2162. Comparison of Plazomicin Disk Diffusion vs. Gradient Diffusion Susceptibility Testing Results Against Drug-Resistant Clinical Enterobacteriaceae Isolates

Abstract

BackgroundPlazomicin, a novel aminoglycoside, is active against carbapenem-resistant Enterobacteriaceae (EB) and is not inhibited by most aminoglycoside modifying enzymes that affect gentamicin and tobramycin. We investigated the activity of plazomicin against resistant EB clinical isolates and compared disk diffusion (DD) vs. gradient diffusion (GD) results.MethodsEB isolates that were carbapenem resistant and/or resistant to both gentamicin and tobramycin were retrieved from the UW Health clinical isolate repository. Each isolate was tested against plazomicin using both DD (MAST Group Ltd. Plazomicin disk 30 µg) and GD (Liofilchem Plazomicin MIC Test Strip 0.16–256 µg/mL) methods according to manufacturer instructions and using FDA clinical breakpoints for interpretation.Results51 isolates were tested: 21 E. coli, 9 P. mirabilis, 7 E. cloacae, 6 K. pneumoniae, 3 K. oxytoca, 3 S. marcescens, 1 K. aerogenes, and 1 C. freundii. Specimen sites included: 29 blood, 8 urine, 8 soft tissue or bone, 5 intra-abdominal, and 1 sputum. Previous phenotypic AST results demonstrated 19 (37%) were CRE, of which 5 were also gentamicin and tobramycin resistant, and 32 (63%) were tobramycin and gentamicin resistant but carbapenem susceptible. Plazomicin zone diameters and minimal inhibitory concentrations (MIC) for all isolates are shown in the figure (data jittered to show frequency). There was a significant correlation between increased MIC and smaller zone diameters (Pearson coefficient −0.443, P = 0.001). However, while all 51 isolates were susceptible by DD breakpoints, only 46 (92%) were susceptible by GD breakpoints. All 5 discordant results were P. mirabilis which had an MIC of 4 µg/mL (intermediate) but zone diameters of 20–21 mm (susceptible).ConclusionConcordance between plazomicin DD and GD susceptibility was only 92%. All 5 discordant results were P. mirabilis. Surveillance studies demonstrate >80% of P. mirabilis have MIC of 2–4 mg/L. Given the DD breakpoint is 16 mm, our data suggest DD was overly active in our sample set. Comparison of DD and GD to reference broth microdilution against a larger set of isolates is warranted to determine which method is optimal; however, our data suggest DD may result in categorical errors for P. mirabilis. DisclosuresAlexander Lepak, MD, Paratek Pharmaceuticals: Research Grant; Tetraphase Pharmaceuticals: Research Grant.