Comparative Analysis of Hospital-

Acquired and Community-Acquired

Pseudomonas aeruginosa Strains in
a Tertiary Care Medical Center
Usamah Hadi, MD*
Mira Chaar, MS†
Rola F. Jaafar, MS†
Ghassan M. Matar, PhD†
Department of Otolaryngology—Head & Neck Surgery, Faculty of Medicine, American
*

University of Beirut Medical Center, Beirut, Lebanon

†Department of Microbiology and Immunology, Faculty of Medicine, American University of

Beirut Medical Center, Beirut, Lebanon

KEY WORDS: molecular methods, Results: These data showed 23 RAPD

nosocomial, Pseudomonas aeruginosa patterns distributed among the 32 P
aeruginosa isolates in patients with otitis
externa and 31 RAPD patterns in
ABSTRACT patients with nosocomially acquired
Objective: Comparison of genotypes of infections. Dendrogram analysis per-
nosocomially acquired and community- formed to compare the 2 P aeruginosa
acquired Pseudomonas aeruginosa iso- populations showed 2 main clusters that
lates in a hospital setting will shed light were genomically distant. Few strains
on the genomic relatedness between the from both populations were closely
2 populations. To that purpose, random related.
amplified polymorphic DNA (RAPD)
analysis was performed to correlate Conclusion: The dendrogram analysis
genotypes of nosocomially identified P confirmed the significant genomic diver-
aeruginosa isolates to community- sity of this organism and demonstrated
acquired isolates from clinical specimens that community-acquired P aeruginosa
encountered in a tertiary care medical infections were genomically different
center in Beirut, Lebanon. from those obtained from nosocomial
infections.
Methods: RAPD was performed on 32
community-acquired and 90 nosocomial- INTRODUCTION
ly acquired isolates of P aeruginosa. Community-acquired Pseudomonas
RAPD patterns were visually observed aeruginosa is the most frequent etiology
and analyzed. Dendrograms were gener- of otitis externa.1 Community-acquired
ated by using GelCompar II software P aeruginosa infections in patients
(Applied Maths NV, St-Martens-Latem, reporting to the emergency room (ER)
Belgium). of a tertiary care medical center in

The Journal of Applied Research • Vol. 7, No. 3, 2007 233

Beirut, Lebanon, were shown to be Thirty-two community-acquired isolates
mostly associated with otitis externa.2 of P aeruginosa were obtained from 22
The patients had either unilateral or patients who presented to the ER with
bilateral otitis externa. A significant the diagnosis of either unilateral or
number experienced severe symptoms bilateral otitis externa.8 Patients had yel-
such as ear discharge, ear canal swelling, lowish to greenish discharge, moderate
pain, periauricular cellulitis, and fever. to severe external auditory canal
P aeruginosa is also one of the most swelling, moderate to severe pain, and
common pathogens that cause nosoco- periauricular cellulitis. None of these
mial infections, particularly in patients patients had intrinsic predisposing fac-
who suffer from immunosuppression.3 P tors.
aeruginosa is a ubiquitous pathogen Consecutive nosocomially identified
prevalent in the hospital environment, P aeruginosa (90) isolates were also
and it can cause severe nosocomial recovered from different patients’ clini-
infections.4 The latter involve a broad cal specimens (1 per patient) submitted
spectrum of infections, including those in for bacteriological investigations at the
the respiratory, gastrointestinal, and uri- Clinical Microbiology Laboratory at the
nary tracts as well as wound infections, American University of Beirut Medical
sepsis, and others.5 Various possible Center.9 Only those patients who
sources of P aeruginosa infection in hos- acquired a nosocomial infection due to
pitals have been identified and include P aeruginosa as determined by clinical
tap water, medical equipment, hospital and laboratory testing and indicated in
personnel, and other patients.4,6 P aerug- their medical records were considered in
inosa accounts for 10% of all hospital- this study. Fever and recovery of P
acquired infections, a site-specific aeruginosa from the site of infection
prevalence that may vary from 1 unit to during the stay at the medical center
another and from study to study.7 constituted the most important criteria
During the last year, the average that defined patient’s infection with this
prevalence of P aeruginosa nosocomial organism. All isolates were identified
infections in the medical center was using standard procedures and tested for
18%. Such a high rate prompted us to susceptibility to a panel of antimicrobial
study the P aeruginosa genotypes circu- agents according to Clinical and
lating in the various units and to corre- Laboratory Standards Institute (CLSI
late them with the genotypes obtained guidelines).
from cases of community-acquired infec-
tions, mainly otitis externa, to determine DNA Extraction and Random
whether there is a genomic relatedness Amplified Polymorphic DNA Analysis
from community-acquired strains and DNA was extracted from P aeruginosa
those endemic in the medical center set- American Type Culture Collection
ting. (ATCC) strain and from all isolates of P
aeruginosa by the GFX Genomic Blood
METHODS DNA Purification Kit (Amersham
Isolation and Identification of PharmaciaBiotech, Uppsala, Sweden)
P aeruginosa according to the manufacturer’s specifi-
Isolates were collected from a tertiary cations. Random amplified polymorphic
care medical center in Beirut, Lebanon, DNA (RAPD) analysis of the clinical
during a 1-year period from patients suf- and environmental isolates using 2 in-
fering from otitis externa (community house oligonucleotide primers, Pa1
acquired) and nosocomial infections. (5’ AGGGGTCTTG 3’) and Pa2

234 Vol. 7, No. 3, 2007 • The Journal of Applied Research

Figure 1. Dendrogram representing the genotypes found in both hospital- and community-
acquired strains.

(5’CTTCTTCAGCTCGACGCGACG3’), for 1 minute, and extension at 72°C for 1

was performed. RAPD was carried out minute for 44 cycles. The cycles were fol-
according to Matar et al8 using the PTC- lowed by a final extension step at 72°C
100 Programmable Thermal Controller for 10 minutes. Amplicons were subject-
(MJ Research, Inc., Watertown, MA). ed to electrophoresis on 2% agarose
Briefly, RAPD analysis was performed gels at 107 volts for 2 hours.
on all isolates in 100-µL reaction mix- The gels were then visualized using
tures, each containing 10 µL of template an ultraviolet UV-transilluminator
DNA, 16 µL of dNTPs (0.2 mM), 10 µL (UVP, Uppland, CA) and photographed
of 10X PCR buffer (100 mM TrisHCl with an Olympus 3.0 camera (Japan)
[pH 8.3], 500 mM KCl, 4mM MgCl2), 1 and the Digidoc-it program (UVP).
µL of primer 1 (0.5 µM), 1 µL of primer Digital TIFF image files were imported
2 (0.3 µg/µL), 2.5 U of Taq DNA poly- into the GelCompar II software
merase, and 61.5 µL of nanopure sterile (Applied Maths NV, St-Martens-Latem,
water. The amplification program includ- Belgium). Gels were normalized using
ed the following steps: denaturation at the DNA ladder markers as reference.
94°C for 3 seconds, annealing at 53°C Band calling of the RAPD patterns was

The Journal of Applied Research • Vol. 7, No. 3, 2007 235

performed by visual inspection of the and opportunistic pathogen in terms of
normalized gel images. A dendrogram its genetics, metabolic potential, and
was generated with GelCompar II statis- mechanisms of virulence. The pathogen-
tical software using the Dice coefficient esis of otitis externa is multifactorial.
and unweighted paired-group methods Both host-related factors as well as
for arithmetic averages algorithm. pathogen factors play major roles in col-
onization and infection.10 The P aerugi-
RESULTS nosa that caused the infection may have
The genotypic analysis of the 32 otitis been part of the patient’s endogenous
externa isolates revealed 23 RAPD pat- flora, originated from other patients, or
terns. Bilateral otitis externa was associ- spread to humans from various environ-
ated with 20 of the 32 isolates. Four of mental sources.11-13 Frequent recombina-
these 20 isolates, obtained from 2 tion of chromosomal genes between
patients, had RAPD 1 pattern, while 8 different isolates of P aeruginosa may
isolates obtained from both ears of 4 have lead to the genetic diversity of the
patients had 4 RAPD patterns. The organism.14 The mechanisms of genetic
remaining 8 isolates from 4 patients with exchange, including transformation,
bilateral otitis each showed a different transduction, and conjugation, affect P
RAPD pattern. Twelve of 32 isolates aeruginosa to adapt to changing condi-
associated with unilateral otitis exhibit- tions by acquiring new genetic informa-
ed all 12 RAPD patterns.8 RAPD analy- tion.15 This versatility is reflected in the
sis of the nosocomial infections have results of this study, whereby 23 differ-
shown 31 genotypes present among the ent RAPD patterns were generated
clinical isolates.9 Thirty-eight of ninety from 32 isolates of patients having otitis
(42%) of the clinical isolates showed externa and 31 genotypes from 90 iso-
genotype 1 to be distributed among the lates obtained from patients with noso-
medical center units. Each of genotypes comial infections with predominance of
2-30 represented from 1% to 8% of the genotype 1.
strains. Dendrogram analysis of nosoco- The predominance of a genotype in
mially and community-acquired strains the medical center indicates that cross
was performed and showed 2 main clus- contamination among patients led to the
ters; the first cluster represented the spread of this genotype among the vari-
genotypes of the nosocomially acquired ous units, possibly through transient
P aeruginosa, while the second showed hand carriage by health care personnel
the community-acquired P aeruginosa due to contact with contaminated sur-
(Figure 1). Most of the strains within the faces or by patient contact with contami-
2 clusters illustrated closer genomic nated surfaces or medical equipment.6
relatedness among each other. Most of These findings suggest that cross colo-
the genotypes from nosocomial infec- nization may be an important means of
tions were highly interrelated whereas P aeruginosa spread and infection espe-
community-acquired genotypes were cially after identification of a potentially
less interrelated. Although similarities virulent clone (genotype 1) of this
between the 2 clusters were seen, both organism that had been propagated in
populations were not related to each various units over a period of 1 year.
other because the genotypes did not The genotypes found in both popu-
overlap. lations were different from one another.
The dendrogram analysis confirmed the
DISCUSSION great genomic diversity of this organism
P aeruginosa is known to be a versatile and demonstrated that P aeruginosa

236 Vol. 7, No. 3, 2007 • The Journal of Applied Research

strains obtained from community- ty patterns from the Sentry Antimicrobial
Surveillance Program. Diagn Microbiol Infect
acquired infections were genomically Dis. 2000;37:115-125.
different from those obtained from
8. Matar GM, Harakeh HS, Ramlawi F, et al.
nosocomial infections. Nevertheless, a Comparative analysis between Pseudomonas
small number of isolates from the 2 aeruginosa genotypes and severity of symp-
sources were closely related as seen in toms in patients with unilateral or bilateral
otitis externa. Curr Microbiol. 2001;42:190-
the middle portion of the dendrogram. 193.
9. Matar GM, Chaar MH, Araj GF, et al.
REFERENCES Detection of a highly prevalent and potential-
1. Brook I, Fraizer E, Thompson D. Aerobic and ly virulent strain of Pseudomonas aeruginosa
anaerobic microbiology of external otitis. Clin from nosocomial infections in a medical cen-
Infect Dis. 1992;15:955-958. ter. BMC Microbiol. 2005;5:29-34.

2. Hadi U, Serhal S, Ghosseini S, et al. Bacterial 10. Hawke M, Wong J, Krajden S. Clinical and
etiology of otitis externa in group of microbiological features of otitis externa. J
Lebanese Individuals. Br Med J. 2000;7:6-7. Otolaryngol. 1984;13:289-295.

3. Zenone T, Souillet G. X linked 11. Cassisi N, Cohn A, Davidson T, Witten BR.

Agammaglobulinemia presenting as Diffuse otitis externa: clinical and microbio-
Pseudomonas aeruginosa septicemia. Scand J logical findings in the course of multicenter
Infect Dis. 1996;28:417-418. study on new otic solution. Ann Otol Rhinol
Laryngol. 1977;86(suppl 39):1–16.
4. Morrison AJ, Wenzel RP. Epidemiology of
infections due to Pseudomonas aeruginosa. 12. Driscoll PV, Drezmer DA, Hicks TA, et al.
Rev Infect Dis. 1984;6:S627-S642. Characteristics of cerumen in diabetic
patients: a key to understanding malignant
5. Pollack M. Pseudomonas aeruginosa. In: external otitis? Otolaryngol Head Neck Surg.
Mandell GL, Bennett JE, Dolin R, eds. 1993;109:676-679.
Principles and Practice of Infectious Diseases.
New York, NY: Churchill Livingstone; 13. Wright DN, Dineen M. A model for the study
1995:1980-2003. of infectious otitis externa. Arch Otolaryngol.
1972;95:243-247.
6. Pittet D, Dharan S, Touveneau S, et al.
Bacterial contamination of the hands of hos- 14. Pasloske B, Joffe M, Sun Q, et al. Serial iso-
pital staff during routine patient care. Arch lates of Pseudomonas aeruginosa from a cys-
Intern Med. 1999;159:821-826. tic fibrosis patient have identical pilin
sequences. Infect Immun. 1988;56:665-672.
7. Jones RN, Croco MA, Kugler KC, et al.
Respiratory tract pathogens isolated from 15. Vasil ML. Pseudomonas aeruginosa: biology,
patients hospitalized with suspected pneumo- mechanisms of virulence, epidemiology. J
nia: frequency of occurrence and susceptibili- Pediatr. 1986;180:800-805.

The Journal of Applied Research • Vol. 7, No. 3, 2007 237