CAROTID ENDARTERECTOMY FOR UNSTABLE AND COMPELLING NEUROLOGIC CONDITIONS - DO RESULTS JUSTIFY AN AGGRESSIVE APPROACH

Purpose: In a retrospective study the outcome of 70 carotid endarterec tomies (CEA) in 68 patients with neurologically unstable conditions or anatomically compelling findings on carotid angiography was examined to more accurately identify patients who might benefit from CEA in thi s setting. Methods: Out of a total of 1734 CEAs performed from 1978 to 1992, five groups of patients were selected: group A, stroke in evolu tion with tight stenosis (n = 5); group C, crescendo transient ischemi c attacks (CTIA) continuing despite heparin (n = 14); group D, CTIA (a bove criteria) ceasing with heparin (n = 21); and group E, anatomicall y compelling situation on carotid angiography (n = 13). Data collected included preoperative and postoperative Neurologic Event Severity Sco re (NESS), CHAT classification, arteriosclerosis risk factors, demogra phics, and long-term overall and transient ischemic attack/stroke-free survival rates. Results: Risk factors and demographics were similar i n all groups. By NESS criteria the conditions of 97.3% of patients in the neurologically unstable groups A to C were improved or stabilized after operation, with one deterioration (2.7%). All patients in group B either stabilized or improved. In group D, one patient's NESS deteri orated, resulting in 3.5% overall morbidity rate and no deaths for gro ups A to D. Follow-up showed an overall survival rate by Kaplan-Meier analysis equivalent to a matched control population, with 85% alive at 5 years. The cumulative TLA/stroke-free survival rate at 5 years was 75%. Conclusions: In this retrospective series, CEA performed for comp elling or unstable neurologic findings carried low morbidity and morta lity rates. Early aggressive surgical therapy of neurologically unstab le patients may be warranted because our results improved on the antic ipated natural history of the conditions studied. Further clarificatio n of proper patient selection is necessary before this principle can b e applied broadly.