RESUMEN
With multidrug-resistant (MDR) Enterobacterales on the rise, a nontoxic antimicrobial agent with a unique mechanism of action such as fosfomycin seems attractive. However, establishing accurate fosfomycin susceptibility testing for non-Escherichia coli isolates in a clinical microbiology laboratory remains problematic. We evaluated fosfomycin susceptibility by multiple methods with 96 KPC-producing clinical isolates of multiple strains and species collected at a single center between 2008 and 2016. In addition, we assessed the presence of fosfomycin resistance genes from whole-genome sequencing (WGS) data using NCBI's AMRFinder and custom HMM search. Susceptibility testing was performed using a glucose-6-phosphate-supplemented fosfomycin Etest and Kirby-Bauer disk diffusion (DD) assays, and the results were compared to those obtained by agar dilution. Clinical Laboratory and Standards Institute (CLSI) breakpoints for E. coli were applied for interpretation. Overall, 63% (60/96) of isolates were susceptible by Etest, 70% (67/96) by DD, and 88% (84/96) by agar dilution. fosA was detected in 80% (70/88) of previously sequenced isolates, with species-specific associations and alleles, and fosA-positive isolates were associated with higher MIC distributions. Disk potentiation testing was performed using sodium phosphonoformate to inhibit fosA and showed significant increases in the zone diameter of DD testing for isolates that were fosA positive compared to those that were fosA negative. The addition of sodium phosphonoformate (PPF) corrected 10/14 (71%) major errors in categorical agreement with agar dilution. Our results indicate that fosA influences the inaccuracy of susceptibility testing by methods readily available in a clinical laboratory compared to agar dilution. Further research is needed to determine the impact of fosA on clinical outcomes.