The purpose of this investigation was to examine the impact of antimicrobial regimens administered for hospital-acquired methicillin-resistant Staphylococcus aureus (MRSA) bacteraemia on the all-cause, 14-day mortality. We retrospectively examined the characteristics of the most effective empiric antimicrobial therapy in 87 consecutive patients, hospitalised at a single institution between April 2003 and March 2008, who presented with clinically and microbiologically confirmed MRSA bacteraemia. The all-cause mortality was measured 14 days after the diagnosis was made. The administration of an effective antimicrobial against MRSA <48 h after the collection of blood cultures was the single, significant predictor of survival (odds ratio 3.85; 95% confidence interval 1.37-10.80; p = 0.01). The survival of patients treated with vancomycin versus other antimicrobial agents was similar. Among subgroups treated with vancomycin, the lowest mortality (6%) was observed among patients treated (a) within 48 h after the collection of blood cultures and (b) with doses sufficient to keep the blood concentrations in the area under the 0-24 h curve >400 μg h/ml (≥2.0 g/day). The empiric administration of antimicrobials effective against MRSA bacteraemia within 48 h after the collection of blood cultures increased the 14-day survival. If vancomycin is chosen, ≥2.0 g/day should be administered, starting within 48 h.