列线图预测急性心肌梗死合并肾功能不全患者全因死亡的影响因素  

Influencing factors of all-cause death among patients with acute myocardial infarction complicated with renal insufficiency by nomogram

作　　者：李建彬 韩晓峰[1] 胡凤英 刘菁 李蓉 郭曦[5] Li Jianbin;Han Xiaofeng;Hu Fengying;Liu Jing;Li Rong;Guo Xi(Department of Interventional,Beijing Anzhen Hospital,Capital Medical University,Beijing Institute of Heart,Lung and Blood Vessel Diseases,Beijing 100029,China;Department of Cardiology,Tangshan Central Hospital,Tangshan 063000,China;Department of Respiratory and Critical Care Medicine,Beijing Ditan Hospital,Capital Medical University,Beijing 100015,China;Department of Ultrasound,Chinese People's Liberation Army Rocket Army General Hospital,Beijing 100088,China;Department of Interventional,Beijing Anzhen Hospital,Capital Medical University,Beijing 100029,China)

机构地区：[1]北京市心肺血管病研究所首都医科大学附属北京安贞医院介入诊疗科,北京100029 [2]唐山中心医院心内科,唐山063000 [3]首都医科大学附属北京地坛医院呼吸危重症科,北京100015 [4]中国人民解放军火箭军总医院超声科,北京100088 [5]首都医科大学附属北京安贞医院介入诊疗科,北京100029

出　　处：《中华胸心血管外科杂志》2022年第11期665-672,共8页Chinese Journal of Thoracic and Cardiovascular Surgery

摘　　要：目的探讨急性心肌梗死合并肾功能不全患者全因死亡的影响因素,并建立临床预测工具。方法回顾性收集2014年1月1日至2019年4月30日在北京安贞医院住院并诊断为急性心肌梗死合并肾功能不全的727例患者。记录患者基线特征、既往病史、本次并发症、实验室及辅助检查结果、治疗方法(包括靶血管重建方式、用药等),随访患者全因死亡的发生情况。首先,在训练集(485例)应用多因素Cox回归分析构建模型。其次,在验证集(242例)绘制受试者工作特征曲线、校准曲线和临床决策曲线,进一步验证预测模型的效果。最后,根据验证后的风险因素开发列线图。结果多因素Cox回归分析发现,急性心肌梗死合并肾功能不全患者全因死亡预后的预测因子有9个,其中危险因素包括女性、共病、直接胆红素高、合并心源性休克、合并呼吸衰竭、合并感染、血滤;保护性因素包括PCI和服用阿司匹林。模型评估结果显示,验证集接受者操作特征曲线(ROC曲线)下面积为0.82;校准线贴近理想校准线,且斜率≈1,截距≈0,拟合效果良好;临床决策曲线在患者全因病死率为35%时,基于Cox模型执行主动干预患者净获益为38%。基于经评估预测效果良好的预后风险因素开发了可视化的列线图模型。结论女性、共病、直接胆红素高、合并心源性休克、合并呼吸衰竭、合并感染、血滤、PCI和服用阿司匹林这9个因素均会影响急性心肌梗死合并肾功能不全患者全因死亡的发生概率,本研究所开发的模型准确性较高。Objective To explore the influencing factors of all-cause death in patients with acute myocardial infarction(AMI)and renal insufficiency(RI),and establish a clinical prediction tool.Methods Collected 727 patients who were hospitalized in Beijing Anzhen Hospital from January 1st 2014 to April 31th 2019,and diagnosed as AMI with RI.Recorded the patients'baseline characteristics,past medical history,current complications,laboratory and auxiliary examination results,treatment methods(included target vessel reconstruction methods,medications,etc.),and follow-up patients for all-cause deaths.Firstly,multivariate Cox regression analysis was used to construct the model in the training set(485 cases).Secondly,the receiver operating characteristic curve(ROC),calibration curve and clinical decision curve analysis(DCA)were drawn in the validation set(242 cases)to further verify the effect of the prediction model.Finally,a nomogram was developed based on the verified risk factors.Results Multivariate Cox regression analysis found that there were nine predictors for the prognosis of all-cause death in patients with AMI&RI,the dangerous factors included women,comorbidities,high direct bilirubin,cardiogenic shock,respiratory failure,co-infection and hemofiltration;the protective factors included PCI and taking aspirin.Model evaluation results showed that the AUC of the validation set was 0.82;the calibration line was close to the ideal calibration line,and the slope≈1,the intercept≈0,and the fitting effect was good.Clinical decision-making when the all-cause mortality rate was 35%,the net benefit of active intervention patients based on the Cox model was 38%.A visual nomogram model was developed based on the prognostic risk factors that have been evaluated and predicted to be good.Conclusion Women,with a history of comorbidities,direct high bilirubin,combined with cardiogenic shock,combined with respiratory failure,combined infection,hemofiltration,PCI and taking aspirin,these nine factors will affect all causes of AMI&RI.Fo

关 键 词：急性心肌梗死 肾功能不全 全因死亡 列线图 

分 类 号：R542.22[医药卫生—心血管疾病] R692.5[医药卫生—内科学]