We have tested the benefit of prophylaxis by intravenous systemic vancomycin among 59 neutropenic patients in a randomized trial. Vancomycin was delivered on day zero of chemotherapy until the resolution of neutropenia in the prophylactic group (vanco+). Empiric antibiotic therapy (piperacillin, ofloxacine) was identical for all patients. The number of days with fever > 38.5 degrees C was significantly higher in the control (vanco-) group than in the vanco+ group (7.4 vs. 3.7, p < 0.02). Zero gram-positive infections occurred in the vanco+ group versus 9 in the vanco- group (p < 0.002). The mean number of days of empiric antibiotic therapy was reduced in the vanco+ group (11.3 vs. 16.3, p = 0.12). However, no benefit was noted between the two groups with regard to mortality or the severity of the infections. The selection of resistant microorganisms after systemic treatment with vancomycin is of potential risk. Such a prophylactic antibiotic regimen does not seem to be justified.