Study of risk prediction model of metabolic dysfunction associated steatotic liver disease among children and adolescents
10.16835/j.cnki.1000-9817.2026107
- VernacularTitle:儿童青少年代谢功能障碍相关脂肪性肝病风险预测模型研究
- Author:
XIANG Fanying, NA Xiaona, AN Xizhou, CHEN Lijing, ZHONG Haiying, LIANG Xiaohua, CHEN Jingyu
1
Author Information
1. Department of Clinical Epidemiology and Biostatistics, Children s Hospital of Chongqing Medical University/ National Clinical Research Center for Children and Adolescents Health and Diseases/ Ministry of Education Key Laboratory of Child Development and Disorders/ Chongqing Key Laboratory of Pediatric Metabolism and Inflammatory Diseases, Chongqing 400014, China
- Publication Type:Journal Article
- Keywords:
Metabolic diseases;
Fatty liver;
Regression analysis;
Child;
Adolescent
- From:
Chinese Journal of School Health
2026;47(4):475-479
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct a risk prediction model for pediatric metabolic dysfunction associated steatotic liver disease (MASLD), so as to provide practical tool for the early identification of high risk children.
Methods:A healthy cohort of children in Southwest China was established from January 2021 to April 2025. A nested case-control study design was used to include 507 cases MASLD group and 507 cases in non MASLD group. Data on physical measurements, blood biochemical parameters, and liver ultrasound indicators were collected. Conditional Logistic regression was used to analyze the relationship between individual variables and MASLD, Lasso regression was applied for multivariable screening, and a high risk prediction model was constructed and presented in the form of a nomogram. Internal validation was performed using 10 repeated ten fold cross validations to assess model discrimination, accuracy, sensitivity, and specificity.
Results:Logistic regression analysis showed that MASLD was associated with central obesity ( OR=22.11, 95%CI =15.62-31.29), apolipoprotein B ( OR=30.24, 95%CI =12.42-73.63), increased hepatorenal echo ( OR=326.00, 95%CI =183.87-578.01), hepatomegaly ( OR=24.98, 95%CI =16.66-37.46) (all P <0.05). The Lasso regression jointly selected 6 key variables, including hepatorenal echo, central obesity, hepatomegaly, right liver lobe inclination, body mass index, and alanine amino transferase. The results of cross validation showed that the average area under the curve (AUC) was 0.999 5, the average accuracy was 98.74%, and the sensitivity and specificity were 98.21% and 99.22% respectively, indicating a good predictive effect of the model.
Conclusion:The risk prediction model for high risk MASLD among children based on ultrasound and clinical indicators has good prediction effect, which is helpful for the early identification and risk stratification of pediatric MASLD.
