One hundred sixty-five patients undergoing primary myocardial revascularization were prospectively entered into a randomized, double-blind, placebo-controlled study, in a single institution, in order to determine the influence of high- and low-dose aprotinin application on early coronary artery bypass graft patency. All patients were operated on by the same team and the three treatment groups were comparable in all demographic data and surgical variables. Postoperative chest tube drainage and transfusion requirements were significantly reduced in patients receiving high or low doses of aprotinin. In all patients vein and internal mammary artery graft patency was assessed by control coronary angiograms 4 to 15 days (median 8.2 days) postoperatively. In the high-dose aprotinin group, 140 of 142 vein grafts and in the low-dose aprotinin group all of the 128 vein grafts were patent compared with 138 of 139 in the placebo group. The difference was not statistically significant (P > 0.05). All pedicled internal mammary artery grafts were patent in the three treatment groups. The prevalence of perioperative myocardial infarction was evaluated by serial creatine kinase-myocardial band (CK-MB) isoenzyme measurements and by electrocardiographic recordings. No additional changes that could be attributed to aprotinin were observed. In conclusion, these results suggest that perioperative myocardial infarction secondary to aprotinin-induced native coronary artery or conduit thrombosis is not increased by aprotinin in patients undergoing primary myocardial revascularization.