Cigarette smoking has been shown to impair endothelium-dependent dilation in arteries. We tested the hypothesis that cigarette smoking also impairs endothelium-dependent venodilation and evaluated changes in this response after smoking cessation in a time-course study using the dorsal hand vein technique. Dose-response curves were constructed in smokers and nonsmokers by infusing bradykinin (1-278 ng/min), an endothelium-dependent vasodilator, and nitroglycerin (0.006-1,583 ng/min), an endothelium-independent vasodilator, into hand veins preconstricted with the selective alpha1-adrenergic agonist phenylephrine. The maximal venodilation induced by bradykinin was 89 +/- 5% in controls (n = 16) and 61 +/- 7% in smokers (n = 18; P = 0.02). No difference in nitroglycerin-induced venodilation was observed between the two groups. Coinfusion of L-arginine (0.33 mg/min) markedly improved the bradykinin-induced venodilation in smokers (52 +/- 7 to 90 +/- 9%; P < 0.01). After acute smoking cessation (n = 7), restoration to normal bradykinin-induced venodilation was observed within 24 h, whereas no change in the response to a maximally effective dose of nitroglycerin (1,583 ng/min) was detected. In a human vein model appropriate for testing vascular functional alterations, this study demonstrates that smoking impairs endothelium-dependent venodilation in heavy smokers. Moreover, this endothelial dysfunction appears to be rapidly reversible after smoking cessation. This model may be useful in studies evaluating mechanisms of endothelial dysfunction and interventions to modify it.