OBJECTIVES: In blood culture-proven pneumococcal infections, streamlining empirical therapy to monotherapy with a penicillin is preferred in order to reduce the use of broad-spectrum antibiotics. However, adherence to this international recommendation is poor, and curiously it is unclear whether antibiotic streamlining may be harmful to individual patients. We investigated whether streamlining in bacteraemic pneumococcal infections is associated with mortality. METHODS: Adults admitted to two Dutch hospitals between 2001 and 2011 with bacteraemic pneumococcal infections were retrospectively included. Detailed clinical data on patient characteristics, comorbidities and severity and outcome of disease were obtained in addition to data on antibiotic treatment. Those eligible for streamlining were selected for further analyses. RESULTS: In the 45.8% of cases (126 of 275) where antibiotic treatment was streamlined, a lower mortality rate was observed (6.3% versus 15.4%, P = 0.021). The decision to streamline was only marginally explained by the 38 determinants accounted for. After correction for potential confounders, the OR for death while streamlining was 0.45 (95% CI: 0.18-1.11, P = 0.082) in all cases and 0.35 (95% CI: 0.12-0.99, P = 0.048) specifically in pneumonia cases. CONCLUSIONS: Our results suggest that streamlining in eligible pneumococcal bacteraemia cases is safe, irrespective of patient characteristics, severity of disease or empirical treatment regimen.