Nigudgi, Hajare, Biradar, and Anandkumar: Evaluation of different phenotypic diffusion methods in the identification of extended spectrum beta lactamase producing uropathogenic Escherichia coli


In spite of the widespread availability of antibacterial drugs, urinary tract infection (UTI) remains one of the major infections in the community and hospital settings.1 Amongst the large number of antibiotic drugs, β-lactams are the diverse and largely used antibiotics contributing above 50% of all systemic antimicrobial agents available.2 The resistance in bacteria towards beta lactam antibiotics is mainly due to the production of beta lactamase enzyme.3 The second and third generation cephalosporin drugs are precisely designed to neutralize the hydrolytic action of beta lactamase enzymes. Nonetheless, the newest in the reserve of these enzymes by the organisms has been the development of extended spectrum beta lactamases. The ESBL enzymes produced largely by the enteric organisms like, E. coli and Klebsiella sps., which hydrolyze oxyimino-cephalosporins leading to the resistance against cephalosporins and monobactams.4

The identification of ESBL producing organisms is difficult for routine diagnostic microbiology laboratories of developing countries without molecular diagnostic facilities. Also screening of ESBL producing organisms by monitoring the decrease in susceptibility to oxyimino-cephalosporin drugs are not a sensitive tool. The recommended methods for screening ESBL producing organisms are based on the decreased susceptibility to cephalosporins in disc diffusion test.5 But, the standard protocol for reliable phenotypic method for detection of ESBL is not available.6 The existing phenotypic methods for ESBL detection are disc diffusion based screening test and double disc synergy test (DDST), Inhibitory potentiated disc diffusion (IPDD) and E-strip confirmatory tests.

According to CLSI guidelines, a screening test for decreased susceptibility to one of the five representative cephalosporin agents, followed by a confirmatory test would increase the chance of identification. Additionally, gene responsible for the production of ESBL enzyme can be detected by molecular methods.7 But these molecular diagnostic facilities will not normally be available in resource constrained routine microbiology laboratories.

The present research work was intended to find out the appropriate method for the identification of ESBL producing urinary Escherichia coli, where the data on the occurrence of ESBL producing E.coli is lacking.

Materials and Methods

This prospective study was conducted in the Department of Microbiology, Shyam Shah Medical College, Rewa, Madhya Pradesh, India. A total of 500 consecutive urine samples were screened from patients with symptomatic UTI. Clean-catch midstream urine samples were collected in sterile disposable container (Uricol, Hi-Media Laboratories Ltd., Mumbai, India) and processed within one hour. Semi quantitative loop (Hi-Media Laboratories Ltd., Mumbai, India) measuring 2.2 mm diameter with a holding capacity of 0.005 ml was employed to culture urine on CLED agar and MacConkey’s agar. The inoculated plates were incubated over night at 37oC. Isolates in significant number (colony count ≥ 105 CFU/ml) were identified by standard procedures.8 Antibiotic susceptibility test was done by Kirby-Bauer disc diffusion method9 using antibiotic discs: ampicillin (10µg), amoxicillin/clavulanic acid (20/10µg), co-trimoxazole (1.25/23.75 µg), amikacin (30µg), imipenem (10µg), gatifloxacin (5µg) and tobramycin (10 µg).

Disc susceptibility test to screen ESBLs

All the isolates were screened for ESBL production by using three indicator cephalosporins, namely ceftazidime (30 µg), cefotaxime (30 µg) and cefpodoxime (30 µg). The isolates were considered to be resistant, if the inhibition zone diameter of ceftazidime, cefotaxime and cefpodoxime were < 22mm, <27mm and <17mm respectively.

The strains which showed resistance to at least one of the three cephalosporins was further included for phenotypic confirmation method.9, 10

Double disc synergy test (DDST)

The Escherichia coli showing decreased susceptibility to any of the three cephalosporins used were further tested for ESBL production by DDST method. Ceftazidime, cefotaxime, cefpodoxime and amoxy-clav (Hi-Media Laboratories Ltd., Mumbai, India) were used in this method.11, 12 Over the lawn cultured Muller-Hinton agar plates, amoxy-clav and third generation cephalosporin discs were placed at a distance of 20mm from the center. The Plates were incubated at 37ºC for 8 hours. The augmentation in the zone of inhibition of cephalosporins towards the amoxy-clav disc was considered to be positive for ESBL. The standard strains of Klebsiella pneumoniae ATCC 700603 and E. coli ATCC 25922 were used as controls.

Inhibitor potentiated disc diffusion test (IPDD)

The turbidity of E.coli in a broth was matched with 0.5 McFarland turbidity standards and inoculated onto two Muller-Hinton agar plates by streak method. Of the two plates, one was supplemented with 0.004 mg/L Potassium clavulanate (Sigma Aldrich Pvt Ltd, Bengaluru) and another without clavulanate. The ceftazidime, cefotaxime and cefpodoxime disks were placed on both of these plates. The inoculated agar plates were then incubated at 37ºC for 8 hours. The inhibition zones of disks were compared between the plates with and without potassium clavulanate. The difference in the zone size of ≥ 10 mm diameter was taken as positive for the production of ESBL.13, 14

ESBL Epsilometer-strip test (E-strip test)

The commercially available ESBL E-strip (make: AB Biomerieux) contains two gradients of antibiotic drugs. At one end, the strip is impregnated with ceftazidime (0.5 to 32 mg/ml) and on the other end is with ceftazidime (0.125 to 8 mg/ml) with clavulanate (4 mg/ml). The overnight growth of E.coli isolate was suspended in saline to match the turbidity with 0.5 McFarland standards and was then inoculated on Muller Hinton agar plate by lawn culture technique. After drying, the E -test strip was placed on the plate and incubated overnight at 37ºC. The MICs on both ends of the E- strip were interpreted as the point of intersection of the inhibition eclipse with the E-test strip edge. The ratio of ceftazidime/ ceftazidime with clavulanate MIC ≥ 8 indicates the presence of ESBL enzymes.7, 15

Statistical analysis

The results of the study were statistically analyzed using SPSS v 16.0 software wherever suitable. The Chi- square test was done to analyze statistical significance. The p-value of less than 0.05 was considered statistically significant.


In our study, 120 symptomatic urinary tract infection cases were diagnosed as significant bacteriuria due to Escherichia coli by calibrated loop culture technique. The isolated E. coli strains were further used for antibiotic susceptibility testing and ESBL detection.

The antibiogram results shown that 110 (91.6%) E.coli isolates were resistant to ampicillin followed by 42 (35.0%) isolates to co-trimoxazole and 41 (34.1%) isolates to gatifloxacin. However, resistance to tobramycin, amoxy-clav and amikacin were recorded as 31.6%, 19.1% and 15.0% respectively. Only one (0.8%) strain has shown resistance to imipenem as depicted in the Table 1.

Table 1

Association between drug resistance pattern and ESBL producing E.coli strains


Resistance pattern (n=120)

ESBL positives strains (n=28)

ESBL negative strains (n=92)

- Value

Ampicillin (A)

110 (91.6%)

28 (100%)

82 (89.1%)


Amoxyclav (AC)

23 (19.1%)

10 (35.7%)

13 (14.1%)


Cotrimoxazole (Co)

42 (35.0%)

15 (53.5%)

27 (29.3%)


Amikacin (AK)

18 (15.0%)

9 (32.1%)

9 (9.7%)


Imipenem (I)

1 (0.8%)

1 (3.5%)


Gatifloxacin (GF)

41 (34.1%)

15 (53.5%)

26 (28.2%)


Tobramycin (Tb)

38 (31.6%)

13 (46.4%)

25 (27.1%)


ESBL screening indicators:


55 (45.8%)

27 (96.4%)

28 (30.4%)


Ceftazidime (CA)

48 (40.0%)

23 (82.1%)

25 (27.1%)


Cefotaxime (CE)

45 (37.5%)

25 (89.2%)

20 (21.7%)


In the DDST and IPDD screening test for ESBL production, 62 (51.6%) isolates were resistant to at least one of the three representative cephalosporin drugs. The highest resistance was observed with cefpodoxime (n=55; 45.8%) followed by ceftazidime (n=48; 40%) and cefotaxime (n=45; 37.5%). Out of the three cephalosporins tested in the study, ceftazidime was found to be the better antibiotic drug for the identification of ESBL production by both DDST and IPDD (Table 2).

Table 2

Comparisonof different diffusion methods for the detection of extended spectrum of beta lactamases



Screening test* (n=120)

Confirmatory tests** (n=62)


IPDD test

E-strip test


Cefpodoxime (CEP)






Ceftazidime (CA)






Cefotaxime (CE)






ESBL positives




[i] * Disc diffusion test

[ii] **The E.coli strain showing resistance to at least one cephalosprin indicator antibiotic is selected for confirmatory test.

[iii] NA – Not Applicable

In the present study, the efficacy of DDST and IPDD disc diffusion tests were compared with ESBL E-strip test. By DDST method, 9 (14.5%) strains were positive for ESBL, one strain was false positive and 18 (29.0%) strains showed false negative result. The IPDD test showed 28 (45.1%) as mentioned in Table 3.

Table 3

Detection of ESBL producing E.coli strains by E-strip test and their respective log2 reduction

No. of Strains

Ceftazidime MIC

TZ/TZL Ratio

MIC log 2



Alone (TZ)

With clavulanate (TZL)































The ESBL positives with augmentation zone of inhibition diameter is ≥ 10mm.The mean zone augmentation (95% CI) was 16.2 (12.8, 21.4) mm for ceftazidime, 13.9 (12.2, 18.0) mm for cefotaxime and 18.6 (12.2, 18.6) mm for cefpodoxime as mentioned in Table 4.

Table 4

Confirmation of screening test positive ESBL producers by inhibitory potentiated

disc diffusion (IPDD) test


Mean Zone diameter± S.D (mm)

Mean Zone augmentation


MH Agar

MH agar + Clavulanate

(mm) (95% CI)

ESBL Positive strains(n=28)




16.2 (12.8, 21.4)




34.2 ±3.9

13.9 (12.2, 18.0)





18.6 (12.2, 18.6)


ESBL Negative strains(n=34)




1.6 (0.8, 2.4)





1.4 (0.4, 2.6)




35.4 ±2.6

1.6 (0.6,2.0)


[i] CI, Confidence interval

[ii] MH– Muller Hinton

The ESBL E-strip test results showed, 28 (45.1%) E.coli isolates were identified as ESBL producers with Ceftazidime/ceftazidime-clavulanate (TZ/TZL) ratio between 8 and 256. Of 28 ESBL positives, 26 isolates showed TZ/TZL ratio of between 32 and 256 with MIC log2 dilution reduction ≥ 5. The remaining 34 (54.8%) E.coli isolates were negative for ESBL production with the ratio less than 8 and log2 reduction less than 3 (Table 4).

In 28 ESBL positives E.coli, 27 (96.4%) isolates have shown resistance to cefpodoxime followed by 25 (89.2%) strains to cefotaxime and 23 (82.1%) strains to ceftazidime. Out of 92 Non-ESBL isolates, 28 (30.4%) were resistant to cefpodoxime followed by 25 (27.1%) strains toceftazidime and 20 (21.7%) strains to cefotaxime. The ESBL positive strains exhibited statistically significant (p<0.05) resistance to tobramycin followed by amoxyclav and amikacin compared to non-ESBL isolates. The resistance to multi drugs was noticeable in ESBL producing (96.7%) isolates compared to non- ESBL producing (52.1%) isolates, which was statistically significant (p<0.05).


In the our study, 62 (51.6%) Escherichia coli strains from UTI cases showed resistant to one of the three representative cephalosporin drugs. Out of these 62 E.coli isolates, 28 (45.1%) were found ESBL producers by IPDD test and E-strip test. However, only 9 (14.5%) strains were positive by DDST method. A study from Hyderabad (TS) reported that, 19.8% Enterobacteriaceae were potential ESBL producers by double disc synergy test with 63.7% of Escherichia coli and 14% of Klebsiella pneumoniae shown ESBL production.16 In another study from Western part India,48.3% ESBL producing urinary isolates were resistant to cefotaxime drug.17

In comparison to DDST, IPDD test appeared to be better methods for confirming ESBLs shown a sensitivity of 100% and specificity of 95.8%. The DDST method was unsuccessful to detect 30.6% ESBL producing isolates with a low sensitivity of 42.2% and positive predictive value of 91.6%. The sensitivity of the DDST test strongly relies on the accurate location of discs on the culture plate.13 The earlier studies have shown that the ESBL E-strip test was comparatively more sensitive, dependable and appropriate method,15 based on which it was used as the gold standard test for confirming ESBL phenotypically in the present study.

The three representative cephalosporin drugs used in IPDD test showed increased zone of inhibition (≥10mm) to ESBL producers. The cefpodoxime showed superior mean zone augmentation (18.6 mm) compared to mean zone augmentation of ceftazidime (16.2 mm) and cefotaxime (13.9 mm). In our study, the IPDD test was more sensitive with ceftazidime than cefotaxime, which is comparable with the previous work by Ho et al.13 The benefit of IPDD test is that the ESBLs could be easily separated from non-ESBL producing organisms by a break point of ≥ 10 mm zone augmentation and more than one cephalosporin drug can be tested on single test plate.

The Ceftazidime is identified as an exceptional substrate for most ESBL enzymes18, 19 and Bush group 2be enzymes can be differentiated from other beta lactamase enzymes by the decrease in ceftazidime MIC in presence of beta lactamase inhibitors like clavulanate.20 Another study suggested that the automated technique like Vitek and Epsilometer–strip tests are sensitive and reliable compared to the disc diffusion tests.21

The disadvantage of these diffusion methods is that, they may not detect inhibitor-resistant beta lactamases. The ESBL confirmatory test is based on the demonstration of inhibition by clavulanate. But, other mechanisms of beta lactam resistance, like AmpC enzymes, change in the porin channel and variants ESBL enzymes may be present or co-exist with ESBL, which interfere in the results of these diffusion tests.

Due to the presence of a large amount of false positives in the screening procedures, the two steps strategy (screening and confirmatory tests) may be adapted. Even the sensitivity of DDST depends on the accurate placement of the discs on the agar plate and interpretation of DDST results is more subjective compared to recording the results of E-strip test and IPDD test. Therefore, IPDD test may be preferred over E-strip test, as it is equally sensitive, cost effective and more than one cephalosporin drug can be used per test.

The drug resistance of ESBL producing E.coli were significantly higher (p <0.05) than the non-ESBL producing isolates. The ESBL producing isolates showed greater resistance to co-trimoxazole (53.5%) and amikacin (32.1%), which is comparable with the study done by Spanu et al., Baby Padmini et al. and Menon et al.22, 23, 24 The cephalosporin resistant organisms have shown resistance to other antibiotics classes like fluoroquinolones and aminoglycosides which are in concordance with the other study reports.10, 25 It was also noticed that, most of the ESBL producing E.coli (96.7%) were resistant to two or more drugs compared to Non-ESBL producing isolates (52.1%). This finding was in complete agreement with a study conducted by Tankhiwale et al., as they also reported the significantly higher multidrug resistancein ESBL producing isolates than in non ESBL producers.17

In our area, the records pertaining to the incidence of ESBL producing isolates is very limited. In addition, confirmatory molecular methods need to be carried out in the identification of ESBL producing isolates to validate the results of different phenotypic diffusion methods. Probably too much reliance and extensive use of third generation cephalosporin drugs in the treatment of enteric gram negative organisms has been the principal factor responsible for increased drug resistance to cephalosporins and other class of antibiotics. The precise identification of ESBL producing isolates, judicious use of broad spectrum antibiotics, periodic surveillance of antibiotic resistance pattern and efforts to decrease empirical antibiotic therapy would go a long way in addressing some of the issues related with ESBL production in clinical isolates.

Source of Funding


Conflict of Interest



We thank the faculty of Dept. of Microbiology Navodaya Medical College, Raichur and M R Medical College, Kalaburagi for their constant inspiration in conducting this work.



S Sharma Current understanding of pathogenic mechanisms in UTIsAnn Natl Acad Med Sci1997331318


JJ Bronson JF Barrett Everninomycin Carbapenem Glycylcycline Quinolone, Everninomycin, Glycylcycline, Carbapenem, Lipopeptide and Cephem antibacterial is in clinical developmentCurr Med Chem20018177593


AA Medeiros Evolution and dissemination of beta lactamases accelerated by generation of beta lactam antibioticsClin Infect Dis1997241945


GA Jacoby AA Medeiros More extended β-lactamasesAntimicrob Agent Chemother19913516971704


European Committee on Antimicrobial Susceptibility Testing. 2010. Breakpoint tables for interpretation of MICs and zone diameters. European Committee on Antimicrobial Susceptibility Testing. Vaxjo, Sweden


H Garrec L Drieux-Rouzet J Gilmard V Jarlier J Robert Comparison of nine phenotypic methods for detection of extended spectrum β- lactamase production by EnterobacteriaceaeJ Clin Microbiol2011493104857


V Gupta An update on newer β-lactamasesIndian J med Res200712641727


JG Collee RS Miles B Watt JG Collee AG Fraser BP Marmion A Simmons Mackie and McCartney Practical Medical Microbiology14th editionChurchill Livingstone IncLondon199613149


Clinical and Laboratory Standards Institute. Performance standards for antimicrobial susceptibility testing: 30th informational supplement. CLSI document M02, M07 & M11. Wayne, PA: CLSI; 2019


S Srisangkaew M Vorachit The Optimum Agent for Screening and Confirmatory Tests for Extended - Spectrum b-Lactamases in Escherichia coliand Klebsiella pneumoniae in Ramathibodi Hospital, ThailandJ Infect Dis Antimicrob Agent200421115


B Duttaroy S Mehta Extended spectrum b lactamases (ESBL) in clinical isolates of Klebsiella pneumoniae and Escherichia coliIndian J Pathol Microbiol2005481458


P Datta A Thakur B Mishra V Gupta Prevalence of Clinical Strains Resistant to Various b-Lactams in a tertiary Care Hospital in IndiaJpn J Infect Dis2004571469


PL Ho KH Chow KY Yuen WS Ng PY Chau Comparison of a novel, inhibitor - Potentiated disc diffusion test with other methods for the detection of extended - spectrum b-Lactamases in Escherichia coli and Klebsiella pneumoniaeJ Antimicrob Chemother1998424954


B Bedenic J Vranes LJ Mihaljevic M Tonkic M Sviben V Plecko Sensitivity and Specificity of various b-lactam Antibiotics and Phenotypical Methods for Detection of TEM, SHV and CTX-M Extended-Spectrum b-LactamasesJ Chemother200719212739


MG Cormican SA Marshall RN Jones Detection of extended spectrum β-lactamase producing strains by the E-test ESBL ScreenJ Clin Microbiol199634818804


MS Kumar V Lakshmi R Rajgopal Occurrence of extended spetrum beta lactamases among Enterobacteriaceaespp. Isolated at tertiary care instituteIndian J Med Microbiol200624320811


SS Tankhiwale SV Jalgaonkar S Ahmed U Hassani Evaluation of Extended spectrum beta lactamase in urinary isolatesIndian J Med Res20041205536


PA Bradford CC Sanders Development of test panel of βlactamases expressed in a common Escherichia coli host background for evaluation of new β-lactam antibioticsAntimicrob Agent Chemother19953930813


GP Katsanis J Sporgo MJ Ferraro I Sutton GA Jacoby Detection of Klebsiella pneumonia and Escherichia coli strains producing extended–spectrum β-lactamasesJ Clin Microbiol1994326916


V Jarlier MH Nicolas G Fournier A Philippon Extended broad spectrum β- lactamases conferring resistance to newer β-lactam agents in Enterobacteriaceae: hospital prevalence and susceptibility patternsRev Infect19881086778


CC Sanders JA Washigton LA Barry C Shubert Assessment of the Vitek ESBL testPrograms and abstracts of the 34thinterscience conference on antimicrobial agents and chemotherapy199444American Society for MicrobiologyWashington DC123


T Spanu F Luzzaro M Perilli G Amicosante A Toniolo G Fadda Occurrence of extended-spectrum beta-lactamases in members of the family Enterobacteriaceae in Italy: implications for resistance to beta-lactams and other antimicrobial drugsAntimicrob Agents Chemother200246119620210.1128/AAC.46.1.196-202.2002


S Babypadmini B Appalaraju Extended spectrum-lactamases in urinary isolates of Escherichia coli and Klebsiellapneumoniae - Prevalence and susceptibility pattern in a tertiary care hospitalIndian J Med20042231724


T Menon D Bindu CPG Kumar S Nalini MA Thirunarayan Comparison of double disc and three dimensional methods to screen for ESBL producers in a tertiary care hospitalIndian J Med Microbiol200624211720


A Subha S Ananthan Extended spectrum b-lactamase (ESBL) mediated resistance to third generation cephalosporins among Klebsiella pneumoniaein ChennaiIndian J Med Microbiol2002202925


© This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

  • Article highlights
  • Article tables
  • Article images

Article History

Received : 09-08-2021

Accepted : 02-09-2021

Available online : 22-09-2021

View Article

PDF File   Full Text Article

Copyright permission

Get article permission for commercial use


PDF File   XML File   ePub File

Digital Object Identifier (DOI)

Article DOI

Article Metrics

Article Access statistics

Viewed: 92

PDF Downloaded: 37

Wiki in hindi