Causes of stroke in carotid endarterectomy.

Carotid endarterectomy performed with the patient conscious under regional anesthesia provides a unique opportunity to determine the time of onset of a neurologic deficit and in deduce a likely cause. If a trial period of carotid occlusion is tolerated without the development of a neurologic deficit (96% of our patients), operation may continue without indwelling shunt. Of the 345 patients who had elective carotid endarterectomies performed without shunt, neurologic deficits lasting longer than 24 hours developed in 6 patients (1.7%), and deficits resolving within 24 hours occurred in 15 patients (4.3%). The neurologic deficit developed during carotid dissection in 3 patients, during carotid occlusion in 1 patient, upon release of carotid occlusion in 2 patients, and in the first 5 postoperative days in 15 patients. Of the 15 postoperative deficits, 9 were transient ischemic attacks similar to preoperative episodes, 4 were strokes, and 2 were visual changes. Twenty of 21 deficits were thromboembolic, reperfusion phenomena or were related to hypotension. Only one (0.3% of 345 cases) could be attributed to cerebral anoxia. We believe comparison of raw stroke rates is not valid in comparing methods of cerebral protection, since most perioperative neurologic deficits are not attributable to hypoperfusion. Furthermore, trial carotid occlusion in the conscious patient is a satisfactory method for determining the need to use a shunt.