Longitudinal Dissociation Within the Posterior AV Nodal Input of the Rabbit
A Substrate for AV Nodal Reentry
Background—Longitudinal dissociation of an anatomic pathway into 2 electrophysiologically distinct pathways has been hypothesized as a basis for localized AV nodal reentry and supraventricular arrhythmias.
Methods and Results—Extracellular bipolar and intracellular microelectrodes were used to record activation in the superfused rabbit AV junction. A subset of rabbit hearts (n=19 of 72) demonstrated dissociation of the posterior AV nodal input into ≥2 functional pathways. Antegrade AH conduction was maintained by a pathway just inferior to the tendon of Todaro. Rate-dependent conduction block was observed in a second pathway just superior to the tricuspid annulus, allowing retrograde activation of the distal portion of the inferior posterior AV nodal input and leading to the formation of apparent “dead-end” pathways. The superior (antegrade) and inferior (retrograde) pathways were separated by a band of well-polarized but poorly excitable transitional cells. Additional decreases in the atrial cycle length progressively increased the AH interval, further delaying retrograde activation of the inferior pathway, and progressively moved the site of conduction block in the inferior pathway proximally, thus extending the length of the retrograde conduction pathway and allowing circus movement within the transitional cells of the posterior AV nodal connection.
Conclusions—Longitudinal dissociation within the posterior AV nodal input can give rise to localized reentry and AV nodal reentrant tachycardia.
Atrioventricular nodal reentrant tachycardias (AVNRTs) are a common cause of supraventricular arrhythmia. Electrode catheter recordings and atrial pacing demonstrate dual AV conduction pathways in a majority of patients with this arrhythmia.1 2 The cellular electrophysiology, tissue composition, and anatomic location of the dual AV pathways are less certain.
Early evidence for a dual AV transmission system as a basis for reentrant arrhythmia was provided by Moe et al3 in the canine heart and by Mendez and Moe4 in the superfused rabbit AV junction. Both experiments demonstrated 2 functional pathways communicating with a final common pathway above the HB. The presumed site of longitudinal dissociation and reentrant atrial echo beats was the upper AV node. Further experiments using the superfused rabbit AV junction have demonstrated echo beats and even sustained supraventricular arrhythmia within the posterior-inferior right atrium, with the reentrant pathway contained either within or partially outside of Koch’s triangle.5 6 7 8 Little information concerning the anatomic site and functional basis for echo beats and sustained supraventricular arrhythmia has been provided, a result of a complicated 3-dimensional anatomy and the absence of an accepted definition for the dimensions and location of the AV node. Both the mechanism and the site for AVNRT in humans have been presumed to be intra-AV nodal for 20 to 30 years, with a paucity of clinical data supporting this hypothesis.9 10 11
Clinical concepts of the basis for AVNRTs have recently been rent asunder by electrode catheter recordings from the posterior-inferior right atrium and localized radiofrequency lesioning.9 10 11 Consistent electrograms can be obtained from tissues remote from the compact AV node, demonstrating decremental conduction and/or block in association with AVNRT. Radiofrequency lesions at critical extranodal sites eradicate 1 excitation pathway without abolishing AV transmission. Atrial extrastimuli no longer initiate sustained AVNRT, and the clinical AVNRT is eliminated.
During the course of study of AV transmission in 72 superfused rabbit AV preparations, 19 preparations demonstrated longitudinal dissociation within transitional cells comprising the posterior input of the compact AV node, commonly termed the slow pathway. Rate-dependent longitudinal dissociation within posterior transitional cells comprising input to the compact AV node was the basis for antegrade and retrograde activation pathways leading to apparent “dead-end” pathways. With additional rate-dependent delays, however, localized reentry in these dissociated pathways resulted in single echo beats and supraventricular tachycardia.
Adult rabbits of both sexes weighing 2.5 to 4 kg were anesthetized with intravenous sodium pentobarbital (30 to 40 mg/kg). The heart was removed and rinsed in modified Tyrode’s solution (in mmol/L: NaCl 130, KCl 4.05, MgCl2 1.0, NaHCO3 20, NaH2PO4 1.0, glucose 5.5, and CaCl2 1.35) bubbled with 0.95 O2/0.05 CO2 at 36°C to 37°C. The AV junction (containing the entire right atrium and interatrial septum and the base of the posterior right ventricle and interventricular septum) was exposed (Hoffman et al12). The tissue was pinned to a 20-mL Lucite chamber and superfused at 20 mL/min with modified Tyrode’s solution. Bipolar electrode recordings (0.10-mm-diameter Teflon-coated silver wires, 1 mm apart) were obtained from the following locations: (1) inferior to the tendon of Todaro and superior to the tricuspid annulus midway between the coronary sinus os and apex of the triangle of Koch (posterior input to the compact AV node), commonly termed the slow pathway [SP]); (2) halfway between the fossa ovalis and the apex of the triangle of Koch along the anterior limbus of the fossa ovalis (anterior input to the compact AV node), commonly termed the fast pathway [FP]); and (3) from the HB anterior to the central fibrous body. Pacing was performed from the high right atrium, the base of the crista terminalis, or the HB with 2-ms-duration square wave pulses at 1 to 3 times middiastolic threshold voltage.
Bipolar electrograms were individually amplified and filtered at 1 to 5000 Hz. Intracellular recordings were obtained with conventional 10- to 25-MΩ resistance glass microelectrodes filled with 3 mol/L KCl and a World Precision Instruments model 773 electrometer. A custom-made differentiator with linear peak-and-hold capability over a range of 10 to 800 V/s was used to determine dV/dtmax. Permanent records of intracellular and extracellular recordings were obtained with a Gould Windograf recorder. Electrophysiological measurements were determined from recordings taken at a paper speed of 100 mm/s. Individual microelectrode recordings were obtained during incremental atrial pacing. Composite illustrations were reconstructed from identical pacing protocols producing identical bipolar electrogram activation patterns in 19 rabbit preparations that demonstrated longitudinal dissociation and in 10 normal rabbit AV preparations.
Data are expressed as mean±SEM. Differences within groups were determined by an ANOVA followed by Scheffé’s test. Differences between normal and dissociated posterior AV nodal input properties were determined by Student’s t test for unpaired data. The criterion for significance was P≤0.05.
Rate-dependent longitudinal dissociation within the posterior AV nodal input was observed in 19 of 72 consecutive rabbit hearts. Dead-end pathways13 14 15 (transitional cells contained within Koch’s triangle having summed antegrade and retrograde conduction times >120% of the AH conduction time) were observed in 3 preparations at a spontaneous cycle length of 497±23 ms. These dead-end pathways provided the substrate for echo beats and nonsustained AVNRT when subjected to rapid atrial pacing.
SP Gradient Between the Tendon of Todaro and the Tricuspid Annulus
The region between the tendon of Todaro (SP-1) to the tricuspid annulus (SP-5) was arbitrarily divided into 5 evenly spaced bands. Intracellular potentials were recorded from each tissue band during incremental high right atrial pacing. At paced cycle lengths >600 ms, uniform activation from the tendon of Todaro to the valve annulus (SP-1 to SP-5) was observed. At shorter cycle lengths, 2 different conduction patterns were apparent: (1) a uniform Wenckebach cycle length for transitional cells of the posterior AV nodal input (Figure 1⇓) (normal SP) and (2) dissociation of the SP into 2 or more pathways with different Wenckebach cycle lengths (Figure 2⇓) (dissociated SPs). In each experiment, only 1 of the pathways maintained HB activation: the superior pathway just inferior to the tendon of Todaro (n=17) or the inferior pathway immediately superior to the tricuspid annulus (n=2). Pacing from the base of the crista terminalis did not alter dissociation within the posterior AV nodal input versus high right atrial pacing.
The AH Wenckebach cycle length did not differ in rabbit hearts with normal (205±8 ms; n=10) and dissociated (207±11 ms; n=19) posterior AV nodal inputs. In addition, the cycle length that produced 2:1 block within transitional cells of the anterior AV nodal input (FP) was not different for hearts with normal (132±3 ms; n=10) and dissociated (134±8 ms; n=19) posterior AV nodal inputs.
In the normal posterior AV nodal input, a gradient of decreasing resting potential, action-potential amplitude, and dV/dtmax was present from the tendon of Todaro (SP-1) to the tricuspid annulus (SP-5) (Table⇓). A different pattern was observed with longitudinal dissociation (Table⇓). A decrease in resting potential, action-potential amplitude, and dV/dtmax was observed from SP-1 to SP-3. Cells near the valve annulus were hyperpolarized, with an increasing gradient of action-potential amplitude and dV/dtmax observed from SP-3 to SP-5 (Table⇓). An inexcitable tissue band (SP-3 or SP-4) was present when longitudinal dissociation produced dual pathways with different Wenckebach cycle lengths (superior to SP-3=227±16 ms versus inferior to SP-3=327±23 ms; P=0.00001).
Dissociated Posterior AV Nodal Inputs at Slow Heart Rates
Two distinct deflections in the local bipolar electrogram were observed in 9 of 19 preparations and correlated temporally with intracellular recordings from the 2 pathways. An example is shown in Figures 3⇓ and 4⇓. Microelectrode and extracellular recordings are shown at paced cycle lengths of 450 and 167 ms for sites SP-2, SP-3, SP-5 proximal (P), SP-5 mid (M), SP-5 distal (D), midposterior AV nodal input (BP-SP), HB (BP-HB), and anterior AV nodal input (BP-FP) (Figure 3⇓). The locations of the recording and stimulation sites are also shown. At a cycle length of 450 ms (Figure 3A⇓), the posterior AV nodal input near the valve annulus was activated retrogradely after HB activation and is reflected as a deflection in the local electrogram. With a decrease in the paced cycle length to 167 ms (Figure 3B⇓), the site of block moved posteriorly along the valve annulus to site SP-5M. SP-3 demonstrated only electrotonic interactions with the 2 SPs. An additional decrease in the cycle length to 150 ms (Figure 4A⇓) produced 5:4 Wenckebach in SP-2, mediating 5:4 AH conduction. Progressive conduction block (beats 1, 2, 3, and 4) and then a reentrant beat were observed at SP-5P (beat 5). With a further decrease in the cycle length to 134 ms (Figure 4B⇓), 3:2 Wenckebach block at SP-2 mediated 3:2 AV conduction. Retrograde activation of SP-5, SP-5M, and SP-5P is not observed in Figure 4B⇓ because AH conduction block prevented retrograde reexcitation of the inferior pathway.
Activation of Apparent Dead-End Pathways
Post-Hisian activation of cells near the tricuspid annulus (SP-5) was recorded during sinus rhythm (n=3) and high right atrial pacing at cycle lengths ≥300 ms (n=5). The location of the cells along the valve annulus posterior to the central fibrous body is consistent with type B dead-end pathways.13 Extracellular potentials were observed coincident with intracellular dead-end recordings (n=2).
In Figure 5⇓, dead-end pathway cell activation correlated with an extracellular potential recorded in the HB electrogram. At cycle lengths ≥275 ms, the intracellular (SP-5) and extracellular potentials occurred before AV nodal and HB activation. An abrupt change from pre-Hisian/pre-AV nodal activation to post-Hisian/post-AV nodal activation occurred at a cycle length of 250 to 275 ms (Figure 5⇓). Conduction block 2:1 of the dead-end action potential occurred at a longer cycle length (200 ms) than that which was necessary to produce 2:1 A-H conduction block (155 ms) (Figure 5⇓). When a 1- to 2-mm-long incision perpendicular to the posterior AV nodal input was made immediately superior to the tricuspid annulus (3 to 4 mm posterior to the central fibrous body) (Figure 6⇓, bottom), activation of the dead-end pathway at cycle lengths >275 ms was observed only subsequent to HB activation (Figure 6⇓). In Figure 7⇓, the response of an apparent dead-end cell to premature stimuli is shown. Before the inferior incision, activation occurred before HB activation at a basic cycle length of 400 ms. Extrastimuli <275 ms shifted dead-end cell activation. After the incision, activation occurred only after AV nodal (Figure 6⇓) and HB activation.
Rate-Dependent Longitudinal Dissociation Within the Posterior AV Nodal Input: A Basis for Localized Reentry
Rate-dependent longitudinal dissociation within the posterior AV nodal input (Figure 2⇑) led to localized reentry incorporating the compact AV node (n=9). Antegrade conduction to the compact AV node at fast atrial pacing rates (n=8) was maintained by a pathway immediately inferior to the tendon of Todaro. Rate-dependent block was observed in a pathway located immediately superior to the tricuspid annulus. With AH prolongation, retrograde activation was then observed immediately superior to the valve annulus. The AH interval for the antegrade limb was 117±14 ms, with an HA interval of 96±12 ms for the retrograde limb (n=8). In only 1 example was antegrade AV nodal activation maintained in the pathway adjacent to the tricuspid annulus and rate-dependent block with retrograde activation observed in the pathway adjacent to the tendon of Todaro. Rate-dependent block in the superior pathway and antegrade conduction in the inferior pathway produced reentrant beats with early activation of the anterior AV nodal input (FP) (AH=123 ms, HA=15 ms; not shown).
An example of localized reentry incorporating dissociated posterior AV nodal inputs and the compact AV node is shown in Figure 8⇓. SP-2 served as the antegrade conduction pathway. SP-4 and SP-5 served as the retrograde conduction pathway. SP-3 (Figure 9⇓) remained well polarized but inexcitable except via intracellular stimulation. Conduction through the posterior input remained associated during the first 2 beats of the pacing sequence (Figure 8⇓). The third beat of the pacing sequence blocked between the distal site SP-4D and the more proximal site SP-4P, and site SP-4D was then retrogradely activated, after SP-4P and HB activation. With the sixth and seventh beats of the pacing train, the site of block moved retrogradely to SP-4P, as evidenced by dual activation of the cell by antegrade and retrograde wave fronts. With the eighth beat, SP-4D activation preceded SP-4P activation. Cessation of pacing after the ninth paced beat produced a reentrant SP beat as seen in the action potential from SP-2 (but not in extracellular recordings). The delay necessary to produce reentry with dissociated SPs resulted from a rate-dependent slowing of conduction within the antegrade SP, slowing of conduction within the compact AV node, and lengthening of the reentrant pathway by posterior movement of the site of conduction block within the retrograde activation pathway. Reactivation of the proximal superior (antegrade) pathway by retrograde activation of the inferior pathway occurred before activation of cells at the base of the crista terminalis (n=5 of 5 preparations tested), which suggests that the turnaround for the reentrant pathway occurred anterior to the coronary sinus os.
The response of the dissociated SP to premature beats is shown in Figures 10⇓ and 11⇓. In Figure 10⇓ (top), a premature stimulus is introduced that splits the local posterior AV input electrogram into 2 components, prolonging the AH interval and producing a nonstimulated atrial beat. Multiple impalements along the tricuspid annulus (inferior pathway) demonstrated a decrementing retrograde impulse that reexcited the SP only and failed to reexcite atrial tissue (Figure 11⇓). Conduction block in the compact AV node (no HB activation) prevented retrograde activation (Figure 10⇓, bottom). A 2-mm transection across the inferior posterior AV nodal input (immediately superior to the valve annulus) prevented echo beats (n=8), nonsustained AVNRT (n=5), and sustained AVNRT (n=2) in 8 of 8 AV preparations.
An example of sustained AVNRT is shown in Figure 12⇓. Intracellular recordings are shown for 5 sites, in conjunction with bipolar recordings from the anterior AV nodal input, posterior AV nodal input, and HB. Double potentials representing antegrade and retrograde conduction pathways are a common finding in microelectrode recordings from the posterior AV nodal input during sustained reentry. The smaller and slower upstroke represents the distant wave front, with the rapid upstroke representing local activation. A poorly excitable tissue band with 2 slow upstrokes representing distant antegrade and retrograde activation wave fronts is also present in Figure 12⇓. The AH interval during sustained tachycardia was 90 ms, and the HA interval was 87 ms. Activation times were measured relative to the atrial electrogram. Conduction from SP-5P to SP-5D represents retrograde conduction down the inferior pathway. The close timing of the antegrade and retrograde wave fronts in the SP-2P and SP-5P electrograms suggests that these sites are near the posterior turnaround for the tachycardia, whereas the prolonged separation of the SP and SP′ electrograms of the bipolar recording suggests a site nearer the center of the reentrant loop. The activation times inferior to the coronary sinus os (23 ms) and within the proximal antegrade pathway are similar, which suggests that activation inferior to the sinus os is a spur from the reentrant pathway.
Reentry in the AV Junction
Single reentrant beats, nonsustained AVNRT, and sustained AVNRT were observed in the superfused rabbit AV preparation. Localized reentry was dependent on the following: (1) longitudinal dissociation and the formation of 2 functional pathways within the posterior AV nodal input, (2) block within a refractory pathway and maintenance of AH conduction in a second pathway, (3) retrograde conduction in a long refractory period pathway, and (4) reexcitation of the proximal posterior AV nodal input. Conduction block terminating reentry was observed in the compact AV node, the retrograde conduction pathway, or the antegrade conduction pathway. A band of well-polarized but poorly excitable transitional cells (SP-3) was observed that separated the 2 dissociated pathways. The reentrant pathway appeared to be contained anterior to the coronary sinus os.
Sustained AVNRT (>30 seconds) was observed 8 times in 3 rabbit AV preparations. Sustained AVNRT could not be reproducibly initiated or terminated, which prevented a systematic study of AVNRT initiation. All 3 rabbit AV preparations that generated sustained AVNRT also demonstrated longitudinal dissociation within the posterior AV nodal input and single reentrant beats or nonsustained AVNRT.
Rate-dependent longitudinal dissociation has been a frequently proposed mechanism for dual AV transmission and AVNRTs in the rabbit,4 5 canine,3 and human heart.1 2 Initial demonstrations of longitudinal dissociation3 4 14 and AVNRT5 6 in the rabbit were assumed to be limited to the AV node. Each study, however, included AN (transitional) cells, ≥4 mm posterior to the compact AV node, as a part of the AV node. A minimal mass of atrial myocardium was also suggested as a necessary component for sustained AVNRT.6
Subsequent studies with superfused rabbit AV preparations by Watanabe and Dreifus15 and Mazgalev and coworkers7 8 recognized the role of perinodal fibers in the formation of echo beats and sustained AVNRT. Mazgalev and coworkers7 8 also described an obligatory role of atrial myocardium superior to the coronary sinus os and tendon of Todaro in AVNRT, stating “in this preparation, AVN reentry circuits always appeared to involve atrial tissue surrounding AVN … Remarkably, reentry confined to the intranodal region was never observed.” The reentrant pathways outlined8 9 contrast with the present studies in which antegrade and retrograde conduction limbs remained inferior to the tendon of Todaro and anterior to the coronary sinus os. As evidenced in Figure 10⇑, a single reentrant echo beat could be observed within the posterior AV nodal input, without right atrial activation. AVNRT, however, demonstrated 1:1 capture of the right atrium.
In the present studies, we distinguish between the compact AV node (knoten) described by Tawara16 and posterior transitional (AN) cells otherwise termed the open node14 or the posterior AV nodal input.17 Dissociation of the posterior AV nodal input providing retrograde and antegrade conduction pathways was observed within transitional (AN) cells. The compact AV node (N cells) was a necessary component of the reentrant circuit but did not need to demonstrate dual-pathway electrophysiology. The posterior AV nodal input was longitudinally divided by a band of well-polarized but poorly excitable transitional cells (SP-3). This is a salient finding because it provides a basis for the division of the posterior AV nodal input into 2 functional pathways. With an anterior or posterior extension to the line of block, the substrate within the posterior AV nodal input could produce early or late activation of the anterior AV nodal input and reproduce AVNRT termed slow/slow, fast/slow, or slow/fast. In subsequent experiments, we have observed 13 rabbit preparations that used the inferior pathway as an antegrade pathway, with rate-dependent block and retrograde conduction in the superior pathway, producing sustained AVNRT with early activation of the anterior AV nodal input (FP). Atrial myocardium was not necessary for single echo beats, although 1:1 atrial capture was observed for both nonsustained and sustained AVNRT. High right atrial stimuli failed to reset the tachycardia, although resetting was observed with stimuli introduced into the posterior AV nodal input (not shown).
Cellular Bases for Longitudinal Dissociation and Functional Block
Conduction block in dissociated posterior AV nodal inputs occurs at paced heart rates inconsistent with action-potential duration in transitional (AN) cells of the posterior input. Preferential block in the inferior pathway is not provided by reductions in dV/dtmax or action-potential amplitude or by prolongation of the action potential (Table⇑) compared with tissue adjacent to the tricuspid annulus in normal SPs. Instead, in the inferior pathway, there is a more rapid loss of source current (dV/dtmax and action-potential amplitude) with incremental pacing. Rate-dependent dissociation of cellular activation/uncoupling of cellular activation (as observed in the HB after anterior septal coronary artery ligation18) is capable of providing for both a decrease in dV/dtmax at decreasing paced cycle lengths and a progressive movement of the site of conduction block from distal to proximal in the inferior pathway.
The type and extent of intercellular connections in transitional cells of the posterior AV nodal input are unknown. Only 25% of transitional cells in the posterior AV nodal input respond to a maximal 5-μA, 4-ms duration intracellular stimulus (unpublished experiments). dV/dtmax and action-potential amplitude for individual transitional cells in associated pathways is identical when stimulated longitudinal to fiber orientation (42±5 V/s, 74±4 mV) or transverse to fiber orientation (42±5 V/s, 74±4 mV) (unpublished observations).
The posterior input as described by Tawara16 consists of an anterior “knoten” and a posterior input comprising small cells running in small bundles separated by connective tissue and gradually merging with atrial myocardium.14 16 17 Racker19 describes a proximal AV nodal bundle, “small fascicles that ran parallel to the AV ring and are contiguous with the AV node,” separate from atrial myocardium. The parallel bundles and physical separation of cells could lead to poor anatomic and physiological coupling of individual lateral muscle bundles and fascicles, potentiating dissociation parallel to fiber axis. The parallel fiber orientation in the posterior AV nodal input close to the tricuspid annulus has also been postulated as an AV nodal bypass tract20 and has been suggested as an anatomic basis for dead-end pathways.14
The present studies suggest that dead-end pathways13 14 (transitional cells having summed antegrade and retrograde conduction times >120% of AH conduction time) share a functional electrophysiological origin with rate-dependent longitudinal dissociation and reentry. Dead-end pathways may be present at very slow rates such that rate dependence remains unrecognized. Each of the superfused rabbit AV preparations that demonstrated dead-end pathways at spontaneous rates also demonstrated reentrant echo beats (n=3) or nonsustained AVNRT (n=1) with high right atrial pacing.
Three different forms of AVNRT can be demonstrated in humans: slow/fast, fast/slow, and slow/slow. The location of the reentrant pathway and the presence of multiple SPs described for slow/slow AVNRT in humans21 are mimicked in the present experiments that used the superfused rabbit AV preparation. Both the antegrade (AH) and retrograde (HA) limbs of echo beats and nonsustained AVNRT of the present experiments are prolonged. Dual SP electrograms were present in a majority of the rabbit AV preparations that demonstrated AVNRT. The failure to observe FP block in association with the initiation of echo beats or nonsustained tachycardia does not suggest a requirement for block in the anterior input. We22 have observed examples of sustained AVNRT consistent with fast/slow (n=4) and slow/fast (n=11) AVNRTs in the same experimental preparation, using the same reentrant circuit, with clockwise rather than counterclockwise activation of 2 dissociated pathways within the posterior input. In 1 preparation, both slow/slow (counterclockwise) and slow/fast (clockwise) AVNRTs could be observed, with a reversal of dissociated antegrade and retrograde SPs for the 2 AVNRTs.
The reentrant pathways in the present experiments had a long axis of 1.0 to 1.2 cm and were contained within the posterior AV nodal input, anterior to the coronary sinus os. The reentrant circuits were consistent with sustained AVNRT previously described by Janse et al5 and Wit et al6 but clearly differed from Mazgalev et al7 and Iinuma et al.8 The larger (incorporating the coronary sinus os and lying superior to the tendon of Todaro) and slower (cycle lengths >250 ms) reentrant circuits described by Mazgalev et al7 and Iinuma et al8 appear to be more consistent with the clinical entity described as slow/slow AV nodal tachycardia despite the presence of both limbs of the reentrant pathway within the posterior AV input (SP and SP′), antegrade AV conduction down the physiological SP, and the absence of involvement of the anterior AV nodal input in initiating or sustaining AVNRT in the present experiments.
This research study was supported by a grant from the American Heart Association, Oklahoma Affiliate, and research funds from the Department of Veterans Affairs. The authors would like to thank Dr Ralph Lazzara for his review of the manuscript and his helpful critique.
- Received March 31, 1998.
- Revision received August 19, 1998.
- Accepted September 2, 1998.
- Copyright © 1999 by American Heart Association
Denes P, Dhingra RC, Chuquimia R, Rosen KM. Demonstration of dual A-V nodal pathways in patients with paroxysmal supraventricular tachycardia. Circulation. 1973;48:549–555.
Wu D, Denes P, Dhingra R, Wyndham C, Rosen KM. Determinants of fast and slow pathway conduction in patients with dual atrioventricular nodal pathways. Circ Res. 1975;36:782–790.
Moe GK, Preston JB, Burlington H. Physiologic evidence for a dual A-V transmission system. Circ Res. 1956;4:357–375.
Mendez C, Moe GK. Demonstration of a dual A-V nodal conduction system in the isolated rabbit heart. Circ Res. 1966;19:378–393.
Janse MJ, van Capelle FJL, Freud GE, Durrer D. Circus movement with the AV node as a basis for supraventricular tachycardia as shown by multiple microelectrode recording in the isolated rabbit heart. Circ Res. 1971;28:403–414.
Wit AL, Goldreyer BN, Damato AN. An in vitro model of paroxysmal supraventricular tachycardia. Circulation. 1971;43:862–875.
Mazgalev T, Dreifus LS, Bianchi J, Michelson EL. The mechanism of AV junctional reentry: role of the atrionodal junction. Anat Rec. 1988;201:179–188.
Jackman WM, Beckman KJ, McClelland JH, Wang X, Friday KJ, Roman CA, Moulton KP, Twidale N, Hazlitt A, Prior MI, Oren J, Overholt ED, Lazzara R. Treatment of supraventricular tachycardia due to atrioventricular nodal reentry by radiofrequency catheter ablation of slow-pathway conduction. N Engl J Med. 1992;327:313–318.
Haissaguerre M, Gaita F, Fisher B, Commenges D, Montserrat P, d’Ivernois C, Lemetayer P, Warin J-F. Elimination of atrioventricular nodal reentrant tachycardia using discrete slow potentials to guide application of radiofrequency energy. Circulation. 1992;85:2162–2175.
Kay GN, Epstein AE, Dailey SM, Plumb VJ. Selective radiofrequency ablation of slow pathway for the treatment of atrioventricular nodal reentrant tachycardia: evidence for involvement of perinodal myocardium within the reentrant circuit. Circulation. 1992;85:1675–1688.
Hoffman BF, Paes de Carvalho A, DeMello WC, Cranefield PF. Electrical activity of single fibers of the atrioventricular node. Circ Res. 1959;7:11–18.
van Capelle FJL, Janse MJ, Varghese PJ, Freud GE, Mater C, Durrer D. Spread of excitation in the atrioventricular node of isolated rabbit hearts studied by multiple microelectrode recording. Circ Res. 1972;31:602–616.
Anderson RH, Janse MJ, van Capelle FJL, Billette J, Becker AE, Durrer D. A combined morphological and electrophysiological study of the atrioventricular node of the rabbit heart. Circ Res. 1974;35:909–922.
Tawara S. Das Reizleitungssytem des saugetier Herzens. Jena, Germany: Gustav Fischer; 1906.
Scherlag BJ, Patterson E, Nakagawa H, Hirao K, Jackman WM, Lazzara R. Changing concepts of A-V nodal conduction: basic and clinical correlates. Prim Cardiol. 1995;21:13–24.
Gonzalez MD, Scherlag BJ, Mabo P, Lazzara R. Functional dissociation of cellular activation as a mechanism of Mobitz type II atrioventricular block. Circulation. 1993;87:1389–1398.
James TN. Anatomy of the cardiac conduction system in the rabbit. Circ Res. 1967;20:638–648.
Jackman WM, Nakagawa H, Heidbuchel H, Beckman K, McClelland J, Lazzara R. Three forms of atrioventricular nodal (junctional) tachycardia. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. Philadelphia, Pa: WB Saunders Co; 1995.
Patterson E, Scherlag BJ. Differential excitation patterns producing sustained slow-slow, slow-fast, and fast-slow patterns of AV nodal reentrant tachycardia in the rabbit. Pacing Clin Electrophysiol. 1998;21:903A. Abstract.