Role of the Purkinje-Muscle Junction on the Ventricular Repolarization Heterogeneity in the Healthy and Ischemic Ovine Ventricular Myocardium

FIGURE 1. Surface versus intramural PMJ activation determined by VF∗ calculation. (A) Schematic representation of the origins of myocardial activation from the Purkinje network in sheep (yellow star). Purkinje mediated-activation showed two different transmural activation profiles, a surface activation or an intramural activation. (B) Mean ± SD of VF∗ or activation number for surface or intramural activation during PF pacing at 1 Hz. Transmural AT was defined as the difference between time at the origin of activation and time at the origin of breakthrough. Dashed line shows VF∗ threshold differentiating origins residing at the tissue surface or deeper myocardium. Statistical differences were determined by paired t-test (∗p < 0.05). (C) Corresponding endocardial AT and VF∗ maps for a short or a long transmural AT during PF stimulation at 1 Hz, by site marked (∗). Arrows show origin of early activation and dashed arrows origin of breakthrough. Isochrones are 10 ms spacing.

FIGURE 2. Influence of pacing location on endocardial and epicardial APD. Endocardial and epicardial AT and APD maps and corresponding AP traces derived from LV preparations when pacing at 1 Hz on the ENDO (A), EPI (B), or PF (C), by site marked (∗). (C) Origins of early activation during PF pacing are localized by arrows. Regions for APD measurements are indicated by the black boxes for PMJs and the red box for ENDO. Dashed arrows show origins of breakthrough. Black circles show PMJs location on AT and APS maps for all pacing sites. Isochrones are 5 ms spacing for AT and APD maps.

FIGURE 3. Dependence of transmural APD heterogeneity on the activation sequence. (A) Frequency-dependence of transmural differences of APD for ENDO (n = 7), EPI (n = 7), and PF (n = 7) stimulation sites. APD measurements were performed at origins of activation or origins of breakthrough on each surface for each stimulus location (black box shown in Figure 2). (B) Transmural differences of APD when pacing ENDO, EPI, and PF at different frequencies. (C) Mean ± SD APD at the PMJ for ENDO, EPI, or PF stimulation at 1 Hz (upper panel) and corresponding AP traces (lower panel). CL, cycle length. Statistical differences were determined by paired t-tests for ENDO and EPI pacing (∗p < 0.05), one-way repeated measures ANOVA for PF stimulation (PMJ vs EPI, ∗p < 0.05; PMJ vs ENDO, #p < 0.05; ENDO vs EPI, aaap < 0.05), and one-way ANOVA for transmural gradient (PF vs EPI pacing, ∗p < 0.05; PF vs ENDO pacing, #p < 0.05; ENDO vs EPI pacing, aaap < 0.05).

FIGURE 4. Impact of ischemia on transmural APD heterogeneity. Endocardial and epicardial AT and APD maps and corresponding AP traces as shown in Figure 2 for ENDO (A), EPI (B), and PF (C) pacing. Stimulation sites are indicated by (∗), arrows show origins of early activation for PF stimulation in C, dashed arrows indicate origin of breakthrough, and black circles show PMJs location on AT and APD maps for all pacing sites. Isochrones are 5 ms spacing for endocardial maps and 10 ms for epicardial maps.

FIGURE 5. Greater resistance of Purkinje network to ischemia. (A) Mean ± SD EPI APD across all pacing locations at 1 Hz, before and after 7 min of ischemia. CTL, control; ISC, ischemia. (B) Mean ± SD APD from ENDO and EPI surfaces for each stimulus location stimulation at 1 Hz (before and after ISC). (C) Transmural differences of APD when pacing ENDO, EPI, and PF at 1 Hz, before and after ISC. (D) Mean ± SD PMJ APD (upper panel) and corresponding AP traces (lower panel), as in Figure 3, are shown for all pacing locations, before and after ISC. Statistical differences were determined by paired t-tests (∗p < 0.05).

FIGURE 6. Impact of pacing sequence on dispersion of repolarization during acute ischemia. (A) RT endocardial and epicardial maps for the different pacing modalities in control conditions and after 7 min of ischemia. Stimulation sites are indicated by (∗) and black circle shows PMJ location for all pacing sites. Isochrones are 10 ms spacing. CTL, control; ISC, ischemia. (B) Mean ± SD global RT dispersion when pacing ENDO, EPI, and PF at 1 Hz, before and after ISC. Statistical differences were determined by two-way ANOVA (∗p < 0.05). (C) Mean ± SD local RT dispersion at PMJs compared to endocardium, before and after ISC. Statistical differences were determined by paired t-test (∗p < 0.05).

FIGURE 7. Transmural APD heterogeneity in human ventricular wedge model. (A) APs derives from human from PF, endocardial, and epicardial ionic models of control heart. Shown are the last of a train of 100 APs stimulated at 1 Hz. AT and APD maps in simulations from transmural cross-sections of wedges for ENDO (B), EPI (C), and PF (D) stimulation at 1 Hz. Right panels show transmural APD profiles intersecting the PMJ and APD profiles along the free-running PF (red line). Isochrones are 2 ms spacing for AT maps and 5 ms for APD maps.

FIGURE 8. Transmural APD heterogeneity in ischemic human model. (A) APs derived from human from PF, endocardial, and epicardial ionic models of ischemic heart. The last of a train of 100 APs are shown. Same PF control AP was used in ischemic condition due to Purkinje resistance to ischemia. AT and APD maps and APD profiles are shown for simulations of ischemia following ENDO (B), EPI (C), and PF (D).

FIGURE 9. Quantification of sheep functional PMJs before and after ischemia. (A) Mean ± SD number of activation origins when pacing PF at 1 Hz. (B) Corresponding endocardial AT maps for three different wedge preparations. Arrows show origins of early activation corresponding to PMJs. Isochrones are 10 ms spacing. Statistical differences were determined by paired t-test (∗p < 0.05).