There are three large randomized trials of early elective coronary artery bypass graft surgery vs early medical therapy in selected patients with stable angina pectoris: the Veterans Administration Study (or VA Study) with enrollment in 1972 to 1974, the European Coronary Surgery Study Group (or European Study) with enrollment in 1973 to 1976 and the Coronary Artery Surgery Study (CASS) with enrollment in 1975 to 1979. Design and other similarities and contrasts for the three trials are presented. For example if 50% luminal diameter narrowing is used to define a significantly diseased vessel and patients with left main coronary artery disease are eliminated from the VA and European studies, the percentages of patients with one-, two-, and three-vessel disease are: VA 16%, 34%, and 50%; European 0%, 43%, and 57%; CASS 19%, 31%, and 51%. The VA and European studies enrolled patients with NYHA anginal classes I through IV, whereas CASS enrolled patients with Canadian Heart Association classes I and II, as well as patients asymptomatic for chest pain after a myocardial infarction. Other study features of note include the differing geographic locations and time of enrollment. In the VA Study, surgery was performed within 2 months on 94% of those randomly assigned to surgery; the mean time to surgery (of those not crossing over to medicine) in CASS was 54 days; and the European Study had a mean delay of 3.9 months. The characteristics of the clinics, socioeconomic environments, and medical care and referral systems may be important to interpreting results, although they were not clearly specified.(ABSTRACT TRUNCATED AT 250 WORDS)