Content uploaded by Ziad Daoud
Author content
All content in this area was uploaded by Ziad Daoud
Content may be subject to copyright.
Extended spectrum bb-lactamase producing Enterobacteriaceae in
Lebanese ICU patients: Epidemiology and patterns of resistance
Ziad Daoud,1,2,* Carole Moubareck,3Noha Hakime,1and Florence Doucet-Populaire3
1Department of Laboratory Medicine, Saint George University Hospital, Beirut, Lebanon
2Faculty of Health Sciences, University of Balamand, Beirut, Lebanon
3Laboratoire de Microbiologie, Faculté de Sciences, Pharmaceutiques et Biologiques,
Université Paris, 5-René Descartes, Paris, France
(Received November 29, 2005; Accepted March 24, 2006)
Our study aims to investigate and describe the epidemiology of the intestinal carriage of ESBL-
PS in intensive care units of five Lebanese hospitals and to analyze the potential risk factors for
the acquisition of these strains. At the same time, we intend to determine the patterns of suscep-
tibility of these strains, exploring therefore the availability of alternative treatment. One thou-
sand, four hundred forty-two fecal samples were collected between January 1, 2003 and March
31, 2003 from 378 patients admitted to the ICUs of five Lebanese tertiary care general hospitals
located in different areas of Lebanon. ESBL production was detected by the double disk synergy
test and antibiotic susceptibility of ESBL-producing strains as well as minimum inhibitory con-
centrations were determined. A paired case-control study was undertaken to identify risk factors
for carriage of ESBL-PS. One hundred eighteen strains isolated from 72 subjects were identified
as ESBL producers, including 95 (80.5%) E. coli, 16 (13.6%) Klebsiella pneumoniae, and 7 (5.6%)
Enterobacter cloacae. A higher rate of multiple ESBL-PS carriage was described among these
acquisition cases (21 double carriages and 3 triple carriages of ESBL-PS compared to only 1
double carriage of ESBL-PS at admission). In general, similar trends of susceptibility were ob-
served in the different hospitals. As expected, the lowest MIC was observed with imipenem for
all E. coli, Klebsiella, and Enterobacter isolates. Ciprofloxacin, followed by trimethoprim-sul-
famethoxazole seem to be associated with the lowest susceptibility. In vitro susceptibility to ce-
foxitin for all isolates was 74.6%; more resistance was associated to ceftazidime (90.7%) than to
cefotaxime (69.7%). Our data agree with other national and international reports showing the in-
crease in ESBL-PS carriage in ICU patients. They demonstrate the endemic character of this car-
riage in Lebanese hospitals and the important risk factors including immunosuppression and ev-
idence of ESBL infection. The highly resistant profile of ESBL-PS to antimicrobial agents avail-
able for treatment reflects the severity of this issue. The significance of this study resides in the
direct correlation between our results and the nationwide increase in multi-drug resistant bacte-
ria and the continuous change in bacterial resistance epidemiology. Our data may have an im-
portant impact on infection control policies in hospitals and on treatment of infectious diseases.
J. Gen. Appl. Microbiol., 52, 169–178 (2006)
*Address reprint requests to: Dr. Ziad Daoud, Head of Clinical Microbiology Laboratory, Saint George Hospital, PO Box: 166378,
Beirut 11002807, Lebanon.
Tel: 00961.1.562108- or 9 ext (22) Fax: 00961.562110
E-mail: ziad.daoud@balamand.edu.lb
Abbreviations: AH, Arz Hospital; ESBL, extended spectrum b-lactamases; ESBL-PS, extended spectrum b-lactamases producing
strain; ICU, intensive care unit; MIC, minimum inhibitory concentration; NN, Nini Hospital; RH, Rizk Hospital; SGH, Saint George
Hospital; SJH, Saint Joseph Hospital.
Full Paper
Introduction
Extended-spectrum b-lactamases (ESBLs) are en-
zymes that confer resistance to penicillins, narrow-
and extended-spectrum cephalosporins, monobac-
tams and related oxyimino-b-lactams (ceftazidime, ce-
fotaxime) and have been shown to be inactive against
cephamycins and carbapenems (Bradford, 2001;
Bradford et al., 1994; Bush et al., 1995; Pitout and
Sanders, 1997). Clavulanate, tazobactam and other b-
lacamase inhibitors are generally effective against
ESBLs although some inhibitor resistant variants have
been described (Bradford, 2001; Bradford et al., 1994;
Bush et al., 1995). Bacteria producing ESBLs are most
commonly Klebsiella pneumoniae and Escherichia
coli, as well as other members of the genus in Enter-
obacteriaceae (Jarlier et al., 1998).
In Lebanon, a study done at a general hospital in
Beirut in 1998 (Araj, 1999) revealed that the preva-
lence of ESBL-producing E. coli and K. pneumoniae
isolates was 3.3% and 6.4%, respectively. At Saint
George University Hospital, a 300-bed hospital partici-
pating in our study, 2% of 4,299 isolates of E. coli and
20% of 1,248 isolates of K. pneumoniae were found to
produce ESBLs between 1997 and 2001 (Daoud and
Hakime, 2003). In fact, the percentage of these had
been experiencing a constant rise over the 5 years
from 7.5% to 24% for Klebsiella pneumoniae and
reached 4% for Escherichia coli (Daoud and Hakime,
2003).
Numerous ESBL enzymes have been described,
with greater than 100 in the TEM class, over 40 in the
SHV family, over 30 in the CTX-M class, and 15 in the
OXA family (Bonnet, 2004; Bradford, 2001). These en-
zymes are coded by plasmids which may also carry re-
sistance to other antibiotics, including aminoglyco-
sides, tetracyclines, chloramphenicol, trimethoprim,
and sulfonamides (Bush et al., 1995). Molecular char-
acterization of strains isolated in a nationwide project
from which this study is derived showed that 83% pro-
duced the CTX-M-15 variant of ESBL enzymes
(Daoud et al., 2005). SHV-5a was detected in the re-
maining isolates. Although most ESBL variants are of
the TEM type, these were not detected in this study
(Daoud et al., 2005). Pulse field gel electrophoresis
(PFGE) phenotypes of E. coli producing ESBL en-
zymes were highly diverse. Patient transfer between
hospitals is commonly practiced in Lebanon and might
have been the impetus for the spread of resistance-
carrying plasmids, thereby producing this diversity in
phenotypes.
In 2002, as reported by the Center for Disease Con-
trol and Prevention, 6.3% of Escherichia coli, 14% of
Klebsiella pneumoniae, and 32.2% of Enterobacter
spp. were resistant to third generation cephalosporins
in intensive care units in the United States (Hacek et
al., 2001; Kaye et al., 2004; National Nosocomial In-
fections Surveillance System Report, 2003). Across
Europe, on the other hand, ESBL production varies
hugely and the percentage of ceftazidime resistant K.
pneumoniae was 20% in non-ICU units and doubled in
ICUs before 2001 (Bradford, 2001). In a study done at
Saint George Hospital in Lebanon, around 1/3 of the
89 ESBL-producing isolates of E. coli and 244 of those
of K. pneumoniae were from patients admitted to the
intensive care unit (Daoud and Hakime, 2003). Fur-
thermore, the number of ESBL-positive E. coli and K.
pneumoniae was undergoing a constant increase until
2001 in ICUs as well as elsewhere (Daoud and
Hakime, 2003).
Information about the epidemiology of ESBL-produc-
ing bacteria in Lebanon is still limited, particularly their
occurrence in ICUs. Health care facilities are increas-
ingly reporting outbreaks of these organisms challeng-
ing their infection control units. Our study aims to ex-
amine the fecal carriage of ESBL-producing bacteria in
patients admitted to intensive care units of five hospi-
tals in Lebanon over a 3-month period.
Materials and Methods
Study design. This study is a prospective surveil-
lance study in which fecal samples collected from hos-
pitalized patients and health care personnel were
screened for the presence of possible ESBL-producing
strains (ESBL-PS). One thousand, four hundred forty-
two fecal samples were collected between January 1,
2003 and March 31, 2003 from 378 patients admitted
to the intensive care units (including the neonatal ICU)
of five Lebanese tertiary care general hospitals located
170 DAOUD et al. Vol. 52
Key Words
—
bacterial resistance; carriage; Enterobacteriaceae; extended spectrum blactamase
in different areas of Lebanon: Rizk Hospital—Beirut
(RH), Saint George Hospital—Beirut (SGH), Arz Hos-
pital—Al Metn (AH), Nini Hospital—Tripoli North
Lebanon (NN), and Saint Joseph Hospital—Beirut
(SJH). Patients were checked for carriage of ESBL-
producing strains (ESBL-PS) upon admission, 72 h
later and then at weekly intervals over a 3-month pe-
riod.
Bacterial strains. All samples were suspended at
participating sites in normal saline in order to form a
10 mg/ml suspension. From these, 100 ml were
streaked on Trypticase soy–5% sheep blood agar con-
taining 6 mg/ml of vancomycin, 2 mg/ml of amphotericin
B, 2 mg/ml of ceftazidime and 1 mg/ml of clindamycin
(VACC medium) (Hacek et al., 2001). After 24 and
48 h of incubation at 35°C, colonies appearing on the
plates were sent to the Saint George Hospital collect-
ing laboratory for strain identification by the API 20E
system (bioMérieux) and for resistance phenotype
analysis and minimum inhibitory concentration (MIC)
determination.
Antibiotic susceptibility testing and ESBL detection.
ESBL production was detected by the double disk syn-
ergy test according to the method described by Jarlier
et al. (1988). Ceftazidime (30 mg), cefepime (30 mg),
aztreonem (30 mg) and cefotaxime (30 mg) disks
(Oxoid) were placed 25 mm (center to center) from the
amoxicillin/clavulanic acid (20/10 mg) disk on Mueller-
Hinton agar plate inoculated with the test organism.
After overnight incubation, enhancement of the inhibi-
tion zone around one of these disks toward the clavu-
lanate-containing disk indicated the presence of
ESBLs. Antibiotic susceptibility of ESBL-PS was deter-
mined by the disk diffusion method according to the
recommendations of the National Committee of Clini-
cal Laboratory Standards (NCCLS) (2000). The follow-
ing disks were tested: amoxicillin/clavulanic acid
(20 mg/10 mg), cefoxitin (30 mg), cefotaxime (30 mg),
ceftazidime (30 mg), gentamicin (10 mg), trimethoprim/
sulfamethoxazole (1.25 mg/23.75 mg), amikacin (10 mg),
and imipenem (10 mg). The minimum inhibitory con-
centrations (MIC) of imipenem, cefoxitin, ceftriaxone,
cefepime, gentamicin, amikacin, amoxicillin/clavulanic
acid, piperacillin/tazobactam, aztreonam, ciprofloxacin,
and ceftazidime, and cefotaxime (with and without
clavaulanate) were determined by using the E-test
(AB-Biodisk). For each subject, if several ESBL-pro-
ducing isolates were detected and all were from the
same species and presented similar antibiograms,
only one was selected for further study and the others
were considered to be repetitive isolation of the same
strain.
Epidemiological survey. The time for ESBL pro-
ducing Enterobacteriaceae acquisition was defined for
each of the 41 patients from whom no ESBL-PS were
isolated at admission by taking into account the date
on which the patient entered the ICU and the date on
which the strain was isolated from the patient’s fecal
material. Only new acquisitions were considered and
the isolation of an ESBL-PS from the same patient
was considered a new acquisition if it was preceded by
at least one negative sample.
A paired case-control study was undertaken to iden-
tify risk factors for carriage of ESBL-PS using the
SPSS 11.0 software. It covered the whole period of the
study and involved 59 patients from which 68 ESBL-
PS were newly isolated (after admission) including pa-
tients who were not colonized at admission and pa-
tients who were recolonized after eradication of the
ESBL-PS (4118 patients respectively). This does not
take into consideration the number of ESBL-PS newly
isolated from the same patient at the same time, but
considers the acquisition as a fact regardless of the
number of strains. Each of these 59 patients was
matched with a control hospitalized in the same unit
and from whom ESBL producing strains were never
isolated. The matching was based on gender and the
presence of the control in the unit at least 1 week be-
fore and 1 week after the date on which the strain was
isolated. Parameters covered by the survey question-
naire are indicated in Table 1. After data collection
from the hospitalization reports, univariate analysis of
risk factors was performed using student’s t-test for
comparision of means and c2test (n5) with Mantel-
Haenszel weighting for other variables. p-values of
0.05 were considered significant.
Results
A total of 1,442 fecal samples were collected during
the whole study period from 378 ICU patients of the
participating 5 hospitals. One hundred eighteen strains
isolated from 72 subjects were identified as ESBL pro-
ducers including 95 (80.5%) E. coli, 16 (13.6%) Kleb-
siella pneumoniae and 7 (5.6%) Enterobacter cloacae.
These were detected in all participating hospitals as
follows: SGH (52 isolates), SJH (22 isolates), AH (20
isolates), RH (17 isolates) and NN (7 isolates). The
2006 Extended spectrum b-lactamase producing Enterobacteriaceae in Lebanese ICU patients 171
characteristics of the enrolled admissions are repre-
sented in Table 1.
Thirty one patients, from whom 32 ESBL-PS were
isolated, were considered as imported cases since
carriage was positive at the time of admission. Among
these, only one patient was noted for the carriage of
more than one ESBL producing isolate. On the other
hand, 41 new patients, for whom a conversion from
negative carriage at ICU admission to positive carriage
after admission was noted, in addition to 18 patients
who were previously colonized (at admission) then re-
colonized after at least 48 h of ESBL-PS eradication,
were considered as acquisition cases (a total of 59 pa-
tients and 86 isolates). A higher rate of multiple ESBL-
PS carriage was described among these acquisition
cases (21 double carriages and 3 triple carriages of
ESBL-PS compared to only 1 double carriage of ESBL-
PS
at admission).
Figure 1 shows the distribution of carriage of ESBL-
PS over the different time periods of ICU stay. The
data here show the actual state of carriage during a
specific time interval. It is obvious that there is a clear
decrease in carriage after admission to ICU in SGH,
SJH, and NN hospitals, while in AH and RH hospitals
the highest percentage of carriage is observed after
72 h. In all hospitals, a sharp decrease in carriage is
observed after 72 h and up to 2 weeks of ICU stay. In
SGH and SJH an increase is re-observed after 15
days.
Antimicrobial susceptibility testing by the disk diffu-
sion method indicated strong synergy between amoxi-
cillin-clavulanate and either cefotaxime or ceftazidime
observed in 103 (87.3%) cases. All ESBL-PS were re-
sistant to extended spectrum cephalosporins (MIC90s
for E. coli, Klebsiella, and Enterobacter were all
greater than 256 mg/ml for ceftazidime. The MIC90 of E.
coli to cefotaxime was 96 mg/ml and that of Klebsiella
and Enterobacter was greater than 256 mg/ml. In the
disk diffusion method, all the tested isolates showed
resistance to at least one of either ceftazidime or cefo-
taxime (Table 2). Strains that showed in vitro resis-
tance to both antimicrobial agents did not exceed 10%
(data not shown). All isolates were susceptible to
imipenem. In the presence of 2 mg/ml of clavulanic acid
the susceptibility of these strains to ceftazidime and/or
cefotaxime was restored.
Table 3 shows the patterns of susceptibility of the
ESBL-PS to different antimicrobial agents. In general,
similar trends of susceptibility were observed in the dif-
ferent hospitals. Exceptions were noted for susceptibil-
ity to piperacillin/tazobactam in SGH and AH hospitals
(showing lower percentages of susceptibility in com-
parison to other hospitals) and for gentamicin,
ciprofloxacine, and trimethoprim-sulfamethoxazole at
NN hospital, where the percentages of susceptibility to
these antibiotics were significantly increased in com-
parison to other hospitals.
The best susceptibility was associated with amikacin
with values ranging between 71.4 and 92%.
Ciprofloxacin, followed by trimethoprim-sulfamethox-
172 DAOUD et al. Vol. 52
Table 1. General characteristics of patients enrolled.
Variable No. (%)
Number of patients 72
Male/female ratio 2.6/1 (52 males and 20 females)
Age, yrs, meanSD 5113.3
Transferred from:
Medicine 231 (29.2)
Surgery 23 (31.9)
Orthopedics and traumatology 7 (9.7)
Obstetrics and gynecology 10 (13.9)
Direct ICU admission from emergency department 11 (15.3)
Recent antibiotic use before ICU entry 81 (70.8)
Nursing home residence 7 (9.7) coming only from one hospital (SGH), this accounts
for 24.1% of the population of the concerned hospital
Transfer in from another hospital 11 (15.2)
Length of stay in ICU, day, meanSD 17.312.3
Fatal outcome in ICU 5 (6.9)
azole seem to be associated with the lowest suscepti-
bility (0 to 10% for ciprofloxacine and 9.5 to 50% for
trimethoprim-sulfamethoxazole) excluding NN hospital
in both cases.
In vitro susceptibility to cefoxitin for all isolates was
74.6% (88 strains). More in vitro resistance was asso-
ciated to ceftazidime (90.7%) than to cefotaxime
(69.7%).
2006 Extended spectrum b-lactamase producing Enterobacteriaceae in Lebanese ICU patients 173
Fig. 1. Distribution of carriage of ESBL producing bacteria in the different hospitals over the period of the study.
SGH, Saint George Hospital; SJH, Saint Joseph Hospital; AH, Arz Hospital; RH, Rizk Hospital; NN, Nini Hospital.
Fig. 2. Percentage of acquisition of ESBL producing strains in patients who were not colonized when admitted to
the ICU or who were recolonized after eradication of ESBL-PS.
Table 2. Non-interpreted antimicrobial activities of cefotaxime and ceftazidime on
118 ESBL-producing bacteria isolated from participating hospitals.
Susceptible Intermediate Resistant
Cefotaxime Ceftazidime Cefotaxime Ceftazidime Cefotaxime Ceftazidime
E. coli (Nb95) 33 (34.7%) 10 (10.5%) 16 (16.84%) 5 (5.3%) 46 (48.4%) 80 (84.2%)
Klebsiella (Nb16) 2 (12.5%) 1 (6.3%) 2 (12.5%) 1 (6.3%) 12 (75.0%) 14 (87.5%)
Enterobacter (Nb7) 1 (14.3%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 6 (85.7%) 7 (100%)
Table 4 shows the different MICs of the ESBL-PS to-
wards the tested antibiotics. As expected, the lowest
MIC, reflecting the best antibacterial effect, was ob-
served with imipenem for all E. coli, Klebsiella, and En-
terobacter isolates. Similar patterns of susceptibility
and MIC values with normal variations were observed
among these different genera.
Discussion
The consequences of infection with ESBL producing
Enterobacteriaceae can be severe and lead to serious
complications. The situation is made worse when such
infections occur in critically ill patients in the ICU,
where medical manipulations are frequent and antibi-
otic use is massive (Itokazu et al., 1996).
It has been demonstrated that infection due to highly
resistant microorganisms is frequently associated with
previous body site colonization (Nseir et al., 2005).
In this context, this multiple center study shows that
the stay in the ICU affects significantly the occurrence
of ESBL-PS carriage in patients. The results of surveil-
lance cultures at ICU entry, and discharge as well as
during ICU stay show that the overall number of pa-
tient carriers of ESBL-PS was increased by approxi-
mately twofold and that the number of ESBL-PS that
were acquired during the course of ICU admission was
approximately increased by threefold. The univariate
analysis of risk factors strongly correlates the cases of
ESBL-PS carriage with the evidence for infection with
this class of bacteria, immunosuppressive factors, and
both general and abnormal surgery. This agrees with
many previous studies that highlighted the importance
of colonization pressure on the risk of acquisition of re-
sistant bacteria (D’Agata et al., 1999; Merrer et al.,
2000; Nseir et al., 2005; Toltzis et al., 1997). A thor-
ough analysis of the changes occurring in ESBL-PS
carriage during ICU stay (Fig. 1) shows that an aver-
174 DAOUD et al. Vol. 52
Table 3. Patterns of susceptibility for 118 ESBL producing isolates to 8 antimicrobial agents.
Cefoxitin Pipera-Tazo Gentamicin Ciprofloxacin TMX Amikacin Cefotaxime Ceftazidime
% (Nb) % (Nb) % (Nb) % (Nb) % (Nb) % (Nb) % (Nb) % (Nb)
SGH (n50)
S66 (33) 70 (35) 64 (32) 10 (5) 22 (11) 92 (46) 32 (16) 6 (3)
I4 (2) 10 (5) 0 (0) 4 (2) 0 (0) 6 (3) 4 (2) 6 (3)
R30 (15) 20 (10) 36 (18) 86 (43) 78 (39) 2 (1) 64 (32) 88 (44)
SJH (n21)
S76.2 (16) 38.1 (8) 42.9 (9) 9.5 (2) 9.5 (2) 71.4 (15) 23.8 (5) 0 (0)
I 9.5 (3) 52.4 (11) 9.5 (2) 0 (0) 4.8 (1) 14.3 (3) 33.3 (7) 4.8 (1)
R14.3 (2) 9.5 (2) 47.6 (10) 90.5 (19) 85.7 (18) 14.3 (3) 42.9 (9) 95.2 (20)
AH (n19)
S84.2 (16) 42.1 (8) 36.8 (7) 0 (0) 26.3 (5) 84.2 (16) 42.1 (8) 10.56 (2)
I0 (0) 15.8 (3) 0 (0) 0 (0) 0 (0) 0 (0) 10.5 (2) 0 (0)
R15.8 (3) 42.1 (8) 63.2 (12) 100 (19) 73.7 (14) 15.8 (3) 47.4 (9) 89.5 (17)
RH (n16)
S81.3 (13) 62.5 (10) 56.3 (9) 6.3 (1) 50 (8) 81.3 (13) 18.7 (3) 25 (4)
I12.5 (2) 31.3 (5) 0 (0) 25 (4) 6.3 (1) 12.5 (2) 25 (4) 6.3 (1)
R6.3 (1) 6.3 (1) 43.8 (7) 68.8 (11) 43.8 (7) 6.3 (1) 56.3 (9) 68.8 (11)
NN (n7)
S71.4 (5) 71.4 (5) 85.7 (6) 85.7 (6) 71.4 (5) 71.4 (5) 42.9 (3) 0 (0)
I0 (0) 14.3 (1) 0 (0) 0 (0) 0 (0) 28.6 (2) 0 (0) 0 (0)
R28.6 (2) 14.3 (1) 14.3 (1) 14.3 (1) 28.6 (2) 0 (0) 57.1 (4) 100 (7)
All isolates were susceptible to imipenem.
SGH, Saint George Hospital; SJH, Saint Joseph Hospital; AH, Arz Hospital; RH, Rizk Hospital; NN, Nini Hospital.
age of 30 to 40% of the isolated strains are observed
on the day of admission; this is observed in all the hos-
pitals with the only exception of AH hospital.
After admission, the following 72 h seem to be asso-
ciated with the highest percentage of acquisition of
ESBL-PS during ICU stay. This is true with all involved
hospitals including SGH and SJH, where 36% and
25% of the isolates occurred after 15 days of ICU stay
respectively. This is explained by the fact that the aver-
age length of stay in ICU is 13 to 15 days; therefore,
the re-increase in carriage in these two hospitals is
taking place after discharge from the ICU. The fact that
the occurrence of ESBL-PS decreases after 72 h and
up to 15 days may be explained by the heavy and sys-
tematic use of carbapenem antibiotics in the ICUs of
these hospitals, and by the emergence of some strains
of imipenem-resistant Pseudomonas aeruginosa and
Stenotrophomonas maltophilia in the fecal material of
the same patients (data not shown).
Another fact that requires special attention is that
the acquisition of new ESBL-PS when compared to im-
ported cases seems to be more diverse and is compli-
cated by the multiple-carriage of resistant strains (21
double carriages and 3 triple carriages of ESBL-PS
versus only one double carriage in the 31 imported
cases).
Our data show that although the acquisition of
ESBL-PS follows a decreasing pattern with the length
of ICU stay, it becomes more complex and the total
number of acquisitions reaches almost double that of
importations. Similar results have been described with
other studies addressing ceftazidime-resistant gram
negative bacilli intestinal carriage (Ho, 2003); however,
the dynamic evolution of carriage could not be de-
scribed in that study because the screening was done
only at the admission and discharge. In all cases, the
use of carbapenems is playing an important role in our
study through the selective pressure exerted on the
patient’s normal intestinal flora.
Preexisting colonization of antibiotic-resistant organ-
isms among ICU patients has implications for the
choices of empirical therapy. In cases where ESBL-PS
has become prevalent, the adequacy of 3rd generation
cephalosporins as empirical antibiotics for serious
gram negative infections in critically ill patients has
been put in doubt. Carbapenems and piperacilline/
tazobactam became the systematic alternative agents
(Babini and Livemore, 2000; D’Agata et al., 1999).
Excessive antibiotic use in the different wards of
2006 Extended spectrum b-lactamase producing Enterobacteriaceae in Lebanese ICU patients 175
Table 4. Antimicrobial activities (MIC50, MIC90, and percentages of susceptibility) of imipenem, cefoxitin, ceftazidim, cefotaxime, ceftriaxone, cefepime, gentamicin, amikacin,
amoxicillin/clavulanic acid£, piperacilline/tazobactam¶, aztreonam, and ciprofloxacin on ESBL producing E. coli, Klebsiella spp., and Enterobacter spp.
E. coli (n95) Klebsiella spp. (n16) Enterobacter spp. (n7)
MIC50 MIC90 MIC range %S MIC50 MIC90 MIC range %S MIC50 MIC90 MIC range %S
Imipenem 0.19 0.25 0.002–0.5 100.0 0.12 0.25 0.016–0.19 100.0 0.12 0.5 0.094–0.75 100.0
Cefoxitin 1.5 16 0.5–32 82.1 6 20 1.5–16 50.0 32 64 16–64 26.8
Ceftazidime 128 256 4–256 10.5 96 256 16–256 6.3 96 256 48–256 0.0
Cefotaxime 32 96 2–256 34.7 16 256 2–256 12.5 48 256 3–256 14.3
Ceftriaxone 256 256 48–256 — 12 256 4–256 — 256 256 4–300 —
Cefepime 16 96 1–256 — 2 64 0.75–64 — 8 96 1–84 —
Gentamicin 4 64 0.25–64 60.0 28 64 0.5–64 31.3 16 48 1–48 42.9
Amikacin 2 4 0.125–16 83.2 4 8 0.25–16 75.0 2 8 0.25–8 100.0
Amx/ Clv£16 32 1–256 — 32 128 24–256 — 32 128 32–48 —
Pip/ Tazo¶4160.016–256 66.3 8 32 4–256 25.0 8 64 2–256 0.0
Aztreonam 24 196 2–256 — 64 256 32–256 — 128 256 64–256 —
Ciprofloxacin 64 64 16–64 7.4 64 64 48–64 18.8 64 64 32–64 28.6
hospitals, mainly in the ICU, has been associated with
the emergence of b-lactam-resistant enteric bacilli
(Babini, 2000; Babini and Livemore, 2000; Masterton,
2000; Monnet, 2000; Waterer and Wunderink, 2001).
The fact that the majority of these isolates (n95)
were Escherichia coli followed by Klebsiella pneumo-
niae (n16) and Enterobacter spp. (n7) is in agree-
ment with previous Lebanese data and international
studies (Araj, 1999; Daoud and Hakime, 2003; Master-
ton, 2000); however, it may be contradictory to other
surveys (Mulvey et al., 2004; Munday et al., 2004a, b).
The non-interpreted antimicrobial activities of the 3rd
generation cephalosporins allows us to study the phe-
notype of resistance and reveals that susceptibility to
ceftazidime (9.3%) is lower than that to cefotaxime
(30.5%) in the studied populations. Due to the large
number of samples analyzed (n1,882), only one se-
lective medium, including ceftazidime rather than cefo-
taxime, was chosen according to previous phenotypic
results revealing a higher percentage of resistance to
ceftazidime than to cefotaxime among ESBL-PS in
Lebanon. Moreover, ceftazidime was reported as the
best single test antibiotic for the detection of ESBL
production (Martin et al., 1996). Nevertheless, the
medium used may have failed to detect some ESBLs
and our results may thus underestimate the spread of
other ESBL types.
The antimicrobial susceptibility patterns of the iso-
lated ESBL strains show consistency among the differ-
ent medical centers involved in this study. This was ex-
pected in view of the small population of Lebanon, the
easy and frequent transfer of patients between areas
and hospitals, and the similar medical and therapeuti-
cal practices. The higher susceptibility of NN hospital
isolates to gentamicin, ciprofloxacin, and trimethoprim-
sulphametoxazole may be explained by the low num-
ber of isolates (n7) or may be due to a different epi-
demiology taking into account that this hospital is lo-
cated in the north of Lebanon, while the other hospitals
are closer to the capital, Beirut. The minimum in-
hibitory concentrations clearly show that imipenem is
still the most effective drug against ESBL-PS. To our
knowledge, resistance has never been reported in
Lebanon to this antibiotic among Enterobacteriacae. A
significant emergence of imipenem resistant Acineto-
bacter baumanii and Pseudomonas aeruginosa seems
to be an important consequence of carbapenem use in
our ICUs (data not shown). Other than imipenem and
among the clinically used antibiotics, amikacin and
piperacilline/tazobactam are associated with a higher
% of susceptibility.
Our results highlight the increasing occurrence of
ESBL-PS in the ICU environment of Lebanese hospi-
tals, showing the serious endemicity of this issue. On
the other hand, it shows clearly that limited therapeuti-
cal alternatives are available for the treatment of infec-
tions produced by ESBL-PS. This is alarming for Infec-
tion Control Units in Lebanese hospitals and medical
centers.
Acknowledgments
We acknowledge Drs. Jacques Mokhbat, Monzer Hamze,
Raymond Rohban, and Hiam Matta for their valuable collabora-
tion and Alexis Cocozaki and Nabil Hanna for their technical as-
sistance.
References
Araj, G. F. (1999) Antimicrobial resistance in Lebanon. APUA
Newsletter, 1, 14.
176 DAOUD et al. Vol. 52
Table 5. Univariate analysis of risk factors for ESBL producing strains’ carriage.
Risk factor Cases Controls pValue
(n59) (n59)
•Mean age (years) 52.615.5 53.713.9 NS
Median 51.2 58.5
•Male/female ratio 1.16/1 1.16/1
•Length of stay in ICU, day, meanSD 18.714.5 16.914.1 NS
Median 14.6 13.7
•Evidence for infection with ESBL producing strain (number) 67.8% (40) 20.3% (12) 0.0001
•Type of feeding (oral/enteral) 41/20 36/25 NS
•Immunosuppresion factors (diseases, treatment) (no/yes) 23/38 44/17 0.0001
•General surgery/abdominal surgery 35/24 28/10 0.0025/0.0001
Babini, G. S. and Livermore, D. M. (2000) Antimicrobial resis-
tance amongst Klebsiella spp. Collected from intensive
care units in Southern and Western Europe in 1997–
1998. J. Antimicrob. Chemother., 45,183–189.
Bonnet, R. (2004) Growing group of extended-spectrum b-lacta-
mases: The CTX-M enzymes. Antimicrob. Agents
Chemother., 48,114.
Bradford, P. A. (2001) Extended-spectrum b-lactamases in the
21st century: Characterization, epidemiology, and detec-
tion of this important resistance threat. Clin. Microbiol., 14,
933–951.
Bradford, P. A., Cherubin, C. E., Idemyor, V., Ramussen, B. A.,
and Bush, K. (1994) Multiply resistant Klebsiella pneumo-
niae strains from two Chicago Hospitals: Identification of
the extended-spectrum TEM-12 and TEM-10 ceftazidime-
hydrolyzing b-lactamases in a single isolate. Antimicrob.
Agents Chemother., 38, 761–766
Bush, K., Jacoby, G. A., and Medeiros, A. A. (1995) A functional
classification scheme for b-lactamases and its correlation
with molecular structure. Antimicrob. Agents Chemother.,
39, 1211–1233
D’Agata, E. M., Venkataraman, L., DeGirolami, P., et al. (1999)
Colonization with broad-spectrum cephalosporin-resistant
Gram-negative bacilli in intensive care units during a
nonoutbreak period: Prevalence, risk factors, and rate of
infection. Crit. Care Med., 27, 1090–1095.
Daoud, Z. and Hakime, N. (2003) Prevalence and susceptibility
patterns of extended-spectrum b-lactamase-producing
Escherichia coli and Klebsiella pneumoniae in a general
university hospital in Beirut, Lebanon. Rev. Esp.
Quimioter., 16, 233–238.
Daoud, Z., Hakime, N. I., Hamze, M., Mangeney, N., Matta, H.,
Mokhbat, J. E., Rohban, R., Sarkis, D. K., Doucet-Popu-
laire, F., and Moubareck, C. (2005) Countrywide spread of
community- and hospital-acquired extended-spectrum
beta-lactamase (CTX-M-15)-producing Enterobacteri-
aceae in Lebanon. J. Clin. Microbiol., 43: 3309–3313.
Hacek, D. M., Bednarz, P., Noskin, G. A., Zembower, T., and
Peterson, L. R. (2001) Yield of vancomycin-resistant Ente-
rococci and multidrug-resistant Enterobacteriaceae from
stools submitted for Clostridium difficile testing compared
to results from a focused surveillance program. J. Clin.
Microbiol., 39, 1152–1154.
Ho, P. L., for the Hong Kong Intensive Care Unit Antimicrobial
Resistance Study (HK-ICARE) Group. (2003) Carriage of
methicillin-resistant Staphylococcus aureus, ceftazidime-
resistant Gram-negative bacilli, and vancomycin-resistant
Enterococci before and after intensive care unit admis-
sion. Crit. Care Med., 31, 1175–1182.
Itokazu, G. S., Quinn, J. P., Bell-Dixon, C., Kahan, F. M., and
Weinstein, R. A. (1996) Antimicrobial resistance rates
among aerobic gram-negative bacilli recovered from pa-
tients in intensive care units: Evaluation of a national post-
marketing surveillance program. Clin. Infect. Dis., 23,
779–784.
Jarlier, V., Nicolas, M. H., Fournier, G., and Philippon, A. (1988)
Extended broad-spectrum beta-lactamases conferring
transferable resistance to newer beta-lactam agents in
Enterobacteriaceae: Hospital prevalence and susceptibil-
ity patterns. Rev. Infect. Dis., 10, 867–878.
Jarlier, V., Nicolas, M., Fournier, G., and Philippon, A. (1998)
Extended broad-spectrum b-lactamases conferring trans-
ferable resistance to newer b-lactam agents in Enetrobac-
teriaceae: Hospital prevalence and susceptibility patterns.
Rev. Infect. Dis., 10, 867–878.
Kaye, K. S., Engemann, J. J., Fraimow, H. S., and Abrutyn, E.
(2004) Pathogens resistant to antimicrobial agents: Epi-
demiology, molecular mechanisms, and management. In-
fect. Dis. Clin. N. Am., 18, 467–511.
Cormican, M. G., Marshall, S. A., and Jones, R. N. (1996) De-
tection of extended-spectrum b-lactamases (ESBL)-pro-
ducing strains by the Etest ESBL Screen. J. Clin. Microb.,
34, 1880–1884.
Masterton, R. G. (2000) Surveillance studies: How can they
help the management of infection? J. Antimicrob.
Chemother., 46 (Suppl. T2), 53–58.
Merrer, J., Santoli, F., Appere-de-Vecchi, C., et al. (2000) “Colo-
nization pressure” and risk of acquisition of methicillin-re-
sistant Staphylococcus aureus in a medical intensive care
unit. Infect. Control Hosp. Epidemiol., 21, 718–723.
Monnet, D. L. (2000) Toward multinational antimicrobial resis-
tance surveillance systems in Europe. Int. J. Antimicrob.
Agents, 15, 91–101.
Mulvey, M. R., Bryce, E., Boyd, D., Ofner-Agostini, M., Chris-
tianson, S., Simor, A. E., and Paton, S. (2004) Extended-
spectrum b-lactamase-producing Escherichia coli and
Klebsiella spp. in Canadian hospitals. Antimicrob. Agents
Chemother., 8,1204–1214.
Munday, C. J., Whitehead, G. M., Todd, N. J., Campbell, M.,
and Hawkey, P. M. (2004a) Predominance and genetic di-
versity of community- and hospital-acquired CTX-M ex-
tended-spectrum b-lactamases in York, UK. J. Antimicrob.
Chemother., 54, 628–633.
Munday, C. J., Xiong, J., Li, C., Shen, D., and Hawkey, P. M.
(2004b) Dissemination of CTX-M type b-lactamases in
Enterobacteriaceae isolates in the People’s Republic of
China. Int. J. Antimicrob. Agents, 23, 175–180.
National Nosocomial Infections Surveillance (NNIS) System Re-
port (2003) Data summary from January 1992 through
June 2003, issued August 2003. Am. J. Infect. Control,
31, 481–498.
NCCLS (2000) Performance Standards for Antimicrobial Disk
Susceptibility Tests, 6th ed., Approved Standard M100-
S9, National Committee for Clinical Laboratory Standards,
Wayne, PA.
Nseir, S., Di Pompeo, C., Soubrier, S., Delour, P., Lenci, H.,
Roussel-Delvallez, M., Onimus, T., Saulnier, F., Mathieu,
D., and Durocher, A. (2005) First-generation fluoro-
quinolone use and subsequent emergence of multiple
drug-resistant bacteria in the intensive care unit. Crit.
2006 Extended spectrum b-lactamase producing Enterobacteriaceae in Lebanese ICU patients 177
Care Med., 33–42.
Pitout, J. D. D., Sanders, C. C., and Sanders, W. E. (1997) An-
timicrobial resistance with focus on b-lactam resistance in
Gram-negative bacilli. Am. J. Med., 103, 51–59.
Toltzis, P., Yamashita, T., Vilt, L., et al. (1997) Colonization with
antibiotic-resistant Gram-negative organisms in a pedi-
atric intensive care unit. Crit. Care Med., 25, 538–544.
Waterer, G. W. and Wunderink, R. G. (2001) Increasing threat
of Gram-negative bacteria. Crit. Care Med., 29, N75–N81.
178 DAOUD et al. Vol. 52
