Electrogram analysis
By using fast Fourier transform analysis, Pachon et al (63) definedcompact  andfibrillar atrial potentials during sinus rhythm. Because fibrillar potentials demonstrated fragmented and heterogeneous conduction properties, authors speculated that fibrillar potentials may be originated from incursions of neural and vascular structures and be used to detect autonomic innervation sites. Lellouche et al (64) analyzed electrogram characteristics based on VRs during radiofrequency application and demonstrated that the best single predictor of VR during radiofrequency application was the presence of at least 4 electrogram deflections at the ablation site. In our initial work, epicardiac ganglia sites were detected through a combination of fast Fourier transform analysis of electrograms, and HFS (65). In accordance with Lellouche’s observations, all the electrograms on the radiofrequency ablation sites demonstrated a fragmented pattern. Because the better resolution afforded by usage of higher high-pass filters allowed better appreciation of electrogram fragmentation (63), in our subsequent work, we used band-pass filter settings of 200-500Hz instead of conventional band-pass filter settings of 30-500Hz during sinus rhythm and targeted all fragmented electrograms in regions which are anatomically consistent with autonomic innervation sites (46) (Table 1). Indeed, this simpler electroanatomical mapping-guided strategy demonstrated an identical success rate in preventing prodromal symptoms and syncope recurrence as compared to the previous combined approach. Figure 5 demonstratesthe anatomical distribution of GPs according to our definition method (49).