Nonlinear synchronization assessment between atrial and ventricular activations series from the surface ECG in atrial fibrillation

Abstract

Atrial fibrillation (AF) is the most common form of arrhythmia encountered in clinical practice. Its presence causes a rapid and irregular ventricular response, being the topic of intensive research in rate control therapies of AF. To this respect, recent studies suggest that ventricular response is notably influenced by atrial activity (AA) temporal organization. However, the interdependency between atrial and ventricular activations has not been adequately explored to date in real-life AF patients. The present work introduces a novel methodology to quantitatively assess synchronization and coupling between real atrial and ventricular activation series. Furthermore, the method operates on surface ECG recordings, thus providing an easy and cost-effective way to be applied. The method is based on a nonlinear index, such as cross-sample entropy (CSE), which estimates the conditional probability to find similar patterns within both activation series. The study has been carried out on patients with paroxysmal and persistent AF in order to be applied over atrial activation series with different properties in their organization. Results showed a statistically significant positive correlation between AA organization and the synchronization between atrial and ventricular activations (R = 0.53, p < 0.01). Furthermore, higher CSE values were observed for persistent (0.759 ± 0.053) than for paroxysmal AF episodes (0.662 ± 0.091), thus suggesting more synchronization between atrial and ventricular activations in paroxysmal AF. As a consequence, CSE provided findings consistent with previous works and could be used to reveal clinically useful information in the improvement of current rate control therapies, which are mainly focused on controlling ventricular rate without paying much attention to the atrial fibrillatory process.