OBJECTIVE Symptomatic cerebrovascular disease is more common in patients who have diabetes mellitus than in the nondiabetic population, even when matched for associated risk factors. Although the safety and efficacy of carotid endarterectomy has been established by NASCET and ACAS, several small studies have noted an increased rate of perioperative neurologic morbidity in patients with diabetes. METHODS Data for all patients who underwent carotid endarterectomy at a single institution from Jan. 1990 to Dec. 1995 were prospectively entered into a computerized vascular registry and form the basis of this report. RESULTS Of 732 carotid endarterectomy procedures performed, 284 (39%) were performed in patients who had diabetes mellitus. Patients with diabetes and without diabetes were matched for clinical presentation (diabetic patients, 45% asymptomatic; nondiabetic patients, 43%) and internal carotid artery percent stenosis (86.6% +/- 10.6% vs 86.4% +/- 10.6%). Patients with diabetes were younger at presentation than patients without (68.8 +/- 8.5 years vs 70.9 +/- 8.5 years; p < 0.005) and were more likely to have a history of coronary artery disease (53% vs 45%; p = 0.04). The mean total length of stay was 6.1 days for patients with diabetes and 4.8 days among patients without (p = 0.01). An adverse postoperative cardiac event (myocardial infarction, congestive heart failure, or arrhythmia) occurred in nine patients with diabetes (3.2%) and in five nondiabetic patients (1.1%; p < 0.05). By logistic regression analysis, however, diabetes was not an independent risk factor for a postoperative cardiac event (p = 0.28). There were 11 perioperative neurologic events (eight cerebrovascular accidents, three transient ischemic attacks) during the entire period (1.5%), of which six were among diabetic patients (2.1%) and five among nondiabetic patients (1.1%; p = NS). Of the eight cerebrovascular accidents, three occurred in diabetic patients (1.0%) and five in nondiabetic patients (1.1%; p = NS). The total operative mortality rate was 0.3% (diabetic patients, 1 of 284, 0.35%; nondiabetic, 1 of 447, 0.2%). CONCLUSIONS Patients with diabetes who undergo carotid endarterectomy are more likely to have coexisting cardiac disease, which may contribute to a higher incidence of postoperative cardiac morbidity. Diabetes mellitus alone, however, is not a risk factor for postoperative cardiac morbidity in patients who undergo carotid surgery. In addition, carotid endarterectomy may be safely performed in patients with diabetes with neurologic morbidity and mortality rates that are comparable with those of the nondiabetic population