AIMS: To verify whether spectral components of atrial electrograms (AE) during sinus rhythm (SR) correlate with cardiac ganglionated plexus (GP) sites. METHODS AND RESULTS: Thirteen patients undergoing atrial fibrillation (AF) ablation were prospectively enrolled. Prior to radio frequency application, endocardial AE were recorded with a sequential point-by-point approach. Electrical stimuli were delivered at 20 Hz, amplitude 100 V, and pulse width of 4 ms. A vagal response was defined as a high-frequency stimulation (HFS) evoked atrioventricular block or a prolongation of RR interval. Spectral analysis was performed on single AE during SR, sampling rate of 1000 Hz, Hanning window. Overall, 1488 SR electrograms were analysed from 186 different left atrium sites, 129 of them corresponding to negative vagal response sites, and 57 to positive response sites. The electrogram duration and the number of deflections were similar in positive and negative response sites. Spectral power density of sites with vagal response was lower between 26 and 83 Hz and higher between 107 and 200 Hz compared with negative response sites. The area between 120 and 170 Hz normalized to the total spectrum area was tested as a diagnostic parameter. Receiver operating characteristic curve analysis demonstrated that an area 120-170/area(total) value >0.14 identified vagal sites with 70.9% sensitivity and 72.1% specificity. CONCLUSION: Spectral analysis of AE during SR in sites that correspond to the anatomical location of the GP is feasible and may be a simpler method of mapping the cardiac autonomic nervous system, compared with the HFS technique.
AIMS: To verify whether spectral components of atrial electrograms (AE) during sinus rhythm (SR) correlate with cardiac ganglionated plexus (GP) sites. METHODS AND RESULTS: Thirteen patients undergoing atrial fibrillation (AF) ablation were prospectively enrolled. Prior to radio frequency application, endocardial AE were recorded with a sequential point-by-point approach. Electrical stimuli were delivered at 20 Hz, amplitude 100 V, and pulse width of 4 ms. A vagal response was defined as a high-frequency stimulation (HFS) evoked atrioventricular block or a prolongation of RR interval. Spectral analysis was performed on single AE during SR, sampling rate of 1000 Hz, Hanning window. Overall, 1488 SR electrograms were analysed from 186 different left atrium sites, 129 of them corresponding to negative vagal response sites, and 57 to positive response sites. The electrogram duration and the number of deflections were similar in positive and negative response sites. Spectral power density of sites with vagal response was lower between 26 and 83 Hz and higher between 107 and 200 Hz compared with negative response sites. The area between 120 and 170 Hz normalized to the total spectrum area was tested as a diagnostic parameter. Receiver operating characteristic curve analysis demonstrated that an area 120-170/area(total) value >0.14 identified vagal sites with 70.9% sensitivity and 72.1% specificity. CONCLUSION: Spectral analysis of AE during SR in sites that correspond to the anatomical location of the GP is feasible and may be a simpler method of mapping the cardiac autonomic nervous system, compared with the HFS technique.