Automatically optimized vectorcardiographic features are associated with recurrence of atrial fibrillation after electrical cardioversion

Automatically optimized vectorcardiographic features are associated with recurrence of atrial fibrillation after electrical cardioversion

Abstract Electrical cardioversion presents one of the treatment options for atrial fibrillation (AF). However, the early recurrence rate is high, reaching ~40% three months after the procedure. Features based on vectorcardiographic signals were explored to find association with early recurrence of A...

Full description

Saved in:
Bibliographic Details
Main Authors: F. Plesinger, S. Hassouna, Z. Carna, J. Vesela, R. Smisek, E. Vargova, V. Sobota, Z. Koscova, P. Nejedly, I. Viscor, F. W. Prinzen, P. Jurak, J. Halamek, P. Osmancik
Format: Article
Language:English
Published: Nature Portfolio 2025-01-01
Series:Scientific Reports
Subjects:
ECG
VCG
Cardioversion
Signal processing
Atrial fibrillation
Online Access:https://doi.org/10.1038/s41598-025-85340-4
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Electrical cardioversion presents one of the treatment options for atrial fibrillation (AF). However, the early recurrence rate is high, reaching ~40% three months after the procedure. Features based on vectorcardiographic signals were explored to find association with early recurrence of AF. Eighty-four patients with non-paroxysmal AF referred to electrical cardioversion were prospectively studied; early AF recurrence was present in 40 (47.6%). Patients underwent 24-h Holter ECG monitoring three months after the procedure to assess AF recurrence. Pre-procedural 12-lead ECGs (10 s, 1 kHz) were recorded and automatically analyzed. We explored associations of VCG-based features with early AF recurrence. Two features were strongly associated with AF recurrence: (1) a mean VCG (y-axis) signal slope in a window starting 145 ms before QRS center, lasting for 190 ms (AUC 0.778, p < 0.001), and (2) a mean VCG (z-axis) signal slope in a window starting 60 ms after QRS center, lasting for 465 ms (AUC 0.744, p < 0.001). These features showed higher association to the outcome than eighteen baseline clinical features. Our approach revealed features based on a slope of vectorcardiographic signals. This work also suggests that state of ventricles strongly affects the AF recurrence after electrical cardioversion.
ISSN:2045-2322

Similar Items