BACKGROUND: Malnutrition is a significant concern in children with β-thalassemia, impacting their growth and overall health. This study aimed to establish optimal thresholds for predicting malnutrition risk in children with β-thalassemia using muscle mass indicators derived from Bioelectrical Impedance Analysis (BIA). METHODS: A cross-sectional study was conducted with 162 pediatric patients diagnosed with β-thalassemia. Nutritional status of them was assessed using the World Health Organization (WHO) Child Growth Standards and references. BIA was performed to obtain fat-free mass (FFM), skeletal muscle mass (SMM), and soft lean mass (SLM). Propensity score matching (PSM) was used to control for age and gender. Receiver Operating Characteristic (ROC) curves were constructed to evaluate the diagnostic performance. RESULTS: SLM-change <
6% demonstrated the highest sensitivity [0.82, 95% confidence interval (CI) 0.72-0.92] and a negative predictive value of 0.83 (95% CI 0.74-0.93), while FFM-change <
4% showed more balanced performance with a sensitivity of 0.58 (95% CI 0.45-0.71) and a specificity of 0.65 (95% CI 0.56-0.74). Percentage change indicators (FFM-change, SLM-change, and SMM-change) exhibited remarkable stability before and after PSM, indicating minimal influence from age and gender. CONCLUSIONS: This study established novel, age-adaptive thresholds (SLM-change <
6% and FFM-change <
4%) for predicting malnutrition risk in children with β-thalassemia. The findings suggest that these thresholds could serve as effective references to assess nutritional status across different age groups, providing new perspectives for personalized nutritional management strategies.