Determining antimicrobial susceptibility in Salmonella enterica serovar Typhimurium through whole genome sequencing: A comparison against multiple phenotypic susceptibility testing methods

Mather A.. (2019)

BMC Microbiology, 19, 148

Background: UK public health organisations perform routine antimicrobial susceptibility tests (ASTs) to characterise the
potential for antimicrobial resistance in Salmonella enterica serovars. Genetic determinants of these resistance mechanisms
are detectable by whole genome sequencing (WGS), however the viability of WGS-based genotyping as an alternative
resistance screening tool remains uncertain. We compared WGS-based genotyping, disk diffusion and agar dilution to the
broth microdilution reference AST for 102 Salmonella enterica serovar Typhimurium (S. Typhimurium) isolates across 11
antimicrobial compounds.
Results: Genotyping concordance, interpreted using epidemiological cut-offs (ECOFFs), was 89.8% (1007/1122) with
0.83 sensitivity and 0.96 specificity. For seven antimicrobials interpreted using Salmonella clinical breakpoints,
genotyping produced 0.84 sensitivity and 0.88 specificity. Although less accurate than disk diffusion (0.94 sensitivity,
0.93 specificity) and agar dilution (0.83 sensitivity, 0.98 specificity), genotyping performance improved to 0.89 sensitivity
and 0.97 specificity when two antimicrobials with relatively high very major error rates were excluded (streptomycin
and sulfamethoxazole).
Conclusions: An 89.8% concordance from WGS-based AST predictions using ECOFF interpretations suggest that WGS
would serve as an effective screening tool for the tracking of antimicrobial resistance mechanisms in S. Typhimurium. For
use as a standalone clinical diagnostic screen, further work is required to reduce the error rates for specific antimicrobials.

BMC Microbiology, 19, 148

View Publication