Performance of polymyxin B Etest in a setting of high prevalence of KPC-producing Klebsiella pneumoniae

Abstract

Objectives: Polymyxin resistance has been increasing in many regions, and appropriate determination ofpolymyxin susceptibility is now a major challenge worldwide. Many clinical laboratories rely on gradientdiffusion methods to assess polymyxin susceptibility, although broth microdilution (BMD) is the onlymethod currently recommended by the CLSI and EUCAST. The aim of this study was to assess theperformance of the polymyxin B (PMB) Etest in a setting with a high prevalence of KPC-producingKlebsiella pneumoniae (KPC-KP).Methods: A commercial Etest susceptibility testing method was evaluated and compared with thereference BMD method, considering isolates with a minimum inhibitory concentration (MIC) 2 mg/L forPMB as susceptible to this drug. A total of 310 clinical KPC-KP isolates were evaluated.Results: Susceptibility was significantly higher by Etest compared with BMD (82.6% vs. 75.8%). The MIC50,MIC90and modal MICs for PMB were 0.25, 32 and 0.25 mg/L (27.1%) by BMD and 0.5, 16 and 0.5 mg/L(49.7%) by Etest, respectively. Although categorical agreement was 90.0%, there was poor essentialagreement (50.6%). A high rate (34.7%) of very major errors (VMEs) and a relatively low rate (2.1%) ofmajor errors were found.Conclusion: The considerable number of resistant isolates in this study allowed an accurate estimation ofVME rates and, consequently, a more comprehensive assessment of susceptibility testing for polymyxins.Etest did not meet fully the acceptance criteria for US FDA requirements. These data do not support theuse of this commercial method for determining PMB MICs in carbapenem-resistant Enterobacterales populations.