Non-linear organization analysis of the dominant atrial frequency to predict spontaneous termination of atrial fibrillation

Abstract

The prediction of spontaneous atrial fibrillation (AF) termination or maintenance could avoid unnecessary therapy and contribute to take the appropriate decisions in the management of the arrhythmia. The aim of this work is to predict if an AF episode terminates spontaneously or not. The prediction was carried out making use of non-linear organization analysis applied to the surface ECG. Sample entropy was selected as organization index, proving that atrial activity (AA) organization increases prior to AF termination. Using the dominant atrial frequency organization analysis, that is the frequency-selected signal produced by the main reentry wandering the atrial tissue, 92% of the terminating and nonterminating analyzed AF episodes were correctly classified. Because noise and ventricular residues degrade AA organization estimation performance, the use of selective filtering to get the dominant atrial frequency was necessary. The obtained outcomes allow to conclude that the dominant atrial frequency, and therefore, the main atrial reentry, contains the most relevant information about spontaneous AF termination.