Pacing in proximity to scar during cardiac resynchronization therapy increases local dispersion of repolarization and susceptibility to ventricular arrhythmogenesis

Abstract

Background: Cardiac resynchronization therapy (CRT) increases the risk of ventricular tachycardia (VT) in patients with ischemic cardiomyopathy (ICM) when the left ventricular (LV) epicardial lead is implanted in proximity to scar.

Objective: The purpose of this study was to determine the mechanisms underpinning this risk by investigating the effects of pacing on local electrophysiology (EP) in relation to scar that provides a substrate for VT in ICM patients undergoing CRT.

Methods: Imaging data from ICM patients (n = 24) undergoing CRT were used to create patient-specific LV anatomic computational models including scar morphology. Simulations of LV epicardial pacing at 0.2-4.5 cm from the scar were performed using EP models of chronic infarct and heart failure (HF). Dispersion of repolarization and the vulnerable window were computed as surrogates for VT risk.

Results: Simulations predict that pacing in proximity to scar (0.2 cm) compared to more distant pacing to a scar (4.5 cm) significantly (P <.01) increased dispersion of repolarization in the vicinity of the scar and widened (P <.01) the vulnerable window, increasing the likelihood of unidirectional block. Moreover, slow conduction during HF further increased dispersion (∼194%). Analysis of variance and post hoc tests show significantly (P <.01) reduced repolarization dispersion when pacing ≥3.5 cm from the scar compared to pacing at 0.2 cm.

Conclusion: Increased dispersion of repolarization in the vicinity of the scar and widening of the vulnerable window when pacing in proximity to scar provides a mechanistic explanation for VT induction in ICM-CRT with lead placement proximal to scar. Pacing 3.5 cm or more from scar may avoid increasing VT risk in ICM-CRT patients.

Keywords: Cardiac resynchronization therapy; Infarct scar; Patient-specific modeling; Ventricular tachycardia.

Figure 1

Pacing locations (green) relative to scar (black). Orange indicates distances from scar. Blue plane indicates the mid-scar plane located 5 cm from the apex.

Figure 2

Location of S1 and S2 stimuli. Left: S1 stimuli locations at the epicardial surface 0.2 and 4.5 cm from the scar (blue spheres).Right: Twenty-one S2 pacing locations (yellow spheres) selected on the endocardial surface within the border zone (BZ).

Figure 3

Activation times (ms) and repolarization gradients (ms/mm) after a point stimulus on the left ventricular epicardial surface of one of the models in our cohort, where the scar was modeled as healthy tissue. Left: Isochrones are 10 ms apart. Location of the epicardial lead is indicated by pink circle. Fiber orientation on the epicardial surface is indicated by the black arrow.Right: Spatial distribution of local repolarization gradients corresponding to the activation sequence shown on the left. Large repolarization gradients (red) spread away from the pacing site in the direction transverse to fibers.

Figure 4

A: Volumes of high repolarization gradients (HRGs) within 1 cm around the scar relative to pacing distance from scar. P values are displayed. n.s. = nonsignificant. B, C: Example of repolarization gradients within the left ventricular (LV) epicardial surface for one of the models when pacing 0.2 cm (B) and 2.5 cm (C) from the scar. White curves indicate the region 1 cm around the scar. The core of the scar is shown in black. Filled white circles indicate the location of the LV epicardial lead.

Figure 5

Example of unidirectional block (A) and normal propagation (B) after an S2 stimulus. C: Vulnerable window (ms) at each S2 pacing location when the left ventricular lead is located 0.2 and 4.5 cm from the scar. An increase in the vulnerable window is observed when pacing in proximity to scar compared to pacing away from it.

Figure 6

Effect of electrophysiological (EP) changes commonly found in heart failure (HF) on the volume of high repolarization gradients (HRG) within 1 cm around the scar when pacing 0.2 and 4.5 cm from the scar. From left to right: “Base model” refers to the model with normal conduction velocity (CV) within the left ventricle, slow CV within the border zone (BZ), and normal action potential (AP) morphology; “Fast CV model” and “Slow CV model” refer to models with 20% faster and slower CV within LV and BZ relative to values of the “Base model”, respectively; and “HF AP model” refers to the model with increased action potential duration . A, B: Examples of the spatial distribution of local repolarization gradients when pacing 0.2 and 4.5 cm from scar, respectively. C: Plots of the volume of HRG within 1 cm around the scar when pacing 0.2 and 4.5 cm from scar for each EP model.