A population-based study of risk factors for stroke after carotid endarterectomy using the ACS NSQIP database

Background: The benefit of carotid endarterectomy (CEA) is heavily influenced by the risk of perioperative stroke. Our objective was to use the American College of Surgeons' 2007 and 2008 National Surgical Quality Improvement Program (NSQIP) database to assess the postoperative stroke and death rate after CEA among the more than 180 NSQIP participating hospitals, and to identify the preoperative risk factors. Materials and Methods: Univariate analysis included 56 preoperative variables. Outcomes were studied for 30 d. Multivariate logistic regression was used for assessment of risk factors. Results: Of 13,316 patients, 7503 (56.5%) were asymptomatic, while 5770 (43.5%) were symptomatic. Combined stroke or death was seen in 262 patients (2.0%). Postoperative stroke occurred in 186 patients (1.4%). One hundred patients (0.8%) died within 30 d. In asymptomatic and symptomatic patients, stroke or death was seen in 1.3% and 2.9% of patients; stroke in 0.9% and 2% of patients; and death in 0.5% and 1.1% of patients, respectively (all P < 0.001). On multivariate analysis for symptomatic patients, dialysis dependence, chronic open wound, impaired sensorium, and dependent functional status were risk factors for stroke or death (all P < 0.05). Among asymptomatic patients, acute renal failure, corticosteroid use, COPD, paraplegia, and dependent functional status were risk factors for stroke or death (all P < 0.05). Conclusions: This prospective database confirms that CEA is currently performed with low peri-procedural stroke rate in participating ACS NSQIP hospitals and provides a contemporary framework for comparison of other treatment modalities to CEA. Identification of the above risk factors may help with risk stratification and patient counseling for CEA.