serafino A, Cirillo D, Esposito R,
Fiore R, Damiano M, Buonadonna L, Cocco F, Di Iorio B: Amino acid losses during CAPD.
Clin Nephrol14: 230-232, 1980.
173. Young Ga, Parsons FM: The effect of peritoneal dialysis upon the amino acids and other
nitrogenous compounds in the blood and dialysates from patients with renal failure. Clin Sci 37:
1-10, 1969.
174. Oreopoulos DG, Marliss E, Anderson GH, Oren A, Dombros N, Williams P, Khanna R,
Rodella H, Brandes L: Nutritional aspects of CAPD and the potential use of amino acid
containing dialysis solutions. Perit Dial Bull 3: SIO-S12, 1983.
175. Giordano C, Capodicasa G, De Santo NG: Artificial gut for total parenteral nutrition through
the peritoneal cavity. Int J Artif Organs 3: 325-330, 1980.
176. Jackson MA, Thomas DW, Talbot S, Lee HA: Prevention of amino acid losses during peritoneal
dialysis. Postgrad Med J 55: 533-536, 1979.
177. Nolph KD, Hopkins CA, Rubin J, Twardowski Z, Popovich R, Van Stone J: Polymer induced
ultrafiltration in dialysis: High osmotic pressure due to impermeant polymer sodium. Trans
ASAIO 24: 162-168, 1978.
178. RubinJ, Klein E, Jones Q, Planch A, Bower J: Evaluation of a polymer dialysate. Trans ASAIO
29: 62-65, 1983.
179. Gjessing J: The use of dextran as a dialyzing fluid in peritoneal dialysis. ACTA Med Scand 185:
237-239, 1969.
180. Jirka J, Kotkova E: Peritoneal dialysis by iso-oncotic dextran solution in anaesthetized dogs.
Intra-peritoneal fluid volume and protein concentration in the irrigation fluid. Proc Eur Dial
Transpl Assoc 4: 141-145, 1967.
181. Fine J, Frank HA, Seligman AM: The treatment of acute renal failure by peritoneal irrigation.
Ann Surg 124: 857-878, 1946.
182. Frank HA, Seligman AM, Fine J: Further experiences with peritoneal irrigation for acute renal
failure. Ann Surg 128: 561-608, 1948.
183. Beam TR, Catalano L, Bentzel CJ, Mookerjee BK: Protamine sulfate (PS) prevents bacterial
peritonitis in an experimental model of CAPD. Abstr Am Soc Nephrol54: 118A, 1983. Kidney
International 25(1): 255, 1984.
184. Low DE, Vas SI, Oreopoulos DG, Manuel MA, Saiphoo MM, Finer C, Dombros N: Prophylac-
tic cephalexin ineffective in chronic ambulatory peritoneal dialysis. Lancet 4: 753-754, 1980.
185. Verger C, Berry JP, Galle P, Lavergne A, Hoang C, Le Charpentier Y: Foreign material
inclusions in the peritoneum of CAPD patients: A study with X-ray microanalysis. Perit Dial
Bull 2: 138-139, 1982.
186. Wilkins JW, Rivecca IN: Measurement of bisulfate in biologic solutions. Clin Chern 15: 997-
1001, 1969.
330

187. Wilkins JW, Greene JA, Weller JM: Toxicity of intraperitoneal bisulfite. Clin Pharm Ther 9:
328--332, 1968.
188. Karanicolas S, Oreopoulos DG, Izatt, S, Shimizu A, Manning RF, Sepp H, de Veber GA, Darby
T: Epidemic of aseptic peJ;itonitis caused by endotoxin during chronic peritoneal dialysis. N Engl
J Med 296: 1336-1337,1977.
189. Mion CM, Boen ST, Scribner P: Analysis of factors responsible for the formation of adhesions
during chronic peritoneal dialysis. Am J Med Sci 250: 675-679, 1965.
190. O'Leary JP, Malik F, Donahoe RR, Johnston AD: The effects of a minidose of heparin on
peritonitis in rats. Surg Gynecol Obstet 148: 571-575, 1979.
191. Bunke CM, Aronoff GR, Luft FC: Pharmacokinetics of common antibiotics used in continuous
ambulatory peritoneal dialysis. Am J Kidney Dis 3: 114-117, 1983.
192. Halstenson CE, Matzke GR, Comty CM: Intraperitoneal administration of tobramycin during
continuous ambulatory peritoneal dialysis (CAPD). Abstr Perit Dial 22: 119A, 1983.
193. Rubin J, Humphries J, Smith G, Bower J: Antibiotic activity in peritoneal dialysate. Amer J of
Kid Disease 3: 205-208, 1983.
194. Guttman RA: Automated peritoneal dialysis for home use. Quarterly J Med 47: 261-280, 1978.
195. Appleby DH, John JF, Jr: Effect of peritoneal dialysis solution on the antimicrobial activity of
cephalosporins. Nephron 30: 341-344, 1982.
196. Glew RH, Pavuk RA: Stability of vancomycin and aminoglycoside antibiotics in peritoneal
dialysis concentrate. Nephron 28: 241-243, 1981.
197. Sewell D, Golper T: Stability of antimicrobial agents in peritoneal dialysate. Antimicrob Agents
and Chemo Mar 21: 3.528-529, 1982.
198. Sewell DL, Golper T A, Brown SD, Nelson E, Knower M, Kimbrough RC: Stability of single
and combination antimicrobial agents in various peritoneal dialysates in the presence of insulin
and heparin. Amer J of Kid Disease III: 3.209-212, 1983.
199. Block RA, Taylor B, Frederick G: Intraperitoneal infusion of streptokinase in the treatment of
recurrent peritonitis: Perit dial bull 3: 162, 1983.
200. Atkins RC, Mion C, Despaux E, Van-Hai N, Julian C, Mion H: Peritoneal transfer of kan-
amycin and its use in peritoneal dialysis. Kidney Int 3: 391-396, 1973.
201. Nielsen HE, Sorensen I, Hensen HE: Peritoneal transport of vancomycin during peritoneal
dialysis. Nephron 24: 274-277, 1979.
202. Rubin J, Oreopoulos DG, Liu IT, Mathews R, deVeber GA: Management of peritonitis and
bowel perforation during chronic peritoneal dialysis. Nephron 16: 220-225, 1976.
203. Black HR, Finkelstein FD, Lee RV: The treatment of peritonitis in patients with chronic
indwelling catheters. Trans ASAIO 20: 115-119, 1974.
204. Rubin J, Rodgers W, Taylor H, Everett E, Prowant B, Fruto L, Nolph K: Peritonitis during
continuous ambulatory peritoneal dialysis. Ann Int Med 92: 1.17-13, 1980.
205. Leigh DA: The treatment of infection in peritoneal dialysis. Br J Hosp Med Sept 12: 389-403,
1974.
206. Golper TA: Peritonitis associated with chronic peritoneal dialysis, a diagnostic and therapeutic
approach. Dial Transpl 7: 1173-1178, 1978.
207. Regamey C, Schabery D, Kirby WMM: Inhibitory effect of heparin on gentamicin concentra-
tions in blood. Antimicrob agents chemother p 329-332, 1972.
208. Furman KL, Gomperts ED, Hockley J: Activity of intraperitoneal heparin during peritoneal
dialysis. Clin nephrol 9: 15-18,1978.
209. Thayssen P, Pindborg T: Peritoneal dialysis and heparin. Scand J Urol Nephrol12: 73-74,1978.
210. Paton TW, Manuel A, Cohen LB, Walker SE: The disposition of cefazolin and tobramycin
following intraperitoneal administration in patients on continuous ambulatory peritoneal di-
alysis. Perit Dial Bull 3: 73, 1983.
211. Loeppky C, Tarka E, Everett ED: Compatibility of cephalosporins and aminoglycosides in
peritoneal dialysis fluid. Perit Dial Bull 3: 128, 1983.
 331

212. Munch R, Steurer J, Luthy R, Siegenthaler W, Kuhlmann U: Serum and dialysate concentra-
tions of intraperitoneal cephalothin in patients undergoing continuous ambulatory peritoneal
dialysis. Clin Nephrol20: 40-43, 1983.
213. Bulger RJ, Bennet JV, Boen ST: Intraperitoneal administration of broad spectrum antibiotics in
patients with renal failure. JAMA 194: 1198-1202, 1965.
214. Perkins RL, Smith EJ, Saslaw S: Cephalothin and cephaloridine: Comparative pharmaco-
dynamics in chronic uremia. Am J Med Sci 257: 116-124,1969.
215. Wilson DE, Chalmers TC, Madoff MA: The passage of cephalothin into and out of ascitic fluid.
Am J Med Sci 93: 449-452, 1967.
216. Sharbaugh RJ, Rambo WM: Cephalothin and peritoneal lavage in the treatment of experimen-
tal peritonitis. Surg Gynecol Obstet 139: 211-214, 1974.
217. Gerding DN, Hall WH: The penetration of antibiotics into peritoneal fluid. Bull N Y Acad Med
51: 1016-1019, 1975.
218. Ritzerfeld VW, Wasterboer S, Trappe H: Cephalexin in serum, dialysat und urin nieren-
geschadigter patienten. Arzneim-Forsch 12: 1881-1884, 1970.
219. Bunke CM, Aronoff GR, Brier ME, Sloan RS, Luft FC: Cefazolin and cephalexin kinetics in
continuous ambulatory peritoneal dialysis. Clin Pharm Ther 33: 66-72, 1983.
220. Singlas E, Boutron HF, Merdjan H, Brocard JF, Pocheville M, Fries D: Moxalactam kinetics
during chronic ambulatory peritoneal dialysis. Clin Pharm Ther 3: 403-407, 1983.
221. Stephens NM, Kronfol NO, Kline BJ, Polk RE: Peritoneal absorption of moxalactam. Anti-
microb Agents Chemother 24: 39-41, 1983.
222. Tourkantonis A, Nicolaidis P: Pharmacokinetics of ceftazidime in patients undergoing per-
itoneal dialysis. J Antimicrob Chemother (Suppl A) 12: 263-267, 1983.
223. Levison ME, Levison SP, Ries K, Kaye D: Pharmacology of cefazolin in patients with normal
and abnormal renal function. J Infect Dis 128: S354-357, 1973.
224. Raine PAM, Courtis GR, Fowler R, Ritchie MA, Hewstone AS: The intraperitoneal use of
cephazolin. I Surg Res 25: 232-235, 1978.
225. Kaye D, Wenger N, Agarwal B: Pharmacology of intraperitoneal cefazolin in patients undergo-
ing peritoneal dialysis. Antimicrob Agents Chemother 14: 318-321, 1978.
226. D'Silva H, Yherberg S, Jasper F, Lasker N: Relative peritoneal clearance of cefazolin. Abstr
Am Soc Nephrol P 40A, 1978.
227. Greaves WL, Kreeft JH, Ogilvie RI, Richards GK: Cefoxitin disposition during peritoneal
dialysis. Antimicrobial Agents & Chemother Feb. 19: 2.253-255, 1981.
228. Heide B, Pierratos A, Khanna R, Pettit I, Ogilvie R, Harrison I, McNeil K, Oreopoulos DG:
Pharmacokinetics of cefamandole in patients undergoing CAPD. Perit Dial Bull 3: 138, 1983.
229. Meyers BR, Hirschman SZ: Pharmacokinetics of cefamandole in patients with renal failure.
Antimicrob Agents Chemother 11: 248-250, 1977.
230. Ahen MJ, Finkelstein FO, Andriole VT: Pharmacokinetics of cefamandole in patients undergo-
ing hemodialysis and peritoneal dialysis. Antimicrob Agents Chemother 10: 457-461, 1976.
231. Local FK, Munro AI, Kerr DNS, Sussman M: Pharmacokinetics of intravenous and intra-
peritoneal cefuroxime in patients undergoing peritoneal dialysis. Clin Neph 16: 1.40-43, 1981.
232. La Greca G, Biasioli S, Chiaramonte S, Fabris A, Feriani M, Pisani E: Pharmacokinetics of
intravenous and intraperitoneal cefuroxime during peritoneal dialysis. Inter J Clin Pharm Thera
Toxicol 20: 2.92-94, 1982.
233. Thomae U, Boos W, Adam D: Transperitoneale resorption von ampicillin, cefuroxim und
gentamicin under kontinuierlicher ambulanter peritonealdialyse. Med Welt 33: 5.182-183,1982.
234. Buck AC, Cohen SL: Absorption of antibiotics during peritoneal dialysis in patients with renal
failure. J Clin Path 21: 88-92, 1968.
235. Shear L, Shinaberger IH, Barry KG: Peritoneal transport of antibiotics in man. N Engl J Med
272: 666-669, 1965.
236. Ruedy I: The effects of peritoneal dialysis on the physiological disposition of oxacillin, am-
332

picillin, tetracycline in patients with renal disease. J Can Med Assoc 94: 257-261, 1966.
237. Deresinski SC, Stevens DA: Clinical evaluation of parenteral dicloxacillin. Curr Ther Res 18:
151-162, 1975.
238. Briedigkeit VH, Schroter R: Antibiotikaspiegelbestimmungen bei Patienten mit Niereninsuf-
fizienz wahrend der Peritonealdialyse. Dtsch Gesundheitswes 23: 337-341, 1968.
239. Thomae U, Boos W, Adam D: Transperitoneale resorption von oxacillin, azlocillin und sis-
omicin unter kontinuierlicher ambulanter peritonealdialyse (CAPD) bei patienten mit und ohne
peritonitis. Med Welt 32: 1365-1367, 1981.
240. Eastwood JB, Curtis JR: Carbenicillin Administration in patients with severe renal failure. Br
Med J 24: 486-487, 1968.
241. Hoffman TA, Cestero R, Bullock WE: Pharmacokinetics of carbenicillin in patients with
hepatic and renal failure. J Infect Dis 122b: S75-77, 1970.
242. Wise R, Reeves DS, Parker AS: Administration ofticarcillin, a new antipseudomonal antibiotic,
in patients undergoing dialysis. Antimicrob Agents Chemother 5: 119-120, 1974.
243. Kampf D, Schurig R, Weihermuller K, Forster D: Effects of impaired renal function, hemo-
dialysis, and peritoneal dialysis on the pharmacokinetics of mezlocillin. Antimicrob Agents
Chemother July 18: 81-87, 1980.
244. Whelton A, Stout RL, Delgado FA: Azlocillin kinetics during extracorporeal haemodialysis and
peritoneal dialysis. J Antimicrob Chemother Suppl. B 11: 89-95, 1983.
245. Malacoff RF, Finkelstein FO, Andriole VT: Effect of peritoneal dialysis on serum levels of
tobramycin and clindamycin. Antimicrob Agents Chemother 8: 574-580,1975.
246. Reinarz JA, McIntosh DA: Lincomycin excretion in patients with normal renal function, severe
azotemia, and with hemodialysis and peritoneal dialysis. Antimicrob Agents Chemother 165:
232-238, 1966.
247. Ayus JC, Eneas JF, Tong TG, Benowitz NL, Schoenfeld PY, Hadley KL, Becker DE,
Humphreys MH: Peritoneal clearance and total body elimination of vancomycin during chronic
intermittent peritoneal dialysis. Clin Nephrolll: 129-132, 1979.
248. Glew RH, Pavuk RA, Shuster A, Alfred HJ: Vancomycin pharmacokinetics in patients under-
going chronic intermittent peritoneal dialysis. Internat J C1in Pharmacol, Thera Toxicol 20:
12.559-563, 1982.
249. Magera RE, Arroyo JC, Rosansky SJ, Postic B: Vancomycin pharmacokinetics in patients with
peritonitis on peritoneal dialysis. Antimicrobial Agents Chemother 23: 710-714, 1983.
250. Pancorbo S, Comty C: Peritoneal transport of vancomycin in 4 patients undergoing continuous
ambulatory peritoneal dialysis. Nephron 31: 37-39, 1982.
251. Krothapalli RK, Senekjian HO, Ayus JC: Efficacy of intravenous vancomycin in the treatment
of gram-positive peritonitis in long-term peritoneal dialysis. The Amer J of Med 75: 345,1983.
252. Harford AM, Sica DA, Tartaglione T, Polk RE, Dalton HP, Poynor WJ: Vancomycin (V)
pharmacokinetics in continuous ambulatory peritoneal dialysis (CAPD ) patients with peritonitis
(P). Abstr The Amer Soc Neph P 120A, 1983.
253. de Martin A, Rottembourg J: Administration intra-peritoneale de sulfamethoxazole-tri-
methoprime dans Ie traitement des infections compliquant les dialyses peritoneales chroniques.
La Nouvelle Presse Medicak Dec 2: 3932, 1978.
254. Adam WR, Brown Dl, Hales P, Dawborn JK: The use of sulphadimidine (sulphamezathine) in
patients with renal failure. Med 1 Aug-May 12: 936-938, 1973.
255. Arfania D, Nolph K, Everett E, Rubin 1: Unusual cases of peritonitis in patients on peritoneal
dialysis. Arch lnt Med 141: 61-64, 1981.
256. Fremont IF, Dkhissi H, Thomas D, Tolani M, Laurence G, Coevoet B, Fournier A, Orfila J:
Traitement des peritonites en dialyse peritonale. Comparison entre lavage continu avec machine
et lavage intermittent par quatre sacs/jour de DPCA. Path Bioi 31: 6.544-547, 1983.
257. lusko WS, Baliah T, Kim KH, Gerbracht LM, Yaffe SI: Pharmacokinetics of gentamicin during
peritoneal dialysis in children. Kidney Int 9: 430-438, 1976.
 333

258. Gary NE: Peritoneal clearance and removal of gentamicin. J Infect Dis 124: S96-97, 1971.
259. Smithivas T, Hyams PJ, Matalon R, Simberkoff MS, Rahal JJ: The use of gentamicin in
peritoneal dialysis. J. Pharmacologic results. J Infect Dis 124: S77-83, 1971.
260. Hyams PJ, Smithivas T, Matalon R, Katz L, SimberkoffMS, RahalJJ: The use of gentamicin in
peritoneal dialysis. II. Microbiologic and clinical results. J Infect Dis 124: S84-89, 1971.
261. Mahon WA, Ezer JL, Fenton SSA, Cattran DC: Studies on gentamicin clearance during
peritoneal and hemodialysis. Abstr Clin Res 21: 1018, 1973.
262. Letter. Gentamicin for treatment of peritonitis in continuous ambulatory peritoneal dialysis.
The Lancet !,ug 22, 1981.
263. Hamann SR, Oeltgen PR, Shank Jr. WA, Blouin RA, Natarajan L: Evaluation of gentamicin
pharmacokinetics during peritoneal dialysis. Thera Drug Monitor 4: 297-300, 1982.
264. Pancorbo S, Comty C: Pharmacokinetics of gentamicin in patients undergoing continuous
ambulatory peritoneal dialysis, Antimicrob Agents Chemother 19: 605-607, 1981.
265. Fuquay D, Koup J, Smith AL: Management of neonatal gentamicin overdosage. J Pediatr 99:
473-476, 1981.
266. Somani P, Shapiro RS, Stockard H, Higgins JT: Unidirectional absorption of gentamicin from
the peritoneum during continuous ambulatory peritoneal dialysis. Clin Pharm Thera 32: 113-121,
1982.
267. Weinstein AJ, Karchmer AW, Moellering RC: Tobramycin concentrations during peritoneal
dialysis. Antimicrob Agents Chemother 4: 432-434, 1973.
268. Bunke CM, Aronoff GR, Brier ME, Sloan RS, Luft FC: Tobramycin kinetics during continuous
ambulatory peritoneal dialysis. Clin Pharm Thera 34: 110--116, 1983.
269. Paton TW, Manuel MA, Walker SE: Tobramycin disposition in patients on continuous ambula-
tory peritoneal dialysis. Perit Dial Bull 2: 179-181, 1982.
270. Keogh JAB, Carr ME, Falkiner FR, Martin P, Grant G, Keane CT: Pharmacokinetics of
netilmicin in CAPD patients. Perit Dial Bull 3: 172-175, 1983.
271. Greenberg PA, Sanford JP: Removal and absorption of antibiotics in patients with renal failure
undergoing peritoneal dialysis. Ann Intern Med 66: 465-470, 1967.
272. Cohn L, Cotlar AM: Intraperitoneal kanamycin. Ann Surg 155: 523-537, 1962.
273. Mullett RD, Keats AS: Apnea and respiratory insufficiency after intraperitoneal administration
of kanamycin. Surgery 49: 530--533, 1961.
274. Madhavan T, YaremchukK, LevinN, PohlodD, Burch K, FisherE, CoxF, QuinnEL: Effect of
renal failure and dialysis on the serum concentration of the aminoglycoside amikacin. Anti-
microb Agents Chemother 10: 464-466, 1976.
275. Regeur L, Colding H, Jensen H, Kampmann JP: Pharmacokinetics of amikacin during hemo-
dialysis and peritoneal dialysis. Antimicrob Agents Chemother 11: 214-218,1977.
276. Matzke GR, Salem N, Bockbrader H, Blevins : The effect of peritoneal dialysis on the
pharmacokinetics of amikacin. Proc Dial Transpl Forum P 302, 1980.
277. Slingeneyer A, Liendo-Liendo C, Despaux E, Balmayer B, Perez C, Mion C, Larmet R:
Transfert peritoneal de l'amikacine son utilisation en dialyse peritoneale de suppleance.
Nouvelle Presse Medicale 8: 3432-3435, 1979.
278. Rose HD, Roth DA, Koch ML: Serum tetracycline levels during peritoneal dialysis, Am J Med
Sci 250: 66-68, 1965.
279. Fabre J, Milek E, Kalfopoulos P, Merier C: Tetracycline kinetics in human being: Excretion,
penetration into normal and inflammatory tissues, behavior in renal insufficiency and hemo-
dialysis. Schweiz Med Wschr 101: 625-633,1971.
280. Moron KC: Peritoneal dialysis in acute poisoning. Clin Pediatr 5: 565-567, 1966.
282. Furman KI, Koomhof HJ, Kilroe-Smith TA, Landless R, Robinson RG: Peritoneal transfer of
thiamphenicol during peritoneal dialysis. Antimicrob Agents Chemother 9: 557-560, 1976.
283. Menz VHP, Hartmann R, V Oldershausen HF: Zur pharmakokinetik von thiamphenicol.
Arzneim-Forsch 24: 102-104, 1974.
334

284. Bush A, Holt IE, Sankey MG, Kaye CM, Gabriel R: Penetration of metronidazole into
continuous ambulatory peritoneal dialysis. Perit Dial Bull 3: 176, 1983.
285. Holten E, Smith-Erichsen N: Concentration of metronidazole in serum during peritoneal
dialysis. Chemother 27: 414-415, 1981.
286. Taylor R, Schofield IS, Ramos JM, Bint AJ, Ward MK: Ototoxicity of erythromycin in
peritoneal dialysis patients. Lancet 2: 935-936, 1981.
287. Goodwin NJ, Friedman EA: The effects of renal impairment, peritoneal dialysis, and hemo-
dialysis on serum sodium colistimethate levels. Ann Int Med 68: 984-994, 1968.
288. MacKay DN, Kaye D: Serum concentrations of colistin in patients with normal and impaired
renal function. N Eng J Med 270: 394-397, 1964.
289. Parisi A, Kaplan MH: Apnea during treatment with sodium colistimethate: JAMA 194: 298-
299,1965.
290. Swick HM, Maxwell E, Charache P, Levin S: Peritoneal dialysis in colistin intoxication. Pediatr
Pharmacol Ther 74: 976-980, 1969.
291. Drouchet E, Babinet P, Chapusot JP, Kleinknechi D: Fluorocytosine in the treatment of
candidiasis with acute renal insufficiency. Biomedicine 19: 408-414, 1973.
292. Andersen KEH, Olsen H: Candida peritonitis in a patient receiving chronic intermittent
peritoneal dialysis. Scand J Infect Dis 10: 91-92, 1978.
293. Holdsworth SR, Atkins RC, Scott DF, Jackson R: Management of Candida peritonitis by
prolonged peritoneal lavage containing 5-fluorocytosine. Clin Nephrol 4: 157-159, 1975.
294. Jones JM, Greenfield RA: Administration offlucytosine to a patient on CAPD. Perit Dial Bull
Jan-Mar 2: 1.46-47, 1982.
295. Lempert KD, Jones JM: Flucytosine-miconazole treatment of Candida peritonitis: Its use during
continuous ambulatory peritoneal dialysis. Arch Intern Med 142: 577, 1982.
296. Peterson LR, Kelty RH, Hall WH, Votava HJ: Therapy of Candida peritonitis: Penetration of
amphotericin B into peritoneal fluid. Postgrad Med J 54: 340--342, 1978.
297. Mandell IN, Ahern MS, Kliger AS, Andriole VT: Candida peritonitis complicating peritoneal
dialysis: Successful treatment with low dose amphotericin B therapy. Clin Nephrol6: 492-496,
1976.
298. Boyer AS, Blumenkrantz MJ, Montgomerie JZ, Coburn JW, Guze LB: Candida peritonitis:
Report of 22 cases. Am J Med 61: 832-840, 1976.
299. Muther RS, Bennett WM: Case reports: Peritoneal clearance of amphotericin Band
5-fluorocytosine. West J of Med 133: 157-160, 1980.
300. McNeely DJ, Vas SI, Dombros N, Oreopoulos DG: Fusarium peritonitis: An uncommon
complication of CAPD. Perit Dial Buill: 6.1981.
301. Hogg RJ, Arant Jr, BS, Houser MT: Candida peritonitis in children on continuous ambulatory
peritoneal dialysis. Intern J Pediatric Nephrol3: 4.287-291, 1982.
302. Letter. Ketoconazole and fungal CAPD peritonitis. Lancet Aug 27: 510--511, 1983.
303. Coco AE, Pazourek LKJ: Acute isoniazid intoxication management by peritoneal dialysis. N
Eng J Med 269: 852-853, 1963.
304. Glogner P, Vogt 0, Lange H: Dialyse bei vergiftung mit isoniazid. Dtsch Med Wochenschr 96:
1307-1309, 1971.
305. Wattel F, Gosselin B, Chopin C, Durocher A: Intoxication massive par isoniazide. La Nouvelle
Presse Medica1e 12: 1134-1135, 1975.
306. Andrew OT, Schoenfeld PY, Hopewell PC, Humphreys MH: Tuberculosis in patients with end-
stage renal disease. Amer J Med 68: 59-65, 1980.
307. Atkins R, Cutting CJ, Mackintosh TF: Acute poisoning by cycloserine. Br Med J April 3: 907-
908,1965.
308. Ing TS, Mahurkar SD, Durea G, Hayashi JA, Dlawans HL, Markey WS: Removal of aman-
tadine hydrochloride by dialysis in patients with renal insufficiency. J Can Med Assoc 115: 515,
1976.
 335

309. Unpublished Observations - author.

310. McCann WP, Permisohn R, Palmisano PA: Fatal chloroquine poisoning in a child: Experience
with peritoneal dialysis. Pediatrics 55: 536-538, 1975.
311. Brooks MH, Hanno JE, Clayton LE, Kazyakl , Barry KG: Quinine extraction during
peritoneal dialysis: The role of nonionic diffusion. Invest Urology 7: 510-516, 1970.
312. McKenzie IFC, Mathew TH, Bailie: Peritoneal dialysis in the treatment of quinine overdose.
Med J Aust 1: 58-59, 1968.
313. Markham TN, Dodson VN, Eckberg DL: Peritoneal dialysis in quinine sulfate intoxication.
JAMA 202: 128-129,1967.
314. Donadio JV, Whelton A, Gilliland PF, Cirksena WJ: Peritoneal dialysis in quinine intoxication.
JAMA 204: 274, 1968.
315. Held H: Effectiveness of peritoneal dialysis in the treatment of quinine poisoning. Dtsch Med
Wschr 97: 1793-1795, 1972.
316. Donadio JV, Whelton A, Kazyak L: Quinine therapy and peritoneal dialysis in acute renal
failure complicating malarial haemoglobinuria. Lancet Feb 24 1: 375-379, 1968.
317. Sabto J, Pierce RM, West RH, Gurr FW: Hemodialysis, peritoneal dialysis, plasmapheresis and
forced diuresis for the treatment of quinine overdose. Clin Nephrol16: 5.264-268, 1981.
318. Yee E, Foss K, Schmidt RW: Use of povidone iodine in continuous ambulatory peritoneal
dialysis (CAPD) - a technique to reduce the incidence of infectious peritonitis. Trans ASAIO
XXVI: 223-224, 1980.
319. Ahrenholz DH, Simmons RL: Povidone iodine in peritonitis. J Surg Res 26: 458-463, 1979.
320. Lavigne JE, Brown CS, Machiedo GW, Blackwood JM, Rush BF: The treatment of experimen-
tal peritonitis with intraperitoneal betadine solution. J Surg Res 16: 307-311, 1974.
321. Gilmore OJA, Reid C, Houang E, Sham EJ: Intraperitoneal povidone iodine in peritonitis. J
Surg Res 25: 471-476, 1978.
322. Stephen RL, Kablitz C, Kitahara M, Nelson J, Duffy DP, Kolff W: Peritoneal dialysis: Per-
itonitis: Saline-iodine flush. Dial Transpl8: 584-595, 1979.
323. Browne MK, MacKenzie M, Doyle PJ: A controlled trial of taurolin in established bacterial
peritonitis. Surg Gynecol Obstet 146: 721-724, 1978.
324. Migliori G, Codinach F, Marsan A, Inglesakis JA: Resultats de l'utilisation de la noxythioline au
cours des peritonites aigues. Ann Anesth Franc XIX: 902-907, 1978.
325. Letter. Somerville PJ, Kaye M: Chlorhexidine is unsuitable for long-term use in patients on
CAPD. Perit Dial Bull 2: 4.195, 1982.
326. Fabbri L, Grimaldi C, Zucchelli P: Peritonitis in CAPD: treatment with chlorhexidine. Dial
Transplant June 11: 483-486, 1982.
327. Buckle RC, Seabridge CE: The effect of chlorhexidine in the peritoneal cavity: An experimental
study. Lancet 1: 193, 1963.
328. Somerville, PJ, Kaye M: Chlorhexidine is unsuitable for long-term use in patients on CAPD.
Perit Dial Bull 2: 4.195,1982.
329. Winchester JF, Gelfond MC, Knepshield JH, Schreiner GE: Dialysis and hemoperfusion of
poisons and drugs-update. Trans ASAIO XXIII: 762-842, 1977.
330. Etteldorf IN, Dobbins WT, Summitt RL, Rainwater WT, Fischer RI: Intermittent peritoneal
dialysis using 5 per cent albumin in the treatment of salicylate intoxication in children. J Pediatr
58: 226-236, 1961.
331. Etchart LW: Peritoneal dialysis in salicylate intoxication. Rocky Mountain Med J Nov 55-56,
1965.
332. Halle MA, Collipp PJ: Treatment of methyl salicylate poisoning by peritoneal dialysis. N York
State J Med June 151788-1789, 1969.
333. Kloss JL, Boeckman CR: Methyl salicylate poisoning. Ohio State Med J 63: 1064-1065, 1967.
334. Schlegel RJ, Alstatt LB, Canales L, Goiser JL, Alexander JL, Gardner LJ: Peritoneal dialysis
for severe salicylism: An evaluation of indications and results. J Pediatr 69: 553-562. 1966
336

335. Buselmeir TJ, Mauer SM, Merino GE, Rodrigo F, Simmens RL, Najarian JS, Kjellstrand CM:
Therapy of severe salicylate intoxication. Current topics in Critical Care Medicine 3rd Int Symp
Rio de Janeiro, Karger Basel, 64-70, 1974.
336. Nahas FF, Giroux 11, Gjessing J, Verosky M, Mark LC: The use of tham in peritoneal dialysis.
Trans ASAIO Organs 20: 345, 1964.
337. MacLean D, Peters TJ, Brown RAG, McCathie M, Barnes GF, Robertson PGC: Treatment of
acute paracetamol poisoning. Lancet Oct 19: 849-852, 1968.
338. Gary NE, Maher JF, DeMyttenacre MH, Liggero SH, Scott KG, Matusiak W, Schreiner GE:
Acute propoxyphene hydrochloride intoxication. Arch Intern Med 121: 453-457, 1968.
339. Karliner JS: Propoxyphene hydrochloride poisoning. JAMA 199: 152-155, 1967.
340. McCarthy WH, Keenan RL: Propoxyphene hydrochloride poisoning. JAMA 187: 460-461,
1964.
341. Robinson RR, Gunnells JC, Clapp JR: Treatment of acute barbiturate intoxication. Modern
Treat 8: 3.461-479, 1971.
342. Hadden J, Johnson K, Smith S, Price L, Giardine E: Acute barbiturate intoxication. JAMA 209:
893-900, 1969.
343. Setter JG, Maher JF, Schreiner GE: Barbiturate intoxication. Arch Intern Med 117: 224-236,
1966.
344. Whiting EG, Barnett 0, Inmon TW: Treatment of barbiturate poisoning. Calif Med 102: 367-
369, 1965.
345. Henderson LW, Merrill JP: Treatment of barbiturate intoxication. Ann Intern Med 64: 876-891,
1966.
346. Knochel JP, Clayton LKE, Smith WL, Barry KG: Intraperitoneal tham: An effective method to
enhance phenobarbital removal during peritoneal dialysis. J Lab Clin Med 64: 257-268, 1964.
347. Setter JG, Freeman RB, Maher JF, Schreiner GE: Factors influencing the dialysis of barbitur-
ates. Trans ASAIO X: 340-344, 1964.
348. Berman LB, Vogelsang P: Removal rates for barbiturates using two types of peritoneal dialysis.
N Eng J Med 270: 77-80, 1964.
349. Campton DS, North JDK: Effect of protein binding of barbiturates on their rate of removal
during peritoneal dialysis. J Lab Clin Med 66: 549-563, 1965.
350. Exaire E, Trevino-Beccera A, Moneon F: An overview of treatment with peritoneal dialysis in
drug poisoning. Contrib Nephrol17: 39-43, 1979.
351. Rosenbaum JL, Mandanas R: Treatment of phenobartitial intoxication in dogs with an anion-
recirculation peritoneal dialysis technique. Trans ASAIO 13: 183-189,1967.
352. Lash RF, Burdette JA, Ozdil T: Accidental profound hypothermia and barbiturate intoxication.
JAMA 201: 123-124, 1967.
353. Fell RN, Gunning AJ, Triger DR: Severe hypothemia as a result of barbiturate overdose
complicated by cardiac arrest. Lancet Feb 24: 392-394, 1968.
354. Ilhan OA, Tannenberg AM: Peritoneal and hemodialysis for acute glutethimide overdosage. N
York State J Med Aug 15: 2076-2079, 1972.
355. DeMyttenaere M, Schoenfeld L, Maher JF: Treatment of glutethimide poisoning. JAMA 203:
885-887,1968.
356. Vhartitzsch B, Pinto MH, Mauer SM, Casali RE, Stem EL, Buselmeier TJ, Kjellstrand CM:
Treatment of glutethimide intoxication: An in vivo comparison of lipid, aqueous and peritoneal
dialysis with albumin. Proc Dial Transpl Forum P 102-106, 1973.
357. McDonald DF, Greene WM, Kretchmar L, O'Brien G: Experiences in acute gluthethimide
(Doriden) intoxication. Invest Urology 1: 127-133, 1963.
358. Polin RA, Henry D, Pippinger CE: Peritoneal dialysis for severe methyprylon intoxication. J
Pediatr 90: 831-833, 1977.
359. Collins JM: Peritoneal dialysis for methyprylon intoxication. J Pediatr 92: 519-520, 1978.
360. Maddock RK, Bloomer HA: Meprobamate overdosage. JAMA 201: 123-127, 1967.
 337

361. Mouton DE, Cohen RJ, Barrett 0: Meprobamate poisoning: Successful treatment with per-
itoneal dialysis. Am J Med Sci 706-709,1967.
362. Castell DO, Sode J: Meprobamate intoxication treated with peritoneal dialysis. III Med J P 298-
299,1967.
363. Hardy W, Pas AT, Nixon RK: Meprobamate intoxication treated by peritoneal dialysis. Henry
Ford Hosp Med Bull 11: 347-350, 1963.
364. Teehan BP, Maher JF, Carey JJH, Flynn PD, Schreiner GE: Acute ethclorvynol (Placidyl)
intoxication. Ann Intern Med 72: 875-882, 1970.
365. Schultz JC, Crowder DG, Medart WS: Excretion studies in ethclorvynol (Placidyl) intoxication.
Arch Intern Med March 117: 409-411, 1966.
366. Proudfoot AT, Noble J, Nimmo J, Brown SS, Cameron JC: Peritoneal dialysis and hemodialysis
in methaqualone (Mandrax) poisoning. Scott Med J 13: 232-235, 1968.
367. DeMarrco V, Bear R, Kapur BM: Peritoneal dialysis in methaqualone overdose: J Can Med
Assoc 113: 823,1975.
368. Espelin DE, Done AK: Amphetamine poisoning. N Eng J Med 278: 1361-1365, 1968.
369. Zalis EG, Cohen RJ, Lundberg GD: Use of peritoneal dialysis in experimental amphetamine
poisoning. Proc Soc Exp Bioi Med 12: 278-281, 1965.
370. Wallace HE, Neumayer F, Gutch CF: Amphetamine poisoning and peritoneal dialysis. Am J
Dis Child 108: 657-661, 1964.
371. Farquharson S: Poisoning by tricyclic drugs. Br Med J Feb 5: 378, 1972.
372. Royds RB, Knight AH: Tricyclic antidepressant poisoning. Practitioner 204: 282-286, 1970.
373. Sunshine P, Yaffe SJ: Mitriptyline poisoning. Am J Dis Child 106: 501-506, 1963.
374. Oreopoulos DG, Lar S: Recovery from massive amitriptyline overdosage. Lancet July 27: 221,
1968.
375. Lipkin D, Kushnick T: Pargyline hydrochloride poisoning in a child. JAMA 201: 135-136, 1967.
376. Setter JG, Singh R, Brackett NC, Randall RE: Studies on the dialysis of methanol. Trans
ASAIO XIII: 178-182, 1967.
377. Keyvan-Larijarn H, Tannenberg AM: Methanol intoxication. Arch Intern Med 134: 293-296,
1974.
378. Keyvan-Larijarni H, Tannenberg AM: Comparison of peritoneal and hemodialysis treatment of
methanol intoxication. Proc Dial Transpl Forum P 107-109, 1973.
379. Wenzl JE, Mills SD, McCall JT: Methanol poisoning in an infant. Am J Dis Child 116: 445-457,
1968.
380. Stinebaugh B: The use of peritoneal dialysis in acute methyl alcohol poisoning. Arch Intern Med
105: 613-617, 1960.
381. Humphrey TJ: Methanol poisoning: Management of acidosis with combined haemodialysis and
peritoneal dialysis. Med J Aust 833-835, 1974.
382. Vale JA, Widdop B, Bluett NH: Ethylene glycol poisoning. Postgrad Med J 52: 598-602, 1976.
383. Joly JB, Iluault G, Frossard C, Fabiani P, Thieffry S: Intoxication aigue par I'ethylene-glycol.
Societe MedicaIe des h6pitaux de Paris 119: 27-45, 1968.
384. Zarembski PM, Rosen SM, Hodgkinson A: Dialysis in the treatment of primary hyperoxaluria.
Br J Urology 41: 530-533, 1969.
385. Dua SL: Peritoneal dialysis for isopropyl alcohol poisoning. JAMA 230: 35, 1974.
386. Grubbauer HM, Schwartz R: Peritoneal dialysis in alcohol intoxication in a child. Arch Toxicol
43: 317-320, 1980.
387. Covey TJ: Ferous sulfate poisoning. J Pediatr 64: 218-226, 1964.
388. Lavender S, Bell JA: Iron intoxication in an adult. Br Med J May 16: 406, 1970.
389. Greengard J: Iron poisoning in children. Clin Toxicol8: 575-597, 1975.
390. Batson R, Petersen JC: Acute mercury poisoning: Treatment with BAL an in anuric states with
continuous peritoneal lavage. Ann Intern Med 29: 278-293, 1948.
391. Robillard JE, Rames LK, Jensen RL, Roberts RJ: Peritoneal dialysis in mercurial diuretic
intoxication. J Pediatr 88: 79--81, 1976.
338

392. Kahn A, Denis R, Blum D: Accidental ingestion of mercuric sulphate in a 4-year-old child. Clin
Pediatr 16: 956-958, 1977.
393. Aronow R, Fleischmann LE: Mercury Poisoning in children. Clin Pediatr 15: 936-945, 1976.
394. Lowenthal DT, Chardo F, Reidenberg MM: Removal of mercury by peritoneal dialysis. Arch
Intern Med 134: 139-141, 1974.
395. Yolken R, Konecny P, McCarthy P: Acute fluoride poisoning. Pediatrics 58: 90-93, 1976.
396. Schiff! H, Weidmann P, Weiss M, Massry SG: Dialysis treatment of acute chromium intoxication
and comparative efficacy of peritoneal versus hemodialysis in chromium removal. Mineral
Electrolyte Metab 7: 28-35, 1982.
397. Wilson ID editor; Editorial-Thallium: Lancet Sept 2: 564-565,1974.
398. Koshy KM: Thallium ingestion with survival: Ineffectiveness of peritoneal dialysis and po-
tassium chloride diuresis. Clin Toxicol18: 5.521-525, 1981.
399. Stanbaugh Jr GH, Holmes AW, Gillit D, Reichel GW, Stranz M: Iron chelation therapy in
CAPD: A new and effective treatment of iron overload disease in ESRD patients. Perit Dial
Bulletin Apr-June 3: 2.99-101, 1983.
400. Falk RJ, Mattern WD, Lamanna RW, Gitelman HJ, Parker NC, Cross RE, Rastall JR: Iron
removal during continuous ambulatory peritoneal dialysis using desferoxamine. Kidney Int 24:
110-112, 1983.
401. Kaufman DB, DiNicola S, McIntosh R: Acute potassium dichromate poisoning. Am J Dis Child
119: 374-376, 1970.
402. Chugh KS, Sharma BK, Singhal PC, Das KC, Datta BN: Acute renal failure following copper
sulphate intoxication. Postgrad Med J 53: 18-23,1977.
403. Hamlyn AN, Gollan JL, Douglas AP, Sherlock S: Fulminant Wilson's disease with haemolysis
and renal failure. Copper studies and assessment of dialysis regimens. Br Med J 2: 660-663,
1977.
404 Cole DE: Peritoneal dialysis for removal of copper. Br Med J 7: 50-51, 1978.
405. Cole DEC, Lirenman DS: Role of albumin-enriched peritoneal dialysate in acute copper
poisoning. J Pediatr 92: 955-957, 1978.
406. Mehbod H: Treatment of lead intoxication. JAMA 201: 152-154,1967.
407. Miroutze J, Mion C, Mathieu-Daude P, Monnier L, Selam JL: Nephropathie chronique au cours
d'un saturnisme. Nouv Presse Med 31: 1642-1644, 1975.
408. Combs RJ, Dentino MM, Lehrman Swed 11: Gold toxicity and peritoneal dialysis. Arthritis
Rheum 19: 936-938, 1976.
409. Dziuk E: Experimental studies on removal of cesium-137 by peritoneal dialysis. Polish Med J III:
6.1596-1601, 1972.
410. Ackerman GL, Doherty JE, Flanigan WJ: Peritoneal dialysis and hemodialysis of tritiated
digoxin. Ann Intern Med 67: 718-723, 1967.
411. Kramer P, Quellhorst E, Horenkamp J, Scheler F: Dialysance und prozentuale elimination
verschiedener herzglykoside wahrend der hamo- und peritonealdialyse. Klin Wschr 50: 609-613,
1972.
412. Doherty JE, Bissett JK, Kane 11, de Soya N, Murphy ML, Flanigan WJ, Dalrymple GV:
Tritiated digoxin: Studies in renal disease in human subjects. Int J Clin Pharmacol12: 89-95,
1975.
413. Gloor HJ, Moore H, Nolph KD: The peritoneal handling of digoxin during CAPD. Perit Dial
Bull 2: 13, 1982.
414. De Paepe M. Belpaire F, Bogaerts Y: Pharmacokinetics of digoxin in CAPD. Clin Exper Dial
Apheresis 6: 65-73, 1982.
415. ViII aba-Pimentel L, Epstein LM, Sellers EM, Foster JR, Bennion LJ, Naoler LM, Nough EW,
Koch-Weser J: Survival after massive procainamide infestion. Am J Cardiol32: 727-730, 1973.
416. Hall K, Meatherall B, Krahn J, Brian P, Rabson JL: Clearance of quinidine during peritoneal
dialysis. Amer Heart J 104: 646-647, 1982.
 339

417. Chin TWF, Pancorbo S, Compty e: Quinidine pharmacokinetics in continuous ambulatory

peritoneal dialysis. Clin Exper Dial Apheresis 5: 391-397, 1981.
418. Parrott KA: Determination of propranolol in peritoneal dialysis fluid by high-performance
liquid chromatography without extraction. J Chromatography 274: 171-178, 1983.
419. Grefberg N, Nilsson P, Andreen T: Sclerosing obstructive peritonitis, beta-blockers, and contin-
uous ambulatory peritoneal dialysis. Lancet 24: 733-734, 1983.
420. Spital A, Scandling JD: Nifedipine in continuous ambulatory peritoneal dialysis. Arch Intern
Med Oct 143: 2025,1983.
421. Harris L, McKenna WJ, Krikler SJ, Holt DW, Hind CRK, Savage C, Storey GCA: Renal
elimination of amiodarone and its desethyl metabolite. Postgrad Med J July 59: 440-442, 1983.
422. Yeh BK, Dayton PG, Waters We: Removal of alpha-methyldopa in man by dialysis. Proc Soc
Exp Bioi Med 135: 840-843, 1970.
423. Sellers EM, Koch-Weser J: Protein binding and vascular activity of diazoxide. N Engl J Med 281:
1141-1145,1969.
424. Roots I, Hilderbrandt AG, Baethke R, Glass V, Motzahn M: Influence of haemo- and per-
itoneal dialysis on pharmacokinetics of diazoxide nauyn-schmiedsbergs. Arch Pharm 287: S89,
1975.
425. Emonds ASG, Driessen OMJ: Treatment of theophylline intoxication: A model study utilizing
peritoneal dialysis. Clin Toxicol13: 505-511, 1978.
426. Miceli IN, Bidani A, Aronow R: Peritoneal dialysis of theophylline. Clin Toxicol14: 539-544,
1979.
427. Miceli IN, Clary B, Fleischmann LE, Sarnaik AP, Aronow R, Done A: Pharmacokinetics of
severe theophylline intoxication managed by peritoneal dialysis. Dev Pharmacol Ther 1: 16-25,
1980.
428. Lee CC, Peterson JC, Marbury TC: Comparative pharmacokinetics of theophylline in per-
itoneal dialysis and hemodialysis. J Clin Pharmacol 23: 274-280, 1983.
429. Weinberger M, Hendeles L: Role of dialysis in the management and prevention of theophylline
toxicity. Dev Pharmacol Ther 1: 26-30, 1980.
430. Blair AAD, Hallpike JF, Lascelles PT, Wingate DL: Acute diphenylhydantoin and primidone
poisoning treated by peritoneal dialysis. J Neurol Neurosurg Psychiat 31: 520-523, 1968.
431. Tenckhoff H, Sherrard DJ, Hickman RO, Ladda RL: Acute diphenylhydantoin intoxication.
Am J Dis Child 116: 422-425, 1968.
432. Lindahl D, Westerling : Detoxification with peritoneal dialysis and blood exchange after
diphenylhydantoin intoxication. Acta Paediatr Scand 71: 665-666, 1982.
433. Orr JM, Abbott FS, Ferguson S, Godolphin WJ: The effects of peritoneal dialysis on the single
dose and steady state pharmacokinetics of valproic acid in a uremic epileptic child. Eur J Clin
Pharmacol 24: 387-390,1983.
434. Knapp RC, St. John E, Bast Jr RC: A review of intraperitoneal therapy of human ovarian
carcinoma. Perit Dial Bull 3: 59-62, 1983.
435. Dedrick RL, Myers CE, Bungay PM, DeVita Jr VT: Pharmacokinetic rationale for peritoneal
drug administration in the treatment of ovarian cancer. Can Treat Reports 62: 1-6, 1978.
436. Wallach RC, Kabakow B, Blinick G: Intravenous chemotherapy with concurrent peritoneal
dialysis: Continued treatment of ovarian carcinoma after drug toxicity. Gynecol Oncol4: 299-
306,1976.
437. Levin DH, Blinick G, Kabakow B, Bernstein Z, Barash P, Antopol W: The fate of double-
labeled radio-active thiotepa (S P32) following intravenous administration and in combination
with peritoneal dialysis in patients with advanced metastatic ovarian cancer. Obstet Gynecol27:
598-599, 1966.
438. Ahmad S, Shen F, Bleyer WA: Methotrexate-induced renal failure and ineffectiveness of
peritoneal dialysis. Arch Intern Med 138: 1146-1147, 1978.
439. Jones RB, Collins JM, Myers CE, Brooks AE, Hubbard SM, Balow JE, Brennan MF, Dedrick
340

RL, DeVita VT: High-volume intraperitoneal chemotherapy with methotrexate in patients. Jan
Can Res 41: 55-59, 1981.
440. Chin KN, Hudson G: Ultrastructural changes in murine peritoneal cells following
cyclophosphamide administration. Br J Exp Path 55: 554--561, 1974.
441. Chin KN, Hudson G: Quantitative study of the effects of cyclophosphamide on peritoneal cell
populations. Acta Haemat 48: 239-244, 1972.
442. Speyer JL, Collins JM, Dedrick RL, Brennan MF, Buckpitt AR, Londer H, DeVita VT, Myers
CE: Phase I and pharmacological studies of 5-fluorouracil administered intraperitoneally.
Cancer Research 40: 567-572, 1980.
443. Ozols RF, Young RC, Speyer JL, Sugarbaker PH, Greene R, Jenkins J: Phase I and phar-
macological studies of adriamycin administered intra peritoneally to patients with ovarian can-
cer. Cancer Research 42: 4266--4269,1982.
444. Howell SB, Pfeifle CL, Wung WE, Olshen RA, Lucas WE, YonJL, Green M: Ann Intern Med
97: 845-851, 1982.
445. Hernandez E, Rosenshein NB, Torrcs JP, Villar J, Grumbine FC, Ettinger DS, Dillon MB,
Klein JL, Order SE: IP immunotherapy and chemotherapy in advanced epithelial ovarian
cancer. Can Treat Reports 66: 1981-1982, 1982.
446. Bast Jr RC, Berek IS, Obrist R, Griffiths CT, Berkowitz RS, Hacker NF, Parker L: Intra-
peritoneal immunotherapy of human ovarian carcinoma with corynebacterium parvum. Can
Research March 43: 1395-1401,1983.
447. Bast RC, Knapp RC, Mitchell AK, Thurston IG, Tucker RW, Schlossman SF: Immunotherapy
of a murine ovarian carcinoma with crynebacterium parvum and specific heteroantiserum. 1
Immunol123: 1945-1951, 1979.
448. Milmen N, Christensen E: Elimination of diatrizoate by peritoneal dialysis in renal failure. Acta
Radiol (Diagn) 15: 265-272, 1974.
449. Brooks HM, Barry KG: Removal of iodinated contrast material by peritoneal dialysis. Nephron
12: 10--14, 1973.
450. Ackrill P, McIntosh CS, Nimmon C, Baker LRI, Cattell WR: A comparison of the clearance of
urographic contrast medium (sodium diatrizoate) by peritoneal and haemodialysis. Clin Sci
Molec Med 50: 69-74, 1976.
451. Lavelle Kl, Doedens Dl, Kleit SA, Forney RB: Iodine absorption in burn patients treated
topically with povidone-iodine. Clin Pharmacol Ther 17: 355-362, 1975.
452. Wainer E, Boner G, Lubin E, Rosenfeld IB: Clearance of Tc-99m DTPA in hemodialysis and
peritoneal dialysis: Concise communication. 1 Nuclear Med 22: 768--771, 1981.
453. Karimeddini MK: Ga-67 Scanning during peritoneal dialysis. 1 Nucl Med May 22: 479-480, 1981.
454. Finberg L, Kiley IE, Luttrell CN: Mass accidental salt poisoning in infancy: A study of a hospital
disaster. JAMA 184: 187-190, 1963.
455. Cundy T, Trafford lA: Efficacy of peritoneal dialysis in severe thiazide-induced hyponatraemia.
Postgrad Med 1 Nov 57: 734--735, 1981.
456. Schmitt GW, Maher JF, Schreiner GE: Ethacrynic acid enhanced bromuresis: A comparison
with peritoneal and hemodialysis. 1 Lab Clin Med 68: 913-922, 1966.
457. Rumpf KW, Fuchs CH, Poser W, Scheler F: Bromureide Intoxication with acute renal failure
and bilateral deep vein thrombosis. Med Klin 72: 993-997, 1977.
458. Humbert G, Fillastre JP, Leroy 1, Maitrot B, Tobelem G, Leroux G, Lavoine A: Intoxication
par Ie lithium. Sem Hop Paris 8 Fevrier 509-514. 1974.
459. Wilson JHP, Donker AJM, VanDerllem G K, Wientjes J: Peritoneal dialysis for lithium poison-
ing. Br Med J 2: 749-750, 1971.
460. Brown EA, Pawlikowski TRB: Lithium intoxication treated by peritoneal dialysis. British J Clin
Pract 35: 90--91,1981.
461. Vaziri NB, Ness RL, Barton CH: Peritoneal dialysis clearance of cimetidine. Am J Gastroen-
terol 71: 572-576, 1979.
 341

462. Kogan FJ, Sampliner RE, Mayersohn M, Kazama RM, Perrier D, Jones W, Michael UF:
Cimetidine disposition in patients undergoing continuous ambulatory peritoneal dialysis. J Clin
Pharmacol 23: 252-256, 1983.
463. Hyneck ML, Murphy JF, Lipschutz DE: Cimetidine clearance during intermittent and chronic
peritoneal dialysis. Am J Hosp Pharm 38: 1760--1762, 1981.
464. Pizzella KM, Moore MC, Schultz RW, Walshe J, Schentag 11: Removal of cimetidine by
peritoneal dialysis, hemodialysis, and charcoal hemoperfusion. Ther Drug Monit 2: 273-281,
1980.
465. Patton TW, Manuel A, Walker SE: Cimetidine disposition in patients on continuous ambulatory
peritoneal dialysis. Perit Dial Bull 2: 73-76, 1982.
466. Graw RG: Chlorpropamide intoxication-treatment with peritoneal dialysis. Pediatrics 45: 106-
109,1970.
467. Black WD, Acchiardo SR: Acetohexamide hypoglycemia: treatment by peritoneal dialysis.
Shouth Med J 70: 1240--1241, 1977.
468. Skoutakis VA, Black WD, Acchiardo SR, Wood GC: Peritoneal dialysis in the treatment of
acetohexamide induced hypoglycemia. Am J Hosp Pharm 34: 68-70, 1977.
469. Fish SS, Pancorbo S, Berkseth R: Pharmacokinetics of epsilon-aminocaproic acid during per-
itoneal dialysis. J Neurosurg 54: 736-739, 1981.
470. Nagaratnam N, Alagaratnam K, Thambapillai AJ, Wijemanne HSR: Acute renal failure follow-
ing potassium ferrocyanide poisoning treated with peritoneal dialysis. Forensic Sci 4: 87-89,
1974.
471. Aviram A, Pfau A, Czaczkes JW, Ullmann TD: Hyperosmolality with hyponatremia: Caused by
inappropriate administration of mannitol. Am J Med 42: 648--650, 1967.
472. Lowenstein E, Goldfine C, Flacke WE: Administration of gallamine in the presence of renal
failure-reversal of neuromuscular blockade by peritoneal dialysis. Anesthesiology 33: 556-558,
1979.
473. Meadow SR, Leeson GA: Poisoning with delayed-release tablets. Arch Dis Child 49: 310--312,
1974.
474. Diekmann L, Hosemann R, Dibbern HW: Pheniramin (avil) intoxication bei einem kleinkind.
Arch Toxikol29: 317-234, 1972.
475. Steyn DG: The treatment of cases of amanita phalloides and amanita capensis poisoning. South
Afr Med J 40: 405-406, 1966.
476. Costantino D, Damia G: l'intoxication phalloidienne. Nouv Presse Med 25: 2315-2317, 1977.
477. Myler RK, Lee JC, Hopper J: Renal tubular necrosis caused by mushroom poisoning. Arch
Intern Med 114: 196-204,1964.
478. Gurr FW: Eucalyptus oil poisoning treated by dialysis and mannitol infusion. Australasian Ann
Med 14: 238-249, 1965.
479. Kann VV, Burgermeister S, Wawschinek 0: Standarddisierte forcierte Diurese und Peri-
tonealdialyse in der Behandlung einer Alkylphosphatvergiftung. Wiener Medizin
Wochenschrift Supplement 129: 54-60, 667-669, 1979.
480. Levinsky WJ, Smalley RV, Hillyer PJ, Shindler RI: Arsine hemolysis. Environ Health 20: 436-
440,1970.
481. Martin GL: Asymptomatic boric acid intoxication, value of peritoneal dialysis. N York State J
Med 71: 1842-1844, 1971.
482. Baliah T, MacLeish H, Drummond KN: Acute boric acid poisoning. Can Med Assoc J 101: 166-
168,1969.
483. Song LC, Heimback MD, Truscott DR, Duncan BD: Boric acid poisoning. Can Med AssocJ 90:
1018-1023, 1964.
484. Segar WE: Peritoneal dialysis in the treatment of boric acid poisoning. N Eng J Med 262: 798-
800,1960.
485. Thomas BB: Peritoneal dialysis and lysol poisoning. Br Med J 3: 720, 1969.
486. Boehm RM, Czajka PA: Hexachlorophene poisoning and the ineffectiveness of peritoneal
342

dialysis. Clin Toxicol14: 257-262, 1979.

487. Klendshoj NC, Burke WJ, Anthone R, Anthone S: Chlorate poisoning. JAMA 180: 1133-1134,
1962.
488. Knight RK, Trounce JR, Cameron JS: Suicidal chlorate poisoning treated with peritoneal
dialysis. Br Med J 3: 601-602, 1967.
489. White N, Snowden SA, Parsons V, Sheldon J, Bewick M: The management ofterminal renal
failure in diabetic patients by regular dialysis therapy. Nephron 11: 261-275, 1973.
490. Mitchell JC, Frohnert PP, Kurtz SB, Anderson CF: Chronic peritoneal dialysis in juvenilel-
onset diabetes mellitus. Mayo Clin Proc 53: 775-781,1978.
491. Flynn CT, Nanson JA: Intraperitoneal insulin in diabetes. Lancet 2: 591,1979.
492. Vrossley K, Kjellstrand CM: Intraperitoneal insulin for control of blood sugar in diabetic
patients during peritoneal dialysis. Br Med J 30: 269-270, 1971.
493. Flynn C, Nanson JA: Intraperitoneal insulin with CAPD: An artificial pancreas. Trans ASAIO
XXV: 114--117, 1979.
494. Twardowski ZJ, Nolph KD, McGary TJ, Moore HL: Nature of insulin binding to plastic bags.
Am J Hosp Pharm 40: 579-582, 1983.
495. Twardowski ZJ, Nolph KD, McGary TJ, Moore HL: Influence of temperature and time on
insulin absorption to plastic bags. Am J Hosp Pharm 40: 583-586, 1983.
496. Rottembourg J, el Shahat Y, Agrafiotis A, Thuillier Y, de Groc F, Jacobs C, Legrain M:
Continuous ambulatory peritoneal dialysis in insulin-dependent diabetic patients: A 40-month
experience. Kidney Int 23: 4~5, 1983.
497. Amidon G, Reichert JE, Johnson CA: Absorption of insulin to the polyvinyl chloride surface of
CAPD solution containers. Proc Dial Transpl Forum, P 296, 1980.
498. Rubin J, Oreopoulos DG, Blair RDG, Chisholm LDJ, Meema HE, De Veber GA: Chronic
peritoneal dialysis in the management of diabetics with terminal renal failure. Nephron 19: 265--
270,1977.
499. Greenblatt DL: Insulin sensitivity in renal failure fatal hypoglycemia following dialysis. NY
State J Med 14: 1040--1041, 1974.
500. Roscoe JM: Practices of insulin administration in CAPD. Perit Dial Bull (Suppl) 2: 27-31, 1982.
501. AmairP, KhannaR, Leibel B, PierratosA, Vas S, MeemaE, ZinggW, Digenis G, Oreopoulos
D: Continuous ambulatory peritoneal dialysis in diabetics with end-stage renal disease. New
Engl J Med 18 306: 625--630, 1982.
502. Putnam WS, Anderson DK, Hanks JB, Jones RS, Lebovitz HE: The effects of portal versus
peripheral insulin administration. Abstr Am Diabetic Assoc, P 59,1980.
503. Schade DS, Eaton RP: The peritoneum- a potential insulin delivery route for a mechanical
pancreas: Diabetes Care 3(2): 229-234,1980.
504. Shapiro DJ, Blumenkrantz MJ, Levin SR, Coburn JW: Absorption and action of insulin added
to peritoneal dialysate in dogs. Nephron 23: 174--180, 1979.
505. Wideroe TE, Smebyy LC, Berg KJ, Jorstad S, Svartas TM: Intraperitoneal (1251) insulin
absorption during intermittent and continuous peritoneal dialysis. Kidney Int 23: 22-28, 1983.
506. Beardsworth SF, Goldsmith HJ, Stanbridge BR: Intraperitoneal insulin cannot correct the
hyperlipidemia of CAPD. Perit Dial Bull 3: 126-127,1983.
507 . Raj a RM, Krasnoff SO, Moros J G, Kramer MS, Rosenbaumm JL: Repeated peritoneal dialysis
in treatment of heart failure. JAMA 28 213: 2268, 1970.
508. Shapira J, Lang R, Jutrin I, Robson M, Ravid M: Peritoneal dialysis in refractory congestive
heart failure. Part I: Intermittent peritoneal dialysis. Perit Dial Bull 3: 130, 1983.
509. Robson M, Biro A, Knobel B, Schai G, Ravid M: Peritoneal Dialysis in refractory congestive
heart failure Part II: Continuous ambulatory peritoneal dialysis. Perit Dial Bull 3: 133, 1983.
510. Mailloux LV, Swartz CD, Onesti GO et al: Peritoneal dialysis for refractory congestive failure.
JAMA 199: 873--878,1967.
511. Cairns KB, Porter GA, Kloster FE et al.: Clinical and hemodynamic results of peritoneal
dialysis for severe cardiac failure. Am Heart J 76: 227-234, 1968.
 343

512. Rae AI, Hopper Jr J: Removal of refractory edema fluid by peritoneal dialysis. Br J Uro140:
336-343, 1968.
513. Rubin J, Kiley J, Ray R, McFarland S, Bower J: Continuous ambulatory peritoneal dialysis-a
treatment for dialysis related ascites. Arch Intern Med 141: 1093-1095, 1981.
514. Beran A V, Taylor WF: Peritoneal dialysis for the support of respiratory insufficiency in rabbits.
Clin Sci 43: 695-703,1972.
515. Mengan BC, Guntheroth WG, Breazeale D, McGough GA: Failure to achieve oxygen supple-
mentation with hydrogen peroxide. Pediatrics 41: 531-533, 1968.
516. Pritchard JA, Adams RH: The fate of blood in the peritoneal cavity. Surg Gynecol Obstet 105:
621-629, 1957.
517. MacDougall LG: Intraperitoneal blood transfusions in children. Br Med J 18: 139-142, 1958.
518. Mellish P, Wolman 11: Intraperitoneal blood transfusions. Am J Med Sci 235: 717-725, 1958.
519. Mengert WF, Cobb SW, Brown WW: Introduction of blood into the peritoneal cavity. JAMA
147: 34-37, 1951.
520. Cole WCC, Montgomery JC: Intraperitoneal blood transfusion. Am J Dis Child 37: 497-510,
1929.
521. Florey LH, Witts LJ: Absorption of blood from the peritoneal cavity. LancetJune 30, 1323-1325,
1928.
522. Herrmann J, Schmidt HJ, Kruskemper HL: Thyroxine elimination by peritoneal dialysis in
experimental thyrotoxicosis. Horm Metab Res 5: 180-183, 1973.
523. Herrmann J, Beisenherz W, Gillieh KH, Jester HG, Kluge R, Nissen P, Kruskemper HL:
Peritoneal dialysis in the treatment of thyrotoxic crisis. Germ Med Mth 14: 616-617, 1969.
524. Schaible UM, Durr F, Kallee E: Acceleniteo elimination of thyroxine by peritoneal dialysis.
Klin Wsch 50: 1112-1113,1972.
525. Herrmann J, Kruskemper HL, Grosser KD, Bohn W: Peritonealdialyse in der behandlung der
thyeotoxischen krise. Dtsche Med Wochenschr 17: 742-745, 1971.
526. Hobolth N, Devabtuer N: Removal of indirect reacting bilirubin by albumin binding during
intermittent peritoneal dialysis in the newborn. Acta Paediatr Scand 58: 171-172, 1969.
527. Grollman ZAP, Odell GB: Removal of bilirubin by albumin binding during intermittent
peritoneal dialysis. N Engl J Med 268: 279-282, 1962.
528. Christoforov B, Ingrand J, Petite JP, Foliot A: Capitation de la bilirubine non conjuguee par des
dialyses peritoneales avec une solution de serum albumine humaine. Path Bioi 17: 985-990,
1969.
529. Siegel NJ, Brown RS: Peritoneal clearance of ammonia and creatinine in a neonate. J Pediatr 82:
1044-1046, 1973.
530. Synderman SE, Sansarieg C, Phansalkar SV, Schact RG, Norton PM: The therapy of hyperam-
monia due to ornithine transcabanylose deficiency in a male neonate. Pediatrics 56: 65-73, 1975.
531. Rey F, Rey J, Cloup M, Feron JF, Dore F, Labrune B, Frezal J: Traitement d'urgence d'une
forme aigue de leucinose par dialyse peritoneale. Arch Franc Ped 26: 113-137, 1969.
532. Wiegand C, Thompson T, Bock FH, Mathis RK, Kjellstrand CM, Mauer SM: The management
of life-threatening hyperammonemia: A comparison of several therapeutic modalities. J Pediatr
97: 142-144, 1980.
533. Russell G, Thorn H, Tarlow MJ, Gompertz D: Reduction of plasma propionate by peritoneal
dialysis. Pediatrics 53: 281-283, 1974.
534. Gaull GE: Pathogenesis of maple-syrup-urine disease. Observation during dietary management
and treatment of coma by peritoneal dialysis. Biochem Med 3: 130-149, 1969.
535. Donn SM, Swartz RD, Thoene JG: Comparison of exchange transfusion, peritoneal dialysis,
and hemodialysis for the treatment of hyperammonemia in an anuric newborn infant. Brit J
Pediatrics 95: 67-68,1979.
536. Wendel U, Becker K, Hildegard P, Bulla M, Monegold C, Mench-Hoinowski A: Peritoneal
dialysis in Maple-Syrup-Urine Disease: Studies on branched-chain amino and keto acids: Eur J
Pediatr 134: 57-63, 1980.
12. Managing the nutritional concerns of the
patient undergoing peritoneal dialysis

MICHAEL J. BLUMENKRANTZ, ISIDRO B. SALUSKY and

R. WILLIAM SCHMIDT

1. Introduction and background

Although less obvious and dramatic than other aspects of the uremic syndrome,
chronic malnutrition and wasting are serious problems for patients with renal
failure [1-5]. The malnutrition syndromes observed resemble both adult Mar-
asmus (caloric malnutrition) and/or adult Kwashiorkor (protein-calorie malnutri-
tion) [6]. For patients undergoing maintenance hemodialysis or peritoneal di-
alysis therapy, recognition of the signs and symptoms of malnutrition is often
difficult because they so closely resemble signs and symptoms of uremia and
inadequate dietary intake. Loss of essential nutrients into the dialysate produce
abnormalities which are superimposed upon and often resemble the metabolic
abnormalities of uremia. The result is often a complex of signs and symptoms that
suggest the patient is 'failing to thrive' but which is generally neither interpreted
correctly nor dealt with. These problems have appeared more likely to occur in
patients undergoing intermittent peritoneal dialysis [7, 8] and may also be signifi-
cant in patients undergoing CAPD or CCPD [9]. It is not clear how and to what
extent malnutrition contributes to the overall morbidity and mortality of patients
with end-stage renal disease, but the fact that malnutrition adversely influences
the outcome of many other acute and chronic illnesses, suggests that careful
attention must be paid to the nutritional management of the peritoneal dialysis
patient.
The patient with chronic renal failure demonstrates a series of multifaceted and
interrelated metabolic problems [10, 11]. Glucose metabolism is abnormal and
peripheral insulin resistance can be demonstrated [12]; plasma glucagon levels are
increased [10, 13]; hepatic gluconeogenisis is increased; muscle release of alanine
and glutamine is increased and muscle intracellular amino acid pools are abnor-
mal [14-16]. Vitamin D deficiency and hyperparathyroidism occur as a result of
the loss of renal mass and, in turn, result in chronic hypocalcemia [17,18]. Patients
are often chronically anemic and usually hypertriglyceridemic [19]. Serum car-
nitine levels may be reduced [20-22] and worsen the dyslipoproteinemia. During
346

acute illnesses, anorexia and malaise couple with tissue breakdown to accelerate
malnutrition and wasting [23, 24]. Increased dialysis to compensate for the
increased urea production from hypercatabolism also contributes by increased
dialysate nutrient loss. Trace mineral depletion may also result when dialysis is
increased and oral intake is curtailed [25].
The patient starting peritoneal dialysis is often suffering from both acute and
chronic malnutrition. It is a sometimes forgotten and often ignored fact that the
patient usually develops end stage renal disease gradually, often in spite of the
physician's best efforts. During this period he is subjected to a protein depleting
diet and treated with a virtual polypharmacy of medications, many of which
worsen anorexia and calorie malnutrition; several drugs interfere with absorption
of specific nutrients [26-27]. Episodic illness from hypertension, congestive heart
failure, recurrent urinary tract infections, pericarditis, pneumonitis or complica-
tions of therapy, particularly corticosteroid therapy, result in the patient initiating
dialysis in an already severely debilitated state.
Peritoneal dialysis itself presents a set of peculiar nutritional and metabolic
problems as well as some possible nutritional advantages [7, 28]. Peritoneal
dialysis has been used extensively for the past 20 years to treat uremia resulting
from acute renal failure. It also has been commonly used as temporary, intermit-
tent treatment for patients awaiting institution of hemodialysis [29]. The initial
experiences with long-term maintenance peritoneal dialysis for the definitive
management of patients with end-stage renal failure generally were not favora-
ble. Although capable of transiently improving uremic symptoms, peritoneal
dialysis was frequently associated with progressive tissue wasting and malnutri-
tion [29, 30]. In the 1960's each treatment usually required an abdominal punc-
ture; dialysis was usually performed infrequently, i.e., every 6-10 days because it
required hospitalization and was labor intensive, see table 1. Unless the patient
had significant residual renal function this frequency of dialysis was usually
inadequate and patients remained symptomatically uremic. Due to weakness,
anorexia and vomiting, nutrient intake was often insufficient during the inter-
dialytic period. In order to minimize azotemic symptoms, protein intake was
often severely restricted. During the long (36-40 h), often uncomfortable dialysis
procedure, most patients had little desire to eat. Dialysate protein, amino acid
and trace mineral losses were substantial and the combination of insufficient
dialysis and inadequate nutrient intake were major factors contributing to the
frequently observed development of malnutrition and wasting. There was a high

Table 1. Nutritional problems with peritoneal dialysis in 1960s and 70s

1. Insufficient dialysis resulting in anorexia and poor intake

2. High dialysate protein and amino acid losses
3. High frequency of peritonitis with increased losses, poor intake and catabolism
4. Failure to provide supplemental nutrients to stable and stressed patients
 347

incidence of peritonitis as a result of the multiple abdominal punctures and

contamination of the sterile system by the repeated manual exchange of bottles of
dialysate. Peritonitis resulted in accelerated catabolism, in increased dialysate
protein loss and in further decreases in food intake, compounding a bad situation
and making it worse.
In the 1970's, the development and use of the Tenckhoff permanent implanted
peritoneal catheters and automated peritoneal dialysis equipment enabled inter-
mittent peritoneal dialysis to be performed safely and at more frequent intervals.
Most commonly patients were started on peritoneal dialysis after demonstrating
intolerance of another treatment modality, usually hemodialysis. Typically this
was the result of hemodialysis access site exhaustion and occurred in the setting of
protracted and profound inadequacy of hemodialysis because of the vascular
problems. Recurrent access site infection was a frequent additional factor. Per-
itoneal dialysis was resorted to as a poor alternative, the 'step-child' of hemo-
dialysis [31]. Despite a marked reduction in the incidence of peritonitis, the same
combination of factors, i.e. inadequate dialysis and insufficient nutrient intake,
often resulted in failure of patients to improve or even to feel well. Not infre-
quently, all patients at a particular dialysis center were placed on a standardized
peritoneal dialysis regimen, (e.g. 8-10 Hrs 3 x/week) as has been customary for
hemodialysis. In many dialysis centers, no account was made for patient size and
residual renal function when the amount of intermittent peritoneal dialysis was
prescribed [32, 33]. The failure to individualize peritoneal dialysis therapy often
had serious consequences. Even when physicians and dietitians prescribed ade-
quate dietary protein and calories, if the amount of dialysis was insufficient, the
patient became anorectic and actual protein intake remained low. Commonly, a
gross discrepancy between prescribed intake and actual intake went undetected.
The result was an incidious, downhill path. Inadequate dialysis resulted in poor
protein and energy intake. The reduced protein intake resulted in a reduced rate
of urea nitrogen appearance (net urea production) and consequently, a reduced
serum urea nitrogen (SUN) level. The low SUN was sometimes misinterpreted as
a 'good' response and neither increased dialysis nor increased protein intake was
prescribed. Muscle wasting decreased the increase in serum creatinine level; an
increase in serum creatinine was often mistakenly attributed to decreased per-
itoneal clearance although this was rare. Patients who started on chronic per-
itoneal dialysis while still retaining significant renal function (GFR's 2-3 mllmin)
often did well initially, but as their endogenous renal function fell, SUN and
serum creatinine became more deranged. The increasingly abnormal blood
chemistries, misinterpreted as increased dietary indiscretion reflected a normal
decline in endogenous renal function. Rather than increasing dialysis, as would
be appropriate, dietary restrictions were heightened. Although SUN levels
would be reduced, the state of malnutrition was begun. The end result is the
gradual development of tissue wasting and general 'failure to thrive' (Fig. 1).
Malnutrition and wasting may contribute to many aspects of the uremic syn-
348

PRESCRIBE 1.3 G PROTEIN/KG BODY wr.

'"
PRoVIDE INSUFFICIENT nIALY~IS

'"
AOOREXIA
'"
ACAALLY INGEST 0.8 G!1<fj

'"
&LNlITRITION AND 'FAIL~E TO THRIVE'

Figure 1. Misperceptions leading to 'failure' of peritoneal dialysis.

drome, including susceptibility to infection, impaired wound healing, anemia,

decreased strength, cardiovascular dysfunction, sexual dysfunction, progressive
neuropathy, failure of rehabilitation and poor quality of life. It is for these reasons
that there is a need to periodically assess the nutritional status of patients
undergoing peritoneal dialysis so that appropriate therapy can be prescribed [5].
Decreases in peritoneal clearance have frequently been invoked as the cause for
this 'failure to thrive syndrome', but seldom documented [32, 34]. Decreases in
peritoneal clearance of urea and creatinine may occasionally occur, but most
commonly the decrease in total solute excretion (milligrams or millimoles per
day) is not due to a decrease in membrane clearance (mllmin) but rather is due to
an insufficient amount of dialysis and/or solute generation. Rarely, it is due to
progressive obliteration of the peritoneal cavity by scar tissue, the latter being the
result of frequent and/or mismanaged episodes of peritonitis [24].
Continuous ambulatory peritoneal dialysis, as an extension of IPD has met
with high levels of patient acceptance and variable levels of physician enthusiasm.
It has rapidly become a common first choice of dialytic therapy particularly for
elderly and for diabetic patients. However, in the 1980's we are seeing the same
misunderstandings and misperceptions of patient management which contributed
to the long-term problems with peritoneal dialysis in the 1960's and 1970's: At
many centers, all patients receive 4 CAPD exchanges/day, usually 2 liter vol-
umes; other centers prescribe 3 two liter exchanges/day for everyone. If the
relationship between protein intake, SUN and total urea clearance is not appreci-
ated and CAPD therapy is not individualized, many patients, especially large
men who have GFR's less than 2 ml/min, may be underdialyzed, develop
anorexia, become proteiu depleted and 'fail to thrive' [35].
Although patients undergoing CAPD are reported to gain edema-free weight
[9], recent data suggest that some may also be prone to protein depletion. Serum
total protein, albumin, transferrin and C3 are below normal, and remain low over
their course of treatment [9]. In one report, extravascular albumin and total
exchangeable albumin were reduced in CAPD patients [36], while in another
study, body albumin pools were normal [37]. In a large series of CAPD patients,
 349

no increase in arm muscle circumference (a measurement of protein mass) was

found even though weight gain occurred [38]. Williams and co-workers
monitored total body potassium and nitrogen in patients undergoing CAPD [39].
Total body potassium remained unchanged or increased in these patients, but
total body nitrogen fell significantly. Other authors have reported a reduction in
total body potassium in CAPD patients who have sustained multiple episodes of
peritonitis [41]. When Lindholm and coworkers performed muscle biopsies in 24
CAPD patients, they found increased muscle intracellular and extracellular
water and increased muscle potassium when expressed per unit of fat-free solids
but not when expressed per liter of intracellular water [41]. Dietary energy and
protein intake in CAPD patients with peritonitis are often below recommended
levels. Hence, it is of particular importance to assess whether the recommended
intakes are appropriate and whether patient compliance is sufficient.

2. Methods to assess the nutritional status of the peritoneal dialysis patient

2.1. History and physical examination

The medical history should be evaluated carefully for the presence of disorders
which are likely to promote malnutrition and wasting. Many of the symptoms and
signs often observed in uremic patients such as anemia, apathy, anorexia, edema,
muscular weakness, depression, congestive heart failure and peripheral neuropa-
thy also occur with chronic malnutrition and wasting. Alcoholism, diabetes
mellitus, severe congestive heart failure, gastrointestinal disorders, and psycho-
social problems may interfere with ingestion and/or assimilation of nutrients.
Certain medications can interfere with absorption or utilization of nutrients [26,
27], while others can produce accelerated protein catabolism. A history of body
weight and habitus before the onset of renal failure is helpful in identifying the
degree of weight loss that has occurred. The usual preuremic weight of the patient
should be established and any recent weight loss should be documented. The
physical signs which are associated with malnutrition have been extensively
reviewed [41-44]. Physical findings associated with extracellular fluid volume
expansion should be carefully analyzed so that an adequate allowance for excess
water weight can be made. It is common for edema fluid to mask decreases in lean
body mass. Anthropometric measurements are essential.
There are several gastrointestinal problems in peritoneal dialysis patients that
can interfere with nutrient intake: anorexia, nausea, vomiting and a sense of
abdominal fullness are not rare [45]. These symptoms are more likely to occur
during the initial few weeks of treatment. They often subside, but may recur in
conjunction with other illnesses. Anorexia can occasionally be relieved by drain-
ing the peritoneal cavity prior to meals and waiting to instill fresh dialysate until
thirty minutes post-prandially. Appetite suppression may sometimes be related
350

to the large caloric intake from glucose absorbed from dialysate; at other times it
presumably is related to the abdominal distention by dialysate. Constipation,
promoted in part by the ingestion of phosphate binding antacids, may result in
abdominal discomfort, anorexia and catheter malfunction. The regular admin-
istration of sorbitol as an osmotic laxative may be preferable to markedly decreas-
ing the intake of dairy products, the ingestion of which is often necessary to
achieve an adequate protein intake.

2.2. Assessment of nutrient intake

Carefully done, dietary intake can be estimated by recall interviews augmented

with dietary diaries [46, 47]. One successful approach is to assess dietary intakes
at frequent intervals e.g. obtain 3-day diet histories and written dietary diaries
every 1 to 4 months (depending on the nutritional status and willingness of the
patient). Dietary histories can be very important in indicating whether a patient is
adhering to the prescribed diet. The patients' family and close associates are often
valuable sources of additional information in determining patient compliance
since frequently patients are unaware of or forget how often they fail to comply.
The rate of urea nitrogen appearance in the CAPD patient provides a good
estimate of dietary protein intake (vida infra).
When assessing nutrient intake, and also when prescribing diets, the physician
and dietician must remember to consider the glucose absorbed during peritoneal
dialysis and the losses of proteins and amino acids into dialysate. The amount of
glucose absorbed from peritoneal dialysate can be accurately predicted by know-
ing the dextrose content of the dialysate and the frequency of exchanges (vida
infra). In the absence of peritonitis, both glucose absorption and protein losses
are generally constant in a given patient.
The constraints of survival on a dialysis regimen, even CAPD, are many. In a
society oriented toward food intake, many of these constraints center about the
restrictions of a diet. Although this is less the case with CAPD, dietary restric-
tions in the dialysis patient may produce intense feelings of anxiety, guilt and
rebellion. Effective intervention by an experienced dietician is of major value in
overall patient management and rehabilitation. It is important that the physician,
nurses, psychosocial staff and other allied health personnel encourage dietary
compliance and cooperation with the dietician (Fig. 2) [47].

2.3. Body composition

Many methods have been used to assess body composition in nonuremic subjects.
Several measurement parameters are useful for the clinical management of the
peritoneal dialysis patients. Tables and standard weight and relative body weight
 351

PHYSICIAN _ _ _ _ _ _ DIETITIAN

SUPPORTING TEAM
SocIAL \t)RKER

2. NrnSE
3. PSYCI-OLOG I ST
PATIENT - FN-lILY ~IT

Figure 2. Team approach to dietary therapy (from Sorenson and KoppJe [41]; reproduced with
permission) .

(RBW) have been derived from measurements of large numbers of normal

people. These tables indicate weight adjusted for sex, height, and age [48, 49].
Desirable (formerly called ideal) body weights are less than the normal weights
for the general population. These weights are derived from the actuarial tables of
the Metropolitan Life Insurance Company and are the weights for a given height,
age and sex which are associated with the greatest longevity [50]. In uremic
patients, RBW is probably a more useful standard than 'desirable' body weight
(vida infra).
Since the fractions of body weight which are fat, muscle or body water can vary
greatly, it is important to evaluate the proportions of each of these constituents.
Virtually all dieticians in the United States are trained in anthropometric mea-
surement. Few are systematically called upon to use these skills. Measurements
of skinfold thicknesses is used widely to estimate body fat [51]. It is an inexpen-
sive, reproducible, relatively easy technique to learn and perform. Moreover, it
can be carried out quickly. The thicknesses of subcutaneous tissue at different
sites of measurement change proportionately with weight gain or loss, and most
observers find a good correlation between subcutaneous tissue thickness and total
body fat. The reproducibility of skinfold measurements is improved if the tech-
nique is standardized rigorously with the same experienced observer performing
all measurements and the same point on the skin measured each time.
Muscle represents a sizable fraction of fat-free body mass and total body
protein. Muscle mass is commonly estimated from measurement of mid-upper
arm circumference. Mid-arm circumference is adjusted for subcutaneous tissue to
estimate mid-arm muscle circumference (MAMC). There are, however, several
recognized problems with interpretation of arm circumference and skinfold
measurements [52, 53]. Desirable MAMC or skinfold thickness for patients with
chronic renal failure are not known and may be very difficult to establish. For
352

example, a man who is normally quite muscular and obese may sustain catabolic
illness and weight loss, but still have a 'normal' MAMC and skinfold thickness.
The most sensitive method for assessing nutritional status would be to compare
the anthropometric measurements of a person to his own parameters prior to the
onset of an illness.

2.4. Serum proteins

Several serum proteins have been used to measure nutritional status [54-56].
Serum albumin and transferrin are the proteins used most extensively in both
uremic and non-uremic patients. Some investigators suggest that serum transfer-
rin may be more sensitive than albumin as an indicator of malnutrition, possibly
because of the shorter half-life of transferrin (8-9 days) as compared to albumin
(18-20 days). However, serum transferrin levels are affected by other factors,
such as iron deficiency which increases transferrin levels [57, 58] or iron-loading
(e.g., with parenteral iron dextran) which may depress the serum transferrin
concentration [59]. Serum albumin levels are determined by a complexity of
factors including rates of synthesis, catabolism, the plasma volume, and compart-
mentalization of albumin. Uremia, per se, appears to affect albumin metabolism,
and this may affect albumin concentrations [60-63]. In renal failure, serum
albumin and transferrin are often decreased. Unfortunately, unless the levels are
very depressed, transferrin and albumin measurements have often failed to
correlate with other parameters of nutritional status. Nevertheless, when they are
abnormal, serial measurement of albumin and transferrin in the same patient
does provide valuable information about protein status and the response to
dietary therapy. Measurement of various other proteins probably adds little to
the evaluation of the nutritional status of the dialysis patient [5].
Studies of body composition in uremic patients have been previously published
[64,65]. These studies may not be applicable to the current patient population.
The data were obtained at a time when prolonged, conservative management
(including severely restricted protein diets) was employed for prolonged times
prior to dialysis and the amount of dialysis accomplished was low, both because
the number of hours on dialysis was low and because of the inefficiency of dialysis
equipment. Recently, an opportunity to re-examine this problem was provided in
conjunction with the Veterans Administration Cooperative Dialysis Study. A
comparison was made between (a) normal volunteers, (b) patients beginning the
study who were thought to represent a relatively healthy subset of the chronic
renal failure population and (c) a group of patients on chronic intermittent
hemodialysis who were felt by their physicians to be unusually healthy and
robust. The patients initiating dialysis had evidence of malnutrition [2].Even the
patients thought to be healthy and robust demonstrated decreases in serum
protein levels [5]. These findings support the clinical impression that biochemical
 353

evidence of malnutrition is common in renal failure, even in populations that are

apparently healthy. It is our clinical experience that uremic patients with reduced
relative body weight, body fat and MAMC are more debilitated, prone to more
complications and less capable of rehabilitation. Objective data on morbidity and
mortality of this group as compared to better nourished patients is much needed.

2.5. Nutritional status of patients undergoing CAPD

Protein and caloric malnutrition is common in patients with ESRD and those
undergoing maintenance hemodialysis [5]. For the reasons discussed above,
protein malnutrition and wasting were common in patients undergoing intermit-
tent PD and has been one of the major reasons for the ultimate failure of this
dialysis modality in many patients. Early reports on the nutritional status of
CAPD patients were quite favorable; however, this was also the case with the
early intermittent PD reports. Therefore, the long-term nutritional evaluations
performed by Oreopoulos and his co-workers are of considerable interest [9]. See
table 2. They evaluated the nutritional status of twenty patients (three diabetics)
for 18 to 24 months. Eight of the patients had been on IPD. Seventeen of the
patients had either one or no bouts of peritonitis. Fifteen of the patients had
residual renal function, 3 ml/min at the start of CAPD and 1.2 mllmin at the end of
evaluation. No mention was made as to the amount of treatment (total urea
clearance). Dietary protein intake which was initially 1.4 g/kg/day decreased to
1.0 g/kg at 1 year; it did not decrease further at 18 months. Total energy intake
decreased from 33 to 29 kcallkg. Serum albumin remained low; transferrin
increased but not significantly. However, C3, another protein indicative of
nutritional status, did increase. The patients' weight increased; the increase in

Table 2. Nutritional characteristics of twenty patients undergoing CAPD'

Months since start 0 6 12 18

Protein (g/kg/d) 1.4 ± 0.3 1.2 ± 0.3 1.0 ± 0.4 1.0 ± 0.3* *
Energy (Kcal/kg/d)b 33 ± 7.6 30 ± 8.4 32 ± 7.0 29 ± 6.0*
BUN (mg/dl) 60± 18 61 ± 17 66± 16 63 ± 21
creatinine (mg/dl) 10.3 ± 4.0 11.1 ±3.2 11.7 ± 3.6 11.6±3.4
albumin (g/dl) 3.3 ± 0.4 3.4 ± 0.5 3.6 ± 0.4 3.5 ± 0.4
transferrin (mg/dl) 138 ± 41 193 ± 70 203 ± 54 197 ± 59
C3 (mg/dl) 79± 16 88 ±32 104 ± 41 105 ± 21 * *
weight (kg) 63 ± 14 65 ± 14 68 ± 15 68±14**
TBK (g) 102 ± 31 105 ± 30 108 ± 27 108 ± 29* *
TBN (g) 1665 ± 42 1570 ± 35 1450 ± 35 1370 ± 30* *

, Adapted from [9].

b Includes calories from dextrose absorbed from dialysate.
* p<O.01; * * p<0.05.
354

total body potassium (TBK) was insignificant but it correlated with the increase in
weight. Paradoxically total body nitrogen (TBN) decreased markedly (about
20%).
It is possible that increases in intracellular water affect TBK and make it an
unreliable index of body protein [66]. The increase in weight was probably due to
an increase in fat and/or body water; unfortunately no measurements of fat stores
or body water were reported. The marked decrease in TBN is probably indicative
of muscle wasting; changes in anthropometric measurements (such as a decrease
in mid-arm muscle circumference) or total creatinine excretion would have
confirmed this supposition. The patients who had one to three episodes of
peritonitis actually had less of a fall in TBN than those without peritonitis.
Dialysate protein losses, which averaged 7 g/24 h at the initiation of CAPD
decreased to 6 g/24 h at 6, 12 and 18 months; protein losses did not differ in the
patients who had peritonitis from those who did not.
Nitrogen balance studies by Blumenkrantz et al. discussed below have shown
that a minimum dietary protein intake to maintain N balance is 1.2 g/kg/day; this
minimum is determined in sedentary patients receiving a high energy intake (42
kcal/kg/day). If energy intakes are less, dietary protein intake may need to be
higher. Dietary protein supplements or the addition of amino acids to dialysate
are needed to increase an inadequate protein intake. Energy supplements should
also be considered if the patient is not obese. The CAPD patients' appetite can be
deminished because of abdominal distension and the absorption of glucose from
dialysate. Studies in non-uremic rabbits [67] have shown that their spontaneous
food intake decreased markedly when they are undergoing CAPD, especially if
4.25% dextrose dialysate is used. To improve appetite, we ask the nonobese
patients to drain before or during meals, and refill after meals - thereby avoiding
the abdominal distension and high glucose infusion that will impair appetite. In
obese patients, we try to impair their appetites by asking them to perform their
exchanges one hour before meals. Even the obese patient must be encouraged to
ingest a high protein diet.
In summary, although standards for evaluating many nutritional parameters
are not well established, particularly in patients with renal failure, serial measure-
ments of several nutritional parameters in the same patient may increase the
sensitivity and accuracy of these evaluations. An evaluation of nutritional status
should be considered a vital component of overall clinical patient management.
The nutritional plan for the peritoneal dialysis patient should be guided by these
findings and frequent reassessment should prove valuable.
 355

3. Alterations in metabolism of specific nutrients by peritoneal dialysis

3.1. Protein

The loss of substantial amounts of protein into dialysate has long been considered
a major disadvantage to chronic peritoneal dialysis. On the basis of recent
experience, particularly with CAPD, this fear seems largely unwarranted. In the
initial experience with chronic intermittent peritoneal dialysis, patients often
developed progressive wasting or malnutrition [19] and the losses of protein into
dialysate may have made a contribution to these problems. With intermittent
peritoneal dialysis, losses were reported to vary from 0.5-4.5 g protein/l of
dialysate exchanged, with 20 to 200 g total protein being lost during a single 24-48
h dialysis [68-70]. Serum protein concentrations decreased transiently during a
single dialysis and progressive decreases in serum total protein and albumin
concentrations developed in many patients undergoing chronic IPD treatment.
Most of the protein loss was albumin although large amounts of immunoglobulins
were also lost. In one report, an average of more than 20 g of immunoglobulins
was found to be removed with a single dialysis [71]. These observations have been
the basis of considerable apprehension about the long term consequences of
protein depletion, both in terms of protein malnutrition and possible immu-
nologic incompetence. It should be noted that these studies were performed upon
patients who were undergoing IPD and who, by present standards, were proba-
bly insufficiently dialyzed. Protein intake was usually restricted or, if not re-
stricted, simply not ingested. Many such patients began peritoneal dialysis after a
long course of renal failure or after having failed on hemodialysis. A separate
catheter was usually inserted into the abdominal cavity for each dialysis; the
dialysate was infused by the manual exchange of 21 bottles with a non-closed
system. Low-grade, indolent bacterial infections were probably common and
recurrent episodes of acute bacterial peritonitis were not infrequent. Such experi-
ence is not representative of recent experience with closed system intermittent
peritoneal dialysis or with CAPD, particularly when efforts are made to effect
both protein repletion and adequate dialysis.
A comprehensive evaluation of the losses of plasma proteins during IPD,
CAPD and acute peritoneal dialysis was performed [72]. With the newer tech-
niques for IPD, dialysate protein losses are considerably reduced. As shown in
Table 3, in patients undergoing IPD an average of 12.9 g total protein and 8.5 g
albumin were lost during each treatment. With treatment performed every other
day the loss amounted to 45 g/week or 6.4 g/day. In individual patients, there was
usually little variation in the quantity of protein lost during successive dialyses;
however, there was subtantial interpatient variability. Approximately 50% of the
protein loss occurred in the initial 2 h due to washout of ascitic fluid which
accumulated during the interdialytic period. This observation has recently been
reconfirmed [73]. Serum protein levels did not decrease during the course of a
356

Table 3. Average protein losses during peritoneal dialysis

Intermittent CAPDa Acute PD

PD (10 hr) (24 hr) (36 hr)

Total protein 12.9 ± 1.1 8.8 ± 0.5 22.3 ± 4.7

Albumin 8.5 ± 1.5 5.7 ± 0.4 13.3 ± 2.4
IgG g/dialysis 1.3 ± 0.23 1.2 ± 0.20 2.9 ± 0.78
IgA 0.29 ± 0.08 0.17 ± 0.02 0.55 ± 0.19
Transferrin 0.26 ± 0.D3 0.33 ± 0.02 ND

Values are mean ± SD.

ND = data not available.
a Losses of IgM, C3 and C4 averaged 71 ± 18,70 ± 7 and 21 ± 2 mg/24 hr respectively (adapted from

(72)).

single peritoneal dialysis or during nine months of follow-up [72]. Protein loss was
higher soon after catheter insertion, but subsequently fell to lower levels. An
increase in dialysate dextrose concentration has been observed to increase pro-
tein loss [72]. Whether this is due to the higher osmolality of the solution or other
factors is not clear. The development of peritonitis during intermittent peritoneal
dialysis markedly increases protein loss to 38-40 g/lOh/dialysis. Occasionally
protein loss exceeds 100 g/day. With IPD, elevated protein losses sometimes
persisted for weeks following an episode of peritonitis. The severe protein
depletion coupled with decreased intake and severe catabolism during the
episode can produce hypoalbuminemia.
With CAPD, losses of total protein average 8.8 g/day and albumin losses
average 5.7 g/day (Table 3). This is similar to the protein losses reported by others
[25, 74-80]. As with IPD, the quantity of protein lost shows considerable interpa-
tient variation, although values remain relatively constant in the same patient
[72]. The dialysate losses of IgG and IgM correlate with serum concentration. No
relationship has been apparent between the quantity of protein loss and the
serum concentration of other proteins. There has been no relationship observed
between the total outflow volume of dialysate per day (which varied from 7.7 to
12.9 l/day) and the quantity of protein lost. Daily dialysate protein losses re-
mained constant in a patient who alternated three and four exchanges/day. This
did not agree with the findings of Rubin et al. [80]. With episodes of mild
peritonitis, dialysate protein losses increase to an average of16.1 ± 3.6 g/day. The
quantity lost usually returns to baseline within a few days following treatment
[72]. Early reports suggested that protein loss remains elevated in patients who
have had peritonitis [74]; however, more extensive recent data by the same group
[9] fails to confirm this initial observation.
The losses of protein with acute peritoneal dialysis (Table 3), using the tempo-
rary insertion of a nylon catheter and the 'manual' exchange of 21 bottles of
dialysate are considerably lower than previously reported [68, 69]; the reason for
the differences is unclear.
 357

3.2. Serum protein levels

Typical serum protein levels for patients undergoing IPD and CAPD are shown
in Table 4 [72]. Most reports have indicated that serum albumin levels are
generally low or low-normal (mean range 3.3-3.8 g/dl) in patients undergoing
maintenance IPD. In the absence of peritonitis, dialysate protein loss probably
has little effect on serum protein levels since the quantity of protein lost in
peritoneal dialysate (Table 3) is similar to the quantity found in only 50-200 ml of
plasma. Poor nutrient intake, the occurance of peritonitis, and other intercurrent
catabolic processes probably contribute more to the persistently low serum
protein levels than does continued loss. In robust, healthy appearing dialysis
patients, including patients on chronic hemodialysis who were not subject to
dialysate protein losses, serum albumin, transferrin and C3 complement levels
were all low [5] suggesting that other casual factors are likely.
The serum albumin concentration is considered an important indicator of
nutritional status in uremic patients (vide supra). However, the serum concentra-
tion is affected by a number of factors including the rates of synthesis, rates of
catabolism, distribution within extravascular (EV) and intravascular (IV) com-
partment. The EV and IV pool sizes are frequently reduced in uremia. In patients
undergoing peritoneal dialysis, losses of albumin into dialysate could alter al-
bumin pool sizes and metabolism. Therefore, kinetic analysis and compartmental
modeling of albumin metabolism were performed in 39 subjects [36]: eight
patients undergoing intermittent peritoneal dialysis, six patients undergoing
CAPD, nine undergoing hemodialysis, six nondialyzed chronically uremic pa-
tients and ten volunteers with normal renal function. Residual renal function and

Table 4. Serum protein levels in patients undergoing peritoneal dialysis"

Intermittent CAPD Normal b

Peritoneal
Dialysis

Total protein { 5.9 ± 0.5' 6.6 ± 0.5 7.4 ± 0.6

Albumin gldl 3.7 ± 0.4' 3.5 ± 0.4' 5.0 ± 0.4
IgG 932 ± 225 1410 ± 420 1139 ± 341
IgA 221 ± 135 220 ± 65 277 ± 139
IgM 112 ± 45' 234 ± 131 186 ± 108
Transferrin mgldl 189 ± 35' 228 ± 40 303 ± 41
C3 95 ± 18' 107 ± 20 138 ± 26
C4 41 ± 5 32 ±6 38 ± 17

Values are mean ± SE.

a Adapted from [72].
b Mean values obtained from 60 male hospital employees.

, Differs from normal, p<[Link].

358

protein intake were similar in the IPD, CAPD, and the hemodialysis patients.
The serum albumin concentrations were low in all patient groups when compared
to the normal volunteers. Albumin levels were lower in the peritoneal dialysis
patients than in the hemodialysis patients. IV albumin pools (glkg body weight)
were not significantly lower than normal in any of the groups of dialysis patients.
However, the three groups of dialysis patients all had low EV albumin pools.
Total exchangeable albumin (glkg body wt) was lower than normal only in the
patients undergoing CAPD; there was no significant difference in this measure-
ment among the intermittent PD, CAPD, and hemodialysis patients. The EV/IV
ratio did not differ from normal in any of the groups of dialysis patients. Thus,
albumin metabolism in peritoneal dialysis patients did not appear to differ in any
important aspect from that in hemodialysis.

3.3. Amino acids

Unlike anthropometric measurements or serum protein concentrations which

reflect chronic nutritional status, the plasma amino acid (PAA) pattern tends to
reflect recent intake of protein and other nutrients. Skeletal muscle contains the
largest pool of intracellular AA and although they may be altered by metabolic
processes, muscle free AA levels may better reflect chronic nutritional status than
PAA [16].
A variable quantity of amino acids have been reported to be lost during
intermittent peritoneal dialysis. Berlyne et al. [81] and Noree and co-workers [82]
each have found AA losses to average 5-6g/dialysis (0.2g/l exchanged). Young
and Parsons calculated that free AA losses average 4.5 g (720 mg of AA nitrogen)
and conjugated amino acid nitrogen losses are about 2.4 g (390 mg of AA
nitrogen) per dialysis [83]. Finkelstein et al. found AA losses to be much greater,
averaging 17 .3g/dialysis treatment in 9 patients undergoing treatment of 36 h each
[84]. The PAA pattern in patients undergoing intermittent peritoneal dialysis
often resembled that seen in patients with malnutrition [85, 86] but as pointed out
previously, patients may have been on low protein diets at the time of study.
Intracellular AA pools were not measured. During intermittent peritoneal di-
alysis treatment Bergstrom and co-workers reported that the plasma concentra-
tions of several AA decreased but muscle intracellular concentration of most AA
did not change significantly [87].
Intravenous administration of essential AA during the last 4 h of a period of
peritoneal dialysis increased the total free AA, the ratio of essential to non-
essential AA and the valine concentration in muscle. Only 4-13% of the infused
AA were lost in dialysate [87]. During intermittent peritoneal dialysis the infu-
sion of essential AA improved nitrogen balance and increased the incorporation
of 15N into muscle protein [88].
The rate of diffusion of most AA across the peritoneal membrane is much less
 359

than the rate measured across artificial hemodialysis membranes [89]. Because of
this lower rate of diffusability, in the absence of peritonitis, peritoneal AA loss is
increased very little during food absorption or during intravenous AA infusion
[90]. It is probable that during peritoneal dialysis parenteral nutrition can be
effected and PAA concentrations improved without causing significant increases
in peritoneal AA losses.
Plasma amino acid levels have been reported abnormal in CAPD patients [91-
94]. In general the abnormalities are similar to those described in nondialyzed
patients with chronic renal failure or in patients undergoing maintenance hemo-
dialysis[95]. In our experience however, both plasma total essential and total
nonessential amino acids in the CAPD patients were similar to normal [96] (Table
5). Other studies of CAPD patients describe both reduced and normal total
essential and nonessential amino acids [92-94]. In our studies there was a direct
correlation between the individual amino acid concentrations in plasma, and in
dialysate (r = 0.83) [96]. The daily losses of individual amino acids (umoles/day)
also correlated with post absorptive plasma concentrations (r = 0.78). Of the
amino acids in dialysate 29.0 ± 3.6% were essential, which is similar to the
proportion in plasma that were essential, 28.4 ± 3.8%. Total free amino acid
losses into dialysate averaged 3.4 ± 0.3 g/24 h, which is slightly higher than the
mean losses reported in other studies. This discrepancy may reflect the higher
plasma amino acid concentrations, greater protein intake, greater body mass and/
or peritoneal surface area of our patients, their more frequent dialysate ex-
changes, or their greater outflow volumes. The fact that PAA levels were normal
and AA losses were greater would suggest that dietary intake was more adequate
than in other studies.
Recently, Williams and associates and Oren an co-workers studied the effects
of substituting amino acids for the dialysate glucose in CAPD patients [97, 98]. In
a dialysate solution containing a 2% mixture of essential and nonessential amino
acids with no glucose, 80-90% of the amino acids were absorbed by six hours.
One hour after instillation of the dialysate, the mean plasma amino acid concen-

Table 5. Plasma amino acids and amino acid losses into dialysate in nine men undergoing CAPO

Plasma amino acids Dialysate amino acids

/Lmoles/l mg/24 hr

CAPO patients Normal men CAPO patients

Number of studies 14 9 14
Total essential 963 ±29 1030 ± 40 1027 ± 96
Total non-essential 2080± 115 2001 ± 135 1949 ±226
Total 3415 ± 134 3260± 145 3355 ±334
Essential/non-essential ratio 0.48 ± 0.03 0.53 ± 0.03 0.59±0.05

Adapted from [96).

360

trations had increased approximately two to three fold. Plasma levels fell to
predialysis values by six hours. A dialysate solution containing 2g1dl of amino
acids without glucose has a sufficient osmolality to promote a volume of ultrafil-
trate that is approximately the same as standard dialysate solution containing 4.25
gldl of dextrose. In the study by Oren et al. nutritional status was monitored in six
CAPD patients for four weeks while they received two two-liter exchanges each
day of a dialysate solution containing a 191dl mixture of essential and nonessential
amino acids in place of the dextrose. The patients also received two additional
daily two-liter exchanges with standard dextrose containing dialysate solutions.
After four weeks of such therapy, there was an increase in serum urea nitrogen
(SUN), total body nitrogen (excluding body urea nitrogen), serum transferrin,
and anion gap. These preliminary studies although very promising, require
confirmation. The optimal mixture of amino acids for peritoneal dialysate still
needs definition. At the present time, peritoneal dialysate solutions that contain
amino acids are not commercially available.

3.4. Dietary protein requirements

The protein requirements of a patient undergoing maintenance hemodialysis

have been assumed to be approximately 1.0 glkg body wt/day. Increased protein,
amino acid and peptide losses occur in patients undergoing peritoneal dialysis
which logically dictates an increased dietary protein intake. At the present time,
it would appear that a protein intake of 1.2 to 1.5 glkg body wt/day is needed in
patients on peritoneal dialysis.
Lindner and Tenckhoff evaluated nitrogen balance in five women undergoing
intermittent peritoneal dialysis [99]. Patients ingested 1.6 g protein/kg body
wt/day (range 1.3-1.9) for 4 weeks. Mean nitrogen balances were markedly
positive during the first two weeks, +3.4 glday. During the second two weeks,
when patients received an androgenic steroid, mean nitrogen balance remained
positive +4.5 g/day. A puzzling finding was that body weight and serum albumin
levels did not increase during the study. More information regarding nitrogen
balance in patients receiving intermittent peritoneal dialysis is needed, par-
ticularly with respect to the requirements needed to permit repletion of body
protein stores after a severe catabolic illness.
Protein requirements in stable patients undergoing CAPD have been studied
by several investigators including ourselves [78, 100-103]. Giordano and co-work-
ers studied 8 patients [100] ingesting 1.2 g protein/kg body wt/day while receiving a
total energy intake (diet + dialysate) of 40-50 kcal/day. Positive nitrogen balance
(greater than +1.5 g N/day) was found in 3 of the 8 patients. One patient was in
negative nitrogen balance and the remaining 4 patients were positive by 191day or
less; if insensible and unmeasured losses are taken into account these latter 4
patients were close to neutral nitrogen balance.
 361

Lindholm et al. evaluated nitrogen balance in 10 patients who had undergone

CAPD for 2-4 months [78, 103]. Protein intake varied from 0.8-2.1 g/kg/day and
total energy intake ranged from 29-53 kcal/kg body wt/day. Nitrogen balance was
positive in all patients, even those patients receiving low total protein intakes.
Positive nitrogen balance correlated with both total protein intake and with total
energy intake. It also correlated well with weight gain.
Gahl and co-workers recently reported the results of nitrogen balance studies
in 5 patients [101] undergoing CAPD. Evaluations were performed while patients
ingested their usual diets at home. Patients were prescribed diets containing 1.2 to
1.5 g protein. Despite the prescription, the mean protein intake was estimated to
be only 0.7--0.9 g/kg. Only 1 of the 5 patients was clearly in positive nitrogen
balance. If unmeasured losses are added the remaining 4 patients were probably
in either neutral or negative balance. Fecal nitrogen losses were not measured
and were assumed to be only 1 g/day.
In order to evaluate protein requirements, we studied protein balances in eight
clinically stable men who underwent 13 metabolic balance studies of 14-33 days
duration in a clinical research center [102]. GFR ranged from 1.0 to 3.0 ml/min in
three patients; the other five patients were anuric. The urea clearance from
dialysate and urine was 10.7 ± 0.3 liters/day . Patients were fed diets that provided
either 0.98 g protein/kg/day (low protein diet) or 1.44 g protein/kg/day (high
protein diet). Five patients received both diets; three ingested the higher protein
diet first. No patient was fed the same diet twice. Total energy intake (diet plus
dialysate) was 41.3 + 1.9 and 42.1 + 1.2 kcal/kg/day with the low and high protein
diets, respectively.
The balance studies for nitrogen and minerals in this study are shown in Table
6. Nitrogen balance, adjusted for changes in body urea nitrogen was +0.35 ± 0.83
g/day with the low protein diet and + 2. 94 ± 0.54 g/day with the high protein diet.
Only the high protein diet resulted in a nitrogen balance significantly different
from zero. Even if nitrogen balances are adjusted by about 1.0 g/day for losses
through skin, respiration, flatus and blood sampling [104], nitrogen balance on
the low protein diet would still be not different from zero. It was still significantly
positive in patients on the high protein intake. There was a curvilinear relation-
ship between dietary protein intake and nitrogen balance in the 13 studies.
Nitrogen balance rose as protein intake increased until protein intake was 1.09
g/kg/day. At this level, there was no further increment in nitrogen balance. The
SUN levels in patients receiving the high protein diet were not greater than
predialysis values commonly found in maintenance hemodialysis patients. SUN
levels averaged 67 ± 5.5 and 91 ± 7.0 mg/dl with the low and high protein diets,
respectively.
Anthropometric data also indicated that patients were anabolic with the diet
providing 1.44 g protein/kg/day. There was an increase in body weight in all
patients, in mid-arm muscle circumference in 5 of 6 patients, and in the sum of the
triceps and subscapular skinfold thickness in 4 of 6 patients with this diet. In
 UJ
0\
N

Table 6. Nitrogen and mineral balances in eight patients undergoing CAPDa

0.98 g/kg/day 1.44 g/kg/day

Diet Dialysate b Urine' Feces Balanced Diet Dialysateb Urine' Feces Balanced

Nitrogen 12.06±0.51 -9.80±0.65 0.97-1.72 1.61±0.09 +0.35±0.83 18.32±0.27 -12.67±0.93 2.61 1.80±0.08f +2.94±0.54g
g/day (7)e (6)h
Potassium 64±4.1(7) -36.6±2.6 2.7-11.6 16.9±1.1 +6.8±4.6 84.1±5.0(6)g -44.5±3.5 8.7 20.4±2. 1 +17.8±4.0
mEq/day
Phosphorus 1047±37(7) -324±23 16-103 468±53 +227±77 1915±117(6)g -332±36 223 839±67g +708±152
mg/day
Magnesium 259±25(6) -48±10 12, 31 158±23 +46±1O 320±11(6)f -44±12 40 203±30 +66±28 f
mg/day
Calcium 769±74(6) + 100±30 2,17 753±110 +112±51 1356±108(6)f +67±18 33 1220±213 + 198±194f .,
mg/day

a Data represent mean ± standard error of data collected during the 14 to 33 days of study. Adapted from (102).
b Minus sign indicates net loss from patient into dialysate; positive sign indicates net uptake from dialysate into patient.
c Indicates values in the three patients studied with the 1.0 g/kg diet and the one patient with the 1.4 g/kg diet who had urine output.

d Nitrogen balance is adjusted for changes in body urea nitrogen content but not for losses from cutaneous structures, respiration, flatus or blood drawing.
e Parentheses indicate the number of men studied with each diet; no patient was studied twice with the same diet.
Significantly different from corresponding values obtained with the 1.0 g/kg protein diet: fp<0.05, gp<[Link], hp<[Link]. Statistics were calculated either by non-
paired t tests of all the data or by paired t tests of values from the five patients studied with both diets.
, One patient received 0.25 ltg/day of 1,25-dihydroxycholecalciferol during the last 12 days of study.
 363

contrast, with the low protein diet, anthropometric measurements did not im-
prove. Serum proteins did not change with either diet.
From the above data it is apparent that there is considerable variability in
protein requirements for the clinically stable CAPD patient. The quantity of
dietary protein required may depend on the energy intake, and probably on the
biological value of the proteins. In most CAPD patients who are not mal-
nourished, an intake of 1.2-1.3 g protein/kg body wt/day is probably sufficient.
Protein requirements are likely to be higher in a malnourished patient or during
recuperation from an illness; such a patient may need a protein intake of 1.4-1.6
glkg in order to become anabolic.

3.5. Mineral balance

Since the independent variable in the studies described above [102] was the
protein intake, it is important to recognize that the mineral balances were
dependent variables, and the dietary intake of potassium, phosphorus, calcium
and magnesium each correlated directly with the nitrogen intake. See table 6.
Hence, it is possible that the balances for a given mineral were influenced by the
intake and balance of nitrogen or other minerals as well as by the intake of the
mineral in question. These factors limit the usefulness of the mineral balances for
determining dietary mineral requirements. Nonetheless, the relationships be-
tween the mineral intakes and balances should have some relevance to dietary
requirements.

Potassium. Potassium intake was greater with the higher protein diet (Table 6).
Dialysate losses accounted for approximately 70% of potassium output and feces
accounted for about 30%. Potassium balance was not different from zero with the
low protein diet and was significantly positive with the higher intake. Potassium
balance (y, mEq/day) correlated directly with potassium intake (x, mEq/day):
y = 0.64x - 35, r = 0.80, p<O.01. When potassium intake was 67 mEq/day or
greater, potassium balance was invariably positive. Serum potassium, obtained
serially throughout the studies in the 13 patients, was not different with the low
and high protein diets and averaged 4.0 ± 0.19 and 4.6 ± 0.31 mEq/l, respec-
tively.

Phosphorus. Dietary phosphorus intake and fecal phosphorus output were each
greater with the higher protein diet, and phosphorus balance was significantly
positive with both diets (Table 6). Serum phosphorus, measured periodically
during the studies, averaged 4.5 ± 0.3 and 4.8 ± 0.4 mgldl with the low and high
protein diets, respectively. The serum phosphorus was slightly increased in
several patients even though the patients ingested an average of 7.8 ± 1.0 glday of
aluminum hydroxide gel during the 13 studies. The doses of aluminum hydroxide
364

used in these studies appeared to increase fecal phosphorus excretion by about

250 to 300 mg/day. Despite the intake of these phosphate binders, the net
intestinal absorption of phosphorus (dietary minus fecal phosphorus) correlated
directly with phosphorus intake (r = 0.87). These observations confirm that
restriction of dietary phosphorus intake is the most effective method for reducing
intestinal phosphorus absorption. The serum phosphorus level was always 5.1
mg/dl or less when phosphorus intake was 1200 mg/day or lower. This data does
not support the contention that hyperphosphatemia may be more easily pre-
vented in uremic patients undergoing CAPD than in those treated with
hemodialysis.

Magnesium. Dietary magnesium intake and balance were also greater with the
higher protein diet, and magnesium balance was significantly positive with both
diets. Mean serum magnesium concentrations were abnormally high, averaging
3.1 mg/dl, with the low and high protein diets. Net intestinal magnesium absorp-
tion ranged from 74 to 200 mg/day except for one patient who had a net intestinal
magnesium loss of 49 mg/day. This patient also had large fecal phosphorus and
calcium losses. The positive magnesium balance and high serum magnesium
levels were related to the rather small magnesium losses into dialysate which
averaged 48 and 44 mg/day with the low and high protein diets, respectively. This
small dialysate output was related to the high magnesium concentrations in the
dialysate inflow, 1.85 mg/dl (1.5 meq/l). Delmez and associates using similar
dialysate magnesium concentrations also reported low dialysate losses of magne-
sium which averaged 31 ± 15 mg/day [105]. In some dialysate preparations, the
magnesium concentrations have now been lowered, and in outpatient studies this
has reduced serum magnesium concentrations, and probably increased the ac-
ceptable upper limits of dietary magnesium intake for CAPD patients.

Calcium. Calcium intake and balance were significantly greater with the 1.4 g/kg
protein diet (Table 6). The major calcium output was from feces. Calcium
balance was not different from zero with the high calcium diet because of the large
fecal calcium losses in the patient with the large fecal phosphorus and magnesium
output. Except for this patient, calcium balance was always neutral or positive
when dietary intake was equal to 720 mg/day or greater. There was net calcium
absorption from dialysate which averaged 84 mg/day in the combined 13 studies.
Net calcium absorption from dialysate was inversely correlated with the serum
calcium concentration which averaged 8.9 mg/dl with the low protein diet and 9.2
mg/dl with the high protein diet.
These findings are consistent with those of Delmez and associates [105] and
Kurtz, et al. [106] who reported that dialysate calcium uptake was affected by both
the serum ionized calcium and the concentration of dialysate glucose, which is a
determinant of the volume of dialysate outflow. Delmez and co-workers studied
ten patients undergoing CAPD who underwent four two-liter exchanges each
 365

day: three exchanges with 1.5% dextrose and one exchange with 4.25% glucose;
dialysate calcium concentration was 3.5 mEqll. The serum ionized calcium aver-
aged 4.9 mg/dl. There was a mean of 9.8 mg/day of calcium taken up from
dialysate. When serum ionized calcium was above normal (greater than 5.0
mg/dl), there was a net calcium loss into dialysate of77 mg/day [105]. When serum
ionized calcium was below normal (less than 4.4 mg/dl) , there was a net uptake of
calcium from dialysate of 44 mg/day. Similarly, with a single 1.5% dextrose
exchange, there was a net calcium uptake from dialysate of 9.8 mg per exchange,
and with a 4.25% dextrose exchange there was a net calcium loss into the
dialysate of 21 mg per exchange. The foregoing observations indicate that with
the calcium concentrations currently employed in peritoneal dialysate, calcium
uptake and losses from dialysate will be small. Most of the calcium taken into the
body will come from the diet.

3.6. Urea and nitrogen metabolism in CAPD

Serum urea nitrogen (SUN) levels have been reported to be lower in CAPO
patients than would be predicted from the estimated protein intake and the
calculated urea clearance by dialysis. Thus, Nolph, Popovich and Moncrief
predicted that an SUN of 70 mg/dl could be maintained during CAPO with five 2L
exchanges/24 hours [107]. In actuality, the SUN averaged 51 ± 5.3 SE mg/dl in 12
patients being treated with CAPO and believed to be ingesting 80 g protein/day
[108]. Moreover, in other studies with patients performing only four 2L ex-
changes/day, SUN levels have averaged 59--63 mg/dl [108, 109]. It was suggested
that patients treated with peritoneal dialysis may excrete N through some unusual
route [110] or that there may be marked anabolism [107]. To resolve these
questions and to assess the relationship between nitrogen intake, nonurea nitro-
gen losses, and urea nitrogen appearance (UNA), urea nitrogen appearance (net
urea generation) and losses of various nitrogenous constituents were measured
during 12 metabolic studies in patients undergoing CAPO [111]. Diets provided
either 1.0 or 1.4 g protein/kg/day, averaging 78 and 115 g protein/day. Total urea
clearance (dialysance plus renal) averaged 10.8 L/day and was similar with both
diets. SUN averaged 62 and 89 mg/dl with the two diets. UNA was 6.3 g/day on
the lower protein diet and 10.1 g/day on the higher protein diet. UNA comprised
only 55% and 68% of total N output (TNO) with the low and high protein diets,
respectively. For the total group studied the dialysate nitrogen accounted for 88%
of TNO. The division of dialysate nitrogen was urea N 70%, protein N 14.2%,
amino acid N 5.4%, creatinine N 4.5%, uric acid N 1.8%. Fecal N accounted for
13.8% ofTNO. In three patients with residual renal function, urine N was 14%.
Except for urea, the fractions did not differ with the two diets. Thus, because of
protein and amino acid losses, urea comprised a relatively low fraction of nitro-
gen losses. This accounts for the 'discrepancy' between predicted and observed
SUN values.
366

As shown in figures 3 and 4 in the CAPO patients there was a close correlation
between TNO and UNA. The relationship is similar to that in normal, uremic or
hemodialysis patients except that for any given TNO, CAPO patients have a
lower UNA by approximately 2 g nitrogen/day (see figure 3): this discrepancy
matches closely the average dialysate nitrogen losses as protein and amino acids.
The fraction of N excreted as urea is known to become smaller with reduced
dietary intake. This tendency may be accentuated in patients treated with CAPO
because of the substantial and relatively constant loss of protein and amino acids
into dialysate. The close correlation between TNO and UNA suggests that UNA
may be used to estimate TNO. Changes in dialysate protein losses and marked
catabolism or anabolism can clearly alter this relationship. When patients are in
neutral N balance, i.e. N intake equal N output, protein intake will also correlate
closely with UNA. Thus, in relatively stable patients undergoing CAPO this data
suggests that protein intake may be estimated from the measurement of UNA.

3.7. Estimation of dietary protein intake from urea nitrogen appearance (UNA)
and the amount of CAPD required

As described above, metabolic balance studies in CAPO patients have demon-

strated a linear and predictable relationship between UNA and total nitrogen
output (TNO) (see figure 4). Since the dialysate/plasma urea nitrogen ratio is 1,
16 •
w
u
z<{ 12
•
a::
<{-
°8
w~
a..
a.. "0
<{'
0'1
Z- 4
<{
w
a::
~ O~-.__.--.-,__,--,__,--.__,-,
4 8 12 16 20

TOTAL NITROGEN OUTPUT(g/day}

Figure 3. Relationship between total nitrogen output and urea nitrogen appearance. The data are
shown for 12 studies in normal subjects (D), 14 studies in patients with chronic renal failure (~) and 26
studies in patients undergoing maintenance hemodialysis (e). The solid line depicts the regression
analysis for these combined data for normal subjects, patients with chronic renal failure, and those
undergoing hemodialysis (112-114) (r = 0.98). The data for the 12 studies in patients undergoing
CAPD are indicated by X (r = 0.96) (from Blumenkrantz et aZ. [111]; reproduced with permission).
 367
,.....
~
17
x
Cl
.......
C)
16 x
~ 15
:::>
a..
~ 14
:::>
0
z 13
lLJ
C)
0 12 y= 84x + 61
a::
~ r = 96
z II
...J x
10
~
g i i i i I I I I I
5 6 7 8 9 10 II 12 13
UREA NITROGEN APPEARANCE (G / DAY)
Figure 4. Relationship between total nitrogen output and urea nitrogen appearances in 12 studies in
patients undergoing CAPD (from Blumenkrantz et al. [111]; reproduced with permission).

UNA, (g/day), in an anuric CAPD patient is simply the serum urea nitrogen
(SUN), (gil), multiplied by the volume of dialysate effluent, (l/day). If the patient
has some residual renal function then the daily urinary urea N clearance should be
added to the peritoneal urea clearance in order to calculate total clearance. In
stable patients TNO is equal to total N protein intake. Even if the patient is
slightly anabolic or catabolic the amount of N accrued or lost by the patient/day
amounts to half a gram of N or less, an amount which will not materially effect the
validity of this estimation. Therefore, urea N appearance (SUN x L/day drai-
nage) is directly related to dietary protein intake and the latter can be estimated
by knowing the former. A nomogram has been constructed to facilitate this
determination (figure 5). A ruler is placed on the line connecting SUN and total
urea clearance; the point at which it intercepts line C is the estimated dietary
protein intake. For example, a SUN of 80 mg/dl in a patient having a total urea
clearance of 12l/day would indicate a dietary protein intake of approximately 100
g/day; a similar SUN level will result if a patient is getting only 61/day of clearance
and ingesting 60 g of protein/day. Thus in the outpatient management of CAPD
patients the SUN concentration, combined with the total urea clearance (dialy-
sate plus renal), can be used to easily estimate dietary protein intake.
The assessment of 'adequacy' of dialysis is complex and multifaceted. The
approach we suggest is not an attempt to define 'adequate' dialysis for the CAPD
patient.
We first determine the patient's dietary protein requirements (1.2 to 1.5 g/kg
depending on nutritional status); next we determine a desired SUN level (!!ener-
368

Figure 5. Normogram describing the relationship between serum urea nitrogen, total urea charance
and dietary proteen intake (see text).

ally 70-100 mg/dl). Then by the use of the nomogram described above (figure 5),
we estimate the total daily urea clearance that will achieve the desired SUN at the
prescribed protein intake. The number of exchanges/day or per week to achieve
the total C urea is then determined (based in part on the amount of daily
ultrafiltration). If some residual renal function exists, it will make a major
contribution to total C urea (Table 7). Dialysate clearance is not dependent on
the number of bags instilled but rather on the amount of dialysate drained out.
Four exchanges/day of a dialysate containing 1.5% dextrose may result in 8.5
L/day of effluent dialysate in one patient and 11 L/day effluent dialysate in another
patient who uses four exchanges of 4.25% dextrose dialysate. The total urea
clearance will differ in the two patients as a function of the total volume of the
effluent dialysate.
The use of the nomogram will enable one to assess the dietary protein intake of
a patient. For example, a stable, well nourished 75 kg man undergoing CAPD,

Table 7. Influence of a decrease in renal urea clearance of 2 ml/min on CAPD requirements

1 ml/min = 1440 milday or 1.44 L/day

2 mllmin = 2.88 or ~3L1day
... effluent dialysate volume must be increased by 3L/day to have the same total C urea
 369

having a daily total urea clearance of 11.5 L/day and a SUN of 80 mg/dl is ingesting
approximately 90 g protein/day which is probably an adequate amount. If his
daily urea clearance is decreased to 6.5 L/day he can still maintain an SUN of 88
mg/dl; however, his protein intake will be only approximately 60 g/day and he
probably will become protein depleted in the long term (Figure 1).
Currently, for our CAPD patients, we employ a flexible dialysis schedule [115].
In order to achieve the desired dialysate C urea, the number of exchanges to be
performed is determined on a weekly basis; a minimum number of exchanges/day
is prescribed as well as the total/week. Since the dialysate urea nitrogen concen-
tration is approximately 80% of the SUN after 2.5 h [107], the dwell time is
permitted to range from 2.5-10 h. We find that a dialysis schedule such as this,
which allows considerable flexibility in the patients daily routine alleviates some
of the boredom and restrictiveness CAPD patients experience on rigid schedules.

3.B. Middle molecules, protein malnutrition and peripheral neuropathy

The permeability characteristics of the peritoneal membrane and its continuous

nature make CAPD quite efficient in the elimination of middle molecules (MM).
Bergstrom and co-workers have observed that, in CAPD, the peritoneal cleara-
nce of four different MM fractions on the average was 89% of the creatinine
clearance, compared to 40% in hemodialysis [116] (Table 8). Plasma 7a and 7c
levels correlate strongly with protein intake. Furthermore, the generation rate of
the MM fraction 7c in CAPD patients was the same as in normal persons. As
shown in table 8 the plasma levels of MM 7a, band c are lower on the average than
those in hemodialysis patients. High plasma levels of 7c, presumably due to
increased generation rate, frequently are associated with 'sickness' in chronic
dialysis patients; a normal 7c generation rate might imply a better metabolic
control of uremia in CAPD patients than in other patients with ESRD [116].
In contrast to the above, the same group observed a significant deterioration of
peripheral nerve function in CAPD patients [117]. These findings seem to contra-
dict the MM hypothesis in light of the decreased MM levels in these patients. The
observed decrease in peritoneal MM clearance with time on CAPD [118] com-

Table 8. Plasma concentration (cm2/ml) of Middle molecules in. hemodialysis & CAPD

Treatment N 7a 7b 7c 7d

Hemodialysis 29
Pre 9.6 ± 9.5 8.2± 6.6 22±37 3.1 ± 3.4
Post 3.6± 4.2 3.2± 2.6 11 ± 18 1.9 ± 2.6
CAPD 14 3.3 ± 1.0 4.9 ± 0.7 0.8±0.1 3.4 ± 0.8

Adapted from Bergstrom [116].

370

bined with a diminishing residual renal function may be of importance in this

regard. Bergstrom et at. have also found high MM peaks in the plasma of six
protein-depleted hemodialysis patients with reduced muscle mass; this is despite
the fact that their SUN and serum creatinine levels were lower [119]. Three of
these patients had severe progressive polyneuropathy. In an earlier study the
same group observed that good dietary control with supplementation of sufficient
amounts of essential amino acids could arrest the progression of subclinical
neuropathy [120]. These findings emphasize the role of protein depletion in the
development of uremic neuropathy.
In so far as the CAPD patient is concerned, the lesson that may be derived from
the above observations is that despite the fact that MM clearances are high in
CAPD patients, inadequate dialysis i.e. too few exchanges in the face of dimin-
ishing residual renal function, associated with decreased dietary protein intake
may result in the development of peripheral neuropathy and failure to thrive.
Adequate CAPD treatment, especially in the larger (>80 kg) patient (this may
mean 11-12 L of total urea clearance/day), careful monitoring of residual renal
function (most easily noted by an increase in serum creatinine on a fixed dialysis
regimen) and careful attention to the provision of a sufficient amount of protein
may prevent long-term problems. The use of oral protein supplements or the
addition of amino acids to dialysate may help in the last regard.

3.9. Carbohydrate

The dialysis solution used for intermittent peritoneal dialysis usually contains 15 g
glucose/I, but concentrations as high as 42.5 gil are occasionally employed and
concentrations as high as 70 gil have been available in the past [121-124]. Anders-
son and co-workers [122] reported that an average of 9.5 and 38 g glucose was
absorbed each hour when peritoneal dialysate containing 15 and 42.5 g glucose/I,
respectively, was used. Glucose absorption in patients undergoing intermittent
peritoneal dialysis has been reported to vary from 1. 7 to 18 g/21 exchanges with
dialysate containing 15 gil [121]. With the extremely hypertonic dialysis solution
(70 g glucose/I), glucose absorption ranged from 25 to 130 glexchange. The wide
range in the rate of glucose absorption has been attributed to differences in
peritoneal membrane permeability [121]. The relatively short duration of each
treatment with intermittent peritoneal dialysis limits the nutritional significance
of glucose absorption from dialysate.
Studies conducted in patients undergoing CAPD have indicated that substan-
tial quantities of glucose are absorbed (78-361 g/day) [78, 125-127]. Because of
this absorbed glucose, carbohydrate is the major source of energy in CAPD
patients. We found that in each patient the amount of glucose absorbed per day
on any given dialysis regimen was quite constant; the average co-efficient of
variation was 4.9% (range, 1.8 to 11.9%) [127]. The relation between the amount
 371

of glucose absorbed each day and the average concentration of glucose in dialy-
sate was so close that the net glucose uptake could be predicted from the
concentration of glucose in the inflow using the equation:
glucose absorbedll dialysate = 11.3 (average daily glucose concn.)-l1.
For example, a CAPD patient performing five exchanges per day with two 2
liter exchanges of 1.5% dextrose (actual measured glucose concentration aver-
ages 1.30 g/dl) and three 2 liter exchanges of 4.25% dextrose (actually 3.86 g/dl)
will have an average daily glucose concentration of 2.80 g/dl. The quantity of
glucose absorbed from each liter of inflow can be predicted as follows:
Glucose absorbed = 11.3 (2.8) - 11 = 20.7 g glucosell dialysate inflow or 20.7
gil X 10 I inflow = 207 glday.
The capability of varying glucose absorption by altering the dialysate glucose
concentration may prove useful; in a malnourished patient, it may be advisable to
increase the number of hypertonic exchanges. Such a patient could increase his
fluid and salt intake to enable the use of an increased number of 4.25% dialysate
exchanges. On the other hand, a patient with obesity andlor an elevated serum
triglyceride level could restrict his salt and fluid intake so as to use fewer
hypertonic exchanges, thereby reducing dialysate glucose absorption. Although
glucose absorbed from dialysate is not inordinately large compared to dietary
carbohydrate intake, exhaustion of the pancreatic beta cells has been postulated
to be a potential hazard of CAPD. With CAPD, most of the glucose absorption
occurs in the 90-120 minutes following instillation of dialysate. Plasma insulin
levels rise and glucagon levels decrease [90]. Lindholm and co-workers per-
formed oral glucose tolerance tests and determined serum insulin and glucagon
levels in 9 patients before and 2-4 months after initiation of CAPD [78]. They
found no impairment in glucose tolerance or insulin and glucagon responses. On
the other hand, Fuchs found an abnormal GTT and impaired insulin release in 6
CAPD patients [128). Further investigation perhaps using the insulin and glucose
clamp method [12] would help to determine if abnormalities in carbohydrate
tolerance develop in CAPD patients.

3.10. Lipids

Cardiovascular death is common in patients receiving long term dialysis [129,

130]. Abnormal lipid metabolism and altered serum lipids are recognized to be
consequent of progressive renal insufficiency [131, 137] and may accelerate athe-
rogenesis. Stimulated by the report of Lindner et al. [133] linking cardiovascular
disease with altered lipid patterns in dialysis patients, considerable research in the
mechanisms of such altered lipid metabolism has been performed and reported.
In patients with chronic end stage renal disease an atypical lipoprotein pattern
is encountered which is similar to that found in non-uremic patients with the type
372

IV dyslipoproteinemia of Frederickson. Although variability amongst authors is

encountered, a concensus appears to exist that such patients have low levels of
high density lipoproteins (HDL) with elevated serum triglycerides. Total serum
cholesterol levels appear to be abnormally high and the triglyceride elevation
appears to consist predominantly of very low density lipoprotein levels (VLDL).
Low density lipoprotein (LDL) triglyceride levels also appear to be slightly
elevated but the changes in low density lipoprotein cholesterol appears to be
variable [134, 135]. The major pathogenetic mechanism behind these changes in
lipid metabolism would appear to be a decreased removal of triglycerides from
the blood stream which is caused by a deminished activity of lipoprotein lipase.
To a lesser extent the defect may be worsened by a diminished hepatic trig-
lyceride lipase activity [136] resulting in increased VLDL production. Hyper-
lipoproteinemia may be a particular problem for patients undergoing peritoneal
dialysis [99-101]. Cattran et al. [137] found that in 64 patients, the average serum
triglyceride level of 195 mg/dl at the start of a single hemodialysis, decreased to 118
mg/dl at the end of treatment. In contrast, serum triglyceride levels were 313 and
312 mg/dl before and after a peritoneal dialysis treatment. Triglyceride removal
was defective in patients undergoing either type of dialysis, but worse in those
treated by peritoneal dialysis [137]. It is possible that the difference in serum
triglyceride levels between patients undergoing intermittent peritoneal dialysis
and hemodialysis may in part be attributable to differences in the patient popula-
tions [138]. With matched pairs of patients undergoing dialysis, subnormal levels
of high density lipoproteins were found in those treated by hemodialysis; this was
not the case in patients undergoing intermittent peritoneal dialysis [139].
Impaired glucose tolerance may worsen the dyslipoproteinemia in uremia and
be a particular problem for patients undergoing both CAPD and CCPD. In-
creased triglyceride synthesis occurs as a result of increased carbohydrate intake.
Even in nondialyzed patients who have advanced renal insufficiency, carbohydr-
ate intake restriction results in reduced serum triglyceride levels. Bailey et al.
[140] have demonstrated in hemodialysis patients a reduction in serum trig-
lycerides by 60% when dialysate glucose concentration is reduced. Patients
undergoing CAPD, CCPD or even IPD absorb large amounts of glucose by the
peritoneal route, especially when the more hypertonic dextrose solutions are
employed to remove excessive body water. Studies of CAPD patients show
substantial quantities of glucose being absorbed from the dialysate and plasma
insulin levels are often elevated.
Increased plasma insulin levels may in themselves be atherogenic by stimulat-
ing smooth muscle proliferation and lipid synthesis in the arterial wall [141]. The
continuous glucose supply provides a substrate for the synthesis of very low
density lipoproteins in the liver and consequently result in increased circulating
serum triglyceride levels. Even in normal individuals a constant high intake of
carbohydrate induces triglyceride synthesis [142]. Turgan et [Link] demon-
strated accelerated hypertriglyceridemia in patients on CAPD [143]. Lindholm et
 373

al. [144] have demonstrated that there is a predictive inverse correlation between
the predialysis rate of clearance of VLDL triglycerides after a fat load and the
subsequent severity of the dyslipoproteinemia on CAPD. Kensch et al. have also
demonstrated that the type IV dyslipoproteinemia profile is worsened in uremic
patients who undertake CAPD [145]. Of particular import is the observation by
Turgan et al. [143] that the changes in triglycerides and cholesterol can be
normalized by carbohydrate restriction in patients undergoing CAPD and that
the therapeutic consequences of worsening triglycerides may be avoided. Lipid
abnormalities in patients undergoing CAPD have been described by others [146-
150].
Lindholm et al. have noted in a study o"f 22 hemodialysis patients and 22
patients undergoing CAPD that although cholesterol, triglyceride, low density
lipoprotein and very low density lipoprotein levels are worse after the initiation of
peritoneal dialysis, they return to normal after a 6-12 month period of time. The
variability in reported alterations of lipoprotein profiles may thus be a con-
sequence of the time at which such measurements are made with respect to
initiation of peritoneal dialysis and additionally be a consequence of the recent
prior interperitoneal glucose absorption. Systematic control and analysis of these
variables is needed to permit interpretation of lipoprotein profiles and their true
clinical meaning.
The use of interperitoneal insulin to manage diabetic patients undergoing
CAPD has been successful in glucose control and in overall successful manage-
ment of these difficult patients [151]. As noted by Beardsworth et al. it has
unfortunately not ameliorated the hyperlipidemia. In their study of non-diabetic
patients all with type IV dyslipoproteinemia the addition of insulin to dialysate
did not result in correction of that defect [152].
Felts et al. [153] have demonstrated that an alpha acid glycoprotein, oroso-
mucoid is lost in peritoneal dialysate. This protein is complexed to heparan
sulfate, a potent activator of lipoprotein lipase, the enzyme primarily responsible
for clearance ofVLDL triglycerides from plasma. It has further been found that a
precursory form of heparan sulfate circulates bound to another protein, a pro-
teoglycan. This proteoglycan and its associated heparan sulfate are also lost into
peritoneal dialysate with resultant reductions in circulating free heparan sulfate
and reductions in circulating lipoprotein lipase (J. M. Felts, personal communica-
tion). Reduction in free heparan sulfate and consequent loss of lipoprotein lipase
activity in plasma most probably contribute to the impeded clearance of VLDL
triglycerides.
Recently considerable interest has been focused on the role of carnitine in
altering the metabolic pathways of lipids in patients undergoing peritoneal di-
alysis [20-22, 154]. Buoncristiani et al. [20] have demonstrated that patients
undergoing continuous ambulatory peritoneal dialysis have low serum carnitine
levels and increased losses of carnitine compared to patients undergoing either
hemodialysis or intermittent peritoneal dialysis. Moreover when interperitoneal
374

carnitine was employed during dialysis by Bazzato et al. a significant fall in

circulating triglycerides was reported. Amair et al. also have studied serum
carnitine levels in patients undergoing CAPD [21]. Although they noted trig-
lyceride levels were abnormal in the CAPD group, no differences in carnitine
levels were found as compared with their control group. Albright et at. [22] also
have studied carnitine levels in CAPD patients. They studied 9 patients of whom
5 were diabetic and found that all had normal carnitine levels in their blood.
Additionally all had normal muscle carnitine concentrations. 4 of the 9 patients
had type IV dyslipoproteinemia [22]. Thus it must be concluded at the present
time that although the possible role of carnitine alterations in causing the dys-
lipoproteinemia is of great interest, it will require further time and effort to
understand its true role.

3.11. Energy requirements

Energy requirements in normal individuals differ with age, sex, height, and
degree of activity. An adequate energy intake is a prerequisite for efficient
utilization of dietary protein to permit growth and maintenance. The recom-
mended dietary allowance (RDA) for normal non-obese adults range from 33-39
kcal/kg body wt/day to provide weight maintenance [155]. Diet surveys have
demonstrated that the average adult hemodialysis and CAPD patient on an
ad libitum diet ingests only 23-28 kcal/kg body wt/day. This caloric intake is
probably inadequate for maintenance and definitely inadequate for long term
repletion of energy stores.
If catabolic illness supervenes, caloric intake usually falls further despite the
acute and considerable increase in energy needs [156, 157]. The energy needs of
the dialysis patient may differ from normal because of various hormonal and
metabolic alterations in uremia. Wasted patients or those with superimposed
illnesses may have increased energy requirements due to the various metabolic
changes which occur in uremia [158]. Oxygen consumption per kilogram body
weight may be increased in uremia [159] reflecting altered cellular metabolism
[10] or the high oxygen requirements of viscera which are disproportionately
preserved during starvation. The energy requirements of an individual patient
may vary greatly on a per kg body weight basis, depending upon how much
muscle and adipose mass had been lost. In very malnourished individuals, or in
the presence of superimposed illness such as peritonitis, caloric intake will have to
be increased to 40-50 kcal/kg body wt if energy balance is to be achieved.
In normal individuals it appears medically important to maintain body weight
within 10-15% of desirable or 'ideal' body weight - an arbitrary weight which can
be shown statistically to be associated with the lowest possible mortality. In adult
men undergoing maintenance hemodialysis, the post dialysis 'dry' body weight is
often similar to desirable body weight, yet these men often have evidence for
 375

malnutrition and wasting. Therefore the desirable body weight for normal adults
may not be desirable for patients with chronic renal failure. How, then, can we set
goals and standards for this population? We currently use either the patient's
usual weight prior to the onset of renal failure as a goal or his relative body weight
(RBW) in conjunction with anthropometric measurements of body fat. We
believe that the goal should be to maintain a patient at no less than 100% of RBW,
and generally do not initiate a fat reduction diet unless the patient is demonstrably
obese with a body weight greater than 115% RBW or 125% desirable (ideal). We
prefer to maintain these relatively high weights, because dialysis patients are
prone to lose weight, particularly when they sustain intercurrent catabolic
stresses, such as peritonitis.
There are few data concerning energy requirements in patients undergoing
peritoneal dialysis. Our current recommendations for adults are shown in Table
9. At the present time it seems prudent that a well nourished patient receive a
total of 35 kcallkg body wt/day. Dietary energy requirements are lower in patients
undergoing CAPD because of the glucose absorption from the peritoneal cavity.
Energy derived from peritoneal glucose can be estimated (vida supra). Grams of
glucose absorbed are multiplied by 3.7 to derive calories [127]. Obese patients

Table 9. Daily dietary allowances for CAPD patients tentative proposal'

Protein 1.2-1.3 glkg normalized body wtb

Energy (oral and dialysate) 35-42 kcal/kg normalized body wt b
Carbohydrate (oral) 35% of ingested calories
Fat Remainder of ingested nonprotein calories
Polyunsaturated:saturated fatty acid ratio 1.5:1.0
Total fiber 20--25 g
Calcium 1000-1400 mg
Phosphorus 700-1200 mg
Magnesium 200-300 mg
Potassium 60-80 mEq
Sodium and water In the non-obese patient sufficient to permit
1.5-2.5L of ultrafiltrate/24H
Supplemental Vitamins
Ascorbic acid 100 mg
Pyridoxine hydrochloride (B6) 10 mg
Thiamin hydrochloride 2.0 mg
Folic acid 1 mg
Other water soluable Recommended daily allowances
vitamins for normal persons
Vitamins A, E and K None
Vitamin D See text
Trace Elements See text

• See text for discussion and recommendations

bPatient's normalized body weight is the average body weight of normal persons of the same age,
height and sex as the patient.
376

(greater than 115% RBW or 125% desirable) should have some caloric restriction
and probably also should be salt restricted so that the frequency of use of higher
dextrose concentrations in dialysate can be minimized. Wasted patients, those
undergoing catabolic stress or patients recuperating from an illness probably
benefit from an increased total energy intake (4(}"'50 kcal/kg body wt/day).

3.12. Vitamins

Loss of water soluble vitamins can occur during peritoneal dialysis. With CAPD,
concern has been expressed that depletion of various vitamins and trace el-
ements, especially those that are protein bound, may occur. Depletion is more
likely in patients who have frequent episodes of peritonitis, when dialysate
protein loss is increased and dietary nutrient intake is inadequate.
Blumberg et al. studied vitamin levels in ten CAPD patients who were eating
normally and not receiving vitamin supplements [160]. Retinol binding protein
and fat soluble vitamins A and E were elevated. They reported that vitamin Bl,
vitamin B6, folic acid, and vitamin C were frequently reduced. Vitamin B2 and
B12 were normal, although serum vitamin B12 levels tended to decline with time
in patients that did not receive supplements of this vitamin. Dietary intake of
several vitamins were often reduced below the recommended allowances for
normal adults. These included vitamin A, vitamin Bl, vitamin B6, vitamin B12,
and nicotinamide. These observations confirm previous reports of low intakes for
many vitamins in chronic renal failure in hemodialysis patients [161].
It therefore appears that low levels of the water soluble vitamins are probably
caused by both low dietary intake and losses into dialysate. Blumberg et al. as well
as Trapas and co-workers [160, 162] demonstrated a high peritoneal clearance of
ascorbic acid. Papadoyanakis [163] et al. noted high peritoneal folate clearance.
In addition to poor dietary intake and peritoneal losses, medicines which inhibit
absorption or impair action of vitamins may promote vitamin deficiency [26, 27].
Finally, it is possible that altered metabolism in uremia may contribute to vitamin
deficiency.
Aloni et al. reported low serum 25-hydroxyvitamin D and 25-hydroxyvitamin D
binding capacity in CAPD patients [164]. There was a mean daily loss of
25-hydroxyvitamin D in peritoneal fluid of 1491 ± 260 ng/day and a loss of
25-hydroxyvitamin D binding capacity into dialysate of 153 ± 28 nmole/day.
Serum 25-hydroxyvitamin D levels were also lower than in maintenance hemo-
dialysis patients. Kurtz and coworkers found very low levels of serum 1,25-
dihydroxyvitamin D levels and 24, 25-dihydroxyvitamin D concentrations in
CAPD patients [106]. Serum 25-hydroxyvitamin D was only slightly decreased
and did not change with 6-12 months of treatment with CAPD. However, Gokal
and associates reported that serum 25-hydroxyvitamin D levels fell with time in
 377

CAPD patients [165]. On the other hand, Delmez and associates reported that
patients treated with CAPD for 6 months had normal serum concentrations of
25-hydroxyvitamin D 21 ± 3 ng/ml, which did not differ from serum values in
maintenance hemodialysis patients (28 ± 6 ng/ml) [105]. Serum 25-hydroX'y-
vitamin D levels were measured serially in five patients and did not change during
the course of treatment with CAPD. Serum vitamin D binding protein levels were
normal in their CAPD patients, 610 ± 21 ug/ml (normal range 400--650 JLg/ml).
These concentrations of vitamin D binding protein were significantly higher than
those of maintenance hemodialysis patients despite the average losses of 6.2 ± 2.2
mg of vitamin D binding protein with each 1.5% dextrose exchange. These
conflicting data suggest that 25-hydroxyvitamin D levels mayor may not be
decreased in CAPD patients; serum 1,25-dihydroxyvitamin D and 24,25-di-
hydroxyvitamin D appear to be reduced in these patients.

3.13. Trace elements

Over the last few years there has been considerable investigation of the altera-
tions in tissue and blood levels of trace elements in hemodialysis patients.
Theoretically, CAPD patients should have lower levels of trace elements than
hemodialysis patients because the volume of dialysate the CAPD patient is
exposed to is less than with hemodialysis, possibly reducing the absorption of
potentially toxic elements [25]. Protein bound trace elements are lost in dialysate.
Thompson and co-workers found a marked reduction in red blood cell concentra-
tions of zinc and copper although plasma concentrations (when adjusted for
decreased albumin concentrations) were normal. Zinc deficiency can cause hypo-
geusia, anorexia and weakness; such symptoms are common in CAPD patients.
Therefore, it would probably be prudent to give zinc supplements to CAPD
patients. Thompson found markedly increased whole blood chromium levels
which may have been related to dialysate contaminations [25]. Thompson et al.
also found normal plasma manganese, RBC lead, whole blood cadmium and
increased plasma aluminum levels [103].
Other researchers have also found increased serum aluminum in CAPD pa-
tients [166, 167] but not to the degree seen in hemodialysis patients. Peritoneal
dialysate contains aluminum 10--50 JLg/L [166]; however the percentage of Al
which is nonabsorbable particulate Al versus that which is soluble is yet to be
determined. Despite this, Al which is protein bound, is removed in dialysate;
Hercz and associates found an average daily loss of 206 ± 23 JLg into dialysate
[167]. Serum aluminum is elevated in CAPD patients who have not received
aluminum binders of phosphate, but serum aluminum levels also correlate with
the total intake of aluminum binders of phosphate. It has been repeatedly claimed
that less phosphate binders are required in CAPD patients than in hemodialysis
378

patients; however, we have shown that intestinal phosphorus absorption is nor-

mal and that patients eating adequate protein diets (1.1-1.5 g/kg) require an
average of 7.8 g of A10H to maintain a mean serum phosphorus level of 4.6
mg/dl. It is therefore quite likely that CAPD patients who require little or no
aluminum hydroxate binders are ingesting insufficient quantities of protein.
Aluminum related osteomalacia has been described in CAPD patient, and some
patients have been treated with desferrioxamine to remove aluminum [166, 167].
In a CAPD patient given 6 g of desferrioxamine intravenously, the aluminum
losses into peritoneal dialysate increased to 3.5 to 4.7 mg in 24 hours [167].
Milman et al. evaluated the iron status in twelve peritoneal dialysis patients
[168]. On the whole they had normal iron absorption capacity. The authors
strongly recommended oral iron supplementation for peritoneal dialysis patients
because of their low dietary iron intake and increased iron losses.
Additional studies need to be done to determine trace element levels in tissue
as well as their concentrations in fresh and effluent dialysate to assess the
importance of absorption or loss of trace elements across the peritoneal
membrane.

4. Other nutritional considerations

Based on the foregoing considerations, it is possible to develop a tentative

proposal for daily nutrient allowances for clinically stable adult CAPD patients
(Table 9). These proposals should be considered to be a first approximation and a
basis for further inquiry. It is clear that additional studies will lead to modifica-
tions in the recommended allowances. In future studies of nutritional allowances
for CAPD patients, the intake of the individual nutrient under evaluation should
be varied independently of others. The optimal intake of energy, protein, vi-
tamins, and minerals should be assessed by comparing the nutrient intake of
patients to such outcome measurements as body composition (e.g., total body
nitrogen, potassium, fat), body nutrient pools (e.g., albumin pools), nitrogen and
mineral balances, biochemical indices of nutritional status, and the physiological
and clinical condition of the patient.
When evaluating the proposed nutritional allowances, the following consider-
ations should be kept in mind: 1) Since excessive and insufficient intakes may each
be hazardous, both the upper and lower ranges of safe nutritional allowances
should be followed. 2) Malnourished patients may require more protein, energy
and other nutrients than well-nourished patients. 3) Patients with moderate or
severe obesity require less energy. 4) Abnormalities in serum lipids and lipopro-
tein concentrations may require changes in the composition of the diet. 5)
Changes in the dialysis regimen or in dialysate composition may also alter dietary
allowances. 6) Nutritional needs may be very different in CAPD patients with
 379

superimposed catabolic illnesses. 7) Finally, the nutrient intake must always be

adjusted according to the clinical response of the individual patient.
The recommended daily protein intake is based upon the observation that
nitrogen balance became more positive as protein intake was increased up to 1.09
g/kg/day and greater protein intakes did not lead to more positive nitrogen
balance. Allowing for individual variation, we propose a daily protein allowance
of 1.2-1.3 g/kg/day (Table 9). If the dietary requirements for essential amino acids
are similar to normal individuals, then allowing 50% of the daily protein intake to
be of high biological value should provide a surfeit of these amino acids. Patients
who are given amino acids in dialysate may have a lower dietary protein require-
ment.
If patients are not obese or hypertriglyceridemic, it is recommended that
attempts should be made to maintain total energy intake (diet plus dialysate) at 42
kcal/kg/day; unfortunately, for many CAPD patients such a high energy intake
will be difficult to attain. It is recommended that dietary carbohydrate should be
composed primarily of complex carbohydrates and should provide about 35% of
ingested calories. Dietary fat should provide the remainder of the ingested
nonprotein calories. The polyunsaturated to saturated fatty acid ratio should be
about 1.5:1.0. These recommendations are based on the observations of San-
felippo and co-workers in nondialyzed chronically uremic and maintenance
hemodialysis patients [169,170] that such dietary modifications may reduce serum
triglyceride levels. (Note: The glucose absorbed from the dialysate will raise the
calories derived from carbohydrates above 35% of total energy intake (diet plus
dialysate). However, further reductions in dietary carbohydrate below 35% of
ingested calories would make dietary compliance too difficult for most patients.)
At the present time, carnitine is not routinely prescribed, but if hypertrigly-
ceridemia or unexplained weakness particularly of the proximal muscles is pres-
ent, serum and possibly muscle carnitine should be measured. If carnitine levels
are low, an oral dose of L-carnitine, 0.5-1.0 g/day may be given.
Dietary fiber may have several health enhancing effects such as improved
glucose tolerance, reduced serum lipids, more normal bowel function, and possi-
bly reduced risk of certain malignancies [172]. Since foods containing fiber are
often high in minerals, particularly magnesium and phosphorus, we recommend a
dietary total fiber intake of 20--25 g/day. This is lower than the high fiber intakes
often recommended for health enhancing diets, but is greater than the total fiber
intake of a typical American diet. Unfortunately, the foregoing modifications in
dietary lipids and carbohydrates and even this quantity of total fiber often make
the diet less palatable and more arduous. Although patients are encouraged to
adhere to these modifications, we do not urge compliance with the same intensity
that is used for the protein, energy, mineral and vitamin intake.
Although the daily recommended dietary phosphorus intake is 700 to 1200 mg,
phosphorus intakes closer to 700 mg/day may decrease the risk of severe hyper-
parathyroidism and the need for large amounts of aluminum binders of phos-
380

phate. Ingestion of aluminum binders of phosphate is considered to increase the

serum levels and body burden of aluminum [173]. Increased serum aluminum
concentrations and aluminum related osteomalacia have been reported in CAPD
patients, and it is therefore considered important to restrict the aluminum intake.
On the other hand, a diet which provides only about 800 mg/day of phosphorus
may be unpalatable and difficult to ingest, particularly with a daily protein intake
of 1.2-1.3 g/kg. We therefore recommend a dietary phosphorus intake as close to
800 mg/day as the patient can comfortably tolerate but do not prescribe more than
1200 mg/day. Aluminum binders of phosphate are prescribed as necessary to
maintain normal serum phosphorus concentrations.
The recommended magnesium intake is based upon the use of dialysate with
magnesium concentrations of 0.5-0.75 mEq/L. A dietary calcium intake of 1000-
1400 mg/day is most easily attained with calcium supplements.
Since sodium and water can be removed easily with CAPD, a liberal salt and
water intake is usually allowed. Indeed, by maintaining a large dietary sodium
and water intake, the quantity of fluid removed from the patient and, hence, the
daily dialysate outflow volume can be increased. This may be advantageous since
with CAPD, the daily clearance of small and middle sized molecules is directly
related to the volume of dialysate outflow. Thus, for some CAPD patients, a
higher sodium and water intake (e.g., 6-8 g/day of sodium and 3 L/day of water)
may enable the patient to use more hypertonic glucose exchanges to increase his
dialysate outflow volume and thereby increase dialysate clearances and glucose
and energy uptake from dialysate. This treatment may be undesirable for obese
or hypertriglyceridemic patients because of the greater need for hypertonic
glucose exchanges.
Dietary requirements for trace elements are not established. At the present
time, unless there is both a low serum or tissue concentration of a trace element
and clinical manifestations of deficiency of this element, trace elements supple-
ments exept for zinc are not given.
CAPD patients are routinely given iron, usually ferrous sulfate 300 mg one to
three times a day approximately one-half hour after meals, unless the patient has
evidence for iron overload. If oral iron therapy causes gastrointestinal symptoms,
changing to another oral medication may alleviate the problem. Otherwise, iron
dextran may be given parenterally on an intermittent basis.
Supplemental vitamin A and E are not recommended because serum levels are
increased. Supplemental vitamin K is not considered necessary unless the patient
is not eating and is receiving antibiotics which may suppress intestinal bacteria
which synthesize vitamin K. Vitamin D analogues are given as with hemodialysis
patients, although there may be a greater need for supplementation [105-106,
165]. Based on observations of nondialyzed chronically uremic patients and
patients undergoing maintenance hemodialysis as well as several studies now
conducted in patients undergoing CAPD, it is recommended that CAPD patients
should receive, as a supplement, the recommended dietary allowance [174] for
 381

each of the water soluble vitamins with the following modifications: pyridoxine
hydrochloride, 10 mg/day [175], thiamin hydrochloride, 2 mg/day, ascorbic acid,
100 mg/day, and folic acid 1.0 mg/day.
Some investigators have questioned the need for folic acid supplements in
maintenance hemodialysis patients. However, the observation that folic acid
intake from foods is often reduced in dialysis patients, that dialysis patients not
uncommonly ingest medicines that are folic acid antagonists, and that without
supplementation, folic acid levels are often low in CAPD patients [163] provides
justification for administering a folic acid supplement. Blumberg, Hanck and
Sander recommend 30--40 mg/day of thiamin (vitamin B1) based on the observa-
tion that with a 16 mg/day supplement, erythrocyte transketolase activity was
reduced [160]. However, the erythrocyte transketolase activity index was normal
in all nine patients receiving this supplement. In nondialyzed chronically uremic
patients and in hemodialysis patients we observed no evidence for thiamin
deficiency; many of these subjects received a daily vitamin supplement that
provided one to several mg/day of thiamin hydrochloride [175]. Since dialysis of
thiamin should not be substantially greater with CAPD than with hemodialysis, it
is not clear why the dietary requirement should be greater with this latter
treatment. Therefore, until further studies resolve this question, we recommend
that patients undergoing CAPD receive 2.0 mg/day of thiamin hydrochloride,
which is slightly greater than the recommended daily dietary allowance for
normal non-pregnant, non-lactating adults [174].

4.1. Exercise

Endurance exercise training can improve some of the metabolic abnormalities

which exist in hemodialysis patients [176, 179]. After training, an increase in work
capacity has been noted and there is a reduction in the amount of medication
required for blood pressure control. Training has been associated with an im-
provement in the rate of glucose disappearance and a reduction in fasting plasma
insulin levels. Plasma triglyceride levels decreased and the concentration of HDL
cholesterol rose. Booth demonstrated an increase in muscle size and strength,
improvement in endurance, and an increase in protein synthesis in response to
training [180]. Various psychological parameters also have been reported to
improve [181]. A reduction in anxiety and depression was noted; in addition,
these patients enjoyed pleasurable activities more frequently and reported feel-
ing more in control of their physical well-being.
Exercise should be considered an integral part of the care of the CAPD patient.
With regular exercise we have observed an improvement in muscle strength,
endurance, and outlook on life. Because of the increased glucose intake from
dialysate, CAPD patients may be particularly prone to hypertriglyceridemia and
obesity; regular exercise may have a salutory affect on these parameters. The
382

degree to which the CAPD patient is able to exercise is of course variable, and
exercise must be carefully designed to meet the specific needs of each patient
without exceeding his or her tolerance. For many patients 15-20 min/day of
exercise on an indoor stationary bicycle should become a regular part of their
daily routine. There is a need for a randomized study in a large number of patients
to assess the long term effects of regular exercise in the CAPD patient.

4.2. Nutritional support during intercurrent illness

Particular attention to nutritional intake should be given to dialysis patients

during times of stress to minimize tissue breakdown. Infections, such as per-
itonitis, or other superimposed illnesses enhance tissue breakdown and increase
protein and energy requirements [23, 24]. Moreover, an increased loss of various
nutrients occur during peritonitis. Intercurrent illnesses are often associated with
marked anorexia, nausea, and decreased food intake. In patients with functional
gastrointestinal tracts encouragement from the medical staff, can often result in
an improved food intake. The aid of an imaginative dietitian is invaluable; meal
plans and recipes can be arranged to satisfy the patient's tastes. Special high
protein, high energy food supplements are generally necessary to meet nutri-
tional needs. Nutrients may have to be administered via an intestinal feeding tube
[182-184].
The administration of protein and energy intravenously may be needed to
supplement an inadequate enteral intake. This may be especially true for patients
being treated for peritonitis. A suggested regimen for alimentation by use of
peripheral vein is shown in Table 10. A solution containing 500cc of 8.5% or 10%
amino acids (both essential and non-essential) and 500 cc of 10% dextrose is used.
Two to three 1 are infused daily, to provide 1.5 g of amino acids/kg body wt/day
(each 1 contains 42.5 to 50 g of amino acids). 500 ml of a 20% lipid emulsion
should be infused continuously at 20-30 ml/h through a Y-infusion set into the
same vein as the amino acids and glucose. The amount of dialysis is intensified;
twelve to sixteen liters of dialysate are exchanged/24 h (e.g. 11 Q 90 min via a
cycler or 21 Q 3 H via a multiprongeal connecting tube). High concentrations of

Table 10. Nutrition via peripheral vein in patients undergoing CAPD

1. 2-3 Llday of DIO 4.25% AA/day with MVI-12, trace elements, insulin, NaP, NaCL, KCL and
heparin
2. 500-750 cc/day of 20% lipid emulsion (administered continuously into same vein as 1.)
3. Provide adequate dialysis - 12-16 L of urea clearance/24 hr, high dextrose conc. + increased
dextrose absorption provide addn. CHO calories
4. Testosterone and insulin for anabolic effect
5. Monitor serum electrolytes, phosphate, glucose & albumin
 383

dialysate dextrose (2.5 - 4.25 mg/dl) are generally needed to maintain fluid
balance. Glucose absorbed across the peritoneum provides a significant amount
of energy (vida supra). Glucose absorption across the peritoneal membrane has
been shown to be enhanced during peritonitis [185]. While this may reduce
ultrafiltration, the need for an increased peritoneal dialysate glucose concentra-
tions may have a beneficial effect in providing additional calories. Water soluble
vitamins, and trace elements are added to the TPN solution. Sufficient insulin is
added to the TPN to maintain plasma glucose levels between 130--160 mgldl; since
insulin is the most anabolic hormone its addition to the TPN solution will reduce
catabolism [186]. We also administer testosterone enanthate (200 mg 1M bi-
weekly) for its anabolic effect, although its beneficial effect is still somewhat
controversial [181]. When patients are undergoing peritoneal dialysis, the ad-
dition of electrolytes to the parenteral solution is often not required. The mal-
nourished patient however, may require parenteral sodium phosphate to prevent
hypophosphatemia. During intravenous nutritional therapy, it is important to
monitor serum glucose, sodium, potassium, bicarbonate and phosphate levels. It
is generally not advisable to use the new low magnesium, high lactate PD
solutions; use of these solutions will in all likelihood result in hypomagnesemia
and alkalosis in the sick patients. If alkalosis does develop, ammonium chloride
can be added to the intravenous fluids. Alternatively, 5% dextrose in 0.9% saline
can be used to substitute for some of the regular dialysate solution. It may also be
necessary to add potassium chloride or sodium chloride to the dialysate. Studies
remain to be done to assess the efficacy of nutritional support by the use of a
peripheral vein in the peritoneal dialysis patient, particularly during hyper-
catabolic events.
In a very catabolic patient sufficient energy cannot be administered through a
peripheral vein; also if prolonged nutritional therapy is required, total parenteral
nutrition should be used [188]. The composition of a typical solution is shown in
Table 11. Two liters of solution containing 42.5-50 g of both essential and non-
essential amino acids and 250 g glucose/l may be administered daily through a

Table 11. Typical composition of solutions for TPN via central vein in patient undergoing CAPD

500-750 m120% lipid emulsion (20-30 mllhr)

1500-2000 ml D254 25% amino acid (60-80 mlIhr)
Additives
Sodium chloride 50-90 mEq/L
Potassium chloride 10-40 mEq/L
Sodium phosphate 5-10 mmlL
trace elements - Zn, Cu, Cr, Mn
insulin 20-80 ulL
MVI-12 1 vial
heparin 1oo0u/L

a An additional 500-1200 kcall24 hrs from glucose absorbed from dialysate.

384

subclavian vein. 500-150 ml of 20% lipid emulsion should be given via peripheral
vein. When the calories from the glucose absorbed from peritonal dialysate is
taken into account such a regimen will deliver up to 3500 kcal and 85-100 g amino
acids; quantities of nutrients which should be sufficient for even the most cata-
bolic patient. Total energy should not exceed 50 kcal per kg. Frequent measure-
ments of plasma glucose are essential. Serum electrolytes must also be monitored
closely. Insulin is generally added to the intravenous solution; as much as 20 to 30
units per hour may be needed to control hyperglycemia.

4.3. Special nutritional problems of the diabetic

Patients with diabetes mellitus who are undergoing maintenance hemodialysis

and intermittent peritoneal dialysis generally have evidence for wasting and
malnutrition [189-192]. Long standing diabetics have many metabolic and func-
tional problems which may contribute to the morbidity and poor rehabilitation
often seen in this population. Inadequate food intake, vomiting and the catabolic
stress associated with intercurrent illness contribute to the nutritional problems.
Diabetic visceroneuropathy, malabsorption and hypoinsulinemia make matters
worse.
Impaired nutrient intake is an important factor promoting malnutrition and is
often overlooked. Diabetic patients may eat less food because of the frequent
occurrence of anorexia, nausea and vomiting as a consequence of their vis-
ceroneuropathy. Blindness, peripheral neuropathy, amputations and other dis-
abling sequelae of diabetes may also impair the patient's ability to obtain, prepare
and ingest food. Given the same degree of illness, the diabetic dialysis patient
appears more likely to develop anorexia or vomiting than the non-diabetic
patient. Thus, a wide variety of intercurrent illnesses, that normally might not
induce vomiting in the non-diabetic may precipitate protracted vomiting in the
diabetic. Vomiting may also be the primary manifestation of insufficient dialysis
therapy. An example is the diabetic patient who is receiving a fixed schedule of
dialysis therapy and whose residual glomerular filtration rate (GFR) decreases.
The SUN may not indicate the need for more dialysis because with anorexia such
patients reduce their protein intake and, as a result, urea generation; the serum
creatinine may be relatively low because of muscle wasting.
A number of diabetic patients who appear to be well-dialyzed manifest another
characteristic vomiting syndrome. The patients do not experience nausea; their
first awareness of the impending emesis is the feeling that food or fluid is being
regurgitated. The vomiting is often forceful. As soon as the vomiting has passed,
the patient feels normal again. The vomiting may occur at any time but is
particularly common at the beginning of a meal. Characteristically, after vomit-
ing, the patients rinse out their mouths and finish eating their entire meal. Many
of these individuals appear otherwise to be relatively healthy. Adequate nutrition
 385

is often difficult to maintain in patients with severe diabetic gastroenteropathy,

when gastric retention, nausea, vomiting and diarrhea are present. Such patients
often require frequent small feedings and oral or intravenous nutritional supple-
mentation (vida supra). Longstreth et at. have found that metaclopramide [193], 5
or 10 mg one hour prior to meals and HS improved gastric emptying and reduces
the frequency of vomiting; it can be administered intravenously to the hospi-
talized patient.
Intercurrent illnesses appear to be a major cause of wasting and malnutrition in
the diabetic dialysis patient, as it is in non-diabetic patients. The higher incidence
of intercurrent illnesses and complications in the diabetic patient makes such
individuals particularly likely to develop malnutrition and wasting. The catabolic
effects of superimposed illnesses appear to be caused by impaired food intake due
to withholding of food for diagnostic tests as well as anorexia and vomiting (vida
supra), increased protein breakdown, as indicated by a rise in urea nitrogen
appearance and, in patients who develop peritonitis, increased protein and amino
acid losses. During these illnesses the serum glucose commonly increases and
fluctuates widely. These episodes, unfortunately, are not uncommon and in our
experience the intercurrent illness and the catabolic effect of insulinopenia fre-
quently leads rapidly to malnutrition and severe weight losses. Some physicians
treat hyperglycemia in the uremic diabetic patient by restricting carbohydrate
intake. Considering the need for good nutrition, it would seem more appropriate
to administer sufficient nutrients, intravenously if necessary, and treat hyper-
glycemia with insulin. In our experience, in hospitalized sick diabetics undergoing
peritoneal dialysis plasma glucose can be most readily controlled by the continu-
ous intravenous infusion of insulin.
A suggested method of nutritional support for the diabetic patient is shown in
Table 12. Since hypertriglyceridemia is common, we generally prescribe a high-
protein diet and attempt to have 50% of the energy ingested as fat with a high
polyunsaturated to saturated ratio. Often diabetic patients do not tolerate
azotemia as well as non-diabetic persons; they may require increased dialysis.
Another problem is an increased thirst, which may lead to excessive fluid intake;

Table 12. Nutritional management of the diabetic on CAPD

1. Protein 1.2-1.5g1kg body weight/day

2. Energy 35 kcallkg relative body weight
3. Frequent small feedings; 6-8 'meals'/day metoclorpramide (Reglan) 5 mg 60 min before meals use
of protein supplements e.g. Propac - or AA in CAPD solution
4. Dialysis: at least 10 L of total urea clearance124 hrs
(don't be misled by low SUN & creatinine)
5. Insulin: added to CAPD bags to maintain BS 120-150 range
6. Catabolic stress (e.g. peritonitis) requires intensive nutritional support - insulin by continuous drip
(see tables 10 and 11)
386

this may be associated with hyperglycemia, or perhaps a raised intracellular

sodium content [194].

5. Nutritional considerations in children

Dietary management is an integral part of the therapeutic regimen for the child
undergoing CAPD. Growth retardation is a significant clinical problem in chil-
dren with chronic renal insufficiency. A number of the factors are implicated
etiologically in the growth retardation of chronic renal failure can be partially or
totally corrected by optimal nutritional management. These include protein and
calorie malnutrition, renal osteodystrophy and acidosis [195, 196]. The use of
CAPD as a primary dialytic treatment modality has led to initial enthusiastic
reports [197, 198]. This modality appears to effectively control the biochemical
consequences of uremia while improving the level of rehabilitation. However,
growth retardation is generally not alleviated with CAPD [189-202]. In addition
to growth retardation optimal dietary management is required to minimize renal
osteodystrophy, hypertension, hyperkalemia and vitamin deficiencies in children
undergoing CAPD.

5.1. Nutritional assessment

Anthropometric parameters have been described for nutritional assessment of

non-uremic children. These can be utilized for assessment of children with
chronic renal failure. There are standard normal values for the American popula-
tion [203]. In the uremic child, measurements should always be performed by the
same person at three monthly intervals. A listing of the methods for monitoring
growth, maturation and nutritional status in children, and their recommended
frequency is shown in Table 13 [190].

5.2. Dietary recommendations

The goal of dietary therapy is to minimize the metabolic consequences of uremia

while maintaining optimal nutrition. Children who are not monitored closely
tend to deviate from their prescribed diets. As with adults, a team approach to
dietary management is critical to ensuring compliance with the dietary rec-
ommendations. All members of the patient care team, including the physician,
dietitian, nursing staff, social worker and psychiatrist are often required to assist
the child and his parents in coping with the prescribed diet. The dietitian should
design a diet which considers the tastes of the patient. The patient's dietary intake
should be evaluated at frequent intervals by the same dietitian, with dietetic
interviews, 3 day dietetic diaries and 3 day weighing technique.
 387

Table 13. Monitoring growth, maturation, and nutritional status

Anthropometry
1. Height (cm) Every 3 months
2. Weight (kg) Every month
3. Skinfold thickness
Subscapular
Triceps Every 3 months
4. Midarm muscle circumference a
additional measures for age 3 years or youngera
5. Recumbent height
6. Head circumference
7. Crown-rump
8. Rump-feet

Biochemical measurements
1. Creatinine appearance/height ratio Every 3 months
2. Serum total protein, albumin Every 3 months
3. Serum transferrin, IgG, IgA, IgM, C3 and C4 Every 6 months

Growth and maturation

1. X-rays - left hand wrist b Every 6 months
2. Sexual maturity' Every 6 months
3. Testicular SAR Every 6 months

a All these parameters should be compared to normal control values of the same chronological age and
height/age.
b Bone age could be determined by the method of Greulich and Pyle or Tanner.

, Sexual maturity and testicular size should be evaluated according to standard technique.

Energy. Energy requirements for uremic children have not been adequately
defined. Simmons et al. [205] found a positive correlation between growth and
calorie intake in children undergoing hemodialysis; unfortunately, the duration
of the study was short. Other investigators have failed to confirm this relationship
[206]. Chantler et at. [159] described an increased basal oxygen consumption in
pediatric patients undergoing hemodialysis. The data suggested that energy
needs in these children are greater than that of normal children. Unfortunately,
the methods utilized for measurement of oxygen consumption were imprecise. In
children undergoing CAPD, the glucose concentration in the dialysate provides
an extra source of calories. In our experience dialysate glucose absorption in
children accounts for approximately 12 percent of the total daily calorie intake
[202]. From current information, the optimal calorie intake for prepubertal
children undergoing dialysis should be at least that determined by the recom-
mended daily allowances (RD A) of the National Academy of Sciences for normal
children of the same height/age and sex. For pubertal and post-pubertal patients,
the prescribed energy intake is similar to the RDA for adolescents: 60 kcal/kg/day
for males and 48 kcal/kg/day for females [193]. Dietary carbohydrate should be
composed primarily of complex carbohydrates and in the CAPD patient provide
about 35 percent of dietary energy intake.
388

In children undergoing CAPD, high serum levels of triglycerides and cho-

lesterol have been observed when compared to age-matched controls [199, 208].
The hypertriglyceridemia is probably due to the impaired clearance of trig-
lycerides as previously described [156]. In addition, the glucose uptake from the
dialysate may increase triglyceride production. The cause of the elevated serum
cholesterol level is unclear. Increased serum cholesterol levels could be related to
the reduced serum albumin levels consequent to albumin losses into dialysate. A
compensatory increase in albumin synthesis by the liver would promote lipopro-
tein production and lead to the rise in the serum cholesterol level [209].
Dietary fat should provide about 50 percent of dietary energy intake. The
polyunsaturated to saturated fatty acid ratio should be about 1.5:1.0. These
recommendations are based on the studies of Sanfilippo et al. which show that
such modifications may reduce serum triglyceride levels in uremic and hemo-
dialyzed patients [169,170].

Protein. Pediatric patients undergoing hemodialysis and CAPD demonstrate

malnutrition and wasting [208, 210, 211]. Low serum albumin, transferrin and
complement levels, abnormal plasma amino acid pattern and low levels of alkali
soluble protein are the biochemical parameters that characterized the malnutri-
tion in these children [208, 210-212].
The optimal protein requirements for children undergoing CAPD are not well
defined. Giordano et al. have shown positive nitrogen balance in children under-
going CAPD. Their protein intake ranged between 1.2 to 2.0 g/kglday in 5
children studied during 3 weeks. We evaluated the mean protein losses into the
dialysate in 12 pediatric patients less than 10 years of age and in 12 patients older
than 10 years. The daily protein losses in the two groups was 3.2 ± 1.2 and
4.2 ± 1.2 glday, and the albumin loss was 2.1 ± 0.6 and 2.6 ± 0.7 glday, respec-
tively. When these losses were expressed as g/kg/day, both the total protein and
albumin losses were significantly higher in the younger patients. Each of these
losses correlated inversely with body weight and with body surface area. These
results are probably related to the proportionally greater peritoneal surface area
in younger patients [208]. Amino acid losses into the dialysate of children [213]
are small.
Preliminary data indicate that the minimum protein intake in children undergo-
ing CAPD can be estimated as follows: less than 5 years old, 3.0 glkglday; 5 years
old until puberty 2.5 glkglday; puberty 2.0 g/kg/day; and post puberty, 1.5 glkg/
day (Table 14). At least 50 percent of the prescribed protein intake should be
protein of high biological value (eggs, milk products, meat and fish) because such
protein contains a higher percentage of essential amino acids. For any patient, the
dietary protein intake may exceed the limitations if the serum urea nitrogen
remains less than 100 mgldl. The prescribed protein intake is monitored by
measurement of the urea nitrogen appearance (net urea production) described
above [23].
 389

Table 14. Prescribed intake" for pediatric patients undergoing CAPD

Energy Prepubertal: RDA for normal children of similar height and sex
Pubertal and postpubertal: 60 KcaVkg/day (in males; 48 Kcal/kg/day in females)

Protein Less than 5 years old: 3.0 g1kg/day

5 years old to puberty: 2.5 g1kg/day
Puberty: 2.0 g/kg/day
Postpuberty: 1.5 glkg/day

" The prescribed dietary intake is evaluated by dietary history using a 3-day weighed technique and
dietetic interviews every 2 months.

5.3. Water and electrolytes

Children undergoing CAPD, with or without residual renal function, are allowed
a relatively unrestricted sodium and water intake, since the fluid can be easily
removed with hypertonic exchanges. The appropriate use of the different dialy-
sate glucose concentrations can maintain excellent control of blood pressure and
body weight. In our experience only three of sixty-eight children undergoing
CAPD required a sodium restricted diet. Dietary potassium also need not be
unrestricted, except possibly in infants.

5.4. Calcium, phosphorous and vitamin D

Children undergoing long-term hemodialysis, without calcium supplementation

or vitamin D therapy, develop bone disease [196]. Baum et al. recently compared
the results of hemodialysis and CAPD in children [201]. Both groups of patients
had elevated serum alkaline phosphatase and immunoreactive parathyroid hor-
mone levels; the levels tended to be higher in the CAPD patients. It is not clear
from their report if the patients received vitamin D therapy or calcium sup-
pIe mentation.
Our experience with 14 children undergoing CAPD for at least 6 months
treated with low doses of vitamin D and phosphate binders showed that in nine
patients bone lesions developed or pre-existing bone disease worsened. Mass
transfer studies using a dialysate solution with 3.5 mEq Calcium/L showed
minimal calcium losses into the dialysate in these children. Patients with higher
serum calcium levels had lower serum iPTH and alkaline phosphatase levels. The
x-ray findings of secondary hyperparathyroidism were significantly less in the
group of patients with higher serum calcium levels. Prevention of bone disease in
children undergoing dialysis requires aggressive treatment.
Phosphate binders should be given after each meal. Calcium carbonate is the
preferred phosphate binder in children. The dose varies in each patient, but the
optimal amount should maintain the serum phosphorus level within normal limits
390

(4-6 mg/dl). Vitamin D supplements with dihydrotachysterol (0.125 to 0.75

mg/day), or 25-hydroxycholecalciferol (25 to 200 mg/day) or 1,25-dihydrox-
ycholecalciferol (0.25 to 1.5 ug/day) are indicated. The dosage of the vitamin D
analogue should be increased in the presence of persistent increased iPTH
secretion with high alkaline phosphatase levels despite normal serum calcium
levels. In this situation, the higher serum calcium levels are probably necessary to
suppress parathyroid gland activity [214].

5.5. Vitamin requirements

Daily requirements for most vitamins are not well defined for children with
chronic renal failure. Uremic children have a tendency to develop deficiency of
the water-soluble vitamins unless supplements are given. Dialysis patients are
susceptible to folate deficiency because of poor dietary intake, destruction of the
vitamin during food preparation and losses during dialysis. Clinical evidence of
folic acid deficiency has been shown in some studies [215] but not in others [216].
A recent study in children undergoing dialysis showed elevated levels of vitamin
A [217].
Current data indicate that children undergoing dialysis should receive the
following supplements: folic acid, 1 mg/day; pyridoxine-B6, 5-10 mg/day; and
ascorbic acid, 75-100 mg/day. For the other vitamins such as thiamin, riboflavin,
niacin, panthothenic acid and biotin, the RDA recommendations should be
prescribed. Since serum retinol binding protein and vitamin A are elevated,
supplements are not indicated; no supplemental vitamin E or K is recommended.

5.6. Trace elements: iron, zinc, copper and aluminum

The current knowledge of trace element toxicity or deficiency in uremic children

is not well established. Iron deficiency may occur in dialysis patients, in associ-
ation with impaired intestinal absorption of iron, dialysis losses or excessive
blood drawing. In addition, uremic patients may have low serum transferrin
levels, secondary to malnutrition [208, 210] which may obscure iron deficiency.
Zinc deficiency has been associated with dysgeusia and impotency in uremic
patients. Improvement of these symptoms may be obtained with zinc supplemen-
tation [218]. Other studies demonstrated normal zinc levels in the red blood cells
in hemodialysis patients [219]. In children undergoing hemodialysis or intermit-
tent peritoneal dialysis, plasma levels of zinc and copper and zinc concentration in
hair are normal [217]. Both zinc and copper are vital elements for normal
metabolism [220]. Zinc deficiency has been associated with growth retardation,
anorexia and weakness. These manifestations should be considered in children
with markedly depressed appetite and poor growth rate. Copper deficiency may
 391

cause anemia, megaloblastic or sideroblastic and leucopenia. Copper depletion

has been demonstrated in serum, skin, kidney, adipose tissue and muscle in non-
dialyzed uremic children [221]. Further studies are required to establish whether
other tissues are similarly zinc and copper depleted.
From the present data it is difficult to assess if uremic children on dialysis
should receive zinc or copper supplements. The average zinc content of mixed
diets consumed by adults is between 10 and 15 mg/day [207]. For preadolescent
children an allowance of 10 mg/day is recommended. No exact data are available
to determine the zinc requirement of infants. The estimated copper requirement
for infants and children of 0.08 mg/kg/day is recommended [207].
The use of aluminum containing water for preparation of dialysate may be
associated with a high incidence of bone disease [222]. Significantly lower plasma
concentrations of aluminum have been shown in adult patients on CAPD and
IPD than those patients with chronic renal failure or on hemodialysis [103]. These
data suggest that oral aluminum hydroxide is an important source of plasma
aluminum in these patients [103]. Baluarte et ai. have recommended that the dose
of aluminum hydroxide be less than 100 mg/kg/day (33 mg/kg of elemental
aluminum) [223]. Children treated with aluminum containing phosphate binders
should be monitored carefully for signs of toxicity. In cases of persistent hyper-
phosphatemia and high serum aluminum levels, calcium carbonate may be used
as a phosphate binder. Several reports have shown the development of aluminum
related osteomalacia in children and adults prior to the initiation of dialysis.
Younger uremic children developed higher serum aluminum levels secondary to
either larger doses of elemental aluminum per kg/day or greater intestinal absorp-
tion of aluminum [222, 223]. At the present time these two factors cannot be
separated. Until new phosphate bindering agents are developed, aluminum
containing phosphate binders should be used very carefully in young children.
Calcium carbonate has been shown as an adequate phosphate binder and should
be utilized as a primary agent of choice in children under five years of age.
Attention should be paid to the development of hypercalcemia when vitamin D
steroids are administered concomitantly with calcium carbonate.

6. Summary

Malnutrition and wasting are common in patients undergoing maintenance di-

alysis. These problems may be more prevalent in patients undergoing peritoneal
dialysis due to inadequate dialysis, poor protein intake, loss of nutrients in
dialysate and tissue breakdown associated with intercurrent illnesses, particularly
peritonitis. Periodic assessment of the nutritional status of these patients should
be used as a guide to appropriate nutritional therapy. Protein and amino acid loss
during CAPO average only 8.8 and 3.3 g/day, respectively; with peritonitis losses
increase. Balance studies indicate that in a well nourished patient, 1.2-1.3 g
392

protein/kg body wt is probably adequate; a malnourished patient should receive

1.4-1.6 g/kg. Sufficient dialysis must be prescribed to enable ingestion of this diet;
residual renal function makes a significant contribution to total clearance. With
CAPD energy intake is supplemented by large amounts of glucose absorption;
this is beneficial except for the obese or hyperlipoproteinemic patient. Nutri-
tional support of the patient with intercurrent illness is crucial; nutrition admin-
isted via peripheral vein is often adequate. Management of the nutritioflal needs
of the diabetic and children presents additional problems often trying the skills of
the clinician.

Acknowledgements

This manuscript was prepared with unbelievable patience and forebearance by

Barbara Boswell. Special thanks to our wives Bonnie Mae, Danielle, and Liddy
for their kind support.

References

1. Coles GA: Body composition in chronic renal failure. 0 J Med 41: 25, 1972.
2. Blumenkrantz MJ, Kopple JD: VA Cooperative Study Participants: Incidence of nutritional
abnormalities in uremic patients entering dialysis therapy. Kidney Int 10: 514, 1976.
3. Kopple JD: Dietary requirements. In: Clinical Aspects of Uremia and Dialysis. Ed. Massry SG,
Sellars A, Springfield, Ill: Charles C. Thomas. 1976, p 453.
4. Banzai VK, Popli S, Pickering J et al.: Protein calorie malnutrition and cutaneous anergy in
hemodialysis maintained patients. Am J Clin Nutr 33: 1608, 1980.
5. Blumenkrantz MJ, Kopple JD, Gutman RA et al.: Methods for assessing the nutritional status of
patients with renal failure. Am J Clin Nutr 33: 1567, 1980.
6. Blackburn G, Bistrian B, Maine B et al.: Nutritional and metabolic assessment of the hospi-
talized patient. J Parenteral Enteral Nutr 1: 11,1977.
7. Giordano C, DeSanto NG: Dietary management of patients on peritoneal dialysis. In: Today's
Art of Peritoneal Dialysis Contributions to Nephrology, Eds. Boen ST, Trevino-Becerra A
(Karger) 17: 11-92,1979.
8. Gutman R, Blumenkrantz M, Chan Y et al.: Controlled clinical trial of patients randomly
assigned to hemodialysis and peritoneal dialysis: Results of the Veterans Administration Coop-
erative study. Kidney Int (in press).
9. Heide B, Pierratos A, Khanna R, et al.: Nutritional status of patients undergoing continuous
ambulatory peritoneal dialysis (CAPD). Perit Dial Bull 3: 138-141, 1983.
10. Feldman H, Singer I: Endocrinology and metabolism in uremia and dialysis: a clinical review.
Medicine 54: 345,1974.
11. Knochel JP, Seldin DW: Metabolic and biochemical disturbances in uremia. In: The Kidney,
Eds. Brenner BM, Rector FC Jr, Phil PA, Saunder WB Co, p 1462, 1976.
12. De Fronzo RA, Tobin JD et al.: Glucose intolerance in uremia: quantification of beta cell
sensitivity to glucose and tissue sensitivity to insulin. J Clin Invest 62: 425, 1978.
13. Soman V, Felig P: Glucagon and insulin binding to liver membranes in a partially nephrec-
tomized uremic rat model. J Clin Invest 60: 224-232,1977.
 393

14. Furst P, Ahlberg M et aL: Principles of amino acid therapy in uremia. Am I Clin Nutr 31: 1744,
1978.
15. Kopple ID, Jones M et al.: Amino acid and protein metabolism in renal failure. Am I Clin Nutr
31: 1532, 1978.
16. Furst P, Alvestrand A, Bergstrom 1: Effects of nutrition and catabolic stress on intracellular
amino acid pools in uremia. Am 1 Clin Nutr 33: 1387, 1980.
11. Kopple ID, Cianciaruso B, Massry SG: Does parathyroid hormone cause protein wasting? In:
Contrib NephroL Eds. Berlyne GM et ai, Brooklyn NY, Karger S, Basel, 1980, p 138-148.
18. Slatapolsky E, Caglar S, et al.: On the pathogenesis of hyperparathyroidism in chronic renal
disease. I Clin Invest 50: 492, 1971.
19. Reaven GM, Swenson RS, Sanfelippo GML: An inquiry into the mechanism of hyper-
triglyceridemia in patients with chronic renal failure. Am I Clin Nutr 33: 1476,1980.
20. Buonocrestiani U, DiPaola N, Carobi C, et a1.: Camitine depletion with CAPD. In: Advances in
Peritoneal Dialysis. Ed. uy Gahl GM, Kessel M, Nolph KD. Excerpta Medica, Amsterdam, p
441-445,1981.
21. Amair P, Gegoriadis A, et al.: Serum carnitine in patients on continuous ambulatory peritoneal
dialysis (CAPD), Perit Dial Bull 2: 11-12, 1982.
22. Albright R, Kram BW et aL: Carnitine status of CAPD patients. Lancet 2(8291): 218, 1982.
23. Grodstein GP, Blumenkrantz 1M, Kopple ]D: Effects of intercurrent illnesses on nitrogen
metabolism in uremic patients. Trans Am Soc Artif. Intern Organs 25: 438, 1979.
24. Grodstein G, Blumenkrantz MI, Kopple ID: Nutritional and metabolic response to catabolic
stress in uremia. Am J Clin Nutr 33: 1411, 1980.
25. Thompson N, Stevens B, Humphery T, Athens R: Comparison of trace elements in peritoneal
dialysis, hemodialysis and uremia. Kidney Int 25: 9-14, 1983.
26. Roe DA: Drug induced nutritional deficiencies. A VI Pub Westport Conn., 1976.
27. Theuer RC, Vitale JJ: Drug and nutrient interactions. In: Nutritional Support of Medical
Practice. Eds. Schneider HA, Anderson CE, Coursin DB. New York, Harper and Row, p297,
1977.
28. Bergstrom I: Potential metabolic problems associated with continuous ambulatory peritoneal
dialysis. In: Continuous Ambulatory Peritoneal Dialysis, Ed M Legrain (Excerpta Medica), p
277,1979.
29. Blumenkrantz MI, Roberts M: Progress in peritoneal dialysis: a historical prospective. In:
Today's art of Peritoneal Dialysis. Contrib Nephrol. Eds: Boen ST, Trevino-Becerra A, Pub
Karger, p 101-110,1979.
30. Palmer RA, Newell IE et al.: Treatment of chronic renal failure by prolonged peritoneal
dialysis. N Eng J Med 274: 248, 1966.
31. Manis T, Friedman E: Dialytic therapy for irreversible uremia. N Eng J Med 301: 1321-1328,
1979.
32. Ahmad S, Gallagher NM, Shen F: Intermittent peritoneal dialysis: status reassessed. Am Soc
Artif. Intern Org 25: 86, 1979.
33. Ghantous WN, Salkin MS et al.: Limitations of peritoneal dialysis in the treatment of ESRD
patients. Am Soc Artif Intern Org. 25: 100, 1979.
34. Blumenkrantz MJ, Moran JK et al.: Peritoneal clearances during prolonged maintenance
peritoneal dialysis in diabetic and non· diabetic patients. Kidney Inter. 14: 671, 1978.
35. Blumenkrantz JM: CAPD: Nutritional concerns - Have we learned from the experiences in the
1960's and 1970's? Proc 2nd Int Symp on CAPD, Austin, Texas, May 1980.
36. Blumenkrantz MJ, Jones MR et aL: Albumin metabolism in patients with chronic renal failure
and those undergoing peritoneal or hemodialysis. Kidney Inter. 14: 670, 1978.
37. Kaysen G, Schoenfeld P: Albumin homeostasis during CAPD. Kidney Int 23: 153, 1983.
38. Bodnar D, Schreiber M, Vidt D: Protein status of patients during the first year on CAPD
(Abstract). Kidney Int (in press), 1984.
394

39. Williams P, Kay R, Harrison J et al.: Nutritional and anthropometric assessment of patients on
CAPD over one year: Contrasting changes in total body nitrogen and potassium. Perit Dial Bull
1: 82-87, 1981.
40. Rubin J, Flynn MA, Nolph KD: Total body potassium-a guide to nutritional health in patients
undergoing continuous ambulatory peritoneal dialysis. Am J Clin Nutr 34: 94-98, 1981.
41. Lindholm B, Bergstrom J: Personal Communication.
42. Sandstead HH, Pearson WN: Clinical evaluation of nutritional status. In: Modern Nutrition in
Health and Disease. Eds. Goodhart RS, Shils ME, Philadelphia, PA: Lea and Febiger, 1980.
43. Waterlow JC, Harper AE: Assessment of protein nutrition. In: Total Parenteral Nutrition. Ad
Ghadimi, New York: John Wiley and Sons, p 231,1975.
44. Davidson S, Passmore R et al.: Assessment of nutritional status. In: Human Nutritiona and
Dietetics. New York: Churchill Livingston, p 558, 1975.
45. Nold JM, Kellman B et al.: The dietary treatment of patients on continuous ambulatory
peritoneal dialysis: Practical aspects. In: Continuous Ambulatory Peritoneal Dialysis, Ed.
Legrain M (Excerpta Medica), p 207,1979.
46. Sorensen MK, Kopple JD: Assessment of adherence to protein, restricted diets during con-
servative management of uremia. Am J Clin Nutr 21: 631, 1968.
47. Sorensen MK, Kopple JD: Dietary instruction of the chronically uremic patient. In: Uremia
Conference on Pathogenesis. Diagnosis and Therapy. Eds. Kluthe R, Berlyn B, Buton N,
Stuttgart, Germany: Georg Thieme Verlag, p 150-163, 1972.
48. Marks HH: Body Weight facts from life insurance records. Hum Bioi 28: 107, 1956.
49. Behnke AR: New concepts of height relationships. In: Obesity. Ed. Wilson NL, Philadelphia,
PA. F.A. Davis Co., p 25,1967.
50. New Weight standards for men and women. Stat Bull Metropol Life Ins Co 40: 1,1959.
51. Durnin JV, Womersley J: Body fat assessed from total body density and its estimation from skin
fold thickness; measurements on 481 men and women aged from 17-72 years. Br J Nutr 32: 77,
1974.
52. Gray GE, Gray MS, Gray LK: Validity of anthropometric norms used in the assessment of
hospitalized patients. J Parent and Ent Nutr 33: 366, 1979.
53. Burgert SL, Anderson CF: An evaluation of upper arm measurements used in nutritional
assessment. Am J Clin Nutr 32: 2136, 1979.
54. Waterlow JC, Cravioto J, Stephen JML: Protein malnutrition in man. In: Advances in Protein
Chemistry. Eds. Anfinsen CB, Anson ML, Bailey K, Edsall JT. New York: Academic Press, p
131,1960.
55. Whitehead RG: Biochemical tests in differential diagnosis of protein and calorie deficiencies.
Arch Dis Child 42: 479, 1967.
56. McFarlane H, Adcock KJ et al.: Biochemical assessment of protein-calorie malnutrition. Lancet
2: 392, 1969.
57. Lane RS: Changes in plasma transferrin levels following administration of iron. Br J Heamatol
12: 249, 1966.
58. Morton A, Hamilton SM et al.: Studies on regulatory factors in transferrin metabolism in man
and the experimental rat. In: Plasma Protein Turnover. Eds Bianci R, Mariani G, McFarlane
AS. Baltimore, Maryland: University Park Press. p 165, 1976.
59. Jontofsohn R, Junders K et al.: Serumtransferrin, eiweibernahrung und eisenstoffwechsel bei
chronisch hemodialyseirten patienten. Klin Wochen 52: 398, 1974.
60. Grossman SB, Yap SH, Shafritz DA: Influence of chronic renal failure on protein synthesis and
albumin metabolism in rat liver. J Clin Invest 59: 869, 1977.
61. James WPT, Sinder PM, Waterlow JC: Nutritional aspects of plasma protein metabolism the
relevance of protein turnover ratios during malnutrition and its remission in man. In: Plasma
Protein Turnover. Eds. Bianchi R, Mariani G, McFarlane AS. Baltimore University Park Press,
p 251, 1976.
 395

62. Coles GA, Peters DK, Jones JH: Albumin metabolism in chronic renal failure. Clin Sci 39: 423,
1970.
63. Fish JC, Remers AR Jr et al.: Albumin kinetics and nutritional rehabilitation in the unattended
home dialysis patient. N Eng J Med 287: 478, 1972.
64. Blainey JD, Hilton DD: The composition of the body in renal failure. Ann Roy Coll Surg Eng
47: 45, 1970.
65. Compty CM: A longitudinal study of body composition in terminal uremics treated by regular
hemodialysis. Can Med Assoc J 98: 482, 1968.
66. McNeill KG, Harrison JE, Mernagh JK et al.: Changes in body potassium and lean body mass
during total parenteral nutrition. J Parenteral Nutrition 6: 106--108, 1982.
67. Oreopoulos D, Martiss E et al.: Nutritional status of CAPD and the potential use of amino acid
containing dialysis solutions. Perit Dial Bull 3: S10-12, 1983.
68. Berlyne GM, Jones JH et al.: Protein loss in peritoneal dialysis. Lancet 1: 738, 1964.
69. Gordon S, Rubini ME: Protein losses during peritoneal dialysis. Am J Med Sci 253: 283, 1967.
70. Baran J, Lebek R et al.: Kinetics of protein loss during peritoneal dialysis. Pol Med J 11: 277,
1972.
71. McKelvey AM, Yeoh HH et al.: Immunoglobin and dextran losses during peritoneal dialysis.
Arch Intern Med 134: 266, 1974.
72. Blumenkrantz MJ, Gahl GM et al.: Protein loss during peritoneal dialysis. Kid Int. 19: 593, 1981.
73. Miller FN, Nolph KD et al.: The influence of solution composition on protein loss during
peritoneal dialysis. Kid Int. 23: 35, 1983.
74. Katirzaglou A, Oreopoulos DG et al.: Reappraisal of protein losses in patients undergoing
CAPD, Nephron 26: 230, 1980.
75. Thompson NM, Walker RG et al.: Continuous ambulatory peritoneal dialysis (CAPD) in the
treatment of end stage renal failure. Proc Eur Dial Trans 16: 171, 1979.
76. Gahl GM, Becker H et al.: Kontinuierliche ambulante peritonealdialyse (CAPD) Schweiz, Med
Wschr 109: 1990, 1979.
77. Oreopoulos DG, Khanna Ret al.: Continuous ambulatory peritoneal dialysis in Canada, Dial &
Trans 9: 224, 1980.
78. Lindholm B, Ahlberg M et al.: Nutritional aspects of continuous ambulatory peritoneal dialysis.
In: Continuous Ambulatory Peritoneal Dialysis. Ed. Legrain M, (Excerpta Medica). p 199,
1979.
79. Kurtz SB, Wong VH: CAPD 3 year experience at Mayo Clinic: MC Proc. 58: 633, 1983.
80. Rubin J, Nolph KD et al.: Protein losses iff CAPD, Nephron 28: 218, 1981.
81. Berlyne GM, Lee HA et al.: Amino acid loss in peritoneal dialysis. Lancet 1: 1339, 1967.
82. Noree LO, Bergstrom J et al.: The effect of essential amino acid administration on nitrogen
metabolism during dialysis. Proc Eur Dial Trans Assoc 8: 182, 1971.
83. Young GA, Parsons PM: The effect of peritoneal dialysis upon the amino acids and other
nitrogenous compounds in the blood and dialysate of patients with renal failure. Clin Sci 37: 1,
1968.
84. Finkelstein AS, Kliger AS et al.: Chronic peritoneal dialysis in diabetic patients with end stage
renal failure. Proc Dial Trans Forum 5: 142, 1975.
85. Young GA, Parsons FM: Plasma amino acid imbalance in patients with chronic renal failure on
intermittent dialysis. Clin Chim Acta 27: 491, 1970.
86. Bergstrom J, Furst P et al.: The effects of peritoneal dialysis on intracellular free amino acids in
muscle from uremic patients. Proc Eur Dial Trans Assoc 9: 393,1972.
87. Bergstrom J, Furst P et al.: Intracellular free amino acids in muscle tissue of patients with chronic
uremia: Effect of peritoneal dialysis and infusion of essential amino acids. Clin Sci and Mol Med.
54: 51, 1978.
88. Furst P, Bergstrom J et al.: The effect of dialysis and administration of essential amino acids on
plasma and muscle protein synthesis, studied with 15 N in uremic patients. Proc Eur Dial Trans
Assoc 7: 175, 1970.
396

89. Wolfson M, Jones MR, Kopple JD: The effect of amino acid infusion during hemodialysis on
amino acid losses. Am Soc Nephr, Kid Int, 16, 1979.
90. Blumenkrantz MJ (unpublished observation).
91. Giordano C, DeSanto NG, Capodicasa G et al.: Amino acid losses during CAPD. Clin Nephrol
14: 230--232, 1980.
92. Randerson DH, Chapman GV, Farrell PC: Amino acid and dietary status in CAPD patients. In:
Continuous Ambulatory Peritoneal Dialysis, Ed. Atkins RC, Thomson NM, Farrell PD,
London, Churchill-Livingstone, p 179-191, 1981.
93. Dombros N, Oren A et al.: Plasma amino acid profiles and amino acid losses in patients
undergoing CAPD. Peritoneal Dial Bull 2: 27-32,1982.
94. Furst P, Alvesstrand A, Bergstrom J: Effects of nutrition and catabolic stress on intracellular
amino acid pools in uremia. Am J Clin Nutr 33: 1387-1395,1980.
95. Flugel-link RM, Jones MR, Kopple JD: Red cell and plasma amino acid concentrations in renal
failure. J Parent and Enteral Nutr 7: 450--456, 1983.
96. Kopple JD, Blumenkrantz MJ et al.: Plasma amino acid levels and amino acid losses during
continuous ambulatory peritoneal dialysis. Am J Clin Nutr 36: 395-402, 1982.
97. Oren A, Wu G, Anderson GH et al.: Effective use of amino acid dialysate over four weeks in
CAPD patients. Perit Dial Bull 3: 66-73, 1983.
98. Williams PF, Marliss E et al.: Amino acid absorption following intraperitoneal administration in
CAPD patients. Perit Dial Bull 2: 124-129, 1982.
99. Lindner A, Tenckhoff H: Nitrogen balance in patients on maintenance peritoneal dialysis. Trans
Am Soc Artif Intern Organs 16: 255, 1970.
100. Giordano C, De Santo NG et al.: Protein requirement of patients on CAPD: a study on nitrogen
balance. Int J Of Artif Organs 3: 11, 1980.
101. Gahl GM, Baeyer HU et al.: Outpatient evaluation of dietary intake and nitrogen removal in
continuous ambulatory peritoneal dialysis. Ann Int Med 94: 643-646, 1981.
102. Blumenkrantz MJ, Kopple JD, Moran JK et al.: Metabolic balance studies and dietary protein
requirements in patients undergoing continuous ambulatory peritoneal dialysis. Kidney Int 21:
849-861, 1982.
103. Lindholm B, Alvestrand A, Furst P et al.: Metabolic effects of continuous ambulatory peritoneal
dialysis. Proc EDTA 17: 283--289,1980.
104. Calloway DH, Odel ACF, Margen S: Sweat and miscellaneous nitrogen losses in human balance
studies. J Nutr 101: 775--786, 1971.
105. De1mez JA, Slatopolsky E, Martin KJ et al.: Minerals, vitamin D, and parathyroid hormone in
continuous ambulatory peritoneal dialysis. Kidney Int 21: 862-867, 1982.
106. Kurtz SB, McCarthy JT, Kumar R: Hypercalcemia in continuous ambulatory peritoneal dialysis
(CAPD) patients: Observations on parameters of calcium metabolism. In: Advances in Per-
itoneal Dialysis. Ed. by Gahl GM, Kessel M, Nolph KD. Excerpta Medica, Amsterdam p 467-
472,1981.
107. Nolph KD, Popovich RP, Moncrief JW: Theoretical and practical implications of continuous
ambulatory peritoneal dialysis. Nephron 21: 117-122, 1978.
108. Moncrief JW, Popovich RP, Nolph KD et al.: Clinical experience with continuous ambulatory
peritoneal dialysis. Am Soc Artif Intern Organs 2: 114-118, 1979.
109. Fenton SSA, Cattran DE et al.: Initial experiences with continuous ambulatory peritoneal
dialysis. Artif Organs 3: 206-209, 1979.
110. VonHartitzsch B, Medlock TR: Automated peritoneal dialysis- the alternative to hemodialysis.
Am Soc Artif Organs 2: 465-469, 1978.
111. Blumenkrantz MJ, Kopple JD et al.: Nitrogen and urea metabolism during continuous ambula-
tory peritoneal dialysis. Kidney Int 20: 78--82, 1981.
112. Steffee WP, Goldsmith RS et al.: Dietary protein intake and dynamic aspects of whole body
nitrogen metabolism in adult humans. Metabolism 25: 281-97, 1976.
 397

113. Borah MF, Schoenfeld PY et al.: Nitrogen balance during intermittent dialysis therapy of
uremia. Kidney Int 14: 491-500, 1978.
114. Cottini EP, Galling DL, Dominquez JM: Urea excretion in adult humans with varying degrees of
kidney malfunction fed milk, egg or amino acid mixture, Assessment of nitrogen balance.
Nutrition 103: 11, 1973.
115. Blumenkrantz MJ: The need for individualization and flexibility in treatment modalities. Con-
temp Dial 4: 12, 1983.
116. Bergstrom J, Asaba H et al.: Middle molecules in chronic uremic patients treated with peritoneal
dialysis in Endocrine Metabolic Implications of CAPD. Perit Dial Bull 3: 57, 1983.
117. Lindholm B, Tegner R, Tranaeus, Bergstrom J: Progress of peripheral neuropathy during
continuous ambulatory peritoneal dialysis. ASAIO Transactions 28: 263, 1982.
118. Randerson DH, Farrell PC: Metabolite generation and clearance and variation in long term
CAPD. In: Moncrief JW, Popovich RP, eds. CAPD Update, Continuous Ambulatory Per-
itoneal Dialysis. New York, Masson Publishing Inc. 75-81, 1981.
119. Asaba H, Alverstrand A, Furst P, Bergstrom J: Clinical implications of uremic molecules in
regular hemodialysis patients. Clin Nephrology, 1982.
120. Bergstrom J, Lindholm U, Noree LO: Preservation of peripheral nerve function in severe
uremia during treatment with low protein high calorie diet and surplus of essential amino acids.
Acta Neurol Scand 51: 99, 1975.
121. Nolph KD, Rosenfeld PS et' al.·: Peritoneal glucose transport and hyperglycemia during per-
itoneal dialysis. Am J Med Sci 259: 272, 1978.
122. Andersson G, Berquist-poppen M et al.: Glucose absorption from the dialysis fluid during
peritoneal dialysis. Scand J Urol Nephrol5: 77, 1971.
123. Boyer J, Gill G, Epstein F: Hyperglycemia and hyperosmolality complicating peritoneal di-
alysis. Ann Intern Med 67: 568, 1967.
124. Brown DJ, Adam WR, Dawborn JK: Glucose absorption and hyperglycemia during peritoneal
dialysis. Aust NZ J Med I: 1, 1973.
125. Lindholm B, Bergstrom J, Karlander S-V: Glucose metabolism in patients on CAPD. In:
Advances in Peritoneal Dialysis, Ed. by Gahl GM, Kessel M, Nolph KD. Excerpta Medica,
Amsterdam p 413-416, 1981.
126. DeSanto NG, Capodicasa F et al.: Glucose utilization from dialysate in patients on continuous
ambulatory peritoneal dialysis. Int J Artif Organs 2: 119, 1979.
127. Grodstein GP, Blumenkrantz MJ et al.: Glucose absorption during continuous ambulatory
peritoneal dialysis. Kid Int 19: 564, 1981.
128. Fuchs C: Round table discussion. In: Continuous Ambulatory Peritoneal Dialysis, Ed. Legrain
M (Excerpta Medica), p 220,1979.
129. Lowrie EC, Lazarus JM, Mocelin AJ: Survival of patients undergoing hemodialysis and renal
transplantation. N Eng J M 288: 863-867,1973.
130. Haire HM, Sherrard DJ, Scardapan D: Smoking, hypertention and mortality in a maintenance
dialysis population. Cardiovascular Med 3: 1163-1168,1978.
131. Bagdade JD, Porte D Jr, Bierman EL: Hypertriglyceridemia: a metabolic consequence of
chronic renal failure. N Eng J M 279: 181-185, 1968.
132. Brunzel JD, Albers 11, Hass JD et al.: Prevalence of serum lipid abnormalities in chronic
hemodialysis. Metabolism 26: 903-910, 1977.
133. Lindner A, Charra B, Sherrard DJ, Scribner BH: Accelerated atherosclerosis in prolonged
maintenance hemodialysis. N Eng J M 290: 697-701, 1974.
134. Norbeck HE, Oro L, Carlson LA: Serum lipid and lipoprotein concentrations in chronic uremia
200: 487, 1976.
135. Norbeck HE: Actamedica Scandinavica Supplement, p 649, 1981.
136. Lindholm B, Karlander S, Norbeck HE, Bergstrom J: Glucose & lipid metabolism in peritoneal
dialysis. Peritoneal Dialysis. Wichtage. Editora Milano Italy, pg 219-229,1982.
398

137. Cattran 0, Fenton S et al.: Defective triglyceride removal in lipemia associated with peritoneal
dialysis and hemodialysis. Annals Int Med 85: 29, 1976.
138. Oreopoulos DG, Karanicolas S et al.: Dialysis and triglycerides. Ann Intern Med 85: 679, 1976.
139. Roxe DM, del Greco F et al.: Final Report: a 3 year prospective comparison of peritoneal
dialysis to hemodialysis. Am Soc Artif Intern Organs 25: 81,1979.
140. Bailey GL: Hemodialysis principles and practice. Ed. Bailey G, New York, Acad Press, 1972.
141. Stout W: Diabetes and atherosclerosis - the role of insulin. Diabetilogica 16: 141, 1979.
142. Ahrens, et al.: Transaction of the Association of American Physicians 74: 134, 1981.
143. Turgan C, Techally J;Bennett S: Accelerated hypertriglyceridemia in patients on CAPD: a
preventable abnormality. Int J of Artif Organs 4: 158-160, 1981.
144. Lindholm B, Karlander SG, Norbeck HE et al.: Glucose and lipid metabolism in peritoneal
dialysis. In: Peritoneal Dialysis. Ed. LaGreca G, Biasioli S, Ronco C. Wichtig Editore, Milano,
p 219-230, 1982.
145. Kensch G, Bammatter F, Mordasini R et al.: Serum lipoprotein concentrations during CAPD.
In: Advances in Peritoneal Dialysis. Ed. by Gahl GM, Kessel M, Nolph KD. Excerpta Medica,
Amsterdam, p 427-429, 1981.
146. Khanna R, Breckenridge C, Roncari D et al.: Lipid abnormalities in patients undergoing CAPD
in endocrine and metabolic implications of CAPD. Perit Dial Bull 3: 513-515, 1983.
147. Monteon F, Trevino A, Gonzalez D et al.: Hypertriglyceridemia in patients on CAPD. In:
Advances in Peritoneal Dialysis. Ed. by Gahl GM, Kessel M, Nolph KD. Excerpta Medica,
Amsterdam, p 437-440, 1981.
148. Gokal R, Ramos JM, McGurk JG et al.: Hypertriglyceridemia in patients on CAPD. In:
Advances in Peritoneal Dialysis. Ed. by Gahl GM, Kessel M, Nolph KD. Excerpta Medica,
Amsterdam, p 430-433, 1981.
149. Cattan DC: The significance of lipid abnormalities in patients receiving dialysis therapy. Perit
Dial Bull 3: 529-532, 1983.
150. Chan MK, Varghese Z, Persaud JW: Hyperlipidemia in patients on maintenance hemo- and
peritoneal dialysis: the relative pathogenetic roles of triglyceride production and triglyceride
removal. Clin Nephrol17: 183-190, 1982.
151. Flynn CT, Nanson JA: Intraperitoneal insulin with CAPD - an artificial pancreas. Trans Am Soc
Artif Intern Organs 25: 114, 1970.
152. Beardsworth S et al.: Interperitoneal insulin cannot correct the hyperlipidemia of continuous
ambulatory peritoneal dialysis. Peritoneal Dialysis Bulletin 3: 126-127, 1983.
153. Felts JM, Staprans I, Zacherle B: Mechanisms of hyperlipidemia in chronic renal failure and in
the nephrotic syndrome. Proc Ann Contractors Conf. AKCUP NIAMDD, NIH 11: 71, 1978.
154. Bartel LL, Hussey JL, Shrago E: Effect of dialysis on serum carnitine, free fatty acids, and
triglyceride levels in man and the rat. Metabolism 31: 944-947, 1982.
155. Munro HN: Nutritional requirements for health: their relevance to tissue function in renal
failure. Am 1 Clin Nutr 22: 1555,1980.
156. Grodstein G, Blumenkrantz MJ, Kopple ID: Nutritional and metabolic response to catabolic
stress in uremia. Am 1 Clin Nutr 33: 1411-6, 1980.
157. Grodstein GP, Blumenkrantz MJ, Kopple JD: Effects of intercurrent illnesses on nitrogen
metabolism in uremic patients. Trans Am Soc Artif Int Organs 25: 438-42,1979.
158. Mehls 0, Ritz E et al.: Effect of vitamin D on growth in experimental uremia. Am J Clin Nutr
31B: 1927, 1978.
159. Chan tier C, El Bishti M, Counahan R: Nutritional therapy in children with chronic renal failure.
Am J Clin Nutr 33: 1682, 1980.
160. Blumberg, Hanck A, Sander G: Vitamin nutrition in patients on continuous ambulatory per-
itoneal dialysis (CAPD). Clin Nephrol20: 244-250,1983.
161. Kopple JD, Swendseid ME: Vitamin nutrition in patients undergoing maintenance hemo-
dialysis. Kid Int 7: S79, 1975.
 399

162. Trapas G, Magoula I, Paletas K, Con co uris L: Effect of peritoneal dialysis on plasma levels of
ascorbic acid. Nephron 33: 34-37, 1983.
163. Papadoyanakis N, Ziroyaris P, Papathanasion E et al.: The effect of peritoneal dialysis on serum
folic acid binding capacity. In: Advances in Peritoneal Dialysis. Ed. by Gahl GM, Kessel M,
Nolph KD. Excerpta Medica, Amsterdam, p 70-72, 1981.
164. Aloni Y, Shany S, Chaimovitz C: Losses of 25-hydroxyvitamin D in peritoneal fluid: possible
mechanism for bone disease in uremia patients treated with chronic ambulatory peritoneal
dialysis. Mineral Electrolyte Metab 9: 82-86, 1983.
165. Gokal R, Ellis HA, Ramos Hvf et al.: Improvement in secondary hyperparathyroidism in
patients on continuous ambulatory peritoneal dialysis. In: Advances in Peritoneal Dialysis. Ed.
by Gahl GM, Kessel M, Nolph KD. Excerpta Medica, Amsterdam p 461-466, 1981.
166. Gokal R, Parkinson IS, Ramos M: Elevated serum aluminum levels in patients on continuous
ambulatory peritoneal dialysi. In: Advances in Peritoneal Dialysis. Ed. by Gahl GM, Kessel M,
Nolph KD. Excerpta Medica, Amsterdam, p 478-482,1981.
167. Hercz G, Milliner D, Shinaberger Jet al.: Aluminum metabolism and removal during CAPD
(Abstract). Kidney Int, 1984.
168. Milman N, Christenson TE, Bartels U, Pedersen NS: Iron status in patients undergoing regular
peritoneal dialysis. Dan Med Bull 27: 291, 1980.
169. Sanfelippo ML, Swenson RS, Reaven GM: Reduction of plasma triglycerides by diet in subjects
with chronic renal failure. Kidney Int 11: 54-61,1977.
170. Sanfelippo ML, Swenson RS, Reaven GM: Response of plasma triglycerides to dietary change
in patients on hemodialysis. Kidney Int 14: 180-186, 1978.
171. Kopple JD, Mercurio K, Blumenkrantz MJ et al.: Daily requirement for pyridoxine HCI
supplements in chronic renal failure. Kidney Int 19: 694-704, 1981.
172. Symposium on Role of Dietary Fiber in Health. Nutrition Coordinating Committee of the
National Institutes of Health, March 1977, Bethesda, Md, Am J Clin Nutr, 31, 1978.
173. Lam M: Influence of aluminum-containing antacids on plasma aluminum. (Abstract) Kidney Int
23: 129, 1983.
174. Recommended dietary allowances. Ninth Revised Edition, National Academy of Sciences,
Washington, D.C., 1980.
175. Kopple JD, Dirige OV, Jacob M et al.: Transketolase activity in red blood cells in chronic
uremia. Trans Amer Soc Artif Int Organs 18: 250, 1972.
176. Goldberg AP, Hagberg J et al.: The metabolic and psychological effect of exercise training in
hemodialysis patients. Am J Clin Nutr 33: 1620, 1980.
177. Greene MD, Zabetakis PM et al.: Effect of exercise on lipid metabolism and dietary intake in
hemodialysis patients. Proc Dial Transpl Forum 8: 80, 1979.
178. Zabetakes P, Gleim G et al.: Long term sub maximal exercise conditioning in hemodialysis
patients. Clin Nephrol18: 17,1982.
179. Kettner A: Exercise in dialysis patients. Int J of Artif Organs 5: 83, 1982.
180. Booth FW: Anabolic effects of exercise on muscle and bone. Artif Kidney Chronic Uremia
Program, Jan 15-17, Bethesda, Md, p 268,1979.
181. Carney R, McKevitt P et al.: Psychological effects of exercise training in hemodialysis patients.
Nephron 33: 179, 1983.
182. Kopple JD, Blumenkrantz MJ: Nutritional therapy of the urological patient. Urol Clin North
Am 3: 403, 1976.
183. Blackburn GL, Bistrain BR: Nutritional care of the injured and/or septic patient. Surg Clin
North Am 56: 1195, 1976.
184. Shils ME: Ed: Defined formula diets for medical purposes, Chicago. Am Medical Assoc, 1977.
185. Rubin J, McFarland S et al.: Peritoneal dialysis during peritonitis. Kidney Int 19: 460-464, 1981.
186. Hinton P, Littlejohn S et al.: Insulin and glucose to reduce catabolic response to injury in burned
patients. Lancet 1: 767, 1971.
400

187. Lewis L, Dahn M, Kirkpatrick J: Anabolic steroid administration during nutritional support: a
therapeutic controversy. J Parent Ent Nutr 5: 64-66, 1981.
188. Kopple JD, Blumenkrantz MJ: Total parenteral nutrition. In: Clinical Disorders of Fluid and
Electrolyte Metabolism. McGraw Hill, 3rd Ed. Eds. Kleemen CR, Maxwell MH, New York, p
413, 1979.
189. Kopple JD, Grodstein GP, Blumenkrantz MJ: Nutritional status of the diabetic patient with
chronic uremia. In: Diabetic Renal-Retinal Syndrome. Eds. Friedman EA, L'Esperance FA jr,
Frune & Stratton, New York, p 239,1980.
190. Comty CM, Leonard A, Shapiro FL: Nutritional and metabolic problems in the dialyzed patient
with diabetes mellitus: Kid Int 6: S51, 1974.
191. Miller D, Levine S et al.: Diagnosis of protein calorie malnutrition in diabetic patients on
hemodialysis and peritoneal dialysis, Nephron 33: 127-132, 1983.
192. Haber M, Pettit J: Nutritional considerations in the diabetic patient on CAPD, Perit Dial Bull 3:
S5G-51, 1983.
193. Longstreth GF, Malagelada JR, Kelly KA: Metoclopramide stimulation of gastric motility and
emptying in diabetic gastroparesis. Ann Intern Med 86: 175,1977.
194. Jones R, Poston L et al: Weight gain between dialysis in diabetics: possible significance of raised
intracellular sodium content. Brit Med J, P 153, 1980.
195. Kleinknecht C, Boroyer M, Gagnadoux MF et al.: Growth in children treated with long-term
dialysis. Maxwell MH (Ed): Advances in Nephrology. Chicago Year Book Medical Publishers,
Inc. pp 133-163, 1980.
196. Fine RN, Isaacson AS, Payn V, Grushkin CM: Renal osteodystrophy in children: the effect of
hemodialysis and renal homotransplantation. J Pediatr 80: 243-249, 1972.
197. Balfe JW, Vigneux A, Willumsen J, Hardy BE: The use of CAPD in the treatment of children
with end-stage renal disease. Peritoneal Dial Buill: 35-38, 1981.
198. Alexander SR, Tseng CH, Maksym KA, Campbell RA, Talwalkar YB, Kohaut EC: Clinical
parameters in (CAPD) for infants and children. Moncrief JW, Popovich RP (eds): CAPD
Update. Texas, Masson Publishing, pp 196-209, 1981.
199. Potter DE, McDaid TK, McHenry K, Mar H: Continuous ambulatory peritoneal dialysis in
children. ASAIO 27: 64-67, 1981.
200. Salusky IB, Lucullo L, Nelson P, Fine RN: Continuous ambulatory peritoneal dialysis in
children. Clin North Am 29: 1005-1012, 1982.
201. Baum M, Powel D, Calvin S et al.: Continuous ambulatory peritoneal dialysis in children.
Comparison with hemodialysis. N Engl J Med 307: 1537-1542, 1982.
202. Salusky IB, Kopple JD, Fine RN: Continuous ambulatory peritoneal dialysis in children - A 20
months' experience. Kidney Int Nov supplement, 1983.
203. Hamill PVV, Drizd TA, Johnson CL, Reed RB, Roche AF: NCHS growth curves for children
birth - 18 years: United States. (Vital and health Statistics. Series 11: Data from the National
Health Survey, no. 165.) Hyattsville, Maryland, National Center for Health Statistics, 1977.
204. Roche AF: Growth assessment in abnormal children. Kidney Int 14: 369-377, 1978.
205. Simmons JM, Wilson CJ, Potter DE, Holliday MA: Relation of calorie deficiency to growth
failure in children on hemodialysis and the growth response to calorie supplementation. N Engl J
Med 285: 653-656, 1971.
206. Betts PR, Magrath G, White PHR: Role of dietary energy supplementation in growth of
children with chronic renal insufficiency. Br J Med 1: 416-418, 1977.
207. The Committee on Dietary Allowances: Recommended Dietary Allowances. Washington,
D.C., National Academy of Sciences, 1980.
208. Salusky IB, Fine RN, Nelson P, Blumenkrantz MJ, Kopple JD: Nutritional status of children
undergoing continuous ambulatory peritoneal dialysis. Am J Clin Nutr 38: 599, 1983.
209. Goldberg ACK, Eliaschewitz FG, Quintao ECR: Origin of hypercholesterolemia in chronic
experimental nephrotic syndrome. Kidney Int 12: 23-27, 1977.
 401

210. Chantler C, Holliday MA: Growth in children with renal disease with particular reference to the
effects of calorie malnutrition: a review. Clin Nephroll: 230-242, 1973.
211. Delaporte C, Bergstrom 1, Broyer M: Variations in muscle cell protein in severely uremic
children. Kidney Int 10: 239-245, 1976.
212. Counahan R, El-Bishti M, Cox BD, Ogg CS, Chantler C: Plasma amino acids in children and
adolescents on hemodialysis. Kidney Int 10: 471-477, 1976.
213. Giordano C, DeSanto NG, Capodicesa G: Amino acid losses during CAPD in children. Int 1
Pediatr Nephrol 2: 85-88, 1981.
214. Brown EM, Wilson RE, Eastman RC, Pallotta 1, Barynick SP: Abnormal regulation of
parathyroid hormone release of calcium in secondary hyperparathyroidism due to chronic renal
failure. 1 Clin End Metab 54: 172-179, 1982.
215. Hampers CL, Streiff R, Nathan DG et al.: Megaloblastic hematopoiesis in uremia and in
patients on long-term hemodialysis. N Engl 1 Med 276: 551-554, 1967.
216. Whitehead VM, Comty CH, Posen GA et al.: Homeostasis of folic acid in patients undergoing
maintenance hemodialysis. N Engl 1 Med 279: 970-974, 1968.
217. Casey CE, Moser MC, Hambridge M, Lum GM: Zinc, copper, and vitamin A in pediatric
dialysis. 1 Pediatr 98: 434-435, 1981.
218. Atkin-Thor, Goddard BW, O'Nion 1, Stephen RL, Kolff Wl: Hypogeusia and zinc depletion in
chronic dialysis patients. Am 1 Clin Nutr 31: 1948-1951, 1978.
219. Mahler Dl, Walsh lE, Haynil GD: Magnesium, zinc and copper in dialysis patients. Am 1 Clin
Path 56: 17, 1971.
220. Busch RE, Hamm MK, Sullivan IF: Newer aspects of the roles of zinc, manganese and copper in
human nutrition. Clin Chern 21: 501-520, 1975.
221. Grupe WE, Kopito LE, Lazarus 1M et al.: Copper and zinc depletion in end-stage renal disease.
Abstract. Pediatr Res 10: 439, 1976.
222. Ward MK, Feest TG, Ellis HA, Parkinson IS, Kerr DNS et al.: Osteomalacic dialysis os-
teodystrophy: evidence for a waterborne aetiological agent, probably aluminum. Lancet 1: 841-
848,1978.
223. Baluarte HG, Gruskin AB, Hiner LB: Eucephalopathy in children with chronic renal failure.
Proc Dial Transplant Forum 7: 95-98, 1977.
13. Peritonitis

STEPHEN I. VAS

1. Introduction

The frequent occurrence of peritonitis - one of the major complications of

peritoneal dialysis - has hindered the development and acceptance of this
technique.
In the mid-1940s, Seligman [1], Frank [3] and Fine [2] reported a method of
continuous peritoneal lavage, but the high incidence of infection and technical
difficulties, as well as the introduction of the artificial kidney, discouraged its
supporters. In 1951, Grollman [4] drew attention again to the value of peritoneal
dialysis in the treatment of acute renal failure when he introduced intermittent
peritoneal lavage. Later, Doolan [5] confirmed its safety and effectiveness in
human sUbjects. Another major advance, the introduction of the nylon catheter
which allowed safe access to the peritoneal cavity [6, 7], made possible intermit-
tent peritoneal dialysis with infection rates of 5.2 to 7.5 episodes of peritonitis per
patient year of dialysis [8, 9].
In 1964 Palmer and associates [10] began to provide long-term intermittent
peritoneal dialysis employing an indwelling silicone rubber irrigating catheter
device which Tenckhoff [11] later modified to its present form. This technique
reduced infection rates to 0.23 to 1.2 episodes of peritonitis per patient year [12-
18].
In 1976 Popovich and colleagues [19] described the technique of continuous
ambulatory peritoneal dialysis (CAPD) for the treatment of chronic renal failure.
The incidence of peritonitis with this method, employing bottled dialysate, was
4.6 episodes of peritonitis per patient-year [20]. This technique was subsequently
modified and replaced the bottled dialysate with a plastic dialysate bag [21]. This
arrangement was more convenient for the patients and substantially reduced the
number of manipulations of the catheter and therefore the infection rate. Sub-
sequently, many workers using this technique have reported rates of peritonitis
varying from 1.2 to 6.3 episodes per patient year [20--26]. As noted above, these
high rates of infection have been the major criticism ofthe procedure [27-30], but
404

with increasing control of peritonitis, CAPD could become the dialysis of choice
for many patients with end-stage renal disease.
In fact with appropriate training, selection of patients and precautionary
measures, peritonitis rates can be reduced to one episode every 18 to 24 months.
Some units report even better results than this.
The US CAPD Registry [31] - which follows data from over 8000 patients-
reports a peritonitis rate of1.6,to 1.8 episodes/patient-year. This rate is an average
rate including peritoneal dialysis units which perform the procedure on very few
patients and very large units. Similarly it includes units which are just starting
with the program as well as units which have many years experience.
It appears that peritonitis rates can be reduced by careful training, use of
appropriate equipment, etc. to a rate of approximately one episode of peritonitis
every two years. If this rate becomes the generally obtained rate, CAPD will have
a much better acceptance for the treatment of endstage renal failure.
Intermittent peritoneal dialysis has a lower rate of peritonitis. It is not unusual
to observe peritonitis rates of one every 3 to 5 years. The reason for this difference
is not quite clear. If one compares certain parameters of intermittent peritoneal
dialysis with continous ambulatory peritoneal dialysis (see Fig. 1) it becomes
obvious that the number of connections made during intermittent peritoneal
dialysis by a cycler is larger and the total volume of fluid used is more. Intermit-
tent peritoneal dialysis patients do not have dialysis fluid in the abdominal cavity
for the major part of their program (usually 2 days off, 12 to 18 hours on). It is
speculative but plausible that peritoneal defense mechanisms are operating better
when the peritoneum does not contain large volumes of fluid (see discussion on
defence mechanisms below).
CCPD or continuous cycling peritoneal dialysis has peritonitis rates reported in
between CAPD and IPD [32], though the recent report of the National CAPD
Registry of the US [31] does not substantiate this claim. This modality of treat-
ment is generally less accepted.

2. Pathogenesis of peritonitis

Initially those who had to deal with peritonitis episodes in peritoneal dialysis
patients perceived peritonitis as similar to the experience with surgical peritonitis
[33]. While this was a reasonable approach, it became clear that peritonitis
developing in dialysis patients has considerable differences from that of surgical
peritonitis.
While small amounts of contamination do not usually cause surgical peritonitis
(as evidenced by the large number of laparotomies which heal without any
evidence of clinical peritonitis), minor contaminations in the peritoneal dialysis
patients will lead to peritonitis. Similarly, surgical peritonitis develops mainly
when the abdominal cavity has a major soil, usually by faecal contents and
 405

ESRD

IPD CAPD

Days of dialysis per year 104 365

No. of connections per year 2600 1460
Flow rate liter per year 8320 2912
Peritonitis per year 0.14 0.63

Figure 1. Comparison of IPD and CAPD at Toronto Western Hospital.

therefore removal of the contaminating material is essential. In peritoneal di-

alysis, such large amounts of contaminants are rare. Finally, in surgical peritonitis
about 30% of the patients will show bacteremia as part of the disease [60], while in
peritonitis of peritoneal dialysis patients positive blood cultures are the rare
exception rather than a frequent event and when positive blood cultures are
observed, they are usually the harbinger of a haematogenous source of peritoneal
infection [70]. It is for such reasons that peritonitis of CAPD patients has now
been accepted as a special disease entity and its management requires a different
approach.
More similar to peritonitis in PD patients is the disease called spontaneous
bacterial peritonitis [34-39] which is occasionally seen in patients suffering from
cirrhosis of the liver with ascites. It is believed that in these patients the lack of
reticuloendothelial function of the liver is the primary reason why these patients
develop spontaneous peritonitis. These patients also have large volumes of fluid
in their abdominal cavity which may also explain some of the similarities.

2.1. Portals of entry

Microbial environment and host are living in a symbiotic relationship. The

occasional penetrations of infectious organisms into an intact host are met with
defence and usually the small number of invaders is destroyed [40]. The outcome
therefore is dependent on a delicate balance of the number of invaders vs.
defense. Bacterial penetration is probably a frequent event (it is known for
example that minor exertions like brushing of teeth or bowel movements will lead
to transient short lasting bacteremias in many normal individuals) and will only
rarely lead to major infections.
We do not know how frequently infectious events occur in peritoneal dialysis
patients. It is well conceivable that these events are more frequent than the
episodes of peritonitis. Since the conditions which lead to peritonitis are not
known, all portals of entry have to be considered seriously and, to reduce
peritonitis, all of them have to be managed with great care.
Surveillance cultures from abdominal skin, throat and hands have been done in
406

our unit on peritoneal dialysis patients before they enter the dialysis program.
The results of such a surveillance are shown in Table 1. These are the areas from
which incidental contaminations of peritoneal dialysis patients can more fre-
quentlyoccur. In fact, if one compares the Staphylococcus aureus phage type or
the Staphylococcus epidermidis biotype isolated from the peritoneal fluid of these
patients during peritonitis episodes with the skin biotype as shown in Table 2, it is
obvious that the patients are at higher risk from their own flora than for acquiring
infections from the environment or other people.
It is therefore possible to generate a probable listing (see Table 3) estimating
from the type of organisms isolated from peritonitis the probable route of entry.
This is a speculative listing making an educated guess on the route through which
infections may occur. For example it is assumed that 2/3 of S. epidermidis
infections occur through the catheter intraluminal route while only 1/2 of S.
aureus infections occur through the same route. Such a listing is helpful (and by
experience close to accurate) in understanding the role of various portals of
entry.

2.1.1. Intraluminal infections

Intraluminal infections occur when bacteria enter the internal pathway of per-
itoneal dialysis tubing through an external lumen or through cracks developed in
the tubing. The commercial dialysis fluids are prepared with great care and they

Table 1. Surveillance cultures of 47 CAPD patients

72 Surveillance cultures Hand Abdomen Nose Total

Coagulase-negative sta-
phylococci
S. epidermidis 46 (70%) 32 (57%) 48 (82%) 126 (70%)
S. warneri 3 (5%) 2 (4%) 1 (1 %) 6 (3%)
S. haemolyticus 7 (11 %) 9 (16%) 4 (7%) 20 (11%)
S. hominis 4 (6%) 9 (16%) 4 (7%) 17 (9%)
S. capitis 1 (1%) 0 0 1 (less than 1% )
S. simulans 2 (3%) 1 (1%) 1 (1 %) 4 (2%)
S. saprophyticus 2 (3%) 2 (4%) 1 (1 %) 5 (3%)
S. xylosus 0 0 0 0
S. cohnii 1 (1 %) 1 (1%) 0 2 (1%)
Total 66 (76%) 56 (69%) 59 (63%) 181 (69%)

S. aureus 3 (3%) 3 (4%) 7 (7%) 13 (5%)

Gram-negative bacilli 3 (3%) 4 (5%) 8 (9%) 15 (6%)
Others: diphtheroids,
fungus, streptococci 15 (17%) 18 (22%) 20 (21%) 53 (20%)

Total 87 81 94 262
 407

Table 2. Correlation of staphylococcal isolates in CAPD patients with peritonitis

Patients with same biotype Patients with different biotype

Caogulase-negative 45 (96%) 2 (4%)
staphylococci (47)
(biotype)
For comparison
Patients with same phage type Patients with different phage
type
Coagulase-positive 19 (86%) 3 (14%)
staphylococci (22)
(phage type)

can be considered sterile. There have been no episodes ascribed to peritoneal

dialysate from commercial sources in the last few years [41]. Additions through
the port to the peritoneal dialysis bag have to be made with sterile precautions
and fresh vials of drugs should be used for each addition.
Most of the intraluminal infections develop after accidental contamination of
the spike by touch contamination or disconnects of the dialysis tubing. Touch
contaminations can be reduced by devices which provide no touch mechanisms
and accidental disconnects have been reduced con~iderably with the introduction

Table 3. Sources of infections in CAPD patients

Route Organism %

Transluminal Staphylococcus epidermidis 23

Staphylococcus aureus 7
Acinetobacter 2
32

Periluminal Staphylococcus epidermidis 11

Staphylococcus aureus 8
Fungus 4
23

Transmural Enteric organisms 26

Anaerobic 3
29

Hematogenous Streptococcus viridans 11

Mycobacterium tuberculosis
12

Other 3
408

of titanium locking connectors. Various other devices (see 11.1.) have been
devised to reduce intraluminal infections.

2.1.2. Periluminal infections

The silastic catheter never forms a completely sealed junction with the skin or
subcutaneous tissues [42]. While the purpose of the cuffs (single or double) was to
reduce the penetration of bacteria around the cathete~, it does not always achieve
its goal. Therefore, bacterial penetration around the catheter is a possibility.
Casual penetration of bacteria around the catheter is probably not a factor; to
develop peritonitis from a periluminal source either an exit site infection or a
tunnel infection has to develop. Studies in our unit have shown that, while
originally the exit site was kept under occlusive dressings, the removal of such
occlusive dressings and permitting the patient to take daily showers without any
cover of the exit site does not lead to an increased incidence of peritonitis. It is
therefore probable that an infectious process has to establish itself in the tissues of
the exit site or the subcutaneous tunnel to produce a peritoneal infection.
Bacteria may enter from these subcutaneous infections occasionally leading to
clinical symptoms.

2.1.3. Transmural (intestinal) infections

The isolation of multiple intestinal organisms from peritoneal dialysis fluid,
especially if they belong to more than one group of bacteria or if anaerobic
organisms are isolated, is indicative of a fecal leak. There is some evidence that
bacteria may migrate rarely through intact intestinal walls [43]; ischemic bowel
disease may lead to more frequent penetration [36]. It is likely that a more
common cause of intestinal leak is preexisting diverticulosis in these patients. It is
known that diverticulosis increases with age and also there is an increased
incidence of diverticulosis in polycystic disease. A study done by us showed that
diverticulosis was a major source of fecal peritonitis [44].

2.1.4. Hematogenous infections

It has been shown that hematogenous peritonitis develops frequently in the
spontaneous bacterial peritonitis of cirrhotics [34, 35]. We have observed in a few
patients that Streptococcus viridans peritonitis developed in patients who had
acute upper respiratory infections. Some of these patients preceeded the develop-
ment of peritonitis with positive blood cultures from which the same organism,
usually Streptococcus viridans, was isolated. It is also probable that tuberculous
peritonitis develops through this route [45].

2.1.5. Other endogenous infections

Rarely, other sources can be implicated in the development of peritonitis. We
[46] and others [47, 48, 69] have observed women who had a vaginal leak of
peritoneal dialysis fluid. Some of these developed Candida peritonitis. Tubal
 409

ligation was indicated in some of these patients after which the vaginal leak
disappeared and there was no recurrence 'of peritonitis.

2.1.6. Environmental infections

Peritonitis from which Pseudomonas maltophilia, Acinetobacter [49] or other
environmental bacteria are isolated, is observed sometimes. It is possible that
these infections develop by contact with water [50] (showering, swimming pool)
that enters through the exit site. In peritoneal dialysis patients, episodes of
infections with Mycobacterium chelonei [51, 52] have been described where the
source was ascribed to tap water entering the peritoneal cavity.

2.2. Inflammatory response

The normal homeostasis of the peritoneal cavity is disrupted if bacteria or

chemical stimuli enter the peritoneal cavity.

2.2.1. Inflammatory mediators

Bacteria combining with normal opsonins present in the peritoneal cavity (in the
presence of complement) result in the release of chemotactic factors. The latter
stimulate the outflow of polymorphonuclear macrophages, thus increasing the
cell number present in the peritoneal cavity and shifting it from a predominantly
mononuclear cell population [143] to a polymorphonuclear cell population [33].
Other inflammatory mediators are released, like histamine and serotonin [53],
which result in vasodilatation, and increased protein in the outflow. Some of the
mediators may contribute to typical peritoneal pain.

2.2.2. Fibrin
The normal peritoneal cavity fluid contains fibrinogen and also fibrinolysin [54]
which breaks down fibrin and maintains the shiny slippery surface of the per-
itoneum. During inflammation, fibrinolysis is impaired while increased amounts
of fibrinogen enter the peritoneal cavity. Resulting fibrin will not be broken down
rapidly enough because of the lack of fibrinolysis. The result is the formation of
fibrin filaments and fibrin clots.
Myhre Jensen [55] and Porter [56] have shown that the fibrinolytic activity of
the mesothelial surface of the peritoneum resides in the mesothelial cells. The
presence of fibrinolytic activities suggests that fibrinolysis may assist in removing
fibrin deposits from these surfaces. However, it has been demonstrated that
certain stumuli such as cuts, abrasions, and ischemia are associated with local
depression of peritoneal fibrinolytic activity [57-59]. Also, Hau et al. [60] were
able to decrease fibrinolytic activity by inducing bacterial peritonitis. Thus it
appears that the depression of fibrinolytic activity of the peritoneum is an
important mechanism in the development of fibrin formation and probably in
intraperitoneal adhesions.
410

2.2.3. Cellular response

The normal peritoneal cell population is primarily mononuclear, probably mac-
rophages of blood origin and some mesothelial cells from the peritoneal lining. In
inflammation, rapid migration of polymorphonuclear cells occurs [143]. The
rapidity of this migration is truly amazing. It may take only a few hours for a
completely clear peritoneal fluid to turn cloudy. The estimation of peritoneal cell
population is a useful adjunct in the diagnosis of peritonitis. The cellular response
also follows the improvement of peritonitis on therapy and is a good clinical guide
to therapeutic success. Occasionally, eosinophilic cells enter the peritoneal cavity
(see below 7.3.2.).
Abscess formation is an infrequent complication of the peritonitis of peritoneal
dialysis patients. Onderdonk et al. [84] demonstrated that intraabdominal abscess
formation appeared to be related to a synergy between anaerobes and gram
negative aerobic bacteria. Since isolation of anaerobes is a rare occurrence in the
peritonitis of peritoneal dialysis patients, abscess formation in this type of per-
itonitis should be suspect as due to faecal flora. Also Staphylococcus aureus is a
organism which is associated with frequent abscess formation due to its effect on
fibrin around the organisms [144].

3. Defense mechanisms of the peritoneum

The peritoneal membrane lines the interior of the abdominal wall (parietal
peritoneum) and the abdominal viscera (visceral peritoneum) forming a potential
space of the peritoneal cavity. The peritoneal membrane consists of a surface
layer of mesothelial cells which lie on a basement membrane with deeper layers of
capillaries and lymphatics. Transport through the peritoneal membrane, moves
from the capillaries through the basement membrane through intercellular junc-
tions. Small particles may have two methods for transport: moving through
cellular junctions or through pinocytosis by mesothelial cells [148, 149]. The
principal route for movement of small particles from the peritoneum is via the
lymphatics, primarily through the lymphatics below the diaphragmatic surface
[150]. The exact mechanism of this movement is not clear although Courtice and
Simmonds [151] have postulated that openings exist between the peritoneal cavity
and the diaphragmatic lymphatics. During peritonitis the primary flow is towards
the peritoneal cavity. This may explain the extremely low rate of bacteremia in
peritoneal dialysis patients during peritonitis. In secondary surgical peritonitis
the rate of bacteremia is 30% [152]' In cirrhotic patients with spontaneous
peritonitis the rate of bacteremia is between 39 to 76% [34,35, 61]. The rate of
bacteremia in patients on intermittent peritoneal dialysis is about 15% [8]. We
have not observed a single positive blood culture in several hundred episodes of
bacterial peritonitis in CAPD patients except as noted above where the bac-
teremia preceeded peritonitis. The defense mechanism of the peritoneum is
 411

probably the single most important factor in the removal of small amounts of
microorganisms from the peritoneal cavity.

3.1. Humoral factors

It has been shown that the immunoglobulins and complement are present in the
peritoneal fluid though the normal level of these components is not established.
The serum immunoglobulin levels of these patients are not severely compromised
[62]. Some patients may have a decreased level of opsonins and probably other
factors needed for immunological reactions in the peritoneal cavity [63]. Patients
who are unable to transport these factors may have increased susceptibility to
peritonitis.

3.2. Cellular factors

While the normal self clearing mechanism of the peritoneum is primarily depen-
dent on mesothelial cells and mononuclear cells in the peritoneal cavity during
inflammation, a large number of active phagocytic polymorphonuclear cells enter
the peritoneal cavity and participate in the removal of bacteria. Under the
conditions of peritoneal dialysis, patients have instead of the normal few milli-
litres of fluid, 2litres of dialysis fluid with a low pH and high osmolality. We have
shown that low pH and high osmolality decrease the efficiency of phagocytic cells
which may not be fully functional at least during the initial periods of peritoneal
dialysis [64]. In addition, urea, creatinine and other low molecular weight sub-
stances enter the peritoneal cavity during peritoneal dialysis. We have examined
the effect of these molecules and did not find them deleterious to phagocytosis in
the concentrations present in the peritoneal dialysis fluid. Similarly, heparin,
which is added to reduce fibrin formation did not appear to inhibit phagocytosis.
The relative ratio of bacteria to cells is also important in the efficiency of
phagocytosis. The large volume present in the peritoneal cavity during peritoneal
dialysis dilutes this ratio and the chance of phagocytosis is diminished. It is
therefore preferable to use smaller volumes of peritoneal dialysis fluid during the
therapy of acute peritonitis.
The role of the eosinophils in peritoneal dialysis fluid is not established. While
some phagocytosis is performed by eosinophilic cells, they probably represent
reaction to inflammatory agents rather than a primary defense mechanism.
The role of lymphocyte mediators in peritoneal defences is not clear. While it is
known that end stage renal failure inhibits cellular immune functions, the role of
such inhibition is not established in peritonitis [65, 66]. Patients with renal failure
are considered more susceptible to infections [72], but the role of various immune
functions requires re-investigation.
412

4. Microbiological diagnosis of peritonitis

4.1. Specimen

In order to establish accurate microbiological diagnosis of peritonitis, the follow-

ing points are important:
a. Cultures should be taken as early as possible from suspected cases of periton-
itis.
b. Large volumes should be concentrated for improving recovery rate.
c. Washing of the specimens with sterile saline may be necessary in patients while
on antibiotic therapy.
d. Aggressive microbiological methods should be used to identify the organisms
as soon as possible to achieve rational antibiotic therapy [67].
In the Toronto Western Hospital, patients are instructed to bring their first
cloudy bag with them to establish diagnosis. This first bag is delivered immedi-
ately to the laboratory. Initially it was believed that culturing of organisms should
be done immediately after drainage of the fluid. In a small study we established
that even several hours or days delay will not decrease the accuracy of bac-
teriological diagnosis from these fluids. Therefore, the sample which arrives at
night can be safely kept until the morning for bacteriological identification [68].

4.2. Gram stain

Gram stain from the sediment of the peritoneal dialysis bag establishes the
presence of microorganisms only in about 20 to 30% of the cases. This is not a
sufficiently high ratio and does not permit rational therapy. While gram stains are
customarily done, the therapy of these patients will have to be started on
speculative therapy rather than based on the gram stain. The only time where a
gram stain may be helpful is the establishment of fungal peritonitis which fre-
quently can be seen in the initial gram stain, thus, preventing the patient from
receiving unnecessary antibiotic therapy.

4.3. Culture procedure

DIALYSIS BAG SAMPLING PROTOCOL

Toronto Western Hospital
Medical Microbiology

Medium used: BBL Thioglycollate (#135C) with SPS (Liquoid) 0.05% added.

I. Cloudy bag procedure

Mix contents of bag well before sampling.
 413

Flood port with methyl alcohol and leave for 60 seconds.

Allow approximately 20 ml to run from sampling port to clear it of alcohol.
A. 1. Aseptically fill two 25 x 150mm screw cap tubes (approximately 50ml each) and 1 tube of
15ml.
2. Centrifuge the 2-50 ml tubes for 30 minutes at top speed.
3. Aseptically decant supernatants. Vortex to resuspend sediment.
4. THE TUBE OF 15 ml GOES TO ROUTINE BENCH.
AAI. Patient not on antibiotics
One tube of 15 ml goes to routine bench AND IS CENTRIFUGED, SEDIMENT planted as
follows:
Blood agar (BA) 02
McConkey (McC) 02
Pre-reduced medium (CMC) -3--4 drops
Brain heart infusion broth (BHI) poured over sediment
Gram stain to be read STAT
Two tubes of 50ml sediments are injected using a sterile 1 ml syringe into 100ml TWH
thioglycollate (sterilize top with alcohol first).
AA2. Patient on antibiotics
Sediment in tubes (Step A2) should be resuspended in 50 ml sterile saline and centrifuged for
30 minutes at top speed.
Aseptically decant the supernatants.
Sediment of two tubes injected into 100ml TWH thioglycollate (BBL #135C).
A gram stain is made of the 100 ml sediment in case organisms are not seen in routine smear.
Tube of 15 ml handled on routine bench as in AA1 (not washed).
11. Clear bag procedure
Mix contents of bag well before sampling.
Flood port with methyl alcohol and leave for 60 seconds.
Allow approximately 20 ml to run from sampling port to clear it of alcohol.
AI. One tube of 15 ml goes to routine bench for planting as follows:
CMC - 3--4 drops
No differenbe in routine planting if antibiotics in use or not.
A2. Use of Addichek (Millipore) filtration system.
Spike Addichek into the sampling port and allow chamber to fill with approximately 100 ml of
effluent. Clamp bag, remove Addichek and replace spike cover.
Remove yellow cap on base of Addichek and connect to vacuum flask. Effluent will flow
through membrane filter.
Clamp off vacuum flask when Addichek is empty.
A3. Patient on antibiotics
Spike a minibag of sterile saline (100 ml) and allow chamber to fill. Connect Addichek to
vacuum flask and allow saline to run through filter. Spike 100 ml bottle of TWH thioglycollate
and allow chamber to fill.
A4. Patient not on antibiotics
A bottle of 100 ml TWH thioglycollate is spiked and chamber allowed to fill.
Incubation and examination
24 hours - 'no growth' reported
48 hours - routine tubes transferred to room temperature and checked daily for growth. Keep
2 weeks.
Anaerobic cultures
CMC - incubated at 36° C 2-3 weeks and checked daily.
Addichek
Incubated at 36°C 2-3 weeks and checked daily.
414

The speed with which bacteriological diagnosis can be established is important

[70]. The concentration method increases not only the possible proper identifica-
tions but also reduces the length necessary for bacteriological cultures. Large
number of cultures will become positive after the first 24 hours and in over 75% of
them, the diagnosis can be established in the initial 3 days.

4.4. Antibiotic sensitivities

Antibiotic sensitivities of the isolated organisms are established by normallabo-

ratory procedures against the primary and secondary antibiotics used in treat-
ment of peritoneal dialysis. It is important to keep current statistics on the
percentage sensitive and resistant organisms isolated from peritoneal dialysis
patients against each antibiotic since these may vary from hospital to hospital.
Such a table is helpful in establishing arbitrary speculative antibiotic therapy in
the first few days before antibiotic sensitivities are available.

4.5. Cell count

Cell counts are routinely established from the first peritoneal dialysis fluid and
preferably daily afterwards to follow the efficiency of therapy [143]. We use
centrifugal (Cytospin) smears for differential cell counts in judging the clinical
course of peritonitis. Normal white cell counts are considered to be between 50 to
100 per cubic millimeter. It is helpful to remember that cell counts are a function
of the abdominal fluid volume and change in volume may result in a change in cell
count not necessarily indicating impairment or improvement in the inflammatory
process.

5. Presenting signs and symptoms of peritonitis

The literature contains only general reference to presenting symptoms and signs
of patients with peritonitis [16]. These manifestations include mild abdominal
pain, low grade temperature, and usually mild abdominal tenderness. Conn [35]
described clinical features of patients with spontaneous peritonitis. Fever was
present in these patients in 81%, abdominal pain in 78% and physical signs of
peritonitis in 65%. 6% of his patients had no symptoms or signs. We have
reviewed presenting symptoms and signs in 103 episodes of peritonitis of CAPD
patients. Fever of greater than 37.5° was present in 53%, abdominal pain was
present on admission on 79%. 31% experienced nausea and 7% complained of
diarrhea. All but 1 patient had cloudy fluid before admission but only 78% had
cloudy fluid on admission. 70% showed abdominal tenderness and 50% rebound
pain.
 415

We use a practical definition [73] of peritonitis which require the presence of

two of the following criteria in any combination:
a. The presence of organisms on gram stain or in culture in the peritoneal dialysis
fluid,
b. Cloudy fluid with inflammatory cells and
c. Symptoms of peritoneal inflammation.
We find this working diagnosis reliable since we have established that above 100
white blood cells per cubic millimeter, the peritoneal dialysis fluid is visibly
cloudy. Therefore no cell count has to be done on an emergency basis.
Besides the above mentioned presenting signs and symptoms patients some-
times present with profound hypotension and shock. This presentation is usually
a sign of either Staphylococcus aureus peritonitis or faecal peritonitis.
Many of the patients will have only very mild symptoms and do not require
hospitalization.

6. Clinical course of peritonitis

6.1. Incubation period

The incubation period of peritonitis in pertoneal dialysis is not well known. It is

estimated from touch contamination incidents that the incubation period usually
is 24 to 48 hours. Occasionally, incubation periods may be as short as 6 to 12
hours.
The appearance of the symptoms may be very rapid [70] and develop during
one peritoneal dialysis period.
The incubation period of endogenous infections is not known but probably is
much shorter, than that of exogenous infections.

6.2. Length of symptoms

In most cases of peritonitis the symptoms decrease rapidly after initiation of

therapy and disappear within 2 to 3 days. During this period the cell counts
decrease and bacterial cultures become negative. In the majority of the cases
positive peritoneal cultures are present only for 3 to 4 days [70]. Any prolongation
of symptoms is indicative of a complicated course or a possible organism which
does not respond well to the antibiotics used and requires further investigation.

6.3. Exit site and tunnel infections

Exit site and tunnel infections are rarely symptomatic. Usually they are dis-
416

covered on routine investigation of the exit site or the patient complaining of

some purulent discharge. The exit site is inflamed with a serous or purulent
discharge and sometimes painful infiltrate can be seen. Tunnel infections are
much more difficult to diagnose if they are present without exit site infections and
recently radioactive scanning [142] has been recommended as a diagnostic pro-
cedure. Exit site and tunnel infections may be present for prolonged periods
without leading to peritonitis but they are always a potential danger for the
development of the disease [74].

6.4. Relapse, recurrence or reinfection

These concepts are not well defined in the peritoneal dialysis population.
Relapse is the reappearance of symptoms, the appearance of positive cultures
after cultures have become negative or an increase of polymorphonuclear cells in
the peritoneal dialysis fluid after they have declined. It indicates either inade-
quate treatment or possibly the opening of an abscess cavity which was previously
inaccessible to treatment.
Recurrence is the term used for reappearance of symptoms of infection after
the therapy has been stopped but within a two week period. It indicates probably
either inadequate therapy or the presence of an endogenous focus like exit site or
tunnel infection from which seeding occurred.
Reinfection is a new peritonitis episode beyond the 2 week period either with
the same organism or a different organism. If reinfection happens with the same
organism as before an internal focus should be suspected.

7. Causative organisms of peritonitis

The overwhelming majority of peritonitis episodes are caused by bacteria. While

a small number (4 to 8%) of peritonitis episodes are caused by fungi, most of them
belonging to the species Candida a few of them caused by Torulopsis and some of
them caused by filamentous fungi (Dermatophyton, Mucor, Penicillium, Fu-
sarium, etc.) [75, 81].
Peritonitis caused by viruses, although theoretically possible, has not been
reported.
No protozoan or parasitic causes of peritonitis have been reported as yet.
Table 4(A) reviews the peritonitis episodes among CAPD patients of the
Toronto Western Hospital between October 1977 and December 1983. It is a
fairly stable patient population being on CAPD for an average of17.2 months and
having a peritonitis rate of one episode every 16.1 patient-months.
 417

Table 4(A). Analysis of peritonitis in patients on CAPD Toronto Western Hospital (October 1977-
December 1983)

Study period 75 months

No. of patients 274
No. of patient-months 4710
X month on CAPD 17.2
No. of peritonitis episodes 292
No. of patients with peritonitis 131
Month/peritonitis 16.1

131 patients or 47.8% of the patient population had peritonitis, a total of 292
episodes, with an average rate of 2.3 episodes per patient. Of these patients,
19.7% had a single episode of peritonitis while 12.0% had 2 episodes, but clearly a
small proportion of the patient population (16.1%) accounts for 58.8% of the
peritonitis episodes (Table 4(B».

Table 4(B)

Peritonitis X month on CAPD No. of % of patients % of episodes

episodes patients

0 12.0 05 143 52.2

1 20.5 p>. 54 19.7 18.5
2 19.1 33 12.0 22.6
3
4
31.9 P>.05
33.9
20
14 73}
;:; 16.1
W5}
~~:; 58.8
5 48.5 6
6+ 54.3 4 1.5 8.9

Table 4(C) shows the distribution of organisms isolated from peritonitis episodes
during this period. This distribution which has been observed in all centres shows
that about 2/3 of the organisms are gram positive originating from the skin. About
114 of them are gram negative (faecal) organisms and only 2.7% of them ana-
erobic. Fungi in this population were present in 3.6%, and 3.3% of them turned
out to be culture negative or 'sterile' peritonitis. Table 4(D) shows that a majority
of the episodes were caused by a single organism (89.7%) while a small number of
them were caused by multiple organisms.
418

Table 4(C). Organisms isolated from peritonitis episodes Toronto Western Hospital (October 1977-
December 1983)

No. of Organisms %

Coagulase-negative Staphylococcus 114 33.7

Staphylococcus aureus 52 15.4
Streptococcus a hemolytic 35 10.4
64.0
Streptococcus species 7 2.1
Diphtheroids 6 1.8
Neisseria species 2 0.6

Escherichia coli 25 7.4

Pseudomonas species 17 5.0
Enterococcus 13 3.8
Proteus species 9 2.7
26.1
Enterobacter 7 2.1
Acinetobacter 6 1.8
Klebsiella 5 1.5
Other gram-negatives 6 1.8

Clostridium 2 0.6
Bacteroides 7 2.1

Fungi 12 3.6

Other organisms 2 0.6

Culture-negative 11 3.3

Table 4(D). Number of organisms isolated from peritonitis episodes (October 1977-December 1983)
Toronto Western Hospital

No. of organisms/episode No. of patients %

262 89.7
2 18 6.2
3 9 3.1
4 3 1.0

7.1. Bacteria

7.1.1. Gram-positive organisms

s. epidermidis. Staphylococcus epidermidis peritonitis is the most frequent event.

It is generally a benign form of peritonitis. Its origin is from the skin by the
 419

transluminal route or from an exit site infection periluminally. It responds well to

appropriate antibiotic treatment and usually it is symptomless within 2 to 3 days.
This is a form of peritonitis most suitable for home treatment with oral antibio-
tics. The organisms belong to the coagulase negative Staphylococcus group
(generally called Staphylococcus epidermidis). Table 5 shows a selection of
coagulase negative staphylococci isolated in our hospital which were biotyped
[76] indicating that Staphylococcus epidermidis proper is the leading cause of
peritonitis in these patients. The role of slime formation of this group of organ-
isms in the attachment to the catheter deserves further study [77, 154].

S. aureus. Staphylococcus aureus peritonitis is a much more alarming infection.

Patients who are admitted with this infection usually are hypotensive, some of
them in outright shock and they complain of extensive abdominal pain. Patients
with Staphylococcus aureus peritonitis showing symptoms of toxic shock syn-
drome have been reported [70, 78]. While the infection presents with alarming
symptoms, it usually responds well to antibiotic treatment. The improvement is
much slower than Staphylococcus epidermidis infections. We have found it useful
to treat these patients with a combination of a pencillin type antibiotic (pencillin
or cloxacillin depending on the sensitivity of the organism) and rifampin which
shows synergy against this organism [70].
The infection subsides slower and sometimes residual abscesses are found
[144]. It is important that peritonitis with exit site infections and tunnel infections
will frequently recur and catheter removal may be necessary.

S. viridans. Streptococcus viridans peritonitis is a milder form of peritonitis

though patients often complain of severe pain. The name of this infection is
probably a misnomer. We analyzed a few alpha haemolytic streptococi isolated
from peritonitis episodes as shown in Table 6 indicating that several alpha

Table 5. Coagulase-negative peritonitis in CAPD patients total of91 peritonitis episodes in 68 patients
from 1979 to 1982

Biotypes #Organisms % of total isolates

isolated

Staphylococcus epidermidis 70 77%

Staphylococcus warneri 6 7%
Staphylococcus haemolYllcuS 5 6%
Staphylococcus hominis 4 4%
Staphylococcus capitis 2 2%
Staphylococcus simulans 2 2%
Staphylococcus saprophyticus 1%
Staphylococcus xylosus 1%

Total 91 100%
420

Table 6. Alpha-hemolytic streptococci isolated from peritonitis in CAPD patients

Organism No. of strains

Streptococcus sanguinis II 12
Streptococcus bovis (VAR) 2
Streptococcus anginosus (Constellatus)
Streptococcus MG (Intermedius)
Streptococcus mitis

haemolytic streptococci may be participating in these infections [79]. We assume

that this infection is caused by hematogenous spread as well as direct intraluminal
infection from the oral flora. While it is possible that this infection is caused by
upper respiratory infections preceeding it, antibiotic prophylaxis for this purpose
has not been investigated.

Enterococcus. Enterococcus, while gram-positive clearly is a faecal organism and

indicates a transmural infection. Peritonitis caused by this organism has no
distinguishing features from gram negative peritonitis (see below).

Diphtheroids. Diphtheroids are skin organisms indicating intraluminal or per-

iluminal infections. While some ofthem are quite resistant to antibiotics, most of
them respond to appropriate antibiotic treatment [80].

7.1.2. Gram-negative organisms

Enterobacteria. Gram-negative enterobacteria are an indication of faecal con-
tamination of the peritoneal cavity. While a small number of gram negative
organisms occur on the skin (see Table 3), it is more likely that peritonitis with
these organisms indicates direct feacal contamination. If more than a single gram
negative organism is isolated from the peritoneal fluid, it is a strong indication of a
perforation. They usually respond well to appropriate treatment with amino-
glycosides or cephalosporins.

Pseudomonas. Pseudomonas infections are usually more resistant to treatment,

often cause multiple abscesses in the patient and therefore require careful evalua-
tion [82]. Patients with this infection are usually strongly hypotensive as well on
admission.

Acinetobacter. Acinetobacter [49], while it has no special distinguishing features,

may be an indication of environmental contamination usually from water. Its
treatment does not represent any problems.

Miscellaneous organisms. Single episodes of peritonitis caused by a large number

 421

of various organisms (Hemophilus, Neisseria, Campylobacter etc.) were dis-

cribed showing that most organisms have the capability to cause infections if
inoculated into the peritoneal cavity of CAPD patients.

7.1.3. Anaerobic organisms

Clostridia and bacteroides. Clostridium and Bacteroides species are isolated from
a small percentage of peritoneal fluids and therefore some centres question the
importance of doing anaerobic cultures on peritoneal fluids [83]. While they are
only present in a small number of infections, our experience with infections
containing anaerobic organisms indicates they are very severe infections usually
requiring laparotomy and there is a high propensity for abscess formation [84].
Aggressive surgical management is necessary [85-87] and therefore we consider it
important to culture peritoneal fluids for these organisms.

7.1.4. Mycobacteria
M. tuberculosis. Mycobacterium tuberculosis, a rare organism causing peritonitis
in peritoneal dialysis patients, requires special consideration. It is not a primary
organism causing peritonitis but usually settles on the peritoneum from a distant
site by hematogenous spread. It occurs in patients who have had a previous
infection with this organism and their disease was inadequately treated. It is
therefore a disease to be considered in high risk groups [88]. It is difficult to
diagnose because the organism grows very slowly on artificial media and there-
fore the microbiology laboratory may not be of much help initially. The index of
suspicion should be high if a patient comes to the hospital with pain and cloudy
fluid which is predominantly mononuclear in composition and repeated cultures
do not yield a bacteriological answer [89, 90]. Since the treatment of the disease
requires long term antituberculous chemotherapy and the removal of the cathe-
ter, the establishment of this diagnosis is rather important. If a high index of
suspicion exists, peritoneal biopsy through direct laparotomy or laparoscopy is
indicated [91, 92]. A histological diagnosis with caseous granulomas with or
without the presence of acid fast organisms is an indication for catheter removal
and chemotherapy.
One should consider antituberculous prophylactic chemotherapy of patients
who have a positive tuberculin test [93]. This consideration is even more impor-
tant in view of the fact that many of the patients who are on peritoneal dialysis will
enter a transplant program later and therefore are at high risk for reactivation of
tuberculosis.

M. chelonei. Peritoneal infections with Mycobacterium chelonei have been ob-

served in intermittent peritoneal dialysis units from contamination of dialyzers
from water sources [51, 52].
422

7.2. Fungi

7.2.1. Yeasts
Yeast are the most common organisms causing fungal peritonitis in peritoneal
dialysis patients [75, 81, 94, 95]. They probably enter the peritoneal cavity
intraluminally or periluminally though in a few cases vaginal infection was noted.
The importance of yeast infections of the peritoneal cavity lies in the fact that
they are very difficult to treat with antifungal antibiotics [75, 95]. These antibio-
tics do not show good penetration into the peritoneal cavity and cannot be
administered intraperitoneally because they are very irritating and painful.
5-fluorocytosin, an antifungal agent often used in the treatment of Candida
cystitis is not suitable for treatment alone since resistance emerges against it fairly
rapidly. Treatment with amphotericin, miconazole, ketoconazole - with or with-
out 5-fluorocytosine - while described in a few cases, is erratic, therefore catheter
removal has to be considered early in these patients [75, 95-98]. After catheter
removal the symptoms subside rapidly with or without chemotherapy. If the
patient cannot be considered for catheter removal and antibiotic therapy has to be
attempted, the placing of the patient on intermittent peritoneal dialysis may be
considered on the basis of increasing peritoneal defenses as discussed above.

7.2.2. Filamentous fungi

Filamentous fungi rarely invade the catheter and cause peritoneal infections [75,
81, 95, 99]. Since most filamentous fungi are resistant to antifungal antibotics
early catheter removal has to be considered.
The importance of fungal infections also lies in the fact that most of the fungi
including Candida can colonize the surface of the silastic material of the catheter
and therefore elimination of this infection without catheter removal appears to be
rather futile [153].

7.3. Cryptogenic

7.3.1. 'Sterile' or aseptic peritonitis

This condition is usually due to inappropriate culture procedures or specimens
taken while the patient was on antibiotics. The incidence of sterile peritonitis
varies among units from 2 to 20% [70,100] depending on the methods used in the
laboratory.
Chemical peritonitis which is an aseptic peritonitis has been described early in
the peritoneal dialysis experience [145].

7.3.2. Eosinophilic peritonitis

This is an alarming complication usually observed early after catheter implanta-
tion [101, 120]. It mayor may not be associated with peripheral eosinophilia. It is
 423

not a true peritonitis since there is no causitive organism isolated. Usually these
patients do not have pain or other signs or symptoms of peritonitis, only cloudy
fluid. The condition subsides in a couple of days without further complications
and without therapy. It is assumed to be associated with chemical stimuli leached
from the catheter or the equipment for peritoneal dialysis [1021; Eosinophilia in
the peritoneal fluid may be observed with use of antibiotics or other drugs in
patients who are hypersensitive to these drugs.

7.3.3. Neutrophilic peritonitis

Diarrhea. Neutrophilia in the peritoneal fluid has been observed during diarrhea
in patients without having any bacteria isolated from the peritoneal fluid.

Endotoxin. Endotoxin is also associated with neutrophilia in the peritoneum

[145].

7.3.4. Bloody fluid

Menstuation, ovulation. During menstruation or ovulation, some patients will
observe bloody peritoneal dialysis fluid. While initially this is alarming it has no
consequences and will be present in a few exchanges.

Intraperitoneal hemorrhage. True intraperitoneal hemorrhage, especially in pa-

tients who are on anticoagulants may be an alarming problem. If patients are on
intraperitoneal heparin it should be suspended for a few days. Careful monitoring
of PT and PTT in patients who are on anticoagulants and peritoneal dialysis is
necessary. If the intraperitoneal hemorrhage is very profuse, intraabdominal
hemostasis through surgical intervention may be considered.

8. Treatment of peritonitis

Treatment of peritonitis has to be initiated in the absence of appropriate diagnos-

tic information and therefore certain arbitrary decisions have to be taken on the
appropriateness of the antibiotic treatment based on the considerations discussed
above on causitive organisms.

S.l. Antibiotics

The antibiotics selected for initial treatment should be effective against the most
frequent organisms observed in peritonitis. It is important to know and develop
such an approach jointly with the infectious disease specialist in each hospital
since certain differences in the sensitivity against antibiotics of various organisms
can be observed depending on the hospital.
424

First choice of antibiotics should cover the majority of the organisms present in
these infections. Therefore it should give coverage against most gram positive
organisms as well as gram negative organisms. Most centres use as initial choice
cephalothin or cephazolin in appropriate concentrations. These antibiotics show
a good coverage for Staphylococcus epidermidis and some coverage for Sta-
phylococcus aureus while covering a certain number of gram negatives. If one
wants to cover the more threatening intestinal organisms the addition of an
amino glycoside is justified. If the patient is having hypersensitivity against
cephalosporins, the use of vancomycin in conjunction with aminoglycoside is the
first choice. After the organisms have been identified and an antibiotic sensitivity
is available, adjustments should be made to the therapy.
Penetration of antibiotics from the peritoneal cavity to serum is good and rapid
[103, 104]. It is therefore not necessary in most cases to give an intravenous
loading dose since peritoneal administration of antibiotics will achieve high
enough concentrations in the serum in a few hours.
In Table 7 the initial loading dose as well as the maintenance dose of several
antibiotics used in our unit is shown.
The pharmaceutical literature generally frowns upon a mixture of antibiotics in
the same fluid. This question has been investigated and it has been found that the
commonly used antibiotics do not interact deleteriously in peritoneal dialysis
fluid. Moreover, other additions like heparin and insulin do not interfere with the
antibiotics [105-109].

8.2. Length of treatment

There is no general agreement on how long peritonitis should be treated. In our

hospital we administer antibiotics for 7 days after the last positive culture has
been obtained. Since antibiotic elimination will be slow after cessation of treat-
ment and this adds another two days of effective therapy to it, the length of
treatment in our patients is probably for 10-14 days. If the cultures turn negative
slower or the patients symptoms subside slowly, more prolonged treatment is
necessary. If no clinical improvement and decrease in cell counts is evident after 1
week or 10 days repeat cultures are necessary and a change in antibiotics should
be considered.

8.3. Side effects

Hypersensitivity reactions against antibiotics have been observed in peritoneal

dialysis patients. If antibiotics are used intraperitoneally in these patients, eosin-
ophilia in the peritoneal fluid may be observed. Skin rash may appear in patients
with antibiotic hypersensitivity from peritoneal application only and the use of
such drugs must be discontinued.
 425

Table 7. The following are J.P. antibiotic dosages on CAPD patients

Loading dose Maintenance dose

Tobramycin 1.7mg/kg BW/bag 8mgll

Cephalothin 500mg/l 250mg/l
Ampicillin 500mg/l 50mgll
Cloxacillin 1000mg/i 100mg/i
Penicillin 1 million VII 50,000 VII
Ticarcillin lOOOmgll 100mg/i
SMZ 400 SMZ 25 II
Septra TMP 80 mg/l TMP 5 mg
Clindamycin 300mg/l 50mg/l
Amikacin 125mg/l 25mg/l
Vancomycin 1000mg/i 30mgll
Amphotericin Ba 5mg/l
5-Fluorocytosina lOOmg/1

a Of questionable value.

Aminoglycosides are known to have nephrotoxic and ototoxic effects as well as

vestibular toxicity. In the concentrations used in our hospital (8 mg/l peritoneal
dialysis fluid) we have not observed nephrotoxicity, though evaluation of nephro-
toxicity in the patients with small residual renal functions is difficult. All attempts
should be made to preserve residual functions since the loss of such functions may
necessitate one extra peritoneal dialysis cycle per day [110].
Ototoxicity has been observed in patients where gentamicin was used. To-
bramycin, netilmicin or amikacin are believed to have less ototoxicity or vestibu-
lar toxicity. We have rarely observed vestibular or ototoxicity except in patients
who accidentaly overdose themselves severalfold with intraperitoneal antibio-
tics. Under normal use such side effects are not frequent.
The use of rifampin occasionally results in elevation of liver enzymes or nausea,
necessitating the discontinuation of the drug. In our hospital we use a dose of
600 mg of rifampin per day which if distributed to 150 mg four times a day has less
toxic effect.
Serious side effects of antibiotics may be seen in peritoneal dialysis patients
resulting in psedomembranous enterocolitis, due to Clostridium difficile - a
serious complication [111]. This necessitates the discontinuation of antibiotic ther-
apy and the institution of oral vancomycin or neomycin or the use of cho-
lestyramine. All efforts should be made to diagnose such condition by cultures
and toxin assay so appropriate management can be instituted.

8.4. Peritoneal lavage

Peritoneal lavage has been instituted in the treatment of peritonitis following

426

experience with this [Link] in the surgical field [112]. The reason for peritoneal
lavage in surgical peritonitis is explained by the need of removing detritus and
faecal contamination from the peritoneal cavity. It has been shown that the effect
of lavage may reduce peritoneal defenses and remove necessary phagocytic cells
[64]. Peritoneal lavage with added iodine has been advocated [113] but the use of
such treatment has not been clinically substantiated [114). Studies show that
peritoneal antibiotic treatment using CAPD protocols is clinically efficacious and
less costly than peritoneal lavage [115-118, 146].

B.5. The role of heparin

Heparin addition to the peritoneal dialysis fluid during peritonitis is important.

Since the inflammatory process will result in the entry of large amounts of
fibrinogen into the peritoneal fluid, the inhibition of formation of fibrin is
necessary. In addition, it appears that heparin will reduce subsequent adhesion of
the peritoneal membrane, therefore reducing postinfectious complications [147].

B.6. Treatment protocols

Development of standard treatment protocols for peritonitis are useful for effi-
cient treatment of patients. It is necessary since the appropriate microbiological
diagnosis will not be available for 24 to 72 hours and until that time arbitrary
antibiotic combinations have to be used to cover the most likely pathogens.
At the Toronto Western Hospital, we use the following protocol. The patient is
instructed at the first sign of peritonitis (pain, cloudy fluid, etc.) to drain the fluid
in the abdomen immediately and put the drainage bag aside for later transporta-
tion to the laboratory. Three quick in and out 1 litre flushes are then applied with
very short dwell times. While the effectiveness of these flushes is not established,
it appears that pain is decreased. The next one litre bag is prepared containing
1. 7 mg/kg present body weight of tobramycin, 1 gram/bag of cephalothin and 2000
units of heparin. The dwell time of this bag is 3 hours. This is considered the
loading dose of the patient. The next exchanges for the next 2 days (exchanges 5
to 11) are 1 litre volumes and the dwell times, 3 hours. These bags contain
appropriate maintenance doses of the same antibiotics as well as heparin (to-
bramycin 8 mg/l, cephalothin 250 mg/l, heparin 2000 units/I). Also, insulin doses
are increased for diabetics.
In the first 24 hours (but after the initiation of antibiotic therapy) the tubing is
changed to prevent reinfection of the abdominal cavity.
Appropriate changes in antibiotics are made when culture and sensitivity data
are available.
If the patients' symptoms are mild, oral cephalosporins can be used for treat-
ment on an out-patient basis.
 427

If the patient reports accidental contamination but no peritonitis, they are

instructed to change their tubing then perform 3 flushes with one litre of per-
itoneal dialysis fluid after which they add tobramycin, 1. 7 mg/kg body weight per
2 litre bag and cephalothin 1 gram per 2 litre bag to the next dialysis fluid. They
hold this fluid for a dwell time of 6 hours. After that they start oral cephalosporin
(cephalexin 500 mg p.o. 112 hour before each bag exchange for a total of 10 days.

B.7. Treatment of exit site and tunnel infections

Exit site infections are a major problem of peritoneal dialysis. Their treatment is
not very successful and often it requires removal of the catheter. Most commonly
exit sites are infected with Staphylococcus epidermidis or Staphylococcus aureus
though occasionally Pseudomonas or Proteus infections can be observed. The
exit site is erythematous and elevated showing draining pus or serous fluid but
generally not painful. For tunnel infections sometimes an abscess can be palpated
under the skin along the cannula tract.
In order to establish microbiological diagnosis a culture swab should be taken
carefully from the depth of the exit site, not touching adjoining skin. Since the
organisms present in tunnel and exit site infections are the same as skin organisms
care has to be exercised to avoid contamination with skin organisms.
Treatment of exit site infections can be attempted with local disinfectants or
oral antibiotics. The use of neomycin ointment should be discouraged since its
effectiveness is questionable and may lead to the emergence of resistant organ-
isms. In addition, the ointment forms a crust over the exit site making cleaning
difficult.
With daily cleaning and local care, exit site infections sometimes can be cured.
If the outer cuff (doublecuffed catheters) appears in the exit site or is extruded,
catheter shaving [74, 119] can be attempted.
Exit site infections are frequent. The 1984 report of the National CAPD
Registry of NIH shows that about 40% of the patients develop exit site infections
within the first year and probably about 1/2 of these patients will require catheter
replacement during this period [31].

B.B. Catheter removal

The most common cause for catheter removal is a persistently infected exit site or
tunnel [121].
Catheter replacement can be done on these patients usually one to two weeks
after catheter removal into a different site.
If the catheter has to be removed because of frequent recurrences of peritonitis
with the same organism or other infectious causes, the catheter usually can be
428

replaced after three weeks of termination of succesful treatment of peritonitis.

Other infectious causes for catheter removal are fungal peritonitis, tuberculous
peritonitis, and faecal peritonitis which requires laparotomy, catheter replace-
ment can be considered after the succesful treatment of peritonitis or the com-
plete healing of the operative wound.
If peritonitis is not responding to adequate therapy and a change in appropriate
antibiotics, catheter removal also should be considered.

9. Complications of peritonitis

9.1. Intestinal perforation and diverticulitis

A small prospective study [44] from our unit has shown the role of diverticulosis; a
risk factor for development of faecal peritonitis. It is an alarming complication
when patients who have pre existing diverticulosis develop faecal peritonitis.
Aggressive treatment, early laparotomy and sometimes surgical excision of the
diverticulum will be necessary. Unfortunately, some of the surgical solutions will
require colostomy which will delay the reinstitution of peritoneal dialysis.

9.2. Adhesions, sclerosing peritonitis

As a consequence of peritonitis, fibrous adhesions may develop between the

peritoneal membranes [140, 141]. This is especially frequent as a consequence of
Staphylococcus aureus peritonitis or faecal peritonitis.
Lysis of adhesions at implantation of the peritoneal catheter may be attempted
by blunt dissection but this procedure increases the risk of producing microper-
forations with resulting peritonitis.
An alarming condition has been described in peritoneal dialysis patients called
sclerosing peritonitis [122-124]. This is really the end result of some process in
which the peritoneum develops a thick fibrous membrane making the exchange
of fluid and solutes impossible. It is not clear at the present moment whether this
is due to repeated injuries to the peritoneum like frequent peritonitis or due to
some chemical injury.

9.3. Mortality

It is difficult to establish the accurate mortality due to peritonitis. One report puts
the mortality at 2 to 3% [125]. While this is certainly high it is not surprising since
the patients with peritonitis suffer from an ultimately fatal disease. Many of the
causes of death during peritonitis are not directly attributable to the inflammation
 429

but may have been triggered by the patients hospitalization or the added stress
(myocardial infarction, metabolic imbalance, etc.) [126].

10. Differential diagnostic problems

Various diseases may mimic peritonitis or may be the initiating event which leads
to peritonitis but without treatment of the cause the peritonitis will not improve.

10.1. Constipation

Constipation may lead to diffuse abdominal pain mimicking peritonitis but will
not lead to cloudy fluid or positive cultures. In the absence of the latter, treatment
with antibiotics should not be initiated. If the patient presents with diffuse
abdominal pain but clear fluid, careful history for bowel habits and a radiographic
examination without contrast material should be done for appropriate diagnosis.
Constipation should be solved by mild laxatives, enema or if necessary,
disimpaction.

10.2. Appendicitis

Appendicitis in a patient on peritoneal dialysis may mimic peritonitis producing

even cloudy fluid with an elevated polymorphonuclear cell count due to inflam-
mation of the bowel wall extending to the serosa. Usually the early diagnosis is
difficult and patients are treated for peritonitis delaying the accurate diagnosis
but not resulting in cure. It is important to diagnose appendicitis in a peritoneal
dialysis patient early since the delay of treatment may lead to perforation and
resulting faecal peritonitis.

10.3. Pancreatitis

Acute pancreatitis or an infected pseudocyst may lead to symptoms of peritonitis.

Elevated serum amylase levels are an early indication of such infection. Some-
times amylase may be measured in the peritoneal dialysis fluid though this test is
not reliable.

10.4. Cholecystitis

Cholecystitis may also mimic peritonitis in dialysis patients. Early ultrasound

430

demonstrating the presence of stones may be a diagnostic clue. Sometimes the

cholecystitis may lead to perforation of the gallbladder resulting in true per-
itonitis. A positive blood culture with a gram negative organism may be an early
sign of cholecystitis or cholangitis since positive blood cultures are nonexistent
during true peritonitis.

10.5. Perforated ulcer

Perforation of a gastric or duodenal ulcer may also lead to peritonitis. Usually the
organisms cultured from such peritonitis are gram positive most commonly alpha
haemolytic streptococci. Surgical management of this complication is essential.

11. Prevention of peritonitis

11.1. Sterile connections

Obviously the most sensible approach to peritonitis is the prevention of infectious

events which may lead to the development of the disease. The observance of strict
sterile conditions during connections and disconnections, the use of disinfectants
on all areas exposed to possible contaminations, the wearing of appropriate face
masks, scrubbing etc. is essential. The single most important prophylactic method
for prevention of peritonitis is the development of appropriate training protocols
and the careful training of patients [127].

11.1.1. U. V. box and sterile weld

A device has been developed (Travenol) which performs the spiking of the bag in
a confined space ('U.V. Box') while the spike and the port are subjected to
sterilizing ultraviolet irradiation. After appropriate time the device signals and a
connection is made by the operator in the confined space.
Another company (Dupont) attempted to solve this problem by removing the
spike from the system and replacing it with an electric device which produces a
sterile weld between the bag and the tubing of the patient [128].

11.1.2. O-Z connection

For patients who have had repeated peritonitis episodes and are high risk for
developing others a special connection has been designed. This connection
maintains the spike between dialysis in a disinfectant and makes a connection in
the presence of this disinfectant (Betadine). Improvement of peritonitis episodes
in high risk patients can be expected [127].
 431

11.1.3. Disinfectant in tubing

Several attempts have been reported where different types of connections were
used to avoid intraluminal infections. One recent such publication advocates the
use of a Y tubing which is filled and rinsed with a sodium hypochlorite solution
before connection. Reduction in peritonitis has been reported by this method
[129].

11.1.4. Millipore filter

To prevent intraluminal infections a bacterial retaining device close to the per-
itoneal surface would be useful. To solve this problem a bacteriological filter
(Peridex, Millipore Corporation) has been prepared which fits into the tubing set
used for peritoneal dialysis and which incorporates a bypass valve to facilitate
drainage of the fluid. The use of such a device is probably indicated only in high
risk patients since it has to be replaced regularly and increases costs. Some
reduction in peritonitis from the use of such a device may be expected [130].

11.1.5. Auxiliary devices for the handicapped

Several auxillary devices have been developed for the use of patients handicap-
ped in movement or vision. The development of such devices is essential to
facilitate self care in these patients.

11.2. Antibiotic prophylaxis

Previous clinical studies have examined the use of prophylactic antibiotics pri-
marily on patients on intermittent peritoneal dialysis [131-133]. It is difficult to
draw conclusions from these studies because of the small number of events and
the otherwise low incidence of peritonitis in these patients. We instituted a
double blind prospective study [134] of the use of oral cephalexin twice daily for
peritonitis prophylaxis. We found that this approach did not decrease the number
of infections. Antibiotic prophylaxis to prevent wound infections during catheter
implantations should be used according to the surgical protocol of each hospital.
If perioperative antibiotic prophylaxis is used during surgical implantation the
wound infection rate due to catheter implantations is exceedingly small.

11.3. Patient selection

It is difficult to predict which patients are going to have frequent peritonitis

episodes. Analysis of patient populations and peritonitis rates do not yield
significant differences for prediction [135]. It is quite obvious that the compliant
patients who have reasonable intellectual capabillties to absorb training and have
good family support are doing better on peritoneal dialysis and have lower
432

peritonitis rates. There are very few absolute contraindications to peritoneal

dialysis at present and it is hoped that future analysis of peritonitis episodes in
large populations of peritoneal dialysis patients may result in some clues as far as
these factors are concerned. Immunological assesment for increased risk of
peritonitis at this time is not justified.

11.4. New catheters

The present peritoneal dialysis catheters whether they are single cuffed or double
cuffed models do not show significant differences in peritonitis rates. It is hope
that new catheters will be developed in the future where the implant will form a
complete integrated seal with the skin, therefore reducing exit site infections due
to periluminal penetration of organisms.

12. Evaluation of peritonitis rate

Statistical evaluation of peritonitis incidents and the establishment of comparable

peritonitis rates is a difficult subject. Initially, the simple ratio of peritonitis
episodes over months was used. While this ratio is a useful initial evaluation it is
not capable of expressing the true differences in peritonitis rates. This lies in the
fact that patients enter and leave peritoneal dialysis programs at different times,
and stay in it for different lengths of time. Their combined experience is different
and therefore they are not suitable for simple statistical maneuvers.
The better statistical approaches [136-139] are based on an actuarial approach
analyzing the time elapsed till the first peritonitis episode of the patient and
therefore expressing the probability at different times when peritonitis develops
in peritoneal dialysis patients (see Figures 6 and 7 in Chapter 18). These show that
the probability to develop the first episode of peritonitis is appr. 45% in the first 6
months while appr. 25% of the patients develop exit site infections during this
period. Using such a statistical approach it has been found that there are no
significant differences in peritonitis rates in the two sexes. The age related
incidence is not significantly different though it appears to be better in the older
age groups and certain diseases (for example diabetes) do not represent increased
risk factors. The use of statistical analysis in larger groups of patients may result in
the future in information helping to reduce peritonitis [139).

13. Future considerations

It appears from experience acquired in the last five years that we have reached a
certain plateau in the frequency of peritonitis and a peritonitis rate of one every
 433

two patient years may be acceptable. Further reduction of this peritonitis rate will
require inordinately large efforts on all fronts. New developments in catheter
technology, improved connections, better understanding of patient selection and
training programs, improved diagnostic and therapeutic methods in the manage-
ment of peritonitis, the possibility of some sort of chemical prophylaxis or drug
prophylaxis of peritonitis and understanding of the infectious and immune pro-
cesses are developments eagerly awaited.

References

1. Seligman AM, Frank HA, Fine 1: Treatment of experimental uremia by means of peritoneal
irrigation. 1 Clin Invest 25: 211-219, 1946.
2. Fine 1, Frank HA, Seligman AM: The treatment of acute renal failure by peritoneal irrigation.
Ann Surg 124: 857-878, 1946.
3. Frank HA, Seligman AM, Fine 1: Further experiences with peritoneal irrigation for acute renal
failure. Ann Surg 128: 561-608, 1948.
4. Grollman A, Turner LB, Mclean lA: Intermittent peritoneal lavage in nephrectomized dogs and
its application to the human being. Arch Intern Med 87: 379-390, 1951.
5. Doolan PD, Murphy WP, Wiggins RA, Carter NW, Cooper WC, Watten RH, Alpen EL: An
evaluation of intermittent peritoneal lavage. Am 1 Med 26: 831-844, 1959.
6. Weston RE, Roberts M: Clinical use of stylet-catheter for peritoneal dialysis. Arch Intern Med
115: 659-662, 1965.
7. Maxwell MH, Rockney RE, Kleeman CR, Twiss MR: Peritoneal dialysis I. Technique and
applications. lAMA 170: 917-924, 1929.
8. Cohen SL, Percival A: Prolonged peritoneal dialysis in patients awaiting renal transplantation.
Br Med 1 1: 409-413, 1968.
9. Leigh DA: Peritoneal infections in patients on longterm peritoneal dialysis before and after
human cadaveric renal transplantation. 1 Clin Pathol 22: 539-544, 1969.
10. Palmer RA, Quinton WE, Gray lE: Prolonged peritoneal dialysis for chronic renal failure.
Lancet 1: 700-702, 1964.
11. Tenckhoff H, Schecter H: A bacteriologically safe peritoneal access device. Trans Am Soc Artif
Intern Organs 14: 181-187, 1968.
12. Tenckhof H, Curtis FK: Experience with maintenance peritoneal dialysis in the home. Trans
Am Soc Artif Intern Organs 16: 90-95, 1970.
13. Brewer TE, Caldwell FT, Patterson RM, Flanigan Wl: Indwelling peritoneal (Tenckhoff)
dialysis catheter. Experience with 24 patients. lAMA 219: 1011-1015, 1972.
14. Lankisch PG, Tonnis Hl, Fernandez-Redo E, Girndt 1, Kramer P, Quellhorst E, Scheller F: Use
of Tenckhoff catheter for peritoneal dialysis in terminal renal failure. Br Med 14: 712-713, 1973.
15. Palmer RA: Peritoneal dialysis by indwelling catheter for chronic renal failure 1963-1968. Can
Med Assoc 1 105: 37&-380, 1971.
16. Rae A, Pendray M: Advantages of peritoneal dialysis in chronic renal failure. lAMA 225: 937-
941, 1973.
17. Devine H, Oreopoulos DG, Izatt S, Matthew R, deVeber GA: The permanent Tenckhoff
catheter for chronic peritoneal dialysis. Can Med Assoc 1113: 219-221, 1975.
18. Petrie 11B, lones EOP, Hartley LC1, Olife KP, Clunie C1A: The use of an indwelling peritoneal
catheter in the treatment of chronic renal failure. Med 1 Aust 2: 119-122, 1976.
19. Popovich RP, Moncrief JW, Decherd lB, Bomar lB, Pyle WK: The definition of a novel
portable/wearable equilibrium peritoneal dialysis technique (abstract). Abstr Am Soc Arlif
Intern Organs 5: 64-68, 1976.
434

20. Popovich RP, Moncrief JW, Nolph KD, Ghods AJ, Twardowski ZJ, Pyle WK: Continuous
ambulatory peritoneal dialysis. Ann Intern Med 88: 449-456, 1978.
21. Oreopoulos DG, Robson M, Izatt S, Clayton S, deVeber GA: A simple and safe technique for
continuous ambulatory peritoneal dialysis (Capd). Trans Am Soc Artif Intern Organs 24: 484-
487,1978.
22. Rubin J, Rodgers WA, Taylor HM, Everett ED, Prowant BF, Fruto LV, Nolph KD: Peritonitis
during continuous ambulatory dialysis. Ann Intern Med 92: 7-13, 1980.
23. Oreopoulos DG: Continuous ambulatory peritoneal dialysis in Canada. Can Med Assoc J 120:
16-19, 1979.
24. Oreopoulos DG, Clayton S, Dombros N, Zellerman G, Katirtzoglou A: Experience with
continuous ambulatory peritoneal dialysis (CAPD). Trans Am Soc Artif Intern Organs 25:
95-97, 1979.
25. Fenton SSA, Cattran DC, Ahlen AF, Rutledge P, Ampil M, Dadson J, Locking H, Smith D,
Wilson DR: Initial experiences with continuous ambulatory peritoneal dialysis. Artif Organs 3:
206-209, 1979.
26. Oreopoulos DG, Khanna R, McCready W, Katirtzoglou A, Vas S: Continuous ambulatory
peritoneal dialysis in Canada. Dial Transpl 9: 224-226, 1980.
27. Blagg Cr, Scribner BH: Long-term dialysis; current problems and future prospects. Am J Med
68: 633-635, 1980.
28. Peritoneal dialysis in chronic renal failure (editorial). Lancet 2: 303, 1978.
29. Moncrief JW: Continuous ambulatory peritoneal dialysis. Dial Transpl8: 1077-1078, 1979.
30. Home peritoneal dialysis for end-stage renal disease. Med Lett Drugs Ther 21: 69-70, 1979.
31. National CAPD Registry of the NIH. Report No.2 for January 1, 1981 through June 30, 1983.
Published January, 1984.
32. Dias-Bruxo JA, Walker PJ, Farmes CD, Chandler JT, Holt KL: Continuous cyclic peritoneal
dialysis - The Nalle Clinic experience. In: JDE Price (ed) Peritoneal Dialysis. The State of the
Art. Communications Media for Education, Princeton, NY, 1983, pp 23-25.
33. Hau T, Ahrenholz DH, Simmons RL: Secondary bacterial peritonitis: The biologic basis of
treatment. In: Current problems in Surgery. Vol 16 No. 10 Year Book Medical Publ Inc Chicago
1979.
34. Correia JP, Conn HO: Spontaneous bacterial peritonitis in cirrhosis: endemic or epidemic? Med
Clin North Am 59: 963-981, 1975.
35. Conn HO, Fessel JM: Spontaneous bacterial peritonitis in cirrhoris: variations on a theme.
Medicine (Bait) 50: 161-197, 1971.
36. Bar-Meir S, Conn HO: Spontaneous bacterial peritonitis induced by intraarterial vasopressin
therapy. Gastroenterology 70: 418-421, 1976.
37. Conn HO: Bacterial peritonitis: spontaneous or paracentric? (editorial). Gastroenterology 77:
1145-1146, 1979.
38. Targan SR, Chow A W, Guze LB: Role of anaerobic bacteria in spontaneous peritonitis of
cirrhosis. Report of two cases and review of the literature. Am J Med 62: 397-403, 1977.
39. Stephen CG, Meadows JG, Kerkering TM, Markowitz SW, Nisman RM: Spontaneous per-
itonitis due to Hemophilus infuenzae in an adult. Gastroenterology 77: 1088--1090,1979.
40. Hau T, Hoffman R, Simmons RL: Mechanisms of the adjuvant effect of hemoglobin in
experimental peritonitis. I. In vivo inhibition of peritoneal leukocytosis. Surgery 83: 223-229,
1978.
41. Stewart WK, Anderson DC, Wilson MI: Hazard of peritoneal dialysis: contaminated fluid. Br
Med J 1: 606-607, 1967.
42. Helfrick GB, Pechau BW, Alijani MR, Barnard WF, Rakowski TA, Winchester JF: The
direction of chatheter complications with lateral placement. Perit Dial Bull Suppl3: 2-4, 1983.
43. Schweinburg FB, Seligman AM, Fine J: Transmural migration of intestinal bacteria. A study
based on the use of radioactive Escherichia coli. N Engl J Med 242: 747-751, 1950.
 435

44. Wu G, Khanna R, Vas S, Oreopoulos DG: Is extensive diverticulosis ofthe colon a contraindica-
tion to CAPD. Perit Dial Bull 3: 180-183,1983.
45. Singh MM, Bhargawa AN, Jain KB: Tuberculous peritonitis: An evaluation of pathogenic
mechanisms, diagnostic procedures and therapeutic measures. New Eng J Med 281: 1091-1094,
1969.
46. Khanna R, Oreopoulos DG, Vas SI, McCready W, Dombros N: Fungal peritonitis in patients
undergoing chronic intermittent or continuous peritoneal dialysis. Proc EDTA 17: 291-296,
1980.
47. Coward RA, Gokal R, Mallick NP: Recurrent peritonitis associated with vaginal leak. Perit Dial
Bull 3: 164--165, 1983.
48. Dias-Buxo JA, Burgess P, Walker PJ: Peritoneovaginal fistula-unusual complication of per-
itoneal dialysis. Perit Dial Bull 3: 142-143, 1983.
49. Abrutyn E, Goodhart GL, Roos K, Anderson R, Buxton A: Acinetobacter calcoaceticus
outbreak associated with peritoneal dialysis. Am J Epidemioll07: 328-335, 1978.
50. Mader JT, Reinarz JA: Peritonitis during peritoneal dialysis. The role of the preheating water
bath. J Chron Dis 31: 635-664, 1978.
51. Baud JD, Ward J, Fraser DW, Peteroon NJ, Silcox VA, Good RC, Ostroy PR, Kennedy J:
Peritonitis due to a mycobacterium chelonei like organism associated with intermittent chronic
peritoneal dialysis. J Inf Dis 145: 9-17, 1982.
52. Poisson M, Beromicide V, Falardeau C, Vega C, Morisset R: Mycobacterium chelonei per-
itonitis in a patient undergoing continuous ambulatory peritoneal dialysis (CAPD). Perit Dial
Bull 3: 86-88, 1983.
53. Majno G, Palade GE: Studies on inflammation. I. The effect of histamine and serotinin on
vascular permeability: an electron microscopic study. J Biophys Cytolll: 571-600,1961.
54. Ellis H: The cause and prevention of post operative intraperitoneal adhesions. Surg Gynecol
Obstet 133: 497-511, 1971.
55. Myhre-Jensen 0, Larsen SB, Astrup T: Fibrinolytic activity in serosal and synovial membrane.
Arch Pathol88: 623-630,1969.
56. Porter JM, McGregor FH, Mullen DC, Silver D: Fibrinolytic activity of mesothelial surfaces.
Surg Forum 20: 80-82, 1969.
57. Ellis H: The aetiology of post-operative abdominal adhesions. An experimental study. Br J Surg
50: 10-16, 1962.
58. Buckman RF, Woods M, Sargent L, Gervin AS: A unifying pathogenetic mechanism in the
etiology of intraperitoneal adhesions. J Surg Res 20: 1-5, 1976.
59. Gervain AS, Puckett CL, Silver D: Serosal hypofibrinolysis. A cause of post-operative adhe-
sions. Am J Surg 125: 80-88, 1973.
60. Hau T, Payne WD, Simmons RL: Fibrinolytic activity of the peritoneum during experimental
peritonitis. Surg Gynecol Obstet 148: 415-418, 1979.
61. Weinstein MP, lannini PB, Stratton CW, Eickhoff TL: Spontaneous bacterial peritonitis. A
review of 28 cases with emphasis on improved survival and factors influencing prognosis. Am J
Med 64: 592-598, 1978.
62. Gilmour J, Tymiansky R, Pierratos A, Vas S, Klein M, Khanna R, Digenis D, Cuff S,
Oreopoulos DG: Changes in some inflammatory proteins during peritonitis in CAPD patients.
Perit Dial Bull 3: 201-204, 1983.
63. Verbough HA, Keane WF, Hoidal JR, Freiberg MR, Elliott GR, Peterson PK: Peritoneal
macrophages and opsonins: Antibacterial defense in patients undergoing chronic peritoneal
dialysis. J Inf Dis 147: 1018-1029, 1983.
64. Duwe A, Vas SI, Weatherhead JW: Effect of composition of peritoneal dialysis fluid on
chemiluminescence, phagocytosis and bactericidal activity in vitro. Infec Immun 33: 130-135,
1981.
65. Collart F, Tielemaus C, Schandene L, Dupont E, Wybrau Y, Dratwe M: CAPD and cellular
immunity: No different than hemodialysis patients. Perit Dial Bull 3: 163-164, 1983.
436

66. Giaccino F, Alloatti S, Guarello F, Coppo R, Pellerey M, Piccoli G: The influence of peritoneal
dialysis on cellular immunity. Perit Dial Bull 2: 165-168,1982.
67. Vas SI, Low DE, Layne S, Khanna R, Dombros N: Microbiological diagnostic approach to
peritonitis in CAPD patients (1981) in Peritoneal Dialysis. Ed: Atkins RC et al. Churchill
Livingstone, Edinburgh. 269-271, 1981.
68. Vas SI: Peritoneal fluid cultures remain positive for days. Perit Dial Bull 2: 144,.1982.
69. Swartz RD, Campbell DA, Stone D, Dickinson C: Recurrent polymicrobial peritonitis from a
gynecological source as a complication of CAPD. Perit Dial Bull 3: 32-33,1983.
70. Vas SI: Microbiologic aspects of chronic ambulatory peritoneal dialysis. Kidney International
23: 83-92, 1983.
71. Dobbelstein H: Immune system in uremia. Nephron 17: 409-414, 1976.
72. Montgomery YZ, Kalmanson GR, Guze LB: Renal failure and infection. Medicine 47: 1-32,
1968.
73. Vas SI: Peritonitis during CAPD. A mixed bag. Perit Dial Buill: 47-49, 1981.
74. Nichols WK, Nolph KD: A technique for managing exit site and cuff infection in Tenckhoff
catheters. Perit Dial Bull Suppl 3: S4-S5, 1983.
75. Khanna R, McNeely DJ, Oreopoulos DG, Vas SI, McCready W: Treating fungal infections:
Fungal peritonitis in CAPD. Br Med J 280: 1147-1148, 1980.
76. Kloos WE, Schleifer KH: Simplified scheme for the routine identification of human staphylococ-
cus species. J Clin Microbioll: 82-88, 1975.
77. Peter G, Locci R, Pulverer G: Adherence and growth of coagulase negative staphylococci on
surfaces of intravenous catheters. J Infect Dis 146: 479-482, 1982.
78. Gregory MC, Duffy DP: Toxic shock following staphylococcal peritonitis. Clin Nephrol20: 101-
104,1983.
79. Facklam RR: Physiological differentiation ofviridans streptococci. J Clin Microbiol5: 184--201,
1977.
80. Pierard D, Lauwers S, Monton MC, Sennesael J, Verbeelen D: Group JK Corynebacterium
peritonitis in a patient undergoing continuyous ambulatory peritoneal dialysis. J Clin Microbiol
18: 1011-1014, 1983.
81. Arfania D, Everett ED, Nolph K, Rubin J: Uncommon causes of peritonitis in patients
undergoing peritoneal dialysis. Arch Int Med 141: 61--64, 1981.
82. Kolmos HJ, Anderson KEH: Peritonitis with Pseudomonas aeruginosa in hospitalized patients
treated with peritoneal dialysis. Scand J Infect Dis 11: 207-210, 1979.
83. Matthews P: Primary anaerobic peritonitis. Br Med J 2: 903-904, 1979.
84. Onderdonk AB, Bartlett JG, Louie T, Sullivan-Seigler N, Gorbach SL: Microbial synergy in
experimental intra-abdominal abscess. Infect Immun 13: 22-26, 1976.
85. Simkin EP, Wright FK: Perforating injuries of the bowel complicating peritoneal catheter
insertion. Lancet 1: 64--66, 1968.
86. Rubin J, Oreopoulos DG, Lio IT, Mathews R, deVeber GA: Management of peritonitis and
bowel perforation during chronic peritoneal dialysis. Nephron 16: 220-225, 1976.
87. Wu G: Review of peritonitis episodes that caused interruption of CAPD. Perit Dial Bull Suppl
Vol 3: S11-S13, 1983.
88. Sasaki S, Aliba T, Suenaga M, Tornura S, Yoobiyama N, Nakagawa S, Shoji T, Sasavka T,
Takenchi J: Ten years survey of dialysis associated tuberculosis. Nephron 24: 141-145, 1979.
89. O'Conner J, MacCormick M: Tuberculous peritonitis in patients on CAPD: the importance of
lymphocytosis in the peritoneal fluid. Perit Dial Buill: 106, 1981.
90. Morford DW: High index of suspicion for tuberculous peritonitis in CAPD patients. Perit Dial
Bull 2: 189-190, 1982.
91. Dineen P, Homan WP, Grafe WR: Tuberculous peritonitis: 43 years experience in diagnosis and
treatment. Am Surg 184: 712-717, 1976.
92. Wolfe JHN, Behn AR, Jackson BT: tuberculom peritonitis and role of diagnostic laparoscopy.
Lancet 1: 852-853, 1978.
 437

93. Vas SI: Editorial comment. Perit Dial Bull 2: 190, 1982.
94. Bayer AS, Blumenkrantz MY, Montgomerie JZ, Galpin JE, Coburn JW, Gruze LB: Candida
peritonitis. Am J Med 61: 832-840, 1976.
95. Kerr CM, Perfect JR, Craven PC, Jorgensen JH, Drutz DJ, Shelburne JD, Gallis HA, Gutman
RA: Fungal peritonitis in patients on continuous ambulatory peritoneal dialysis. Ann Int Med
99: 334-337, 1983.
96. Holdsworth SR, Aktkins RC, Scott DF, Jackson R: Management of Candida peritonitis by
prolonged peritoneal lavage containing 5-f1uorocytosine. Clin Nephrol4: 157-159, 1975.
97: Lempert KD, Jones JM: Flucytosine - miconazole treatment of Candida peritonitis: Its use
during continuous ambulatory peritoneal dialysis. Arch Int Med 142: 577-578, 1982.
98. Chapman JR, Warnoch DW: Ketoconazole and fungal CAPD peritonitis. Lancet II: 510--511,
1983.
99. Pearson JG, McKinney TD, Stone WJ: Penicillium peritonitis in a CAPD patient. Perit Dial Bull
3: 20--21, 1983.
100. NIH CAPD Patient Registry Report No. 83-1. Published April 1983.
101. Steiner R: Clinical observations on the phathogenesis of peritoneal dialysate eosinophilia. Perit
Dial Bull 2: 118-119,1982.
102. Verger C, Berry JP, Galle P, Lavergue A, Hoang C, LeCharpentier Y: Foreign material
inclusions in the peritoneum of CAPD patients: a study with X-ray microanalysis. Perit Dial Bull
2: 138-139, 1982.
103. Williams P, Khanna R, Simpson H, Vas SI: Tobramycin blood levels of CAPD patients during
peritonitis. Perit Dial Bull 2: 48, 1982.
104. Manuel MA, Paton TW, Cornish WR: Drugs and peritoneal dialysis. Perit Dial BuIl3: 117-125,
1983.
105. Sewell DL, Golper TA: Stability of antimicrobial agents in peritoneal dialysate. Antimicrob
Agents Chemother 21: 528-529, 1982.
106. SeweIl DL, Golper TA, Brown SD, Nelson E, Knower M, Kimbrough RC: Stability of single
and combination antimicrobial agents in various peritoneal dialysates in the presence of insulin
and heparin. Am J Kidney Dis 3: 209-212, 1983.
107. Rubin J, Humphries J, Smith G, Bower J: Antibiotic activity in peritoneal dialysate. Am J
Kidney Dis 3: 205-208, 1983.
108. Bunke CM, Aronoff GR, Luft C: Pharmacokinetics of common antibiotics used in continuous
ambulatory peritoneal dialysis. Amer J Kidney Dis 3: 114-117, 1983.
109. Vas SI: Letter. Perit Dial BuIll: 67, 1981.
110. Gokal R, Vas SI: Risk of tobramycin use in CAPD patients with peritonitis. Perit Dial Bull 2:
139-141, 1982.
111. Silva J, Fekety R: Clostridia and antimicrobial entocolitis. Am Rev Med 32: 327-333, 1981.
112. Antibiotic lavage for peritonitis. Editorial (1979) Brit Med Jour Sept 22,1979,691-692.
113. Stephen RL, Kablitz C, Kitahara M, Welson JA, Duffin DP, Kolff WJ: Peritoneal dialysis:
peritonitis: saline iodine flush. Dial Transpl8: 584-595, 1979.
114. Nolph KD: In: cOIitinuous Ambulatory Peritoneal Dialysis, ed. by Legrain M. Excerpta
Medica, New York, p 272,1980.
115. Digenis GE, Khanna R, Pierratos A, Vas S: Morbidity and mortality after treatment of
peritonitis with prolonged exchanges and intraperitoneal antibiotics. Perit Dial Bull 2: 45-46,
1982.
116. Cantaluppi A, Scalamogna A, Guerra L, Graziani G, Ponticelli C: Treatment of peritonitis in
patients on CAPD. Perit Dial BuIl2: 142,1982.
117. DeGrocF, RottembourgJ, Jacq D, JaslierV, N'Guyen J, Legrain M: Les peritonites au coursde
la dialyse peritoneal continue ambulatoire. Traitment par lavage ou non? Etude prospective,
Nephrologie (SWZ) 4: 24-27, 1983.
118. De TremontJF, Khissi H, ThomasD, TolainM, Lawrence G, Coevoet B, Fournier A, OrficaJ:
438

Traitment des peritonites ou dialyse peritoneal. Comparison entre lavage continue avec machine
et lavage intermittent par quatre sac/jour de DPCA. Pathol Bioi (Paris) 31: 544-547, 1983.
119. Helfrich GB, Winchester JF: Shaving of external cuff or peritoneal catheter. Perit Dial Bull 2:
183,1982.
120. Digenis GE, Khanna R, Pantalony D: Eosinophylia after implantation of the peritoneal cathe-
ter. Perit Dial Bull 2: 98-99, 1982.
121. Vas SI: Indications for removal of peritoneal catheter. Perit Dial Buill: 145-146, 1981.
122. Gandhi VC, Humayun HM, IngTS, DaugirdasJT, JablokowVR, Iwantsuki S, Geis WP, Hano
JE: Sclerotic thickening of the peritoneal membrane in maintenance peritoneal dialysis patients.
Arch Int Med 140: 1201-1203, 1980.
123. Schmidt RW, Blumenkrantz M: Peritoneal sclerosis. A sword of Democles for peritoneal
dialysis. Arch Int Med 141: 1265-1267, 1981.
124. Sclerosing Peritonitis, Letters to the editor (1983): Lancet July 9, August 13, September 3,
September 24, November 5, 1983.
125. Fenton SSA: Peritonitis related deaths among CAPD patients. Perit Dial Bull Suppl Vol3 No 3:
S9-S11, 1983.
126. Wu G: Cardiovascular deaths among CAPD patients. Perit Dial Bull Suppl Vol 3 No 3: S23-S26,
1983.
127. Oreopoulos D, Vas S, Khanna R: Prevention of peritonitis during continuous ambulatory
peritoneal dialysis. Perit Dial Bull Suppl Vol 3 No 3: Sl8-S20, 1983.
128. Hamilton RW, Disher BA, Dillingham GA, Nicholas AF: The sterile weld a new method for
connection in continuous ambulatory peritoneal dialysis. Perit Dial Bull Suppl Vol 3 No 4:
S8-SlO, 1983.
129. Buonchristiani U, Cozzari M, Quintiliani G, Carobi C: Abatement of exogenous peritonitis
using the Perugia CAPD system. Dial Transplant 12: 14-25, 1983.
130. Ash SR, Hoswell R, Heefer EM, Bloch R: Effect of the Peridex filter on peritonitis rates in a
CAPD population. Perit Dial Bull 3: 89-93, 1983.
131. Eremin J, Marshall VC: The place of prophylactic antibiotic in peritoneal dialysis. Aust Ann
Med 18: 264-266, 1969.
132. Sharma BK, Smith EC, Rodriguez H, Pillay UKG, Gandhi VC, Dunea G: Trial of oral
neomycin during peritoneal dialysis. Am J Med Sci 262: 175-178, 1971.
133. Axelrod J, Meyers BR, Hirschman SZ, Stein R: Prophylaxis with cephalothin in peritoneal
dialysis. Arch Intern 132: 368-371, 1973.
134. Low DE, Vas SI, Oreopoulos DG, Manuel RA, Saiphoo CS, Finer C, Dombros N: Randomized
Clinical trial of prophylactic cephalexin in CAPD. Lancet 2: 753-754, 1980.
135. Corey PN, Steele C: Risk factors associated with time to first infection and time to failure on
CAPD. Perit Dial Bull Suppl Vol 3 No 3: Sl4-S17, 1983.
136. D'Apice AJF, Atkins RC: Analysis of peritoneal dialysis data. Peritoneal Dialysis Edited:
Atkins RC, Thomas NM, Farrell PC, Churchill Livingstone, Edinburgh, pp 440-444, 1981.
137. Randerson DH, Farrell PC: Analysis of peritonitis data in CAPD 2nd Int. Symposium in
Peritoneal Dialysis, Berlin, p 52, 1981.
138. Corey P: An approach to the statistical analysis of peritonitis data from patients on CAPD. Perit
Dial Bull Suppl Vol 1 No 6: S29-S32, 1981.
139. Pierratos A, Amair P, Corey P, Vas SI, Khanna R, Oropoulos DG: Statistical analysis of the
incidence of peritonitis in continuous ambulatory peritoneal dialysis. Perit Dial Bull 2: 32-36,
1982.
140. Ryan GB, Grobety J, Majno G: Post-operative peritoneal adhesions. A study of the mecha-
nisms. Am J Pathol 65: 117: 117-138, 1971.
141. Mion CM, Boen ST, Scribner P: Analysis of factors responsible for the formation of adhesions
during chronic peritoneal dialysis. Am J Med Sci 250: 675-679, 1965.
142. Steiner RW, Kipper S, Savoia MC, Witztum KF: Identification of peritoneal dialysis catheter
 439

tunnel infection by scanning with Indium-111 labelled leukocytes. Amm Int Med 99: 44-45, 1983.
143. Williams P, Pantalony D, Vas SI, Khanna R, Oreopoulos DG: The value of dialysate cell count
in the diagnosis of peritonitis in patients on continuous ambulatory peritoneal dialysis. Perit Dial
Buill: 59-62, 1981.
144. Kapral FA, Godwin JR, Dye ES: Formation of intraperitoneal abscesses by Staphylococcus
aureus. Infec Immun 30: 204-211, 1980.
145. Karanicolas S, Oreopoulos DG, Frath SH, Shiminer A, Manning RF, Sepp H, deVeber GA,
Darby T: Epidemic of aseptic peritonitis caused by endotoxin during chronic peritoneal dialysis.
N Engl J Med 296: 1336-1337,1972.
146. Williams P, Khanna R, Vas S, Layne S, Pantalony D, Oreopoulos DG: Treatment of peritonitis
in patients on CAPD: To lavage or not. Perit Dial Buill: 14-17, 1980.
147. O'Leary JP, Malik FS, Donahoe RR, Johnston AD: The effects of a minidose of heparin on
peritonitis in rats. Surg Gynecol Obstet 148: 571-575, 1979.
148. Cotran RS, Karnowsky MJ: Ultrastructural studies on the permeability of the mesothelium to
horse radish peroxidase. J Cell Bioi 37: 123-137, 1968.
149. MacCallum WG: On the mechanism of absorption of granular materials from the peritoneum.
Bull Johns Hopkins Hosp 14: 105-110, 1903.
150. Casley-Smith JR: An electron microscopical study of the passage of ions through the endo-
thelium of lymphatic and blood capillaries, and through the mesothelium. Quart J Exp Physiol
52: 105-113, 1967.
151. Courtice FC, Simmonds WJ: Physiological significance oflymph drainage of the serous cavities
and lungs. Physiol Rev 34: 419-448, 1954.
152. Lorber B, Swenson RM: The bacteriology of intraabdominal infections. Surg Clin North Am 55:
1349-1354, 1975.
153. McNeely D, Vas SI, Dombros N, Oreopoulos DG: Fusarium peritonitis: an uncommon com-
plication. Perit Dial Buill: 94-96, 1981.
154. Marrie TJ, Noble MA, Costerton JW: Examination of the morphology of bacteria adhering to
peritoneal dialysis catheters by scanning and transmission electron microscopy. Jour Clin
Microbiol18: 1388-1398, 1983.
14. Complications of peritoneal dialysis other than
peritonitis

RAMESH KHANNA and DIMITRIOS G. OREOPOULOS

A. ACUTE AND INTERMITTENT PERITONEAL DIALYSIS

1. Complications of peritoneal dialysis

Ever since Ganter [1] published the first experience in 1923, much progress has
been made in the techniques and technology of peritoneal dialysis [2-6]. De-
velopments in these five decades gave rise to the three forms of peritoneal dialysis
known today, i.e. (a) acute, (b) chronic intermittent, and (c) continuous ambula-
tory peritoneal dialysis. Patients with a life-threatening disease who receive these
treatments are exposed to multiple risks, some of which are serious. They are
prone to complications of the uremia itself, to those arising from the dialysis
procedure and to those associated with a primary illness that may have produced
the renal failure. Patients maintained on periodic dialysis for a long time face the
complications of uremia in a modified form, and also must accommodate to the
long-term effects of dialysis itself. Finally, these patients are also prone to the
other illnesses which are a part of everyday life. Because of the variety of these
risks, the physician who takes responsibility for patients on long-term dialysis
must confront multifactorial problems.
Although peritoneal dialysis is assumed to be safe, it may be associated with
significant morbidity and mortality [7-11]. Most of the complications can be
prevented and, when they do develop, can be brought under control by appropri-
ate measures. Units with an experienced staff can handle the technical problems
and rarely are forced to discontinue dialysis.
Because the complications observed in acute, chronic intermittent or continu-
ous ambulatory peritoneal dialysis generally are specific to each method, they will
be described separately.

2. Complications of acute peritoneal dialysis

The problems encountered in acute peritoneal dialysis are mechanical, infectious

and medical (Table 1). Infectious complications are examined in detail elsewhere
in this volume and hence they will be mentioned in this chapter only in passing.
442

Table 1. Complications of acute peritoneal dialysis

Mechanical complications:

Abdominal pain
Bleeding
Dialysate leak
Poor drainage
a. loss of siphon effect
b. one way obstruction
c. wrong placement of catheter
d. fluid loculation
Perforation or laceration of internal organs
Intraperitoneal loss of peritoneal catheter

2.1. Mechanical complications

Mechanical complications are quite frequent, especially when inexperienced staff

members carry out catheter insertions and dialysis-bag exchanges. Most of these
problems are catheter-related.

2.2. Abdominal pain

This complication may be encountered in as many as 75% of acute dialyses.

Vaamonde reported that 73% of his acute dialyses were associated with pain [12].
This is not surprising because in his series persons working outside a specialized
renal unit bore the major responsibility for peritoneal dialysis. In another series
[13], 56% of dialyses were associated with pain. This is most common during
inflow and outflow of dialysate. Inflow pain, which usually is crampy, starts
immediately and may increase in intensity as the flow continues, but tends to
subside gradually during the dwell time. Inflow pain is believed to be due to
dialysate acidity (pH 5.5).
Dialysate dextrose concentration, if hypertonic, also may irritate the per-
itoneum [14]. In patients who have persistent pain, we have adjusted dialysate pH
by adding sodium bicarbonate, with some success. Some patients benefit by the
addition of 2% xylocaine solution to the dialysate, in amounts ranging from 3 to
5 mlll of dialysate. Overdistention of a small abdomen is another cause of periodic
pain. Usually this pain does not appear until after the full volume of dialysate has
gone in, and is relieved with initiation of drainage. In such a situation, the
physician should attempt to carry out the dialysis with a smaller volume of fluid.
Outflow pain, which is difficult to explain, is much less common. It may be due
to entrapment of omentum in the catheter during the siphoning action of fluid
 443

drainage. Irrigation of the catheter with saline generally relieves this pain.
Constant pain during the dialysis is usually related to the catheter, which by
impingement on intra-abdominal organs often produces continuous rectal or low
back pain. This complaint calls for an adjustment in catheter position.
These patients frequently complain of shoulder pain, referred from the di-
aphragm, or costal margin pain, referred from rectus muscle. These pains are
usually not severe and do not signify any intra-abdominal pathology. The collec-
tion of free air under the diaphragm towards the end of dialysis may produce
shoulder or intrascapular pain when the patient assumes the upright position.
This may be due to inadequate deaeration of reverse osmosis (RO) water used
during the dialysis or accidental infusion of air during the procedure. A small
volume of air will be absorbed quickly, usually will not produce symptoms and
may go unnoticed. Large volumes of air may cause symptoms that may persist for
days; hence the discovery of air in the peritoneal cavity does not necessarily mean
a bowel perforation. Svetvilae [15] showed that peritoneal air is not completely
absorbed until seven days after the termination of dialysis. If the symptoms are
severe and intolerable to the patient, the air must be removed by filling the
peritoneal cavity with 1 to 21 of dialysate. The patient is then manipulated into the
Trendelenburg or knee-chest position, so as to expel the air out through the
catheter [16].

2.3. Bleeding

In 30-32% of dialyses, bloody effluent will be noted from the first exchange after
catheter implantation [12, 13]. This bleeding (usually minor) comes from the small
vessels in the abdominal wall. If the catheter is inserted in the midline, even this
minor bleeding can be kept to a minimum. After three to four exchanges,
bleeding usually subsides unless the procedure has damaged a major vessel or the
patient has a bleeding disorder. Pressure applied over the catheter insertion site
usually controls minor bleeding. Severe hemorrhage is uncommon [17-18] but, if
not recognized early, it can be fatal [19]. Occasionally, a transfusion of fresh blood
will stop the bleeding. On rare occasions, either aorta or mesenteric artery may
be perforated accidentally, but removal of the catheter may be all that is neces-
sary. Significant hemorrhage from the site of open renal biopsy [12, 18] may be
related to clot disruption by the dialysate. If the bleeding is copious, it may
obstruct the catheter; in such a situation it is common practice to add 1000 units of
heparin to each liter of dialysate to minimize the risk of catheter obstruction.
Because heparin is negatively charged, very little is absorbed from the peritoneal
surface and hence this practice does not increase the risk of systemic bleeding
[20]. Serious hemorrhage should not occur from a recent incision when peritoneal
dialysis is carried out in the post-operative period [21, 22]. Some workers [23, 24]
have recommended the infusion of streptokinase (250000 units intra-
444

peritoneally), but we have found it ineffective; in addition, this practice involves

the risk of reaction to a foreign protein.

2.4. Dialysate leak

This common complication of acute dialysis is encountered in 14% to 36% of

patients in various series [12, 13, 19]. Frequent manipulation of the catheter to
improve drainage promotes dialysate leak from the catheter exit site. It may also
occur when the catheter is not properly secured to the skin. The risk of external
leak is higher in elderly or debilitated patients who have lax abdominal walls.
Such leaks can be minimized by inserting a tightly fitting catheter through a small
incision. The presence of a large mass, such as a polycystic kidney(s), will raise
the intra-abdominal pressure to high levels and promote an external dialysate
leak when the standard 2-1 volume is instilled. In such a situation, it may be wise to
employ smaller volumes and a shorter dwell time.
Fluid may extravasate into the abdominal wall, particularly in patients who
have had previous abdominal surgery or multiple catheter insertions [12, 21, 25].
This usually results from tears in the peritoneum or represents an infusion of
dialysate into the potential space between the layers of abdominal wall. If
undetected, this dissection of fluid may spread to the thigh, scrotum or labia or,
rarely, up towards the chest wall. Although these patients may experience
discomfort, the problem is not serious. The collected fluid is absorbed in two or
three days once the dialysis is terminated. The operator should wait for 8 to 10
days before attempting dialysis again, in order to give any peritoneal tear time to
heal. If such patients need dialysis in the interval, they should have hemodialysis
through acute subclavian vascular access [26]. When peritoneal dialysis is at-
tempted in a patient who has had recent abdominal surgery, the fluid may leak
from the incision or through drains [25, 27, 28]. In addition, wound dehiscence is
a real hazard in such a patient [12]. It is exceedingly uncommon, however, for
dialysis fluid to enter the pleural cavity; however, isolated reports of such
misadventures have appeared in the literature [19, 29-35 B]. In such cases,
peritoneal dialysis is discontinued and the patients are switched to hemodialysis.
Acute hydrothorax results from a traumatic or a congenital defect in the dia-
phragm.

2.5. Poor drainage

Inadequate drainage, a frequent problem during acute dialysis, is due to one or

more of the following factors [12, 13, 19, 21-27]: loss of siphon effect, one-way
obstruction, and incorrect placement of the catheter. The siphon effect is lost
when an air bubble is trapped in the drainage system. Such bubbles can be
 445

expelled from the system by running in a small quantity of dialysate. The

interruption of siphonage can be avoided by leaving behind a liter or two of
dialysate as a reservoir at the end of the drainage. This reservoir fluid is drained
off completely only at the end of dialysis.
One-way catheter obstruction, which generally implies obstruction to outflow
of fluid, may have multiple causes. Fibrin or blood clots may be trapped in the
catheter and block the terminal holes, especially when dialysis is complicated by
major hemorrhage or peritonitis. Patients who tend to form fibrin clots, or those
in whom bleeding is prolonged, may require the addition of heparin to the
dialysate. If this measure fails to relieve the obstruction, the catheter is irrigated
with saline in an attempt to expel the clot, fibrin or omentum stuck in the
catheter. If poor drainage still persists, the operator should insert a new catheter.
Poor outflow may also reflect extrinsic pressure on the catheter from adjacent
organs such as sigmoid full of feces. For this reason, a cleansing enema should be
given before insertion of the catheter. Pressure of feces in the colon can easily be
verified on a flat plate of the abdomen. If this cause is suspected on clinical
grounds, the patient should receive a high cleansing tap-water or oil-retention
enema. On rare occasions, an unrecognized full bladder may cause similar
problems.
Occasionally, placement of the catheter in front of the peritoneum may cause
poor drainage [17]. In his series, Vaamonde [12] reported two patients whose
dialysis was unsuccessful because the catheter had lodged in the abdominal wall,
and at autopsy it was shown that the catheter had not entered the peritoneal
cavity. In such a situation, continued infusion of dialysate causes further dissec-
tion of the abdominal wall [27] and this fluid is not available for drainage. If
necessary, the catheter can be inserted again at another site after a few hours.
Loculation of fluid, another cause of poor drainage [2,17,19,36], is encoun-
tered in patients who have had previous intra-abdominal surgery or peritonitis.
Such loculation not only diminishes the surface area available for dialysis but may
seriously reduce ultrafiltration capacity. Relocation of the catheter may improve
outflow, but frequently it is necessary to terminate dialysis because of loss of
efficiency. Then the patient may have to be transferred to hemodialysis.

2.6. Perforation or laceration of internal organs

Although uncommon, this complication is feared by all who carry out acute
catheter insertion with a stylet, especially beginners. Every peritoneal dialysis
unit will have encountered this complication at some time; the incidence is low,
varying between 0.5% and 1.3% [37-39]. Organs which have been perforated or
lacerated included the bowel, bladder, liver, a polycystic kidney, aorta, mes-
eneric artery and hernial sac [18, 19,37-47]. Perforation of the gut is most likely to
occur in those who have had previous abdominal surgery with intra-abdominal
446

adhesions or previous peritoneal dialysis complicated by peritonitis [38, 48].

Adhesions may fix segments of bowel to the abdominal wall or to each other,
producing fixed loops which are held in the path of the advancing catheter during
insertion.
Patients with abdominal distention due to paralytic ileus or bowel obstruction
also may be predisposed to bowel perforation [38, 42]. Those who are uncon-
scious, cachectic or heavily sedated are also at high risk [27, 37, 48, 49]. Clinical
evidence of bowel perforation includes sudden, sharp or severe abdominal pain
followed by watery diarrhea, and poor drainage of dialysate, which may be
cloudy, foul-smelling or mixed with fecal material. Such a situation may require
only prompt removal of the catheter, and the perforation seals off completely in
about 12 h [37].
Many treatments have been proposed for bowel perforation, and reputable
physicians currently are using opposing methods. Simkin and Rubin [38, 50] have
recommended insertion of the catheter at another site and continuation of
dialysis, while Maher and Denovales [19, 44] have suggested hemodialysis in such
a case. Generally, perforation seals off gradually, although it may be ac-
companied by localized evidence of peritonitis. Many workers recommend sys-
temic and/or intraperitoneal antibiotics [37, 50], and others undertake surgical
exploration in some cases [39]. Henderson described five cases of perforation,
three managed conservatively and two by surgical exploration. Two of the three
patients managed conservatively responded, and the other three died [37]. Blad-
der perforation is uncommon [19, 40, 45, 51], but several authors [18, 44] have
reported isolated experience. After bladder perforation, the patient will com-
plain of bladder distention and urgency. Current treatment consists of bladder
drainage with a Foley catheter, reinsertion of the PD catheter at a new site and
continuation of dialysis [18, 27]. Careful preparation before catheter insertion -
i.e., emptying the bladder completely and infusion of 1 to 21/dialysate intraperi-
toneally with a large-bore needle - should eliminate this complication [12].

2.7. Intraperitoneal loss of peritoneal catheter

Loss of part or all of the catheter has been reported [12, 37,3952-54] following its
manipulation with the trocar in place. Its distal end may be amputated after intra-
abdominal kinking of the catheter, followed by manipulation. However, the
presence of broken catheters within the abdominal cavity does not cause symp-
toms or ill effects. During laparoscopy, broken catheters have been found lying
freely in the peritoneal cavity without causing a peritoneal reaction [53, 54] or
have been found walled off by mesentery without an inflammatory reaction [52].
On routine postmortem examination, Stein [55] discovered such a catheter in a
patient who had had previous peritoneal dialysis. Exploration to retrieve the
catheter is unnecessary because laparotomy is more hazardous than leaving the
catheter in place in a severely ill patient [12]. The incidence of catheter loss into
 447

the peritoneal cavity has been greatly reduced since the introduction of a design
which incorporates a metal disc with a central hole; this not only allows the
catheter to pass through the wall but also holds the catheter snugly to the skin of
the abdominal wall.

2.8. Dialysis in patients with polycystic kidney disease

Acute peritoneal dialysis may be carried out safely in patients with polycystic
kidneys. We have encountered no serious problems in 25 such patients, and
Henderson [37] reported similar success in 15 patients.

2.9 Complications of acute peritoneal dialysis after major abdominal surgery

When dialysis is necessary in the immediate postoperative period, the peritoneal

route should be avoided because of potential wound dehiscence and the increased
risk of peritonitis [56-59]. Hemodialysis is preferred to PD whenever the pos-
terior peritoneum has been breached, i.e. after retroperitoneal surgery or trau-
ma, because of inadequate drainage and a tendency to dissection of fluid behind
the peritoneum and into the scrotum [21, 22, 60]. However, the experience of Aye
[25], Nienhuis [40], Tzamaloukas [61] and Kanter [62] suggests that peritoneal
dialysis is a good alternative to hemodialysis for those who develop severe acute
renal failure in the immediate postoperative period. Tzamaloukas [61] described
14 patients with recent abdominal surgery treated with acute peritoneal dialysis.
In all but one the dialysis was effective and the technical complications were
minor except that one developed wound dehiscence. The complications observed
were a small leak around a gastrostomy, a colostomy, and drains, dissection of
fluid into the scrotum, positive cultures (of dialysate) in patients with pre-existing
peritonitis, positive culture without signs of peritonitis, intraperitoneal bleeding,
fluid retention and leakage around a drain, gastronomy, colostomy or catheter.

2.10. Overheated dialysate

Before modifications were introduced in the thermostat system of the automatic

cyclers, we encountered three instances in which the patient was infused with
overheated dialysate. All complained of severe abdominal pain; in two the
symptoms subsided after drainage of the dialysate (dialysis was continued with a
dialysate at body temperature). The third patient, who was exposed to very hot
dialysate, developed a paralytic ileus that lasted for three days. After this period,
her bowels started moving and she developed severe metabolic acidosis, which
improved after two days. Then the patient was started again on chronic peritoneal
dialysis.
448

2.11 Hypernatremic dialysate

This complication developed in a patient dialysed with the reverse osmosis

machine. Following a malfunction of the conductivity meter (which is supposed
to ensure an adequate dilution of the concentrated solution), the sodium concen-
tration rose to a level in excess of 155 mEq/1. The patient went into coma after
being exposed to this type of dialysate for at least four hours. She was continued
on dialysis for two more hours, at which time her serum electrolytes showed
severe hypernatremia (over 155 mEq/I). The remainder of the electrolytes were
normal; the high concentration of blood sugar (500 mg/dl) also may have contrib-
uted to the hyperosmolar state. The patient did not awaken from her coma and
died four hours later. Subsequently it was found that the conductivity meter had
not been working at its extreme range when the dialysate temperature was below
36° C. This experience suggests that. whenever a patient on dialysis develops
unexplained symptoms, dialysis shouid be discontinued until another cause for
the symptom is found. The remaining dialysate should be kept for analysis of its
contents.

2.12. Medical complications

Introduction of dialysate into the abdominal cavity induces many changes in

systemic hemodynamics. Most of the medical complications (Table 2) encoun-

Table 2. Complications of acute peritoneal dialysis

Medical complications:

Cardiovascular
a. hypovolemia
b. fluid overload, heart failure and pulmonary edema
c. arrhythmia and hypokalemia
d. ischemic heart disease
Pulmonary complications
Neurological complications
a. disequilibrium syndrome
b. convulsion
c. mental confusion
Metabolic complications
a. hyperglycemia
b. hypoglycemia
Hypematremia
Alkalosis
Persistent metabolic acidosis
Hyperkalemia
 449

tered during acute peritoneal dialysis are related to the rapidity with which these
changes are brought about.

2.13. Cardiovascular complications

Hypovolemia, hypernatremia, hyperglycemia, and alkalosis of varying severity

are common during acute peritoneal dialysis, particularly in those subjected to
repeated peritoneal dialysis. These symptoms develop when large volumes of
fluid are removed too rapidly and can usually be prevented by anticipating them
and by attending to them promptly once they develop.

2.13.1. Hypovolemia
De Pacifico [63] has demonstrated a decrease in cardiac output, an increase in
pulmonary artery pressure and an increase in peripheral vascular resistance after
dialysate is introduced into the abdominal cavity. Rapid removal of sodium and
water may deplete the intravascular volume and may further aggravate the
hemodynamic changes. The patient may become restless and develop tachycardia
and hypertension. Hypovolemia, a frequent complication, occurs more often
when one uses hypertonic dialysis solutions; however, it can also occur with 1.5%
solutions [64]. Patients on maintenance dialysis during acute renal failure seldom
have high serum osmolality. Under these circumstances, the patient may develop
a negative fluid balance during a 24 to 36 h dialysis with 1.5% solution, which is
expected to have an osmolality of 347 mOsm/kg of water. In these cases, the
operator should use a 0.5% dextrose solution so as to maintain the ideal weight
[64]. An alternative method of preventing excessive fluid loss uses a solution of
equal amounts of the standard 1.5% dextrose and Ringer's lactate [65]. If the
patient develops an extreme degree of hypovolemia, e.g., after hyperosmolar
solutions are used for a long period, peripheral vascular collapse with shock may
ensue [17). One must use extreme caution when administering 7% glucose
solution; indeed, the use of such a hyperosmolar solution is rarely indicated [21].
Hypovolemia can be corrected by appropriate salt and water replacement.

2.13.2. Fluid overload, heart failure and pulmonary edema

This common complication of acute peritoneal dialysis [12, 17-19, 66, 67] develops
as a result of positive fluid balance in a hypertensive or overhydrated patient.
Inadequate drainage may make the situation worse. In desperate situations, one
may have to resort to hemodialysis. Paradoxically, pulmonary edema is rarely
seen in dehydrated patients [64, 66]; when it does occur, it is attributed to
increased sodium retention [63]. If these patients are not managed carefully, they
may die [64, 68].
450

2.13.3. Arrhythmia
Tachyarrhythmias are frequent during dialysis, especially in patients with under-
lying heart disease or those receiving digitalis. Valk [13] reported 24 dialyses in 17
patients in which the procedure was complicated by tachycardia. Eleven of these
patients had underlying heart disease, five had wide fluctuations in serum po-
tassium levels, and in one the arrhythmia was unexplained. Valk [13] also re-
ported six patients who developed cardiac arrest while on dialysis; five of these
had underlying heart disease. Patients on digitalis for any reason may develop
serious arrhythmia during dialysis [12, 42] and digitalis intoxication may by
masked by hyperkalemia. Since digitalis is poorly dialysed [69], the correction,
during dialysis, of hyperkalemia, acidosis, hypocalcemia and hyponatremia may
precipitate digitalis intoxication. Several workers [19, 42, 70] have reported fatal
arrhythmias. Patients on digitalis who have underlying heart disease should
receive a dialysate containing at least 3 mEq/1 potassium. In addition, close
monitoring for cardiac rhythm is advisable.

2.13.4. Ischemic heart disease and myocardial infarction

Patients with ischemic heart disease tolerate hemodynamic fluctuations poorly.
Hence rapid ultrafiltration accompanied by a sudden fall in blood pressure may
precipitate anginal pain and, in some, frank myocardial infarction.

2.14. Pulmonary complications

Basal atelectasis, acute purulent bronchitis, pneumonia and pleural effusions

have occurred as a direct consequence of peritoneal dialysis [51, 71]. Repeated
upward displacement of the diaphragm following overdistention of the peritoneal
cavity with dialysate may produce atelectasis of the basal pulmonary segments
[71]. Berlyne has suggested that the short 'drain-out' period usually allowed for
these patients may not be long enough to permit the atelectic segment to re-ex-
pand, hence continued dialysis may decrease the vital capacity. In a severely
uremic patient, the level of consciousness may fluctuate, the cough reflex is
impaired and bronchial secretions may pool in dependent pulmonary segments;
these factors all predispose to purulent bronchitis and/or pneumonia. The risk of
these complications increases as the duration of dialysis is prolonged. Berlyne [71]
described three patients who developed atelectasis and pneumonia, two of whom
died. In Vaamonde's series of 14 patients with pneumonia, three died [12].
Pulmonary infection associated with asymptomatic positive peritoneal fluid cul-
tures played a significant role in the death of two patients [13]. Measurement of
pulmonary gas exchange during peritoneal dialysis indicated that P02 of arterial
blood falls by 3 to 26 mm Hg when the abdomen is distended with 21 dialysate [72,
73]. The P0 2 returns to basal level when the fluid is drained out. Aspiration of
gastric contents may be fatal during peritoneal dialysis. Sudden right-sided
 451

pleural effusion, which develops during dialysis, has been mentioned earlier.
Pulmonary complications carry a significant morbidity and mortality, but in the
absence of prospective pulmonary function studies in patients undergoing per-
itoneal dialysis, the morbidity is difficult to assess. The deaths due to pulmonary
complications can be greatly reduced by a general awareness of these problems.
The appropriate preventive measures include the use of small volumes of dialy-
sate per exchange, the use of a cycle exchange ofless than one hour, the reduction
of the duration of dialysis to less than 36h, and performance of the procedure
with the patient in a more upright position.

2.15. Neurological complications

Assessment of such complications is difficult during dialysis because of their close

similarity to the uremic syndrome [74-76]. In any event, the frequency of neu-
rological complications is much lower with peritoneal dialysis than with hemo-
dialysis, chiefly because PD is a slower process [74].

2.15.1. Disequilibrium syndrome

This syndrome is applied to the clinical signs and symptoms observed mainly in
hemodialysed patients; the cerebral edema responsible for it develops in response
to an osmotic gradient set up following the rapid removal of urea and other
solutes from the extracellular compartment during dialysis [77-82]. The patient
complains of headache, vomiting, hypertension, convulsions and coma; the last
may be fatal. This syndrome has rarely been reported in patients treated with
peritoneal dialysis [12, 19, 40]. The relatively slow transfer of urea across the
peritoneal membrane (compared with that during hemodialysis) probably ex-
plains the rare occurrence of disequilibrium syndrome during peritoneal dialysis.
Pabico [83] studied changes in cerebrospinal fluid compositon during peritoneal
dialysis and demonstrated that they were minor as compared with those during
hemodialysis.

2.15.2. Convulsion
In most series, this complication has been infrequent [12, 13, 68, 74] and many of
these represent the onset of dialysis eqUilibrium. However, because diseq-
uilibrium syndrome is rare during peritoneal dialysis, the patient who develops
convulsions should be studied carefully before this diagnosis is accepted. Sei-
zures, which usually occur after dialysis is completed, may be due to hypo-
glycemia, hypertension and pre dialysis overhydration. During a total of 184
dialyses involving 104 patients, Vaamonde [12] encountered convulsions on
twelve occasions (6.5%). The patients who developed convulsions were not
known to have a history of seizures. In two patients these episodes were ac-
companied by hypoglycemia and hypertension. On EEG, one of these patients
452

had a focus of hyperactivity in the left parietal area and had a venous hum over the
left mastoid. Lee [84] has attributed post-dialysis convulsion to lactic acidemia.
While the seizures are controlled with anticonvulsants, the attending physician
should make every effort to identify the cause. Peritoneal dialysis need not be
terminated in the pre sense of seizures.

2.15.3. Mental confusion

Mental confusion may be a manifestation of persistent uremia or an exaggerated
response to hypnosedatives, used so frequently in these patients. Methyldopa
when first used to control hypertension may cause mental clouding. Hypo- and
hyperglycemia may be a frequent accompaniment of such episodes [85]. The
clinical picture may be confused by markedly elevated blood pressure. Finally,
systemic acidosis with change in CSF pH can cause mental deterioration [86]. In
summary, it seems apparent that the diagnosis of disequilibrium syndrome can
only be accepted in a patient on peritoneal dialysis when all known causes are
excluded.

2.16. Metabolic complications

2.16.1. Hyperglycemia
The high levels of blood glucose observed during peritoneal dialysis [19, 87-89]
reflect the large amounts of glucose absorbed from the dialysate [88-90]. This
phenomenon is understandable in patients with diabetes or glucose intolerance,
but it is surprising in those who are not diabetics. The impaired hepatic clearance
of glucose and a state of insulin resistance in the non-diabetic uremic patient may
explain the hyperglycemia observed in nondiabetic uremics undergoing per-
itoneal dialysis. Vaamonde [12] observed hyperglycemia in 12 of 139 dialysis and
noted a relationship with the use of hypertonic dialysis solutions. Prolonged use
of a 7% solution may induce hyperglycemia in non-diabetic uremics when a
4.25% solution does not.

2.16.2 Hypoglycemia
This preventable complication is always a result of the administration of excessive
amounts of insulin to diabetics undergoing peritoneal dialysis. Thus it is essential
to monitor the blood sugar, especially during the first few dialyses. Hypoglycemia
may induce seizures and irreversible brain damage if it is not treated immediately.
Post-dialysis hypoglycemia can be avoided by omitting insulin from the last three
or four exchanges of dialysis if 0.5 or 1.5% solutions are used or by reducing the
insulin dose considerably if 4.25% solutions are used.
Sorbitol has been substituted for dextrose during the dialysis of diabetics [91],
but with it Tenckhoff [92] encountered nausea, vomiting and eventually coma in
several patients. Subsequently its use was discontinued at the request of the U.S.
Food and Drug Administration [93, 94].
 453

2.17. Hypernatremia

Serum sodium concentrations over 160mEq/1 have been recorded [7, 66, 89, 95,
96] during this relatively common complication of peritoneal dialysis; however,
mild degrees of hypernatremia usually pass unnoticed. Following ultrafiltration
using markedly hypertonic solutions, sodium is held back in the blood, for
reasons probably related to the charge on the ion rather than its size. Because the
filtrate so formed is hyponatremic compared to plasma [96], more water than
sodium is lost from plasma and extracellular fluid. This phenomenon also pro-
duces increases in serum chloride and bicarbonate. In addition, markedly hyper-
tonic solutions, i.e. 7% dextrose exaggerate the problem [64, 96]. Also, when the
glucose has been metabolized, the water shifts back into the cells, increasing the
concentration of the serum electrolytes. The hypernatremia which develops
during peritoneal dialysis is accentuated by the relatively high sodium concentra-
tions of certain commercial dialysate solutions (140-146mEq/I); thus, the use of
dialysis solutions containing sodium at 130-135 mEq/1 would reduce the frequency
of this complication. To prevent hypernatremia, Nolph [96] recommended the
replacement of one-half the net dialysis fluid loss with water intravenously or by
mouth, or by the use of a solution with a sodium concentration calculated to yield
a 'sieving coefficient' near 1.0. Substitution of 5% glucose and water for every
sixth or eighth liter of dialysate may also prevent hypernatremia [65]. The serum
chloride concentration usually follows the change in serum sodium concentra-
tion.

2.18. Alkalosis

Patients on peritoneal dialysis may develop either respiratory or metabolic al-

kalosis. Respiratory alkalosis may appear during the initial stages of dialysis when
hyperventilation due to a low spinal fluid pH may continue to reduce pC02 while
serum bicarbonate increases as bicarbonate is generated to correct the extracellu-
lar acidosis. Since the bicarbonate ion diffuses slowly across the blood brain
barrier, cerebral fluid pH does not change initially [48, 86] and the respiratory
centre (responding to the acid pH) continues to direct hyperventilation, further
reducing pC02 • This may be a minor problem in peritoneal dialysis because
lactate or acetate conversion to bicarbonate in the liver is slow enough to permit
the cerebral fluid pH to adjust. In view of this cycle of events, the infusion of
bicarbonate to rapidly correct acidosis should also be restrained. However, no
harmful effects have been attributed to temporary respiratory alkalosis [48].
Metabolic alkalosis develops slowly in patients on peritoneal dialysis, and
during intermittent peritoneal dialysis it increases progressively over a period of
three to four weeks [64], if they receive significant amounts of 7% dextrose
454

solution. Much of the time, these patients exhibit lethargy or drowsiness. The
changes in serum bicarbonate roughly parallel those in sodium, which suggests
that the rise in bicarbonate and the resultant metabolic alkalosis are due in part to
removal of relatively more water than bicarbonate [64]. The relatively slow
movement of the bicarbonate ion across the peritoneal membrane tends to
maintain a high level in the blood and thus produces a contraction alkalosis. This
complication can be prevented by avoiding hyperosmolar solutions, especially
7% dextrose.

2.19. Persistent metabolic acidosis

Uremics with liver failure may have difficulty in converting the lactate of the
dialysate to bicarbonate, and as a result the blood lactate rises [12, 84]. This
development, which will aggravate the metabolic acidosis already present in
these patients [12, 22, 28], can be corrected by infusion of bicarbonate [19, 22] or
prevented by the use of acetate instead of lactate in the dialysate [17, 97].

2.20. Hyperkalemia

In patients submitted to peritoneal dialysis, Boen [48] noted a sudden rise in

plasma potassium concentrations after dialysis is terminated; he attributed this to
a breakdown of glycogen stored during dialysis and the associated release of
potassium. Vaamonde [12] made a similar observation in 11 dialysis patients in his
series. The potassium rarely reaches a level requiring a repeat dialysis, and in
most instances the accumulation can be controlled with cation exchange resins.

2.21. Mortality rates due to complications

The mortality associated with acute peritoneal dialysis is difficult to determine;

the literature reports it to be between 5% and 12% [8,12,18,98]. Considering that
most of these patients are seriously ill, this high rate is not surprising. Obviously,
although it is a safe procedure, peritoneal dialysis may be associated with con-
siderable morbidity. Most of the complications can be controlled by appropriate
treatment, and only rarely does it become necessary to terminate the dialysis. In
an experienced unit, the great majority of these complications can be prevented
or kept at a low level.
 455

3. Chronic intermittent peritoneal dialysis

Interest in chronic peritoneal dialysis was renewed in the late 1960s and early
1970s following the introduction of implantable, bacteriologically safe peritoneal
dialysis catheters [99,100]. Since then several centers have treated large groups of
patients with end-stage renal disease with chronic intermittent peritoneal dialysis
and have established its reliability and also its superiority over hemodialysis or
transplantation in certain patients or circumstances [101-105]. With the increasing
use of chronic intermittent peritoneal dialysis at a center or at home, we have
come to recognize several complications, some serious and others less so. These
complications are listed in Table 3. This chapter will deal with complications
other than those due to infection, which are described in Chapter 13.

3.1. Catheter-related complications

These complications are either early or late. Early complications associated with
catheter insertion and its function are much the same as those seen with the acute
peritoneal dialysis except that malfunction may be due to catheter tip displace-
ment or internal obstruction due to omental incarceration. Catheter malfunction
usually presents as one-way or outflow obstruction, i.e. free inflow of dialysate
which cannot be drained out. To avoid this, Tenckhoff [16] recommends that
during implantation the intra-abdominal segment should be directed caudally by
pushing the subcutaneous catheter tract in a cephalad direction or by giving it an
arched subcutaneous course. He also observed that this complication is most
frequent with surgical catheter implantation when the surgeon does not appreci-
ate the importance of giving the catheter the desired caudal direction.

Table 3. Complications of chronic intermittent peritoneal dialysis

Catheter-related problems:
a. early catheter malfunction
b. late complications
Depletion syndrome
Neurological complications:
a. peripheral neuropathy
b. dialysis disequilibrium
c. EEG abnormalities
d. neuropsychiatric problems
Osteodystrophy
Ascites
Constipation and bowel perforation
Kidney stones
456

The frequency of catheter tip displacement can also be minimized by using the
TWH catheter [106], which has two flat silicone rubber discs in the distal end of its
intraabdominal part to stabilize it in the abdominal cavity. In addition, the
distance between the two cuffs is only 2.5 cm, so that the subcutaneous tunnel is
short and the catheter exits at or near the midline. A prospective controlled trial
has demonstrated the superiority of this catheter over those designed by Ten-
ckhoff and Goldberg. Sixty-three patients, 21 in each group, were assigned to one
of the three catheters. During the course of the study, 28% of the Tenckhoff
catheters, 19% of the Goldberg catheters and 8% of the TWH catheters had to be
replaced because of one-way obstruction. A radiological study of the catheters
one month after implantation showed that 33% of the Tenckhoff catheters, 23%
of the Goldberg catheters and 7% of the TWH catheters had migrated out of the
pelvis [106, 107].
Entrapment of omental tissue in the catheter lumen, the other major cause of
catheter malfunction and displacement, produces one-way catheter obstruction
which usually develops within a few days of implantation. The patient may
experience localized or diffuse abdominal discomfort. A radiograph after injec-
tion of a radiopaque material may show the displaced catheter and the internal
obstruction as one or multiple round shadows within the lumen. The frequency of
omental or tissue entrapment is related to the size of the catheter's terminal side
holes [16]: the larger the size, the greater the frequency of tissue incarceration.
Such obstructions are usually permanent and cannot be forced out. Temporary
success may occasionally be achieved by expelling the tissue with the 'Italian
corkscrew,' an instrument devised to remove clots from arteriovenous shunts
[108]. Heparin irrigation may also relieve the blockage temporarily. However,
this incarceration is a recurrent problem and eventually requires catheter repla-
cement.
Catheter obstruction also may follow an episode of partially treated asymp-
tomatic peritonitis. Intra-abdominal adhesions or fibrosis around the catheter tip
may produce loculation of fluid, increasing resistance to the dialysate inflow, and
pain. In such instances, dye introduced through the catheter will show that the
fluid does not diffuse freely and that a sac has formed around the catheter tip.
However, the catheter is usually patent. In this situation the catheter must be
replaced under antibiotic coverage. External compression of the intra-abdominal
portion of the catheter by distended rectum or sigmoid is encountered chiefly in
patients who are habitually constipated; usually they present with one-way
obstructions. Here, a flat plate of the abdomen shows large fecal shadows in the
rectosigmoid region and normally positioned catheter. Stimulation of vigorous
bowel movements with laxatives, suppositories or enemas usually restores the
free flow of dialysate.
Late catheter-related complications usually are secondary to infection. We
observed one patient who developed a hernia through the catheter exit site four
years after intermittent peritoneal dialysis. The hernial sac, which was embedded
 457

in the layers of abdominal wall, had strangulated and brought the patient to
emergency surgery.

3.2. Depletion syndrome

If not compensated, protein, amino acid and vitamin losses into the dialysate over
a long period may produce significant depletion in some patients [109-115],
Tenckhoff described three patients with depletion syndrome; all had extremely
poor caloric and protein intake, although he did not observe this syndrome in
others who were able to compensate for these losses [116-118]. He also observed
that more prolonged protein starvation associated with intercurrent illness,
especially in the presence of peritonitis, will provoke the depletion syndrome;
these patients require parenteral protein supplementation. With conventional
intermittent peritoneal dialysis, protein losses range from 0.68 to 4.49 gil dialy-
sate [48, 111]. The predominant loss is albumin; however, the dialysate contains
all components of plasma protein [48, 119, 120]. The small volume of ascitic fluid
that collects in the peritoneal cavity between dialyses is rich in protein and the
drainage of this protein-rich fluid during the first few exchanges accounts for
nearly one-half of the protein lost during a dialysis. Protein loss during the latter
half of dialysis is considerably less [112, 121-123]. Strauch [121] found that protein
loss was affected by both the duration of dialysis and the dextrose concentration
of the dialysate. Aminoacid losses in the dialysate are in the range of 2-3 g/day
[124-125]. According to Kowalewski [126], it seems that hydroxyproline-contain-
ing proteins are removed faster than other proteins.
One can compensate for the protein loss by maintaining a protein intake of 1 g/
kg/day. In addition, these patients should receive a water-soluble, multiple-
vitamin preparation. Finally, they should ingest sufficient calories to prevent
protein catabolism and enhance growth [16].

3.3. Neurological complications

3.3.1. Peripheral neuropathy

Prevalence and progression of uremic neuropathy in patients undergoing long-
term peritoneal dialysis is similar to that in patients on hemodialysis [16, 127, 128].
In the past, it had been reported that the peripheral neuropathy following
peritoneal dialysis was less than that after hemodialysis [129], and subsequently
we had conduded that it was more frequent after PD than after hemodialysis
[130]. In 1975, we described four patients who developed severe neuropathy in
contrast to 17 other patients who showed only minimal progression [131]. Elec-
trophysiological studies in two of these four patients showed significant changes
in two peripheral nerves; a progressive decrease in motor conduction velocity in
458

the peroneal nerve, and an increase in the distal sensory latency in the median
nerve. These changes did not correlate with the level of BUN or serum creatinine.
Blumenkrantz suggests that conflicting observations can be explained by pre-
existing differences in the populations under comparison and by differences in
treatment regimens. From his review of the literature, he concluded that nerve-
conduction velocities do not decrease in patients undergoing either HD or PD
except in underdialysed or severely malnourished patients [127].

3.3.2. Dialysis disequilibrium

Dialysis disequilibrium is rare in patients treated by chronic peritoneal dialysis
[12,19,40]. This complication was discussed in detail earlier in this chapter.

3.3.3. EEG abnormalities

Electroencephalographic studies of patients undergoing chronic PD showed that
they have a more normal background rhythm and photic driving response than
those undergoing hemodialysis [132]. The latter showed more normal activity in
latency and amplitude of visual-evoked response. The significance of these
findings and of the difference between the two groups is unclear.

3.3.4. Neuropsychiatric problems

The literature concerning neuropsychiatric disturbances in dialysis patients is
scanty. Colotla [133], who carried out psychological comparisons of patients
undergoing peritoneal or hemodialysis, suggested that manual dexterity declines
in those undergoing peritoneal dialysis. However, Roxe [132] did continuous
performance task tests on 10 matched pairs of PD and HD patients, and found no
differences in cognitive functioning between the two groups.

3.4. Osteodystrophy

In chronic renal failure, abnormalities in calcium and phosphorus homeostasis

are often manifested in an elevation of serum phosphorus and a decrease in serum
calcium. Hypocalcemia stimulates parathyroid secretion, and the secondary hy-
perparathyroidism so induced produces osteitis fibrosa in 82% of bone biopsies
from patients with renal failure [134]. Hyperphosphatemia is a frequent finding in
patients on chronic peritoneal dialysis because dialysis by this route achieves poor
control of phosphorus levels [135]. As a result, these patients commonly have the
osteitic form of bone disease and, for the same reason, have a high incidence of
vascular and soft-tissue (including articular) calcifications [136]. The latter may
give rise to a variety of articular complaints presenting as gout or pseudogout.
Commonly, the hyperparathyroidism, which is progressive, is difficult to treat
with vitamin D or its analogues because of the presence of hyperphosphatemia.
These patients often require parathyroidectomy. In 1973, we reported that the
 459

osteomalacia form of renal osteodystrophy was rare among patients undergoing

chronic peritoneal dialysis; however, since then we have encountered several
who developed osteomalacia and pseudofractures [128]. Cangiano et al. [137] also
reported a lower incidence of osteomalacia and no fractures or soft-tissue cal-
cification in patients undergoing maintenance peritoneal dialysis [137]. They used
a dialysate containing 7.0mg % of calcium and noted a decrease in the levels of
serum alkaline phosphatase. In summary, it may be concluded that the pro-
gression of renal osteodystrophy can be prevented as long as patients on mainte-
nance peritoneal dialysis are adequately dialysed against a dialysate calcium of
6.5-7.0 mg %; vitamin D or its analogues are administered to maintain serum
calcium at high normal values; and serum phosphorus is maintained under 6 mg
% with phosphatase binders.

3.5. Ascites

Ascites, which is not uncommon in patients who have received peritoneal di-
alysis, appears especially after termination of peritoneal dialysis and during
hemodialysis or transplantation. Among the causes of this complication, Ten-
ckhoff [16] cites the use of hypertonic and very acidic dialysate. He believes it to
be a response to some form of peritoneal irritation. Response to treatment is
poor. Control of ascites has been attempted using reinfusion of ascitic fluid,
sequential ultrafiltration with hemodialysis, drainage into the thoracic duct and
instillation of intraperitoneal triamcinolone acetate.

3.6. Constipation and bowel perforation

Chronic constipation constitutes a serious problem in these patients because they

are given large amounts of aluminum-containing phosphate binders. We encoun-
tered six chronic peritoneal dialysis patients who developed bowel obstruction
secondary to fecal impaction [138]; four of these had bowel perforation. All six
were receiving 2-4 g/day aluminum hydroxide in addition to oral calcium in a dose
of 2 g/day. These patients presented with crampy abdominal pain and vomiting
and, surprisingly, two had watery bowel movements. All had diffuse abdominal
tenderness and palpable masses. Radiological investigation revealed large areas
of fecal shadows. Five of the six developed fecal peritonitis. The sigmoid colon
perforated in three and the rectum in one patient. All had fecoliths, which caused
ulcerations with gangrene. It is interesting that in four of these six patients
polycystic kidney disease was the primary cause of renal failure. Five of the six
died of fulminant sepsis. As a result of this and other experience, we avoid the use
of aluminum-containing phosphate binders and calcium carbonate in peritoneal
dialysis patients. Phosphate binders should be used sparingly in elderly patients
460

because they have an increased frequency of diverticulosis [139]. Indeed, a much

higher incidence of diverticulosis has been reported in patients with polycystic
kidney disease, and this may explain their great predisposition to serious con-
stipation [139].

3.7. Kidney stones

This complication is unexpected because these patients pass only small amounts
of urine which contains extremely low concentrations of calcium. We first de-
scribed [140] two patients who had recurrent renal colic and passed small stones.
On analysis the stones were composed of calcium oxalate. Since then we have
observed six more of these patients. These complications should be suspected in
any dialysis patient who complains of renal colic or dysuria. Calculi also have
been described in patients on chronic hemodialysis, but here the stones are
composed of a proteinaceous material [141]. Our finding of calcium oxalate stones
may be related to the increased oxalate concentration in the plasma of these
patients [140].

References

1. Ganter G: Uber die Benseitigung giftiger stoffe aus dem Blute durch Dialyse. Muench Med
Wochenschr 50: 1478, 1923.
2. Maxwell MH, Rockney RE, Kleeman CR, Twiss MR: Peritoneal dialysis. I Technique and
applications. lAMA 170: 917-924, 1959.
3. Boen ST, Mion CM, Curtis FK, Shilipeter G: Periodic peritoneal dialysis using the repeated
puncture technique and an automatic cycling machine. Trans Am Soc Intern Artif Organs 10:
409,1964.
4. Tenckhoff H, Schechter H: A bacteriologically safe peritoneal access device. Trans Am Soc
ArtifIntern Organs 14: 181-187, 1968.
5. Popovich RP, Moncrief lW, Nolph KD, Ghods AJ, Plye WK: Continuous ambulatory per-
itoneal dialysis. Ann Intern Med 88: 449--456, 1978.
6. Oreopoulos DG, Robson M, Izatt S, Clayton D, deVeber GA: A simple and safe technique for
continuous ambulatory peritoneal dialysis (CAPD). Trans Am Soc Artif Intern Organs 24: 484-
489,1978.
7. Odel HM, Ferris DO, Power MH: Peritoneal lavage as an effective means of extrarenal
excretion. Am J Med 9: 63--77,1950.
8. Barry KG, Swartz FD, Hano JE, Schrier RW, Canfield C: Peritoneal dialysis: Current appli-
cation and recent developments. Proc 3rd Int Cong Nephrol3: 288--304, (Karger, Basel), 1967.
9. Cameron IS, Ogg C, Trounce JR: Peritoneal dialysis in hypercatabolic acute renal failure.
Lancet 1: 118-119,1967.
10. Stott RD, Ogg CS, Cameron IS, Bewick M: Why the persistently high mortality in acute renal
failure? Lancet 2: 75-79, 1972.
11. Orofino L, Lampreabe I, Munio R, Sancho JL, Villar F, Gomez VP, Montenegro 1, Garcia DR:
Supervivenia del fracaso renal agudo (FRA) sometido a dialysis. Revision de 82 pacientes. Rev
Clin Esp 140: 243--249, 1976.
 461

12. Vaamonde CA, Michael VF, Metzger RA, Carrol KE: Complications of acute peritoneal
dialysis. J Chron Dis 28: 637-659, 1975.
13. Valk TW, Swartz RD, Hsu CH: Peritoneal dialysis in acute renal failure: Analysis of outcome
and complications. Dial Transpl 9: 54-58, 1980.
14. Pauli HG, Buttigafer E, Varburger C: Clinical experience with peritoneal dialysis. Helv Med
Acta 33: 51-58, 1966.
15. Svetvilas C, Tannenberg AM: Pneumoperitoneum with peritoneal dialysis. Lancet I: 910-911,
1971.
16. Tenckhoff H: Home peritoneal dialysis. In Massry SG, Sellers AL (eds), Clinical Aspects of
Uraemia and Dialysis. Charles C. Thomas, Springfield, Illinois, 1976, p 583.
17. Ribot S, Jacobs MG, Frankel HJ, Bernstein A: Complications of peritoneal dialysis. Am J Med
Sci 252: 505-517, 1966.
18. Matalon R, Levine S, Eisinger RP: Hazards in routine use of peritoneal dialysis. NY State J Med
71: 219-224, 1971.
19. Maher JF, Scheiher GE: Hazards and complications of dialysis. N Engl J Med 273: 370-377,
1965.
20. Furman KI, Gomperts ED, Hockley J: Activity of intraperitoneal heparin during peritoneal
dialysis. Clin NephrollO: 15-18, 1978.
21. Stewart JH, Tuckwell LA, Sinnet IF, Edward KDG, Whyte HM: Peritoneal and hemodialysis:
A comparison of their morbidity and of the mortality suffered by dialysed patients. Quart J Med
139: 407-420, 1966.
22. Burns RO, Henderson LW, Hager EB, Merril JP: Peritoneal dialysis. N Engl J Med 267: 1060-
1066,1962.
23. Thompson N, Uldall R: A problem in peritoneal dialysis. Lancet 2: 602-603, 1969.
24. Ladefoged J, Steiness I: Intra-abdominal bleeding complicating peritoneal dialysis. Lancet 1:
190,1970.
25. Aye MM, Kalatilke AK, Shakmen R: Peritoneal dialysis in surgery. Proc Eur Dial Transpl
Assoc 2: 49-52, 1965.
26. Uldall PR, Dyck RF, Woods F, Merchant N, Martin GS, Cardella CJ, Sutton D, de Veber GA:
A subclavian cannula for temporary vascular access for hemodialysis or plasma-phoresis. Dial
Transpl 8: 963-968, 1979.
27. Barry KG, Swartz FD: Peritoneal dialysis: current status and future applications. Pedial. Clin
North Am 11: 593-609, 1964.
28. Hager EB, Merril JP: Peritoneal dialysis and acute renal failure. Surg Clin North Am 43: 883-
896, 1963.
29. Firmat J, Zucchini A: Peritoneal dialysis in acute renal failure. Today's art of peritoneal dialysis.
Contrib Nephrol17: 33-38, (Krager, Basel), 1979.
30. Edward SR, Unger AM: Acute hydrothorax a new complication of peritoneal dialysis. JAMA
199: 853-855, 1967.
31. Finn R, Jowett EW: Acute hydrothorax: complication of peritoneal dialysis. Br. Med J 2: 94,
1970.
32. Holm J, Lieden B, Lindqrist B: Unilateral pleural effusion - a rare complication of peritoneal
dialysis. Scand J Urol NephroI5:84-85, 1971.
33. Haberli R, And Stucki P: Akuter hydro-thorax als Komplikation bei peritoneal dialyse. Praxis
60: 13-13-14, 1971.
34. Fehmirling E, Christensen E: Hydrothorax under peritoneal dialyse. Ugeskr Laeg 137: 1650-
1651, 1975.
35. Alquier P, Achard J, Bonhome R: Hydrothorax aign au loure de dialyses peritoneales. La Nouv
Press Med 4: 192,1975.
35a. Rudnick MR, Coyle JF, Beck H, McCurdy DK: Acute massive hydrothorax complicating
peritoneal dialysis, report of 2 cases and a review ofthe literature. Clin Nephrol12: 38-44,1980.
462

35b. Milutinovic J, Shyong W, Lindholme DD, Leroy N: Acute massive unilateral hydrothorax as
result of complications of peritoneal dialysis. South Med J 78: 827, 1980.
36. Mion CM, Boen ST: Analysis of factors responsible for the formation of adhesions during
chronic peritoneal dialysis. Am J Med Sci 250: 675-679, 1965.
37. Henderson LW: Peritoneal dialysis. Clinical aspects of uraemia and dialysis. In: Massry SG,
Sellers AL (eds.), Charles C Thomas, Springfield Illinois, 1976, p 574.
38. Simkin EP, Wright FK: Perforating injuries of the bowel complicating peritoneal catheter
insertion. Lancet 1: 64-67, 1968.
39. Chugh KS, Bhattacharya K, Amaresan MS, Sharma BK, Bansal VK: Peritoneal dialysis our
experience based on 550 dialyses. J Assoc Physicians India 20: 215-221, 1972.
40. Nienhuis LI: Clinical peritoneal dialysis. Arch Surg 93: 643-653, 1966.
41. Pauli HG, Billikofer E, Vorburger C: Clinical experience with peritoneal dialysis. Helv Med
Acta 33: 51-58, 1966.
42. Palmer RA, Maybee TK, Henry EW, Eden J:Peritoneal dialysis in acute and chronic renal
failure. Can Med Assoc J 88: 920-927, 1963.
43. Krebs RA, Burtiss BB: Bowel perforation. lAMA 198: 486-487, 1966.
44. Denovales EL, Avendano LN: Risks of peritoneal catheter insertion (letter). Lancet 1: 473,
1968.
45. Dunea G: Peritoneal dialysis and hemodialysis. Med Clin North Am 55: 155-175,1971.
46. Rigalosi RS, Maher JF, Schriener GE: Intestinal perforation during peritoneal dialysis. Ann
Intern Med 70: 1013-1015,1964.
47. Edwards DB, Gardner RD, Williams DG: Rupture of a hernial sac: A complication of per-
itoneal dialysis. J Uro1108: 255, 1972.
48. Boen ST: Peritoneal dialysis in clinical medicine. Springfield Illinois, Charles C Thomas, 1964.
49. Vaamonde CA: Peritoneal dialysis current status. Postgrad Med 62: 148-156, 1977.
50. Rubin J, Oreopoulos DG, Liu IT, Mathews R, de Veber GA: Peritonitis and bowel perforation
in peritoneal dialysis: Ann Intern Med 81: 403, 1974.
51. Lee HA: Evaluation of peritoneal dialysis and hemodialysis in pediatrics. I J Med Sci 3: 28-44,
1967.
52. Wertheimer M, Nielsen KD, Stevens LE: Clinical investigation of the dynamics of peritoneal
dialysis. AM J Surg 114: 685-690,1967.
53. Goldsmith HJ, Edwards EC, Moorhead PJ et al.: Difficulties encountered in intermittent
dialysis for chronic renal failure. Br I Urol 38: 625-634, 1966.
54. Smith E, Chamberlain MJ: Complications of peritoneal dialysis. BR Med J 1: 126-127,1965.
55. Stein MF Jr: Intraperitoneal loss of dialysis catheter. Ann Intern Med 71: 869-870, 1969.
56. Berne TW, Barbour BH: Acute renal failure in general surgical patients. Arch Surg 102: 594-
597,1971.
57. Day SW et al.: Management of post operative renal failure by peritoneal dialysis. Am Surg 30:
814-818,1964.
58. Kulatilke AE: The artificial kidney in acute renal failure following surgery. Proc R soc Med 59:
42-45, 1966.
59. Miller A: Peritoneal dialysis. Proc R Soc Med 59: 40-42, 1966.
60. Thomson WB, Buchanan A, Doak PD, Pert WS: Peritoneal dialysis. Br Med 11: 932-935, 1964.
61. Tzamaloukas HH, Garella S, Chazan A: Peritoneal dialysis for acute renal failure after major
abdominal surgery. Arch Surg 106: 639-643, 1973.
62. Kanter A, Nadler N, Vertel RM etal.: Peritoneal dialysis: indications and techniques in surgical
patients. Surg Clin North AM 98: 47-58, 1968.
63. Pacifico AD, Lasker N, Frank MJ, Levinson GE: Cardiovascular function in peritoneal dialysis.
Trans Am Soc Artif Intern Organs 11: 86-88,1965.
64. Gault MH, Ferguson El, Sidhu IS, Corbin RP: Fluid and electrolyte complications of peritoneal
dialysis. Ann Intern Med 75: 253-262, 1971.
 463

65. Moriarty MV: Hypertonic peritoneal dialysis (letter). Ann Intern Med 72: 146, 1970.
66. Swales JD: Sodium uptake in peritoneal dialysis. Br Med J 3: 345-347, 1967.
67. Shear L, Shinaberger JH, Barry KG: Peritoneal transport of antibiotics in man. N Engl J Med
272: 666--669, 1965.
68. Pingle A, Smith E: Daily peritoneal dialysis in renal failure. Br J Urol 36: 493-500, 1964.
69. Ackerman GL, Doherty JE, Flanigan WJ: Peritoneal dialysis and hemodialysis of tritiated
digoxin. Ann Intern Med 67: 718-723,1967.
70. Fleishman SJ, Goldberg B, Luonsine AJ: Peritoneal dialysis: the results of its use in 55 patients.
S Afr Med J 39: 435-439, 1965.
71. Berlyne GM, Lee HA, Ralston AJ, Woodlock JA: Pulmonary complications of peritoneal
dialysis. Lancet 2: 75-78,1966.
72. Freedman S, Maberly DJ: Pulmonary gas exchange during dialysis. Br Med J 3: 48, 1971.
73. Goggin MJ, Joekes AM: II Pulmonary gas exchange during peritoneal dialysis. Br Med J 2: 247-
248, 1971.
74. Tyler HR: Neurological complications of dialysis, transplantation and other forms of treatment
in chronic uremia. Neurology 15: 1081-1088, 1965.
75. Hampers CL, Doak PB, Callaghan MN, Tyler HR. Merril JP: The electroencephalogram and
spinal fluid during hemodialysis. Arch Intern Med 118: 340-346, 1966.
76. Locke S, Merril JP, Tyler HR: Neurologic complications of acute uraemia, Arch Intern Med 108:
519-530, 1961.
77. Sitpria V, Holmes JH: Preliminary observation on intracranial pressure and intraocular pressure
during hemodialysis. Tr Am Soc Artif Intern Organs 10: 340-344, 1964.
78. Gilliland KG, Hegstrom RM: Effect of hemodialysis on cerebrospinal fluid pressure in uremic
dogs. Trans Am Soc Artif Intern Organs 8: 44-50, 1963.
79. Kennedy AC, Linton AL, Eaton JC: Urea in the cerebrospinal fluid after hemodialysis. Lancet
1: 410-511, 1962.
80. Dossetor 18, Oh JH, Dayes L, Pappins HM: Urea and water changes with rapid hemodialysis of
uremic dogs. Am Soc Artif Intern Organs 10: 323-327, 1964.
81. Kennedy AC, Linton AL, Luke RG, Renfrew S: Electroencephalographic changes during
hemodialysis. Lancet 1: 408-411, 1963.
82. Setprija V, Holmes J: Intracranial and intraocular pressure during hemodialysis. Clin Res 12:
258,1964.
83. Pabico RC, Atanacio BC, McKenna BA: Peritoneal dialysis and its effect on cerebrospinal fluid
composition. Trans Am Soc Artif Intern Organs 16: 97, 1970.
84. Lee HA, Hill LF, Hewill V, Ralston AJ. Berlyne GA: Lactic acidemia in peritoneal dialysis.
Proc Eur Dial Transpl Assoc 4: 150-155, 1967.
85. Whang R: Hyperglycaemic nonketotic coma induced by peritoneal dialysis Lancet 87: 453-456,
1967.
86. Posner JB, Pluim F: Spinal fluid pH and neurological symptoms in systemic acidosis. New Engl J
Med 277: 605-613,1967.
87. Hutchings RH, Hegstrom RM, Scribner BH: Glucose intolerance in patients with long term
intermittent dialysis. Ann Intern Med 65: 275-285, 1966.
88. Boen ST: Kinetics of peritoneal dialysis: a comparison with the artificial kidney. Medicine (Bait)
40: 243, 1961.
89. Boyer J, Gill GN, Epstein FH: Hyperglycemia and hyperosmolality complicating peritoneal
dialysis. Ann Intern Med 67: 568-572, 1967.
90. Doolan PD, MurphyWP, Wiggens RA, Carter NW, CooperWC, Wattern RH. Elphen EL: An
evaluation of intermittent peritoneal lavage. Am J Med 26: 831, 1959.
91. Yutuc W, Ward G. Shilipeter G, Tenckhoff H: Substitution of sorbitol for dextrose in peritoneal
irrigation fluid. Trans Am Soc Artif Intern Organs 13: 168-171,1967.
92. Tenckhoff H: Choice of peritoneal dialysis solution (Editorial). Ann Intern Med 75: 313-314,
1971.
464

93. Vidt DG: Recommendation on choice of peritoneal dialysis solution. Ann Intern Med 78: 144-
145,1973.
94. Gault MH: Peritoneal dialysis solutions. CMAJ 108: 325-327, 1973.
95. Miller RB, Tassistro CR: Peritoneal dialysis. N Engl J Med 281: 945-949, 1969.
96. Nolph KD, Hano JE, Teschan PE: Peritoneal sodium transport during hypertonic peritoneal
dialysis. Ann Intern Med 70: 931-941, 1969.
97. Sheppard JM, Lawrence JR, Oon RCS, Thomas DW, Row PG, Wise PH: Lactic acidosis:
recovery associated with use of peritoneal dialysis. NZJ Med 4: 389-392, 1972.
98. Coltheart G, Tiller DJ, Hurley BP, Duggin G, Johnson JR, May J: Peritoneal dialysis. A review
of two years experience. Med J Aust 2: 314-316, 1973.
99. Tenckhoff H, Schechter H: A bacteriologically safe peritoneal access device. Trans Ann Soc
Artif Intern Organs 14: 181-187, 1968.
100. Striker GE, Tenckhoff H: A transcutaneous prosthesis for prolonged access to the peritoneal
cavity. Surgery 69: 70--74, 1971.
101. Brewer TE, Caldwell Fr, Paterson RM, Hanigan WJ: Indwelling peritoneal dialysis catheter.
Experience with 24 patients. JAMA 219: 1011-1015, 1972.
102. Blumenkrantz MJ, Miller JH, Barslay ME, Kopple JD, Friedler RM, Coburn JW: Chronic
peritoneal dialysis for the management of chronic renal failure. Clin Dial Transpl Forum Nov
17-18,1973 Washington DC p 117.
103. Oreopoulos D: Home peritoneal dialysis. Proc Eur Dial Trans Assoc 12: 139-145.
104. Sherrard DJ, Curtis FK, Lindner A, Scollard D, Merrit A: Peritoneal dialysis, a feasible
alternative. Clin Dial Transpl Forum Nov 17-18, 1973 Washington DC, p 114.
105. Fenton SS, Cattran DC, Barnes NM, Waugh K: Home peritoneal dialysis. Trans Am Soc Artif
Intern Organs 23: 194-200,1977.
106. Oreopoulos DG, Izatt S, Zellerman G, Karanicolas S, Mathews RE: A prospective study of the
effectiveness of three permanent catheters. Proc Dial Trans Forum 6: 96-97, 1967.
107. Oreopoulos DG, Zellerman G, Izatt S: The Toronto Western Hospital permanent peritoneal
catheter and continuous ambulatory peritoneal dialysis connector. Proc Int Symp CAPD, Paris
Nov 2-3, 1979, Excerpta Medica 73-78.
108. Haberstrob PB, Uniyal B, Trivedi H: A clot screw. Dial Transpl Dec/Jan 27,1974.
109. Palmer RA, Newell JF, Gray EJ, Quinton WE: Treatment of chronic renal failure by prolonged
peritoneal dialysis N Engl J Med 274: 2293, 1970.
110. Lasker N, Harvey A, Baker H: Vitamin levels in hemodialysis and intermittent peritoneal
dialysis. Trans Am Soc Artif Intern Organs 9: 51, 1963.
111. Berlyne GM, Jones JH, Hewitt V, Nilwarangkur S: Protein loss in peritoneal dialysis. Lancet 1:
738--741, 1964.
112. Gordon S, Rubin ME: Protein losses during peritoneal dialysis. Am J Med Sci 253: 283, 1967.
113. Young GA, Parsons FM: The effect of peritoneal dialysis upon the aminoacids and other
nitrogenous compounds in the blood and dialysate from patients with renal failure. Clin Sci 37:
1-10,1969.
114. Berlyne GM, Lee HA, Giordano C, Delascale C, Esposito R: Aminoacid loss in peritoneal
dialysis. Lancet 1: 1339-1341, 1967.
115. Sullivan JF, Eisenstein AB, Motola OM, Mittal AK: The effect of dialysis on plasma and tissue
levels of Vitamin C. Trans Am Soc Artif Intern Organs 18: 227-280, 1972.
116. Tenckhoff H, Curtis FK: Experience with maintenance peritoneal dialysis in the home. Trans
Am Soc Artif Intern Organs 16: 90-95, 1970.
117. Lindner A, Tenckhoff H: Influence of anabolic steroids on nitrogen balance in chronic per-
itoneal dialysis. Nephron 9: 77-85, 1972.
118. Lindner A, Tenckhoff H: Nitrogen balance in patients on maintenance peritoneal dialysis. Trans
Am Soc Artif Intern Organs 16: 255-259, 1970.
119. Miller RB, Tassistro CR: Peritoneal dialysis. N Engl J Med 281: 945 -949,1969.
 465

120. Henderson LW, Merril JP, Crane C: Further experience with the inlying plastic conduit for
chronic peritoneal dialysis. Trans Am Soc Artif Intern Organs 9: 108,1963.
121. Strauch M, Walzvn P, Van Henning GE, Roettger G, Christ H: Factors influencing protein loss
during peritoneal dialysis. Trans Am Soc Artif Intern Organs 13: 172-175, 1967.
122. Bastl C, Goffinet JA, Hendler ED, Longnecker RE: Protein losses in patients with silastic
peritoneal dialysis catheters. Clin Dial Transpl Forum Nov 17-18, (Abstract), 1973.
123. Baran H, Lebek R, Spett K, Twardowski Z: Kinetics of protein loss during peritoneal dialysis.
Pol. Med J 11: 277-280, 1972.
124. Glessing J: Addition of aminoacids to peritoneal dialysis fluid. Lancet 2: 812, 1968.
125. Kobayaski K, Manj T, Hiramatsu S et al.: Nitrogen metabolism in patients on peritoneal dialysis.
Today's art of peritoneal dialysis, Trevino-Becerra SA, Boen FST, published in Basel, p 93,
1979.
126. Kowalewski J, Tomaszewski J, Hanzlik J, Zawislak H, Zbikowska A: The elimination of free,
peptide-bound and protein bound hydroxyproline into dialysate during peritoneal dialysis in
patients with renal failure. Clin Chern Acta 34: 123,1971.
127. Blumenkrantz MJ, Lindsay RM: Comparison of hemodialysis and peritoneal dialysis: A review
of literature: Contr Nephrol17: 20-29, (Krager, Basel), 1979.
128. Oreopoulos DG: Chronic peritoneal dialysis. Clin nephrol 9: 165-173, 1978.
129. Fernandez E, Quellhorst E, Meitzsch G: Treatment of chronic uraemia. A comparison of
intermittent peritoneal dialysis and hemodialysis. Abstract 13th Congr Eur Dial Transpl Assoc,
p. 1367, 1976.
130. Oreopoulos DG, Blair GRD, Meema HE, deVeber GA: Overall experience with peritoneal
dialysis. Dial Transpl 7: 783-787, 1978.
131 Oreopoulos DG, Blair GRD, Meema HE, deVeber GA: Evolution of uremic neuropathy and
renal osteodystrophy in patients undergoing chronic peritoneal dialysis. Abstr. Am Soc Artif
Intern Organs 4: 47, 1975.
132. Roxe DM: Comparisons of maintenance peritoneal and hemodialysis. Dial Transpl7: 800-802,
1978.
133. Colotla VA, Campbell EM, Oreopoulos DG, Blair GRD, Fenton SS: Neuropsychological
compari~on of patients on home peritoneal dialysis (HPD) and on home hemodialysis (HHD).
Abstr Am Soc Nephrol, p 28, 1975.
134. Oreopoulos DG, Izatt S, Ogilvie R: Phosphate kinetics in patients undergoing peritoneal
dialysis. Abstr 3rd Int workshop on phosphate and other minerals, Madrid, Spain, 1977.
135. Oreopoulos DG, Rabinovich S, Meema HE et al.: Correlation of histological and radiological
assessment of ostietis fibrosa in patients with renal failure. Clin Res 18: 749, 1970.
136. Meema HE, Oreopoulos DG, de Veber GA: Arterial calcification in severe chronic renal failure
disease and their relationship to dialysis treatment,renal transplant and parathyroidectomy.
Radiology 121: 315-321, 1976.
137. Cangiano JL, Ramirez-Gonzales RE, Ramirez-Muxo 0: Bone disease and soft tissue calcifica-
tion in chronic peritoneal dialysis. Am J Med Sci 264: 301-306, 1972.
138. Karanicolas S, Oreopoulos DG, Dombros N, Pierratos A, Mathews RE, Vas S: Intestinal
obstruction and bowel perforation in patients undergoing chronic peritoneal dialysis, (Abstract)
16 EDTA meeting, Amsterdam, 1979.
139. Almy TA, Howel DA: Diverticular disease of the colon. N Engl J Med 302: 324-331, 1980.
140. Oreopoulos DG, Silverberg S: Calcium oxalate urinary tract stones in patients on maintenance
dialysis. N Eng J Med 290: 1438-1439, 1974.
141. Bommer J, Ritz E, Waldher WTR et al.: Urinary matrix calculi consisting of microfillar protein
in patients on maintenance hemodialysis. Kidney Int 16: 722-728, 1979.
466

B. CONTINUOUS AMBULATORY PERITONEAL DIALYSIS

1. Introduction

Continuous ambulatory peritoneal dialysis (CAPD) has been found to be as

effective as hemodialysis for maintaining patients with end-stage renal disease for
at least four to five years. Because of the many medical and social advantages of
CAPD, there has been, during the past five years, an exponential growth in the
number of patients so managed. As of 1983, over 10000 ESRD patients were
being treated with CAPD in the USA and Canada alone. In the process, we have
identified many technical and medical complications of the procedure and de-
vised means for their prevention and effective treatment. However, other com-
plications persist and are serious enough to force the discontinuation of CAPD.
This subchapter will describe most of the complications (other than peritonitis)
observed during CAPD and we will recommend ways of prevention on treatment
and, wherever known, will discuss in detail the pathophysiology of these com-
plications. The growth of CAPD has reduced sharply the number of patients
receiving maintenance intermittent peritoneal dialysis (IPD). As a result, since
our previous publication [1], only a few papers have appeared describing com-
plications of IPD.

2. Complications in patients on CAPD

Complications in CAPD may be grouped under five headings*

Complications associated with the peritoneal catheter
2. Complications due to the peritoneal dialysis solution
3. Complications due to increased intra-abdominal pressure
4. Metabolic complications
5. Persistence and/or progression of complications associated with renal failure

2.1. Complications associated to the peritoneal catheter

Access to the peritoneal cavity is obtained through a permanent indwelling

silastic catheter. The dialysis solution, contained in a plastic bag, is run through a
connecting tube and the catheter into the peritoneal cavity. During the early
development of CAPD, accidental contamination was frequent while making the
various connections in the system, and generally was attributed to the patient's
weakness, lack of coordination and/or poor vision. In recent years, modifications
in the connection technology such as luer-Iock connections [2], Titanium adapters
[3], the Beta-cap system [4], the Perugia system [5] and the Oreopoulos-Zeller-

* Authors wish to acknowledge the help of Dr. Zbylut Twardowski in suggesting the scheme of this
classification.
 467

man connector [6] have reduced dramatically the incidence of accidental contam-
ination.
Accidental contamination, if not treated, usually leads to peritonitis within 24
to 48 hours. Details of the protocol for managing accidental contamination may
vary from center to center but most recommend a tube change and a single dose of
broad spectrum antibiotic, either orally or intraperitoneally. Chapter 13 in this
book gives details concerning the management of this complication.

2.1.1 Direct catheter problems

The key factor in the success of long-term peritoneal dialysis is access to the
peritoneal cavity using an indwelling permanent and trouble-free catheter. The
introduction of CAPD for an increasing proportion of patients with end-stage
renal disease (ESRD), made even more essential a simple and safe access to the
peritoneal cavity.
Since Palmer et al. introduced their indwelling silicone-rubber peritoneal
catheter in 1964 [7, 8] which subsequently was modified by Tenckhoff [9], several
workers have proposed further modifications of the original design directed
chiefly at: (a) improving the anchorage of the catheter into the pelvis to prevent
its free movement inside the peritoneal cavity; (b) decreasing the frequency of
complications, such as one-way (outflow) obstruction; (c) minimizing the pain
induced by pressure from the catheter tip or due to the jet effect of dialysate
inflow; and (d) preventing periluminal or exit site infection, and dialysate leak.
Catheter-related abdominal pain may be due to: (A) pressure of the tip on an
internal organ, especially if the catheter is made of a stiff material; (B) a 'jet
effect' whereby the rate of inflow is rapid through a wide lumen, producing severe
inflow pain; and (C) inflammatory peritoneal reaction due to chemical irritation
or infection.
Catheter obstruction may effect either inflow or outflow. Inflow obstruction
develops when the operator produces an acute angle (kink) along the length of
the catheter during its insertion. This kink may occur either at the catheter's
subcutaneous portion or in the free intraperitoneal portion. Occlusion of the
lumen due to fibrin or blood clot also may impede dialysate inflow. Usually
omental wrapping does not cause inflow obstruction. Outflow or one-way ob-
struction (Le. failure to drain the dialysis solution) occurs when the catheter tip is
displaced away from the pelvis into the subhepatic or sub splenic space and/or
when omentum has wrapped around the catheter. Omentum can be 'sucked in'
through the small perforations along the terminal third (intra-abdominal) portion
of the catheter. To avoid this the catheter tip should be placed deep inside the
minor pelvis, away from the omental tissue.
To date, we have more experience with the Tenckhoff catheter than with any
other peritoneal access device. During the IPD era, the Tenckhoff catheter was
mainly used. However, with the growth of CAPD and the large numbers of
patients requiring peritoneal access, use of the Tenckhoff catheter has been
468

associated with a high incidence of complications such as intra-abdominal migra-

tion of the catheter tip, exit site leaks and infections.
Recently Slingeneyer et at. [10] reported their experience with 315 straight
Tenckhoff catheters in 247 patients maintained on IPD and CAPD between
September 1973 and September 1980. The cumulative duration of treatment was
410 patient-years of treatment. They observed the following catheter complica-
tions: bleeding into the subcutaneous tissue or peritoneum 1.9%, dialysate leak in
3.5%, and skin exit site infection in 10.5%. Skin exit site infection was more
frequent in diabetic than in non-diabetic patients. They reported a 5.3% inci-
dence of one-way obstruction requiring either catheter revision or replacement.
Subcutaneous cuff erosion necessitating cuff repositioning or catheter replace-
ment occurred in 2.2%. Fifteen patients (4.7%) had persistent localized abdomi-
nal pain resulting in either replacement [5] or revision (10 catheters). Incisional
hernias were observed in five patients. Cumulative catheter survival was 79.9% at
one year and 69.6% at two years. From this large experience they concluded that,
despite limitations due to exit site infections and one-way obstruction, the Ten-
ckhoff catheter provides adequate access for peritoneal dialysis. Most of Slinge-
neyer's patients were on IPD treatment. During intermittent peritoneal dialysis,
the peritoneal cavity is empty most of the time, whereas during CAPD, this cavity
is full nearly all the time. Therefore, catheter tip displacement is seen more
frequently in CAPD.
Rubin et at. [11] prospectively evaluated the complications encountered with
the Tenckhoff catheter in CAPD patients between August 1981 and May 1983.
They inserted 97 single-cuff catheters into 90 patients, and 118 double-cuff
catheters into 92 patients. Within 40 days of insertion of the single-cuff catheters,
25% had an associated complication that did not require catheter removal for
correction, and 19% had a complication that required catheter removal. With the
double-cuff catheters, 24% had an associated complication that did not require
catheter removal, while 28% had catheter-related problems that required re-
moval. In their long-term patients, the primary reason for catheter removal was
failure of peritonitis to resolve. Tables 1 and 2 show the early and late Tenckhoff
catheter complications reported by Rubin et at. [11], who had a catheter life span
of 38% at 22 months for both single- and double-cuff catheters (Fig. 1). Collec-
tively, for both single- and double-cuff Tenckhoff catheters, they reported an exit
site infection rate of 5.1%, an obstruction rate of 21.3% , and a leak rate of9.3%.
Due to its design, the intraperitoneal portion of the straight Tenckhoff catheter
is free to move about within the peritoneal cavity. Therefore, it is not surprising
with this design to find a high incidence of catheter displacement and one-way
obstruction. The curled catheter, the original Palmer design, which presently is
manufactured by Quinton, may overcome the problem of displacement (Fig. 2).
Recently Rottembourg et at. [12] described their large experience with this curled
catheter. It is inserted using a technique similar to that ofthe straight Tenckhoff
catheter. At the time of positioning the curled portion is kept straight with the
 469

help of a long obturator. Thus straightened, the catheter is advanced gently

towards the deeper pelvic area until it meets resistance or until the patient notes
rectal discomfort; then the obturator is slowly withdrawn and the catheter is
pushed downwards so that its curled section remains in the pelvis. Between
August 1978 and January 1980, Rottembourg inserted 48 straight Tenckhoff

Table1. Early Tenckhoff catheter complications (from Rubin et al. [11]; reproduced with permission)

Complication Single-cuff (N = 97) Double-cuff (N = 118)

Frequency Catheter Frequency Catheter

removed removed

Exit site infection 6 o 5

Gram (+) 4 3
Gram (-) 2 2
Peritonitis 10 o 14 2
Gram (+) 5 6
Gram (-) 3 3
No growth 2 2
Fungus 3
Obstruction 15 8 31 22
Leak 10 7 10 2
Leak with other complications 4 3 2 2

n = number of dialysis catheters

Table 2. Late Tenckhoff catheter complications leading to catheter removal in the CAPD patient
(from Rubin et al. [11]; reproduced with permission)

Complication Single-cuff Double-cuff

(N = 19) (N = 56)
number removed number removed

Peritonitis 7 12
Gram (+) 2 4
Gram (-) 2 4
No growth 1
Fungus 3 3
Infected subcutaneous catheter tunnel l' 5b
Days after insertion 235 ± 5()C 179 ± 52c
Median 215 264
Obstruction 5
Days after insertion 141 74±4c
Leak 1
Days after insertion 340 225

aAssociated with an infected exit site.

bFour infections were secondary to a gram-positive organism; one was secondary to a gram-negative
organism.
cMean ± SEM.
470

100

•• 80 \
q"
0
..J '0 ,
.,...
--
s:
0
.,
60

40
'a..
/ .........""'0
Single-cuff 'b-----
u
~ 20

0
140 280 420 560 100
Time in Days
Figure 1. Life span of single- and double-cuff Tenckhoff catheters over a 22-month period at the
University of Mississippi Medical Center (from Rubin et al. [11]; reproduced with permission) .

Figure 2. Different peritoneal catheters that are currently available for use. From top to bottom:
curled Tenckhoff, single-cuff Tenckhoff, Toronto Western , column-disc, Yali and Gortex catheters.
 471

catheters; these they compared with 95 curled catheters inserted between Febru-
ary 1980 and April 1983 . The most important difference between the two groups
was the incidence of outflow obstruction: of the straight Tenckhoff catheters,
41.6% became dislodged and 85% of these had to be replaced; on the other hand
only 10% of the curled catheters became dislodged and, of these, only 20% had to
be replaced. Except for peri-operative pain, which is higher with the curled
catheters, the frequency of other complications such as infection, dialysate leak-
age, exit site and tunnel infection and cuff extrusio~, were similar in the two
groups. For the straight Tenckhoff catheters, the cumulative catheter survival
was 65% at one year and 60% at two years; for the curled catheters, these rates
were 83% at one year and 78% at two years.
In 1976 Oreopoulos and Zellerman [13] designed the Toronto Western Hospital
catheter in order to overcome the problems of outflow obstruction and dialysate
leak observed with the Tenckhoff device. The main distinguishing feature of the
TWH type 1 catheter is the two flat silastic rubber discs (1 mm thick, 28 mm in
diameter, and 5 cm apart) on the intra-abdominal portion ofthe catheter (Fig. 2).
By preventing free movement of the tip within the peritoneal cavity, these flat
discs help to keep the catheter in the minor pelvis. In the TWH type 2 catheter,
two alterations were made at the level of the peritoneal Dacron cuff to reduce the
incidence of immediate and late dialysate leaks and the risk of incisional hernias.
First, a Dacron disc, 1 cm in diameter was added at the base of the cuff; this disc is
placed between the peritoneum and the fascia at the time of insertion to seal the
peritoneal incision. Second, a silastic ring was placed 1 mm distal to the Dacron
disc, thus creating a groove into which the surgeon can tie the peritoneum tightly.
Because of the presence of the silastic discs on the peritoneal portion of the
catheter, it is necessary to insert the Toronto Western Hospital catheter in the
operating room. The details of this implantation are as follows: a lateral trans-
verse lower abdominal incision is performed. The rectus fascia is incised transver-
sely. The rectus muscle is split longitudinally. The peritoneum is incised and,
under direct visualization and, manipulation the cannula is inserted, into the
pelvis. The peritoneum is closed around the cannula using Dexon 1 suture. The
muscle is dropped back over the Dacron disc and the catheter is brought out
through a separate stab wound in the anterior rectus sheath superior to the
transverse incision. Through a subcutaneous tunnel and a skin stab wound, the
catheter is brought out superior to the skin incision.
One-half hour before catheter implantation, the patient receives 1.7 mglkg of
Tobramycin and 1 g of Cephalothin intravenously as a bolus injection, to prevent
peri-operative infections.

2.1.2. Break in technique

During the immediate post-operative period, the catheter is checked for patency
by flushing the peritoneal cavity with 500 to 1000 ml of dialysate solution. Three
rapid (in-out) exchanges are carried out until the effluent return is clear of blood.
472
100

-it'0 8090
>
5
TWH-2(n=83)
.:;
~
::J 70
en

--
~

0
••
~

0
60
50
8 7

40
f I I I I
0
6 12 18 24
Months
Figure 3. One and two year cumulative catheter survival in the four groups of patients (TWH-l, TWH-
2, Tenckhoff medical insertion and Tenckhoff surgical implantation).

The first dialysis is delayed for at least 24 hours except in high-risk patients - those
on steroids, diabetics, mUltiparous women etc., in whom dialysis is delayed for
several days. During this period and until dialysis is begun, the catheter is
irrigated every 12 hours on the first day and daily thereafter with 10 ml of
heparinized (1000 Vlml) solution. After irrigation, the catheter lumen is filled
with 3000 units of undiluted heparin solution (1000 Vlml) and the external tip is
covered with a sterile cap. To avoid incisional leakage, every patient is main-
tained on intermittent peritoneal dialysis (IPD) with small volumes (500 ml the
first time and 1000 ml the second) twice a week for at least two weeks before
commencing CAPD training. During this period and between dialyses, the
catheter is filled with 3000 units of heparin solution as before.
A prospective controlled trial [13] in IPD patients at the Toronto Western
Hospital demonstrated the superiority of TWH catheter over the Tenckhoff
catheters and the Goldberg catheter* which has an inflatable balloon in the
middle of its intra-abdominal part. Sixty-three patients, 21 in each group, were
assigned in rotation to one of the three catheters. During the study, 28% of the
Tenckhoff, 19% of the Goldberg, and 8% of the TWH catheters had to be
removed because of outflow obstruction. One month after implantation, a radi-
ological study showed that 33% of the Tenckhoff, 20% of the Goldberg, and 7%
of the TWH catheters had migrated out of the pelvis.
A 1981 study [14] compared the survival rate and peri-operative and long-term
complications of TWH-1 and TWH-2 catheters with those of the Tenckhoff
catheters inserted medically or surgically. Most of these patients were on CAPO.
Figure 3 and Table 3 show the cumulative catheter survival rates at the end of the

• This catheter is no longer available.

 473

Table 3. One-year catheter survivals in the four types of catheters. The numbers indicate the catheter
studied and the numbers in parentheses, the percent survival at one year. TS = surgically implanted
Tenckhoff; TM = medically implanted Tenckhoff catheter

Catheters

TS TM TWH-l TWH-2 Total

All catheters 57 37 90 83 267

(76) (58) (74) (78) (73.7)
Catheters in diabetics 10 8 14 11 43
(73) (40) (83) (73) (67)
Catheters in patients over 60 11 11 36 39 97
(49) (70) (57) (82) (67)
First catheters 47 34 50 52 183
(82) (58) (79) (79) (76)

first and second year for the four groups. The TWH-2 catheters had a significantly
(p<0.05) better survival than the medically inserted Tenckhoff catheters, but
there was no significant difference between the other groups. Among the various
subgroups comparison of catheter survival data showed the following: a trend
towards lower survival of the catheters placed in patients older than 60 years
(67%) and diabetics (67%) compared to the younger group (78%) and non-
diabetics (75%) (Figs. 4, 5). However, these differences were not statistically
significant. Considered together all first-time catheters had significantly (p< =
0.05) better one-year survival- 76% than all second-time or subsequent catheters
- 65% (Fig. 6).
Tables 4 and 5 show the incidence of the various complications among the four
groups and their subgroups. Significant differences were observed only between
the following: catheters in all patients older than 60 years had a higher incidence
of leakage (42.3%) than younger patients (26.9%) (p<0.01). Also, catheters in
all women had a higher incidence of leakage (36.2%) than those in men (18.6%)
(p< = 0.01). Leakage was more common among second or subsequent catheters
(42.7%) than in the first catheters (26.9%) (p< = 0.01). Surgically inserted
Tenckhoff catheters had a significantly (p< = 0.01) lower incidence of leakage-
17.5% and exit site infection - 1.7%, than the TWH-1 catheter - 32.Z% and
12.2%, respectively. In the TWH-1 group, catheters in patients older than 60 had
a higher leakage rate - 5Z.8%, than those in younger patients in the same group-
18.5% (p< = 0.05).
In addition to the complications described in Table 4, we observed in two
patients with TWH-2 catheters, a bloody effluent, which lasted three to four days,
and in a patient with the TWH-1 catheter, a hematoma of the abdominal wall.
Five patients complained of abdominal or back pain but it always subsided, and
no catheter had to be removed because of this complication.
Using a lateral insertion 93 TWH-Z catheters were inserted in 90 patients
474

~
\
90
--~
\
i'
',--
\ (107)
\

80

.
(44)
.; (33)
(48,-........
~
70
~
....... _-+ 4-

..
CJ)
(35) --(21)............
.,
~
60
' (12)
.,
..
~
c 50
0
40 .--e
0-0 Catheters In Pts.<60(n=173)
Catheters in Ptl.>60(n=94)

1 I I I I
0 6 12 18 24
Months

Figure 4. Cumulative survival of catheters implanted in patients older vs younger than 60 years of age.

100 (224)
(43)

-iig
\
90 \
\
\ (138)
\
.s; 80 ~-~
~ (7),
" ~---A---A--~
:::I
en 70

--
~
ID
ID
.s=
60
un (5)

0
0
50 A--"
l::r-A Non-diabetiet (n=224)
Diabetic. (n=43)

0
t I I I I
6 12 18 24
Months

Figure 5. Cumulative survivals of catheters implanted in diabetics vs non-diabetic patients.

 475

-
•
\
90 \
\
~

a
80 ""
.>
> ""'e....
70 ~"''''
........
~
~ (42)

--
CJ)

.,.,
~ 60 (25) ...........
.....
(9)
.&;
50
a 0--0 First [Link] (n = 183)
0
40 .---e Second or Subsequent Catheters (n=84)
(p·O.05)
:(
I I I I
0 12 18
6 24
Months

Figure 6. Cumulative survivals of first catheters and second or subsequent catheters.

Table 4. Overall incidence of catheter·related complications. Numbers in parentheses indicate per-

centage. TS = surgically implanted Tenckhoff; TM = medically implanted Tenckhoff catheter

Complication Catheters

TS TM TWH-l TWH-2 Total

(N = 57) (N = 37) (N = 90) (N = 83)

Peri-operative pain 3 1 5 9
(8.1 ) (Ll) (6.0) (3.4)
Peri-operative infections 1 1 1 4 7
(peritonitis and/or (1.7) (2.7) (Ll) (4.8) (2.6)
skin exit infections)
Poor drainage 10 8 10 14 42
(17.5) (21.6) (ILl) (16.6) (15.7)
Leakage 10 9 29 24 72
(17.5) (24.3) (32.2) (28.9) (27)
Long-term. exit site and I 2 11 8 22
tunnel infections (1.7) (5.4) (12.2) (9.6) (8.2)
Extruded cuff 1 6 3 10
(1.7) (6.7) (3.6) (3.7)
One-way obstruction 10 6 9 8 33
(major cause of (17.5) (16.2) (10.0) (9.6) (12.4)
catheter removal)
476

Table 5. Incidence of leakage in the various subgroups. Numerator indicates the number of catheters
that leaked and the denominator the total number of patients in each subgroup. Number in paren-
theses indicates percentage. TS = surgically implanted Tenckhoff; TM = medically implanted
Tenckhoff catheter

Catheters

TS TM TWH-l TWH-2 Total

Diabetics 3/10 2/8 8/14 2/11 15/43

(30.0) (25) (57.1) (18) (37.9)
Patients over 60 years 3/10 3/11 19/36 16/39 41197
(27.3) (27.3) (52.8) (41) (42.3)
Females 7/27 4/15 20/51 15/34 461127
(25.9) (26.7) (39.2) (44.1) (36.2)
Second and subsequent 3111 18/40 14/31 35/82
catheters (27.2) (45.0) (45.2) (42.7)

between February 1982 and June 1983. These 93 patients were maintained on
either IPD - twice a week, 20 hours each time, or CAPD - three or four bag
exchanges/day. At the end of 6 and 12 months, cumulative catheter survival rates
were 95% and 87%, respectively (Fig. 7). In calculating survival, the removal of
catheters because of outflow obstruction, skin exit site infection or persistent
peritonitis, were considered as 'end events'. Functioning catheters in patients
who died or underwent transplantation were considered as 'lost to follow-up'.
During a 16-month period, nine of these 93 catheters were removed. Three of
these nine failed because of one-way obstruction from the outset. These three
were implanted in three patients who had had previous abdominal operations
with numerous intra-abdominal adhesions; one had a perforated appendix with
fecal peritonitis, the second had had a cystectomy with ureteric diversion through
the ileal conduct, and the third returned to peritoneal dialysis after renal trans-
plantation had failed. During the course of the study, 22 patients were withdrawn
from peritoneal dialysis with a functioning catheter.

2.1.3. Post-operative leak

Twelve of the 93 catheters developed a post-operative dialysate leak through the
incision during the first dialysis (incidence = 12.9% ). These patients had a variety
of associated conditions such as diabetes mellitus, many previous pregnancies,
were receiving steroids or were older than 60 years. In every case, our manage-
ment of early dialysate leak (see below) was successful, and no catheter failed
because of this complication. One diabetic patient experienced leakage from the
exit site six months after catheter implantation and his exit site became infected
with Staphylococcus aureus. Eventually, peritonitis developed involving the
same organism, and the catheter had to be replaced.
 477

(93)
(58)
100 (31)

....... 80
(23) (12)

'ii
.>...> 60
:J
en 40
....
...
II

.,
II
.s: 20
()

0
3 6 9 12
Duration of Treatment
(Mos)

Figure 7. Cumulative catheter survival rates for TWH-2 at the end of 6 and 12 months using lateral
insertion technique (Feb. 82-June 83).

2.1.4. Management of early and late dialysate leaks

If this complication occurs during the first dialysis, peritoneal dialysis should be
discontinued. Three to four days later, dialysis is started again with a small
volume, i.e., 500ml. Late dialysate leaks are managed by discontinuing CAPD
temporarily and maintaining the patient on IPD for two weeks. If the leak recurs
the catheter should be replaced electively.

2.1.5. Skin exit site infection

Nine catheters (9.6%) were associated with persistent exit site infection. Five of
these were removed when peritonitis developed. One catheter was removed
electively owing to persistent infection with S. aureus. The remaining three
infections responded to treatment as described below. During the 16-month
follow-up period, we observed neither cuff extrusion nor incisional hernias.

2.1.6. Management of skin exit site infection

Development of infection (i.e., redness, induration or purulent discharge) is
managed by antibiotics administration and one of the following measures: 1) daily
dressings with povidone-soaked gauze placed around the exit site; 2) daily
showers followed by application of a dressing; or 3) daily showers with frequent
paintings of the exit site with povidone iodine. Occasionally, shaving away the
infected subcutaneous cuff may help clear the infection [14a]. If peritonitis with
the same organism develops, or if the exit site infection persists despite these
measures, the catheter should be removed and replaced at the opposite site three
to five days later.
478

Our experience with the Toronto Western Hospital catheter supports the
following conclusions: 1) use of the paramedian approach for catheter insertion
has further improved the cumulative catheter survival and reduced the incidence
of post-operative dialysate leaks; 2) use of the TWH-2 catheter has practically
eliminated the incidence of outflow obstruction; and 3) this catheter design has
not decreased the incidence of exit site infection and tunnel abscess.
Grefberg [15] of Sweden reported their comparative experience with Ten-
ckhoff and Toronto Western Hospital catheters. Catheters were randomly se-
lected and both were surgically inserted. Fifty-nine Tenckhoff catheters were
observed for 592 treatment months and 24 TWH catheters for 220 treatment
months. At 18 months, the cumulative life span of both catheters were similar at
80%. With regards to complications, 11 of the 59 Tenckhoff catheters become
obstructed as opposed to one of 24 TWH catheters. The Swedish workers believe
this high incidence of Tenckhoff catheter blockage was due to inexperience and
that this complication would disappear with experience. With regard to exit site
infection, tunnel abscess and dialysate leak, there was no difference between the
two groups. Despite its obvious advantages, they abandoned the use of the TWH
catheter because laparotomy was needed whenever the catheter was removed,
and because the bowel was perforated during the removal of two TWH catheters.
Recently Hogg et al. [16] used the TWH catheter with considerable success in
children on long-term CAPD. Six of the TWH catheters were inserted in children
who earlier had had either obstruction or leakage with one to four Tenckhoff
catheters. Overall, they used fifteen TWH catheters in 12 children and compared
the results to 23 Tenckhoff catheters in nine children. The rate of obstruction with
TWH catheters (7% ) was much lower than that with Tenckhoff catheters (45%).
Problems of slow outflow, one-way obstruction and tip displacement from the
pelvis, which are inherent in the Tenckhoff catheter and its modifications, led
Ash et al. [17] to design the 'column-disc' catheter (Fig. 2). This catheter was
made of a pair of silicone discs, 13/4 inch in diameter, which are separated by 16
silicone columns - 1/4 inch high. A silicone drainage tube enters at the center of
one disc into the space between the discs. Two pieces of Dacron felt, which are
wrapped around the catheter to form 'cuffs' similar to those on the Tenckhoff
catheter, are attached with silicone adhesive. One cuff is placed 1f4 inch from the
base of the disc, and the other approximately 41/2 inches further away (Fig. 2). A
recent publication [18] has summarized a multicenter experience with the 'col-
umn-disc' catheter. A total of 89 column-disc catheters were placed at three
centers. Twenty were placed in patients with previous failures of Tenckhoff
catheter. Table 6 shows the number of failures of column-disc catheters and their
causes. Outflow failure, which is the main cause of early failure is less frequent
after one month. Similarly, subcutaneous leak and pericatheter herniation occur
rarely, if at all, after one month. Life-table analysis (Fig. 8) revealed that
compared to Tenckhoff catheters, the 'column-disc' catheter is more likely to fail
in the early months but over the long-term is much less likely to fail.
 479

Table 6. Nature of complications leading to removal of column-disc and Tenckhoff catheters (from
Ash et al. [18]; reproduced with permission)

Column-disc catheters (89) Tenckhoff catheters (61)

Early «1 mo)Late (>lmo) Early «1 mo)Late (>lmo)

Exit site or subcutaneous infections 0 0 3

Recurrent peritonitis 0 9 1 6
Outflow failure 6 2 2 3
Subcutaneous leak or hernia 0 0 0 3

Another advantage of the column-disc catheter is that the dialysate flows out
much more rapidly. In a study of catheter function in Montpellier [18], three
patients who had Tenckhoff catheters and who had persistent pain during infu-
sion, were fitted with the column-disc catheter. The outflow rate with the new
catheters was 11 to 37% faster than that with the previous catheter for the first six
minutes of outflow. The decrease of inflow pain with the column-disc catheter
probably relates to the diminished 'jet effect' of fluid entering the abdomen and
its direction along the parietal peritoneum.
Overall, the preliminary experience with the column-disc catheter appears
promising. It has been beneficial in high-risk patients; however, its advantages in
an unselected population remains to be established.
It is not clear why some patients with peritonitis refractory to appropriate
antibiotic therapy respond promptly to catheter removal. Some have attributed

1.0 ~ Tenckhoff
....
iL .---. Column-Disc Cothet.r

.ae
( ) No. [Link] Mol at Risk
0.8
~ (4)

j 0.6

8
---
Column-Disc
.................- .............-e-e-e
~ 0.4 / (4) (2) (2) (I)

~
>-
:a 0.2
...
(15)

.8
e
0..

2 6 10
MOl After Implantation
Figure 8. Life-table analysis of probability of catheter failure for Tenckhoff and column-disc catheters
(from Ash et al. [18]; reproduced with permission).
480

this to 1) the presence of infected tissues or clots within the peritoneal catheter, 2)
proteinaceous material containing bacteria and coating the catheter, 3) infected
internal cuffs, 4) associated occult tunnel infections, or 5) foreign body effects.
Catheters removed from patients with refractory peritonitis have shown a variety
of structural defects [19]; small cracks in the silastic may harbor organisms [19].
Indium III scans have been used to identify catheter tunnel infections [20]. Often
it is difficult to determine whether an exit site infection involves only the skin
surrounding the exit, or whether it extends deeper into the tunnel. We need
further experience with Indium scans. The Gortex peritoneal catheter has a
Gortex disc attached to the catheter immediately under the skin; the disc is
alleged to promote skin ingrowth and adherence of cells to the prosthesis and seal
the exit site [21]. To date experience with this catheter is limited. Other reports
have described new methods of catheter placement said to prevent displacement
[22-25].

2.2. Complications due to the peritoneal dialysis solution

These complications include peritoneal eosinophilia, peritoneal fibrosis and ul-

trafiltration failure, and sclerosing peritonitis. Elsewhere in this book Vas will
discuss the first complication in detail.

2.2.1. Peritoneal fibrosis and ultrafiltration failure and sclerosing peritonitis

The peritoneal membrane is a delicate structure composed of capillaries, intersti-
tial tissue and a single layer of mesothelial cells. Its primary function is to line the
abdominal wall and the surface of the intra-abdominal organs to provide a
smooth surface. It was not designed for peritoneal dialysis. Early experience with
acute peritoneal dialysis suggested that this membrane could be used for this
purpose without serious side effects. This experience prompted pioneers to
attempt chronic peritoneal dialysis and surprisingly, in most cases, the membrane
continued to provide this service for long periods. However, as the experience
with chronic peritoneal dialysis increased, workers began to report that, under
certain conditions, the peritoneal membrane may fail. Thus, an episode of severe
peritonitis may provoke the formation of adhesions, which produce loculation of
the fluid and mechanical interference with its flow. The adhesions appear to be
localized and various centers have reported that peritoneal dialysis can be re-
sumed after surgical lysis. In addition to adhesion formation, peritoneal thicken-
ing is a complication of chronic peritoneal dialysis [26]. However, most patients
with a thickened peritoneum do not have any difficulty with peritoneal dialysis.
The syndrome of ultrafiltration failure began to emerge when we described a
patient who was unable to continue on peritoneal dialysis after a period of eight
years [27]. In this condition, solute (urea-creatinine, etc.) transport remains
intact but when ultrafiltration of water fails further peritoneal dialysis is
impossible.
 481

Faller and her colleagues from Colmar, France [28] first expressed concern
about gradual failure of ultrafiltration in CAPD patients. They observed that,
with time on dialysis, 22 patients on long-term CAPD needed an increasing
number of hypertonic exchanges and eventually after 30 months all required four
hypertonic exchanges daily. While other centers in France - Montpellier (Mion),
Paris (Legrain), confirmed this observation North American centers with equal
or even larger and longer experience with CAPD reported this phenomenon only
infrequently.
This complication of peritoneal-membrane thickening and ultrafiltration
failure became even more disturbing when French investigators [29] described
the syndrome of progressive sclerosing peritonitis in patients on chronic intermit-
tent or continuous ambulatory peritoneal dialysis. Usually this syndrome de-
velops in patients who discontinue peritoneal dialysis because of ultrafiltration
failure and transfer to hemodialysis. While on hemodialysis and after periods of a
few to 20 months, the patients develop nausea, vomiting, malabsorption and
eventually complete bowel obstruction. At operation the peritoneal membrane is
greatly thickened and encases all the bowel loops in an orange-like mass. The
loops are released only with difficulty and most of the patients die. Surprisingly,
no one has been able to demonstrate a definite correlation between this syndrome
and the type of dialysis (IPD or CAPD), or the number and type of previous
episodes of peritonitis or the use of any particular drug.
In North America this syndrome was reported first by Gandhi et al. [30] in five
patients on chronic intermittent peritoneal dialysis with the reverse osmosis
machine. The two IPD patients described by Backenroth-Maayan et al. [31] from
New York may have been similar. However, in North America this complication
has rarely, if ever, been observed among CAPD patients.
Our search for an understanding of membrane failure is complicated by the
observation that solute transport remains intact at a time when water removal is
impaired. In order to understand the pathogenesis of ultrafiltration failure, we
studied four patients with this complication and six controls. Hypertonic (4.25%)
Dianeal was left in place for six hours. The average amount of ultrafiltration
(expressed as percentage of instilled volume) was 9.5% (i.e. absorption) for the
patients, and + 32.4% for the controls. Two of the patients showed a rapid drop in
dialysate glucose and osmolality to levels lower than the controls. In contrast
dialysate glucose and osmolality in the other two remained at much higher levels
than the controls. The average dialysate/plasma ratio of urea and creatinine was
1.0 in the first two patients, and in the other two was 0.84 and 0.75 respectively. It
appears that two mechanisms operate in this complication: 1) rapid absorption of
glucose resulting in rapid dissipation of the osmotic gradient, 2) decreased
glucose, solute and water movement because of loss of peritoneal permeability or
surface area.
Chief among the possible mechanisms underlying ultrafiltration failure is the
loss of the ability to maintain an osmolality gradient across the peritoneal mem-
482

brane secondary to rapid glucose absorption. Across a perfectly semipermeable

membrane, a difference of one milliosmole will produce enough force to maintain
a 19 mm column of mercury. Since the peritoneal membrane is not perfectly
semipermeable with regard to solutes present in the peritoneal fluid (Le. they are
absorbed), we must apply a correction factor in calculating osmolality gradients.
The driving force for glucose depends on its concentration in the peritoneal
dialysate. Thus, for a 1.5 g/dl solution, the osmolality gradient is 78 milliosmoles
(1482mmHg) and for a 4.25g/dl solution it is 230 milliosmoles (4381mmHg).
These maximum values are never reached because of the rapid absorption of
glucose from the peritoneal cavity into the blood stream. Figures 4 and 5 in
chapter 5 show that ultrafiltration proceeds rapidly for about 100 to 160 min
following a 1.5% and 4.25% solution respectively. The rate then falls as the
osmolality approaches 300mOSm. The maximum volume of ultrafiltrate re-
moved with a 4.25% solution is about 1300 ml, and with a 1.5% solution about
200 ml though there are individual differences. It is clear that rapid dissipation of
the glucose gradient will reduce this rate with loss of ultrafiltration, as may be
seen without peritonitis. The temporary loss of ultrafiltration and rapid glucose
absorption, seen during peritonitis, support this contention. Also it is known that
hypertonic solutions induce vasodilatation [33]. The osmolality of most solutions
commonly used for peritoneal dialysis exceeds that of the plasma. Acetate in
concert with high osmolality may contribute to the vasodilatation because acetate
reduces the peripheral vascular resistance [34]. It does not appear that a change of
pH from 5.8 to 7.4 has any effect on vasodilatation [33]. *

2.2.2. The acetate story

Most of the dialysis solutions in France use acetate as a buffer whereas those in
North America contain lactate. Is acetate responsible for all or at least part of the
ultrafiltration failure reported from France? The five patients with sclerosing
peritonitis reported by Gandhi et al. [30] from Chicago and the two patients of
Backenroth-Maayan et al. [31] were dialyzed with acetate solution made with the
reverse osmosis machines; these machines are sterilized with formaldehyde and it
is possible that this substance may contribute to this complication. Further, one of
our patients [27], who failed to 'ultrafilter' after eight years of IPD, also was
dialyzed with a solution containing acetate as a buffer. Some patients with
progressive sclerosing peritonitis [35] were dialyzed with lactate-containing solu-
tions but all of them had severe peritonitis.
We believe that the cause of ultrafiltration failure and sclerosing peritonitis is
multifactorial. The multicenter cooperative trial currently underway should shed
light on this serious problem.

* This paragraph is reproduced with permission of M.A. Manuel et al. (33a).

 483

2.3. Complications due to increased intra-abdominal pressure

These include dialysate leaks, hernias, and cardiopulmonary compromise.

2.3.1. Dialysate leaks

External dialysate leaks are common in high-risk patients and when an inap-
propriate break in technique is applied after catheter insertion, as discussed
earlier in the section on catheter-related complications. However, both external
(through the skin exit site) and internal leaks i.e. infiltration of dialysis fluid into
the abdominal wall, penis, scrotum and vulva may occur at any time after catheter
implantation. The most common site of leak is around the catheter entrance into
the peritoneum, but leaks may also occur through a previous surgical incision, or
via pinholes too small to localize.
The sudden reduction in the volume of dialysate effluent and/or sudden and
rapid weight gain may give early evidence of internal dialysate leak. When the
dialysis solution leaks into the layers of the abdominal wall, and diffuses evenly
throughout, it may be extremely difficult to differentiate it from the abdominal
wall edema which is part of systemic edema. In such situations, ultrasound and
radiographic examinations including the use of contrast material into the dialy-
sate are of little value in localizing the leak. However, computerized tomography
aided by contrast material in the dialysis solution has been successful [36] (Fig. 9).
Twardowski et al. [36] recommend these steps before undertaking the study: (a)
the contrast material should be well mixed in the dialysis fluid, (b) encourage the
patient to increase the intra-abdominal pressure by any activity so as to increase
the fluid collection, (c) CT images should be made one or two hours after the
injection of contrast.
The incidence of late dialysate leaks has ranged between 2-10% [14]. Our
experience suggests that patients over age sixty, multiparous women, diabetics
and those receiving steroids are more prone than others to dialysate leaks.

2.3.2. Massive hydrothorax

A rare but serious complication is trans diaphragmatic dialysate leakage into the
pleural cavity. In 1967 Edwards and Unger [37] described the first massive
hydrothorax during peritoneal dialysis, and since then many more cases have
been reported in both adults and children. Almost all have been on the right side,
and most, with some exceptions [46, 48, 50], are discovered within hours to a few
days after the onset of peritoneal dialysis. The composition of the pleural fluid is
similar to peritoneal dialysate. Glucose concentration of the pleural fluid is very
high and protein content low. However, if the composition of pleural fluid is to be
of any diagnostic assistance one must obtain the fluid from the pleural cavity as
soon as the complication is noted. If the pleural tap is delayed the composition of
the fluid will alter as a result of equilibration of solutes across the pleural
membrane.
484

Figure 9. A.C.T. image through the pelvis shows extravasation of specified fluid into the abdominal
wall. The peritoneal cavity herniates anteriorly on the left and suggests the side of the tract connecting
to the abdominal wall fluid collection (from Twardowski et al. [36]; reproduced with permission).
 485

Other diagnostic tests have been recommended to confirm the diagnosis of

hydrothorax due to pleuroperitoneal communication. Injection of methylene
blue into the dialysate (20 mg/liter of dialysate) promptly colors the pleural fluid
as well. However, rapid absorption of methylene blue through lymphatics may
also stain the pleural fluid. Although the methylene-blue test may demonstrate
that the pleural fluid is coming from peritoneal dialysate, it does not localize the
route of communication. Moreover, in several patients the intraperitoneal use of
methylene blue has been associated with intense peritoneal reaction and chemical
peritonitis [52]. Therefore, this test is not recommended for this complication.
Spadaro et al. [53] demonstrated pleuroperitoneal communication by adding
Technetium TC 99 m tagged human albumin to the dialysate and carrying out
radionucleide imaging with a large Gamma camera. Thirty minutes after instilla-
tion, the isotope was detected in the right hemithorax confirming the presence of
a communication. This technique is quick, painless, and non-invasive. Since then,
radionuclide scan has been used by others to demonstrate pleuro-peritoneal
communication [56]. Spadaro et al. [53] recommends the following technique:
lyophilized macro aggregated human albumin, reconstituted by the addition of a
10 mci dose of sodium pertechnetate 99 m TC, is administered intraperitoneally
through a Tenckhoff catheter and flushed with 10 ml of 1.5% dialysis solution.
After the radionuclide has been distributed throughout the peritoneum one liter
of 1.5% dialysate solution is administered, followed by a second liter 1 hour later.
The spread of the fluid is monitored continuously with a Searle LFOV camera
and cathode ray tube. Four hours after radio nuclide administration, simul-
taneous samples of blood, pleural fluid and peritoneal fluid are obtained and
count rates are done. Pleural fluid counts are about 24 times higher than blood or
background counts.
The pathogenesis of these massive effusions is still unclear. Enhanced move-
ment of peritoneal fluid to the pleural space via the diaphragmatic lymphatics, or
chronic congestive heart failure and increased pulmonary-artery pressure usually
produce small bilateral pleural effusions but, by themselves, do not explain the
acute hydrothorax under discussion here.
The acute onset and equally rapid resolution of hydrothorax after discontinua-
tion of peritoneal dialysis and other changes mentioned above indicate a signifi-
cant anatomical communication between pleural and peritoneal space. In ad-
dition, there is probably a pre-existing structural defect in the diaphragm. It has
been proposed that, during the first trimester of gestation, a congenital diaphrag-
matic hernia develops through the pleuroperitoneal canal (foramen of Boch-
dalek). Patients on dialysis, who develop acute hydrothorax may have this
anomaly but the literature does not give surgical or autopsy evidence of its
existence. Rudnick and his colleagues suggest that the defect probably is too small
to detect on gross examination [58]. Patients with cirrhosis and massive ascites
may have a pleural effusion, which is believed to be part of peritoneal fluid [59].
At autopsy in such patients, Lieberman et al. [59] could find no gross anatomic
486

defects but saw several small diaphragmatic fluid-filled blebs with overlay separa-
tions of collagen bundles in the tendinous diaphragm. Recently Grefberg et al.
[57] described an autopsy in a patient with right-sided hydrothorax complicating
peritoneal dialysis. Before the autopsy was begun, the peritoneal cavity was filled
with dialysis solution stained blue with methylthionine. Careful dissection re-
vealed about 20 defects in the central tendon of the right diaphragm (Fig. 10).
Some of the holes were partly covered by a thin membrane, which histologically
consisted of peritoneum and pleura but lacked the tendinous part of the central
tendon. They speculated that the acute and late onset of this complication was
due to congenital and acquired defects.
Small effusions need no treatment except observation. However, most au-
thorities recommend stopping peritoneal dialysis in the presence of massive ,
effusion and respiratory distress, and also transfer to hemodialysis. On stopping
the PD, the fluid would be reabsorbed, however if the patient exhibits respiratory
distress acute thoracentesis may be necessary. When for medical reasons, hemo-
dialysis is not feasible it becomes extremely difficult to manage patients with
acute hydrothorax. In the past, several such patients have died in uremia [37,41].
Pleurodesis has been proposed to prevent recurring pleural effusions [41] and, of
the several available methods, talcage and tetracycline seem to be the most used
(46,60). Whenever one selects pleurodesis, thoracoscopy should be done be-
forehand in an attempt to detect anatomical defects. Some patients who have
developed this complication while on CAPD have been maintained on IPD [56].

Figure 10. Autopsy finding in a case with pleuro-peritoneal communication during peritoneal dialysis
(from Grefberg et al. [57); reproduced with permission),
 487

Available experience suggests that acute hydrothorax developing in patients on

peritoneal dialysis almost always appears on the right side. The pleuroperitoneal
communication may be through either congenital or acquired anatomical defects.
The appropriate treatment seems to be stopping peritoneal dialysis and con-
tinuing such patients on hemodialysis. In special cases where hemodialysis is not
feasible, pleurodesis may be considered.

2.3.3. Abdominal hernias

The presence of two liters of dialysate in the peritoneal cavity produces an
increase in intra-abdominal pressure, and Gotloib et at. [61] demonstrated a
positive correlation between the volume of the intraperitoneal solution and the
intra-abdominal pressure. Thus, it is not surprising that patients on CAPD are
prone to develop hernias, particularly those with weak areas in the abdominal
wall from previous pregnancies or old age.
At the Toronto Western Hospital, the incidence of hernias among CAPD
patients was 11.5% [62]. The mean age of those developing hernias was signifi-
cantly higher than those who did not develop hernias. The group with specific
primary kidney disease did not show predisposition to develop hernia. Approx-
imately one half of the hernias developed through a healed (midline) incision at
the site of the catheter insertion. Two thirds of patients with incisional hernias
had a temporary dialysate leak at the time of catheter insertion. The most
common initial presentation was painless swelling. A few patients presented with
acute bowel obstruction. Those with a previous hernia repair tended to develop
hernia again but at a different site. About one half of the patients who developed
hernia had had previous abdominal surgery other than catheter insertion. On the
average women developing hernia had had about three pregnancies. There seems
to be no correlation between time on dialysis and the development of hernia.
There was a predominance of women among those who developed hernias
(73%). Similarly, of all patients in the CAPD program at the Toronto Western
Hospital, 18% of the women and 6% of the men developed this complication.
Hernia at the catheter insertion site was common in susceptible individuals
probably because the catheter was inserted through an incision in the midline - a
weak area of the abdominal wall, which over time yields to the continuously
increased pressure. For this reason we abandoned the midline as a site of catheter
implantation and now use a paramedian incision through the rectus muscle [63].
Since then, the incidence of dialysate leak has been reduced, and also incisional
hernias have become uncommon. In our experience, the development of a hernia
does not necessitate discontinuation of CAPD; all but two of our patients
returned to CAPD after a temporary interruption of 10 to 15 days for surgical
repair. During this period, these patients can be maintained on intermittent
peritoneal dialysis with small volumes or hemodialysis through a subclavian
catheter. Our policy for the management of hernias is a.s follows: when the hernia
is recognized before CAPD is inititated the hernia is repaired promptly but
488

electively unless the patient presents with acute bowel obstruction and then an
emergency repair is carried out. Patients in whom an operation is not con-
templated or is contraindicated should be fitted with a special supportive corset,
similar to that used by pregnant women, and observed closely for signs of
incarceration or bowel obstruction.
Hernias and related complications are reported frequently in CAPO patients
[64-73]. Unusual sites of such hernias include: Richter's hernia [66], obturator
hernia [72], hydrocele [73], hernia of Morgagni [64], incisional hernias [69] and
diaphragmatic hernia [69]. Hiatus hernia may become symptomatic following
CAPO therapy [62].
In summary, hernias should be looked for carefully and repaired before
peritoneal dialysis is begun and, if found during CAPO, should be repaired
electively. Almost all of these patients should be able to return to CAPO after the
repair.

2.3.4. Cardiopulmonary compromise

The infusion of dialysis solution into the peritoneal cavity increases the intra-
abdominal pressure, indeed Twardowski et al. [74] have shown a linear relation
between intraperitoneal volume and pressure in the supine, upright and sitting
positions. The pressures are higher in the upright and sitting positions than in the
supine. Also they showed that each liter of intraperitoneal fluid increases the
pressure by 2 to 3 cm water. Every movement increases intra-abdominal pres-
sure, especially in the vertical position. Weight lifting or jumping may raise the
pressure to 200mm Hg.
In most CAPO patients the vital capacity does not change or increases only
slightly with progressive increases in intra abdominal volume. At the Toronto
Western Hospital [75] pulmonary function was measured in eight patients under-
going CAPO, using 2 liters of fluid in all but one who was receiving 3 liters. All
were free of lung disease and were at stable dry weight. Lung volumes were
measured by body plethysmographic and helium dilution techniques; forced vital
capacities were measured before and after removal of dialysate. After drainage
functional residual capacity was significantly higher (mean 590ml) in eight pa-
tients using 2 liters of fluid. There were no significant differences before and after
drainage in maximum expiratory flow rates. From this study we concluded that
peritoneal dialysis exerted no detrimental effect on diaphragmatic configuration
or function.
Rebuck speculated that Starling's law applies to the diaphragm as it does to the
heart; that is, as the distending pressure becomes excessive, the diaphragm will
no longer be able to increase its contractible force [76]. Hence, depending upon
the integrity of the diaphragm and intercostal muscles, some patients should be
able to tolerate larger volumes of abdominal fluid better than others. We had an
opportunity to test this hypothesis in 17 CAPO patients using 3-liter volumes.
Spirometric studies were done in the sitting position. Lung volume was measured
 489

with the abdomen empty and then at progressively increasing intra-peritoneal

volumes of 2, 3 and 4 liters. Figure 11 shows that for the 17 patients the mean
percent prediction of FEVt , did not change significantly at 2-liter volumes.
Thereafter, it decreased exponentialy with three and four liters. However, even
at 4 liters, the mean predicted FEVt was in the normal range. Based on the FEVt
changes seen with increasing volumes, these patients could be separated into
three groups:
Group I (N = 4), FEVt increased significantly up to 3L volumes and decreased
slightly at 4 liters (Fig. 11). Group II (N = 8), no change in FEVt up to 2L and then
decreased gradually; Group III (N = 5), no change in FEVt even up to 4 liters. It
is interesting to observe that, in nine patients of Groups I and III, FEVt did not
change or else increased when intra-abdominal pressure was raised gradually
with increasing volumes of fluid. These findings in nine patients tend to support
Rebuck's hypothesis that diaphragm muscle functions in a fashion similar to that
of heart muscle; up to certain distending pressure, the diaphragm will increase its

/\4
120
All Patientl Group I

>
~
1o
110
~
l 100

90

0 I 2 3 4 o I 2 3 4

130

Graupm
120 N-S
>
1&1

..
IL

1 110
0
:;

..•
!
Il. 100

.
c:
0

:. 90

80
0 2 3 4 o I 234

Figure 11. Changes in predicted percent FEV) with progressive increase in intraperitoneal volume in 17
patients. Group I (N = 4): increase in FEV) up to a certain volume and then decline. Group II (N = 8);
progressive decline in FEV). Group III (N = 5); no change in FEV) up to 4-liter volume.
490

force of contraction, then when the pressure becomes excessive, the function
declines.
In a similar study, Twardowski et al. [74] showed that in most CAPD patients,
the vital capacity does not change or increases only slightly with moderate
volumes of intra-abdominal fluid. In the supine and sitting positions, the vital
capacity decreases only with intra-peritoneal volumes in excess of 3L and 4.5L
respectively. In the supine position and with intraperitoneal volumes exceeding
2.5L, some patients have severe dyspnea and show a dramatic deterioration in the
vital capacity. For similar levels of intra-abdominal pressure, patients had a
decrease in forced vital capacity and others actually had an increase. Probably
these variations can be explained by variations in the initial and final radii of the
diaphragm conforming to Starling's law as suggested by Rebuck.
Thieler et al. [77] studied the FVC in CAPD patients after infusion of 2 liters of
dialysate compared to that with the abdomen empty. They found a decrease of
3.4% only in the supine position. Up to volumes of 2 liters they observed no
significant deterioration or improvement in pulmonary function in the sitting and
upright positions. In four IPD patients studied in a semirecumbent position
Gotloib et al. [61] found a 62% reduction in FVC with 2.5 liters of intraperitoneal
fluid. Other studies reported a much smaller reduction in FVC in IPD patients
[38, 78, 79]. It is apparent that patients treated with CAPD have much better
pulmonary function with intraperitoneal fluid than those treated with IPD. Some
of the decrease in pulmonary function noted in IPD patients probably reflects
their inefficient control of fluid as compared to CAPD patients.
Besides altering pulmonary function, elevated intra-abdominal pressure may
also induce changes in cardiac function. In cirrhotic patients drainage of ascitic
fluid has produced dramatic improvement in cardiac function, concomitant with a
fall in the right and left atrial pressure [80]. However, after 3 liters of dialysate
infusion, Schurig et at. [81] did not find a significant change in the right atrial
pressure, pulmonary artery pressure and cardiac index. Others have reported up
to a 20% decrease in cardiac index with 2 liters of intraperitoneal fluid [82, 83].
We need more detailed studies of cardiac function in CAPD patients.
In summary, raised intra-abdominal pressure induces changes in pulmonary
function only beyond a critical point. The supine position seems to exert an
adverse influence on pulmonary function even at a low intra-abdominal pressure.

2.4. Metabolic complications

2.4.1. Amino acids and protein losses

Some CAPD patients develop malnutrition through a combination of decreased
appetite, and hence decreased intake of nutrients, aggravated by an increased
loss of nutrient in the dialysate. In 13 patients whose daily protein and caloric
intake at the beginning of CAPD was 1.5 g per kg, and 35 kcal/kg respectively,
 491

intake decreased spontaneously after one year on CAPD even though they were
urged during clinic visits to maintain the initial protein and caloric intake (Table
7) [84]. CAPD patients decrease their food intake, presumably because of a
decrease in appetite and a feeling of fullness, although such factors are difficult to
quantify.
The decrease in appetite results in a reduced intake of many nutrients including
protein. This combined with a significant dialysate loss - 6-12 g of protein per day
mostly as albumin [85] may induce protein malnutrition. Low serum albumin and
transferrin have been reported frequently in CAPD patients [84]. In 20 CAPD
patients Heide et at. [86] found that total body nitrogen, when measured by
neutron-activation analysis, had decreased over a period of two years. Sur-
prisingly, total body potassium, which some believe is a good index of lean body
mass, increased at a time when total body nitrogen was decreasing [86]. We
believe that only in the healthy state is total body potassium a good index of lean
body mass and that in disease states, like malnutrition, a better index of the long-
term nitrogen balance is total body nitrogen.
If poor appetite combined with high protein losses leads to malnutrition in a
CAPD patient, we must find another way (than oral) to increase protein intake;
for example, we can add protein to the dialysate as amino acides (Amino
Dianeal) [87].

2.4.2. Daily amino acid losses and plasma amino acid abnormalities in CAPD
patients
In six non-diabetic patients on CAPD, we found an average daily amino acid loss
of2 g per day [88], findings which agree with those of previous studies [89-91]. On
a weekly basis, the amino acid losses on CAPD are slightly lower than those on
hemodialysis (20--25 g/day) [93-96] and intermittent peritoneal dialysis (11-53 g/
day) [97-99].

2.4.3. Plasma amino acid abnormalities in patients on CAPD

Figure 12 shows the plasma profile of the essential amino acids in six non-diabetic

Table 7. Changes in protein and caloric intake from diet and dialysate glucose absorption at 0,6 and 12
months in CAPD patients

Months

0 6 12

Protein intake (g/kg) 1.46 ± .09 1.15 ±.11 1.06 ± .09b

Kcals intake from diet + dialysate 2179 ± 113 1858± 88 1877 ± 78"
Kcals/kg body weight 36.5 ± 2.3 30± 2.3 28.9 ± l.3 b

" 0-12 months, p = 0.05.

b 0-12 months, p = 0.01.
492

200 • Normals IN- 6)

• Non-Diabetics IN - 6)

-•
c
u
150

100
•
~

Cl.

50

c; •
! .5 I c •c .;sc .J:cCI c
• • •
..
...
E .;s
c; ·0::I ·0 ·0
>- CI 0-
~
z
0 0
...
II > ::I
II II
0
...J Ci i
.....J: ] ...J
>. >- X
0-
c ...
4.
•
.J:
....

Figure 12. Profile of plasma essential amino acids in six non-diabetic CAPD patients compared with six
matched normal controls (percent of normal).

CAPD patients compared with sex-matched normal controls. The mean values of
almost all the essential amino acids were significantly lower than that of controls.
Figure 13 shows the profile of the non-essential amino acids in this same group.
The mean levels of some amino acids were significantly lower than that of the
controls, whereas citrulin and half-cystine were higher than in the normal con-
trols. It has been reported that 3-methylhistidine, which was not measured in our
patients, is elevated in the plasma of CAPD patients [90].

2.4.4. Amino acid absorption via the peritoneal route

We infused six non-diabetic patients over six hours with two liters of a dialysate
solution (Amino Dianeal), which contains 2% amino acids as a mixture of
essential and non-essential amino acids [100]. We then compared the plasma
amino acid changes produced by this solution with those produced by the infusion
of 4.25% Dianeal solution with dextrose. These two solutions have a similar
osmolality.
The Amino Dianeal appears to be an effective osmotic agent and at the same
time suitable for nutritional supplementation. It has an osmotic effect double that
of glucose at the same concentration; this is not surprising because the molecular
weight of the average amino acid is approximately one half that of glucose.
Amino Dianeal has the advantage that it can provide nutrition and, at the same
time, remove excess fluid, unlike total parenteral nutrition, which has to be
infused in large volumes, and causes fluid overload.
In a study [101] of Amino Dianeal given over four weeks, we showed an
improvement of nutritional status of CAPD patients as judged by a significant
increase in the total body nitrogen, which usually decreases in CAPD patients.
 493

300 _ Normals
11\1 Non-Diabetics
250

200
-
c
CD
u
~
150
~
100 -------------- ------------

50

• • • • • ., • • .5• • :;•c
c; c c c c

-.. -.. e - • ! •c
- -
c c
...0 .;0 .;: 'u~ :is .5-; ·S "i::::t
E 0
...
0 '2 E
z ...
>-
•
(I) 's.
... ... c...
·S
~
0
o..
E
0 >-
u C
0
0

•
(!) G-
~ 0 u C
:::t
(!) ....c;
:::t
c;
:z:
Figure 13. Profile of plasma non-essential amino acids in six non-diabetic CAPD patients compared
with six matched controls (percent of normal).

The serum transferrin also rose but not the serum albumin. Serum transferrin has
been used as an index of acute changes in visceral protein synthesis; due to its
shorter half-life, it is thought to provide a better measure of nutritional changes
than serum albumin in short-term studies. The use of amino acid exchanges
decreased the glucose load and tended to reduce the serum triglyceride level.
Initially, there was a significant but transient increase in HDL-cholesterol and at
the end of the four weeks HDL-cholesterol levels were at the baseline levels.
Amino Dianeal did not affect the patient's appetite. Long-term use did not lead to
an accumulation of amino acids in plasma and the patients tolerated it well. No
episodes of peritonitis were seen during this short trial.
In conclusion, amino acid-containing solutions can improve the nutrition of
malnourished CAPD patients who cannot increase their protein intake by other
means.

2.4.5. Glucose absorption and lipid abnormalities

During CAPD, patients absorb 160 to 200 g of glucose from the dialysate each
day. This glucose load may have a significant influence on the lipid metabolism,
because carbohydrate (sugar more than starch) has been shown to induce
hypertriglyceridemia.
The leading cause of death in patients receiving long-term dialysis is car-
diovascular disease [102-103]. Lipid abnormalities, especially those related to
494

triglycerides, are an important consequence of renal failure [104, 105] and persist
in many patients on dialysis [104, 106, 107] and after renal transplantation [108-
110]. Since Lindner's report of accelerated atherosclerosis in dialysis patients
[111], lipid metabolism has been studied extensively in patients with renal failure.
Plasma triglycerides are elevated in 20 to 70% of patients on dialysis, regardless of
the type of dialysis [104-110]. The predominant lipoprotein pattern is the type IV
phenotype - Frederickson's classification [112]. These patients have shown an
increase in the triglyceride contents of very-low-density lipoprotein (VLDL) and
low-density lipoprotein (LDL) and decrease in the cholesterol as high-density
lipoprotein (HDL). It is becoming obvious that not only do uremic patients have
high plasma triglycerides, but they also have reduced HDL cholesterol levels
[113]. This combination of lipoprotein abnormalities probably contributes to an
increased risk of cardiovascular disease.
Cattran etal. [114] studied 14 patients (mean age 55 years) on chronic peritoneal
dialysis; they had significantly higher and more sustained mean (± 50) plasma
triglyceride levels (313 ± 57 mgldl) than patients on hemodialsyis (195 ± 19 mg/
dl). Ten of these 14 had type IV lipoprotein phenotype. However, 21 patients
(mean age 50.6 years) on chronic peritoneal dialysis at the Toronto Western
Hospital [115], had a similar high serum triglyceride level (260.4 ± 12.4 mgldl) ,
but did not differ from 15 patients (mean age 36.3 years) on hemodialysis
(263.5 ± 9.1mg/dl).
Soon after the introduction of contino us ambulatory peritoneal dialysis, it
became apparent that many patients develop hypertriglyceridemia within two to
three months on CAPD [116]. Figure 14 shows our experience with lipid changes
among 16 patients on CAPD for over 30 months. The mean serum triglyceride at
the initiation of CAPD was 260.4 ± 97.6. After six months it increased to
362 ± 289.3 and thereafter, remained at this high level throughout the study.
During the same period, from an initial value of 251.1 ± 64.4 mgldl, the mean
serum cholesterol increased to 283.2 ± 64.9 mg/dl at six months but returned to
the pretreatment level by the second year and remained at this low level there-
after.
Despite the increase in the mean serum triglyceride level, not all patients
develop hypertriglyceridemia during CAPD. At the beginning of CAPD, 51 % of
37 patients, who completed one year of treatment, had serum triglyceride levels
below 230mg/dl; after one year of CAPD, 41% still had serum triglyceride levels
below this level. This clearly indicates that factors other than the glucose ab-
sorbed from the dialysate must be responsible for these abnormalities because
they are not present in all patients. These other factors may be the drugs used by
these patients - androgens, estrogens, beta-adrenergic blocking agents, or hor-
monal factors and genetic susceptibility to hyperlipidemia. In addition, over time
certain patients show a large variation in serum triglycerides with no change in
therapy; Figure 15 shows the course of serum triglycerides in such a patient. This
50-year old woman had been on CAPD for nearly two and one-half years. She
 495

SERUM CHOLESTEROL DURING CAPO

(N - 16)
300
~
! 280

e
~
260

••
& 240
en
I I I I I
o 6 12 18 24 30
MOl on CAPO Study

SERUM TRIGLYCERIDE DURING CAPO

(N-16)

'a 4!S0
.....

-r 400

I I I I I
o 6 12 18 24 30
MOl on CAPO Study

Figure 14. Serial mean serum cholesterol (above) and triglyceride (below) initiation and at every six
months during 30 months follow-up on CAPD.

1750
Mrs. V.T. Clofibrate - 0.5 Q q2 Days
on CAPO
'a
.....
go
E

••
'a

•u>-
.~

a.
.~

l-
E
...
~
250
~
0
9 II I 3 5 7 9 II I 3 5 7 9 II I 3 5
178 179 lao 181
Figure 15. Serial serum triglyceride level in a 50-year old female uremic on CAPD over a period of 32
months.
496

was using, on average, six liters of 1.5% dextrose solution and two liters of
4.25 g% dextrose solution every day, and was consuming a diet containing 60 g
protein and 2000 calories. Her serum triglyceride level fluctuated widely with no
identifiable cause.
To characterize the lipid and lipoprotein abnormalities in <;::APD patients, we
undertook a prospective study of 28 CAPD patients. Fifteen were males and 13
were females (mean age 53.6 years). They received CAPD for a period of four to
six months. None of the patients was receiving drugs, which could affect lipid
metabolism. All consumed a diet which consisted of 1.5 g/kg of body weight of
protein with a total of 30 to 35 Kcl/kg body weight per day. Fluid was not
restricted. Most patients were asked to observe moderate phosphate restriction.
These 28 patients were divided into three groups: Group I included non-diabetics
with initial serum triglyceride levels less than 230 md/dl; Group II included non-
diabetics with initial serum triglycerides greater than 230 mg/dl; and Group III
included all diabetics. Two patients in Group II and two in Group III received
small amounts of heparin (250-500 VII) in the dialysis solution to prevent clot
formation. Heparin in these concentrations is not absorbed in significant amounts
through the peritoneum. Plasma lipids and lipoproteins were estimated at the
beginning of CAPD and after three to six months of therapy. As shown in Table
8, when compared to normal values in the Toronto area, all groups had low mean
HDL cholesterol (HDL-C) concentrations (33 to 37 mg/dl) when they started on
CAPD. The LDL level remained within normal limits even in the presence of
elevated VLDL triglycerides characteristic of Type IV hyperlipoproteinemia.
After three to six months, there was a significant (p<0.01) increase in HDL
cholesterol (32.2 to 42.8 mg/dl) in those with normal, initial, plasma triglyceride
levels. In contrast, HDL-C concentration in hyperlipidemic or diabetic subjects
showed no significant changes after treatment. In all groups plasma cholesterol
and LDL-C levels remained relatively unchanged after dialysis while the plasma
triglycerides and VLDL-C were most variable in the hyperlipidemic patients.
While the diabetics showed no consistent trend, there was a suggestion that
HDL-C concentrations declined with time on dialysis in the hyperlipidemic

Table 8. Changes in lipoprotein cholesterol during four to six months of CAPD

Group I Group II Group III

HDL (mg/dl) Initial 32.2 ± 8.4 34.0 ± 5.5 37.0 ± 10.2

Post 42.8 ± 10.6" 30.9 ± 5.8 39.3 ± 10.3
LDL (mg/dl) Initial 105.6 ± 37.4 136.3 ± 28.7 131.7 ± 29.3
Post 117.8 ± 43.5 162.9 ± 48.4 129.0 ± 15.3
VLDL (mg/dl) Initial 21.5 ± 7.1 65.4 ± 30.5 32.9 ± 13.3
Post 26.2 ± 11.3 102.1 ± 63.1 35.0 ± 15.9

a p<[Link].
 497

group. This may reflect the tendency for triglyceride concentrations to increase in
these subjects during therapy.
HDL-C concentrations increased only in patients with initial normal trig-
lyceride levels and did not change in those with initially elevated plasma trig-
lycerides nor in the diabetics. The reason for this increase is not clear. Although,
the increase in HD L-C after dialysis may be a function of improved catabolism of
VLDL [117], this study found no correlation between HDL increment and a
decrease in the static concentrations of plasma VLDL.
Subjects who were hypertriglyceridemic before CAPD remained hypertrig-
lyceridemic and, in some cases, the plasma triglycerides increased further during
CAPD. Although the reason for this effect is unclear, it is possible that some of
these patients have a genetic susceptibility to hyperlipidemia and suffer a second-
ary aggravation of this metabolic disorder as a result of renal failure and caloric
overloading from glucose absorbed from the dialysis solution [118, 119].
A cross-sectional survey of 41 CAPD patients at the University of Missouri
[120] showed that 76% of the patients had above-normal triglyceride levels and
24% had levels within normal range. Moreover, in 15%, triglyceride levels were
more than five times the upper limits of normal. Table 9 shows the mean serum
concentration of albumin, triglyceride, cholesterol and HDL cholesterol. These
patients, whose mean age was 52.6 ± 14 years, had been on CAPD for a mean
duration of 21. 7 ± 13.4 months. Only 37% of the patients had below-normal HDL
cholesterol; and 85% of these patients had a serum albumin concentration within
the normal range. Cholesterol levels showed a significant inverse correlation with
height, and HDL cholesterol with weight. Triglyceride concentration correlated
highly with cholesterol. Interestingly, triglyceride concentration did not correlate
with time on CAPD.
Other centers have reported that hyperlipidemia is common in CAPD patients
[121-126]. Lindholm et al. [121] suggested that the increases in serum triglyceride
and cholesterol are related to an increase in very low-density lipoprotein (VLDL)
triglyceride and VLDL cholesterol. Gokal et al. [123] noted a high prevalence of

Table 9. Mean serum concentration of albumin, triglyceride, cholesterol and HDL cholesterol in 41
CAPD patients at the University of Missouri Health Science center (from Nolph et al. [120];
reproduced with permission)

Albumin Triglyceride Cholesterol HDL cholesterol

(GM/DL) (MGIDL) (MGIDL) (MG/DL)

Mean 4.2 500 245 27

Standard deviation 0.49 591 113 14
Range 3.1-5.5 101-2578 98-710 6-70
Median 4.2 283 218 25

Normal 3.5-4.7 30-175 150-300 23-73

498

hyperlipidemia in CAPD patients, and found hypertriglyceridemia and hyper-

cholesterolemia with equal frequency. However, these abnormalities were per-
sistent chiefly in those who had abnormal levels at the start of CAPD.

2.4.6. Changes in life style

Weight control, increased physical activity, alcohol restriction, and fat restriction
are primary therapy in the presence of elevated triglyceride levels. These mea-
sures should be used to the fullest extent possible before considering drug
therapy. The three drugs - clofibrate, gemfibrozil and nicotinic acid - are be-
lieved to be effective in lowering triglyceride in non-uremics. Several workers
have studied the effect of clofibrate in lowering triglyceride in uremic patients
[127-129]. The major drawback is its toxicity; in CAPD patients, it has been
associated with cholelithiasis, myalgias, CPK enzyme elevations and life-threat-
ening hyperkalemias [130-133]. In our experience clofibrate, even in a dose of
0.5 gm per week, may produce severe muscle toxicity. Therefore, patients receiv-
ing this drug should be monitored for alterations in the serum levels of creatinine
phosphokinase, lactic dehydrogenase, potassium, and phosphorus. Lameire et al.
[134] and Moncrief et al. [135] have suggested that insulin given intraperitoneally
may lower blood lipid levels in non-diabetic CAPD patients. Beardsworth et al.
[136] tested this hypothesis in 10 of their most hyperlipidemic non-diabetic CAPD
patients, and despite inducing severe hypoglycemia, found no significant change
in any of the lipids. Therefore, one cannot recommend intraperitoneal insulin for
its lipid-lowering effect at this time.
In summary, hypertriglyceridemia is frequent among patients on CAPD,
especially in those with abnormal triglycerides at the beginning of dialysis.
Usually HDL cholesterol is low at the initiation of CAPD and does not decrease
further with time on CAPD. In those who have initial normal triglyceride levels,
HDL cholesterol seems to increase with time on CAPD. Diet and weight control
are the mainstays of treatment. If used, the dose of clofibrate should be severely
reduced to avoid muscle toxicity.

2.4.7. Vitamin losses during CAPD

Deficiencies of several vitamins have been demonstrated in chronically hemo-
dialyzed patients, but few studies have been done to chart the changes during
peritoneal dialysis. Blumberg et al. [137] measured blood levels of the vitamins A,
E, B-complex, and C in 10 patients who had been on CAPD for a mean of 8.75
months. They found elevated vitamin A levels in all patients; retinol binding
protein (RBP) was elevated even more, causing the ratio of retinol RBP to drop.
Vitamin E levels were also high. Vitamin B1 was low or borderline in five
patients, Vitamin B6 was decreased in three and erythrocyte pyridoxal phosphate
in eight patients. Folic acid was low or borderline low in six patients, whereas
vitamin B2 and BJ2 were normal in all. Vitamin C was diminished in four patients.
Concerning the vitamin concentration in dialysate, they found that Vit. C was
 499

62%, Vit. A 1.4-1.6%, and Vit. E 0.7% of respective plasma concentrations.

Dietary evaluation showed that intakes of vitamin B1, B6 and Bl2 were below the
recommended range. After supplementation of water-soluble vitamins for seven
weeks, the vitamin A and E remained elevated, Bj remained low, B6 and Chad
become normal in all, and folic acid was markedly elevated. Blumberg et al. [137]
concluded that, during CAPD, a combination of inadequate dietary intake and
losses in the dialysate leads to a decrease the concentrations of water-soluble
vitamins. They recommend the following tentative guidelines for CAPD pa-
tients: the vitamin A and E should not be included in daily vitamin supplements,
but they suggest a daily supplement of 30--40 mg of vitamin B1, 10-15 mg of
vitamin B6 , 0.5-1 mg of folic acid and 100-300 mg of vitamin C.

2.5. Persistence and/or progression of complications associated with renal failure

Dialysis replaces the excretory kidney functions only partially. No form of dialysis
replaces renal metabolic and endocrine functions. Therefore, ESRD patients on
dialysis frequently show either progression or the development of new complica-
tions arising from the incomplete replacement of kidney functions. Complica-
tions in this category include: (a) hematological disorders, (b) renal osteodystro-
phy, (c) hormonal imbalance, (d) immunodeficiency, (e) serositis, (f) neuro-
logical disturbances, (g) electrolyte and fluid problems, and (h) decreased drug
disposition. Problems arising from immunodeficiency and drug disposition are
covered elsewhere in this book.

2.5.1. Hematologic disorders

As renal excretory function decreases, there is a similar reduction in erythropoie-
tic function, which leads to normochronic, normocytic anemia. The mechanisms
postulated in the anemia of chronic renal failure include [138-140]: (a) lack of or
decrease in erythropoietin production by the kidneys, (b) hemolysis due to
prevailing extracorpuscular factors, and (c) inhibition of bone-marrow, red cell
production by circulating uremic toxins.
Several workers have reported a noticeable improvement in hematocrit and
hemoglobin concentration in patients on CAPD [141-143] while others were
unable to confirm this observation [144]. DePaepe et al. attributed the increase in
hematocrit to a decrease in plasma volume secondary to increased ultrafiltration
during CAPD, rather than to an increase in red cell mass [141]. Spinowitz et al.
[144] noted no improvement in hematocrit in their patients maintained on CAPD;
however, in comparison with a group of hemodialysis patients, their CAPD
patients less frequently required transfusions, anabolic steroids and iron supple-
ments. They noted a small but statistically significant increase in hematocrit
among their CAPD patients transferred from hemodialysis. Spinovich et at. also
observed that those on CAPD did not show an elevation of oxygen affinity, as
500

measured by P50, previously seen in hemodialysis patients [145]. They speculated

that the improved acid-base balance and a more stable hemodynamic status of
CAPD patients may make unnecessary this adaptive mechanism directed to
increased tissue oxygen delivery. In 22 patients on CAPD up to 21 months,
Lamperi et al. [143] observed a rise in hematocrit, hemoglobin and reticulocyte
values, which showed a significant correlation with a recovery of erythroid-cell
proliferative activity. They found no change in the serum erythropoietin level.
The improvement of bone-marrow function, they speculated, is due to a better
clearance by CAPD of substances, which inhibit the marrow's response to
erythropoietin. They also observed a progressive increase in reticulocyte re-
sponse to oral iron, indicating that oral iron administration is useful in these
patients, especially those with serum ferritin levels lower than 50 ng/ml.
The red cells of patients on both hemodialysis and peritoneal dialysis show
increased glucose-6-phosphate dehydrogenase (G6PD) activity (146), which is
attributed to the younger mean red cell age resulting from shorter red cell half-life
in patients with uremia, compared to the red cell age in healthy individuals.
CAPD may result in increased iron removal following the intravenous admin-
istration of desferoxamine [147-148]. The red cell survival in CAPD patients is
shortened and correlates with red cell mass [149]. CAPD does not eliminate the
hemolytic component of anemia.

2.5.2. Platelet function

One of the complications of uremia is impaired platelet function causing an
enhanced bleeding tendency [150]. Both intermittent hemodialysis and peritoneal
dialysis are reported to improve bleeding time and platelet function [151]. The
mechanism(s) behind these dysfunctions as yet are unknown but it is believed to
be related to unidentified toxins because these disturbances disappear on their
removal following dialysis [152]. Intermittent peritoneal dialysis appears to cor-
rect these aberrations, especially defects in platelet aggregation, better than
hemodialysis [153-154]. Hemodialysis does not always correct the abnormal
bleeding time [155]. Recently, Arends et al. [156] investigated bleeding time,
platelet aggregation and the platelet count in patients on CAPD. The mean
bleeding time in 11 uremic patients improved from 11.2 (range 3.5-20) before, to
5.8 (range 2.5-20) minutes during CAPD treatment (p<0.025). Bleeding times
became normal in six out of seven patients with prolonged bleeding times before
CAPD. There was a concomitant improvement in platelet aggregation. Mean
platelet count rose from 195 (range 117-414) to 311 (range 148-522) x 101O/L
(p<0.01). They found a significant correlation between the change in bleeding
time and in platelet aggregation induced by ristocetin, but not between the
change in bleeding time and the platelet count. Although this study, was done on
a small number of patients it suggests that CAPD has a beneficial effect on
bleeding time and platelet function.
It is postulated that its beneficial effect on platelet function is due to removal of
 501

middle molecule toxins, although no such toxin has yet been identified. Differen-
ces in platelet function between patients on hemodialysis and peritoneal dialysis
may indicate toxic effects of substances used during hemodialysis such as heparin
[157].

2.5.3. Renal osteodystrophy

Renal osteodystrophy becomes established early in the course of renal failure and
its incidence and severity progress with its advancement [158]. Prolongation of
life in patients with end-stage renal disease (ESRD) with chronic dialysis allows
renal osteodystrophy more time for evolution. In addition the disease process
may change in response to dialysis-related factors, such as the dialysate concen-
tration of calcium or its aluminum content.
While we know much about the evolution of renal osteodystrophy in patients
undergoing chronic hemodialysis [159], studies of this disease in patients undergo-
ing continuous ambulatory peritoneal dialysis (CAPD) have been of short dura-
tion and have provided conflicting results [160-162].
In 1980 we described [161] the evolution of renal osteodystrophy in our patients
on short-term CAPD; at that time we had studied 28 patients on CAPD for
periods of 6-23 months. The radiologically diagnosed osteitis fibrosa element of
renal osteodystrophy progresses in CAPD patients whereas osteomalacia seems
to improve. At that time we did no studies of bone histology. The mean plasma
iPTH level, which remained high, was accompanied by high levels of serum
alkaline phosphatase. Arterial calcifications did not progress. The next year,
Tielman et al. [163], who studied 15 patients on CAPD for seven to 28 months,
also found that osteitis fibrosa progressed. They speculated that this might be
related to a simultaneous progressive decrease in serum Ca and serum 25 (OH)
vitamin D3 levels. In contrast to our findings and those of Tielman, Gokal et al.
[160] studied 40 patients on CAPD during their first year of treatment and
observed an improvement of osteitis fibrosa and osteomalacia. Also they con-
firmed the decline in serum 25 (OH) Vitamin D3 with time on CAPD. They
ascribed the difference between their results and ours to the difference in dialy-
sate calcium between the two groups (7 mg/dl- Gokal et al. vs 6 mg/dl- Calderaro
et al.). Finally, Teitelbaum et al. [164] studied bone histomorphology by repeated
bone biopsies in six patients on CAPD, reported an improvement in os-
teomalacia, while the evolution of osteitis fibrosa depended on the levels of
plasma iPTH.
Recently, Nolph et al. [120] reported a cross-sectional survey of serum
25(OH)D3 and 1,25(OH)2D3 in 41 CAPD patients. Their mean age was 52.6 ± 14
years and they had been on CAPD for a mean duration of 21.7 ± 13.4 months.
The levels of 25(OH)D3 and 1,25(OH)2D3 were correlated with the serum C and
N terminal parathyroid hormone, calcium, phosphorus, alkaline phosphatase,
magnesium and residual renal kidney function (Tables 10, 11). Each day the
patients took a tablet containing 400 I. U. of ergocalciferol and 200mg of calcium.
502

Table 10. Mean serum concentrations of vitamin D, calcium, phosphorus, PTH, and alkaline phos-
phatase in 41 CAPD patients (from Nolph et al. ]120]; reproduced with permission)

25-0H 1.25-0H Calcium Inorganic C-terminal N-terminal Alkaline

Vit. D Vit. D (mg/dl) phosphorus PTH (% of PTH (% of phosphatase
(ng/ml) (pg/ml) (mg/dl) norm. norm. (U/L)
maximum) maximum)

Mean 19.5 32.1 9.8 3.9 653 307 92

Standard 10.8 18.4 0.9 1.4 746 321 55
deviation
Range 5-53 6-81 8--11 0.8--8.3 100--2500 78--1250 34--261
Median 19.0 26.5 9.8 4.0 381 160 77

Normal 15-60 37-69 8.5-10.1 2.3-4.7 100 100 36-92

Table 11. Correlation among measurements in 41 CAPD patients (from Nolph et al. [120]; reproduced
with permission)

Months 25-0H 1.25- Calcium Inorganic C-term. N-term. Alkaline

on Vit. D (OH)2 phos- PTH PTH phospha-
CAPD Vit. D phorus tase

Months on 1.0 -0.09 -0.09 -0.16 0.10 0.41 b 0.32b 0.36'

CAPD

25-0H 1.0 -0.10 0.10 0.15 0.001 -0.10 -0.06

Vit. D

1.25-(OH)2 1.0 0.14 -0.03 0.15 0.18 -0.15

Vit. D

Calcium 1.0 0.13 -0.10 -0.05 0.04

Inorganic 1.0 0.20 0.31' 0.26

phosphorus

C-terminal 1.0 0.79 c 0.59'

PTH

N-Terminal 1.0 0.50c

PTH

Alkaline 1.0
phosphatase

a p<0.05.
b p<[Link].
c p<[Link].
 503

Phosphate binders were used as necessary to maintain serum phosphorus be-

tween 2.3-4.7 mg%. Significant correlations (p<0.05) were found between PTH
and alkaline phosphatase and time on CAPD, phosphorus and alkaline phos-
phatase and C and N-PTH (Table 11). This survey showed that most CAPD
patients have low 1,25(OH)P3' and elevated C and N terminal PTH levels.
A more recent study [165] updated the status of renal osteodystrophy in 27
patients maintained on CAPD for over three years at the Toronto Western
Hospital. Ten men and 17 women with a mean age of 49.5 years had been on
CAPD for a mean of 45 months. Dialysate calcium was 6 mg%; in the USA
solutions usually contain 7 mg%. For all the patients oral dietary calcium intake
was low (500mg/day). Al(OH)3' vitamin D and/or 1,25(OH)P3 were prescribed
to maintain serum calcium and phosphorus at normal levels. Patients were
divided into two groups according to the radiological findings at the beginning of
treatment. The results are summarized in Table 12. Group A (10 patients)
included those who had no subperiosteal resorption, and Group B (17 patients)
those with subperiosteal resorption at the initiation of CAPD. In Group A,
subperiosteal resorption grade remained normal in eight patients and progressed
in two. In Group B, subperiosteal resorption remained unchanged or increased in
14, while it improved in the remaining three.
Two patients with no subperiosteal resorption (Group A) showed severe
osteomalacia with positive aluminum staining on bone biopsy. In two Group B
patients, bone biopsy showed severe osteitis fibrosa with negative aluminum
staining. Of the 27 patients, five developed spontaneous fractures; these, how-
ever, healed with callus formation after adjustments in treatment. This study
again confirmed the earlier observation that hyperparathyroidism persists in
patients on long-term CAPD; this may be attributed to a low dialysate calcium
and low oral calcium intake.
This is the first long-term study of renal osteodystrophy in CAPD patients.
Most of our patients (16/27) had evidence of subperiosteal resorption at the
beginning of CAPD and a similar number (15/27) had abnormal subperiosteal
resorption at the end of the study. There was a good correlation between the
mean level of iPTH and the severity of subperiosteal resorption - the iPTH being
nearly normal in those with normal grades of subperiosteal resorption (Group
AI) throughout the study. However, osteomalacia was the main finding in the
biopsies of two of these patients, both of whom had developed spontaneous
fractures before biopsy examination; at the same time, both showed decreased
bone-mineral mass and density in the radius. Generally the dietary calcium of our
patients was low mainly because we were aiming at a low phosphorus diet, which
by necessity is low in calcium as well. Dietary calcium was the lowest in Group
A2, who were receiving the highest mean amounts of vitamin D and 1,25(OH)2
vitamin D 3, and who also had high plasma levels of iPTH and markedly increased
superiosteal resorption. It seems that, in the presence of low dietary calcium
intake, high doses of 1,25(OH)2 vitamin D3 may increase serum calcium by
 Ul
0
.j>..

Table 12. Age, sex, duration of CAPD, calcium intake, spontaneous fracture, bone pain, average Vit. D & 1,25(OH)2D3 i~t~ke per month and bone histology in
the four categories of CAPD patients at the Toronto Western Hospital .

Subperiosteal No. Age Sex Months Ca in,ake Spont Bone Average Bone
resorption at the x±SD on CAPD Mg/day fract pain 1,25(OH)2 D3 intake histology
x±SD x±SD (pt) mcg/pt. Vit. D3
Beginning End
Month U/day DTH

Normal Normal 8 50± 12 M(4) 46±6 590 ± 170 1 0.7 8820 3 pts.
(AI) 52± 17 F(4) osteomalacia
(aluminum stain
+)
1 patient normal
findings
Normal Abnormal 2 32 M 44± 12 355 ± 135 2 4.8 28570
(A2) 58 F
Abnormal Abnormal 14 52± 26 M(5) 45 ±7 475 ± 182 3 most of 0.4 3690 0.12 mg 2 pts. severe
(Bl) 47± 13 F(9) the pts. (7pt) osteitis fibrosa
(aluminum stain
- ve)
Abnormal Normal 3 44±8 F(3) 44±8 423 ± 168 2 4.5 1667 1 pt. severe
(B2) osteomalacia
(aluminum stain
+ ve)
 505

increasing bone resorption. If this speculation is correct, we may have to give

supplemental calcium to these patients; however, the risk of severe constipation
is high when calcium supplements are administered with aluminum phosphate
binders. Another alternative may be to increase dialysate calcium to 6.5 or 7 mg/
dl, as suggested by Gokal et al. [160].
Although CAPD removes significant amounts of iPTH [166], the control of
serum iPTH seems to depend on changes in serum calcium. In our patients
(Group B1), hyperparathyroidism persisted despite an increase in serum calcium
(1O.2-1O.6mg/dl) which, however, did not reach levels that seem critical in
suppressing PTH secretion. If high PTH levels should persist despite hyper-
calcemia, this indicates the presence of autonomous or tertiary hyperparathyroid-
ism and the patient requires parathyroidectomy [160, 167].
Bone pain is frequent in patients with abnormal subperiosteal resorption and
sometimes is due to spontaneous fractures; these were present in five of our
patients. Such fractures may be the result of severe osteitis fibrosa or co-existent
osteomalacia, but only a bone biopsy can identify the latter abnormality. Alumi-
num staining (seen in three of our six biopsies) suggests that A1(OH)3 given by
mouth will lead to aluminum absorption by the gut and significant accumulation
in bone. The aluminum content of the dialysate is very low [168-170]. However, it
should be stressed that the kinetics of aluminum transport across the peritoneal
membrane have not yet been clarified.
In summary, radiologically diagnosed osteitis fibrosa - a frequent complication
in patients entering CAPD, remains unchanged in most of them after 35-56
months of treatment. This may be the result of an increase in serum calcium levels
that is inadequate to turn off parathormone secretion. The incomplete control of
serum calcium in these patients, may be due to inadequate oral calcium intake
and, in Canada, a relatively low dialysate calcium (6 mg/dl), despite large
amounts of 1,25(OH)2 vitamin D 3 .

2.5.4. Back pain

Back pain, one of the most frequent musculoskeletal afflictions of humans,
accounts for millions of dollars in lost workdays and much personal and family
suffering. The appearance of back pain in a patient on CAPD may result in less-
than-optimal dialysis because of the decrease in dialysate volume, and eventually
may force discontinuance of CAPD.
Mailis [171] surveyed the Toronto Western Hospital CAPD patient population
with regard to back pain. During a five-year period, eight patients developed
back pain while on CAPD. Table 13 shows the history, physical examination and
postural findings in five of these patients - four women and one man. The mean
age was 63 years with a range of 52 to 74 years. Four of these five had a history of
back disorders of three to 20 years' duration. All had degenerative spinal disease
or metabolic bone disease and/or extraspinal disease affecting the normal spinal
mechanics, such as obesity, arthritis of the hip with flexion contractures, multiple
506

Table 13. Analysis of CAPD patients with back pain at the Toronto Western Hospital

Patient Previous Extra sp. Spinal Postural Effect

back HX disease disease changes of CAPD

QG.M. (60) 20 yrs R.A. osteodystrophy iT. kyphosis t B. pain, CAPD

psoriasis i L. lordosis inadequate
QA.F. (52) 27 yrs multiple ? very obese t B. pain
abdominal abdomen - with sciatica
operations t L. lordosis
QL.H. (72) 10 yrs bilateral severe t T. kyphosis 30% tb. pain
total hips DDD LSS- TIL scoliosis
osteoporosis bi!. hip flex
contractures
d'S.P. (74) -ve proximal ? t L. lordosis development of
leg muscles DDD LSS -protuberant mod. b. pain
weakness abdomen
QE.C. (57) 3 yrs neuropathy DDD L4-5 ~ L. lordosis- i b. pain with
t T. Kyphosis- sciatica
-shortet L.
leg

abdominal operations with weak abdominal muscles, neuropathy or myopathy

with weak proximal muscles. On physical examination, all had postural changes
and abnormal spinal biomechanics - kyphosis, scoliosis, leg-length discrepancy or
protruberant abdomen. Four had suffered exacerbation of back pain on CAPD
(two with sciatica) and one developed back pain de novo. In three, back pain was
the major reason for failure of CAPD while, in the other two, pain combined with
other factors necessitated decreased volume and thereby clearence on CAPD and
made alternative treatment necessary.
Patients on continuous peritoneal dialysis constitute a special population; most
are at middle age or older. By itself, age is associated with an increased frequency
of degenerative changes in the spine, i.e., degenerative disc disease, facet hyper-
trophy, spondylolysis or spondylolisthesis. In addition, the CAPD patients suffer
from a systemic disease, i.e., uremia and other concomitant illnesses - athe-
rosclerotic heart disease, diabetes; as a result, their levels of activity are low and
result in poor exercise tolerance, poor posture and poor muscle tone, including
the tone of abdominal muscles (Table 14). Also these patients often have underly-
ing neuromusculoskeletal disease. For example, osteoporosis with compression
fractures increases the thoracic kyphosis and, as a compensation, lumbar lord-
osis. Hip or knee involvement in different arthritides can create flexion con-
tractures. Neuropathies or myopathies may affect the strength of proximal or
distal musuculature. Multiple abdominal operations or obesity weaken the ab-
dominal muscles or shift the line of gravity further forward. All these factors alter
the normal spinal mechanics and produce abnormal posture. On such a back-
 507

Table 14. Etiology of low back pain in CAPD patients

poor muscle tone

L Deconditioning poor posture
poor exercise tolerance

spinal (degeneration or
2. Neuromusculoskeletal metabolic bone disease)
diseases hip-knee arthritis
neuropathies
myopathies
abd. hernias
3. Altered spinal mechanics by dialysis fluid

ground, the addition of two or three liters of dialysate in the abdomen pulls
forward the center of gravity and may aggravate back strain by increasing or
attempting to increase the lumbar lordosis (Fig. 16).
Low-back pain in patients on CAPD seems to be due to many factors. It is the
patient with a long-standing history of back ailments and/or evidence of disease

A B

Figure 16. Normal posture (A); addition of the dialysate increases the lumbar lordosis (B).
508

who shows aggravation of back pain while on CAPD. Prevention starts with the
recognition of the high risk patient, following appropriate history, physical
examination and spinal X-rays. Primary prevention teaches the patient how to
protect his back during the strains of daily activity; back education teaches him
how to strengthen his abdominal muscles and improve his posture. Primary
prevention requires an exercise program 'custom-made' for the individual back; it
will take into account the patient's physical condition, motivation and exercise
tolerance. The simplest exercise - pelvic tilt can be performed by any person well
enough to be on CAPD. Special abdominal supports (binders) may assist the
patient with abdominal hernias or multiple abdominal scars. The most appropri-
ate time for back education is during the training period, which the patient spends
in the home dialysis unit while he becomes familiar with the CAPD procedure.
The Physical Medicine Department can be of special help in designing an appro-
priate program for each patient.

2.5.5. Endocrine disturbances

Uremic patients have a variety of endocrine disturbances. The spectrum of
disease may range from clearly recognizable endocrinopathies to laboratory
changes only. Radioimmunoassay techniques permit us to measure very low
concentrations of certain hormones in body fluids. While such sensitive indices
help to understand the pathophysiology, one must interpret such data carefully in
clinical and research settings. Secondary hyperparathyroidism, osteomalacia
related to abnormal vitamin-D metabolism and decreased erythropoietin produc-
tion and anemia were discussed earlier in this chapter. We will now discuss
thyroid function, growth hormone, reproductive function and the complex sys-
tem of renin and aldosterone in CAPD patients.

2.5.6. Thyroid function

Not uncommonly, patients with advanced renal failure will develop symptoms
compatible with hypothyroidism. However, the bulk of clinical evidence suggests
that the incidence of hypothyroidism is not increased in patients with uremia
[172-184]. On the other hand, an abnormally high incidence of thyromegaly has
been described in certain hemodialysis patients and frequently, thyroid function
studies suggest hypothyroidism [185-187]' A low serum total triiodothyronine
(T3) concentration - the second most common thyroid abnormality in uremia
[172, 174, 17~179, 181-183] occurs in about 30% of patients. Furthermore, free
serum thyroxine levels are usually normal or only slightly altered [173, 174, 183-
184], although a few workers have reported a significant reduction. The total
serum concentration of reverse triiodothyronine (rT3) appears to be normal [173,
188--191], but, on the other hand, free rT3 may be increased. Most investigators
report normal or below-normal values of thyroxine-binding globulin (TBG) in
hemodialysis and peritoneal dialysis patients [176, 184, 185, 187, 190-192]. The
free T4 index (FT4I) provides an indirect estimate of free T4 concentration in
 509

circulating plasma. In most series, uremic patients show low Ff4I because the
total T4 is decreased out of proportion to RT3U [172, 176, 178, 179, 181, 182, 184].
The great majority of investigators report normal mean basal concentration of
TSH despite decreased mean levels of total T3, total T4, and FT41. Also, most
investigators have not been able to demonstrate the expected negative correla-
tions between circulating levels ofthyroid hormones and TSH [173, 183, 185, 193].
Frequently, TSH responses to bolus injections of exogenous thyrotropin-releas-
ing hormone (TRH) are diminished [173, 175, 178, 185,194]. Production rates of
TSH in uremics are greatly reduced.
We know little about the effect on thyroid function of CAPD, with its trans-
peritoneal protein losses, steady state chemistries and better removal of middle
molecular. Most studies tend to show a pattern similar to that seen in uremics
[198-210]. Charytan et al. [198] report that patients on CAPD have markedly
reduced total T3, but otherwise did not differ significantly from hemodialysis
patients. Similarly, Selgas et al. [210] noted a significant decrease in total T3 and T4
in plasma, but no change in free T4 or TSH. They did not find appreciable
amounts of the hormone in the dialysate. A retrospective analysis of 104 CAPD
patients at the Toronto Western Hospital showed [208] that before starting
CAPD 14 of them had been treated with L-thyroxine and 13 more had had
elevated TSH but all other indices were normal. T3 resin uptake was increased in
11 patients and normal in the remainder. Total T4 and free T4 were normal.
During CAPD, mean TSH increased significantly (p<0.01) and an additional
seven patients, whose T4 and free T4 were also decreased, had to be treated for
hypothyroidism. In patients in whom T3RU was increased before CAPD, this
returned to normal during dialysis. Total T4 FfI remained normal. Total T3 was
low in 60% of the euthyroid patients in whom it was measured.
In summary, the spectrum of thyroid function abnormalities noted during
CAPD is similar to that observed in uremics. More study is needed to determine
the significance and long-term effects of these changes, and to follow their
response to prolonged CAPD therapy.

2.5.7. Growth hormone and prolactin

Patients on long-term hemodialysis, are reported to increase serum levels of
growth hormone (GH) and prolactin (PRI) [211-212]. Such increases are believed
to be due, in part, to the reduction in the renal handling ofthese hormones [213].
However, Sievertsen et al. [213] noted pitutary prolactin hypersecretion in
uremia. Furthermore, Tolis et al. [212], who showed abnormal growth hormone
and prolactin responses to stimulatory or suppressive agents, suggested that
uremics had an abnormal hypothalamic-pituitary axis. Recently Nicoletti et al.
[214] showed elevated serum levels of prolactin in patients on hemodialysis and
CAPD, whereas transplanted patients had near-normal serum levels. In control
and transplanted patients, TRH-LHRH administration was followed by a prompt
increase in prolactin levels in both hemodialyzed and CAPD patients, the re-
510

sponse was blunted - similar to that often seen in acromegaly, liver cirrhosis,
depressive illness, and anorexia nervosa - conditions presumed to be associated
with abnormal brain catecholamine function [215]. There was no significant
difference in the basal level of growth hormone in any of the groups, including
controls. However, after TRH-LHRH stimulation, they showed a paradoxical
increase in the GH level. Semple et al. [216] evaluated pitutary-testicular function
in 18 patients with chronic renal failure - nine treated by hemodialysis and nine by
CAPD, and compared them with non-uremic controls. Serum total testosterone
and the free testosterone index were significantly low in both dialysis groups.
Basal FSH and LH levels were elevated, both FSH and LH responses to the IV
administration of LRH were normal. There were no significant differences
between the CAPD and hemodialysis groups in the hormonal parameters stud-
ied.
Based on the available literature neither hemodialysis or CAPD seem to
correct growth hormone, prolactin levels and hypothalamopitutary axis aberra-
tion noted in uremics but transplantation seems to return these functions to
normal.

2.5.B. Reproductive functions

In the female, chronic renal failure frequently is associated with reproductive
dysfunction [217], with such manifestations as: menstrual irregularities, anovula-
tory cycles and ovarian cystic disease. The general experience is that these
reproductive irregularities are not easily reversible during chronic hemodialysis.
Whenever it returns, menstruation is irregular and complicated by menometror-
rhagia and anovulatory cycles [218, 219]. Often regular cycles are associated with
an abnormal luteal phase and inadequate progesterone increments [220]. Several
workers have described pregnancy in women on hemodialysis [220--226]. Con-
trary to the observations concerning hemodialysis, several women on CAPD
have had normal menstruation with ovulatory cycles [227]. Three authors have
reported pregnancy in women on CAPD [228-230]. Galler et al. [230a] compared
menstrual histories, vaginal hormonal smears and ovulatory curves in 12 women
under the age of 45 years - on hemodialysis vs seven on CAPD. Eighty-six per
cent of the women on CAPD and 25% of those on hemodialysis had regular
menses. Upon transfer to CAPD, two amenorrheic women resumed regular
menses within 21 months. Only two patients (one in each group) had ovulatory
cycles. All had a vaginal estrogen effect. Galler et al. [230a] speculated that the
return of regular menses in CAPD patients may be due to improved clearances of
middle molecules with secondary restoration of normal hormonal balances.
Studies in uremic females usually show hormonal disturbances at several sites
along the hypothalamic - pituitary - ovarian axis. Lim et at. [231] demonstrated
primary ovarian dysfunction in certain uremic women who could not respond
with adequate estrogen production after clomiphene stimulation, or after a
normal endogenous LH elevation or ovulatory surge. Several workers attributed
 511

these dysfunctions to hypothalamic anovulation [231-232]. In uremic women,

pituitary dysfunction is manifested by high prolactin levels, which correlate with
the ovarian dysfunction [217, 232, 233]. We need similar studies in women on
CAPO to elucidate the nature of their dysfunction.

2.5.9. Response of blood pressure and the renin-angiotensin aldosterone system to

CAPD
Continuous ambulatory peritoneal dialysis (CAPD) often gives good control of
hypertension, allowing a reduction or discontinuation of antihypertensive drugs.
Most of this hypotensive response occurs during the first weeks on CAPO, but a
further gradual fall occurs over the next few months (Fig. 17) [234]. The decline in
blood pressure may be marked and lead to symptomatic hypotension [234]. This
hypotensive effect may be related to excessive sodium and water removal via the
dialysate inducing to both volume and sodium depletion. Failure to adjust the
'ideal dry weight' to the rapid increase in body weight observed after initiation of
CAPO can result in excessive sodium and fluid removal and chronic hypo-
volemia. Because hypoproteinemia is frequent in these patients, the extravascu-
lar fluid pooling may occur in the upright posture leading to decreased venous
return and cardiac filling and thus orthostatic hypotension. This phenomenon
also may cause peripheral edema, which may be misinterpreted as intravascular
fluid, overload. Nevertheless, when faced with hypovolemia, compensatory
mechanisms should maintain blood pressure. However, uremic autonomic neu-
ropathy may interfere with the normal responses of both the sympathetic nervous
system and the renin-angiotensin system. Furthermore, on its own, sodium
depletion can interfere with vascular reactivity and sympathetic nerve activity.
Previous studies have shown that considerable amounts of sodium can be lost via

110
D
X
E 105
E
100
...

'., l' '

CD

t\,,
~
en
en 95
f

r "'.\
a.. 90

8
iii
85 I

c: '4. !
1
80 0--0 Supine
e
CD ..-.... Erect
~
75
I I I I I I I I I I I I I
0 2 4 6 8 10 12 14 16 18 2022 24
Mos on CAPO
Figure 17. Changes in mean blood pressure in 132 patients during 24 months of CAPD.
512

the dialysate [235], amounts which easily can exceed the dietary sodium intake,
especially in patients on a low sodium diet.
To explore the role of sodium per se, we treated five patients with symptomatic
orthostatic hypotension on CAPD by salt-loading (oral administration of NaCl
capsules); this increased body sodium stores without inducing a rise in total body
water [236]. In CAPD patients this dissociation between sodium stores and water
volume is possible by increasing sodium intake without allowing a concomitant
increase in body weight. In these five patients supine blood pressure increased
significantly after salt-loading (from 94/67mmHg to 121-78mmHg) and symp-
tomatic orthostatic hypotension disappeared. The mechanisms leading to the
improvement after salt-loading were studied in three of these patients. In these
three, plasma norepinephrine levels before salt-loading were within the high
normal range in the supine and standing position. Somewhat surprisingly, com-
pared to their baseline, supine and standing norepinephrine levels were increased
in the three patients studied after salt-loading. The dose response for intra-
venously administered norepinephrine shifted to the left in two of the three
indicating an increase in pressure reactivity. For example, the increase in systolic
blood pressure in response to infusion of 25 ng/kg/min of angiotensin changed
from 5 to 18 mmHg, and from 8 to 28 mmHg in two patients, and did not change in
the third. Figure 18 shows the effects of salt-loading on blood pressure and

z::: '1F
MAP (mmHQ) Heart Rate (Beats/Min)

:~
70
60
50 II I I 80" I I
01 5 15 0 I 5 15
Min Min

Plasma NoreRineRhrine (!)(I/mU Pressor-Reactivitl.

1600 ~y'stolic BP (mmHgl
~ __ At--------~ +25
800 +20
+15
400 +10
+5
200 nll--~I------~I I I I
01 5 15 125 25 50
Standing Norepinephrine (ng/kg/min)
e---e Salt-loading
0--0 [Link].

Figure 18. Effect of salt-loading on supine and standing blood pressure and on sympathetic tone in one
of the CAPD patients (from Leenen et al. [236]; reproduced with permission).
 513

sympathetic tone in one of these patients. These results suggest that, in some
CAPD patients, hypotension or orthostatic hypotension may be controlled by
oral salt-loading. FurthermOl'e, the study shows that salt-loading can be done
without a gain in body weight, i.e., without increasing total body water. More
detailed studies in three of the five suggest that the beneficial effect of salt-loading
depends upon at least two mechanisms: 1) an increase in ECV and probably BV,
2) an increase in sympathetic tone, as assessed by plasma norepinephrine levels,
and in pressor reactivity to norepinephrine.
In summary, although this study does not prove that the late development of
hypotension or orthostatic hypotension in CAPD patients is due to sodium
depletion, it does show that oral salt-loading may give effective control of this
complication.
Besides the influence of volume and sympathetic tone on blood pressure, the
renin-angiotensin-aldosterone system has a significant modulating effect on vas-
cular tone. Plasma renin activity is of prime importance in hypertension in dialysis
patients that is not due to hypervolemia. In chronically dialyzed patients, the
levels of renin-angiotensin-aldosterone have been inappropriately high for the
degree of sodium and volume excess. Most of these studies were done in hemo-
dialysis patients. In a longitudinal study of seven patients during CAPD, Glasson
et al. [237] investigated the renin-angiotensin-aldosterone system to see how it
was modified by treatment and to examine its relationship to blood pressure.
They found steady increases in plasma renin activity and aldosterone with good
correlation between the two. During the observation period, plasma electrolytes,
renin substrate and body weight did not change significantly. Angiotensin II
perfusion studies showed a relative vascular resistance to angiotensin II. These
changes could not be explained by a decrease in body weight or by changes in
serum electrolytes, because these did not change significantly during the study.
Despite the observation of Osmond et al. [238], that a large daily loss of renin
substrate occurs via the peritoneum, Glasson et al. [237] noted no change in the
plasma level of renin substrate. For the stimulation of renin-angiotensin system
and good control of blood pressure in CAPD patients, they postulated two
mechanisms: (a) a decreased vascoconstrictor effect of angiotensin II, as ex-
plained by altered vascular sensitivity, leads to increase in plasma renin activity
(b) CAPD may remove an unknown vasopressor substance responsible for the
inhibition of the renin-angiotensin system. This hypothesis, however, needs
further study.

2.5.10. Pericarditis
Until the advent of dialysis and transplantation, pericarditis usually was a termi-
nal event in chronic renal failure and invariably had a fatal outcome. The
frequency of pericarditis in patients with chronic renal failure undergoing any
form of dialysis has ranged from 15 to 20% [239-241]. OUf patients on mainte-
nance peritoneal dialysis had an incidence of 4.3, compared to 8.4 per 100 patient-
514

years in hemodialyzed patients [239]. The incidence of pericarditis in patients

treated with CAPO was less than 2% in the Toronto Western Hospital program.
Of our dialysis patients with pericarditis who developed tamponade, none were
receiving peritoneal dialysis; this supports Cohen's recommendation that patients
developing pericarditis during hemodialysis be transferred to peritoneal dialysis
to prevent further effusion or tamponade [242]. Clinically, pericarditis presents
with chest pain in 63%, with a pericardial friction rub in 91%, with low-grade
temperature in 79%, and mild leukocytosis in 35% [239]. We find needle pericar-
diocentesis effective in treating pericardial tamponade and reserve surgical per-
icardectomy for recurrent cardiac tamponade. Pericarditis without tamponade
responds to conservative management, namely increased dialysis, salt restriction
and strict control of fluid intake. Autopsy studies suggests that subclinical pericar-
ditis is common in uremic patients and that despite clinical resolution several
years before death, pathological changes in pericardium will persist.
The pathogenesis of pericarditis in renal failure is not clearly understood.
There is a recent finding [243] of significantly elevated levels of circulating
immune complexes in dialysis patients with pericarditis. This study also showed a
very low incidence of serositis in CAPO patients compared to those on hemo-
dialysis. The levels of circulating immune complexes were significantly lower in
patients maintained on CAPO. Moreover, these workers were able to detect
immune complex-like material in the peritoneal effluent of some patients. This
finding may suggest the better clearance by CAPO of these 'uremic toxins' - a fact
that may be responsible for relative infrequency of pericarditis in patients on
CAPO.

References

1. Oreopoulos DG, Khanna R: Complications of peritoneal dialysis other than peritonitis. In: KD
Nolph (ed) Peritoneal Dialysis, Martinus Nijhoff Publishers, The Hague, 1981, pp 309-343.
2. Moncrief JW, Rutherford CE, Sorrels PAJ, Bailey A, Popovich RP: Technical aspects of
continuous ambulatory peritoneal dialysis new connection devices developed by Baxter Trav-
enol Laboratories. M Legrain (ed), Excerpta Medica, Amsterdam, 1980, pp 79-81.
3. Nolph KD: Complications during continuous ambulatory peritoneal dialysis. Proceedings of the
first international symposium on CAPD. M Legrain (ed), Excerpta Medica, Amsterdam, 1980,
pp 258-262.
4. Mavichak V, Moriarty MV, Cameron EC et al. : Three and a half years experience with CAPD
using the Beta cap technique. Trans ASAIO 28: 253-258, 1982.
5. Di Paolo N, Buoncristiani V, Frigato C, De Mia M, Rubegni M: A new system for CAPD. In:
Proc. of Second International Symposium on Peritoneal Dialysis. GM Gahl, M Kessel, KD
Nolph (eds), Excerpta Medica, Amsterdam, 1981, pp 105-110.
6. Oreopoulos DG, Zellerman G, Izatt S: The Toronto Western Hospital permanent peritoneal
catheter and CAPD connector in Proc. 1st Int. Symposium CAPD. M Legrain (ed), Excerpta
Medica, Amsterdam, 1980, pp 73-78.
7. Palmer RA, Quinton WE, Gray EJ: Prolonged peritoneal dialysis for chronic renal failure.
Lancet 1: 700, 1966.
 515

8. Palmer RA et al.: Treatment of chronic renal failure by prolonged peritoneal dialysis. N Engl J
Med 274: 248, 1966.
9. Tenckhoff H, Schechter H: A bacteriologically safe peritoneal access device. Trans Am Soc Art
Intern Organs 14: 181,1968.
10. Slingeneyer A, Balmes M, Mion C: Surgical implantation of the Tenckhoff catheter in peritoneal
dialysis. G LaGreca, S Biasioli, C Roneco (eds), Wichtig Editore, Milano, 1983, pp 133-136.
11. Rubin J, Adair C: Peritoneal access using the Tenckhoff catheter. Perspective in Peritoneal
Dialysis 1: 2-3, 1983.
12. Rottembourg J, de Groc F: Peritoneal access using the curled Tenckhoff catheter. Perspectives
in Peritoneal Dialysis 1: 7-9, 1983.
13. Oreopoulos DG, Izatt S, Zellerman G et al.: A prospective study of the effectiveness of three
permanent peritoneal catheter. Proc Dial Transplant Forum p. 96, 1976.
14. Ponce SP et al.: Comparison of the survival and complications of three permanent peritoneal
dialysis catheters. Perit Dial Bull 2: 82, 1982.
14a. Nichols WK, Nolph KD: A technique for managing exit site and cuff infection in Tenckhoff
catheters. Perit Dial Bull 3: 24-5, 1983.
15. Grefberg N: Clinical aspects of CAPD. Scand J Urol and Nephrology S72: 7-38, 1983.
16. Hogg RJ, Coin D, Chang J et al.: The Toronto Western Hospital catheterin a pediatric dialysis
program. Am J of Kidney Disease 3: 219-223,1983.
17. Ash SR et al.: The column-disc peritoneal catheter: A peritoneal access device with improved
drainage. ASAIO J 3: 109, 1980.
18. Ash SR, Slingeneyer A, Scchardin KE: Peritoneal access using the column-disc catheter.
Perspective in Peritoneal Dialysis 1: 9-11, 1983.
19. Roxe DM, Santhanam S: Structural defects in chronic peritoneal catheters contributing to
peritonitis. Nephron 34: 267-268, 1983.
20. Steiner RW, Kipper S, Savoia MC et al.: Identification of peritoneal dialysis catheter tunnel
infection by scanning with Idium-III-labelled leukocytes. Ann Int Med 99: 44-45, 1983.
21. Erlich LF, Powell SL: Care of the patient with a Gortex peritoneal dialysis catheter. Dial &
Transplant 12: 572-577,1983.
22. Korten G, Arendt R, Brugmann E et al.: A new procedure for the revision and reposition of
functionless peritoneal dialysis catheters. Z Urol Nephrol75: 885-887,1982.
23. Davis R, Young J, Diamond D et al.: Management of chronic peritoneal catheter malfunction.
Am J Nephrol 2: 85-90, 1982.
24. MarichaJJF, Faller B, Wagner D et al.: Continuous ambulatory peritoneal dialysis: implantation
of the catheter. Nephron 30: 192, 1982.
25. Cerilli J, Walker J, Bay W: A new technique for placement of catheters for peritoneal dialysis.
Surg Gynecol Obstet 156: 663-664, 1983.
26. Singer H, Rapoport A, Crassweller PO: A case of terminal renal disease maintained for 134 days
by intermittent peritoneal dialysis. CMAJ 90: 1318-1321, 1964.
27. Oreopoulos DG, Gotloib L, Calderaro V, Khanna R: For how long can peritoneal dialysis be
continued? 124: 12-13, 1981.
28. Faller B, Marichal JF: Loss of ultrafiltration in continuous ambulatory peritoneal dialysis:
clinical data. In: GM Gahl, M Kessel, KD Nolph (eds), Advances in Peritoneal Dialysis.
Excerpta Medica, 1981, pp 227-232.
29. Slingeneyer A, Canaud B, Mourad G et al.: Sclerosing peritonitis: Late and severe complication
of long term home perioneal dialysis. Abstract Amer Soc Artif Intern Organs Vol 12, p. 64.
Toronto, April 28-30, 1983.
30. Gandhi VC, Humayun HM, Ing TS et al.: Sclerotic thickening of the peritoneal membrane in
maintenance peritoneal dialysis patients. Arch Intern Med 140: 1201-1203, 1980.
31. Backenroth-Maayan R, Longnecker R, Kamboroo D: Failure of the peritoneal membrane
during chronic intermittent peritoneal dialysis. In: GM Gahl, M Kessel, KD Nolph (eds),
516

Advances in Peritoneal Dialysis. Exerpta Medica, 1981, pp 208--213.

32. Slingeneyer A, Canaud B, Mion C: Permanent loss of ultrafiltration capacity of the peritoneum
in long term peritoneal dialysis: An epidemiologic study. Nephron (submitted).
33. Miller FN, Nolph KD, Harris PO, Rubin J et al.: Microvascular and clinical effects of altered
peritoneal dialysis solutions. Kidney Intern 15: 638--639, 1979.
33a. Manuel MA: Failure of ultrafiltration in patients on CAPO. Peritoneal Dialysis Bulletin 3: S
38--40, 1983.
34. Liang C, Lowenstein JM: Metabolic control of the circulation. J Clin Invest 62: 1029-1038,1978.
35. Slingeneyer A, Mion C, Mourad B et al.: Progressive sclerosing peritonitis: A late and severe
complication of maintenance peritoneal dialysis. Trans ASAIO 29: 633-640, 1983.
36. Twardowski ZJ, Tully RJ, Nichols WK, Sunderjan S: Computerized tomography in the diag-
nosis of subcutaneous leak sites during CAPO. Perit Dial Bull (accepted for publication), 1984.
37. Edwards SR, Unger AM: Acute hydrothorax - A new complication of peritoneal dialysis. JAm
Med Assoc 199: 853,1967.
38. Berlyne GM, Lee HA, Ralston AJ, Woolwock JA: Pulmonary complications of peritoneal
dialysis. Lancet 2: 75, 1966.
39. Maher JF, Schriener G E: Hazards and complications of dialysis. New Engl J Med 273: 370, 1965.
40. Aye AM, Kulatilake AK, Schackman R: Peritoneal dialysis in surgery. Proc Eur Dial and
Transplant Assoc II: 49, ??
41. Finn R, Jowett EW: Acute hydrothorax complicating peritoneal dialysis. Br Med J 2: 94, 1970.
42. Holm J, Lieden B, Lindqvist B: Unilateral pleural effusion - A rare complication of peritoneal
dialysis. Scan J Urol Nephrol 5: 84, 1971.
43. Haberli R, Stucki P: Akuter hydrothorax als komplikation bei peritoneal dialyse. Praxis 1: 13,
1971.
44. Alquier Ph, Achard J, Bonhomme R: Hydrothorax aigue au cours de dialyses peritoneales. A
propos de 5 cas. Nour Presse Med 4: 192, 1975.
45. Rudnick MR, Coyle FJ, Beck LH, McCurdy OK: Acute massive hydrothorax complicating
peritoneal dialysis, report of two cases and a review of the literature. Clin Nephrol12: 38, 1979.
46. Posen GA, Sachs HJ: Treatment of recurrent pleural effusions in dialysis patients by talc
insufflation (abstract). Am Soc Artif Intern Organs 8: 75, 1979.
47. Khanna R: Questions and answers. Perit Dial Buill: 17, 1980.
48. Von Thieler H, Schmidt U, Schafer W: Akuter hydrothorax unter peritoneal dialyse. Dt
Gesundh Wesen 35: 2043, 1980.
49. Milutinevich J, Shyong W, Lindhom DO et al.: Acute massive unilateral hydrothorax: A rare
complication of peritoneal dialysis. South Med J 73: 827, 1980.
50. Kuehnel E: Massive pleural effusion secondary to CAPO (,abstract). Kidney Int 152, 1981.
51. Lorentz WB: Acute hydrothorax during peritoneal dialysis. J Pediatr 94: 417, 1979.
52. Steiner RW: Adverse effects of intraperitoneal methylene blue. Perit Dial Bull 3: 43, 1983.
53. Spadaro JJ, Thakur V, Nolph KD: Technetium- 99 -m-labelled macroaggregrated albumin in
demonstration oftransdiaphragmatic leakage of dialysate in peritoneal dialysis. Am J Nephrol2:
36-38, 1983.
54. Scheldewaert R, Bogerts Y, Pauwels R et al.: Management of a massive hydrothorax in a CAPO
patient: A case report and a review of literature. Perit Dial Bull 2: 69--73, 1982.
55. Seebaran AR, Patel PL: Acute massive hydrothorax - A rare complication of peritoneal dialysis.
A case report. S Afr Med J 60: 827-828, 1981.
56. Townsend R, Fragola JA: Hydrothorax in a patient receiving CAPO a successful treatment with
intermittent peritoneal dialysis. Arch Intern Med 142: 1571-72, 1982.
57. Grefberg N, Danielson BG, Benson L, Pitkanen P: Right sided hydrothorax complicating
peritoneal dialysis. Nephron 34: 130--134, 1983.
58. Rudnick MR, Coyle JF, Beck LH et at.: Acute massive hydrothorax complicating peritoneal
dialysis. Clin Nephrol12: 38, 1979.
 517

59. Lieberman FL, Hidemura R, Peters RL et al.: Pathogenesis and treatment of hydrothorax
complicating cirrhosis with ascites. Ann Intern Med 64: 34,1929.
60. Jones GR: Treatment of recurrent malignant pleural effusion by iodized talc pleurodesis. Thorax
24: 69, 1969.
61. Gotloib L, Mines M et al.: Hemodynamic effects of increasing intra-abdominal pressure in
peritoneal dialysis. Perit Dial Buill: 41-43, 1981.
62. Khanna R, Oreopoulos DG et al.: Continuous ambulatory peritoneal dialysis (CAPD) after
three years: Still a promising treatment. Perit Dial Buill: 24--34, 1981.
63. Ponce SP, Pierratos A et al.: Comparison of the survival and complications of three permanent
peritoneal dialysis catheters. Perit Dial Bull 2: 82-86, 1982.
64. Ramos TM, Burke DA, Veitch PS: Hernia of Morgagni in patients on continuous ambulatory
peritoneal dialysis. Lancet I: 161, 1982.
65. Chan MK, Baillod RA et al.: Abdominal hernias in patients receiving continuous ambulatory
peritoneal dialysis. Br Med J 283: 826, 1981.
66. Power DA, Edward N et al.: Richter's hernia: An unrecognized complication of chronic
ambulatory peritoneal dialysis. Br Med J 283: 528, 1981.
67. Rubin J, Raju S, Bower J: Abdominal hernia among patients undergoing continuous ambula-
tory peritoneal dialysis (CAPD). Kidney Int 21: 177,1982.
68. Madden MA, Beirne GJ et al.: Acute bowel obstruction: An unusual complication of chronic
peritoneal dialysis. Am J Kidney Dis 1: 219--221, 1982.
69. Digenis GE, Khanna R, Mathews R et al.: Abdominal hernias in patients undergoing CAPD.
Perit Dial Bull 2: 115-118, 1982.
70. Francis DM, Schofield I, Veitchh PS: Abdominal hernias in patients treated with continuous
ambulatory peritoneal dialysis. Br J Surg 69: 409, 1982.
71. Rubin J, Raju S, Teal N et al.: Abdominal hernia in patients undergoing continuous ambulatory
peritoneal dialysis. Arch Intern Med 142: 1453-1455, 1982.
72. Lee AB, Waffle! CM, Trebbin WM et al.: Clostridial myonecrosis. Origin from an obturator
hernia in a dialysis patient. JAMA 246: 1232-1233, 1981.
73. King C: Hydrocele: A complication of CAPD. Nephrol Nurse 3: 37-39, 1981.
74. Twardowski ZJ, Prowant BF, Nolph KD etal.: High volume low frequency CAPD. Kid Int 23:
64--70, 1983.
75. Epstein SW, Inouye T, Robson M et al.: Effect of peritoneal dialysis fluid on ventilatory
function. Perit Dial Bull 2: 120-123, 1982.
76. Rebuck AS: Peritoneal dialysis and the mechanics of the diaphragm. Perit Dial Bull 2: 109--111,
1982.
77. Thieler H, Riedel E, Pielesch W et al.: CAPD and pulmonary function. Proc EDTA 17: 333-
336,1980.
78. Odriozola J, Bahlmann J, Fabel H: Einfluss der peritoneal dialyse auf die Lungenfunktion. Klin
Wochenschr 49: 484-488, 1971.
79. Prefaut CH, Monteil AL. Ramonatxo M et al.: Closing volume and pulmonary gas exchange
during peritoneal dialysis. Bull Eur Physiopathol Respir 14: 755-764, 1978.
80. Guazzi M, Polese A, Magrini F et al.: Negative influences of ascites on cardiac function of
cirrhotic patients. Am J Med 59: 165-170, 1975.
81. Schurig R, Gahl G, Schartl M et al.: Central and peripheral hemodynamics in long term
peritoneal dialysis patients. Proc EDTA 16: 165-169, 1979.
82. Swartz C, Onesti G, Mailloux L et al.: The acute hemodynamic and pulmonary perfusion effects
of peritoneal dialysis. Trans ASAIO 15: 367-372, 1969.
83. Acquatella H, Perez-Rozas M, Burger B, Guinand-Baldo A: Left ventricular function in
uremia: A hemodynamic and echocardiographic study. Nephron 22: 160-174,1978.
84. Williams P, Kay R, Harrison J et al.: Nutritional and anthropometric assessment of patients on
CAPD over one year: contrasting changes in total body nitrogen and potassium. Perit Dial Bull
1: 82-87, 1981.
518

85. Katirtzoglou A, Oreopoulos DG, Husdan H et al. : Appraisal of protein loss in patients undergo-
ing continuous ambulatory peritoneal dialysis (CAPD). Nephron 26: 230-233, 1980.
86. Heide B, Pierratos A, Khanna R et al.: Nutritional status of patients undergoing CAPD. Perit
Dial Bull 3: 138-141, 1983.
87. Oreopoulos DG, Crassweller P, Katirtzoglou A et al. : Amino acids as an osmotic agent (instead
of glucose) in CAPD. In: M Legrain (eds), Continuous ambulatory peritoneal dialysis. Excerpta
Medica, Amsterdam, 1979, pp 335-340.
88. Dombros N, Oren A, Marliss E et al.: Plasma amino acid profiles and amino acid losses in
patients undergoing CAPD. Perit Dial Bull 2: 27-32, 1982.
89. Giordano C, DeSanto NG, Capodicase G et al.: Amino acid losses during CAPD. Clin Nephrol
14: 230-232, 1980.
90. Randerson D, Chapman GV, Farrell PC: Amino acid and dietary status in CAPD patients. In:
Atkins et al. (eds), Peritoneal Dialysis. Churchill Livingstone: 179-191, 1981.
91. Giordano C, DeSanto NG, Capodicasa G et al.: Studies on amino acids in diabetic patients
undergoing CAPD. Intern J Artif Organs 4: 62-67, 1981.
92. Tepper T, Van der Hem GK, Tume GJ et al.: Loss of amino acids during hemodialysis:
Qualitative and quantitative investigations. Chem NephrollO: 16-20, 1978.
93. Rubin ME, Gordon S: Individual plasma free amino acid in uremics: Effect of hemodialysis.
Nephron 5: 339-351, 1968.
94. Doyle DG, O'Dwyer WF, Marmody M, Gallagher V: Quantitative and qualitative changes in
plasma amino nitrogen and amino acids during hemodialysis. Ir J Med Sci 3: 475-482, 1970.
95. Gylyassy DFG, Peters JH, Lin SC, Ryan DM: Hemodialysis and plasma amino acid composition
in chronic renal failure. Am J Clin Nutr 5: 565-573, 1968.
96. Hecking E, Eohler H, Zobel R et al.: Treatment with essential amino acids in patients on chronic
hemodialysis: A double blind cross-over study. Am J Clin Nutr 1821-1826,1978.
97. Berlyne CM, Lee HA, Giordano C et al.: Amino acid loss in peritoneal dialysis. Lancet 1339-
1341, 1967.
98. Gjessing J: Addition of amino acids to peritoneal dialysis fluid. Lancet 812, 1968.
99. Kobayashi K, Manji T, Hiramatsu S et al.: Nitrogen metabolism in patients on peritoneal
dialysis. Contr Nephrol17: 93--100, 1979.
100. Williams FP, Marliss EB, Anderson G H et al. : Amino acid absorption following intraperitoneal
administration in CAPD patients. Perit Dial Bull 2: 124-130, 1982.
101. Oren A, Wu G, Anderson H et al.: Effective use of amino acid dialysate over four weeks in
CAPD patients. Trans ASAIO 29: 604-609, 1983.
102. Lowrie EC, Lazarus JM, Mocelin AJ: Survival of patients undergoing chronic hemodialysis and
renal transplantation. N Engl J Med 288: 863--867,1973.
103. Haire HM, Sherrard DJ, Scardapan D: Smoking, hypertension and mortality in a maintenance
dialysis population. Cardiovasc Med 3: 1163-1168, 1978.
104. Bagdale JD, Porte D Jr, Bierman EL: A metabolic consequence of chronic renal failure. N Engl
J Med 279: 181-185, 1968.
105. Brunzell JD, Albers n, Haas LD et al.: Prevalence of serum lipid abnormalities in chronic
hemodialysis. Metabolism 26: 903--910, 1977.
106. Ibels LD, Reardon MF, Nestel PJ: Plasma postheparin lipolytic activity and triglyceride cleara-
nce in uremia and hemodialysis patients and renal allograft recipients. J Lab Clin Med 27: 648,
1976.
107. Novarini A, Zuliani U, Bandini L et al.: Observations on lipid metabolism in chronic renal
failure during conservative and hemodialysis therapy. Eur J Clin Invest 6: 473, 1976.
108. Ibels LS, Simons LA, King JO et al.: Studies on the nature and causes of hyperlipemia,
maintenance dialysis and renal transplantation, 1975.
109. Lowrie EC, Lazarus JM, Mocelin AJ et al.: Survival of patients undergoing chronic hemodialysis
and renal transplant. N Engl J Med 288: 863,1973.
 519

110. Casaretto A, Marchioro T, Bagdade J: Hyperlipidemia following renal transplantation. Trans

Am Soc Artif Intern Organs 19: 154, 1973.
111. Lindner A, Charra B. Sherrard DJ, Scribner BH: Accelerated atherosclerosis in prolonged
maintenance hemodialysis. N Engl J Med 290: 697-701, 1974.
112. Frederickson DS, Levy RI, Lees RS: Fat transport in lipoproteins - an integrated approach to
mechanisms and disorders. N Engl J Med 276: 34-44,94-103,215-225,273-281,1967.
113. Norbeck HE: Serum lipoprotein in chronic renal failure. Acta Med Scand (Suppl) 649: I, 1981.
114. Cattran DC, Fenton SSA, Wilson DR, Steiner G: Defective triglyceride removal in lipemia
associated with peritoneal dialysis and hemodialysis. Ann Intern Med 85: 29-33, 1976.
115. Oreopoulos DG, Karanicolas S, Izatt S, deYeber GA: Dialysis and triglycerides. Ann Intern
Med 85: 679-680, 1976.
116. Oreopoulos DG, Khanna R, McCready W et al.: Continuous ambulatory peritoneal dialysis in
Canada. Dial and Transplant 9: 224-226, 1980.
117. Goldberg A, Sherrard DJ, Brunzell JD: Adipose tissue lipoprotein lipase in chronic hemo-
dialysis: Role in plasma triglyceride metabolism. J C1in Endocrinol47: 1173-1182,1978.
118. Cantaluppi A, Scalamogna A. Guerra L, Graziani G, Ponticelli C: Plasma lipid and lipoprotein
levels in patients treated with CAPD. Perit Dial Bull 2: 99, 1982.
119. Triolo G, Boggio-Bertinet D, Salomone M, Yercellone A: Changes in serum lipids with
prolonged CAPD. Perit Dial Bull 2: 192-193, 1982.
120. Nolph KD, Ryan L, Prowant B, Twardowski ZJ: A cross sectional study of serum vitamin D and
triglyceride concentrations in a CAPD population. Perit Dial Bull (Accepted for publication),
1984.
121. Lindholm B, Karlander SG, Norbeck HE: Glucose and lipid metabolism in peritoneal dialysis.
In: Peritoneal Dialysis, (LaGreca et al. (ed», Wichtig Editore, Milano, pp 219-230,1983.
122. Lindholm B, Bergstrom J, Norbeck HE: Lipoprotein metabolism in patients on continuous
ambulatory peritoneal dialysis. In: Advances in Peritoneal Dialysis, (CM Gahl et al. (ed»,
Excerpta Medica, Amsterdam/Oxford/Princeton, pp 434-436, 1981.
123. Gokal R, Ramos JM, McGurth JG, Ward MK, Kerr DNS: Hyperlipidemia in patients on
continuous ambulatory peritoneal dialysis. In: Advances in Peritoneal Dialysis. (GM Gahl et al.
(ed», Excerpta Medica, Amsterdam, pp 461-466,1981.
124. Turgan C, Feehally J, Bennet S, Davies TJ, Walls J: Effect of continuous ambulatory peritoneal
dialysis (CAPD) on lipid metabolism. In: GM Gahl et al. (ed), Advances in Peritoneal Dialysis.
Excerpta Medica, Amsterdam/Oxford/Princeton, 1981, pp 417-420.
125. deFremont JF, Bataille P, Morinvere P, Kaczmareck p, Fievet P, Fourmier A: Metabolic
tolerance of continuous ambulatory peritoneal dialysis (CAPD). In: GM Gahl et al. (ed),
Advances in Peritoneal Dialysis. Excerpta Medica, Amsterdam/Oxford/Princeton, 1981, pp
446-448.
126. Splendiani G, Acitelli S, Albano Y, Gianforte A, Tancredi M: Metabolic aspects of CAPD. In:
GM Gahl et al. (ed), Advances in Peritoneal Dialysis. Excerpta Medica, Amsterdam/Oxford/
Princeton, 1981, pp 446-448.
127. DiGuilio S, Boulu R. Druecke T et al.: Clofibrate treatment of hyperlipidemia in chronic renal
failure. Clin Nephrol3: 304, 1977.
128. Goldberg A, Brunzell J, Holmes J, Sherrard D: Clofibrate effective treatment for uremic
hypertriglyceridemia. Kidney Int 8: 412, 1975.
129. Kurokawa K: Clofibrate in hemodialysed patients. N Engl J Med 296: 942 (letter), 1977.
130. Bridgman JF, Rosen SM, Thoyer 1M: Complications during clofibrate treatment of nephrotic
syndrome hyperlipoproteinemia. Lancet 2: 506, 1972.
131. Pierides AM, Alvarez-Ude F, Kerr DNS: Clofibrate induced muscle damage in patients with
chronic renal failure. Lancet 2: 1279,1975.
132. Kijima Y, Sasaoka T, Kanayama M. Kubota S: Untoward effects of clofibrate in hemodialysis
patients. N Engl J Med 296: 515 (letter), 1977.
520

133. Demedts W, Desager JP, Bilpaire F et al.: Life threatening hyperkalemia associated with
clofibrate-induced myopathy in a CAPO patients, 1983.
134. Lameire N, Ringoir S: An overview of peritonitis and other complications of CAPO. In: M
Legrain (ed), Proc Int Symp. Excerpta Medica, Amsterdam, 1979, pp 229-237.
135. Moncrief JW, Pyle WK, Simon P, Popovich RP: Hypertriglyceridemia, diabetes mellitus and
insulin administration in patients undergoing CAPO. In: CAPO update eds. Moncrief JW,
Popovich RP, Mason Publishers: New York, 1981, pp 143-165.
136. Beardsworth SF, Goldsmith HJ, Stanbridge BR: Intraperitoneal insulin cannot correct the
hyperlipidemia of CAPO. Perit Dial Bull 3: 126-127, 1983.
137. Blumberg A, Hanek A, Sander G: Vitamin nutrition in patients on CAPO. Clin Nephrol20:
244-250, 1983.
138. Fried W: Anemia in patients with chronic renal failure. Int J Artif Organs 32: 62-64, 1980.
139. Erslev AJ: Anemia of chronic renal disease. Arch Int Med 126: 774-780, 1970.
140. Koch KM, Patyna WB, Shaldon S, Hendler ED: Anemia ofthe regular hemodialysis patient and
its treatment. Nephrol12: 405-410, 1974.
141. DePaepe M, Lameire N, Schelstraete K, Ringoir S: Changes in red cell mass, plasma volume,
and hematocrit in patients on CAPO. Proc EDTA 18: 286-289, 1981.
142. Diaz-Buxo JA, Farmer CD, Walker PJ et al.: Continuous cyclic peritoneal dialysis - A prelimi-
nary report. Artif Organs 5: 157-161, 1981.
143. Lamperi S, Carozzi S, Icardi A: In vitro and in vivo studies of erythropoiesis during CAPO. Perit
Dial Bull 3: 94-96, 1983.
144. Spinowitz BS, Sherwood J, Galler M, Chary tan C: Anemia and oxygen affinity in patients on
CAPO. Perit Dial Bull S3: 33--35, 1983.
145. Mittman N, Sherwood JB, Chang H: Erythropoietin levels in chronic dialysis patients deter-
mined by radio immune assay. Kidney Int 21: 174, 1982.
146. Mittman N: Erythrocyte glucose-t-phosphate dehydrogenase in chronic renal failure and after
renal transplantation. Scand J Hematol24: 307-314, 1980.
147. Falk RJ, Mattern WD, Lamanna RW etal.: Iron removal during CAPD. Kidney Int 24: 110--112,
1983.
148. Stanbaugh GH, Holmes A W, Gillit 0 et al.: Iron chelation therapy in CAPO: a new and
effective treatment of iron overload disease in ESRD patients. Perit Dial Bull 3: 99-102, 1983.
149. Hefti JE, Blumberg A, Marti HR: Red cell survival and red enzymes in patients on CAPO. Clin
Nephrol19: 232-235, 1983.
150. Castaldi PA, Rozenberg MC, Stewart JH: Bleeding disorders of uremia. Lancet 2: 66,1966.
151. Stewart JH, Castaldi PA: Uremic bleeding: A reversible platelet defect corrected by dialysis.
Quart J Med 36: 409-423, 1967.
152. Horowitz HI: Uremic toxins and platelet function. Arch Int Med 126: 823, 1970.
153. Nenci GG, Berrettini M, Agnelli G et al.: Effect of peritoneal dialysis, hemodialysis and kidney
transplantation on blood platelet function. Nephron 126: 287, 1979.
154. Lindsay RM, Koens F, Linton AL, Oreopoulos DG, deVeber GA: Platelet function in patients
on long term peritoneal dialysis. Clin Nephrol 6: 335, 1976.
155. Remuzzi G, Livio M, Marchiaro G et al.: Bleeding in renal failure: altered platelet function in
chronic uremia only partially corrected by hemodialysis. Nephron 22: 347,1978.
156. Arends JP, Krediet RT, Boeschoten EW et al.: Improvement of bleeding time, platelet aggrega-
tion and platelet count during CAPO treatment. Proc EDT A 18: 280--285, 1981.
157. Heiden 0, Mielke CH, Rodvien R: Impairment by heparin of primary hemostasis and platelet
(14C) 5-hydroxytryptamine release. Br J Hematol36: 427-436, 1977.
158. Avioli LV, Teitelbaum SL: The renal osteodystrophies. In: LE Earley, CW Gottschalk(eds),
Disease of the Kidney, Little, Brown and Co. Boston, 1979, pp 307-370.
159. Coburn JW: Renal osteodystrophy. Kidney Int 17: 677-693, 1980.
160. Gokal R, Ellis HA, Ramos JM et al.: Improvement in secondary hyperparathyroidism in
 521

patients on CAPD. In: GM Gahl, M Kessel, KD Nolph (eds), Advances in Peritoneal Dialysis.
Excerpta Medica, Amsterdam, 1981, pp 461-466.
161. Calderaro V, Oreopoulos DG, Meema HE et al.: The evolution of renal osteodystrophy in
patients undergoing CAPD. Proc EDTA 17: 533-542, 1980.
162. Nolph KD, Sorkin MI, Arjanin D et al.: CAPD - three years experience at a single center. Ann
Intern Med 92: 609-612, 1980.
163. Tielmans C, Aubry C, Dratwa M: The effects of continuous ambulatory peritoneal dialysis
(CAPD) on renal osteodystrophy. In: GM Gahl, M Kessel, KD Nolph (eds), Advances in
Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1981, pp 455-460.
164. Teitelbaum S, Fallon MD, Gearing BK et at.: The effects of CAPD on bone histomorphology.
Kidney Int 21: 180, 1982.
165. Digenis G, Khanna R, Pierratos A et al.: Renal osteodystrophy in patients maintained on CAPD
for more than three years. Perit Dial Bull 3: 81-86, 1983.
166. Dalmez J, Martin K, Harter H et al.: Effects of continuous ambulatory peritoneal dialysis on the
removal of iPTH in renal failure (Abstract). Am Soc Artif Intern Organs 9: 44, 1980.
167. Dawborn JK, Brown DJ, Douglas MC et al.: Parathyroidectomy in chronic renal failure.
Nephron 33: 100-105, 1983.
168. Gokal R, Parkinson I, Ramos M et al.: Elevated serum aluminium levels in patients on CAPD.
In: GM Gahl, M Kessel, KD Nolph (eds), Advances in Peritoneal Dialysis. Excerpta Medica,
Amsterdam, 1981, pp 478-482.
169. Sorkin MI, Nolph KD, Anderson HO et al.: Aluminium mass transfer during CAPD. Perit Dial
Buill: 91-94, 1980.
170. If A, Graf H, Pengerra WF et af.: Serum aluminium and CAPD. Ann Int Med d92: 130-131,
19EO.
171. Mailis HA: Pathogenesis and treatment of back pain in peritoneal dialysis patients. Perit Dial
Bull 3: (S3): 41-42, 1983.
172. Kolendorf K, Moller BB, Rogowski P: The influence of chronic renal failure on serum and
urinary thyroid hormone levels. Acta Endocrinol 88: 80-88, 1978.
173. Gomez F, de la Cueva R, Wauters J-P et al.: Endocrine abnormalities in patients undergoing
long-term hemodialysis. The role of prolactin. Am J Med 68: 522-530, 1980.
174. Gomez-Pan A, Alvarez-Ude F, Yeo PPB et al.: Function of the hypothalamo-hypophysial-
thyroid axis in chronic renal failure. Clin Endocrinolll: 567-574, 1979.
175. Hasegawa K, Matsushita Y, Otomo S et al.: Abnormal response of thyrotrophin and growth
hormone to thyrotrophin releasing hormone in chronic renal failure. Acta Endocrinol (Copenh)
79: 635-643, 1975.
176. Hershman JM, Krugmaan LG, Kopple JD et al.: Thyroid function in patients undergoing
maintenance hemodialysis: unexplained low serum thyroxine concentration. Metabolism 27:
755-759, 1978.
177. Joasoo A, Murray IPC, Parkin J et al.: Abnormalities of in vitro thyroid function tests in renal
disease. Q J Med XLIII: 245-261, 1974.
178. Kalk WJ, Morley JE, Gold CH et al.: Thyroid function tests in patients on regular hemodialysis.
Nephron 25: 173-178, 1980.
179. Karunanikhi A, Kanagasbapathy AS, Shastry JCM et al.: Thyroid function in patients with
chronic renal failure. Indian J Med Res 69: 792-797, 1979.
180. Savdie E, Stewart JH, Mahony JF et al.: Circulating thyroid hormone levels and adequacy of
dialysis. Clin Nephrol 9: 68-72,1978.
181. Silverberg DS, Ulan RA, Fawcett DM et al.: Effects of chronic hemodialysis on thyroid function
in chronic renal failure. Can Med Assoc J 109: 282-286, 1973.
182. Sofferr 0, Pelet D, Segal S et al.: Thyroid function in hemodialysis. Isr J Med Sci 15: 836-839,
1979.
183. Spector DA, Davis PJ, Helderman JH et al.: Thyroid function and metabolic state in chronic
renal failure. Ann Int Med 85: 724-730, 1976.
522

184. Wassner SJ, Buckingham BA, Kershnar AJ et al.: Thyroid function in children with chronic
renal failure. Nephron 19: 236-241, 1977.
185. Lim VS, Fang VS, Katz AI et al.: Thyroid dysfunction in chronic renal failure. A study of the
pituitary-thyroid axis and peripheral turnover kinetics of thyroxine and triiodothyronine. J Clin
Invest 60: 522-534, 1977.
186. Neuhaus K, Baumann G, Walser A et al.: Serum thyroxine and thyroxine-binding protein in
chronic renal failure without nephrosis. J Clin Endocrinol Metab 41: 395-398, 1975.
187. Ramirez G, Jubiz W, Gutch CF et al.: Thyroid abnormalities in renal failure. A study of 53
patients on chronic hemodialysis. Ann Intern Med 79: 500--504, 1973.
188. Chopra IJ, Chopra U, Smith SR et al.: Reciprocal changes in serum concentration of 3,3' ,5'-
triiodothyronine (reverse T 3) and 3,4'5-triiodothyronine (T3) in systemic illnesses. J Clin EnQo-
crinol Metab 41: 1043-1049, 1975.
189. Faber J, Kirkegaard C, Lumholtz IB et al.: Measurements of serum 3' ,5'-diiodothyronine and
3,3'-diiodothyronine concentrations in normal subjects and in patients with thyroid and non-
thyroid disease: studies of 3' ,5'-diiodothyronine metabolism. J Clin Endocrinol Metab 48: 611-
617,1979.
190. Nicod P, Burger A, Staeheli V et al.: A radioimmunoassay for 3,3' ,5'-triiodo-L-thyronine in
unextracted serum: method and clinical results. J Clin Endocrinol Metab 42: 823-828, 1976.
191. Melmed S, Geola FL, Reed A Wet al.: A comparison of methods for assessing thyroid function in
nonthyroidal illness. J Clin Endocrinol Metab 54: 300--306, 1982.
192. Oddie TH, Flanigan WJ, Fishere DA: Iodine and thyroxine metabolism in anephric patients
receiving chronic peritoneal dialysis. J Clin Endocrinol 31: 277-282, 1970.
193. Felicetta JV, Green WL, Haas LB et al.: Thyroid function and lipids in patients with chronic
renal disease treated by hemodialysis: with comments on the 'free thyroxine index'. Metabolism
28: 756-763, 1979.
194. Pokroy N, Epstein S, Hendricks S et at.: Thyrotrophin response to IV TRH in patients with
hepatic and renal disease. Horm Metab Research 6: 132-136, 1974.
195. Bakke JL, Lawrence NL, Roy S: Disappearance rate of exogenous thyroid-stimulating hormone
(TSH) in man. J Clin Endocrinol Metab 22: 352-363, 1962.
196. Beckers C, Machiels J, Soyez C et al.: Metabolic clearance rate and production rate of thyroid-
stimulating hormone in man. Horm Metab Res 3: 34-36, 1971.
197. Cuttelod S, Lemarchand-Beraud T, Magnenat P et al. : Effect of age and role of kidneys and liver
on thyrotropin turnover in man. Metabolism 23: 101-113, 1974.
198. Charytan C, Thysen B, Gatz M, Freeman R, Alpert BE: Thyroid function tests in uremic
patients on maintenance dialysis: a comparison of CAPD and hemodialysis. Perit Dial Bull 3:
(SI): 27-29,1983.
199. Semple CG, Beastall GH, Henderson IS et al.: Thyroid function and continuous ambulatory
peritoneal dialysis. Nephron 32: 249-252, 1982.
200. Herrmann J, Kruskemper HL, Grosser KD et at.: Peritoneal Dialyse in der Behandlung der
Thyreotoxishen Ktise. Dtsch Med Wochenschr 96: 742-745, 1977.
201. Herrmann JH, Schmidt HJ, Kruskemper HL: Thyroxine elimination by peritoneal dialysis in
experimental thyrotoxicosis. Horm Metab Res 5: 180--183, 1973.
202. Oddie TH, Flanigan WJ, Fisher DA: Iodine and thyroxine metabolism in anephric patients
receiving chronic peritoneal dialysis. 31: 277-282, 1970.
203. Ramirez G, O'Neill W Jr, Jubiz W, Bloomer HA: Thyroid dysfunction in uremia: evidence for
thyroid and hypophyseal abnormalities. Ann Intern Med 84: 672-676, 1976.
204. Farrington K, Boss AMB, Varghese Z et at.: Thyroid function tests in dialysis patients. Dial
Transplant 9: 846-846,873,1980.
205. Herrmann J, Schmidt HJ, Kruskemper HL: Thyroxine elimination by peritoneal dialysis in
experimental thyrotoxicosis. Horm Metab Res 5: 180--183,1973.
206. Bonoumini V: Hypothyroidism secondary to chronic peritoneal dialysis (CAPD) (Abstract).
Proc EDTA 12, 1980.
 523

207. Nolph KD, Sorkin MI, Rubin J et al.: Continuous ambulatory peritoneal dialysis: Three years
experience at one center. Ann Intern Med 92: 609-613, 1980.
208. Walker F, From G, Khanna R, Digenis GE, Oreopoulos DG: Study of thyroid function in
patients on CAPD. 15th Ann Meeting ASN 15: 70A, 1982.
209. Boero R, Quarello F, Bellardi P, Piccoli G: Blunted response to TRH stimulation in CAPD
patients. Perit Dial Bull 3: 213, 1983.
210. Selgas R, Albero R, Beberide JM et al.: Evaluation of thyroid function in patients treated with
CAPD. Perit Dial Bull 3: 25-29,1983.
211. Cowden EA, Ratcliffe WA, Ratcliff JG, Dobbie JW, Kennedy AC: Hyperproloactinemia in
renal disease. Clin Endocrinol 9: 241-246, 1978.
212. Tolis G, Goltzman D, Guyda H, Mountokalakis T: Hormonal derangements in uremia. J
Endocrinol Invest 3: 83-97, 1980.
213. Sievertsen GD, Lim VS, Nakavatase C, Frohman LA: Metabolic clearance and secretion rates
of human prolactin in normal subjects and in patients with chronic renal failure. J Clin Endo
Metab 50: 846-852, 1980.
214. Nicoletti I, Buocristiani U, Filipponi P et al. : Prolactin and growth hormone dynamics in chronic
renal failure: Effect of hemodialysis, peritoneal dialysis and kidney transplantation. IRCS Med
Sci Biochem 9: 879-880,1981.
215. Martin JB: Pathophysiology of growth hormone regulation. Clinical Neuroendocrinology. G
Tolis (ed). New York, Raven Press, 1979, pp 269-277.
216. Semple CG, Beastall GH, Henderson IS et al. : The pituitary testicular axis of uremic subjects on
hemodialysis and CAPD. Acta EndocrinollOl: 464-467, 1982.
217. Gomez F, De La Gueva R, Wauters JP, LeMarchand-Beraud T: Endocrine abnormalities in
patients undergoing long-term hemodialysis. Am J Med 68: 322-350, 1980.
218. Rice GG: Hypermenorrhea in the young hemodialysis patient. Am JObst Gynecol116: 539-543,
1973.
219. Goodwin NJ, Valenti C, Hall JE et al.: Effects of uremia and chronic hemodialysis on the
reproductive cycle. Am JObst Gynecol100: 528-535, 1968.
220. Pepperell RJ, Adam WR, Dawborn JK: Hemodialysis in the management of chronic renal
failure during pregnancy. Austr NZ Obstet GynaecollO: 180-186,1970.
221. Unzelman RF, Alderfer GR, Chojnacki RE: Pregnancy and chronic hemodialysis. Trans Am
Soc Artif Intern Organs 19: 144-149, 1973.
222. Sheriff MH, Hardman M, Lamont CA et al.: Successful pregnancy in a 44-year-old hemodialysis
patient. Br J Obstet Gynaecol 5: 386-389, 1978.
223. Kobayashi H, Matsumoto Y, Otsubo 0, Naito T: Sucessful pregnancy in a patient undergoing
chronic hemodialysis. Obstet Gynaecol 57: 382-386, 1981.
224. Ackrill P, Goodwin FJ, Marsh FP et al.: Successful pregnancy in patient on regular dialysis. Br
Med J 2: 172-174, 1975.
225. Wing AJ, Brunner FP, Brynger H et al.: Combined report on regular dialysis and transplant in
Europe VIII. Prwoc EDTA 15: 4-76, 1978.
226. Brunner FP, Brynger H, Chantler C et al.: Combined report on regular dialysis and transplanta-
tion in Europe IX. Proc EDTA 16: 2-73, 1979.
227. Winchester JF, Foegh M, Kloberdanz N et al.: Return of menstruation and improvement in
sexual function as a result of CAPD. Abstracts II, International Symposium on Peritoneal
Dialysis, Berlin, (West)Germany 75, 1981.
228. Rubin J: Can a female patient on CAPD become pregnant - Perit Dial Buill: 44, 1981.
229. Cattran DG, Benzie RJ: Pregnancy in a CAPD patient. Perit Dial Bull 3: 3-16, 1983.
230. Kioko EM, Shaw KM, Clarke AD, Warren DJ: Successful pregnancy in a diabetic patient
treated with CAPD. Diabetes Care 6: 298-300, 1983.
[Link] M, Spinowitz B, Charyton C, Kabadi M, Freeman R: Reproductive function in dialysis
patients. CAPD vs. Hemodialysis. Perit Dial Bull 3: 530-32, 1983.
524

231. Lim VS, Henriquez C, Sievertsen G, Frohman LA: Ovarian function in chronic renal failure:
Evidence suggesting hypothalamic anovulation. Ann Intern Med 93: 21-27, 1980.
232. Lim VS, Kathpalia SC, Frohman LA: Hyperprolactinemia and impaired pituitary response to
suppression and stimulation in chronic renal failure: reversal after transplantation. J Clin
Endocrinol Metab 48: 101-107, 1979.
233. Peces R, Horcajada C, Lopez-Novos JM et al.: Hyperprolactinemia in chronic renal failure:
impaired responsiveness to stimulation and suppression. Nephron 28: 11-16, 1981.
234. Khanna R, Oreopoulos DG, Dombros N et al.: Continuous ambulatory peritoneal dialysis
(CAPD) after three years: still a promising treatment. Perit Dial Buill: 24--34, 1981.
235. Nolph KD, Sorkin MI, Moore H: Autoregulation of sodium and potassium removal during
continuous ambulatory peritoneal dialysis. Trans Am Soc Artif Intern Organs 6: 334--337, 1980.
236. Leenen FHH, Shah P, Boer WH, Khanna R, Oreopoulos DG: Hypotension on CAPD. An
approach to treatment. Perit Dial Bull 1983 3S: 533-35, 1983.
237. Glasson LF, Riondel RH, Vallotton MB: Plasma renin activity and plasma aldosterone in
CAPD. Clin Sci 63: S207-209, 1982.
238. Osmond DH, Loh A Y, Dombros N, Oreopoulos DG, Rapoport A: Effects on CAPD on the
renin-angiotensin system. Clin Res 26: 870A, 1978.
239. Silverberg S, Oreopoulos DG, Wise DJ et al.: Pericarditis in patients undergoing long-term
hemodialysis and peritoneal dialysis. Incidence, management and complications. Am J Med 63:
874--878, 1976.
240. Ribot S, Franket HJM, Gelchinsky I et al.: Treatment of uremic pericarditis. Clin Neprhol 2:
127-130, 1974.
241. Comty CM, Cohen SL, Shapiro FL: Pericarditis in chronic uremia and its sequels. Ann Intern
Med 75: 173--183, 1971.
242. Cohen GF, Burgess JH, Kaye M: Peritoneal dialysis for the treatment of pericarditis in patients
on chronic hemodialysis. Can Med Assoc J 102: 1363--1368,1970.
243. Twardowski ZJ, Alpert MA, Gupta RC, Nolph KD, Madsen BT: Circulating immune com-
plexes: possible toxins responsible for serositis in renal failure 35: 190-195,1983.
15. Peritoneal dialysis in children

STEVEN R. ALEXANDER

1. Introduction

Peritoneal dialysis has been a mainstay in the treatment of uremic infants and
children for more than twenty years. The intrinsic simplicity and safety of the
technique and the relative ease with which it can be adapted for use in patients of
widely divergent ages and sizes have resulted in widespread acceptance of per-
itoneal dialysis as the dialytic treatment of choice for infants and children with
acute renal failure (ARF) [1,2]. Most nephrologists who routinely treat pediatric
patients have had extensive clinical experience with peritoneal dialysis in the
setting of ARF. Until very recently, however, a role for peritoneal dialysis in the
treatment of large numbers of children with chronic renal failure seemed doubt-
ful, despite the fact that successful maintenance peritoneal dialysis had been
described in children as early as 1973 [3]. Of the 823 children under the age of 15
years who were alive on dialysis in the member countries of the European dialysis
and Transplant Association on December 31,1979, only 38 (4.6%) were being
treated with peritoneal dialysis [4]. A similar situation existed in the United
States where of the 1819 children under the age of 20 years who received
maintenance dialysis during the last six months of 1980, only 128 (7%) were
treated with peritoneal dialysis [5].
This situation now appears to be changing as a direct result of the discovery of
continuous ambulatory peritoneal dialysis (CAPO) by Popovich and associates in
1976 [6]. It is not surprising that pediatric nephrologists have been enthusiastic
participants in the renaissance of peritoneal dialysis which began with the intro-
duction of CAPO. A dialysis innovation which offers to combine the familiar
simplicity, safety and adaptability of acute peritoneal dialysis with such attractive
chronic features as near steady-state biochemical and fluid control, greatly re-
duced dietary restrictions and freedom from repeated needle punctures has
obvious potential advantages for pediatric patients.
In the five years which have elapsed since the first child was treated with
CAPO, some of the original enthusiasm for the use of the technique in children
526

appears to have been justified. The safety and effectiveness of CAPD in children,
at least for the short term, has been well established [7]. Whether CAPD will be
as successful over the long term is unknown, but for the present the use of CAPD
in children appears to be increasing at a rapid rate [1,7]. To those responsible for
the care of uremic infants and children, a knowledge of peritoneal dialysis has
never been more important than it is today.
This chapter will attempt to review what is presently known about the use of
peritoneal dialysis in pediatric patients. Attention will be focused on clinical
considerations and technical issues; occasionally specific information will be
offered which unavoidably reflects only the author's current approach to certain
clinical problems. The overriding goal of a chapter such as this is to provide some
practical assistance to those who are actively involved in the treatment of infants
and children with peritoneal dialysis.

2. Historical notes

The use of peritoneal dialysis in the treatment of children with renal failure was
first described by Swan and Gordon in 1949 [8] at a time when worldwide clinical
experience with peritoneal dialysis did not total 100 patients [9].Swan and Gor-
don described three acutely an uric children, 9 months, 3 years and 8 years of age
whom they treated with a technique known as continuous peritoneal lavage.
Modeled on methods first described by Fine et al. [10], continuous peritoneal
lavage required surgical implantation of two rigid peritoneal sumps (which were
actually operating room suction tips with multiple perforations), one in the upper
abdomen, the other into the pelvis. Dialysis was accomplished by the continuous
infusion of fresh dialysate into one sump while draining from the other, resulting
in unidirectional flow of large volumes of dialysate through the peritoneal cavity.
The children treated by Swan and Gordon received fresh dialysate at an average
rate of 1.4 liters per hour for treatment periods which continued 24 hours per day
for from 5 to 12 consecutive days.
Results were generally encouraging, despite technical limitations (e.g., dialy-
sate temperature was modulated by varying the number of 60-watt light bulbs in a
box placed over the inflow water bath). Two of the three children regained
normal renal function following continuous peritoneal lavage for 9 and 12 con-
secutive days. The third child (a 9-month old infant) was sustained for 28 days
before she succumbed to obscure complications. Peritonitis occurred only once
and responded to antibiotics. Removal of urea was excellent and fluid balance
was maintained by adjusting the dextrose content of the dialysate between 2 gm %
and 4 gm%. The authors also noted the substantial nutritional contribution
obtained from absorbed dialysate dextrose.
In summary, peritoneal dialysis as described by Swan and Gordon was time-
consuming, technically difficult, and, for 1949, very costly; however, their experi-
 527

ence demonstrated that children of almost any·age could be treated in this manner
and that for some of these children treatment with peritoneal dialysis could be
life-saving [8].
Despite this promising beginning, the use of peritoneal dialysis in children was
not reported again for over a decade. During that period the continuous lavage
technique was abandoned in favor of the intermittent techniques more familiar
today.
In 1960 Segar described the use of peritoneal dialysis in the treatment of three
severely poisoned neonates who, as the result of a horrifying error, had received
hospital nursery feedings prepared from a solution of boric acid [11]. When two of
these infants survived, the role of peritoneal dialysis in the treatment of certain
pediatric intoxications was established. The following year Etteldorf and associ-
ates in Memphis [12], and Segar and associates in Indianapolis [13], almost
simultaneously reported the successful treatment of salicylate intoxication with
peritoneal dialysis. Both of these groups also documented the effectiveness of
short -term intermittent peritoneal dialysis in the treatment of infants and children
with acute renal failure (ARF) [13, 14] Subsequent reports from many parts of the
world have established the pre-eminent role of peritoneal dialysis in the treat-
ment of ARF in infants [15, 16, 17] and children [18, 19, 20, 21].
Such widespread reliance upon acute dialysis in children might have been
expected to lead to extensive use of peritoneal dialysis to maintain children with
end-stage renal failure. Until recently, this has not been the case. The manual
peritoneal dialysis technique used throughout the 1960s required re-insertion of
the dialysis catheter for each treatment; this made prolonged use in infants and
small children essentially impossible. In the only published pediatric series,
Feldman and associates in 1968 described their results in 7 children who were
maintained on intermittent peritoneal dialysis (IPD) for Fi2 to 8 months while
awaiting transplantation [22]. None ofthese children required dialysis more often
than once every 7 to 12 days.
The development of a permanent peritoneal catheter, which culminated in the
1968 report by Tenckhoff and Schecter [23], made chronic IPD an acceptable
form of treatment for pediatric patients. In Seattle, Tenckhoff, Hickman and
their associates combined permanent peritoneal catheters with an automated
dialysate generation and delivery system which could be used in the home. In 1968
this group established the first pediatric home chronic IPD program [3].
For the better part of the next decade, a period of intense interest in hemo-
dialysis during which chronic peritoneal dialysis was all but forgotten elsewhere,
the Seattle group made steady progress in the science of long-term peritoneal
dialysis in children [24]. The importance of these fundamental contributions to
the current state-of-the-art of pediatric peritoneal dialysis cannot be over-
emphasized.
Additional limited experience with chronic IPD was reported from several
other centers during the next ten years [25, 26, 27, 28], but enthusiasm for the
528

technique was never great. IPD seemed to mandate many of the undesirable
features of hemodialysis (dietary restrictions, fluid limits, immobility during
treatments, complex machinery requiring extensive parental supervision) with-
out providing the one great advantage of hemodialysis - efficiency.
With the description of CAPO by Popovich, Moncrief and associates in
1976[6], a new era in the treatment of renal failure in children was begun. CAPO
was first used to treat children in Toronto in 1978 [29, 30]. Additional early
experience was soon reported from growing pediatric CAPO programs in Port-
land [31], Birmingham [32], San Francisco [33], Paris [34], and Los Angeles [35].
Despite widespread interest in the technique, there were probably fewer than
50 pediatric CAPO patients prior to the summer of 1980. This was largely due to
the availability of dialysate in the United States only in 2000 ml plastic containers
prior to July 1980. The early pediatric CAPO programs dealt with this problem in
several different ways, each of which had important drawbacks. Parents were
taught to discard most of the fluid from the 2-liter container before infusing the
remainder (San Francisco) [33], or to prepare small volume bags at home by
filling empty blood bank transfer packs to the appropriate volumes from 2-liter
containers (Portland) [36]; hospital pharmacies were also used to periodically
prepare and ship supplies of small volume bags to individual patients (Toronto
[37], Birmingham [38]). These wasteful, expensive and potentially risky tech-
niques could finally be abandoned for most older children in the summer of 1980
when dialysate in 500 ml and 1000 ml plastic containers became commercially
available in the United States. Subsequent addition ot"250 ml, 750 ml and 1500 ml
containers completed a range of standardized dialysate volumes which will ac-
commodate most pediatric patients. For those occasional children for whom no
standard size is appropriate, e.g., small infants (especially neonates) and young
children who have grown beyond their present exchange volume but who are still
too small to tolerate the next standard size, customized patient-specific exchange
volumes can be obtained from the manufacturer on a prescription basis [39]. Thus
it is now relatively easy to obtain the supplies necessary to treat a child of almost
any size with CAPO, from premature neonates to adolescents.
The most recent modification of peritoneal dialysis mechanics to attain the
status of a separate treatment modality is continuous cycling peritoncal dialysis
(CCPD). Price and Suki were probably first to describe the use of CCPD in a
child, although they called the technique PO PO ('prolonged dwell peritoneal
dialysis') [40]. At present, published experience with CCPD in children is sparse,
but recent reports suggest that this technique is gaining popularity among pedi-
atric dialysis programs [41, 42], especially for those adolescent patients who
comply poorly with the more intrusive daytime exchange regimen required by
CAPO.
Additional variations on the CAPO theme have been proposed to meet the
special requirements of very young infants [43], and further developments aimed
at selected patient groups are likely to appear so long as interest in peritoneal
dialysis for children remains at its present intensity.
 529

3. Developmental aspects of peritoneal dialysis kinetics

It is frequently stated that the peritoneal cavity of the child is a more efficient
dialysis system than that of the adult. The origins of this concept can be found in
studies of peritoneal anatomy performed over a century ago. In a paper read
before the Siberian Branch of the Russian Geographic Society in 1884 (surely one
of the more obscure moments in the history of peritoneal dialysis), P. V. Putiloff
presented direct oiled paper tracings of peritoneal contents with which he had
precisely measured the total peritoneal surface areas of a newborn infant and an
adult [44]. Putiloff found that the infant had almost twice the peritoneal surface
area relative to body weight as the adult.
Earlier measurements in adults had suggested that peritoneal surface area
approximated total skin surface area [45]. Putiloff's adult data supported this
relationship; in his measurements on the infant the correlation between skin and
peritoneal surface areas was not as good, but the general trend was as expected.
That is, since the ratio of skin surface area to body weight is greater in infants than
in adults, infants might also be expected to exhibit a greater peritoneal surface
area to body weight ratio, assuming peritoneal and skin surface areas to be similar
throughout life.
The clinical implications of these relationships were first addressed by
Esperanca and Collins in a now familiar report published in 1966 [46]. These
investigators measured peritoneal surface areas during autopsies performed on
six neonates and six adults. Their measurements confirmed Putiloff's observation
that the newborn infant's peritoneal surface area, relative to body weight, was
roughly twice that of the adult. This led Esperanca and Collins to postulate that
' ... peritoneal dialysis should be twice as efficient in the infant'.
To test this hypothesis they compared peritoneal urea clearances obtained
during dialysis performed on uremic puppies and adult dogs. Average urea
clearance observed in the puppies was nearly three times greater than that
observed in the adult animals, relative to body weight. To complete their studies,
Esperanca and Collins measured peritoneal urea clearance in an ll-day old
infant, finding it to be 49.8 mllmin170 kg, or ' ... roughly twice what is usually
accomplished in adult practice' [46).
Based on little more than these studies by Putiloff, Esperanca and Collins, the
notion that peritoneal dialysis was more efficient in children than in adults gained
wide acceptance. It should be noted that the only pediatric subjects in these
studies were neonates; while Esperance and Collins were careful to limit their
conclusions accordingly, others have routinely extended enhanced peritoneal
efficiency to children of all ages [19]. Interpretation of the urea clearance data
obtained by Esperance and Collins is also difficult, because different dialysis
mechanics were used in each study. Dialysate flow rates averaged 128 cC/kg/hr in
the puppies, while adult dogs were studied using only 42 cc/kg/hr, a difference in
flow rate which closely parallels the observed difference in urea clearances.
530

Similarly, exchange volumes used in the infant were several times greater than
the usual adult volumes (85 cc/kg in the infant vs 28 to 30 cc/kg in adults).
Comparable study conditions for urea clearance could only have been obtained if
exchange volume in the adults approached six liters.
Several subsequent studies of peritoneal mass transfer in infants have em-
ployed highly variable dialysis mechanics [15, 19]. This fact alone prevents mean-
ingful comparison with adult studies in most instances.
Comparisons of peritoneal mass transfer may be more easily understood after a
brief review of the physiology of transperitoneal solute movement. Current
knowledge holds that if comparative studies were to show peritoneal dialysis to be
'more efficient' in children, one or more of the following should be involved: [47]
1. Effective peritoneal surface area could be greater in children relative to body
weight than in adults.
2. Peritoneal membrane pcrmeability could be greater in children.
3. Peritoneal capillary blood flow could be greater in children than in adults.!
4. Ultrafiltration could be disproportionately greater in children than adults.
5. Differences in dialysis mechanics could produce apparent differences in effi-
ciency. 'Dialysis mechanics' refers to the following:
a. Dialysate flow rate (exchange volume per unit weight and exchange cycle
timing).
b. Physical characteristics of the dialysate (temperature, pH, osmolality, chem-
ical composition).
c. Distribution of dialysate within the peritoneal cavity (stagnant fluid film
thickness, dialysate pooling, catheter function, loculations, etc.).
Additional factors which might influence comparisons of dialysis efficiency in
children and adults include age-related differences in solute distribution volumes
and generation rates relative to body weight. For example, total body water in the
neonate approaches 70% of body weight [48], compared to the adult figure of
50% ± 5% ;thus, more urea per unit body weight must be removed from the
neonate than from the adult to produce an equivalent decline in BUN. Solute
generation rate is usually considered to be proportional to metabolic rate, which
is highest, relative to body weight, in infants, and then declines with advancing
age [49]. It could be argued that greater dialysis efficiency is needed by the young
to offset higher solute generation rates.
Infants grow at a much faster rate than older children and may generate less
solute for removal by dialysis during periods of vigorous anabolism. In fact, one
authority has referred to the rapid growth of the neonate as the infant's 'third
kidney' [50].
Recent systematic studies of peritoneal dialysis kinetics have sought to clarify
the concept of age-related differences in peritoneal function. Elzouki and associ-
ates determined peritoneal dialysance values for urea (DJ and inulin (DJ in six
puppies less than one month of age and five adult dogs [51]. Dialysance was
chosen for these comparative studies because it is a pure measure of the per-
 531

itoneal membrane area-permeability product when dialysis mechanics are held

constant and ultrafiltration does not occur. Simultaneous determination of Du
and D, allows calculation of the permeability index or dialysance ratio (DJ The
Dr is a dimensionless index of membrane permeability independent of effective
surface area [52].
Elzouki and associates found that Du and D, per kilogram of body weight were
1.66- and 2.8-fold greater, respectively, in the puppies than in the adult dogs [51].
Dr was also greater in the young dogs. These data suggest that both effective
membrane surface area and permeability are increased in young animals relative
to body weight and that this reflects intrinsic age-related differences in peritoneal
mass transfer.
Studies in humans have yet to confirm these findings; systematic studies of
peritoneal transport kinetics in neonates have not been reported. In studies
performed in nine children, 4 months to 18.5 years of age, Gruskin and associates
constructed peritoneal diffusion curves describing time-related changes in dialy-
sate to blood concentration ratios for seven different solutes [53]. Diffusion
curves were found to be fundamentally similar to adult reference curves, and
when dialysis mechanics were held constant in relation to body weight there were
no demonstrable differences in diffusion curves obtained for children of widely
divergent ages and weights. Mean peritoneal urea and creatinine clearances
scaled for weight were no different in children less than two years of age
compared to older children and were within the range of values reported for
adults.
Gruskin and associates rigidly controlled dialysis mechanics in these studies,
demonstrating both theoretically and actually the perturbations generated by
even minor alterations in exchange volume, relative to body weight and dwell
time [53]. Based on their results, they postulated that if dialysis mechanics were
held constant relative to body weight, any age-related differences in membrane
area-permeability products must be reflected in differences in dialysate to blood
solute concentration ratios at various times during an exchange. Unable to find
such differences among the nine children studied, these investigators concluded
that apparent age-related differences in dialysis efficiency observed in other
studies were in fact spuriously produced by the use in those studies of different
dialysis mechanics among the different age groups [53].
Using the methodology of Pyle et al. [54], Popovich and associates derived mass
transfer-area coefficients (MTACs) for various solutes from studies performed in
four children, 17 months to 6 years of age [55]. Pyle's model allows calculation of
the MTAC independent of dialysis mechanics and distinct from the mass transfer
contributions of ultrafiltration. The MTAC is then a measure of the maximum
rate of transperitoneal mass transfer and is determined solely by the product of
the effective peritoneal surface area and the membrane permeability to the solute
in question (i. e. , the area-permeability product) [54]. Convective transport in this
model is represented by the reflection coefficients computed for each solute (see
Chapter 5 for a more complete discussion of these concepts).
532

Popovich and associates found no significant difference between mean MTACs

observed in their young subjects and adult reference values when scaled for body
weight [55]. Similar agreement was obtained when maximum ultrafiltration rates
were compared.
While it is possible to conclude from Popovich's data that MT ACs are similar in
adults and children beyond about 2 years of age, when scaled according to body
weight, these data should be interpreted with caution. Only four children were
studied by Popovich and associates, and calculations for one of the four (the
youngest) gave substantially different MTAC results for three of the four solutes
studied [55].
In a similarly constructed study, the results of which are available only in
abstract form, Morgenstern and associates obtained somewhat different MT AC
results in five children, 1.5 to 18 years of age [56]. Nonetheless, these authors
reached conclusions similar to those of Popovich and associates; i.e., MTACs in
children older than about two years are not greater than in the adult when scaled
for body weight. They also found that reflection coefficients computed in children
were similar to adult values, suggesting that convective transport is also similar in
older children and adults.
In summary, the limited available systematic studies of peritoneal clearance,
peritoneal diffusion curves, and mass transfer-area coefficients have shown that
peritoneal mass transfer relative to body weight is similar among adults and
children beyond the age of about 4 months, suggesting that peritoneal membrane
effective surface area and permeability relative to body weight are similar beyond
the first few months of life [51]. Only in very young animals has an increased
peritoneal mass transfer efficiency been observed [51]. However, should these
observations in puppies be confirmed in adequately controlled studies in human
neonates, the original postulates of Esperanca and Collins will have been verified
and properly limited to neonates.

4. Peritoneal dialysis for acute renal failure

4.1. Oliguric and non-oliguric ARF

Acute renal failure (ARF) can be defined as a sudden (and frequently reversible)
deterioration in the ability of the kidneys to maintain homeostasis of body fluids
[2]. Earlier definitions emphasized oliguria as a cardinal feature of the syndrome
[58]; recent experience has documented that although retention of nitrogenous
wastes is a consistent finding in ARF, oliguria need not be present [59].
Non-oliguric ARF has not been as well characterized in pediatric patients as in
adults. Many contemporary pediatric studies have persisted in the use of oliguria
(usually defined as urine output less than about 1.0 cC/kg/hr) as the principle
clinical criterion upon which the diagnosis of ARF is entertained. In three recent
 533

studies of ARF in neonates and older infants where oliguria was not mandated by
study design, the combined incidence of non-oliguric ARF was found to be 12%
[21, 60, 61]. Prior to 1970, the incidence of non-oliguric ARF in adults was also
thought to be only about 10% of all cases of ARF [59]. Several large adult series
published in the last decade have shown that non-oliguric ARF now accounts for
30%-59% of adult ARF [62, 63]. Whether non-oliguric ARF actually became
more prevalent in adults during the 1970s, or simply more readily recognized is
controversial. However, a similar increase in the frequency with which non-
oliguric ARF is encountered in infants and children might be anticipated during
the 1980s if (to paraphrase Anderson and Schrier) [59] pediatricians could be
persuaded to abandon their time-honored but potentially misleading habit of
monitoring only urine flow as an index of renal function.

4.2. Incidence

In many parts of the world, ARF is a common pediatric disorder, occurring

primarily as a consequence of infectious diarrhea, dehydration, bronchopneumo-
nia, and sepsis [64]. In the United States, Canada and Europe, the incidence of
ARF is relatively low in the pediatric population as a whole, yet it occurs with
alarming frequency in certain patient groups who appear to be at increased risk.
Two such high-risk groups have been well described; they are: neonates who are
admitted to Newborn Intensive Care Units (ARF incidence =6%), [65], and
infants who undergo major cardiac surgery during the first year of life (ARF
incidence = 8% [61] to 10.5% [66]).

4.3 Etiology

The list of diseases and conditions which can cause ARF in infants and children is
quite extensive. It is possible to focus the differential diagnosis to some degree
based on the age of the child [2]. For example, neonates usually develop ARF
following perinatal hypoxia and shock, but congenital malformations, obstructive
lesions and renal vein thrombosis are all frequent causes of ARF in neonates
which rarely occur in other age groups. In developed countries the most com-
monly reported cause of ARF in children between 6 months and 3 to 5 years of
age is the hemolytic-uremic syndrome (HUS). Only in older children do the
primary glomerular diseases commonly occur.
A full review of the classification and differential diagnosis of ARF, and the
relative incidence of the various disorders that produce renal failure in children at
different ages is not the purpose of this chapter; excellent reviews of these
subjects are available in pediatric nephrology texts [67, 68].
534

4.4. Diagnosis

The diagnostic approach to ARF in all age groups is similar and has been
reviewed in detail elsewhere [21, 69]. Initial studies are usually directed at
identifying patients suffering from so-called 'pre-renal' azotemia in whom it is
possible to restore normal renal function by correcting renal hypoperfusion.
Clinical judgment based on a good history and a thorough physical examination
will accurately separate azotemic patients into 'pre-renal' and 'intrinsic renal
failure' categories in most cases, but occasionally the status of renal perfusion is
not readily determined from clinical criteria alone. In these situations in adult
patients the use of urinary diagnostic indices has gained popularity [70]. Recent
studies have attempted to adapt these concepts to the evaluation of the azotemic
neonate [71, 72, 73]. In neonates, normal renal tubular function might be mis-
taken for evidence of renal damage when compared to norms established for
older children and adults. For example, in neonates the fractional excretion of
sodium has been shown to correlate inversely with gestational age [74]. It might
be expected that this parameter as well as the renal failure index (see below)
might be higher in oliguric infants with 'pre-renal azotemia' who by virtue of renal
tubular immaturity cannot respond to renal hypoperfusion by reducing urinary
sodium concentration to the degree seen in older children and adults. Table 1 lists
representative values for various indices of renal function obtained in studies of
azotemic neonates, along with adult reference values.
The usefulness of diagnostic indices in infants and children beyond the neonatal
period has not been systematically studied. For most clinical purposes, adult
norms appear to be adequate [2].

4.5. Indications for dialysis

The conservative management of ARF in infants and children employs basically

the same approach to therapy as is used in adult patients [75]. Meticulous
attention to every detail of fluid and electrolyte therapy is essential in these small
patients in whom relatively minor errors can have serious consequences. Dietary
adjustments, phosphate binders, diuretics, bicarbonate, calcium, anti-hyperten-
sives and sodium-potassium exchange resins all can playa role in forestalling or
evading dialysis in some patients who continue to have sufficient urine output.
However, several factors are at work in children which tend to defeat even the
best laid conservative management plans. Children have higher metabolic rates
and, as a result, generate harmful solutes more rapidly. In the oliguric child,
relatively greater caloric requirements must be met within stringent limitations on
allowable fluid intake. Major catabolic activity as is seen in children with HUS
and in infants who have undergone major surgery or survived multiple perinatal
insults is likely to result in an increased rate of accumulation of potassium,
 535

Table 1. Urinary diagnostic indices in adults and neonates

Diagnostic Pre renal azotemia Oliguric acute renal failure Non-oliguric acute renal
indices failure
Adultc Neonatea. d Adult c Neonatea. d
Adultc Neonate"

UOs m e 518 ± 35 369 ± 20 343 ± 17

U Na + 31 ± 19 63 ±34
U/P",ea 18±7 30 ± 18 3±0.5 6±3 7±1
U IP creatmme 45 ±6 29 ± 16 17 ± 2 1O±4 17±2 9±2
RFI <1 <2.5 >1 >2.5 >1 >2.5
(0.6)b (1.9) (10) (11) (4) (9)
FENa (%) <1 <2.5 >1 >2.5 >1 >2.5
(O.4)b (1.4) (7) (5.8) (3) (6)

From Refs. [62, 70, 71, 72, 73].

a Not valid for infants less than 32 weeks gestational age.
b Value in parentheses is the average index in each category calculated from data in the references.
Generalizations useful in clinical practice are given above each calculated index.
C Mean ± 1 SEM.

d Mean±1 SD.

e mOsmlkg H 2 0
f mEq/L

(U/P)Na
FENa = fractional excretion of filtered sodium = (U/P) x 100.
creatinine

U
RFI = renal failure index (U/P) Na
creatimne

phosphate, urea and other solutes in the extracellular fluid which can reach
harmful proportions at surprising rates. Consequently, dialysis tends to be used
more frequently and sooner in children, especially in neonates and young infants
[76].
Widely accepted indications for dialysis in children are not much different from
those used to guide treatment in adults and are listed in Table 2 [2, 21, 76]. Such a
list fails to properly emphasize the need to consider the rate at which conditions
are approaching dialysis criteria as well as the need to assess the overall status of
the patient. A marginally acceptable clinical situation should not be tolerated for
long in children when the prompt institution of peritoneal dialysis can be relied
upon to improve nutritional status and provide better control of metabolic
parameters. The convenience and relative safety of peritoneal dialysis in children
allow the nephrologist to begin dialysis as soon as it is needed without undue
alarm over possible complications associated with the procedure itself.
536

Table 2. Indications for dialysis in acute renal failure in children a

Hyperkalemia (serum K+ >7.0 mEq/L)

Intractable acidosis
Fluid overload - usually with hypertension, congestive heart failure, or pulmonary edema
Severe azotemia (BUN >150 mg/dl)
Symptomatic uremia (encephalopathy, pericarditis, intractable vomiting, hemorrhage)
Hyponatremia, hypocalcemia, hyperphosphatemia (severe, symptomatic)
Fluid removal- for optimal nutrition, transfusions, infusions of cardiac drugs, etc.

a Please see text for qualifying comments; each case must be individualized.

4.6. Mortality rates and the influence of early dialysis

Nearly one half of all acute renal failure victims will die, death coming usually
within a few days to weeks of the appearance of renal insufficiency. Almost as if in
defiance of advancing dialysis technology, a mortality rate figure of :::::50% recurs
with sobering regularity in series after series of adult and pediatric patients
published during the past 15 years [61, 65, 66, 71, 72, 77]. Mortality rates tend to
vary among infants and children according to patient age and etiology of ARF [61,
65,72,77]. This fact, more than any real differences in management, is probably
responsible for the occasional series in which an unusually high or low mortality
rate is reported [21, 61, 78].
Early use of dialysis has been advocated as a means by which mortality rates
might be reduced among selected patient groups. With early dialysis comes
control of biochemical and fluid status and maximum nutritional support; the
logic inherent in such an approach is compelling and tends to obscure the paucity
of objective data available on the subject.
In 1975 Chesney and associates observed a 65% mortality rate among infants
less than a year of age who developed ARF following major cardiac surgery [61].
Only 6 of the 19 oligo-anuric infants were treated with peritoneal dialysis, and this
only after relatively protracted periods of conservative [Link] and
associates recently reported the results of a much more aggressive approach to the
use of peritoneal dialysis in 24 children who developed ARF following major
cardiac surgery [66]. Sixteen of these children were less than one year of age and
form a group which can be compared to that described in the report by Chesney
and associates [61]. For Rigden, anyone of the following criteria was considered
to be sufficient indication for peritoneal dialysis: (1) plasma potassium :::::6.0 mEq/
L; (2) BUN :::::240 mg/dl; (3) oliguria (urine output <1.0 cC/kg/hr) resistant to
volume expansion, dopamine, and furosemide and persistent for 4 hours; or (4)
fluid overload with pulmonary edema and increased atrial pressures. Nine of the
16 infants under one year of age died, for a mortality rate of 56%, only a small
improvement over the 65% mortality observed by Chesney and associates.
Book and associates reported somewhat better results in a smaller group of
 537

infants who, following cardiac surgery, received peritoneal dialysis 'within hours'
after urine output of less than 1.0 cc/kg/hr was found to be resistant to volume
expansion and/or dopamine infusion [79]. There were two deaths among the 7
infants less than a year of age in their series, for a mortality rate of 28.5%.
Comparisons among these studies are limited by the small number of patients
involved, and the inability to control for such variables as cardiac diagnosis and
intra-operative events. It seems reasonable to conclude that early dialysis clearly
does not reduce and may increase the chance for survival among oliguric infants
following cardiac surgery. Delaying dialysis much beyond the time required to
demonstrate the ineffectiveness of volume expansion, dopamine and diuretic
therapy seems unwarranted at this time.
Early peritoneal dialysis has also become generally recommended for children
with the hemolytic-uremic syndrome who remain anuric or severely oliguric [80].

4.7. Technical considerations

Catheters - temporary vs permanent. Nephrologists have traditionally relied on

percutaneously placed, polyethylene catheters for peritoneal dialysis in the acute
setting. Surgical placement of Tenckhoff silicone catheters in ARF patients has
recently become increasingly popular [1]. The choice between a percutaneously
placed, temporary catheter and the Tenckhoff catheter placed under direct vision
is usually somewhat arbitrary in the individual patient, although a number of
considerations obviously bear on the decision.
It is generally held that percutaneously placed catheters should be removed
after 72 hours due to the increased infection risk associated with longer usage
[20]. Thus, if the nephrologist could reliably predict the time at which adequate
renal function was going to return, the choice of a Tenckhoff catheter for all
patients expected to require dialysis for more than 3 to 5 days would be obvious.
Unfortunately, dialysis dependency varies from less than 48 hours to more than 3
weeks in children with ARF, and there are no reliable indicators of the coming
duration of dialysis present at the time the choice of a catheter must be made.
The advantages gained from catheter placement under direct vision in which
good immediate function can be assured must be weighed in the individual
patient against the risks and delays which occur when surgical placement requires
a general anesthetic in the operating room. For infants in cardiac recovery and
neonatal intensive care units, surgical catheter placement at the bedside using
local anesthesia is the preferred approach in our program and others [81, 82].

Temporary catheters. The most widely used percutaneous catheter is the Tro-
cath®2 which comes in pediatric, adult, and infant [83] sizes. A number of
modified catheter systems have been proposed for use in neonates, some of which
make use of commonly available devices such as intravenous catheters [81, 84].
538

While occasionally ingenious, these devices should rarely be necessary. The

pediatric or the adult Trocath®, with its intraperitoneal portion appropriately
trimmed, has been used successfully in infants weighing as little as 800 grams [85].
For these small babies the fenestrated distal segment is trimmed to a length of
only 2 cm to ensure that all side holes rest within the peritoneal cavity after
catheter insertion.

Tenckhoff catheters. Standard straight and coiled Tenckhoff catheters may be

used in the acute or chronic setting; catheters with single cuffs, useful in either
setting are the prevailing preference among many pediatric CAPD programs at
the moment. We have the manufacturer3 glue a single 1.0 cm Dacron®felt cuff at a
point 3 cm above the first side holes on all adult size straight Tenckhoff catheters
which we then use for acute or chronic dialysis in infants weighing 5 kg or more.
For smaller infants we prefer the pediatric size Tenckhoff which has a slightly
smaller outer diameter. The manufacturer provides a smaller Dacron® felt cuff
(0.6 cm) glued at a point 2.5 cm above the first side holes. Although even smaller
catheters are available, we have not found them necessary, nor is it particularly
cost-effective to maintain large supplies of many different catheter models and
sizes.

Percutaneous catheter placement technique. Percutaneous catheter placement is a

simple procedure which, nonetheless, can have the sort of life-threatening com-
plications one might expect to be associated with penetrating abdominal trauma.
The bladder is first emptied with a small urinary catheter which is then removed
to reduce infection risks. Precise urine output determinations are no longer
particularly important once the oliguric patient begins 'dialysis,
Adequate sedation of older infants and children prior to the procedure is
essential. Adults can be instructed to valsalva at the moment of trocar insertion,
but children are rarely able to cooperate to this degree. The practice of utilizing
the ', .. vocally induced abdominal wall tone' [86] (translation: 'terrified
screams') of the infant or child to act as a valsalva-producing maneuver is to be
deplored. Children old enough to describe it remember the catheter placement
procedure as frightening and painful [67]. In conscious children we use standard
pediatric doses of meperidine and hydroxyzine hydrochloride for preoperative
sedation. Small doses of ketamine (0.5 mg/kg i.v.) can then be used if needed to
keep the child asleep throughout the procedure. Insistance on good sedation may
increase the intraperitoneal fluid volume required to safely perforate the per-
itoneum in the relaxed child, Careful attention must be given to cardiorespiratory
status throughout the procedure to avoid problems of respiratory embarrass-
ment.
We prefer to use a 16 gao polyethylene over-the-needle catheter (intracath)4 for
infusion of dialysate to distend the abdomen. An assistant grasps the skin on
either side of the catheter insertion point and stretches it upward to provide
 539

resistance to the insertion of the intracath. Attaching the intracath to the dialysate
inflow line, the nephrologist inserts the intracath below the skin at a point in the
midline 2 to 3 cm below the umbilicus; another assistant now opens the inflow
clamp. By watching the drip chamber in the inflow line, dialysate can be seen to
pass drip by drip into the subcutaneous tissue. The intracath is advanced until a
steady stream of dialysate is observed in the drip chamber, demonstrating free
flow into the peritoneal cavity. The intracath is then advanced a bit farther, the
line interrupted, the needle withdrawn, the line reattached and the remaining
plastic catheter advanced until it is well into the peritoneal cavity. At least 30
cc/kg of warmed dialysate is infused, while close attention is given to the vital
signs of the child. Neonates may require additional circulatory support at this
stage. The abdomen should be well distended to provide adequate peritoneal
resistance to insertion of the dialysis catheter and trocar. This may require up to
50 cc/kg, the actual limit in each case determined by the point at which respiratory
fluctuations in the inflow stream become apparent.
A pre-trimmed pediatric Trocath® is now inserted after a small stab wound has
opened a path through skin and subcutaneous tissue. The cut edge of the catheter
can be passed through a flame to smooth the edges, or the edges beveled with iris
scissors. We estimate ideal intraperitoneal length to be 1 cm less than the distance
from xyphoid to umbilicus and trim the catheter to ensure that the first side holes
will reside at least 3 cm inside the peritoneal cavity. Short catheters perform
better than long ones.
When good in-and-out dialysate flow has been demonstrated the extra-abdom-
inal portion of the catheter is trimmed so that only 4-6 cm extend above the
abdominal wall; the catheter is then secured in place with a silk purse-string
suture and water-resistant tape.
Poor catheter function is a common problem with percutaneous catheters.
Omental envelopment or obstruction is usually responsible for catheter malfunc-
tions. When this occurs it is probably best to replace the catheter with a Tenckhoff
catheter under direct visualization.

Surgical catheter placement technique. In our program there is no difference in the

technique used to place Tenckhoff catheters whether the patient is thought to
have acute or chronic renal failure. The same single cuff catheters are used in both
groups. Because dialysis must begin immediately, emphasis is placed on creation
of a near water-tight seal at the peritoneal entry site. The operative procedure is
depicted in Figure la. While this illustration is of a neonate in whom the pediatric
size Tenckhoff has been used, the procedure described is essentially the same for
older infants and children as well [87].
The operative procedure begins with a 2 to 4 em incision made over the mid-
portion of the rectus muscle at or slightly above the level of the umbilicus. The
anterior rectus sheath is incised and the rectus muscle is separated bluntly. The
peritoneum is exposed, fixed by two temporary sutures, and then incised. Digital
540

examination assures that no bowel is adherent to the peritoneum. A small patch

of omental tissue which is readily available though the incision is then excised.
The Tenckhoff catheter is threaded over a lubricated catheter guide with a gentle
distal curve, and then guide and catheter are passed just beneath the anterior
abdominal wall into the contralateral deep pelvis . At this point care is taken to
trim the intraperitoneal portion so that the catheter tip rests in the anterior deep
pelvis but does not come in contact with the lateral or posterior abdominal wall
and is not imbedded in adhesions, omentum or loops of bowel. The catheter
which is left too long can result in painful infusion as well as poor function. The
peritoneum is now closed. A purse-string of non-absorbable suture is placed in
the peritoneum surrounding the catheter with the suture being passed through
the cuff at its lowest point in 3 or 4 places (Fig. Ib). When secured, this suture
pulls a collar of peritoneum around the base of the cuff creating a nearly water-
tight seal and anchoring the catheter in position. The anterior rectus sheath is
then purse-stringed around the cuff with the suture passing through the upper
aspect of the cuff in the same manner (Fig. 1c). A subcutaneous tunnel approx-
imately 4 to 5 cm in length is then created with care taken that the skin exit
aperture is small and fits tightly onto the catheter. Dialysate is now infused to
document the absence of leaks and good catheter function. The skin incision is
then closed and dialysis continued without interruption.

o.

4em (;" /
/
. ~
" ,.....
,
.

o.6em{"(" d:-.,;1/ .-"",

2.5em :' / ~)
.. ,
,

~
~.'.
\:<', .
',: .~

4 em ... :. ~ ~::

Figure 1. Schematic drawing of the surgical technique for Tenckhoff catheter placement which can be
used for either acute or chronic peritoneal dialysis. The use of a customized pediatric Tenckhoff
catheter in an infant weighing less than 5 kg is shown. For larger infants and older children , an adult-
size Tenckhoff catheter is placed using this same technique . Please see text for surgical details .
 541

Dialysate. Dialysate is commercially available from a growing number of manu-

facturers in standardized hydrous dextrose concentrations of 1.5,2.5 and 4.25%.
These solutions contain sodium, chloride, calcium, magnesium and lactate in
concentrations which are usually appropriate for use in the treatment of children
with ARF. A notable exception is the hypoxic neonate in whom oxidation of
lactate to bicarbonate may not be adequate [88]. The result may be a progressive
metabolic acidosis with accumulation of lactate and loss of bicarbonate into the
dialysis fluid. In this situation, or whenever a child on peritoneal dialysis develops
a progressive metabolic acidosis with a large anion gap, the dialysate should be
re-formulated to contain bicarbonate instead of lactate. A representative bicar-
bonate-containing dialysate formulation is given in Table 3. Note that calcium
must be given intravenously when bicarbonate-containing dialysis solutions are
used.

Dialysis mechanics. Usual peritoneal dialysis exchange volumes vary from 30 to

45 cc/kg, although we prefer somewhat smaller volumes for the first 24 to 48 hours
after surgical catheter placement. Exchange volumes of 50 cC/kg or more may
result in respiratory embarrassment [76]. Hydrothorax has also been reported in
association with large volumes [89], and low volumes «25 cc/kg) have been
associated with poor ultrafiltration in neonates [90]. There is rarely any reason to
use more than 40 cc/kg/exchange.
Initial stabilization on dialysis usually requires 24 to 48 hours of frequent
exchanges (40 to 60 minutes each, depending on catheter function) in order to
remove accumulated solutes and excess body fluid. This period corresponds to a
traditional IPD regimen. Once the child has been stabilized, if a Tenckhoff
catheter is in place, dialysis may then be continued indefinitely, using longer
dwell periods and fewer daily exchanges until a typical maintenance CAPD
regimen is reached.

Table 3. Peritoneal dialysis solution containing bicarbonate a

ml Na+ Ci- Mg++ S04~ HC0 3 - Hydrous

(mEq) (mEq) (mEq) (mEq) (mEq) dextrose
(grams)

NaCl (0.45%) 896.0 69 69

NaCl (2.5 mEq/ml) 12.0 30 30
NaHC0 3 (1.0 mEq/ml) 40.0 40 40
MgS04 (10%) 1.8 1.5 1.5
DsoW 50.0 25

Totals 999.8 139 99 1.5 1.5 40 25

Calculated osmolality = 423 mOsm/kg H]O.

aModified from Ref. [88).
542

Familiarity with the use of prolonged dwell periods in CAPD and CCPD has
stimulated adaptation of the CAPD regimen for use in ARF. In 1980 Poser and
Luisello described their use of this approach in 20 adults with oliguric ARF[91].
Abbad and Ploos van Amstel recently reported their success with a CAPD-type
dialysis regimen in the treatment of 5 anuric infants, 3 weeks to about 3 years of
age whose ARF was due to HUS [92]. CAPD was maintained in these children
for 10 to 33 days; eventually all regained renal function. After an initial stabiliza-
tion perod using IPD, the authors found the CAPD regimen to have many
benefits. The infants improved rapidly once on CAPD despite continued anuria.
They received unlimited diets averaging 2.1 grams of protein/kg and 81 kcal/kg per
day. Fluid and biochemical management was relatively easy despite unregulated
intake. Four of the five infants developed peritonitis, but there were no other
major complications.
In Oregon we have enjoyed similar success with a CAPD approach to ARF
patients, such that this is now our standard form of peritoneal dialysis once
stabilization with IPD is achieved. The steady-state biochemical and fluid control
characteristic of CAPD is particularly beneficial to small infants whose car-
diovascular status can be precarious. Further systematic studies are needed to
properly evaluate this method, but having become familiar with the many advan-
tages of the CAPD approach to ARF, any other dialysis regimen now seems to
me to be unnecessarily difficult.

4.8. Complications

Peritonitis is the most frequent serious complication of acute peritoneal dialysis

[2]. The incidence of infection appears to be directly proportional to length of
time on dialysis [20]. However, peritonitis occurs only rarely in some series [21].
Most authorities warn against prophylactic antibiotics and urge instead meticu-
lous technique and a high index of suspicion [21].
Massive intraperitoneal hemorrhage caused by deep trauma at the time of
percutaneous catheter insertion has been reported in up to 5% of cases [20, 61].
Adequate abdominal distension at the time of trocar insertion is essential; if
trocar insertion is not going well, it is probably advisable to place a catheter
surgically rather than risk repeated abdominal punctures.
A number of potentially serious metabolic derangements can occur in small
infants on peritoneal dialysis. Leumann and associates described 6 neonates with
ARF who developed the following complications during peritoneal dialysis:
hypophosphatemia, hypercalcemia, hypomagnesemia, hyponatremia, hyper-
glycemia, and hypoproteinemia [81]. Hypokalemia was avoided by adding KCI to
the dialysate. Frequent laboratory studies are necessary in these infants to
minimize metabolic complications. The use of a CAPD-like regimen once initial
stabilization has been accomplished should reduce the likelihood of serious
metabolic derangements [92].
 543

5. Peritoneal dialysis for end-stage renal disease in children

5.1. Intermittent peritoneal dialysis (IPD)

Reported experience with chronic IPD in children is esentially limited to four

series published since 1973 [3, 25, 26, 28]. Potter and associates recently sum-
marized this experience [28], suggesting that additional reports were unlikely
because CAPD has begun to supersede IPD in most pediatric dialysis pro-
[Link], reported experience with IPD in children has not been unfavora-
ble. Combining the results from four pediatric centers [3, 25, 26, 28], a total of 64
children were treated with IPD for an average of15 months. Five children died, 20
received renal transplants and 8 were transferred to hemodialysis, 7 of the 8
transfers occurring at a single center [25]. Peritoneal catheter survival averaged 12
months and peritonitis occurred at a rate of one episode every 15.8 patient-
months. Normal growth occurred in 6 of 30 children evaluated for short-term
growth. Psychological benefits were obtained from having dialysis performed at
home, and children treated with both in-center hemodialysis and home IPD
uniformly preferred IPD. Decreasing dialysis efficiency was cited as responsible
for the transfer to hemodialysis of more than one third of the patients in one series
[25]. Over the short term (therapy lasting one year or less) IPD compared
favorably to hemodialysis for children of comparable ages.
When compared to CAPD, however, IPD is not very appealing. CAPD is
simpler and provides greater weekly clearances of small and middle molecular
weight solutes when compared to standard 40 hours per week of IPD [93]. In
point of fact, very few children can be expected to spend 40 hours each week
attached to the IPD machinery unless dialysis treatments are undertaken more
often than thrice weekly [28]. Thus it would appear that IPD is rapidly becoming
relegated to a subordinate position to CAPD and its mechanized variant, CCPD.
In a very few years it seems likely that IPD will be used almost exclusively in the
treatment of acute renal failure in hospitalized patients, its use in chronic home
therapy having become another important, yet completed chapter in the annals of
peritoneal dialysis in children.

5.2. Continuous ambulatory peritoneal dialysis (CAPD)

In the five years which have elapsed since the first child was treated with CAPD,
the impact of this new therapy on the care of infants and children with ESRD has
been enormous. Although precise data are unavailable, there has probably been
at least a ten-fold increase in pediatric CAPD patients worldwide during the past
four years [2]. It is not surprising that CAPD has become a popular pediatric
dialysis technique. Because CAPD proceeds continuously, body fluid composi-
tion and volume change slowly, resulting in what could almost be considered a
544

'steady state'. The dysequilibrium syndrome experienced by many children dur-

ing hemodialysis does not occur in children on CAPD. Very few dietary restric-
tions are imposed on the pediatric CAPD patient; most children are encouraged
to eat an essentially unlimited diet, high in protein with generous fluid and
sodium allowances. The simplicity and safety of the technique allow it to be
performed at home by parents and older children, thereby returning the child
with ESRD to regular school attendance and allowing family vacations and other
normal childhood activities. CAPD avoids the many difficulties associated with
maintenance of vascular access in children and eliminates the need for frequent
painful needle punctures. CAPD has been shown to be particulary successful
when used in very small infants and young children, thereby extending routine
renal replacement therapy to an age group or whom dialysis and renal transplan-
tation have not been readily available in the past [94].

Patient selection. Prior to the introduction of CAPD, most pediatric ESRD

treatment programs were reluctant to accept infants for maintenance therapy
because treatment was technically demanding and often unsuccessful. As experi-
ence with CAPD in infants has grown, it has become more widely recognized that
CAPD can be a practical and effective maintenance therapy for very small infants
[95]. The ramifications of this new treatment strategy for infants with ESRD are
extensive and involve neonatologists, pediatric urologists, pediatricians and
many others who will soon be aware that their youngest and smallest patients with
irreversible renal failure now may survive on CAPD and then undergo renal
transplantation after one or two years of dialysis. A life expectancy of several
decades or more in this scenario does not seem unduly optimistic. New technol-
ogy is often a mixed blessing. The complex social, ethical and even legal issues
raised by this new treatment strategy for infants with renal failure demand serious
consideration; fortunately for the author, such considerations are beyond the
scope of this chapter.
Infants with omphalocele, gastroschisis or diaphragmatic hernia may be the
only children in whom CAPD is absolutely contraindicated [96]. Even neonates
with massive polycystic kidneys could undergo CAPD following unilateral
nephrectomy. In several large series [97, 98] and in our own experience [96, 99],
the following conditions have been compatible with successful management on
CAPD: prematurity, obstructive uropathy with cutaneous ureterostomies or
vesicostomy; prune belly syndrome, bilateral Wilms' tumor, recent abdominal
surgery, concurrent cancer chemotherapy or corticosteroid therapy, radi-
otherapy involving a portion of the peritoneum, and extensive intra-abdominal
adhesions requiring lysis to create an adequate intraperitoneal space. We have
found that age, sex, prior ESRD therapy, primary renal disease and renal
transplantation status have had no apparent influence on CAPD outcome.
We believe that there are at least 3 major criteria which should be met before
CAPD can be recommended for most children:
 545

1. The child should be an eventual candidate for renal transplantation.

2. The family should be motivated to learn and comply with the home dialysis and
chronic care program. Motivation and ability to cope with the rigors of the
home dialysis procedure may be the most important determinants of CAPD
success [100]. No objective criteria have been established by which family
motivation and coping abilities may be reliably predetermined. For example,
we have found that sufficiently motivated single working parents and parents
with limited intelligence and/or non-traditional life styles have performed
CAPD with great success. This has led our program to give a trial of CAPD to
any family requesting it regardless of their circumstances. In our five-year
experience in Oregon with over 50 CAPD/CCPD families there have been no
dropouts as a result of inability to cope with the procedure. That is not to say
that coping with CAPD has been easy for these families. An aggressive
approach to the 'parent fatigue syndrome' has been an essential element in the
continued success of these families (see below).
3. The facility must be able to provide the necessary multidisciplinary training and
intensive ongoing care and support the child and family will continue to require
while on CAPD. CAPD for children is provided most successfully by a spe-
cialized team which usually includes pediatric nephrologists, urologists, CAPD
nurses, renal dieticians, medical social workers, and child life therapists.
Children require a greater investment of time from each team member: In
Oregon we have found that one CAPD nurse can properly follow no more than
six pediatric patients, although this same nurse can follow additional adult
patients. When child/nurse ratios are higher than 6:1, the quality of care
received by all suffers and turnover within the nursing staff increases.

Peritoneal access. A reliable peritoneal catheter is the cornerstone of successful

CAPD. Surgical placement techniques suitable for use in infants and children
have been suggested by several authors [34, 37, 87, 101]. There is general
agreement on a few points: single-cuff catheters are preferred; the cuff must be
securely sutured to the peritoneum at the entry site; the intraperitoneal portion of
the catheter must be trimmed to fit the individual infant.
The catheter placement procedure we currently use has been outlined earlier in
this chapter and is depicted in Figure 1. We have found that peritoneography at
the time of catheter placement can identify hernias and potential hernias which if
unrepaired will later become clinically significant as a result of the increased
intraperitoneal hydraulic pressures seen in active children on CAPD [87, 102].
A summary of the reported experience with pediatric CAPD catheters through
the end of 1982 is contained in a recent review [7]. To this experience, which
involves only Tenckhoff catheters, should be added the first description of the use
of the Toronto Western Hospital catheter in children [103].
546

Mechanics of dialysis. As was seen earlier in this chapter, recent data suggest that
peritoneal efficiency is similar in adults and children beyond the first few months
of life when scaled on the basis of body weight. Despite this, a theoretical
framework for CAPD in children has not been well defined. Preliminary efforts
by Popovich and associates [55] and Gruskin and associates [51] are steps in the
right direction, but the complexities and assumptions involved in their methods
limit the usefulness of such approximations in the individual clinical setting.
Current practice has evolved empirically. Published guidelines from several
different pediatric CAPD programs are strikingly similar despite diverse patient
populations. Table 4 lists CAPD regimens used in three different pediatric
CAPD programs [33, 35, 96] along with adult reference values [104].
In Oregon our current practice consists of adjusting exchange volume and
frequency to achieve a total urea clearance offrom 210 to 250 cc/kg/day. Residual
renal function can proportionately reduce the amount· of dialysis needed accord-
ing to this scheme, although we have only rarely prescribed less than four daily
exchanges. Individual exchange volumes have varied in our patients from 35 to 45
cc/kg. Infants who are not yet 'ambulatory' are more tolerant of larger exchange
volumes than older infants and children in whom abdominal distension may limit
physical activity at exchange volumes approaching 50 cc/kg.
Careful nitrogen balance studies in stable adults on CAPD have shown that
when total urea clearance is 138 cc/kg/day and protein intake is 1.4 grams/kg/day
SUN averages 89 mg/dl and nitrogen balance is slightly positive [104]. Similar
studies have not been done in children. The observations described in Table 4
where similar average SUN is maintained in children receiving almost twice the
protein intake of the reference adult can be explained at least in part by the
different dialysis mechanics employed in adults and children. A CAPD regimen
that provides a 52% increase in urea clearance over that of the adult when scaled
by weight might be expected to provide sufficient dialysis for a protein intake

Table 4. Representative prediatric CAPD regimens

Reference Prescribed Mean total daily Mean SUN Mean daily

exchange urea clearance (mg/dl) protein intake
volume (ml/kg) (mllkg) (grams/kg)

Salusky et al. [33] 43 238 77 2Ab

Potter et al. [35] 35-50 164-224a 70 2.0b
Alexander et al. [96] 35-40 211 77 2.3 b
Adult Guidelines 25 238 89 1.4'
[104]

a Estimated.
b Estimated from dietary histories.
, Actual, from metabolic balance studies.
 547

which is increased by roughly the same order of magnitude. Only with appropri-
ate metabolic balance studies in children on CAPD will these relationships be
properly understood.
It should be emphasized that the CAPD regimen described above has evolved
from empiric observations and cannot be offered as providing 'adequate' dialysis.
Until this complex issue is better understood CAPD regimens for children will
remain largely empiric.

Ultrafiltration (UF) and water balance. Early observations suggested that ultra-
filtration (UF) was more difficult in infants and younger children on CAPD owing
to more avid dextrose absorption from the dialysate [37]. In one study children
under three years of age showed more rapid declines in dialysate dextrose
concentration and osmolality when compared to older children and adults [105].
However, this study did not control for relative differences in exchange volumes.
Subsequent studies have failed to demonstrate any age-related differences in UF
between pediatric and adult patients [55, 56].
Our current approach to fluid balance is to adjust dialysate dextrose concentra-
tion and dwell times to provide 35 to 40 cellOO kcallday of ultrafiltration in the
anephric child. Insensible and stool water losses in children usually amount to
about 50 ccllOO kcallday; total fluid turnover in these children on CAPD then
should average 85 to 90 ccllOO kcal/day. Ad lib fluid intake in our young patients
rarely exceeds this total. Infants require fortification of basic formula feedings to
achieve nutritional goals within these fluid limits [106]. On the other hand,
adolescents who ingest large quantities of salt and fluids must have more daily
ultrafiltration to avoid edema and hypertension.
Control of ultrafiltration must also be used to protect against dehydration. In
our experience, ultrafiltration is not significantly diminished even in the face of up
to 10% dehydration. Whenever there are increased body fluid losses as with
diarrhea, emesis, fever or diminished intake, as can be seen during episodes of
peritonitis, the CAPD regimen must be properly adjusted.

Growth. CAPD offers children several theoretical advantages over other dialysis
methods which were hoped would lead to improved growth, including essentially
unrestricted diets and the absorption of a continuous carbohydrate energy boost
from the dialysate. Steady-state biochemical and fluid control was considered
more conducive to growth than the fluctuations seen with hemodialysis.
Growth in children on CAPD, while better than that seen in hemodialyzed
children, did not meet expectations in early reports. Although normal growth was
seen in a few children, statural growth averaged only 50% to 68% of expected
growth in children on CAPD for relatively short periods of time [37,32,33,96].
Kohaut recently reported better growth in a group of 11 children, 3 months to 16
years of age at onset of CAPD, each of whom had been treated with CAPD for at
least 12 months when studied [107]. Five of the 11 children exhibited growth rates
548

in excess of 100% of expected and three others grew at rates ~88% of expected.
Fennell and associates recently compared growth among 58 children with
ESRD of whom 9 were on CAPD, 15 were on hemodialysis, and 34 had received a
renal transplant [108]. Children receiving CAPD grew better than those treated
by hemodialysis, and grew as well as the children who received a kidney trans-
plant when all children up to 15.5 years of age were considered. Average growth
velocity in the CAPD and transplant groups was 77% and 80% of predicted
growth velocity respectively. However, when only children less than 11 years of
age were considered, near normal to normal growth was obtained in no hemo-
dialysis patients, 33% of CAPD patients and 63% of transplant recipients.
Somewhat better growth was seen in Canadian children treated with CAPD for
an average of 15 ± 2 months [109]. Average growth was 82% of predicted for the
group as a whole and 5 children exhibited catch-up growth.

Nutrition. Improved nutritional status has been documented in children who are
begun on CAPD, but normal nutritional status has only rarely been seen [110].
Dietary regimens have remained somewhat arbitrary but in most programs
protein intake of at least 2.0 grams/kg/day is considered important as is a daily
energy intake of at least 100% of the Recommended Dietary allowance for a
normal child of the same height and sex [95, 98, 107]. Daily losses of protein into
the dialysate, which average 0.3 grams/kg/day in younger infants, must be con-
sidered when assessing nutritional therapy [37]. Similarly, carbohydrate ab-
sorbed from the dialysate may provide 8 to 20 kcal/kg/day in supplemental energy
[33, 35]. Unfortunately, children with ESRD often have poor appetites. In
infants less than 2 years of age, growth failure is especially costly in terms ofloss of
ultimate height potential [111]. We now institute nasogastric tube feedings in all
such infants when ad lib intake falls below recommended levels. Formula feed-
ings are supplemented with glucose polymers and medium chain triglycerides or
vegetable oils; we have also used protein supplements and amino acids in several
cases [95]. The optimum nutritional therapy for children on CAPD is currently
the subject of intense investigation in several centers.

Complications. Peritonitis is, not surprisingly, the most frequently encountered

complication of CAPD in children. When data on peritonitis incidence from six
early reports from pediatric CAPD programs were compiled, overall peritonitis
incidence was found to be 1. 70 episodes per patient-year [7]. This figure is similar
to that obtained for the largely adult CAPD population included in the NIH
CAPD Registry during 1981 and 1982 [112].
A detailed description of one large pediatric CAPD program's experience with
peritonitis was recently published [113]. Fine and associates observed 20 episodes
of peritonitis among 12 of 29 children undergoing CAPD during a 20-month
observation period, for an incidence of 0.96 episodes of peritonitis per patient-
year. This incidence compares favorable with that reported among children on
 549

home IPD [3, 26, 28]. No relationship was found between episodes of peritonitis
and the length of CAPO treatment. Presenting signs and symptoms included:
cloudy dialysate (18/20 episodes), abdominal pain (15/20), fever >38°C (11120),
peritoneal WBC count >501jLL (18/20), and >1001jLL (16/20), peritoneal WBC
differential count >50% neutrophils (18/20). The authors recommend that when
anyone of the classic triad of peritonitis symptoms occurs in a child on CAPO
(fever, abdominal pain, or cloudy fluid), peritonitis should be considered and
appropriate diagnostic studies obtained. Even if that peritoneal cell count is
<100/JLL the authors believe that treatment for presumptive peritonitis should
proceed until culture results are available. Four of 12 episodes of culture-proven
peritonitis in their series presented with peritoneal cell counts <1001jLL [113].
Recent reports have called attention to the improvement in anemia observed in
children treated with CAPO. In one study, children on CAPO required 0.16
transfusion per month to maintain hematocrits at an average of 21. 9% whereas
children treated with hemodialysis required almost five times as many transfu-
sions to maintain hematocrits at 19.6% [100].
It has been our impression in Oregon that younger children have diminished
energy and activity levels and eat even less when hematrocrits are allowed to
approach 20% [95]. Accordingly, we have maintained hematocrits 2::23% in our
younger patients. This has required an average of 0.33 transfusions (10 cclkg
packed red blood cells) per month [96]. Acute elevations in blood pressure are
frequent complications of transfusions in infants and children whose blood pres-
sures are normally well controlled.
Early reports described dramatic improvement in control of hypertension when
children were treated with CAPO. In one study, only 1 of 26 children required
antihypertensive therapy following institution of CAPO [35]. The beneficial
effects of CAPO on blood pressure control have not yet been elucidated, but may
relate to the maintenance of the child near his dry weight.
Persistence of renal osteodystrophy in children treated with CAPO has been a
disturbing finding in several reports [110, 114]. Hyperparathyroidism has been
only minimally affected by the CAPO procedure itself and essentially all pediatric
CAPO patients require vitamin 0 analogues.
Markedly elevated serum parathyroid hormone (PTH) levels have been re-
ported in some children on CAPO [100]. Other investigators report successful
control of the PTH levels in a majority of patients [107]. There is little agreement
currently on the manner in which renal osteodystrophy may best be followed and
treated in infants and children on CAPO. Care must be taken to ensure adequate
calcium intake in infants who may be on special formulas, or may have subnormal
intakes. Similarly, adequate phosphate must be provided to allow normal
growth. Dietary phosphate binders are usually unnecessary during periods of
rapid growth in infants on CAPO.
550

Renal transplantation. Renal transplantation has been performed in over 100

children on CAPD. In one large study, the outcome was no different in children
on CAPD than it was in children on hemodialysis or on no dialysis [115]. The
peritoneal catheter is usually left in place for up to 3 months following transplan-
tation; the catheter can then be used for dialysis during acute rejection or ATN
episodes and for the removal of ascitic fluid which has been found to accumulate
to a clinically significant degree in up to 25% of children transplanted on CAPD
[115].

Survival. It is difficult to evaluate the survival of children treated with CAPD so

early in the history of the technique. Most pediatric dialysis programs actively
pursue renal transplantation for their young patients which results in relatively
short periods on dialysis for each child. Short-term actuarial survival has been
shown to be as good with CAPD as with in-center hemodialysis in limited studies
[100].

Quality of life of the child and family on CAPD. There is little doubt that CAPD
offers children and their families a better quality of life than hemodialysis. Praise
for the beneficial effects of CAPD on the emotional health of patients and
families has been a consistent feature of published reports. Older children who
have experienced both treatment modalities unanimously prefer CAPD over
hemodialysis [100]. Greater freedom, absence of painful needle punctures, op-
portunities for regular school attendance and other peer group activities, in
addition to a more normal family life are obviously attractive features of CAPD.
As with other home therapies for serious illnesses, however, the advantages
derived from greater independence and self-reliance are not achieved without
cost. The stresses experienced by families of children on CAPD are substantial,
especially during the first 12 months of therapy. We have identified a syndrome of
'parent fatigue' which has been present to a variable degree in all of the families
we have seen to date; it has been particularly evident in the families of young
infants.

Parent fatigue. Mothers of our CAPD patients typically assume total responsi-
bility for home care which quickly relegates the best-intentioned fathers to minor
supporting roles. When added to the usual burdens of caring for a small child, the
CAPD procedures can seem overwhelming. Most parents find it difficult to
acknowledge their fatigue and frustration; they will rarely spontaneously com-
plain about the demands of the home dialysis regimen. In our experience, parent
fatigue can lead directly to breaks in technique followed by repeated episodes of
peritonitis [36]. The intense guilt experienced by parents when their child de-
velops peritonitis further increases family tension and fatigue. This cycle of
increased anxiety and perceived failure must be prevented or interrupted if
CAPD is to successfully continue.
 551

We now insist that mothers allow other adults to share the regular CAPD care
of their children. Fathers, grandparents, preschool teachers, visiting nurses,
babysitters and neighbors have all been trained by us to provide basic CAPD care
in support of the primary care provided by the child's mother. The importance of
extensive involvement of the entire CAPD medical team in this supportive
function cannot be overly emphasized. Regular telephone contact by CAPD
nurses and social workers has been helpful; more frequent clinic follow-up visits
are also important during the first year on CAPD.
After the first year on CAPD, we have observed a general reduction in the level
of stress experienced by these families. By the second year of treatment the
CAPD regimen has long since been adjusted to accommodate family activities.
Parents have developed confidence in themselves and close ties to the medical
team members. Vacations, extremely important events for most families, have
taken place on schedule.
Underlying this more relaxed approach may be a fundamentally reduced fear
of serious errors in the CAPD care of the child. It may take many months, but
most families eventually come to rely on the inherent safety of CAPD.

5.3. Continuous cycling peritoneal dialysis (CCPD)

CCPD was developed in an attempt to combine the benefits of continuous

peritoneal dialysis with the convenience of an automated delivery system [116].
There have been only a few reports describing the use of CCPD in children and
these only in preliminary form [41, 42]. In general, results appear to be as good
with CCPD as with CAPD. Two observations deserve additional comment:
contrary to earlier predictions [116], peritonitis was found to occur in children on
CCPD as often as in children on CAPD [41]. Preliminary studies also found that
biochemical control among pediatric CAPD and CCPD patients was equivalent
in all parameters except mean serum creatinine concentration, which was found
to be significantly higher in CCPD patients [41, 42]. These data are difficult to
interpret since details of dialysis mechanics are not available.
It has also been observed that small infants may not tolerate nighttime CCPD
cycling with its attendant ultrafiltration; dehydration can rapidly develop in some
infants unless drip feedings via nasogastric tubes are also continued throughout
the night [117].
Pediatric experience with CCPD is increasing and it is likely that this technique
will become more important in the future.
552

6. Peritoneal dialysis for intoxications, inborn errors of metabolism and other

miscellaneous disorders in children

6.1. Intoxications

Since the implementation of the Poison Prevention Packaging Act of 1970 there
has been a dramatic decline in the incidence of accidental ingestions of regulated
products by young children. Between 1974 and 1981 ingestions of aspirin, aspirin
substitutes, oven cleaners and other lye-containing products, lighter fluids, and
anti-freeze by children under 5 years of age decreased from 2.9 per 1000 children
under 5 years to less than 2.0 per 1000 [118]. Morbidity and mortality from
accidental ingestions of aspirin and acetaminophen in this age group declined
sharply during the same period. The death rate from accidental aspirin ingestion
by young children decreased by 69% between 1970 and 1978 [119].
As gratifying as these statistics may be, the fact remains that many uninten-
tional intoxications and deaths still occur each year among young children. In 1981
nearly 100000 children under 5 years of age were seen in emergency rooms in the
United States because of accidental ingestion of hazardous substances [118].
Treatment of intoxications in small children remains an important if infre-
quently tested area of expertise for the nephrologist who is likely to be consulted
regarding the advisability of dialysis in these situations. For many years, per-
itoneal dialysis has played an important role in the treatment of small children
who have been poisoned with substances removable by dialysis [120]. The use of
peritoneal dialysis in the treatment of poisoning has been reviewed in Chapter 11
and will not be presented again here. For detailed information regarding specific
intoxications the reader in urgent need of this information is advised to contact
the nearest Poison Control Center or to call the Rocky Mountain Poison Control
Center in Denver, Colorado (303-629-1123). Several excellent general reviews
will be of interest to those whose questions are less urgent [121, 122, 123].
In recent years the use of peritoneal dialysis to treat intoxications in children
has almost disappeared in our center. Several factors seem to be responsible for
this phenomenon. As was noted above, the incidence of serious salicylate intox-
ication in young children has been declining steadily throughout the United
States. In addition, a better understanding of the pathophysiology of salicylate
intoxication has led to more successful use of forced diuresis and other therapeu-
tic maneuvers in the child whose renal function remains intact [124]. Finberg has
noted that of 600 children hospitalized in Brooklyn following single large aspirin
ingestions, none required dialysis and all survived apparently undamaged [125].
Improvements in acute hemodialysis techniques and equipment which have
been specifically developed for use in very small children have also reduced the
frequency with which peritoneal dialysis must be used to treat intoxications [126];
successful hemoperfusion techniques for small children have recently been de-
scribed [127]. Reliable percutaneous vascular access procedures and catheters
 553

designed for use in small children 5 as well as single-needle hemodialysis machin-

ery have become widely available.
As a result of these and other developments, emergency hemodialysis is now
available for infants and small children in many pediatric centers. It should be
emphasized that regardless of the patient's age or size, hemodialysis is many times
more effective than peritoneal dialysis at removing dialyzable drugs and poisons
[122, 128]. When a child has ingested a potentially lethal amount of a dialyzable
poison, hemodialysis should be used whenever possible. Peritoneal dialysis is an
acceptable second choice only for those children too small to receive hemodialysis
at the facility in which they are being treated and too unstable to be safely
transported to a pediatric dialysis center where hemodialysis could be performed
expeditiously.

6.2. Congenital hyperammonemia and other inborn errors of metabolism

Congenital urea cycle enzymopathies are characterized by a reduced capacity to

synthesize urea, which leads to accumulation of ammonium and other nitro-
genous urea precursors [129]. Severely affected neonates develop vomiting,
lethargy, seizures, and coma within the first few days of life. The central nervous
system symptomatology in these disorders is thought to be solely due to the
effects of increased blood ammonium concentration [129]; thus, emergency treat-
ment must be aimed at rapid and sustained removal of accumulated ammonium.
Peritoneal dialysis has emerged as the treatment of choice for infants with
congenital hyperammonemic coma [130].
The superiority of peritoneal dialysis over exchange transfusion W0S recently
demonstrated. In studies performed in 53 episodes of hyperammonemic coma,
ammonium was removed more rapidly with peritoneal dialysis than with ex-
change transfusion, and the rebound hyperammonemia which often follows
treatment with exchange transfusion did not occur in babies treated with per-
itoneal dialysis [130]. Hemodialysis is the most efficient method for removal of
ammonium [131] but treatments with hemodialysis must be limited to several
hours, whereas endogenous ammonium production in these babies is persistent
early in the course of treatment [130]. Peritoneal dialysis can be continued
indefinitely, providing time in which the diagnosis of the specific urea cycle
enzymopathy can be made ano appropriate therapy instituted [129]. Peritoneal
dialysis removes ten times more nitrogen as glutamine than as ammonium; it has
been suggested that the effectiveness of peritoneal dialysis in hyperammonemic
infants may be due in part to the continuous removal of both ammonium and its
precursors (glutamine, glutamate, and alanine) [130].
Peritoneal dialysis has also been useful in the acute management of several
other congenital metabolic defects which do not always present with hyperam-
monemia. Successful treatment has been reported in cases of maple syrup urine
554

disease [132], proprionic acidemia [133], and other congenital organic acidemias
[134].

6.3. Miscellaneous pediatric disorders in which treatment with peritoneal dialysis

has been attempted

Many of the serious afflictions of infants and children have been treated at one
time or another with peritoneal dialysis. In 1966 Nora and associates demon-
strated the effectiveness with which peritoneal dialysis removed fluid from chil-
dren in intractable congestive heart failure [135]. Today such children would
probably be as successfully treated with one or more powerful diuretic agents
unknown to Dr Nora 20 years ago. The early and only marginally successful use of
peritoneal dialysis in the treatment of sodium chloride poisoning in infants has
received little attention in recent years [136]. Fortunately, salt poisoning has
become extremely unusual now that infant formula is manufactured rather than
mixed by hand in hospital nursuries from basic ingredients [137].
Peritoneal dialysis has not been shown to be of sufficient benefit in the
treatment of children with the following disorders to warrant continued use:
hyaline membrane disease [138]; neonatal hyperbilirubinemia [139]; Reye's syn-
drome [140]; hepatic coma [141].

Acknowledgements

The author wishes to thank Edward S. Tank, M.D., for his assistance with the
descriptions of surgical procedures and Ms Norma Fritz who prepared the manu-
script.

Notes

1 Of the parameters listed, an age-related difference in peritoneal capillary blood flow is the least

likely to be important. Peritoneal blood flow in adults is great enough under most circumstances to
exceed requirements for maximum observed peritoneal mass transfer [47]. Further increases in
peritoneal blood flow would not be expected to be able to accelerate peritoneal mass transfer in
children unless relative effective membrane area and permeability were dramatically different in the
young. There are no data available on the subject of peritoneal blood flow in the young.
2 Trocath, McGaw Laboratories, Los Angeles, California.
3 Quinton Instrument Company, Seattle, Washington.
4 Intracath, Deseret Medical, Inc., Sandy, Utah.

5 Argon Medical Corp., Athens, Texas.

 555

References

1. Fine RN: Peritoneal dialysis update. J Pediatr 100: 1-7, 1982.

2. Dobrin RS, Kjellstrand C: The management of acute renal failure in pediatrics. In: A Chapman
(ed), Acute Renal Failure. Churchill Livingstone, New York, 1980, pp 99-101.
3. Counts S, Hickman R, Garbaccio A, TenckhoffH: Chronic home peritoneal dialysis in children.
Trans Am Soc Artif Intern Organs 19: 157-167, 1973.
4. Donckerwolcke RA, Chantler C, Broyer M et al.: Combined report on regular dialysis and
transplantation of children in Europe, 1979. Proc EDT A 17: 89-115, 1980.
5. Report on distribution of patients by outpatient dialysis modality (March 14, 1981) ESRD
Program Medical Information System, HCFA, Washington, DC.
6. Popovich RP, Moncrief JW, Decherd JW et al.: The definition of a novel portable/wearable
equilibrium peritoneal dialysis technique. Trans Am Soc Artif Intern Organs 5: 64, 1976.
7. Alexander SR: Pediatric CAPD update -1983. Perit Dial Bull Suppl3: SI5-S22, 1983.
8. Swan H, Gordon HH: Peritoneal lavage in the treatment of anuria in children. Pediatrics 4: 586-
595, 1949.
9. Odel HM, Ferris DO, Power MH: Peritoneal lavage as an effective means of extra-renal
excretion. Am J Med 9: 63-77, 1950.
10. Fine J, Frank HA, Seligman AM: The treatment of acute renal failure by peritoneal irrigation.
Ann Surg 124: 857, 1946.
11. Segar WE: Peritoneal dialysis in the treatment of boric acid poisoning. N Engl J Med 262: 798-
800,1960.
12. Etteldorf IN, Dobbins WT, Summitt RL et al.: Intermittent peritoneal dialysis using 5 per cent
albumin in the treatment of salicylate intoxication in children. J Pediatr 58: 226-236, 1961.
13. Segar WE, Gibson RK, Rhamy R: Peritoneal dialysis in infants and small children. Pediatrics 27:
603-613, 1961.
14. Etteldorf IN, Dobbins WT, Sweeney MJ et al.: Intermittent peritoneal dialysis in the manage-
ment of acute renal failure in children. J Pediatr 60: 327-339, 1962.
15. Lloyd-Still JD, Atwell JD: Renal failure in infancy, with special reference to the use of
peritoneal dialysis. J Pediatr Surg 1: 466-475, 1966.
16. Manley GL, Collipp PJ: Renal failure in the newborn: treatment with peritoneal dialysis. Am J
Dis Child 115: 107-110, 1968.
17. Lugo G, Ceballos R, Brown Wet al.: Acute renal failure in the neonate managed by peritoneal
dialysis. Am J Dis Child 118: 655-659, 1969.
18. Gianantonio CA, Vitacco M, Mendelbarzee J et al.: Acute renal failure in infancy and child-
hood. J Pediatr 61: 660-678, 1962.
19. Wiggelinkhuizen J: Peritoneal dialysis in children. S Afr Med J 45: 1047-1054, 1971.
20. Day RE, White RHR: Peritoneal dialysis in children: review of 8 years' experience. Arch Dis
Child 52: 56-61, 1977.
21. Chan JCM: Peritoneal dialysis for renal failure in childhood. Clin Pediatr 17: 349-354, 1978.
22. Feldman W, Baliah T, Drummond KN: Intermittent peritoneal dialysis in the management of
chronic renal failure in children. Am J Dis Child 116: 30-36, 1968.
23. Tenckhoff H, Schecter H: A bacteriologically safe peritoneal access device. Trans Am Soc Artif
Intern Organs 14: 181-186,1968.
24. Hickman RO: Nine years' experience with chronic peritoneal dialysis in childhood. Dial Trans-
plant 7: 803,1978.
25. Brouhard BH, Berger M, Cunningham RJ et al.: Home peritoneal dialysis in children. Trans Am
Soc Artif Intern Organs 25: 90-94, 1979.
26. Baluarte HJ, Grossman MB, Polinsky MS et al. : Experience with intermittent home peritoneal
dialysis (IHPD) in children (Abstract). Pediatr Res 14: 994, 1980.
27. Lorentz WB, Hamilton RW, Disher B et al.: Home peritoneal dialysis during infancy. Clin
Nephrol 15: 194, 1981.
556

28. Potter DE, McDaid TK, Ramirez JA: Peritoneal dialysis in children. In: RC Atkins, NM
Thomson, PC Farrell (ed), Peritoneal Dialysis. Churchill Livingstone, New York, 1981, pp 356-
361.
29. Oreopoulas DG, Katirtzoglou A, Arbus G, Cordy P: Dialysis and transplantation in young
children (Letter). Br Med J 1: 1628-1629,1979.
30. Balfe JW, Irwin MA: Continuous ambulatory peritoneal dialysis in pediatrics. In: M Legrain
(ed), Continuous Ambulatory Peritoneal Dialysis; Proc Int Symp, Paris, November 2 and 3,
1979. Excerpta Medica, Amsterdam, 1980, pp 131-136.
31. Alexander SR, Tseng CH, Maksym KA et al.: Early clinical experience with continuous
ambulatory peritoneal dialysis (CAPD) in infants and young children (Abstract). Clin Res 28:
l3lA,1980.
32. Kohaut EC: Continuous ambulatory peritoneal dialysis: a preliminary pediatric experience. Am
J Dis Child 135: 270-271, 1981.
33. Potter DE, McDaid TK, McHenry K et al.: Continuous ambulatory peritoneal dialysis (CAPD)
in children. Trans Am Soc Artif Intern Organs 27: 64-67, 1981.
34. Guillot M, Clermont M-J, Gagnadoux M-F, Broyer M: Nineteen months' experience with
continuous ambulatory peritoneal dialysis in children: main clinical and biological results. In:
GM Gahl, M Kessel, KD Nolph (eds), Advances in Peritoneal Dialysis. Excerpta Medica,
Amsterdam, 1981, pp 203-207.
35. Salusky lB, Lucullo L, Nelson P, Fine RN: Continuous ambulatory peritoneal dialysis in
children. Pediatr Clin North Am 29: 1005-1012, 1982.
36. Alexander SR, Tseng CH, Maksym KA et al.: Clinical parameters in continuous ambulatory
peritoneal dialysis for infants and young children. In: JW Moncrief, RP Popovich (eds), CAPD
Update. Masson Publ, New York, USA, 1981, pp 195-209.
37. Balfe JW, Vigneux A, Williamson J et al.: The use of CAPD in the treatment of children with
end-stage renal disease. Perit Dial Buill: 35-38, 1981.
38. Shmerling J, Kohaut EC, Perry S: Cost and social benefits of CAPD in a pediatric population.
In: JW Moncrief, RP Popovich (eds), CAPD Update. Masson Publ, New York, USA, 1981, pp
189-193, 1981.
39. Travenol Laboratories, Inc, Deerfield, IL 60015, USA.
40. Price CG, Suki WN: Newer modifications of peritoneal dialysis: options in the treatment of
patients with renal failure. Am J Nephrol1: 97-104, 1981.
41. Alliapoulos JC, Salusky lB, Fine RN: Preliminary results with the use of CCPD in children. A
comparison to CAPD (Abstract). Program and Abstracts, National Kidney Foundation 13th
Annual Scientific Meeting, Nov. 30-Dec. 4, 1983, Washington DC, p 1,1983.
42. Hogg RJ, Lum GM: A multicenter collaborative study of continuous ambulatory (CAPD) and
continuous cycling peritoneal dialysis (CCPD) in children (Abstract). Program and Abstracts,
National Kidney Foundation 13th Annual Scientific Meeting, Nov. 30-Dec. 4, 1983, Washington,
DC, p 11, 1983.
43. Conley SB, Brewer ED, Gandy S et al.: Normal growth in very small children on peritoneal
dialysis: 18 months' experience (Abstract). Program and Abstracts, National Kidney Foundation
12th Annual Clinical Dialysis and Transplant Forum, Dec. 8-13, 1982, Chicago, Illinois, p 8.
44. Putiloff PV: Materials for the study of the laws of growth of the human body in relation to the
surface areas of different systems; the trial on Russian subjects of planigraphic anatomy as a
means for exact anthropometry - one of the problems of anthropology. Report of Dr PV Putiloff
at the meeting of the Siberian Branch of the Russian Geographic Society, October 29, 1884,
Omsk (summarized in Ref. 46), 1886.
45. Wegner G: Chirurgische Bemerkurgen uber die peritoneal Hohle, mit besonderer
BeIilcksichtigung der Ovariotomie. Arch Klin Chir 20: 51, 1877.
46. Esperanca MJ, Collins DL: Peritoneal dialysis efficiency in relation to body weight. J Pediatr
Surg 1: 162-169, 1966.
 557

47. Nolph KD: Peritoneal anatomy and transport physiology. In: W Drukker, FM Parsons, JF
Maher (eds), Replacement of Renal Function by Dialysis, 2nd ed. Martinus Nijhoff, Boston,
1983, pp 440-456.
48. Driscoll JM, Heird We: Maintenance fluid therapy during the neonatal period. In: RW Winters
(ed), The Body Fluids in Pediatrics. Little, Brown, Boston, 1973, p 266.
49. Holliday MA: Metabolic rate and organ size during growth from infancy to maturity and during
late gestation and early infancy. Pediatrics 47: 169-179,1971.
50. McCance RA: The maintenance of chemical stability in the newborn in chemical exchange. Arch
Dis Child 34: 361, 1959.
51. Elzouki A Y, Gruskin AB, Baluarte HJ et al.: Developmental aspects of peritoneal dialysis
kinetics in dogs. Pediatr Res 15: 853-858, 1981.
52. Henderson LW: The problem of peritoneal membrane area and permeability. Kidney Int 3: 409-
410, 1973.
53. Gruskin AB, Cote ML, Buluarte HJ: Peritoneal diffusion curves, peritoneal clearances and
scaling factors in children of differing age. International J Pediatr Nephrol 3: 271-278, 1982.
54. Pyle WK, Moncrief JW, Popovich RP: Peritoneal transport evaluation in CAPD. In: JW
Moncrief, RP Popovich (eds), CAPD Update. Masson Publ, New York, USA, 1981, pp 35-52.
55. Popovich RP, Pyle WK, Rosenthal DA et al.: Kinetics of peritoneal dialysis in children. In: JW
Moncrief, RP Popovich (eds), CAPD Update. Masson Publ, New York, USA, 1981, pp 227-
242.
56. Morgenstern B, Pyle WK, Gruskin AB et al.: Transport characteristics of the pediatric per-
itoneal membrane (Abstract). Kidney Int 25: 259, 1984.
57. Gruskin AB, Morgenstern BZ, Perlman SA: Kinetics of peritoneal dialysis in children. In: RN
Fine, AB Gruskin (eds), End-stage renal disease in children. WB Saunders, Philadelphia, 1984,
pp 95-114.
58. Swann RC, Merrill JP: The clinical course of acute renal failure. Medicine 32: 215, 1953.
59. Anderson RJ, Schrier RW: Clinical spectrum of oliguric and nonoliguric acute renal failure. In:
BM Brenner, JH Stein (eds), Acute Renal Failure. Churchill Livingstone, New York, 1980, pp
1-16.
60. Medani CR, Davitt MK, Huntington DF et at: Acute renal failure in the newborn. Contrib
Nephrol15: 47-54, 1979.
61. Chesney RW, Kaplan BS, Freedom RM et at.: Acute renal failure: an important complication of
cardiac surgery in infants. J Pediatr 87: 381-388, 1975.
62. Anderson RJ, Linas SL, Berns AS et at.: Nonoliguric acute renal failure. N Engl J Med 296:
1134-1138, 1977.
63. Schrier RJ: Acute renal failure. Kidney Int 15: 205-216, 1979.
64. Gordillo-Panigua G, Velasquez-Jones L: Acute renal failure. Pediatr Clin North Am 23: 817-
828, 1976.
65. Norman ME, Asadi FK: A prospective study of acute renal failure in the newborn infant. Pediatr
63: 475-479, 1979.
66. Rigden SPA, Barratt TM, Dillon MJ et at.: Acute renal failure complicating cardiopulmonary
bypass surgery. Arch Dis Child 57: 425-430, 1982.
67. Lieberman E: Acute renal failure. In: E Lieberman (ed), Clinical Pediatric Nephrology. JB
Lippincott, Philadelphia, 1976, pp 272-293.
68. Broyer M: Acute renal failure. In: P Royer, R Habib, H Mathieu, M Broyer (eds), Pediatric
Nephrology. WB Saunders, Philadelphia, 1973, pp 343-349.
69. Ellis D, Gartner JC, Galvis AG: Acute renal failure in infants and children: diagnosis, complica-
tions, and treatment. Crit Care Med 9: 607-617, 1981.
70. Miller TR, Anderson RJ, Linas SL et at.: Urinary diagnostic indices in acute renal failure: a
prospective study. Ann Intern Med 89: 47-50, 1978.
71. Ellis EN, Arnold We: Use of urinary indexes in renal failure in the newborn. Am J Dis Child
136: 615-617, 1982.
558

72. Mathew OP, Jones AS, Janes E et al.: Neonatal renal failure: Usefulness of diagnostic indices.
Pediatrics 65: 57-60, 1980.
73. Grylack L, Medani C, Hultzen C et al.: Nonoliguric acute renal failure in the newborn: a
prospective evaluation of diagnostic indexes. Am J Dis Child 136: 518-520, 1982.
74. Oh W: Renal functions and clinical disorders in the neonate. Clin Perinatol8: 215-223, 1981.
75. Chan JCM: Acute renal failure in children: principles of management. Clin Pediatr 13: 686-695,
1974.
76. Groshong T: Dialysis in infants and children. In: JC Van Stone (ed), Dialysis in the Treatment of
Renal Insufficiency. Grune & Stratton, New York, 1983, pp 234-236.
77. Hodson EM, Kjellstrand CM, Mauer SM: Acute renal failure in infants and children: outcome
of 53 patients requiring hemodialysis treatment. J Pediatr 93: 756-761, 1978.
78. Cannahan J, Cameron JS, Ogg CS et al.: Presentation, management, complications, and
outcome of acute renal failure in childhood: five years experience. Br Med J 1: 599-602, 1977.
79. Book K, Ohgvist G, Bjork VO etal.: Peritoneal dialysis in infants and children after open heart
surgery. Scand J Thorac Cardiovasc Surg 16: 229-233, 1982.
80. Fong JSC, deChadaravian J-P, Kaplan BS: Hemolytic-uremic syndrome: current concepts and
management. Pediatr Clin North Am 29: 835-856, 1982.
81. Leumann EP, Knecht B, Dangel P et al.: Peritoneal dialysis in newborns: technical improve-
ments: In: M Bulla (ed), Renal Insufficiency in Children. Springer-Verlag, New York, 1982, pp
147-150.
82. Borzotta A, Harrison HL, Groff DB: Technique of peritoneal dialysis cannulation in neonates.
Surg Gynecol Obstet 157: 73-74, 1983.
83. Anand SK, Northway JD, Gresham EL: Peritoneal dialysis catheter for small infants (Letter). J
Pediatr 86: 985, 1975.
84. Shapiro RS: A home-made catheter for peritoneal dialysis in the neonate (Letter). J Pediatr 87:
160,1976.
85. Kanarek KS, Root E, Sidebottom RA, Williams PR: Successful peritoneal dialysis in an infant
weighing less than 800 grams. Clin Pediatr 21: 166-169, 1982.
86. Searle M, Lee HA: Peritoneal dialysis in infants (Letter). Br Med J 286: 1353, 1983.
87. Alexander SR, Tank ES: Surgical aspects of continuous ambulatory peritoneal dialysis in
infants, children and adolescents. J Uro1127: 501, 1982.
88. Nash MA, Russo JC: Neonatal lactic acidosis and renal failure: the role of peritoneal dialysis. J
Pediatr 91: 101-105, 1977.
89. Lorentz WB: Acute hydrothorax during peritoneal dialysis. J Pediatr 94: 417-419, 1979.
90. Kohaut EC: Effect of dialysate volume on ultrafiltration in small children (Abstract). Int J Ped
Neph (in press).
91. Posen GA, Luisello J: Continuous equilibration peritoneal dialysis in the treatment of acute
renal failure. Perit Dial Buill: 6-7, 1980.
92. Abbad FCB, Ploos van Amstel SLB: Continuous ambulatory peritoneal dialysis in small
children with acute renal failure. Proc EDTA 19: 607-613,1982.
93. Popovich RP, Pyle WK, Moncrief JW: Kinetics of peritoneal transport. In: Kd Nolph (ed),
Peritoneal Dialysis. Martinus Nijhoff, Boston, 1981, pp 116-117.
94. Alexander SR: The treatment of infants with ESRD. In: RN Fine, AB Gruskin (eds), End-stage
renal disease in children. WB Saunders, Philadelphia, 1984, pp 17-27.
95. Alexander SR: Continuous ambulatory peritoneal dialysis in infants less than one year of age.
In: RN Fine, AB Gruskin (eds), End-stage renal disease in children. WB Saunders, Phila-
delphia, 1984, pp 149-171.
96. Alexander SR, Lubischer JT: Continuous ambulatory peritoneal dialysis in pediatrics: 3 years'
experience at one center. Nefrologia (Madrid) 11: Suppl 2, 53-62, 1982.
97. Balfe JW, Irwin MA, Oreopoulos DG: An assessment of continuous ambulatory peritoneal
dialysis (CAPD) in children. In: JW Moncrief, RP Popovich (eds), CAPD Update. Masson Publ
USA, New York, 1981, pp 211-220.
 559

98. Salusky IB, Kopple JD, Fine RN: Continuous ambulatory peritoneal dialysis in pediatric
patients: a 20-month experience. Kidney Int 24(Suppl)15: S-101-S-105, 1983.
99. Alexander SR, Tank ES, Hatch D: Continuous ambulatory peritoneal dialysis. Dialogues in
Pediatr Urol 6: 1-8, 1983.
100. Baum M, Powell D, Calvin S et al.: Continuous ambulatory peritoneal dialysis in children:
comparison with hemodialysis. N Engl J Med 307: 1537-1542,1982.
101. Vigneux A, Hardy BE, Balfe JW: Chronic peritoneal catheter in children - one or two dacron
cuffs? (Letter) Perit Dial Buill: 151, 1981.
102. Alexander SR, Tank ES: Technical considerations in the implantation of Tenckhoff catheters
for continuous ambulatory peritoneal dialysis in children. Nefrologia (Madrid) 11(Supp1): 2:
49-52, 1982.
103. Hogg RJ, Coin D, Chang J et al.: The Toronto Western Hospital Catheter in a pediatric dialysis
program. Am J Kidney Dis 3: 219-223, 1983.
104. Blumenkrantz MJ, Koppel JD, Moran JK et al.: Nitrogen and urea metabolism during continu-
ous ambulatory peritoneal dialysis. Kidney Int 20: 78--82, 1981.
105. Kohaut EC, Alexander SR: Ultrafiltration in the young patient on CAPD. In: JW Moncrief, RP
Popovich (eds), CAPD Update. Masson Publ USA, New York, 1981, pp 221-226.
106. Warren S, Conley SB: Nutritional considerations in infants on continous peritoneal dialysis
(CPD). Dial Transplant 12: 263-266, 1983.
107. Kohaut EC: Growth in children with end-stage renal disease treated with CAPD for at least one
year. Perit Dial Bull 2: 159-160, 1982.
108. Fennell RS, Orak JK, Hudson T et al. : Growth in children with various therapies for end-stage
renal disease. Am J Dis Child 138: 28--31, 1984.
109. Stefanidis CJ, Hewitt IK, Balfe JW: Growth in children receiving continuous ambulatory
peritoneal dialysis. J Pediatr 102: 681-685,1983.
110. Salusky IB, Fine RN, Nelson P et al.: Nutritional status of pediatric patients undergoing CAPD
(Abstract). Kidney Int 21: 177, 1982.
111. Betts PR, Magrath G: Growth Pattern and dietary intake of children with chronic renal
insufficiency. Br Med J 2: 189, 1974.
112. Patient population demographics and selected outcome measures for the period January 1, 1981
through December 31, 1982. The National Institutes of Health CAPD Patient Registry Report
Number 83-1, April 15, 1983.
113. Fine RN, Salusky IB, Hall T et al.: Peritonitis in children undergoing continuous ambulatory
peritoneal dialysis. Pediatrics 71: 806-809, 1983.
114. Hewitt IK, Stefanidis C, Reilly BJ et al.: Renal osteodystrophy in children undergoing continu-
ous ambulatory peritoneal dialysis. J Pediatr 103: 729-734, 1983.
115. Stephanidis CJ, Balfe JW, Arbus GS et al.: Renal transplantation in children treated with
CAPD. Perit Dial Bull 3: 5-7, 1983.
116. Diaz-Buxo JA, Walker PJ, Farmer CD et al.: Continuous cyclic peritoneal dialysis. Trans Am
Soc Artif Intern Organs 27: 51-54, 1981.
117. Conley SB: Personal communication.
118. Unintentional poisoning among young children - United States (March 11, 1983). Morbidity and
Mortality Weekly Report 32: 117-118.
119. Clarke A, Walton WW: Effect of safety packaging on aspirin ingestion by children. Pediatrics
63: 687-693, 1979.
120. Chan JCM, Campbell RA: Peritoneal dialysis in children: a survey of its indications and
applications. CIin Pediatr 12: 131-139, 1973.
121. Burgess ED, Blair AD, Cutler RE: Dialysis and hemoperfusion of drugs and poisons. In: HC
Gonick (ed), Current Nephrology, Vol 5. John Wiley & Sons, New York, 1982, pp 309-332.
122. Van Stone JC: Hemodialysis. In: HC Gonick (ed), Current Nephrology, Vol 7. John Wiley &
Sons, New York, 1984, pp 87-105.
560

123. Winchester JF, Gelford MC, Knepshield JH et af.: Dialysis and hemoperfusion of poisons and
drugs. Trans Am Soc Artif Intern Organs 23: 762-842, 1977.
124. Temple AR: Acute and chronic effects of aspirin toxicity and their treatment. Arch Intern Med
141: 364-369, 1981.
125. Finberg L: Electrolyte problems and poisonings: salicylates and phosphates. In: L Finberg, RE
Kravath, AR Fleishman (eds), Water and Electrolytes in Pediatrics. WB Saunders, Phila-
delphia, 1982, pp 201-205.
126. Mauer SM: Pediatric renal dialysis. In: CM Edelmann Jr (ed), Pediatric Kidney Disease. Little
Brown & Co, Boston, 1978, pp 493-502, 1982.
127. Papadopoulou ZL, Novello AC: The use of hemoperfusion in children, past, present, and
future. Pediatr Clin North Am 29: 1039-1052, 1982.
128. Rubin J: Comments on dialysis solution composition, antibiotic transport, poisoning and novel
uses of peritoneal dialysis. In: KD Nolph (ed), Peritoneal Dialysis. Martinus Nijhoff, The
Hague, p 253 (Table).
129. Shih VE: Congenital hyperammonemic syndromes. Clin Perinatol3: 3-4, 1976.
130. Batshaw ML, Brusilow SW: Treatment of hyperammonemic coma caused by inborn errors of
urea synthesis. J Pediatr 97: 893-900, 1980.
131. Donn SM, Swartz RD, Thoene JG: Comparison of exchange transfusion, peritoneal dialysis and
hemodialysis for the treatment of hyperammonemia in an anuric newborn infant. J Pediatr 95:
67-70, 1979.
132. Sallan SE, Cottom D: Peritoneal dialysis in maple syrup urine disease. Lancet 2: 1423, 1969.
133. Russell G, Thorn H, Tarlow MJ et af.: Reduction of plasma proprionate by peritoneal dialysis.
Pediatrics 53: 281-283, 1974.
134. Mahoney MJ: Organic acidemias. Clin Perinatol 3: 61-78,1976.
135. Nora 11, Trygstad CW, Mangos JA: Peritoneal dialysis in the treatment of intractable congestive
heart failure of infancy and childhood. J Pediatr 68: 693,1966.
136. Miller NL, Finberg L: Peritoneal dialysis for salt poisoning. N Engl J Med 263: 1347, 1960
137. Finberg L, Kiley J, Luttrell CN: Mass accidental salt poisoning in infancy. A study of a hospital
disaster. JAMA 184: 187, 1963.
138. Boda D, Muranyi L. Altorjay I, and Veress I: Peritoneal dialysis in the treatment of hyaline
membrane disease of newborn premature infants. Acta Paediatr Scand 60: 90-92, 1971.
139. Hobolth N, Devantier M: Removal of indirect reacting bilirubin by allumin binding during
intermittent peritoneal dialysis in the newborn. Acta Paediatr Scand 58: 171, 1969.
140. Pross DC, Bradford WD, Krueger RP: Reye's syndrome treated by peritoneal dialysis. Pedi-
atrics 45: 845, 1970.
141. Krebs R, Flynn M: Treatment of hepatic coma with exchange transfusion and peritoneal dialysis.
JAMA 199: 430, 1967.
16. Peritoneal dialysis in diabetics

MARCEL LEG RAIN and JACQUES ROTIEMBOURG

1. Introduction

Diabetes is the only growing cause of end-stage renal disease (ESRD) in all of the
industrialized countries [1, 2]. Everyone agrees that exclusion of diabetic patients
from renal replacement therapy is no longer acceptable when the socio-economic
environment is adequate and treatment facilities are available [3, 4). Diabetics
should be offered all dialysis methods and transplantation within an integrated
program. Despite the very encouraging results observed with transplantation[5),
because of age and lack of either related or cadaver kidney donor, the majority of
diabetic patients with ESRD are treated either exclusively or temporarily by
dialysis methods [6, 7, 8, 9).
Recent developments in peritoneal dialysis treatment, mainly in relation with
the continuous ambulatory peritoneal dialysis procedure (CAPD) have opened
new therapeutic alternatives. From the experience gained in different countries
since 1978 peritoneal dialysis is an appealing dialysis procedure in diabetic pa-
tients both type I and type II [8, 9,10,11,12]. Time has come for a critical appraisal
of various forms of peritoneal dialysis versus hemodialysis and transplantation.

2. Continuous ambulatory peritoneal dialysis

Many papers have shown that an excellent control of both uremia and hyperten-
sion is possible using CAPD and reports by Flynn [13, 14), and different groups [8,
15,16,17,18] have proved that the intraperitoneal administration of insulin can
restore the plasma glucose level to near-normal values without increased risk of
infection. For such reasons since 1978 various institutions have selected CAPD as
the preferred mode of therapy for diabetics with ESRD. Table 1 gives the choice
of the first dialysis method offered since 1973 to 124 insulin-dependent diabetics
(IDD) with terminal renal failure, in the Department of Nephrology of the
Hopital de la Pitie. Results in such patients at two and three years are now
becoming available [8, 19] and allow one to make early conclusions.
562

Table 1. H6pital de la Pitie, Paris, Dialysis methods. First choice. 124 insulin-dependent diabetics,
1973-1983.

HD IPD CAPD CCPD Total

1973-1974 9 9
1975-1976 13 1 14
1977-1978 16 2 19
1979-1980 19 13 32
1981-1982 9 18 4 31
1983 4 13 2 19

70 3 45 6 124

2.1. The method. Blood glucose control

CAPD procedure in diabetic patients is similar to the one used in non-diabetic

cases. CAPD is conducted in our program through a double-cuff Tenckhoff
catheter with a preferential use of the curled type [20] or with a Toronto Western
Hospital catheter [21]. The average training period is around 20 days. Most
patients perform four exchanges per day routinely using three 2-liter bags with a
1.5% dextrose concentration during the daytime and one 2-liter bag with a 4.2 to
4.5% dextrose concentration overnight. A 2.5% dextrose concentration can be
occasionally used [99].
Insulin requirements of diabetic patients are influenced by uremia. Modifica-
tions of daily insulin doses when dialysis is started are unpredictable [22] although
frequently increased. Type II uremic diabetic patients may require insulin when
dialysed.
Patients on CAPD are taught to add the insulin to the bags [8, 13] or to inject
the insulin into the connecting tube before instillation of the following bag as
previously described [17, 18]. On the average 15 to 20 U of regular insulin are
needed with the 1.5% dextrose concentration and 30 to 40 U with the 4.5%
concentration. The doses of insulin administered through the catheter are either
increased or decreased by 2 to 4 U according to blood sugar levels measured the
previous day. The aims of insulin requirements are to obtain fasting blood glucose
level around 7mmolll (126 mg/dl) and to avoid 2 h post-prandial sugar levels of
over 11 mmol/l (198 mg/dl). Patients and their relatives are trained to check the
blood sugar level with the 'finger prick technique'. During the initial training
period, blood sugar is measured routinely 5 to 10 minutes before draining the bag
and occasionally [Link] after meals. At home two such readings are performed
daily, one during fasting and the second before supper. A series on six readings
during 24 hours are obtained once a month. Some patients may be reluctant to
perform numerous daily glucose measurements and one should be cautious that
 563

the 'finger prick' technique in some diabetics with severe vascular lesions can
induce necrotic lesions. Monthly determination of glycosylated hemoglobin lev-
els can be performed when patients are attending the out-patient clinic [23].
Considering a daily peritoneal absorption between 100 and 125 g of dextrose
and the daily loss of about 10 g of protein through the dialysate effluent, patients
are asked to eat a diet with a carbohydrate content ranging between 130 and 160
g/day and a protein content of about 1.5 g/kg of body weight/day. Table 2 gives an
estimation of insulin requirements and daily carbohydrate intakes from food and
dialysate in IDD patients treated by CAPD at the Hopital de la Pitie. Water and
salt intakes are adjusted according to the residual renal function, clinical status of
hydration and blood pressure values. Furosemide administration, 250 to 500
mg/day, given orally can possibly contribute to the preservation of residual renal
function [18].
Partial or total blindness constitutes a real handicap for using CAPD, the
patient being in charge of his own treatment. Nevertheless Flynn and other
groups have shown that totally blind patients can handle the technique very well
[13].

2.2. Clinical and biological status

The clinical status of a majority of diabetics, most of them insulin-dependent,

treated by CAPD can be judged as very good. After corrections of the predialysis
overhydration a steady weight gain can be observed during the first years of
treatment requiring careful dietary control and adequate insulin management to
avoid further increase in body weight. Hypertension is easily controlled despite
reduction or discontinuation of antihypertensive drugs over time. Unfortunately
postural hypotension can be a major problem in some patients sometimes favored
by an overuse of high dextrose concentration dialysate inducing hypovolemia.

Table 2. Estimation of insulin requirements and daily carbohydrate intakes from food and dialysate in
diabetics treated by CAPD. Hopital de la Pitie, Paris

Timing of bag exchanges and meals Total

7 a.m. 12 a.m. 4 p.m. 8p.m.

Concentration of the 1.5 1.5 1.5 4.5

2-1 bag (%)
Routine dextrose absorption 25 20 20 60 125
per 2-1 bag (g)
Routine carbohydrate intake 30 70 20 40 160
per meal (g)
Regular insulin injected (U) 18 26 14 30 88
564

The evolution of some major clinical and biological parameters is satisfactory

as shown in various recent reports [8, 9] and in Table 3 dealing exclusively with
insulin-dependent diabetics (IDD). Persistence of a relatively high hematocrit
value contributes to the feeling of well-being of the patients. The persistence of
residual renal function although slowly decreasing with time is commonly, but not
always [19], observed at least until the third year (Table 3). Such a difference from
what is routinely observed with either hemodialysis or intermittent peritoneal
dialysis [24, 25] may be related to the stable high plasma osmolality, the absence
of acute fluid shifts with rapid reduction in extra-cellular volume and perhaps the
routine administration of furosemide.

2.3. Technical complications

Acute peritonitis remains the main complication of the method. This complication
is in most cases benign and easily cured by local administration of antibiotics
without lavage and hospitalization (see Chapter 13). The incidence of peritonitis
is 1 episode every 10 to 20 patient-months [8, 17,19]. This rate is either similar to or
lower than that of a non-diabetic population treated in the same center. The
traditional good training on diabetic patients in aseptic precautions may explain

Table 3. Evolution of major clinical and blood parameters

Months 0-1 12 24 30

Patients (n) 46 31 14 8

Weight (kg) 63.6± 13,9 66.9 ± 13.4 61.1 ± 9.5 62.4 ± 7.5
Blood pressure
- Syst. mmHg 173 ± 42 149 ± 30 146 ± 32 139 ± 32
- Diast. 96±27 8D± 17 83 ± 16 81 ± 16
S. albumin (gil) 33 ±6 32± 5 30±5 30±3
Creatinine (/Lmol/l) 890 ±230 830 ± 243 861 ± 229 871 ± 134
P. bicarbonates (mmolll) 21 ± 4 25 ±3 25 ±3 25 ±2
Calcium (mmolll) 2.16 ± 0.50 2.31 ± 0.18 2.21 ± 0.15 2.26 ± 0.12
Phosphorus (mmol/l) 1.9 ± 0.8 1.6 ± 0.5 1.6±0.4 1.63± 0.4
Cholesterol (mmol/l) 5.8 ± 1.8 6.3 ± 1.7 5.9 ± 1.9 6.1 ± 1.8
Triglycerides (mmolll) 2.6 ± 1.5 3.0 ± 1.8 2.4 ± 1.9 2.0 ± 1.9
Hemoglobin (g%) 8.8 ± 2.2 11.4 ± 1.8 10.3 ± 1.6 10.1 ± 1.9
Diuresis (mllday) 1060 ± 350 680± 360 700 ± 350 650 ± 280
Residual renal creat.
clearance (mllmin) 4.3 ± 2.5 3.8 ± 2.1 3.4 ± 2.7 3.2 ±2.1
Peritoneal creat.
clearance (mllmin) 4.6 ± 1.3 4.4 ± 1.3 4.6 ± 1.5
Protein losses (g/day) 10.8 ± 1.5 9.8 ± 2.5 9.7 ± 2.6
 565

the satisfactory results obtained sometimes even with blind patients [13]. The
organisms responsible are predominantly staphylococci either epidermidis or
aureus. Peritonitis may lead to fatal complications in relation with a perforated
bowel, the organisms involved such as fungi, or frequent recurrence inducing
severe malnutrition. Infection of the tunnel or at the exit site of the dialysis line is
the most common indication for catheter replacement.
Other abdominal complications, not influenced by the diabetic status, can
occur. Decreased ultrafiltration can be observed with or without a past history of
recurrent peritonitis. Such a complication exposes the patient to chronic over-
hydration and requires transfer to HD or in some cases IPD [17,26,27]. Recovery
of normal UF can be observed after a few weeks on an alternative procedure [27].
Sclerosing peritonitis is a rare, but severe, complication of CAPD that can occur
in diabetics as well as in non-diabetic patients. Symptoms are nausea, vomiting,
abdominal pain and intermittent partial obstruction. A decreased UF rate is
commonly observed. The real cause of such a severe complication remains
unknown. The use of a dialysate with an acetate buffer instead of the lactate
buffer commonly used may be a contributing (although not exclusive) risk factor
for both decreasing ultrafiltration and sclerosing peritonitis [27, 28, 29].

2.4. Metabolic and nutritional problems

Blood glucose control. An excellent control of blood glucose levels, the best
obtained among diabetics with ESRD, is made possible by using as proposed by
Flynn [14] the intraperitoneal route four times a day for administration of insulin
associated with frequent blood sugar monitoring. The injection of insulin in an
empty peritoneal cavity 15 to 30 minutes before the meal and the introduction of
the dialysate offers the highest plasma free insulin levels and allows post-prandial
blood levels close to normal values [18, 30]. Such a procedure requires an
injection site on the line and forces the patient to eat while the bag is being hung.
Some patients are reluctant to follow rigidly such a technique. The demonstration
of true advantages of the injection of insulin into an empty peritoneal cavity will
require a careful long-term follow-up. Glycosylated hemoglobin (Hb Ale) values
decrease after starting treatment by CAPD and after a few months, tend to
diminish by 10% although still above normal values [18]. High HbAle values are
reported in non-diabetic patients treated by CAPD. Interpretation of HbAlc
values in uremic patients should be cautious because of the hemoglobin car-
bamylation process in relation with the increased blood urea and of the frequent
decreased red blood cell life span [31].
Lipid values. A marked increase in very low-density lipoprotein (VLDL)-
bound triglyceride and decreased levels of high-density lipoprotein (HDL) cho-
lesterol concentrations are routinely observed in patients with ESRD dialysed or
not [32]. Serum triglycerides are frequently increased in diabetic patients treated
566

by CAPD [33, 34, 35] mainly during the first 6 months. The elevations remain
moderate or even disappear if blood glucose levels are well controlled using the
addition of insulin to the dialysate and if the daily use of dialysate with 4.5%
dextrose concentrations is restricted to one 2-liter bag. The added potential risk
of mild permanent hypertriglyceridemia among diabetic patients with ESRD
remains debatable.
Malnutrition (see also Chapter 12). Adequate nutrition is difficult to maintain
in diabetic patients with gastroenteropathy leading to nausea, vomiting and
diarrhea. CAPD can ex age rate gastrointestinal disorders requiring transfer to
another method. Malnutrition can also be induced by recurrent peritonitis. High
protein losses in the dialysate outflow can rapidly lead to severe hypo-
albuminemia. Severe weight losses can be masked by a positive sodium balance
and overhydration. Such a situation requires the administration of nutriment in
large amounts. The Ol;al route is often inadequate and intravenous administra-
tion, for example through a subclavian catheter, can be required. Larger caloric
intake will require adjustment of insulin doses and careful control of water-
electrolyte balance, often difficult because of severe thirst. Hospitalization in an
intensive care unit may be required.
Replacement of glucose in the dialysate. To avoid some disadvantages of the use
of glucose as the osmotic agent in peritoneal dialysis, including caloric load and
hyperlipidemia, the search for alternative osmotic agents has justified active
recent research. Diabetic patients could be the best candidates for such replace-
ment. Replacement of glucose by xylitol, fructose, sorbitol, dextran and amino
acid mixtures has been tested both in animals and in men [36]. Bazzato et al. have
used xylitol in four diabetic patients treated by CAPD for six months [37].
Although xylitol can induce a lower need for insulin and a better control of
hyperlipidemia, the administration of high doses of xylitol can induce an increase
in plasma uric acid and lactic acid levels and deterioration of liver function.
Dialysis with glycerol by de Paeppe et al. was also disappointing [38]. A major
reduction in caloric load could not be achieved because of the lower ultrafiltration
capacity of the glycerol solution. The use of more concentrated dialysate resulted
in a complete disappearance of the caloric advantages. Dialysis with amino acid-
containing solutions can improve protein nutrition and provide effective ultra-
filtration. Such costly dialysates could be used for short periods, for example
during peritonitis episodes, or in patients with malnutrition. They are still under
investigation [39].

2.5. Micro- and macro-angiopathy

Blood pressure. After three months on CAPD most diabetic patients will remain
normotensive without drugs [8, 9]. If hypertension persists, close attention to
hydration status should be given, diabetics being very sensitive to overhydration
 567

and positive sodium balance. If antihypertensive drugs are required, overdosage

can lead to severe postural hypotension when hypovolemia is induced by using
high dextrose concentration dialysates.
Visual status. Improvement and stabilization, but also in a few cases deteriora-
tion of visual acuity, have been observed. Good control of blood glucose and
blood pressure and also adequate specialized care are required to improve visual
status when lesions are reversible. Vitreous surgery and pan retinal pho-
tocoagulation can be highly beneficial and sight preserving even in dialysed
patients [40]. Table 4 gives the evolution of visual acuity in 22 IDD patients
treated by CAPD for at least one year.
Peripheral arteritis. Severe peripheral vascular disease leading to gangrene and
requiring amputations is a still too frequent complication observed in diabetics
with ESRD. Such complications with a risk of sepsis and death (see Table 5) in
malnourished patients occur with all forms of renal replacement therapy includ-

Table 4. H6pital de la Pitie, Paris. Evaluation of visual acuity in 22 IDD patients treated by CAPD for
at least one year.

At start of CAPD Last examination

G I Good V. A. of both eyes 8 _____10

______ (2)
G II Good V. A. of one eye 7~ 7
(V. A. 20/200 to 40/40) - - - - __ (4)~
~- __ j2)
G III Legally blind
(V. A. 20/200)
6~
--- - --._
~3

(1)
G IV Totally blind ---..... 2

Table 5. H6pital de la Pitie, Paris 51 insulin-dependent diabetic patients treated by peritoneal dialysis,
45 by CAPD, 6 by CCPD. Causes of transfer and death

Transfer to hemodialysis a Deaths

Unable to handle the technique .......... 5 Myocardial infarct ..................... . 3

Recurrent peritonitis ................... 4 Cerebro-vasc. Accident ................ . 1
Sclerosing peritonitis ................... 2 Gangrene and sepsis ................... . 6
Malnutrition .......................... . Bowel perforation ..................... .
Bowel perforation ..................... . Malnutrition .......................... .
Loss of ultrafiltration .................. . Liver insufficiency ..................... .

Total ................................. 14 13

a 3 patients died during the two months following transfer.

568

ing successful transplantation [41]. CAPD has been considered a contributing

factor to the acceleration of peripheral vascular disease [42]. 11 patients out of 46
patients treated by continuous peritoneal dialysis have required amputations, a
percentage close to 25% observed among hemodialysed patients [9]. A reduction
of the amputation rate is certainly possible through adequate prevention includ-
ing a foot care program [3, 43]. Skin necrotic lesions were observed in four of our
patients.

2.6. Neurological status

Peripheral neuropathy is an almost constant finding in diabetic patients with

ESRD induced by both diabetes and uremia. Reduction of peritoneal nerve
conduction velocity is present in about 25% on the patients and walking impair-
ment, often mild, is present in 40 to 50% of cases. On CAPD progressive clinical
improvement and stabilization are possible, but deterioration is also observed.
When measurable and despite so-called 'adequate dialysis' , low nerve conduction
velocity values persists in most cases [9].
Autonomic nervous system dysfunction with clinical symptoms are encountered
in about 10% of cases; gastroenteropathy with nausea, vomiting and diarrhea can
contribute to severe malnutrition. Micturition disorders are in most cases
asymptomatic but careful evaluation is required if the patient is a candidate for
transplantation.

2.7. Hospitalization and rehabilitation

Duration of hospitalization and quality of rehabilitation will partly rely on

selection criteria and the mean age of the patients considered. The overall
hospitalization rate in recent published series [9,10] is between 30 and 40 days per
patient year including hospitalization for training, complications and social rea-
sons. This is about twice the rate of hospitalization of non-diabetic patients of the
same age. Peritonitis remains the main cause of hospitalization even if standard
treatment of most acute peritonitis episodes is performed at home. The ambula-
tory treatment of peritonitis requires great care by the patient and the relatives
and careful attention by the staff in charge. In our experience [9] the treatment of
each peritonitis episode requires a mean of 4.2 ± 1.4 extra consultations and
3.5 ± 1.2 days of hospitalization in a specialized out-patient unit.
Rehabilitation depends on many factors, not only medical. Only a minority of
patients can resume their original occupation. In a series of 37 unselected diabetic
patients trained for CAPD with a mean age of 49.8 years (range 26 to 70yr)
Khanna et al. observed that 21 were able to carry out normal activity and seven of
these were gainfully employed [19]. For retired persons CAPD offers a unique
means of home dialysis treatment avoiding tiring and costly travel to centers.
 569

2.8. Causes of death and transfer

The main causes of death, increasing with age, are of vascular origin including
cerebrovascular accidents, myocardial infarctions, and also arteritis complicated
by gangrene and sepsis often favored by malnutrition (see Table 5). Peritonitis
can be life threatening according to the pathogens involved (fungi for example),
or related to anatomical lesions (such as bowel perforation), or secondary to
severe malnutrition induced by recurrent peritonitis episodes [17]. Many series
report some cases of cessation of treatment decided in agreement with the rela-
tives when facing a situation where multiple complications including dementia
lead to permanent and definitive hospitalization.
Transfer, mainly to hemodialysis, remains a frequent course among diabetic
patients treated by CAPD (see Table 5). The main reasons are either peritoneal
complications such as recurrent and sclerosing peritonitis or severe malnutrition.
Both conditions can be associated [9]. In the juvenile population a main cause of
transfer should be transplantation [8]. Unfortunately in most countries the lack of
a donor, either living related, or cadaver is still a reality [6].
Actuarial survival rates on treatment by CAPD of diabetic patients during a
period of 2 or 3 years have been recently published [8, 17, 19]. The number of
patients treated remains small and long-term follow-up is not yet available. In
many series both actuarial survival rates and technique success rates in diabetics
are lower than in non-diabetic populations of the same age [3, 4]. The actuarial
survival rates of all our insulin-dependent diabetics treated between July 1978 and
December 1983 by continuous peritoneal dialysis (CPD), with a mean age of
52 ± 13.5 years was 85, 65 and 55% at 1, 2 and 3 years respectively while the
maintenance rate on CPD was for the same periods 80,55 and 40% (see Fig. 1).
100
, 38 46 Diabetics
........ ,
-....
90
' mean age 52.3±13.5y
....... 33 • __ • Survival rate
.... .... . . . . Maintenance on CPO
80 ....... 24

70 " "...
...
... " 16
~" , 10 7
...... -----e

40

30

i i I i
6 12 24 36
Time. months
Figure 1. Actuarial survival rate and maintenance rate on continuous peritoneal dialysis (CPD) in 46
insulin-dependent diabetic patients (40 on CAPD, 6 on CCPD).
570

Such results are close to the most recent data published by Khanna et al. [19]
dealing with 29 type I diabetics with a mean age of 47 yr. As in a diabetic
population treated by hemodialysis [9] age remains the main risk factor, at two
years the patient survival rate is respectively 84% in patients under 50 years and
only 49% in patients over 50 years (see Fig. 2).
Despite the encouraging results observed with CAPD in diabetics of either
type I or type II, such a method is not suitable to all patients. To offer to each
patient the best, other forms, either continuous or intermittent, of peritoneal
dialysis should be available.

3. Other forms of peritoneal dialysis

3.1. Intermittent peritoneal dialysis

Until 1977 intermittent peritoneal dialysis (IPD) was the only method of
peritoneal dialysis available to treat insulin-dependent as well as non-insulin-
dependent diabetics with ESRD, either at home or at the hospital. Very few
patients were on treatment but early results were encouraging [42, 45]. Since
1978, the rapid development of CAPD as the preferential choice for many units to
treat diabetics has restricted the indication of IPD. Nevertheless results recently
published [24] have clearly emphasized that IPD can be an adequate mode of
therapy in high risk patients including diabetics.

3.1.1. Dialysis method and blood glucose control

The peritoneal access and the dialysis technique are similar to those used in a non-

16
100
13 11
90

80

70
ii
>
.~ 60
:I
II)

* 50
46 Diabetics
40 ...... 19 patients under 50
mean age 39.1±8y 4
30 0--<> 27 patients IMIr 50
mean age 61.5±8y
f i i i i i
6 12 18 24 30
TIme. months

Figure 2. Actuarial survival rate in 46 insulin-dependent diabetic patients treated by continuous

pentoneal dialysis (CPO) in relation with age (40 on CAPO, 6 on CCPO).
 571

diabetic population (see Chapter 7). Two types of equipment to allow closed-
circuit delivery of the dialysis fluid are routinely used. The patients are treated
either with an automatic cycler using commercially available dialysis solution
usually prepared in 10-1 plastic containers or with a dialysis system using reverse
osmosis-treated water and concentrate solution to prepare the adequate fluids
required for different situations [24]. Standard dialysis schedules include 3 di-
alysis sessions of 10 to 14 hours duration per week. Large quantities of dialysis
fluid, between 40 to 601 per session, are required. Most patients, at least at home,
are dialysed overnight. The composition of commercial dialysis solution routinely
used are in mmolll: Na 130, KOt02, C195, acetate orlactate 35, Ca1.75, MgO.75,
glucose 83 or 220.
Control of blood glucose requires careful monitoring using the 'finger prick'
method. Most patients are on 2 to 3 daily subcutaneous injections of a mixture of
regular and long-acting insulin. Determination of the extra dose of insulin re-
quired on the dialysis day is determined empirically and is a function of the
dextrose concentration in the dialysis solution required to maintain an adequate
water and electrolyte balance. Control of blood sugar can be achieved, either by
supplementary subcutaneous injection or by addition of regular insulin to the
dialysate [44].

3.1.2. Results
The largest series of diabetic patients treated by IPD recently reported by Mion et
at. [24] emphasizes the encouraging results which can be obtained in the long-
term in a high-risk population. Actuarial survival rates and technique succes rates
differ according to the type of diabetes and age. Among type I diabetics with a
mean age on 33.7 ± 5.5 yrthe patient survival rate was 95% at two years and 83%
at 3 years which compares very favorably with the recent results obtained in
similar patients with hemodialysis [46]. Nevertheless because of a high rate on
transfer to CAPD or transplantation the technique success rate was much lower,
respectively 65% at one year, 55% at 2 years and 48% at 3 years. Among type II
diabetics with a mean age of 61.2 ± 7.9 yr IPD was almost the exclusive treatment
and technique success rate and survival rate were almost identical with percen-
tages of79, 68 and 41 at 1,2 and 3 years, respectively. The main cause of death was
vascular.
Because of repeated and important shifts in body fluids, good control of blood
pressure among patients on IPD is often difficult requiring in most cases the use of
anti-hypertensive drugs. Thirst, commonly observed among non-diabetic pa-
tients on IPD (in relation with hypernatremia occurring late during dialysis and/
or immediately thereafter) (see Chapter 7), can be an important problem in
diabetics leading to excessive water drinking and over weight. Rapid reduction in
extracellular overload frequently encountered might explain the frequent de-
crease in residual renal function after starting dialysis [24], in contrast with what is
observed among patients on CAPD.
572

Because of high diet requirement to avoid malnutrition and low peritoneal

clearance, patients on IPD are at risk of inadequate dialysis. Decline on nerve
conduction velocity on IPD has been reported [24]. Great attention should be
paid to increase dialysis time up to 50 hours per week in patients with a large
surface area.
Visual status can improve on treatment but deterioration has also been ob-
served. As with any dialysis treatment, regular control of eye lesions is required
leading in some cases to successful surgery or retinal photocoagulation. The main
advantage of IPD remains the very low rate of peritonitis, observed also among
diabetic patients. Eradication of peritonitis was virtually obtained in both juve-
nile and type II diabetics treated at home. A rate as low as one episode per 12
patient-years has been observed in a group of 19 patients with a mean age of 33 yr
and a cumulative duration of treatment of 60 years [24]. Such results may be
favored by the routine use of bacteriologic filters [47]. A higher rate of infection,
although still very low, one episode per 5 patient-years, is observed in the non-
insulin-dependent elderly group.
Severe abdominal complications are rarely observed on IPD. Nevertheless,
sclerosing peritonitis has been reported in diabetics as in non-diabetics [29].
Among various potential risk factors, the role of an excessive use of hypertonic
solution and/or use of an acetate-buffered dialysate have been underlined [28,
29]. Such a complication can be associated with a progressive decrease of per-
itoneal ultrafiltration. Surprisingly patients on CAPD with a decreased trans-
peritoneal ultrafiltration rate can recover completely after a few weeks of IPD as
observed in two of our patients and by other groups [27].
Despite major handicaps, mainly costs and duration of dialysis, IPD can offer a
safe and efficient home dialysis method to patients who for various reasons
(including high vascular risk, recurrent peritonitis, loss of ultrafiltration) could
not be treated by either home hemodialysis or CAPD.

3.2. Continuous cyclic peritoneal dialysis

The multiple connections required during CAPD are the true limitation of the
method. They are time consuming and increase the risk of peritonitis. Continu-
ous cyclic peritoneal dialysis (CCPD) was designed by Diaz-Buxo et at. [84] to
avoid the several daily exchanges of CAPD but to retain the physiologic advan-
tages of the method. Large series of diabetic patients treated by CCPD are not yet
available. Early results including our own experience dealing with 6 patients
during a 90 patient-months period are encouraging.

3.2.1. Method
An automatic peritoneal cycler is required. Commercial dialysates with dIfferent
glucose concentrations are used. Selection of dialysate is made according to the
 573

patient's need for ultrafiltration. Our routine schedule is four short nocturnal
cycles using a 1.5% glucose solution. During day time peritoneal cavity is filled
with either a 4.2 or 4.5% glucose solution. Drainage is performed before the first
nocturnal cycle.
Insulin can be administered to patients on CCPD, using the subcutaneous
route. The peritoneal route is also possible and two methods have been proposed
[49]. One schedule eliminates all subcutaneous insulin injections and uses an
intraperitoneal dose of regular insulin that is equivalent to two times the previous
24-hour total subcutaneous dose. Fifty percent of the regular insulin is added to
the diurnal bag (4.25% or 4.5% glucose concentrations) and the other 50% is
equally divided and injected into the bags used for the nocturnal exchanges. The
total dose of insulin required to obtain good blood glucose control is often close to
three times the previous dose administered subcutaneously. The other schedule
maintains the subcutaneous injection of 50% of the previous subcutaneous
insulin dose and adds 15 units on regular insulin to bags with 4.25% or 4.5%
glucose concentrations and 10 units to bags with a 1.5% glucose concentration.
Further gradual increments of intraperitoneal insulin doses with consequent
reductions of subcutaneous insulin doses are made according to blood glucose
concentrations routinely measured using the 'finger prick' method.

3.2.2. Acceptance
Patient acceptance has been recorded as excellent with enjoyment of uninter-
rupted day activities and automated exchanges at night. CCPD allows fewer
manual connections than CAPD under aseptic conditions with the consequent
potential for a lower risk of infection. Indeed the rate of peritonitis recorded by
Diaz-Buxo was only one episode per 26 patient-months [50]. In our series, the
rate was one episode per 18 patient-months. 3 patients out of 6 were totally free of
peritoneal infection. Other clinical and biological parameters are very similar of
those observed among patients treated by CAPD, but the series are too small and
the follow-up too short to authorize definite conclusions.

4. Peritoneal dialysis versus other forms of treatment of ESRD in diabetics. The

best buy

A valid comparison of results obtained in a diabetic population with the different

treatments available remains difficult [4, 9, 51]. The series reported are often
smaller with a too short follow-up. We have stressed how many factors as age,
type of diabetes, severity of vascular lesions can influence the prognosis. Social
and economic factors play a significant role in the utilization of any mode of
treatment. The choice of which renal replacement therapy should be used is
largely influenced by the medical team's experience and preference as well as by
the technical facilities available in a given country or even region. Negative and
574

positive selection are a common consequence of such a reality. Nevertheless a

rational approach to what should be the indication for peritoneal dialysis for a
given patient is already possible.
To offer to each patient the best buy treatment of ESRD of diabetic patients
requires an integrated program with all available transplantation and dialysis
methods. The appropriate decision requires a careful medical psychological and
social investigation including long and multiple conversations with the patient
and his relatives. Before considering to put a diabetic patient on a peritoneal
dialysis program two main questions must be answered:
- Is the patient a candidate for transplantation?
- Should home dialysis be considered?

4.1. Transplantation and peritoneal dialysis

Along with many we consider that transplantation should be the first choice for
juvenile diabetics less than forty years old. Nevertheless, because of the shortage
of cadaver donors, the frequent absence of living potential donors and the
difficult ethical problem raised by transplantation between related persons,
dialysis, at least in many European countries, will be required while the patient
waits for a transplant [6].
In such a group if home dialysis is considered and the patient is able to handle
the technique himself, we propose CAPD as a first choice, which has many
advantages. The CAPD technique does not require any machinery, offers full
independence from relatives and allows, by using the intraperitoneal route for
insulin, the best control of blood glucose level. CAPD does not jeopardize the
chance of a successful transplantation [19]. The method can be used in the post-
operative phase to control uremia in relation with acute tubular necrosis or early
rejection.
If for some reason, mainly partial or total blindness, the patient cannot handle
the CAPD technique, we believe that the exchange is difficult to impose on
relatives and consider CCPD as the appropriate alternative. This technique can
offer adequate care without overtaxing the patient. With CAPD and CCPD the
major risk remains peritonitis. Transplantation should not be performed within a
short time following an episode of acute peritonitis. A delay of three to four
weeks after treatment seems reasonable although to our knowledge no controlled
data on this topic are available.
If center dialysis must be considered while waiting for a transplant, hemo-
dialysis, because of a shorter duration of dialysis, should be selected first, while
intermittent peritoneal dialysis should be kept for patients who for various
reasons, mainly vascular ones, cannot be hemodialysed.
 575

4.2. Hemodialysis versus peritoneal dialysis

Once again where to dialyse the patient remains the first question to answer
before selecting the dialysis method.
If home dialysis is considered, CAPD can be the first choice for many diabetic
patients of either type I or type II including the elderly. Recent reports [8, 9] have
underlined the high survival rate, the quality of life at least for many patients, the
excellent control of blood glucose and an incidence of peritonitis almost equal and
sometimes inferior to what is observed in a non-diabetic population. Nevertheless
the true shortcomings of CAPD should not be forgotten. Some patients will never
be adequately trained and the help of a relative is not always available. The main
drawbacks of CAPD remain the two different types of complications often
implicated in relation to the technique: malnutrition and recurrent peritonitis. A
decreasing ultrafiltration rate or sclerosing peritonitis although not specific for
diabetic patients raise the problem of the long-term function of the peritoneal
membrane [27, 28, 52].
Intermittent and continuous cyclic peritoneal dialysis should be available to
treat some diabetic patients at home either as a first choice or if transfer from
CAPD to hemodialysis is required. Both techniques offer adequate control of
uremia, allow home dialysis in diabetics with high vascular risk, diminish the risk
of peritonitis compared to CAPD, reduce the technical burden either for the
patient and/or the relatives and allow normal activities during the day, the
machine or the cycler doing the dialysis during night.
Malnutrition present in some diabetic patients with gastroenteropathy and
uremia should be considered as a relative contra-indication to starting peritoneal
dialysis. In suchJpatients high protein losses through the dialysate outflow can
precipitate hypoalbuminemia and expose the patients to severe complications.
Treatment by hemodialysis must sometimes precede, at least for a while, any
therapy by various modes of peritoneal dialysis.
Results of peritoneal dialysis, sometimes excellent, but with many drawbacks,
should not darken the regularly improving benefits offered to diabetics by hemo-
dialysis mainly performed in centers, but also at home [9, 46, 53]. On a world
basis hemodialysis remains the method most frequently used to treat insulin- or
non-insulin-dependent diabetics with ESRD. An appropriate evaluation of all
forms of treatment will require a few more years. To adapt to each case all
methods should be available [4, 53, 54, 55]. Adequate facilities offering early
transfer, when one method has failed, are part of an efficient end-stage renal
failure program for diabetics as well as for non-diabetics.

4.3. When to start peritoneal dialysis

In diabetics high rates of vascular complications, mainly retinopathy and pe-

576

ripheral arteritis, often jeopardize the otherwise excellent results obtained either
with transplantation or dialysis methods. Uncontrolled high blood pressure is a
frequent symptom of the terminal phase of end-stage renal failure leading to
severe ocular lesions and to lethal cardiac or cerebral vascular complications. For
such seasons diabetics could be considered as excellent candidates for early
dialysis. However the true benefit of such a therapeutic approach versus the
detriment to the patient and the community is hard to evaluate.
Criteria to start dialysis are similar for all dialysis methods. One advantage of
peritoneal dialysis is the possibility of starting dialysis immediately after insertion
of the catheter, while the A V fistula may not be used immediately requiring, if
necessary, other temporary routes to vascular access.
For diabetic patients with easily controlled blood pressure and good nutritional
status, starting dialysis when the creatinine clearance reaches 7 to 5 ml/min seems
appropriate. Earlier dialysis can be considered only for patients who present
refractory edema and/or severe hypertension despite high doses of antihyperten-
sive drugs.

4.4. Conclusion

Peritoneal dialysis has proved to offer excellent treatment and a unique oppor-
tunity to treat insulin- or non-insulin-dependent diabetics at home. CAPD, as far
as simplicity, control of blood glucose and cost are concerned, offers the best but
should not rule out the other peritoneal dialysis techniques. IPD or CCPD are for
some patients excellent modes of treatment, either permanent or transient. All
forms of peritoneal dialysis should not be considered as the exclusive technique
for the diabetic population but only an important part of an integrated program
including hemodialysis and transplantation. Peritoneal dialysis is not the ultimate
solution for an obviously desperate situation.

References

1. Avram MM: Diabetic renal failure. Nephron 31: 285-290, 1982.

2. Friedman EA: Clinical imperatives in diabetic nephropathy. Kidney Int 23 (Suppl) 14: sl6-s19,
1982.
3. Friedman EA: Overview of diabetic nephropathy. In: H Keen, M Legrain (eds), Prevention and
treatment of Diabetic Nephropathy. MTP Ltd, Lancaster, 1983, pp 3-19.
4. Legrain M: Diabetics with end stage renal disease: the best buy. Editorial. Diabetic Nephropathy
2: 1-3, 1983.
5. Sutherland DE, Morrow CE, Fryd D, Ferguson R, Simmons R, Najarian J: Improved patient and
primary renal allograft survival in uremic diabetic recipients. Transplantation 34: 319-325, 1982.
6. Jacobs C, Brunner F, Brynger H, Challah S, Kramer P, Selwood N, Wing A: The first five
thousand diabetics treated by dialysis and transplantation in Europe. Diabetic Nephropathy 2:
11-16, 1983.
 577

7. Kjellstrand C, Whitley K, Comty C, Shapiro F: Dialysis in patients with diabetes mellitus.

Diabetic Nephropathy 2: 5-14, 1983.
8. AmairP, KhannaR, LeibeIB,PierratosA, VasS, MeemaE, BlairG, ChisholmL, VasM,Zingg
W, Dibenis G, Oreopoulos DG: Continuous ambulatory peritoneal dialysis in diabetics with end
stage renal disease. N. Eng!. J. Med. 306: 625-630,1982.
9. Legrain M, Rottembourg J, Bentchikou A, Poignet JL, Issad B, Barthelemy A, Strippoli P:
Dialysis treatment of insulin dependent diabetic patients. A ten years experience. Clinical
Nephrology 21: 72-81, 1984.
10. Lameire N, Dhaene M, Matthijs E, de Paepe M, Vereerstraeten P, Dratwa M, Ringoir S:
Experience with CAPD in diabetic patients. In: H Keen, M Legrain (eds), Prevention and
Treatment of Diabetic Nephropathy. MTP Ltd, Lancaster, 1983, pp 289-297.
11. Polla-Imhoof B, Pirson Y, Lafontaine JJ, Vandenbroucke JM, Cosyns JP, Swhifflet JP, Alex-
andre G, van Ypersele de Strihou C: Resultats de l'hemodialyse chronique et de la transplantation
renale dans Ie traitement de l'uremie terminale du diabetique. Nephrologie 3: 80-84, 1983.
12. Thomson NM, Simpson RW, Hooke D, Atkins RC: Peritoneal dialysis in the treatment of
diabetic end stage renal failure. In: RC Atkins, NM Thomson, PC Farell (eds), Peritoneal
Dialysis. Churchill Livingstone, New York, 1981, pp 345-355.
13. Flynn C: The diabetic on CAPD. In: E. Friedman, F. L'Esperance (eds), Diabetic Renal Retinal
Syndrome 2. Grone and Stratton, New York, 1982, pp 320-330.
14. Flynn C, Hibbard J., Dohrman B: Advantages of continuous ambulatory peritoneal dialysis to the
diabetic with renal failure. Proc Eur Dial Transplant Assoc 16: pp 184-190, 1979.
15. Madden MA, Zimmerman S, Simpson DP: Continuous ambulatory peritoneal dialysis in diabetes
mellitus. Am J Nephrol2: 133-139, 1982.
16. Legrain M, El Shahat Y, Rottembourg J, Balducci A, Jacobs C: Continuous ambulatory per-
itoneal dialysis (CAPD) versus other treatment modalities in end stage diabetic nephropathy. In:
G Gahl, M Kessel, K Nolph (eds), Advances in Peritoneal Dialysis. Excerpta Medica, Amster-
dam, 1981, pp 365-368.
17. Rottembourg J, El Shahat Y, Agrafiotis A, Thuillier Y, de Groc F, Jacobs C, Legrain M:
Continuous ambulatory peritoneal dialysis in insulin dependent diabetics: a 40 month experience.
Kidney Int 23: 40-45,1983.
18. Rottembourg J, Issad B, Poignet JL, Strippoli P, Balducci A, Slama G, Gahl G: Residual renal
function and control of blood glucose levels in insulin-dependent diabetic patients treated by
CAPD. In: H. Keen, M Legrain (eds), Prevention and Treatment of Diabetic Nephropathy. MTP
Ltd, Lancaster, 1983, pp 339-352.
19. Khanna R, Wu G, Ghisholm L, Oreopoulos D: Update: Further experience with CAPD in
diabetics with end stage renal disease. Diabetic Nephropathy 2: 8-12,1983.
20. Rottembourg J, Jacq D, Vonlanthen M, Issad B, El Shahat Y: Straight or curled Tenckhoff
peritoneal catheter for continuous ambulatory peritoneal dialysis. Perit Dial Buill: 123-124, 1981.
21. Oreopoulos DG, Zellerman G, Izatt S: The Toronto Western Hospital permanent puritoneal
catheter and continuous ambulatory peritoneal dialysis connector. In: M Legrain (ed). Continu-
ous Ambulatory Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1979, pp 73-78.
22. Avram MM, Paik SK, Okanya D, Rajpal K: The natural history of diabetic nephropathy:
unpredictable insulin requirements. A further clue. Clin Nephrol21: 36-38, 1984.
23. Friedman E, Levits C, Hirsch S, Butt K: Feasibility of self blood glucose monitoring in uremic
diabetics. In: E. Friedman, F L'Esperance (eds), Diabetic Renal Retinal Syndrome 2. Grone and
Stratton, New York, 1982, pp 437-445.
24. Mion C, Slingeneyer A, Canaud B, Gules R, Branger G, Chong G, Mourad G: Home intermit-
tent peritoneal dialysis in the treatment of end-stage diabetic nephropathy. 1982 Update. In: H
Keen, M Legrain (eds), Prevention and Treatment of Diabetic Nephropathy. MTP ltd, Lancas-
ter, 1983, pp 263-277.
25. Rottembourg J, Issad B, Gallego JL, Degoulet P, Aime F, Gueffaf B, Legrain M: Evolution of
578

residual renal function in patients undergoing maintenance hemodialysis or continuous ambula-

tory peritoneal dialysis. In: Proc Eur Dial Transpl Assoc 19: 397-401. Pitman, London, 1982.
26. Faller B, Marichal JF: Loss of ultrafiltration in CAPD: clinical data. In: G Gahl, M Kessel, K,
Nolph (eds), Advances in Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1981, pp 227-232.
27. Slingeneyer A, Canaud B, Mion C: Permanent loss of ultrafiltration capacity of the' peritoneum in
long term peritoneal dialysis: an epidemiological study. Nephron 33: 133-138, 1983.
28. Rottembourg J, Gahl G, Poignet JL, Mertani E, Strippoli P, Langlois P, Tranbaloc P, Legrain M:
Severe abdominal complications in patients undergoing continuous ambulatory peritoneal di-
alysis. In: Proc Eur Dial Transpl Assoc 20: 236-242. Pitman, London, 1983.
29. Slingeneyer A, Mion C, Mourad G, Canaud B, Faller B, Beraud JJ: Pr}lgressive sclerozing
peritonitis: a late and severe complication of maintenance peritoneal dialysis. Trans Am Soc Art
Int Organs 29: 633-640, 1983.
30. Balducci A, Slama G, Rottembourg J, Baumelou A, Delage A: Intraperitoneal insulin in uraemic
diabetics undergoing continuous ambulatory peritoneal dialysis Br Med J 283: 1021-1023,1981.
31. Gabbay K: Glycohaemoglobin. A measure on glycaemic cruising altitude. In: H Keen, M Legrain
(eds), Prevention and Treatment of Diabetic Nephropathy. MTP Ltd, Lancaster, 1983, pp 145-
153.
32. Norbeck H, Oro L, Carlson LA: Serum lipid and lipoprotein concentrations in chronic uremia.
Act Med Scand 200: 487-495, 1976.
33. Gokal R, Ramos JM, McGurk JG, Ward MK, Kerr DNS: Hyperlipidaemia in patients on
continuous ambulatory peritoneal dialysis. In: G Gahl, M Kessel, K Nolph (eds), Advances in
Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1981, pp 430-433.
34. Moncrief JW, Pyle WK, Simon P, Popovich RP: Hypertriglyceridemia, diabetes mellitus and
insulin administration in patients undergoing continuous ambulatory peritoneal dialysis. In: J
Moncrief, R Popovich (eds), CAPD Update. Masson, New York, 1981, pp 143-165.
35. Norbeck H: Lipid abnormalities in continuous ambulatory peritoneal dialysis. In: M Legrain (ed),
Continuous Ambulatory Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1979, pp 298-301
36. Wu G: Osmotic agents for peritoneal dialysis solutions. Perit Dial Bull 2: 154-158, 1982.
37. Bazzato G, Coli U, Landini S, Frascasso A, Morachiello P, Righetto F, Scanferla F: Xylitol and
low dosages of insulin: new perspectives for diabetic uraemic patients on CAPD. Perit Dial Bull 2:
161-165, 1982.
38. De Paepe M, Matthijs E, Peluso F, Dokkart R, Lameire N: Experience with glycerol as the
osmotic agent in peritoneal dialysis in diabetic and non-diabetic patients. In: H Keen, M Legrain
(eds), Prevention and Treatment of Diabetic Nephropathy. MTP Ltd, Lancaster, 1983, pp 299-
311.
39. Oren A, Wu G, Anderson G, Marliss E, Khanna R, Pettit J, Mupas L, Rodella H, Brandes L,
Roncari D, Kakis G, Harrison J, Mc Neil K, Oreopoulos DG: (1983). Effective use of amino-acid
dialysate over four weeks in CAPD patients. Trans Am Soc Art Int Organs 29: 604-610, 1983.
40. Kohner E, Chahal P: Retinopathy in diabetic nephropathy. In: Prevention and Treatment of
Diabetic Nephropathy. MTP Ltd, Lancaster, 1982, pp 191-196.
41. Sutherland D, Fryd DS, Simmons RL, Ferguson RM, Najarian JS: Current status of kidney
transplantation in uremic diabetics at the University of Minnesota. In: E Friedman, F I'Esperance
(eds), Diabetic Renal Retinal Syndrome 2. Grune and Stratton, New York, 1982, pp 373-383.
42. Brown PM, Johnston KW, Fenton SSA, Cattran DC: Symptomatic exacerbation of peripheral
vascular disease with chronic ambulatory peritoneal dialysis. Clin Nephrol16: 258-261, 1981.
43. Rdusher H, Levitz CS, Butt KMH, Friedman E: Pediatric contribution to limb preservation in
diabetic renal transplant recipients. In: E Friedman, F I'Esperance (eds), Diabetic Renal Retinal
Syndrome 2. Grune and Stratton, New York, 1982, pp 419-426.
44. Ahmad S, Gallager N, Shen F: Intermittent peritoneal dialysis: Status reassessed. Trans Am Soc
Artif Intern Organs 25: 86-90,1979.
45. Goldberg M, Baugham KJ, Wombolt DG: Home intermittent peritoneal dialysis in high risk
 579

patients. In: G Gahl, M Kessel, K Nolph (eds), Advances in Peritoneal Dialysis. Excerpta
Medica, Amsterdam, 1981, pp 138-143.
46. Shapiro FL: Haemodialysis in diabetic patients. In: H Keen, M Legrain (eds), Prevention and
Treatment of Diabetic Nephropathy. MTP Ltd, Lancaster, 1983, pp 247-259.
47. Slingeneyer A, Mion C, Despaux E, Perez C, Duport J, Dansette AM: Use of a bacteriologic
filter in the prevention of peritonitis associated with peritoneal dialysis: Long-term clinical results
in intermittent and continuous ambulatory peritoneal dialysis. In: RC Atkins, NP Thompson, PC
Farrell (eds), Peritoneal Dialysis. Churchill Livingstone, Edinburgh, 1981, pp 301-312.
48. Diaz-Buxo JA, Walker P, Farmer CD, Chandler JT, Holt KL, Cox P: Continuous cyclic
peritoneal dialysis. Trans Am Soc Artif Int Organs 27: 51-53, 1981.
49. Diaz-Buxo J, Walker P, Farmer C, Chandler J, Burgess W, Holt K: Diabetic nephropathy
experience with various dialystic therapies. Contemp Dialysis 4: 9-13, 1983.
50. Diaz-Buxo J, Chandler JT, Farmer CD, Walker P, Holt KL, Burgess WP, Orr SL: Long term
observation of peritoneal clearances in patients undergoing peritoneal dialysis. ISAO 6: 21-25,
1983.
51. Farrell PC, Randerson DH: Comparison of CAPD with HD and IPD. In: G Gahl, M Kessel, K
Nolph (eds), Advances in Peritoneal Dialysis. Excerpta Medica, Amsterdam, 1981, pp 131-137.
52. Oreopoulos D: Peritoneal membrane: handle with care. Perit Dial Bull 3: 111-113, 1983.
53. Kjellstrand C, Whitley K, Comty C, Shapiro F: Dialysis in patients with diabetes mellitus.
Diabetic Nephropathy. 2: 5-17, 1983.
54. Mion C: Integration of peritoneal dialysis in a regional end stage renal disease programme: a
French experience in Languedoc-Roussillon. In: CAtkins, N Thomson, P Farell (eds), Peritoneal
Dialysis. Churchill Livingstone, Edinburgh, 1981, pp 395-403.
55. Legrain M, Rottembourg J, Gahl G, Bentchikou A, Strippoli P: The treatment of renal failure in
diabetic patients. The best buy. In: H Keen, M Legrain (eds), Prevention and Treatment of
Diabetic Nephropathy. MTP Ltd, Lancaster, 1983, pp 361-376.
17. Peritineal membrane stability and the kinetics
of peritoneal mass transfer

PETER C. SPENCER and PETER C. FARRELL

1. Introductin

Long-term CAPD is dependent upon the ability of the peritoneum to maintain

adequate water and solute clearances; the latter are a function of both the
permeability of the peritoneal membrane and the mass transfer area available.
Normally these two factors cannot be separated and are grouped together as the
mass transfer area coefficient (MTC) which determines the maximum clearance
rate for a particular solute [1].
When faced with changing blood biochemistries or increased morbidity, the
clinician must decide whether these changes are related to problems with the
peritoneum or to other factors such as dietary changes or intercurrent illness. If
peritoneal mass transfer capacity has been lost, the question must be asked how
much capacity has been lost and whether the damaged conditions will still allow
"adequate" dialysis.
Blood solute levels and clearances are significantly affected by many factors,
apart from a solute's peritoneal mass transfer characteristics. These factors
include: (a) total water removal (dependent on daily dialysate dwell times,
dialysate dextrose levels, blood osmolarity and glucose transfer chacteristics); (b)
residual renal clearance; and (c) net solute generation rates.
Where membrane characteristics have changed to the point where dialysis is no
longer considered adequate, the following questions become relevant.
1. Is the change permanent?
2. Is the change continuous or was it a step change?
3. Can required clearances be met by varying the daily dialysis regimen (e.g.
varying dwell times, dialysate volume or dextrose)?
These questions can only be adequately addressed by modeling patient/dialysate
solute/water transfer.
Many models have been used to estimate solute mass transfer coefficients.
These models most often require the use of a computer as well as knowledge of
numerical methods. We have applied such a modeling procedure to assess
582

variations in maximum solute clearances with time on CAPD. Our results showed
that during CAPD treatment of up to 2 years in 15 patients, 3 of the patients lost
membrane permeability and/or transfer area whilst 12 showed no evidence of
membrane deterioration with time [2]. Unfortunately, the application of model-
ing in the way we applied it is somewhat onerous and requires considerable effort
on the part of the dialysis staff, despite external help. However, not only are
people with the requisite mathematical and other skills required to be on staff,
there also has to be willingness to commit scarce resources to applying these
techniques. The task in collecting and analyzing the data may not now be so
onerous. A recent model proposed by Garred et al. [3] yieldsttc values more
simply, it can be programmed onto a hand-held calculator or the values simply
determined by graphical procedures. Hence, a simplified yet sufficiently accurate
technique exists for following long-term changes in the peritoneum. With appro-
priate manipulation this model can also be used to estimate the change in
expected blood-solute levels due to variations in MTC values, daily dialysis
regimen, solute generation rate and solute renal clearance.
Our model is based on the concept of membrane-limited diffusion. As outlined
by Dedrick et al. [4], this does not allow for the spatially distributed capillary
network surrounding the peritoneal cavity. However, the validity of our sim-
plification of the physical process, as with all models, can only be tested by
investigation of its predictive capacity.
Variations in the peritoneal membrane caused by CAPD fall into 2 categories.
Some patients suffer from decreasing MTCs with time due to the formation of
adhesions and/or fibrosis. In the previously mentioned study of 15 patients over 1
to 2 years, we found decreasing MTCs for urea, creatinine and vitamin B12 in 3
cases. Two of these patients died and autopsies showed in one, fibrous adhesions
and in the other dense collagenous connective tissue infiltrated by chronic inflam-
matory cells and a layer of fibrin [2]; other groups have shown similar results [5].
The other category includes patients who lose the ability to remove water by
ultrafiltration during normal 4-10 hour dialyses. This may be the result of a large
glucose MTC and a concomitantly rapid decrease in osmotic gradient during
dialysis, due to increased peritoneal membrane blood flow and/or permeability;
lymphatic water removal may also playa role in this phenomenon. Although we
found no patients in our study who suffered permanently from this problem,
several patients in renal units associated with our studies have encountered it.
Possible causes of changes in peritoneal membrane solute and water transfer
characteristics have been suggested as follows [6]:
* low dialysate pH
* irritation caused by particulate matter in the dialysate
* chemical irritation, both by chemicals placed in the peritoneum and drugs
taken orally
* the hypertonicity of the dialysate
* catheter irritation
* peritonitis
 583

In our studies, all patients tested used the same dialysate composition (apart
from dextrose) from a local supplier (Travenol Laboratories, Toongabbie,
NSW). Since only some patients have shown marked variation in membrane
properties over 1 to 2 years, this suggests that pH and particulate matter effects
are negligible or, if involved, are vastly different in their manifestation from
patient to patient.
Patient numbers, followed by us for any length of time, are too small to infer
anything about the influence of chemical irritants (e.g. low levels of betadine) or
orally ingested drugs. New connection systems utilizing UV sterilization and heat
sealing will undoubtedly decrease the incidence of accidental betadine dialysate
contamination. We have examined the acute effects of hypertonicity by compar-
ing KBDS evaluated with high (4.25%) and low dextrose solutions (1.5%); we
found no discernible effect (vide infra). The long-term consequences of using
hypertonic solutions remain to be elucidated.
The effects of catheter irritation may, in the future, be determined by compari-
son of patients using catheters which allow less movement and mechanical
irritation than those, such as the free-floating Tenckhoff catheters which are
known to irritate some patients. The need for high dextrose solution (4.25%) to
remove fluid is a factor often cited as possibly causing membrane changes. For
patients with this problem, it may often be possible to manipulate dialysate dwell
times so that required solute clearances and water removal can be obtained using
solutions of lower hypertonicity (e.g. 1.5% dextrose). Finally, we have found no
correlation between patients' long-term membrane status and the number of
treated peritonitis cases or recurrent episodes.
The discussion to follow outlines our current work on assessing the status of the
peritoneal membrane on a long-term basis and also shows how a simple MTC
model can be used to estimate the influence on solute blood levels of the following
factors:
1. changes in daily dialysis regimen
2. solute MTC variation
3. solute generation variation
4. residual renal clearance variation
Finally, a technique is proposed for a simple analysis of peritoneal membrane
viability based on rate of equilibration between dialysate and blood. If future
work proves the validity of this approach, it will aid in determining whether loss of
membrane permeability/mass transfer area is the cause of resultant patient mor-
bidity or solute blood-level variations. Irregardless, the technique is simple
enough to be applied on a routine basis.

2. Methods

Patients' MTCs were estimated for urea and creatinine in studies conducted
584

during the second exchange of the day, using a patient's normally recommended
dextrose bag content. Blood samples of 10 mls were taken during inflow and
outflow of the test bag; the latter was always pre-warmed to 37° C. During dwell, 6
dialysate samples were taken at approximately 20-minute intervals, with samples
taken from the bag injection port by running 100-200ml into the bag from the
peritoneum during dialysis; a 7-ml sample was then taken from this pool. Sample
dilution effects were minimized by infusing all dialysate from the bag and, as far
as possible, from the infusion lines into the peritoneum, between and before each
dialysate sampling step. Sampling times were recorded, all times being measured
from the midpoint of dialysate infusion.
Urea and creatinine in plasma and dialysate samples were determined in
duplicate using a Roche Cobas auto analyser, and pre- and post-dialysate volumes
were measured during each test by weighing. The data collection is therefore
relatively straightforward.

3. Theory

3.1. Measurement of KBD

The simple model used in this study has been described in detail elsewhere [3].
The equation used to determine MTC values (given as K BD ) is as follows:

In VD (CB-CD) = _ KBD . t (1)

VDO(C B- CDC) VD
Sensitivity analysis shows that errors are small if the dialysate volume, VD' during
dialysis is estimated from the initial and final (drained) dialysate volume. This is
especially so if low dextrose dialysate (1.5%) is used during a test (i.e. low total
ultrafiltration volumes). As an alternative methodology, a dialysate marker
molecule can be used to estimate dialysate volume during dialysis. However, it is
more complex and introduces many new variables into the procedure. The
approximate change in dialysate volume during dialysis can be estimated using an
exponential model [1]:
VD = VD o + A (l-e- at) (2)
where the time constant a is taken as 0.0192 min-1 which assumes that 90% of the
ultrafiltration is over in 2 hours. The effect of this assumption on KBD values, if the
90% ultrafiltration point is reached at 1 or 4 hours, is an error in KBD of
approximately 5 %. The constant A is found by inserting the known values V D (t O

= 0) and V D at time t(min) to the end of dialysis. Average dialysate volume is then
given by
V - f~VD' dt = V °+ A + ~ (e- at - 1) (3)
D - J~dt D a' t
 585

The time t (mins) is taken from a point half-way through dialysate infusion. By
obtaining the dialysate solute concentration at known points during dialysis a plot
»
of In(VD(CB-CD versus t yields a straight line of slope -KBDIVD from which KBD
can be calculated. A KBD evaluation can be done over a short dialysis dwell
(-1h). Where CD approaches CB (~D >0.9) the model is sensitive to small solute
B
concentration analysis errors, hence paired {CD,t} data are best omitted if they
are within this range.
Once a KBD value has been determined for a given solute it may then be used to
estimate the effects on patient blood-solute levels of the various factors pre-
viously outlined.

3.2. Other model applications

By rearranging equation [1), a value for CD at the end of dialysis can be found as
follows:

(4)

In contrast, ifthe daily amount of solute removed via dialysis is known (i.e. solute
generation rate), then equation (4) can be used to estimate the resultant blood
level of the solute for a given set of patient/dialysis conditions based on an
iterative solution to equations (4) and (6) (vide infra).
The amount of solute (e.g. urea, creatinine) removed can be readily obtained
from a 24-hour dialysate/urine collection where the daily net generation rate is
given by mass balance:
n
G =(L VdFdJ + VuCu + Vs [WPt- W,C,D (5)
J= 1

A steady-state blood solute level is reached when the amount removed, both
during dialysis and by residual renal function (KR ) equals net solute generation
rate; the blood solute level is at steady-state and the inventory term can be
ignored. Hence:
n n
G = L . VdJ C dJ + VuCu = L VdJCdJ + KRCB (6)
J= 1 J=1

where n is number of exchanges per day. Equation (6) assumes negligible

contribution of skin/fecal solute removal; these are not included in the calculation
of G.
If net solute generation and residual renal solute clearance are known or
assumed, then CB can be found from equations (4) and (6) for any combination of
586

K BD , dwell times, dialysate volumes and daily dialysis frequency. These factors as
well as the generation rate and renal clearance, may be manipulated to study the
sensitivity of CB to anyone or all of these parameters. The calculation simply
involves assuming a value of CB , finding CD from equation (4) for each dialysis of
the day, then calculating G from equation (6). CB is manipulated until Gcalcul-
ated = Gactual. All the procedures above can be accomplished with a hand-held
calculator (see appendix 1 for a worked example).
In the case of urea, generation rates can also be used to estimate patient dietary
protein intake as we have reported [7].

4. Results and discussion

A relevant concern in following long-term membrane viability by means of solute

MTCs, is whether or not the latter change with variations in dialysate volume and
dextrose content. If so, this would have to be taken into account in a long-term
study; ideally tests on each patient should be constant with respect to initial
dextrose levels and dialysate volumes.
A study completed by us recently showed significant differences in MTC values
for 2-liter versus 1-liter exchanges. Twelve patients were used as their own
controls, with urea and creatinine MTCs being estimated during a 2-liter dwell,
followed within a few days by a 1-liter exchange or vice-versa using a constant
concentration of 1.5% dextrose. The average combined ratio of 2-liter to 1-liter
MTCs for the 2 solutes was 1.2 ± 0.3 (p<0.05) (urea 1.3 ± 0.4; creatinine 1.2 ± 0.3
(mean ± SD)) suggesting an average 20% greater transfer area for 2-liter versus
1-liter exchanges. A similar study in 5 patients comparing solute transfer using
4.25% to 1.5% dextrose showed no significant MTC differences for either urea or
creatinine. These results suggest that for a given total daily dialysate exchange
volume, clearances of non-equilibrating solutes can be maximized by optimizing
the separate exchange volumes making up the daily total, e.g. 3 x 3 liters/day may
be better than 4 x 2.25 liters/day. Regarding long-term studies, where test dialy-
sate input volume and dextrose content remain constant, this factor should have
little effect as average dialysate test volume will vary much less than the factor of 2
used in our 2-liter versus 1-liter study.
Initial MTC results for 7 patients involved in our most recent study of long-term
variation in KBD are given in Table 1.

Patients 1-3 were in their first year of CAPD when tested, while patients 4-7 were
in their second or third year of CAPD. There were no significant differences
between the urea and creatinine MTCs of the 2 groups. However, patient 7
showed a significant decline in urea KBD over the test period (variance test on K BD
slope, p<0.05), and the creatinine values followed a similar trend which was not
however significant. This patient's first test was done within 2 weeks of finishing
 587

Table 1. Sequential urea/creatinine mass transfer coefficient versus time on CAPO

Test result KBD (ml/min) versus months on CAPO' Mean ± SO

Patient 1 2 3 4

1 urea 26 (0) 16 (2) 20 (8) 21 ±5

creatinine 14 8 15 12±4
2 urea 23 (0) 19 (4) 21 (7.5) 23 (11) 22±2
creatinine 17 15 11 21 16±4
3 urea 15 (0) 15 (4) 17 (9) 16± 1
creatinine 6 3 6 5±2
4 urea 24 (17) 26 (20) 18 (24) 23±4
creatinine 6 6 5 6±1
5 urea 21 (19.5) 23 (21.5) 16 (24.5) 18 (27.5) 20±3
creatinine 13 9 10 8 1O±2
6 urea 23 (13) - (16.5) 19 (19) 21 ±3
creatinine 10 16 13 13±3
7 urea 32 (14) 18 (21) 15 (24) 22±9
creatinine 14 10 8 11 ±3

• Time in months, after commencement of treatment when MTC was measured, is given in paren-
theses.

antibiotic treatment for peritonitis, hence there is the possibility that these KBD
values reflect a more permeable peritoneum as the aftermath of the effect of
peritonitis. Further tests are being done on this patient to determine whether an
actual downward trend exists; no other patients showed significant KBD varia-
tions. Tests by us of day to day patient KBD variations gave coefficients of
variation in the range 5-25%, this fact should be considered when analyzing
patient data.
A further patient (No.8) who had initially been evaluated in 1978-79 was
re-evaluated after approximately 5 years of CAPO; results are shown in Table 2.

Table 2. Patient 8 mass transfer coefficient evaluation

Time on CAPO (months) 1 5 8 58

Weight (kg) 48.9 52.0 53.6 50.2
Urea
serum (mmol/L) 28.0 22.5 21.8 27.8
generation rate (ILmoIlL) 132 98 117 172
P.c.R. G/24°/kg 1.0 0.8 0.8 1.2
KBD mllmin 17 31 29 32
Creatinine
serum (mmol/L) 0.87 0.82 0.85 0.90
excretion (mg/24°/kg) 18.4 10.4 10.0 16.7
KBD (ml/min) 12 19 14 20
Ultrafiltration rate mllmin 0.4 -0.7 1.0 0.9
588

Initially this patient, who experienced difficulty with 2-liter exchanges, was doing
9 x 1-liter exchanges per day (this continued for evaluations conducted through 8
months of CAPD), at the last evaluation she had reduced this to 7 x 1-liter
exchanges per day. Both the urea and creatinine MTC (KBD ) values have been
remarkably stable over her 5 years of CAPD, suggesting no significant changes in
peritoneal properties. The only significant change in monitored parameters
during the 5-year period has been an apparent increase in dietary protein intake,
as reflected by her increased urea generation rate at the last evaluation; protein
catabolic rate (PCR) was calculated from urea generation rate [7]. This patient
had had 3 bouts of peritonitis to the date of her last evaluation.
Concomitant with the long-term membrane permeability studies summarized
in Tables 1 and 2, one of the renal units participating in our studies did 1 or more
24-hour urea/creatinine total clearance/generation tests (dialysate + urine) on
their CAPD patients. These data were analyzed by means of equation (5) and
consisted of up to 1 year's worth of data collection per patient. These data have
allowed us to compare and contrast actual with predicted values of serum urea
and creatinine values, based upon mean MTC values and mean clearance/
generation rates collected on the same patients. In other words, the simplified
model was used to predict each patient's urea and creatinine blood levels at the
time of the clearance tests and these values were then compared with actual
values obtained. Tables 3 and 4 provide the data on 5 patients.

These results were calculated assuming average dwell times of 3 x 4.5 hours and
1 x 9 hours (not recorded at time of test). Except for patient 6, creatinine and
urea blood levels were predicted within ~ 10% of the actual recorded levels.
These limited data suggest that CB predictions made by the model, where G and
KBD are known or assumed, will be within an acceptable and clinically useful
range.

Table 3. Observed versus predicted urea plasma levels

Patient CB urea observed CB urea predicted Predicted/observed ratio

(mmoI/L) (mmoI/L)*

4 20 21 1.05
6 20 21 1.05
7 26 28 1.08
7 27 27 1.00
7 29 28 0.97
7 20 22 1.10
9 28 28 1.00
10 18 17 0.94
Mean±SD 1.02 ±0.06

• From equations (4) and (6) (see Appendix 1).

 589

Table 4. Observed versus predicted creatinine plasma levels

Patient CB creatinine observed CB creatinine predicted Predicted/observed ratio

(mmoIlL) (mmoIlL)

6 1.37 1.11 0.81

7 0.63 0.60 0.95
7 0.83 0.74 0.89
7 0.86 0.79 0.92
7 0.86 0.83 0.97
9 0.90 0.91 1.01
10 0.68 0.73 1.07
Mean ± SO 0.95 ± 0.08

For these calculations it was also assumed that each patient had a residual
volume in their peritoneum of 100 ml after dialysate outflow and that, with
respect to urea and creatinine, it was at equilibrium with blood levels. This
assumption was used to calculate CD ° in equation (4). In actuality errors are
negligible if CDO is taken as zero, but increase as the KBD value decreases. At a KBD
of 5ml/min, the calculated C B varies by about 3% if CDo = 0 compared to the
assumption used in the calculations. The model also assumes a constant daily
blood solute level, this is not in fact the case as small swings occur during dialysis
and calculated CB values can vary from analyzed values by about 5% due to these
daily variations.
Faced with changing blood biochemistries a clinician must decide whether or
not these changes are the result of peritoneal membrane variations. Table 5
shows predicted blood urea levels for patients of various weights with varying KBD
values. Daily water removal during 4 X 2-liter dialyses is assumed to be 1.2 liters
with a dialysis regimen consisting of 3 x 4.5 hours and 1 x 9 hours. Patients in
Table 5 are assumed to have no residual renal function and to be eating 1 g
protein/day/kg; generation rates are calculated from a correlation described
elsewhere [6].

Table 5. Predicted urea CB (mmolll) for 1 g/day/kg protein intake

KBD (ml/min) Weight (kg)

40 50 60 70 80

5 18 26 35 44 53
10 13 20 26 33 39
15 12 18 23 29 35
20 11 17 22 28 33
30 11 16 21 27 31
40 11 16 21 26 31
590

Table 6 shows the predicted effect of residual renal clearance on blood urea
levels for a 60-kg patient (same dwell times and assumptions as for Table 5).

Table 6. Predicted urea C B (mmolll) versus residual renal clearance (60-kg patient)

KBD (mllmin) Renal clearance (ml/min)

0 2 3

5 35 28 23 20
10 26 22 19 17
15 23 20 17 15
20 22 19 17 15
30 21 18 16 14
40 21 18 16 14

In our experience normal KBD values for urea are in the range of14-35 ml/min.
Table 5 shows that for large changes within this range, much smaller variations in
blood urea levels will be seen. In contrast, variations in urea CB are much more
sensitive to dietary or renal clearance variations, as shown in Tables 6 and 7.
Large permeability or mass transfer area variations can occur without resultant
loss of daily urea clearance, provided water removal rates remain constant (i.e.
manipulated by varying dialysate dextrose concentration).
Predictions can also be made as to how dialysate dwell time and volume
variations will affect CB levels. As an example, Table 8 shows what would be
expected if the 80-kg patient in Table 5 began using 3-liter dialysate bags/
exchange.

Table 8 outlines an obvious use for the model considering the range of dialysate
bag volumes available. The lower CB values for the 3 X 3 L compared to the 4 X 2
L regimen are only maintainable if the ultrafiltration volume of 1.2 L can be
maintained over the longer 7-hour dwells.

Table 7. Predicted urea C B (mmolll) versus DPI (60-kg patient)

KBD (mllmin) Dietary protein intake (gm/24°/kg)

0.8 1.0 1.2

5 24 35 45
10 18 26 34
15 16 23 30
20 16 22 28
30 15 21 27
40 14 21 27
 591

Table 8. Predicted urea CB (mmolll) versus dialysate volume (80-kg patient)

KBD (mllmin) Daily dialysate infusion volume-

4 x 2 liters 3 x 3 liters 4 x 3 liters

5 53 49 46
10 39 36 33
15 35 31 28
20 33 29 25
30 31 28 23
40 31 28 22

Time regimen 3 x 4.5 hours 2 x 7 hours 3 x 4.5 hours

1 x 9 hours 1 x 9 hours 1 x 9 hours

* Assuming daily ultrafiltration volume of 1.2 liters.

Another example of possible use for this model concerns patients who cannot
remove sufficient water over 4-9-hour dialyses. One such patient who had this
problem when initially on CAPD, had KBDS for urea and creatinine of 46 and
32 ml/min respectively. At four hours of dialysis, using 1.5% dextrose dialysate,
he normally absorbed 100 ml dialysate (i.e. net 1.9 L in outflow). Blood urea and
creatinine levels were around 26-29 and 0.70-l.00mmol/L respectively. At 2.5
hours of dialysis using 1.5% dextrose dialysate he could remove 200 mls of water
(i.e. net 2.2 L out). The model predicts that if dialysed for 4 x 2.5 hours daily with
14 hours/day off CAPD, his blood urea and creatinine levels would have been
20 ± 2 and 0.68 ± 0.05 mmol/L (mean ± off dialysis fluctuation). (These levels
assume urea and creatinine generation rates calculated for this patient on the day
his MTC values were estimated.)
This patient eventually was able to remove water dialysing for 4-9 hours per
dialysis, so the theoretical regimen outlined above was not actually tested. A
question arising from this prediction is whether or not large molecules will follow
suit with urea and creatinine by moving rapidly into the dialysate over 2.5 hours.
In the above case, the patient had a vitamin BI2 KBD of 23 ml/min, suggesting that
large waste products would be cleared sufficiently in this time. The effect of a
12-14-hour per day rest from CAPD for such people has yet to be determined.
Such rest may allow repair of the peritoneum, if it is damaged. Another possible
option for such patients would be to dialyse them using CCPD instead of CAPD.
The main point is that with circumspect use, a simple model derived from
acceptable assumptions, can be used both to assess with reasonable accuracy,
long-term viability of the peritoneum and to examine the effects on patient
chemistries of such treatment options as frequency of exchange, volume of
exchange, and the CAPD versus CCPD option.
Predictions of creatinine blood levels can be made for variations in KBD and
592

residual clearance. The data in Table 9 for patient 10 show the relatively greater
effect of decreased residual clearance on creatinine levels compared to urea levels
at low KBD values.
It is noted that mass transfer studies and 24-hour clearance measurements gave
a KBD value for patient 10 of 4.4 ml/min with a residual renal clearance of 3.3 ml/
min as shown in Table 9.

For a decrease of -3mllmin in K BD, the creatinine blood level will increase by
approximately 40% if the residual renal clearance is 3.3 ml/min, while a 100%
increase will be observed if KBD remains constant and renal clearance drops to
zero. The large effect of residual renal function on blood levels is seen for
creatinine since at low K BD values the effect of a small amount of renal clearance is
large relative to that diffusive based dialysate clearance; the mean dialysate
clearance is considerably less than the maximal clearance of 4.4 ml/min (KBD ) and
the RRF value of 3.3 mllmin.
Tables 5-9 highlight the effects on solute blood level of changing generation
rates, residual clearance and dialysis regimen. The question remaining for clini-
cians is how to separate these effects from those caused by peritoneal membrane
KBD variations. One obvious way is to determine KBD for urea and creatinine as
outlined above. Patients in our study with urea KBDs of 14-35 ml/min and
creatinine KBDs of 4-20mllmin fit the normal criteria for 'adequate' dialysis (i.e.
eating well, feeling well). Another reason to do KBD studies is that periodic KBD
determination allows a continual assessment of peritoneal membrane function. If
this is not possible an alternative procedure is to determine whether or not urea
and creatinine KBDs are in the normal range. This can be done by measuring
dialysate and blood urea and creatinine levels at the end of dialysis, noting dialysis
time, and dialysate pre/post volumes. Equation (4) can be rearranged as follows

CD = 1- V Do (1- C Do e f KBD • tIVD (7)

CB VD CB

Table 9. Creatinine C B for patient 10 as a function of KBD and renal clearance variations (constant
daily generation rate)

KBD (mllmin) Predicted creatinine CB (mmol/L)

3.58 1.03
3 1.86 0.82
4" 1.49 0.73
6 1.26 0.68
10 1.01 0.60
15 0.84 0.55
Renal clearance (mllmin) o 3.3>

• Patient 10 estimated values.

 593

For given KBDS and dialysis timeslvolumes the CDIC B ratio can be calculated.
The ratio CDIC Bis independent of residual clearances and generation rates. Table
10 shows calculated ratios for varying dwell times and KBDS; it is equally applica-
ble both to urea and creatinine. Table 10 assumes CD = 0 and that for a 2-liter
Q

dialysis 350 ml of ultrafiltrate is removed; for more accurate results actual patient
volume data should be used.
If a patient's blood levels of urea or creatinine have recently risen without
explanation the test described will ascertain whether or not an abnormal per-
itoneum may be the problem. As long as the patient is removing sufficient water a
creatinine KBD above 3-4 mllmin and a urea K BD above 10-13 mllmin in our
experience is sufficient to allow 'adequate' dialysis. The underlying assumption is
that other higher molecular weight solutes will also be cleared sufficiently with
urea and creatinine KBDS at these levels; to date this seems to be the case.
Although this technique is open to error it is sufficient to determine whether or
not a patient is in the normal range, on the borderline or below normal with
respect to peritoneal membrane permeability and mass transfer area capacity.
Four patients who have undergone KBD determinations have on separate occa-
sions had their creatinine CDIC B ratios measured post dialysis. KBDS calculated
from the 2 techniques are shown in Table 11.

Table 10. CDICs ratio

Dwell time (hrs) KSD mllmin

2 4 6 8 10 15

4 0.23 0.31 0.44 0.55 0.64 0.71 0.83

6 0.27 0.38 0.55 0.67 0.76 0.83 0.93
8 0.31 0.44 0.64 0.76 0.84 0.90 0.97
10 0.35 0.50 0.71 0.83 0.90 0.94 0.98
12 0.38 0.55 0.76 0.88 0.93 0.97 0.99

Table 11. KSD comparison

Patient Creatinine KSD mllmin

CdCs method using Table 10 KSD determination

A 12 9-13
B 4 5-6
C 8 7-9
D 7 10-16
594

Patient D shows marked variation between the 2 methods, apart from error
due to the simplification used in Table 10; the fact that the CJCB ratio came from
overnight dwell dialysate may in part be responsible. (The KBD obtained at night
with the patient resting could be markedly different to that obtained during
daytime due to patient movement and eating.) However, both methods give KBDS
in the normal range, showing that in the cases presented this simple method
fulfills the requirements outlined above concerning decisions of adequacy of
dialysis. To gain most accuracy for urea KBDS, dwell times should be used that give
a CD/CB ratio <0.90 as, above this ratio errors greatly affect the results.
Finally, as noted previously, KBDS can also be estimated from daily urea and
creatinine generation/excretion rates and blood levels. This method has advan-
tages over 2-4-hour assessments as the results estimated represent 'daily' values.
The method is usually inaccurate where KBDS are large, such as for urea, as the
technique gives large changes in predicted K BD values for small measured CB urea
variations. However, even for urea this technique is accurate enough to deter-
mine whether KBDS are in the normal range. Table 12 gives KBDS calculated by this
method for 3 patients for whom daily clearances, residual renal function and
dialysate volumes were measured. KBD determinations were also done for these 3
patients and are shown in Table 12. Hence, as can be seen from the results, when
time is not available to determine KBD accurately, a good estimate can be
obtained from 24-hour dialysate and urine collections.

S. Conclusions

The causes of permanent changes to the solute mass transfer characteristics of the
peritoneal membrane are still uncertain. The reasons why a particular patient can
be treated by CAPD for over 5 years without apparent peritoneal deterioration
while others may suffer gross peritoneal permeability and/or mass transfer area

Table 12. KBD from daily clearances

Patient KBD ml/min

Urea Creatinine

2 2

E 4 6
F* 22±9 22±9 11 ±3 8±1
G 32 >30 18 21

1 = KBD determination; 2 = KBD from clearances.

• Mean ± SD given for 3 KBD determinations and 4 KBD estimations from daily clearance measure-
ments.
 595

changes, after less than 2 years dialysis, need further study. Another question to
be answered concerns whether or not resting 'damaged' peritoneums for an
extended period will have a reparative effect allowing patients to return to
CAPD.
While the answers to these questions are yet to be clarified there is the need to
determine if a given patient's peritoneum allows 'adequate' dialysis or is capable
of coping with a change in daily dialysis regimen. The model and tables presented
are a useful aid in such determinations. At the basest level, urea/creatinine
dialysate concentrations at the end of dialysis can be used to demonstrate if a
patient fits into the normal solute maximum clearance (KBD ) range, as deter-
mined by us in tests conducted on 40 CAPD patients. At a more complicated
level, the model presented can be used with the aid of a personal computer, or
even a hand-held calculator, to obtain more accurate KBD estimates to compare
and contrast the effects of dietary, residual renal clearance or daily dialysis
regimen variations on patients' blood solute levels.

Nomenclature

V DO, V D dialysate volume at beginning of dialysis (VDO) and time t(VD) (ML)
cDo, CD dialysate solute concentration at beginning of dialysis (C DO) and time t(CD ) (mmoI/L)
CB average blood solute concentration during dialysis (mmollL)
VD average dialysate volume during dialysis (ML)
G generation rate (mmol/day)
KR residual renal clearance (ml/min)
Vd, dialysate exchange volume for jth exchange (L)
Vu pooled urine volume (L)
Cd} pooled dialysate solute concentration for jth exchange (mmoIlL)
Cu pooled urine solute concentration (mmoIlL)
Wt, W, finallinitial patient weights (kg)
Ct, C, final/initial solute blood concentration (mmol/L)
V, solute body distribution volume fraction (-0.57 for urea and creatinine)
a time constant (min-I)

Appendix

Sample calculation using equations (4) and (6)

Assume first that a patient, when initially being trained for CAPD, has a urea KBD of 20 ml/min and
residual renal function of 2 ml/min. Assume further the patient's blood urea concentration has risen
sharply to 35 mmol/l from about 20 mmolll during a couple of weeks. It is uncertain whether this urea
level rise is due to dietary factors, residual renal function deterioration or variations in K BD • To assess
the cause of the rapid rise in blood urea concentration the following calculation can be undertaken:
(1) A 24-hour dialysate and urine collection is completed. The results are:
* Gure• = 250mmollday
* KR = 2mllmin (unchanged from previous tests)
596

* dwell times 3 x 4.5 hours, 1 x 9 hours

* average ultrafiltration volume 300 mls/exchange
e
(2) Start with an initial guess for the value of B in equation (4) of 20 mmol/l. It is known that V n° =
2L, Vn = 2.3L, and it is assumed that Vn "'" 2.2L and that CDo = O. The derived results are:
* CD (4.5 hours) = 18.5mmolll
* CD (9 hours) = 19.9mmolll
4
* L V",C"j = 173 mmollday
J:= 1

* KR X C B x 1.44 = 58mmollday
Therefore total daily urea excretion (G mea ) = 173 + 58 = 231 mmollday.
(3) Since Gealeulated (231 mmollday) is less than Gartual (250 mmollday) use a higher estimate for e
B at
22 mmol/l. Recalculation gives:
* Cn (4.5 hours) = 20.4mmol/l
* CD (9 hours) = 21. 9 mmolll
4
* =LI Vd,Cd, =
j
191 mmol/day

* KR X CB x 1.44 = 63 mmollday
Therefore total daily urea excretion (G m ,,) = 191 + 63 = 254 mmollday.
(4) Gcaleulated (254mmol/day) = Gactual (250mmollday) «5% difference) therefore the expected value
of C B , if KBD is still 20 mllmin, is 22 mmol/l.
The conclusion for the above example is that the high urea blood level of 35 mmol/l is due to a
decrease in KBD since the expected blood level is 22 mmolll for a KBD of 20 mllmin. An alternative
e
technique would be to set B at 35 mmolll and decrease the value of KBD in equation (4) until Gealcul"ed
= Gee,ual' The calculated K Bn is -3 mllmin - a significantly decreased K BD .

References

1. Randerson DR, Farrell PC: Mass transfer properties of the human peritoneum. ASAIO J 3: 140,
1980.
2. Randerson DR, Farrell PC: Long-term clearance variations in CAPO. In: RC Atkins, PC Farrell,
N. Thomson (eds), Peritoneal Dialysis. Churchill-Livingstone, Edinburgh, 1981, pp 22-29.
3. Garred LJ, Canaud B, Farrell PC: A simple kinetic model for assessing peritoneal mass transfer in
CAPO patients. ASAIO J 6: 131, 1983.
4. Dedrick RL, Flessner MF, Collins JM, Schultz JS: Is the peritoneum a membrane? ASAIO J 5: 1,
1982.
5. Scott OF, Marshall VC: Insertion and complications of Tenckhoff catheters - surgical aspects. In:
RC Atkins, N Thomson, PC Farrell (eds), Peritoneal Dialysis. Churchill-Livingstone, Edinburgh,
1981, pp 62-72.
6. Farrell PC, Garred LJ: Long-term studies on the human peritoneum. Proc 1st Int Course on
Peritoneal Dialysis, Vicenza, Italy, May 25-28,1982, pp 97-107.
7. Farrell PC, Randerson DR: Long-term nutritional and clearance status in CAPO patients. Con-
temporary Dialysis 2: 45,1980.
18. The USA CAPD Registry
Characteristics of participants and selected outcome measures
for the period January 1, 1981, through June 30, 1983

SETH M. STEINBERG, SIDNEY J. CUTLER, JOEL W . NOVAK and

KARL D. NOLPH

Preface

The National Institutes of Health have supported a CAPD Registry. Since there
is no USA Registry for patients undergoing all forms of chronic dialysis therapy,
the USA CAPD Registry was created to monitor certain outcome measures
during the rapid growth and development of this relatively newer form of
therapy. Although newer Registry reports will have been released by the time this
book is published, a copy of the report released in early 1984 is enclosed for
several reasons. First, the report represents the status of the Registry at the time
this book was prepared. This may clarify to some extent the state of affairs
impacting on the thinking of respective authors. Secondly, the report will provide
the reader with an example of the type of information that can be expected to
evolve from the Registry.
We are pleased to have this example of a Registry report included in this book.

Gladys Hirschman, M.D.

Director, Chronic Renal Disease Program,
National Institute of Arthritis, Diabetes &
Digestive & Kidney Diseases,
National Institutes of Health,
Bethesda, Maryland USA

Karl D. Nolph, M.D.

Director, Division of Nephrology,
Professor of Medicine, Department of Medicine,
University of Missouri Health Sciences Center,
VA Hospital & Dalton Research Center,
Clinical Coordinator of the CAPD Registry,
Columbia, Missouri, USA
598

1. Introduction

1.1. An overview of continuous ambulatory peritoneal dialysis

By 1984, it is expected that over 60000 Americans will be receiving dialysis for
end-stage renal disease. For the past 25 years, dialysis has mainly been accom-
plished with the use of 'artificial kidneys' which utilize the process of hemo-
dialysis. The hemodialysis patient usually needs treatment two to three times a
week for four to five hours per treatment.
An alternative maintenance therapy for the patient with end-stage renal dis-
ease is peritoneal dialysis. Here, the blood is cleansed of wastes across a living
membrane, the peritoneum. With peritoneal dialysis, the dialyzing fluid is infu-
sed into the peritoneal cavity.
Continuous ambulatory peritoneal dialysis (CAPD) was first described in 1976
[1] but was not widely used until 1978, when the technique was improved by the
introduction of plastic dialysate bags. CAPD is a closed system composed pri-
marily of the peritoneal cavity, a chronic in-dwelling catheter, 42-inch connecting
tubing, and a collapsible plastic dialysate bag. Dialyzing fluid is gravity-infused
into the patient's peritoneal cavity over a period of about ten minutes. The fluid is
usually allowed to dwell in the peritoneal cavity for four to eight hours, after
which the patient drains the peritoneal cavity by placing the dialysate bag lower
than the abdomen, allowing gravity to refill the bag. The filled bag is discarded, a
fresh bag of solution connected to the system, and the cycle repeated. Most
patients perform four exchanges per day. Primarily because of its convenience as
a home technique, it is expected that the use of CAPD will be increasing in the
future.
Continuous cyclic peritoneal dialysis (CCPD) is a recently introduced offshoot
of CAPD. With CCPD, three exchanges take place at night with an automatic
cycling machine. The dialysate remains in the abdomen during one IS-hour cycle.

1.2. The National CAPD Registry and the Registry report

In 1980, when the CAPD Advisory Committee of the National Institute of

Arthritis, Diabetes, and Digestive and Kidney Diseases recommended that the
Chronic Renal Disease Program initiate a registry of CAPD patients, there was
no single source of data that could be used to describe generalized experience
with CAPD. Such a program was needed to provide basic information regarding
the characteristics of patients receiving such therapy and the distribution of
centers offering it, and selected outcome measures such as treatment-related
complications. The National CAPD Registry was established by NIADDK to
assemble such information. In addition, the Registry collects and disseminates
information regarding the efficacy, safety, acceptability, and associated risks of
 599

peritoneal dialysis in different subgroups of patients. The Registry also registers

and follows patients using CCPD.
The Registry is managed by a Clinical Coordinating Center (CCC) under the
direction of Karl D. Nolph, M.D., at the University of Missouri-Columbia, and a
Data Coordinating Center (DCC) which is located at The EMMES Corporation
in Potomac, Maryland. The Clinical Coordinating Center is a liaison between the
Registry and the medical community, and specifically assists in new center
recruitment and information dissemination. The primary mission of the Data
Coordinating Center is to operate the data collection and processing system,
provide epidemiological and biostatistical support, and produce technical reports
such as this, in collaboration with the Clinical Coordinating Center.
Staff of the Kidney-Urology Branch (DKUHD), NIADDK, are also active
contributors to the project as are a Data Monitoring Committee, a CAPD
Advisory Committee, and an Executive Committee of user groups. (See Appen-
dix 1.)
At this writing, there are 259 clinical centers in the United States, including a
number of end-stage renal disease networks, that participate in the Registry and
that routinely submit data on their patients.
This is the first Registry report produced by the Data Coordinating Center .
located at The EMMES Corporation. It contains operational information con-
cerning the number of participating centers and the growth of the Registry over
time, as well as descriptions of patient characteristics and treatment outcomes.
Questions which the reader may have which are not answered by this report
may be addressed to:
Dr Seth M. Steinberg, Project Director
Data Coordinating Center, National CAPD Registry
The EMMES Corporation
11325 Seven Locks Road, Suite 214
Potomac, MD 20854, USA

2. Characteristics of participating centers

The National CAPD Registry is a multi-center effort. Although participation

from all centers offering CAPD is encouraged, it is not known to what extent the
participating centers and their patients are representative of all CAPD centers
and patients.
The Registry began collecting information on a pilot basis in January 1981. As
indicated in Figure l(a), only 15 centers registered patients and submitted follow-
up information in the first and second calendar quarters of 1981. In fact, only nine
of these original participants were performing the actual registration and follow-
up during the quarters mentioned, while the other six provided the necessary
information retrospectively.
 600

300
a
250

200

150

100

50
15
EiJ o
81-02 81-03 81-04 82·01 82-02 82·03 82·04 83-01 83-02
CALENDAR PERIOD
250
b
200

150

100

50
15 15 23
~
E:J ~
~ ~ o
•• 81·01 81·02 81·03 81·04 82·01 82·02 82·03 82-04 83-01 83·02
CALENDAR PERIOD
• Entry Into the Registry Program IS defined as submission of one or more patient follow·up repons
(Pallent Status Form) to addition to submiSsion of patient registration forms .
•• 81.01 refers to the first quaner of 1981 .
••• These figures refle<:t a lag In reponing; documents recelveO alter 9/30183 are not counted in this repan.

Figure 1. (a) Cumulative number of centers that entered the Registry program' , by calendar period.
(b) Number of centers that submitted patient status forms in each calendar period.

Figure l(a) details the growth of the Registry. By the last quarter of 1981,184
centers had entered the program. Since that time, the number of new centers
participating has been increasing, but at a much slower rate. By the end of the
second calendar quarter of 1983, 259 centers had provided patient follow-up
information . Since then, 28 additional centers indicated an intent to participate ,
and while some submitted patient registrations, no follow-up information had
been received from any of the 28 centers as of the end of September 1983, the cut-
off date for this report.
 601

Where Figure l(a) demonstrates the cumulative size of participation in the

Registry, Figure l(b) presents the actual number of centers that submitted patient
status forms for each quarter. Information for the two quarters of 1983 are
considered incomplete, due to a lag in submission of reports.
Twenty-nine centers that had been submitting patient follow-up reports are no
longer active in the program. The largest number of dropouts occurred after the
second quarter of 1982, at which time 10 centers ceased submitting follow-up
reports.
If we categorize centers on the basis of the numbet of patient status reports
submitted, we find that 'small centers', those that submitted fewer than 25 initial
status reports for 1981 and 1982, account for 70% of all centers in the Registry.
Twenty-four of the 183 'small centers' (13%) have dropped out of the program
since its beginning, while only 6% of the 'larger centers' have stopped sending in
follow-up data for their patients.
An analysis was conducted to determine whether the centers that failed to
continue participation had similar profiles with respect to last reported patient
status as centers that are continuing in the program. The relative frequencies of
the last reported status of patients from centers that continued and from those
which failed to continue participation were very similar.

3. Number and types of patients followed, and data currency

The Registry is designed to follow both CAPD and CCPD patients. Registrations
began in January 1981 and those received at the Data Coordinating Center
through the end of September 1983 are included in this report. As Figure 2 shows,
the Registry has now received registration forms for a total of 8075 patients -
7734 on CAPD and 341 on CCPD. At the end of 1982, 5725 patients had been
registered, indicating an average growth of about 260 new patient registrations
per month during 1983: (8075-5725) divided by 9 months. CCPD patients
accounted for 7.6% of all patients registered in 1983, whereas only 1.5% of
registrations were for CCPD patients in 1981.
Table 1 presents data on patient registrations and related follow-ups. Over 12%
of patients registered through March 1983 have never had a status report submit-
ted, and, of 8 075 total patients ever registered, follow-up information is currently
available on 6706. A comparison of the cohorts of patients registered in 1981,
1982, and the first quarter of 1983 reveals a marked decrease in the percentage of
registered patients with no follow-up whatsoever - from 19.7% to 10.2% to 4.1 %.
Patients participating in the Registry were found to vary widely in their
experience on CAPD or CCPD and whether they received prior ESRD therapy.
It was felt that patients with or without experience on CAPD or CCPD prior to
being followed by the Registry, and patients with or without experience using
other therapies, should be analyzed separately, as these previous treatment
602

NUMBER OF PATIENTS
8,000 ---------------.:...:..:..~

6,000 - - - - - - - - - - - - - . ,

4,000 - - - - - - - - - - /

2,000-----~

1981 1982
Figure 2. Cumulative number of patients registered by calendar period according to type of dialysis.

Table 1. Number of patients registered and number and percent with no follow-up

Calendar period Number registered Number without Percent without

follow-up follow-up

1
J
1981 2290 451 19.7
1982 3435 349 10.2 12.7
1982-Ql 846 35 4.1
1983-Q2 783 140a 17.9"
1983-Q3 721 394" 54.6"

Total 8075 1369 17.0%

a These figures reflect a lag in reporting.

 603

experiences may affect outcomes. To this end, a classification scheme was de-
vised to categorize patients with respect to their CAPD or CCPD experience and
prior therapy for end-stage renal disease (ESRD) at the time of their registration.
Briefly, the classification scheme is as follows:

No other Prior other

ESRD therapy ESRD therapy

New to CAPD/CCPD at time of registration Class lA Class 1B

Prior experience with CAPD/CCPD at time of registration Class 2A Class 2B

Figure 3 presents the breakdown in terms of numbers of patients on each therapy,

according to class of case and type of therapy. As some of the 6706 patients had
incomplete information, it was not possible to classify every patient into a class of
case. A total of 6656 were categorized into one of the four classes, and these
patients will hereafter be considered 'available for analysis'. Approximately two
thirds of the patients were new to CAPD or CCPD at the time of registration (IA
or lB) and approximately two thirds of the patients had other ESRD therapy
before beginning CAPD or CCPD (lB or 2B).
Over time, there have been changes in the distribution of patients in the various
classes of case. While 58% of patients registered in 1981 and 1982 are class lA or
class lB patients, 87% of patients thus far registered in 1983 are class lA or class
lB. And, while 30% of patients registered in 1981 and 1982 had no ESRD therapy
prior to CAPD or CCPD, the corresponding proportion is 37% for the first three
quarters of 1983. Thus, we conclude that the Registry is registering more patients
.at the time they start CAPD/CCPD therapy and, similarly, that more patients are
being registered with CAPD or CCPD as their first therapy.
Figure 4 illustrates graphically the pattern of initial status reports received over
time. As with registrations, a significant number of first status reports were
received in 1981, and since then a smaller but relatively steady number of new
patients have been added to follow-up.
By definition, patients in classes lA and lB started on CAPD/CCPD at time of
registration. In contrast, patients in classes 2A and 2B started on CAPD/CCPD
some time prior to registration. Clinics entering the program are requested to
register all patients on CAPD or CCPD; most of these patients would then be
class 2A or 2B patients. The distribution of times elapsed between the start on
CAPD/CCPD and entry into the Registry for patients in classes 2A and 2B
combined is presented in Table 2. As the table shows, class 2A and 2B patients
were generally registered within one year of their time of starting therapy;
relatively few patients were registered two years after beginning therapy.
As indicated above, 29 centers dropped out of active participation in the
Registry. This has resulted in incomplete follow-up information on some pa-
604

tients. Of the 6656 patients available for analysis, 4425 were last reported to be
continuing on CAPD or CCPD. However, the last patient status report received
for 435 of the 4425 patients (9.8%) reflected information prior to 1983. The
predominant proportion of the 435 patients with incomplete follow-up had been
registered by clinics that dropped out of the program.
The volume of patient status reports received, pertaining to each quarter, is

CAPO - . - - - - 4 1 7 6 - - - - - l - - - -

2.000

1,000

o
lA 18 2A 28

CCPO I ~·----161----~-----73 ----~'50

108
100

50

• Pallents witll one or more Status Repons.

CI... 01 C41se Dncr1ptlon
tA New to CAPO/CCPO at time of reglstrallon; no otller ESRO tllerapy.
1B New to CAPO/CCPO at time of registration: pnor other ESRO therapy.
lC Continuing on CAPO/CCPO al time of reglstraUon: no other ESRO therapy.
2B Continuing on CAPO/CCPO al time of reglstrallon; pnor other ESRO therapy.

Figure 3. Number of patients available for analysis' by class of case ' * according to type of dialysis.
 605

presented in Figure 5. From this chart it·is evident that the number of patients
followed is increasing over time. In fact, the number of patients followed cur-
rently is approximately twice what it was at the end of 1981. The increase is slightly
more rapid for CCPD than for CAPO patients , although many more CAPO
patients than CCPD patients are being followed.
Table 3 is presented in order to identify the number of patients followed by

2.000
CAPO
1756

1.500

1.000

1[J·
«
301

L .,'::'
:~.

~ I E£)
171

81.01 81.02 81.03 8"1.04 82.01 82.02 82.03 82.04 83-01

CALENDAR PERIOD

cepo 44
41
40

2,9" 1
;~ I
~20

~ ~
81.01
=
81.02 81.03 81-04 82·01 82·02 82·03
CALENDAR PERIOD
82·04 83.01 83·02
lo

Figure 4. Number of patients who became available for analysis * in each calendar period according to
type of dialysis .
606

various sizes of centers. The centers were placed into five separate size classifica-
tions depending on how many initial patient status forms were submitted on
patients for 1981 and 1982. During 1981 and 1982, 171 centers initiated follow-up on
one to 24 patients. This group of centers accounts for two-thirds of all clinics. As
Table 3 shows, while the 'small' clinics account for 183 (71%) of the 259 clinics in

Table 2. Number of class 2A and 2B CAPD/CCPD patients by calendar period began CAPD/CCPD
according to calendar period patients became available for analysis

CAPDI
CCPD Became available for analysis
begun 81-Q1 81-Q2 81-03 81-Q4 82-01 82-Q2 82-Q3 82-Q4 83-01 83-02 Total

Before
1981 197 63 539 108 77 19 10 29 10 1053
81-01 27 265 41 25 8 7 14 2 389
81-Q2 330 40 29 16 10 13 2 440
81-03 52 46 15 14 15 3 145
81-Q4 39 18 14 12 8 91
82-01 33 18 21 10 82
82-02 33 31 9 73
82-Q3 24 10 34
82-Q4 12 12

Total 197 90 1134 241 216 109 106 159 66 2319

Table 3. Number of patients available for analysis by size of center"

Size of center Total

1 2 3 4 5

Clinics in each size class 12 171 47 22 7 259

(4.6%) (66.0%) (18.1%) (8.5%) (2.7%) (100%)
Patients available for analysis 158b 2196 1977 1548 777 6656c
(2.4% ) (33.0%) (29.7%) (23.3%) (11.7%) (100%)

" Based on initial patient status forms for 1981 and 1982.
Size class Definition
No initial status forms in 1981 and 1982
2 1-24 initial status forms in 1981 and 1982
3 25-49 initial status forms in 1981 and 1982
4 50-74 initial status forms in 1981 and 1982
5 75+ initial status forms in 1981 and 1982
b These patients were initially followed in 1983. The sizes of the clinics cannot be deduced from such a

figure.
c 6656 = total patients available for analysis.
 607

the program, they only account for approximately 2350 (35%) of the 6656
patients.
The 12 centers that submitted no initial status reports during 1981 and 1982 are
centers that joined the program in 1983 .

4066"
CAPO

3:.Q1 81·Q2 81·Q3 81·Q4 82·01 82·02 82·03 82·04 83·01 83·C2
CALENDAR PERIOD

CCPO 159"

140 150

109
~oo

57

~
.. 50
"

.. j('

~ y
~.:

y..
2 ..;.... 6 1 o
81·01 81·02 81·03 81·Q4 82·01 82·02 82·03 82·04 83-01 83-02
CALEN DAR PERIOD

• These :I9ures rerlee! a lag In reoonong.

Figure 5. Number of patient status reports received by calendar period according to type of dialysis.
608

4. Patient characteristics

As indicated in the preceding section, 6656 patients - 6422 CAPD patients and
234 CCPD patients - were considered to be available for analysis in this report,
there being sufficient information available to classify them into one of four
'classes of case'. This section will categorize patients within each class of case by
their age, sex, race, and type of primary renal disease.

4.1. Age and sex distribution

In Table 4, the age and sex distribution of patients available for analysis, by type
of dialysis, is presented. Approximately 50% of patients on CAPD were under 53
years of age at entry to follow-up. Very few were over 80 years old, and only
about 2% were under 10 years of age. Forty-four percent of CAPD patients are
women; 45% of CCPD patients are women. The ratio of men to women on
CAPD increases as age increases: from 1.0 at ages 21-30 to 1.5 at ages 71-80.
Approximately the same pattern holds true for CCPD patients. A much greater
proportion of CCPD patients than CAPD patients are under 20 years of age.

4.2. Age and race distribution

Table 5 presents the age and race distribution of patients available for analysis by
type of dialysis. Among CAPD patients, 77% are white, 16% are black, and 7%
are classified into other categories. Approximately the same proportion of non-
whites is seen for patients on CCPD. A greater proportion of blacks than whites
(47% vs 37.2 %) are between 40 and 60 years of age. The distribution for CCPD
patients demonstrates some greater racial disparity among the age groups, but the
small numbers involved may make these differences meaningless.

4.3. Primary renal disease

Finally, the number of patients available for analysis by type of primary renal
disease is presented in Table 6. The three most commonly identified renal
diseases among CAPD and CCPD patients in the Registry are chronic glomeru-
lonephritis, diabetic glomerulosclerosis, and hypertensive renal disease. Con-
sidering that the overall proportion of women is approximately 44%, it is note-
worthy that a reversal in these proportions occurs in some cases. For example,
women account for about 67% of the diagnosed cases of chronic pyelonephritis,
and 72% of diagnosed cases of systemic immunological disease with renal invol-
vement. Almost 9% of patients had a disease reported as 'other'.
 609

Table 4. Age and sex distribution of patients available for analysis by type of dialysis

CAPD

Age Male Female Total

group in
years No. % Cum. % No % Cum. % No. % Cum. %

:s1O 62 1.7 1.7 53 1.9 1.9 115 1.8 1.8

11-20 111 3.1 4.8 103 3.7 5.8 214 3.4 5.2
21-30 286 8.0 12.8 282 10.1 15.9 568 8.9 14.1
31-40 541 15.2 28.0 459 16.4 32.0 1000 15.7 29.8
41-50 595 16.7 44.7 387 13.8 45.8 982 15.4 45.3
51-60 806 22.6 67.3 699 25.0 70.8 1505 23.7 69.0
61-70 823 23.1 90.5 596 21.3 92.1 1419 22.3 91.3
71-80 324 9.1 99.6 213 7.6 99.7 537 8.4 99.7
81-90 13 0.4 100.0 8 0.3 100.0 21 0.3 100.0

Total 3561 (100%) 2800 (100%) 6361 (100%)

% by
sex 56% 44%

CCPD

Age Male Female Total

group in
years No. % Cum. % No. % Cum. % No. % Cum. %

:s1O 11 8.7 8.7 4 3.8 3.8 15 6.5 6.5

11-20 13 10.2 18.9 21 20.0 23.8 34 14.7 21.2
21-30 7 5.5 24.4 9 8.6 32.4 16 6.9 28.1
31-40 19 15.0 39.4 10 9.5 41.9 29 12.5 40.6
41-50 12 9.4 48.8 11 10.5 52.4 23 9.9 50.5
51-60 21 16.5 65.4 20 19.0 71.4 41 17.7 68.2
61-70 32 25.2 90.6 26 24.8 96.2 58 25.0 93.1
71-80 12 9.4 100.0 4 3.8 100.0 16 6.9 100.0

Total 127 (100%) 105 (100%) 232 100%

% by
sex 54.7% 45.3%

Note: information not available for 63 patients.

610

Table 5. Age and race distribution of patients available of analysis by type of dialysis

CAPD

Age group White Black Other Total

in years
No. % Cum. % No. % Cum. % No. % Cum. % No. % Cum. %

:510 89 1.8 1.8 12 1.2 1.2 14 3.3 3.3 115 1.8 1.8
11-20 151 3.1 4.9 36 3.5 4.7 27 6.5 9.8 214 3.4 5.2
21-30 438 9.0 13.9 77 7.4 12.1 51 12.2 22.0 566 8.9 14.1
31-40 764 15.6 29.5 178 17.1 29.2 58 13.9 35.9 1000 15.8 29.9
41-50 696 14.2 43.7 205 19.7 48.9 75 17.9 53.8 976 15.4 45.3
51-60 1125 23.0 66.7 283 27.3 76.2 95 22.7 76.5 1503 23.7 69.0
61-70 1158 23.7 90.4 191 18.4 94.6 70 16.7 93.3 1419 22.3 91.3
71-80 457 9.3 99.5 53 5.1 99.7 26 6.2 99.5 536 8.4 99.7
81-90 16 0.3 100.0 3 0.3 100.0 2 0.5 100.0 21 0.3 100.0

Total 4893 (100%) 1038 (100%) 418 (100%) 6349 (100%)

% by
race 77.1% 16.3% 6.6%

CCPD

Age White Black Other Total

group
in years No. % Cum. % No. % Cum. % No. % Cum. % No. % Cum. "<0

:510 8 4.6 4.6 2 6.7 6.7 5 18.5 18.5 15 6.5 6.5

11-20 24 13.7 18.3 3 10.0 16.7 7 25.9 44.4 34 14.7 21.2
21-30 15 8.6 26.9 3.3 20.0 0 0 44.4 16 6.9 28.1
31-40 19 10.9 37.8 8 26.7 46.7 2 7.4 51.8 29 12.5 40.6
41-50 21 12.0 49.8 3.3 50.0 1 3.7 55.5 23 9.9 50.5
51-60 31 17.7 67.5 5 16.7 66.7 5 18.5 74.1 41 17.7 68.1
61-70 44 25.1 92.6 7 23.3 90.0 7 25.9 100.0 58 25.0 93.1
71-80 13 7.4 100.0 3 10.0 100.0 0 0 100.0 16 6.9 100.0

Total 175 (100%) 30 (100%) 27 100% 232 (100%)

% by
race 75.5% 12.9% 11.6%

Note: information not available for 75 patients.

 611

Table 6. Patients available for analysis by type of primary renal disease, according to type of dialysis'

Renal disease CAPD CCPD

M F Total % M F Total %

Chronic glomerulonephritis 843 545 1388 (22.0) 20 17 37 (15.9)

Diabetic glomerulosclerosis 719 557 1276 (20.2) 32 30 62 (26.6)
Hypertensive renal disease 612 424 1036 (16.4) 8 10 18 (7.7)
Polycystic kidney(s) 243 249 492 (7.8) 9 5 14 (6.0)
Chronic pyelonephritis 79 160 239 (3.8) 3 4 (1.7)
Systemic immunological disease with
renal involvement 53 135 188 (3.0) 7 5 12 (5.2)
Interstitial nephritis 81 101 182 (2.9) 3 4 7 (3.0)
Obstructive uropathy 118 38 156 (2.5) 6 5 11 (4.7)
Rapidly progressing
glomerulonephritis 95 53 148 (2.3) 6 3 9 (3.9)
Familial nephritis 47 37 84 (1.3) 2 2 4 (1.7)
Aplastic/hypoplastic kidney(s) 23 14 37 (0.6) 3 3 6 (2.6)
Stone-forming renal disease 16 20 36 (0.6) 0 1 (0.4)
Amyloidosis with renal involvement 22 14 36 (0.6) 1 2 (0.9)
Renal infarct, secondary to vascular
occlusion 21 9 30 (0.5) 1 1 2 (0.9)
Nephrectomy, secondary to cancer 19 9 28 (0.4) 0 0 0 (0.0)
Gouty nephropathy 11 12 (0.2) . 1 0 (0.4)
Bilateral cortical necrosis 3 4 7 (0.1) 0 (0.4)
Other 291 256 547 (8.7) 16 11 27 (11.6)
Unknown 243 154 397 (6.3) 10 5 15 (6.4)

Total 3539 2780 6319 100% 126 107 233 100%

• Information not available for 104 patients.

5. Treatment outcomes

The foregoing discussion described the characteristics of all registered patients

classified as 'available for analysis'. In this section, outcomes of treatments are
presented for these patients, with each class of case being discussed separately.
Since patients in class 2A and 2B were registered after an unspecified length of
time on CAPD/CCPD therapy, those patients who are registered need to be
considered as 'survivors' for a larger population of patients who started therapy in
prior years, an unknown number of whom did not survive therapy long enough to
be registered.
(i12

5.1. Last reported status

Table 7 presents the last reported status on CAPD patients according to the
number of months on therapy by class of case. As the exact date on which the last
status was reported is uncertain, the length of time on CAPD was approximated
by using the dates on which CAPD was begun and midpoint of the last quarter
reported. The data in Table 7 indicate that a large proportion of deaths occurred

Table 7. Number of patients by last reported status according to months on CAPD and class of case

Class lA

Last reported status Months completed Total

0-5 6-11 12-17 18-23 24+

CAPD patient at this center 491 263 154 71 12 991

CAPD patient, transferred to another 25 8 3 0 0 36
center
Transferred to hemodialysis 84 36 16 5 0 141
Transferred to intermittent peritoneal 3 4 2 1 0 10
dialysis
Not on dialysis (no return of kidney 6 3 0 0 0 9
function)
Kidney function returned, not on dialysis 13 7 0 0 0 20
Received transplant, not on dialysis 44 29 14 0 0 87
Died 89 53 14 6 2 164

Total 755 403 203 83 14 1458

Class 1B

Last reported status Months completed Total

0-5 6-11 12-17 18-23 24+

CAPD patient at this center 732 493 341 172 27 1765

CAPD patient, transferred to another 48 21 7 2 79
center
Transferred to hemodialysis 218 123 49 19 3 412
Transferred to intermittent peritoneal 24 9 2 2 0 37
dialysis
Not on dialysis (no return of kidney 6 3 0 0 10
function)
Kidney function retuned, not on dialysis 15 5 2 0 0 22
Received transplant, not on dialysis 67 35 14 1 0 117
Died 142 71 37 9 260

Total 1252 760 453 205 32 2702

 613

Table 7. (Continued)

Class 2A

Last reported status Months completed Total

0-5 6-11 12-17 18--23 24+

CAPD patient at this center 27 32 40 70 179 348

CAPD patient, transferred to 5 6 3 5 20
another center
Transferred to hemodialysis 4 18 16 20 27 85
Transferred to intermittent 2 2 1 7
peritoneal dialysis
Not on dialysis (no return of 0 0 0 0 0 0
kidney function)
Kidney function returned, 0 0 0 2
not on dialysis
Received transplant, not on 2 8 9 6 6 31
dialysis
Died 4 24 23 20 41 112

Total 39 89 96 122 259 605

Class 2B

Last reported status Months completed Total

0--5 6-11 12-17 18--23 24+

CAPD patient at this center 8 77 98 202 559 944

CAPD patient, transferred to 11 12 16 12 52
another center
Transferred to hemodialysis 8 58 74 79 91 310
Transferred to intermittent 2 7 10 4 8 31
peritoneal dialysis
Not on dialysis (no return of 0 0 3
kidney function)
Kidney function returned, 0 3 0 1 5
not on dialysis
Received transplant, not on 19 16 12 11 59
dialysis
Died 3 36 58 58 75 230

Total 23 212 270 372 757 1634

614

early in the therapeutic regimen, and many patients were transferred off the
CAPD program when they had been on CAPD for less than six months. The
numbers for the cla'ss lA and 1B patients on CAPD longer than 18 months are
small because these patients were entered prior to the last quarter of 1981, before
the majority of centers began to participate in the Registry. The distribution of
last reported status as a function of time on CAPD does not vary greatly between
class lA and class 1B patients.
The status profile. of Class 2A patients with under six months of therapy
resembles that of class lA patients with less than six months of therapy. However,
there is a tendency for a greater percentage of class 2A and 2B patients to have
transferred off CAPD or died than for class lA and 1B patients to have trans-
ferred off or died. This is explained in part by the much higher proportion of class
2A and 2B patients completing more than two years of therapy through the date
of last report.

5.2. Reasons for leaving CAPDlCCPD

The reasons for leaving CAPD or CCPD are available for patients who termin-
ated CAPD or CCPD and transferred to hemodialysis, or intermittent peritoneal
dialysis (IPD), or who were removed from dialysis with no return of kidney
function. Patients are considered to be transferred to another therapy if there is
no intention of returning to CAPD or CCPD in the near future, or if they have
been off therapy for more than 28 consecutive days as of the end of the report
period. These data are presented in Table 8, according to the reasons they
terminated CAPD or CCPD, and class of case. Approximately one-fourth of
patients terminated CAPD or CCPD because of peritonitis (reason 3), while half
ofthe class lA patients left for 'other medically indicated reasons' (reason 2). A
greater proportion of class 1B patients (with prior ESRD therapy) than class lA
patients left CAPD on the basis of patient or family choice. Approximately 10%
left CAPD because of its inability to meet fluid or biochemical standards; 21 % of
CCPD patients (6 of 29) left CCPD for that reason. The proportions of class 2A
and 2B CAPD patients leaving for various reasons are similar to those of class lA
and 1B patients.

5.3. Complications

There are three major types of complications which are reported to the Registry:
peritonitis, exit site and tunnel infections, and catheter replacements. Only
complications occurring to class lA and 1B patients will be reported. This is
because the Registry has no knowledge of the extent of problems occurring to
class 2A and 2B patients prior to registration. For Class lA and 1B patients, a
 615

Table 8. Number of patients leaving CAPD or CCPD by reason for terminationa according to last
reported status, class of case, and type of dialysis

Class Reason for termination Total

2 3 4 5 6

CAPD
Class 1A
Transferred to hemodialysis 12 72 34 17 5 141
Transferred to intermittent peritoneal 0 5 2 3 0 0 10
dialysis
Not on dialysis (no return of kidney 2 0 5 0 9
function)
Class 1B
Transferred to hemodialysis 45 155 107 72 6 27 412
Transferred to intermittent peritoneal 16 9 9 2 0 37
dialysis
Not on dialysis (no return of kidney 2 0 4 0 3 10
function)
Class 2A
Transferred to hemodialysis 12 30 29 10 1 3 85
Transferred to intermittent peritoneal 0 3 0 2 7
dialysis
Not on dialysis (no return of kidney 0 0 0 0 0 0 0
function)
Class 2B
Transferred to hemodialysis 31 130 97 33 3 16 310
Transferred to intermittent peritoneal 5 9 14 0 2 31
dialysis
Not on dialysis (no return of kidney 0 0 0 3 0 0 3
function)
CCPD
All classes
Transferred to hemodialysis 6 3 5 3 0 0 17
Transferred to intermittent peritoneal 0 2 7 0 2 12
dialysis
Not on dialysis (no return of kidney 0 0 0 0 0 0 0
function)
Total 114 429 298 169 14 60 1084
Overall percent 10.5% 39.6% 27.5% 15.6% 1.3% 5.5% (100%)

a Reasons for termination are:

1 = CAPD/CCPD not able to meet fluid/biochemical standards;
2 = other medically indicated reasons;
3 = non-compliance with technique/excessive peritonitis;
4 = patient/family choice;
5 = socio-economic factors (e.g. distance to center too great, unsuitable home environment, etc.);
6 = other or unknown.
616

complete record of complications since beginning therapy is theoretically avail-

able for analysis. In Table 9, the patients on CAPD are classified by class of case,
and then the number of episodes each patient has experienced is cross-classified
with the length of time on CAPD. Many patients did not experience any per-
itonitis. Generally, as patients are on CAPD longer, the number of episodes
increases. Several patients reported more than 10 episodes each. Over 70% of
class 1A patients, and 90% of class 1B patients, with more than two years on
CAPD, have experienced at least one episode of peritonitis. The data for class 1B
demonstrate that multiple episodes of peritonitis are not uncommon, as 60 of 690
of these patients (8.7%) with more than one year on CAPD have had six or more
episodes of peritonitis.
Another major problem is exit site or tunnel infections. As these are difficult to
identify separately in the usual clinical setting, their occurrence was classified into
one type of infection. Table 10 is presented in an identical manner to Table 9. It is
clear that most followed patients do not have an exit site or tunnel infection
during their first year on CAPD, but that nearly half of all class 1A or class 1B
patients eventually have one or more exit site/tunnel infections if they are on
CAPD for one year or longer. Class 1A and 1B patients do not greatly differ in
their distributions for these types of infections.
Finally, catheter replacements are detailed in Table 11. Approximately 85% of
patients with one year of experience on CAPD will still be using their original
catheter; about half have the original catheter after two years. The probability of
requiring three or more catheter replacements is quite small. Catheter replace-
ments occur with approximately equal frequency for the patients with and with-
out prior ESRD therapy.

5.4. Complication rates

A convenient way to summarize events whose frequency is a function of time is in

terms of rates of occurrence per patient-years under observation. This procedure
facilitates comparison of various risks with one another and comparison with
other sets of data. However, meaningful comparisons with other reports are
possible only if the methods for calculating the rates are the same. In this report,
for class 1A and 1B patients, person-years under observation account for the total
time elapsed from the date of entry onto CAPD or CCPD to the last report,
indicating that the patient was either continuing on CAPD/CCPD or that this
type of dialysis had been terminated. No deduction is made for periods during
which patients may be temporarily off CAPD/CCPD. The date of the last contact
is estimated as the midpoint of the last calendar quarter for which a patient status
report was received. For class 2A and 2B patients, person-years under observa-
tion is considered to extend from the beginning of the first quarter from which a
status report was submitted until the midpoint of the last quarter with follow-up
information. Thus their rates are actually rates while under observation.
Table 9. Number of patients by number of peritonitis episodes experienced according to class of case and number of months on CAPD

Months Number of episodes Total

on CAPD
0 1 2 3 4 5 6+

Class 1A
<3 346 (73.6) 93 (19.8) 20 (4.3) 7 (1.5) 2 (0.4) o (0) 2 (0.4) 470 (100%)
3-5 176 (62.0) 62 (21.8) 31 (10.9) 11 (3.9) 2 (0.7) 1 (0.4) 2 (0.7) 284 (100%)
6-8 107 (46.7) 56 (24.5) 35 (15.3) 16 (7.0) 7 (3.1) 2 (0.9) 6 (2.6) 229 (100%)
9-11 79 (45.1) 47 (26.9) 20 (11.4) 15 (8.6) 4 (2.3) 3 (1.7) 7 (4.0) 175 (100%)
12-14 39 (31.5) 27 (21.8) 28 (22.6) 14 (11.3) 8 (6.5) 5 (4.0) 3 (2.4) 124 (100%)
15-17 19 (23.2) 14 (17.1) 19 (23.2) 10 (12.2) 9 (11.0) 4 (4.9) 7 (9.8) 82 (100%)
18-20 18 (31.6) 10 (17.5) 13 (22.8) 10 (17.5) 2 (3.5) 3 (5.3) 1 (1.8) 57 (100%)
21-23 10 (38.5) 4 (15.4) 4 (15.4) 3 (11.5) 3 (11.5) 1 (3.9) 1 (3.9) 26 (100%)
24+ 4 (28.5) 3 (21.4) 2 (14.3) 1 (7.1) 2 (14.3) 1 (7.1) 1 (7.1) 14 (100%)

Class 1B
<3 553 (76.2) 125 (17.2) 28 (3.9) 13 (1.8) 6 (0.8) 1 (0.1) o (0) 726 (100%)
3-5 306 (57.2) 132 (24.7) 56 (10.5) 22 (4.1) 10 (1.9) 5 (10.9) 4 (0.7) 535 (100%)
6-8 181 (43.6) 108 (26.0) 58 (14.0) 33 (7.9) 18 (4.3) 6 (1.5) 11 (2.9) 415 (100%)
9-11 121 (34.8) 82 (23.6) 59 (17.0) 37 (10.6) 23 (6.6) 13 (3.7) 13 (3.7) 348 (100%)
12-14 73 (27.1) 71 (26.4) 51 (19.0) 26 (9.7) 23 (8.6) 11 (4.1) 14 (5.1) 269 (100%)
15-17 46 (25.0) 45 (24.5) 30 (16.3) 24 (13.0) 13 (7.1) 10 (5.4) 16 (8.8) 184 (100%)
18-20 37 (26.4) 38 (27.1) 19 (13.6) 11 (7.9) 13 (9.3) 10 (7.1) 12 (8.6) 140 (100%)
21-23 16 (24.6) 11 (16.9) 11 (16.9) 6 (9.2) 6 (9.2) 5 (7.7) 10 (15.4) 65 (100%)
24+ 3 (9.4) 5 (15.6) 7 (21.9) 3 (9.4) 6 (18.8) o (0) 8 (25.0) 32 (100%)

Note: fi~ures within parentheses are percentages.

0\
.......
-....I
 0\
......
00
Table 10. Number of patients by number of exit site/tunnel infections experienced according to class of case and number of months on CAPD

Months Number of episodes Total

on CAPD
0 2 3 4 5+

Class 1A
<3 394 (83.8) 63 (13.5) 11 (2.4) 2 (4.3) o (0) o (0) 470 (100%)
3-5 215 (75.7) 43 (15.1) 23 (8.1) 3 (Ll) o (0) o (0) 284 (100%)
6-8 153 (66.8) 45 (19.7) 20 (8.7) 6 (2.6) 1 (0.4) 4 (1.8) 229 (100%)
9-11 118 (67.4) 27 (15.4) 8 (4.6) 6 (3.4) 7 (4.0) 9 (5.2) 175 (100%)
12-14 77 (62.1) 25 (20.2) 11 (8.9) 3 (2.4) 4 (3.2) 4 (3.2) 124 (100%)
15-17 42 (51.2) 17 (20.2) 15 (18.3) 2 (2.4) 4 (4.9) 2 (2.4) 82 (100%)
18--20 27 (47.4) 13 (22.8) 9 (15.8) 4 (7.0) 1 (1.8) 2 (3.5) 57 (100%)
21-23 14 (53.9) 7 (26.9) 1 (3.9) 2 (7.7) 1 (3.9) 1 (3.9) 26 (100%)
24+ 6 (42.9) 4 (28.6) 2 (14.3) o (0) 2 (14.3) o (0) 14 (100%)

Class 1B
<3 614 (84.6) 101 (13.9) 9 (1.2) 2 (0.3) o (0) 0(0) 726 (100%)
3-5 408 (76.3) 88 (16.5) 31 (5.8) 5 (0.9) 2 (0.4) 3 (0.5) 535 (100%)
6-8 280 (67.5) 83 (20.0) 34 (8.2) 12 (2.9) 3 (0.7) 3 (0.7) 415 (100%)
9-11 223 (64.1) 69 (19.8) 33 (9.5) 9 (2.6) 7 (2.0) 7 (2.0) 348 (100%)
12-14 146 (54.3) 74 (27.5) 25 (9.3) 14 (5.2) 4 (1.5) 6 (2.2) 269 (100%)
15-17 107 (58.2) 34 (18.5) 17 (9.2) 14 (7.6) 5 (2.7) 7 (3.8) 184 (100%)
18--20 75 (53.6) 33 (23.6) 17 (12.1) 5 (3.6) 5 (3.6) 5 (3.6) 140 (100%)
21-23 35 (53.9) 10 (15.4) 7 (10.8) 7 (10.8) 1 (1.5) 5 (7.7) 65 (100%)
24+ 13 (40.6) 5 (15.6) 6 (18.8) 3 (9.4) 3 (9.4) 2 (6.3) 32 (100%)

Note: figures within parentheses are percentages.

Table 11. Number of patients by number of replacement catheters according to class of case and number of months on CAPD

Months Number of replacement catheters Total

on CAPD
0 2 3 4+

Class 1A
<3 428 (91.1) 35 (7.4) 6 (1.3) 1 (0.2) o (0) 470 (100%)
3-5 253 (89.1) 23 (8.1) 5 (1.7) 3 (Ll) o (0) 284 (100%)
6-8 193 (84.3) 25 (10.9) 8 (3.5) 1 (0.5) 2 (0.9) 229 (100%)
9-11 147 (84.0) 22 (12.6) 5 (2.9) 0(0) 1 (0.6) 175 (100%)
12-14 96 (77.4) 15 (12.1) 10 (8.1) 2 (1.6) 1 (0.8) 124 (100%)
15-17 56 (68.3) 13 (15.9) 11 (13.4) 1 (1.2) 1 (1.2) 82 (100%)
18-20 45 (79.0) 9 (15.8) 1 (1.8) 2 (3.5) o (0) 57 (100%)
21-23 21 (80.8) 4 (15.4) 1 (3.9) 0(0) o (0) 26 (100%)
24+ 7 (50.0) 4 (28.6) 3 (21.4) 0(0) o (0) 14 (100%)

Class 1B
<3 665 (91.6) 47 (6.5) 10 (1.4) 3 (4.1) 1 (1.4) 726 (100%)
3-5 478 (89.3) 47 (8.8) 8 (1.5) 2 (0.4) o (0) 535 (100%)
6-8 348 (83.9) 57 (13.7) 8 (1.9) 1 (0.2) 1 (0.2) 415 (100%)
9-11 291 (83.6) 45 (12.9) 12 (3.5) 0(0) o (0) 348 (100%)
12-14 209 (77.7) 49 (18.2) 8 (3.0) 3 (1.1) o (0) 269 (100%)
15-17 143 (77.7) 34 (18.5) 6 (3.3) 0(0) 1 (0.5) 184 (100%)
18-20 106 (75.7) 26 (18.6) 6 (4.3) 2 (1.4) o (0) 140 (100%)
21-23 48 (73.9) 16 (24.6) 1 (1.5) o (0) o (0) 65 (100%)
24+ 17 (53.1) 11 (34.4) 3 (9.4) 0(0) 1 (3.1) 32 (100%)

Note: figures within parentheses are percentages.

0\
\0
-
620

Table 12 presents rates for peritonitis, exit site and tunnel infections, and
catheter replacements. Separate rates are calculated for classes lA, 1B, and a
combination of 2A and 2B for both CAPO and CCPD. Each of the rates is
obtained by totalling the number of relevant episodes/occurrences and dividing
by the number of patient-years of exposure to the risk discussed. Examining
Table 12, it is clear that class 1A and 1B CAPO and CCPD patients have similar
rates of occurrence for most events. Among class 1B patients, the risk of exit site
tunnel infections and catheter replacements are higher for those on CCPD than
on CAPO. Class 2A and 2B rates are not greatly different from those of 1A and
1B patients. This is in spite of the fact that the class 2A and 2B patients are
'survivors' from a larger group of patients that began CAPO prior to their
registration. Rates for diabetics and non-diabetics were also considered for class
1A and 1B patients combined. In general, these rates did not differ greatly by
diabetic condition. The most noticeable difference was in total hospital days per
patient-year, with diabetics experiencing 36 hospital days per patient-year and
non-diabetics 23.
It should be noted that the Registry's CAPO patients have many more patient-
Table 12. Rates of occurrence" of selected events and average months between episodes by class of
case
Class

lA IB 2A and 2B

CAPD CCPD CAPD CCPD CAPD CCPD

Rates
Peritonitis 1.6 1.7 1.8 1.5 1.8 1.8
Exit/tunnel infection 0.9 0.9 0.8 1.1 0.7 0.9
Catheter replacement 0.4 0.4 0.3 0.4 0.4 0.5

A verage months between

episodes
Peritonitis 7.5 7.0 6.7 7.8 6.5 6.5
Exit/tunnel infection 14.0 14.0 15.5 11.1 18.5 12.8
Catheter replacement 33.4 31.7 42.0 31.4 38.8 25.6

Based on
Patients 1,463 53 2,713 108 2,246 73
Patient-years 868.1 31.6 1,796.0 65.5 2,208.7 51.1

Average No. days hospitalizedh

Due to CAPD/CCPD 11.0 13.0 10.4 13.9 10.2 14.7
complications
All causes 29.3 33.2 24.2 30.9 20.7 23.9

d Number of episodes per patient-year of observation.

h Per patient-year of observation.
 621

years of experience than the CCPD patients in the Registry. This means that the
CCPD values are more likely to change with additional patients and time. In
short, comparisons made at this point are mainly speculations.
Average months between consecutive episodes of these occurrences may be
obtained by using the relationship:

12
months between occurrences = .
rate per patient-year

This measures how long, on the average, a patient will remain on CAPD or
CCPD between two episodes of the particular type. These values also appear in
Table 12. This measure does not represent actual time between episodes, but an
average value which does not take into account how many episodes the patient
has already experienced.

5.5. Life-table analyses

Several life-table type analyses for CAPD patients (class 1A and 1B combined)
are worth examining because they provide, in a pictorial manner, useful informa-
tion about time until the first occurrence of an event. Class 2A and 2B patients are
excluded because there is no knowledge about events occurring prior to follow-up
by the Registry. Eight life-table graphs are presented. Each of these graphs
indicates the probability of having had a particular event occur by the time noted.
For convenience, values read off the graph are tabled under the plotted curve.
For every three-month interval, the number of patients still after observation and
who have not experienced the first episode of the stated condition are listed in the
second column. The third column reports the number of patients that have
experienced the specified event (i.e., died, developed peritonitis, transferred off,
etc.) during the interval. The fourth column presents the number of patients
withdrawn from observation during the interval. The exact meaning of 'with-
drawn from observation' will vary from table to table and is explained by a
footnote with each table. The fifth column contains the probability, expressed as
a percent, of having had the first episode by the beginning of each three-month
interval. The last column presents the probability of not having had the first event
of the particular type occur by the beginning of the time interval. The first life-
table, Figure 6, presents the probability of experiencing the first episode of
peritonitis as a function of how long the patient is on CAPD. Only 31.6% of
CAPD patients remained free of peritonitis for 12 months; only 21.5% for 18
months. The figures beyond 18 months are based on less than 100 patients
remaining under observation, and thus are likely to change with additional
patients.
The next life-table, Figure 7, presents the probability of contracting the first
622

exit site or tunnel infection by different periods of time. According to the table,
51.5% of the considered patients will have had either an exit site or tunnel
infection by the end of 18 months of CAPD, and 48.5% will not have had these
infections.
Figure 8 presents the probability of first replacement of the catheter. As the
chart (and the graph) indicate, after one year on CAPD 20.9% of patients were
no longer using their original catheter; 28.7% were no longer using their original
catheter after 18 months of follow-up. In a future report, the time to the second
and later episodes of these conditions will be reported.
The probability of having an infection (peritonitis or exit site/tunnel infection)
or having a catheter replacement is presented in Figure 9. Nearly 80% of patients
needed to have their cather replaced or developed peritonitis or an exit site or
tunnel infection by the end of the first year.
To determine the probability of transferring off CAPD onto hemodialysis or
onto IPD, or leaving dialysis without return of kidney function, consult Figure 10.
After one year, 22.6% of patients remaining alive have transferred off CAPD;
31.2% have transferred off CAPD within 18 months. The number of patients
under observation for more than 1.5 years is too small to yield a statistically
reliable figure. When the probability of transferring off CAPD is examined by the
year the patient was trained to performed CAPD, the data suggested that those
trained earlier have about the same chance of remaining on CAPD for one year as
those trained later.
Analyses for age and diabetic/non-diabetic (diabetic glomerylosclerosis) sub-
groups of the population are also presented in the tabular portion of the life-table.
Since these subgroup analyses exclude patients on whom the primary renal
disease was not reported, or the age was not reported or the age was under 20, the
subgroup figures in the table may appear inconsistent with the overall figure
presented. The identified subgroups exhibited little variation with respect to this
outcome.
Another question of interest concerns time to first hospitalization for CAPD
complications. By examining Figure 11, it can be noted that 57.7% of patients had
been hospitalized for CAPD-related complications by the end of their first year
on therapy; 71.0% had been hospitalized for this reason within 18 months on
CAPD. There is a slight tendency for those over 60 to be hospitalized earlier than
those between 20 and 60.
The probability of dying while on CAPD is of major consequence, and is
depicted in Figure 12. Patients are considered to have died while on CAPD if they
die within two weeks after transferring to either hemodialysis or intermittent
peritoneal dialysis (IPD). The 'death-table' includes deaths that mayor may not
be CAPD related. They may be related more to age, the nature of the renal
disease, and other diseases rather than to the quality or type of therapy. Also,
these values should not be compared to values reported elsewhere on other
treatments because the CAPD deaths are not broken out separately. According
 623

to the life-table, 14.5% of registered patients placed onto CAPO will have died by
the end of their first year on the therapy; 20.8% of followed patients will have
died within 18 months on the therapy.
There is a noticeable tendency for diabetics to die sooner while on CAPO than
the non-diabetics. As well, the older patients on CAPO die sooner than the 20-
through 59-year olds on CAPO. Of course, this latter effect may be due in part to
age per se. Patients between 20 and 59 years old who are not diabetic have
demonstrated the best survival pattern. After one year, 6.6% of the patients will
have died, 9.9% after 18 months.
Mortality classified according to the year the patient began CAPO therapy was
also examined. There did not appear to be any major survival differences be-
tween the different cohorts. Additional years of data may change that conclu-
sion.
The last question addressed concerns how long a patient remains alive and on
CAPO. Figure 13 presents the probability of transferring off CAPO (tranferring
to hemodialysis, IPO, or leaving dialysis without return of kidney function) or
dying, in the form of a life-table. The table reveals that there is 66.3% probability
that a patient who started CAPO will be able to remain on CAPO for one year - a
54.9% chance of remaining on CAPO for 18 months. When all reasons for leaving
CAPO are considered (transfers to hemodialysis or IPO, leaving dialysis with or
without return of kidney function, transplants, and death), it was found that
59.7% of patients remain on for 12 months and 46.6% for 18 months.
In summary, the results presented in this section describe the experience of
patients followed since beginning CAPO. Additional years of data collection
from greater numbers of centers should improve the stability and generalizability
of the figures discussed.
624

PROBABILITY
100------

00----------
oo--------------------~----

ro-------------=~~--------

oo---------~~---------------

M-------.~----------------

~-----~-----------------

~---~~--------------------

20--~--------------------

10---F-----------------------------

r-r-......-r-rr-r-r--o--TTT iii i lb . iii I 1~ iii , I Jl

MONTHS

Figure 6. Probability of first peritonitis.

Data for figure 6

Number of Number of Number Number Cumulative Cumulative

months from non-failing developed first withdrawn probability probability
placement patients on peritonitis from (%) of first (%) of
onto CAPD CAPD at during observation peritonitis remaining free
beginning of interval during episode by of peritonitis
interval interval- beginning of by beginning
interval of interval

::;3 3977 791 899 0 100

4-6 2287 551 474 25.8 74.2
7-9 1262 283 288 45.6 54.4
10--12 691 123 198 59.6 40.4
13-15 370 64 112 68.4 31.6
16-18 194 24 65 74.4 25.6
19-21 105 13 55 78.5 21.5
22-24 37 4 25
25+ 8 7

_ Includes patients who were taken off CAPD or died, and patients on CAPD who had no
peritonitis through date of last report.
 625

PROBABILITY
100--------------------------

00----------------------------------------------

~----------------------------------------------

70------------------------------------------
M----------------------------------------------
ro-----------------------------------~~~------

40----------------------~~~------------------

~--------------~~----------------------------

~------~~---------

10---7'L----------

I, I I I • I ii, (i i i. I I I " I I I I I I I I I' I I I , I I I I , I I

3 6 9 12 15 18 21
MONTHS

Figure 7. Probability of first exit site/tunnel infection.

Data for figure 7

Number of Number of Number Number Cumulative Cumulative

months from non-failing developed withdrawn probability probability
placement patients on first exit site/ from (%) of first (%) of
onto CAPD CAPD at tunnel observation exit site/ remaining
beginning of infection during tunnel free of exit
interval during interval" infection by site/tunnel
interval beginning of infection by
interval beginning of
interval

s3 4028 518 1005 0 100

4-6 2505 258 611 17 83
7-9 1636 148 437 26.8 73.2
10--12 1051 78 333 34.6 65.4
13-15 640 49 223 40.4 59.6
16--18 368 28 150 46.4 53.6
19-21 190 16 102 51.5 48.5
22-24 72 3 48
25+ 21 2 19

" Includes patients who were taken off CAPD or died, and patients on CAPD who had no exit
site or tunnel infection through date of last report.
626

PROBABILITY
100----------------------------------------------

90

00----------------------------------------------

70----------------------------------------------

00----------------------------------------------

50----------------------------------------------

40----------------------------------------------

30 -

2~",=
o 3 6 g
MONTHS

Figure 8. Probability of first catheter replacement.

12 15 18 21

Data for figure 8

Number of Number of Number Number Cumulative Cumulative

months from non-failing replacing first withdrawn probability probability
placement patients on catheter from (%) of (%) of using
onto CAPD CAPD at during observation replacing original
beginning of interval during catheter by catheter by
interval interval a beginning of beginning of
interval interval

::;3 4050 210 1082 0 100

4--6 2758 112 719 8.2 91.8
7-9 1927 76 544 12.3 87.7
10-12 1307 58 430 16.4 83.6
13--15 819 30 305 20.9 79.1
16--18 484 25 203 24.4 75.6
19-21 256 10 148 28.7 71.3
22-24 98 1 68
25+ 29 5 24

a Includes patients who were taken off CAPD or died, and patients on CAPD who had not
replaced the original peritoneal catheter through date of last report.
 627

PROBABILITY
100----------------------------------------------
001----------------------------------------------

oo'------------------------~~~----------------

70-------------~~~==--------------------------

oo----------~----------------------------------

~,--------~------------------------------------

40-------t/---------------------------------------

~,----~-------------

~--~------------------

...,-,--r--r-T~.,.--r-rrl I ' I I I I I • iii j I I I I i I Ii' I I i I

3 6 9 12 15 18 21
MONTHS

Figure 9. Probability of first peritonitis, exit site or tunnel infection, or catheter replacement.

Data for figure 9

Number of Number of Number Number Cumulative Cumulative

months from non-failing having first withdrawn probability probability
placement patients on peritonitis or from (%) of having (%) of not
onto CAPD CAPD at exit site/ observation peritonitis, having
beginning of tunnel during exit site/ peritonitis,
interval infection or interval- tunnel inf. or exit site/
replacing first new cath. by tunnel info or
catheter beginning of new cath. by
during interval beginning of
interval interval

:53 3819 1187 733 0 100

4-6 1899 596 368 39.7 60.3
7-9 935 262 192 70.7 39.3
10-12 481 95 131 73.2 26.8
13-15 255 69 64 79.6 20.4
16-18 122 16 42 85.7 14.3
19-21 64 11 31 88.1 11.9
22-24 22 18
25+ 3 1 2

_ Includes patients who were taken off CAPD or died, and patients on CAPD who had no
peritonitis, exit site, or tunnel infection, or had not replaced their original catheter through date of
last report.
628

PROBABILITY
100------------------------------------------------

~------------------------------------------------

00-----------------------------------------------

ro------------------------------------------------
00----------------------------------------------

~------------------------------------------------

~[Link]S;"", ,
o 3 6 9
MONTHS
12 15 18 21

Figure 10. Probability of transferring to hemodialysis or IPD, or leaving dialysis with no return of
kidney function.

PROBABILITY
100------------------------------------------------

~------------------------------------------------

00------------------------------------------------

70--------------------------------------:~--------

M----------------------------:~~----------------

~------------------~~~------------------------

~------------~~------------------------------

~--------~'-------------------------------------

ro----~~--------------------------------------

i I' i i' iii iii iii iii iii iii j I

9 ~ ffi ~ ~
U0t!THS

Figure 11. Probability of first hospitalization for CAPD-related complications.

 629

Data for Figure 10

Number of Number of Number Number Cumulative Cumulative

months from non-failing transferring to withdrawn probability probability
placement patients on hemodialysis, from (%) of (%)ofnot
onto CAPD CAPD at IPD, or observation transferring transferring
beginning of leaving during off CAPD by offCAPD by
interval dialysis with interval" beginning of beginning of
no return of interval interval
kidney
function
during
interval

:53 4103 194 983 0 100

4-6 2926 138 672 6.3 93.7
7-9 2116 123 532 11.1 88.9
10--12 1461 75 434 17.0 83.0
13-15 952 51 336 22.6 77.4
16-18 565 24 234 27.5 72.5
19-21 307 17 163 31.2 68.8
22-24 127 9 79
25+ 39 3 36

a Includes patients who were transplanted during interval, had return of kidney function, died, or

who had not been transferred off CAPD through date of last report.

Subgroup analysis a of cumulative probability (%) of transferring off CAPD

Number of All patients Diabetics b Non-diabetics Age 20--59 Age ~60

months from only
placement
onto CAPD

0 0 0 0 0 0
3 6.3 2.6 2.6 2.6 2.6
6 11.1 6.8 6.9 7.1 6.4
9 17.0 14.0 12.9 12.8 13.7
12 22.6 19.6 18.8 18.4 20.9
15 27.5 25.0 23.9 24.0 25.3
18 31.2 28.9 27.8 28.7 27.7

a These subgroup probabilities are not based on the full data set. See Text.
b Patients with diabetic glomerulosclerosis as primary renal disease.
630

Data for Figure 11

Number of Number of Number first Number Cumulative Cumulative

months from non-failing Hospitalized withdrawn probability probability
placement patients on due to CAPD- from (%) of first (%) of not
onto CAPD CAPD at related observation CAPD-related having been
beginning of problems during hospitalization hospitalized
interval during interval" by beginning by beginning
interval of interval of interval

,-;3 3945 685 890 0 100

4-6 2370 421 474 23.3 76.7
7-9 1475 239 328 38.4 61.6
10-12 908 121 261 49.8 50.2
13-15 526 69 167 57.7 42.3
16-18 290 49 98 64.6 35.4
19-21 143 16 71 71.0 29.0
22-24 56 5 37
25+ 14 6 8

a Includes patients who were taken off CAPD or died, and patients on CAPD who had not been
hospitalized for CAPD-related complications through date of last report.

Subgroup analysis' of cumulative probability (%) of first hospitalization for CAPD complications

Number of All patients Diabetics b Non-diabetics Age 20-59 Age 2:60

months from only
placement
onto CAPD

0 0 0 0 0 0
3 23.3 18.2 16.9 16.3 17.3
6 38.4 36.1 31.6 31.4 33.0
9 49.8 47.4 44.3 43.2 46.3
12 57.7 55.6 53.1 50.6 57.5
15 64.6 63.2 60.8 57.7 65.7
18 71.0 70.0 67.9 66.3 70.2

a These subgroup probabilities are not based on the full data set. See Text.
b Patients with diabetic glomerulosclerosis as primary renal disease.
 631

PROBABILITY
1001------------------------

00'-----------------------------------------------
~I---------------------------------------------------

70'---------------------------------------------------

M,---------------------------------------------------
~,------------------------------------------------

~,---------------------------------------------------

~---------------------------------------------------

:~::="~"
o 3 6 9
MONTHS
12 15 18
, J
21

Figure 12. Probability of dying while on CAPO.

PROBABILITY
100---------------------------------------------------

00---------------------------------------------------
~----------------------------------------------

ro---------------------------------------------------
M---------------------------------------------------
~------------------------------------------------

~--------------------------------~~~=------------

~----------------------~~~--------------------

~------------~~~------------------------------

10 -------.~~---------

~TO_."TO_.""_.rr,, r'T"','_.'TJT'-~rrT'.J"~'rT·TO"1"rr'T'TO"J
12 15 18 21
MONTHS

Figure 13. Probability of transferring to hemodialysis or IPD, or leaving dialysis with no return of
kidney function or dying while on CAPO.
632

Data for Figure 12

Number of Number of Number dying Number Cumulative Cumulative

months from patients on during withdrawn probability probability
placement CAPD at interval from (%) of death (%) of not
onto CAPD beginning of observation by beginning dying by
interval during of interval beginning of
interval" interval

:53 4135 99 1080 0 100

4-6 2956 114 689 3.1 96.9
7-9 2153 76 573 7.4 92.6
10-12 1504 47 465 11.2 88.8
13-15 992 37 356 14.5 85.5
16-18 599 15 254 18.6 81.4
19-21 330 8 186 20.8 79.2
22-24 136 6 84
25+ 46 3 43

" Includes patients who were taken off CAPD and patients on CAPD through date of last report.

Subgroup analysis" of cumulative probability (%) of death

Number of All patients Diabeticsb Non-diabetics Age 20-59 Age 2:60

months from only
placement
onto CAPD

0 0 0 0 0 0
3 3.1 2 0.9 0.8 1.9
6 7.4 7.1 4.5 3.4 8.8
9 11.2 13.6 7.7 5.8 15.8
12 14.5 18.5 10.8 8.0 21.4
15 18.6 29.5 13.4 11.7 26.9
18 20.8 31.6 15.6 13.4 30.5

" These subgroup probabilities are not based on the full data set. See text.
b Patients with diabetic glomerulosclerosis as primary renal disease.
 633

Data for Figure 13

Number of Number of Number Number Cumulative Cumulative

months from patients on transferring to withdrawn probability probability
placement CAPD at hemodialysis from (%) of leaving (%) of
onto CAPD beginning of or IPD, observation CAPD by remaining on
interval leaving during beginning of CAPD by
dialysis interval" interval beginning of
without return interval
of kidney
function, or
dying

,,;3 4088 293 869 0 100

4-6 2926 251 559 9.2 90.8
7-9 2116 196 459 17.7 82.3
10--12 1461 118 391 26.2 73.8
13-15 952 86 301 33.7 66.3
16-18 565 37 221 40.8 59.2
19-21 307 24 156 45.1 54.9
22-24 127 15 73
25+ 39 6 33

" Includes patients who received transplants, had kidney function return, or remained on CAPD
without transferring off or dying through date of last report.

Subgroup analysis" of cumulative probability (%) of transferring or death

Number of All patients Diabetics b Non-diabetics Age 20--59 Age 2:60

months from only
placement
onto CAPD

0 0 0 0 0 0
3 9.2 4.6 3.4 3.4 4.4
6 17.7 13.6 11.1 10.3 14.6
9 26.2 25.7 19.5 18 27.1
12 33.7 34.6 27.3 25.1 37.2
15 40.8 46.4 34.0 32.9 44.8
18 45.1 49.9 39.1 37.8 49.3

" These subgroup probabilities are not based on the full data set. See text.
b Patients with diabetic glomerulosclerosis as primary renal disease.
634

Acknowledgements

The authors wish to thank all participating centers for providing the data on which
this report is based.
We gratefully acknowledge the assistance of the following staffs of The
EMMES Corporation and the University of Missouri, without whom the report
could not have been completed:
Jeanette Leroux Maureen Santiago
Jeanne Novak Mark Stempko
Kathleen O'Reagan Alice Thompson
Warren Pendleton Tamara Voss
Jane Redmond Robin Walters
Walter Saiko Sue Ward
Questions regarding this report and requests for additional copies should be
addressed to:
National CAPD Registry,
Data Coordinating Center,
The EMMES Corporation,
11325 Seven [Link] Road, Suite 214,
Potomac, MD 20854, USA.
800--638-2578; in Maryland, 301-299-8655
For other information about the Registry project, contact:
Karl D. Nolph, M.D., or Ms Susan Cooper,
Division of Nephrology (M472),
University of Missouri Health Sciences Center,
Columbia, MO 65212, USA.
314-882-7991

Reference

1. Popovich RP, Moncrief JW et al.: The definition of a novel portable/wearable equilibrium per-
itoneal dialysis technique (abstract). Abstr Am Soc Artif Intern Organs 5: 64,1976.
 635

Appendix 1

NIH Project Officer

Gladys Hirschman, M.D.

Chronic Renal Disease Program
Kidney-Urology Branch DKUHD)
NIADDK
Bethesda, MD

CAPD Registry Committee Members

CAPD Advisory Committee

Christopher R. Blagg, M.D. Robert G. Luke, M.D.

Northwest Kidney Center University of Alabama in Birmingham
Seattle, WA Birmingham, AL

John A. Goffinet, M.D. John F. Maher, M.D.

VA Hospital Uniform Services University of Health Sciences
West Haven, cr Bethesda, MD

Lee W. Henderson, M.D. George W. Williams, Ph.D.

VA Hospital Cleveland Clinic Foundation
San Diego, CA Cleveland, OH

Joanne Hoover", M.D., MPH

University of Washington
Seattle, WA

CAPD Data Monitoring Committee

Joanne Hoover", M.D., MPH William F. Krol, Ph.D.

University of Washington Maryland Medical Research Institute, Inc.
Seattle, WA Baltimore, MD

Bruce A. Kottke, M.D., Ph.D. Edward H. Wagner, M.D.

Mayo Clinic Group Health Cooperative of Puget Sound
Rochester, MN Seattle WA

CAPD Executive Committee

Robert Hamilton, M.D. Steven B. Kurtz, M.D.

Bowman Gray School of Medicine Mayo Clinic
Winston-Salem, NC Rochester, MN

Richard Hamburger, M.D. Wadi N. Suki, M.D

Indiana University Medical Center Baylor College of Medicine
Indianapolis, IN Texas Medical Center
Houston, TX
" Dr Hoover is on two committees
19. Peritoneal dialysis results in the EDTA
Registry

A.J. WING, R. MOORE, F.P. BRUNNER, C. JACOBS, P. KRAMER

and N.H. SELWOOD

1. The EDT A Registry

The Registry of the European Dialysis and Transplant Association (EDTA)

collects data from 32 European countries with a base population of 574 millions.
This provides a computerised data base of over 140000 individually registered
patients, 80000 of whom were alive on dialysis or with a functioning transplant on
31st December, 1982 [1].
Every patient record contains the treatment sequence described by a number
code for each change in the method of treatment and a date for the change. This
date sequence is used for incidence and prevalence statistics and for calculating
survival rates.
At the close of 1982 almost 5% of European patients were being treated by
CAPD (Fig. 1, Table 1). Patients on CAPD had a mean age of 51.4 years and
represented 6.69 patients per million population (pmp). Over 8% of all patients
were aged over 65 at the start of treatment for end-stage renal failure (ESRF),
and 10% ofthese patients were on CAPD with a mean age of 71.9 years (Table 2).
Children who commenced treatment under the age of 15 years accounted for just
over 3% of the total patients and 5% of these were on CAPD with a mean age of
10.4 years, in contrast to those on haemodialysis or with a functioning transplant
who, although they had commenced therapy when aged under 15 had mean ages
above 15 on 31st December, 1982. It can be seen, therefore, that CAPD makes a
significant contribution to the treatment of ESRF in Europe.
The extent of this contribution varies widely between countries (Table 3). In
the United Kingdom, where hospital haemodialysis facilities have not expanded
in recent years almost 90% of centres were performing CAPD at the close of 1982,
whereas little CAPD was performed in the Eastern European countries (Fig. 2).
In the mid-1970s intermittent peritoneal dialysis (IPD) began to make a small
contribution to the dialysis programme and was promptly overtaken by the
growth of CAPD from 1978, and the subsequent decline in the proportion of
patients alive on IPD or haemodialysis (Fig. 3). There was a rapid growth in the
638

" C>
~ Hom ..
Hospita l
Hopmod,alys,s

~ Hapmodlalysis

~ I PD
~CAPD

n = 78.637
~ Transplan t

Mpan Age . "ypors

Figure 1. Proportions of live patients recorded by the EDTA Registry on each mode of therapy on 31st
December, 1982.

Table 1. Methods of treatment of live registered patients on 31st December, 1982

All patients

Nos . % PMP Mean age

CAPO 3896 4.9 6.7 51.4

IPD 855 1.1 1.5 56.9
Hospital haemodialysis 50210 63.8 86.2 49.5
Home haemodialysis 7988 10.2 13.7 45 .1
Transplant 15688 19.9 26.9 38.5

Total Registry 78637 100.0 47.0

Table 2. Methods of treatment of live elderly patients (aged over 65 at first treatment) and of children
(aged under 15 at first treatment)

Age at start of treatment

Less than 15 years More than 65 years

Nos . % PMP Mean Nos. % PMP Mean

age age

CAPO 140 5.4 1.1 10.4 671 10.2 1.1 71.9

IPD 20 0.8 0 .1 9.1 284 4.3 0.5 72.8
Haemodialysis 1212 47.0 9.2 15.3 5500 84.0 9.4 72 .2
Transplant 1206 46.7 9.1 16.1 89 1.4 0.1 76.9

Total Registry 2578 3.3 15.3 6544 8.3 72.3

 639

Table 3. Registered patients known to be alive on 31st December, 1982 in 32 countries (reproduced
from Wing et al [1]

Country Centres IPD CAPO On haemodialysisl With Total Per

replied haemofiltration func. mill.
_ _ _ _ _ _ _ _ _ _ trans. pop.
%
Hos- home of
pital Home Total total

Algeria 0 741 5 20.0 o 12 0.7

Austria 24 4 18 917 61 978 6.2 241 1241 165.5
Belgium 50 2 129 1594 116 1710 6.8 851 2692 274.7
Bulgaria 27 0 o 445 19 464 4.1 14 478 53.1
Cyprus 2 0 o 85 0 85 0 24 109 181.7
Czechoslovakia 21 0 10 650 1 651 0.2 189 850 55.9
Denmark 11 49 95 341 61 402 15.2 343 889 174.3
Egypt 19 o 385 2 387 0.5 30 418 10.7
Fed. Rep.
Germany 248 109 194 10 485 1820 12305 14.8 1673 14281 233.4
Finland 24 13 69 169 4 173 2.3 293 548 114.2
France 189 208 637 8436 1 835 1027l 17.9 1932 13048 244.3
German. Oem.
Rep. 51 6 5 1008 1009 0.1 406 1426 84.9
Greece 38 5 5 917 3 920 0.3 210 1140 122.6
Hungary 10 16 3 205 0 205 0 94 318 29.7
Iceland 1 0 o 15 0 15 0 7 22 110.0
Ireland 4 2 26 150 42 192 21.9 195 415 125.8
Israel 26 54 54 688 59 747 7.9 221 1076 283.2
Italy 259 143 863 9817 913 10730 8.5 1135 1287l 226.6
Lebanon 3 0 o 42 0 42 0 o 42 15.6
Libya 2 0 o 64 0 64 0 4 68 23.4
Luxembourg 5 0 o 69 2 7l 2.8 4 75 187.5
Netherlands 43 90 1587 158 1745 9.1 804 2640 188.6
Norway 16 0 25 200 5 205 2.4 472 702 171.2
Poland 35 26 6 449 13 462 2.8 156 650 18.4
Portugal 17 2 2 783 0 783 0 33 820 83.7
Spain 156 90 332 5727 343 6070 5.7 883 7375 199.3
Sweden 31 35 125 584 129 713 18.1 898 1771 213.4
Switzerland 32 3 170 834 198 1032 19.2 640 1845 283.8
Tunisia 7 2 o 138 0 138 0 141 22.7
Turkey 15 3 6 239 240 0.4 116 365 8.3
United
Kingdom 59 73 1017 1416 2184 3600 60.7 3676 8366 149.7
Yugoslavia 52 8 8 1767 17 1784 1.0 143 1943 87.9

Total Registry 1478 855 3896 50210 7988 58198 13.7 15688 78637 135.0
640

UK
SWITZERLAND

FRANCE
SWEDEN
BELGIUM
TOTAL REGlSTRY- NETHERLANDS

SPAIN FRG

YUGOSLAVIA

GDR

Figure 2. Per cent of centres which performed CAPD on 31st December 1982 in 11 selected countries .

• ~
78
77
76
75
74
73

:J
h
I

.-----~~-----..--~
PATS.

+-------•.-------•
17
~
5
4
3
2
1
0
YEAR '76 '77 '78 '79 '80 '81 '82
Figure 3. Per cent of patients alive on different modes of therapy on 31st December in each of the years
1976-1982. Note interrupted scale.
 641

proportion of patients on CAPD in the United Kingdom and Italy, which still
continues, whilst in other countries the growth has plateaued (Fig. 4). Only 1% of
the large number of patients on therapy in the Federal Republic of Germany were
on CAPD. Eastern European countries commenced CAPD later and there has
been little growth.

2. Demography: centres

Analysing the number of centres that commenced a particular mode of therapy

year by year demonstrates clearly the 'explosion' in CAPD that occurred in the
late 1970s and peaked in 1980 (Fig. 5). There was a subsequent decline which is
also seen in the number of centres which commenced haemodialysis or transplan-
tation for the first time in each year.
This indicates that expansion of populations depended in recent years on the
development of existing centres rather than the opening of new ones. Separate
analysis for six selected countries (Fig. 6) reveals similar patterns except for the
noted lack of CAPD in the Eastern block and a late peak in the centres commenc-
ing CAPD in the Federal Republic of Germany.

15

10

%
PATS.

OL-_.......,._ _.... =-
YEAR '76 '77 '78 '79 '80 '81 '82

Figure 4. Per cent of all living patients treated by CAPD in eight selected countries, 1976-1982.
642

3. Demography: patients

Analysing the number of patients alive on each mode of therapy enables us to see
the growth of CAPD in correct relationship (Fig. 7). Whilst the number alive on
CAPD has grown, so has the number alive with a functioning transplant and on
haemodialysis; moreover, they have increased at an equal rate. The different
patterns in growth of numbers of patients alive per million population on each
mode of therapy in a selection of six European countries are shown in Figure 8. In
all countries, except the United Kingdom, the numbers on haemodialysis or with
a transplant have grown at the same or at a greater rate than those on CAPD. In
the United Kingdom limited hospital haemodialysis facilities have forced physi-
cians to expand the programmes of transplantation and CAPD. The percentage
of live patients on CAPD correlates in each country with the number of patients

• 200
TOTAL REGISTRY

* CAPD
D HAEMODIALYSIS
.. TRANSPLANT
• IPD *

No. OF
CENTRES
150

I *
100

YEAR '76 '77 '78 '79 'SO '81 '82

Figure 5. Commencement of modes of treatment in Europe, 1976-1982. Numbers of centres are

plotted according to the year in which they claimed to have treated their first patient on each therapy.
 BI!LGIUM SWEDEN GDR *CAPD
CD 10 10 10
D HAEPI>DIALYSIS

110 ....
CE .....
5 5
/D

°/"0 l"- V *~*_

'-.o--?--?
'fA. o'76
LFRANCE
'77 ']8
• '79 •
i ,., 'II '12 '76
.r:/~ .\->
UK
'77 '71 '79 '., 'II '82
oU'~f3.
'16

F R G
'77 '71 '79 '., 'II '82

30 30

20
110 ....
CE .....
*/\ 20 /* ~ //*
.I *
10[;-./.*'-. \ 10 10 O'-.O'f*
/ *
---\/1
* 0
'j'"""'>?-y /'
'U.
. I ~~
'76 77
,* , '78 .'79 •., -
,·Jl 'k
O~L '76 '77 '78 '79 '.,
y\
'81 '82 '76 '77 '78 '79 '., '81 '82

Figure 6. Commencement of CAPD and haemodialysis in six European countries, 1976-1982. Numbers of centres are plotted according to the year in which they
0\
claimed to have treated their first patient on each therapy. .j:>.
VJ
644

0 ,0

60
I H"~,, "''''
TRANSPLANT
P. O.

50

~O
PATS.
X 10 3
30

20

10

0
YEAR '78 '79 'SO '81 '82

Figure 7. Numbers of patients alive on haemodialysis, with a functioning graft and on CAPD, total
Registry on 31st December, 1978-1982.

on CAPD per million population (Fig. 9). The exception is the United Kingdom
where the number treated by CAPD represents a larger percentage of the total
than in other countries because of the restrictive numbers of haemodialysis
patients in the United Kingdom.
The rate of acceptance of patients for CAPD in different countries in 1982,
expressed per million population, is shown in Figures 10 and 11. Switzerland and
the United Kingdom accepted the most patients for CAPD and the Eastern block
countries the least. Sweden and Switzerland were ahead of the other countries in
accepting diabetic patients whilst the other Western European countries were
more closely grouped together. When rate of acceptance according to age was
considered Switzerland and the United Kingdom accepted the most children
(aged 0-15) but the United Kingdom came behind France, Switzerland, Sweden,
Belgium and Italy in accepting patients aged 65 and above .
These demographic data show that although CAPD has been deployed speed-
ily during recent years in Europe it is used only for a small proportion of patients
overall and its recent popularity has been achieved neither by reduction in
numbers of patients on haemodialysis nor in the rate of increase of the
haemodialysis population. It has notably augmented the programmes for ESRF
therapy and in no country more than the United Kingdom.
 645

0. .HOIU ..

* •
CAPO
TRANSPLANT
o HA£MOD JALYS I S
• IPO
''''EOEN

III) 0 OR

1--/--
50 ./

.-- --
t --.--. _ _ .

._-.
..
O~ , I t-.-~-..
. " 'n ' 11 ' /'I ' ., ' 11 ' 12
"""ANCE

'au .
I'IF
110

.1
/
100 100

,.""
o f~r'
' ]L
",

'n
•
..--. "."..---...--.
-:;:-:;"-0_
'1'
T I
' 1'9
0I- .-*-
' .:I "U
I
•.,
, '--rf~"""'''''''
'n ' /1 ." •.,
.-
' Il ' 11

Figure 8. Numbers of patients alive per million population on 31st December, 1976-1982 according to
mode of therapy and in six European countries.

4. Selection of patients for CAPD

Our analysis of the population of patients selected for CAPD is based on patient
characteristics recorded for all cases on the Registry. The treatment sequence was
used to deduce whether a patient had received CAPD as first therapy or after
transfer from some other treatment. Reasons for choice for CAPD were recorded
by number code for all cases up till the end of 1981. Age at commencement of
646

• 12

10
·U.K.

.CH

8
% .Sw
I·
LIVE PATS.
6
-B
.E .F
4
-NL

2
-FRG
Y
..GDR
0
0 5 10 15 20 25
No. PATS. PER MILLION POPULATION

Figure 9. Contribution of CAPD to national programmes on 31st December, 1982, showing correla-
tion between per cent of all patients on CAPD and number of CAPD patients per million population.

FRANCE
NETHERLANDS

SPAIN

UK
SPA I N
UK

S,ll TZERLAN;)
ITALY

DIABETIC NOfl-DIABETIC

Figure 10. Rate of acceptance of diabetics and of non-diabetics for CAPD during 1982 in ten selected
countries.
 647

YUGOSL.V I A~ I GOR

, - - FRG

' - - - SPAIN
"--- BE LG I UM
15 - 35

Figure ll. Rate of acceptance of patients aged 0-15 and 65 plus for CAPO during 1982 in ten selected
countries.

CAPO was computed from the recorded birth date, and the diagnosis of primary
renal disease was available through a two-digit number code.
The proportion of patients who commenced CAPO in 1982, with CAPO as the
first therapy ('new' patients) or having transferred from a previous mode of
therapy is shown in Figure 12. Similar proportions amongst diabetic and non-
diabetic patients transferred from haemodialysis and IPD but very few after a
failed transplant. Almost 70% of diabetic patients commenced CAPO as 'new'
patients, compared to 58% non-diabetics, suggesting a preference for CAPO in

•
PREVIOUS TREATMENT IN PATIENTS COMMENCING CAPD IN 1982

Non - Diabeolic Diabeot ic

No of
patil"nts 2,178 352

C>Nl"W C> Haeomod ialys is

~ IPD • Transplant

Figure 12. Previous treatment in diabetic and non-diabetic patients who commenced CAPO in 1982.
'New' patients are those for whom CAPO was the first recorded mode of therapy.
648

diabetic patients. However, only 15% of diabetics on the Registry were treated by
CAPD. Most were on haemodialysis (66%) and 19% had a functioning
transplant.
The reasons for choice of CAPD were separated into three categories; those
patients in whom CAPD was regarded as temporary, 'enforced' or as 'first choice'
treatment. CAPD was considered first choice treatment for over 50% of all age
groups and both sexes (Fig. 13). There was a tendency for its preference as first
choice treatment to increase with age and also in males. CAPD was more likely to
be used as a temporary measure in younger patients awaiting transfer to
haemodialysis or transplantation. The proportion of patients for whom CAPD
was said to be enforced treatment because of the non-availability or failure of
alternative therapy, or medical and technical contra-indications was consistent
throughout the age groups, but was larger in females than in males .
The percentage for whom CAPD was considered to be enforced in the Federal
Republic of Germany was strikingly larger than most other countries (Fig. 14).
Sweden and the United Kingdom considered CAPD the first choice treatment in
over 70% of patients for whom it was used . CAPD was more often considered
treatment of first choice in diabetic patients than non-diabetics, and was less often
considered an enforced treatment (Fig. 15). The proportions thus varied between
countries with replies from the Federal Republic of Germany reflecting different
physician attitudes in that country. In almost 90% of diabetic patients on CAPD
in the United Kingdom, CAPD was considered the treatment of first choice.
The age distribution of all patients who were treated by transplantation,

0-14 15-34 35-54 55-64 65+ MALE FEMALE ALL

8 100
TEMPORARY

80 ENFORCED
VI

~ 60
8. FIRST
~ 40 CHOICE
v
a;
"-
20

0
n 129 612 1.553 1.341 769 2.446 1,939 4,404

Figure 13. Reason for choice of CAPO according to age and sex in 1981. 'Temporary' indicates that
CAPO was carried out whilst waiting for transfer to haemodialysis or transplantation : 'enforced'
indicates that CAPO was necessary because haemodialysis and transplantation were technically or
medically impossible or had failed.
 649

haemodialysis and CAPO at any time shows that a large proportion of male and
female patients treated by CAPO were aged 55 and over at commencement of
renal replacement therapy (Fig. 16). The age distribution of all patients with
diabetic nephropathy treated at any time by haemodialysis and CAPO reveals a
preponderance of patients aged 35-54 with similar age distribution within the two
treatments (Fig. 17) .

• 100

80
BELGIUM FRG FRANCE ITALY SPAIN SWEDEN SWITZ UK
TEMPORARY

'"
~ 60
0
c. FIRST
~ 40 CHOICE

Q;
Q.

0
n 157 265 730 791 189 155 155 1.044

Figure 14. Reason for choice of CAPO according to country in 1981. For meaning of 'temporary' and
'enforced' see legend Figure 13.

FRG FRANCE UK
NonDM OM Non OM OM Non OM OM
100
TEMPORARY
ENFORCED

80
FIRST
c CHOICE
~ 60
;;
c.

~v 40
Q;
0..
20

o
n 191 74 639 91 960 84

Figure 15. Reasons for choice of CAPO in diabetics and non-diabetics in 1981. For meaning of
'temporary' and 'enforced' see legend Figure 13.
650

• METHOD OF TREATMENT ACCORDING TO AGE, FEMALES

so ~ Grall n = 12,288

40
Em HO n= 51,320

CAPO n = 3,135

III 30
c:
~

'[ 20
C
~
... 1
8:
~-----­
~
o
Age 0 - 14 15 - 34 35 - 54 55 - 64 65 +

METHOD OF TREATMENT ACCORDING TO AGE, MALES

SO
~ Grafl n =20,547

40
Illi] H0 n = 76,361

III CAPO n = 3,960

c:
1; 30
o
a.
C
~ 20

'"
Cl.
10

o
Age 0 - 14 15 - 34 35 -54 55 - 64 65 +

Figure 16. Distribution of patients according to treatment mode and age at start of RRT. This analysis
counted only the first entry to each mode for each patient and shows males (upper panel) and females
(lower panel) .

The proportion of patients whose end-stage renal failure was caused by diabe-
tes, reno-vascular disorders (benign and malignant hypertension, polyarteritis,
Wegener's, etc.) and other multisystem diseases (myeloma, amyloid, SLE ,
Goodpasture's, scleroderma, etc.) was higher in the patients treated by CAPO
(Fig. 18). Patients on haemodialysis had a higher percentage of ESRF caused by
intrinsic 'renal' conditions such as glomerulonephritis and pyelonephritis. This
may in part be due to the selection policy of patients suitable for CAPO and in
part the preponderance of older patients put on to CAPO.
 651

• 50

40
•
lIT]
CAPO n:: 957

HO n:: 4,891

~
c
C1I
30
0
Co

c:
C1I 20
u

ct
10

0 -
Age 0-14 15-34 35-54 55-64 65+

Figure 17. Distribution of diabetic patients according to dialysis mode and age at start of RRT. This
analysis counted only the first entry to each mode for each patient.

•
6 40

30

[ill) HO n = 128,125

• CAPOn= 7, 153

10

o
CRF
Aetiology
Uncerta in
GN PN Drug
ephropothy
Cyst ic
•Heredo- Renal Doabetes Mult i-
Familial Vascular System
Other

Figure 18. Primary renal disease in patients treated by CAPD compared to haemodialysis .
652

These considerations demonstrate that the CAPD population in Europe differs

in many important respects from the haemodialysis and transplant populations
[2]. These differences result from variation in clinical policy and practice and from
contrasting attitudes to the role of CAPD in our different countries. Results of
treatment by the different methods of renal replacement therapy, therefore,
should not be compared in these populations. Patients have been selected, not
randomly assigned to therapies.

5. Management of patients on CAPO

The section of the Registry questionnaire dedicated to CAPD information in-

cluded questions on the number of catheter insertions during 1982, the numbers
of episodes of peritonitis and days of hospitalisation for peritonitis and other
causes. The main reason for abandonment of CAPD was given by a numbered
code. Death was recorded by a record of date of death and the cause of death was
recorded using a two-digit number code.
The proportions of patients who commenced CAPD in 1982 and who had 1, 2,
3, or 4 plus catheter insertions were similar in both sexes and in different age
groups (Fig. 19). The initial catheter insertion was the only one necessary in 80%
of patients. The proportion of patients who commenced CAPD before 1982 and
did not require a single catheter re-insertion was over 70% (Fig. 20). There was a
relation between the numbers of episodes of peritonitis and the number of
catheter insertions (Fig . 21). In patients with no episodes of peritonitis, 89%

Malp
n= \,732

Fpmalp
n = 2,160
1
:::::i::'
;:::::::::
=\

-- 2
=3
= 4+

15- 35
n = 451

55 - 65
n = 1,718
~~~~~~~~~~.w

I I I I I i
o 20 40 60 80 100
Ppr cpnt pat I~nts

Figure 19. Numbers of catheter insertions during 1982 in patients on CAPO throughout 1982 according
to sex and age.
 653

Insertions in 1982

Started CA PO in 1982 • 0
n • 1.837 • 1
=2
=3
Started CAPObefore 1982
n •2.141 • 4+

, , I I I
o 25 50 75 100
Per cent patients

Figure 20. Numbers of catheter insertions during 1982 in patients on CAPO on 31st December , 1982
according to whether the patient commenced CAPO before or after 1st January, 1982.

Peritonit is
Episodes in Patients (n l Inse rt ions in 1982

o 860

1
509

2 240 • 2
0

• 3 '

120 • 4+

4+ 102

o 25 50 75 100
Per cent patients

Figure 21. Numbers of catheter insertions during 1982 related to number of episodes of peritonitis in
patients who commenced CAPO in 1982 and were on CAPO on 31st December, 1982.

required only the initial catheter placement, but in patients with 4 or more
episodes of peritonitis, 32% required catheter replacements.
The proportions of non-diabetic or diabetic patients who commenced CAPO in
1982 and who had 0,1,2 , 3 or 4 plus episodes of peritonitis were the same in both
groups, almost 50% had no episodes and only 7% had 4 or more episodes (Fig.
22). The likelihood of having multiple episodes of peritonitis increased with the
654

time spent on CAPD (Fig. 23). Over 40% of patients who were on CAPD for
between six and 12 months, having commenced in 1982, had two or more episodes
of peritonitis, whilst only 6% of those patients who had been on treatment for
three months or less had multiple episodes. It is encouraging to note that 60% of
patients on treatment for six months or more during 1982 had no more than a

Non - Diabet ic Diabet ,c

No of
pat,ents 1,989 326

No of
episodes

Figure 22. Proportions of non-diabetic and diabetic patients throughout 1982 which had 0,1,2,3 or 4
plus episodes of peritonitis during 1982.

0 - 3 Months 3-6 Months 6-12 Months

No of
pat ,ents 656 568 1,069

No of
ep isodes

Figure 23. Proportions of patients treated for 0-3 months, 3-6 months and 6-12 months respectively
which had 0, 1, 2, 3, or 4 plus episodes of peritonitis during 1982.
 655

single episode of peritonitis. Peritonitis affected a iarger proportion of patients in

whom CAPO was considered the treatment of first choice, occurring in over 60%
of those patients on CAPO throughout 1981 (Fig. 24).
Diabetics spent longer in hospital because of peritonitis in 1982 than non-
diabetic patients. Almost 30% of diabetic patients spent between two and six
weeks in hospital compared to 20% of the non-diabetics (Fig. 25). Over 40% of
each group spent no time hospitalised for peritonitis. More time was spent in
hospital for peritonitis as the length of time on CAPO increased (Fig. 26) . Eighty
per cent of patients avoided hospitalisation during the first three months after
commencement of peritoneal dialysis, but only half of those treated for longer did
so . There was little difference in the pattern of hospitalisation according the age
of patients. The proportions admitted for up to two weeks, two to six weeks, and
more than six weeks do not suggest that older patients are any more costly to treat
in this respect than younger ones.
Peritonitis was given as the main cause of abandonment of CAPO in 1982 in
both male and female patients (Fig. 27) . Other abdominal complications and
inadequate dialysis contributed to the reasons given for abandonment in the
remaining proportion. Females appear more likely to be unable to cope with the
demands of CAPO. The proportion of patients who abandoned because they
were unable to cope with CAPO or because of their or of their family's request
was larger in patients on treatment for less than 12 months than in those on
treatment for more than 12 months. Peritonitis and inadequate dialysis accounted
for a larger proportion of reasons for abandonment after 12 months' treatment.

1st choice (580 pIS) Not Is1choice (315pts)

1.7% 1%

.,.
53.3
.,.
43.2

Figure 24. Proportions of patients who were selected for CAPD as treatment of first choice or not first
choice and who were on CAPD throughout 1981 which had O. 1-3, 4-6 or more than 6 episodes of
peritonitis .
656

Non- DIabet ic Diabet ic

No. of
patIents 1,521 182

Length 01
hospItalizat ion C>=o ~ = 0-2 ~ = 2 - 5 .......... = 5+
~ weeks ~ weeks .......- weeks

Figure 25. Proportions of non-diabetic and diabetic patients on CAPO throughout 1982 which had 0,
0-2, 2-6 and 6 plus weeks of hospitalisation for peritonitis during 1982 .

•<3 m
onlhs
Aye <40 40-60 >60

n ' l68 n Z90 n 240

3 - 12 months

n 480

Hospitalizahon C>- a [» <2

"eeks
~ 2-6 ~ >6
~ \',eeks ...........- ',eeks

Figure 26. Proportions of all patients who were on CAPO for less than three months or between three
and 12 months during 1982 which had 0, 0-2, 2-6 and 6 plus weeks hospitalisation for peritonitis during
1982.
 657

CD
M F <12mon >12mon

f'72J (77A Other ~ IZZ3

In'7!
= Family Request IZZl I:ZZI

IZI r:zI Patient Request [:ZI [77;1

f'72J ~ Unable to Cope

~ IZZ3

~ ~
Inadequate
I«J DialysIs ~

U Abdominal
Complications
h

~ =10%

n 629 423 451 255

Figure 27. Reasons for abandonment of CAPD in 1982 in male and female patients and those treated
for less than and more than 12 months.

There appears to be no striking differences between the proportions of diabetic

and non-diabetic patients who attributed the abandonment of CAPD to these
reasons, nor is there any marked difference between the age groups, although
younger patients appear more likely to request the cessation of CAPD (Fig. 28).
Whether CAPD was the treatment of first choice or not did not influence the
reasons for abandonment. Peritonitis and other abdominal complications ac-
counted for 70% in both groups (Fig. 29).
The largest proportions of death on CAPD in 1982 were caused by cardiac or
vascular complications which include myocardial infarction, hypertensive and
other causes of cardiac failure, pericarditis, pulmonary embolus, and cerebro-
vascular accidents (Fig. 30). Cardio-vascular causes accounted for half of all
deaths on haemodialysis, the slightly lower proportion attributed to these reasons
amongst CAPD patients being due to the larger proportion of death from
infective causes [3]. A significant number of deaths were attributed to peritonitis,
especially in patients on treatment for over 12 months in whom over 15% of
deaths were caused by peritonitis. Patients on treatment for less than 12 months
were more likely to request withdrawal of their therapy, commit suicide or stop
therapy for other reasons. Almost 5% of deaths were due to these 'social' causes
in this group.
658

D.M NonQM 15-34 55-64

IZZZI ~ Other rzzI rm

IZZZI cz::z:I Family Request II::Z:OI ICI:Z:I

~ ~ Patient Request ~ ~

~ f23 Unable to Cope ~ ~

f:a ~
Inadequate
DialysIs ~ ~

~
Abdominal
Complications
U
Peritonitis

~=10'lo
n 92 1,035 141 490

Figure 28. Reasons for abandonment of CAPD in 1982 in diabetics and non-diabetics and in patients

•
aged 15-34 and 55-64.
Not
I'tchoice I'tcholce
1190 pIs) (118 pIs)
Family's
~ __ Request

Patient's
~D Request
Inability
~D to Cope

D Inadequate
Dialysis

C
II
E
c
Peritonitis
o
't>
C

"
.0
~

LI Other
Abdominal

D Do OTHER

Figure 29. Reasons for al]andonment of CAPD in 1981 in patients for whom CAPD was treatment of
first choice and not first choice.
..."*'30
On CAPO for
659

E2lI <12 Months, n = 348

m >12 Months, n=316

g 20
o
'"
"0

c
'"
u
~
CL 10

o~ Uncertain Cardiac Vascular Infection Pentonilis Hepatic Social Mlsc Malignancy Other

Figure 30. Causes of death on CAPD in 1982 according to whether the patients had been on CAPD for
less than or more than 12 months.

In the management of patients on CAPO, therefore, peritonitis has been

documented as the major problem. The scored number of episodes mark painful
illnesses punctuating the patients' lives. There is some indication in our Registry
data that as centres become more experienced they keep these events to a
minimum and do not have to hospitalise patients for so long [4]. Peritonitis
remains the commonest reason for having to abandon CAPO and also takes its
toll as a cause of death. These findings emphasise that CAPO requires back-up by
haemodialysis and should playa part in integrated treatment programmes rather
than be offered as sole available therapy in any programme or for anyone
particular patient.

6. Survival of patients on CAPD

Patient survival was calculated using actuarial methodology according to the

convention adopted by the Registry. The follow-up clock was zeroed from time of
commencement of CAPO and antecedent therapy was discarded. Age-related
survivals were calculated using age at commencement of CAPO. Patients were
censored as 'lost to observation' from the date of any change in therapy. By this
method of calculation the numbers at risk diminished more steeply with each
interval than if patients were included from the date of commencement of CAPO
irrespective of any subsequent changes in therapy. We have termed the former
660

convention survival on and the latter survival after (starting) CAPD [1]. Both
methods make assumptions concerning deaths which occur after the patient has
changed to another mode of therapy (e.g. haemodialysis or transplantation) and
which might be due to complications acquired during CAPD, such as peritonitis.
The period immediately following return to haemodialysis from CAPD is one of
high risk and mortality is greater amongst patients returning to haemodialysis
after (failed) CAPD than it is in patients starting haemodialysis without prior
CAPD [1].
Interval mortality is the converse of survival. We have chosen to display
interval mortality because it demonstrates the time on CAPD at which mortality
has occurred and shows more clearly than cumulative survival curves whether the
rates have changed with duration of therapy.
Treatment change rate is analogous to interval mortality and shows the inter-
vals at which change in therapy occurred most frequently. The computation
employs actuarial methodology, the patients at risk being diminished at each time
interval by 50% of those 'lost to observation' during the interval. We have chosen
to treat death on CAPD as reason for censoring the patients as 'lost to observa-
tion' from the date of their death.
Calculation of survival, interval mortality and treatment change rates given
here were based on the first period of CAPD in each patient's treatment se-
quence. Second or subsequent periods after a recorded interruption on IPD,
haemodialysis or transplantation were not included. Instructions accompanying
the questionnaires indicated that temporary changes in therapy should not be
recorded and analysis showed that very few patients had interruptions in their
CAPD of less than one month reported.
Non-diabetic patients had a significantly better survival on CAPD compared to
diabetics with over 70% survival at two years (Fig. 31). The small number of
patients who had switched from haemodialysis to CAPD had a worse survival
than 'all' CAPD patients considered together. Amongst the former patients are
those who were forced to switch because of vascular access or cardiac problems
and this probably explains the increased mortality in this group.
Diabetic patients on CAPD had an increased mortality rate at each interval
compared to non-diabetics in whom the mortality rate remained constant (Fig.
32). We have previously noted an increase in the interval mortality rate at about
15 to 18 months [1], and this can be seen to be particularly accentuated in diabetic
patients.
Patient survival on CAPD was not unexpectedly related to age (Fig. 33) but
even those patients who were over 65 years when they commenced CAPD had a
two year survival of 60%.
The patient survival when analysed according to the year that CAPD was
started has shown poor initial figures with a deterioration in 1980 and a steady
improvement since [4]. This accounts for the bad result shown in our initial
reports ofthis mode oftherapy [6, 7]. Moore et al. [4] reported a manipulation of
 661

•
0 - - 0 NON DN 'ALL' N=6,429
e - - e DM 'ALL' N= 992
100 O-----IJ NOll DMSwl rCHED N= L[,23
e----e DM SWITCIf~D N= 140

90

80
%
SURVIVAL
7C

50

50

110
~

0 I I I I I I
0 3 )2 1) B 21 LI~

Momll:

Figure 31. Patient survival on CAPD in 'all' diabetic and 'all' non-diabetic patients compared to
survival in those who had switched from previous haemodialysis treatment.

• 10 -
o NON-DIABETIC

~ DIABETIC

u
0
-
%
ltlTERVIlL

6-
'iORTALITY
RATE

4 -
2 -
0
--
3
~

6 --
9 --
12
MONTHS
~

IS -- -- --
18 21 24

Figure 32. Three-monthly interval mortality rate in patients on CAPD: diabetic and non-diabetic
patients compared.
662

CD
100

gO

80
%
SURVIVAL

70

£0

50

o r
0
I
G
I
12
I
15
I
18
I
21
I
24
MONTHS

Figure 33. Patient survival on CAPD according to age at start of treatment.

EDTA Registry data in which CAPD patients were divided into two subpopula-
tions, one group located in 'experienced' centres which were arbitrarily defined
as those who commenced their CAPD programmes before 1st January, 1981 and
who had 25 or more patients alive on CAPD on 31st December, 1982, and a
second group located in the remaining centres, termed 'inexperienced'. The
comparison was instructive (Table 4). In 'inexperienced' centres survival for
patients who commenced CAPD in 1978 and 1979 was poor when compared to
those treated in 'experienced' centres in the same years. In 1980 both groups of
centres had notably poor results. The poor figures for 1980 appear to coincide
with a rapid expansion in the CAPD population. In both groups of centres
survival subsequently improved and the figures for 1982 show the best 12-month
survival to date.
Patients who changed to an alternative treatment, whether this change repre-
sents failure of CAPD or an elective move to a preferred mode of therapy were
more likely to do so in the first three months following commencement of CAPD
(Fig. 34). The treatment change rates (TCR) calculated at three-month intervals
were higher for diabetic patients than non-diabetics but in both groups there was
a tendency for the rates to decline with time except for a slight increase in
diabetics at 15 to 18 months.
Treatment change rates analysed according to the year that CAPD was started
demonstrate clearly that 1980 was the least successful year because of high initial
TCRs although the later rates for 1980 patients were better (Fig. 35). The TCRs
 663

Table 4. Patient survival in 'experienced' and 'inexperienced' centres according to year in which
patients started CAPO (reproduced from Moore et al. [4].

Per cent patient survival

Year Nos. at risk Months

3 6 9 12 15 18 21 24

"0 1978 204 92 88 82 76 73 71 70 68

<!)

c<!) 1979 427 96 91 87 85 80 76 73 71

u

.t: 1980 1239 96 92 87 83 78 72 68 67

<!)
Q.. 1981 1554 96 92 88 83 78 74 71 65

-
:<
c 1982
<!)
1396 97 93 88 85 + + + +

"0
1978 92 99 97 93 91 91 88 86 +
<!)
u 1979 190 98 96 96 95 94 93 88 82
c<!) 1980 432 97 94 84 79
.t:
92 89 75 71
<!)
Q..
1981 649 96 94 91 89 87 83 80 +
:<
UJ 1982 540 96 93 93 91 + + + +

+ : Numbers at risk less than 30 .

• 10
o NON -DIABETIC
~ DIABETIC

8
%
TREATMENT
CHANGE b
RATE
4

0
3 6
MONTHS

Figure 34. Three-monthly treatment change rate in patients on CAPO: diabetic and non-diabetic
patients compared.
664

8
TREATi-lE T

o
MoNTHS 3

Figures 35. Three-monthly treatment change rate in patients on CAPD according to year started , 1978
to 1982.

for 1978 and 1979 were consistently worse than for 1981 and 1982 and , in particu-
lar, demonstrate higher rates in patients on treatment for 18 to 21 months . 1982
has been the best year yet with a treatment change rate for the fourth three-month
interval of less than 0.25%.
From the above it is now possible to conclude that patient survival on CAPD
has improved since the early years' experience of this method in Europe [6, 7] .
Calculations based on pooled data from nearly 600 European centres give a
12-month patient survival of 87% for all patients who commenced CAPD in 1982.
Treatment change rates for 1982 patients were around 1% per three-month period
resulting in a technique survival better than 95% at the end of the year.

7. Summary

CAPD now makes a significant contribution to the treatment of end-stage renal

failure in Europe, but the extent of the integration of CAPD into the dialysis
programmes of different countries varies . For selected patients CAPD is widely
accepted as the treatment of first choice and it is also an important back-up to
haemodialysis in other cases. After the poor results in patients put on to CAPD in
1978 to 1980 the patient and technique survival rates have improved so that for
 665

patients who commenced CAPO in 1982 they were comparable with those
provided by alternative treatments.

Acknowledgements

The Registry gratefully acknowledges the collaboration of the directors of the

European Dialysis and Transplant centres who have contributed their data year
by year. It is funded partly by the parent Association, partly by European
governments and partly by commercial companies amongst whom Travenol
Laboratories have particularly supported work on CAPO.

References

1. Wing AJ et al.: Combined Report XIII, 1982. Proc EDT A - ERA 20: 2, 1983.
2. Wing AJ et al.: The contribution of continuous ambulatory peritoneal dialysis in Europe. ASAIO J
6: 214, 1983.
3. Wing AJ et al.: Cardio-vascular related causes of death and the fate of patients with reno-vascular
disease. Proc Strasbourg Conf on Cardio-vascular aspects of renal failure, 1983. Ed Jahn (in press).
4. Moore R et al.: Comparison of the results of CAPD treatment in 'experienced' versus 'inex-
perienced' centres in Europe. Am J Kid Dis (in press).
5. Brunner FP et al.: Combined Report Y, 1974. Proc EDTA 12: 2, 1975.
6. Jacobs C et al.: Combined Report XI, 1980. Proc EDT A 18: 2, 1981.
7. Kramer P et al.: Combined Report XII, 1981. Proc EDTA 19: 2, 1982.
20. Quality of life and psychosocial aspects of
chronic peritoneal dialysis
R.M. LINOSA Y, H.J. BURTON and S.A. KLINE

1. Introduction

In recent years the increasing use of peritoneal dialysis, particularly continuous

ambulatory peritoneal dialysis (CAPO), has been an important factor in the
management of patients with end-stage renal disease (ESRO) outside the hospi-
tal environment. While the efficiency of this innovative peritoneal technique is
well documented with numerous publications describing physiological and pro-
cedural advantages, little attention has been given to the influence of psychologi-
cal and social factors upon the rehabilitation and quality of life that the CAPO
patient experiences.
It may well be that psycho-social factors are more important than physiological
ones in determining outcome measures, such as success, failure, and even sur-
vival, of the home dialysis patient [1, 2]. There is no question that the success rates
of various forms of home or self-care dialysis programs are extremely good; but
equally there is no doubt that such programs are associated with considerable
stresses to patients and their families. The appreciation of the stresses that such
patients are constantly encountering and, hence, the possibility of their preven-
tion or correction may contribute to the well-being of individual patients and,
hence, to the success rates of regional home dialysis programs.
By the end of 1982, 46.5% of all dialysis patients in Canada were on some form
of home or self-care dialysis program. This very substantial figure makes Canada
a world leader in the use of home and self-care dialysis and this success is, in part,
due to the advent of CAPO. Indeed, 58.8% of the home or self-care dialysis
population were being treated by this modality at that time [3]. It has been
estimated that the cost of home hemodialysis is perhaps one half that of in-center
hemodialysis. While the true costs of CAPO are not yet established, it has been
estimated that CAPO is at least as cost-effective as home hemodialysis [4, 5].
Thus, in the current climate of economic restraint, home and self-care dialysis
programs must expand if all suitable patients with ESRO are to be treated.
Further, study of the Canadian Renal Failure Registry [3] shows that survival
668

for non-diabetic ESRD patients at 2 years is similar for hemodialysis and CAPD
(80%); on the other hand, excluding transplantation, there was a significant
patient drop-out from CAPD (35%) over the 2-year period. These patients
mostly returned to (high cost) in-hospital hemodialysis units. Thus, only 50% of
patients who commence CAPD will be on that form of therapy in 2 years. The
Canadian data is similar to that reported by the US Peritoneal Dialysis Registry
[6] and to the latest European data [7]. If home and self-care dialysis programs
have to expand (for economic reasons) then drop-out rates must be reduced.
A currently ongoing Ontario study is designed to identify all factors which
influence both outcome, and quality of life on home dialysis. The study incorpor-
ates two stages: a retrospective phase to include patients who commenced a home
training program between 1975 and 1978; and a prospective phase following
patients entered into such programs between 1978 and 1981, with continuing
follow up to a maximum of 60 months.
Data from the retrospective phase of this study [8] suggested that the success
and failure rates for CAPD and home hemodialysis were very similar over a short
period (18 months) when corrected for age and sex. This data showed a 15%
failure rate, a 66% success rate, and 19% death rate in the CAPD-treated group.
Current life-table analysis of data from the prospective phase of this study (RM
Lindsay and HJ Burton, unpublished data) shows that the failure rate (death plus
drop-out) from CAPD is approximately 10% higher than that from home hemo-
dialysis at all time periods beyond 12 months. The data parallels the Canadian
Registry figures showing a 50% failure rate (for CAPD) at 24 months and by 48
months the failure rate is approximately 80%. Further analysis of this population
clearly shows that age and risk (diabetes, vascular disease, etc.) are strong
influencing factors of outcome. Unfortunately, attempts to control data for age,
sex, and risk factors simultaneously are difficult at present because of small cell
sizes. As the study proceeds better information will become available. It is the
authors' opinion, however, that if such variables are allowed for then CAPD and
home hemodialysis treatments are interchangeable as far as outcome is con-
cerned. Furthermore, the Ontario study also indicates that over the short term (1
year) there is little difference in any hematological, biochemical or clinical
parameter between patients on CAPD or home hemodialysis [9].
On the other hand, the Ontario study has shown that the type of dialysis
treatment used may influence the quality of life experienced by individual pa-
tients and that this, in turn, may influence outcome (success, failure, or death) [1,
2]. Thus, it is suggested that the choice of a particular treatment modality should
take into account the life style and personality profile of a given patient and such
an approach may improve the overall success rate of home dialysis programs.
Having stressed the importance that psycho-social factors have on the success
of home or self-care dialysis programs a detailed examination of such factors is
now necessary. Thereafter, the chapter will attempt to give practically useful
information regarding the determinants of success in long-term CAPD patients
 669

together with suggestions as to how to choose a dialysis therapy for a given

patient.

2. A psycho-social perspective on ESRD and its treatment by dialysis

While the issue of quality of life for those affected with kidney disease has been
generally ignored in the health psychology literature [10, 11] it is of special interest
for those in the field of nephrology [12]. Concern by members of the renal team as
to the quality of life of their patients is not the least surprising considering the
multitude of problems renal failure inflicts upon the patients and their families.
The dependence on artificial life support devices, adherence to a strict medical
regime, an increase in financial hardships and a decrease in physical and social
functioning frequently results in feelings of depression, anxiety, worthlessness
and hostility. Singularly or collectively they can cause major life style disruptions
and require a considerable psychological and social accommodation.

2.1. Living and dying

The dilemma for the renal patient is a difficult one. It is a dilemma that concerns
itself as much with living as with dying. On the one hand, the patient fears life will
be shortened by premature death.! He begins to question the value of continued
existence and wonders whether life is worth living. On the other hand, he is
fearful that if he continues to live by artificial means life may not be acceptable to
himself and hi~ family. He also fears that his physical health and his level of
activity will decline and that his daily existence will be plagued by the constant
threat of death and its accompanying stress. Peritoneal dialysis, he knows, will
prolong his life but it is hard for him to accept that it will not return him to his
former level of functioning.
For most patients the conflicts of living and dying are resolved through positive
adjustments. For others, the conflict is solved by suicide. Resolution by suicide is
not surprising if one accepts the premise that all people have self-destructive
mechanisms [14]. Everyone controls these mechanisms until the environment
invites one to stop living, an invitation that usually takes the form of a subtle
message questioning the meaning and value of one's very being. It is theoretically
simple for ESRD patients to end their lives because of the intrinsic nature of their
disease and their potential control over their immediate environment; what is
remarkable is that so few choose to terminate life.
Suicide ideation and behaviour, none-the-Iess, are serious problems among
dialysis patients [15]. Self-destruction can be a preoccupation and those that are
so preoccupied typically have difficulty facing the implications of their illness and
question constantly the value of continuing to struggle for a minimal existence.
670

The notion that patients often think about taking their life is generally accepted
by the treatment team. They are in less agreement, however, as to the number
who actually do. One estimate, often quoted, reported suicide rates for dialysis
patients as 100 to 400 times that of the general population [16]. More recently the
rate has been estimated to be 4 to 5 times greater than population norms [17].
Rate discrepancies are to a large degree based on differing opinions as to what
constitutes genuine suicidal behaviour. Diet violations, for example, often are
seen as the primary means of self-destruction, but there is some question whether
they should be evaluated from a suicidal framework.
Irrespective of how abuse of treatment regime is interpreted, the fact remains
that significant numbers of patients find being chronically ill and in need of
constant medical care almost as intolerable as the fear of imminent death. Their
simple, almost fatalistic, attitude paradoxically seems to act as an adequate
psychological prop against lethal self-destructive behaviour [18].

2.2. Compliance vs non-compliance

Adherence to a medical treatment is a major concern for most chronically ill

populations. A review of the literature shows compliance to have a positive
correlation with social support, knowledge of disease or therapy, stresses and
strains, patient motives and perceived utility, and the demographic variables of
age, education, social class, income, and occupational status [19].
Non-adherence, especially to nutritional requirements is one of the most
significant medical complications attendant on long-term dialysis. Dietary restric-
tions not only block the accustomed stress outlets such as eating, alcohol con-
sumption or smoking excessively, but also require a major readjustment from
previous habits of consumption. The consequences of non-adherence as we have
noted can be serious. For a number of patients it becomes the vehicle for acting
out conflicting feelings towards dialysis and it often jeopardizes their lives. This is
equally true of CAPD as for hemodialysis. CAPD may be less complex and the
dietary and fluid restrictions less limiting, but it still requires major changes in the
patient's life style and coping strategies. Data from the Ontario 'Adaptation to
Home Dialysis' study indicates that the achievement of compliance positively
influences physiological status and reduces dialysis-related stresses [20].
It is commonly accepted that all dialysis patients have problems complying. At
some time they cheat by not adhering to nutritional requirements or by altering
their dialysis procedures. Variability in willingness to comply can, in part, be
attributed to personality factors, low frustation tolerance and primary and sec-
ondary gains from the 'sick role'2 are common factors associated with non-
compliance [21]. Physicians can use personality features to assure successful
outcomes for CAPD patients. Meticulous even obsessive-compulsive patients,
for example, do well with their regular charting of exchanges, blood pressure and
 671

drugs, as well as with thorough care of the catheter [22]. Case studies have
reported that patients who are compulsive and compliant and who manage their
treatment accordingly, do extremely well on CAPD [23].

2.3. Loss and threat of loss

2.3.1. Employment
Actual loss of employment for ESRD patients varies from patient to patient. The
threat of loss, however, is always present. The majority of the patients experience
a reduction in occupational status within their community and significant num-
bers are forced to either alter their job responsibilities or leave or change their
employment. For those who are engaged in occupations which are less physically
demanding and/or amenable to flexibility, it is not as difficult to maintain employ-
ment [24]. In certain occupations the necessity for regular bag changes creates a
problem for those on CAPD [25].
It is difficult to draw firm conclusions as to the vocational rehabilitation of
peritoneal patients, since most studies include those on hemodialysis [26-30].
One study that included a substantial number of center and home peritoneal
patients showed only 40% of the patients were employed. Of those employed
64% were white collar workers; 36% were unskilled labourers [31]. In the
Ontario study on home dialysis (retrospective phase) of which 37% of the
patients were'on peritoneal dialysis, 47% were employed. Of those employed,
half were males, half were under 40 years of age, and 75% were married. This
suggests that a large number are unemployed due to the disease and treatment,
and not by choice [8].
Analysis specifically of the CAPD patients revealed that only 8% were able to
remain at the same premorbid job; 20% moved to a different job and 72% had to
leave their place of employment because of their chronic illness. Almost 75% of
those employed worked less than a 40-hour week.
Those patients who remained employed were then analyzed according to their
treatment modality. Those on CAPD reported more problems in obtaining salary
increases, and in receiving and accepting promotions. On the other hand, they
experienced fewer problems with their employers and fellow workers and were
less likely to transfer to a new job. In terms of continuity of employment before
and after ESRD failure therapy, those on hemodialysis had greater job stability
than those on CAPD. This was equally true of housewives who were more likely
to resume their household duties [9, 32]. These findings when published were the
subject of much controversy but have since been substantiated [33, 34].
For ESRD patients rehabilitation generally includes plans by the staff and the
patient for their return to full-time employment. The goal of full-time employ-
ment inherently reflects the North American value that work is an important
indicator of well-being, and therefore, is viewed as a treatment goal worth
672

striving for . Unfortunately disincentives to work related to eligibility for benefits

is a counterproductive force which frequently undermines the efforts of voca-
tional rehabilitation.
A case in point are those individuals who in the initial stages of treatment and
while adjusting to their condition choose to receive municipal, state or federal
assistance or long-term disability benefits. At some time in the future if they seek
out part-time work as an initial step to full employment, they and their families
can face financial difficulties. Many governmental programs are inflexible and fail
to consider earnings from part-time work. Benefits like medication and transpor-
tation may have to be assumed by the patient when he becomes employed. These
bureaucratic disincentives when compounded by the time requirements of di-
alysis and the possibility that a bag change may need to occur on the job often
create problems for the patient that far outweigh the advantages of being em-
ployed.
Almost one-third of the CAPD patients in the Ontario study reported a
disability pension, private pension, or some other governmental assistance pro-
gram as their primary source of household income. Twenty-two percent had no
personal income and only 20% reported a personal income over $12500 [8].

2.4. Body image, self-image and self-esteem

The body image is the central and personal representation of body parts and of
the body as a whole. It influences what a person does and does not do, attitudes
and opinions, even more so than would the realistic image of the body [35].
Perceptions of distorted body image tend to arouse great emotional tension, to
disrupt habitual patterns of daily behaviour, to reduce mental efficiency and
produce painfully unpleasant effects [36].
According to Deutsche [37], the stress associated with fear of loss of bodily part
is rarely, if ever, definitively solved. For patients with chronic renal failure these
fears are particularly common. Loss of body function or part of the body has been
recorded as a major cause of stress by Cummings [38], De-Nour and Czaczkes
[39], and Wright and colleagues [38-40]. Abram [41], Kestenbaum [42], Shea and
co-workers [43], and Kemph [44] have also given special attention to the unplea-
sant effects that follow disturbance in body image.
Patient-machine relationships and their influence on body image received
much attention in the late 1960s and early 1970s as did concern with the ar-
teriovenous shunts and their effects on general appearance [39]. While the
problems related to body image are still common they occur less frequently and
with less severity as in the earlier days of dialysis. Changes in body image,
however, still remain a major source of stress for those who are attached to
artificial devices or have kidneys removed [45]. Diminution of body image is also
evident for those on CAPD. It occurs as a frequent reaction to the presence of a
 673

permanent catheter, the wearing of an empty CAPD bag [25], the noises pro-
duced by the fluid and the sensation of feeling full [22].
Distortions of the self-image are quite common in dialysis patients. Their
perception of themselves in relation to their environment is a major determinant.
This is especially true as it relates to shifts in social roles, patient-staff interac-
tions, and the physical changes related to renal failure. Perceptive clinicians are
cognizant that increased dependency on family and staff and perceived loss of
stature are particularly difficult for those patients who put a high premium on self-
sufficiency and independence. Loss of self-image is especially a problem felt by
previously independent males [46]. Self-image is negatively correlated with an
ability to adequately perform employment or household tasks, economic status,
aspirations for the future, physical appearance, and the feeling of attractiveness
[38].
As for loss of esteem, a number of events are contributing factors. Some people
associate it with the loss of intellectual functioning secondary to uremic toxins,
with feelings of estrangement and loneliness and with the loss of interest in the
immediate environment [38, 47, 48]. Others maintain self-esteem is seriously
affected by loss of membership in social groups, failures of plans or ventures,
decrease in financial and occupational status, frustation of drives, and by the role
reversal which frequently occurs as a consequence of inactivity [40, 41]. Shame
and feelings of failure and weakness are also concomitant with feeling of loss of
self-esteem [50].
In summary, peritoneal dialysis patients behave similarly to those on hemo-
dialysis and to others with chronic illness. Each patient handles the threat of loss,
or actual loss, of body image, self-image, and self-esteem as an outgrowth of his
previously existing personality development, of the ways in which he coped with
previous conflicts and anxieties, and the meaning he attaches to this current
illness [51].

2.4.1. Sexual relations

Information of patients' sexual relations was sparse in the early years of dialysis.
More recently, a wealth of data on this subject has been published and a clearer
picture of that aspect of adjustment does emerge. The most frequently cited
observation is the deterioration of sexual function in both sexes after dialysis
despite improvement in general physical condition and in hormonal balances
[52]. Men and women both report a decrease in libido and a marked reduction in
frequency of intercourse [53]. Significant correlations have been reported be-
tween potency and libido, and vocational rehabilitation, depression, self-image,
sick role, dominance, and age at the time of diagnosis [54-56].
Our experience from the Ontario study is that both CAPD and hemodialysis
patients complain frequently about marital and sexual concerns [57]. Hemo-
dialysis patients, however, experience significantly more problems of sexual
dysfunction than those on CAPD. They are more concerned by their loss of libido
674

and avoid sexual activity out of fear of damage to their blood access devices. This
appears especially true of females. Compared with CAPD patients, those on
hemodialysis also complain more often of loss of sexual identity and attractive-
ness. Even those between 60 and 79 years of age on hemodialysis indicate greater
changes in sexual activity and loss of sexual attractiveness than their counterparts
on CAPD.
These findings should not minimize the unique sexual problems faced by
patients on CAPD. Sexual difficulties do arise from the belief that the 'bag' is ugly
and undesirable [22]. The Ontario study showed a 30% increase in those who no
longer engaged in intercourse after having commenced CAPD and there was a
four-fold increase of men who reported difficulty maintaining an erection [58].
It is the authors' opinion however that the se]{ual functioning of dialysis
patients can be normal when those factors that interact and influence sexual
functioning are also normal. Sexual activities will continue only if the patient is
physically sound, his or her marital relations are satisfactory and if psychiatric
complications, especially depression, are minimized.

2.4.2. Sociallleisure activities

Social participation and deployment of leisure time are non-work activities
known to be strongly related to a sense of well-being [59]. This is a belief held
widely by renal staff. Those dialysis patients and families who pursue outside
interests are seen as better adjusted emotionally. The staff believe outside
interests provide the opportunity for those on dialysis to become absorbed in
activity which distracts from self-preoccupation and excessive introspection.
Keeping personal concerns at a distance, they argue, appears to recharge the
patient's emotional and intellectual apparatus to deal more effectively and objec-
tively with his/her personal problems. This assumed relationship between social
activity and dialysis adjustment has yet to be proven.
Evidence to date indicates that social interaction and recreation are modified
by the dialysis regimen and are among the greatest threats to patients' sense of
well-being [60, 61]. In one study the authors found that roughly 50% of patients
were involved in sports, or active recreation including swimming, golf, hunting,
and fishing [31]. Other studies have shown a marked decrease in such extra-
familial activities because of the complex treatment regimen and its effects [62,
63]. A comparative study of home and in-center dialysis patients reported a
significant decrease in social functioning in both groups [64].
In the Ontario study we compared leisure time activities at the start of home
dialysis and after three months. CAPD patients showed little decrease in physical
activities as compared with hemodialysis patients. Those on CAPD spent signifi-
cantly less time on sedentary activities while those on hemodialysis spent more
time. Neither group, however, altered patterns of sociability as a result of
dialyzing at home [65].
All available data support the hypothesis that social activities for those with
 675

chronic kidney disease are limited. The majority continue with passive recreation
- listening to the radio, reading newspapers or watching television. Few are
inclined to socialize out of the house and their participation in clubs and associ-
ations is almost non-existent. Despite the gloomy picture there is a ray of hope.
Those who continue in their occupation also continue with their leisure activities
without a great change [66, 45].

2.5. Coping with illness

2.5.1. Anxiety and depression

People respond to illness and disability with varying degrees of frustration and
pleasure. In general, physical disability may be considered as a negative, frustrat-
ing, and anxiety-provoking event. Individual reactions to the loss of kidney
function and the debilitating consequences of this loss support this generalization.
Anxiety and depression are among the most frequently observed reactions of
dialysis patients. This is particularly evident during the first three to six months
after beginning dialysis when the majority of patients experience intense aware-
ness of and, vulnerability to, the stresses inherent in their situation [41-44].
Much of the anxiety is related to the fear of dying, fear of peritonitis, psycho-
logical stresses, financial worries, changes in life style, and the distance between
the patient and the hospital support staff [67]. The Ontario data shows that
females are more likely to be anxious than are men. This holds true irrespective of
modality. Females on CAPD, however, are significantly less anxious than those
on hemodialysis, and show a marginal decrease in their anxiety during the initial
home experience. This is in contrast to the marked increase in anxiety for females
on hemodialysis [57]. These findings have implications for patient selection for a
home program.
Another means by which dialysis patients often adapt to stress and anxiety is
through depression. Depression is one of the most serious emotional problems
associated with chronic renal disease [68, 7]. Levels of depression vary consider-
ably. Elevated depression was characteristic of ESRD patients in the 1960s.
Lefebvre and his colleagues observed that depression may be manifested by a
'giving up' syndrome in which patients no longer actively participate in their own
care, and disregard their medical instructions [70]. In these circumstances severe
complications and death frequently occur. Depression as the primary problem of
dialysis is currently under review [71]. A number of studies more recently have
failed to find elevated levels in their dialysis subjects [72-78].
There is disagreement as to be reasons for the existence of depression. Depres-
sion, some have argued, is a function of patients' realization of the complex
responsibilities in their 'new life' [79]. Others state it is a gradually developing
entity, evolving from dialysis patients' continuous struggle to cope with day to day
stresses [80]. It is also speculated that depression originates from two sources,
676

guilt over regime violations and shame related to physical appearance, difficulties
caused to others, and to the nature of the illness [81]. Regardless of etiology, most
patients suffer severe depressive reactions, even after they have been stabilized
medically through dialysis. The depressive reactions are usually accompanied by
apathy and suicidal ideation. They are of sufficient severity to impede the
patient's cooperation in self-care and adherence to the medical treatment to
which they are assigned.
The Ontario study indicates ESRD patients are significantly more depressed
than the general population [57], that depression increases marginally during the
initial home experience and that depression is not a factor of treatment selected
[57]. Of vital importance to the treatment team is the finding that among all
psycho-social parameters examined, depression distinguished more clearly those
who failed, succeeded, or died on a home program [1, 2].
From the literature it is apparent that depression creates management prob-
lems of great magnitude. If not resolved it may endanger patients' lives. ESRD
patients with marked depression grow chronically more preoccupied with feel-
ings of unworthiness, failure or hopelessness. They lose initiative and interest and
lapse into repetitive expressions of futility with their situation. Is it no wonder
that for a significant number denial becomes the major mechanism of defence.

2.5.2. Adaptive vs maladaptive denial

To some extent all patients use denial [41, 49, 82, 83]. Denial, nonetheless, has
adaptive aspects. Not only is it a necessary tool for coping with uncertainties and
unfortunate realities in their lives, but also it is a useful effective mental mecha-
nism to help patients deal with their continuing unsatisfactory physical condition
[84].
Massive denial is a maladaptive mechanism which, in some cases, is so extreme
that it disrupts basic reality. It can cause a patient to view any sort of psychiatric
intervention with suspicion, negativity, and indifference. Excessive denial is a
potential killer when the individual refuses to accept the demands and limitations
of his disease. There are case studies that suggest massive denial can be of
delusional proportions [85]. Denial may prevent the patient from hearing instruc-
tions and information, from recognizing emotional conflicts and difficulties and
may cause the patient to ignore necessary medical procedures.
The use of denial to cope with chronic renal failure can be a blessing for the
patient and no attempt should be made to substitute a more realistic attitude [62].
Used up to a point, denial permits the patient to live with his illness. It is a
characteristic reaction to stress and in fact is used more by dialysis patients than
any other group of chronically ill. Denial should be viewed as a constant mecha-
nism which protects the patient from experiencing intensive feelings of helpless-
ness when they are depressed. It is our experience that denial reduces the
appreciation of stresses and concerns and indeed was found to positively correlate
with success on home hemodialysis in males over the age of 45 [1].
 677

2.6. Stresses and concerns

The importance of recognizing the role of stress and concerns relative to adapta-
tion to chronic renal failure is clearly documented in the literature. At least six
factors are described as sources of stress: losses, restrictions, dependency, an
increase in aggression, threat of death, and changes in body image and self-image.
For the most part, little attention has been given to the stresses associated with
the treatment process and activities of daily living. Recently data from the
Ontario study has been published identifying stresses associated with home
dialysis [8, 9, 32].
Contrary to what one would expect CAPD patients were found to be more
concerned about having to alter or cancel vacation plans (62%) than their feelings
of dependence on others (51%), their financial problems they were facing (59%)
or their decrease in sexual ability (48%). They were least concerned about
marriage strain (11%), loss offamily roles (29%) and care of children (19%).
A similar analysis of peritoneal dialysis-related stresses show that fear of
infection (50%) was of paramount concern, followed closely by feelings of being
physically weak (49%), fear of death (24%), cramps (23%), and an inability to
sleep (20%). Headaches (9%), itchiness (11 %), fear of blood clotting (12%), pain
during dialysis (14%), dietary restriction (17%), and fear of losing dialysis sites
(18%) were mentioned less frequently.
Patients on home hemodialysis were likewise studied for their appreciation of
stresses and concerns. Comparative analysis by treatment modality showed that
the frequency of stresses and concerns is much more relevant to those treated by
hemodialysis, especially in the younger age group. It was interesting to note that
older patients experience less stress from their dialysis treatment. While males in
either group did not differ in the frequency of experienced stress, men on CAPD
did indicate less concern. Females on hemodialysis reported significantly more
stress and concern than those on CAPD.

2.7. Dependency vs independency

Dependence on dialysis, on the medical tean, and on family members is a major

source of stress and a likely encumberance to successful adaptation to renal
failure. All dialysis patients, are reliant on someone or something for assistance
and emotional support. Contingent on the degree and nature of his physical and
emotional impairment, the patient may be unable to take care of many of his
medical needs and become dependent. Most patients handle this situation well.
Some react through passive dependence and others deny the realities of the
situation.
Those who adopt the dependent role obtain gratification in being cared for and
resist learning to care for themselves. Rarely, in extreme cases, patients may
678

become almost parasitic and find themselves in a role at odds with the strongly
valued attribute of 'self-reliance'.
In attempts to develop self-reliance, dialysis patients should be reminded that
much of their overall physical and mental adjustment depends upon how well
they accept responsibility of their own treatment. Paradoxically CAPD purports
to offer total patient independence, yet large numbers of patients have dialysis
assistants and complain of being dependent on others [8].

2.8. Family and marital relationship

Families are invariably affected by the illness of the dialysis patient, and must
learn to cope with the latter's fluctuating health, physical weakness and threat of
death. They may also face a loss of income and have to alter or forego vacations.
Families often deny or refuse to accept the realities of the illness [87].
Adjustment for the spouse is particularly difficult. Many are troubled by
depression and grief and some feel hostility because of the patient's increased
dependency [88]. The spouse occasionally tries to assume multiple roles in the
family, including those of the patient and can become tired and depressed [88].
Fortunately, the majority of families are able to cope. A comprehensive study
of center and home dialysis showed that nearly 50% of the patients felt the quality
of their marriage had not been affected by dialysis. Indeed approximately 40%
thought the illness had strengthened their marriage. Only 6% reported the illness
was responsible for their separation [31]. In the Ontario study, CAPD patients
indicated more support from their household members and greater satisfaction
with interspousal relationships than did their counterparts on home hemodialysis.
Yet, they were as dependent on their spouses as were the patients on hemo-
dialysis [57].

3. What factors determine successful adaptation to long-term CAPD?

In an attempt to answer this very important question the authors have used
information from the ongoing prospective phase of the Ontario Home Dialysis
Study (RM Lindsay and HJ Burton unpublished data). Data on physiological and
psycho-social functioning of patients at the end of their first 3 months of CAPD
therapy was taken and was examined in relationship to two end-points. One end-
point was 'successful adaptation' to CAPD as of December 31, 1983; 'successful
adaptation' implied that the patient was still at home on CAPD 17 to 48 months
after commencing dialysis and that the patient described himself (herself) as
doing 'very well' taking 'everything into account' (physically, socially, emo-
tionally, etc.), where the patient had the options of describing himself (herself) as
doing 'very well', 'fairly well', 'fairly poorly', and 'poorly'. 'Qualified success' was
 679

defined as the patient being at home on CAPD for the time period mentioned but
not describing himself as doing 'very well'. The second end-point was 'failure',
which meant either death or return to the in-hospital hemodialysis center, at any
time beyond the 3-month examination period and up to December 31,1983.
On this basis 40 patients were identified as being 'successfully adapted' to their
CAPD program; similarly there were 79 'failures'. A further 31 patients were
adapted to CAPD with 'qualified success' but for this analysis were not con-
sidered further.
The sex of the patient appeared to influence 'successful adaptation' in that
38.3% of the total female population were successful as opposed to 26.5% of the
males. The age of the patient, however, was of no influence. The patients who
were successful were significantly less depressed than those who returned to
in-center hemodialysis or who died (p<0.035) and they were also less anxious.
Interestingly enough, the patients who were successful showed more social
introversion than those who failed and they had better marital cohesion and
showed better interpersonal relationships with other members of the household
family (usually children) (p<0.02). However, success was not correlated with
relationships involving extended family or friends and the successful patients
were less involved than the failures in community activities. Using the Rotter
Internality/Externality Scale [89] it was found that the successful patient scored
more internally than those who returned to in-center hemodialysis or who sub-
sequently died. The successful patients were certainly physiologically better as
judged by routine clinical hematological and biochemical examinations and had
less morbidity as defined by hospital admissions. These successfully adapted
patients only had 0.09 hospital admissions/patient/month whereas those who
returned to in-center hemodialysis or died showed, respectively, 0.29 and 0.25
admissions per patient month (p<O.0085).
This study, once again, indicates the importance of psycho-social factors as well
as physiological parameters in the outcome of a CAPD program and, further-
more, suggests that psychometric testing of patients early in their treatment
regime may be of prognostic value and, also, may indicate where intervention by
trained personnel is necessary.

4. The choice of a treatment modality for home dialysis

Review of the presented information indicates that the frequency and degree of
stresses and concerns are much more relevant to those patients treated by
hemodialysis especially in the younger age groups. Patients on either peritoneal
or hemodialysis procedures both experience stress and concern which is directly
related to the degree of depression and anxiety that the individual has as part of
his/her basic personality make up. Thus, it can be assumed that patients who
demonstrate high levels of depression or anxiety will find home hemodialysis a
680

very stressful form of therapy; a patient who is not unduly (for a dialysis patient)
depressed and who has a 'healthy' level of denial on the other hand may easily
cope with that treatment mode. CAPD on the other hand appears to interfere
with jobs and household work and is likely to hamper an individual's return to
employment especially if the patient is male and of the older age group. Thus,
males who are strongly motivated to continue working, particularly if over 45
years of age, might best be treated by hemodialysis. However, the young female
patient who is not employed will likely prefer CAPD; a treatment modality
associated with less stress and concern. This preference would be enhanced if her
personality profile showed elevated levels of depression, anxiety, self-deprecia-
tion, or social introversion.
Obviously, the management of patients with ESRD is not the sole domain of
any particular treatment modality; rather a very flexible attitude should be
adopted by renal units and patients should be guided towards the type of dialysis
therapy that best suits them. It does appear that the stresses of home dialysis are
not a uniform problem and they may be influenced by the type of dialysis therapy
in selected groups of patients. Thus, information obtained from psycho-social
profiling may lead to the appropriate selection of treatment modality for any
given patient.
What is becoming clear to nephrologists and their health-care teams is that
every geographic region should have all forms of peritoneal and hemodialysis
modalities at its disposal. This is becoming more important as financial pressures
are forcing those teams to maintain as large a percentage of patients outside the
hospital setting as possible. Thus, it is obviously imperative to be cognizant of
non-physiological parameters which may influence the survival and probabilities
of success for dialysis patients. More importantly, identification of these key
determinants may help to minimize unnecessary mortality and morbidity by early
appropriate intervention and will likely improve patients' adaptation to their
illness and influence their quality of life.

Acknowledgements

We would like to thank Miss E. Lyons for the preparation of this manuscript.
Robert Lindsay and Howard Burton also acknowledge grant support by the
Ministry of Health of Ontario and by Physicians Services Inc. for the 'Ontario
Adaptation to Home Dialysis Study' which provides much of the current knowl-
edge of psycho-social problems in the dialysis field.
 681

Notes

1. The estimated average expected survival for all patients starting treatment between 1970 and 1980
was just under seven years [13].

2. As a concept, the sick role offers an interesting explanation of the formation of dependency needs.
The theory underlying the concept 'sick role' is that others' expectations of an ill person, regardless of
socio-economic status, are determined most heavily by an uncertain or worsening prognosis. That is,
with a serious prognosis, there is increasing social pressure which discourages normal behaviour and
treats the individual as dependent, exempting him from social responsibilities insulating him from
daily activities and emphasizing his dependence on others for his continued well-being. The individual
according to this theory may accept this role, even to the point of exaggerated demands for more help
or personal attention.

References

1. Richmond JM, Lindsay RM, Burton HJ et aZ. : Psychological and physiological factors predicting
the outcome on home hemodialysis. Clin Nephrol17: 109, 1982.
2. Wai L, Richmond J, Burton H et al.: Influence of psychosocial factors on survival of home dialysis
patients. Lancet 2: 1155, 1981.
3. Posen GA, Jeffery JR, Arbus GS et aZ.: Canadian Renal Failure Register: 1982 Report. Published
by the Kidney Foundation of Canada, 1983.
4. Robson MD, Oreopoulos DG: A revolution in the treatment of chronic renal failure. Dialysis and
Transplant 7: 999, 1978.
5. Feton SSA, Cattran DC, Allen AF et af.: Initial experience with continuous ambulatory per-
itoneal dialysis. Artif Organs 3: 206, 1979.
6. Steinberg SM, Cuter SJ, Novak JW: The Report of the National CAPD Registry of the National
Institutes of Health. University of Missouri-Columbia, 1984
7. Wing AJ, Broyer M, Brunner FP et aZ. : Combined report on regular dialysis and transplantation
in Europe, XIII, 1982. In: AM Davison, PJ Guillou (eds), Proceedings of the European Dialysis
and Transplant Association - European Renal Association, Vol 20, Pitman Publ, London,
England, 1983, pp 2-75.
8. Burton HJ, Canzona L, Lindsay RM et aZ.: Adaptation to home dialysis: The Health Care
Research Unit Report. University of Western Ontario Press, 1980.
9. Lindsay RM, Oreopoulos DG, Burton HJ et af.: The effect of treatment modality (CAPD vs
hemodialysis) in adaptation to home dialysis in Canada. In: RC Atkins, NM Thomson, PC Farrell
(eds), Peritoneal Dialysis. Churchill, Livingstone, Edinburgh, 1981, pp 385-394.
10. Millon T, Green C, Meagher R: Handbook of Clinical Health Psychology. Plenum Press, New
York, 1982.
11. Stone GC, Cohen F, Adler NE: Health Psychology - A Handbook. Jossey-Bass, Washington,
1980.
12. Levy NB: Psychonephrology 11. Plenum Publ, New York, 1981.
13. Hutchinson TA, Thomas DC, MacGibbon B: Predicting survival in adults with end-stage renal
disease: an age equivalence index. Ann Intern Med 96: 417,1982.
14. Jourard SM: The invitation to die. In: ES Scheidam (ed), On the Nature of Suicide. Jossey-Bass,
New York, 1969.
15. Gelfman M, Wilson EJ: Emotional reactions in a renal unit. Comp Psychiatry 13: 283, 1972.
16. Abram HS, Moore GL, Westervelt FB Jr: Suicidal behaviour in chronic dialysis patients. Am J
Psychiatry 127: 1199, 1971.
682

17. Haenel T, Brunner F, Battegay R: Renal dialysis and suicide: occurrence in Switzerland and in
Europe. Comp Psychiatry 21: 140, 1980.
18. Norton CE: Attitudes toward living and dying in patients on chronic hemodialysis. Ann NY Acad
Sci 164: 720, 1969.
19. Haynes RB, Sackett DL: A Workshop Symposium: Compliance with Therapeutic Regimens,
Annotated Bibliography. McMaster University Medical Centre, Hamilton, Ontario, 1974.
20. Lindsay RM, Burton HJ, Heidenheim P: Compliance, denial and dialysis success. In: Levy N
(ed), Psychonephrology IV, Plenum Publ, New York, 1983.
21. De-Nour K, Czaczkes JW: Personality factors in chronic hemodialysis patients causing non-
compliance with the medical regime. Psychosomatic Medicine 34: 333, 1972.
22. Gonsalves-Ebrahim L, Gulledge AD, Miga D: Continuous ambulatory peritoneal dialysis:
psychological factors. Psychosomatics 23: 944, 1982.
23. Blagg C: Continuous ambulatory peritoneal dialysis. Am J Kidney Dis 2: 378, 1982.
24. Kutner NG, Cardenas DD: Rehabilitation status of chronic renal disease patients undergoing
dialysis: variations by age category. Arch Phys Med Rehabil62: 626, 1981.
25. Oreopoulos DG, Khanna R: The present and future role of continuous ambulatory peritoneal
dialysys (CAPD). Am J Kidney Dis 11: 381,1982.
26. Calsyn DA, Sherrard DJ, Hyerstay BJ et al.: Vocation adjustment and survival on chronic
hemodialysis. Arch Phys Med Rehabil 62: 483, 1981.
27. Gutman RA, Stead UW, Robinson RR: Physical activity and employment status of patients on
maintenance dialysis. N Engl J Med 304: 309, 1981.
28. Rosenbaum I, Atcherson E, Corry RJ: Rehabilitation and the transplant patient. Dialysis and
Transplant 10: 136, 1981.
29. Tews HP, Schreiber WR, Huber W, Zeit J, Ritz E: Vocational rehabilitation in dialyzed patients:
a cross sectional study. Nephron 26: 130, 1980.
30. Palmer S, Canzoma L, Conley J, Wells G: Vocational adaptation of patients on home dialysis- its
relationship to personality, activities and support received. J Psychosomatic Res 27: 201, 1983.
31. Hagarty SF, Hagarty LM: Beyond Survival: A Study of the Rehabilation of Kidney Dialysis and
Transplant Patients. Laidlaw Foundation, Toronto, Canada, 1977.
32. Lindsay RM, Oreopoulos D, Burton HJ et al.: Adaptation to home dialysis: a comparison of
continuous ambulatory peritoneal dialysis and hemodialysis. In: M Legrain (ed), Continuous
Ambulatory Peritoneal Dialysis. Excerpta Med, Amsterdam, 1980, pp 120-130.
33. Lindsay RM, Burton HJ, Heidenheim P: Is CAPD for everyone? In: N Levy (ed), Psycho-
nephrology IV. Plenum Publ, New York, 1983.
34. Wiltshire S: The Quality of Life of Patients on Self-Care Hemodialysis and Continuous Ambula-
tory Peritoneal dialysis. Ph.D. Dissertation, School of Profession Psychology, UCLA, Berkeley,
1982.
35. Fisher S, Cleveland SE: Body Image and Personality, Van Nostrund, New York, 1958.
36. Cameron N: Personality Development and Psychopathology. Houghton Mufflin Company,
Boston, 1963.
37. Deutsche H: Some psychoanalytic observations in surgery. Psychosomatic Med 4: 105,1942.
38. Cummings J: Hemodialysis; the pressures and how patients respond. Am J Nursing 70,1970.
39. De-Nour KA, Czaczkes SW: Psychological and psychiatric observations on patients in chronic
haemodialysis. Proc Eur Dialysis Transplant Assoc 5: 67, 1969.
40. Wright RG, Sand P, Livingston G: Psychological stress during hemodialysis for chronic renal
failure. Ann Intern Med 64: 611, 1966.
41. Abram HS: The psychiatrist, the treatment of chronic renal failure and the prolongation oflife: II.
Am J Psychiatry 126: 157,1969.
42. Kestenbaum JM: Psychological adjustment to hemodialysis. Paper presented at the American
Psychological Association, Chicago, 1975.
43. Shea EJ, Bogdan A, Freeman RB etal.: Hemodialysis for chronic renal failure IV: Psychological
considerations. Ann Intern Med 62: 558, 1965.
 683

44. Kemph JP: Renal failure, artificial kidney and kidney transplants. Am J Psychiatry 122: 1270,
1966.
45. Czaczkes JW, De-Nour KA: Chronic Hemodialysis As A Way of Life, Brunner/Mazel Publ, New
York, 1978.
46. Schreiner CE: The Artificial Kidney. 8: 42-45, 1959.
47. Menzies IG, Stewart WK: Psychiatric observations on patients receiving regular dialysis treat-
ment. Br Med J 1: 544, 1968.
48. McKegney FP, Lange D: The decision to no longer live on chronic hemodialysis. Am J Psychiatry
128: 267, 1971.
49. Short MJ, Wilson WP: Roles of denial in chronic hemodialysis. Arch General Psychiatry 20: 433,
1969.
50. Wittkower E: Psychological aspects of physical illness. Can Med Assoc J 66: 220, 1952.
51. Ballak L: Psychology of Physical Illness. Grune and Stratton, New York, 1952.
52. Levy NB: Psychological reactions to machine dependency. Psychiat Clin North Am 4: 351,1981.
53. Larsen NA: Sexual problem of patients on RDT and after transplantation. Proc Eur Dialysis and
Transplant Assoc. The Pitman Press, Bath, 9: 271, 1972.
54. Abram HS, Hester LR, Sheridan WF et al.: Sexual functioning in patients with chronic renal
failure. J Nervous Mental Dis 160: 220, 1975.
55. De-Nour KA, Czaczkes JW: Personality factors influencing vocational rehabilitation. Arch
General Psychiatry 32: 573, 1975.
56. Holcomb JL, MacDonald RW: Social functioning of artificial kidney patients. Social Sci Med 7:
109, 1973.
57. Burton HJ, Klatt H, Conley JA et al.: Stresses associated with sexual adjustment of end-stage
renal disease commencing home dialysis: A comparison between hemodialysis and CAPD.
Contemporary Dialysis, 25, 1981.
58. Kline SA, Burton HJ, Lindsay RM: Sexual problems of patients on Continuous Ambulatory
Peritoneal Dialysis. P.D. Bulletin (submitted), 1984.
59. Campbell DT, Converse P, Rogers W: The Quality of American Life. Russell Sage Foundation,
New York, 1976.
60. Short MJ, Alexander RJ: Roles of denial in chronic hemodialysis. Arch General Psychiatry 20:
433, 1969.
61. Friedman EA, Goodwin NJ, Chaudhry L: Psychosocial adjustment to maintenance hemodialysis:
II NY State J Med 70: 767, 1970.
62. Glassman RM, Siegel A: Personality correlates of survival in a long-term hemodialysis program.
Arch General Psychiatry 22: 566, 1970.
63. Goldberg RT: Rehabilitation of patients on chronic hemodialysis and after renal transplantation:
a comparative study. Scand J Rehabil Med 6: 65, 1974.
64. Mock LA: Psychosocial aspects of home and in centre dialysis. P.D. Dissertation, University of
Houston, 1975.
65. Burton HJ: Survival without the machine: a comparison of the psycho-social effectiveness of
continuous ambulatory peritoneal dialysis (CAPD) and home hemodialysis. Dr P.H. Disserta-
tion, University of Pittsburgh, 1981.
66. Shulman R, Pacey I, Diewold P: The quality of life on home hemodialysis. Presented at the
European Conference on Psychosomatic Research, Edinburgh, 1974.
67. Brand L, Komorita NJ: Adapting to long term hemodialysis. Am J Nursing 66: 1178, 1966.
68. De-Nour KA, Shaltield J et al.: Emotional reactions of patients on chronic hemodialysis. Psycho-
somatic Med 30: 521, 1968.
69. Blatt B, Tsushima WT: A psychological survey of uremic patients being considered for the chronic
hemodialysis program: Intellectual and emotional patterns in uremic patients. Nephron 3: 206,
1966.
70. Lefebvre P, Nobert A, Crombez JC: Psychological and psychopathological reactions in relation to
chronic hemodialysis. Can Psychiatric Assoc J 17: 55,1972.
684

71. Binik Y: Coping with chronic life threatening illness: Psychosocial perspectives on end-stage renal
disease. Can J Behav Sci Rev Can Sci Comp 15: 373, 1983.
72. Burke HR: Renal patients and their MMPI profiles. J PsychollOl: 229,1979.
73. Trieschmann RB, Sand PL: WAIS and MMPI correlates of increasing renal failure in adult
medical patients. Psychol Rep 29: 1251, 1971.
74. Devins GM, Binik YM, Gorman P et al.: Perceived self-efficacy outcome, expectancies, and
negative mood states in end-stage renal disease. J Abnormal Psycho I 91: 241, 1982.
75. Devins GM, Binik YM, Hollomby B et al.: Helplessness and depression in end-stage renal
disease. J Abnormal Psychol 90: 531, 1981.
76. Livesley WJ: Psychiatric disturbances and chronic hemodialysis. Br Med J 2: 306, 1979.
77. Livesley WJ: Factors associated with psychiatric symptoms in patients undergoing chronic hemo-
dialysis. Can a J Psychiatry 26: 562, 1981.
78. Lowry MR: Frequency depressive disorder in patients entering home hemodialysis. J Nervous
Mental Dis 167: 199, 1979.
79. Abram HS, Wadlington W: Selection of patients for artificial and transplanted organs. Ann
Intern Med 69: 615, 1968.
80. Cramond WA, Knight PR, Lawrence JR: The psychiatric contribution to a renal unit undertaking
chronic hemodialysis and renal homotransplantation. Br J Psychiatry 113: 1201, 1967.
81. Reichsman F, Levy NB: Problems in adaptation to maintenance hemodialysis. Arch Intern Med
130:859, 1972.
82. Beard BH: Fear of death and fear of life. Arch General Psychiatry 21: 373, 1969.
83. Freyberger H: Six years' experience as a psychosomaticist in a hemodialysis unit. Psychotherapy
and Psychosomatic 22: 226, 1973.
84. MacNamara M: Psychosocial problems in a renal unit. Br J Psychiatry 113: 1231, 1967.
85. Coogan JP: Psychiatry for the slowly dying. Report on a program at Los Angeles Country
Hospital. SK and F Psychiatric Reporter. 45: 4, 1969.
86. Lindsay RM, Burton HJ: Psychosocial factors. In: S Fenton, M Kaye, J Price (eds), The
Treatment of End-Stage Renal Disease by Peritoneal Dialysis. Princeton Junction, New Jersey,
Communications Media for Education, 1982.
87. Shambaugh PW, Kanter SS: Spouses under stress. Group meetings with spouses of patients on
hemodialysis. Am J Psychiatry 126: 928, 1969.
88. Shambaugh PW, Hampers CL, Bailey G et al. : Hemodialysis in the home - emotional impact on
the spouse. Trans Am Soc Artif Intern Organs 13: 41, 1967.
89. Rotter JB: Some problems and misconceptions related to the construct of internal vs external
control of reinforcement. J Consult Clin Psychol 43: 56, 1975.
 INDEX OF SUBJECTS

A Cardiovascular complications 449-450

Cgmplications other than peritonitis 441-
Abdominal complications, diabetic patient on 454,460-465
CAPD 565 Convulsions 451-452
Abdominal cramps 234 Dialysate leak 444
Abdominal discomfort 235, 350 Disequilibrium syndrome 451
Abdominal distension 350, 354, 442 Fluid overload 449
Abdominal fullness 349 Heart failure 449
Abdominal hernia 233, 260, 487-488 Hyperglycemia 452
Abdominal inflammation 221 Hyperkalemia 454
Abdominal pain Hypernatremia 452
Acute peritoneal dialysis 442-443 Hypernatremic dialysate 448
Catheter-related 467, 468 Hypoglycemia 452
Hypovolemia 449
Children 549
Intraperitoneal loss of peritoneal catheter
Continuous ambulatory peritoneal dialysis
446-447
228-229, 235, 468
Ischemic heart disease 450
Diabetics 565
Laceration of internal organs 445-446
Peritonitis 414
Mechanical complications 442-447
Abdominal punctures 347
Medical complications 448-454
Abdominal surgical procedures 221, 444-
Mental confusion 452
445, 544
Metabolic complications 452
Complications of acute peritoneal dialysis Mortality rates due to complications 454
after 447 Myocardial infarction 450
Abdominal tenderness 414 Neurological complications 451-452
Abdominal wall infection 186, 222 Overheated dialysate 447
Abnormal lipid metabolism, 198-199 Perforation of internal organs 445-446
Abrasions 409 Persistent metabolic acidosis 454
Abscess formation 410 Polycystic kidney disease 447
Accelerated catabolism 347 Poor drainage 444-445
Access site infection 347 Pulmonary complications 450-451
Acetaminophen 552 Pulmonary edema 449
Acetate 109, 287, 302-303, 482 Acute peritonitis 411, 464-465
Dialysis solutions 4, 81-85 Acute purulent bronchitis 450
Acetycholine 69 Acute renal failure 2, 115, 316
Acetyl CoA 84 Children 525, 527, 532-542
Acidosis 3, 302, 386, 450, 452 Bicarbonate in dialysis solution 541
Acinetobacter 409, 420 Catheters, considerations relating to
Acinetobacter peritonitis 306 537-540
Acute bacterial peritonitis 355 Complications of dialysis 542
Acute hemolysis 319 Continuous ambulatory peritoneal di-
Acute hydrothorax 444 alysis 542
Acute malnutrition 346 Defined 532
Acute peritoneal dialysis Diagnosis 534
Abdominal pain 442-443 Dialysate 541
Abdominal surgery 447 Etiology 533
Alkalosis 453-454 Incidence 533
Arrhythmia 450 Indications for dialysis 534-536
Bleeding 443-444 Influence of early dialysis 536-537
686

Intennittent peritoneal dialysis 186 Diabetics 566

Intraperitoneal hemorrhage 542 Amino Dianeal 492-493
Mechanics of dialysis 541-542 Aminoglycoside antibiotics 236, 238
Metabolic complications 542 Aminoglycosides 308, 311, 424-425
Mortality rates 536-537 Aminophylline 85, 279
Non-oliguric 532-533 Amiodarone 317
OJiguric 532-533 AMP 84
Percutaneous catheter placement tech- Amphetamine poisoning 316
nique 538-539 Amphotericin 313
Peritonitis 542 Amphotericin B 215, 274, 288-289, 425
Permanent catheters 537 Ampicillin 310, 425
Surgical catheter placement technique Amputations 568
539-540 Anabolism 365
Technical considerations 537-542 Anaerobes 410
Temporary catheters 537-538 Anaerobic infections 312
Tenckhoff catheters 538-540 Anaerobic organisms causing peritonitis 421
Trocath catheter 537-539 Anaerobic peritonitis 236
Intennittent peritoneal dialysis 179-180, Analgesics 315
185-187 Anatomical resistance barriers, see Barriers
Mortality rate 185 of solute exchange between vascula-
Poisoning 319 ture and peritoneal cavity
Acute subclavian vascular access 444 Andenosine 65
Acute uremia 2 Anemia 348-349, 391,499-500
Acyclovir 313 Continuous ambulatory peritoneal dialysis
Adaptive versus maladaptive denial 676 233
Adenine nucleotides 84-85 End-stage renal disease in children 549
Adenosine 84-85 Angiotensin 65, 216, 284
Adequacy of dialysis 193-194, 367-369 Angiotensin II 513
Adhesions 7, 109 Anion gap 360, 541
Adrenergic nerve terminal 65 Anorexia 346, 382
Adrenergic receptor stimulation 69 Alterations in metabolism of nutrients 377
Adriamycin 318 Assessment of nutritional status 349-350
Adsorbents 276 Children 390
Alanine 345 Continuous ambulatory peritoneal dialysis
Albumin 33, 161, 164, 169-170,315,317, 211, 232
321 Diabetics 385
Leakage 71 Intennittent peritoneal dialysis 193
Albumisol 316 Transport kinetics and 139
Alcoholism 349 Anovulatory cycles 510
Alcohol restriction 498 Anthranilic acid 298
Alcohols 316 Anthropometric measurements 349, 352,
Aldomet 317 354, 361, 363
Alkaline dialysis solution 303 Children 386-387
Alkalosis 383, 449, 453-454 Anti-arrhythmic agent 317
Alpha-adrenergic receptors 272, 281 Antibiotics, see also specific types 306-319,
Alpha-adrenergic receptor stimulation 78 422
Aluminum 377-380, 391 Aminoglycosides 311
Aluminum staining 256 Antimicrobial activity 307
Amanitine 319 Cephalosporins 308-309
Amantadine 313 Chloramphenicol 312
Amikacin 311, 425 Clindamycin 310
Amino acid absorption via peritoneal route Compatibility with dialysate solutions 307-
492-493 308
Amino acid pools 345 Continuous ambulatory peritoneal dialysis
Amino acids 174, 305, 358-360, 490-493 228
 687

Erythromycin 312 Azotemia 385

Half-life 307 Azotemic symptoms 346
Heparin 308
Intermittent peritoneal dialysis 185
Lincomycin 310 8
Metronidazole 312
Penicillins 310 Back pain
Peritonitis 238, 423-424, 431 Continuous ambulatory peritoneal dialysis
Polymyxins 312 222,233
Sensitivity to 414, 424-425 Continuous cyclic peritoneal dialysis 264
Sulfas 311 Renal failure 505-508
Tetracycline 312 Back-up dialysis 229
Therapeutic recommendations 308-310 Bacteremias 405
Thiamphenicol312 Bacterial infections 355
Vancomycin 311 Bacterial peritonitis, see also specific types
Antidepressants 316 of bacteria 185,409,418-421
Anti-epileptics 317 Bacteroides 421
Antifungal agents 312-313 Barbiturates 289-290, 315
Antihypertensive medication 216, 228, 567, Barium enema 229
571 Barriers of solute exchange between vascula-
Anti-malarials 313 ture and peritoneal cavity 59-65
Antiseptics 3 13-314 Arterioles 59-60
Antituberculos agents 313 Basement membrane 62-63
Anti-viral agents 313 Capillaries 60-62
Anxiety 350, 675-676 Endothelial lining 64-65
Aortic aneurysm repair 221, 230 Endothelium 60-62
Apathy 349, 676 Interstitium 63-64
Apnea 312 Mesothelium 63-64
Appendicitis 429 Post capillary and collecting venules 62
Appetite suppression 349-350 Basal atelectasis 450
Arachidonic acid 283-285, 298 Basal lamina 60, 62, 103
Arcades 55-56 Basement membrane 25, 29, 62-63, 103
Arrhythmias 194, 450 Bathing, peritoneal catheters, 225
Arsine exposure 319 "Bath to blood" 175
Arterial blood supply 53 Belligerent patients 222
Arterial capillaries 104 Beta-adrenergic blocking agents 72
Arteriolar dilators 276 Beta-adrenergic receptors 272
Arterioles 59-60, 66, 104 Beta-adrenergic receptor stimulation 78
Artery to artery anastomoses 55 Beta-blockers 317
Arthritis 222 Beta-cap system 466
Artificial kidneys 598 Bicarbonate 3-4, 79, 302-303, 541
Ascites 53, 58, 273, 297, 459 Bicarbonate dialysis 195
Aseptic peritonitis 306, 422 Bilateral Wilms' tumor 544
Aspirin 234, 284, 552 Bilirubin 321
Asthma 229 Biochemical parameters, continuous cyclic
Atherosclerosis 494 peritoneal dialysis 253-255
ATP 84 Biological status, diabetic patients on CAPD
Atropine 53 563-564
Autacoid effects in peritoneal vasculature 65- Biotransformation 317
66,72 Bisulfite 306
Automated peritoneal dialysis machines 8- Bladder perforation 446
10,347 Bleeding complications 185-186, 196,468,
Autonomic nervous system dysfunction 568 500
Auxiliary devices for handicapped 431 Acute peritoneal dialysis 443-444
Az10cillin 310 Blindness 221, 563
688

Blood administration 321 Balance, continuous ambulatory peritoneal

Blood-brain barrier 135 dialysis 218
Blood circulation in peritoneal cavity 52-54 Children 389-390
Blood flow patterns, factors affecting 51 Concentration 30 I
Blood flow regulatory mechanisms 51 Continuous cyclic peritoneal dialysis
Blood glucose control 562-563, 565, 570- 256-257
571 Influx 73
Blood pressure 8, 13, 68, 135 Mobilization 72
Control 216 Nutritional concerns 362, 364-365
Continuous ambulatory peritoneal di- Calcium carbonate 217-218
alysis 230 Calcium oxalate stones 460
Continuous cyclic peritoneal dialysis Caloric malnutrition 345, 353, 386
254, 257-258 Cancer 51,318,544
Diabetic patients on CAPD 566-567 Candida cystitis 422
Renal failure 511-513 Candida peritonitis 312-313, 408-409
Blood solute levels, factors affecting 581 Candida species 312
Bloodstream 116 Capillaries 55, 60-62, 96, \03
Blood sugar control 200-201 Capillary dimensions, peritoneal dialysis sys-
Blood supply 54 tem and hollow fiber dialyzer com-
"Blood to bath" 160 pared 33-35
Blood urea nitrogen (BUN) 8, 131-132 Capillary endothelium 29, 116
Dialysate/plasma ratio in children on Capillary filtration coefficient 276
CAPD 142 Carbenicillin 310
Generation rate 136 Carbohydrates 231, 370-371
Levels 137-139, 141 Carcinoma 53
Blood vessels \03-104 Cardiac surgery 537
Bloody dialysate, continuous ambulatory Cardioactive agents, see also specific types
peritoneal dialysis 234 317
Bloody fluid 423 Cardiomyopathy 194
Body composition 350-352 Cardiopulmonary compromise 488-490
Body fat 353 Cardiovascular complications 185-186, 200,
Body image 672-673 449-450
Boen's closed-system peritoneal dialysis Cardiovascular death 371
equipment 180-181 Cardiovascular disease 493
Bone disease 389, 458 Cardiovascular dysfunction 348
Bone pain 505 Cardiovascular factors 65
Bone problems 199 Carnitine 373-374, 379
Boric acid intoxication 319 Catabolic stress 384
Bovine grafts 200 Catabolism 356-357
Bowel obstruction 446 Catecholamines 68-69, 78, 281-282
Bowel perforation 446, 459-460 Catheters, see also specific types
Bradykinin 65 Break-in technique 473-476
Branched-chain amino acid 322 Broken 446
Bromide 318 Children 537-540
Bronchopneumonia 533 Chronic management 219
Buffer anions 73-74, 81-85 Column-disc 478-479
Complications, see also Mechanical com-
plications 455-457
c Continuous ambulatory peritoneal di-
alysis 466-480
Cachexia 194 First period (1923-1962) 2-3
Caffeine 53 Intraperitoneal loss 446-447
Calcifications 458 Malfunction 350, 455
Calcium 380 Multiple insertions 444
Absorption, factors affecting 30\ New 432
 689

Nylon, 3, 8, 10, 403 Developmental aspects of peritoneal di-

Obstruction 467 alysis kinetics 529-532
One-way obstruction 444-445, 455-456, Dietary recommendations 386-389
467 Efficiency of peritoneal dialysis 529-530
Outflow obstruction 261 Electrolytes 389
Paramedian approach for insertion 478 Emergency hemodialysis 553
Peritoneal, see Peritoneal catheters End-stage renal disease, see also End-
Permanent 537 stage renal disease, children 543-551
Plugging, prevention of 308 Energy requirements 387-388
Polyethylene plastic tubes 3 Exchange volumes 530
Polyvinyl chloride 3 Historical notes on use of peritoneal di-
Post-operative leak 476 alysis in 526-528
Removal 422,427-428 Inborn errors of metabolism 553-554
Rubber 2 Intermittent peritoneal dialysis 197-198,
Silastic II, 408 527-528
Site drainage 224-225, 229 Intoxications 552-553
Temporary 537-538 Mass transfer 530
Tip displacement 456, 468 Mass transfer-area coefficients 531-532
Catheter site hernia 224 Membrane area-permeability products 530-
Cefazolin 309 531
Cefotaxime 307 Nutritional assessment 385
Cefoxtin 309 Nutritional considerations 386-391
Ceftazidime 309 Peritoneal cavity 529
Cefuroxime 309 Peritoneal dialysis in 525-560
Celiac arteries 53 Acute renal failure 532-542
Cell counts 414 Developmental aspects of kinetics 529-
Cellular factors in peritonitis 411 532
Cellular response 410 End-stage renal disease, 543-551
Cephalexin 309 Historical notes 526-528
Cephaloridine 309 Inborn errors of metabolism 553-554
Cephalosporin 228, 236, 238, 308-309 Intoxications 552-553
Cephalothin 307, 425 Miscellaneous disorders 554
Cephamandole 309 Peritoneal surface area 529
Cerebral edema 451 Phosphorus 389-390
Cerebrovascular accidents 222 Protein intake 388-389
Cesium 317 Salicylate intoxication 315
Cetyl trimethyl ammonium chloride 287 Single-cuffed Tenckhoff catheter 224
Chelates 276 Sodium chloride poisoning 554
Chelation 84 Trace elements 390-391
Chemical peritonitis 422 Urea clearances 529
Chemotactic factors 409 Vitamin D 389-390
Chemotherapeutic agents 318 Vitamin requirements 390
Chest pain 514 Water intake 389
Children 13 Chloramphenicol 312
Acute renal failure, see also Acute renal Chlorate 319
failure, children 525, 527, 532-542 Chlorhexidine 314
Calcium 389-390 Chloride 164, 166, 172
Comparison to adult 529-532 Chloroquine 313
Congenital hyperammonemia 553-554 Choice of treatment modality for home di-
Congestive heart failure 554 alysis 679-680
Continuous ambulatory peritoneal dialysis Cholecystitis 429-430
in 141-145,221,525-526,528 Cholecystokinin 286
Continuous cyclic peritoneal dialysis 265, Cholelithiasis 498
628 Cho1estyramine 425
Continuous peritoneal lavage 527 Chromium 316
690

Chronic abdominal wall infections 222 Clostridium difficile 425

Chronic anemia 345 CloxaciIlin 310, 425
Chronic end stage renal disease 371 Coeliac arteries 95
Chronic hemodialysis 357 Cold peritoneal lavage 179
Chronic hypocalcemia 345 Colistin 312
Chronic intermittent peritoneal dialysis Collecting lymphatics 56-57
Ascites 459 Colloid osmotic pressure 297
Automatic machines 8-10 Colostomies 222, 447
Bowel perforation 459-460 Coma 451-452,553
Catheter-related complications 455-457 Combined diffusive and convective mass
Clinical results II, 13 transfer models 122-123
Complications other than peritonitis, see Comparison of peritonal dialysis and hemo-
also other SUbtopics hereunder 455- dialysis 6
465 Compliance versus non-compliance 670-671
Constipation 459-460 Complications, see also specific types
Depletion syndromes 457 Acute peritoneal dialysis, see also Acute
Devices for access into peritoneal cavity peritoneal dialysis 441-454,460-465
7-8 Acute renal failure in children 542
Dialysis disequilibrium 458 Chronic intermittent peritoneal dialysis
EEG abnormalities 458 455-465
History, see also History of peritoneal di- Continuous ambulatory peritoneal dialysis,
alysis 7-13 see also Continuous ambulatory peri-
Indwelling catheters 10-12 toneal dialysis 228-239, 466-524
Kidney stones 460 Continuous cyclic peritoneal dialysis 258-
Neurological complications 457-458 261
Neuropsychiatric problems 458 End-stage renal disease in children 548-
Osteodystrophy 458-459 549
Peripheral neuropathy 457-458 Intermittent peritoneal dialysis 185-186
Chronic malnutrition, see also Malnutrition Other than peritonitis, see also specific
346 types 441-524
Chronic obstructive pUlmonary disease 251 Peritoneal dialysis 441
Chronic peritoneal dialysis 116, 135 Peritonitis, see Peritonitis
Chronic renal failure, see also Renal failure Renal failure 499-514
180,345 Composition of dialysate in first period
Children 197-198 (1923-1962) 3-5
Intermittent peritoneal dialysis, see also Concentration polarization 168
Intermittent peritoneal dialysis 187- Concentration profiles 148-149, 166-169
201 Confusion 312
Congenital hyperammonemia 553-554
Chronic renal insufficiency 386
Congenital malformations 533
Chronic uremia 58
Congenital organic acidemias 554
Cimetadine 232, 318
Congenital urea cycle enzymopathies 553
Circulatory pathways in peritoneal cavity 52-
Congestive heart failure 277, 321, 346, 349
54 Children 554
Circumflex arteries 96 Constipation 229, 350, 429, 459-460, 505
Cirrhosis 53 Contaminants of dialysis solutions 306
Cisplatin 318 Contamination 347, 467
Clearance measurements 120 Continuous ambulatory peritoneal dialysis
Clearances 42, 45-46, 581 (CAPO) 118-119, 130-131, 135, 184-
Clindamycin 310, 425 185, 209-246
Clinical experience, continuous cyclic perito- Abandonment, reasons for 657-658
neal dialysis 253-258 Abdominal hernias 487-488
Clinical status, diabetic patients on CAPD Abdominal pain 228-229, 467
563-564 Acute renal failure in children 542
Clofibrate 498 Adaptation to, factors determining success
Clostridium 421 in 678-679
 691

Advantages over hemodialysis 141 Glucose absorption 216, 493-498

Amino acid absorption via peritoneal route Glucose as osmotic agent 214-215
492-493 Heart failure patients 321
Amino acids 490-493 Hemorrhoids 233
Anemia 233 Hernia 222, 233
Anorexia 232 High-risk patient population 220
Back pain 222, 233 History 13-15, 209-211
Blindness 221 Hospital back-up 220
Blood pressure control 230 Hyperglycemia 232-233
Bloody dialysate 234 Hypertension 230
Break-in technique for catheter insertion Hypotension 230
473-476 Immunosuppression 222
Calcium balance 218 Intra-abdominal pressure, complications
Calcium concentration 30 I due to 483-490
Cardiopulmonary compromise 488-490 Intravascular volume depletion 215
Catheter leaks 229 Laboratory tests 227-228
Catheter obstruction 467 Life style changes 498
Catheter-related complications 466-480 Lipid abnormalities 493-498
Children 141-145,221,525-526,528
Management of patients on 652-659
Clinical aspects, see also other SUbtopics
Massive hydrothorax 483, 485-487
hereunder 219-228
Materials used 211
Complications, see also other subtopics
Medications 228
hereunder 228-239
Metabolic complications 490-499
Complications during training 228-229
Complications other than peritonitis, see Middle molecule hypothesis 215
also other subtopics hereunder 466- Middle molecule removal 214
524 Molecular size 212-214
Concepts 211-219 Motivation 220-221, 231
Containers for delivery of dialysis solution Neurological defects 222
210-211 Nitrogen metabolism 365-366
Containers for dialysis solution 234 Non-compliance with protocols 231-232
Contra-indications 221-222 Number of exchanges 215
Creatinine removal 213 Nutritional status of patients undergoing
Daily dietary allowances 375 353-354
Depression 231 Orthostatic hypotension 215-216
Diabetic patients, see also Diabetic pa- Ostomies 222
tients 233, 348, 561-570 Overview of 598
Dialysate flow 34-36 Patient acceptance 348
Dialysate flow rate 211-212 Patient selection 220-221, 645-652
Dialysate leaks 476-477, 483-484 Peritoneal catheters, see also Peritoneal
Dialysis solution tonicity 215 catheters 222-225
Dietary management 218-219 Peritoneal fibrosis 480-482
Direct catheter problems 467-471 Peritoneum during 105-11 0
Distension 228-229 Peritonitis, see also Peritonitis 234-239,
Elderly 348 403-405
End-stage renal disease in children 543- Personnel 219
551 Phosphate balance 217
Establishment of program 219-220 Plasma amino acid abnormalities 491-492
European Dialysis and Transplant Associ- Pleuroperitoneal communication 485-487
ation Registry of patients on, see Post-operative leak 476
EDTA registry Potassium balance 217
Exchange rate criteria 137-139 Protein loss 217-218, 356, 490-491
Fatigue 231 Psychological problems 222
First choice 348 Psychosocial aspects, see also Quality of
Follow-up 227 life 667-684
692

Quality of life, see also Quality of life Hematologic parameters 253-255

667-684 Hernias 260-261
Recommended nutrient allowances for High-volume dialysate exchange 251
adult patients undergoing 378-381 Indications 263-265
Removal of patients from 238 Infection rate 198
Renal failure, see also Renal failure 499- Intra-abdominal pressure 265
514 Medical circumstances 264
Sclerosing peritonitis 480-482 Nutritional status 257
Site of training 219-220 Partner preference 264
Skin edema 230-231 Patient and technique survival 261-263
Skin exit site infection 477-480 Patient preference 264
Social problems 222 Peritonitis 255, 258-260, 404
Sodium and water balance 216 Phosphorus 256-257
Solution complications 480-482 Physiologic considerations 250-253
Statistics on patients on 209-2\0 Primary objective 247
Success of 194 Relationship between dialysate volume
Success rates 667-669 and intra-abdominal pressure 251-252
Survival of patients on 659-664 Renal osteodystrophy 256-257
Team work 220 Secondary goal 247
Telephone contact with patient 227 Solutions 248
Theoretical basis 136-137 Steady physiologic state 252-253
Total drain volume 212-214 Technique 247-249
Training of patient 226 Tunnel infections 261
Transfer surface area 221 Ultrafiltration 250, 261
Transport limitations, intermittent perito- Urea clearance 250-251
neal dialysis distinguished 151-153 Continuous endothelium 60-61
Triglycerides 222 Continuous flow compound peritoneal di-
Ultrafiltration failure 480-482 alysis 117-118
Ultrafiltration rate 214-215 Continuous flow intermittent peritoneal di-
Urea clearances 36, 211-212 alysis 122
Urea metabolism 365-366 Continuous flow peritoneal dialysis 117-118
Urea removal 212-2\3 Continuous flow recirculation peritoneal di-
USA registry, see also USA CAPD regis- alysis 117-118
try 597-635 Continuous flow technique 3
Vitamin losses 498-499 Continuous peritoneal lavage, children 527
Vomiting 232 Contractile tone, level of 66
Water flux data, clinical application of Contraction alkalosis 454
215-216 Contra-indications for dialysis
Continuous cyclic peritoneal dialysis (CCPD) Continuous ambulatory peritoneal dialysis
14,119,184-185,194,197,247- 221-222
266,598 Intermittent peritoneal dialysis 186-187
Back pain 264 Contrast agents 318
Biochemical parameters 253-255 Convective mass transfer mechanism 123
Blood pressure 254, 257-258 Convective solute mass transfer 166-170
Calcium 256-257 Convective transport 38-39, 124, 272-274
Catheter exit site infections 261 Effects of 145-147
Catheter outflow obstruction 261 Convulsions 451-452
Children 265, 528, 551 Coping with illness 675-676
Clinical experience 253-258 Copper 317,377,390-391
Complications 258-261 Coronary artery disease 194
Convenience provided by 263 Correction factor 160, 165
Creatinine clearance 250-251 Corticosteroid therapy 346, 544
Cyclers 247-249 Cost function 122
Diabetic patients 257, 262-263, 572-573 Cost of systems 197
Glucose solutions 248 Coulter-counter 235
 693

CPK enzyme elevations 498 Death, causes of 569

Creatinine 68, 164 Hospitalization 568
Concentration profiles 131-132 Hypertension 563
Continuous cyclic peritoneal dialysis 250- Insulin requirements 562-563
251 Macro-angiopathy 566-568
Equilibration 140-141 Metabolic problems 565-566
Removal 213 Method 562-563
Cremaster muscle 54-55, 68, 73, 75 Micro-angiopathy 566-568
Cuff infection 229 Neurological status 568
Cuts 409 Nutritional problems 565-566
Cyclers for continuous cyclic peritoneal di- Peripheral arteritis 567-568
alysis 247-249 Peripheral neuropathy 568
Cyclopeptides 319 Postural hypotension 563
Cyclophosphamide 318 Procedure 562-563
Cycloserine 313 Rehabilitation 568
Cytochalasin D 287 Sclerosing peritonitis 565
Cytoplasm 10 I, 104 Survival rates 569-570
Technical complications 564-565
D Transfer, causes of 569-570
Visual status 567
Continuous cyclic peritoneal dialysis 262-
Daily dietary allowances, 375 263, 572-573
Dairy products 350 End-stage renal disease in 561, 573-576
Death, see Mortality; Mortality rates
Intermittent peritoneal dialysis 570-572
Debendox poisoning 319
Malnutrition 575
Decreased strength 348
Nutritional concerns of 384-386
Decreased ultrafiltration 565
Other forms of peritoneal dialysis 570-573
Defense mechanisms of peritonitis 410-411
Peritoneal dialysis in, see also other sub-
Dehydration 533
topics hereunder 561-579
Denial in coping with illness 676
Dependency versus independency 677-678 Peritonitis 575
Depletion syndromes 457 Diabetic retinopathy 197
Depression 231, 349, 675-676 Diabetic vascular disease 25
Desferoxamine 317, 378, 500 Diabetic visceroneuropathy 384
Desirable body weight 351 Dialysate
Desmosome 103 Bloody 234
Devices for access into peritoneal cavity, Channel features 30-31
chronic intermittent peritoneal di- Children 541
alysis 7-8 Cloudy 549
Dextran 566 Contamination 8
Diabetes mellitus 232, 277, 349, 384 Factory made 4
Diabetic gastroenteropathy 385 First period (1923-1962) 3-5
Diabetic nephropathy and intermittent perito- Flow rates 34-36, 151-152, 270-271
neal dialysis 199-20 I Continuous ambulatory peritoneal di-
Diabetic patients 13-14,233,257,304,561- alysis 211-212
579, 655, 657 Limited 151
Continuous ambulatory peritoneal dialysis Techniques to improve 211-212
348,561-570, 575 Glucose replacement for diabetic patients
Acute peritonitis 564-565 on CAPD 566
Biological status 563-564 Leaks 483-484
Blindness 563 Acute peritoneal dialysis 444
Blood glucose control 562-563, 565 Management 477
Blood pressure 566-567 Post-operative 476
Cessation of treatment 569 Turbidity 236
Clinical status 563-564 Utilization 122
694

Volume, relationship between intra-ab- Dietary protein intake

dominal pressure and 251-252 Children 388-389
Dialysis membrane 23 Urea nitrogen appearance 366-369
Dialysis solutions 298-306 Dietary protein requirements 360-361, 363
Acetate 81-85, 287, 302-303 Dietary recommendations for children 386-
Acidity 73, 77 389
Alkaline 303 Dietary restrictions, noncompliance with 231
Amino acids 305 Diethylenetriaminepentacetate 318
Antibiotics, compatibility of 307-308 Differential diagnosis problems in peritonitis
Bicarbonate 302-303 429-430
Calcium 301 Diffuse mass transfer models 120-121
Commercial 73, 79, 83-84, 116 Diffusion 269-270
Complications due to 480-482 Diffusive transport rates 274
Containers for delivery of 210-211, 234 Digitalis 228, 450
Contaminants 306 Digoxin 277, 317
Continuous cyclic peritoneal dialysis 248 Dihydrotachesterol 221
Electrolytes 300 Dihydroxycholecalciferol 221
Glucose 300, 303-304
Dilantin 317
Hazards associated with 299-300
Dimercapropanol 317
Hyperosmolar 74-75
Hypertonic 76, 141, 304 Dimethyl sulfoxide 287
Hypocapnia 73, 77 Dioctyl sodium sulfosuccinate 287
Isotonic 161, 170 Diphtheroid peritonitis 420
Lactate 81-85,287,302-303 Diplopia 312
Magnesium 300-301 Dipyridamole 85,215,277,282-283
Microcirculation, effects on 73-85 Peritoneal efficiency affected by, 280
Non-glucose osmotic agents 305 Discontinuous endothelium 60
Osmolality 73-77 Disequilibrium ratio 149
Osmotically active agents 303-304 Disequilibrium syndrome 135, 451-452, 458
pH and PC0 2 77-81 Children 544
Polymers 305-306 Disinfectant in tubing 431
Potassium 302 Dissociation 159-160
Risks associated with 299-300 Distension 215, 228-229
Sodium 301-302 Diuretics 228
Solute clearances, effects on 73-85 Diverticulitis 428
Sorbitol 304-305 Diverticulosis 408, 460
Tonicity in continuous ambulatory perito- Docusate sodium 287
neal dialysis 214-215 Dopamine 68, 288
Dialysis solution chamber, capacities and
Dopaminergic receptors 272, 281
pressures in 44-46
Double-cuffed catheter 223
Dianeal 73-74,82, 125-126, 140,210
Doxylamine succinate 319
Diaphragmatic contractions 57
Diaphragmatic defect 444 Drainage problems, acute peritoneal dialysis
Diaphragmatic hernia 222, 488, 544 444-445
Diaphragmatic lymphatics 101 Drug effects in peritoneal vasculature, 65-73
Diaphragmatic strength 45 Permeability effects 68-73
Diarrhea 234, 414, 423, 566, 568 Vasoactive effects 66-68
Diazoxide 282-283,317 Drug intoxications 314
DicloxacilIin 310 Duodenal ulcers 232
Dicyclomine hydrochloride 319 Dwell time 212-215, 221, 224, 248
Dietary fiber 379 Dysgeusia 390
Dietary intake, assessment of 350 Dyslipoproteinemia 345, 372
Children 386 Dyspepsia 232
Dietary management, continuous ambulatory Dyspnea 44
peritoneal dialysis 218-219 Dysuria 460
 695

E Quality of life 550

Renal osteodystrophy 549
Early clinical experience 1-2 Renal transplantation 550
Edema 53-54, 72, 215, 321, 349, 547 Survival 550
EDTA registry 637-665 Ultrafiltration 547
Demography 641-645 Water balance 547
Function 637 Continuous ambulatory peritoneal dialysis
Management of patients on CAPO 652- as method of treatment of 209
659 Diabetes as growing cause of 561
Patient selection for CAPO 645-652 Diabetic patients, see also Diabetic pa-
Patients treated by CAPO 637-641 tients 573-576
Survival of patients on CAPD 659-664 Gradual development of 346
EEG abnormalities 458 Insulin supplementation 319-320
Efficiency of peritoneal dialysis 51-52 Malnutrition 353
Cardiovascular factors influencing 65 Peritonitis, see also Peritonitis 234
Children versus adult 529-530 Psychosocial aspects, see also Quality of
Peritoneal vasculature, role of 54 life 667-684
Elderly patients 197, 348 Quality of life, see also Quality of life
Electrolyte disturbance 186 667-684
Electrolytes 300 End stage renal failure 159
Children 389 Enemas 445
Emergency hemodialysis 553 Energy requirements 374-376
Encapsulating and sclerosing peritonitis 109 Children 387-388
Endocrine disturbances 508 Enterobacteria peritonitis 420
Endogenous infections 408-409 Enterococcus peritonitis 420
Endogenous mediators, effects of 68-70 Environmental infections 409
Endoplasmic reticulum 101, 104 Eosinophilic cells 410
Endothelial cell contraction 84 Eosinophilic peritonitis 422-423
Endothelial cells 59 Eosinophils 411
Endothelial gaps 61-62, 73, 84 Epigastric arteries 96
Endothelial lining 64-65 Epigastric distress 232, 235
Endothelial resistance 30 Epsilon-aminocaproic acid 318
Endothelium 60-62 Equilibrium peritoneal dialysis technique 13
Endotoxins 306, 423 Erosive gastritis 232
End-stage renal disease (ESRD), see also Eructation 232
Renal failure 141 Erythrocyte transketolase activity index 381
Children 543-551 Erythromycin 312
Anemia 549 Esophageal reflux 232
Complications 548-549 Esophageal varices 53
Continuous ambulatory peritoneal di- Establishment of programs, continuous am-
alysis 543-551 bulatory peritoneal dialysis 219-220
Continuous cyclic peritoneal dialysis Ethambutol 313
551 Ethanol 316
Growth 547-548 Ethochlorvynol315-316
Hypertension 549 Ethylene glycol 316
Intermittent peritoneal dialysis 543 Eucalyptus oil intoxication 319
Mechanics of dialysis 546-547 European Dialysis and Transplant Associa-
Nutritional concerns 548 tion Registry, see EDTA registry
Parathyroid hormone levels 549 Examination of peritoneum 96-100
Parent fatigue 550-551 Excess body water removal 159
Patient selection 544-551 Exchange transfusion 553
Peritoneal access 545 Exchange volumes, children versus adult 530
Peritoneal catheter 545 Excitement 51
Peritoneal dialysis for 543-551 Exercise 51, 381-382
Peritonitis 548-549 Exit site infections 261, 415-416, 565
696

Treatment 427 G
Extracellular fluid concentrations 38
Extracellular fluid volume expansion 349 Gallamine 319
Gallium-67 318
Gas clearances 28, 40-41
F Gastrodialysis 180
Gastroenteropathy 385, 566, 568
Failure rates 668 Gastrointestinal disorders 349
Failure to thrive 345, 347-348 Gastrointestinal hemorrhage 186
Family relationship 678 Gastrointestinal problems 179
Fatigue, continuous ambulatory peritoneal Gastronomy 447
dialysis 231 Gastroschisis 544
Fat restriction 498 Gelatin 305
Fecal peritonitis 408, 415 Gemfibrozil 498
Fenestrae 61, 104 Gentamicin 307-308,311,425
Fenestrated endothelium 60-61 Gibbs-Donnan equilibrium 38
Ferritin 63 Giving up syndrome 675
Fever 235, 414 Glomerular filtration rate (GFR) 384
Fibrin 409-410 Glucagon 68, 285-286, 371
Glucose 3-4, 318
Fibrin clots 409, 445
Absorption 53,350,370-371,383,493-
Fibrin filaments 409
498
Fibrin formation 225
Continuous ambulatory peritoneal di-
Fibrinogen 409
alysis 216
Fibrinolysin 409 Lipid abnormalities and 493-498
Fibrinolytic activities 409 Rate 42
Fibroblasts 103 Caramelization, avoidance of 4
Fibrosis 456 Continuous cyclic peritoneal dialysis 248
Ficks' Law of Diffusion 121 Electrolytes and 300
Filamentous fungal peritonitis 422 Hypertonic solutions to remove salt and
First period (1923-1962), see History of water 219
peritoneal dialysis Metabolism 345
Flucytosine 312-313 Objections to 303-304
Fluid exchange 26 Osmolality regulation with 159
Fluid film resistances 116 Osmotic agent in continuous ambulatory
Fluid films 28-29 peritonal dialysis 214-215
Urea clearance limited by 30-32 Solute 160-161
Fluid overload 449 Ultrafiltration and 159-174
Fluid reabsorption rates 128 Glucose-6-phosphate dehydrogenase activity
Fluid transfer parameters 126-130 500
Fluoride 316-317 Glutamine 345
Gluthethimide 315
Fluoroacetate 84
Glycerine 305
5-Fluorocytosin 425
Glycine 53
5-Fluorouracil 318
Glycosylated hemoglobin values 565
Fogarty catheters 225 Gold 317
Foley catheter 446 Golgi apparatus 101, 104
Folic acid 381, 390, 498-499 Gortex peritoneal catheter 480
Follow-up, continuous ambulatory peritoneal Gram negative aerobic bacteria 410
dialysis 227 Gram-negative organisms causing peritonitis
Forced diuresis 315 420-421
Fructose 304, 566 Gram-positive organisms causing peritonitis
Fungal peritonitis 236, 239, 313, 412, 422 418-420
Furosemide 274, 287-288, 315 Gross anatomy of peritoneum 95-96
 697

Growth hormone 509-510 Irrigation 456

Growth of children, end-stage renal disease, Treatment of peritonitis 426
peritoneal dialysis for 547-548 Heparin sulfate 373
Growth retardation 386, 390 Hepatic coma 554
Guilt 350 Hepatic congestion 277
Hepatic gluconeogenisis 345
Hepatic triglyceride lipase activity 372
H Hernia 213, 487-488
Continuous ambulatory peritoneal dialysis
Headache 135, 211, 451 222,233
Head trauma 186 Continuous cyclic peritoneal dialysis 260-
Health Care Financing Administration 261
(HCFA) 184 Hernia of Morgagni 488
Heart disease, underlying 450 Hexachlorophene 319
Heart failure 321, 449 Hiatus hernia 488
Heavy metals, see also specific types 316- High density lipoproteins (HDL) 372, 494-
317 497,565
Hematocrit 8, 13 High protein diets 217-219
Hematogenous infection 408 High-risk patient popUlation, continuous am-
Hematologic disorders 499-500 bulatory peritoneal dialysis 220
Hematologic parameters, continuous cyclic Histamine 65, 72, 282-283,409
peritoneal dialysis 253-255 Albumin leakage 71
Hematologic problems 196 Local release of 33
Hemocytometer 235 Histamine-induced change in vascular
Hemodialysis 13, 135,315,317,347,446- permeability 69
447 History of peritoneal dialysis 1-22
Advantages of continuous ambulatory per- Children, use in 526-528
itoneal dialysis over 141 Chronic intermittent peritoneal dialysis, 7-
Clearances 191 13
Comparison with peritoneal dialysis 6 Continuous ambulatory peritoneal dialysis
Congenital hyperammonemia 553 (CAPO) 13-15
Conversion to II First period (1923-1962) 1-7
Diabetic patients 575 Hollow fiber dialysis 23
Dialysate leak 444 Hollow fiber dialyzer 33-38, 165
Excess body water removal 159 Hollow fiber dialyzer fluid 28
Intermittent peritoneal dialysis chosen in- Hollow fiber kidney, dialysate channel fea-
stead of 186-187 ture for 30-31
Return to 238 Home dialysis, choice of treatment modality
Teflon arteriovenous shunt 180 for 679-680
Transfer to continuous ambulatory perito- Home peritoneal dialysis 196-197
neal dialysis 222, 231 Hormonal manipulation 297
Uremia treatment 146-150 Hormones 51
Hemodynamic instability 194-195, 200 Horseradish peroxidase 62
Hemodynamic stress 200 Hospital back-up, continuous ambulatory
Hemolytic-uremic syndrome (HUS) 277, peritoneal dialysis 220
533, 536-537 Humoral factors in peritonitis 411
Hemorrhage, see Bleeding Hyaline membrane disease 554
Hemorrhagic hypotension 277 Hydralazine 282-283
Hemorrhoids 53, 222, 233 Hydrocele 488
Heparin Hydrostatic pressure 276
Bleeding and 443 Hydrothorax 483,485-487
Continuous ambulatory peritoneal dialysis Hyperactivity 452
224-225, 234 Hyperammonemia 322
Dialysate solutions and 307-308 Hypercalcemia 179, 218, 256-257, 301,
Intermittent peritoneal dialysis 185, 197 391,542
698

Hypercapnia 78 Hypopermeability 109

Hypercatabolism 186, 267-268, 346 Hypophosphatemia 383,542
Hyperchloridemia 173-174 Hypoproteinemia 511, 542
Hypercholesterolemia 498 Hypotension 24-25, 141, 235, 304
Hyperglycemia 304, 384-386, 449, 452, 542 Continuous ambulatory peritoneal dialysis
Continuous ambulatory peritoneal dialysis 230
232-233 Hypothalamic anovulation 511
Hyperkalemia 217,227,267,319,386,450, Hypothalamopituitary axis aberration 510
454,498 Hypothermia 315
Hyperlacticacidemia 303 Hypothyroidism 508
Hyperlipidemia 257,304,373,496-498,566 Hypovolemia 449, 511-513, 567
Hypernatremia 173, 195,216,248,301,
318,449,453
Hypernatremic dialysate 448 I
Hyperosmolality 304, 319
Hyperosmolar coma 305 Ileostomies 222
Hyperosmolar dialysis solutions 74-75, 179 Ileus 186, 235
Hyperosmolarity 195 Paralytic 446-447
Hyperparathyroidism 256, 345, 379, 505 Iliac arteries 96
Hyperpermeability 109 Immunoglobulins 411
Hyperphosphatemia 198, 218, 256, 391, 456 Immunosuppression 222, 277
Hyperpotassemia 2 Immunosuppressive drugs, see
Hypersensitivity reactions 424-425 Immunosuppression
Hypertension 51, 58, 346, 386, 449, 451- Immunotherapy 318
452, 511 Impotency 390
Children 547 Inborn errors of metabolism 553-554
Continuous ambulatory peritoneal dialysis Incisional hernias 488
215-216, 230 Indications for dialysis
Continuous cyclic peritoneal dialysis 257 Acute renal failure in children 534-536
Diabetic patients 563, 566-567 Continuous cyclic peritoneal dialysis 263-
End-stage renal disease in children 549 265
Intermittent peritoneal dialysis 198 Intermittent peritoneal dialysis 186-187
Hypertonic dialysis solutions 30, 76, 141, Indomethacin 69, 72, 284-285, 298
304 Indwelling catheters, chronic intermittent
Hypertonic peritoneal dialysis 24 peritoneal dialysis 10-12
Hypertriglyceridemia 345,372,379,381, Indwelling devices 7, 8
385, 388, 493-498 Infants, see Children
Continuous ambulatory peritoneal dialysis Infections, see also specific types 382
219, 222 Risk of II
Continuous cyclic peritoneal dialysis 257- Susceptibility to 348
258 Infectious diarrhea 533
Diabetics 566 Inflammation 33, 42
Intermittent peritoneal dialysis 198 Inflammatory mediators 409
Hypervascularization 105 Inflammatory response 409
Hypnosedatives 452 Inflow pain 442
Hypnotics 315-316 Infusion pain, see Pain
Hypoalbuminemia 356, 566 Inguinal hernia 222, 233
Hypocalcemia 450, 458 Inpersol 73, 78-79, 82
Hypocapnia 73, 77 Insulin 232, 257-258, 307, 319-320,371,
Hypogeusia 377 452,498
Hypoglycemia 451-452 Clearances 32
Hypoinsulinemia 384 Diabetic patients 562-563, 573
Hypokalemia 227,315,542 Dosages 228
Hypomagnesemia 542 Insulin-dependent diabetics
Hyponatremia 318-319, 450, 542 Clinical and blood parameters 564
 699

Comparison of dialysis methods 561-562 Renal replacement therapy, see also other
Survival rate 569 subtopics hereunder 179-208
Insulinopenia 385 Transport limitations, continuous ambula-
Intercellular gap junction 60 tory peritoneal dialysis distinguished
Intercell ular gaps 41-42, 44 151-153
Intercellular junctions 275, 287 Urea clearances 35-36
Intercostal arteries 96 Intermittent recirculation peritoneal dialysis
Intercurrent illness, nutritional support dur- 117-118
ing 382-384 Intermittent technique, see also Intermittent
Intermittent partial obstruction 565 peritoneal dialysis 3
Intermittent peritoneal dialysis (IPD), see Interstitial cellular infiltration 44
also Chronic intermittent peritoneal Interstitial diffusion path 116
dialysis 13, 130, 184-185 Interstitial resistance, urea clearance limited
Acute renal failure 185-187 by 30-32
Children 527-528 Interstitium 26-27,29-31,44,63-64, 116
Chronic renal failure 187-201 Transmission electron microscopy 103
Abnormal lipid metabolism 198-199 Intestinal motility 51
Adequacy of dialysis 193-194 Intestinal perforation 428
Biased patient selection 192-193 Intoxications, see also specific agents
Changes in dialysis regimens 191 Peritoneal dialysis for children with 552-
Children 197-198 553
Diabetic nephropathy, see also Diabetic Intraabdominal adhesions 445-446, 456, 544
nephropathy and intermittent perito- Intra-abdominal catastrophes 225
neal dialysis 199-20 I Intra-abdominal disease 186
Elderly patients 197 Intra-abdominal pressure 44-46, 222, 229
Exhaustion of vascular access sites 194 Complications due to 483-490
Hematologic problems 196 Continuous cyclic peritoneal dialysis 265
Hemodynamic instability 194-195 Relationship between dialysate volume
Home peritoneal dialysis 196-197 and 251-252
Important aspects of IPD 193-201 Intraluminal infections 406-408
Logistics of out-patient IPD 197 Intraperitoneal adhesions 308, 409
Long-term survival 187-189 Intraperitoneal hemorrhage 423, 542
Malnutrition 198 Intraperitoneal insulin 320
Osteodystrophy 199 Intraperitoneal nitroprusside, see
Peritonitis 198 Nitroprusside
Possible reasons for overoptimism about Intra-peritoneal volume 44-46
IPD 192 Intraperitoneal volume profile 126-128, 143
Protein loss 198 Intravascular fluid overload 319
Theoretical considerations regarding Intravascular volume depletion 215
cause of IPD failure 189-192 Inulin 164, 166-167
Thirst 195 Concentration profiles 131, 133
Unsophisticated methodology 192 Iodine 318
Clearances 189-191 Iron 316-317,378,380,390
Diabetic patients 570-572 Deficiency 352
Dialysate flow 34-36 Loading 352
Efforts to increase efficiency of 211-212 Ischemia 409
End-stage renal disease in children 543 Ischemic bowel disease 408
Future of 202-203 Ischemic heart disease 450
Historical perspecti ve 179-185 Isoniazid 3 13
Impact of other forms of peritoneal di- Isopropyl alcohol 316
alysis on 202 Isoproterenol 66-67, 69, 71, 278
Outlook for future of 203 Enhancement of peritoneal mass transport
Patients who should be treated by 202 278-279
Peritonitis 404-405 Isotonic dialysis solutions 161, 170
Protein loss 355-356 Italian corkscrew 456
700

Itching 218 Low density lipoproteins (LDL) 372-373,

494-497
Luer-Iock connections 466
K Lumbar arteries 96
Lumbosacral spine pain 233
Kanamycin 311 Lymphatic capillaries 116
Keto-acids, removal of 322 Lymphatic circulation 52
Ketoconazole 313 Lymphatic lacunae 101
Kidney stones 460 Lymphatic system 53-54
Kinetic studies of dialysis 145-150 Lymphatic vessels 54, 56, 104
Kinetics of peritoneal mass transfer 581-595 Lymph formation 62
Developmental aspects in children 529- Lymphocyte mediators 411
532 Lysol319
Measurement of KBD 584-585
Methods 583-584 M
Other model applications 585-586
Results and discussion 586-594
Theory 584-586 Macro-angiopathy, diabetic patients on
Krebs dialysis solution 75-76, 82-83, 287 CAPO 566-568
Kwashiorkor 345 Macromolecular barrier to protein movement
Kyphosis 506 60
Macromolecular clearance 57
Macromolecular leakage 58, 69
L Macromolecular loss 73
Macromolecular movement 62
Macrophages 105, 107-108
Laboratory tests, continuous ambulatory per-
Macula adhaerans 103
itoneal dialysis 227-228
Magnesium 380
Laceration of internal organs 445-446
Nutritional concerns 362, 364
Lactate 109, 287, 302-303
Magnesium solution 300-301
Dialysis solutions 81-85
Maintenance hemodialysis 345, 359-360
Lactic acidemia 452 Malabsorption 384
Lactic acidosis 303, 316 Malaise 194, 346
Laryngeal edema post anesthesia 229 Malignant hypertension 25, 277
Lead 317 Malnutrition 198, 255, 345-349, 355, 384-
Leg-length discrepancy 506 385, 388, 391
Lethargy 139, 235, 553 Diabetic patients 566, 575
Leucinosis 322 Protein 369-370
Leucopenia 391 Mannitol 319
Leukocytes 107 Maple-Syrup-Urine disease 322, 553-554
Leukocytosis 514 Marasmus 345
Life style changes 498 Marital relationship 678
Light microscopy of peritoneum 96-98, 105 Massive hydrothorax 483, 485-487
Lincomycin 310 Mass transfer
Lipid abnormalities 493-498 Causes of changes in 582
Lipids 371-374 Children versus adult 530
Diabetic patients on CAPO 565-566 Kinetics of, see also Kinetics of peritoneal
Transport of 274 mass transfer 581-595
Lipoprotein lipase, loss of 373 Limited 151
Lithium 318 Rate 123
Living and dying 669-670 Resistance 116
Local diseases 51 Mass transfer area coefficient (MTAC) 40-
Loculation of fluid 445 41, 120-122, 124, 126, 130-131,
Loss and threat of loss 671-672 139-140, 152, 581
Low back pain 233, 443, 507-508 Children 142-145, 531-532
 701

Mass transfer/ultrafiltration volume/plasma Metabolic complications 186, 452

water concentrations 40 Acute renal failure in children 542
Mass transport, see Peritoneal transport Continuous ambulatory peritoneal dialysis
McGaw dialysis solution 73, 75, 80-82 490-499
Mean pore size 36-37 Diabetic patients on CAPO 565-566
Mechanical complications, see also Catheter- Metabolic problems 345
related complications 185 Metabolic products 51
Acute peritoneal dialysis 442-447 Metabolite generation 124
Diabetic patients on CAPO 564-565 Metabolites 317
Mechanics of dialysis Metaclopramide 385
Children 541-542 Metarterioles 59
Diabetic patients on intermittent peritoneal Metastatic calcifications 256
dialysis 570-571 Methanol 316
End-stage renal disease in children 546- Methaqualone 316
547 Methicillin 310
Medical complications of acute peritoneal di- Method of dialysis in first period (1923-
alysis 448-454 1962) 2-3
Medical history of patient 349-350 Methotrexate 318
Medications 349 Methyldopa 452
Continuous ambulatory peritoneal dialysis Methylprednisolone 69
228 Methylprylon 315
Mefenamic acid 69, 72 Metronidazole 312
Megaloblastic 391 Mezlocillin 310
Membrane surface-active agents 287-288 Miconazole 313
Membrane transport limitations 138-141 Micro-angiopathy, diabetic patients on
Difference between CAPO and IPD 151- CAPO 566-568
153 Microbiological filter 235
Membrane-transport problems 152-153 Microcirculation, see Peritoneal
Menometrorrhagia 510 microcirculation
Menstrual irregularities 510 Micropinocytotic vesicles 59-61
Menstrual periods 234 Microvilli 44, 97-98, 101
Menstruation 423 Micturition disorders 568
Mental confusion 452 Mid-arm muscle circumference (MAMC)
Meprobamate 315 351-353
Mercury 316-317 Middle molecular hypothesis 215
Mesenteric arteries 53, 95 Middle molecule removal, continuous ambu-
Mesenteric blood flow 271-272, 285 latory peritoneal dialysis 214
Mesenteric vasodilation 281 Middle molecules 369-370
Mesentery 53 Milk and milk products 218
Mesentery mesothelium 64 Milky spots 10 I
Mesentery vasculature 67 Millipore filter 431
Mesocecum 66-67 Mineral balance 361-365
Mesothelial cells 25-26, 10 I, 116, 411 Misperceptions leading to failure of perito-
Mesothelial desquamation 107, 109 neal dialysis 348
Mesothelial intercellular junctions 10 I-I 02 Mitochondria 10 I, 104
Mesothelial intercellular resistance 30 Model of peritoneal dialysis 123-126
Mesothelial layer 29, 10 1-103 Molecular size, continuous ambulatory peri-
Mesothelial regeneration 107-108 toneal dialysis 212-214
Mesothelial resistance 37, 44 Mononuclear cells 411
Mesothelium 63-64, 116 Morbidity 441, 451
Metabolic acidosis 447,454,541 Mortality 441
Metabolic alkalosis 303, 453-454 Acute peritoneal dialysis 454
Metabolic alterations of specific nutrients by Continuous ambulatory peritoneal dialysis
peritoneal dialysis, see also Nutri- patients 657, 659
tional concerns 355-378 Diabetic patients on CAPO 569
702

Peritonitis 428-429 Neuromuscular blockade 319

Pulmonary cqmplications 451 Neuropathy 193, 200-201
Rate 668 Neuropsychiatric problems 458
Acute renal failure 185 Neutrophilic peritonitis 423
Acute renal failure in children 536-537 New catheters 432
Complications as factor 454 Nicotinic acid 498
Motivation for continuous ambulatory perito- Nifedipine 317
neal dialysis 220-221, 231 Nightly automated peritoneal dialysis 184
Movement disorders 222 Nitrogen balance 218, 360-362
Moxalacatam 309 Nitrogen in diet 362-363
Mucopolysaccharides 63 Nitrogen metabolism in CAPD 365-366
Mucositis 193 Nitroprusside 25,32,66-68,70-72,277,
Muscle cramps 211 282-283
Muscle mass 351-352 Augmentation of peritoneal mass transfer
Muscle toxicity 498 280
Muscle wasting 193-194, 347, 354, 384 Rate of dilation 75
Muscular weakness 349 Nocardia peritonitis 311
Mushroom poisoning 319 Nocturnal hypoglycemia 233, 320
Myalgias 498 Non-compliance with protocol, continuous
Mycobacteria causing peritonitis 421 ambulatory peritoneal dialysis 231-
Myobacterium chelonei 198,409,421 232
Mycobacterium tuberculosis 421 Non-glucose osmotic agents 305
Myocardial infarction 450 Non-oliguric acute renal failure 532-533
Myoendothelial junctions 59-60 Norepinephrine 65, 68,216,277,283,512
N-Myristyl-j3-amino-proprionate 287 I-Norepinephrine 281-282
N-Myristyl-beta-aminopropionic acid 298 Norepinephrine contracted ring 82-83
Normal peritoneum 96-98, 105
Normal regulatory mechanisms 65
N Novel applications of peritoneal dialysis 319-
322
Nafcillin 307 Nutrient loss 346
Nausea 135, 139, 349, 382, 414, 452 Nutritional concerns 345-40 I
Children 565-566, 568 Alterations in metabolism of specific nu-
Continuous ambulatory peritoneal dialysis trients by peritoneal dialysis, see also
211,235 other subtopics hereunder 355-378
Neomycin 425 Amino acids 358-360
Neonatal hyperbilirubinemia 554 Assessment of nutrient intake 350
Neonates, see Children Body composition 350-352
Nephrostomies 222 Calcium 362, 364-365
Nephrotic syndrome 2 Carbohydrate 370-371
Nephrotoxicity 425 Children, see also Children 386-391
Nerve conduction 568 Continuous cyclic peritoneal dialysis 257
Net electrolyte sieving effects with peritoneal Daily dietary allowances 375
ultrafiltration 38-40 Diabetic patients 384-386
Netilmicin 311, 425 Diabetic patients on CAPD 565-566
Net ultrafiltration 25-26 Dietary protein intake, urea nitrogen ap-
Neural inputs 51 pearance 366-369
Neurological complications 186, 451-452 Dietary protein requirements 360-361, 363
Chronic intermittent peritoneal dialysis End-stage renal disease in children 548
457-458 Energy requirements 374-376
Neurological defects, continuous ambulatory Exercise 381-382
peritoneal dialysis 222 History of patient 349-350
Neurological disease 222 Intercurrent illness, nutritional support
Neurological status of diabetic patients on during 382-384
CAPD 568 Lipids 371-374
 703

Magnesium 362, 364 Osmotic pressure 39

Methods to assess nutritional status of per- Osmotic pressure gradient 42
itoneal dialysis patient 349-354 Osteitis fibrosa 256, 501, 505
Middle molecules 369-370 Osteodystrophy, see Renal osteodystrophy
Mineral balance 362-365 Osteomalacia 199, 256, 380, 391,459,501,
Misperceptions leading to failure of peri- 505,508
toneal dialysis 348 Osteopenia 256
Nitrogen 362-363 Osteoporosis 506
Nitrogen metabolism in CAPO 365-366 Ostomies, continuous ambulatory peritoneal
Nutritional status of CAPO patients 353- dialysis 222
354 Ototoxicity 312, 425
Peripheral neuropathy 369-370 Outflow failure 478
Peritoneal dialysis in 1960s and 1970s 346 Outflow obstruction 455, 467
Phosphorus 362-364 Outflow pain 442-443
Physical examination of patient 349-350 Out-patient intermittent peritoneal dialysis,
Potassium 362-363 logistics of 197
Protein loss 355-356 Out-patient irrigation technique 223-224
Protein malnutrition 369-370 Ovarian cystic disease 510
Recommended nutrient allowances for Overheated dialysate 447
adult patients undergoing CAPO 378- Ovulation 423
381 Oxacillin 310
Serum protein levels 357-358 Oxalate 316
Serum proteins 352-353 Oxygen administration 321
Trace elements 377-378 O-Z connection 430, 466-467
Urea metabolism in CAPO 365-366
Vitamins 376-377
p
Nutritional status of patients, see Nutritional
concerns
Pancreatitis 221, 429
Pan retinal photocoagulation 567
o Papaverine 66-67, 69
Pappenheimer theory of restricted diffusion
Obesity 219,231,304,371,378,381,505 across capillaries 273
Obstruction of portal system 53 Paracetamol 315
Obstructive lesions 533 Parathyroidectomy 458
Obstructive uropathy 544 Parathyroid hormone levels 549
Obturator hernia 488 Parathyroid hormone secretion, stimulation
Oliguria 2 of 218
Oliguric acute renal failure 532-533 Parent fatigue, children with end-stage renal
Omenta 53 disease on CAPO 550-551
Omental tissue entrapment 456 Pargyline hydrochloride intoxication 316
Omphalocele 544 Parietal lymph nodes 53
One-way catheter obstruction 444-445, 455- Parietal mesothelium 64
456,467 Parietal peritoneum 23-24, 52, 95-96, 116
Opsonins 411 Arterial blood supply to 53
Oral hypoglycemic agents 318 Patient-machine relationships 672-673
Orange juice 217 Patient-peritoneal dialysis system, theoretical
Organo-phosphorus poisoning 319 analysis of 119-123
Orosomucoid 373 Patient selection
Orthostatic hypotension 215-216, 230, 511- Continuous ambulatory peritoneal dialysis
513 program 220-221
Osmolality 73-77, 159 EOTA registry statistics relating to CAPO
Osmotic agents 122 645-652
Osmotically active agents 303-304 End-stage renal disease in children, perito-
Osmotic laxative 350 neal dialysis for 544-551
704

Peritonitis prevention 431-432 Peritoneal capillary blood flow 24-29

Patient training 219 Resistances for large and small solutes 36-
Pe02 dialysis solutions 77-81 38
Pediatric enzyme deficiencies 322 Urea clearances, limitation of 30-32
Pediatric patients, see Children Vascular permeability as major resistance
Penicillamine 317 for larger solutes 32-33
Penicillin 307, 310, 425 Vesicles vs intercellular gaps 41-42
Percutaneous catheter placement technique in Peritoneal fibrosis 480-482-
children 538-539 Peritoneal inflammation 33
Perforated bowel 236, 238 Peritoneal lavage 425-426
Perforated ulcer 430 Peritoneal mass transfer, see Mass transfer
Perforation of hollow viscus 186 Peritoneal membrane 23, 410
Perforation of internal organs 445-446 Black box description 160
Pericarditis 193,200-201,346,513-514 Correction factor 160
Pericatheter herniation 478 Organs covered 51
Pericytes 62, 104 Permeability differences 297
Periluminal infections 408 Peritoneal microcirculation 25, 33,44, 51-
Perinatal hypoxia 533 93,96
Peripheral arteritis 567-568, 575-576 Autacoid effects in peritoneal vasculature
Peripheral edema 511 65-66
Peripheral insulin resistance 345 Circulation of blood in peritoneal cavity
Peripheral neuropathy 349, 369-370, 457- 52-54
458,568 Dialysis solutions, effects of, see also Di-
Peripheral vascular disease 568 alysis solutions 73-85
Peritoneal absorption 298 Drug effects in peritoneal vasculature, see
Peritoneal access, CAPD for children with also Drug effects in peritoneal vascu-
end-stage renal disease 545 lature 65-73
Peritoneal adhesions 428 Peritoneal vasculature, architecture of 54-
Peritoneal biopsy 96 58
Peritoneal cancers 58 Peritoneal pain 229
Peritoneal capillary blood 24-29 Peritoneal permeability, factors influencing
Peritoneal catheters, see also Catheters 222- 297
225, 229, 545 Peritoneal surface area, child versus adult
Peritoneal cavity 116, 410
529
Barriers of solute exchange between peri-
Peritoneal transport
toneal vasculature and, see also Bar-
Acceleration of, mechanisms of 274-277
riers of solute exchange between
Acceleration of specific solutes 289-290
vasculature and peritoneal cavity 59-
Amphotericin B 288-289
65
Arachidonic acid 283-285
Child versus adult 529
Augmented 267-268
Circulation of blood in 52-54
Barbiturates 289-290
Peritoneal clearances 7, 24-25, 200-201
Bolus administration versus continuous in-
Peritoneal dialysis, see specific topics
Peritoneal dialysis clearance 136-137 fusion 286
Peritoneal dialysis optimizations 121-122 Catecholamines, influence of 281-282
Peritoneal dialysis system 23-50 Cholecystokinin 286
Capillary dimensions 33-35 Convective 272-274
Capillary kidney 23 Dialysate flow rate 270-271
Comparisons in animals and humans 40- Diazoxide 282-283
41 Diffusion 269-270
Dialysate flow 34-36 Dipyridamole 282-283
Dialysis solution chamber, capacities and Dipyridamole as affecting efficiency 280
pressures in 44-46 Electrolyte concentrations 287
Mesothelial resistance, importance of 44 Glucagon 285-286
Net electrolyte sieving effects with perito- Histamine 282-283
neal ultrafiltration 38-40 Hydralazine 282-283
 705

Increase above normal values 278 Endogenous mediators 68-70

Inefficient 267-268 Macromolecular leakage 58
Isoproterenol enhancement of 278-279 Peritoneography 545
Lipids 274 Peritoneum, see also Peritoneal ultrastructure
Mechanisms of 268 116
Membrane surface-active agents 287-288 Peritonitis, see also specific types 2, 7, 13,
Mesenteric blood flow 271-272 42-44,51, 105,313,403-439,447,
Nitroprusside 282-283 655,657
Nitroprusside augmentation of 280 Acute renal failure in children 542
Norepinephrine 283 Adhesion formation 186, 189,428
Osmotic gradient 289 Antibiotic sensitivities 414
Other hormones and drugs affecting 286- Antibiotic therapy 238, 308, 431
287 Auxiliary devices for handicapped 431
Pharmacologic manipulation of, see also Bacteria, see also specific types 418-421
other subtopics hereunder 267-296 Bacterial diagnosis 414
Phentolamine 282-283 Bloody fluid 423
Phenytoin 290 Causative organisms 416-423
Procaine 286 Cell counts 414
Prostaglandin modulation of 283-285 Cellular factors 411
Rationale for augmenting transport rates Cellular response 410
267-268 Chemical 422
Restoration of decreased rates toward nor- Clinical course 415-416
mal 277-278 Complications 428-429
Salicylates 290 Contamination of dialysate 8
Secretin 286, 288 Continuous ambulatory peritoneal dialysis,
Serotonin 286 see also other subtopics hereunder
Streptokinase 286 221,225,231,234-239,403-405
Theophylline 279 Continuous cyclic peritoneal dialysis 255,
Tolazoline 282-283 258-260, 404
Ultrafiltration rate 272-274, 288-289 Continuous peritoneal lavage 403
Vasodilator gastrointestinal hormones 285- Cryptogenic 422-423
286 Culture procedure 412-414
Vasodilators 278-283 Defense mechanisms 410-411
Vasopressin 286 Definition 415
Peritoneal tumors 58 Diabetic patients 575
Peritoneal ultrastructure 95-113 Diagnosis 235
Continuous ambulatory peritoneal dialysis Differential diagnosis problems 429-430
105-110 Disinfectant in tubing 431
Gross anatomy 95-96 Diverticulitis 428
Light microscopy 96-98, 105 Early treatment 236
Normal peritoneum 96-98, 105 End-stage renal disease in children 548-
Scanning electron microscopy 97-10 I , 549
107-108 Environmental infections 409
Technics of examination 96-110 Evaluation 432
Transmission electron microscopy 10 1-104 Exit site infections 415-416
Uremic patients 104-110 Fibrin 409
Peritoneal vasculature, architecture of 54-58 Frequent occurrence of 403
Altered 58 Fungal 422
Autacoid effects 65-66 Future considerations 432-433
Barriers of solute exchange between peri- Gram stain 412
toneal cavity and, see also Barriers of Hematogenous infections 408
solute exchange between vasculature Humoral factors 411
and peritoneal cavity 59-65 Incubation period 415
Drug effects, see also Drug effects in per- Inflammatory mediators 409
itoneal vasculature 65-73 Inflammatory response 409
706

Intermittent peritoneal dialysis 198, 200, Phagocytic polymorphonuclear cells 411

404-405 Phagocytosis 306, 411
Intestinal perforation 428 Phalloidone 319
Intraluminal infections 406-408 Pheniramine maleinate 319
Intraperitoneal hemorrhage 423 Phenosulfonphthalein 297
Length of symptoms 415 Phenoxybenzamine 272, 277
Menstruation 423 Phentolamine 281-283
Microbiological diagnosis 412-414 Phenytoin 290
Microbiological filter 235 Phosphate balance, continuous ambulatory
Millipore filter 431 peritoneal dialysis 217
Mortality 428-429 Phosphate binding agents 228
New catheters 432 Phosphate binding antacids 350
New developments 234-235 Phosphate dialysis 217
Nutrition and 347-348, 356, 382 Phospholipase A2 73
Nylon catheter 403 Phosphorus 377, 380
Other endogenous infections 408-409 Children 389-390
Ovulation 423 Concentration, continuous cyclic perito-
O-Z connection 430 neal dialysis 256-257
Pathogenesis 404-410 Nutritional concerns 362-364
Patient selection 431-432 Physical examination of patient 349-350
Periluminal infections 408 Physiological tissue differences 51
Pharmacologic manipulation 267-268 Physiologic considerations, continuous cyclic
Portals of entry 405-406 peritoneal dialysis 250-253
Prevention 235-236, 430-432 Pinocytotic vesicles 98, 101, 298
Recurrence 416 Pituitary dysfunction 511
Reduction in incidence of 258-260 Plasma amino acid (PAA) abnormalities 491-
Reinfection 416 492
Relapse 416 Plasma amino acid (PAA) pattern 358-359
Removal of patients from CAPD 238 Plasma glucagon levels 345
Sclerosing 428 Plasmalemma 102
Signs 414-415 Plasma protein 164
Sources of infection 407 Platelet function 500-501
Specimen 412 Platelet thrombi 277
Staff-induced 238-239 Pleural effusions 450
Staphylococcal isolates 406-407 Pleuroperitoneal communication 485-487
Sterile connection device 235, 430 Pneumonia 450
Sterile weld 430 Pneumonitis 346
Surveillance of cultures 406 Poisoning, see also specific poisons 179,
Symptoms 235, 414-415 314
Tenckhoff catheter 403 Peritoneal dialysis for children subject to
Transmural (intestinal) infections 408 552-553
Treatment, see also Treatment of peritoni- Poison Prevention Packaging Act of 1970
tis 235-238 552
Tunnel infections 415-416 Polycystic disease 408
Ultrafiltration and 141 Polycystic kidney disease 447, 460
Ultraviolet sterilization chamber 234-235 Polycystic kidneys, 444, 544
Unresponsive 236, 238 Polymers 174, 305-306
U.V. box 430 Polymyxins 312
Permeability effects of drugs 68-73 Pore area 29
Personnel Portal hypertension 53
Continuous ambulatory peritoneal dialysis Portal vein 53, 95
program 219 Portal venous pressure 277
Training 219-220 Portal venous system 53
Perugia system 466 Post capillary and collecting venules 55, 62,
pH, dialysis solutions 77-81 104
 707

Post-dialysis hypoglycemia 320 Peritonitis 42

Postural hypotension 257, 563, 567 Malnutrition 353, 369-370, 386
Potassium 164, 166, 172 Transfer 62
Nutritional concerns 362-363 Protein-calorie malnutrition 345
Potassium balance, continuous ambulatory Protuberant abdomen 506
peritoneal dialysis 217 Prune belly syndrome 544
Potassium chloride 302 Pseudofractures 459
Potassium concentration 25-26 Pseudomembranous enterocolitis 425
Potassium dichromate 317 Pseudomonas infections 420
Potassium ferrocyanide poisoning 319 Pseudomonas maltophilia 409
Povidone-iodine 314 Psychological problems, continuous ambula-
Predialysis overhydration 451 tory peritoneal dialysis 222
Prednisone 277 Psychosocial aspects of chronic peritoneal
Pregnancy 510 dialysis, see also Quality of life 667-
Premature infants 544 684
Preparation of dialysate, first period (1923- Psychosocial problems 349
1962) 4 Psychotic patients 222
Pre-renal azotemia 534 Pulmonary complications 186, 450-451
Primary arteriovenous fistulae 200 Pulmonary edema 2, 449
Procainamide 317 Puncture technique 8
Procaine 286 Purinergic nervous system 84-85
Progesterone 53
Progressive hyperparathyroidism 257
Progressive neuropathy 348 Q
Prolactin 509-510
Prolonged swell peritoneal dialysis (PDPD) Quality of life 348, 667-684
528 Adaptation to long-term CAPD, factors
Prophylactic antibiotics 306 determining success in 678-679
Propionic acidemia 322, 554 Adaptive versus maladaptive denial 676
Propoxyphene hydrochloride 315 Anxiety 675-676
Prostaglandin E 66 Apathy 676
Prostaglandins 65, 72, 272 Body image 672-673
Modulation of peritoneal transport 283- Children with end-stage renal disease 550
285 Choice of treatment modality for home di-
Synthesis 72-73 alysis 679-680
Protamine 107,215,306 Compliance versus non-compliance 670-
Protein 54 671
Binding 310 Coping with illness 675-676
Catabolism 349 Denial 676
Concentration profile 131, 133-134 Dependency versus independency 677-678
Dietary intake Depression 675-676
Children 388-389 Employment 671-672
Urea nitrogen appearance (UNA) 366- Failure rates 668
369 Family relationship 678
Dietary requirements 360-361, 363 Giving up syndrome 675
Leakage 68-69, 71, 72 Living and dying 669-670
Loss 8, 60, 76, 298, 346, 355-356, 490- Loss and threat of loss 671-672
491,566 Marital relationship 678
Assessment of nutritional status of pa- Mortality rate 668
tient 350 Patient-machine relationships 672-673
Catheters and 457 Self-esteem 672-673
Continuous ambulatory peritoneal di- Self-image 672-673
alysis 217-218 Sexual relations 673-675
Intermittent peritoneal dialysis 198 Social/leisure activities 674-675
Peritoneal inflammation 33 Stresses and concerns 677
708

Success rates 667-669 Children with end-stage renal disease 550

Suicide 669-670, 676 Diabetic patients 574
Quantitative data, first period (1923-1962) Renal vein thrombosis 533
4,6-7 Renin 216
Quinidine 317 Renin-angiotensin-aldosterone system 511-
Quinine intoxication 313 513
Repeated puncture technique 8, 10-11
Reproductive dysfunction 51O-511
R Resistance to solute movement 28-29
Resistances for large and small solutes 36-38
Radioimmunoassay techniques 508 Respiratory alkalosis 453
Radiotherapy 544 Respiratory distress 229
Rapid intermittent peritoneal dialysis 117- Respiratory embarrassment 213
1I8 Respiratory insufficiency 186
Rebellion 350 Retinal ischemia 200
Rebound pain 414 Retinopathy 199-200, 575
Rebound tenderness 235 Reverse osmosis automatic machine 10, 183-
Reciprocating peritoneal dialysis 118-120 184, 300, 307, 320, 448
Recommended daily allowances for children Reye's syndrome 554
390 Ribosomes 10 I, 104
Recommended dietary allowances (RDA) Richter's hernia 488
374 Rifampin 425
Rectal pain 229, 443 Ringer's lactate 125-126
Recurrent urinary tract infections 346 Ringer's solution 3
Reflection coefficient 124-126, 133-135, Ruptured viscus 221
160, 164
Refractory peritonitis 479-480
Regurgitation 232 s
Rehabilitation 568, 667
Relative body weight (RBW) 350-351, 353, Salicylate intoxication 315
375-376 Salicylates 290
Renal colic 460 Salt-loading 512-513
Renal failure 352, 499-514 Scanning electron microscopy (SEM), peri-
Back pain 505-508 toneum 97-10 I, 107-108
Blood pressure 511-513 Scar tissue 348
Complications 499-514 Scleroderma 25
Endocrine disturbances 508 Sclerosing obstructive peritonitis 317
Growth hormone 509-510 Sclerosing peritoneal membrane 109
Hematologic disorders 499-500 Sclerosing peritonitis 14, 109, 428, 480-482
Malnutrition 353 Diabetic patient on CAPO 565
Pericarditis 513-514 Scoliosis 506
Platelet function 500-501 Scrotal edema 230
Prolactin 509-510 Secondary hyperparathyroidism 256,458,
Renal osteodystrophy 501-505 508
Renin-angiotensin aldosterone system 511- Secretin 286, 288
513 Sedatives 315-316
Reproductive functions 510-511 Seizures 211, 451-452, 553
Thyroid function 508-509 Self-destruction, see Suicide
Renal hypoperfusion 534 Self-dialysis, see Continuous ambulatory
Renal osteodystrophy 198-199, 386,458-459 peritoneal dialysis
Continuous cyclic peritoneal dialysis 256- Self-esteem 672-673
257 Self-image 672-673
End-stage renal disease in children 549 Semi-continuous semi-ambulatory peritoneal
Renal failure 501-505 dialysis 14-15
Renal transplantation 2, 188 Sensorium, changes in 235
 709

Sepsis 533 Barriers between vasculature and perito-

Sequential ultrafiltration 195 neal cavity, see also Barriers of sol-
Serotonin 65,69, 286, 409 ute exchange between vasculature and
Serum albumin 352-353, 357-358,497 peritoneal cavity 59-65
Levels 8 Number required for continuous ambula-
Serum calcium levels, suppression of 218 tory peritoneal dialysis 215
Serum carnitine levels 345 Solute polarization 168
Serum cholesterol 257 Solute transfer, factors affecting 51, 130-135
Serum creatinine level 8, 347 Solute transport across peritoneum 117
Serum proteins 352-353 Somatic vasoconstriction 281
Levels 357-358 Somnolence 211
Serum transferrin 352-353, 357, 360, 493 Sorbitol 304-305, 350, 452, 566
Serum urea nitrogen (SUN) 347, 360, 365- Splanchnic blood flow 271-272
367 Spontaneous peritonitis 414
Severe hypernatremia 38 Standard intermittent peritoneal dialysis 117-
118
Severe nephrotic syndrome 321
Staphylococcus aureus 410, 565
Severe systemic vasculitis 25
Infections 476-477
Sexual activity 224 Peritonitis 415,419
Sexual dysfunction 348 Staphylococcus epidermidis 565
Sexual relations 673-675 Peritonitis 418-419
Shock 211, 533 Starling forces 53, 297
Shock patients 186 Starling's law 488
Shoulder pain 443 Staverman's reflection coefficient 160
Sideroblastic 391 Sterile connection device 235, 430
Sieving coefficient 163-166 Sterile peritonitis 314, 417, 422
Single body pool assumption 136 Sterile technique 219
Single-cuffed catheter 223-224 Sterile weld 430
Siphon effect 444-445 Steroid therapy 222
Site of training, continuous ambulatory peri- Stomatae 57, 10 I
toneal dialysis program 219-220 Stop-go blood flow pattern of capillary bed
Skin edema, continuous ambulatory perito- 61
neal dialysis 230-231 Streptococcus viridans 408
Skin exit site infection 468, 477-480 Peritonitis 419-420
Skinfold thicknesses 351 Streptokinase 286, 443
Skin necrotic lesions 568 Stresses and concerns 677
Stylet-catheter 8
Skin rash 424
Subcutaneous cuff erosion 468
Slurred speech 312
Subcutaneous leaks 478
Smooth muscle cells 59
Subcutaneous peritoneal devices 7
Smooth muscle contraction 78 Subperiosteal resorption 256, 503, 505
Social/leisure activities 674-675 Success rates 667-669
Social problems, continuous ambulatory per- Suicide 315, 657, 669-670, 676
itoneal dialysis 222 Sulfamethoxazole 311
Sodium 164, 166, 172 Sulfas 311
Concentrations 301-302 Sulphamidine 311
Intoxication 318 Surface-acting agents 287-288
Sodium and water balance 216, 380 Surgical catheter placement technique in
Sodium chloride poisoning 554 children 539-540
Sodium fusidate 312 Survival rate II, 14, 187-189, 201
Solute clearances 149 Children with end-stage renal disease 550
Dialysis solutions, effects of, see also Di- Continuous ambulatory peritoneal dialysis
alysis solutions 73-85 patients 659-664
Solute delivery, factors affecting 51 Continuous cyclic peritoneal dialysis, 261-
Solute exchange 24-27 263
710

Diabetic patients on CAPD 569-570 Tolazoline 282-283

Swimming, peritoneal catheters 225 Tonofilaments 102
Sympathetic tone, level of 65-66 Toronto Western Hospital catheters 478
Synthetic grafts 200 Total body nitrogen (TBN) 354, 360
Systemic blood coagulation 308 Total body potassium (TBK) 354
Systemic disease 51 Total creatinine clearance II, 13
Systemic lupus erythematosis 25, 222, 277 Total drain volume, continuous ambulatory
Systemic system 53 peritoneal dialysis 212-214
Systemic toxicity 236 Total nitrogen output (TNO) 365-366
Trace elements 377-378
Children 390-391
T Trace mineral depletion 346
Training of patients
Tachyarrhythmias 450 Complications during 228-229
Tachycardia 449-450 Continuous ambulatory peritoneal dialysis
Tamponade 514 226
Tap water 409 Transendothelial cell water movement 3
Taurolin 314 Transendothelial channels 61
Team work, continuous ambulatory perito- Transfer surface area, continuous ambulatory
neal dialysis program 220 peritoneal dialysis 221
Technetium-99m 318 Transfusion of fresh blood 443
Technical developments 14 Transmission electron microscopy (TEM)
Technique in first period (1923-1962) 3 101-104
Telephone contact with patient, continuous Transmural (intestinal) infections 408
ambulatory peritoneal dialysis 227 Transperitoneal metabolite concentration gra-
Temperature, see also Fever 414, 514 dient 120
Tenckhoff's peritoneal dialysis system using Transperitoneal ultrafiltration profile 127-129
heat sterilization 180, 182 Transplantation, see Renal transplantation
Tenckhoff silicone rubber catheter 8, 10-12, Transport kinetics 115-158
116,223,347,403,467,478-479 Clinical measurement of model parameters
Children 538-540 125-126
Complications related to 468-471 Clinical protocols 117-119
Combined diffusive and convective mass
Terbutaline 69
transfer models 122-123
Terminal arterioles 59
Continuous ambulatory peritoneal dialysis,
Tetracycline 312
135
Thallium 316
Children 141-145
Theophy IIine 85, 279, 317
Exchange rate criteria 137-139
Thiamin 381
Theoretical basis 136-137
Thiamphenicol 312
Definition of model of peritoneal dialysis
Thiotepa 318 123-125
Thirst 195, 216, 385 Diffuse mass transfer models 120-121
Thoracic duct 96 Fluid transfer parameters 126-130
Three-pool model 123 Historical perspective 115-116
Thromethamine 316 Kinetic studies of dialysis 145-150
Thyroid function 508-509 Membrane transport limitations 138-141
Thyromegaly 508 Peritoneal dialysis optimizations 121-122
Thyrotoxic crisis 321 Physiological principles 116-117
Thyrotropin-releasing hormone 509 Solute transfer parameters 130-135
Thyroxine 321 Theoretical analysis of patient-peritoneal
Thyroxine-binding globulin 508 dialysis system 119-123
Ticarcillin 310, 425 Transport limitations, see Membrane trans-
Tight junctions 102-103 port limitations
Tissue breakdown 382 Transport rates, see Peritoneal transport
Titanium adapters 466 Treatment of peritonitis, see also Peritonitis
Tobramycin 307, 311, 425 235-238, 423-428
 711

Antibiotics 423-424 Continuous ambulatory peritoneal di-

Catheter removal 427-428 alysis 214-215
Exit site infections 427 Dialysis solution tonicity 214
Heparin 426 Factors affecting 214
Hypersensitivity 424-425 Hydration of peritoneal membrane 214
Length of 424 Increase in 288-289
Peritoneal lavage 425-426 Osmotic gradient between blood and di-
Protocols 426-427 alysate sate 214
Side effects 424-425 Peritoneal membrane thickness 214
Tunnel infections 427 Peritoneal transport 272-274
Tricarboxylic acid cycle metabolites 84 Rapid 216
Triglycerides 222, 372 Status of vasoconstriction 214
Trimethoprim 311 Total infused volume 214
Trocars 225 Ultraviolet sterilization chamber 234-235
Trocath catheter 537-539 Umbilical hernia 222
Tromethamine 313, 315 Uncooperative patients 222
TSH 509 Under-dialysis 152, 191, 193-194
Tuberculosis peritonitis 236, 239,408 Underlying heart disease 450
Tumor cells 58 United States continuous ambulatory perito-
Tunnel infection 225, 229, 238, 261, 415- neal dialysis registry, see USA
416, 565 CAPD registry
Treatment 427 Urea 164-168
Two-pool compartmental model 120-122 Clearance 7-8, 27-32, 68, 85, 122
Two-pool patient model 147 Children versus adult 529
Type IV dyslipoproteinemia 371-374 Continuous ambulatory peritoneal di-
Tyrode's solution 3 alysis 211-212
Continuous cyclic peritoneal dialysis
250-251
u Dialysate flow 35-36
Fluid films limiting 30-32
Ultrafiltration Interstitial resistance limiting 30-32
Capacity, loss of 14 Metabolism in CAPD 365-366
Failure 480-482 Removal, continuous ambulatory perito-
Peritoneal dialysis and 38-40, 74, 124, neal dialysis 212-213
159-177 Urea cycle enzyme deficiencies 322
Concentration profiles 166-169 Urea nitrogen appearance (UNA) 365-369,
Continuous cyclic peritoneal dialysis 385
250 Uremia 1-2, 115,357,372,374,386
Convective solute mass transfer 166-170 Albumin metabolism 352
Correction factor 160 Classic treatments of 146
Critique of present concepts 170-172 Complications 441
Description of system 159-165 Continuous ambulatory peritoneal dialysis
Dissociation 159-160 as method of treatment of 210
Driving force 159-160 Prevention 212, 221
End-stage renal disease in children 547 Uremic patients, peritoneum of 104-110
Excessive 304 Uremic symptoms 139-140
Loss of 108-109, 261 Uremic syndrome 345,347-349,451
Measuring sieving coefficient 165-166 Ureteral obstruction 179
Reflection coefficient 164 Ureter blockage I
Sieving coefficient 163-164 Ureterostomies 544
Speculations 170-172 Uric acid 8, 164
Therapeutic applications of 173-174 USA CAPD registry 597-635
Transport effects of 123 Characteristics of participating centers
Rate 123-125 599-601
Children on CAPD 143-145 Data currency 601-607
712

Establishment of 598-599 Vesicles 41-42, 104

Organization 599 Vesicostomy 544
Overview of continuous ambulatory peri- Vesicular channel 62
toneal dialysis 598 Visceral circulation 55
Patient characteristics 608-611 Visceral lymphatics 96
Patients followed, number and types of Visceral lymph nodes 53
601-607 Visceral mesentery 23-24, 42
Reason for creation of 597 Visceral peritoneum 23-24, 52, 82-83, 95,
Report 599 116
Treatment outcomes Arterial blood supply to 53
Complications 614, 616-621 Visceroneuropathy 384
Last reported status 612-614 Visual status
Life-table analysis 621-633 Diabetic patients on CAPD 567
Reasons for leaving CAPD/CCPD 614- Diabetic patients on intermittent peritoneal
615 dialysis 572
Uterine carcinoma 1, 179 Vitamin B'2 clearance values 13
U.V. box 430 Vitamin D 256
Children 389-390
v Deficiency 345
Vitamin D3218
Vitamin deficiency 376, 386
Vaginal leak of peritonal dialysis fluid 408-
Vitamin losses 498-499
409
Vitamins 228, 376-377, 380, 390
Valproic acid 317
Vancomycin 307, 311, 424-425 Vitreous hemorrhages 200
Vascular access 200 Vitreous surgery 567
Vascular access sites, exhaustion of 194 Volumes of distribution of metabolite on
Vascular calcification 256 either side of peritoneum 120
Vascular disease 25, 277 Vomiting 135, 139,346,349,384-385,451-
Vascular endothelial injury 277 452, 553, 565-566, 568
Vascular permeability 27, 32-33, 51, 65, 69 Continuous ambulatory peritoneal dialysis
Vascular tone 66 211,232,235
Vascular wall, resistances for large and
small solutes 36-38
Vasoactive drugs 32 w
Vasoactive effects
Autacoids 65-66 Wasting 345-347, 349, 353, 355, 384-385,
Drugs 66-68 388, 391
Vasoconstriction 65, 272 Water balance, end-stage renal disease in
Vasoconstrictors 24-25, 68 children 547
Vasodilation 33, 44, 65, 72-73, 272, 275, Water flux. data, clinical application of con-
278-280, 298, 409, 482 tinuous ambulatory peritoneal dialysis
Vasodilator gastrointestinal hormones 285- 215-216
286
Water intake of children 389
Vasodilators 24-25, 27, 30, 62, 66-68, 276
Weakness 377, 379, 390
Peritoneal transport affected by 278-283
Vasopressin 69, 286 Weight control 498
Vein to vein anastomoses 55 Weight losses 385
Venodilation 27 White blood cell count 235-236, 238, 549
Venous capillaries 104 Wound dehiscence 444, 447
Venular dilation 298 Wound healing, impairment of 348
Venular endothelium 62
Venular permeability 32
Venules 104
x
Very low density lipoproteins (VLDL) 372-
373,494-497,565 Xylitol 304, 566
 713

y z
Yeasts causing peritonitis 422 Zinc 377, 390-391
Zonula adhaerans 102
Zonula occIudentes 103