Abstract
OBJECTIVE: This study aimed to evaluate the role of Systemic Immune-Inflammation Index (SII) in predicting the diagnosis of severe pneumonia in children. METHODS: A retrospective analysis included 595 pediatric community-acquired pneumonia (CAP) patients from the Affiliated Hospital of North Sichuan Medical College (January 2024-July 2025). Patients were randomly divided into development (70%) and validation (30%) sets. General clinical data and SII were collected. Statistical analysis included Mann-Whitney U, chi-square, multivariate logistic regression, and ROC analysis. A predictive nomogram was developed and evaluated for calibration, discrimination, and clinical utility. RESULTS: SII was an independent risk factor for severe pneumonia. An SII ≥738.0 significantly increased severe pneumonia risk. A combined model including SII, infection status, length of hospital stay, and ICU admission showed higher predictive accuracy than SII alone. Conclusion: SII is a useful biomarker for predicting severe pneumonia in children. The nomogram integrating SII with clinical factors demonstrates good predictive performance.