Evaluation of the MicroScan Colistin Well and Gradient Diffusion Strips for Colistin Susceptibility Testing in Enterobacteriaceae

Abstract

Many laboratories are unable to perform colistin susceptibility testing. Diffusion-based antimicrobial susceptibility testing methods are not recommended, and not all laboratories have the capacity to perform broth microdilution (BMD). Using a multistep tiered approach, we investigated whether the adapted use of the MicroScan colistin well (4 μg/ml) could enhance laboratory capacity for the detection and subsequent molecular characterization of colistin-resistant Enterobacteriaceae For the MicroScan colistin well, categorical agreement with BMD was 92.7%, and the very major error rate was 10.7%. For gradient diffusion strips, the categorical agreement was 86.4%, and the very major error rate was 53.6%. The MicroScan colistin well detected all isolates carrying mcr-1 or mcr-2 genes (n = 16), but gradient diffusion strips identified an MIC of ≥4 for colistin for only 62.5% of these isolates. A 6-month prospective phenotypic and genotypic study performed at a single clinical microbiology laboratory assessed isolates growing in the MicroScan colistin well for concordance. While 37 of 39 isolates growing in the MicroScan colistin well displayed a colistin MIC of ≥4 by BMD, all were determined to be negative for the mcr-1 and mcr-2 genes by PCR. A retrospective review of all Escherichia coli, Klebsiella spp., and Enterobacter spp. tested by MicroScan at this laboratory in 2016 identified 260 of 7,894 (3.3%) isolates that grew in the MicroScan colistin well. Based on the data presented, clinical and public health laboratories could use the MicroScan colistin well as a first screen for the detection of isolates displaying elevated colistin MICs, which could then undergo further characterization.

Keywords: AST; MicroScan; antimicrobial susceptibility testing; colistin; polymyxin.