Introduction

The extended spectrum beta lactamase (ESBL) enzymes are plasmid-mediated enzymes capable of hydrolyzing and inactivating a wide variety of beta lactams, including third generation cephalosporins, penicillins and aztreonam.(₁) Plasmids responsible for ESBL production carry resistance to many antibiotics like aminoglycosides, fluoroquinolones, tetracyclines, chloramphenicol and co-trimoxazole.(₂,₃) The ESBL producing organisms are reported in increasing numbers worldwide.(₄,₅,₆,₇) National Committee for Clinical Laboratory Standards (NCCLS), now called Clinical Laboratory Standards Institute (CLSI) recommends screening for ESBL production among E.coli, K.pneumoniae and K.oxytoca.(₈) However, other organisms reported to produce ESBL less frequently are Enterobacter species, Proteus species, Morganella morganii, Serratia marcescens and Pseudomonas aeruginosa.(₉,₁₀)

Materials and Methods

Between July 2004 to Dec 2004, a total of 1889 clinically significant Gram negative bacilli belonging to Enterobacteriaciae isolated from various clinical specimens were subjected to ESBL screening. The isolates were identified by standard techniques. Antimicrobial susceptibility testing was performed by Kirby-Bauer method and interpretation of results was as recommended by NCCLS.(₈)

ESBL production was tested by using ceftazidime (30mcg) and ceftazidime plus clavulanic acid (30/10mcg) discs on Mueller-Hinton agar. Organism was considered as an ESBL producer if there was a ≥5mm increase in zone diameter around ceftazidime/clavulanic acid disc compared to zone around ceftazidime disc alone. ESBL production was tested in parallel with the antibiotic susceptibility testing on a separate Mueller Hinton agar plate. As per NCCLS guidelines, an isolate was reported as resistant to all penicillins, cephalosporins and aztreonam, if it was an ESBL producer. Klebsiella pneumoniae ATCC 700603 strain was used as ESBL producing control strain, E.coli ATCC 25922 was used as ESBL negative control strain. Sample source, patient location and other relevant details were noted. Sample source distribution of ESBL producers is shown in Table 1.

Figure 1

Table 1: Sample source distribution of ESBL producers

Statistical analysis

Results

The most frequent Enterobacteriaciae isolated was E.coli (48.2%) followed by Klebsiella species (34.5%) and Enterobacter species (7.8%) [Table 2]. ESBL production was noted in 60.98% (n=1152/1889) of the isolates tested. Among the ESBL producers, urinary isolates were 535 and 617 isolates were from other samples. Percentage of ESBL producers within a species was highest in Enterobacter species 70.9% (n=105/148) followed by Klebsiella species 67.4% (n=439/651) and E.coli 62.3% (n= 568/911). Locationwise distribution of ESBL producers and predominant organism from different locations of the hospital was analysed. E.coli was the most frequent ESBL producer from ambulatory and from ward samples. Klebsiella species was the most frequent ESBL producer from intensive care unit samples [table 3].

Figure 2

Table 2 : Frequency of Enterobacteriaciae and ESBL producers

Figure 3

Table 3: Locationwise distribution of ESBL producers

Sensitivity to meropenem was 99.8% (n=1150/1152), piperacillin tazobactam 96.8% (n=1115/1152), cefoperazone sulbactam 93.3% (n=1075/1152), cotrimoxazole 21.7% (n=250/1152), ciprofloxacin 16.6% (n=191/1152), gentamicin 13.3% (153/1152), amikacin 53% (610/1152). Urinary isolates were tested against nitrofurantoin, norfloxacin, nalidixic acid. These showed susceptibility rate of 66.35% (n=355/535), 16.4% (n=88/535) and 12.3% (n=66/535) respectively [Table 4].

Figure 4

Table 4: Antibiotic susceptibility pattern of ESBL positive isolates

Discussion

The frequency of ESBL producers of 60.98% in our study is comparable to previous Indian studies.(₁₁,₁₂,₁₃) The most frequent isolates in our study were E.coli and Klebsiella species. However, the highest ratio of ESBL production was in Enterobacter species 70.9% (n=105/148). This demonstrates that ESBL screening should not be limited only to E.coli and Klebsiella species. Other species which showed ESBL production were Citrobacter species 30.5% (n=25/82) and Proteus species 19.4% (n=13/67). However, these were infrequent isolates. None of the Salmonella species, Serratia marcescens and Morganella morganii showed ESBL production. Indian studies so far are limited to screening ESBL production among E.coli and K.pneumoniae.(₁₄,₁₅)

Our study shows that there is a significant difference in the type of organism isolated from different locations of the hospital (p=0.001). The highest number of ESBL producers were E.coli in ambulatory and ward samples, while from intensive care units it was Klebsiella species. This demonstrates that E.coli and Klebsiella species pose a major challenge in hospitalized patients during selection of empirical antibiotic therapy.

It is an established fact that, ESBL producers show cross resistance to other antibiotics also, thus limiting the therapeutic choice. We have noted this in our study as well. Sensitivity to meropenem was almost absolute (99.8%). Sensitivity to piperacillin tazobactam (96.8%) and cefoperazone sulbactam (93.3%) was also good. Whereas sensitivity to gentamicin, ciprofloxacin, co-trimoxazole, nalidixic acid and norfloxacin was poor and not suitable for empirical selection. Urinary isolates showed >50% susceptibility to amikacin and nitrofurantoin. These are the alternative cheaper antibiotics which can be considered for empirical therapy. The routine susceptibility testing by clinical laboratories fail to detect ESBL positive strains and shows false invitro sensitivity to cephalosporins.(₁₆) Screening for ESBL production as a routine procedure in clinical laboratories gives valuable information to the clinician in appropriate selection of antibiotics. Hence, we conclude based on our study that, there is a high prevalence rate of ESBL producers among Enterobacteriaciae. E.coli and Klebsiella species pose a major concern among these. Based on the prevalence rate of the ESBL producers in a healthcare facility, antibiotic policy of the institution can be tailored to achieve superior therapeutic outcome and bring about a reduction in healthcare costs. It also eliminates misuse of conventional cephalosporins in a significant proportion of patients.

Correspondence to

Dr.Shashikala.S.,M.D Asst. Professor, Department of Microbiology, Amrita Institute of Medical Sciences and Research Centre, Kochi, Kerala-PIN-682 026 India Tel : 04842801234, 04845008010 Fax : 04842802020 shashikalas@aims.amrita.edu