Completeness of revascularization after coronary artery bypass operation has been shown to improve short- and medium-term outcome. The purpose of this study was to assess the independent contribution of completeness of revascularization to long-term outcome. A total of 2057 patients with multivessel disease with complete revascularization and 803 with incomplete revascularization, mean age 57 +/- 9 years, was studied. The patient groups were similar except for more prior myocardial infarctions, worse left ventricular function, and more three-vessel disease in the incomplete revascularization group. Complications of perioperative infarction and stroke were not different between those having complete versus incomplete revascularization. The hospital death rate for patients having complete revascularization during the period of study was 0.7% versus 1.5% for those having incomplete revascularization (p = 0.06). Length of hospital stay for the two groups of patients also was not different. At late follow-up (mean 11.7 years for complete and 10.8 years for incomplete) patients who had incomplete revascularization had a significantly higher prevalence of recurrent angina. Multivariate analysis demonstrated the strongest predictors of incomplete revascularization to be number of vessels diseased and left ventricular function (ejection fraction). The multivariate correlates of survival were older age, left ventricular dysfunction, and completeness of revascularization. Completeness of revascularization correlated with improved overall patient survival, as well as survival in patients with normal left ventricular function. Furthermore, the curves continued to separate over time, such that the difference was greater at 8 years than at 4 years, although by 12 years the curves started to converge.