Skip to main content
NCBI home page
Elsevier Sponsored Documents logoLink to Elsevier Sponsored Documents
. 2015 Feb;12(2):397–408. doi: 10.1016/j.hrthm.2014.10.027

Fractionation of electrograms is caused by colocalized conduction block and connexin disorganization in the absence of fibrosis as AF becomes persistent in the goat model

Senthil Kirubakaran *,⁎, Rasheda A Chowdhury †, Mark CS Hall *, Pravina M Patel †, Clifford J Garratt *, Nicholas S Peters †
PMCID: PMC4315883  PMID: 25444850

Abstract

Background

Electrogram fractionation and atrial fibrosis are both thought to be pathophysiological hallmarks of evolving persistence of atrial fibrillation (AF), but recent studies in humans have shown that they do not colocalize. The interrelationship and relative roles of fractionation and fibrotic change in AF persistence therefore remain unclear.

Objective

The aim of the study was to examine the hypothesis that electrogram fractionation with increasing persistence of AF results from localized conduction slowing or block due to changes in atrial connexin distribution in the absence of fibrotic change.

Methods

Of 12 goats, atrial burst pacemakers maintained AF in 9 goats for up to 3 consecutive 4-week periods. After each 4-week period, 3 goats underwent epicardial mapping studies of the right atrium and examination of the atrial myocardium for immunodetection of connexins 43 and 40 (Cx43 and Cx40) and quantification of connective tissue.

Results

Despite refractoriness returning to normal in between each 4-week period of AF, there was a cumulative increase in the prevalence of fractionated atrial electrograms during both atrial pacing (control and 1, 2, and 3 months period of AF 0.3%, 1.3% ± 1.5%, 10.6% ± 2%, and 17% ± 5%, respectively; analysis of variance, P < .05) and AF (0.3% ± 0.1%, 2.3% ± 1.2%, 14% ± 2%, and 23% ± 3%; P < .05) caused by colocalized areas of conduction block during both pacing (local conduction velocity <10 cm/s: 0.1% ± 0.1%, 0.3% ± 0.6%, 6.5% ± 3%, and 6.9% ± 4%; P < .05) and AF (1.5% ± 0.5%, 2.7% ± 1.1%, 10.1% ± 1.2%, and 13.6% ± 0.4%; P < .05), associated with an increase in the heterogeneity of Cx40 and lateralization of Cx43 (lateralization scores: 1.75 ± 0.89, 1.44 ± 0.31, 2.85 ± 0.96, and 2.94 ± 0.31; P < .02), but not associated with change in connective tissue content or net conduction velocity.

Conclusion

Electrogram fractionation with increasing persistence of AF results from slow localized conduction or block associated with changes in atrial connexin distribution in the absence of fibrotic change.

Abbreviations: AERP, atrial effective refractory period; AF, atrial fibrillation; ANOVA, analysis of variance; CV, conduction velocity; Cx40, connexin 40; Cx43, connexin 43; MRI, magnetic resonance imaging; NPV, negative predictive value; PPV, positive predictive value

Keywords: Atrial fibrillation, Fractionation, Connexin, Conduction

Introduction

Wijffels et al1 initially demonstrated in a landmark article that artificial maintenance of atrial fibrillation (AF) in the chronically instrumented goat model leads to a shortening of the atrial effective refractory period (AERP) associated with an increase in arrhythmia stability and inducibility. However, the time course of AERP shortening (48–72 hours) differs from the time course of progression to persistent AF (>7 days), suggesting that additional myocardial changes are responsible for the delayed increase in AF stability.2 In that these changes are not of the action potential or its causative ion channels, they are likely to be structural. The mechanism of myocardial remodeling leading to persistence of AF and its prevention and treatment are a major focus in this field.

Early studies highlighted the association between electrogram fractionation in the ventricles and slow conduction and its relevance to arrhythmogenesis.3–5 Later, mapping studies in human atria demonstrated greater electrogram fractionation during persistent AF6,7 and, together with studies showing modest short-term results for catheter ablation of these complex fractionated atrial electrograms,8 supported the concept that areas of increased fractionation represent areas of slow atrial conduction that are critical to the maintenance of AF. Although a number of other causative mechanisms for atrial electrogram fractionation have since been described, including anchor points for reentry circuits, sites of high-frequency sources, and autonomic innervation,6,7,9,10 the findings of histopathological and imaging studies of atrial fibrosis in patients with persistent AF11 have supported the widely held concept that fibrosis causes slow and discontinuous atrial conduction and is therefore responsible for the observed increased electrogram fractionation and arrhythmia stability.12

However, recent clinical studies have shown that areas of fractionation during AF and atrial fibrosis detected using gadolinium-enhanced magnetic resonance imaging (MRI) do not colocalize,7,13 and hence not only do the relative roles of fractionation and fibrotic change, and their relationship, in AF stability and persistence remain unclear, but alternative contributory mechanisms of conduction slowing need be sought.

The aim of the study was to examine the hypothesis that electrogram fractionation with increasing persistence of AF results from localized conduction slowing and block in the absence of the need for fibrotic change, and resulting from changes in atrial connexin distribution.

Methods

Twelve adult female goats (mean weight 67 ± 8 kg) were used for the study. The experiments were conducted in accordance with a project license issued by the UK Home Office under the Animals (Scientific Procedures) Act 1986 and Directive 2010/63/EU of the European Parliament. The animals were allowed free access to food and water and were unrestrained in their pens throughout the experiment.

Study protocol

Each of the 12 animals had 2 pacing systems implanted as described previously.14 In this model, specially designed software in the pacemakers deliver 50 Hz of atrial burst pacing through the atrial lead in the right atrial appendage when 2 beats of sinus rhythm are detected; conversely, burst pacing is inhibited when AF is detected. Nine goats underwent maintenance of AF for up to 3 consecutive 1-month periods, each period separated by 1 week of sinus rhythm that was sufficient for electrical remodeling to reverse between periods of pacing-maintained AF. Measurements were made of AERP, AF cycle length, ventricular cycle length, and AF stability (duration of individual episodes of induced AF) at t = 0, 4, 8, 12, and 24 hours, twice daily until 120 hours, and then every 24 hours during each 1 month of burst pacing.14 AF cycle length was measured by direct recordings of the intracardiac atrial electrogram from the pacemaker from the lead in the right atrial appendage and therefore consistent within and between each goat. At the end of each 1-month period, 3 goats underwent epicardial mapping of the right atrium as described below. Atrial tissue samples were taken from the right atrial free wall and right atrial appendage for connexin 40 and 43 (Cx40 and Cx43) and histological examination.

Three control goats with pacemakers implanted and no atrial burst pacing underwent epicardial mapping 3 months after pacemaker implantation.

Pacemaker implantation

Animal preparation and epicardial mapping

At the end of each 1-month period, the goats were anesthetized and cardioverted to sinus rhythm. A thoracotomy was then made to expose the right atrial free wall. A rectangular array (53 mm × 21 mm) consisting of 112 unipolar electrodes with a 3.57-mm interelectrode separation was manually and consistently positioned on the right atrial free wall such that the sulcus terminalis, the visible epicardial line of the crista terminalis, traversed the center of the array, which therefore spanned right atrial myocardium only. Electrograms were acquired using a Cardiomapp system (Prucka Engineering) at a sampling rate of 1 kHz and band-pass filtering between 0.2 and 300 Hz during atrial pacing at a cycle length of 400 ms from the right atrial appendage and AF. AF was induced by the introduction of early atrial premature beats or by the initiation of atrial burst pacing (see Online Supplemental Methods).

During off-line analysis, for each mapping study, 5 consecutive beats of atrial pacing and 2 seconds of AF were selected for detailed analysis, creating a database totaling 28,958 atrial unipolar electrograms, which were examined manually. After mapping studies and while the animals were under general anesthesia, the hearts were explanted for tissue examination.

Epicardial right atrial mapping

Conduction velocity

Isochronal activation maps were constructed from the local activation times at each electrode during pacing from the right atrial appendage at a cycle length of 400 ms. Activation at an electrode was taken to coincide with the steepest negative deflection of the atrial unipolar electrogram, and atrial activation maps were constructed from the local activation times by drawing isochronal lines of 10-ms activation intervals.

Overall atrial conduction velocity was calculated from the time taken for a wavefront to travel across the mapping array relative to the distance traveled measured perpendicular to the isochrones.

Local atrial conduction velocity was calculated directly from the activation times assigned to each electrode using the triangulation method15,16 by which local velocities and vectors of conduction are estimated from the difference in activation times of each small triangle of electrodes (Figure 1). By using this method, we estimated a propagation vector and velocity on a millimeter scale, although precise myocardial conduction pathways at the cellular level cannot be precisely determined.

Figure 1.

Figure 1

Open in a new tab

Method for calculating local conduction velocity. Diagram shows activation times of 3 neighboring electrodes (T1, T2, and T3). Interval X can then be calculated using trigonometry. The conduction velocity in the direction of propagation would then be the interelectrode distance divided by interval X. AF = atrial fibrillation.

During AF, we used the same triangulation method to determine local velocities and vectors of conduction. During AF, wavefront collision within a triangle can occasionally result in falsely high apparent velocities; so, as in previous studies, apparent velocities of >160 ms were discarded.17

Electrogram morphology

In each goat, all the right atrial unipolar electrograms from 5 consecutive beats of atrial pacing at each electrode were analyzed and characterized into 3 groups (Figure 2)10: single electrograms exhibiting 1 negative deflection; double electrograms with 2 negative deflections, with the amplitude of the smaller deflection being at least 25% of the amplitude of the larger; and fractionated electrograms consisting of 3 or more negative deflections. Unipolar electrogram–based analysis was performed during AF, as previous studies have shown a clear relationship between atrial unipolar electrogram morphologies and atrial conduction patterns in goats and humans.10,18,19 Single electrograms were found to represent uniform fast atrial conduction, double electrograms represented areas of conduction block or areas of collision, and fractionated electrograms represented areas of slow atrial conduction and nonuniform anisotropic conduction.

Figure 2.

Figure 2

Open in a new tab

A: Consecutive right atrial unipolar electrogram recordings during mapping of the right atrium during atrial pacing at a cycle length of 400 ms. Representative electrograms are shown as single, double, and fractionated. B: Consecutive right atrial unipolar electrogram recordings during mapping of the right atrium during AF in representative goats from each group. Atrial electrograms during AF in the control goat had discrete sharp negative deflections, with goats in the remaining groups displaying variable morphologies. AF = atrial fibrillation.

The spatial distribution of single, double, and fractionated electrograms was mapped for each goat.

Atrial electrograms during AF were classified and mapped using the previous criteria into single, double, and fractionated for each goat during 2 seconds of AF.

Isochronal atrial activation mapping

We used Konings classification17 to define the complexity of atrial activation from the constructed isochronal maps. Type 1 AF was defined as a single broad wavefront propagating across the right atrium with only small areas of slow conduction not disturbing the main course of the fibrillation wave. Type 2 AF was defined as either 1 wavefront propagating with marked local conduction delay or 2 separate wavefronts present. Type 3 AF was defined as 3 or more fibrillation waves present. The proportion of type 1, 2, and 3 AF was calculated during 2 seconds of AF in each goat.

In the isochronal maps during pacing and AF, and in keeping with convention, conduction block was defined as local conduction velocities of <10 cm/s, combined with the discontinuity of conduction across the line of block.15,17 However, given the arbitrary nature of this convention, and the likelihood, variability, and relevance of propagation in partially excitable tissue in AF, which may manifest as slow rather than block to conduction, as well as mapping the distribution of conduction velocities <10 cm/s, we also mapped the distribution of conduction velocities <20 and <30 cm/s, which represent slow conduction.

Atrial myocardial architecture

Cx40 and Cx43 expression and distribution

Immunolabeling and Western blotting of Cx40 and Cx43 were performed using sections taken from the right atrial appendage, as described previously20 (see Online Supplemental Methods).

Cx40 and 43 distribution quantification

The whole area of the sections on the slide were viewed on an epifluorescence microscope and scored. Where the majority of the tissue showed normally distributed Cx43 (<10% of the sample showing lateralization), a score of 1 was given; a score of 5 was given if more than 40% of the sample showed lateralization (Figure 3; see Online Supplemental Methods).

Figure 3.

Figure 3

Open in a new tab

Tissue fiber orientation is indicated using double-headed dotted arrows. Examples of the Cx43 label in the region of the intercalated disc are shown by vertical arrows, and those of Cx43 along the lateral membrane are shown by horizontal arrows. A: In the image for score 1, the majority of Cx43 immunolabel is confined to the ends of the myocytes in the location of the intercalated disc. B: In the image for score 5, the immunolabel is found along the length of the lateral membrane. Cx43 = connexin 43.

Fibrosis

Elastin and collagen autofluoresce under 488-nm wavelength. Ten micrometer sections were fixed in frozen methanol for 5 minutes, Vectashield Mounting Media (Burlingame, CA, USA) and imaged by using confocal microscopy. The ImageJ program (National Institutes of Health, Bethesda, Maryland, USA) was used to quantify percentage of fluorescence in 6 random fields measuring 230.34 by 230.43 micrometers.

Hypertrophy

Ten micrometer sections underwent hematoxylin and eosin staining. The cell diameter of 20 cells was measured through the plane of the nucleus.

Statistical analysis

Results are expressed as mean ± SD. A nonparametric comparison between groups was performed using the Mann-Whitney U test, and parametric data were analyzed using the unpaired t test. Multiple groups were compared using 1-way analysis of variance (ANOVA). A 2-tailed P value of ≤.05 was considered statistically significant. Correlations between AF stability, connexin content and distribution, conduction velocities, electrogram morphology, and complexity of activation were analyzed using Pearson correlation test; a Pearson r value was obtained. The relationship between electrogram fractionation and areas of conduction block and slow conduction was expressed by the positive predictive value (PPV) and negative predictive value (NPV).

Results

All 12 goats successfully completed the protocol. Successive 1-month periods of AF resulted in a cumulative increase in AF stability (Figure 4), with no difference in the rate of decrease in AERP (time for AERP to decrease to half of nadir: first AF period 22.3 ± 4.9 hours, second AF period 16.6 ± 2.7 hours, and third AF period 18.7 ± 5.3 hours; P = .7) or the rate of decrease in AF cycle length. Ventricular cycle length during AF episodes remained constant (after 1 week 429 ± 39 ms, at the end of the first AF period 433 ± 25 ms, second AF period 478 ± 47 ms, and third AF period 458 ± 58 ms; P = 0.68).

Figure 4.

Figure 4

Open in a new tab

Changes in mean duration of burst pacing required to induce AF episodes. Results are expressed as mean values obtained from all goats. AF = atrial fibrillation; ANOVA = analysis of variance.

Epicardial right atrial mapping

During atrial pacing

The prevalence of fractionated atrial electrograms during atrial pacing showed a progressive increase from the control value (0.3%) with the increase in AF durations (1 month 1.3%; 2 months 10% ± 2%; and 3 months 17% ± 5%; ANOVA, P < .05), associated with a colocalized increase in the proportion of areas of conduction block and slow conduction by all definitions (conduction velocity [CV] <10 cm/s: control 0.1% ± 0.1%, first AF period 0.3% ± 0.6%, second AF period 6.5% ± 3%, and third AF period 6.9% ± 4%, P < .05; CV <20 cm/s: control 1.4% ± 0.5%, first AF month 5.4% ± 1.2%, second AF month 11% ± 4.3%, and third AF month 12% ± 5.4%, P < .05; CV <30 cm/s: control 3.6% ± 2.3%, first AF month 9.3% ± 6.2%, second AF month 13% ± 7.4%, and third AF month 15% ± 8.7%, ANOVA, P < .05).

However, there was no change in the time taken for activation to traverse the entire electrode array (the overall atrial conduction velocity: control 81.7 ± 11.7 cm/s, first AF period 92 ± 5.3 cm/s, second AF period 86.3 ± 18 cm/s, and third AF period 85 ± 12 cm/s; ANOVA, P = .78).

During AF

Figure 2B shows consecutive atrial electrograms recorded during AF in representative goats from each group. The prevalence of fractionated electrograms during AF showed a similar progressive increase with each month of AF (0.3% ± 0.1%, 2.3% ± 1.2%, 14% ± 2%, and 23% ± 3%; ANOVA, P < .05), associated with an increase in the proportion of areas of conduction block and slow conduction during AF (CV <10 cm/s: control 1.5% ± 0.5%, first AF month 2.7% ± 1.1%, second AF month 10.2% ± 1.2%, and third AF month 13.6% ± 0.4%; CV <20 cm/s: control 5.3% ± 3.2%, first AF month 6.3% ± 3.2%, second AF month 16.2% ± 5.6%, and third AF month 21.5% ± 12.4%; CV <30 cm/s: control 15.2% ± 5.4%, first AF month 23.2% ± 12.6%, second AF month 29.2% ± 8.8%, and third AF month 30.2% ± 11.2%; ANOVA, P < .05).

Spatial distribution of abnormal atrial electrograms and areas of localized conduction block and slow conduction

Figures 5 and 6 illustrate fragmentation maps of each goat during atrial pacing and AF, demonstrating the spatial distribution of single, double, and fractionated electrograms with areas of localized conduction block and slow conduction. Areas with electrogram fractionation was a predictor of regional conduction block and slow atrial conduction during atrial pacing (local CV <30 cm/s: PPV 0.87; NPV 0.89) and during AF (local CV <30 cm/s: PPV 0.61; NPV 0.81). After the third AF month, there appeared to be increased fragmentation and slow or blocked conduction in the middle of the array (Figures 5 and 6, in the region of the crista terminalis); however, this was not consistent in all the animals.

Figure 5.

Figure 5

Open in a new tab

Fragmentation maps of each goat, showing the distribution of single, double, and fractionated atrial electrograms and localized areas of conduction block and slow conduction during epicardial mapping of the right atrium during atrial pacing. AF = atrial fibrillation.

Figure 6.

Figure 6

Open in a new tab

Fragmentation maps of each goat, showing the spatial distribution of atrial electrogram morphology and localized areas of conduction block and slow conduction during epicardial mapping of the right atrium during AF. AF = atrial fibrillation.

Complexity of AF

Figure 7 shows examples of 10 consecutive isochronal maps of AF in a single representative goat from each group. In the control goat (goat 1), atrial activation during AF was uniform, with a single wave crossing from right to left of the mapping array with no areas of conduction block. After each 1-month period of AF, the activation patterns during AF became progressively more complex. After the second AF period (goat 7), several lines of conduction block were present, fractionating the wavefronts into multiple wavelets. There were infrequent “focal” areas of activation within the mapping array (beat 8). After the third AF period (goat 11), the atrial activation patterns became even more complex with large areas of conduction block, multiple wavelets present, and more apparently focal areas of activation (beats 1, 5, and 8).

Figure 7.

Figure 7

Open in a new tab

Isochronal maps of right atrial activation during AF in a representative goat from each group during 10 consecutive time frames. Isochronal lines are drawn at 10-ms intervals. Thick lines represent arcs of conduction block, and dashed lines represent arcs of collision. Arrows indicate the direction of activation. Asterisks represent areas of epicardial breakthrough. AF = atrial fibrillation.

Accordingly, large differences were observed in the classification type of atrial activation of the right atrium during AF. In control and after the first AF period, atrial activation was mostly by a single broad wavefront, with the proportion of type 1 activation being 77% ± 15% and 72% ± 1%, respectively (P = .73). However, after the second and third AF periods, the proportion of type 1 activation was significantly reduced (25% ± 5% and 0%, respectively), corresponding to a progressive increase in complexity and type 3 activation (control 0%, second AF period 33% ± 15%, and third AF period 83% ± 21%; ANOVA, P < .05).

Underlying atrial myocardial structural changes

Cell diameter increased progressively after each 1-month period of AF (control 17.0 ± 2.5 μm, first AF period 19.25 ± 2.8 μm, second AF period 20.1 ± 3.1 μm, and third AF period 23.9 ± 3.6 μm; P = .04). There was no detectable difference in the immunoquantification of Western blots of Cx40 or Cx43 (either total expression or functional docked and phosphorylated proportions), relative ratio of Cx40/(Cx40 + Cx40), or the degree of fibrosis at any time point. Consistent with previous reports,21 the Cx40 label was more inhomogeneous after 3-month period of AF compared to control, with an increase in the total area devoid of the label (52.1% ± 18.7% vs 24.2% ± 6.3%; P = .02) and in the variability between fields in the same animal (11.2% ± 1.8% vs 7.0% ± 2.4%; P = .03). Although Cx43 was homogeneously distributed across the tissue, after the first 1-month period of AF there was an increase in lateralization of Cx43 (control 1.8 ± 0.9, first AF period 1.4 ± 0.3, second AF period 2.9 ± 1.0, and third AF period 2.9 ± 0.3; P = .02), signifying a redistribution of Cx43 from the normal location of the intercalated discs to the lateral wall of the myocytes (Figure 3). Quantification of disc size and Cx43 content of the disc using en face images showed no differences in the expression of Cx43 in the intercalated disc.

Correlation between AF stability, distribution of Cx43, conduction block, electrogram morphology, and complexity of atrial activation during AF

Cx43 lateralization scores correlated with both the proportion of conduction block and atrial electrogram fractionation during both atrial pacing (r = 0.53; P < .05 and r = 0.42, P < .05, respectively) and AF (r = 0.59, P < .05 and r = 0.46, P < .05, respectively), as well as with complexity scores of AF activation (r = 0.68, P < .05; Figure 8).

Figure 8.

Figure 8

Open in a new tab

Correlation of Cx43 redistribution scores with both proportion of areas of conduction block fractionated electrograms during atrial pacing and AF, as well as with complexity scores of atrial activation during AF. AF = atrial fibrillation; Cx43 = connexin 43.

Discussion

The principal findings of this study are that increasing AF stability and the concurrent increase in the complexity of fibrillatory conduction are associated with an increase in atrial electrogram fractionation caused by colocalized areas of slow conduction and conduction block and an increase in the heterogeneity of Cx40 and lateralization of Cx43 in the absence of atrial fibrosis.

The functional association between the complexity of fibrillatory conduction and AF stability in the goat is in agreement with the study of Verheule et al.22 In this study, we have shown a redistribution of Cx40 and increased lateralization of Cx43.

Relationship between electrogram fractionation, conduction block, and connexin distribution

Although many studies, including previous clinical studies, have shown a similar consistency of fractionation during both pacing and AF, it is not a universal finding, and in a recent study by Jadidi et al,7 a high proportion of fractionated atrial electrograms were located in normal voltage areas, showing no correlation of fractionation during AF with that during sinus rhythm or coronary sinus pacing. Findings such as these indicate that the majority of fractionated potentials seen during AF in patients not only are functionally determined but also do not occur in significantly fibrotic myocardium. In the present study, the progressive increase in AF duration and fractionation with colocalization of conduction block was associated with the redistribution of Cx40 and increased lateralization of Cx43 in the absence of atrial fibrosis.

Previous studies that have variously and inconsistently focused on conduction, fractionation, myocardial architecture, but not all 3 simultaneously, have reported discrepant results,23 probably owing to differences in experimental models, duration of AF, and inclusion of different types of patients, with different clinical patterns of AF.

In the present study, we sought to unify measures of conduction, fractionation, and myocardial architecture, as well as the time course of their change in a single well-established model of AF, and have demonstrated that with increasing duration of AF, there is a progressive increase in the prevalence of atrial electrogram fractionation, similar to that observed in human AF.6,7 The greater prevalence of fractionation and conduction block during AF compared to relatively slow (400 ms) atrial pacing at all time points indicates that these phenomena are at least in part functionally determined. However, the fact that atrial electrogram fractionation was colocalized during slow atrial pacing and AF in each heart indicates that slow discontinuous conduction and regional conduction block are also in part secondary to an underlying fixed substrate (Figures 5 and 6).

In that this is the first study to examine changes in connexins in parallel with assessment of atrial conduction and electrogram morphology, although causation cannot be proved, this study provides evidence for gap-junctional remodeling as a potential mechanism for colocalized slow or blocked conduction and increased electrogram fractionation.

The effect of rapid atrial rates on redistribution of connexins in the chronic AF goat model has been studied previously by van der Velden et al.24 As in the present study, these authors identified AF-induced inhomogeneity of Cx40 distribution, with areas of low-density Cx40 located next to areas of higher (normal) density Cx40; these changes became progressively more pronounced with increasing durations of maintained AF up to a maximum at 16 weeks. These authors hypothesized that increasing AF stability could be due to the generation of small foci of intra-atrial conduction block. The present study has shown direct evidence for such foci of conduction block in conjunction with increased electrogram fractionation and also that this change correlates closely in time with the redistribution of Cx43. Lateralization of Cx43 has been postulated to be arrhythmogenic in a number of settings25 by promoting nonuniformity and anisotropy of cell-to-cell conduction and microscopic conduction block.

Recent findings have identified progressive electrical dissociation between the endocardium and the epicardium in the goat model and in humans26,27 and that in the diseased atrium there may be differential remodeling of Cx40 and Cx43 between the endocardium and the epicardium, associated with heterogeneous slow conduction and sustained atrial arrhythmias.28 Differential uncoupling and slowing of conduction between layers of the myocardium during burst pace–initiated AF may lead to electrical dissociation of the epi- and endocardial layers, which may have an important role in the substrate for arrhythmia persistence and stability.26,27

Other potential substrates responsible for altered atrial conduction

This and previous studies in the atrially burst–paced goat model have shown that the development of significant atrial fibrosis does not occur in this model,29 even when AF is maintained for several months,22 and that the absence of fibrotic change is consistent with many studies of human AF.20,30 The model used in the present study can be considered as a model of “lone” AF,1 distinct from models in which there is ventricular dysfunction and fibrosis.31 A study by Dosdall et al32 evaluated left ventricular function using cardiac MRI after 6 months of maintained AF in different animal models and found no significant change in left ventricular function in the goat model (in comparison to other animal models of AF). Although such studies have indicated that fibrosis plays a role both in the perpetuation of AF and in slowing of atrial conduction under conditions of ventricular myocardial disease and dysfunction,33 fibrosis is not an essential feature of stability of AF.

Although atrial fibrosis may be seen in patients with persistent AF,11,34 it is not a universal finding, and in particular its significance in the generation of fractionation seen during AF has been unclear. A recent study by Jadidi et al13 demonstrated that areas of fractionation observed during AF were not associated with areas of fibrosis or scarring detected using late gadolinium cardiac MRI in patients with persistent and long-standing AF. Our study supports these findings and offers an alternative mechanism for electrogram fractionation seen in patients with persistent AF.

Study limitations

Although this study has indicated a role of connexin redistribution in the self-perpetuation of AF and colocalization of atrial conduction block and areas of electrogram fractionation, causality cannot be proven in the absence of primary manipulation of connexin distribution. However, recent complementary work using adenoviral transfer of connexins in a porcine model has demonstrated that enhanced connexin expression can have substantial AF-suppressing activity,35 although the precise mechanisms by which this occurred were also uncertain.23

Although progressive electrical dissociation between the endocardium and the epicardium has been demonstrated in the goat model and in humans, our study was not designed to look specifically at connexin heterogeneity and lateralization in different myocardial layers.

Clinical Perspectives.

Electrogram fractionation and atrial fibrosis are hallmarks of evolving persistence of atrial fibrillation (AF). In the goat model of persistent AF, we show that electrogram fractionation results from colocalization of slow conduction and changes in atrial connexin distribution in the absence of fibrotic change. Currently, a class of antiarrhythmic peptides that act by increasing cardiac gap junction intercellular communication represents a potential new antiarrhythmic strategy in AF that has reached phase 2 clinical studies. This study adds to the impetus for gap-junctional coupling as an antiarrhythmic therapeutic target and further development of drugs that modulate gap-junctional function and organization.

Conclusion

Electrogram fractionation with increasing persistence of AF results from colocalization of atrial conduction block and slow conduction associated with changes in atrial connexin distribution in the absence of fibrotic change.

Acknowledgments

We thank Medtronic and Guidant for providing the pacemakers, leads and software. In addition we would like to acknowledge the ElectroCardioMaths Programme of the British Heart Foundation Centre at Imperial College and the NIHR Biomedical Research Centre Programme.

Footnotes

Dr Kirubakaran and Dr Chowdhury contributed equally as joint first authors, and Dr Garratt and Dr Peters contributed equally as joint last authors.

This work was supported and funded by the British Heart Foundation (grant nos. PG/03/135 and BHF RG/10/11/28457).

Appendix

Supplementary data

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.hrthm.2014.10.027.

Appendix

Supplementary data

Supplementary Material

mmc1.docx (16.5KB, docx)

References

  • 1.Wijffels M.C., Kirchhof C.J., Dorland R., Allessie M.A. Atrial fibrillation begets atrial fibrillation: a study in awake chronically instrumented goats. Circulation. 1995;92:1954–1968. doi: 10.1161/01.cir.92.7.1954. [DOI] [PubMed] [Google Scholar]
  • 2.Allessie M. The “second factor”: a first step toward diagnosing the substrate of atrial fibrillation? J Am Coll Cardiol. 2009;53:1192–1193. doi: 10.1016/j.jacc.2009.01.008. [DOI] [PubMed] [Google Scholar]
  • 3.Berbari E.J., Lazzara R., El-Sherif N., Scherlag B.J. Extracardiac recordings of His-Purkinje activity during conduction disorders and junctional rhythms. Circulation. 1975;51:802–810. doi: 10.1161/01.cir.51.5.802. [DOI] [PubMed] [Google Scholar]
  • 4.Klein H., Karp R.B., Kouchoukos N.T., Zorn G.L., Jr, James T.N., Waldo A.L. Intraoperative electrophysiological mapping of the ventricles during sinus rhythm in patients with previous myocardial infarction: identification of the electrophysiological substrate of ventricular arrhythmias. Circulation. 1982;66:847–853. doi: 10.1161/01.cir.66.4.847. [DOI] [PubMed] [Google Scholar]
  • 5.El-Sherif N., Scherlag B.J., Lazzara R., Hope R.R. Re-entrant ventricular arrhythmias in the late myocardial infarction period, 1: conduction characteristics in the infarction zone. Circulation. 1977;55:686–702. doi: 10.1161/01.cir.55.5.686. [DOI] [PubMed] [Google Scholar]
  • 6.Stiles M.K., Brooks A.G., Kuklik P. High-density mapping of atrial fibrillation in humans: relationship between high-frequency activation and electrogram fractionation. J Cardiovasc Electrophysiol. 2008;19:1245–1253. doi: 10.1111/j.1540-8167.2008.01253.x. [DOI] [PubMed] [Google Scholar]
  • 7.Jadidi A.S., Duncan E., Miyazaki S. Functional nature of electrogram fractionation demonstrated by left atrial high-density mapping. Circ Arrhythm Electrophysiol. 2012;5:32–42. doi: 10.1161/CIRCEP.111.964197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Nademanee K., McKenzie J., Kosar E., Schwab M., Sunsaneewitayakul B., Vasavakul T., Khunnawat C., Ngarmukos T. A new approach for catheter ablation of atrial fibrillation: mapping of the electrophysiologic substrate. J Am Coll Cardiol. 2004;43:2044–2053. doi: 10.1016/j.jacc.2003.12.054. [DOI] [PubMed] [Google Scholar]
  • 9.Zlochiver S., Yamazaki M., Kalifa J., Berenfeld O. Rotor meandering contributes to irregularity in electrograms during atrial fibrillation. Heart Rhythm. 2008;5:846–854. doi: 10.1016/j.hrthm.2008.03.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Konings K.T.S., Smeets J.L., Penn O.C., Wellens H.J., Allessie M.A. Configuration of unipolar atrial electrograms during electrically induced atrial fibrillation in humans. Circulation. 1997;95:1231–1241. doi: 10.1161/01.cir.95.5.1231. [DOI] [PubMed] [Google Scholar]
  • 11.Oakes R.S., Badger T.J., Kholmovski E.G. Detection and quantification of left atrial structural remodeling with delayed-enhancement magnetic resonance imaging in patients with atrial fibrillation. Circulation. 2009;119:1758–1767. doi: 10.1161/CIRCULATIONAHA.108.811877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Jacquemet V., Henriquez C.S. Genesis of complex fractionated atrial electrograms in zones of slow conduction: a computer model of microfibrosis. Heart Rhythm. 2009;6:803–810. doi: 10.1016/j.hrthm.2009.02.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Jadidi A.S., Cochet H., Shah A.J. Inverse relationship between fractionated electrograms and atrial fibrosis in persistent atrial fibrillation—a combined MRI and high density mapping. J Am Coll Cardiol. 2013;62:802–812. doi: 10.1016/j.jacc.2013.03.081. [DOI] [PubMed] [Google Scholar]
  • 14.Todd D.M., Fynn S.P., Walden A.P., Hobbs W.J., Arya S., Garratt C.J. Repetitive 4-week periods of atrial electrical remodeling promote stability of atrial fibrillation: time course of a second factor involved in the self-perpetuation of atrial fibrillation. Circulation. 2004;109:1434–1439. doi: 10.1161/01.CIR.0000124006.84596.D9. [DOI] [PubMed] [Google Scholar]
  • 15.Konings K.T.S., Dorland R., Mast F., Allessie M.A. Maastricht: Maastricht University; 1999. Mapping of Electrically Induced Atrial Fibrillation in Humans [PhD thesis] pp. 87–111. [Google Scholar]
  • 16.Kanagaratnam P., Rothery S., Patel P., Severs N.J., Peters N.S. Relationship between connexins and atrial activation during human atrial fibrillation. J Am Coll Cardiol. 2002;39:116–123. doi: 10.1016/s0735-1097(01)01710-7. [DOI] [PubMed] [Google Scholar]
  • 17.Konings K.T., Kirchhof C.J., Smeets J.R., Wellens H.J., Penn O.C., Allessie M.A. High-density mapping of electrically induced atrial fibrillation in humans. Circulation. 1994;89:1665–1680. doi: 10.1161/01.cir.89.4.1665. [DOI] [PubMed] [Google Scholar]
  • 18.Jacquemet V., Virag N., Ihara Z., Dang L., Blanc O., Zozor S., Vesin J.M., Kappenberger L., Henriquez C. Study of unipolar electrogram morphology in a computer model of atrial fibrillation. J Cardiovasc Electrophysiol. 2003;14:S172–S179. doi: 10.1046/j.1540.8167.90308.x. [DOI] [PubMed] [Google Scholar]
  • 19.Eckstein J, Maesen B, Linz D, Zeemering S, van Hunnik A, Verheule S, Allessie M, Schotten U. Time course and mechanisms of endo-epicardial electrical dissociation during atrial fibrillation in the goat. Cardiovasc Res 89:816–824. [DOI] [PubMed]
  • 20.Kanagaratnam P., Rothery S., Patel P., Severs N.J., Peters N.S. Relative expression of immunolocalized connexins 40 and 43 correlates with human atrial conduction properties. J Am Coll Cardiol. 2002;39:116–123. doi: 10.1016/s0735-1097(01)01710-7. [DOI] [PubMed] [Google Scholar]
  • 21.van der Velden H.M., van Kempen M.J., Wijffels M.C., van Zijverden M., Groenewegen W.A., Allessie M.A., Jongsma H.J. Altered pattern of connexin40 distribution in persistent atrial fibrillation in the goat. J Cardiovasc Electrophysiol. 1998;9:596–607. doi: 10.1111/j.1540-8167.1998.tb00940.x. [DOI] [PubMed] [Google Scholar]
  • 22.Verheule S., Tuyls E., van Hunnik A., Kuiper M., Schotten U., Allessie M. Fibrillatory conduction in the atrial free walls of goats in persistent and permanent atrial fibrillation. Circ Arrhythm Electrophysiol. 2010;3:590–599. doi: 10.1161/CIRCEP.109.931634. [DOI] [PubMed] [Google Scholar]
  • 23.Kato T., Iwasaki Y.K., Nattel S. Connexins and atrial fibrillation: filling in the gaps. Circulation. 2012;125:203–206. doi: 10.1161/CIRCULATIONAHA.111.075432. [DOI] [PubMed] [Google Scholar]
  • 24.van der Velden H.M., Ausma J., Rook M.B., Hellemons A.J., van Veen T.A., Allessie M.A., Jongsma H.J. Gap junctional remodeling in relation to stabilization of atrial fibrillation in the goat. Cardiovasc Res. 2000;46:476–486. doi: 10.1016/s0008-6363(00)00026-2. [DOI] [PubMed] [Google Scholar]
  • 25.Gutstein D.E., Morley G.E., Vaidya D., Liu F., Chen F.L., Stuhlmann H., Fishman G.I. Heterogeneous expression of gap junction channels in the heart leads to conduction defects and ventricular dysfunction. Circulation. 2001;104:1194–1199. doi: 10.1161/hc3601.093990. [DOI] [PubMed] [Google Scholar]
  • 26.Eckstein J., Coppini R., Verheule S., Allessie M.A., Schotten U. Three dimensional organization of atrial fibrillation: first insights from simultaneous epi-endocardial high density mapping in goats. Heart Rhythm. 2007;4:S175. [Google Scholar]
  • 27.de Groot N.M., Houben R.P., Smeets J.L., Boersma E., Schotten U., Schalij M.J., Crijns H., Allessie M.A. Electropathological substrate of longstanding persistent atrial fibrillation in patients with structural heart disease: epicardial breakthrough. Circulation. 2010;122:1674–1682. doi: 10.1161/CIRCULATIONAHA.109.910901. [DOI] [PubMed] [Google Scholar]
  • 28.Ryu K., Li L., Khrestian C.M., Matsumoto N., Sahadevan J., Ruehr M.L., Van Wagoner D.R., Efimov I.R., Waldo A.L. Effects of sterile pericarditis on connexins 40 and 43 in the atria: correlation with abnormal conduction and atrial arrhythmias. Am J Physiol Heart Circ Physiol. 2007;293:H1231–H1241. doi: 10.1152/ajpheart.00607.2006. [DOI] [PubMed] [Google Scholar]
  • 29.Ausma J., Wijffels M., Thone F., Wouters L., Allessie M., Borgers M. Structural changes of atrial myocardium due to sustained atrial fibrillation in the goat. Circulation. 1997;96:3157–3163. doi: 10.1161/01.cir.96.9.3157. [DOI] [PubMed] [Google Scholar]
  • 30.Todd D.M., Skanes A.C., Guiraudon G., Guiraudon C., Krahn A.D., Yee R., Klein G.J. Role of the posterior left atrium and pulmonary veins in human lone atrial fibrillation: electrophysiological and pathological data from patients undergoing atrial fibrillation surgery. Circulation. 2003;108:3108–3114. doi: 10.1161/01.CIR.0000104567.72914.BF. [DOI] [PubMed] [Google Scholar]
  • 31.Li D., Fareh S., Leung T.K., Nattel S. Promotion of atrial fibrillation by heart failure in dogs: atrial remodeling of a different sort. Circulation. 1999;100:87–95. doi: 10.1161/01.cir.100.1.87. [DOI] [PubMed] [Google Scholar]
  • 32.Dosdall DJ, Ranjan R, Higuchi K, Kholmovski E, Angel N, Li L, Macleod R, Norlund L, Olsen A, Davies CJ, Marrouche NF. Chronic atrial fibrillation causes left ventricular dysfunction in dogs but not goats: experience with dogs, goats, and pigs. Am J Physiol Heart Circ Physiol 305:H725–H731. [DOI] [PMC free article] [PubMed]
  • 33.Xu J., Cui G., Esmailian F., Plunkett M., Marelli D., Ardehali A., Odim J., Laks H., Sen L. Atrial extracellular matrix remodeling and the maintenance of atrial fibrillation. Circulation. 2004;109:363–368. doi: 10.1161/01.CIR.0000109495.02213.52. [DOI] [PubMed] [Google Scholar]
  • 34.Boldt A., Wetzel U., Lauschke J., Weigl J., Gummert J., Hindricks G., Kottkamp H., Dhein S. Fibrosis in left atrial tissue of patients with atrial fibrillation with and without underlying mitral valve disease. Heart. 2004;90:400–405. doi: 10.1136/hrt.2003.015347. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Igarashi T., Finet J.E., Takeuchi A., Fujino Y., Strom M., Greener I.D., Rosenbaum D.S., Donahue J.K. Connexin gene transfer preserves conduction velocity and prevents atrial fibrillation. Circulation. 2012;125:216–225. doi: 10.1161/CIRCULATIONAHA.111.053272. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary Material

mmc1.docx (16.5KB, docx)

RESOURCES