Abstract A085: Development of the Medium and Distal Mechanical Thrombectomy (MAD-MT) Score to Optimize Patient Selection for Endovascular Treatment

Background and Objective: While endovascular thrombectomy (EVT) is well-established for large vessel occlusion strokes, its role in distal and medium vessel occlusion (DMVO) strokes remains uncertain. In this study, we aimed to develop a Medium and Distal Mechanical Thrombectomy (MAD-MT) score that integrates clinical need and procedural risk to guide patient selection for DMVO EVT. Methods: This was a retrospective cohort analysis of an international registry of patients with DMVO strokes (M2/M3/M4, A2/A3, or P1/P2/P3 occlusion) across 37 sites. Patients were treated with either medical management (MM) or EVT. Independent predictors of poor functional outcome (modified Rankin Scale [mRS] >2 at 90 days) in the MM cohort and predictors of EVT failure or complications in the EVT cohort were identified using multivariable logistic regression. Predictors of poor outcome in the MM cohort were assigned positive weights (clinical need), while predictors of EVT-related procedural failure or harm were assigned negative weights (procedural risk), and weights were summed to create the MAD-MT score. Interaction analyses were used to assess heterogeneity of EVT treatment effect as measured by 90-day mRS across MAD-MT score levels. Results: A total of 1,007 patients were included (EVT: 822; MM: 185); 71.5% had M2 occlusion, 13.8% had M3/M4 occlusion, and 9.9% had PCA occlusion. Higher NIH Stroke Scale (+1 per point) and lack of intravenous thrombolysis (+7 points) were associated with clinical need, and older age (−1 point per 15 years above age 25), lack of hypertension (-2 points), and lack of atrial fibrillation (-2 points) were associated with procedural risk. Overall, the MAD-MT score (range -8 to 49) significantly modified the effect of EVT vs. MM on 90-day functional independence (p-interaction=0.032; Figure 1A) and bedridden state or death (p-interaction=0.024; Figure 1B). Among patients with high MAD-MT score (≥15, n=293, 29%), EVT was associated with better 90-day mRS compared to MM (median 3 vs. 4, p=0.009; Figure 2); in contrast, EVT was associated with worse 90-day mRS compared to MM in low MAD-MT score patients (<15, n=710, 71%; median 2 vs. 1, p=0.014; Figure 2). Conclusions: The MAD-MT score is a pragmatic, clinically interpretable tool that identifies DMVO stroke patients most likely to benefit from EVT while minimizing risk. This score may promote patient-centered decision-making and inform future trial design in this heterogeneous population.