Novel Vascular Territory Mapping Algorithm as a Predictive Tool for Identification of Antegrade Flow in Middle Cerebral Artery Occlusion

Background: Vascular territory mapping (VTM) software estimates which intracerebral vessel provides predominant arterial flow to a brain voxel. The presence of antegrade flow in the setting of acute middle cerebral artery (MCA) occlusion is associated with improved outcomes. We identify whether VTM software is a determinant of antegrade flow in patients with proximal MCA occlusion.

Methods: Consecutive patients with the first branch of MCA (M1) occlusion were analyzed from the International Stroke Perfusion Imaging Registry with their computed tomography perfusion reprocessed through VTM software. Authors reviewed dynamic 4-dimensional computed tomography angiography for the presence of an acute M1 thrombus and also for the presence of residual antegrade flow. The VTM software assigned regions of the brain to an estimated feeding vessel (anterior cerebral artery, MCA, and posterior cerebral artery). A binomial logistic regression was performed to determine the effects of VTM, ischemic core (cerebral blood flow <30), and perfusion lesion (delay time >3) on the likelihood that patients had the presence of any antegrade flow in the MCA territory. A secondary analysis was performed to assess the relationship between imaging variables and 3-month modified Rankin Scale outcomes.

Results: The final data set included 130 patients with M1 occlusion. The median age of participants was 74 years (interquartile range, 62-81) with an onset-to-scan time of 2.1 hours (interquartile range, 1.4-3.8) and a National Institutes of Health Stroke Scale score of 15 (interquartile range, 12-20). Eighteen patients were identified with antegrade flow on 4-dimensional digital subtraction angiography. Patients with antegrade flow had significantly larger VTM volume normal side MCA, 101 (72-180) mL, compared with those with complete occlusion, 41 (21-71) mL. VTM volume normal side MCA volume significantly predicted antegrade flow and outcome, and 1 mL VTM volume normal side MCA volume increased odds of antegrade flow by 1.024 (95% CI, 1.013-1.036). Ischemic core and the perfusion lesion volumes did not predict antegrade flow.

Conclusions: VTM software was more effective than traditional perfusion parameters in the detection of antegrade flow. The results demonstrate a potential clinical utility for VTM; however, larger cohorts will be required to detect whether VTM can predict clinical outcome after reperfusion treatment.