The relative influences of revascularization status and baseline characteristics on long-term outcome were examined in 867 patients with multivessel coronary disease who had undergone successful coronary angioplasty. These patients represented 83% of a total of 1,039 patients in whom angioplasty had been attempted with an in-hospital mortality and infarction rate of 2.5% and 4.8%, respectively. Emergency coronary bypass surgery was needed in 4.9%. Of the 867 patients, 41% (group 1) were considered to have complete revascularization and 59% (group 2) to have incomplete revascularization. Univariate analysis revealed major differences between these two groups with patients in group 2 characterized by advanced age, more severe angina, a greater likelihood of previous coronary surgery and infarction, more extensive disease and poorer left ventricular function. Over a mean follow-up period of 26 months, the probability of event-free survival was significantly lower for group 2 only with respect to the need for coronary artery surgery (p = 0.004) and occurrence of severe angina (p = 0.04). The difference in mortality was of borderline significance (p = 0.051) and there were no significant differences between groups 1 and 2 in either the incidence of myocardial infarction or the need for repeat angioplasty. Multivariate analysis identified independent baseline predictors of late cardiac events that were then used to adjust the probabilities of event-free survival. This adjustment effectively removed any significant influence of completeness of revascularization on event-free survival for any of the above end points including the combination of death, myocardial infarction and need for coronary artery surgery. Therefore, late outcome in these patients is not significantly influenced by revascularization status but depends more on baseline patient characteristics.