Theoretical Models and Computational Analysis of Action Potential Dispersion for Cardiac Arrhythmia Risk Stratification

Abstract

Reentrant cardiac arrhythmias such as atrial fibrillation (AF) and ventricular fibrillation (VF) are common cardiac arrhythmias that account for substantial morbidity and mortality throughout the world. However, the mechanisms and optimal ablation treatment strategies for such arrhythmias are still unclear. Using 2D optical mapping of a mouse model with AF and VF, we have identified regional heterogeneity of the action potential duration (APD) in the atria and ventricles of the heart as key drivers for the initiation and persistence of reentry. The purpose of this paper is to discuss theoretical patterns of dispersion, demonstrate patterns of dispersion seen in our mouse model and discuss the computational analysis of APD dispersion patterns. These analyses and discussions may lead to better understanding of dispersion patterns in patients with these arrhythmias, as well as help comprehend whether and how reducing dispersion can lead to arrhythmia risk stratification and treatment strategies for arrhythmias.

Keywords: action potential dispersion; action potential duration; cardiac arrhythmia; heart; optical mapping of calcium and action potentials.

Figure 1

Theoretical models of APD dispersion proposed.

Figure 2

Representative optical mapping images in the atria of three TG mice (A–C) with AF. The column on the left shows the APD maps of the epicardial surface of the atria captured during optical mapping. The middle column displays the corresponding APD dispersion maps. The column on the right demonstrates the binary APD dispersion patterns obtained when applying a threshold for areas with APD differences >10 ms in the atria. The mice were mapped after hyperkalemic induced sinus conversion and 10 Hz pacing.

Figure 3

Representative optical mapping images in the ventricles of three TG mice (A–C) with VT/VF. The column on the left shows the APD maps of the epicardial surface of the ventricle captured upon optical mapping. The middle column displays the corresponding APD dispersion maps. The column on the right demonstrates the binary APD dispersion patterns obtained when applying a threshold for areas with APD differences >10 ms in the ventricle. The three mice with spontaneous and sustained VT/VF were captured in normal sinus rhythm.

Figure 4

Analysis of the APD maps showing the percentage of areas with dispersion > 10 ms over the total epicardial area of the left atria imaged by optical mapping, in TG mice with AF. One sample t-test p < 0.05, n = 4 control and 22 TG mice. The average area of TG atrial tissue with dispersion > 10 ms was 10.24 ± 1.15%, while the control mice group with no AF was significantly different (***p < 0.001, Student's t-test) and did not exhibit any dispersion.

Figure 5

Analysis of the APD maps showing the percentage of areas with dispersion > 10 ms over the total epicardial area of the ventricle imaged by optical mapping, in TG mice with spontaneous VT/VF. One sample t-test p < 0.05, n = 4 control and 5 TG mice. The average area of TG ventricular tissue with dispersion > 10 ms was 12.9 ± 1.4%, while the control mice that exhibited no VT/VF were significantly different (***p < 0.001, Student's t-test) and did not show any dispersion.