Risk Factors and Construction of Risk Prediction Model for Postoperative Chronic Heart Failure in Patients Undergoing Heart Valve Replacement

CAO Dandan; WANG Lei; MENG Shuping

doi:10.3969/j.issn.1004-437X.2026.06.008

Abstract：Objective To explore the risk factors for chronic heart failure after heart valve replacement surgery and construct a risk prediction model.Methods Data of 273 patients with heart valve disease who underwent heart valve replacement surgery at Fuwai Central China Cardiovascular Hospital from December 2021 to February 2024 were collected.According to whether chronic heart failure occurred during the 6-month postoperative follow-up,they were divided into the occurrence group(51 cases)and the non-occurrence group(222 cases).Logistic regression analysis was conducted to identify the risk factors for the occurrence of chronic heart failure,and a risk prediction model was constructed based on these risk factors.Results Univariate and logistic regression analysis showed that preoperative pulmonary hypertension,preoperative levels of tumor necrosis factor-α(TNF-α)and homocysteine(Hcy),intraoperative blood loss,and postoperative use of adrenaline may be risk factors for chronic heart failure after heart valve replacement surgery(P＜0.05).Column chart prediction model was constructed based on risk factors,with an ideal fit between the calibration curve and the ideal curve within the model.The receiver operating characteristic(ROC)curve was plotted.The area under the curve(AUC)of the model for predicting chronic heart failure after heart valve replacement was 0.829,which had a high predictive value.Conclusion The occurrence of chronic heart failure after heart valve replacement surgery may be related to preoperative pulmonary hypertension,preoperative TNF-α and Hcy levels,intraoperative blood loss,and postoperative use of adrenaline.Based on this,a column chart model was constructed to evaluate the accuracy.