Effect of Atrial Fibrillation Load and Medication on the Beat-to-Beat P-wave Morphology Variability during Sinus Rhythm

Dimitris Filos1, Dimitris Tachmatzidis2, Vassilios Vassilikos2, Ioanna Chouvarda1
1Lab of Computing, Medical Informatics and Biomedical Imaging Technologies, School of Medicine, Aristotle University of Thessaloniki, 23rd Cardiology Department, Aristotle University of Thessaloniki


Atrial Fibrillation (AF) is the most common cardiac arrhythmia while its progressive nature is associated with gradual modifications of the electrical properties of the atria and the presence of structural remodeling in the later and more permanent stages. The P-wave in the surface Electrocardiogram (ECG) reflects the atrial activation while the modification of the atrial pathophysiological properties leads to P-wave morphology (PWM) alternations. In paroxysmal AF (pAF), the modifications of the PWM may not appear permanently, whereas they may be present randomly in the ECG signal. The analysis of the P-waves, during sinus rhythm, in a beat-to-beat basis, has revealed the existence of at least two PWM whereas the wavelet characteristics of the P-wave matching the main morphology can distinguish accurately the patients with pAF from healthy volunteers. In this work, we examine the hypothesis that there is an effect on beat-to-beat PWM alternations of a) the AF load, computed as the product between the frequency of AF episodes during the last moths and their duration, and b) anti-arrhythmic medication. ECG signals of high frequency (1000Hz), in the three orthogonal leads, were collected for 77 pAF patients of high and low load, 41 of them receiving antiarrhythmic medication treatment, and from 58 healthy volunteers. Kruskal-Wallis test was performed, and the preliminary results denote the existence of statistically significant differences between the groups suggesting that the PWM and its temporal variability reflect both the effect of AF load, as a result of the remodeled substrate, and the effect of medication. A multiclass Support Vector Machine classifier was trained, using the forward wrapper approach, resulting a high overall classification accuracy (≈90%). This analysis is a step towards the improvement of our understanding on the effect of the underline electrical remodeling extent and that of the medication on the variability of P-wave.