A retrospective evaluation of recently introduced EUCAST rapid antimicrobial susceptibility testing (RAST) breakpoints for Salmonella enterica in a high burden setting

Le résumé fourni par la source

PURPOSE: In July 2024, EUCAST released zone diameter breakpoints for Salmonella enterica isolates for direct antimicrobial susceptibility testing from positively flagged automated blood culture bottles (+aBCBs). In this study, we evaluated the diagnostic accuracy of these breakpoints on previously unreported Salmonella enterica isolates in a setting where the EUCAST rapid AST (RAST) method has been implemented since 2020. METHODS: The study was done on Salmonella enterica serotype Typhi and Paratyphi A received between January 2021-June 2024. Interpretive category in RAST methodology, at 8- and 16-20 h was compared with the interpretive category obtained by VITEK®-2 using EUCAST breakpoints after full incubation, the latter being considered as the comparator method. Diagnostic accuracy for cefotaxime, piperacillin-tazobactam, meropenem, imipenem and trimethoprim-sulphamethoxazole were compared in terms of categorical agreement (CA), major error (ME), very major error (VME) and positive predictive values (PPV). RESULTS: Of 149 isolates, 121 (81.2%) and 28 (18.8%) were serotypes Typhi and Paratyphi A, respectively. Of 422 and 592 drug-bug combinations evaluated at 8- and 16-20 h, the overall rates for CA, ME, VME & PPV were 91.9% (95%CI: 89-94), 1.6% (95%CI: 1-3), 61.4% (95%CI: 47-74), 92.7% (95%CI: 90-95) and 90% (95%CI: 87-92), 2.5% (95%CI: 1-4), 66.1% (95%CI: 53-77), 91.6% (95%CI: 89-94), respectively. Poor agreements and PPV were noted for piperacillin-tazobactam at both time points and cefotaxime at 16-20 h. CONCLUSION: EUCAST RAST method for Salmonella enterica was found to be diagnostically accurate in our study. If implemented in routine clinical laboratories, it has the potential to provide an early clinical report within 24 h of positive blood culture. This may improve patient outcomes and promote antimicrobial de-escalation, especially in resource-limited settings.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.