Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2012 May 14;590(Pt 18):4537–4551. doi: 10.1113/jphysiol.2012.229088

Three-dimensional mechanisms of increased vulnerability to electric shocks in myocardial infarction: Altered virtual electrode polarizations and conduction delay in the peri-infarct zone

Lukas J Rantner 1,2, Hermenegild J Arevalo 1,2, Jason L Constantino 1,2, Igor R Efimov 3, Gernot Plank 4, Natalia A Trayanova 1,2
PMCID: PMC3477756  PMID: 22586222

Abstract

Defibrillation efficacy is decreased in infarcted hearts, but the mechanisms by which infarcted hearts are more vulnerable to electric shocks than healthy hearts remain poorly understood. The goal of this study was to provide insight into the 3D mechanisms for the increased vulnerability to electric shocks in infarcted hearts. We hypothesized that changes in virtual electrode polarizations (VEPs) and propagation delay through the peri-infarct zone (PZ) were responsible. We developed a microanatomically detailed rabbit ventricular model with chronic myocardial infarction from magnetic resonance imaging and enriched the model with data from optical mapping experiments. We further developed a control model without the infarct. The simulation protocol involved apical pacing followed by biphasic shocks. Simulation results from both models were compared. The upper limit of vulnerability (ULV) was 8 V cm−1 in the infarction model and 4 V cm−1 in the control model. VEPs were less pronounced in the infarction model, providing a larger excitable area for postshock propagation but smaller transmembrane potential gradients to initiate new wavefronts. Initial post-shock transmural activation occurred at a later time in the infarction model, and the PZ served to delay propagation in subsequent beats. The presence of the PZ was found to be responsible for the increased vulnerability.

Key points

  • Defibrillation is known to be less efficient in infarcted than in healthy hearts.

  • In a rabbit model of myocardial infarction, altered 3D distribution of virtual electrodes and propagation delay in the peri-infarct zone caused increased vulnerability to electric shocks in infarcted hearts.

  • The infarct scar alone – without the presence of a peri-infarct zone – did not cause an increase in vulnerability.

  • The results help us to understand the mechanisms of increased vulnerability and decreased defibrillation efficacy in infarcted hearts.

Introduction

Myocardial infarction (MI) as a consequence of coronary artherosclerosis is the underlying cause of up to 80% of ventricular tachyarrhythmias leading to sudden cardiac death (SCD) (Huikuri et al. 2001). Defibrillation, the application of a strong electric shock to the heart, is the only known way of terminating these disturbances in heart rhythm and, thus, preventing SCD. The use of shocks to treat malignant cardiac arrhythmias is associated with a host of adverse effects that include cellular injury from electroporation (Tung et al. 1994; Al-Khadra et al. 2000), cardiac conduction disturbances (Eysmann et al. 1986), mechanical dysfunction (Tung et al. 1994; Tokano et al. 1998; Mollerus & Naslund, 2007), increased mortality (Moss et al. 2002; Larsen et al. 2011), and pain and psychological trauma (Maisel, 2006). Better understanding the mechanisms of defibrillation, and specifically in infarcted hearts, is likely to lead to an optimization of the delivery of defibrillation in patients with MI. Indeed, research on defibrillation mechanisms has focused predominantly on the healthy heart (Dixon et al. 1987; Efimov et al. 1998; Rodriguez et al. 2005; Ashihara et al. 2008), with limited insights under the conditions of MI. It has been shown experimentally that vulnerability to shock-induced ventricular tachycardia (VT) increases in the infarcted rabbit heart (Li et al. 2005). However, the 3D mechanisms responsible for the increased vulnerability to electric shocks in MI remain unknown, and the interaction of the electric shock with the electrophysiological and structural substrate of MI in the depth of the ventricular wall is poorly understood.

A key component of the distinct electrophysiological substrate of MI is the border or peri-infarct zone (PZ). Tissue in the PZ is electrically and structurally remodelled. This leads to abnormal cell-to-cell coupling, increased anisotropy, and changes in ion channel properties, all of which can form a substrate for reentry (for review, see Nattel et al. 2007). It is, therefore, likely that the remodelled tissue of the PZ could also play a role in the substrate for post-shock arrhythmia formation and maintenance.

Post-shock arrhythmogenesis results from the specific distribution of virtual electrode polarization (VEP; depolarizing and hyperpolarizing changes in membrane potential in response to the applied electric field) established at the end of the shock (Rodriguez et al. 2005; Ashihara et al. 2008; Constantino et al. 2010); VEP is a function of the electrophysiological state and the structure of the tissue (Trayanova et al. 1998a; Trayanova, 2001). Hence, electrically and structurally remodelled tissue in the PZ along with the infarct scar is expected to establish a specific 3D VEP distribution and subsequent post-shock propagation patterns that could result in post-shock arrhythmogenesis significantly different from that in the healthy heart.

The goal of this study is to ascertain the mechanisms for the increased vulnerability to external electric shocks observed experimentally in infarcted rabbit hearts (Li et al. 2005) and to test the hypotheses that it is due to (1) altered 3D distribution and magnitude of VEPs in the zone of infarct, altering the immediate post-shock transmural propagation, and (2) subsequent propagation delay in the PZ. The 3D activity in the rabbit ventricles both during and after external electric shocks needs to be analysed in order to achieve the aim of the study and test these hypotheses. Because current optical mapping techniques cannot resolve electrical activity in the depth of the ventricular wall, we use an established computer modelling approach here, in which the model is enriched with experimental data.

Methods

A detailed description of the methods can be found in the online Supplemental Material in Supplementary Methods. A brief description is below.

Ethical approval

The animal protocol was approved by the Washington University Institutional Animal Care and Use Committee. Details on anaesthesia and killing of the animal involved in this study can be found in a previous publication (Li et al. 2005).

Model development

A New Zealand White rabbit at 7.5 weeks post-infarction was used to gather the needed input for the computational model of the same heart. Ex-vivo optical mapping studies were performed according to established protocols (Li et al. 2009; Ripplinger et al. 2009) (for details see Supplementary Methods).

The infarcted rabbit heart subsequently underwent high-resolution (61 μm × 61 μm × 60 μm) magnetic resonance imaging (MRI) and diffusion tensor MRI (DTMRI) to acquire data for model generation; a representative long-axis slice of the MR image stack is shown in Figure 1A. From the MRI scans, a highly detailed finite-element geometric model of the rabbit ventricles was constructed, following the model generation pipeline developed recently (Vadakkumpadan et al. 2009, 2010). The scar and PZ were segmented from the healthy myocardium (Fig. 1B) (McDowell et al. 2011). The segmented images were used to generate the finite element mesh of the ventricles using the meshing package Tarantula (CAE Software Solutions, Eggenburg, Austria); the methodology is described in (Prassl et al. 2009). Each element of the 3 million node mesh (average edge length 135 μm) was assigned a fibre orientation based on the primary eigenvector and a sheet normal orientation based on the tertiary eigenvector of the diffusion tensor as obtained from the DTMRI data (Fig. 1C) (Scollan et al. 1998).

Figure 1. The model.

Figure 1

Open in a new tab

A, ex-vivo MRI scan of the rabbit heart with healed myocardial infarction. B, left panel, anterior view of the ventricles submerged in a perfusing bath and placed between plate electrodes (blue, grounding electrode; red, shock electrode). The infarct scar is shown in blue, the PZ is shown in green. The pink square on the apex shows the location of the pacing electrode. The right insert shows the highly detailed structure of the scar and the PZ. The left insert shows the details of the computational mesh. Right panel, anterior view of the ventricles cut by two coronal planes. Vm traces of nodes marked by pink disks are shown in Suppl. Fig. S1. C, anterior view of the fibre orientations in the ventricles.

The membrane kinetics of the ventricular myocytes in the normal portion of the ventricles were represented by the species-specific Mahajan–Shiferaw et al. model of the rabbit ventricular action potential (Mahajan et al. 2008). The ionic model and the conductivity tensors in the PZ were altered, in accordance with experimental data, to represent PZ electrophysiological and structural remodelling (details can be found in Supplementary Methods) (Litwin & Bridge, 1997; Pu & Boyden, 1997; Jiang et al. 2000; Cabo & Boyden, 2003; Yao et al. 2003). The action potential duration (measured at 90% repolarization, APD90) in the PZ cell was prolonged compared to the healthy cell (207 ms vs. 152 ms at 240 ms pacing cycle length, PCL), which is in agreement with experimental recordings from rabbit PZ cells (Litwin & Bridge, 1997). The necrotic infarct scar was modelled as an insulator. Our choice was based on the fact that while there has been speculation regarding the role of coupled fibroblasts in conduction in the infarct scar (Walker et al. 2007), conclusive experimental data is lacking.

Mathematical description of current flow in cardiac tissue in the rabbit ventricles was based on the bidomain representation (Plonsey, 1988). The bidomain equations were solved with the Cardiac Arrhythmia Research Package (CARP) (Vigmond et al. 2002, 2003). The specific numerical techniques used in CARP have been described extensively elsewhere (Plank et al. 2007, 2008; Vigmond et al. 2008).

A control rabbit ventricular model was also developed where the ionic properties and conductivities in the entire ventricles were set to the values used for healthy tissue. The model was used to elucidate the role of the PZ and infarct scar in post-shock arrhythmogenesis. In order to determine if the infarct scar without the electrophysiologically remodelled PZ would cause an increased vulnerability to electric shocks, we created an additional model that did feature the infarct scar, but that had the ionic and conduction properties of healthy tissue in the area of the PZ. We referred to this model as the ‘no PZ’ model. Furthermore, we created a model where the entire infarct region (scar and PZ) was modelled as scar (‘scar only’ model) in order to determine if increased scar size (occupying the entire zone of infarct) would lead to increased vulnerability. Finally, we created two additional models, one where only the conductivity values were changed in the PZ but no ionic remodelling was represented (‘PZ anisotropy’ model), and another where the PZ was remodelled in terms of ionic currents, but conductivity values were not changed (‘PZ ionic’ model). These two models were used to determine if either conductivity or ionic remodelling alone would cause an increase in vulnerability.

Enriching the model with electrical activation experimental data

While the geometry and fibre orientation were acquired experimentally in this study, and the ionic model was species-specific, the bidomain conductivity values to be assigned to the model are not known for the rabbit heart. Uncertainties in bidomain tissue conductivities lead to erroneous conduction velocities (CVs) and thus unrealistic post-shock activation patterns. Therefore, we used experimental data to match the model and experimental activation patterns and adjusted the set of bidomain conductivities as described in Supplementary Methods; similar approaches were used in previous studies by our group (Rodriguez et al. 2005; Bishop et al. 2007; Narayan et al. 2008). Activation maps following pacing at PCL of 300 ms were compared between optical mapping recordings and computer simulations for three 1 cm × 1 cm recording windows on the epicardial surface of the same rabbit heart.

With the adjusted bidomain conductivity values, the activation times and patterns for the three epicardial recording windows were consistent between experiment and model (Fig. 2A: 18–38 ms in experiment vs. 21–38 ms in model; Fig. 2B: 24–45 ms in experiment vs. 24–47 ms in model; Fig. 2C: 35–46 ms in experiment vs. 35–50 ms in model). Activation of the entire ventricles took 53 ms in the experiment and 57 ms in the model. With the activation patterns corresponding well between experiment and model, the 3D rabbit ventricular model was then used to explore the shock-induced behaviour in the depth of the ventricular wall not accessible by fluorescence recordings using the protocols below.

Figure 2. Enriching the model with experimental data.

Figure 2

Open in a new tab

Left panels, the excised rabbit heart. Black squares mark 1 cm × 1 cm areas analysed in the right panels. Right panels, experimental and simulated epicardial pacing activation maps. A, left lateral views. B, right lateral views. C, posterior views.

Simulation protocol for shock-induced vulnerability studies

Sustained arrhythmias following electric shocks could not be induced in the model with the conductivity values determined as described above. This is consistent with experimental results from infarcted rabbit hearts, where sustained VT could not be induced without the administration of flecainide (Li et al. 2009; Ripplinger et al. 2009). Therefore, CV in the model was decreased by 30% by reducing the conductivity values, thereby promoting arrhythmogenesis. A 30% decrease in CV mimics the effects of flecainide administration on CV (Coromilas et al. 1995).

The ventricles were paced at the apex at a PCL of 240 ms, which was the shortest PCL that did not induce voltage or Ca2+ transient alternans (see Supplementary Methods for details). Biphasic plate-electrode shocks (see Fig. 1B for electrode set-up) of both polarities and 7 ms duration (3.5/3.5 ms, 50% tilt) (Efimov et al. 1998) were delivered at varying coupling intervals (CIs) and shock strengths in order to determine the ventricles’ vulnerability to electric shocks.

The upper limit of vulnerability (ULV) was defined as the highest shock strength that induced a sustained arrhythmia. A post-shock arrhythmia was considered sustained if five or more post-shock beats were observed (Li et al. 2005; Ashihara et al. 2008). The ULVs were determined in the infarction and control models. Three-dimensional maps of VEPs (measured as ΔVm, the difference between the transmembrane potential Vm if no shock was given and the shock-end Vm) (Trayanova et al. 2003) and post-shock propagation patterns were compared between infarction and control models in order to determine the mechanisms responsible for changes in ULV in MI and to test our hypotheses.

The applied electric field was referred to as RV− when the left ventricular (LV) electrode was the anode. If the LV electrode was the cathode, the shock was referred to as LV− (Rodriguez et al. 2005, 2006). Shock strength was defined as the leading-edge value of the applied electric field.

Results

Vulnerability to electric shocks

Figure 3 shows the grids representing shock-induced vulnerability for the computational infarction and control models following biphasic RV− and LV− shocks. The ULV for RV− shocks was 8 V cm−1 in the infarction model and 4 V cm−1 in the control model, while the ULV for LV− shocks was 4 V cm−1 in infarction and 2 V cm−1 in control. In the exploration of mechanisms responsible for the elevated ULV in MI below, we used, without loss of generality, simulations with RV− shocks, since the ULV was higher in the infarction model than in control irrespective of the polarity of the applied electric field.

Figure 3. Vulnerability grids.

Figure 3

Open in a new tab

Vulnerability grids following biphasic shocks. Filled circles refer to sustained post-shock VT, open circles to unsustained post-shock VT, ‘–’ mark lack of post-shock activity, the red dotted lines mark the ULV. A, vulnerability grid after RV− shocks for the infarction model. B, vulnerability grid after RV− shocks for the control model. C, vulnerability grid after LV− shocks for the infarction model. D, vulnerability grid after LV− shocks for the control model.

Virtual electrode polarizations are altered in the infarcted heart

Figure 4 compares the pre-shock states and the shock responses of the infarction and control models to an electric shock (RV−) with a shock strength of 6 V cm−1, applied at a CI of 200 ms. This shock resulted in sustained post-shock arrhythmia in the infarcted heart, whereas in control it resulted only in non-sustained arrhythmia. Transmembrane potential traces of the three nodes portrayed in the right panel of Figure 1B are shown in Supplementary Figure S1 for this shock.

Figure 4. Virtual electrode polarization following a 6 V cm−1 shock delivered at a CI of 200 ms.

Figure 4

Open in a new tab

Anterior views of the ventricles, with the scar shown in black and the PZ shaded in grey. A, pre-shock Vm. The APD in the PZ was prolonged (green arrows) and conduction was slowed in the PZ. Far left panel, Vm difference between infarction and control models (calculated as control Vm minus infarction Vm). White colour marks areas where control Vm was identical to infarction Vm, blue colour marks areas where control Vm was lower than infarction Vm, and red colour marks areas where control Vm was higher than infarction Vm. B, shock-end ΔVm. ΔVm was less positive in the infarction model (green arrows). Far right panel, ΔVm difference between infarction and control models, calculated as control ΔVm minus infarction ΔVm. C, shock-end Vm. Less tissue was excited in the infarction model (purple arrows). Far left panel, Vm difference between infarction and control models, again computed as control Vm minus infarction Vm. D, shock-end Vm gradient maps. There were smaller Vm gradients close to the LV epi- and endocardium and in the midmyocardium close to the PZ (red arrows) in the infarction model. Far right panel, Vm gradient difference between infarction and control models (control Vm gradient minus infarction Vm gradient).

At shock-onset, Vm was higher in the infarction model than in control, specifically in the area of the PZ and in the LV (see Fig. 4A; the small panel on the far left shows the Vm differences between infarction and control, calculated as Vm values in control minus Vm in the infarction model). Due to decreased CV in the PZ – mean CV in the PZ was 0.17 ± 0.09 m s−1 compared to 0.31 ± 0.08 m s−1 in the control heart at locations corresponding to the PZ (P value <2.2 × 10−16) – overall activation of the ventricles by the apical pacing stimulus took longer in the infarction model (88 ms) than in control (79 ms). This slowed propagation resulted in less repolarized tissue at shock-onset in the PZ and in the LV in the infarction model as compared to control. In the PZ, this was further exacerbated by the prolonged APD90 caused by the ionic remodelling there, leading to a mean pre-shock Vm of −64.52 ± 18.15 mV in the PZ and −77.84 ± 0.49 mV in the corresponding region in control (P < 2.2 × 10−16; green arrows in Fig. 4A).

Figure 4B presents the distribution of ΔVm and shows the depolarizing (virtual cathodes; red areas) and hyperpolarzing membrane effects (virtual anodes; blue areas) of the electric shock. The small panel on the far right shows the differences in ΔVm between infarction and control (as control minus infarction). It is evident from the figure that VEPs were less positive in infarction than in control in the PZ and on the LV endocardium (marked by the green arrows). Mean ΔVm was 23.36 ± 29.49 mV in the area of the PZ, and 61.41 ± 32.42 mV in the same area in the control ventricles (P < 2.2 × 10−16), showing that, on average, there was stronger shock-induced depolarization in the healthy tissue than in the PZ. Transmembrane potential traces at a node lying in healthy tissue on the endocardium, a node in the PZ adjacent to the scar, and a node in the epicardial PZ illustrate this observation and are shown in Supplementary Figure S1 and described in Supplementary Results.

Figure 4C shows the shock-end distribution of Vm. Shock-end Vm dictates post-shock behaviour, and is itself determined both by pre-shock state and shock-induced VEP. Differences in Vm between the infarction and control models (control values minus infarction values) are portrayed in the small panel on the far left. Mean shock-end Vm was −46.31 ± 28.88 mV in the PZ and −16.99 ± 32.38 mV in the same area in control (P < 2.2 × 10−16). Tissue in the PZ and on the LV endocardium was less depolarized in infarction than in control (see purple arrows in Fig. 4C), providing less of an excitatory stimulus for post-shock propagation in those areas.

Figure 4D shows shock-end Vm gradients, reflecting the fact that cells depolarized by the shock can supply current to nearby hyperpolarized cells. The panel on the far right displays the differences in Vm gradients between infarction and control. Gradients were smaller in the infarction model than in control in areas close to the LV epi- and endocardium, as well as in areas in the midmyocardium close to the PZ (red arrows), on the borders of the areas where shock-end Vm was lower in infarction than in control (purple arrows in Fig. 4C). The smaller post-shock Vm gradients in the infarction model provide a lesser stimulus current, as compared to control, to elicit post-shock propagation into nearby repolarized areas.

Immediate activation following shock end

Activation maps of the initial 80 ms after the shock are shown in Figure 5A. The propagating wavefront travelled in the LV towards the base, then excited the septum followed by the right ventricle (RV). The immediate activation of the LV following the shock started 32 ms after shock onset in the infarction model and 11 ms post-shock in control. It initiated in the areas where the Vm gradients were larger in control than in infarction (red arrows in Fig. 4D). Earliest propagation in the infarction model was delayed compared to control and originated from areas with high Vm gradients on the LV septum and the LV epicardium. Note that there were also shock-end activations in other areas with high Vm gradients in both infarction and control models (blue areas in RV wall, septum, and LV epicardium in Fig. 5A). These activations did not propagate because of a lack of excitable tissue nearby. Figure 5B shows the spatial extent of the post-shock wavefronts, calculated as the percentage of nodes that were newly activated at every millisecond. Initially, the wavefronts were smaller in infarction compared to control (difference marked by yellow area), emphasizing that the global activation onset was delayed in infarction.

Figure 5. Immediate post-shock propagation following a 6 V cm−1 shock delivered at a CI of 200 ms.

Figure 5

Open in a new tab

A, activation maps 0–80 ms post-shock. The ventricles are shown in an anterior view, the scar is shown in black and the PZ is shaded in grey. Grey arrows mark direction of propagation. B, the percentage of nodes that were newly activated at every millisecond 0–50 ms post-shock. The grey vertical line marks shock-end. Wavefronts were larger immediately after the shock in the control model (yellow area), but wavefronts in the infarction model became larger after the initial delay (blue area). C, percentage of excitable nodes (Vm≤−70 mV) out of the total number of nodes along with the mean value of INa inactivation gates h×j. More nodes were excitable in the infarction model than in control (yellow area).

However, the propagation of the shock-induced wavefront was faster in the infarction model than in control. The delay in the initial post-shock activation in the infarction model provided the tissue with more time to recover from the shock, becoming fully excitable. The post-shock excitable volume for the initial 50 ms after the shock is portrayed in Figure 5C (as percentage of nodes where Vm≤−70 mV; difference between infarction and control shown as yellow area). The larger excitable volume in the infarction model is confirmed by elevated mean values of the sodium current (INa) inactivation gates h×j, showing less inactivation and, therefore, more post-shock excitability in infarction compared to control (Fig. 5C). This large excitable volume in the infarction model ensured rapid propagation through the LV, evident from the fact that the earliest activation in the septum occurred 26 ms after the immediate post-shock propagated activation in the LV in the infarcted ventricles, but 37 ms in control (see Fig. 6A). Over time the spatial extent of the wavefronts (or the merged wavefront) in the infarcted ventricles increased due to the rapid propagation, becoming larger than in control after 38 ms post-shock (blue area in Fig. 5B).

Figure 6. Propagation delay in the peri-infarct zone following a 6 V cm−1 shock delivered at a CI of 200 ms.

Figure 6

Open in a new tab

A, activation maps 80–160 ms post-shock (anterior views of the ventricles, with the scar shown in black and the PZ shaded in grey). The red arrow points to an instance of conduction block in the infarct region, marked by the grey line. B, Vm maps 440 ms post-shock (posterior views). Small panels, three-dimensional activation wavefronts are shown as purple surfaces (posterior views). VT was sustained in the infarction model, but not in control.

Post-shock propagation delay in the infarct region

Transmural activation maps 80–160 ms after the shock are shown in Figure 6A. Propagation was delayed in and around the infarct region in this time interval. Activation wavefronts were located at roughly the same location in both models at 80 ms post-shock, but after 160 ms only small sections of the LV were activated in the infarcted ventricles whereas the propagating wavefront had almost reached the base of the control ventricles. Between 80 ms and 160 ms after the shock, mean CV was 0.12 ± 0.08 m s−1 in the PZ and 0.22 ± 0.13 m s−1 in the same area in control (P < 2.2 × 10−16), indicating a significant post-shock conduction slowing in the PZ.

The delay in initial post-shock activation of the LV along with the subsequent faster propagation through the LV in the infarcted ventricles (see Fig. 5A) resulted in tissue being still refractory near the endocardium when the wavefront of the second post-shock VT beat arrived there. Thus, conduction was blocked locally (see red arrow in Fig. 6A and black arrow in Suppl. Fig. S1A), facilitating arrhythmogenesis.

Figure 6B shows Vm maps 440 ms after the shock, which demonstrate that the arrhythmia was sustained in the infarction model. In control, on the other hand, the propagating wavefront encountered only refractory tissue in the LV after three post-shock beats and, therefore, self-terminated (see Suppl. Fig. S3 for activation maps for the first 500 ms post-shock where repeated localized conduction block and propagation delay was observed in the infarction model, but not in control).

Peri-infarct zone is responsible for the increased vulnerability to electric shocks

As shown above, propagation delay in the PZ played an important role in sustaining VT in the infarction model. However, it remained unclear whether or not the electrophysiologically remodelled PZ alone caused the increased vulnerability to electric shocks in the infarction model, or if the infarct scar also contributed to the increase. The vulnerability grid for the no PZ model following RV− shocks is shown in Figure 7A. The ULV for the no PZ model was identical to the ULV of the control model (4 V cm−1). Thus, the presence of the infarct scar alone did not increase vulnerability to electric shocks.

Figure 7. The role of the peri-infarct zone in vulnerability to electric shocks.

Figure 7

Open in a new tab

A, vulnerability grid after RV− shocks for the no PZ model. B, anterior views of the ventricles, with the scar shown in black. Differences in shock-end ΔVm between infarction and no PZ models (left panel) and between control and no PZ models (right panel) after a 6 V cm−1 shock delivered at 200 ms CI. Areas with ΔVm differences between infarction and no PZ models (green arrows) were larger than areas with differences between control and no PZ models (red arrows). C, histograms of absolute differences in ΔVm between infarction and no PZ models (red) and between control and no PZ models (black). D, activation maps for the no PZ model 0–160 ms following a 6 V cm−1 shock delivered at 200 ms CI (anterior views of the ventricles, scar shown in black).

Figure 7B shows differences in shock-end ΔVm between the infarction and the no PZ models (left panel) and between the control and the no PZ models (right panel). In the area of the PZ, in the septum towards the apex, on the LV endo- and epicardium towards the base, ΔVm was increased in the no PZ model compared to the infarction model (green arrows). On the other hand, on the LV epicardium close to the infarct scar and on the LV endocardium, ΔVm was decreased in the no PZ model compared to control (red arrows). Histograms of absolute differences in ΔVm between infarction and no PZ models, and between control and no PZ models are shown in Figure 7C. It is evident from the histograms that ΔVm differences between infarction and no PZ were much larger than between control and no PZ models.

Figure 7D shows transmural activation maps for the no PZ model. When comparing these maps to those in the right panels of Figures 5A and 6A, it is evident that the initial post-shock propagation pattern was virtually identical between the control and no PZ models, demonstrating that the small differences in ΔVm between control and no PZ did not result in different post-shock propagation patterns. Also, as expected, propagation through the infarct region was not delayed in the no PZ model compared to propagation through the same area in control, as is manifested by virtually identical positions of the activating wavefronts in the no PZ and control models at both 80 ms and 160 ms after the shock.

The ‘scar only’ and the ‘PZ anisotropy’ models experienced sustained arrhythmia after a 200 ms CI, 4 V cm−1 shock, but not after a 200 ms CI, 6 V cm−1 shock, exactly like the ‘no PZ’ model above. On the other hand, a 200 ms CI, 6 V cm−1 shock led to sustained arrhythmia in the ‘PZ ionic’ model. These results show that neither the size of the infarct region alone nor the conductivity changes in the PZ without ionic remodelling increased vulnerability to electric shocks, whereas ionic remodelling in the PZ did cause an increased ULV. The presence of the scar alone (even if the scar occupied the entire zone of infarct, including the PZ, as in the ‘scar only’ model) was not sufficient to cause large VEP changes and to delay propagation, and thus did not increase vulnerability. Therefore, it was the ionic remodelling in the PZ that was responsible for increasing the vulnerability of infarcted hearts to electric shocks by altering VEP and Vm gradients and by causing propagation delay.

Discussion

This study provided insights into the 3D mechanisms that underlie increased vulnerability to electric shocks in infarcted hearts and that could predispose infarcted hearts to defibrillation failure. Affirming our hypotheses, the main findings were:

  1. Vulnerability to electric shocks in infarcted hearts was increased for biphasic waveforms of both polarities.

  2. Decreased magnitude of the positive and increased magnitude of the negative VEPs in infarcted hearts resulted in the formation of a larger post-shock excitable gap compared to healthy hearts.

  3. Smaller shock-end Vm gradients in infarcted hearts provided smaller excitatory currents for post-shock propagation and thus caused a delayed initial post-shock activation.

  4. Propagation delay in the PZ during post-shock propagation allowed for the increase in excitable areas, which in turn facilitated reentry sustenance.

  5. The presence of the PZ but not of the scar alone caused increased vulnerability to electric shocks.

These findings provide valuable insights into the defibrillation mechanisms in the setting of MI, since increased vulnerability to electric shocks translates into increased defibrillation threshold (DFT). This is because induction of arrhythmia with electric shocks (i.e. vulnerability to electric shocks) and defibrillation failure are driven by the same mechanisms (Fabiato et al. 1967; Chen et al. 1991; Malkin et al. 1995; Rodriguez et al. 2005; Ashihara et al. 2008), also supported by the close correlation between ULV and DFT (Chen et al. 1986). Therefore, understanding the mechanisms of vulnerability to electric shocks provides a means to understand the mechanisms of defibrillation and arrhythmogenesis by failed shocks. Examining the vulnerability to electric shocks by computational means is also a cheaper route to uncovering the mechanisms by which an electric shock interacts with the 3D ventricles than by conducting defibrillation studies. In view of the fact that the present study used a high-resolution MRI-based model of the rabbit ventricles, reasonable computational tractability necessitated simulating vulnerability to electric shocks rather than defibrillation in uncovering these mechanisms.

The rabbit bidomain model of infarction presented here employed a recently published pipeline for the creation of image-based models (Vadakkumpadan et al. 2009, 2010). The model featured realistic structure, including infarct scar and PZ geometries, and fibre architecture. Thus, the model had three distinct regions (normal tissue, scar and PZ), consistent with previous histological studies that have shown the presence of a region between infarct scar and normal tissue, composed of complex interdigitations of healthy, remodelled and scar tissue (Factor et al. 1978, 1981). Our modelling approach treated this intermediate zone as a region with averaged remodelled electrophysiological properties resulting in decreased CV. This assumption is reasonable as demonstrated by other simulation studies showing that a microscopic heterogeneous mixture of viable myocardium and scar results in a global decrease in CV (Engelman et al. 2010). Structure and fibre orientations in our model were acquired from high resolution MRI and DTMRI scans. The ionic model used here incorporated Markovian Ca2+ handling (Mahajan et al. 2008); it was modified based on published data to represent cellular behaviour in the electrophysiologically remodelled PZ. The model was enriched with experimental data acquired from the same heart that was used for model development.

The model utilized in this study represents the first attempt to model delivery of electric shocks in the structurally remodelled intact ventricles. Previous attempts to model vulnerability to electric shocks in the diseased ventricles were limited to global ischaemia only (Rodriguez et al. 2004).

Reports regarding the effect of infarction on vulnerability to shocks and on defibrillation have been scarce and contradictory. An early study found an increase in defibrillation energy and current requirements at the onset of acute infarction (Babbs et al. 1983). In the study by Chang et al. (1986), hearts with healed infarcts were found to require lower peak voltage but not less total energy to terminate ventricular fibrillation (VF). Wharton et al. (1990) reported that the presence of infarction was not found to affect the ULV and DFT. In a recent optical mapping study, Li et al. (2005) found, in rabbit hearts with healed infarction, an increase in the ULV for monophasic shocks. We demonstrated that the ULV was higher in infarction than in control for biphasic shocks, implying the need for stronger defibrillation shocks – which are usually biphasic in clinical practice – in infarction than in control.

Li et al. (2005) suggested that post-shock arrhythmias in the infarcted rabbit heart mainly originated from break excitations in the endo- and midmyocardial PZ, leading to breakthrough on the epicardium near the PZ; they also reported diminished epicardial VEPs near the infarct region. Both Ripplinger et al. (2009) and Li et al. (2009) reported that, in infarcted rabbit hearts, long-term VT was maintained by spiral waves centred at the infarct region. Phase singularities clustered in the infarct region and, especially, on the edge of the infarct (Ripplinger et al. 2009). Our results underscore the importance of the PZ in post-shock arrhythmia induction. Remodelled ionic currents prolonged APD90 in the PZ, while reduced conductivities along with the altered ionic currents slowed propagation through the PZ. Both of these mechanisms changed the pre-shock state so that the LV was less repolarized in infarction compared to control. The pre-shock Vm in the ‘no PZ’ model was virtually identical to the pre-shock Vm in the control model (not shown). This demonstrates that the differences in pre-shock state between the control and infarction models were caused by electrophysiological remodelling and ensuing conduction delay in the PZ, and not by a possibly longer conduction pathway around the scar. In addition, reduced transverse conductivities – representing remodelled gap junctions – altered the anisotropy ratios in the PZ. The changes in pre-shock state and passive tissue properties resulted in changes in VEP, consistent with results from previous studies (Trayanova et al. 1998a, b). Furthermore, the remodelled ionic currents and gap junction distribution slowed conduction in the PZ, which resulted in post-shock propagation delay. Altered VEP and altered post-shock propagation acted synergistically, resulting in increased vulnerability to electric shocks in infarcted hearts. We further showed that the presence of an infarct scar alone did not increase vulnerability – even if the scar was enlarged to the size of the entire infarct region – and did not change post-shock propagation. Indeed, the scar alone did not cause a large change in VEP compared to control, and post-shock propagation patterns were virtually identical between the control model and the model with the scar but without the PZ (‘no PZ’ model). Thus, the remodelled PZ was the culprit for increased vulnerability in infarcted hearts.

It has been shown that collagen is upregulated after infarction (Fraccarollo et al. 2002), and that fibroblast density in the infarction area is elevated, but decreases over time (Camelliti et al. 2004). While we did not incorporate a specific distribution of collagen and fibroblasts in our model, we included conductivity remodelling in the PZ, which could arise from collagen upregulation. Since the results presented here demonstrate that only ionic remodelling in the PZ increases the ULV, whereas only conductivity remodelling or a scar occupying the entire infarct zone without any PZ does not increase vulnerability, we speculate that different degrees of collagen upregulation or distribution in the PZ would not have a major effect on vulnerability to electric shocks.

Limitations

The model used in this study did not incorporate the Purkinje system. This choice was based on the finding by Deo et al. (2009), who showed that the Purkinje system played an important role in prolonged post-shock VT, but not in arrhythmogenesis caused by electric shocks. Furthermore, Dosdall et al. (2010) demonstrated that the Purkinje system changed the defibrillation mechanisms after long-duration VF, but that the Purkinje system was not involved in defibrillation mechanisms after short-duration VT/VF.

Some specific values presented in this study such as ULV, activation times, ΔVm, and Vm gradients might be dependent on the size and location of the infarct region as well as on the specific extent of the PZ. Nevertheless, the mechanisms (e.g. altered VEPs, propagation delay) causing the increase in vulnerability in infarcted hearts would hold true regardless of the specific infarct geometry. This model does represent a typical example of an infarcted rabbit heart (Li et al. 2005; Kim et al. 2008), which in turn is a good model for infarction in human hearts because, like the human heart, the rabbit heart develops both an endocardial and an epicardial PZ (Li et al. 2005).

Although the choice of plate electrodes represents a simplified shock electrode configuration, its use in vulnerability and defibrillation studies (both experiment and simulation) is an established technique (Efimov et al. 2000; Ashihara et al. 2008; Ambrosi et al. 2011) since it resembles transthoracic defibrillation (the electric field in the region of the heart is close to uniform). Some specific values such as the ULV would likely be different for implantable cardioverter–defibrillator (ICD) electrode configurations, but we expect that the overall findings (increased vulnerability in infarction, VEP changes in infarction, propagation delay through the PZ) would remain valid in these cases.

Acknowledgments

L.J.R. is a recipient of a DOC fellowship of the Austrian Academy of Sciences at the Department of Biomedical Engineering, Johns Hopkins University. This study was supported by National Institutes of Health (NIH) grant R01-HL067322 to I.R.E., by NIH grants R01-HL082729 and R01-HL067322 to N.A.T., and by NIH fellowships F31-HL099275 to H.J.A., and F31-HL103090 to J.L.C.

Glossary

3D

three-dimensional

APD90

action potential duration at 90% repolarization

CI

coupling interval

CV

conduction velocity

DFT

defibrillation threshold

DTMRI

diffusion tensor magnetic resonance imaging

LV

left ventricle

MI

myocardial infarction

MRI

magnetic resonance imaging

PCL

pacing cycle length

PZ

peri-infarct zone

RV

right ventricle

SCD

sudden cardiac death

ULV

upper limit of vulnerability

VEP

virtual electrode polarization

VF

ventricular fibrillation

Vm

transmembrane potential

VT

ventricular tachycardia

Author contributions

Conception and design of the experiments: L.J.R. and N.A.T. conceptualized and designed the computational studies (at Johns Hopkins University, Baltimore), and I.R.E. was responsible for the optical mapping experiments (at Washington University in St. Louis). Collection, analysis and interpretation of data: L.J.R. performed the computer simulations, H.J.A. and J.L.C. contributed to the execution of the computational studies (at Johns Hopkins University, Baltimore), G.P. adapted the Mahajan–Shiferaw for use in whole organ simulations (at Medical University Graz and at Johns Hopkins University, Baltimore), and L.J.R. and N.A.T. analysed the data (at Johns Hopkins University, Baltimore). Drafting the article or revising it critically for important intellectual content: L.J.R. drafted the manuscript, and H.J.A., J.L.C., and N.A.T. critically revised the manuscript (at Johns Hopkins University, Baltimore). All authors approved the final version of the manuscript.

Disclosures

G.P. and N.A.T. have partial ownership of CardioSolv, LLC. CardioSolv, LLC was not involved in this research.

Supplementary material

S.1 Supplementary Methods

Figure S1

Figure S2

Figure S3

tjp0590-4537-SD1.pdf (405.9KB, pdf)

References

  1. Al-Khadra A, Nikolski V, Efimov IR. The role of electroporation in defibrillation. Circ Res. 2000;87:797–804. doi: 10.1161/01.res.87.9.797. [DOI] [PubMed] [Google Scholar]
  2. Ambrosi CM, Ripplinger CM, Efimov IR, Fedorov VV. Termination of sustained atrial flutter and fibrillation using low-voltage multiple-shock therapy. Heart Rhythm. 2011;8:101–108. doi: 10.1016/j.hrthm.2010.10.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ashihara T, Constantino J, Trayanova NA. Tunnel propagation of postshock activations as a hypothesis for fibrillation induction and isoelectric window. Circ Res. 2008;102:737–745. doi: 10.1161/CIRCRESAHA.107.168112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Babbs CF, Paris RL, Tacker WA, Jr, Bourland JD. Effects of myocardial infarction on catheter defibrillation threshold. Med Instrum. 1983;17:18–20. [PubMed] [Google Scholar]
  5. Bishop MJ, Rodriguez B, Qu F, Efimov IR, Gavaghan DJ, Trayanova NA. The role of photon scattering in optical signal distortion during arrhythmia and defibrillation. Biophys J. 2007;93:3714–3726. doi: 10.1529/biophysj.107.110981. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cabo C, Boyden PA. Electrical remodeling of the epicardial border zone in the canine infarcted heart: a computational analysis. Am J Physiol Heart Circ Physiol. 2003;284:H372–384. doi: 10.1152/ajpheart.00512.2002. [DOI] [PubMed] [Google Scholar]
  7. Camelliti P, Devlin GP, Matthews KG, Kohl P, Green CR. Spatially and temporally distinct expression of fibroblast connexins after sheep ventricular infarction. Cardiovasc Res. 2004;62:415–425. doi: 10.1016/j.cardiores.2004.01.027. [DOI] [PubMed] [Google Scholar]
  8. Chang MS, Inoue H, Kallok MJ, Zipes DP. Double and triple sequential shocks reduce ventricular defibrillation threshold in dogs with and without myocardial infarction. J Am Coll Cardiol. 1986;8:1393–1405. doi: 10.1016/s0735-1097(86)80313-8. [DOI] [PubMed] [Google Scholar]
  9. Chen PS, Feld GK, Mower MM, Peters BB. Effects of pacing rate and timing of defibrillation shock on the relation between the defibrillation threshold and the upper limit of vulnerability in open chest dogs. J Am Coll Cardiol. 1991;18:1555–1563. doi: 10.1016/0735-1097(91)90690-b. [DOI] [PubMed] [Google Scholar]
  10. Chen PS, Shibata N, Dixon EG, Martin RO, Ideker RE. Comparison of the defibrillation threshold and the upper limit of ventricular vulnerability. Circulation. 1986;73:1022–1028. doi: 10.1161/01.cir.73.5.1022. [DOI] [PubMed] [Google Scholar]
  11. Constantino J, Long Y, Ashihara T, Trayanova NA. Tunnel propagation following defibrillation with ICD shocks: hidden postshock activations in the left ventricular wall underlie isoelectric window. Heart Rhythm. 2010;7:953–961. doi: 10.1016/j.hrthm.2010.03.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Coromilas J, Saltman AE, Waldecker B, Dillon SM, Wit AL. Electrophysiological effects of flecainide on anisotropic conduction and reentry in infarcted canine hearts. Circulation. 1995;91:2245–2263. doi: 10.1161/01.cir.91.8.2245. [DOI] [PubMed] [Google Scholar]
  13. Deo M, Boyle P, Plank G, Vigmond E. Arrhythmogenic mechanisms of the Purkinje system during electric shocks: a modeling study. Heart Rhythm. 2009;6:1782–1789. doi: 10.1016/j.hrthm.2009.08.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dixon EG, Tang AS, Wolf PD, Meador JT, Fine MJ, Calfee RV, Ideker RE. Improved defibrillation thresholds with large contoured epicardial electrodes and biphasic waveforms. Circulation. 1987;76:1176–1184. doi: 10.1161/01.cir.76.5.1176. [DOI] [PubMed] [Google Scholar]
  15. Dosdall DJ, Osorio J, Robichaux RP, Huang J, Li L, Ideker RE. Purkinje activation precedes myocardial activation following defibrillation after long-duration ventricular fibrillation. Heart Rhythm. 2010;7:405–412. doi: 10.1016/j.hrthm.2009.11.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Efimov IR, Aguel F, Cheng Y, Wollenzier B, Trayanova N. Virtual electrode polarization in the far field: implications for external defibrillation. Am J Physiol Heart Circ Physiol. 2000;279:H1055–1070. doi: 10.1152/ajpheart.2000.279.3.H1055. [DOI] [PubMed] [Google Scholar]
  17. Efimov IR, Cheng Y, Van Wagoner DR, Mazgalev T, Tchou PJ. Virtual electrode-induced phase singularity: a basic mechanism of defibrillation failure. Circ Res. 1998;82:918–925. doi: 10.1161/01.res.82.8.918. [DOI] [PubMed] [Google Scholar]
  18. Engelman ZJ, Trew ML, Smaill BH. Structural heterogeneity alone is a sufficient substrate for dynamic instability and altered restitution. Circ Arrhythm Electrophysiol. 2010;3:195–203. doi: 10.1161/CIRCEP.109.890459. [DOI] [PubMed] [Google Scholar]
  19. Eysmann SB, Marchlinski FE, Buxton AE, Josephson ME. Electrocardiographic changes after cardioversion of ventricular arrhythmias. Circulation. 1986;73:73–81. doi: 10.1161/01.cir.73.1.73. [DOI] [PubMed] [Google Scholar]
  20. Fabiato A, Coumel P, Gourgon R, Saumont R. [The threshold of synchronous response of the myocardial fibers. Application to the experimental comparison of the efficacy of different forms of electroshock defibrillation] Arch Mal Coeur Vaiss. 1967;60:527–544. [PubMed] [Google Scholar]
  21. Factor SM, Okun EM, Kirk ES. The histological lateral border of acute canine myocardial infarction. A function of microcirculation. Circ Res. 1981;48:640–649. doi: 10.1161/01.res.48.5.640. [DOI] [PubMed] [Google Scholar]
  22. Factor SM, Sonnenblick EH, Kirk ES. The histologic border zone of acute myocardial infarction – islands or peninsulas? Am J Pathol. 1978;92:111–124. [PMC free article] [PubMed] [Google Scholar]
  23. Fraccarollo D, Galuppo P, Bauersachs J, Ertl G. Collagen accumulation after myocardial infarction: effects of ETA receptor blockade and implications for early remodeling. Cardiovasc Res. 2002;54:559–567. doi: 10.1016/s0008-6363(02)00256-0. [DOI] [PubMed] [Google Scholar]
  24. Huikuri HV, Castellanos A, Myerburg RJ. Sudden death due to cardiac arrhythmias. N Engl J Med. 2001;345:1473–1482. doi: 10.1056/NEJMra000650. [DOI] [PubMed] [Google Scholar]
  25. Jiang M, Cabo C, Yao J, Boyden PA, Tseng G. Delayed rectifier K currents have reduced amplitudes and altered kinetics in myocytes from infarcted canine ventricle. Cardiovasc Res. 2000;48:34–43. doi: 10.1016/s0008-6363(00)00159-0. [DOI] [PubMed] [Google Scholar]
  26. Kim SC, Vasanji A, Efimov IR, Cheng Y. Spatial distribution and extent of electroporation by strong internal shock in intact structurally normal and chronically infarcted rabbit hearts. J Cardiovasc Electrophysiol. 2008;19:1080–1089. doi: 10.1111/j.1540-8167.2008.01201.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Larsen GK, Evans J, Lambert WE, Chen Y, Raitt MH. Shocks burden and increased mortality in implantable cardioverter-defibrillator patients. Heart Rhythm. 2011;8:1881–1886. doi: 10.1016/j.hrthm.2011.07.036. [DOI] [PubMed] [Google Scholar]
  28. Li L, Nikolski V, Wallick DW, Efimov IR, Cheng Y. Mechanisms of enhanced shock-induced arrhythmogenesis in the rabbit heart with healed myocardial infarction. Am J Physiol Heart Circ Physiol. 2005;289:H1054–1068. doi: 10.1152/ajpheart.01253.2004. [DOI] [PubMed] [Google Scholar]
  29. Li W, Ripplinger CM, Lou Q, Efimov IR. Multiple monophasic shocks improve electrotherapy of ventricular tachycardia in a rabbit model of chronic infarction. Heart Rhythm. 2009;6:1020–1027. doi: 10.1016/j.hrthm.2009.03.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Litwin SE, Bridge JH. Enhanced Na+-Ca2+ exchange in the infarcted heart. Implications for excitation-contraction coupling. Circ Res. 1997;81:1083–1093. doi: 10.1161/01.res.81.6.1083. [DOI] [PubMed] [Google Scholar]
  31. Mahajan A, Shiferaw Y, Sato D, Baher A, Olcese R, Xie LH, Yang MJ, Chen PS, Restrepo JG, Karma A, Garfinkel A, Qu Z, Weiss JN. A rabbit ventricular action potential model replicating cardiac dynamics at rapid heart rates. Biophys J. 2008;94:392–410. doi: 10.1529/biophysj.106.98160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Maisel WH. Pacemaker and ICD generator reliability: meta-analysis of device registries. JAMA. 2006;295:1929–1934. doi: 10.1001/jama.295.16.1929. [DOI] [PubMed] [Google Scholar]
  33. Malkin RA, Idriss SF, Walker RG, Ideker RE. Effect of rapid pacing and T-wave scanning on the relation between the defibrillation and upper-limit-of-vulnerability dose-response curves. Circulation. 1995;92:1291–1299. doi: 10.1161/01.cir.92.5.1291. [DOI] [PubMed] [Google Scholar]
  34. McDowell KS, Arevalo HJ, Maleckar MM, Trayanova NA. Susceptibility to arrhythmia in the infarcted heart depends on myofibroblast density. Biophys J. 2011;101:1307–1315. doi: 10.1016/j.bpj.2011.08.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Mollerus M, Naslund L. Myocardial stunning following defibrillation threshold testing. J Interv Card Electrophysiol. 2007;19:213–216. doi: 10.1007/s10840-007-9150-9. [DOI] [PubMed] [Google Scholar]
  36. Moss AJ, Zareba W, Hall WJ, Klein H, Wilber DJ, Cannom DS, Daubert JP, Higgins SL, Brown MW, Andrews ML, Multicenter Automatic Defibrillator Implantation Trial III Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med. 2002;346:877–883. doi: 10.1056/NEJMoa013474. [DOI] [PubMed] [Google Scholar]
  37. Narayan SM, Bayer JD, Lalani G, Trayanova NA. Action potential dynamics explain arrhythmic vulnerability in human heart failure: a clinical and modeling study implicating abnormal calcium handling. J Am Coll Cardiol. 2008;52:1782–1792. doi: 10.1016/j.jacc.2008.08.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Nattel S, Maguy A, Le Bouter S, Yeh YH. Arrhythmogenic ion-channel remodeling in the heart: heart failure, myocardial infarction, and atrial fibrillation. Physiol Rev. 2007;87:425–456. doi: 10.1152/physrev.00014.2006. [DOI] [PubMed] [Google Scholar]
  39. Plank G, Liebmann M, Weber dos Santos R, Vigmond EJ, Haase G. Algebraic multigrid preconditioner for the cardiac bidomain model. IEEE Trans Biomed Eng. 2007;54:585–596. doi: 10.1109/TBME.2006.889181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Plank G, Zhou L, Greenstein JL, Cortassa S, Winslow RL, O’Rourke B, Trayanova NA. From mitochondrial ion channels to arrhythmias in the heart: computational techniques to bridge the spatio-temporal scales. Philos Transact A Math Phys Eng Sci. 2008;366:3381–3409. doi: 10.1098/rsta.2008.0112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Plonsey R. Bioelectric sources arising in excitable fibers (ALZA lecture) Ann Biomed Eng. 1988;16:519–546. doi: 10.1007/BF02368014. [DOI] [PubMed] [Google Scholar]
  42. Prassl AJ, Kickinger F, Ahammer H, Grau V, Schneider JE, Hofer E, Vigmond EJ, Trayanova NA, Plank G. Automatically generated, anatomically accurate meshes for cardiac electrophysiology problems. IEEE Trans Biomed Eng. 2009;56:1318–1330. doi: 10.1109/TBME.2009.2014243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Pu J, Boyden PA. Alterations of Na+ currents in myocytes from epicardial border zone of the infarcted heart. A possible ionic mechanism for reduced excitability and postrepolarization refractoriness. Circ Res. 1997;81:110–119. doi: 10.1161/01.res.81.1.110. [DOI] [PubMed] [Google Scholar]
  44. Ripplinger CM, Lou Q, Li W, Hadley J, Efimov IR. Panoramic imaging reveals basic mechanisms of induction and termination of ventricular tachycardia in rabbit heart with chronic infarction: implications for low-voltage cardioversion. Heart Rhythm. 2009;6:87–97. doi: 10.1016/j.hrthm.2008.09.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Rodriguez B, Eason JC, Trayanova N. Differences between left and right ventricular anatomy determine the types of reentrant circuits induced by an external electric shock. A rabbit heart simulation study. Prog Biophys Mol Biol. 2006;90:399–413. doi: 10.1016/j.pbiomolbio.2005.06.008. [DOI] [PubMed] [Google Scholar]
  46. Rodriguez B, Li L, Eason JC, Efimov IR, Trayanova NA. Differences between left and right ventricular chamber geometry affect cardiac vulnerability to electric shocks. Circ Res. 2005;97:168–175. doi: 10.1161/01.RES.0000174429.00987.17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Rodriguez B, Tice BM, Eason JC, Aguel F, Trayanova N. Cardiac vulnerability to electric shocks during phase 1A of acute global ischemia. Heart Rhythm. 2004;1:695–703. doi: 10.1016/j.hrthm.2004.08.018. [DOI] [PubMed] [Google Scholar]
  48. Scollan DF, Holmes A, Winslow R, Forder J. Histological validation of myocardial microstructure obtained from diffusion tensor magnetic resonance imaging. Am J Physiol Heart Circ Physiol. 1998;275:H2308–2318. doi: 10.1152/ajpheart.1998.275.6.H2308. [DOI] [PubMed] [Google Scholar]
  49. Tokano T, Bach D, Chang J, Davis J, Souza JJ, Zivin A, Knight BP, Goyal R, Man KC, Morady F, Strickberger SA. Effect of ventricular shock strength on cardiac hemodynamics. J Cardiovasc Electrophysiol. 1998;9:791–797. doi: 10.1111/j.1540-8167.1998.tb00118.x. [DOI] [PubMed] [Google Scholar]
  50. Trayanova N. Concepts of ventricular defibrillation. Philos Transact A Math Phys Eng Sci. 2001;359:1327–1337. [Google Scholar]
  51. Trayanova N, Skouibine K, Aguel F. The role of cardiac tissue structure in defibrillation. Chaos. 1998a;8:221–233. doi: 10.1063/1.166299. [DOI] [PubMed] [Google Scholar]
  52. Trayanova N, Skouibine K, Moore P. Virtual electrode effects in defibrillation. Prog Biophys Mol Biol. 1998b;69:387–403. doi: 10.1016/s0079-6107(98)00016-9. [DOI] [PubMed] [Google Scholar]
  53. Trayanova NA, Gray RA, Bourn DW, Eason JC. Virtual electrode-induced positive and negative graded responses: new insights into fibrillation induction and defibrillation. J Cardiovasc Electrophysiol. 2003;14:756–763. doi: 10.1046/j.1540-8167.2003.03042.x. [DOI] [PubMed] [Google Scholar]
  54. Tung L, Tovar O, Neunlist M, Jain SK, O’Neill RJ. Effects of strong electrical shock on cardiac muscle tissue. Ann N Y Acad Sci. 1994;720:160–175. doi: 10.1111/j.1749-6632.1994.tb30444.x. [DOI] [PubMed] [Google Scholar]
  55. Vadakkumpadan F, Arevalo H, Prassl AJ, Chen J, Kickinger F, Kohl P, Plank G, Trayanova N. Image-based models of cardiac structure in health and disease. Wiley Interdiscip Rev Syst Biol Med. 2010;2:489–506. doi: 10.1002/wsbm.76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Vadakkumpadan F, Rantner LJ, Tice B, Boyle P, Prassl AJ, Vigmond E, Plank G, Trayanova N. Image-based models of cardiac structure with applications in arrhythmia and defibrillation studies. J Electrocardiol. 2009;42:157.e151–110. doi: 10.1016/j.jelectrocard.2008.12.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Vigmond EJ, Aguel F, Trayanova NA. Computational techniques for solving the bidomain equations in three dimensions. IEEE Trans Biomed Eng. 2002;49:1260–1269. doi: 10.1109/TBME.2002.804597. [DOI] [PubMed] [Google Scholar]
  58. Vigmond EJ, Hughes M, Plank G, Leon LJ. Computational tools for modeling electrical activity in cardiac tissue. J Electrocardiol. 2003;36(Suppl):69–74. doi: 10.1016/j.jelectrocard.2003.09.017. [DOI] [PubMed] [Google Scholar]
  59. Vigmond EJ, Weber dos Santos R, Prassl AJ, Deo M, Plank G. Solvers for the cardiac bidomain equations. Prog Biophys Mol Biol. 2008;96:3–18. doi: 10.1016/j.pbiomolbio.2007.07.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Walker NL, Burton FL, Kettlewell S, Smith GL, Cobbe SM. Mapping of epicardial activation in a rabbit model of chronic myocardial infarction. J Cardiovasc Electrophysiol. 2007;18:862–868. doi: 10.1111/j.1540-8167.2007.00858.x. [DOI] [PubMed] [Google Scholar]
  61. Wharton JM, Richard VJ, Murry CE, Dixon EG, Reimer KA, Meador J, Smith WM, Ideker RE. Electrophysiological effects of monophasic and biphasic stimuli in normal and infarcted dogs. Pacing Clin Electrophysiol. 1990;13:1158–1172. doi: 10.1111/j.1540-8159.1990.tb02174.x. [DOI] [PubMed] [Google Scholar]
  62. Yao JA, Hussain W, Patel P, Peters NS, Boyden PA, Wit AL. Remodeling of gap junctional channel function in epicardial border zone of healing canine infarcts. Circ Res. 2003;92:437–443. doi: 10.1161/01.RES.0000059301.81035.06. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

tjp0590-4537-SD1.pdf (405.9KB, pdf)

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES