Population health impact of extended window thrombectomy in acute ischemic stroke

Background: Recent trials support endovascular thrombectomy (EVT) in select patients beyond the conventional 6-hour window. Objective: In this work, we estimate the impact of extended window EVT on procedural volumes and population-level clinical outcomes using Monte Carlo simulation. Methods: We simulated extending EVT eligibility in a system comprising an EVT-incapable primary stroke center (PSC) and EVT-capable comprehensive stroke center (CSC) using routing paradigms that initially direct patients to (1) the nearest center, (2) the CSC, or (3) either CSC or nearest center based on stroke severity. EVT eligibility and outcomes are based on HERMES, DEFUSE-3, and DAWN studies in the 0-6, 6-16, and 16-24 hour windows, respectively. Probability of good clinical outcome is determined by type and timing of treatment using clinical trial data. Results: Relative increase in EVT volume in the three tested routing paradigms was 15.7-15.8%. The absolute increase in the rate of good clinical outcome 0.4% in all routing paradigms. NNT for extended window EVT was 239.9-246.4 among the entire stroke population. Conclusion: Extended window EVT with DEFUSE-3 and DAWN criteria increases EVT volume and modestly improves population-level clinical outcomes.