Abstract T MP1: The Pittsburgh Collateral Ratio: A Novel Method to Assess Collateral Circulation Status in Acute Mca Occlusions

Introduction: Collateral status is a robust predictor of recanalization and clinical outcome in acute ischemic stroke and numerous collateral grades exist to evaluate it (many of which are qualitative or semi-quantitative) being CT angiography (CTA)-based collateral grading the most frequently used. We attempt to quantitatively assess collateral grade based on a comparison between contrast filling proximal and distal to a medial cerebral artery (MCA) occlusion. Material and methods: We retrospectively reviewed the data of patients with M1 MCA occlusions referred to our center between January 15, 2006 and January 15, 2014. The Pittsburgh Collateral Ratio (PCR) was determined by calculating the mean Hounsfield Units (HU) in regions of interest (ROI) in the MCA, both pre and post-thrombus at the time of maximal opacification of the vessel (PCR = HU post-thrombus/ HU pre-thrombus). Ipsilateral (iPCR) and contralateral (cPCR) ratios were calculated. Two independent reviewers (CA and WD) graded pre-treatment CTA collateral status as good (the entire MCA reconstitutes with contrast) and poor (partial or no reconstitution of the distal MCA). PCR was compared to CTA collateral status and associated with good outcome (mRS 0-2 at 3 months) by logistic regression analysis. Results: Fifty one patients were included (mean age 65±3 yrs, median initial NIHSS 17 [IQR 14-21]). Strong correlation between PCR and CTA collateral grade was observed (iPCR: Pearson’s R=-0.796, p<0.001, cPCR: Pearson’s R=-0.64, p<0.001). iPCR was highly predictive of good collateral status (AUC 0.89, p<0.001) and iPCR ≥ 0.6 had a 100% sensitivity and 64% specificity for good collateral status. In multivariate analysis after controlling for age, NIHSS and ivTPA, iPCR was an independent predictor of good outcome (p=0.02). Conclusions: The PCR is a simple and accurate method to evaluate collateral circulation status in MCA occlusion patients and further validation of this score in larger cohorts is warranted. ![][1] [1]: /embed/graphic-1.gif