Development and Validation of a Predictive Model for Heart Failure in the People with Hypertension
10.11969/j.issn.1673-548X.2025.04.015
- VernacularTitle:高血压合并心力衰竭预测模型的构建及验证
- Author:
Mengyuan WU
1
;
Zichen DU
;
Jiao LI
Author Information
1. 301617 天津中医药大学研究生院
- Publication Type:Journal Article
- Keywords:
Hypertension;
Heart failure;
Predictive model;
Nomogram
- From:
Journal of Medical Research
2025;54(4):77-83
- CountryChina
- Language:Chinese
-
Abstract:
Objective To establish and validate a risk model for predicting heart failure(HF)in patients with hypertension.Methods The clinical data of 1500 patients with hypertension in Department of Cardiology of Tianjin Union Medical Center from January 2019 to December 2021 were retrospectively collected.The dataset was randomly partitioned into a training set(n=1050)and a valida-tion set(n=450)at a ratio of 7∶3.Then,the patients of training set were segmented into heart failure group(n=144)or non-heart failure group(n=906).The univariate Logistic regression,least absolute shrinkage and selection operator(LASSO)regression model and multivariate Logistic regression analysis were used for screening the risk factors and developing model.The R software was used to con-struct a nomogram.The discrimination ability of the model was determined by the area under the receiver operating characteristic curve(AUC),the calibration degree was evaluated by the calibration plot and Hosmer-Lemeshow goodness-of-fit test.Decision curve analy-sis(DCA)was performed to assess the clinical utility.Results This study showed that age,coronary heart disease,uric acid(UA),red blood cell distribution width(RDW),blood urea nitrogen/albumin ratio(BAR)and neutrophil to lymphocyte ratio(NLR)are inde-pendent predictors of the development of HF in patients with hypertension.A predictive model was constructed using the above six predic-tive factors.In the training set,the model had an AUC value of 0.864(95％CI;0.831-0.896),Hosmer-Lemeshow goodness of fit test showedx2=10.29,P=0.25.In the validation set,the AUC value was 0.842(95％CI:0.794-0.891),Hosmer-Lemeshow goodness of fit test showedx2=10.48,P=0.23.The clinical decision analysis also showed that the nomogram model had a better clinical perform-ance and could bring net clinical benefits to the patients.Conclusion This study successfully constructs a predictive model for the risk of HF in adults with hypertension and it can effectively identify patients at high risk of HF and formulate targeted intervention measures,thereby providing a basis for performing early prevention and treatment and improving prognosis.
