Arrhythmogenic Remodeling of the Left Ventricle in a Porcine Model of Repaired Tetralogy of Fallot

Abstract

Background: Ventricular arrhythmias are frequent in patients with repaired tetralogy of Fallot (rTOF), but their origin and underlying mechanisms remain unclear. In this study, the involvement of left ventricular (LV) electrical and structural remodeling was assessed in an animal model mimicking rTOF sequelae.

Methods: Piglets underwent a tetralogy of Fallot repair-like surgery (n=6) or were sham operated (Sham, n=5). Twenty-three weeks post-surgery, cardiac function was assessed in vivo by magnetic resonance imaging. Electrophysiological properties were characterized by optical mapping. LV fibrosis and connexin-43 localization were assessed on histological sections and protein expression assessed by Western Blot.

Results: Right ventricular dysfunction was evident, whereas LV function remained unaltered in rTOF pigs. Optical mapping showed longer action potential duration on the rTOF LV epicardium and endocardium. Epicardial conduction velocity was significantly reduced in the longitudinal direction in rTOF LVs but not in the transverse direction compared with Sham. An elevated collagen content was found in LV basal and apical sections from rTOF pigs. Moreover, a trend for connexin-43 lateralization with no change in protein expression was found in the LV of rTOFs. Finally, rTOF LVs had a lower threshold for arrhythmia induction using incremental pacing protocols.

Conclusions: We found an arrhythmogenic substrate with prolonged heterogeneous action potential duration and reduced conduction velocity in the LV of rTOF pigs. This remodeling precedes LV dysfunction and is likely to contribute to ventricular arrhythmias and sudden cardiac death in patients with rTOF.

Keywords: arrhythmias; disease model; heart ventricles; swine; tetralogy of Fallot.

Figure 1.

Action potential duration (APD) from Sham and repaired tetralogy of Fallot left ventricles (LVs). A, Representative optical action potentials from the LV endocardium and epicardium of Sham (black) and rTOF (grey) pigs. B, Endocardial and epicardial APD80 were prolonged in the anterior LV midwall of rTOF pigs (grey) compared with Sham (black) when paced at 1 Hz. C, LV effective refractory period tended to be longer on the rTOF epicardium than in Sham pigs. Data are means±SD. Sham, n=5; rTOF, n=6. *P<0.05.

Figure 2.

Dynamic action potential duration (APD) restitutions in Sham and repaired tetralogy of Fallot (rTOF) left ventricles (LVs). A, Mean APD80 at varying basic cycle lengths (BCL) on the LV endocardium and the epicardium. APD80 was prolonged on the epicardium and endocardium at different cycle lengths of rTOF (grey lines) and Sham (black lines) pigs. B, There was a trend for an increase in epicardial APD restitution slope in rTOF compared with Sham pigs. Data are means±SD. Sham, n=5; rTOF, n=6. *P<0.05, **P<0.01.

Figure 3.

Dispersion of repolarization in Sham and repaired tetralogy of Fallot (rTOF) left ventricles (LVs). A, Representative epicardial APD80 maps showing heterogeneous action potential duration (APD) distribution in rTOF LVs paced at 1 Hz. B, APD80 dispersion was increased in rTOF compared with Sham in the epicardium but not the endocardium (C). D, There was a trend for an increase in repolarization time dispersion in rTOF LV epicardium but not in the endocardium (E). Data are means±SD. Sham, n=5; rTOF, n=6. **P<0.01.

Figure 4.

Left ventricular (LV) activation and conduction velocity in Sham and repaired tetralogy of Fallot (rTOF) preparations. A, Representative epicardial activation maps (5 ms spaced isochrones) from Sham and rTOF LVs stimulated at 1 Hz. B, Longitudinal conduction velocity (CV_L) was significantly decreased in rTOF LVs compared with Sham, whereas transverse conduction velocity (CV_T) remained unchanged (C). D, This resulted in a trend for a decreased CV_L to CV_T ratio in rTOF animals. Data are means±SD. Sham, n=4; rTOF, n=4. *P<0.05.

Figure 5.

Dynamic conduction velocity restitution curves in Sham and repaired tetralogy of Fallot (rTOF) left ventricles (LVs). Epicardial restitution of conduction velocity measured at various cycle lengths (BCL) in the longitudinal (A) and transverse (B) directions in rTOF (grey lines) and Sham (black lines) LVs. Data are means±SD. Sham, n=4; rTOF, n=4. *P<0.05.

Figure 6.

Membrane localization and expression of Connexin-43 in Sham and repaired tetralogy of Fallot (rTOF) left ventricles (LVs). A, Immunohistochemistry for Cx43 (connexin-43) performed on 8 µm LV base tissue sections revealed Cx43 Lateralization (arrows) in rTOF LVs as opposed to Sham LVs in which Cx43 was localized at the intercalated discs. B, Cx43 expression was unchanged in rTOF LVs compared with Sham. Data are means±SD. Sham, n=3; rTOF, n=3.

Figure 7.

Collagen content in Sham and repaired tetralogy of Fallot (rTOF) left ventricles (LVs). A, Representative histological sections of Sham and rTOF LVs stained with Masson trichrome. In addition to diffuse fibrosis, large fibrotic regions are visible in the LV base and apex of rTOF pigs. B, Collagen content, expressed as a percentage of total section area, was increased in rTOF LV base and apex compared with Sham. C, A positive correlation was found between LV collagen content and right ventricular end-diastolic volume indexed to body surface area (RV-EDVi), and (D) a negative correlation was found between RV-EDVi and LV longitudinal conduction velocity (CV_L). Data are means±SD. Sham, n=4; rTOF, n=4. **P<0.01, ***P<0.001.