Predicting major adverse cardiovascular events after orthotopic liver transplantation using a supervised machine learning model:A cohort study

作     者：Jonathan Soldera Leandro Luis Corso Matheus Machado Rech Vinícius Remus Ballotin Lucas Goldmann Bigarella Fernanda Tomé Nathalia Moraes Rafael Sartori Balbinot Santiago Rodriguez Ajacio Bandeira de Mello Brandão Bruno Hochhegger 

作者机构：Post Graduate Program at Acute Medicine and GastroenterologyUniversity of South WalesCardiff CF371DLUnited Kingdom Postgraduate Program in PathologyFederal University of Health Sciences of Porto Alegre(UFCSPA)Porto Alegre 90050-170Brazil Department of EngineeringUniversidade de Caxias do SulCaxias do Sul 95070-560Brazil School of MedicineUniversidade de Caxias do SulCaxias do Sul 95070-560Brazil Postgraduate Program in HepatologyFederal University of Health Sciences of Porto Alegre(UFCSPA)Porto Alegre 90050-170Brazil 

出 版 物：《World Journal of Hepatology》 (世界肝病学杂志（英文版）（电子版）)

年 卷 期：2024年第16卷第2期

页      面：193-210页

学科分类：1002[医学-临床医学] 100210[医学-外科学(含：普外、骨外、泌尿外、胸心外、神外、整形、烧伤、野战外)] 10[医学] 

主　　题：Liver transplantation Major adverse cardiac events Machine learning Myocardial perfusion imaging Stress test 

摘      要：BACKGROUND Liver transplant(LT)patients have become older and *** rate of post-LT major adverse cardiovascular events(MACE)has increased,and this in turn raises 30-d post-LT *** cardiac stress testing loses accuracy when applied to pre-LT cirrhotic *** To assess the feasibility and accuracy of a machine learning model used to predict post-LT MACE in a regional *** This retrospective cohort study involved 575 LT patients from a Southern Brazilian academic *** developed a predictive model for post-LT MACE(defined as a composite outcome of stroke,new-onset heart failure,severe arrhythmia,and myocardial infarction)using the extreme gradient boosting(XGBoost)machine learning *** addressed missing data(below 20%)for relevant variables using the k-nearest neighbor imputation method,calculating the mean from the ten nearest neighbors for each *** modeling dataset included 83 features,encompassing patient and laboratory data,cirrhosis complications,and pre-LT cardiac *** performance was assessed using the area under the receiver operating characteristic curve(AUROC).We also employed Shapley additive explanations(SHAP)to interpret feature *** dataset was split into training(75%)and testing(25%)*** was evaluated using the Brier *** followed Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis guidelines for ***-learn and SHAP in Python 3 were used for all *** supplementary material includes code for model development and a user-friendly online MACE prediction *** Of the 537 included patients,23(4.46%)developed in-hospital MACE,with a mean age at transplantation of 52.9 *** majority,66.1%,were *** XGBoost model achieved an impressive AUROC of 0.89 during the training *** model exhibited accuracy,precision,recall,and F1-score values of 0.84,0.85,0.80,and 0.79,***