Abstract WP57: Perfusion Angiography Improves Clinical Outcomes Prediction in Endovascular Stroke Therapy

Abstract

Background: Successful reperfusion is a key determinant of outcome in endovascular stroke therapy (ET). However, present methods of grading treatment success using the Thrombolysis in Cerebral Infarction (TICI) scale fail to measure quantitative changes in cerebral blood flow and volume (CBF and CBV) and as such, may not fully represent treatment effect. Methods: From our prospectively maintained institutional registry, we identified patients treated with ET between February 2014 and May 2016. CBF and CBV maps were calculated automatically for both AP and Lateral projections and regions of interest (ROIs) were drawn by two experienced neuroimagers over the middle cerebral artery territories. Delta CBF and CBV scores were determined by subtracting pre- from post-intervention maps and averaging over the ROIs. Non-linear regression was used to calculate correlations against clinical outcome (modified Rankin scale at discharge). Results: Among 104 patients treated with ET, average age was 70, 50% were female, and median presentation NIHSS was 16 (IQR 10-19). Target occlusion location was ICA in 14%, M1 in 67%, and M2 in 18%. TICI scores ranged from 0 (4%), 1 (13%), 2a (2%), 2b (22%), 3 (58%). Relative increases in CBF and CBV ranged from 0.4-17% (CBF) and 0.3-14% (CBV). Delta CBF and CBV maps correlated well with angiographic TICI (CBF p<0.05, CBV p<0.05). TICI alone did not correlate significantly with outcome (r=0.24, p=0.14). However, including delta CBF and CBV with TICI resulted in a stronger correlation (r=0.37, p<0.05) against outcome. Conclusions: TICI is an important determinant of outcome in EST. The incorporation of perfusion angiography measurements (CBF and CBV) improves the predictive power of angiography for clinical outcome.