肥厚型心肌病患者长期全因死亡风险预测列线图模型的建立

Establishment of a nomogram model for long-term all-cause mortality risk assessment in patients with hypertrophic cardiomyopathy

目的:探讨肥厚型心肌病(HCM)患者长期全因死亡的危险因素,并构建个体化预后风险预测模型。方法:选取2014年6月至2022年6月就诊于郑州大学第一附属医院的HCM患者346例,收集患者的临床基线特征、实验室和影像学指标以及随访资料,采用Cox回归筛选HCM全因死亡的影响因素。采用R软件基于筛选出的影响因素构建HCM患者全因死亡预测的列线图模型,使用Bootstrap法对模型进行内部验证,采用一致性指数和校准曲线评价模型的区分度和校准度,使用ROC曲线评价预测效果。结果:Cox回归分析显示左心室流出道压力阶差、血红蛋白、N末端B型脑钠肽原、非持续性室速是HCM患者全因死亡的影响因素。所构建的列线图模型内部验证一致性指数(95%CI)为0.757(0.684~0.855),校准曲线显示模型表现出较好的预测符合度。ROC曲线结果显示模型预测5 a、7 a全因死亡的AUC(95%CI)分别为0.772(0.695~0.863)、0.818(0.707~0.933)。结论:所构建的HCM患者长期全因死亡风险预测列线图模型具有较高的区分度和良好的拟合度。

Aim:To explore the risk factors of long-term all-cause mortality in patients with hypertrophic cardiomyopathy,and develop a prediction model to provide individualized prognostic risk assessment.Methods:A total of 346 patients with hypertrophic cardiomyopathy who visited the First Affiliated Hospital of Zhengzhou University from June 2014 to June 2022 were selected,and the clinical baseline characteristics,laboratory and imaging indicators,and follow-up data of the patients were collected.Cox regression analysis was used to screen the influencing factors of all-cause mortality in patients with hypertrophic cardiomyopathy.A nomogram model was developed based on the screened influencing factors using R software and verified by Bootstrap method.C index and calibration curve were used to evaluate the discrimination and calibration of the nomogram model,and ROC curve was used to evaluate the predictive efficacy.Results:Cox regression analysis showed that the left ventricular outflow tract gradient,hemoglobin,NT-proBNP,and non-sustained ventricle tachycardia were influencing factors for all-cause mortality in patients with hypertrophic cardiomyopathy.C index(95%CI)of the constructed model was 0.757(0.684-0.855).The calibration curve showed that the constructed model revealed a good predictive conformity.The ROC curve showed that AUC(95%CI)of the model for predicting 5 or 7 year all-cause mortality was 0.772(0.695-0.863)and 0.818(0.707-0.933).Conclusion:The constructed nomogram model for risk prediction of long-term all-cause mortality in patients with hypertrophic cardiomyopathy has high discrimination and good fitting degree.