The detrimental impact of silent cerebral infarcts on asymptomatic carotid endarterectomy outcome.

Background: Silent cerebral infarctions (SCIs) can be identified by preoperative computed tomography (CT) scans in patients with severe carotid stenosis being considered for carotid endarterectomy (CEA). It is unknown whether this finding has any effect on perioperative complications or long-term outcome. This study investigates the influence of SCI on early and late complications in asymptomatic patients undergoing CEA.

Methods: All consecutive CEAs undertaken for asymptomatic severe carotid stenosis from 2005 to 2013 were retrospectively evaluated for clinical and anatomic characteristics. SCI was defined as cerebral embolic infarcts in the anterior or middle cerebral artery territory, ipsilateral to the target carotid stenosis, identified on preoperative CT. The end points of the study were to compare the 30-day and long-term stroke and death rate after CEA in patients with and without SCI. All patients were followed yearly through duplex ultrasonography and clinical assessment. Statistical methods used were Cox regression (hazard ratio) and Kaplan-Meier for life-table analysis.

Results: A total of 743 CEAs were performed in asymptomatic patients during the study period of which all had CT scans, and 97 (13.1%) demonstrated SCI. All patient stroke and death outcomes at 30 days were 0.5% and 0.7%, respectively. Patients with SCI had a significantly higher 30-day stroke outcome (3.1% vs 0.2%; P = .001; odds ratio, 16.39; 95% confidence interval, 1.33-201.4; P = .02) but not death or stroke/death outcome (0% vs 0.8%; P = .19 and 3.1% vs 0.9%, P = .06, respectively) compared with those without SCI. In addition, at a mean follow-up of 44.3 ± 23.9 months, the patients with SCI had a significantly worse 5-year ipsilateral stroke or any stroke-/death-free survival compared with patients without SCI (86.7% vs 99.0%; P = .001; and 76.9% vs 87.7%; P = .004). SCI was confirmed as an independent predictor of late any stroke/death by Cox regression (hazard ratio, 2.45; 95% confidence interval, 1.29-4.67; P = .006).

Conclusions: Patients who have SCI in the presence of severe carotid stenosis and undergo CEA have significantly worse perioperative stroke and long-term stroke/death outcomes. This data would suggest that asymptomatic patients undergoing CEA who have CT scan evidence of a cerebral infarct have worse prognosis than those with normal CT scans.