Related Experiment Video
Updated: Jan 12, 2026
08:10
Estimating Bilateral Atrial Function by Cardiovascular Magnetic Resonance Feature Tracking in Patients with Paroxysmal Atrial Fibrillation
Published on: July 20, 2022
2.0K
Discriminative Accuracy of CHA2DS2-VASc Score, and Development of Predictive Accuracy Model Using Machine Learning
Waqas Ullah1, Abhinav Nair2, Eric Warner3
1Department of Interventional Cardiology, University of Massachusetts, Worcester, MA, USA.
Cardiology Research
|October 31, 2025
Summary
The CHA2DS2-VASc score underestimates stroke risk in cardiac amyloidosis with atrial fibrillation (CA-AF). A new E-CHADS score, incorporating ESRD, dementia, and cancer, shows improved prediction accuracy for ischemic stroke in CA-AF patients.
Area of Science:
- Cardiology
- Neurology
- Geriatrics
Background:
- Cardiac amyloidosis (CA) with atrial fibrillation (AF) poses a significant risk for ischemic stroke.
- The current CHA2DS2-VASc score is insufficient for accurately predicting stroke risk in CA-AF patients.
Purpose of the Study:
- To evaluate the predictive accuracy of the CHA2DS2-VASc score for ischemic stroke in CA-AF patients.
- To develop and validate a novel, more accurate stroke risk prediction model for CA-AF patients.
Main Methods:
- Utilized the National Readmission Database (NRD) to compare outcomes between CA-AF and non-CA-AF patients.
- Employed multivariate regression and the AutoScore machine learning framework to develop a new stroke risk model.
- Assessed model performance using receiver operating characteristic analysis and area under the curve (AUC).
Main Results:
- The CHA2DS2-VASc score demonstrated poor discriminative accuracy for 30-day stroke risk in CA-AF (AUC 49%).
- The novel E-CHADS score (ESRD, CHF, HTN, cancer, dementia, DM) achieved excellent predictive ability for 30-day ischemic stroke risk in CA-AF (AUC 80%).
- CA-AF patients exhibited significantly higher adjusted odds of mortality, stroke, and hemorrhage compared to non-CA-AF patients.
Conclusions:
- The CHA2DS2-VASc score is inadequate for predicting ischemic stroke in patients with CA and AF.
- The proposed E-CHADS score, incorporating ESRD, dementia, and cancer, offers superior discriminative accuracy for ischemic stroke risk in this population.
