Treatment-Associated Stroke in Patients Undergoing Endovascular Therapy in the ARUBA Trial

Background and Purpose: Since the publication of ARUBA trial (A Randomized Trial of Unruptured Brain Arteriovenous Malformations), outcomes in treated and untreated patients with unruptured arteriovenous malformation have been thoroughly compared. However, no prior analysis of ARUBA patients has sought to identify risk factors for perioperative stroke. Improved understanding of risks within the ARUBA cohort will help clinicians apply the study’s findings in a broader context. Methods: The National Institute of Neurological Disorders and Stroke database was queried for all data relating to ARUBA patients, including demographics, interventions undertaken, and timing of stroke. Retrospective cohort analysis was performed with the primary outcome of perioperative stroke in patients who underwent endovascular intervention, and stroke risk was modeled with multivariate analysis. Results: A total of 64 ARUBA patients were included in the analysis. One hundred and fifty-ninth interventions were performed, and 26 (16%) procedures resulted in stroke within 48 hours of treatment. Posterior cerebral artery supply (adjusted odds ratio, 4.42 [95% CI, 1.23–15.9], P =0.02) and Spetzler-Martin grades 2 and 3 arteriovenous malformation (adjusted odds ratio, 7.76 [95% CI, 1.20–50.3], P =0.03; 9.64 [95% CI, 1.36–68.4], P =0.04, respectively) were associated with increased perioperative stroke risk in patients who underwent endovascular intervention. Patients treated in the United States or Germany had a significantly lower stroke risk than patients treated in other countries (adjusted odds ratio, 0.18 [95% CI, 0.04–0.82], P =0.02). Conclusions: Knowing patient and lesion characteristics that increase risk during endovascular treatment can better guide clinicians managing unruptured brain arteriovenous malformation. Our analysis suggests risk of perioperative stroke is dependent on Spetzler-Martin grade and posterior-circulation arterial supply. Differences in regional treatment paradigms may also affect stroke risk.