Determinants of outcome after carotid endarterectomy

Background: The efficacy of carotid endarterectomy for selected patients ha s been evaluated with randomized controlled clinical trials. The generaliza bility of these studies to average surgical practice remains an important p ublic health concern. Objective: The objective of the study was to determine the predictors of ou tcome after carotid endarterectomy on a regional basis. Patients and Methods: The study was designed as a retrospective cohort stud y and included all consecutive patients presented for carotid endarterectom y at the 8 University of Toronto-affiliated hospitals in the period from Ja nuary 1, 1994, to December 31, 1996. The main outcome measure was 30-day po stoperative stroke or death rate. Results: During the study interval, 1280 primary carotid endarterectomies w ere performed. The overall combined stroke and death rate was 6.3% for all patients who underwent endarterectomy (4.0% for patients who were asymptoma tic). The significant predictors of poor outcome were the following: presen ting symptoms (odds ratio, 1.74; 95% confidence interval [CI], 0.96, 3.12), low surgeon volume (<6 cases per year; odds ratio, 3.98; 95% CI, 1.65, 9.5 8), and left-sided surgery (odds ratio, 1.72; 95% CI, 1.07, 2.76). Conclusion: These data suggest that adoption of the recommendations of the symptomatic carotid endarterectomy trials is appropriate. However, endarter ectomy for asymptomatic lesions remains of uncertain benefit on a regional basis and must be individualized to the experience of the specific surgeon. The surgeon volume/outcome relationship that is identified in this study s uggests a need for a minimum volume threshold for this procedure.