A prospective, randomized, and double-blind study comparing teicoplanin with vancomycin in the initial management of febrile neutropenic patients was conducted. Teicoplanin was administered at 6 mg per kg of body weight every 24 h (q24h) intravenously (i.v.) after initial loading at 6 mg/kg q12h for three doses. Vancomycin was administered at 15 mg/kg q12h i.v. Patients also received piperacillin (3 g q4h i.v.) and tobramycin (1.5 to 2.0 mg/kg q8h i.v.). Of 53 patients enrolled, 50 were judged to be evaluable. Among these, 25 received teicoplanin and 25 received vancomycin. At enrollment, both groups were comparable in age, sex, renal function, underlying hematologic condition, and concurrent therapy. Both groups had similar sites of infection and microbial pathogens. Empirical antimicrobial therapy resulted in the cure of or improvement in 23 (92%) teicoplanin patients and 21 (84%) vancomycin patients (P = 0.67). Failures occurred with two vancomycin patients but no teicoplanin patients. Clinical response was indeterminate for two patients in each group. Adverse reactions occurred significantly more often in the vancomycin group than in the teicoplanin group (P = 0.01), and these reactions required the termination of the study regimens of 6 vancomycin versus 0 teicoplanin patients (P = 0.02). Nephrotoxicity was observed more frequently in the vancomycin group (10 versus 2 patients; P = 0.02). Subgroup analysis revealed a significant deterioration of renal function when vancomycin and cyclosporin A, but not teicoplanin and cyclosporin A, were used concurrently (P = 0.02). Among patients who received vancomycin and amphotericin B or teicoplanin and amphotericin B concurrently, deterioration in renal function was equivalent in both groups. Teicoplanin in the dosage employed was tolerated better than vancomycin in the empirical treatment of fever and neutropenia in our patient population.