Abstract 119: Serial Analysis of Perfusion Angiography in Endovascular Intervention for Acute Stroke With Proximal Middle Cerebral Artery Occlusion

Abstract

Background: Perfusion imaging is often used to predict tissue fate during acute stroke in clinical practice and in research trials. During endovascular interventions, conventional angiography provides perfusion information at the actual time of reperfusion therapy yet scoring reperfusion remains operator-dependent and limited by scale properties. Here, a quantitative analysis was conducted to evaluate serial changes in perfusion during treatment. Methods: Angiography was analyzed from 118 acute MCA occlusion cases with attempted endovascular reperfusion at UPMC or UCLA. Parametric maps for MTT, TTP, Tmax, CBV, and CBF were extracted with perfusion angiography software (PerfAngio) from angiograms acquired at the time of treatment. Delta maps were created representing the serial change in perfusion parameters across treatment. Region of interest analysis (ROI) performed on the MCA distribution quantified the average change in perfusion. Results: Delta maps allowed for spatial characterization of the change in perfusion that occurs as a consequence of endovascular therapy as shown in the figure. ROI analysis performed for the first 35 cases demonstrated a wide variance in perfusion change ranging from 52% decrease in CBF to 476% increase in CBF. Conclusions: Serial changes extracted using perfusion angiography provide a quantitative, continuous measure of tissue reperfusion resulting from an endovascular therapy. Such perfusion data are more spatially informative and may be more amenable to statistical analysis than the standard categorical scale of reperfusion (TICI) following endovascular intervention.