A web-based dynamic nomogram for predicting readmission in patients with heart failure with preserved ejection...

Area of Science:

• Cardiology
• Medical Informatics
• Predictive Analytics

Background:

• Heart failure with preserved ejection fraction (HFpEF) poses a significant rehospitalization burden.
• Accurate prediction of 1-year HF-related rehospitalization is crucial for patient management.
• Existing predictive models may not fully capture dynamic risk factors in HFpEF.

Purpose of the Study:

• To develop and validate a web-based dynamic nomogram for predicting 1-year HF-related rehospitalization in HFpEF patients.
• To assess the predictive efficacy and clinical utility of the developed nomogram.
• To identify key clinical variables associated with HF rehospitalization in this population.

Main Methods:

• A dynamic nomogram was constructed using Least Absolute Shrinkage and Selection Operator (LASSO) and multivariate logistic regression on training data from two centers.
• Risk factors identified included age, BMI, atrial fibrillation, triglyceride-glucose index, LVEF, E/e, and ACEI/ARB use.
• Model performance was evaluated using Area Under the Receiver Operating Characteristic Curve (AUC-ROC), calibration plots, and Decision Curve Analysis (DCA) on independent test sets.

Main Results:

• The nomogram incorporated seven significant predictors of 1-year HF rehospitalization in HFpEF.
• The model demonstrated good predictive ability with AUC-ROC values of 0.801 (training) and 0.773 (test) datasets.
• Calibration plots showed excellent agreement, and DCA confirmed high clinical effectiveness across a relevant threshold probability range (10%-80%).

Conclusions:

• The developed dynamic nomogram is an effective tool for predicting 1-year HF-related rehospitalization risk in HFpEF patients.
• The nomogram facilitates the identification of high-risk patient categories, enabling personalized risk stratification.
• This web-based tool has the potential to improve clinical decision-making and resource allocation in HFpEF management.