This randomised double-blind, placebo-controlled, clinical trial investigated the effect of 3 months of treatment with calcium dobesilate on endothelium-dependent vasodilation, markers of endothelial function, blood pressure, and markers of oxidation in obese, male smokers. Vascular effects may depend on the type of vessel and we, therefore, investigated both smaller arteries, i.e. resistance arteries and small arterioles, and large conduit arteries. Vascular function was measured by acetylcholine- and sodium-nitroprusside-mediated vasodilation, and capillary recruitment, in the skin microcirculation; by forearm blood flow (FBF) responses to several agonists and to N-G-monomethyl L-arginine (L-NMMA) in the forearm vascular bed; by flow-mediated vasodilation in the brachial artery; and by determination of soluble levels of vascular cell adhesion molecule-1 (sVCAM-1), intercellular adhesion molecule-1 (sICAM-1) and E-selectin. Twenty-eight individuals received dobesilate and 24 placebo. No effect of calcium dobesilate on endothelial function, blood pressure or markers of oxidation was observed compared with placebo. The difference in acetylcholine-mediated vasodilation in the microcirculation was -52.1%-point (95% confidence interval -132.8 to 28.1); in sodium-nitroprusside-mediated vasodilation in the microcirculation, 2.6%-point (-95.1 to 100.2); in capillary recruitment, 2.5%-point (-6.8 to 11.7); in acetylcholine-induced increases in FBF (n=28), 23%-point (-173 to 126); in L-NMMA-induced reduction of basal FBF, -2.8%-point (-29.3 to 23.8); in flow-mediated vasodilation of the brachial artery, 0.3%-points (-2.7 to 3.3); in 24-h systolic blood pressure, 2.1 mmHg (-1.3 to 5.5); in soluble VCAM-1, 54 ng/ml (-8 to 115); in soluble ICAM-1, 9 ng/ml (-49 to 67); in sE-selectin, -17 ng/ml (-44 to 11); in ketocholesterol 5 nM (-17 to 26); and in oxidised LDL -1.6 U/l (-6.7 to 3.5). We have shown that endothelial function, blood pressure, and markers of oxidation were not affected by 3 months of treatment with calcium dobesilate in mildly obese, smoking men. Thus, our data provide no evidence of an effect on vascular function of calcium dobesilate in humans.