Trimethoprim-sulfamethoxazole compared with vancomycin for the treatment of Staphylococcus aureus infection.

To compare trimethoprim-sulfamethoxazole (TMP-SMZ) and vancomycin regarding efficacy and safety in the therapy of serious Staphylococcus aureus infections. Randomized, double-blind comparative trial. A tertiary-care hospital. One hundred and one intravenous drug users hospitalized with S. aureus infection. Cure and failure rates; blood and wound cultures; minimum inhibitory and bactericidal concentrations; serum inhibitory and bactericidal titers; temperature; leukocyte count; durations of treatment and hospitalization; and toxicity. Of 228 intravenous drug users, 101 had S. aureus infection and were included in the efficacy analysis (43 received TMP-SMZ and 58 received vancomycin). Methicillin-resistant S. aureus (MRSA) accounted for 47% of S. aureus isolates, and 65% of patients were bacteremic. Infections were cured in 57 of 58 vancomycin recipients and in 37 of 43 TMP-SMZ recipients (P less than 0.02). Failure occurred mostly in patients with tricuspid valve endocarditis and only in those with infection caused by methicillin-sensitive S. aureus (MSSA). The mean duration of bacteremia was 6.7 days in TMP-SMZ recipients and 4.3 days in vancomycin recipients. Among 222 subjects hospitalized for at least 24 hours, toxicity rates were similar for TMP-SMZ (23%) and vancomycin (20%) recipients; nausea and vomiting were associated with TMP-SMZ and inflammation at the intravenous site was associated with vancomycin. Forty-four percent of TMP-SMZ recipients and 29% of vancomycin recipients experienced side effects in the efficacy cohort (P greater than 0.05). Vancomycin is superior to TMP-SMZ in efficacy and safety when treating intravenous drug users who have staphylococcal infections. However, all treatment failures occurred in patients with MSSA infection at any site. Therefore, TMP-SMZ may be considered as an alternative to vancomycin in selected cases of MRSA infection.