Cefamandole nafate, cephapirin, and cephalothin were administered intravenously in crossover fashion to 12 volunteers, in dosages of 2 g every 6 h for 16 doses. Mean peak levels of cefamandole were approximately 50% higher than those of the other agents. The serum concentration curves appeared to decline bi-exponentially, suggesting that a two-compartment model was most applicable for pharmacokinetic analysis; accordingly, the t1/2 of cefamandole was significantly longer when the serum peak was omitted from the analysis (0.86 versus 0.73 h, P < 0.05). The half-lives of cephalothin and cephapirin, 0.34 and 0.36 h, respectively, were probably underestimates reflecting the inclusion of distribution-phase values in the calculation. Repeated dosing had no effect on the peak serum levels, half-life, serum clearance, or apparent volume ofdistribution with one exception: peak serum levels ofcephapirin were significantly lower after the sixteenth than after the first dose. Marked variations within a given subject were noted in the half-life and apparent volume of distribution ofcefamandole in several instances. Renal clearances ofcefamandole exhibited saturation kinetics similar to those of penicillin G.Cefamandole is a new cephalosporin with an extended spectrum against gram-negative bacilli (4). Recently, some aspects of its pharmacology have been reported; these suggest that it produces serum levels slightly higher than those of cephalothin, possesses a half-life similar to that of cephalothin, and displays the kinetics of a one-compartment model (5, 9, 12). The pharmacological behavior of cephapirin, while not extensively studied, appears to be similar to that of cephalothin; however, direct crossover comparisons of the two drugs are lacking (1,10).The present investigation was carried out as part of an examination ofthe comparative phlebitogenic potential of cefamandole, cephapirin, and cephalothin ini healthy volunteers (3). The design of the study afforded an opportunity to compare the pharmacokinetics of the three drugs after single and multiple doses, the extent of intra-subject variability in these parameters, and the applicability of a one-compartment model to the kinetic analysis. In addition, the findings emphasize the effect of the serum level of cefamandole on its apparent renal clearance.MATERIALS AND METHODS The design of the present study, which was carried out during an investigation of the phlebitogenic potential of these cephalosporins, has been described (3). Briefly, 12 healthy volunteers were given cefamandble nafate, sodium cephapirin, or buffered sodium cephalothin in a dosage of 2 g every 6 h for 4 days; the drugs were administered intravenously over 30 min using an infusion pump. Each subject, with two exceptions, received all three drugs in randomized crossover fashion with a 24-h lapse between courses of therapy. Specimens of blood were obtained at 0.5 (peak), 1.5, 2.5, 4.5, and 6 h after the beginning of the first and last (sixteenth) infusions of each agent. Corresponding urine specimens w...