Results in animals suggest favourable coronary vasomotor actions of isoflavones; however, the effects of isoflavones on the human coronary circulation have not been determined. In the present study, we therefore investigated the effects of short-term isoflavone-intact soya protein ingestion on basal coronary arterial tone and stimulated vasoreactivity and blood flow in patients with CHD (coronary heart disease) or risk factors for CHD. Seventy-one subjects were randomized, double-blind, to isoflavone-intact soya protein [active; n=33, aged 58+/-8 years (mean+/-S.D.)] or isoflavone-free placebo (n=38, aged 61+/-8 years) for 5 days prior to coronary angiography. In 25 of these subjects, stimulated coronary blood flow was calculated from flow velocity, measured using intracoronary Doppler and coronary luminal diameter before and after intracoronary adenosine, ACh (acetylcholine) and ISDN (isosorbide dinitrate) infusions. Basal and stimulated coronary artery luminal diameters were measured using quantitative coronary angiography. Serum concentrations of the isoflavones genistein, daidzein and equol were increased by active treatment (P<0.001, P<0.001 and P=0.03 respectively). Basal mean luminal diameter was not significantly different between groups (active compared with placebo: 2.9+/-0.7 compared with 2.73+/-0.44 mm, P=0.31). There was no difference in luminal diameter, flow velocity and volume flow responses to adenosine, ACh or ISDN between groups. Active supplement had no effect on basal coronary artery tone or stimulated coronary vasoreactivity or blood flow compared with placebo. Our results suggest that short-term consumption of isoflavone-intact soya protein is neither harmful nor beneficial to the coronary circulation of humans with CHD or risk factors for CHD. These results are consistent with recent cautions placed on the purported health benefits of plant sterols.