OBJECTIVE The aim of this study was to determine the long-term survival and control of angina in patients with coronary artery disease and sequentially decreased ejection fractions (EF) after first-time coronary artery bypass grafting. METHODS Between 1981 and 1995, 156 (1.3%) patients with an EF less than 0.25 (group 1), 588 (5%) patients with an EF of 0.25 to 0.34 (group 2), 2,438 (20.6%) patients with an EF of 0.35 to 0.49 (group 3), and 8,648 (73.1%) patients with an EF equal to or greater than 0.50 (group 4) underwent coronary artery bypass grafting. The EFs were determined by uniplanar or biplanar left ventriculography. For each group, the clinical and angiographic characteristics and the operative and outcome data were compared. Survival curves were derived and compared for each group. Correlates of angina, and of early (30-day) and long-term mortality, for all groups were analyzed. RESULTS For all groups the mean age was approximately 60+/-10 years. Group 1 had the highest percentage of patients who were men (88%), had congestive heart failure (34%), had hypertension (53%), and had left main coronary artery disease (24%). Groups 1 through 3, compared with group 4, had a lower percentage of complete revascularization (p < 0.0001), a lower percentage of internal mammary artery grafts (p < 0.0001), and a greater use of intraaortic balloon pump (p < 0.0001), but had similar cross-clamp and cardiopulmonary bypass times, number of grafts, incidences of myocardial infarction, and stroke. Hospital mortality for groups 1, 2, 3, and 4 was 3.8% (n = 6), 3.4% (n = 20), 3% (n = 72), and 1.6% (n = 134), respectively. Groups 1 through 3, compared with group 4, had similar incidences of angina during follow-up (31% to 40% versus 33%, respectively; p < 0.06). Survival was greatest for group 4 compared with groups 1 through 3 at 1, 5, and 10 years (p < 0.0001). Patients in group 1 had 1-, 5-, and 7-year survivals of 90%, 64%, and 49%. Multivariate correlates of early mortality were advanced age, female sex, decreased EF, hypertension, diabetes, and emergency operation. Multivariate correlates of long-term mortality included severity of preoperative angina class, congestive heart failure, number of diseased vessels, and incomplete revascularization. The strongest correlates of angina at follow-up were younger age, female sex, previous myocardial infarction, lower ejection fraction, and incomplete revascularization. The absence of an internal mammary artery graft did not predict the occurrence of angina or influence long-term survival. CONCLUSIONS In the long term there is a higher mortality in patients with sequentially decreased left ventricular function undergoing coronary artery bypass grafting, although more than 60% of patients with an EF less than 0.25 were alive and had good control of angina at 5 years despite having a higher percentage of risk factors, poorer functional status, and more complex coronary disease. Failure of symptom control and survival beyond 5 years appeared to be influenced by preexisting medical conditions and factors that affect the ability to completely revascularize the myocardium. These results suggest that in selected patients with ischemia and poor left ventricular function, coronary artery bypass grafting may preserve remaining viable myocardium, provide relief of symptoms, and offer survival greater than 60% at more than 5 years.