Wang Peihe, Lu Meiling, Huang Yu, Sun Lu, Han Zhen
Department of Cardiovascular Surgery, Peking University Shenzhen Hospital, Shenzhen, China.
Shantou University Medical College, Shantou, China.
J Thorac Dis. 2024 Jul 30;16(7):4174-4185. doi: 10.21037/jtd-24-24. Epub 2024 Jul 17.
Extracorporeal circulation auxiliary to open cardiac surgery (ECAOCS) is one of the most complex surgical procedures and carries a very high risk of death. We developed a nomogram from a retrospective study to predict the risk of death during patient hospitalization.
All clinical data were extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. We extracted clinical variables for the first 24 hours after admission to the intensive care unit (ICU) in a total of 880 patients who underwent ECAOCS. All patients were randomly divided into training and validation cohort in a ratio of 7:3. All variables included in the study were subjected to univariate logistic regression analysis. In order to prevent overfitting and to address the problem of severe covariance, all factors with P<0.05 in the univariate logistic regression analysis were analyzed using the least absolute shrinkage and selection operator (LASSO) regression. A multivariate logistic regression model was developed based on the factors output from the LASSO regression and a nomogram was plotted. The receiver operating characteristic (ROC) curve was constructed and the area under the curve (AUC) was calculated in training and validation cohort. Finally, the evaluation of the model was performed by calibration curves and Hosmer-Lemeshow goodness-of-fit test (HL test) and decision curve analysis (DCA) was performed.
Indicators included in the nomogram were anion gap (AG), central venous pressure (CVP), glucose, creatinine (Cr), prothrombin time (PT), activated partial thromboplastin time (APTT), bicarbonate ion (HCO ), cerebrovascular disease (CVD), peripheral vascular disease (PVD), and acute myocardial infarction (AMI).
Our study developed a model for predicting postoperative hospital mortality in patients underwent ECAOCS by incorporating AG, CVP, glucose, Cr, APTT, HCO , CVD, AMI, and PVD from the first 24 hours after admission to the ICU.
Extracorporeal circulation; cardiac surgery; intensive care; nomogram; prediction model.
体外循环辅助心脏直视手术（ECAOCS）是最复杂的外科手术之一，死亡风险极高。我们通过一项回顾性研究开发了一种列线图，以预测患者住院期间的死亡风险。
所有临床数据均从重症监护医学信息数据库IV（MIMIC-IV）中提取。我们提取了总共880例行ECAOCS患者入住重症监护病房（ICU）后最初24小时的临床变量。所有患者按7:3的比例随机分为训练队列和验证队列。对研究中纳入的所有变量进行单因素逻辑回归分析。为防止过度拟合并解决严重共线性问题，对单因素逻辑回归分析中P<0.05的所有因素采用最小绝对收缩和选择算子（LASSO）回归进行分析。基于LASSO回归输出的因素建立多因素逻辑回归模型并绘制列线图。构建受试者工作特征（ROC）曲线，并在训练队列和验证队列中计算曲线下面积（AUC）。最后，通过校准曲线和Hosmer-Lemeshow拟合优度检验（HL检验）对模型进行评估，并进行决策曲线分析（DCA）。
列线图中纳入的指标包括阴离子间隙（AG）、中心静脉压（CVP）、血糖、肌酐（Cr）、凝血酶原时间（PT）、活化部分凝血活酶时间（APTT）、碳酸氢根离子（HCO）、脑血管疾病（CVD）、外周血管疾病（PVD）和急性心肌梗死（AMI）。
我们的研究通过纳入入住ICU后最初24小时的AG、CVP、血糖、Cr、APTT、HCO、CVD、AMI和PVD，开发了一种预测行ECAOCS患者术后医院死亡率的模型。
体外循环；心脏手术；重症监护；列线图；预测模型。
