The Coronary Artery Surgery Study (CASS) includes 780 patients with mild or moderate stable angina pectoris or asymptomatic survivors of a myocardial infarction who were randomized to either medical or surgical therapy and 1,319 patients who were eligible for randomization but were not randomized (randomizable patients). There were no substantial aggregate differences observed in any of the survival comparisons after 10 years of follow-up study between the randomized and randomizable patients assigned to the medical (79% versus 80%) or surgical (82% versus 81%) groups or in patient subgroups stratified according to coronary artery disease extent and left ventricular ejection fraction. Cox regression analyses were done with independent variables known to be predictors of survival, including surgical versus medical therapy and randomized versus randomizable group, to test the null hypothesis of a mortality difference between medical versus surgical assignment according to group assignment (randomized versus randomizable). In no case did the initial group category enter as a significant predictor of survival. The results in the randomizable group reinforce those in the randomized group with respect to the medical versus surgical comparison. Two subgroups are identified with a significant surgical advantage: 1) patients with proximal left anterior descending coronary artery stenosis greater than or equal to 70% and an ejection fraction less than 0.50, and 2) patients with three vessel coronary artery disease and an ejection fraction less than 0.50. In both groups, coronary bypass surgery had a statistically significant beneficial effect on survival (p less than 0.05). After a decade of follow-up, the CASS randomizable patients confirm conclusions reached on the basis of the CASS randomized trial.