Construction and validation of a risk prediction model for postoperative severe bradyarrhythmia in pa-tients with valvular heart disease
10.3969/j.issn.1008-0074.2025.04.23
- VernacularTitle:心脏瓣膜疾病患者术后并发严重缓慢型心律失常风险预测模型的构建与验证
- Author:
Xiao-xiao LI
1
;
Qing LING
1
Author Information
1. 南京医科大学附属南京医院(南京市第一医院)心胸血管外科,江苏南京 210000
- Publication Type:Journal Article
- Keywords:
Heart valve disease;
Arrhythmias,cardiac;
Risk factors
- From:
Chinese Journal of cardiovascular Rehabilitation Medicine
2025;34(4):559-565
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct a risk prediction model of postoperative severe bradyarrhythmia for patients with valvular heart disease and verify the predictive efficacy of the model.Methods:A total of 636 patients undergoing heart valve surgery in Nanjing First Hospital between June 2023 and March 2024 were retrospectively selected.Ac-cording to the need for permanent pacemaker implantation after the surgery,the patients were divided into severe bradyarrhythmia group(n=80)and non-severe bradyarrhythmia group(n=556),then univariate and multivariate Logistic regression analysis were conducted to analyze the influencing factors of postoperative severe bradyarrhyth-mia in patients with valvular heart disease.R software was used to draw a nomogram prediction model for postoper-ative severe bradyarrhythmia in patients with valvular heart disease,and to verify the predictive effect of the model.Results:The incidence of postoperative severe bradyarrhythmia in patients with valvular heart disease was 12.6％in this study.Multivariate Logistic regression analysis showed that BMI＞28kg/m2(OR=2.094,95％CI 1.047～4.189,P=0.037),preoperative conduction block(OR=6.507,95％CI 2.909～14.552,P＜0.001),tricuspid valve surgery(OR=1.831,95％CI 1.094～3.063,P=0.021)and low cardiac output after surgery(OR=3.186,95％CI 1.310～7.751,P=0.011)were independent risk factors of postoperative severe bradyarrhythmia in these patients,while NYHA class Ⅲ～Ⅳ was its independent protective factor(OR=0.328,95％CI 0.194～0.555,P＜0.001).The nomogram model was constructed based on the influencing factors.H-L test in model group showed that x2=3.667,P=0.886,AUC under ROC curve was 0.744(95％CI 0.685～0.803),H-L test in validation group showed that x2=9.314,P=0.316,AUC under ROC was 0.718(95％CI 0.618～0.818),and the accuracy was 72.0％.Conclusion:The independent predictors for postoperative severe bradyarrhythmia in patients with val-vular heart disease include BMI,preoperative conduction block,NYHA class,tricuspid valve surgery and low cardi-ac output after surgery.A prediction model based on the above-mentioned risk factors may have good performance in predicting postoperative severe bradyarrhythmia in this population.
