Reperfusion-dependent treatment effects of thrombectomy in patients with large ischemic infarcts

Background: While thrombectomy benefits patients with large infarcts, it is unclear whether this benefit persists across different levels of reperfusion. Aims: This study investigates how the degree of reperfusion influences the effectiveness of endovascular thrombectomy (EVT) combined with best medical treatment (BMT), compared to BMT alone, in patients with large infarcts. Methods: This post hoc analysis of the TENSION trial, a randomized controlled study, assessed EVT versus BMT in patients with extensive infarction (Alberta Stroke Program Early CT Score (ASPECTS) 3–5). Primary outcome was the modified Rankin Scale (mRS) score at 90 days. Secondary outcomes included infarct volume at 24 h, mortality, and symptomatic hemorrhage. Outcomes were stratified by final reperfusion level, measured with the modified thrombolysis in cerebral infarction (mTICI) scale. Confounder-adjusted common odds ratios (cORs) and average treatment effects (ATEs) were estimated using inverse probability weighting with regression adjustment. Results: A total of 246 patients (median age, 74 years (interquartile range (IQR), 65–80); median baseline ASPECTS, 4 (IQR, 3–5)) were included. Compared to BMT alone, unsuccessful EVT (mTICI ⩽ 2a) was not associated with worse functional outcomes (cOR:1.2, 95% CI, 0.95 to 1.52; p = 0.131), higher mortality (ATE: –11.6%; 95% CI, –28.82 to 5.61; p = 0.187), or larger infarct volumes on follow-up (ATE:0.99 mL; 95% CI, –45.30 to 45.32; p = 0.965). First-pass complete reperfusion (mTICI 3) showed the greatest treatment benefit, significantly improving all endpoints, with a cOR of 4.85 (95% CI, 3.74–6.31; p < 0.001) for improved mRS scores and a 29% absolute reduction in mortality. Conclusion: In this post hoc analysis of the TENSION trial, unsuccessful EVT did not worsen outcomes compared to BMT alone. The highest benefit of EVT occurred with first-pass complete reperfusion, emphasizing the importance of achieving optimal reperfusion in this vulnerable stroke subgroup. These findings do not justify general treatment recommendations.