The systemic immune-inflammation index as a superior predictor of short-term prognosis in acute ischemic stroke after mechanical thrombectomy: a retrospective cohort study

Objective: This study evaluated the role of machine learning models based on the Systemic Immunity Index (SII) in predicting short-term prognosis after mechanical thrombectomy (MT) in acute ischemic stroke (AIS). Methods: Data from 387 AIS patients who underwent MT were retrospectively analyzed, including clinical variables, inflammatory markers such as SII, platelet lymphocyte ratio (PLR), neutrophil lymphocyte ratio (NLR）and 90-day modified Rankin Scale (mRS) scores. Patients were categorized into good and poor prognosis groups based on mRS scores. Univariate and multifactorial logistic regression models were constructed to identify risk factors and compare predictive performance. Four models were developed: clinical baseline, SII+clinical baseline, PLR+clinical baseline, and NLR+clinical baseline. Model performance was assessed using ROC curves, NRI, IDI, calibration curves, and decision curve analysis (DCA). Results: Results showed that SII outperformed PLR and NLR, with AUCs of 0.834 (uncorrected) and 0.841 (corrected). The optimal model (SII+clinical baseline) achieved an AUC of 0.863, significantly improving prognosis prediction. SHAP analysis confirmed SII as the most influential variable (74.2%). The model demonstrated good fit, clinical utility, and effectiveness in identifying poor prognosis patients at a 15% probability threshold. Conclusion: In conclusion, SII-based models provide superior prognostic accuracy compared to traditional markers, offering a valuable tool for clinical decision-making in AIS patients post-MT.