Insulin-dependent diabetic patients found to have substantial coronary artery disease at the time of assessment for renal transplantation have 2-year survival of less than 50%. Because most of these patients have no angina symptoms their management is controversial. We tried to find out whether coronary artery revascularisation in such patients might decrease the combined incidence of unstable angina, myocardial infarction, and cardiac death. 151 consecutive insulin-dependent diabetic candidates for renal transplantation underwent coronary angiography. 31 had stenoses greater than 75% in one or more coronary arteries, atypical chest pain or no chest pain, and a left ventricular ejection fraction greater than 0·35. Of these, 26 agreed to be randomly assigned medical treatment (a calcium-channel-blocking drug plus aspirin) or revascularisation (angioplasty or coronary bypass surgery). 10 of 13 medically managed and 2 of 13 revascularised patients had a cardiovascular endpoint within a median of 8·4 months of coronary angiography (p<0·01). 4 medically managed patients died of myocardial infarction during follow-up. Thus, revascularisation decreased the frequency of cardiac events in insulin-dependent diabetic patients with chronic renal failure and symptomless coronary artery stenoses. These findings suggest that diabetic renal transplant candidates should be screened for silent coronary artery disease, because revascularisation may decrease cardiac morbidity and mortality in this population.