Abstract P505: Prediction of Infarct Volume at the 24 Hours After Late Window Presentation With Perfusion Imaging in Patients With Anterior Circulation Large Vessel Occlusion

Abstract

Introduction: We sought to evaluate the accuracy of perfusion-weighted imaging (PWI) in late presenting patients for estimating the infarct volume at 24 hours after presentation. Methods: This is a secondary analysis of DEFUSE 3, which included stroke patients with anterior circulation occlusion within 6-16 hours of last known normal. The primary outcome is the final infarct volume on a 24-hour MRI scan (volume of DWI positive tissue), adjusted for the baseline infarct volume. We censored 3 patients with 24-hour follow-up MRI infarct volumes >300 mL, which we considered non-physiologic for a hemispheric stroke. The primary predictors are the baseline volume of Tmax >6s, Tmax >10s, and hypoperfusion intensity ratio (HIR: Tmax10/Tmax6) on CT/MR perfusion at hospital admission. We stratified the cohort into 4 categories (untreated, TICI 0-2a, TICI 2b, and TICI3) and fit linear regression models to each of our predictors. Results: We included 147 patients, of which 69 were untreated, 17 had TICI 0-2a, 46 had TICI 2b, and 15 had TICI 3. In untreated patients, both HIR and Tmax10 volume were predictive of adjusted 24-hour follow-up infarct volume (Table). In treated patients, there were no consistent relationships between the perfusion imaging variables and adjusted final infarct volume (Table). Conclusion: For patients with late window anterior circulation large vessel occlusion stroke, HIR and Tmax10 volume appear to be reliable predictors of subsequent infarct volume in untreated patients. For patients treated with thrombectomy, further research is warranted to better understand the more complex relationship between baseline perfusion imaging and the 24 hour, and beyond, infarct volume.