Contralateral Occlusion and Concomitant Procedures Drive Risk of Non-ipsilateral Stroke After Carotid Endarterectomy

Abstract

Stroke after carotid endarterectomy (CEA) has been assessed widely. However, factors enhancing non-ipsilateral stroke risk are poorly defined. The aim of this study was to identify drivers of 30 day non-ipsilateral stroke after CEA in the Vascular Quality Initiative (VQI) and assess long-term survival based on laterality of post-operative stroke. The VQI was queried between April 1, 2003, and March 31, 2017, for all CEA. Bilateral carotid procedures within 30 days were excluded. Thirty day non-ipsilateral strokes were identified. Factors were examined to discriminate between patients with and without non-ipsilateral stroke. Univariable analysis followed by multivariable logistic regression was performed. Kaplan-Meier and log rank methods were used to estimate and compare survival. During this 14 year period, 80,230 CEA in 74,928 patients met the criteria. The average age was 70.3±9.3 years. Most were male (48,506; 60%), Caucasian (73,967; 92%), smokers (60,543; 76%), and asymptomatic (43,074; 54%). Contralateral stenosis ≥70% was present in 8033 (10%) with 2239 (3%) having contralateral occlusion. In 491 (0.6%) patients, peri-operative non-ipsilateral stroke occurred. After characterising univariable associations, logistic regression identified independent drivers of non-ipsilateral stroke after CEA. Operative urgency (p=.001), symptomatic disease (p<.001) and contralateral occlusion (p=.001) were pre-operative drivers. Operative predictors included shunt use (p=.008), CEA with cardiac surgery (p=.013), and CEA with concomitant proximal ipsilateral endovascular intervention (p=.01). Use of dextran (p=.005) and anti-angiotensin therapy (p=.03) were protective. Reperfusion syndrome (p<.001), re-exploration (p<.001), myocardial infarction (p<.001), and intravenous treatment of hypotension (p<.001) or hypertension (p<.001) were post-operative correlates. Non-ipsilateral stroke 30 day mortality was less than ipsilateral stroke (6.1% vs. 10.3%; p=.007). Five year survival after non-ipsilateral stroke was 73%, and no different from ipsilateral stroke 76% (p=.16). Both were worse than without stroke (88%; p<.001). Non-ipsilateral stroke after CEA is rare. Features driving risk surround global disease burden, combined procedures, and haemodynamic fluctuations. Contralateral occlusion independently increases non-ipsilateral stroke risk. Regardless of laterality or location, effects of stroke after CEA on long-term survival are similar.