Background: The classification of atrial fibrillation (AF) currently is based on clinical characteristics, however, classifying underlying electropathology would assist in selecting appropriate therapy. To find differences in electropathology between paroxysmal and persistent AF patients, we measured atrial conduction parameters in sinus rhythm (SR) using an intra-operative high resolution epicardial mapping approach in persistent and paroxysmal AF patients.
Objective: To study whether the clinical classification is related to heterogeneity in conduction.
Methods: We included 47 paroxysmal (PAF) and 24 persistent AF (persAF) patients, who were scheduled for valvular and/or coronary artery bypass surgery. Five seconds of SR were recorded intra-operatively at the right atrium (RA), Bachmann’s Bundle (BB), the left atrium (LA) and the pulmonary vein area (PV). Various conduction parameters were calculated, including number of continuous conduction block and delay (cCBCD) lines, total activation time (TAT), and orientation of conduction block (CB) lines at BB. Additional to the comparison between PAF and persAF patients, the comparison between patients with and without a history of spontaneous termination of AF episodes was made.

Results: Biatrial activation times as well as the TAT of the LA and BB were significantly longer in the persAF group, p = 0.034; p= 0.010; p = 0.017, respectively. When adjusting for confounders, only TAT of BB was significantly longer in persAF patients, which was caused by more CB line parts orientated perpendicular to the conduction direction and a higher number of cCBCD lines. Other conduction characteristics at BB and other atrial areas showed no difference between PAF and persAF patients. Patients without spontaneous termination of AF episodes had a longer TAT of BB than patients with spontaneous termination (p = 0.033).
Conclusion: Patients with persAF show a longer activation time at Bachmann’s Bundle compared to patients with PAF, even when correcting for confounders such as electrical remodeling. Other conduction parameters show significant overlap between these two groups at BB and in other atrial areas, leading to the conclusion that the clinical classification does not match the arrhythmogenic substrate.