Association of Workflow Times and Clinical Outcome in Acute Ischemic Stroke With and Without Intravenous Alteplase: DIRECT-MT Results

Background: DIRECT-MT demonstrated the noninferiority of endovascular thrombectomy (EVT) alone compared to combination-therapy in large vessel occlusions stroke patients presenting directly at EVT capable centers. We aim to assess the influence of time from stroke onset to reperfusion (TOR), from stroke onset to randomization (TORnd), and form randomization to reperfusion (TRR) on the treatment effect of intravenous thrombolysis (IVT) prior to EVT. Methods: We used ordinal logistic regression with multiplicative interaction terms to assess the influence of TOR, TORnd, and TRR on the treatment effect. For binary outcomes, absolute risk differences were calculated as differences of adjusted outcome proportions from ordinal logistic regression models. Findings: Among all 656 patients, The TORnd was 171 minutes (interquartile range [IQR], 125-210), TRR was 98 minutes (IQR, 72·5-136), and TOR was 273 minutes (IQR, 225-326·5). Every 30-minute increase in TOR was associated with higher levels of disability (odds ratio [OR],0·88 [95% confidence interval, CI 0·79, 0·97] in the thrombectomy-alone group and 0·80[0·72, 0·89] in the combination-therapy group). The acOR of a less-disabled outcome with thrombectomy-alone became statistically significant when TOR was >5·2 hours (acOR[95%CI], 1·73[1·01, 2·45]) or time from randomization to successful reperfusion was > 2·0 hours (1·89[1·01, 2·77]). An interaction between time from randomization to successful reperfusion (extended Thrombolysis in Cerebral Infarction score≧2b) and treatment allocation existed (p =0·036), but not between TOR or TORnd and treatment (P =0·092 and 0·505).  Interpretation: Clinical outcomes declined more steeply with increasing reperfusion-relevant interval times in patients receiving combination therapy compared to thrombectomy alone, and the difference in outcomes in favor of thrombectomy-alone became statistically significant when the TOR exceeded 5·2 hours and when the randomization to successful reperfusion time exceeded 2·0 hours. Funding Statement: This study was funded byStroke Prevention Project of the National Health Commission of the People’s Republic of China and Wu Jieping Medical Foundation. Declaration of Interests: PY reports grants from National Natural Science Foundation of China, grants from Shanghai Municipal Health Commission, outside the submitted work. JL reports grants from Stroke Prevention Project, National Health Commission of the People’s Republic of China, grants from Wu Jieping Medical Foundation, during the conduct of the study; grants from National Natural Science Foundation of China, grants from Shanghai Municipal Health Commission, grants from Shanghai Science and Technology Committee, outside the submitted work. MG reports grants from Stryker, personal fees from Medtronic, personal fees from Microvention, personal fees from Mentice, personal fees from GE Healthcare, outside the submitted work. CM reports grants from CVON/Dutch Heart Foundation, grants from European Commission, grants from Dutch Health Evaluation program, grants from TWIN foundation, grants from Stryker, outside the submitted work; and Shareholder of Nico-lab, of company that focuses on the use of artificial intelligence for medical image analysis (modest). YR reports grants from CVON/Dutch Heart Foundation, grants from Stryker, grants from Dutch Health Evaluation program, outside the submitted work; and Shareholder of Nico-lab, of company that focuses on the use of artificial intelligence for medical image analysis (modest). All other authors declare no competing interests. Ethics Approval Statement: The detailed protocol was previously published, and was approved by all relevant local ethics committees and research boards. Written informed consent was obtained before enrollment from all the patients or their legal representatives.