There is a wide variation in the reported results of endarterectomy in conjunction with coronary artery bypass grafting. Operative mortality ranges from 0 to 10 per cent, perioperative infarction ranges from 5 to 30 per cent, and patency rates range from 38 to 100 per cent, with 74 to 95 per cent being asymptomatic or improved. This wide discrepancy in clinical outcome occurs for a number of reasons. First, there is nonhomogeneity of patient populations resulting from a lack of standardized patient selection criteria. Second, the operative experience of most published reports spans a decade throughout which major advances in cardiac surgery have occurred. Finally, endarterectomy is a technically challenging procedure with its own learning curve. Differences in technique and expertise almost certainly contribute to the variability of results. Patients with diffuse coronary artery disease pose a challenge to the cardiac surgeon. Endarterectomy entails the risks of increased morbidity and mortality and therefore should be done only if conventional bypass grafting is precluded. However, the risks of these complications must always be weighed against the possible benefits. At the present time carefully selected patients can benefit from this procedure. Unfortunately, until such time as controlled randomized studies are carried out on a prospective basis, statistical support for this procedure will not be available. Given that such a study is questionable from an ethical point of view, clinical decisions for this select group of patients must be done on a case-by-case basis. The best that can currently be done is to carefully follow these patients and scrutinize existing data to ensure optimal clinical management.