Abstract W MP6: Pre-treatment Neuroimaging Is Critical to Identify Acute Stroke Patients With Large Infarcts Unlikely to Respond to Endovascular Therapy

Abstract

Purpose: The Penumbra START Trial was a multi-center, prospective trial with an aim of testing if core infarct size on pre-treatment neuroimaging predicts clinical response to IA stroke therapy. Presented herein are final results of this study. Methods: As prespecified, infarcts were trichotomized into small [lesion volume <50 cc (CTP, DWI) or ASPECTS 8-10 (CTA)], medium (volume 50-100 cc or ASPECTS 5-7) or large (volume >100 cc or ASPECTS 0-4). In total, 146 patients were enrolled at 27 centers, including 115 patients meeting study criteria. Core infarct volumes were tiered by ranking imaging results by DWI > CTP > CTA. Results: Mean age was 66. Median NIHSS score was 19. Overall rate of TICI 2b-3 revascularization was 71% (post vs pre-procedure p <0.0001). Forty-five percent of patients had a 90-day good outcome (mRS ≤2); 27% died. Core infarct volumes were 24% small, 52% medium and 24% large; the number of patients for each imaging modality was 5% DWI, 38% CTP and 57% CTA. In tiered volume analysis, the good outcome rate was 68% in small, 46% in medium and 17% in large infarcts ( p =0.0004), despite similar recanalization rates (74% small, 73% medium, 64% large, p =0.4186). Patients with large infarcts had significantly higher admission NIHSS ( p =0.00782). Within 24 hours of intervention, the procedural SAE rate was 22% in small, 27% in medium and 46% in large infarcts ( p =0.0497). Mortality and sICH rates also showed relationships with infarct volumes (11% small, 23% medium, 50% large, p =0.0011, and 4% small, 7% medium, 21% large, p =0.0245, respectively). Conclusion: Pre-treatment neuroimaging is important to identify patients with large infarcts who are less likely to have favorable outcomes and more likely to suffer mortality, sICH and procedural SAEs. These findings support the use of stringent imaging criteria in patient selection.