Spectral analysis of the ECG to guide optimal endpoint in catheter ablation of atrial fibrillation

Thanks to its high rates of success, catheter ablation (CA) is nowadays considered as the first-line therapy to treat atrial fibrillation (AF). However, a procedural consensus beyond pulmonary veins isolation (PVI) still lacks for patients with persistent AF. Nonetheless, the main goal of every CA procedure is to ablate the least amount of tissue able to render patients free from AF at long-term. Hence, the identification of an optimal CA endpoint is an interesting clinical challenge. For that purpose, the present work explores the ability of some spectral features obtained from the atrial activity (AA) reflected on the surface ECG. More precisely, the variations experienced during the procedure by the dominant atrial frequency (DAF), its 3 dB bandwidth and the median atrial frequency (MAF) are analyzed from lead V1 in 15 patients undergoing CA. The obtained results show a strong agreement between the DAF and MAF, which are able to anticipate with an accuracy between 70 and 80% the patients' rhythm after a follow-up of 3 months. As a consequence, monitoring AA spectral features could be helpful in quantifying the atrial substrate alteration provoked by CA procedures.