.- ,* > * - - - -- ?

-=fix<. .-f---mf- ,* Physiology and Biochemistry v

P~UW3 T w F V l t % - 1 7 - m 9 H M I ~ E Tyzw 225

The Effect of Different Dosages of Caffeine

on Endurance Performance Time
W.1. Pasman', M.A. van Baakl, A. E.]eukendnlpi, A. de Hoan2
Department of Human Biology. University of Urnburg. Maastricht, The Netherlands
Department of Muscle and Exercise Physiology. Vrije Universiteit, Amsterdam. The Netherlands

W.J. Pasman. M. A. van Baak, A. E.]eukendrup and A. de Haan. Introduction

The Effect of Different Dosages of Caffeine on Endurance Per-
forrnanceTime. lnt. J. Sports Med..Vol. 16. No. 4, pp. 225-230, The importance o f the amount of glycogen stored in muscle
1995. cells for the endurance performance o f athletes is well ac-

Downloaded by: Queen's University. Copyrighted material.

cepted (4,7,9), Increased fat metabolism during exercise ena-
Accepted after revision: October 30. 1994
bles subjects to perform longer, since glycogenolysis is
decreased and glycogen is spared thereby delaying the mo-
The effect of different dosages of caffeine (0- 5-9- 13 mg ment of complete glycogen depletion and fatigue (3.4).
kg body weight-') on endurance performance was examined.
Nine well-trained cyclists participated in thisstudy (V0,max 65.1 I t has been suggested that caffeine is an ergogenic aid that in-
k2.6 ml . kg-' . rnin-'). Caffeine capsules were administered in creases lipolysis and fat oxidation and reduces glycogen break-
random order and double-blind. One hour after capsule inges- down. Because of its supposed ergogenic properties caffeine is
tlon. subjects cycled until exhaustion at 80%Wmax on an elec- on the doping list of the International Olympic Committee.
tromagnetically braked cycle ergometer. Blood samples were However, the data on the metabolic and ergagenic effects of
taken before. during and after the exercise test Before and after caffeine are equivocal. Costill et al. (8). Essig et al. (13) and Ivy
the test a urine sample was obtained. A significant increase in et al. (17) demonstrated that caffeine ingestion resulted in a
endurance performance was found for all caffeine tests com- reduction of the exercise-induced breakdown of glycogen and
pared t o placebo (endurance time 47f 13, 5 8 k 11, 59f 12 and enhanced use of blood borne free fatty acids (FFA) and/or
58 ? 12 rnin for 0.5.9 and 13 m g . kg-' body weight, respectively). muscle triglycerides. Other researchers, however. were unable
No differences were found in endurance performance between to confirm the metabolic effects of caffeine (6.20,24,25).
the three caffeine dosages which indicates that no dose-response
relation o f caffeine and endurance performance was found. An The effects of caffeine on exercise performance are also con-
increased free fatty acid and glycerol concentration was found troversial. Some studies found a stimulating effect o f caffeine
after caffeine consumption compared with placebo. The mean (8.17). others found n o effect (20.24.25). Recent studies by
urinary caffeine concentrations after exercise were 4.8 f [Link] Graham, Spriet and coworkers (13.26) showed a large increase
f 5.2 and 14.9 k 6.9pg . ml-' urine for 5, 9 and 13 mg of caffeine of endurance performance after ingestion of a high dose o f caf-
kg-' body weight. Only the lowest dose o f caffeine resulted in feine (9rng.l<g1 body weight). Although the mean caffeine
urine caffeine concentrations below the doping limit of the In- concentration i n urine after exercise was below the doping
ternational Olympic Committee of 1 2 ~ 9rnl-' . urine in all in- limit of the International Olympic Committee (IOC), some uri-
dividuals. It is concluded that caffeine is an ergogenic aid that nary caffeine concentrations equalled or exceeded the doping
stimulates endurance performance. Adose-response relation be- limit o f 12pgLgmol-' urine.
tween caffeine and endurance time was not found for the dose-
range investigated. The stimulating effect o f caffeine was already I n order to investigate whether the inconsistent findings de-
apparent at the lowest dose of caffeine given (5 mg . kg-'). At this scribed above were due to differences i n the dosage of caffeine
dose urinary caffeine concentration remained below the doping administered, the effect o f three different dosages of caffeine
limit in all subjects. on endurance performance and metabolic parameters was
studied. I n addition, urinary caffeine concentrations were
!MKey words: Caffeine, cycling exercise, doping, dose-response, measured after exercise, i n order to establish the relationship
endurance performance. urinary caffeine concentration between the amount of caffeine ingestion and urinary caffeine
concentration after exercise more clearly. The urinary caffeine
concentrations found i n this study are discussed i n relation to
the doping limit of the IOC.

Int. J. Sports Med.16 (1995) 225-230

8 Georg Thieme Verlag Stuttgart . New York
226 Int. I. Sports Med. 16 (1995) W. 1. Posrnon, M. A van Boak,A. E. leukendrup. A deHoon

Methods feine ingested before exercise. Besides a placebo that contained

no caffeine (CAF) (0mg CAF. kg body weight) the doses of caf-
Subjects feine tested were 5 - 9 - 13 mg CAF. kg body weight-' (CAF. kg
BW-').The capsules were prepared and assigned by a chemist.
Nine healthy well-trained subjects were studied. Physical char- The capsules were administered double-blindly and in random
acteristics are given in Table [Link] experimental procedures order.
and potential risks of the study were explained to each subject
both verbally and in writing. All subjects gave informed con- Subjects arrived at 9.00 h or 13.00 h at the laboratory. Subjects
sent. The experiment was approved by the Ethics Committee who were tested in the morning had fasted, those who cycled
of the University of Wmburg. in the afternoon were asked to fast from 11.00 h on. The sub-
jects produced a urine sample immediately after arrival. A
Physical characteristics of the subjects participating this catheter was placed in an antecubital vein. One hour before
Table 1
-
study (n 9).
in
exercise (t = - 60 min) a blood sample was taken and a capsule
was ingested with water. After capsule ingestion subjects
Physical characteristics mean + SD rested for an hour.
Height (cm) Before starting the exercise a blood sample was obtained (t =
Weight (kg) 0 rnin). Subjects cycled at 80 % Wmax until exhaustion. The test
Fat (%)
was ended when pedalling frequency was lower than 50rpm.
Age (yean)
VOzrnax(I. min-I) During the first 15 min of exercise a continuous sampling of
V0,max (ml . min-I . kg-') expired gas was carried out and blood was sampled at 5. 10

Downloaded by: Queen's University. Copyrighted material.

Wmax (W) and 15 minutes. ARer 15 min blood and gas samples were ob-
80 % Wmax (W) tained every 10 rnin until exhaustion. Fifteen minutes after fin-
HRmax (beats . rnin-I) ishing the cycling exercise (recovery) another blood sample
Wmax. BW-I ( W . kg-') was obtained and urine was collected.

Analyses
Pre-experimental protocol
Expired gas samples were analyzed every 20sec for fractions
Each subject visited the laboratory before the start of the actual of 0, and C02with an on-line system (Sensormedics 2900. Sen-
study and performed an incremental V0,max test on an elec- sormedics Cooperation, USA). The blood samples were mixed
tromagnetically braked cycle ergometer (Lode Excalibur. with EDTA to prevent clotting and were centrifuged immedi-
Groningen. The Netherlands) to determine maximal power ately. Plasma was frozen in liquid N, and stored at - 20'C until
output (Wmax) and maximal oxygen uptake (VOzmax) (21). further analysis. The urine samples were also stored at - 20'C.
The mean V0,max for the group and mean Wmax are shown
in Table 1. The [FFA] was determined with an in vitro enzymatic method
for quantification of the IFFA] in plasma (Wako NEFAC kit;
instructions to thesubjects Wako Chemicals. Germany). The analyses were carried out
with a Cobas-Bio semi-automated centrifugal spectrophoto-
The subjects were instructed to maintain their normal training meter. This apparatus was also used for quantification of the
program during the study and were asked to incorporate the glycerol concentration by standard enzymatic method. Caf-
experimental test into their training program as a hard work feine concentrations in urine were determined using an HPLC
out. The subjects were asked to refrain from training 24 h method (18).
before the test. A diary was kept to record food consumption
and activities during the three days before the first test. The Data anolysis
subjects were asked to eat and train before the following tests
as they had done before the first test to minimize variation in Data are presented as meankstandard deviation (SD). Data of
glycogen concentration in muscle and liver cells. An almost all subjects were available between - 60 and 35 rnin and these
similar procedure has been shown to result in similar pretest were used for statistical analysis. Data of the moment of ex-
glycogen concentrations (23). During these three pre-testing haustion (EXH)and during recovery (REC)were also analyzed.
days subjects were asked to refrain from caffeine consumption
(coffee/tea/chocolate(milk)/cola). as was done in the studies of An analysis of variance (ANOVA)with repeated measurements
Graham et al. and Spriet and colleagues (14.27). to make sore was used for statistical analysis of the response differences be-
that no caffeine was present in the body before the exercise tween four dosages of caffeine. When significant differences
test. All subjects were caffeine users invaryinggradations. Five were found ( p i 0.05) the Scheffe F-test was used to test for
subjects consumed between 100-250 rng caffeine per day and differences between pairs of dosages. Pearson correlation
four subjects consumed above 250 mg caffeine per day. coefficient (r) was calculated for relation between the dosage
of caffeine and post-exercise urinary caffeine concentration.
Experimental protocol
Results
Subjects performed the four endurance tests over four succes-
sive weeks. Day of the weekand time ofday were kept constant Time to exhaustion differed significantly among the caffeine
in each [Link] tests differed only in the amount of caf- and placebo tests. The endurance performance was signifi-
The Effect of Different Dosages of Caffeine on Endurance Performance Time Int. 1. Sports Med. 16 (1995)

Table 2 The individual endurance time data for the four different con-
ditions in minutes and seconds. Mean and standard deviations of the
four tests of the nine subjects are presented, too.
Subjects Omg 5mg 9rng 13mg
1 36.05 42.47 51.50 37.55
2 52.47 85.1 5 65.00 59.30
'p < 0.05 placebo 3 56.55 63.20 73.10 79.12
versus cdfeine
4 45.20 52.10 64.40 58.33
5 35.25 66.20 57.45 70.54
6 66.38 73.25 76.49 69.47
7 40.57 44.50 40.55 46.48
8 57.15 57.17 66.47 66.35
9 28.34 35.05 33.17 36.20
Fig. 1 Endurance performances with different caffeine concentra-
tions. The cycling time (min) is averaged for 9 subjects. The caffeine mean 47 58 59 58
performances were significantly better than the placebo performance. SD 13 11 12 12
No significant differences among the caffeine tests were found.

-
-
-
Orng
2
m 2000 Fig. 2 Free fatty acid
la00 1800 rw (FFA) concentrations
1600 16w 3 9 mg during the 4 exercise

Downloaded by: Queen's University. Copyrighted material.

;
A

. .1
14M) 5 I-t 13 mg tests.
MH: exhaustion
g , , -B
-- m
12W
REC: recovery
d I lox 2
y.
Y Bcm 803
MX) 600

4W 400
p < 0.05 placebo
2W 2 0
versus caffeine
0 0
- 60 o 5 10 15 25 35 EX H REC time (mid
1 10 min I10 min I 1 15 min I

Fig. 3 Glycerol con-

centrations during
the 4 exercise tests.
a00 EXH: exhaustion
A

2 REC: recovery

-$E!
3,

a
L
-8 200

100
p c 0.05 placebo
versus caffeine
0
-61) 0 5 10 15 25 35 EXH
I10 min I10 min I 1 15min I

cantly increased (average 27%) when caffeine was ingested nificantly higher after caffeine ingestion during the whole test
compared to the placebo test (Fig. 1).Among t h e caffeine tests and after the exercise test. The similarity of glycerol concen-
no significant differences in cycling time were found. In- trations after caffeine intake is shown in Fig.3. [The concen-
dividual endurance performances are shown in Table 2. trations for the three caffeine concentrations are almost the
same and therefore only one line is seen.]
Sixty minutes after caffeine ingestion ( t = 0 min) FFA concen-
trations were significantly different from placebo (Fig.2). The caffeine concentrations in urine following exercise showed
Duringthe first 15 rnin of exercise significantly higher FFAcon- a significant linear relationship with t h e amount of caffeine in-
centrations were found in all caffeine tests compared to gested before exercise ( r = 0.65) (p < 0.001). The mean urinary
placebo. No differences in FFA concentrations were found be- caffeine concentration obtained after exercise was 4.8k 1.8,8.9
tween the caffeine dosages. Similar to the FFA concentrations. f 5.2 and 14.9f6.9 C L ml-'~ for 5, 9 and 13m g CAF. kg BW-'.
the glycerol concentrations were significantly higher 60min The individual urinary caffeine concentrations after exercise
after caffeine ingestion (Fig.3) compared to placebo. Unlike the are shown in Fig.4 for the different caffeine dosages.
FFA concentrations, the glycerol concentrations remained sig-
*-'=
Int. I. S~ortsMed. 16 (1995) W. I. Posmon, M. A van Book,A. E. leukendrup. A. de Haan

subjects Fig. 4 Individual caffeine concentrations

in urine obtained 15 minutes after finishing
I
exercise. The doping level of the IOC is shown
2 at 12 pg. mi-' urine. Mean urinary caffeine
concentrations obtained after exercise were
4.8 k 1.8. 8.9 t 5.2 and 14.9k 6.9 pg . rnl-' for
o 4 5 , 9 and 13 rng . kg-' caffeine.
5
- IZ doping limit
5'
E:
9 rn
-5 10 - I

Discussion 27 % in performance time is the same as found by Spriet et al.

Downloaded by: Queen's University. Copyrighted material.

(27) for cyclingexercise with a similar protocol. The differences
In the past decades the effect of caffeine on exercise perform- in absolute endurance time found in this study and the study
ance has been examined in many studies. The studies showed of Spriet et al. (27) (placebo 47 5 13 min versus 76k5 and for
equivocal results with respect to the proposed ergogenic effect caffeine 58 k 12 versus 96 f 9. respectively), might be a con-
of caffeine. Coscill(8). Ivy (17) and coworkers found a stimu- sequence of the methods used to determine the workload
lating effect of caffeine on endurance performance. However. during the test. In our study the subjects cycled at 8 0 %Wmax.
other investigators like Knapik et al. (20). Perkins and Williams Spriec et al. (27) determined Wmax and 30 minutes later sub-
(24) and Powers and coworkers (25) did not find an ergogenic jects cycled for 10-15min at a power output to elicit ~ 8 0 %
effect of caffeine. V02max. Because a levelling-off of the oxygen uptake at max-
imal work load is known, 80%W0,max might result in a lower
work load than 8 0 %Wmax. It is possible that the mean work-
load differed considerably in these studies as a consequence of
Recent findings of Graham and Spriet (14.27) have again the methods used. In this study subjects cycled on average
opened the discussion about the stimulating effect of caffeine 296Wk22. This workload is high for an endurance perform-
during endurance performance. In their studies an endurance ance. The mean work load of Spriet and coworkers (27) was
exercise test was performed at 80-85% V0,max after a high not reported. A lower power output might explain the longer
caffeine dose (9 mg. kg BW-I) or placebo ingestion. For all sub- absolute endurance performance rimes as was shown by Spriet
jects an increase in endurance performance was found, aver- and colleagues (27).
aging 44 and 51 % for running and cycling, respectively. The
subjects were highly trained or recreational athletes, men and The increase in performance time found in this study was
women, caffeine users and [Link] positive effect found lower than found by Graham and Spriet (14) for running
may have been a consequence of the caffeine dose given before (45 %) and cycling exercise (.; 31 %, without subject 5 who
the exercise (14). showed an increase of 156%).The results in this study are in
agreement with the findings of Cadarette and coworkers (5).
The effect of the caffeine dosage on endurance performance They investigated a dose-response relation (0-2.2-4.4-
was examined in the present study. Some earlier studies in- 8.8 mg CAFr kgBW-') during running at 80%V0,max. The re-
vestigating a dose-response relation for caffeine and en- searchers concluded that no dose-response relation exists, al-
durance performance failed to show a relation (20.24). How- though a significant difference between placebo and all caf-
ever, many f e s e a r c h e ~still suggest that a dose-response rela- feine trials was found. Studies carried out with lower amounts
tion does exist (15,19). of caffeine (< 4 mg . kg BW-') showed conflicting results on en-
durance performance. Positive effects of caffeine were found
In this study a wider range of caffeine dosages was used (O- by Wiles (29) using 150 to 200mg caffeine, and Cadarette (5)
5-9-13 mg CAF. kg BW-I) to examine whether the 1 3 m g k g using 200 to 300 mg caffeine, while Titlow (28) found no caf-
BW-' dose was more effective than the already high dose of feine effect on submaximal exercise when 200 mg caffeine was
9 mg CAF . kg BW-I dosage. Some professional cyclists use these ingested.
large amounts of caffeine in competition (pers. comm.).
Mechanism of action of caffeine
We found a stimulating effect of caffeine on cycling time for
all three caffeine dosages (an average increase in performance The increase in endurance performance after ingestion of caf-
time of 27 %).No differences were found between the caffeine feine is thought to be caused by an enhanced fat oxidation (3,
tests: all caffeine dosages tested were equally effective with 8,10,11,14.15). An increase in fat oxidation during exercise will
respect to endurance performance. The relative increase of delay glycogen depletion and fatigue. The glycogen-sparing is
The Effect of Different Dosages of Caffeine on Endurance Performance Time Int. 1. Sports Med. 16 (1995) TP

suggested to be especially effective in the first 15 min of exer- performed after caffeine consumption were easier to carry out
cise, because glycogenolysis is normally v e y high in this pe- than after placebo consumption.
riod (11). A simultaneous down-regulation of the CHO-meta-
bolism (2) could result in glycogen-sparing which was also Side effects of caffeine intake
shown by Spriet et dl. (27). In the present study FFA and glyc-
erol concentrations in blood were increased after caffeine in- The known side effects of caffeine like dizziness. headache,
gestion which is in agreement with previous tindings in other tremor. hunger sensations. insomnia and diuresis (4.17.20)
caffeine studies (8,13.27). A n increased lipolysis is implicated were reported during and after the test. especially for the 9
in the mechanism of action. In this study no dose-response re- and 13mg CAF. kg BW-' dosages, but disappeared within a few
lationship for FFA and glycerol was found. hours.

The lipolysis might be increased via a direct or indirect caffeine In conclusion, an ergogenic effect of caffeine is found for all
effect. The cellular mechanisms of caffeine action were recently caffeine dosages [Link] cyclingtime increased on aver-
reviewed by Dodd et al. (12). The most probable cellular mech- age 27 % for all caffeine dosages. In the dose range investigated
anism of caffeine is inhibition of adenosine receptors, thereby no dose-response relation was found. A maximal beneficial
reducing the inhibition of lipolysis by adenosine. Lindinger et effect of caffeine was already found with 5 mg CAF. kg BW-I.
al. (22) suggested that the ergogenic effect of caffeine was At this dosage all subjects had urinary caffeine concentrations
caused by attenuation in plasma [I<+] during exercise. Main- below the doping limit of the IOC.
tenance of membrane potential by Na-K pump activity may
contribute to the stimulating effect of caffeine. Acknowledgements

Downloaded by: Queen's University. Copyrighted material.

Ingestion of caffeine in competition The authors thank M. Snel for analysis of the urine samples
and W. Jennen for preparation of the caffeine capsules.
The results of the present study suggest that caffeine consurnp-
tion before an endurance exercise positively influences per- References
formance. The maximal stimulating effect of caffeine was al-
ready found with the lowest dosage of caffeine (5mg CAF. kg I Arag6n. J. J., Tornheim K.,Lowenstein J. M.: On a possible role
BW-'). The urinary caffeine concentration was below the of IMP in the regulation of phosphorylase activity in skeletal
doping limit of 12 pg. ml-' urine of the IOC for all subjects for muscle. FEBS Lett 117: K56-K64. 1980.
the lowest dose of caffeine tested. This study showed that Astrup A.. Breum L. Toubro S.. Hein P.. Quaade F.: The effect
higher dosages of caffeine did not result in a better endurance and safety of an ephedrinelcaffeine compound compared 10
performance, but did result in high urinary caffeine concentra- ephedrine, caffeine and placebo in obese subjects on an energy
tions. The urinary caffeine concentrations were too high with restricted diet. A double blind trial. IntJ Obes 16: 269-277.1991.
respect to the doping limit for some individuals with the 9 rng Bjarntorp P.: Importance of fat as a support nutrient for energy:
CAF. kg BW-' and for most of the subjects at the highest dosage metabolism of athletes.] Sports Sci 9: 71 -76, 1991.
4 Bergstriimj., Hermansen L, Hultman E., Saltin 0.: Diet, muscle
given (see Fig.4). There were large interindividual variations glycogen and physical performance. Acta Physiol Scond 71:
in the amount of caffeine in urine after exercise between the 140- 150. 1967.
different caffeine dosages. Because caffeine distributes in the 5 Cadarette B. S., Levine L.. Berube C. L, Posner B. M.. Evans W.
tissues in approximate proportion of the water content, the I.: Effects of varied dosages of caffeine on endurance exercise
physiological response is influenced by hydration status and to fatigue. In: Biochemistry of Exercise. Knuttgen H. G., Vogel
exercise duration (sweat production). This means that body H.,Poorrmans J. (eds) Champaign. IL: Human Kinetics, 1983. pp
composition (far content), hydration state and caffeine dose in- 871 - 876.
gested (10.15.26) are important factors with respect to the ef- Casal, D. C.. Leon A. S.: Failure of caffeine to affect substrate
fect of caffeine. The amount of dehydration among the subjects utilization during prolonged running. Med Sn' Sports Exerr 17:
may have differed during the exercise test. Caffeine further 174- 179.1985.
stimulates diuresis resulting in jncreased urine production. Costill D. L: Carbohydrates for exercise: dietary demands for
which could result in decreased caffeine concentration in optimal performance. lnt J Sports Med 9: 1 - 18, 1988.
urine. Another important aspect is the different endurance 8 Costill D. L.. Dalsly G. P.. Fink W. J.: Effects of caffeine ingestion

times resulting in different moments of urine sampling after on metabolism and exercise performance. Med Sci Sports Exerc
exercise. Within 15 minutes after finishing the test urine was 10: 155-158.1978.
[Link] might affect the amount of caffeine measured in Costill D. L.. Fink W. J., Getchell L H.. Ivy J. L. Witzmann F. A.:
Lipid metabolism in skeletal muscle ofendurance-trained males
urine. The acceptability of using urinary caffeine concentra-
and females.J Appl Physiol 47: 787, 1979.
tions for competition screening has been questioned because l o Clarckson P. M.: Nutritional ergogenic aids: Caffeine. 1ntJSports
of the variation in concentration measured after exercise (10). Nutr 3: 103-111, 1993.
The doping limir of the 10C could be reached by ingesting 8 Dyck D. J., Putman C. T..Heigenhauser C. J. F.. Hultman E.. Spriet
cups of coffee (100 [Link]-'), indicating that athletes have to L. L.: Enhanced FFA availability decreases muscle glycogenolysis
refrain from caffeine consumption before competition, be- during intense aerobic cycling. Physiologist 35: 191, 1992.
cause the ergogenic drug could otherwise be measured in a l2 Dodd 5. L, Herb R. A.. Powers 5. I(.: Caffeine and exercise per-
urine sample. formance. An update. Sports Med 15: 14-23. 1993.
l 3 Essig D., Costill D. [Link] Handel P.J.: Effects of caffeine ingestion
Caffeine consumption not only affects substrate metabolism. on utilization of muscle glycogen and lipid during leg ergometer
jt is also known to effect mood and feelings of perceived exer- cycling. lntJ Sports Med 1: 86-90. 1980.
tion. In this study too, subjects reported that the exercise tests
i b ^ y sInt.]. Sports Med. 16 (1995) W,j. Posmon, M. A. von Book. A. E, jeukendrup. A. de Haan

'4 Graham T. E., Spriet L. L: Performance and metabolic responses Corresponding Author
to a high caffeine dose during prolonged exercise.] Appl Physiol
71 : 2292 - 2298,1991. W. J. Pasman
' 5 Handel P. V.: Caffeine. In 'Ergogenic Aids in Sport'. Williams M. Department of Human Biology
(ed), Champaign. Ill. Human Kinetics, 1983, Chapter 5. pp University of Limburg
128- 163. [Link] 616
' 6 Hickson R. C.. Remie M. J.. Conlee R. K., Winder W. W.. Holloszy NL-6200 MD Maastricht
J. 0.:Effect of free fatty acids on glycogen utilization and en- The Netherlands
durance. ] Appl Physiol43: 829-833, 1977.
l7 Ivy J. L., Costill D. L., Fink W. J., Lower R. W.: Influence of caffeine
and carbohydrate feedings on endurance performance. Med Sci
Sports Exerc 11: 6-11, 1979.
l8 I.S.0 report - 10095. Coffee, determination of caffeine contents
- Methods of using HPLC. 60,first ed 01.15,1992.
l9 James E J.: Caffeine and Health. Academic Press. Norfolk, 1991.
Knapik J. J., Jones B. H., Toner M. M.. Daniels W. L. Evans W. 1.:
lnfluence of caffeine on serum substrate changes during running
in trained and untrained individuals. In: Biochemistry of Exer-
cise. Icnuttgen H. G., Vogel H.,Poortmans (eds), Champaign, Ill.,
Human Kinetics, 1983, pp 514-519.
21 Kuipers H.: Variability of Physiological Responses to Exercise.
Haarlem, de Vrieseborch: 1983, pp 19-22.

Downloaded by: Queen's University. Copyrighted material.

22 Lindinger M. I.,Graham T. E., Spriet L . L: Caffeine attenuates
the exercise-induced increase in plasma [I<+]in humans.)Appl
Physiol74: 1149-1155, 1993.
23 Miller J. M., Coyle E. F.. Sherman W. M., Hagberg J. M.. Costill
D. L. Fink W.]., Terblanche S. E.. Holloszy J. 0.:Effect of glycerol
feeding on endurance and metabolism during prolonged exer-
cise in men. Med Sci Sporrs Everc 15: 237-242. 1983.
24 Perltins R., Williams M. H.: Effect of caffeine upon maximal
muscular endurance of females. Med Sci Spotts Everc 7: 221 -
224. 1975.
25 Powers S. K.. Byrd R. J., Tulley R. Callender T.: Effects of caffeine
ingestion on metabolism and performance during graded exer-
cise. EurJ Appl Physiol50: 301 -307. 1983.
26 Schroder P.. Schumann C.,Maassen N.: Kaffeekonsurn und Kof-
feinkonzentration irn Urin. Dtsch ZSpomed 44: 602 -604,1993.
27 Spriet L. L, MacLean D. A.. Dyck D. J., Hultman E., Cederblad G.,
Graham T. E: Caffeine ingestion and muscle metabolism during
prolonged exercise in humans. A m ) Physiol262: E891 -E898.
1992.
28 Titlow L W.. lshee J. H., Riggs C. E.: Failure of caffeine to affect
metabolism during 60min submaximal exercise. J Sports Sci 9:
15-22, 1991.
29 Wiles J. D.. Bird S. R, Hopl<insJ.. Riley M.: Effect of caffeinated
coffee on running speed. respiratory factors, blood lactate and
perceived exertion during 1500-m treadmill running. BrJ Sports
Med 26: 116-120.1992.