Development and validation of a nomogram-based model for predicting postoperative pulmonary complications after coronary artery bypass grafting with cardiopulmonary bypass.

IF 0.6 Q4 CARDIAC & CARDIOVASCULAR SYSTEMS

Indian Journal of Thoracic and Cardiovascular Surgery Pub Date : 2025-10-01 Epub Date: 2025-08-13 DOI:10.1007/s12055-025-02011-9

Ying Ji, Jingjing Liu, Tao Shan, Ruoyu Jia, Hong-Guang Bao, Hong-Yu Wang, Jing Hu, Yan Shen, Qian Zhao, Yongjun Li

{"title":"Development and validation of a nomogram-based model for predicting postoperative pulmonary complications after coronary artery bypass grafting with cardiopulmonary bypass.","authors":"Ying Ji, Jingjing Liu, Tao Shan, Ruoyu Jia, Hong-Guang Bao, Hong-Yu Wang, Jing Hu, Yan Shen, Qian Zhao, Yongjun Li","doi":"10.1007/s12055-025-02011-9","DOIUrl":null,"url":null,"abstract":"Objective: Patients undergoing coronary artery bypass grafting (CABG) with cardiopulmonary bypass (CPB) are at high risk of developing postoperative pulmonary complications (PPCs). This study aimed to develop and validate a clinical prediction model for these complications after CABG.Methods: In total, 849 patients were randomly divided into training (n=594) and validation (n=255) sets in a 7:3 ratio. We used least absolute shrinkage and selection operator (LASSO) regression to identify predictive variables, incorporated them into a multivariable logistic regression model, and developed a nomogram. Model performance was assessed through discrimination (receiver operating characteristic (ROC) curve analysis, area under the curve (AUC)), calibration (calibration curves, maximum calibration error (Emax), average calibration error (Eavg)), and clinical utility assessment (decision curve analysis).Results: Five predictive indicators were selected: age, smoking history, diabetes mellitus, emergent surgery, and anesthesia duration. The model demonstrated excellent predictive performance, with an AUC of 0.902 (0.859-0.945) for the training set and 0.864 (0.811-0.917) for the validation set. Calibration curve results showed non-significant P-values from the unreliability test (P = 0.861 for training set, P = 0.741 for validation set), indicating excellent calibration. Emax and Eavg values were 0.042 and 0.013 for the training set, and 0.046 and 0.009 for the validation set, respectively, showing a strong agreement between the predicted values and actual observations.Conclusion: An original nomogram accurately predicted PPCs after CABG with CPB, which enables clinicians to rapidly assess PPC risk for individual patients without complex calculations, providing objective, quantitative evidence for preoperative risk evaluation, informed consent discussions, and perioperative management.Graphical abstract: Supplementary information: The online version contains supplementary material available at 10.1007/s12055-025-02011-9.","PeriodicalId":13285,"journal":{"name":"Indian Journal of Thoracic and Cardiovascular Surgery","volume":"41 10","pages":"1396-1407"},"PeriodicalIF":0.6000,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12450204/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Indian Journal of Thoracic and Cardiovascular Surgery","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1007/s12055-025-02011-9","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/8/13 0:00:00","PubModel":"Epub","JCR":"Q4","JCRName":"CARDIAC & CARDIOVASCULAR SYSTEMS","Score":null,"Total":0}

引用次数: 0

Abstract

Objective: Patients undergoing coronary artery bypass grafting (CABG) with cardiopulmonary bypass (CPB) are at high risk of developing postoperative pulmonary complications (PPCs). This study aimed to develop and validate a clinical prediction model for these complications after CABG.

Methods: In total, 849 patients were randomly divided into training (n=594) and validation (n=255) sets in a 7:3 ratio. We used least absolute shrinkage and selection operator (LASSO) regression to identify predictive variables, incorporated them into a multivariable logistic regression model, and developed a nomogram. Model performance was assessed through discrimination (receiver operating characteristic (ROC) curve analysis, area under the curve (AUC)), calibration (calibration curves, maximum calibration error (Emax), average calibration error (Eavg)), and clinical utility assessment (decision curve analysis).

Results: Five predictive indicators were selected: age, smoking history, diabetes mellitus, emergent surgery, and anesthesia duration. The model demonstrated excellent predictive performance, with an AUC of 0.902 (0.859-0.945) for the training set and 0.864 (0.811-0.917) for the validation set. Calibration curve results showed non-significant P-values from the unreliability test (P = 0.861 for training set, P = 0.741 for validation set), indicating excellent calibration. Emax and Eavg values were 0.042 and 0.013 for the training set, and 0.046 and 0.009 for the validation set, respectively, showing a strong agreement between the predicted values and actual observations.

Conclusion: An original nomogram accurately predicted PPCs after CABG with CPB, which enables clinicians to rapidly assess PPC risk for individual patients without complex calculations, providing objective, quantitative evidence for preoperative risk evaluation, informed consent discussions, and perioperative management.

Graphical abstract:

Supplementary information: The online version contains supplementary material available at 10.1007/s12055-025-02011-9.

查看原文本刊更多论文

基于形态学图的预测冠状动脉搭桥术合并体外循环术后肺部并发症模型的建立和验证。

目的：行冠状动脉旁路移植术（CABG）合并体外循环（CPB）的患者术后发生肺部并发症（PPCs）的风险较高。本研究旨在建立并验证CABG术后并发症的临床预测模型。方法：849例患者按7:3的比例随机分为训练组（n=594）和验证组（n=255）。我们使用最小绝对收缩和选择算子（LASSO）回归来识别预测变量，将它们纳入多变量逻辑回归模型，并开发了nomogram。通过鉴别（受试者工作特征（ROC）曲线分析、曲线下面积（AUC））、校准（校准曲线、最大校准误差（Emax）、平均校准误差（Eavg））和临床效用评估（决策曲线分析）来评估模型的性能。结果：选择年龄、吸烟史、糖尿病、急诊手术、麻醉时间5项预测指标。该模型具有良好的预测性能，训练集的AUC为0.902(0.859-0.945)，验证集的AUC为0.864（0.811-0.917）。校准曲线结果显示，不信度检验的P值无显著性（训练集P = 0.861，验证集P = 0.741），表明校准良好。训练集的Emax和Eavg值分别为0.042和0.013，验证集的Emax和Eavg值分别为0.046和0.009，表明预测值与实际观测值具有较强的一致性。结论：原始的nomogram图能准确预测CABG合并CPB后的PPC，使临床医生能够快速评估个体患者的PPC风险，无需复杂的计算，为术前风险评估、知情同意讨论和围手术期管理提供客观、定量的证据。图片摘要：补充信息：在线版本包含补充资料，可在10.1007/s12055-025-02011-9获得。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Indian Journal of Thoracic and Cardiovascular Surgery CARDIAC & CARDIOVASCULAR SYSTEMS-

CiteScore

1.20

自引率

14.30%

发文量

141

期刊介绍： The primary aim of the Indian Journal of Thoracic and Cardiovascular Surgery is education. The journal aims to dissipate current clinical practices and developments in the area of cardiovascular and thoracic surgery. This includes information on cardiovascular epidemiology, aetiopathogenesis, clinical manifestation etc. The journal accepts manuscripts from cardiovascular anaesthesia, cardiothoracic and vascular nursing and technology development and new/innovative products.The journal is the official publication of the Indian Association of Cardiovascular and Thoracic Surgeons which has a membership of over 1000 at present.DescriptionThe journal is the official organ of the Indian Association of Cardiovascular-Thoracic Surgeons. It was started in 1982 by Dr. Solomon Victor and ws being published twice a year up to 1996. From 2000 the editorial office moved to Delhi. From 2001 the journal was extended to quarterly and subsequently four issues annually have been printed out at time and regularly without fail. The journal receives manuscripts from members and non-members and cardiovascular surgeons. The manuscripts are peer reviewed by at least two or sometimes three or four reviewers who are on the panel. The manuscript process is now completely online. Funding the journal comes partially from the organization and from revenue generated by subscription and advertisement.