PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
June 27, 2026EP EuropaceOpen Access

Artificial intelligence-powered ECG is superior to NT-proBNP in predicting left ventricular systolic dysfunction in atrial fibrillation with rapid ventricular response

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Risk stratification for tachycardia-induced cardiomyopathy remains poorly understood, motivating evaluation of AI-ECG versus NT-proBNP in predicting systolic dysfunction during atrial fibrillation with rapid ventricular response.

Does AI-ECG improve the prediction of reduced LVEF (<35%) compared to NT-proBNP in patients with atrial fibrillation and rapid ventricular response?

Population

11,801 adult patients with atrial fibrillation and rapid ventricular response without known HFrEF

Comparison

AI-ECG vs NT-proBNP

Design

Retrospective study

Key result

AI-ECG was superior to NT-proBNP in predicting left ventricular ejection fraction <35% among patients with atrial fibrillation and rapid ventricular response (AUC 0.72 vs 0.59; accuracy 77% vs 47%).

Authors

OBO BaqalCYC YeeJQJ Quillen

Discussion

Loading...

Member takes

Overview

AI ECG may aid LVEF screening from routine ECGs; leaves open prospective validation before clinical adoption.

Key Points

  • The study aims to evaluate how well AI-powered ECG predicts left ventricular systolic dysfunction compared to NT-proBNP in patients with atrial fibrillation and rapid ventricular response.
  • Conducted a retrospective analysis of 11,801 adult patients with atrial fibrillation and rapid ventricular response.
  • Compared AI-ECG low EF probability results to NT-proBNP levels and echocardiogram findings.
  • Assessed performance using ROC curve analysis to determine predictive accuracy.
  • AI-ECG showed an AUC of 0.72 for predicting LVEF <35%, while NT-proBNP had an AUC of 0.59.
  • AI-ECG achieved a sensitivity of 66% and specificity of 78% at a probability cutoff of 17%.
  • Mean NT-proBNP level in available cases was 4090 pg/mL, with only 47% accuracy in predicting reduced LVEF.

Study Design

Type

Observational (n=11,801)

Multicenter

Yes

Structured PICO

Does AI-ECG improve the prediction of reduced LVEF (<35%) compared to NT-proBNP in patients with atrial fibrillation and rapid ventricular response?

P
Population
11,801 adult patients with atrial fibrillation and rapid ventricular response, without known heart failure with reduced ejection fraction, evaluated retrospectively.
E
Exposure
Artificial intelligence-powered 12-lead ECG (AI-ECG)
C
Comparator
NT-proBNP levels
O
Outcome
Prediction of reduced left ventricular ejection fraction (LVEF <35%)surrogate

Main Result

Absolute Event Rate: 77% vs 47%

AI-ECG is superior to NT-proBNP for predicting left ventricular systolic dysfunction in patients presenting with atrial fibrillation and rapid ventricular response, offering a rapid screening tool to guide management.

Cite This Study

Baqal et al. (2026) conducted an observational in Atrial fibrillation with rapid ventricular response (n=11,801). AI-ECG vs. NT-proBNP was evaluated on Predicting LVEF <35%. AI-ECG was superior to NT-proBNP in predicting left ventricular ejection fraction <35% among patients with atrial fibrillation and rapid ventricular response (AUC 0.72 vs 0.59; accuracy 77% vs 47%).

synapsesocial.com/papers/6a3f9785125782b61d865764https://doi.org/10.1093/europace/euag105.1203
View Full Paper
Ask AI
Bookmark
Share