Differences Studied in Referral Patterns for CAS Versus CEA

January 11, 2011—Ryan B. Longmore, DO, et al have published findings from a study of clinical referral patterns for carotid artery stenting (CAS) versus carotid endarterectomy (CEA) in the Carotid Artery Revascularization and Endarterectomy Registry. The study was published online ahead of print in Circulation: Cardiovascular Interventions.

Although randomized clinical trials are the gold standard for assessing the relative benefits of CAS versus CEA, the investigators noted that observational research is necessary for determining real-world effectiveness. Current recommendations limit the application of CAS to high-risk patients, undermining the ability of physicians to balance the characteristics of patients treated with either approach. In this study, the investigators compared the clinical profiles of patients who were referred for CAS versus CEA in a large national database.

According to the investigators, clinical characteristics of 12,701 patients who were referred for CAS or CEA in the National Cardiovascular Data Registry–Carotid Artery Revascularization and Endarterectomy were compared for 44 clinical and demographic variables. To investigate the comparability of CAS and CEA patients, the cohort was stratified into quintiles of the propensity score for referral for CAS.

The investigators reported that among 8,069 patients who were referred for CAS and 4,632 patients who were referred for CEA, the CAS patients had significantly more comorbidities. The propensity model balanced most covariates but the pooled standardized differences (≥ 10%) suggested persistent imbalance for ischemic heart disease, recent myocardial infarction, and restenosis of prior CAS/CEA, all of which were more common in the CAS group. After stratification of propensity scores by quintile, CEA patients comprised only 14% of the upper two quintiles.

The investigators concluded that characteristics of patients who were referred for CAS differed markedly from those who were referred for CEA. Because of extreme clinical disparities between these patients, generalizable comparative effectiveness analyses of observational data will be difficult.