Cefepime (1 g), ceftazidime (1 g), and cloxacillin (2 g) were administered intravenously to 10 volunteers each. After infusion of a single dose over 30 min, blood samples were obtained at 0.5, 1, 2, 3, 4, 5, 6, 8, 10, and 12 h (for ceftazidime at 0.5 and 4 h) after dosing. Drug levels were determined by the bioassay method. Serum bactericidal activity against five clinical isolates of cloxacillin-susceptible Staphylococcus aureus were determined by the microdilution method according to the National Committee for Clinical Laboratory Standards guidelines. The mean peak serum level was 76.88 ± 24.71 mg/L for cefepime, 42.8 ± 15.98 mL/L for ceftazidime, and 92.81 ± 24.7 mg/L for cloxacillin. Concentrations of cefepime were detected during the whole testing period (mean trough level, 1.43 ± 0.9 mg/L at 12 h), whereas concentrations of cloxacillin were measurable up to 5 h after administration (mean trough level, 0.90 ± 0.97 mg/L). The mean peak reciprocal bactericidal titers were 29.41 for cefepime, 5.6 for ceftazidime, and 377 for cloxacillin. Effective bactericidal titers were detected as long as 5 h for cefepime (~40% of the dosing interval) and 3 h for cloxacillin (at least 50% of the dosing interval). For ceftazidime, serum bactericidal activity was markedly lower compared with that of cefepime. Although cefepime has demonstrated an improved antistaphylococcal bactericidal activity compared with ceftazidime, it was somewhat lower than that of cloxacillin. Copyright (C) 1999 Elsevier Science Inc.