Multidimensional risk prediction model for infant atopic dermatitis: A prospective cohort study.

Area of Science:

• Immunology
• Pediatrics
• Genetics

Background:

• Atopic dermatitis (AD) is a complex inflammatory skin condition with increasing prevalence.
• Early identification of infants at high risk for AD is crucial for timely intervention.
• The interplay between maternal-fetal immunity, genetic predisposition, and environmental factors is implicated in AD development.

Purpose of the Study:

• To construct a multi-dimensional risk prediction model for neonatal atopic dermatitis (AD).
• To integrate maternal-fetal immune axis markers, genetic risk factors, and environmental exposures into a predictive tool.
• To identify key predictive factors for early-onset AD in a prospective neonatal cohort.

Main Methods:

• Prospective enrollment of 503 full-term newborns with 1-year follow-up for AD incidence.
• Collection of parental allergic history, maternal/cord blood biomarkers (IL-4, IL-13, IL-31, IL-33, IgE, TSLP), and dust mite exposure levels.
• Development of a nomogram model using LASSO and multivariable logistic regression, validated through cross-validation and decision curve analysis.

Main Results:

• A final cohort of 456 infants included 106 with AD.
• Six significant risk factors were identified: cord blood immunoglobulin E (IgE) and thymic stromal lymphopoietin (TSLP), maternal blood interleukin-4 (IL-4) and IL-33, maternal allergic history, and dust mite exposure.
• The constructed predictive model demonstrated good discriminatory ability, calibration, robustness, and clinical applicability.

Conclusions:

• Cord blood IgE and TSLP, maternal IL-4 and IL-33, maternal allergic history, and dust mite exposure are valuable predictors of neonatal AD.
• The developed model shows promise for identifying infants at risk of developing AD.
• Further multicenter studies are necessary to validate the model's generalizability.