早发冠心病危险因素分析及列线图模型构建  

Analysis of risk factors for premature coronary artery disease and construction of a nomogram model

作　　者：陈润林 廖江波 皮建彬 赵华云[2] CHEN Runlin;LIAO Jiangbo;PI Jianbin;ZHAO Huayun(The Eighth Clinical Medical College of Guangzhou University of Chinese Medicine,Foshan 528000,China;Department of Cardiovascular Medicine,Foshan Hospital of Traditional Chinese Medicine,Foshan 528000,China)

机构地区：[1]广州中医药大学第八临床医学院,广东佛山528000 [2]佛山市中医院心血管内科,广东佛山528000

出　　处：《现代医学》2025年第4期522-529,共8页Modern Medical Journal

基　　金：广东省医学科学技术研究基金项目(B2024005)。

摘　　要：目的:探讨早发冠心病(PCAD)的危险因素并构建相关的列线图模型。方法:收集2023年1月至2024年7月佛山市中医院收治的111例早发冠心病患者(男63例,女48例)作为PCAD组,并按照性别进行频数匹配,纳入同期在该院诊疗的222例非冠心病患者(男126例,女96例)作为非PCAD组。对比两组患者的一般资料、实验室指标及甘油三酯葡萄糖指数(TyG指数)。采用Lasso回归筛选变量,并将筛选出的变量纳入Logistic回归模型。绘制列线图模型,并通过受试者工作特征(ROC)曲线及其曲线下面积(AUC)和C指数评估模型的区分度。进行Hosmer-Lemeshow(H-L)拟合优度检验并绘制校准曲线以评价校准度,同时绘制临床决策曲线以分析模型的临床适用性。结果:心电图异常、心脏彩超异常、高血压病病史、低密度脂蛋白胆固醇(LDL-C)、空腹血糖(FBG)是预测PCAD的危险因素(P<0.05)。基于危险因素成功构建预测PCAD的列线图模型。训练集的C指数为0.926(95%CI 0.888~0.965),验证集为0.940(95%CI 0.893~0.987)。ROC曲线分析显示,组合模型预测PCAD具有优秀区分度,训练集AUC为0.926(95%CI 0.888~0.964);验证集AUC为0.949(95%CI 0.906~0.992)。H-L检验结果显示良好,训练集结果:χ^(2)=3.1041,P=0.928;验证集结果:χ^(2)=2.0201,P=0.980。训练集和验证集的实际曲线和校正曲线的轨迹均具有较强一致性。决策曲线分析显示,组合模型具有良好的临床适用性,训练集在0.00~0.90概率内,组合模型净收益>0;验证集在0.00~0.94概率内,组合模型净收益>0。结论:心电图异常、心脏彩超异常、高血压病病史、LDL-C、FBG为早发冠心病的危险因素,基于上述因素建立的列线图模型具有良好的预测价值。Objective:To investigate the risk factors for premature coronary artery disease(PCAD)and develop a nomogram model.Methods:A total of 111 PCAD patients(63 males,48 females)admitted to Foshan Hospital of Traditional Chinese Medicine between January 2023 and July 2024 were enrolled as the PCAD group.A control group(non PCAD group)of 222 non-coronary heart disease patients(126 males,96 females)was frequency-matched by sex.General characteristics,laboratory indices,and the TyG index were compared between the two groups.Variables were screened using LASSO regression,and a Logistic regression model was constructed to develop a nomogram model.Discriminative ability was evaluated via receiver operating characteristic(ROC)curves,area under the curve(AUC),and the concordance index(C-index).Calibration was assessed using the Hosmer-Lemeshow test and calibration curves.Clinical utility was analyzed through decision curve analysis(DCA).Results:Independent risk factors for PCAD included electrocardiogram(ECG)abnormalities,echocardiographic abnormalities,hypertension history,LDL-C,and fasting blood glucose(FBG)(P<0.05).The nomogram demonstrated excellent discrimination in both the training set(C-index:0.926,95%CI 0.888-0.965;AUC:0.926,95%CI 0.888-0.964)and validation set(C-index:0.940,95%CI 0.893-0.987;AUC:0.949,95%CI 0.906-0.992).The Hosmer-Lemeshow test indicated strong calibration(training set:χ^(2)=3.1041,P=0.928;validation set:χ^(2)=2.0201,P=0.980),with calibration curves showing close alignment between predicted and observed probabilities.DCA confirmed clinical applicability,with net benefits>0 across risk thresholds of 0.00-0.90(training set)and 0.00-0.94(validation set).Conclusion:ECG abnormalities,echocardiographic abnormalities,hypertension history,LDL-C,and FBG are significant risk factors for PCAD.The nomogram model based on these factors exhibits robust predictive performance and clinical utility.

关 键 词：早发冠心病 危险因素 列线图模型 

分 类 号：R541.4[医药卫生—心血管疾病]