Variation of P-wave indices in paroxysmal atrial fibrillation patients before and after catheter ablation. (April 2021)

• Record Type: Journal Article
• Authors: Ortigosa, Nuria; Ayala, Guillermo; Cano, Óscar
• Is Part Of: Biomedical signal processing and control. Volume 66(2021)
• Publication Date: 2021-04
• Subjects: Electrocardiogram (ECG) -- Atrial fibrillation -- Catheter ablation -- Pulmonary vein isolation; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2021.102500
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: Effects of pulmonary vein isolation in the ECG are studied. Classification methods of atrial fibrillation recurrence using logistic regression with P-wave features as predictors are proposed. Early detection of AF recurrence can be done in a non-invasive way by ECG markers. Abstract: The effects of pulmonary vein isolation in the surface electrocardiogram of patients with paroxysmal atrial fibrillation are analyzed in this paper using non-invasive markers for early detection of the arrhythmia recurrences. Several features have been extracted on P-waves of V1 lead for paroxysmal atrial fibrillation patients who underwent catheter ablation of pulmonary veins for restoring sinus rhythm permanently. Surface ECG was simultaneously recorded along with intracardiac recordings starting from the beginning of the intervention until half an hour after the catheter ablation successfully ended. Significant difference between the means before and after catheter ablation have been observed for the cross-correlation index, kurtosis and amplitude dispersion of P-waves. A logistic regression has been applied to all the descriptors and pointed to the amplitude dispersion index, as well as the minimum gradient joint with kurtosis of P-waves prior to catheter ablation as good predictors of recurrence of atrial fibrillation (78% accuracy). It is important to note how using a few descriptors good classification results are achieved. This study opens a door to early detection of atrial fibrillation recurrences using markers obtained by non-invasive methods.