作者：姚元滨1 ，李卫 2 ，刘彦民 2

单位：

1.810007 青海省西宁市，青海省人民医院急诊内科　；2.810007 青海省西宁市，青海省人民医院心内科；通信作者：刘彦民，E-mail：13997245699@163.com

关键词：

心肌病，肥厚性；高血压，肺性；影响因素分析；预测模型；预测价值

关键词（英文）：

Cardiomyopathy，hypertrophic；Hypertension，pulmonary；Root cause analysis；Predictive model；Predictive value

中图分类号：

R 542.2　R 544

DOI：

基金项目：

摘要：

背景　老年肥厚型心肌病（HCM）并肺高血压患者的病死率高，预后差，而目前有关老年 HCM 患者并发肺高血压的影响因素报道较少。目的　分析老年 HCM 患者并发肺高血压的影响因素及其预测模型的预测价值。方法　选取 2016 年 6 月—2018 年 12 月于青海省人民医院收治的老年 HCM 患者 80 例，根据是否并发肺高血压分为观察组（HCM 并肺高血压患者，n=22）和对照组（单纯性 HCM 患者，n=58）。比较两组患者一般资料 { 包括性别、年龄、体质指数、病程、收缩压（SBP）、舒张压（DBP）、基础疾病〔包括慢性阻塞性肺疾病（COPD）、高血压、糖尿病〕}、心功能指标〔包括左心室舒张末期内径（LVEDD）、左心房内径（LAD）、左心室收缩末期内径（LVEDS）、左心室射血分数（LVEF）、纽约心脏病协会（NYHA）分级〕、实验室检查指标〔包括总胆固醇（TC）、白蛋白（Alb）、超敏 C 反应蛋白（hs-CRP）、N 末端脑钠肽前体（NT-proBNP）、胱抑素 C（Cyc-C）〕。老年 HCM 患者并发肺高血压的影响因素分析采用多因素 Logistic 回归分析；绘制受试者特征工作曲线（ROC 曲线）以分析预测模型对老年 HCM 并发肺高血压的预测价值。结果　两组患者性别、体质指数、病程、SBP、DBP、基础疾病、LVEDS、TC比较，差异无统计学意义（P>0.05）；观察组患者年龄、hs-CRP、NT-proBNP、Cyc-C 高于对照组，LVEDD、LAD 长于对照组，LVEF、Alb 低于对照组，NYHA 分级劣于对照组（P<0.05）。多因素 Logistic 回归分析结果显示，LVEDD〔OR=1.773，95%CI（1.205，2.609）〕、LVEF〔OR=0.716，95%CI（0.585，0.876）〕、NYHA 分 级〔OR=1.907，95%CI（1.354，2.686）〕、NT-proBNP〔OR=3.117，95%CI（1.329，7.311）〕是老年 HCM 患者并发肺高血压的影响因素（P<0.05）。根据多因素 Logistic 回归分析结果建立预测模型：P=1/〔1+e （7.041-0.573X2+0.334X4-0.646X5-1.137X9） 〕，e 为自然对数，X2 为 LVEDD，X4 为 LVEF，X5 为 NYHA 分级，X9 为 NT-proBNP。预测模型预测老年 HCM 患者并发肺高血压的 ROC 曲线下面积（AUC）为 0.755〔95%CI（0.645，0.866）〕，灵敏度为 0.727，特异度为 0.603。结论LVEDD、LVEF、NYHA 分级、NT-proBNP 是老年 HCM 患者并发肺高血压的影响因素，且预测模型对患者具有一定预测价值。

英文摘要：

Background　Elderly patients with hypertrophic cardiomyopathy （HCM） complicated with pulmonaryhypertension have high mortality and poor prognosis，but there is few reports about influencing factors on elderly HCM patientscomplicated with pulmonary hypertension.Objective　To analyze influencing factors of pulmonary hypertension in elderlypatients with HCM and predictive value of its predictive model.Methods　A total of 80 elderly patients with HCM in QinghaiProvincial People's Hospital from June 2016 to December 2018 were selected，they were divided into observation group （HCMcomplicated with pulmonary hypertension，n=22） and control group （HCM without pulmonary hypertension，n=58）.The general data〔including gender，age，BMI，course of disease，SBP，DBP and basic diseases（including COPD，hypertension，diabetes）〕，cardiac function indexes（including LVEDD，LAD，LVEDS，LVEF，NYHA grade）and laboratory inspection indexes（including TC，Alb，hs-CPR，NT-proBNP and Cys-C） between the two groups werecompared. The influencing factors of pulmonary hypertensionin in elderly patients with HCM were analyzed by multivariateLogistic regression.ROC curve was drawn to evaluate the predictive value of prediction model of pulmonary hypertension in elderlypatients with HCM.Results　There was no statistically significant difference in gender，BMI，course of disease，SBP，DBP，basic diseases，LVEDS or TC between the two groups（P>0.05）；age，hs-CRP，NT-proBNP and Cys-C in observationgroup were statistically significantly higher than those in control group，LVEDD and LAD in observation group were statisticallysignificantly longer than those in control group，LVEF and Alb in observation group were statistically significantly lower thanthose in control group，while NYHA grade in observation group was statistically significantly worse than that in control group（P<0.05）.Multivariate Logistic regression analysis results showed that，LVEDD 〔OR=1.773，95%CI（1.205，2.609）〕，LVEF 〔OR=0.716，95%CI（0.585，0.876）〕，NYHA grade 〔OR=1.907，95%CI（1.354，2.686）〕 and NT-proBNP〔OR=3.117，95%CI（1.329，7.311）〕 were influencing factors of pulmonary hypertensionin in elderly patients withHCM（P<0.05）.Predictive model was established according to the results of multivariate Logistic regression analysis：P=1/〔1+e （7.041-0.573X2+0.334X4-0.646X5-1.137X9） 〕，e was natural logarithm，X2 was LVEDD，X4 was LVEF，X5 was NYHA grade andX9 was NT-proBNP.ROC curve showed that，AUC，sensitivity and specificity of predictive model in elderly HCM patientscomplicated with pulmonary hypertension was 0.755〔95%CI（0.645，0.866）〕，0.727 and 0.603，respectively.Conclusion　LVEDD，LVEF，NYHA grade and NT-proBNP are influencing factors of elderly HCM patients complicated with pulmonaryhypertension，and prediction model has predictive value on elderly HCM patients complicated with pulmonary hypertension.

参考文献：