Abstract 125: Marked Variation in Perfusion Imaging Definitions of No-Reflow Phenomenon after Thrombectomy

Introduction: The no-reflow phenomenon (persistent microvascular hypoperfusion despite macrovascular angiographic reperfusion) represents an explanation to poor outcome despite successful thrombectomy. There remains no universally-accepted definition to standardise future studies. We aim to compare the clinical features and outcomes of patients identified as having no-reflow using different perfusion MRI/CT definitions. Methods: We performed a pooled analysis of thrombectomy patients who underwent 24-hour follow-up perfusion MRI or CTs in the EXTEND-IA, EXTEND-IA TNK part 1 and 2 RCT. Presence of no-reflow was defined according to four definitions identified from a meta-analysis of 13 studies (Definition A = eTICI2c-3 and >15% asymmetry in CBV or CBF within the infarct on follow-up perfusion MRI/CT; definition B = mTICI2c-3 and >40% CBF asymmetry, definition C = mTICI2b-3 and presence of a Tmax>6s lesion; Definition D = mTICI2b-3 and >90% reduction of baseline Tmax>6s lesion). Receiver Operating Characteristics (ROC) analysis was performed with the outcome variable being poor functional outcome at 90 days (mRS≥3). Results: Of 325 patients analysed, the prevalence of no-reflow varied between definitions from 1.9 to 29.3% (p<0.001). There was poor agreement between definitions (kappa 0.062-0.745, 5 out 6 comparisons <0.196). Among patients identified as exhibiting no-reflow by any definition, there were significant differences in the intralesional interside differences in CBF (p=0.006), CBV (p<0.001) and MTT (p=0.005). Definition A yielded the highest Area Under the ROC Curve (AUC=0.679) for discrimination of 90-day functional outcome (Definitions C=0.649, D=0.597, B=0.515; p<0.0001). Sensitivity analyses testing across the eTICI≥2b, eTICI≥2c and eTICI3 strata showed consistent results. Conclusions: Existing imaging definitions of no-reflow varied significantly in prevalence and post-treatment perfusion imaging profile, suggesting that patients classified as having no-reflow by various definitions differ in their underlying pathophysiological processes. Definition A (eTICI2c-3 & >15% CBV/CBF asymmetry) discriminated prognostic performance best, supporting its use as the reference no-reflow imaging definition.