Are Atrial Fibrillation Dominant Frequency Areas the Source of Dominant Excitation Patterns? a Left Atrial Panoramic View

Frederique Vanheusden1, Gavin Chu2, Xin Li2, Joao L Salinet3, Peter Stafford2, G. André Ng2, Fernando S. Schlindwein4
1Nottingham Trent University, 2Department of Cardiovascular Sciences, University of Leicester, 3Federal University of ABC, 4Department of Engineering, University of Leicester


Introduction: Atrial areas of dominant frequency (DF) have been considered sources of atrial fibrillation, and therefore targets for guiding catheter ablation. However, DF areas on their own do not provide information about the propagation of AF patterns, which would enhance understanding of the complex behaviour of AF. Causality analysis has previously shown that DF areas drive dominant AF excitation patterns. These studies used contact mapping, which only provides information on local DF behaviour. The aim of this study is to extrapolate the correlation between DF and dominant excitation patterns over the entire atrial surface. Methods: Left atrial virtual electrograms (VEGM, 64 channels) were collected from 10 subjects with persistent AF undergoing first-time catheter ablation. VEGMs were collected for at least 5 minutes. After QRST subtraction, DF analysis was performed on 4-second windows with 50% overlap between successive windows. For each VEGM, the DF was defined as the frequency with highest power between 4 and 10 Hz. The highest DF (HDF) for each window was defined as the highest DF frequency over all VEGMs. Causality at this HDF was analysed using the direct transfer function (DTF). High causality recurrence locations and causality propagation direction were overlapped with identified HDF locations to determine if both algorithms identified similar source regions. Results and conclusion: Regions with high causality recurrence overlapped with HDF regions in only 19.3% (±8.0%) of windows over all patients. Regions of high causality recurrence generally encompassed smaller regions than the HDF region (73.2% ± 16.3% of windows) The main causality propagation direction showed complex behaviour, with direction being towards the HDF site in 31.7% (±8.3%) of windows. These results suggest that causality indices might identify alternative areas of interest for guiding catheter ablation, and that a combined frequency and causality analysis might improve AF source identification.