Abstract DP191: Endovascular therapy takes longer for MeVO EVT than LVO EVT. This delay is related to prolonged decision-to-treat rather than procedural time.

Introduction: Recent trials of endovascular therapy (EVT) for medium vessel occlusion (MeVO) failed to show benefit of EVT. Interestingly, MeVO EVT may be disadvantaged compared to Large Vessel Occlusion (LVO) EVT because of longer time to achieve reperfusion as was observed in recent MeVO trials (Goyal et al NEJM 2025, Psychogios et al NEJM 2025). However, the relative contribution of decision time and procedure time has not been systematically evaluated. Here we we examined a real world sample to determine if MeVO EVT takes longer to achieve than LVO EVT. If so, is this related to longer decision-making or longer procedural time? Methods: We performed a retrospective analysis of patients with acute ischemic stroke from a multi-state registry. Patients discharged between January 2018 and May 2025, age ≥18, with large vessel occlusion (LVO) or middle vessel occlusion (MEVO) were included. To assess differences in DTD times for MEVO vs. LVO patients, a mixed effects linear regression model was used, adjusting for age, sex, race, medical history, patient mode of arrival, administration of IV thrombolytic, and last known well to arrival in minutes. To investigate whether any difference in DTD time was related to longer procedure or decision-making times respectively, we also compared median differences in door to groin (DTG) and groin to device (GTD) times in minutes using Wilcoxon rank sum tests. P-values were considered significant at p = .05 Results: The study population included 3,050 patients; 1,066 (35%) were MEVO and 1,984 (65%) were LVO. DTD times were 9.0 minutes longer (95% CI = 5.9, 12.0, p < .001) for patients with MEVO compared to those with LVO. While median DTG times were also significantly longer for MEVO at 88.0 [67.0, 110.0] minutes vs. 80.0 [63.0, 102.0] minutes for LVO ( p <.001), there was no difference in median GTD times between the two groups (18.0 [11.0, 28.0] minutes for MEVO vs. 17.0 [11.0, 26.0] minutes for LVO; p = .06). Conclusions: Door to device times were longer for MeVO EVT compared to LVO EVT. This delay persisted in the model which corrected for potential confounders (Figure 2). Furthermore, this delay is entirely accounted for by longer decision workflow as measured by DTG. Procedure times on the other hand were the same from MeVO EVT and LVO EVT. Future attempts to maximize benefit of MeVO EVT may be best focused on improving time-to-treatment decisions.