On the Pacing Frequency Dependence of Cardiac Excitation Conduction via Fibroblasts

Introduction. While myocardial fibrosis is widely recognised as a primary contributor to arrhythmogenesis, underlying mechanisms that extend beyond formation of conduction barriers remain unclear. Non-myocytes, such as fibroblasts, not only contribute to extracellular matrix accumulation, but can also facilitate electrotonic cross-talk between cardiomyocytes in regions of fibrosis. In prior experiments, we combined advanced imaging techniques to quantify and correlate macro-scale cardiac electrophysiology with 3D micro-scale structural reconstructions of whole ventricles in an arrhythmogenic cardiomyopathy mouse model. We observed that often, conduction through fibrotic tissue is present at low pacing rates, yet it fails at high stimulation frequencies, promoting re-entrant arrhythmias. Computational reconstruction of cardiac structure and in silico experiments were conducted to elucidate this phenomenon. Here, we study the frequency dependence of the observed trans-lesional propagation of excitation with an in silico model.

Methods. We modelled a 1D chain of myocytes and fibroblasts because state-of-the-art bidomain models homogenise processes occurring on a cell-to-cell scale. The chain consisted of ten cardiomyocytes, the first five of which were stimulated at different pacing frequencies, followed by varying numbers of fibroblasts, and a final downstream myocyte to assess whether excitation was conducted and captured successfully. We used the Bondarenko and Sachse mouse models for the myocytes and fibroblasts, respectively.

Results and Discussion. As expected for a passively conducting cable between the proximal and distal cardiomyocytes, downstream myocyte excitation failed when too many fibroblasts were introduced; this conduction failure occurred earlier (i.e. at lower fibroblast numbers) if the pacing rate was elevated, see Fig. (1a). One mechanism contributing to this frequency dependence is a reduction in conduction velocity when the pacing frequency is raised (Fig. 1b; for an example see Fig. 2). In follow-up studies, we will additionally include lateral fibroblast coupling to cardiomyocytes to model the scar border zone, experimentally shown to decelerate the approaching wave, and we will extend our models to 2D to connect the observed phenomena with experimentally identified contributions to arrhythmogenesis.