7 patients with chronic renal insufficiency (4 of whom were on maintenance hemodialysis) and 6 healthy controls received single 15-mg oral doses of oxazepam for 7 consecutive days. Multiple venous blood samples drawn during the 72 h after the final dose were analyzed for concentrations of oxazepam and its glucuronide metabolite, and the extent of oxazepam protein binding. In hemodialysis patients, the final dose was taken just prior to dialysis; arterial (inflow) and venous (outflow) plasma was analyzed as well as dialysate. Mean steady-state plasma concentrations (C<sub>ss</sub>) of total (free and bound) oxazepam in nondialysis renal failure patients (133 ng/ml) were lower, and clearance of total oxazepam (2.5 ml/min/kg) higher, than the corresponding values in controls (330 ng/ml and 1.2 ml/min/kg, respectively). However, these differences were attributable to the greatly increased oxazepam free fraction in renal insufficiency patients (12.0 vs. 4.5% unbound). After correction for individual values of free fraction, C<sub>ss</sub> and clearance of pharmacologically active, unbound oxazepam were essentially identical between groups (14.6 vs. 14.4 ng/ml; 27 vs. 28 ml/min/kg). Compared to controls, C<sub>ss</sub> of the inactive glucuronide metabolite of oxazepam among renal failure patients was greatly increased (2,377 vs. 372 ng/ml) and clearance greatly reduced (0.16 vs. 0.89 ml/min/kg) since excretion of the glucuronide depends on glomerular filtration. Dialyzability of oxazepam and oxazepam glucuronide was minimal. Arterial and venous plasma concentrations were essentially identical, and dialysate concentrations were a small fraction of those in plasma. Thus renal insufficiency does not alter the clearance or C<sub>ss</sub> of pharmacologically active, unbound oxazepam.