Abstract P321: The Cerebral Collateral Cascade: Rethinking the Assessment of Vascular Pathways in Acute Ischemic Stroke Patients

Abstract

Background and Purpose: Robust collateral blood flow in patients with acute ischemic stroke due to large vessel occlusion (AIS-LVO) has been correlated with favorable outcomes. Collaterals are commonly assessed by the number of arteries present on non-invasive CT angiography (CTA) overlying ischemic brain, but blood transit from these arteries through the brain tissue and the venous drainage is not assessed by this technique. We hypothesized that a comprehensive imaging analysis of the cerebral collateral cascade (CCC) would predict clinical and radiological outcomes in patients with AIS-LVO patients. Materials and Methods: Multicenter retrospective cohort study of AIS-LVO patients undergoing thrombectomy triage. CCC was determined on pre-treatment imaging by scoring for pial arterial collaterals, tissue-level collaterals (TLC), and venous outflow. Pial arterial collaterals were determined by CTA (Tan scale), TLC were assessed on CT perfusion data using the Hypoperfusion Intensity Ratio, and venous egress was assessed on CTA using the cortical vein opacification score system. 3 groups were defined: CCC+ (good pial collaterals, TLC, and venous perfusion), CCC- (poor pial collaterals, TLC, and venous perfusion) and CCCmixed (reminder of patients). Primary outcome was a good functional outcome (modified Rankin Scale [mRS] 0-2 at 90 days). Secondary outcome was final infarct volume. Results: 647 patients met inclusion criteria: 176 CCC+, 345 CCCmixed and 126 CCC-. Multivariate ordinal logistic regression showed that CCC+ predicted good functional outcomes (mRS 0-2: OR=20.8 [95% CI 9.3-46.8]; p<0.001) compared to CCC- and CCCmixed. CCCmixed (β: 27.1, SE: 7.7; p<0.001) and CCC- (β: 86.6, SE: 9.9; p<0.001) profiles were associated with higher final infarct volumes after treatment compared to CCC+. Conclusion: Comprehensive assessment of the collateral blood flow cascade in AIS-LVO patients is a strong predictor of clinical and radiological outcomes in AIS-LVO patients.