成爽 1邹昆 1周武 1张洪来 1董丽娟 2赵代鑫 3商洪才4
作者信息
- 1. 广州中医药大学医学信息工程学院(广州 510006)
- 2. 广东省中山市中医院护理部(广东 528400)
- 3. 广州中医药大学第二附属医院肾病科(广州 510120)
- 4. 广州中医药大学医学信息工程学院(广州 510006)||北京中医药大学东直门医院中医内科学教育部和北京市重点实验室(北京 100700)||北京中医药大学东方医院(北京 100078)
- 折叠
摘要
Abstract
Objective To investigate the risk factors for coronary atherosclerotic heart disease(CHD)in patients undergoing maintenance hemodialysis(MHD)and to construct and validate a clinical prediction model integrated Chinese medicine(CM)and Western medicine.Methods A retrospective study was conducted,including MHD patients admitted to Zhongshan Hospital of Traditional Chinese Medicine(176 cases,training set)and Guangdong Provincial Hospital of Traditional Chinese Medicine University Town Hospital(160 cases,validation set)from January 2024 to June 2025.Patients in the training set were assigned into CHD and non-CHD groups based on whether major adverse CHD events occurred.Demographic data,CM syndrome patterns,tongue parameters(color and texture features),and laboratory indicators were collected.Univariate analysis was performed on the training set to preliminarily screen potential risk factors.After collinearity diagnosis,multivariate Logistic regression was used to identify independent risk factors for CHD in MHD patients and construct a predictive model.Model performance was evaluated in the training set using the area under the receiver operating characteristic curve(AUC),Hosmer-Lemeshow test,stratified ten-fold cross-validation,and the AUC of the validation set.Results Univariate analysis showed that a history of diabetes mellitus(DM),disease's location(Shen,Gan,Xin,Pi),syndrome elements(yin deficiency,blood deficiency,blood stasis),tongue parameters[tongue_H(hue),tongue_A(red-green chromaticity)],and low-density lipoprotein cholesterol(LDL-C)levels were associated with the occurrence of CHD(P＜0.05).Multivariate Logistic regression analysis identified a history of DM,disease's location Shen,disease's locationn Xin,syndrome elements blood stasis,tongue_H,tongue_A,and LDL-C as independent risk factors.In the training set,the prediction model demonstrated excellent discriminative ability(AUC=0.910,95％CI:0.865-0.954)and good calibration(Hosmer-Lemeshow test:x2=3.834,P=0.872).In stratified ten-fold cross-validation,the average test set AUC was 0.890±0.060(95％CI:0.849-0.925).In the external validation set,the model maintained stable generalizability(AUC=0.802,95％CI:0.723-0.881).Conclusions This study constructed an integrated Chinese and Western medicine predictive model,which was validated through internal cross-validation and external validation.This model can help identify risk factors for MHD combined with CHD at an early stage,enabling timely adoption of more proactive treatment strategies for intervention.关键词
维持性血液透析/冠状动脉粥样硬化性心脏病/影响因素/中西医结合/临床预测模型/模型验证Key words
maintenance hemodialysis/coronary heart disease/risk factors/integrative medicine/clinical prediction model/model validation
