Construction and validation of a risk prediction model for sarcopenia in elderly patients with type 2 diabetes mellitus
10.3969/j.issn.1006-6187.2025.11.006
- VernacularTitle:老年2型糖尿病患者合并肌少症风险预测模型构建与验证的研究
- Author:
Xiang XIAO
1
;
Feng LIU
;
Ying ZHANG
;
Lai WANG
;
Yuanhui CHENG
Author Information
1. 401331 重庆医药高等专科学校临床医学院
- Publication Type:Journal Article
- Keywords:
Diabetes mellitus,type 2;
Sarcopenia;
Nomogram prediction model;
LASSO-Logistic regression
- From:
Chinese Journal of Diabetes
2025;33(11):827-832
- CountryChina
- Language:Chinese
-
Abstract:
Objective To establish a nomogram prediction model for the risk of sarcopenia among elderly patients with type 2 diabetes mellitus(T2DM)by LASSO-Logistic regression.Methods From January 2021 to December 2023,2986 elderly T2DM patients who were admitted to The Fifth People's Hospital of Chongqing were enrolled.Based on the 2019 criteria of the Asian Sarcopenia Working Group,they were divided into a T2DM combined with sarcopenia(Sar,n=867)group and simple T2DM group(n=2119).All patients were randomly divided into a model group(n=1990)and a validation group(n=996)in a 2∶1 ratio.Another 528 patients with T2DM admitted to our hospital from January to October 2024 were selected as the external validation group.LASSO regression was used to screen variables.Logistic regression was used to analyze the influencing factors of sarcopenia.R language was used to establish a nomogram model.Receiver operating characteristic(ROC)curve and Hosmer-Lemeshow curve were used to validate the prediction model.Results The age,proportions of drinking,hypertension,hyperlipidemia,DM complications,DM duration,fasting plasma glucose(FPG),2-hour postprandial plasma glucose,hemoglobin A1c(HbA1c),triglycerides,total cholesterol,and low-density lipoprotein cholesterol in the Sar group were higher than those in the T2DM group(P<0.05).The proportion of male individuals,body mass index,and high-density lipoprotein cholesterol in the Sar group were lower than those in the T2DM group(P<0.05).LASSO regression identified 8 factors,including gender,age,drinking,hypertension,DM complications,DM duration,FPG,and HbA1c,as predictive variables for T2DM with sarcopenia.Logistic regression analysis showed that gender,age,DM duration,HbA1c,and DM complications were influencing factors for sarcopenia in elderly T2DM patients.The nomogram prediction model showed that the highest scores corresponding to DM complications,DM duration,HbA1c,age,and gender were approximately 100,70,52,43,and 30 points,respectively,with a maximum total score of approximately 295.The area under the ROC curve of the model group,validation group,and external validation group were 0.914,0.822,and 0.777,respectively.The Hosmer-Lemeshow calibration curve showed that the maximum offsets for the model group,validation group,and external validation group were 0.031,0.049,and 0.056,respectively(P=0.929,0.802,and 0.782,respectively).Conclusions Elderly T2DM patients are prone to developing sarcopenia.A nomogram model based on LASSO-Logistic analysis,which includes DM complications,DM duration,HbA1c,age,and gender,can effectively predict the risk of sarcopenia among elderly T2DM patients.
