Reader response: Lower carotid revascularization rates after stroke in racial/ethnic minority-serving US hospitals.

Abstract

I read the article by Faigle et al.1 about the overall low rates of revascularization for symptomatic carotid stenosis in all groups of patients. Even under the best scenario of whites in a predominantly white-serving hospital, the predicted rates were <25% (table 3),1 suggesting a broad underutilization of this procedure nationwide and even more so in underserved hospitals and groups in the population. Our local practice includes carotid endarterectomy (CEA)/carotid artery stenting (CAS) in all high-grade symptomatic patients within 3 days after an infarct, unless the stroke is major and disabling or the patient refuses. With this in mind, and because rates were so low in this case series, methodologic issues in the study may exist. There likely is misclassification of patients as having symptomatic high-grade internal carotid artery (ICA) stenosis that requires intervention by using International Classification of Diseases, Ninth Revision, Clinical Modification codes 433.11 and 433.31 only—the validity of this code for high-grade symptomatic stenosis was not described, and this would also have been helpful information for codes on CEA/CAS. These codes likely include low-grade stenosis, which can be used by coders to justify 433.xx, as well as potentially intracranial ICA stenosis, for which revascularization is not indicated, and are more likely to be present in nonwhites. The probability of misclassification of mild stenosis or of intracranial artery stenosis (patients who should not be undergoing revascularization; more likely nonwhite), as CEA/CAS candidates, is a likely source of bias and could explain the results. Whites are also more likely to undergo CEA/CAS if asymptomatic,2 potentially due to socioeconomic factors, and the 433.xx has been used for asymptomatic patients to justify treatment or after developing a postoperative stroke.3 Last, patients who were transferred were excluded, but the reason for transfer was not explored; if they had symptomatic disease that needed surgery, they were indeed treated appropriately by the hospital that was otherwise low performing.