Cigarette smoking has been shown to increase the clearance of several drugs, including pentazocine, theophylline, and phenacetin. This effect presumably is mediated through enzyme induction, resulting from inhaled polycyclic aromatic hydrocarbons. Because the O-demethylation pathway for codeine is similar to that of other drugs known to be influenced by cigarette smoking, a study was conducted to compare the pharmacokinetics and metabolism of codeine in smokers and nonsmokers. Twelve volunteers with no history of cigarette smoking and ten volunteers who smoke cigarettes were studied; an open, two-treatment, simultaneous parallel and crossover study design was used. Each volunteer received a single dose of codeine sulfate 60 mg po and codeine phosphate 60 mg im, in random treatment order, at one-week intervals. Serial blood samples were collected up to 12 hours after dosing, and plasma codeine and morphine concentrations were measured by radioimmunoassay. There was no significant difference between smokers and nonsmokers in either codeine or morphine areas under the plasma concentration-time curves (AUCs), with either route of administration. The relative codeine bioavailability in these groups was 54.8 +/- 4.9 percent and 50.2 +/- 2.1 percent (mean +/- SE), respectively. Smoking was associated with greater variability in plasma concentrations. Interestingly, greater morphine:codeine AUC ratios were observed in both groups after oral than after intramuscular administration. Cigarette smoking should have no clinically important influence on codeine absorption or disposition.