Association Between Atopic Dermatitis and Systemic Immune-Inflammation Index: Evidence From NHANES 1999-2006.
Jie Han, Ge Du, Shuping Guo, Jianhua Hao, Yuqi Wang, Rui Li, Xiaoqing Lang, Yingjie Zhang, Xiulan Zhu, Hongzhou Cui
Abstract
Open AccessBackground: Clinical studies have demonstrated that the systemic immune-inflammation index (SII) is widely used to assess immunity and inflammation in patients. However, the association between SII and atopic dermatitis (AD) remains unclear. This study, based on the National Health and Nutrition Examination Survey (NHANES) database, aims to explore the relationship between SII and AD. Methods: This study utilized NHANES data from 1999 to 2006, with a total of 8194 subjects included in the final analysis. We examined the associations between AD, SII, and other covariates by analyzing baseline characteristics and performing correlation analyses. Multivariate generalized linear models (GLMs) were used to analyze the correlation between AD and SII risk. A weighted multivariate logistic regression model was applied to examine the association between SII and AD. Additionally, a nomogram was constructed to predict the risk of developing AD. The eXtreme Gradient Boosting (XGBoost) algorithm was employed to evaluate feature importance. Finally, subgroup analysis was performed to further explore the relationship between SII and AD across different subpopulations. Results: Significant differences were observed between the AD and control groups in terms of race, SII, SII group, and other variables. Furthermore, the p-values for SII (Q2 and Q3 groups) in all three models were less than 0.05, indicating that the influence of SII on the outcome was not significantly affected by other covariates. The weighted multivariate logistic analysis revealed that SII was strongly associated with AD as a risk factor. The nomogram demonstrated good predictive value for AD, and the XGBoost algorithm further confirmed the high predictive value of SII in AD diagnosis. Finally, subgroup analysis highlighted the significance of the association between SII and specific forms of dermatitis in various subpopulations. Conclusion: Elevated SII is independently associated with increased AD risk. Although the cross-sectional design precludes causal inference, SII represents a cost-effective biomarker for AD risk stratification. Critically, emerging evidence positions SII as a predictor of therapeutic response-particularly to JAK inhibitors and biologics-highlighting its dual utility in risk assessment and precision management of AD.