帕金森病患者合并肌少症的风险预测模型构建  

Risk prediction model construction of patients with Parkinson’s disease complicated with sarcopenia

作　　者：陈思 杨霄鹏[1] 楚广磊 陈静[1] 奚志 白宏英[1] CHEN Si;YANG Xiaopeng;CHU Guanglei;CHEN Jing;XI Zhi;BAI Hongying(The Second Affiliated Hospital of Zhengzhou University,Zhengzhou 450003,China)

机构地区：[1]郑州大学第二附属医院,郑州河南450003

出　　处：《中国实用神经疾病杂志》2024年第6期696-700,共5页Chinese Journal of Practical Nervous Diseases

基　　金：河南省医学科技攻关项目(编号:LHGJ20210411)。

摘　　要：目的研究构建帕金森病患者合并肌少症的风险模型,为早期干预治疗提供依据。方法选择郑州大学第二附属医院神经内科2020-09—2023-02收治的170例帕金森病患者为研究对象。收集患者的临床资料并依据亚洲肌少症工作组诊断标准对肌少症筛查进行诊断,通过机器学习筛选构建帕金森病患者合并肌少症的风险预测模型,并通过接受者操作特性曲线和临床决策分析验证模型的稳定性。结果帕金森病患者合并肌少症71例(41.8%),肌少症组和非肌少症组患者性别比例、体重指数、Hoehn-Yahr分级以及日常生活活动能力评分差异有统计学意义(均P<0.05)。LASSO回归模型分析显示Hoehn-Yahr分级、体重指数、白蛋白、维生素D、同型半胱氨酸、肌酐、日常生活活动能力、叶酸和血红蛋白是帕金森病患者合并肌少症的危险因素。通过多因素Logistic回归分析,以P<0.1进一步研究发现,Hoehn-Yahr分级、白蛋白、维生素D、同型半胱氨酸、肌酐和日常活动能力评分构建的帕金森病患者合并肌少症风险预测模型的C指数为0.819,受试者工作特征曲线下面积为0.819,显示较好的预测价值。决策曲线分析显示,概率值为8%~82.0%时,使用此模型对患者临床净获益率最高。结论帕金森病患者合并肌少症率高,本研究构建的列线图模型可有效筛查帕金森病患者合并肌少症的风险,为尽早进行干预提供依据。Objective To construct a risk model of sarcopenia in patients with Parkinson’s disease(PD),and to provide a basis for early intervention and treatment.Methods A total of 170 Parkinson’s disease patients admitted to the Department of Neurology of the Second Affiliated Hospital of Zhengzhou University from September 2020 to February 2023 were enrolled as the study subjects.The clinical data of patients were collected and the patients were screened and diagnosed with sarcopenia according to the diagnostic criteria of the Asian Sarcopenia Working Group,and the risk prediction model of patients with Parkinson’s disease complicated with sarcopenia was constructed through machine learning screening,and the stability of the model was verified by the receiver operating characteristic curve and clinical decision analysis.Results There were 71 patients(41.8%)with sarcopenia in Parkinson’s disease,and there were significant differences in sex,body mass index(BMI),Hoehn-Yahr grade and activities of daily living(ADL)score between the two groups(all P<0.05).LASSO regression model analysis showed that Hoehn-Yahr grade,body mass index,albumin,vitamin D,homocysteine,creatinine,ability to perform activities of daily living,folic acid and hemoglobin were risk factors for sarcopenia in patients with Parkinson’s disease.Through multivariate Logistic regression analysis,further studies with P<0.1 showed that the C-index of the risk prediction model for patients with Parkinson’s disease complicated with sarcopenia constructed by Hoehn-Yahr grade,albumin,vitamin D,homocysteine,creatinine and daily activity ability score was 0.819,and the area under the curve was 0.819,showing good predictive value.Decision curve analysis showed that when the probability value was 8%-82.0%,the clinical net benefit rate of patients was highest using this model.Conclusion Patients with Parkinson’s disease have a high proportion of sarcopenia,and the nomogram model constructed in this study can effectively screen the risk of sarcopenia in patie

关 键 词：帕金森病 肌少症 风险因素 预测模型 

分 类 号：R742.5[医药卫生—神经病学与精神病学]