Association between the systemic immune-inflammation index and mortality risk in myocardial infarction patients: a retrospective cohort study.

BACKGROUND: The systemic immune-inflammation index (SII) has shown potential as a prognostic biomarker in several diseases. This study aimed to investigate the relationship between SII and mortality in patients with myocardial infarction (MI). METHODS: A retrospective cohort study was conducted from the MIMIC-Ⅳ database (n = 774) and Harbin Medical University Cancer Hospital (n = 482). Patients were divided into four quartiles based on SII, and mortality at 28, 90, and 180 days was analyzed. Kaplan-Meier survival curves and Cox regression models adjusted for confounders were used to assess the relationship between SII and mortality. Additionally, restricted cubic splines (RCS) were used to explore the nonlinear relationship between SII and 28-day mortality. RESULTS: RCS analysis revealed a significant nonlinear J-shaped association between SII and mortality risk, with the lowest mortality risk observed at approximately 770 × 10⁹/L of SII. In quartile analysis, the highest SII (Q4) had significantly higher mortality rates at 28 days (30.93% vs. 8.29%, P < 0.001), 90 days (38.14% vs. 11.92%, P < 0.001), and 180 days (42.78% vs. 12.44%, P < 0.001), while the lowest SII group (Q1) demonstrated a modest increase compared to the second quartile (Q2). Multivariate analysis confirmed SII as an independent predictor of 28-day mortality (adjusted HR for Q4: 3.04, 95% CI 1.30-7.10, P = 0.005). Subgroup analysis showed the strongest association in patients with congestive heart failure (HR: 1.48, 95% CI 1.26-1.88, P = 0.005). CONCLUSIONS: SII exhibits a threshold effect on post-MI mortality, where values exceeding 770 × 10⁹/L independently predict adverse outcomes, suggesting its potential as a prognostic tool for early risk stratification.