Influence of hospital capabilities and prehospital time on outcomes of thrombectomy for stroke in Japan from 2013 to 2016.

To determine whether increasing thrombectomy-capable hospitals with moderate comprehensive stroke center (CSC) capabilities is a valid alternative to centralization of those with high CSC capabilities. This retrospective, nationwide, observational study used data from the J-ASPECT database linked to national emergency medical service (EMS) records, captured during 2013-2016. We compared the influence of mechanical thrombectomy (MT) use, the CSC score, and the total EMS response time on the modified Rankin Scale score at discharge among patients with acute ischemic stroke transported by ambulance, in phases I (2013-2014, 1461 patients) and II (2015-2016, 3259 patients). We used ordinal logistic regression analyses to analyze outcomes. From phase I to II, MTs increased from 2.7 to 5.5%, and full-time endovascular physicians per hospital decreased. The CSC score and EMS response time remained unchanged. In phase I, higher CSC scores were associated with better outcomes (1-point increase, odds ratio [95% confidence interval]: 0.951 [0.915-0.989]) and longer EMS response time was associated with worse outcomes (1-min increase, 1.007 [1.001-1.013]). In phase II, neither influenced the outcomes. During the transitional shortage of thrombectomy-capable hospitals, increasing hospitals with moderate CSC scores may increase nationwide access to MT, improving outcomes.