中老年冠心病心绞痛患者中医虚证诊断模型的探索与构建  

Exploration and construction of a diagnostic model for traditional Chinese medicine deficiency syndrome in middle-aged and elderly patients with angina pectoris of coronary heart disease

作　　者：薛鹏[1] 林玲 朱亚娟 谢桂香[1] 王亮 XUE Peng;LIN Ling;ZHU Yajuan;XIE Guixiang;WANG Liang(Department of Geriatric Cardiology,Taizhou Hospital Affiliated to Nanjing University of Chinese Medicine,Jiangsu Province,Taizhou225300,China;Department of Gynecology,Taizhou People’s Hospital,Jiangsu Province,Taizhou225300,China;Department of Education and Teaching,Affiliated Hospital of Nanjing University of Chinese Medicine,Jiangsu Province,Nanjing210029,China)

机构地区：[1]南京中医药大学泰州附属医院老年心血管病科,江苏泰州225300 [2]江苏省泰州市人民医院妇科,江苏泰州225300 [3]南京中医药大学附属医院教育教学处,江苏南京210029

出　　处：《中国医药导报》2024年第12期27-31,共5页China Medical Herald

基　　金：江苏省卫生健康委员会老年健康科研项目(LK2021058、LD2021040)。

摘　　要：目的 构建中老年冠心病心绞痛中医虚证的诊断预测模型。方法 选取2021年1月至2022年12月南京中医药大学泰州附属医院住院的160例冠心病心绞痛中老年(50~85岁)患者的临床资料,统计年龄、性别、体重指数、吸烟、饮酒、血常规、血脂、肝肾功能、冠状动脉血管病变程度,以及高血压病、糖尿病、胃肠道疾病、慢性阻塞性肺疾病、脑血管疾病等合并症作为自变量,每位患者均给出虚实辨证,其结果作为因变量。应用单因素logistic回归筛选变量,多因素logistic回归分析建立模型,可视化模型。使用受试者操作特征曲线、校准曲线及决策曲线分析评估模型。结果 160例患者实证83例,虚证77例。多因素回归结果显示:年龄>67~85岁[OR=2.776,95%CI(1.219,6.321),P=0.015]、糖尿病[OR=2.230,95%CI(0.960,5.178),P=0.062]、Genisi评分4.50~7.50[OR=10.126,95%CI(2.574,39.843),P=0.001]、Genisi评分8.00~15.75 [OR=13.510,95%CI(3.441,53.040),P<0.001]、Genisi评分16.00~80.00[OR=53.032,95%CI(11.769,230.027),P<0.001]是冠心病心绞痛虚证的预测因子。模型方程为:Logit(虚证概率)=-4.231+1.021×年龄(>67~85岁)+0.802×患糖尿病+2.315×Gensini等级(4.50~7.50)+2.603×Gensini等级(8.00~15.75)+3.952×Gensini等级(16.00~80.00)。该模型受试者曲线下面积值为0.852(95%CI:0.795~0.910)。校准度曲线提示准确度较好,决策曲线分析表现在该预测变量在6%~85%广大风险范围内展现出较好的正向收益率。结论 本研究构建了一个中老年冠心病心绞痛虚证诊断预测模型,有较好的临床预测能力。Objective To construct a diagnostic prediction model of traditional Chinese medicine(TCM) deficiency syndrome of angina pectoris in middle-aged and elderly patients with coronary heart disease.Methods Clinical data of 160 middleaged and elderly(50-85 years) patients with angina pectoris of coronary heart disease and who were hospitalized in Taizhou Hospital Affiliated to Nanjing University of Chinese Medicine from January 2021 to December 2022 were selected.Age,gender,body mass index,smoking,drinking,blood routine,blood lipids,liver and kidney function,degree of coronary artery disease,and hypertension,diabetes,gastrointestinal disease,chronic obstructive pulmonary cerebrovascular disease were counted as independent variables.Each patient was given the deficiency and excess syndrome differentiation,and the results were used as dependent variables.Univariate logistic regression was used to screen variables,and multivariate logistic regression analysis was used to establish a model.A nomogram was used to present the model.Receiver operating curve,calibration curve,and decision curve analysis were used to evaluate the model.Results Among 160 patients,83 cases were positive syndrome and 77 cases were deficiency syndrome.Multivariate regression analysis showed that age >67-85years [OR=2.776,95%CI(1.219,6.321),P=0.015],diabetes mellitus[OR=2.230,95%CI(0.960,5.178),P=0.062],Genisi score 4.50-7.50 [OR=10.126,95%CI(2.574,39.843),P=0.001],Genisi score 8.00-15.75 [OR=13.510,95%CI(3.441,53.040),P<0.001],Genisi score 16.00-80.00[OR=53.032,95%CI(11.769,230.027),P<0.001] were predictors of angina pectoris deficiency syndrome.The model equation was as follows,Logit(deficiency probability)=-4.231+1.021×age(>67-85 years) +0.802×diabetes +2.315×Gensini level(4.50-7.50) +2.603×Gensini level(8.00-15.75) +3.952×Gensini level(16.00-80.00).The area under the curve was 0.852(95%CI:0.795-0.910).Calibration curve showed good accuracy,and decision curve analysis showed that the predictor variable showed a good positive rate of return i

关 键 词：心绞痛 中医分型 虚证 LOGISTIC回归分析 预测模型 列线图 

分 类 号：R541[医药卫生—心血管疾病]