Perfusion Imaging Collateral Scores Predict Infarct Growth in Non-Reperfused DEFUSE 3 Patients

Objective: This study evaluated the associations of perfusion imaging collateral profiles with radiographic and clinical outcome in late presenting, non-reperfused patients in the DEFUSE 3 clinical trial. Methods: Non-reperfused patients in both treatment arms were included. Baseline ischemic core, Tmax >6s, and Tmax >10s perfusion volumes were calculated with RAPID software; infarct volumes obtained 24 hours after randomization were manually determined from DWI or CT. Substantial infarct growth was defined as a >25mL increase between baseline and 24-hour follow-up. Hypoperfusion Intensity Ratio (HIR) was defined as the proportion of the Tmax >6s lesion with Tmax >10s delay; CBV index was calculated by RAPID from mean CBV values within the Tmax >6s lesion compared to regions of normal CBV. Results: Eighty-four patients were included. ROC analysis showed HIR >0.34 (AUC=0.68) and CBV index <0.74 (AUC=0.72) optimally predicted substantial infarct growth in follow-up. Median growth was 23.4 versus 73.2mL with HIR threshold of 0.34 (p=0.005), and 24.3 versus 58.7mL with CBV index threshold of 0.74 (p=0.004). If baseline HIR and CBV index were both favorable, median growth was 21.7mL, 40.9mL if one was favorable, and 108.2mL if both were unfavorable (p<0.001). Baseline perfusion profile was not associated with 90-day functional outcome. Conclusions: Perfusion collateral scores forecast infarct growth in late presenting, non-reperfused ischemic stroke patients. These parameters may be useful for guiding transfer decisions, such as need for repeat imaging upon thrombectomy center arrival, and may help identify slow progressing patients more likely to have persistent salvageable ischemic tissue beyond 24 hours.