2型糖尿病合并非酒精性脂肪肝的预测模型研究

Prediction of non-alcoholic fatty liver in patients with type 2 diabetes mellitus

目的构建中老年2型糖尿病(T2DM)病例合并非酒精性脂肪肝(NAFLD)风险的预测模型,为T2DM合并NAFLD早期筛查和预防提供依据。方法选择2021年在新疆维吾尔自治区人民医院克拉玛依医院确诊的≥45岁T2DM病例为研究对象,收集病例基本信息、血液生化指标等资料,按照7∶3比例将病例随机纳入训练组(n=3241)和验证组(n=1389)。采用LASSO回归和多因素logistic回归模型筛选预测因子;建立T2DM合并NAFLD风险预测模型,采用受试者工作特征(ROC)曲线、校准曲线和决策曲线(DCA)评估预测效能。结果纳入T2DM病例4630例,其中合并NAFLD 1279例,占27.62%。LASSO回归和多因素logistic回归分析结果显示,性别、年龄、舒张压、体质指数、血清谷丙转氨酶、三酰甘油、低密度脂蛋白胆固醇和血小板计数是T2DM合并NAFLD的风险预测因子。训练组和验证组模型建立的ROC曲线下面积分别为0.823(95%CI:0.814~0.832)和0.809(95%CI:0.799~0.818),Hosmer-Lemeshow拟合优度检验显示模型拟合度较好(P>0.05),DCA结果显示当病例的风险阈值概率为0.27~0.85时,使用该预测模型预测NAFLD风险的临床净收益较高。结论本研究构建的预测模型对≥45岁T2DM病例合并NAFLD风险具有较好的评估价值。

Objective To construct a prediction model of non-alcoholic fatty liver disease(NAFLD)in middle-aged and elderly patients with type 2 diabetes mellitus(T2DM),so as to provide basis for early screening and prevention of T2DM complicated with NAFLD.Methods Patients aged 45 years and above and diagnosed with T2DM in Karamay Hospital of People's Hospital of Xinjiang Uygur Autonomous Region in 2021 were collected as the study subjects.The data of general demographic characteristics and biochemical test results were collected.The patients were randomly divided into training group(n=3241)and validation group(n=1389)according to the ratio of 7∶3.LASSO regression and multivariable logistic regression model were used to select predictive factors.The nomograph model for prediction of NAFLD risk in T2DM patients was established.The predictive value of the model was evaluated using the receiver operating characteristic(ROC),adjusted curve and decision clinical analysis.Results Totally 4630 T2DM cases were included,including 1279 cases(27.62%)complicated with NAFLD.LASSO regression and multivariable logistic regression analysis identified gender,age,diastolic blood pressure,body mass index,alanine transaminase,triglycerides,low density lipoprotein cholesterol and platelet count as risk prediction factors for NAFLD in T2DM patients.The area under the ROC curve was 0.823(95%CI:0.814-0.832)for the training group and 0.809(95%CI:0.799-0.818)for the validation group,and Hosmer Lemeshow test showed a good fitting effect(P>0.05).Decision curve analysis showed higher net clinical benefit of using the predictive model to predict NAFLD risk when the risk threshold probability was 0.27 to 0.85.Conclusion The nomogram model established has a good predictive value for the risk of NAFLD in T2DM patients aged 45 years and above.