Empirical use of β-lactam/β-lactamase inhibitor combinations does not increase mortality compared with cloxacillin and cefazolin in methicillin-susceptible Staphylococcus aureus bacteraemia: a propensity-weighted cohort study

Abstract

Abstract Objectives To evaluate the effectiveness of empirical therapy with β-lactam/β-lactamase inhibitor combinations (BL/BLICs) for MSSA bacteraemia. Methods We conducted a post hoc analysis of all adult patients with MSSA bacteraemia who were hospitalized at a Spanish university hospital between 2013 and 2018. We compared 30 day mortality among patients receiving initial therapy with BL/BLICs (de-escalated to cloxacillin or cefazolin within 96 h) versus cloxacillin or cefazolin, using propensity score analysis with the inverse probability of treatment weighting (IPTW) method. Results We evaluated 373 patients with MSSA bacteraemia. Among them, 198 patients met the eligibility criteria, including 127 patients in the BL/BLICs group and 71 patients in the cloxacillin/cefazolin group. Patients in the BL/BLICs group had a higher Charlson comorbidity index (median, 2 [IQR, 1–4.5] versus 2 [IQR, 0–4]); an increased proportion of high-risk sources (i.e. endocarditis, respiratory sources and bacteraemia of unknown origin [34.6% versus 18.3%]); and an earlier start of antibiotic treatment (median, 0 days [IQR, 0–0] versus 1 day [IQR, 1–2]). Thirty day mortality did not significantly differ between the BL/BLICs and the cloxacillin/cefazolin groups (27 patients [21.3%] versus 13 patients [18.3%]; IPTW-adjusted OR = 0.53 [95% CI, 0.18–1.51]). For secondary outcomes, 7 day mortality and 90 day relapse were not statistically different between study groups (8.7% versus 5.6% [P = 0.62] and 6.2% versus 3.8% [P = 0.81], respectively). Conclusions BL/BLICs might be an effective empirical treatment for MSSA bacteraemia when de-escalated to cloxacillin or cefazolin within 96 h from the index blood culture.