Abstract 2217: Magnetic Resonance Blood-Brain Barrier Permeability Measurements To Predict Hemorrhagic Transformation In A Rat Model Of Ischemic Stroke

Abstract

Objectives: To evaluate the predictive value of blood-brain barrier permeability (BBBP) measurements extracted from perfusion-weighted MRI for hemorrhagic transformation (HT) in stroke. Methods: Spontaneously hypertensive rats and Wistar rats with unilateral, 2h filament occlusion of the right MCA underwent imaging during occlusion, post reperfusion at 4h and 24h. BBBP was imaged by perfusion-weighted MRI and quantified by Patlak analysis. Cresyl-violet staining was used to detect and characterize hemorrhage on histology in sacrified rats at 24 hours, immediately following the last imaging study. Permeability changes were evaluated at baseline, after reperfusion and at 24 hours, in animals with or without macroscopic hemorrhage. Receiver-operating characteristic curve (ROC) analysis was performed to determine the most accurate permeability threshold to predict HT. Findings: In animals that showed macroscopic hemorrhage at 24 hours, the 95th BBBP percentile values on the infarcted side were 0.323 [0.260, 0.387], 0.685 [0.385, 0.985], and 0.412 [0.210, 0.613] ml/min·100g (marginal mean [95%CI]) during occlusion, at 4h post reperfusion and at 24h post reperfusion, respectively. The 95th BBBP percentile values on the nonischemic, contralateral side were 0.280 [0.212, 0.348], 0.266 [0.065, 0.665], and 0.203 [0.080, 0.483] respectively. The BBBP values on the infarcted side and the contralateral side were significantly different at 4h post reperfusion (p=0.034) and at 24h post reperfusion (p=0.031). The predictive value of BBBP in terms of macroscopic hemorrhage was highest on imaging obtained 4 hours after reperfusion (ROC area under the curve = 0761). The negative predictive value of BBBP measurements was high, while the positive predictive value was limited. For instance, for a BBBP threshold of 0.35 ml/min·100ml positive and negative predictive values for imaging 4 hours post reperfusion were 7.2 % and 98.6%, respectively. Conclusion: Altered BBBP is a necessary but not sufficient condition to cause HT in rats with an infarct. Further research is needed to identify those additional risk factors that are required for HT to develop in the setting of an ischemic stroke.