Abstract 135: Correlation of TICI Reperfusion with MR Reperfusion, Infarct Growth and Clinical Outcome in the DEFUSE 2 Trial

Abstract

Background: The Thrombolysis In Cerebral Infarction (TICI) score is a widely used angiographic score in endovascular stroke studies. Assessment of reperfusion based on perfusion weighted MRI (PWI reperfusion) has been used more commonly in patients treated with intravenous thrombolysis. This analysis of the DEFUSE 2 study data was undertaken to 1) determine the association between TICI and PWI reperfusion and 2) assess the association between TICI-reperfusion and clinical and radiographic outcomes. Methods: Patients undergoing acute endovascular stroke therapy of anterior circulation strokes were enrolled in a prospective multi-center study (DEFUSE 2) if an MRI could be obtained within 90 minutes before endovascular treatment and repeated within 12 hours after the intervention. Only patients with a TICI score of 0 or 1 on baseline digital subtraction angiography (DSA) were included in this analysis. A single blinded reader at the core imaging facility determined pre- and post-procedure TICI scores. TICI-reperfusion was defined as a TICI score of 2B or 3. PWI lesion volumes were assessed using fully automated software (RAPID). PWI-reperfusion was defined as a reduction in PWI(Tmax>6s) lesion volume of >50% between baseline and early follow-up. Infarct growth was defined as the difference between baseline DWI and 5-day FLAIR lesion volume. Favorable clinical response was defined as a NIHSS score of 0-1 at day 30 or an improvement in NIHSS score of ≥8 points between baseline and day 30. Results: This preliminary analysis includes 68 of 101 patients who underwent endovascular therapy and had adequate PWI data to assess reperfusion (final results will be presented at the meeting). At completion of endovascular treatment 30% of the patients remained TICI 0 or 1, 27% improved to TICI 2A, 29% to TICI 2B, and 13% had complete reperfusion (TICI 3). Better TICI-reperfusion scores were associated with higher rates of reperfusion assessed by PWI. PWI-reperfusion was seen in 32% of patients who remained TICI 0-1, 53% with TICI 2A, 98% with TICI 2B, and 100% with TICI 3 reperfusion. Agreement between TICI-reperfusion and PWI-reperfusion was moderate (kappa 0.51). The incidence of favorable clinical response increased with higher TICI scores: 35% with TICI 0-1, 44% with TICI 2A, 72% with TICI 2B, and 67% with TICI 3. Patients who met pre-specified DEFUSE 2 criteria for reperfusion (TICI 2B/3) were more likely to have a favorable clinical response (70% vs 40%; p=0.015), and had less median [IQR] lesion growth (10 [2-56] ml vs 67 [28-122] ml; p=0.001) than patients without TICI-reperfusion. Conclusion: TICI 2B or 3 reperfusion following endovascular therapy for acute anterior circulation stroke is highly correlated with PWI reperfusion. Patients with TICI 2B or 3 reperfusion show less infarct growth and are more likely to have a favorable clinical response.