Systemic inflammation markers (SII and SIRI) as predictors of cognitive performance: evidence from NHANES 2011–2014
Xiaoyue Wang, Qinghua Wen, Yujie Li, Huanhuan Zhu, Fengyin Zhang, Simin Li, Zhan Lin, Juan Li
Abstract
Background Neuroinflammation is linked to cognitive function. However, epidemiological research on two emerging inflammation markers—the systemic immune-inflammation index (SII) and the systemic inflammation response index (SIRI)—remains limited in the context of cognitive performance. This study investigates the relationship between SII, SIRI, and cognitive performance in older adults. Methods This cross-sectional analysis included 2,194 participants from the 2011–2014 National Health and Nutrition Examination Survey (NHANES) who met eligibility criteria. Logistic regression, subgroup analysis, and restricted cubic spline modeling were used to assess the associations between cognitive performance and inflammation markers, specifically SII and SIRI. Results After adjusting for population weights, participants with low cognitive function had an SII of 541.54 (95% CI: 360.00–796.50, p = 0.037) and an SIRI of 1.28 (95% CI: 0.82–2.18, p = 0.031). In fully adjusted models, higher levels of both SII (OR = 0.858, 95% CI: 0.856–0.859) and SIRI (OR = 0.891, 95% CI: 0.889–0.892) were significantly associated with lower odds of normal cognitive function, indicating an increased risk of cognitive impairment. Neutrophil-related markers (NC, NLR, SIRI) exhibited the strongest inverse associations. Subgroup analysis showed more consistent associations for SIRI across demographic and behavioral factors, while SII displayed fewer. RCS analysis indicated a stronger non-linear relationship for SIRI ( p = 0.005) compared to SII ( p = 0.018) after full adjustment. Conclusion This study suggests a positive association between SII, SIRI, and cognitive function, with a more pronounced relationship for SIRI. These findings highlight the potential of SIRI as a novel, accessible marker for predicting cognitive impairment risk.