Wavelet transform predicting success electrical cardioversion

Abstract

Atrial fibrillation (AF) is the most common arrhythmia in clinical practice. It is a significant source of morbidity, prolongs hospital stay and increases the costs of the treatment. The recurrence rate of AF after effective cardioversion is high. The study aimed to identify predictors for successful electrical cardioversion and maintenance of sinus rhythm after a first electrical cardioversion in patients with persistent AF. In the study were analysed episodes in ECG registers AF patients before cardioversion. At 4 weeks after cardioversion, only 15 (41,6%) of 36 patients maintained sinus rhythm. The response, method estimated the predictive power with respect to defibrillation outcome of main peak frequency and other parameters extracted from different coefficients obtained with Continuous Wavelet Transform applied over only one lead attempt to convert AF to sinus rhythm with a higher acute (VI) of the considered ECG register. Receiver operating characteristic (ROC) curve was used to evaluate the usefulness of these parameters in diagnosing recurrence of AF versus non recurrence of this arrhythmia. For the diagnosis of recurrence of AF the area under the ROC curve was 0.874 (95% CI 0.76-0.991). This technique can be used as an important independent predictor for maintenance of sinus rhythm after electrical cardioversion in patients with persistent AF. It complements existing clinical parameters with other extracted from a signal originating in only one lead of surface ECG, and contributes to a better interpretation of AF recurrence risk.