The effect of random cell decoupling on electrogram morphology near the percolation threshold in microstructural models of cardiac tissue

Random clusters of decoupled cells caused by cell death in diseased cardiac tissue can increase structural heterogeneity and give rise to abnormal patterns of propagation. The objective of this study was to use a 2-D 0.6 cm × 0.6 cm randomly generated model of a ventricular monolayer (LRd membrane model) (dx=dy=10 μm) to study the relationships between the underlying cardiac microstructure, the pattern of wavefront activation, and electrogram morphology as the substrate transitions from well-connected to disconnected. A fraction of individual cells (~100μm ×30 μm) between 0 and 0.50 were randomly uncoupled from neighboring cells. Simulated unipolar electrograms were computed at 121 distinct sites at the center of the tissue during longitudinal and transverse propagation. Compared to propagation in the longitudinal direction, conduction block in the transverse direction occurred at a lower percolation threshold (0.44 vs. 0.50) and was associated with a greater fraction of negative asymmetric electrograms. Directional differences in the percolation threshold and electrogram asymmetry were found to be correlated with slow, delayed propagation through isthmuses between decoupled cells. Analysis of electrogram morphology may be useful for identifying regions of tissue that may be vulnerable to long conduction delays and wavefront reentry.