Randomized trial develops IMPACT-EVT risk score to predict in-hospital mortality after mechanical thrombectomy, indicating significant clinical utility.
Background and aims Mechanical thrombectomy (MT) improves outcomes in acute ischemic stroke due to large-vessel occlusion, yet in-hospital mortality remains substantial. We aimed to develop a clinically intuitive risk score—IMPACT-EVT—to predict in-hospital mortality after MT by integrating baseline clinical severity, neuroimaging tissue vulnerability, and early procedural outcomes. Methods IMPACT-EVT was derived from published multicenter registries (MR CLEAN, HERMES) and recent thrombectomy trials. Variables with independent associations to in-hospital mortality were selected: age, baseline NIHSS, admission hyperglycemia, ASPECTS, collateral status, infarct core volume on CTP/DWI, recanalization success (mTICI ≥2b), symptomatic intracranial hemorrhage (sICH), and NIHSS at 24 hours. Infarct core volume was stratified as <30 ml, 30–70 ml, and >70 ml. Each variable was assigned points proportional to effect sizes reported in the literature. Total scores (0–18) were stratified into four risk categories: Low (0–4), Moderate (5–8), High (9–12), Very High (≥13). Results Each domain independently contributed to mortality prediction. Baseline clinical factors, particularly age ≥80 and NIHSS >20, strongly influenced risk. Neuroimaging markers, including low ASPECTS, poor collaterals, and increasing core volume, enhanced discrimination. Procedural and early post-procedure factors, especially sICH and NIHSS at 24 h, were the most potent predictors. Risk categories corresponded with reported in-hospital mortality: Low (<5%), Moderate (5–15%), High (20–40%), Very High (>50%). Conclusions IMPACT-EVT provides a practical, literature-based framework for early in-hospital mortality risk stratification after MT. Incorporating quantitative infarct core volume improves prognostic accuracy. External validation is required prior to clinical implementation. Conflict of interest Nothing to disclose Figure 1 - belongs to Results
No takes yet. Share an insight, caveat, or question.
Hrynykha et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: