Abstract W P22: Examining the Effects of Intra-Arterial Therapy in Acute Posterior Circulation Occlusions

Abstract

Background: Recent trials have failed to show benefit of intra-arterial therapy (IAT) in acute ischemic stroke (AIS) patients. Patients with posterior circulation occlusions, however, were vastly underrepresented (only 4 patients in IMS3, MR-RESCUE and SYNTHESIS combined). IAT is often pursued, due to the inherent poor outcome associated with brainstem infarction. We sought to evaluate potential benefit with IAT in this population. Methods: A US multicenter retrospective cohort of 106 patients with AIS and LAO in the vertebrobasilar distribution presenting 01/05-3/14 within 12 hours from stroke onset was examined. Patients were divided into two groups (IAT vs No IAT). Logistic regression with good and poor (discharge mRS 0-2 and 4-6) outcome was performed. Univariate and multivariate analyses evaluated independent predictors of good outcome after IAT. Results: Patients in the IAT group (n=57) had more severe strokes (NIHSS 24 vs 12) than the medical management group (n=49); otherwise the two groups were comparable (Table 1). IAT was not associated with improved odds of good outcome (OR 0.76, 95%CI 0.27-2.14, p=0.60), even after adjusting for admission NIHSS, recanalization, and time to IAT start (OR 1.35, 95%CI 0.39-4.68, p=0.63). The subgroup with basilar occlusions also did not benefit from IAT. Stroke severity predicted treatment with IAT; each additional point in NIHSS was associated with 5% increased chances of IAT (P 0.07). There were more deaths in the IAT group (Figure 1); 73% of these, however, were associated with care withdrawal decisions. Conclusion: While vascular neurologists intervene more often on severe posterior circulation strokes, our results suggest poor outcome was not averted, even after adjusting for stroke severity. While our results are affected by the inherited limitations of a retrospective cohort and selection bias of a non- randomized design, it addressed an underrepresented stroke population. RCTs are needed to validate these results.