Study of unipolar electrogram morphology in a computer model of atrial fibrillation.

Electrograms exhibit a wide variety of morphologies during atrial fibrillation (AF). The basis of these time courses, however, is not completely understood. In this study, data from computer models were studied to relate features of the signals to the underlying dynamics and tissue substrate. A computer model of entire human atria with a gross fiber architecture based on histology and membrane kinetics based on the Courtemanche et al. atrial model was developed to simulate paced activation and simulated AF. Unipolar electrograms were computed using a current source approximation at 256 sites in right atrium, to simulate a mapping array. The results show the following: (1) In a homogeneous and isotropic tissue, the presence of highly asymmetric electrograms is rare (<2%), although there is a marked variability in amplitude and symmetry. (2) The introduction of anisotropy increases this variability in symmetry and amplitude of the, electrograms especially for propagation across fibers. The percentage of highly asymmetric electrograms increases to 12% to 15% for anisotropy ratios greater than 3:1. (3) Multiphasic and fractionated electrograms are rarely seen in the model with uniform properties but are more common (15%-17%) in a model including regions with abrupt changes in conductivity. Beat-to-beat variations in the occurrence of multiphasic signals are possible with fixed anatomic heterogeneity, due to beat-to-beat variations in the direction of the wavefront relative to the heterogeneity. Analysis of the amplitude and symmetry of unipolar atrial electrograms can provide information about the electrophysiologic substrate maintaining AF.