Phenytoin kinetics during long-term alcohol use and withdrawal were studied in 11 male alcoholics with a history of withdrawal seizures and no evidence of chronic liver disease. Ethanol, 20% v/v, was given for 6 days after admission to maintain the blood alcohol level between 500 and 800 mg/l and phenytoin suspension, 150 mg, was given orally or intravenously (on three occasions) every 12 hr for 20 days. The mean (+/- SD) total phenytoin clearance in 9 of 11 subjects was 0.023 +/- 0.006 l/kg/hr during the alcohol ingestion period. Clearance rose to 0.033 +/- 0.013 l/kg/hr (P less than 0.05) during alcohol withdrawal. Total steady-state concentration after 3 wk ranged from 3.4 to 29.9 mg/l, while the weight-corrected dose range was only 3.7 to 5.5 mg/kg/day. Inter- and intra-subject variation in bioavailability was small (0.93 to 1.03). Phenytoin free fractions ranged from 9.09% to 17.75% and changes in total and free phenytoin concentration correlated (r2 = 0.92, P less than 0.001). The data are compatible with the hypothesis that increased phenytoin clearance during alcohol withdrawal is due to the increased metabolic rate of the drug secondary to enzyme induction by ethanol, which becomes unmasked on cessation of drinking. In most alcoholics standard-dose phenytoin (300 mg/l) will induce lower than usual plasma concentrations.