From a large cooperative prospective randomized study, data relating to a subgroup of 113 patients with angina pectoris and a significant lesion of the left main coronary artery were analyzed. Of these patients, 53 had been randomly allocated to a medical treatment group and 60 to a surgical treatment group. The former group received conventional medical treatment, while the surgical treatment group received one or more aortocoronary saphenous vein bypass grafts. Important risk factors were approximately uniformly distributed between the two groups. Both are being followed up to 60 months (average follow-up, 30 months). To date, 12 of 60 surgical patients (20%) and 19 of 53 medical patients (36%) died (P less than 0.06). The operative (30-day) mortality declined from a rate of 25% for the first 2 years of the study to 7% for the last 3 years. Of patients randomized in the latter 3 years of the study, 12 of 41 medical patients (29%) and three of 42 surgical patients (7%) died (P less than 0.01). The average follow-up period in this group was 24 months. The proportion surviving 24 months was clearly larger in the surgically treated group (P less than 0.02). The difference in the proportion of patients surviving after surgery as compared with medical treatment was greatest in patients with additional significant disease involving the right coronary artery, with or without left ventricular dysfunction. Relief of angina as assessed by an "anginal score" was also better in surgical patients to a significant degree. Graft-patency rates correlated well with relief of angina, but objective studies including treadmill testing are not yet available.