Abstract WMP7: Poor Reperfusion After Endovascular Therapy is Associated With Worse Outcomes Than No Intervention in the HERMES Analysis of Seven Randomized Trials

Abstract

Introduction: Although the role of successful reperfusion in improved outcome after endovascular treatment (EVT) for stroke is known, there has been limited evaluation of the effects of poor reperfusion. We compared results in the HERMES collaboration between standard care and those in the EVT group with poor reperfusion. Methods: Patient-level data were pooled in this meta-analysis of seven randomized trials comparing endovascular thrombectomy with standard care in anterior ischemic stroke. Functional outcome was assessed by ordinal analysis of the modified Rankin scale (mRS) at 90 days, comparing standard care versus EVT subjects achieving mTICI of 0 or 1 post-procedure. Analyses were adjusted for baseline prognostic variables to correct for potential imbalances. Results: The meta-analysis included 877 subjects in the standard care group and 78 in the EVT group with mTICI 0-1 per the HERMES central imaging core laboratory. At baseline, the EVT cohort had higher NIHSS (median 19 vs 17, p=0.011), but was not significantly different from standard care in other characteristics including age, sex, ASPECTS, time to randomization, site of occlusion and alteplase administration. Subjects with poor reperfusion in the EVT group had worse mRS at 90 days than standard care, unadjusted (p=0.003) and after adjustment for baseline characteristics (common odds ratio 0.59 (0.38-0.91), p=0.016). Fewer subjects in the EVT poor-reperfusion cohort achieved mRS 0, mRS 0-1 and all other dichotomized mRS cutpoints (Figure 1). Symptomatic intracranial hemorrhage was not different between groups (3.9% vs 3.5%, p=0.75). Conclusion: In recent endovascular trials, poor reperfusion after EVT was associated with worse outcomes than standard care in recent endovascular trials. This suggests that additional efforts to achieve reperfusion by EVT should be encouraged if deemed safe to do so.