Fifty North American centers have combined to evaluate the benefit of carotid endarterectomy in randomized patients who have experienced symptoms related to arteriosclerotic stenosis of the carotid artery and who have received either best medical therapy alone or best medical therapy plus carotid endarterectomy. The outcome events are nonfatal and fatal stroke or death. A three-tier system identifies and adjudicates the type, severity, and location of each stroke and the cause of any death. Data about patients submitted to carotid endarterectomy outside the trial are compiled at the Nonrandomized Data Center at the Mayo Clinic. Between December 27, 1987, and October 1, 1990, 1,212 patients were randomized, 596 to medical therapy, 616 to carotid endarterectomy. Cross-over from the medical to the surgical arm has been low (4.2%). Patients eligible for the trial, but not randomized totaled 1,044; their characteristics were similar to those randomized so that, for the type of symptomatic patient in this study, our conclusions about the benefit of carotid endarterectomy can be generalized. Patients excluded by medical criteria totaled 679. Another 1,591 had carotid endarterectomy, but either lacked the disease under study, were asymptomatic, or received inadequate investigation to meet entry criteria. We set sample size at 1,900 patients, with continuing enrollment. The Monitoring Committee reviews at intervals the confidential analyses performed on the groups with moderate (30-69%) and severe (70-99%) stenosis. Stopping rules will be invoked for one or both groups if unequivocal benefit or harm is identified.
