Establishment of a Risk Prediction Model for Non-alcoholic Fatty Liver Disease in Type 2 Diabetes

摘要

IntroductionNon-alcoholic fatty liver disease (NAFLD) is becoming more prevalent in patients with type 2 diabetes mellitus (T2DM) and can contribute to serious liver damage in this patient population. The aim of this study was to develop a risk nomogram for NAFLD in a Chinese population with T2DM.MethodsA questionnaire survey, physical examination and biochemical indicator testing were performed on 874 patients with T2DM, and the collected data were used to evaluate the risk to develop NAFLD in T2DM patients. The least absolute shrinkage and selection operator (LASSO) regression analysis method was used to optimize variable selection by running cyclic coordinate descent with k -fold (tenfold in this case) cross-validation. Multivariable logistic regression analysis was applied to build a predictive model by introducing the predictors selected from the LASSO regression analysis. The nomogram was developed based on the selected variables visually. A calibration plot, receiver operating characteristic curve (ROC) and decision curve analysis (DCA) were used to validate the model, with further assessment by external validation.ResultsA total of nine predictors, namely sex, age, total cholesterol (TC), body mass index (BMI), waistline, diastolic blood pressure (DBP), serum uric acid (SUA), course of disease and high-density lipoprotein-cholesterol (HDL-C), were identified by LASSO regression analysis from a total of 24 variables studied. The model constructed using these nine predictors displayed medium prediction ability, with an area under the ROC of 0.848 in the training set and 0.809 in the validation set. The DCA curve showed that the nomogram could be applied clinically if the risk threshold was between 48 and 91%, which was found to be between 44 and 82% in the external validation.ConclusionIntroducing sex, age, TC, BMI, waistline, DBP, SUA, course of disease and HDL-C into the risk nomogram increased its usefulness for predicting NAFLD risk in patients with T2DM.