OBJECTIVES No randomized controlled trials have evaluated the comparative outcomes of cefazolin versus nafcillin for methicillin-susceptible Staphylococcus aureus (MSSA) bacteraemia.

METHODS A prospective observational cohort study including all S. aureus bacteraemia was conducted at 10 hospitals. Patients (≥15 years) with MSSA bacteraemia who received cefazolin or nafcillin as definitive antibiotics were included. The rates of treatment failure (premature discontinuation of antibiotics because of adverse effects, switching of antibiotics because of clinical failure, all-cause mortality within 1 month, or recurrence) were compared between the cefazolin and nafcillin groups. Propensity score matching analyses were performed to balance the factors influencing the selection of antibiotics.

RESULTS Among the 242 included cases, the bones and joints (36.8%) were the most common sites of infection and 60.7% of the patients had sepsis. The overall treatment failure rate was 43.8% (106/242). All-cause mortality within 1 month was 6.2% (15/242). After propensity score matching, the treatment failure rate of cefazolin was lower than that of nafcillin (30.4% (24/79) vs. 49.4% (39/79), p 0.015) because of a higher rate of discontinuation caused by adverse events. When the data were limited to patients with sepsis, the treatment failure rates of both groups were not significantly different. Approximately 22% (24/110) of MSSA isolates exhibited a cefazolin-inoculum effect (CIE) that had significant impact on the failure rate and mortality of the cefazolin group.

CONCLUSIONS Cefazolin might be recommended as an adequate and better-tolerated treatment for MSSA bacteraemia in the absence of CIE.