Comparative Evaluation of Four Phenotypic Methods for Detection of Class A and B Carbapenemase-Producing Enterobacteriaceae in China

The objective of this study was to evaluate the performance of four phenotypic methods in the detection of carbapenemase-producing Enterobacteriaceae (CPE) in China. We evaluated the performance of four carbapenemase detection methods, the modified Hodge test (MHT), the Carba NP test, the meropenem hydrolysis assay (MHA) with 1- and 2-h incubation, and the modified carbapenem inactivation method (mCIM) with meropenem, imipenem, and ertapenem, on 342 carbapenem-resistant Enterobacteriaceae isolates (CRE) in China. PCR was used as the gold standard. The 2-h-incubation MHA performed the best in carbapenemase detection (overall sensitivity, specificity, positive predictive value, and negative predictive value all 100%). Second was the Carba NP test, with a sensitivity of 99.6%. The 1-h-incubation MHA performed poorly in Klebsiella pneumoniae carbapenemase (KPC) detection (sensitivity, 71.3%). For mCIM, the best performance was observed with the meropenem disk. The MHT exhibited the worst performance, with a specificity of 88.8%. All assays except 1-h-incubation MHA, which failed to identify 68 KPC-2s, had a sensitivity of >98% in the detection of 172 KPCs. Likewise, all assays had a sensitivity of >95% in the detection of 70 class B carbapenemases, except for MHT (82.9%). The 2-h-incubation MHA significantly improved the accuracy in CPE detection compared with that for 1-h incubation and performed the best in the detection of class A and B carbapenemases. Our findings suggest that the MHA is the most practical assay for carbapenemase detection. For those who cannot afford the associated equipment, both the Carba NP test and mCIM are good alternatives with regard to the practical requirements of time and cost.