@article{QIN2025, 
author = {Jiayi QIN and Xueqian ZHOU and Xiangyong MENG and Wensheng CHEN},
title = {Construction and validation of a risk prediction model for lean metabolic dysfunction-associated steatotic liver disease},
year = {2025},
journal = {Journal of Army Medical University},
volume = {47},
number = {9},
pages = {969-979},
keywords = {lean metabolic dysfunction-associated steatotic liver disease, risk factors, prediction model},
url = {https://www.sciopen.com/article/10.16016/j.2097-0927.202412024},
doi = {10.16016/j.2097-0927.202412024},
abstract = {ObjectiveTo investigate the risk factors for the occurrence of metabolic dysfunction-associated steatotic liver disease (MASLD) in lean individuals and construct a risk prediction model.MethodsBased on the National Health and Nutrition Examination Survey (NHANES) database in the United States (from January 2017 to March 2020), 1 123 adult individuals were included in this study. Then the participants were randomly divided into a training set (n=561) and a validation set (n=562) through simple random sampling. Data on their demographics, anthropometrics, lifestyle, underlying diseases, and laboratory test results were collected. LASSO regression analysis was used to screen potential variables in the training set, and multivariate logistic regression was employed to identify independent risk factors for lean MASLD. Based on these risk factors, a prediction model for lean MASLD was constructed (LMPM). To evaluate the clinical value of the LMPM, it was compared with two commonly used prediction models for non-alcoholic fatty liver disease, the fatty liver index (FLI) and the hepatic steatosis index (HSI). The performance of the model was evaluated and internally validated using the area under the receiver operating characteristic curve (AUC), net reclassification index (NRI), integrated discrimination improvement (IDI), calibration curve, decision curve, and clinical impact curve.ResultsAge, waist circumference, and triglycerides (TG) were identified as independent risk factors for the development of MASLD in lean individuals. The LMPM, constructed based on these indicators, demonstrated good discriminative ability in both the training and validation sets, with AUC values of 0.86(95%CI: 0.82~0.89) and 0.81(95%CI: 0.77~0.85), respectively, which were significantly better than those of FLI [training set: AUC=0.83(95%CI: 0.79~0.87); validation set: AUC=0.74(95%CI: 0.70~0.79)] and HSI [training set: AUC=0.71(95%CI: 0.66~0.76); validation set: AUC=0.71(95%CI: 0.65~0.76)]. Compared with FLI and HSI, the LMPM showed improvements in NRI and IDI in both the training and validation sets. The calibration curve demonstrated its high accuracy, and both the decision curve analysis and the clinical impact curve analysis indicated that the LMPM provided greater clinical benefits.ConclusionAge, waist circumference, and TG are independent risk factors for lean MASLD. Based on these factors, a prediction model, named LMPM, is developed to assess the risk of MASLD in lean individuals, which exhibits good predictive performance and has certain guiding significance for the timely identification of high-risk populations.}
}