ABSTRACT: Our objective was to examine whether empirical antimicrobial therapy (EAT) against methicillin-susceptible Staphylococcus aureus bacteremia (MS-SAB) with piperacillin-tazobactam (TZP), cefuroxime or combination therapy with one of these was differentially associated with 7-, 30-, and 90- day all-cause mortality or MS-SAB relapse. A multicenter retrospective cohort study of adults with MS-SAB from 2009 through 2018 was used, and 7-, 30-, 90-day mortality and relapse within 90 days were assessed and expressed as hazard ratio (HR) with a 95% confidence interval (95% CI) using Cox proportional hazard regression analysis. Matching of the two monotherapy groups was performed using propensity score matching. In total, 1158 MS-SAB cases were included and received one of three EAT regimens: TZP (n = 429), cefuroxime (n = 337), or TZP or cefuroxime with one or more additional effective antimicrobial (n = 392). The overall 30-day mortality was 28.0% (25.5 to 30.3%). After adjustment and matching, there was no significant difference in 7-, 30-, or 90-day mortality between the therapy groups. The matched HR of death was 0.81 (95% CI, 0.38 to 1.76) at 7 days, 0.82 (95% CI, 0.47 to 1.46) at 30 days, and 0.81 (95% CI, 0.50 to 1.32) at 90 days for TZP compared with cefuroxime. Adjusted HR of 90-day relapse was insignificant between the three therapy groups: TZP: 1.55 (95% CI, 0.54 to 4.43); combination therapy: 1.73 (95% CI, 0.62 to 4.80) compared to cefuroxime. There was no significant difference in 7-, 30-, or 90-day mortality or relapse between MS-SAB patients treated with empirical TZP or cefuroxime after adjustment and matching of covariables. IMPORTANCE This multicenter retrospective matched cohort study evaluated the effect of empirical antimicrobial therapy on the clinical outcome of methicillin-susceptible Staphylococcus aureus bacteremia (MS-SAB) in >1100 adult patients. To the best of our knowledge, this is the largest study to date evaluating the effect of empirical treatment on the MS-SAB outcome. Importantly, the study found no significant difference in either short- or long-term mortality nor relapse between patients with MS-SAB receiving empirical treatment with cefuroxime or piperacillin-tazobactam. As such, this study provides crucial contemporary data supporting the widespread clinical practice of initiating empirical antimicrobial therapy of sepsis with β-lactam-β-lactamase-inhibitor.
