The impact of routine intraoperative carotid angiography was evaluated by comparing 206 procedures without such angiograms with our last consecutive 100 endarterectomies with completion angiography. No significant age or sex differences were observed between the two groups. Exploratory surgery was repeated in five cases for a stenosis greater than 40% or for an intimal flap. This protocol reduced operative mortality (2.9% to 1%), the permanent stroke rate (1.9% to 1%), and the temporary stroke rate (6.3% to 1%). Furthermore, a second angiogram was performed in these 100 cases (at a mean interval of 19.2 months later) and the incidence and evolution of both residual and recurrent carotid lesions were analyzed. Five internal carotid artery lesions that had been immediately repaired because of intraoperative angiographic defects remained normal. Of 58 normal internal carotid arteries at the completion of surgery, two became stenotic during the next year. In addition, three spastic internal carotid arteries became normal. Of 20 internal carotid arteries with modest irregularities, 16 became normal and four were stenosed. Of three internal carotid arteries with intimal flaps, two became normal and one was stenosed. Among 13 internal carotid arteries with modest stenosis (40%), eight became normal, two became severely stenotic, and three became thrombosed. Among 21 instances of a proximal common carotid artery "shelf," 17 resolved and four progressed to less than 50% stenosis. Of 67 normal external carotid arteries, late stenosis was seen in one case. Of 33 external carotid arteries with residual stenosis, 17 became normal, 14 remained unchanged, and two were thrombosed.(ABSTRACT TRUNCATED AT 250 WORDS)