Use of frequency analysis on the ECG for the prognosis of low energy cardioversion treatment of atrial fibrillation

Electric cardioversion is the most effective therapy for restoring sinus rhythm in patient with atrial fibrillation (AF), however, there is not a guiding criteria for advising on when and in whom it will be successful. The objective of this study was to employ frequency analysis on the surface electrocardiogram (ECG) to predict the outcome of low energy internal cardioversion in patients with AF. Thirty nine patients with AF, for elective DC cardioversion were included in this study. One catheter was positioned in the right atrial appendage and another in the coronary sinus. A voltage step-up protocol (50-300 V) was used for patient cardioversion. Prior to shock delivery, residual atrial activity signal (RAAS) was derived from 60 seconds of surface ECG from defibrillator pads, by bandpass filtering and ventricular activity (QRST) cancellation. Dominant atrial fibrillatory frequency (DAFF) was estimated from the RAAS power spectrum as the dominant frequency within the 3-12 Hz band. DAFF was calculated from whole 60 seconds segment (DAFF_L) and from the finals 10 seconds segment (DAFF_S) of the RAAS. Lower DAFF_L and DAFF_S were found in successfully cardioverted patients than in those nonsuccessful ones, with energy les3 and les6 joules. Therapy result (employing 3J or less) was predicted in 35/39 (89.7%) patients with DAFF_L=5.40 Hz, and DAFF_L was ges5.75 Hz in a 100% of noncardioverted patients. In conclusion, frequency analysis of the RAAS could be useful for predicting success of low energy internal cardioversion of patients with atrial fibrillation.