The rising prevalence of antibiotic resistance in foodborne Staphylococcus aureus underscores the critical importance of accurate and rapid antimicrobial susceptibility testing in guiding effective clinical therapy.

A comparative evaluation of the diagnostic accuracy of E-test and whole-genome sequencing (WGS) was performed for the detection of clinically significant antimicrobial resistance in foodborne Staphylococcus aureus.

Seventy-two foodborne Staphylococcus aureus were collected from the foodborne disease surveillance programs conducted by the Beijing Chaoyang District Center for Disease Prevention and Control and Xinjiang Uygur Autonomous Region Center for Disease Prevention and Control between March and November 2025, then tested in parallel against eight key antimicrobial agents (oxacillin, vancomycin, daptomycin, linezolid, rifampicin, ciprofloxacin, levofloxacin, and moxifloxacin) using both the E-test and WGS. The broth microdilution method (BMD), performed according to CLSI guideline M07, served as the reference standard for evaluating the diagnostic performance of both methods.

When evaluated against the BMD reference method, high sensitivity (80.0%-100.0%), specificity (94.4%-100.0%), and inter-method agreement (Kappa=0.86-1.00) were observed for the E-test in detecting resistance to oxacillin, linezolid, rifampicin, levofloxacin, and moxifloxacin. Specificity was absolute (100.0%) for both daptomycin and vancomycin. A notable exception was ciprofloxacin, for which the E-test exhibited low sensitivity (57.1%) alongside 100.0% specificity and substantial agreement (Kappa=0.70). Genotypic analysis by WGS accurately identified major resistance determinants (mecA, 23S rRNA G2576T, rpoB H481N), which correlated highly with the phenotypic resistance profile. The predictive performance of WGS was excellent for ciprofloxacin (sensitivity: 100.0%; specificity: 96.9%; Kappa=0.85). However, for levofloxacin and moxifloxacin, predicted phenotypic resistance, while perfectly sensitive (100.0%), showed only moderate specificity (88.4%) and fair agreement (Kappa=0.38) with the BMD reference.

Notwithstanding its reliable performance against the BMD standard for key antibiotics, the E-test is compromised by specific shortcomings, namely: a tendency to overestimate the minimal inhibitory concentration (MIC) of daptomycin, potential failure to detect intermediate resistance to vancomycin, and inadequate sensitivity for identifying low-level ciprofloxacin resistance mediated by the parC S80F mutation. While the performance of WGS is highly concordant for predicting resistance to oxacillin, linezolid, rifampicin, daptomycin and vancomycin via accurate genotyping, it is imperative to integrate and interpret mutations from multiple loci to achieve reliable phenotypic prediction for fluoroquinolones (ciprofloxacin, levofloxacin, moxifloxacin).