Predictors of in-hospital mortality after mechanical thrombectomy for acute ischemic stroke: A single-center cohort study

Introduction/Objective. In-hospital mortality after mechanical thrombectomy (MT) for anterior-circulation large vessel occlusion (LVO) remains insufficiently characterized. We aimed to identify independent predictors and to test whether complete recanalization (mTICI 3) rather than successful recanalization (mTICI 2b-3) carries independent prognostic weight for this interventional radiological procedure. Methods. In a single-center retrospective cohort, 100 consecutive patients with acute ischemic stroke due to anterior-circulation LVO treated with MT were analyzed. The primary outcome was in-hospital mortality estimation (modified Rankin Scale [mRS] score of 6 at discharge). Independent predictors were identified by binary logistic regression using three pre-specified variables: pre-stroke mRS, complete recanalization (mTICI 3), and ASPECTS. Sensitivity analyses and bootstrap validation were performed. Results. In-hospital mortality was 30% (95% CI 21.9-39.6). Independent predictors were a higher pre-stroke mRS (aOR 4.33; 95% CI 1.85-10.13; p < 0.001) and complete recanalization (aOR 0.21; 95% CI 0.07-0.62; p = 0.004), whereas a higher ASPECTS showed a protective direction of borderline significance (aOR 0.63; 95% CI 0.39-1.02; p = 0.059). Among successfully reperfused patients, mTICI 3 remained independently associated with lower mortality than mTICI 2b-2c (aOR 0.17; 95% CI 0.05-0.60), an exploratory subgroup finding. The model showed good discrimination (AUC 0.82; optimism-corrected 0.80) and satisfactory calibration (Hosmer-Lemeshow p = 0.89). Conclusion. In-hospital mortality was high despite successful reperfusion. A simple model based on premorbid status, ASPECTS, and completeness of recanalization may enable post-procedural risk stratification and warrants external validation.