Safety of Carotid Endarterectomy for Symptomatic Stenosis by Age: Meta-Analysis With Individual Patient Data
Leung YYR., Bera K., Urriza Rodriguez D., Dardik A., Mas J-L., Simonte G., Rerkasem K., Howard DPJ.
Background: There is uncertainty whether elderly patients with symptomatic carotid stenosis have higher rates of adverse events following carotid endarterectomy. In trials, recurrent stroke risk on medical therapy alone increased with age, whereas operative stroke risk was not related. Few octogenarians were included in trials and there has been no systematic analysis of all study types. We aimed to evaluate the safety of carotid endarterectomy in symptomatic elderly patients, particularly in octogenarians. Methods: We did a systematic review and meta-analysis of studies (from January 1, 1980 through March 1, 2022) reporting post carotid endarterectomy risk of stroke, myocardial infarction, and death in patients with symptomatic carotid stenosis. We included observational studies and interventional arms of randomized trials if the outcome rates (or the raw data to calculate these) were provided. Individual patient data from 4 prospective cohorts enabled multivariate analysis. Results: Of 47 studies (107 587 patients), risk of perioperative stroke was 2.04% (1.94–2.14) in octogenarians (390 strokes/19 101 patients) and 1.85% (1.75–1.95) in nonoctogenarians (1395/75 537); P =0.046. Perioperative death was 1.09% (0.94–1.25) in octogenarians (203/18 702) and 0.53% (0.48–0.59) in nonoctogenarians (392/73 327); P <0.001. Per 5-year age increment, a linear increase in perioperative stroke, myocardial infarction, and death were observed; P =0.04 to 0.002. However, during the last 3 decades, perioperative stroke±death has declined significantly in octogenarians (7.78% [5.58–10.55] before year 2000 to 2.80% [2.56–3.04] after 2010); P <0.001. In Individual patient data multivariate-analysis (5111 patients), age ≥85 years was independently associated with perioperative stroke ( P <0.001) and death ( P =0.005). Yet, survival was similar for octogenarians versus nonoctogenarians at 1-year (95.0% [93.2–96.5] versus 97.5% [96.4–98.6]; P =0.08), as was 5-year stroke risk (11.93% [9.98–14.16]) versus 12.78% [11.65–13.61]; P =0.24). Conclusions: We found a modest increase in perioperative risk with age in symptomatic patients undergoing carotid endarterectomy. As stroke risk increases with age when on medical therapy alone, our findings support selective urgent intervention in symptomatic elderly patients.