Abstract
OBJECTIVE: This study aimed to analyze the differential expression patterns of serum cytokines across severity-stratified pediatric adenovirus pneumonia cases and evaluate their clinical utility as predictive biomarkers for disease severity. METHODS: This retrospective study included 88 pediatric adenovirus pneumonia cases and 54 healthy children between January 2021 and December 2024. Cases were stratified by disease severity into mild (n = 59) and severe (n = 29) pneumonia groups. We collected baseline clinical characteristics and measured pretreatment levels of 12 serum cytokines. Comparative analyses of cytokine profiles were performed across different severity groups (mild and severe) and healthy controls. The diagnostic performance of these cytokines for severe pneumonia detection was evaluated using receiver operating characteristic (ROC) curve analysis, with particular focus on the area under the curve (AUC) values. RESULTS: Severe pneumonia cases exhibited a more pronounced clinical course, characterized by significantly longer hospitalization, fever duration, and cough persistence, a higher incidence of wheezing and tachypnea, and marked elevations in inflammatory markers (CRP, ESR, LDH, fibrinogen, D-dimer). Of the 12 cytokines elevated versus controls, only interleukin (IL)-6, IL-8, and IL-10 exhibited statistically significant, severity-dependent increases (p < 0.05) and were correlated with longer hospitalization and persistent cough. ROC analysis revealed IL-6 and IL-10 as superior predictors for disease severity, with AUC reaching 0.88 (95% CI: 0.83-0.94). At optimal cutoffs, IL-6 showed 93% sensitivity and 71% specificity, while IL-10 demonstrated 90% sensitivity and 83% specificity. IL-8 showed moderate diagnostic value (AUC = 0.79, with 69% sensitivity and 83% specificity). CONCLUSION: These findings demonstrate distinct cytokine expression patterns across varying severity levels of pediatric adenovirus pneumonia. The robust performance of IL-6 and IL-10 underscores their potential as biomarkers for early severity stratification in clinical practice.