Pretreatment of patients with diazepam has been reported to reduce the plasma half-life of epidurally administered bupivacaine. The concentrations of bupivacaine used to estimate its plasma half-life may be influenced by input into the vascular system (rate and extent of absorption from the epidural space) and disposition (distribution and elimination). We conducted a study in Rhesus monkeys in which the input function was known and the influence of intravenous diazepam on the disposition of intravenous bupivacaine could be delineated. Results indicated no differences in the primary pharmacokinetic parameters (i.e., total clearance and volume of distribution) for the disposition of bupivacaine. To determine whether the apparent disparity between these results and a previous study might be due to a diazepam-induced alteration in the input function, the influence of intravenous diazepam or saline on the mean absorption time of epidurally administered bupivacaine was also studied in monkeys. Epidural bupivacaine was completely absorbed whether the animals received intravenous saline or intravenous diazepam. No significant differences in pharmacokinetic parameters were found between animals pretreated with intravenous saline or intravenous diazepam and receiving an epidural injection of bupivacaine. The mean residence time of bupivacaine in the body (P less than 0.01) and plasma elimination half-life (P less than 0.005) were significantly longer after epidural administration than after intravenous administration. Intravenous diazepam does not alter the pharmacokinetics of intravenously or epidurally administered bupivacaine.