Extended double disc synergy testing reveals a low prevalence of extended-spectrum b-lactamases in Enterobacter spp. in Vienna, Austria

Transcription

1 Journal of Antimicrobial Chemotherapy (2007) 59, doi: /jac/dkm060 Advance Access publication 8 March 2007 Extended double disc synergy testing reveals a low prevalence of extended-spectrum b-lactamases in Enterobacter spp. in Vienna, Austria Petra Apfalter 1 *, Ojan Assadian 2, Florian Daxböck 2, Alexander M. Hirschl 1, Manfred L. Rotter 1 and Athanasios Makristathis 1 1 Department of Clinical Microbiology, Institute of Hygiene and Medical Microbiology, Medical University Vienna, Vienna, Austria; 2 Department of Hospital Hygiene, Institute of Hygiene and Medical Microbiology, Medical University Vienna, Vienna, Austria Received 8 August 2006; returned 13 November 2006; revised 29 January 2007; accepted 5 February 2007 Objectives: The aims of this study were to determine the prevalence of extended-spectrum b-lactamases (ESBLs) in AmpC-carrying Enterobacter spp. in a tertiary care university hospital in Vienna, Austria, and to implement a cost-effective strategy to detect ESBLs in this particular genus on a routine basis. Methods: Clinical Enterobacter isolates (n 5 208) were investigated by means of (i) an inhibitor-potentiated diffusion test using cefpodoxime, (ii) an expanded double disc diffusion synergy test (discs of cefotaxime, ceftazidime, cefpodoxime and cefepime placed around amoxicillin/clavulanic acid), (iii) the Etest ESBL screening method and (iv) the cefoxitin cefotaxime antagonist test. Cefepime MICs were determined by separate Etests. Results: Of 208 isolates, 76 (37%), 18 (9%) and 92 (44%) were derepressed, partially derepressed and inducible AmpC producers, respectively. Eight (4%) ESBL-producing Enterobacter strains could be detected, all of which would have been detected using disc-based tests. Six out of eight strains were genetically not related, as assessed by random amplification of polymorphic DNA. Typing results were confirmed by means of enterobacterial repetitive intergenic consensus PCR. The MIC 90 of cefepime was not different in ESBL carriers (range 2 4 mg/l), and was especially low in inducible AmpC producers (0.125 mg/l). More than half of all Enterobacter isolates (n 5 110; 53%) were partly derepressed or fully inducible AmpC producers. In the absence of cefoxitin, they appeared susceptible or intermediately susceptible to cefazolin (n 5 8; 9%), cefuroxime (n 5 75; 81.5%), ceftazidime (n 5 91; 99%), cefotaxime (n 5 92; 100%), cefpodoxime (n 5 75; 81.5%) and cefepime (n 5 91; 99%). Conclusions: Susceptibility to third-generation cephalosporins would have been falsely assumed in more than half of all Enterobacter isolates, but ESBL in Enterobacter is currently rare in our institution. Integration of multiple double disc tests into the routine antibiogram seems a reliable approach to screen for emerging resistance mechanisms. Etests did not provide additional information in this study. Keywords: Etest, AmpC, third-generation cephalosporins, fourth-generation cephalosporins Introduction Enterobacter spp. can cause severe, difficult to treat nosocomial infections, 1 and antimicrobial susceptibility testing of Enterobacter spp. can be challenging for numerous reasons. First, production of chromosome-encoded AmpC b-lactamases is a typical mechanism of extended-spectrum cephalosporin resistance in Enterobacter spp. 2 Although derepressed AmpC producers are easily detectable due to their in vitro resistance to third-generation cephalosporins, inducible AmpC-producing... *Correspondence address. Department of Clinical Microbiology, Institute of Hygiene and Medical Microbiology, Vienna General Hospital, Waehringer Guertel 18-20/5P, 1090 Vienna, Austria. Tel: þ ; Fax: þ ; petra.apfalter@meduniwien.ac.at # The Author Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 Disc tests for ESBL detection in Enterobacter spp. strains might be overlooked when susceptibilities are read in the absence of an inducer agent. 3 Thus, susceptibility may falsely be assumed. Secondly, even though CLSI (formerly NCCLS) guidelines do not recommend screening Enterobacter spp. for extended-spectrum b-lactamase (ESBL) production on a routine basis, 4 ESBL-producing Enterobacter has been increasingly reported. 5 8 Thirdly, both resistance mechanisms can occur simultaneously. In such situations, the use of clavulanic acid to inhibit an ESBL may induce high-level expression of the chromosomal AmpC enzyme and may then antagonize rather than protect the antibacterial activity of the partner b-lactam, masking the synergistic effect required to detect ESBL production. 9 Currently, there are no recommendations from CLSI for the detection of ESBL in AmpC b-lactamase-producing Enterobacteriaceae. 4 When administering empirical antimicrobial therapy, knowledge on local resistance mechanisms is essential. Although fourth-generation cephalosporins might be a therapeutic option in Enterobacter carrying only AmpC, they are not recommended in the case of an infection by ESBL-producing strains. 10 The aims of this study were (i) to determine the prevalence of derepressed, partially derepressed and inducible AmpC as well as simultaneous ESBL production in Enterobacter strains in our institution, (ii) to evaluate our current strategy not to report results of second- and third-generation cephalosporins in the case of in vitro susceptibility, and indicating on the report that these agents are clinically not indicated, and (iii) to establish an easy and cost-effective procedure for the routine microbiology laboratory to detect ESBL in AmpC-carrying Enterobacter spp. on a routine basis. Materials and methods Strains Between October 2004 and March 2005, a random sample of 208 clinically relevant, non-duplicate Enterobacter isolates were routinely collected in the microbiology laboratory of the Vienna General Hospital and stored at 2808C using the Cryobank system (Mast Diagnostica, Reinfeld, Germany). All strains were obtained from hospitalized patients and were isolated from blood (n ¼ 5; 2%), lower respiratory tract (collected by bronchoalveolar lavage; n ¼ 48; 23%), urinary tract (n ¼ 46; 22%), deep surgical wound infections (n ¼ 73; 35%) and faecal swabs (n ¼ 36; 17%) from bone marrow transplant recipients. The isolates were thawed and subcultured once over night at 358C on Columbia agar supplemented with 5% sheep blood (Becton Dickinson, Heidelberg, Germany). Strains were retested for species identification using the VITEK-GN card, as recommended by the manufacturer (Vitek 2 system, BioMérieux, Marcy l Étoile, France). Antimicrobial susceptibility testing Antimicrobial susceptibility testing was performed by means of the agar disc diffusion method following the respective CLSI (formerly NCCLS) guidelines. 11 The following antimicrobial agents (all supplied from Oxoid, Hampshire, UK) were tested: ampicillin (10 mg), amoxicillin/clavulanic acid (20 mg/10 mg), cefazolin (30 mg), cefuroxime (30 mg), cefoxitin (30 mg), cefotaxime (30 mg), ceftazidime (30 mg), cefpodoxime (10 mg), cefpodoxime/clavulanic acid (10 mg/ 1 mg), cefepime (30 mg), imipenem (10 mg), gentamicin (10 mg), amikacin (30 mg), ciprofloxacin (5 mg) and trimethoprim/ sulfamethoxazole (1.25 mg/23.75 mg). ESBL production was tested by comparing the inhibitory zone diameter of a cefpodoxime disc and a cefpodoxime/clavulanic acid disc. A difference of 5 mm in the zone diameter was considered as a positive result. This procedure is recommended by CLSI 12 for ESBL screening in Klebsiella pneumoniae, Klebsiella oxytoca, Escherichia coli and Proteus mirabilis, and thus is integrated in the routine antibiogram tested for Enterobacteriaceae in our institution. For this study, all strains were additionally subjected to an expanded double disc diffusion synergy test. Preliminary experiments were performed to evaluate disc-to-disc distances of 20, 25 and 30 mm (centre to centre). A distance of 25 mm between discs was regarded as optimal to observe a keyhole phenomenon. This distance also allowed incorporation of double disc synergy testing into routine susceptibility testing with commercially available disc dispensers (Oxoid): discs of cefotaxime, ceftazidime, cefpodoxime and cefepime were placed around an amoxicillin/clavulanic acid disc at a distance of 25 mm, as described previously. 13,14 A keyhole phenomenon was regarded as positive for ESBL production. In addition, all isolates were also tested by the Etest ESBL screening method using cefepime strips with and without clavulanic acid (AB Biodisk, Solna, Sweden). 15 Etest reading and interpretation were carried out according to the manufacturer s instruction. Briefly, cefepime MIC reduction by three or more 2-fold dilutions with clavulanic acid, phantom zones or deformation of the inhibition ellipse was indicative of ESBL production. To study the inducibility of the AmpC enzyme, the cefoxitin cefotaxime antagonist test was performed as described recently, 16 and the b-lactamase inducibility was confirmed by the presence of a blunted cefotaxime zone adjacent to cefoxitin. The latter test, as well as disc diffusion synergy testing, was done as displayed in Figure 1. Cefepime MICs were determined by means of separate Etest strips. Definition of resistance types According to the characteristics of b-lactamase production, resistance types were defined as follows: 5 (i) Group 1, derepressed AmpC producers were resistant to cefoxitin (zone diameter 14 mm), resistant or intermediate susceptible to cefotaxime (22 mm), had a negative cefoxitin cefotaxime antagonist test and a negative ESBL production; (ii) Group 2, partially derepressed AmpC isolates were resistant to cefoxitin (14 mm), resistant or intermediate susceptible to cefotaxime (22 mm), had a positive cefoxitin cefotaxime antagonist test and a negative ESBL production; (iii) Group 3, ( partly) derepressed AmpC producers with ESBL production were resistant to cefoxitin (14 mm), had a negative cefoxitin cefotaxime antagonist test and produced ESBL (see also Figure 1); (iv) Group 4, strains were susceptible to cefoxitin (18 mm) and had a positive ESBL production defined by an increase of 5mm in the zone diameter of a cefpodoxime/clavulanic acid disc compared with the diameter of cefpodoxime alone, a keyhole phenomenon in the double disc synergy test to at least one of the broad-spectrum cephalosporins surrounding the amoxicillin/clavulanic acid disc and/or a positive cefepime/cefepime clavulanate Etest; and (v) inducible AmpC producers were susceptible to cefoxitin (18 mm), had a positive cefoxitin cefotaxime antagonist test and a negative ESBL production. Typing of ESBL-producing Enterobacter isolates Random amplification of polymorphic DNA (RAPD) was applied for molecular typing. Enterobacterial repetitive intergenic consensus (ERIC)-PCR was performed for confirmation. The same DNA 855

3 Apfalter et al. (a) (b) CTX FOX FEP AMC CPD CLA preparations were used for both assays, whereby DNA was extracted from pure cultures as described previously. 17 RAPD was performed using the primer P15 (5 0 -AAT GGC GCA G-3 0 ). 18 The PCR mixture included a Ready-to-Go RAPD Analysis Bead (Amersham Biosciences, Little Chalfont, UK), 2 ml (50 pmol) of primer P15, 3 ml of DNA solution and sterile distilled water for a final volume of 25 ml. The amplification conditions included an initial denaturation step (948C for 5 min) followed by 40 cycles, each consisting of 948C for 30 s, 368C for 30 s and 728C for 1 min. For ERIC-PCR, the primer ERIC-2 (5 0 -AAG TAA GTG ACT GGG GTG AGC G-3 0 ) was applied. 19 The amplification profile consisted of an initial denaturation of 4 min at 948C, 35 cycles (948C for 45 s, 588C for 1 min and 728C for 2 min) and a final elongation of 10 min at 728C. The PCR products were separated by electrophoresis in 2% TopVision TM LE GQ agarose (Fermentas Life Sciences, St Leon-Rot, Germany) for 3.5 h (2.5 V/cm) and visualized by ethidium bromide staining. The patterns were interpreted individually by two researchers, whereby single-band differences were ignored. 20 Results CAZ 2.5 cm CPD Figure 1. Derepressed AmpC-hyperproducing E. cloacae strain producing ESBL. (a) Striped arrow: cefoxitin cefotaxime antagonist test to check for AmpC inducibility; black arrows: expanded double disc diffusion synergy test (cefotaxime, cefpodoxime, ceftazidime and cefepime around amoxicillin/clavulanic acid); white arrow with bold black edges: testing for ESBL production by comparing the inhibitory zone diameter of a 10 mg cefpodoxime disc and a 10 mg cefpodoxime/1 mg clavulanic acid disc. AMC, amoxicillin/clavulanic acid; CAZ, ceftazidime; CLA, clavulanic acid; CPD, cefpodoxime; CTX, cefotaxime; FEP, cefepime; FOX, cefoxitin. (b) White arrows indicate keyhole phenomena due to double disc synergy testing. Species identification of the 208 Enterobacter isolates collected for this study revealed 179 Enterobacter cloacae (86%), 27 Enterobacter aerogenes (13%) and one each Enterobacter sakazakii and Enterobacter asbinae. Susceptibility test results by means of agar disc diffusion were as follows: all strains were susceptible to imipenem; resistance rates for amikacin, gentamicin, ciprofloxacin and trimethoprim/sulfamethoxazole were 0.5% (1/208; 3 intermediate susceptible), 5.7% (12/208; 1 intermediate susceptible), 11.5% (24/208; 3 intermediate susceptible) and 11.5% (24/208; 7 intermediate susceptible), respectively. Enterobacter AmpC producers Of 208 Enterobacter spp. isolates, 76 (37%), 18 (9%) and 92 (44%) were derepressed, partially derepressed and inducible AmpC producers, respectively. Fourteen strains could not be categorized (discussed subsequently). Thus, 95% (197/208) of Enterobacter isolates were definitely AmpC carriers. All derepressed strains were resistant to cefuroxime and cefpodoxime, and. 90% were resistant to cefotaxime (71/76) and ceftazidime (74/76). Only 3 out of 76 were also resistant to cefepime by disc diffusion, but none had an MIC.32 mg/l. In the group of the partially derepressed strains, one strain (1/18) was susceptible to cefuroxime and 40% strains were susceptible to cefotaxime (7/18) and ceftazidime (8/18). All strains were resistant to cefpodoxime and susceptible to cefepime. Nearly half of all Enterobacter isolates (n ¼ 92; 44%) were fully inducible AmpC producers. In the absence of cefoxitin, they appeared susceptible or intermediate susceptible to cefazolin (n ¼ 8; 9%), cefuroxime (n ¼ 75; 81.5%), ceftazidime (n ¼ 91; 99%), cefotaxime (n ¼ 92; 100%), cefpodoxime (n ¼ 75; 81.5%) and cefepime (n ¼ 91; 99%). Figure 2 summarizes all derepressed, partially derepressed and inducible AmpC producers with and without additional ESBL production with reference to their MICs of cefepime. The MIC 90 of cefepime was not significantly different for ESBL-producing derepressed AmpC producers (4 mg/l) when compared with strains without ESBL production (2 mg/l), whereas for inducible AmpC producers, it was mg/l. According to current CLSI breakpoints for Enterobacteriaceae, all isolates would have been in the susceptible range of cefepime. ESBL-producing Enterobacter spp. In the present study, eight (4%) ESBL-producing Enterobacter strains could be detected. Of these, seven were also fully or partly derepressed AmpC producers. The single cefoxitinsusceptible ESBL only isolate probably was falsely defined as ESBL by the cefepime/cefepime clavulanate Etest due to a cefepime MIC reduction by three or more 2-fold dilutions. ESBL production could not be confirmed by the other methods used in this study. The concordance of test results for the other seven 856

4 Disc tests for ESBL detection in Enterobacter spp AmpC producers carrying ESBL is displayed in Table 1. In all strains with a significant MIC reduction by Etest in the presence of clavulanate (n ¼ 4), an increase of 5 mm could be observed in the zone diameter of a cefpodoxime/clavulanic acid disc when compared with the diameter of cefpodoxime alone. All of the seven strains would have been detected by disc-based tests: six out of seven showed a keyhole phenomenon in the double disc synergy test to at least cefepime. The 14 strains that could not be categorized were confirmed as Enterobacter by GN and API 20 (BioMérieux). Seven of 14 were susceptible to cefoxitin, only 2 of 14 and 1 of 14 were resistant to cefazolin and cefuroxime, respectively, and none was resistant to cefotaxime, ceftazidime, cefpodoxime and cefepime. All ESBL tests were negative and MICs of cefepime were mg/l in all 14 strains. Genetic relationship between ESBL-producing Enterobacter isolates (n ¼ 8) Six different genetic profiles (Types A F) could be determined and confirmed by RAPD and ERIC-PCR, respectively (data not shown). Strains 1 and 3 were genetically identical and assigned to Type A. These strains were isolated from different patients on different wards at different points in time. Strains 3 and 4 were also identical and assigned to Type C. The latter strains were isolated from very low birth-weight infant twins, who were nursed at the same neonatal intensive care unit at the same time. Thus, it is very likely that cross-contamination had occurred in this special case. Discussion MIC (mg/l) Inducible AmpC Partly derepressed AmpC Derepressed AmpC AmpC + ESBL Enterobacter spp. can cause difficult to treat nosocomial infections in critically ill patients. 1 Owing to mutations, the expression of a chromosomal gene encoding the AmpC b-lactamase 2 can be increased during therapy with third-generation cephalosporins Figure 2. Distribution of cefepime MICs in inducible, partly derepressed and derepressed AmpC and AmpC producers with ESBL production. The x-axis and y-axis display the MIC of cefepime as determined by Etest and the number of isolates tested, respectively. such as cefotaxime, ceftazidime and ceftriaxone. 3,21 As to what extent fourth-generation cephalosporins are affected by this mechanism is not fully understood. 1,8,22 24 As of 2006, CLSI guidelines for antimicrobial susceptibility testing recommend to test repeated isolates of Enterobacter to detect developing resistance during prolonged therapy with third-generation cephalosporins, because initially susceptible strains may become resistant within 3 4 days after initiation of therapy. Whether this is also true for fourth-generation agents is not mentioned in the CLSI guidelines. In daily clinical routine, however, testing of more than one isolate is not always feasible and empirical as well as calculated antimicrobial therapy often is based on broadspectrum cephalosporins. In addition, ESBL-producing Enterobacter has been increasingly reported. 5 8 In this scenario, penicillins, aztreonam and all extended-spectrum cephalosporins should be reported as resistant. 4 Some data derived from clinical studies in which patients with an invasive ESBL-producing enterobacterium have been treated with fourth-generation cephalosporins showed an unfavourable outcome when compared with patients treated with carbapenems. 25 To the best of our knowledge, data on the in vitro susceptibility of Enterobacter spp. in Austria are scant, 26,27 both for the frequency of occurrence of diverse AmpC producers and for ESBL-producing strains. Thus, in the case of empirical therapy, we would have to rely on data derived from other institutions, some of which report up to 30% ESBL-producing Enterobacter. 8 According to this assumption, first-line empirical Enterobacter therapy would have to rely on carbapenems in any case. The median number of blood cultures analysed at the Department of Clinical Microbiology of the Vienna General Hospital, a 2200 bed university-affiliated, tertiary care hospital with 242 intensive and intermediate care unit beds, is over per year, with a mean positivity rate of 12%. Besides E. coli and Klebsiella spp., E. cloacae was the Gram-negative organism most frequently isolated from positive blood cultures between 1998 and 2005, causing 2700 bacteraemia episodes during this 8 year period. Retrospective review of all blood cultures positive for E. cloacae revealed that resistance rates for cefotaxime, which was tested during the whole period, increased from 11% in 1998 to 96% in As a consequence and independent from their phenotypic susceptibility pattern, broadspectrum second- and third-generation cephalosporins were reported as clinically not indicated from 2003 onwards in the case of in vitro susceptibility. In 2005, 65% of Enterobacter bloodstream isolates were resistant to cefotaxime. Knowledge of the current situation with respect to resistance data, however, is of critical importance in our institution, since empirical Gram-negative therapy is largely based on third- and fourthgeneration cephalosporins. Although susceptibility to third-generation cephalosporins was evidently not given in partly and fully derepressed AmpC producers (94 strains; 46%), it would have been falsely assumed in 92 isolates (44%), which were in vitro susceptible to thirdgeneration cephalosporins, but fully inducible AmpC producers. Thus, our strategy to report third-generation cephalosporins in Enterobacter isolates as being clinically not indicated was retrospectively viewed on the safe side. In addition, partially derepressed strains would have been missed as AmpC producers if they were tested in the absence of the inducer cefoxitin. According to current CLSI breakpoints for Enterobacteriaceae, all isolates would have been susceptible to cefepime, irrespective 857

5 Apfalter et al. Table 1. Concordance of test results of different phenotypic methods for ESBL detection in seven Enterobacter cloacae isolates and one Enterobacter aerogenes isolate Disc diffusion synergy test with AMC b No. Species AmpC status CPD/CPD CLA disc diffusion a CPD CTX CAZ FEP FEP MIC (mg/l) FEP/FEP CLA Etest c 1 E. cloacae derepressed pos (R) pos (I) pos (R) pos (S) pos 1.5 pos 2 E. cloacae derepressed neg (R) neg (R) neg (R) neg (S) pos 4 NA 3 E. cloacae derepressed neg (R) neg (R) neg (R) neg (S) pos 0.75 NA 4 E. cloacae derepressed neg (R) neg (R) neg (R) neg (S) pos 1 NA 5 E. cloacae derepressed pos (R) neg (R) pos (R) neg (S) pos 3 pos 6 E. cloacae derepressed pos (R) pos (I) pos (R) pos (S) pos 1.5 pos 7 E. cloacae unknown neg (S) neg (S) neg (S) neg (S) neg 0.75 pos 8 E. aerogenes derepressed pos (R) neg (R) pos (S) neg (R) neg 8 pos FEP, cefepime; FEP/FEP CLA, cefepime/cefepime clavulanic acid; CPD, cefpodoxime; CPD/CPD CLA, cefpodoxime/cefpodoxime clavulanic acid; CTX, cefotaxime; CAZ, ceftazidime; AMC, amoxicillin/clavulanic acid. a pos, positive, a 5 mm (neg, negative, a,5 mm ) increase in the zone diameter in combination with clavulanic acid versus tested alone. b pos, positive, a keyhole phenomenon at a distance of 25 mm between AMC and the respective cephalosporin; neg, negative, no keyhole phenomenon at a distance of 25 mm between AMC and the respective cephalosporin; S, I or R in brackets indicates susceptibility according to CLSI agar diffusion breakpoints (S, susceptible; I, intermediate susceptible; R, resistant). c pos, positive, cefepime MIC reduction by three or more 2-fold dilutions with clavulanic acid, phantom zones or deformation of the inhibition ellipse; NA, not applicable, no endpoint could be determined due to the scaling of the strip; clavulanic acid may have induced high-level expression of AmpC chromosomal enzyme and antagonized rather than protected the bacterial activity of the partner b-lactam. whether they were derepressed, partially derepressed or inducible AmpC producers with and without additional ESBL production. The MIC 90 of cefepime was similar in ESBL-producing derepressed AmpC producers (4 mg/l) when compared with strains without ESBL production (2 mg/l) and was especially low for inducible AmpC producers (0.125 mg/l). Detection of an inducible AmpC producer might give an additional hint to an especially low MIC for cefepime. In addition, a recently published review article indicates that the use of cefepime to treat serious nosocomial infections (e.g. bacteraemia, pneumonia and urinary tract infections) may be associated with clinical success, even in ESBL-carrying Enterobacteriaceae. 28 Jacoby et al. 29 have recently proposed a similar disc-based scheme that could be used to detect various b-lactamases in E. coli and K. pneumoniae. Application of combined multiple double disc tests integrated in the routine antibiogram seems to us an easy and reliable approach to do so, especially in AmpC carriers such as Enterobacter spp., which might act as hidden ESBL reservoirs. Incorporation of these double disc tests into the antibiotic testing panel for Enterobacteriaceae saves time and financial resources. There were only eight ESBL producers in Enterobacter spp., which currently seem to play a minor role in our institution. It is, however, important to continuously monitor emerging resistance mechanisms on a routine basis since it may be difficult to draw this conclusion with this small number of isolates. In contrast to what others reported, 15 Etest strips did not provide additional information when compared with disc-based tests in this study. Acknowledgements The excellent technical assistance of Cornelia Üblauer is deeply appreciated. Transparency declarations We have no conflicts of interest or financial support to disclose. References 1. Goethaert K, Van Looveren M, Lammens C et al. High-dose cefepime as an alternative treatment for infections caused by TEM-24 ESBL-producing Enterobacter aerogenes in severely-ill patients. Clin Microbiol Infect 2006; 12: Jacoby GA, Munoz-Price LS. The new b-lactamases. N Engl J Med 2005; 352: Livermore DM, Brown DFJ. Detection of b-lactamase-mediated resistance. J Antimicrob Chemother 2001; 48: S Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing: Sixteenth Informational Supplement M100-S16 (Volume 26, Number 3. Zone diameter interpretative standards and equivalent minimal inhibitory concentration breakpoints for Enterobacteriaceae). CLSI, Wayne, USA, Pai H, Hong JY, Byeon JH et al. High prevalence of extendedspectrum b-lactamase-producing strains among blood isolates of Enterobacter spp. collected in a tertiary hospital during an 8-year period and their antimicrobial susceptibility patterns. Antimicrob Agents Chemother 2004; 48: Canton R, Oliver A, Coque TM et al. Epidemiology of extendedspectrum b-lactamase-producing Enterobacter isolates in a Spanish hospital during a 12-year period. J Clin Microbiol 2002; 40: Quinteros M, Radice M, Gardella N et al. Extended-spectrum b-lactamases in Enterobacteriaceae in Buenos Aires, Argentina, public hospitals. Antimicrob Agents Chemother 2003; 47: Rossi F, Baquero F, Hsueh PR et al. In vitro susceptibilities of aerobic and facultatively anaerobic Gram-negative bacilli isolated from patients with intra-abdominal infections worldwide: 2004 results from SMART (Study for Monitoring Antimicrobial Resistance Trends). J Antimicrob Chemother 2006; 58:

6 Disc tests for ESBL detection in Enterobacter spp. 9. Thomson KS. Controversies about extended-spectrum and AmpC b-lactamases. Emerg Infect Dis 2001; 7: Szabó D, Bonomo RA, Silveira F et al. SHV-type extendedspectrum b-lactamase production is associated with reduced cefepime susceptibility in Enterobacter cloacae. J Clin Microbiol 2005; 43: National Committee for Clinical and Laboratory Standards. Performance Standards for Antimicrobial Susceptibility Testing: Fourteenth Informational Supplement M100-S14. NCCLS, Wayne, USA, Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing: Sixteenth Informational Supplement M100-S16 (Volume 26, Number 3. Table 2A. Screening and confirmatory tests for ESBLs in Klebsiella pneumoniae, K. oxytoca, Escherichia coli and Proteus mirabilis). CLSI,Wayne,USA, Tzelepi E, Giakkoupi P, Sofianou D et al. Detection of extendedspectrum b-lactamases in clinical isolates of Enterobacter cloacae and Enterobacter aerogenes. J Clin Microbiol 2000; 38: Thomson KS, Sanders CC. Detection of extended-spectrum b-lactamases in members of the family Enterobacteriaceae: comparison of the double-disc test and three-dimensional test. Antimicrob Agents Chemother 1992; 36: Stürenburg E, Sobottka I, Noor D et al. Evaluation of a new cefepime-clavulanate ESBL Etest to detect extended-spectrum b-lactamases in an Enterobacteriaceae strain collection. J Antimicrob Chemother 2004; 54: Sanders CC, Sanders WE, Goering RV. In vitro antagonism of b-lactam antibiotics by cefoxitin. Antimicrob Agents Chemother 1982; 21: Coenye T, Liu L, Vandamme P et al. Identification of Pandoraea species by 16S ribosomal DNA-based PCR assays. J Clin Microbiol 2001; 39: Denton M, Todd NJ, Kerr KG et al. Molecular epidemiology of Stenotrophomonas maltophilia isolated from clinical specimens from patients with cystic fibrosis and associated environmental samples. J Clin Microbiol 1998; 36: Maroye P, Doermann HP, Rogues AM et al. Investigation of an outbreak of Ralstonia pickettii in a paediatric hospital by RAPD. J Hosp Infect 2000; 44: Renders N, Romling Y, Verbrugh H et al. Comparative typing of Pseudomonas aeruginosa by random amplification of polymorphic DNA or pulsed-field gel electrophoresis of DNA macrorestriction fragments. J Clin Microbiol 1996; 34: Livermore DM. b-lactamases in laboratory and clinical resistance. Clin Microbiol Rev 1995; 8: Fernandez-Cuenca F, Rodriguez-Martinez JM, Martinez- Marthinez L et al. In vivo selection of Enterobacter aerogenes with reduced susceptibility to cefepime and carbapenems associated with decreased expression of a 40 kda outer membrane protein and hyperproduction of AmpC b-lactamase. Int J Antimicrob Agents 2006; 27: Pfaller MA, Sader HS, Fritsche TR et al. Antimicrobial activity of cefepime tested against ceftazidime-resistant Gram-negative clinical strains from North American Hospitals: report from the SENTRY Antimicrobial Surveillance Program ( ). Diagn Microbiol Infect Dis 2006; 56: Goldstein FW. Cephalosporinase induction and cephalosporin resistance: a longstanding misinterpretation. Clin Microbiol Infect 2002; 8: Paterson DL, Ko WC, Von Gottberg A et al. Antibiotic therapy for Klebsiella pneumoniae bacteremia: implications of production of extended-spectrum b-lactamases. Clin Infect Dis 2004; 39: Krause R, Mittermayer H, Feierl G et al. In vitro activity of newer broad spectrum b-lactam antibiotics against Enterobateriaceae and non-fermenters: a report from Austrian intensive care units. Austrian Carbapenem Susceptibility Surveillance Group. Wien Klin Wochenschr 1999; 30: Bujdáková H, Klimáčková J, Allerberger F et al. Spectrum and transferability of b-lactam resistance in hospital strains of Enterobacter isolated in Bratislava and Innsbruck. Int J Antimicrob Agents 2000; 16: Ramphal R, Ambrose PG. Extended-spectrum b-lactamases and clinical outcomes: current data. Clin Infect Dis 2006; 42: S Jacoby GA, Walsh KE, Walker VJ. Identification of extendedspectrum, AmpC, and carbapenem-hydrolyzing b-lactamases in Escherichia coli and Klebsiella pneumoniae by disk tests. J Clin Microbiol 2006; 44: