Comparison of clinical methods for detecting carbapenem-resistant Enterobacteriaceae

We evaluated detection of carbapenem-resistant Enterobacteriaceae (CRE) by routine minimal inhibitory concentration (MIC) testing, polymerase chain reaction (PCR) using Xpert® Carba-R assay, hydrolysis of ertapenem and imipenem detected by matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS), and hydrolysis by colorimetry using the EPI-CRE assay. Ninety-six Enterobacteriaceae isolates possessing carbapenemase genes and 29 carbapenem-susceptible Enterobacteriaceae were available for testing. The sensitivity and specificity of each assay was determined. For sensitivity, discrepant results from each assay compared to reference genotype were arbitrated with MIC and/ or PCR testing to assess loss of plasmid-mediated resistance. Xpert Carba-R was evaluated for resistance genes in their FDA claim (i.e., the genes encoding KPC; NDM; VIM; IMP; and OXA-48). The sensitivity for the assays was: MIC (N=96), 96.8%, (discrepant analysis to 98.9% [2 cured plasmids]); Xpert Carba-R (N=85), 97.6% (discrepant analysis to 100% % [2 cured plasmids]); EPI-CRE (N=96), 91.7% (discrepant analysis to 91.7%); MALDI-TOF MS (N=96) ertapenem hydrolysis using Compass software for interpretation (2 h incubation), 92.7% (discrepant analysis to 94.7% % [2 cured plasmids]); MALDI-TOF MS (N=96) imipenem hydrolysis (1 h incubation), 97.9% (discrepant analysis to 98.9% % [1 cured plasmid]). The specificity for each assay was: MIC (N=29), 100%; EPI-CRE (N=29), 96.6%; MALDI-TOF MS ertapenem hydrolysis (N=29), 100%; MALDI-TOF MS imipenem hydrolysis (N=29), 96.6%. All isolates tested to ensure specificity demonstrated susceptible MIC results for carbapenems and did not qualify for testing with Xpert Carba-R. No single assay detected all of the known genetic markers of carbapenem hydrolysis.