Association between systemic immune-inflammation index and ICU-acquired infection in critically Ill patients without infection: a competing risks analysis from a retrospective cohort

Abstract Background Healthcare-associated infections and their complications represent a predominant cause of morbidity and mortality in intensive care units (ICUs) globally. Several studies have reported that the systemic immune-inflammation index (SII) is associated with an increased risk of infection; however, its association with ICU-acquired infection (ICU-AI) remains unclear. We aimed to examine the association between SII and the subsequent risk of ICU-AI. Methods We conducted a retrospective cohort study using the MIMIC-IV database, including adult ICU patients without suspected infection at baseline. SII was calculated as platelet × neutrophil / lymphocyte count, based on values from the first 24 h of ICU admission. The primary outcome was ICU-AI, defined as suspected infection occurring after 48 h of ICU admission. Two competing risk regression models were employed to evaluate the association between SII and ICU-AI. Results Among the 5,459 patients included, 627 (11.5%) developed ICU-AI. Higher SII (per 1,000-unit) was independently associated with an increased risk of ICU-AI in both models: adjusted cause-specific hazard ratio (CSHR), 1.06 (95% CI, 1.03–1.08; p < 0.001) and subdistribution hazard ratio (SHR), 1.05 (95% CI, 1.03–1.07; p < 0.001). When analyzed by quartiles, a significant dose-response relationship was observed (p for trend < 0.001). Compared with patients in the lowest quartile (Q1, SII ≤ 594 × 10⁹/L), those in Q3 (SII 1125–2201 × 10⁹/L) and Q4 (SII ≥ 2202 × 10⁹/L) showed progressively higher risks of infection, with adjusted CSHR of 1.44 (95% CI 1.11–1.86) and 1.76 (95% CI 1.37–2.26), and adjusted SHR of 1.51 (95% CI 1.17–1.93) and 1.96 (95% CI 1.54–2.49), respectively. Time-to-event curves showed distinct and consistent cumulative risk trajectories across SII quartiles. Conclusion Elevated SII at ICU admission was independently associated with an increased risk of developing suspected secondary infection in infection-naïve critically ill patients.