Biochemical and Inflammatory Profiles in Paediatric Obesity, Undernutrition and Normal Weight: A Single-Centre Retrospective Cohort Study

Background: To compare biochemical, micronutrient-related and complete blood count (CBC)-derived inflammatory profiles among children with obesity, undernutrition and normal weight, and to explore the discriminatory performance of selected routine laboratory markers for the obesity phenotype. Methods: In this single-centre retrospective comparative study, medical records of children aged 2–17 years evaluated between January 2023 and December 2025 were reviewed. Participants were classified as obese, undernourished or normal weight using World Health Organization (WHO) body mass index-for-age (BMI-for-age) references. Liver enzymes, lipid parameters, iron-related indices, micronutrient levels, zinc, insulin and CBC-derived inflammatory indices were compared across groups using non-parametric omnibus tests. Exploratory receiver operating characteristic (ROC) analyses were performed for selected markers. Results: A total of 690 children were included: 100 with obesity, 90 with undernutrition and 500 with normal weight. The obesity group was significantly older than the normal-weight group (median 11.0 vs 7.5 years; P < 0.001). Obesity was characterised by higher alanine aminotransferase (ALT), insulin, triglycerides, neutrophil-to-lymphocyte ratio (NLR), systemic immune-inflammation index (SII) and systemic inflammation response index (SIRI), together with lower high-density lipoprotein (HDL) cholesterol, 25-hydroxyvitamin D [25(OH)D] and aspartate aminotransferase-to-alanine aminotransferase (AST/ALT) ratio (all P < 0.05). In age-unadjusted exploratory ROC analysis, the AST/ALT ratio showed the strongest discriminatory performance for obesity (area under the curve [AUC] 0.859; 95% confidence interval [CI] 0.823–0.895; optimal cut-off ≤1.52; sensitivity 77.0%; specificity 77.6%), followed by SIRI (AUC 0.751; 95% CI 0.700–0.802). Conclusions: Paediatric obesity was associated with a distinct liver enzyme-metabolic-inflammatory laboratory profile, whereas undernutrition exhibited a different micronutrient-biochemical pattern. Because the study was retrospective and age-unadjusted, these findings should be interpreted as exploratory and hypothesis-generating. Prospective studies with age-balanced sampling, pubertal staging and liver imaging are required before clinical implementation.