Effectiveness of an In-Hospital Stroke Bypass System for Endovascular Treatment: A Comparative Study.

Objective: Rapid reperfusion is essential in acute ischemic stroke (AIS). The Saitama Stroke Network (SSN) facilitates the direct transfer of mechanical thrombectomy (MT) candidates to MT-capable hospitals. However, some candidates are misclassified and transferred to a general emergency department, delaying treatment. To address this, we implemented an "in-hospital stroke bypass" system. In this system, emergency medical technicians (EMTs) in the emergency department perform the emergent large vessel occlusion (ELVO) screen and directly activate the MT team, bypassing emergency physicians. This study aimed to evaluate the effectiveness of an in-hospital stroke bypass system in reducing treatment delays in patients with AIS who were initially transported as general emergency cases. Methods: We retrospectively analyzed 140 consecutive patients with AIS who underwent emergent endovascular reperfusion therapy between December 2019 and December 2021. Patients were categorized into the SSN group (n = 94), stroke bypass (SB; n = 17) group, or non-SB (n = 29) group. Baseline characteristics, door-to-puncture (D2P) and door-to-reperfusion (D2R) times, angiographic results, and modified Rankin Scale (mRS) scores at 90 days were compared among the groups. Results: Baseline demographics were similar among the groups; however, baseline National Institutes of Health Stroke Scale scores differed significantly among the 3 groups: 21 (15-26), 25 (21-28), and 27 (12-35) in SSN, SB, and non-SB groups, respectively (p = 0.04), although no significant differences were observed in post hoc pairwise comparisons. Median D2P times were 50 (38-65), 56 (50-69), and 133 (94-196) min in SSN, SB, and non-SB groups, respectively; median D2R times were 108 (80-145), 125 (98-150), and 217 (166-270) min in SSN, SB, and non-SB groups, respectively. Both times were significantly longer in the non-SB group (p <0.01). Successful reperfusion (modified thrombolysis in cerebral infarction [mTICI] grade 2b-3) occurred in approximately 75% of the patients in all groups. Functional independence (mRS 0-2 at 90 days) was achieved in 41.5% (SSN), 23.5% (SB), and 27.6% (non-SB) of patients, with no significant differences observed. Coma and the absence of conjugate gaze deviation were associated with in-hospital stroke bypass failure. Conclusion: In-hospital stroke bypass conducted by trained EMTs has effectively reduced the treatment times for MT patients who were previously overlooked, bringing them in line with the treatment times of those transferred via SSN. Establishing refined prehospital criteria for comatose patients without gaze deviation may further enhance early identification and timely reperfusion therapy.