Variable performance of different commercial systems for testing carbapenem susceptibility of KPC carbapenemase-producing Escherichia coli

Abstract

Objectives: The aim was to evaluate different methods for testing carbapenem susceptibility of Escherichia coli producing KPC-type carbapenemase.

Methods: Susceptibility to imipenem, meropenem and ertapenem was assayed using the reference broth microdilution method and several commercial methods (Vitek2, MicroScan, Etest, MIC Test Strip) starting from the same bacterial suspension. Susceptibility to imipenem and meropenem was also tested by Sensititre and disc diffusion (Bio-Rad). Results were interpreted according to EUCAST clinical breakpoints. Essential agreement (EA), category agreement (CA) and error rates were calculated as described by the International Organization for Standardization (ISO) guidelines and also considering the new EUCAST definitions. Genotypic diversity of isolates was evaluated with a RAPD profiling protocol.

Results: Of 54 KPC-positive E. coli isolates, 5.6%, 7.4% and 0% were susceptible standard dosing regimen (S), 55.6%, 72.2% and 0% susceptible increased exposure (I), and 38.9%, 20.4% and 100.0% resistant (R) to imipenem, meropenem and ertapenem, respectively, using the reference broth microdilution method. CA lower than 90% were observed with all systems for imipenem and meropenem using both the ISO and the modified EUCAST criteria. With ertapenem, CA >90% was observed with all methods except Vitek2. RAPD profiling revealed a remarkable genotypic diversity of the isolates, supporting that results were not biased by an oligoclonal nature of the collection.

Conclusions: Commercial methods can be unreliable for testing susceptibility to carbapenems of KPC-producing E. coli. Susceptibility should be confirmed by reference broth microdilution.

Keywords: Carbapenemase; Category agreement; Discrepancy; Essential agreement; MIC.