Risk factors for postoperative respiratory failure in patients undergoing cardiovascular surgery and construction of a prediction model

ZHOU Fuchao; LIU Xiang; LU Kaizhi; SHAO Jianliang

doi:10.16016/j.2097-0927.202507047

您当前的位置：

首页 >

文章列表页 >

Risk factors for postoperative respiratory failure in patients undergoing cardiovascular surgery and construction of a prediction model

Clinical Medicine | 更新时间：2025-08-25

- Risk factors for postoperative respiratory failure in patients undergoing cardiovascular surgery and construction of a prediction model
- Journal of Army Medical University Vol. 47, Issue 16, Pages: 1970-1980(2025)
- 作者机构：
  
  陆军军医大学（第三军医大学）第一附属医院麻醉科，重庆
- 作者简介：
  
  LU Kaizhi， E-mail：lukaizhi2010@163.com；
  SHAO Jianliang， E-mail：654708721@qq.com
- 基金信息：
- DOI：10.16016/j.2097-0927.202507047
  CLC： R181.32;R563.806;R654.06
- Received：15 July 2025，
  
  Revised：01 August 2025，
  
  Published：30 August 2025
- 稿件说明：
移动端阅览
周辅超，刘祥，鲁开智，等.心血管手术患者术后呼吸衰竭的危险因素及预测模型构建［J］.陆军军医大学学报，2025，47（16）：1970-1980.DOI：10.16016/j.2097-0927.202507047.

ZHOU F C，LIU X，LU K Z，et al.Risk factors for postoperative respiratory failure in patients undergoing cardiovascular surgery and construction of a prediction model［J］.J Army Med Univ，2025，47（16）：1970-1980.DOI：10.16016/j.2097-0927.202507047.
周辅超，刘祥，鲁开智，等.心血管手术患者术后呼吸衰竭的危险因素及预测模型构建［J］.陆军军医大学学报，2025，47（16）：1970-1980.DOI：10.16016/j.2097-0927.202507047. DOI：

ZHOU F C，LIU X，LU K Z，et al.Risk factors for postoperative respiratory failure in patients undergoing cardiovascular surgery and construction of a prediction model［J］.J Army Med Univ，2025，47（16）：1970-1980.DOI：10.16016/j.2097-0927.202507047. DOI：

摘要

目的

基于机器学习算法识别心血管手术患者术后呼吸衰竭（postoperative respiratory failure， PORF）的危险因素，并构建特异性的风险预测模型。

方法

回顾性的纳入INSPIRE数据库2011到2020年间行心血管手术的1 623例患者，进行数据质量分析后，使用多重插补处理缺失数据，并采用Boruta筛选相关特征变量。基于筛选指标构建8种机器学习模型，包括GBM、GLM、XGBoost、KNN、NNET、NB、SVM及RF，以曲线下面积（area under the curve， AUC）、灵敏度、特异性等指标评估模型性能，并基于排列重要性算法识别对PORF有重要影响的变量。

结果

PORF总体发生率为27.05%（439/1 623），PORF组患者院内死亡率显著高于非PORF组（12.98%

. 1.60%，

0.001）。建立的预测模型中，SVM模型效果较好，在测试集中，该模型AUC达到0.705，其灵敏度、特异度、阳性和阴性预测值分别为0.481、0.825、0.504、0.812。根据特征重要性计算，对术后呼吸衰竭预测影响最大的前十个变量为麻醉时长、动脉血二氧化碳分压、钙、淋巴细胞百分比、体外循环时长、术中出血量、年龄、肌酐、天门冬氨基酸氨基转移酶和活化部分凝血活酶时间。

结论

本研究成功构建了预测心血管手术患者术后呼吸衰竭的模型，该模型可以识别出高危患者及其发生呼吸衰竭的概率，促进数据驱动的临床决策。

Abstract

Objective‍　‍To identify risk factors for postoperative respiratory failure (PORF) in cardiovascular surgery patients using machine learning algorithms and to construct a specific risk prediction model. Methods‍　‍A retrospective cohort was conducted on 1 623 patients undergoing cardiovascular surgery between 2011 and 2020 from the INSPIRE database. Following data quality analysis

multiple imputation was employed to

handle missing data

and the Boruta algorithm was used for feature selection. Eight machine learning models were constructed based on the selected features

including Gradient Boosting Machine (GBM)

Generalized Linear Model (GLM)

Extreme Gradient Boosting (XGBoost)

K-Nearest Neighbors (KNN)

Neural Network (NNET)

Naive Bayes (NB)

Support Vector Machine (SVM)

and Random Forest (RF). Model performance was evaluated using metrics including the area under the curve (AUC)

sensitivity

and specificity. Variables significantly influencing PORF were identified using the permutation importance algorithm. Results‍　‍The overall incidence of PORF was 27.05% (439/1 623). The in-hospital mortality rate was significantly higher in the PORF group than the non-PORF group (12.98%

1.60%

0.001). Among the developed models

the SVM model demonstrated the best performance

achieving an AUC of 0.705 in the testing set

with a sensitivity

specificity

positive predictive value (PPV)

and negative predictive value (NPV) of 0.481

0.825

0.504

and 0.812

respectively. Based on feature importance analysis

the top 10 variables most predictive of PORF were anesthesia duration

arterial partial pressure of carbon dioxide (PaCO₂)

calcium level

lymphocyte percentage

cardiopulmonary bypass duration

intraoperative blood loss

age

creatinine level

aspartate aminotransferase (AST) level

and activated partial thromboplastin time (aPTT). Conclusion‍　‍A predictieon model for PORF following cardiovascular surgery is successfully developed. This model can identify high-risk patients and estimate their probability of developing respiratory failure

thereby facilitating data-driven clinical decision-making.

关键词

Keywords

references

NTELIOPOULOS G ， NIKOLAKOPOULOU Z ， CHOW B H N ， et al . Lung injury following cardiopulmonary bypass： a clinical update ［J］. Expert Rev Cardiovasc Ther ， 2022 ， 20 （ 11 ）： 871 - 880 . DOI： 10.1080/14779072.2022.2149492 http://dx.doi.org/10.1080/14779072.2022.2149492 .

MILLS G H . Respiratory complications of anaesthesia ［J］. Anaesthesia ， 2018 ， 73 （ Suppl 1 ）： 25 - 33 . DOI： 10.1111/anae.14137 http://dx.doi.org/10.1111/anae.14137 .

FERNANDEZ-BUSTAMANTE A ， FRENDL G ， SPRUNG J ， et al . Postoperative pulmonary complications， early mortality， and hospital stay following noncardiothoracic surgery： a multicenter study by the perioperative research network investigators ［J］. JAMA Surg ， 2017 ， 152 （ 2 ）： 157 - 166 . DOI： 10.1001/jamasurg.2016.4065 http://dx.doi.org/10.1001/jamasurg.2016.4065 .

PEI J Q ， CAI L ， WANG F ， et al . LPA（2） contributes to vascular endothelium homeostasis and cardiac remodeling after myocardial infarction ［J］. Circ Res ， 2022 ， 131 （ 5 ）： 388 - 403 . DOI： 10.1161/CIRCRESAHA.122.321036 http://dx.doi.org/10.1161/CIRCRESAHA.122.321036 .

JIA P ， JI Q ， ZOU Z ， et al . Effect of delayed remote ischemic preconditioning on acute kidney injury and outcomes in patients undergoing cardiac surgery： a randomized clinical trial ［J］. Circulation ， 2024 ， 150 （ 17 ）： 1366 - 1376 . DOI： 10.1161/circulationaha.124.071408 http://dx.doi.org/10.1161/circulationaha.124.071408 .

MISKOVIC A ， LUMB A B . Postoperative pulmonary complications ［J］. Br J Anaesth ， 2017 ， 118 （ 3 ）： 317 - 334 . DOI： 10.1093/bja/aex002 http://dx.doi.org/10.1093/bja/aex002 .

LI W ， LIAO Y ， CHEN J ， et al . Ischemia - Reperfusion injury： a roadmap to precision therapies ［J］. Mol Aspects Med ， 2025 ， 104 ： 101382 . DOI： 10.1016/j.mam. 2025.101382 http://dx.doi.org/10.1016/j.mam.2025.101382 .

STONE P H ， LIBBY P ， BODEN W E . Fundamental pathobiology of coronary atherosclerosis and clinical implications for chronic ischemic heart disease management-the plaque hypothesis： a narrative review ［J］. JAMA Cardiol ， 2023 ， 8 （ 2 ）： 192 - 201 . DOI： 10.1001/jamacardio.2022.3926 http://dx.doi.org/10.1001/jamacardio.2022.3926 .

GUPTA P K ， GUPTA H ， SUNDARAM A ， et al . Development and validation of a risk calculator for prediction of cardiac risk after surgery ［J］. Circulation ， 2011 ， 124 （ 4 ）： 381 - 387 . DOI： 10.1161/circulationaha.110.015701 http://dx.doi.org/10.1161/circulationaha.110.015701 .

LIM L ， LEE H ， JUNG C W ， et al . INSPIRE， a publicly available research dataset for perioperative medicine ［J］. Sci Data ， 2024 ， 11 （ 1 ）： 655 . DOI： 10.1038/s41597-024-03517-4 http://dx.doi.org/10.1038/s41597-024-03517-4 .

STOCKING J C ， UTTER G H ， DRAKE C ， et al . Postoperative respiratory failure： an update on the validity of the Agency for Healthcare Research and Quality Patient Safety Indicator 11 in an era of clinical documentation improvement programs ［J］. Am J Surg ， 2020 ， 220 （ 1 ）： 222 - 228 . DOI： 10.1016/j.amjsurg.2019.11.019 http://dx.doi.org/10.1016/j.amjsurg.2019.11.019 .

JAMMER I ， WICKBOLDT N ， SANDER M ， et al . Standards for definitions and use of outcome measures for clinical effectiveness research in perioperative medicine： European Perioperative Clinical Outcome （EPCO） definitions： a statement from the ESA-ESICM joint taskforce on perioperative outcome measures ［J］. Eur J Anaesthesiol ， 2015 ， 32 （ 2 ）： 88 - 105 . DOI： 10.1097/EJA.0000000000000118 http://dx.doi.org/10.1097/EJA.0000000000000118 .

VAN BUUREN S ， GROOTHUIS-OUDSHOORN K . mice ： multivariate imputation by chained equations in R ［J］. J Stat Soft ， 2011 ， 45 （ 3 ）： 1 - 67 . DOI： 10.18637/jss.v045.i03 http://dx.doi.org/10.18637/jss.v045.i03

HUANG J Z ， LIU W C . Comparison of machine learning models for predicting stroke risk in hypertensive patients： Lasso regression model， random forest model， Boruta algorithm model， and Boruta algorithm combined with Lasso regression model ［J］. Medicine ， 2025 ， 104 （ 22 ）： e42690 . DOI： 10.1097/MD.0000000000042690 http://dx.doi.org/10.1097/MD.0000000000042690 .

KURSA M B ， RUDNICKI W R . Feature selection with the Boruta Package ［J］. J Stat Soft ， 2010 ， 36 （ 11 ）： 1 - 13 . DOI： 10.18637/jss.v036.i11 http://dx.doi.org/10.18637/jss.v036.i11

AGYEKUM E A ， ZHANG Y Z ， FANG Y ， et al . Ultrasound-based classification of follicular thyroid Cancer using deep convolutional neural networks with transfer learning ［J］. Sci Rep ， 2025 ， 15 （ 1 ）： 21708 . DOI： 10.1038/s41598-025-05551-7 http://dx.doi.org/10.1038/s41598-025-05551-7 .

STEYERBERG E W ， VERGOUWE Y . Towards better clinical prediction models： seven steps for development and an ABCD for validation ［J］. Eur Heart J ， 2014 ， 35 （ 29 ）： 1925 - 1931 . DOI： 10.1093/eurheartj/ehu207 http://dx.doi.org/10.1093/eurheartj/ehu207 .

STOCKING J C ， DRAKE C ， ALDRICH J M ， et al . Outcomes and risk factors for delayed-onset postoperative respiratory failure： a multi-center case-control study by the University of California Critical Care Research Collaborative （UC（3）RC）［J］. BMC Anesthesiol ， 2022 ， 22 （ 1 ）： 146 . DOI： 10.1186/s12871-022-01681-x http://dx.doi.org/10.1186/s12871-022-01681-x .

LIVINGSTONE S A ， WILDI K S ， DALTON H J ， et al . Coagulation dysfunction in acute respiratory distress syndrome and its potential impact in inflammatory subphenotypes ［J］. Front Med ， 2021 ， 8 ： 723217 . DOI： 10.3389/fmed.2021.723217 http://dx.doi.org/10.3389/fmed.2021.723217 .

CASSELMAN F P A ， LANCE M D ， AHMED A ， et al . 2024 EACTS/EACTAIC guidelines on patient blood management in adult cardiac surgery in collaboration with EBCP ［J］. J Cardiothorac Vasc Anesth ， 2025 ， 39 （ 8 ）： 1964 - 2018 . DOI： 10.1053/j.jvca.2024.10.006 http://dx.doi.org/10.1053/j.jvca.2024.10.006 .

DING X C ， CHENG D ， SUN B ， et al . Nomogram and risk calculator for severe hypoxemia after heart valve surgery ［J］. Front Cardiovasc Med ， 2022 ， 9 ： 972449 . DOI： 10.3389/fcvm.2022.972449 http://dx.doi.org/10.3389/fcvm.2022.972449 .

SMETANA G W ， LAWRENCE V A ， CORNELL J E . Preoperative pulmonary risk stratification for noncardiothoracic surgery： systematic review for the American College of Physicians ［J］. Ann Intern Med ， 2006 ， 144 （ 8 ）： 581 - 595 . DOI： 10.7326/0003-4819-144-8-200604180-00009 http://dx.doi.org/10.7326/0003-4819-144-8-200604180-00009 .

TAKAGAKI M ， YAMAGUCHI H ， IKEDA N ， et al . Post-cardiotomy venovenous extracorporeal membrane oxygenation without heparinization ［J］. Gen Thorac Cardiovasc Surg ， 2019 ， 67 （ 11 ）： 982 - 986 . DOI： 10.1007/s11748-018-0990-2 http://dx.doi.org/10.1007/s11748-018-0990-2 .

BOUABDALLAOUI N ， STEVENS S R ， DOENST T ， et al . Society of thoracic surgeons risk score and EuroSCORE-2 appropriately assess 30-day postoperative mortality in the STICH trial and a contemporary cohort of patients with left ventricular dysfunction undergoing surgical revascularization ［J］. Circ Heart Fail ， 2018 ， 11 （ 11 ）： e005531 . DOI： 10.1161/CIRCHEARTFAILURE.118.005531 http://dx.doi.org/10.1161/CIRCHEARTFAILURE.118.005531 .

KUMAR A ， SATO K ， NARAYANSWAMI J ， et al . Current society of thoracic surgeons model reclassifies mortality risk in patients undergoing transcatheter aortic valve replacement ［J］. Circ Cardiovasc Interv ， 2018 ， 11 （ 9 ）： e006664 . DOI： 10.1161/CIRCINTERVENTIONS.118.006664 http://dx.doi.org/10.1161/CIRCINTERVENTIONS.118.006664 .

MAZO V ， SABATÉ S ， CANET J ， et al . Prospective external validation of a predictive score for postoperative pulmonary complications ［J］. Anesthesiology ， 2014 ， 121 （ 2 ）： 219 - 231 . DOI： 10.1097/ALN.0000000000000334 http://dx.doi.org/10.1097/ALN.0000000000000334 .

TONER A J ， CORCORAN T B ， VLASKOVSKY P S ， et al . Inflammation risk before cardiac surgery and the treatment effect of intraoperative dexamethasone ［J］. Anaesth Intensive Care ， 2024 ， 52 （ 1 ）： 28 - 36 . DOI： 10.1177/0310057X231195098 http://dx.doi.org/10.1177/0310057X231195098 .

KESKINIDOU C ， VASSILIOU A G ， PAPOUTSI E ， et al . Dysregulated coagulation and fibrinolysis are present in patients admitted to the emergency department with acute hypoxemic respiratory failure： a prospective study ［J］. Biomedicines ， 2024 ， 12 （ 5 ）： 1081 . DOI： 10.3390/biomedicines12051081 http://dx.doi.org/10.3390/biomedicines12051081 .

ZHANG J R ， ZHANG H ， WANG J ， et al . The organ-protective effects of nitric oxide in adult patients undergoing cardiac surgery with cardiopulmonary bypass： a systematic review and meta-analysis ［J］. BMC Anesthesiol ， 2025 ， 25 （ 1 ）： 335 . DOI： 10.1186/s12871-025-03207-7 http://dx.doi.org/10.1186/s12871-025-03207-7 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Construction and validation of a prediction model for central lymph node metastasis of papillary thyroid carcinoma based on contrast-enhanced venous phase CT radiomics

Pregnancy probability prediction models based on 5 machine learning algorithms and comparison of their performance

Construction and validation of a risk prediction model for high altitude de-acclimatization syndrome

Construction of a prediction model of histologic chorioamnionitis based on machine learning

Construction and validation of a Klotho-based machine learning model for predicting all-cause mortality in chronic kidney disease

Related Author

HE Xingyun

LIU Chen

DU Junze

LI Qian

CHEN Kang

FAN Rui

WANG Jian

REN Chao

Related Institution

Department of Radiology

Department of Breast and Thyroid Surgery， First Affiliated Hospital， Army Medical University （Third Military Medical University）， Chongqing

Department of Radiology， No. 9

Hospital of PLA Army

Institute of Toxicology， Faculty of Military Preventive Medicine， Army Medical University （Third Military Medical University）

⁰