AbstractAtrial fibrillation (AF) is a degenerative cardiac arrhythmia characterised by high incidence rate and limited effectiveness of clinical treatments. Fibrosis is one of the major pathological factors linked with AF progression. However, mechanistic links between this pathology and AF arrhythmogenesis are incompletely understood. This project investigates the diverse arrhythmogenic effects of fibrosis using computational model of the human atria, which integrates electrophysiological and structural changes associated with fibrosis. Structural data is reconstructed based on magnetic resonance imaging (MRI) from AF patients.
The computational model of the 3D human atria integrated the Visible Female cardiac geometry, rule-based fibre orientation, region-specific atrial electrophysiology and its changes due to AF-induced ionic remodelling. Fibrosis was modelled by integrating (i) a novel electrophysiologically detailed model for a single atrial fibroblast, (ii) electrotonic myocyte-fibroblast (M-F) coupling, (iii) structural effects of fibrosis on the anisotropy of atrial tissue and (iv) either random or patient-specific distributions of fibrosis in the 3D atrial model. Patient-specific distributions were reconstructed from late-gadolinium enhanced (LGE) MRI.
At the single-cell level, electrotonic M-F coupling via gap junctions resulted in changes of the atrial myocyte electrophysiological properties, including the resting membrane potential, action potential duration (APD) and its restitution, regional APD heterogeneity, effective refractory period and cell excitation threshold. At the 3D atria level, these changes translated into the altered susceptibility for atrial re-entry. Additional changes of the tissue anisotropy and heterogeneity, both associated with non-uniform fibrosis in the atria, resulted in the breakdown of re-entry into multiple rotors and wavelets, which is characteristic of AF.
Application of LGE MRI enabled the segmentation of patchy fibrosis from AF patients. An image processing and modelling pipeline was developed to maps the distributions of fibrosis into the 3D atrial model and investigate the effects of regional fibrosis patches on the genesis of AF. Simulations the patient-specific model revealed pinning of re-entrant waves primarily at the border zone of dense fibrosis patches. Patient-specific atrial wall thickness was also reconstructed using novel PSIR MRI protocol and integrated into the 3D model. This non-uniform structural feature provided additional locations for the wave pinning in regions of minimum atrial wall thickness.
In summary, the developed MRI processing and computational modelling pipeline has been applied for dissecting the multiple effects of fibrosis in the genesis of AF. Novel insights provided by the image-based models pave a way for understanding of the disease and providing treatment on a patient-by-patient basis.
|Date of Award
|Kawal Rhode (Supervisor) & Oleg Aslanidi (Supervisor)