Basilar artery occlusion management: specialist perspectives from an international survey

Abstract

Background and purpose: Two early basilar artery occlusion (BAO) randomized controlled trials did not establish the superiority of endovascular thrombectomy (EVT) over medical management. While many providers continue to recommend EVT for acute BAO, perceptions of equipoise in randomizing patients with BAO to EVT versus medical management may differ between clinician specialties.

Methods: We conducted an international survey (January 18, 2022 to March 31, 2022) regarding management strategies in acute BAO prior to the announcement of two trials indicating the superiority of EVT, and compared responses between interventionalists (INTs) and non-interventionalists (nINTs). Selection practices for routine EVT and perceptions of equipoise regarding randomizing to medical management based on neuroimaging and clinical features were compared between the two groups using descriptive statistics.

Results: Among the 1245 respondents (nINTs = 702), INTs more commonly believed that EVT was superior to medical management in acute BAO (98.5% vs. 95.1%, p < .01). A similar proportion of INTs and nINTs responded that they would not randomize a patient with BAO to EVT (29.4% vs. 26.7%), or that they would only under specific clinical circumstances (p = .45). Among respondents who would recommend EVT for BAO, there was no difference in the maximum prestroke disability, minimum stroke severity, or infarct burden on computed tomography between the two groups (p > .05), although nINTs more commonly preferred perfusion imaging (24.2% vs. 19.7%, p = .04). Among respondents who indicated they would randomize to medical management, INTs were more likely to randomize when the National Institutes of Health Stroke Scale was ≥10 (15.9% vs. 6.9%, p < .01).

Conclusions: Following the publication of two neutral clinical trials in BAO EVT, most stroke providers believed EVT to be superior to medical management in carefully selected patients, with most indicating they would not randomize a BAO patient to medical treatment. There were small differences in preference for advanced neuroimaging for patient selection, although these preferences were unsupported by clinical trial data at the time of the survey.