Role of Cardiac Microstructure Variability on Ventricular Arrhythmogenesis

Abstract

The propagation of cardiac electrical excitation is influenced by tissue microstructure. Reaction-diffusion computational models of cardiac electrophysiology incorporating both dynamic action potential (AP) behaviour and image-based myocardial architecture provide an approach to study the complex organisation of excitation waves within variable myocardial structures. The role of tissue microstructure (cardiomyocyte and sheetlet orientations) on organ-scale arrhythmic excitations was investigated. Five healthy rat ventricle datasets were obtained using diffusion tensor MRI (DTI). The Fenton-Karma minimal AP model was modified to reproduce the rat AP duration and restitution. Re-entrant scroll waves were initiated in the five anatomical models at ten locations for three microstructure scenarios: (i) isotropic; (ii) anisotropic; and (iii) orthotropic. Variability in anatomy and microstructure caused simulated scroll waves to self-terminate, remain tachycardia-like, or degenerate into fibrillatory activity. Whilst inclusion of DTI-based microstructure increased total scroll wave filament length to differing extents between the five hearts, overall mean filament dynamics were quantitatively similar under anisotropic and orthotropic conditions. This study highlights the important role of inter-subject structural variability.

Investigation of the Role of Myocyte Orientations in Cardiac Arrhythmia Using Image-Based Models

“…An assessment of arrhythmia dynamics arising from 10 different re-entry simulations (i.e., averaged over the 10 different re-entry initiation sites) revealed that, for these three DTI-based data sets and the idealized RB case, scroll wave filament dynamics were quantitatively similar under conditions of different myocyte orientations, and no statistically significant differences in measures of the filament dynamics between the different scenarios were found. This is consistent with a preliminary work, in which we also showed that measures of arrhythmia dynamics were not significantly different between similar-sized hearts, each with sample-specific myocyte orientations ( 24 ). Thus, in the limited case of investigating three hearts and an idealized case only, no clear intersubject variability in overall sustainability of arrhythmia was observed; such results, however, cannot be extrapolated to general statements about population variability nor to potential arrhythmia variability in diseased conditions, in which intersubject variation of myocyte orientations may be larger than in control conditions.…”

“…This leads to a significant asymmetry in the activation-recovery cycle and eventual development of re-entry for DTI1 but not DTI2. consistent with a preliminary work, in which we also showed that measures of arrhythmia dynamics were not significantly different between similar-sized hearts, each with sample-specific myocyte orientations (24). Thus, in the limited case of investigating three hearts and an idealized case only, no clear intersubject variability in overall sustainability of arrhythmia was observed; such results, however, cannot be extrapolated to general statements about population variability nor to potential arrhythmia variability in diseased conditions, in which intersubject variation of myocyte orientations may be larger than in control conditions.…”

“…As myocyte orientations in cardiac modeling studies are typically based on two values of the diffusion coefficient (longitudinal and transverse), axially symmetric anisotropy was assumed for all DTI-based myocyte orientation scenarios (i.e., based on the primary eigenvector alone). In a preliminary study, we showed that simulated arrhythmia dynamics under anisotropic and orthotropic conditions were quantitatively similar ( 24 ), supporting the appropriateness of this simplification. Histograms of the helix and transverse angles for the newly created DTI1, DTI2, and DTI3 myocyte orientation scenarios are shown in Fig.…”

“…Previously, five healthy rat hearts were perfusion and immersion fixed ex vivo using low osmolality Karnovsky’s fixative and then imaged using a diffusion-weighted fast spin echo sequence at a resolution of 100 μ m isotropic ( 23 ). Myocyte and sheetlet orientation angles were subsequently reconstructed from these data sets using methods described in detail elsewhere ( 23 , 24 ). Ventricular geometries were segmented out, and variability in ventricular myocyte orientations (helix, transverse, and sheetlet angles) were quantified, as shown in Fig.…”