Abstract
BACKGROUND: Chronic low-grade inflammation (metaflammation) is a central feature of obesity and contributes to metabolic complications. Complete blood count (CBC)-derived inflammatory indices have been proposed as accessible markers of obesity-related inflammation, but their biological meaning in pediatric obesity remains unclear. OBJECTIVE: To determine whether CBC-derived composite inflammatory indices primarily reflect adiposity-related inflammation or metabolic impairment in children with different obesity severities. METHODS: This retrospective cross-sectional study included 417 children aged 5-18 years, categorized into four groups according to body mass index (BMI) percentiles: normal weight, overweight, obese, and morbidly obese. The CBC-derived inflammatory indices, including the systemic immune-inflammation index (SII), systemic inflammation response index (SIRI), aggregate inflammation index (AISI), and monocyte-to-high-density lipoprotein cholesterol ratio (MHR), were calculated. Insulin resistance was assessed using the Homeostasis Model Assessment for Insulin Resistance (HOMA-IR). Trend, correlation analyses were performed. Multivariate linear regression models based on a predefined clinical rationale were created, and their discriminative performances were evaluated using ROC analyses. RESULTS: SII, SIRI, AISI, and MHR levels increased as the severity of obesity increased, and a significant difference was observed, especially between the normal weight and morbidly obese groups. In multivariate models, the significant relationship between BMI-SDS and SII, AISI, and MHR persisted, whereas no relationship was found with SIRI. HOMA-IR was only associated with MHR. In ROC analyses, the ability of the inflammatory indices to distinguish between morbid obesity and insulin resistance was found to be limited (AUC approximately 0.55-0.68). CONCLUSION: CBC-derived inflammatory indices in pediatric obesity appear to primarily reflect adiposity-related inflammatory burden rather than isolated insulin resistance. However, owing to their limited explanatory power and low discriminative performance, they should not be interpreted as standalone diagnostic tools but as biomarkers supporting clinical assessment.