Abstract DP023: Quantitative Biomarker Profiling of No-Reflow Regions Following Excellent Reperfusion

Background: Despite successful macrovascular reperfusion following endovascular treatment patients with large vessel occlusion (LVO) acute ischemic stroke (AIS), persistent microvascular hypoperfusion known as “no-reflow” can occur in 20–30% of patients and is associated with poor outcomes. We aimed to characterize the MRI biomarker profile of no-reflow regions on posttreatment MRI and evaluate the predictive value of individual and combined parameters. Methods: In this retrospective study, patients were included if they had anterior circulation LVO-AIS, who achieved excellent reperfusion (TICI ≥2c) and had posttreatment MRI including dynamic susceptibility contrast (DSC) perfusion within 48 hrs of reperfusion. DSC was processed with a novel vascular model using Bayesian voxel-wise transit time distribution to generate perfusion maps including capillary transit-time heterogeneity (CTH) and cerebral metabolic rate of oxygen (CMRO 2 ). Final infarct was segmented on diffusion-weighted MRI. Segmentation masks and perfusion maps were coregistered via linear registration, and voxel values extracted. All voxel values were normalized to median value of contralateral region-of-interest to calculate relative measures. No-reflow was defined as voxel values with relative CBV (rCBV) and CBF (rCBF) ≥15% below contralateral median. Group comparisons, multivariate logistic regression, and ROC analyses were performed. Results: A total of 53 patients met the inclusion criteria. Among 42,761 voxels defined as infarction on final MRI, a total of 14,765 (34%) designated as no reflow voxels. Ischemic tissue with no-reflow had significantly higher rCTH and Tmax, and significantly lower rCMRO 2 (all p < 0.001) compared to the completely reperfused tissue. In multivariable logistic regression, lower rCMRO 2 (OR = 0.14; 95% CI, 0.14–0.15; p < 0.001), longer Tmax (OR = 0.99; 95% CI, 0.98–1; p=0.04), and higher rCTH (OR = 2.4; 95% CI, 2.3–2.5; p < 0.001) were independently associated with no-reflow. The combined model of these 3 predictors demonstrated good discriminative performance (AUC = 0.87; 95% CI, 0.86–0.87). The optimal Youden threshold corresponded to rCMRO 2 ≤ 0.98, Tmax ≥ 0.2, and rCTH ≥ 1.2, yielding a sensitivity of 80% and specificity of 78%. Conclusion: No-reflow regions exhibit a distinctive MRI biomarker profile of elevated Tmax and CTH, as well as reduced CMRO 2 . A combined multi-parametric model provides superior discrimination compared to individual predictors.