The pharmacokinetics of cefotaxime, moxalactam, and ceftazidime were investigated in six human volunteers who received in a crossover fashion doses of 0.5, 1.0, and 2.0 g of each drug by a 5-min infusion. Doses of 1.0 g were repeated after the administration of probenecid. Serum and urine concentrations were assayed with an agar diffusion method. Serum concentrations of moxalactam exceeded those of ceftazidime at all times and were distinctly higher than those of cefotaxime. The normalized area under the concentration time curve (defined as the ratio of the area under the curve per dose) reflects this relationship: compared with cefotaxime the normalized area under the curve of moxalactam was 3 to 4 times higher, and that of ceftazidime was 2 to 3 times higher. By intra-individual comparisons, the area under the curve of moxalactam was 44% larger than that of ceftazidime. With increasing doses, cefotaxime exhibited a nonlinear increase of the area under the curve. The half-lives of moxalactam, ceftazidime, and cefotaxime were 2.34, 1.95, and 1.16 h, respectively. The volume of distribution averaged 0.20 +/- 0.03, 0.23 +/- 0.02, and 0.25 +/- 0.04 liters per kg, and the mean total body clearance was 84, 131, and 328 ml/min for moxalactam, ceftazidime, and cefotaxime, respectively. The 24-h urinary recovery was highest for moxalactam (75 +/- 4%) followed by ceftazidime (68 +/- 11%) and cefotaxime (53 +/- 6%). The influence of probenecid on serum concentrations, half-life, area under the curve, and clearance was most apparent with cefotaxime, whereas the pharmacokinetics of moxalactam and ceftazidime were only slightly affected. After the 0.5- and 2.0- g doses of cefotaxime, desacetyl-cefotaxime activity (determined by high-pressure liquid chromatography) reached a peak of 2.7 and 9.9 mug/ml and declined with a half-life of 1.9 and 1.4 h. The ratio of the R(-) and S(-) epimers of moxalactam, which could be differentiated by high-pressure liquid chromatography, fell rapidly from 0.81 at 0.17 h to 0.5 at 5 h, indicating the presence of twice as much of the microbiologically less active S(-) epimer. From a pharmacokinetic standpoint it appears reasonable to conclude that moxalactam and possibly ceftazidime could be administered twice daily and that cefotaxime could be administered three or even four times daily.