Comparison of the carbapenem inactivation method (CIM) and modified carbapenem inactivation method (mCIM) for the detection of carbapenemase-producing organisms

Abstract

The phenotypic detection of carbapenemase-producing organisms (CPO) is an important adjunct to genotypic detection methods.1Viau R. Frank K.M. Jacobs M.R. et al.Intestinal carriage of Carbapenemase-producing organisms: current status of surveillance methods.Clin Microbiol Rev. 2016; 29: 1-27Crossref PubMed Scopus (0) Google Scholar Phenotypic methods such as the well described Carba-NP test are utilised in diagnostic laboratories that do not have access to more expensive gene detection platforms, and for the detection of CPO harbouring novel resistance genes.2Nordmann P. Poirel L. Dortet L. Rapid detection of carbapenemase-producing Enterobacteriaceae.Emerg Infect Dis. 2012; 18: 1503-1507Crossref PubMed Scopus (504) Google Scholar The Carbapenem Inactivation Method (CIM) was first described by Van der Zwaluw et al. in 2015 and detects the presence of carbapenemase activity in bacteria via the hydrolysis of meropenem.3Van der Zwaluw K. de Haan A. Pluister G.N. et al.Carbapenem Inactivation Method (CIM), a simple and low cost alternative to the Carba NP test to assess phenotypic carbapenemase activity in Gram-negative rods.PLoS One. 2015; 10: e0123690Crossref PubMed Scopus (220) Google Scholar However when the Clinical and Laboratory Standards Institute (CLSI) updated its performance standard in 2017, a modified version of this test was recommended: the modified CIM (mCIM).4Clinical and Laboratory Standards Institute (CLSI) Performance Standards for Antimicrobial Susceptibility Testing.27th ed. Clinical and Laboratory Standards Institute, Wayne, PA2017Google Scholar The CIM and mCIM offer several advantages over other phenotypic detection methods (e.g., Carba-NP): they are simple to perform, use readily available, inexpensive consumables and do not have a subjective colorimetric endpoint.4Clinical and Laboratory Standards Institute (CLSI) Performance Standards for Antimicrobial Susceptibility Testing.27th ed. Clinical and Laboratory Standards Institute, Wayne, PA2017Google Scholar, 5Madkour L.A. Soliman M.S. Hassan D.M. et al.Detection of carbapenemase producers: evaluating the performance of the carbapenem inactivation method and Carba NP test versus multiplex PCR.J Glob Antimicrob Resist. 2017; 9: 10-14Crossref PubMed Scopus (7) Google Scholar To date, there has been extensive evaluation of the CIM,6Aguirre-Quinonero A. Cano M.E. Gamal D. Evaluation of the carbapenem inactivation method (CIM) for detection carbapenemase activity in Enterobacteriaceae.Diagn Microbiol Infect Dis. 2017; 88: 214-218Crossref PubMed Scopus (19) Google Scholar, 7Atkas E. Malkocoglu G. Otlu B. et al.Evaluation of the carbapenem inactivation method for detection of carbapenemase-producing Gram-negative bacteria in comparison with the RAPIDEC Carba NP.Microb Drug Resist. 2017; 23: 457-461Crossref PubMed Scopus (21) Google Scholar, 8Tijet N. Patel S. Melano R.G. Detection of carbapenemase activity in Enterobacteriaceae: comparison of the carbapenem inactivation method versus the Carba NP test.J Antimicrob Chemother. 2016; 71: 274-276Crossref PubMed Scopus (58) Google Scholar, 9Gauthier L. Bonnin R. Dortet L. et al.Retrospective and prospective evaluation of the carbapenem inactivation method for the detection of carbapenemase-producing Enterobacteriaceae.PLoS One. 2017; 12: e0170769Crossref PubMed Scopus (25) Google Scholar, 10Yildiz S. Kaskatepe B. Avcikucuk H. et al.Performance of CarbaNP and CIM tests in OXA-48 carbapenemase-producing Enterobacteriaceae.Acta Microbiol Immunol Hung. 2017; 64: 9-16Crossref PubMed Scopus (9) Google Scholar limited published studies evaluating the mCIM test,11Pierce V.M. Simner P.J. Lonsway D.R. Modified carbapenem inactivation method (mCIM) for phenotypic detection of carbapenemase production among Enterobacteriaceae.J Clin Microbiol. 2017; 55: 2321-2333Crossref PubMed Scopus (149) Google Scholar and only one head-to-head comparison of the two methods.12Miller S.A. Hindler J.A. Chengcuenca A. Use of ancillary carbapenemase tests to improve specificity of phenotypic definitions for carbapenemase producing Enterobacteriaceae.J Clin Microbiol. 2017; 55: 1827-1836Crossref PubMed Scopus (11) Google Scholar The aim of our study was to directly compare the performance of the mCIM and CIM in the setting of a local blaKPC outbreak, previously described by Chang et al.,13Chang L.W. Buising K.L. Jeremiah C.J. et al.Managing a nosocomial outbreak of carbapenem-resistant Klebsiella pneumoniae: an early Australian hospital experience.Intern Med J. 2015; 45: 1037-1043Crossref PubMed Scopus (33) Google Scholar and then by using a broader range of reference laboratory acquired isolates. During the previously described blaKPC outbreak,13Chang L.W. Buising K.L. Jeremiah C.J. et al.Managing a nosocomial outbreak of carbapenem-resistant Klebsiella pneumoniae: an early Australian hospital experience.Intern Med J. 2015; 45: 1037-1043Crossref PubMed Scopus (33) Google Scholar 67 isolates with an elevated minimum inhibitory concentration (MIC) to meropenem or resistance to multiple antibiotic classes were referred from St Vincent's Pathology to the Microbiological Diagnostic Unit (MDU) for molecular characterisation. Forty-seven of these isolates were found to harbour a carbapenemase gene. Table 1 provides details of these isolates and an additional 28 isolates obtained from reference laboratories in Victoria and South Australia. The additional 28 (reference laboratory) isolates included 20 CPOs (predominantly blaOXA and blaVIM) which were selected to compliment the range of carbapenemase producing isolates in the St Vincent's Pathology set and a further eight isolates with multiple antibiotic class resistance but no carbapenemase gene. The combined isolate sets included 67 CPOs and 29 non-CPOs. The non-CPO isolates were selected for inclusion as they had been referred to a reference laboratory after displaying resistance to multiple classes of antibiotics or a meropenem MIC ≥0.5 μg/mL on initial Vitek (bioMérieux, France) antimicrobial susceptibility testing. All 29 non-CPO isolates therefore had been subjected to an in-house broad range polymerase chain reaction (PCR) for carbapenemase genes and had no carbapenemase gene detected. All non-CPOs had subsequent antimicrobial susceptibility performed by E-test (bioMérieux); 11 isolates had a meropenem MIC ≤2 μg/mL and 18 isolates had a meropenem MIC >2 μg/mL (range 0.125 μg/mL to >32 μg/mL). A head-to-head comparison of the CIM and mCIM was performed using all 96 isolates (67 carbapenemase-producing isolates and 29 non-carbapenemase producing isolates). The 67 St Vincent's Pathology isolates were subcultured from original isolates stored in glycerol at –70°C. Isolates were subcultured onto MacConkey agar for determination of purity and then a repeat subculture was made on Columbia agar with 5% sheep blood (SBA) for testing. Isolate identification was confirmed by MALDI-TOF MS (Bruker Daltonics, Germany) and repetitive subculturing of isolates was avoided to prevent resistance gene dropout. The provenance and number of passages of the reference laboratory isolates was unknown. The CIM was performed according to the original method as described by Van der Zwaluw et al.3Van der Zwaluw K. de Haan A. Pluister G.N. et al.Carbapenem Inactivation Method (CIM), a simple and low cost alternative to the Carba NP test to assess phenotypic carbapenemase activity in Gram-negative rods.PLoS One. 2015; 10: e0123690Crossref PubMed Scopus (220) Google Scholar In brief, using a 10 μL loop the test isolate was suspended in 400 μL of sterile water in an Eppendorf tube. The suspension was vortexed for 10–15 seconds, after which a 10 μg meropenem antibiotic disk (Oxoid, UK) was added to the tube. The Eppendorf tube was then incubated in ambient air at 35–37°C. After 2 h incubation the meropenem disk was carefully removed from the Eppendorf tube with sterile forceps, squeezed gently to remove excess moisture and placed on Mueller–Hinton agar (Oxoid) pre-inoculated with a lawn of susceptible Escherichia coli (ATCC strain 25922). After 12–18 h incubation in ambient air at 35–37°C the zone of inhibition around the meropenem disk was measured using Vernier callipers. Results were confirmed by a second reader. Both readers were blinded to the carbapenemase status of the isolates. Carbapenemase production was recorded as present when there was no zone of inhibition. A positive control (Klebsiella pneumoniae ATCC BAA-1705) and a negative control (Klebsiella pneumoniae ATCC BAA-1706) were tested in parallel with the isolates. Isolates subsequently found to have produced false negative results were retested under the same conditions once and then subsequently using a longer vortexing time (2 min) and a disk incubation period of 6 h. The mCIM was performed according to the method described in CLSI M100, Performance Standards for Antimicrobial Susceptibility Testing.4Clinical and Laboratory Standards Institute (CLSI) Performance Standards for Antimicrobial Susceptibility Testing.27th ed. Clinical and Laboratory Standards Institute, Wayne, PA2017Google Scholar In brief, a 1 μL loop of test isolate was suspended in 2 mL of Trypticase Soy Broth (TSB) in a test tube. The test tube was then vortexed for 10–15 seconds after which time a 10 μg meropenem antibiotic disk (Oxoid) was added to the tube. The test tube was incubated in ambient air for 4 h at 35–37°C. After 4 h incubation, the meropenem disk was carefully removed with sterile forceps, squeezing excess moisture from the disk. The 10 μg meropenem disk was then applied to a Mueller–Hinton agar (Oxoid) plate pre-inoculated with a lawn of susceptible Escherichia coli (ATCC 29522). After incubation in ambient air at 35–37°C for 18–24 h the zone of inhibition was read using Vernier callipers. Results were confirmed by a second reader. Both readers were blinded to the carbapenemase status of the isolates. As per the CLSI standard, a carbapenemase positive result was recorded for any isolate exhibiting a zone of inhibition of 6–15 mm or presence of growth within a zone of 16–18 mm. If the zone of inhibition was 19 mm or more the isolate was determined to be carbapenemase negative. For isolates with a zone size of 16–18 mm the carbapenemase status was recorded as indeterminate. A positive control (Klebsiella pneumoniae ATCC BAA-1705) and a negative control (Klebsiella pneumoniae ATCC BAA-1706) were tested in parallel. Isolates producing false negative results were retested under the same conditions and then subsequently using a longer vortexing time (2 min) and a disk incubation period of 6 h.Table 1Carbapenemase-producing isolates and non-carbapenemase-producing isolates subjected to the CIM and mCIM testOrganismResistance geneaAs determined by reference laboratory PCR.No. isolatesNo. positiveCIMmCIMCarbapenemase-producing isolatesAmbler Class AKPC-2252525KPC-3111Ambler Class BNDM-1222NDM-4111NDM-510109NDM-7222VIM-1100VIM-2110VIM-5100IMP-1111IMP-4111IMP-14222Ambler Class DOXA-48554OXA-23, OXA-51110OXA-24, OXA-40100OXA-181665Mixed Ambler ClassNDM-5, OXA-232555NDM-5, OXA-232, VIM-2111Total676459Non-carpabenemase producing isolatesAmbler Class A (non-carbapenemase)blaCIT200blaDHA200Non-β-lactamase method of resistancend2511Total2911a As determined by reference laboratory PCR. Open table in a new tab Evaluated against the 67 St Vincent's Pathology isolates referred to MDU, the CIM test and mCIM both produced a false negative result for one Klebsiella pneumoniae blaVIM-5. However, in addition to this isolate the mCIM gave false negative results for a Acinetobacter baumannii blaOXA-23/OXA-51 and Escherichia coli blaNDM-5 (Table 1). The CIM test and the mCIM test both produced a false positive result for an Enterobacter cloacae that harboured no resistance genes by reference laboratory PCR. The sensitivity of the CIM test was 97.9% and the sensitivity of the mCIM test was 93.6%. The specificity of both tests was 96.5%. Evaluated against the combined sets of isolates (Table 1) the CIM test produced a further two false negative results (Klebsiella pneumoniae blaVIM-1, Acinetobacter baumannii blaOXA-24/-40) and the mCIM produced a further three false negative results (the Klebsiella pneumoniae blaVIM-1 and Acinetobacter baumannii blaOXA-24/-40 mentioned above and a Pseudomonas aeruginosa blaVIM-2) and two indeterminate results (Klebsiella pneumoniae blaOXA-181, Klebsiella pneumoniae blaOXA-48), reducing the sensitivity to 95.5% and 88.0%, respectively. The specificity for both tests remained unchanged. Modifying the CIM by increasing the vortexing time to 2 min and prolonging the disk incubation time to 6 h resulted in one less false negative result (Acinetobacter baumannii blaOXA-24/-40). Increasing the vortexing time to 2 min and prolonging the disk incubation time to 6 h for the mCIM resulted in one less negative result (Pseudomonas aeruginosa blaVIM-2) and the two indeterminate results became clearly positive (Klebsiella pneumoniae blaOXA-181, Klebsiella pneumoniae blaOXA-48). Tested against isolates obtained during enhanced surveillance for a local blaKPC-2 outbreak, the CIM offered superior test performance (sensitivity 97.9%, specificity 96.5%). When challenged against a larger set of blaOXA producing isolates the sensitivity of the CIM test decreased to 95.5%. Our study has some limitations that warrant discussion. Having been performed in the setting of a local outbreak 38% of the isolates tested were blaKPC harbouring Enterobacteriaceae. This is significant as isolates harbouring these resistance genes are very strong carbapenemase producers and are more readily detected than blaIMP, blaVIM and blaOXA harbouring isolates. Having noted this, Miller et al.,12Miller S.A. Hindler J.A. Chengcuenca A. Use of ancillary carbapenemase tests to improve specificity of phenotypic definitions for carbapenemase producing Enterobacteriaceae.J Clin Microbiol. 2017; 55: 1827-1836Crossref PubMed Scopus (11) Google Scholar the only study to demonstrate superiority of the mCIM test, evaluated a set of isolates of which over 50% harboured blaKPC. We attempted to correct this bias by obtaining isolates with different Ambler class carbapenemase genes from two reference laboratories, however blaIMP and blaVIM harbouring isolates and non-Enterobacteriaceae remained underrepresented. Additionally, we did not test any isolates containing the less abundant blaGES and blaSME carbapenemases. We found that increasing the vortexing and incubation time resulted in detection of weak carbapenemase harbouring isolates that initially tested negative, however as this methodology was not applied to all isolates we could not be certain that this would result in an overall improvement in test performance. Further studies should specifically address this issue. The CIM and mCIM are effective, simple and cost effective methods for the phenotypic detection of CPO. Both methods are easy to follow and utilise materials readily available in most diagnostic microbiology laboratories. In our hands the CIM test demonstrated superior sensitivity when directly compared to the CLSI recommended mCIM test and took less time to perform. We propose that further head-to-head studies are needed to compare the CIM and mCIM. The authors would like to thank Kerrie Stevens from Microbiological Diagnostic Unit, Public Health Laboratory, University of Melbourne at the Peter Doherty Institute for Infection and Immunity, and Jan Bell, University of Adelaide, for providing reference isolates for this study. Microbiological Diagnostic Unit performed the in-house resistance gene PCR testing for all of the 67 St Vincent's Pathology isolates.