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

Zhang Kaiyue; Ding Yuan; Li Chuang; Niu Lidan; Zhang Qiang; Pang Xinxin

doi:10.3969/j.issn.1002-1949.2024.08.008

Abstract：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 care Ⅳ(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 calibration and clinical decision curves.Conclusions Age,lactate,non-invasive systolic blood pressure,arterial oxygen partial pressure,and the presence of pneumonia complications are independent risk factors for adverse outcomes in patients with chronic cor pulmonale complicated by heart failure.The nomogram and online calculator based on these factors demonstrate good predictive ability,assisting the clinicians the in risk stratification and clinical intervention.