Abstract Objectives Standard once-daily dosing of ceftriaxone may not lead to adequate antibiotic exposure in all cases of Staphylococcus aureus bacteraemia (SAB). Therefore, we compared clinical effectiveness of empirical antibiotic treatment with flucloxacillin, cefuroxime and ceftriaxone in adult patients with MSSA bacteraemia Methods We analysed data from the Improved Diagnostic Strategies in Staphylococcus aureus bacteraemia (IDISA) study, a multicentre prospective cohort study of adult patients with MSSA bacteraemia. Duration of bacteraemia and 30 day SAB-related mortality were compared between the three groups using multivariable mixed-effects Cox regression analyses. Results In total, 268 patients with MSSA bacteraemia were included in the analyses. Median duration of empirical antibiotic therapy was 3 (IQR 2–3) days in the total study population. Median duration of bacteraemia was 1.0 (IQR 1.0–3.0) day in the flucloxacillin, cefuroxime and ceftriaxone groups. In multivariable analyses, neither cefuroxime nor ceftriaxone were associated with increased duration of bacteraemia (respectively HR 1.08, 95% CI 0.73–1.60 and HR 1.22, 95% CI 0.88–1.71) compared with flucloxacillin, nor were the cephalosporins associated with higher 30 day SAB-related mortality (respectively, subdistribution HR (sHR) 1.37, 95% CI 0.42–4.52 and sHR 1.93, 95% CI 0.67–5.60). Conclusions In this study, we could not demonstrate a difference in duration of bacteraemia and 30 day SAB-related mortality between patients with SAB empirically treated with flucloxacillin, cefuroxime or ceftriaxone. Since sample size was limited, it is possible the study was underpowered to find a clinically relevant effect.