Introduction: Outcome of patients undergoing catheter ablation of atrial fibrillation (AF) varies widely. We sought to investigate whether parameters derived from the spectral analysis of surface ECG and intracardiac AF electrograms can predict outcome in patients referred for pulmonary vein isolation (PVI).
Methods: We performed spectral analysis on the surface ECG and intracardiac electrograms from patients referred for AF ablation. After filtering and QRST subtraction, we measured the dominant frequency (DF), regularity index (RI) and the organizational index (OI) of fibrillatory electrograms and determined their value for predicting AF recurrence after ablation. A subjective, blinded prediction based on the surface ECG was also performed.
Results: We analyzed data from 153 PVI procedures in 140 patients (67.1% with persistent or longstanding AF). In a multivariable model, DF in the right atrium (RA) and distal coronary sinus (CSd)-to-RA DF gradient predicted AF recurrence (OR, 3.52, P = 0.023 and OR, 0.2, P = 0.034, respectively). DF in RA and CSd to RA DF gradient had a good predictive value for PVI outcome (area under the curve [AUC] of 0.73, P = 0.007 and 0.74, P = 0.007, respectively). These performed better than the subjective predictions of experienced electrophysiologists ( P = 0.2).
Conclusions: Higher RA DF, lower CSd to RA DF gradient predicted recurrence after AF ablation. These spectral measures suggest a more remodeled atrial substrate and may provide simple tools for risk stratification or predict the need for additional substrate modification in patients referred for AF ablation.
Keywords: atrial fibrillation; catheter ablation; dominant frequency; organizational index; pulmonary vein isolation; signal processing.
© 2018 Wiley Periodicals, Inc.