Abstract WP304: Real-World Experience with the RAPID Platform to Screen Acute Ischemic Stroke Patients at a U.S. Comprehensive Stroke Center

• Published in: Stroke (1970) 2020-02, Vol.51 (Suppl_1)
• Main Authors: Assadi, Rami-James, Henn, Julia, Varmaz, Ajlana, Panagos, Peter, Miller-Thomas, Michelle, Ford, Andria, Goyal, Manu, Kansagra, Akash P, Heitsch, Laura
• Format: Article
• Language: English
• ISSN: 0039-2499; 1524-4628
• DOI: 10.1161/str.51.suppl_1.WP304

Abstract only Introduction: Mechanical thrombectomy (MT) is an important part of acute ischemic stroke (AIS) treatment. Recent trials of MT beyond the 6-hour window have utilized RAPID perfusion imaging for patient selection. The utility of this method is established in patients with large vessel occlusions (LVO) but screening efficiency in real-world practice remains unknown. We present the experience of a single, large volume, Comprehensive Stroke Center (CSC) utilizing RAPID to screen patients for LVO and MT. Methods: We performed a retrospective analysis of prospectively collected consecutive patients who presented to our emergency department (ED) between 01/2018 to 06/2019 with suspected LVO. Protocol was based on 2018 AHA guideline Level IA recommendations and followed DAWN and DEFUSE-3 time and imaging parameters. Patients who underwent RAPID imaging were selected for inclusion. Results: 865 patients met criteria for RAPID perfusion imaging (median age 67, females 52%, outside hospital transfers 29%). Of these, 178 (21% of total) were confirmed to have an LVO (40% ED presentation, 10% inpatient, 50% transfer). For patients presenting to the ED (N=509), 14% had an LVO (median NIHSS 13 [IQR 8-19]), of which 41% underwent MT. Mean CTP core and penumbra volume was 25mL and 100mL respectively. Number needed-to-screen in the ED cohort was 7 to detect LVO and 17 to perform MT. Transfer patients showed no significant difference in LVO detection or MT rates compared to ED patients (56%, p=0.3). Conclusions: In ED-presenting patients at a CSC, the number of RAPID perfusion imaging studies needed to detect an additional case of LVO was 7.1, and to perform an additional MT was 17.4. Current AHA Class IA recommendations for evaluation and treatment of AIS yield a reasonably high rate of LVO detection and subsequent MT in real-world practice. Additional multicenter data will be useful to establish benchmarks and improve screening efficiency.