94 patients with more than 50% obstruction of the left main coronary artery (LM) were studied between February 1972 and August 1975. 18 patients (19%) had one-vessel disease; 21 (22%) two-vessel disease; and 52 (55%) three-vessel disease in addition to LM disease. Only 3 patients (3.1%) had solitary involvement of the LM, 38 (40%) presented a clinical picture of accelerated or recent onset of angina pectoris, and 56 patients (60%) had moderate to severe stable angina pectoris. By means of cinefluoroscopy, calcification was seen in 23 patients (24%). The testing electrocardiogram (ECG) was normal in only 5 patients (5.3%); in 55 patients (58%), the resting ECG showed nonspecific ST-T changes, and in 34 (36%) it showed evidence of an old infarction. Treadmill exercise electrocardiography was performed in 9 patients; the ECG was markedly positive in 7. Atrial pacing produced angina pectoris and ECG changes or an increase in left ventricular end-diastolic pressure (LVEDP) accompanied by a decrease in stroke work in 17 of 18 patients (94%). The left ventriculogram was normal in 18 patients (19%); in 76 patients (81%) it showed variable degrees of hypokinesia or dyskinesia, or both. The LVEDP was normal (12 mm Hg or less) in 41 patients (44%) and between 12 and 20 mm Hg in 33 patients (35%); in 20 (21%) it was higher than 20 mm Hg. Of these 20 patients with markedly elevated LVEDP, 16 had markedly abnormal ventriculograms, and were not considered candidates for surgery. In contrast to our earlier experience, there was no morbidity or mortality associated with cardiac catheterization. 54 patients (57%) underwent saphenous vein bypass surgery, with an operative mortality of 9 patients (17%). We concluded from the follow-up of our patients and our review of the literature that surgery offers symptomatic relief superior to that with medical therapy. However, the question of whether surgery prolongs life is not yet answered, and certainly further studies are needed.