ASPIRE: Aspiration Thrombectomy as the Primary Modality in Endovascular Stroke Treatment: Pooled Analysis of the 3D Stent-Retriever, ASTER, THERAPY, COMPASS and MR CLEAN Trials.

Background: Given the overwhelming evidence for the benefit of stent-retriever thrombectomy, a future trial of primary aspiration thrombectomy (PAT) against best medical therapy (BMT) alone has become unethical and impractical. Therefore, alternative approaches are needed to further investigate the impact of PAT in the outcomes in large vessel occlusion (LVO) strokes. The aim of this project is to assess the effectiveness of PAT plus BMT (including IV t-PA if indicated) compared to BMT alone in patients with anterior circulation LVO strokes. Methods: A systematic search of the medical literature and trial registries (EMBASE, Pubmed Medline, Cochrane trial registry, clinical trials.gov and ISRCTN) was conducted to trace randomized clinical trials (RCT) that evaluated PAT as a first treatment approach for LVO strokes. Four prospective RCTs employing aspiration thrombectomy as a primary treatment modality in one arm were identified: THERAPY (n=108; 55 PAT), Penumbra 3D Stent-Retriever (n=198; 100 PAT), ASTER (n=381; 192 PAT) and COMPASS (n=270; 134 PAT). Data from the PAT arms from these four RCTs and controls from the MR CLEAN (n=500; 267 controls) and THERAPY (53 controls) trials will be pooled. Treatment effect on the primary outcome (90-day mRS ordinal shift analysis) will be analyzed with multivariable ordinal regression analysis in the propensity matched dataset, after adjusting for remaining differences in distribution of baseline prognostic factors, to increase power. Adjustments will be made for age, sex, baseline NIHSS, collateral score, pre-treatment ASPECTS, admission glucose levels, time from stroke onset to qualifying imaging. The effect will be expressed as an adjusted common odds ratio, with 95% confidence interval. Treatment effects for secondary outcomes will be estimated in a similar approach, with the appropriate regression method. Results: The data will be analyzed by an independent statistical group (Erasmus MC, University Medical Center Rotterdam, the Netherlands). The results will be available and reported at the time of the ISC 2019. Conclusion: This pooled data project will provide unique and high-level evidence concerning the effect of aspiration thrombectomy on functional outcome after acute ischemic stroke.