Abstract Number: Esoc2026a623 From Delay to Delivery: Improving Thrombolysis Delivery in a Uk Rural District General Hospital

Abstract Background and aims Delivering timely hyperacute stroke care depends on how well complex systems work together. In district general hospitals, fragmented workflows, competing priorities, and unclear role allocation frequently delay time-critical treatment. At our centre, these system inconsistencies resulted in prolonged door-to-needle times and thrombolysis rates below national benchmarks. We set out to redesign the hyperacute stroke service, focusing on reliability rather than individual performance. Methods Using a systems-of-care improvement approach, a multidisciplinary team mapped the entire stroke pathway, identifying key points of delay including communication handovers, task overload, inconsistent imaging access, and variable decision-making. We implemented a standardised Code Stroke “pit-stop” model, integrating pre-hospital pre-alerts, predefined multidisciplinary roles, prioritised CT access, treatment delivery at scanner, and early stroke unit admission. Performance data were reviewed prospectively and fed back in real time to support iterative pathway refinement. Results Following implementation, Code Stroke activations increased progressively from 18 at six weeks to 36 by fifteen weeks, reflecting improved pathway utilisation and reliability. Median door-to-needle time reduced from a baseline average of 75 minutes to 36 minutes at early review and 34 minutes at 15 weeks. More than half the patients were scanned within 20 minutes. At organisational level, thrombolysis rates increased from 8% to 12%, indicating sustained system-level improvement. Conclusions Redesigning the hyperacute stroke system of care, rather than targeting isolated steps, delivered rapid and sustained improvements in treatment timeliness in a rural stroke centre. This work demonstrates how standardisation, multidisciplinary coordination, and continuous feedback can strengthen stroke services and offer a scalable model for similar services. Conflict of interest Junaid Khan: Nothing to disclose, Sarah Hunter: Nothing to disclose, Nicola Hewett: nothing to disclose, Carmen Constantin: nothing to disclose, Abduelbaset Elmarimi: nothing to declare