Quality of life indexes were assessed in 780 patients 10 years after randomization to medical therapy (n = 390) or coronary artery bypass graft surgery (n = 390) in the Coronary Artery Surgery Study. At 10 years, mortality was 21.8% in the medical group and 19.2% in the surgical group (p = NS), and 144 (37%) of the medical group had undergone surgery because of increasing chest pain. At study entry, 22% of medical and surgical patients were angina free; at 1 and 5 years after entry, the frequency of asymptomatic patients was 66% and 63% in the surgical group and 30% and 38% in the medical group. However, by 10 years after entry, the proportion of patients free of angina had fallen to 47% in the surgical group and to 42% in the medical group. Activity limitation and use of β-blockers and long-acting nitrates were less in the surgical than the medical group at 1 and 5 years after entry but little different from the medical group at 10 years after entry. Throughout follow-up, recreational status, employment status, frequency of heart failure, use of other medications, and hospitalization frequency were similar between the two groups. Thus, indexes of quality of life such as angina relief, increased activity, and reduction in use of antianginal medications initially appear superior in patients with stable manifestations of ischemic heart disease assigned to surgery, but by 10 years after entry, these advantages are much less apparent. Although the observed similarities of the medically and surgically assigned groups at 10 years reflect return of symptoms in the surgical group to some extent, a more important explanation is the performance of late surgery in a late proportion of the medically assigned patients, rendering them asymptomatic.