Abstract
Introduction
Myocardial fibre orientation and heterogeneity of action potential (AP) characteristics across different atrial tissues can play an important role in the mechanisms of atrial fibrillation. However, few computational studies of the entire atria have accounted for these in detail.
Methods
A 3D canine atrial geometry was generated using contrast-enhanced 36-μm resolution microCT images, from which fibre orientation was also estimated using structure tensor analysis. The atria were segmented into the left atrium (LA), right atrium (RA), pulmonary vein (PV) sleeves and Bachmann's bundle (BB). Specific canine cell models based on recently published experimental data were created and assigned to each tissue. The monodomain equation was solved using a finite differences method in a Cartesian grid. Simulations in physiological conditions were performed by stimulating the sino-atrial node region at a cycle length (CL) of 1 s. To study arrhythmia initiation, the PV region was stimulated at 192 ms.
Results
The AP duration (APD90) of the developed cell models agreed well with published experimental data in the dog (RA: 184, LA: 170, BB: 235, PV: 135 ms, CL = 1 s). Tissue activation sequence and timings in physiological conditions also showed good agreement with experimental data with a total activation time of 76 ms and a mean conduction velocity of 96 cm/s. Fast pacing near the PVs gave rise to rapid re-entrant activity, due to the high local fibre anisotropy and the APD gradient between the PV sleeves and the surrounding LA tissue.
Conclusion
We developed a detailed 3D model of the canine atria with realistic anatomical and electrophysiological properties. This model can be used to further investigate the mechanisms underlying atrial arrhythmias, namely the characteristics of the rapid activation sources observed clinically near the PV sleeves during atrial fibrillation.
Myocardial fibre orientation and heterogeneity of action potential (AP) characteristics across different atrial tissues can play an important role in the mechanisms of atrial fibrillation. However, few computational studies of the entire atria have accounted for these in detail.
Methods
A 3D canine atrial geometry was generated using contrast-enhanced 36-μm resolution microCT images, from which fibre orientation was also estimated using structure tensor analysis. The atria were segmented into the left atrium (LA), right atrium (RA), pulmonary vein (PV) sleeves and Bachmann's bundle (BB). Specific canine cell models based on recently published experimental data were created and assigned to each tissue. The monodomain equation was solved using a finite differences method in a Cartesian grid. Simulations in physiological conditions were performed by stimulating the sino-atrial node region at a cycle length (CL) of 1 s. To study arrhythmia initiation, the PV region was stimulated at 192 ms.
Results
The AP duration (APD90) of the developed cell models agreed well with published experimental data in the dog (RA: 184, LA: 170, BB: 235, PV: 135 ms, CL = 1 s). Tissue activation sequence and timings in physiological conditions also showed good agreement with experimental data with a total activation time of 76 ms and a mean conduction velocity of 96 cm/s. Fast pacing near the PVs gave rise to rapid re-entrant activity, due to the high local fibre anisotropy and the APD gradient between the PV sleeves and the surrounding LA tissue.
Conclusion
We developed a detailed 3D model of the canine atria with realistic anatomical and electrophysiological properties. This model can be used to further investigate the mechanisms underlying atrial arrhythmias, namely the characteristics of the rapid activation sources observed clinically near the PV sleeves during atrial fibrillation.
Original language | English |
---|---|
Pages (from-to) | e10-e11 |
Number of pages | 2 |
Journal | Journal of Electrocardiology |
Volume | 46 |
Issue number | 4 |
DOIs | |
Publication status | Published - Jul 2013 |