Abstract WP60: Later Imaging More Accurately Captures Infarct Growth in DEFUSE 3

Abstract

Background: DEFUSE 3 assessed thrombectomy in patients with evidence of salvageable tissue on brain imaging in the 6 to 16 hour treatment window. Despite substantially better functional outcomes as measured by the modified Rankin Scale at day 90, the study did not find a significant reduction in infarct growth between baseline and 24-hour follow up with endovascular therapy. The aim of this study was to determine if infarct growth captured with scans performed at later timepoints could demonstrate differences between treatment groups. Methods: We included DEFUSE 3 patients in whom infarct volume could be measured at baseline, at 24-hour follow up, and at one additional later time point using either CT or MRI. Early infarct growth (increase in lesion volume between baseline imaging and 24 hours) and late infarct growth (increase in lesion volume between 24 hours and a later time point) were calculated. Differences in infarct growth between the medical and endovascular groups were assessed using Wilcoxon rank-sum tests. Results: For the 70 patients who met inclusion criteria, the average age was 69 ± 14 years, average NIHSS 17 ± 6, and the average time of late imaging was 5.3 ± 3.1 days. There were no significant differences in these variables between the medical (n=43) and endovascular (n=27) cohorts. Median early infarct growth was not different between the medical 32 mL (IQR: 24 - 79) and the endovascular groups 24 mL (IQR: 15 - 91; p = 0.359). Median late infarct growth was greater in the medical group 11 mL (IQR: 2 - 29) compared to the endovascular group 2 mL (IQR: -10 - 9; p = 0.003) (see figure). Discussion: Imaging performed at a mean of 5 days post-stroke demonstrates continued infarct growth in the medical arm and minimal additional growth in the endovascular arm. These data correspond to the clinical benefit of endovascular therapy observed in DEFUSE 3 and suggest that imaging beyond 24 hours is preferred for assessing final infarct volume in endovascular therapy trials.