Development and validation of an online nomogram for screening metabolic-associated fatty liver disease in obese children

Background Metabolic-associated fatty liver disease (MAFLD) has emerged as a critical pediatric health concern, particularly among children with obesity. However, its diagnosis poses substantial challenges, especially in the use of non-invasive methods. Our goal was to construct an online nomogram for screening MAFLD in obese children. Methods We designed a retrospective cross-sectional study involving 2,512 obese children. Detailed anthropometric data and laboratory parameters were collected. The study dataset was randomly allocated into training (n = 1758) and validation (n = 754) sets at a 7:3 ratio. To identify MAFLD risk factors, we conducted logistic regression analyses, from which a web-based predictive nomogram was constructed. Using receiver operating characteristic (ROC) curves and area under the curve (AUC), the nomogram's performance was assessed and contrasted with the triglyceride glucose (TyG) index, Zhejiang University (ZJU) index, and Korean NAFLD (K-NAFLD) score. The goodness-of-fit of the nomogram was evaluated using calibration plots, and the nomogram's clinical value was assessed using decision curve analysis (DCA). Results A total of 1,344 participants (53.50%) were diagnosed with MAFLD by ultrasound. Age, gender, BMI Z-score, waist circumference (WC), homeostatic model assessment for insulin resistance (HOMA-IR), and alanine aminotransferase (ALT) were identified as independent factors influencing MAFLD in obese children. These six variables were selected for the construction of the nomogram. ROC analysis revealed that the nomogram had superior diagnostic performance for MAFLD detection compared to the other three models, with AUC values of 0.874 (95% confidence interval [CI]: 0.858-0.890) in the training set and 0.870 (95% CI: 0.845-0.895) in the validation set. Calibration plots indicated a good fit of the nomogram in both datasets. Furthermore, DCA demonstrated its strong clinical applicability. Conclusions This study developed an online nomogram that demonstrates robust diagnostic accuracy and clinical utility for assessing obese children's MAFLD risk.