预测慢性肺心病合并心力衰竭患者不良预后的列线图和网络计算器  

To predict adverse outcomes of the patients with chronic cor pulmonale complicated by heart failure by using the nomogram and Web calculator

作　　者：张凯月 丁源 李闯[1] 牛丽丹[1] 张强[1] 庞欣欣 Zhang Kaiyue;Ding Yuan;Li Chuang;Niu Lidan;Zhang Qiang;Pang Xinxin(Emergency Department,the First Affiliated Hospital of Xinxiang Medical University,Xinxiang 453000,China)

机构地区：[1]新乡医学院第一附属医院急救医学科,河南新乡453000 [2]新乡医学院第一附属医院呼吸与危重症医学科二病区,河南新乡453000

出　　处：《中国急救医学》2024年第8期703-710,共8页Chinese Journal of Critical Care Medicine

摘　　要：目的开发和验证用于预测慢性肺心病合并心力衰竭患者发生死亡的列线图并制作在线网络计算器。方法回顾性纳入美国重症监护医学信息数据库(MIMIC-Ⅳ)中2012~2018年间慢性肺心病合并心力衰竭患者资料,按7∶3比例随机分为训练集(n=820)和内部验证集(n=352),采用最小绝对值收缩和选择算子(LASSO)回归算法,单因素及多因素Logistic回归分析患者不良预后的独立危险因素,绘制预测患者不良预后的列线图并建立网络计算器。收集2022~2023年间入住于新乡医学院第一附属医院急救医学科重症监护室和呼吸与危重症医学科重症监护室的慢性肺心病合并心力衰竭患者121例作为外部验证集,通过受试者工作特征(ROC)曲线下面积(AUC)、校准曲线及临床决策曲线评价该预测模型的预测效能。结果经过LASSO及单因素及多因素Logistic回归分析发现,年龄(OR=1.026,95%CI 1.006~1.049)、乳酸(OR=1.308,95%CI 1.042~1.393)、无创收缩压(OR=0.988,95%CI 0.978~0.998)、动脉血氧分压(OR=0.998,95%CI 0.996~1.000)及是否合并肺炎并发症(OR=1.883,95%CI 1.183~2.990)是导致慢性肺心病合并心力衰竭患者出现不良预后的独立危险因素;以此构建列线图预测模型,在训练集、内部验证集及外部验证集中AUC分别为0.730(95%CI 0.679~0.781)、0.856(95%CI 0.792~0.920)及0.816(95%CI 0.740~0.892),校准曲线及临床决策曲线分析该列线图模型具有较好的预测能力。结论年龄、乳酸、无创收缩压、动脉血氧分压及是否合并肺炎并发症是导致慢性肺心病合并心力衰竭患者不良预后的独立危险因素,基于其构建的列线图及网络计算器对慢性肺心病合并心力衰竭患者具有良好的预测能力,有助于临床医师更快的进行危险分层并实施临床干预。Objective To develop and validate a nomogram for predicting the mortality of patients with chronic cor pulmonale complicated by heart failure and to create an online calculator.Methods The data of patients with chronic cor pulmonale complicated by heart failure from the medical information mart for intensive careⅣ(MIMIC-IV)database between 2012 and 2018 were collected retrospectively.The patients were randomly divided into a training set(n=820)and an internal validation set(n=352)in a 7∶3 ratio.Independent risk factors for adverse outcomes were identified by using the least absolute shrinkage and selection operator(LASSO)regression,univariate and multivariate Logistic regression analysis.A nomogram was then developed to predict adverse outcomes,and an online calculator was created.Additionally,121 patients with chronic cor pulmonale complicated by heart failure admitted to the Intensive Care Unit of Emergency Medicine and the Intensive Care Unit of Respiratory and Critical Care Medicine in the First Affiliated Hospital of Xinxiang Medical University between 2022 and 2023 were collected as an external validation set.The predictive performance of the model was evaluated by using the area under the receiver operating characteristic(AUC)curve,calibration curve and clinical decision curve.Results The LASSO regression,univariate and multivariate Logistic regression analysis identified age(OR=1.026,95%CI 1.006-1.049),lactate(OR=1.308,95%CI 1.042-1.393),non-invasive systolic blood pressure(OR=0.988,95%CI 0.978-0.998),arterial oxygen partial pressure(OR=0.998,95%CI 0.996-1.000),and the presence of pneumonia complications(OR=1.883,95%CI 1.183-2.990)were independent risk factors for adverse outcomes in the patients with chronic cor pulmonale complicated by heart failure.The nomogram model showed AUCs was 0.730(95%CI 0.679-0.781)in the training set,0.856(95%CI 0.792-0.920)in the internal validation set,and 0.816(95%CI 0.740-0.892)in the external validation set,indicating good predictive ability as evidenced by calibratio

关 键 词：慢性肺源性心脏病 心力衰竭 列线图 网络计算器 预后 

分 类 号：R541[医药卫生—心血管疾病]