RS05 - Re-evaluating Current Post-carotid Endarterectomy Surveillance: Long-term Results from a Large Integrated Regional Health System

Objectives: The appropriate post-carotid endarterectomy (CEA) surveillance protocol is actively debated. Current guidelines recommend five duplex ultrasound studies in the first two postoperative years, followed by annual scans. Yet, robust clinical trial data shows low incidence of ≥70% restenosis following CEA and annual incidence of ipsilateral ischemic events of 0.5-1.8% per year. We evaluated the long-term utility of surveillance after CEA in preventing stroke and identifying restenosis.

Methods: This retrospective study included patients that underwent CEA for severe (70-99%) carotid artery stenosis from 2008-2012 with follow-up through 2019 within a large integrated regional healthcare system. Patients with prior ipsilateral intervention or who died within 30 days postoperatively were excluded. All routine postoperative carotid artery imaging was assessed; studies for acute postoperative issues were excluded. Primary outcomes were ipsilateral severe restenosis or occlusion, and ipsilateral stroke. A competing risks analysis was used to quantify freedom from the primary outcomes.

Results: Overall, 1,923 carotid arteries in 1,816 patients underwent CEA during the study period, of which 831 (43.2%) were for symptomatic indication. Mean follow-up was 6.6 years (SD 3.3 years). There were 5,457 postoperative imaging studies performed in 1,571 patients (1,663 arteries); and 249 (13.7%) patients (260 arteries [13.5%]) had no imaging during the follow-up period. Among patients with imaging follow-up, the median number of studies was 3 (interquartile range 1-4) and the median time to first study was 3.4 months (interquartile range 1.3-7.1 months). Severe restenosis was seen in 124 arteries (6.5%), and occlusion was observed in 20 arteries (1.0%). The mean time to restenosis or occlusion was 1.9 (SD 2.4) years. Only 27 (1.4%) arteries underwent reintervention. Of the 71 (3.7%) postoperative ipsilateral strokes, only 3 (4.2%) and 2 (2.8%) of these occurred in the setting of a known severe stenosis or occlusion, respectively. The cumulative risk of restenosis, occlusion, or ipsilateral stroke at five years were 6.0%, 0.9%, and 2.7% respectively (Figure 1). Among arteries with an unremarkable first study (1,396/1,663) the cumulative risk of stenosis, occlusion and stroke at five years were 2.3%, 0.3%, 2.4%, respectively (Figure 2).

Conclusions: The risks of severe restenosis or occlusion after CEA were low and did not appear to be associated with the low rate of ipsilateral ischemic stroke. Patients with unremarkable first surveillance studies had even lower rates. These results challenge the notion that patients substantially benefit from the current postoperative surveillance guidelines and suggest that these may lead to over-surveillance.