Accurate characterization of rotor activity during atrial fibrillation depends on the properties of the multi-electrode grid

Multi-electrode array systems are increasingly being used to study atrial electrical excitation in humans and are shedding new light on the mechanisms of atrial fibrillation (AF). However the mapping systems that are currently being used to characterize the rotors that are postulated to drive AF have not been systematically analyzed for accuracy. Computer simulations of chronic AF (cAF) were carried out using a realistic 3D model of the atria. Extracellular electrical potentials were calculated on intra-atrial electrode arrays placed at the driving rotor area. The array-to-endocardial wall distance (d_aew) was set at 0.2. 1, 2 and 5 mm and uniform interelectrode distances (d_ie) were set at 1, 2 or 5 mm. The instantaneous location of the rotor meandering were analyzed on interpolated phase movies based on the Hilbert transform. At d_ie of 1 mm and d_aew ≤2 mm, the rotor location on the array was not significantly diferent from that on the wall, although meandering on the array was greater than in the atrium. At d_ie = 2 mm, increasing the d_aew decreased the detected meandering area, but remained higher than in the atrium. Finally, when the spatial resolution was 5 mm, in most cases the array recognized the presence of the rotor, but not its trajectory accurately.