ABSTRACT NUMBER: ESOC2026LB152 RELIABILITY AND GENERALISABILITY OF A FLYING INTERVENTION TEAM FOR ENDOVASCULAR THROMBECTOMY: LESSONS FROM THE FIRST 1000 DEPLOYMENTS

Abstract Background and aims The Flying Intervention Team (FIT), implemented in 2018 within the telestroke network in Southeast Bavaria, provides on-site endovascular thrombectomy for 15 regional hospitals. By August 2025, 1,000 FIT deployments had been completed. While the FIT trial showed reduced treatment times and improved functional outcomes, performance under routine large scale conditions remains uncertain. We therefore assessed the reliability and generalisability of FIT across its first 1,000 deployments. Methods We analysed all consecutive deployments between February 2018 and August 2025. Predefined time targets were decision-to-puncture <90 min and first-image-to-puncture <120 min. Additional metrics included thrombectomy rates, successful reperfusion and time to reperfusion. Stratified analyses were performed across patient and system-level variables, including deployment phase, age, baseline NIHSS, distance to the FIT hub, receiving hospital, hospital stroke volume, on/off hours, weekday/weekend, season, and telemedical vs. on-site evaluation. Exploratory multivariable regression analyses were conducted to examine associations. Results A total of 1,000 patients were included (54.6% female, median age 78). Thrombectomy was performed in 87.8%. Among these patients, successful reperfusion was achieved in 88.4%. Median decision-to-puncture time was 65 min, with 90.6% achieving <90 min. Median first-image-to-puncture time was 98 min with 76.4% achieving <120 min. Performance metrics were mostly stable across deployment phases and most subgroups. Distance to the FIT hub and weekend presentation were associated with longer decision-to-puncture times. Conclusions Across 1,000 real-world deployments, FIT maintained reliable performance under diverse patient, temporal, geographic, and structural conditions, and did not deteriorate with scale. These findings support the reliability and generalisability of the FIT model. Conflict of interest