Objective: To compare the long-term outcomes of patients treated with carotid endarterectomy and carotid-artery stenting.
Background: Evidence for the long-term safety and efficacy of carotid-artery stenting compared with endarterectomy is accumulating from randomized trials. However, comparative data on the long-term outcomes of carotid revascularization strategies in real world practice are lacking.
Methods: We conducted a population-based, multicenter, observational cohort study using validated linked databases from Ontario, Canada. We identified all individuals treated with carotid endarterectomy and stenting (2002-2014), and followed them up to 2015. We compared long-term (up to 13 years) and 30-day outcomes of each strategy using multilevel multivariable Cox proportional-hazards models, and conducted confirmatory analyses using propensity-score matching methods.
Results: In all, 15,525 patients received carotid-artery revascularization. Rate of the primary composite outcome of 30-day death, stroke, or myocardial infarction plus any stroke during 13-year follow-up was higher with stenting (16.3%) compared with endarterectomy (9.7%) [adjusted hazard ratio (HR) 1.57, 95% confidence interval (CI) 1.43-1.73, P < 0.001). The increased risk with stenting was observed regardless of age, sex, intervention year, carotid-artery symptoms, or diabetes. The primary outcome was driven by higher rates of 30-day stroke (adjusted HR 1.59, 95% CI 1.29-1.95), 30-day death (adjusted HR 2.62, 95% CI 2.20-3.13), and long-term stroke >30 days after the procedure (adjusted HR 1.47, 95% CI 1.36-1.59) with stenting; 30-day myocardial infarction was lower with stenting (adjusted HR 0.70, 95% CI 0.57-0.86). These results were confirmed with 1:2 propensity-score matching (HR for primary composite outcome with stenting 1.55, 95% CI 1.31-1.83, P < 0.001).
Conclusions: Compared with carotid endarterectomy, stenting was associated with an early and sustained approximately 55% increased hazard for major adverse events over long-term follow-up. Although nonrandomized, these results raise potential concerns about the interchangeability of carotid endarterectomy and stenting in the context of actual clinical practice.