关键词: 列线图, 糖尿病, 2型, 非肥胖, 四肢骨骼肌指数, 危险因素

Abstract: Background　Muscle mass loss increases the risk of hyperglycaemia and sarcopenia in patients with type 2 diabetes mellitus (T2DM), and Chinese adults with T2DM are predominantly non-obese, who are more likely to be associated with muscle mass loss than the obese. Aims　To establish an individualized prediction nomogram model for the risk factors of muscle mass loss in non-obese patients with T2DM. Methods　A retrospective analysis was conducted on 905 non-obese patients with T2DM admitted to the Affiliated Drum Tower Hospital, Nanjing University from January 2018 to September 2023. The patients were divided into a training set（n=633）and a validation set（n=272）using simple random sampling at a ratio of 7∶3. Univariate logistic regression analysis was used to determine risk factors for muscle mass loss in the training set. A nomogram was constructed based on the results of the multifactorial analysis.The predictive value and clinical utility of the nomogram were evaluated using Receiver Operating Characteristic（ROC）curve，Hosmer-Lemeshow calibration（H-L）curve，and Decision Curve Analysis（DCA），respectively. Results　The prevalence of muscle mass loss in non-obese patients with T2DM was 42.3%. The logistic regression analysis in the training set revealed that age（OR=1.037，95%CI 1.009~1.067，P=0.010）、male（OR=3.464，95%CI 2.185~5.490，P<0.001）、Body mass index<23.5 kg/m2（OR=18.583，95%CI 10.865~31.783，P<0.001）、HbA1c（OR=1.172，95%CI 1.063~1.292，P=0.001）、Visceral Fat Area（OR=1.021，95%CI 1.011~1.032，P<0.001）were independent risk factors for muscle mass loss in non-obese patients with T2DM. The AUC for predicting the risk of muscle mass loss in the training and validation sets were 0.825（95%CI 0.793～0.856，P<0.001）and 0.806（95%CI 0.753～0.859，P<0.001）, respectively. Hosmer-Lemeshow test indicated that this model fit the data well（χ2=11.822，P=0.159 in the training set；χ2=8.189，P=0.415 in the validation set）. Calibration charts of the model drawn by Bootstrap method showed that the calibration curves fit well with the standard curves. DCA showed that the use of nomogram prediction model was more beneficial in predicting muscle mass loss in non-obese patients with T2DM when the threshold probability of patients was 0.06～0.94. Conclusion　The nomogram model established in this study enables individualized prediction of the risk of muscle mass loss in non-obese patients with T2DM，which help to facilitate early identification of high-risk individuals and support the development of personalized intervention measures.

Key words: Nomogram, Diabetes mellitus, type 2, Non-obesity, Appendicular skeletal muscle index, Risk factors