New Insights Regarding the Atrial Flutter Reentrant Circuit
Studies in the Canine Sterile Pericarditis Model
Background—We studied atrial activation during induced atrial flutter in the canine sterile pericarditis model to test the hypothesis that the atrial flutter reentrant circuit includes a septal component.
Methods and Results—We studied 10 episodes of induced, sustained (>5 minutes) atrial flutter in 9 dogs. In all episodes, the reentrant circuit included a septal component. In 6 episodes, there were 2 reentrant circuits, one in the right atrial free wall and the second involving the atrial septum, Bachmann’s bundle, and the right atrial free wall; both circuits shared a pathway in the right atrial free wall (figure-of-eight). The direction (superior or inferior) of the septal wave front of the second circuit correlated with the direction (clockwise or counterclockwise, respectively) of the right atrial free-wall circuit. A line of functional block in the right atrial free wall was part of both reentrant circuits. In the other 4 atrial flutter episodes, only 1 reentrant circuit was present, with activation in an inferior-to-superior direction in the septum and a superior-to-inferior direction in the right atrial free wall in 2 episodes and in the opposite direction in the other 2 episodes. In all atrial flutter episodes, the flutter wave polarity in ECG lead II was determined by the direction of activation in the left atrium; polarity was positive when the direction was superior to inferior and negative when the direction was inferior to superior.
Conclusions—In this model of atrial flutter, the reentrant circuit (1) always included a septal component, (2) did not always require a right atrial free-wall reentrant circuit, (3) demonstrated figure-of-eight reentry when a reentrant circuit was present in the right atrial free wall, and (4) was associated with a line of functional block in the right atrial free wall.
We previously reported mapping studies of induced atrial flutter in the canine sterile pericarditis model. The initial studies, using the technique of sequential site mapping, demonstrated the presence of a reentrant circuit in the free wall of the right atrium, although it was suggested that the septum also might be involved.1 Subsequent studies of the atrial flutter reentrant circuit from our laboratory were performed using simultaneous multisite mapping from an array of 190 electrodes placed on the right atrial free wall.2 3 4 5 These studies also demonstrated a reentrant circuit on the right atrial free wall, as did subsequent studies from other laboratories.6 7 8 9 None of the latter studies included activation mapping of the atrial septum.
More recently, we demonstrated that the sterile pericarditis canine model can serve as a model for atrial fibrillation.10 The latter study included simultaneous, multisite mapping of both atria and the interatrial septum from 384 to 396 electrodes. While using the latter recording techniques to understand why sometimes atrial flutter is induced and sometimes atrial fibrillation is induced in this model, it became clear that septal activation was an important part of the atrial flutter reentrant circuit. This article is a more complete characterization of the reentrant circuit in this model, and it provides important new insights into the mechanism of atrial flutter.
Studies were performed 2 to 4 days after the creation of sterile pericarditis in 9 adult mongrel dogs weighing 18 to 27 kg. All studies were performed in accordance with standard guidelines on the use of laboratory animals.
Creation of the Sterile Pericarditis Model
Sterile pericarditis was created as described previously by our laboratory.11 At the initial surgery, 3 pairs of stainless-steel wire electrodes, Teflon-coated except at the tip, were sutured on Bachmann’s bundle, the right atrial appendage, and the posterior-inferior left atrium close to the proximal portion of the coronary sinus.1 2 3 4 5 10 11 12 A fourth pair was sutured on the right ventricular apex for ventricular pacing after His-bundle ablation was performed during subsequent studies. These electrodes were then brought out through the chest wall and exteriorized posteriorly in the midline of the neck. After completion of surgery, the dogs were given antibiotics and analgesics and allowed to recover.
On the second (4 dogs), third (2 dogs), or fourth (3 dogs) postoperative day, the heart was exposed for placement of the electrode arrays using standard techniques. Each dog’s body temperature was kept within the normal physiological range throughout the study by using a heating pad and keeping intravenous saline administration at body temperature. Before opening the chest but during the anesthetized state, radiofrequency ablation of the His bundle was performed10 to create complete AV block to minimize temporal superimposition of ventricular activation with atrial activation during atrial flutter. Then, ventricular pacing (60 to 100 bpm) was initiated.10
Induction of Atrial Flutter
Atrial flutter induction was attempted using previously described pacing protocols.1 11 12 During these studies, ECG lead II and bipolar electrograms obtained from the stainless-steel wire electrodes were recorded and monitored.10 Atrial flutter was identified using standard criteria.2 3 4 5 7 Only episodes of sustained atrial flutter (>5 minutes) were studied. After successful induction of sustained atrial flutter and simultaneous multisite mapping, the pacing protocol was repeated to confirm the reproducibility of the arrhythmia induction.
Simultaneous Multisite Mapping
Our multiplexing system can record continuously for 30 minutes.10 For studies of the sequence of right and left atrial activation, an electrode array containing 372 unipolar electrodes arranged in 186 bipolar pairs was used.10 There were 95 pairs for the right atrium and 91 pairs for the left atrium; the latter included 14 pairs placed separately on Bachmann’s bundle.10 The interatrial septum was mapped simultaneously with the atrial epicardium using a 24-pole electrode catheter (interelectrode distance, 1 mm).10 The distal tip of the electrode catheter was placed in the orifice of the coronary sinus for stability, so we only used 22 of the 24 electrodes.10 Thus, using 394 electrodes, electrograms were simultaneously recorded for up to 30 minutes from the right atrial free wall (190 electrodes), the left atrial free wall (154 electrodes), Bachmann’s bundle (28 electrodes), and the atrial septum (22 electrodes).
Data Acquisition and Analysis
For all studies, atrial electrograms from all electrode sites in both atria, the interatrial septum, a marker channel, and ECG lead II were recorded during induced, stable atrial flutter. Data recording and processing were performed using 2 cardiac mapping systems, as previously described.10 Mapping data were also analyzed as previously described.2 3 4 5 10
Induction of Atrial Flutter
In all 9 dogs studied, sustained atrial flutter was successfully induced: in 3 dogs on postoperative day 2, in 3 dogs on postoperative day 3, and in 3 dogs on postoperative day 4. In dog 7, 2 kinds of atrial flutter were reproducibly induced (the reentrant circuits and the atrial flutter cycle lengths were the same, but the directions of the reentrant wave fronts were opposite). In all 10 episodes, a transitional rhythm preceded the onset of atrial flutter.
The Atrial Flutter Reentrant Circuit(s)
The location and course of the atrial flutter reentrant circuits were identified by analyzing each activation map in each atrial flutter episode. A reentrant circuit was present on the right atrial free wall in 6 of the 10 atrial flutter episodes studied, as we2 3 4 5 12 and others6 7 8 9 previously described. However, in these 6 episodes, a second reentrant circuit involving the atrial septum and the right atrial free wall was also present. These 2 reentrant circuits shared activation of the pectinate muscle region of the right atrial free wall in common, thereby forming a figure-of-eight reentrant circuit. In the other 4 episodes, a right atrial free-wall circuit was not observed. Rather, only a single reentrant circuit involving the atrial septum, Bachmann’s bundle, and the right atrial free wall was observed.
Figure⇓-of-Eight Type Reentry
Figure 1A⇑ shows a representative example of the activation sequence map during induced, sustained atrial flutter due to a figure-of-eight reentry in dog 1. A counterclockwise reentrant circuit in the right atrial free wall circulates around a line of functional block. Also, another reentrant circuit exists; it includes atrial septal activation in a superior-to-inferior direction and epicardial breakthrough at 94 ms in the peri-inferior vena caval region (near site a). The latter wave front then proceeds in an anterior-superior direction, joining the free-wall reentrant circuit in the pectinate muscle region. Thus, site b is a part of the shared pathway of the reentrant circuits. Then, activation from this shared reentrant pathway bifurcates (point c), with 1 limb continuing toward Bachmann’s bundle, where it reenters the atrial septum to complete the reentrant circuit (superior-to-inferior septal activation and inferior-superior right free-wall activation). The left atrium is activated by daughter waves from the figure-of-eight reentrant circuit, as is the right atrial appendage.
Figure 1B⇑ shows a representative example of the activation sequence map during induced, sustained atrial flutter due to figure-of-eight reentry (in dog 8) in which the reentrant circuits travel in the opposite direction shown in Figure 1A⇑. The shared pathway remains in the pectinate muscle region of the right atrial free wall.
Note in both Figures 1A⇑ and 1B⇑ that the location of the line of functional block is similar but not the same. The length and location of this line of functional block permitted a pivot point at either end. Also, the line of functional block was never straight. Similar findings were present for the other 4 figure-of-eight reentrant circuits. Figure 1C⇑ shows the relative activation sequence, recorded from selected sites, in the right atrial free-wall reentrant circuit for the episode of atrial flutter shown in Figure 1A⇑. Figure 1D⇑ shows the relative sequence of activation recorded from selected sites in the other reentrant circuit during the same episode.
Of the 6 figure-of-eight reentry episodes (Table 1⇓), 4 had clockwise rotation of the right atrial free wall reentrant circuit. The other reentrant circuit demonstrated superior-to-inferior activation of the atrial septum and inferior-to-superior activation in the right atrial free wall; epicardial breakthrough from the septum was always in the peri-inferior vena caval region, septal “reentry” was always at Bachmann’s bundle, and the shared portion of the reentrant circuits was in the pectinate muscle region of the right atrial free wall (Table 1⇓). In the other 2 episodes, the figure-of-eight reentrant circuits were reversed. In all 6 of the figure-of-eight reentrant circuits, the shared pathway was in the pectinate muscle region of the right atrial free wall, and the functional line of block in the right atrial free wall served both reentrant circuits, one as a central line of block around which one reentrant circuit circulated, and the other as a boundary for that portion of the other reentrant circuit that was in the right atrial free wall.
Single Reentrant Circuit
Figure 2A⇓ illustrates a representative example of the activation sequence during induced, sustained atrial flutter in dog 5, in which no right atrial free wall reentrant circuit existed. Rather, atrial flutter was caused by a reentrant circuit that traveled in a superior-to-inferior direction in the atrial septum and in an inferior-to-superior direction in the right atrial free wall. The latter reentrant wave front reenters the atrial septum via Bachmann’s bundle, and the septal reentrant activation wave front breaks through to the atrial epicardium in the peri-inferior vena caval region. Note the presence of a line of functional block that once again acts critically as a boundary for the reentrant circuit. However, the location of this line of block is more posterior than in the examples of figure-of-eight reentry. Note also that the inferior aspect of the functional block is located so that an inferior pivot point, such as was seen during figure-of-eight reentry, is not possible. This likely explains why a free-wall reentrant circuit does not form.
Figure 2B illustrates the atrial flutter reentrant circuit in dog 7a. In this example, reentrant activation is in the opposite direction to that shown in Figure 2A⇑. Figure 2C⇑ shows electrograms recorded from the selected epicardial atrial sites A-H in Figure 2B⇑. Note that activation proceeds from site H, low in the right atrium, to left atrial site G, from which point it travels in an inferior-to-superior direction via sites F, E, and D. The wave then collides with a wave front that started at site A on Bachmann’s bundle, traveled to site B in the high right atrium, and then to site C just inferior to the superior vena cava. Because of the collision of these 2 wave fronts, the superior end of the line of block could not be a pivot point for a reentrant circuit.
There were 4 episodes of induced, sustained atrial flutter in which only 1 reentrant circuit existed (Table 2⇓). In 2 episodes, activation of the septum was in an inferior-to-superior direction, epicardial breakthrough was at Bachmann’s bundle, reentry into the septum was in the peri-inferior vena caval region, and right atrial free-wall activation was in a superior-to-inferior direction. In the other 2 episodes, the reverse was true. In all 4 episodes, the line of functional block provided an important boundary for the atrial flutter circuit.
F Wave Polarity During Atrial Flutter
We previously related the sequence of atrial activation during induced, sustained atrial flutter in the canine sterile pericarditis model to flutter wave polarity in the ECG, but septal mapping was not performed.1 Data from the present study are completely consistent with what was previously shown. Thus, a positive flutter wave in ECG lead II (Figure 3A⇓) was associated with early activation of Bachmann’s bundle compared with the posterior-inferior left atrium, with the left atrium being activated mainly in a superior-to-inferior direction (Figure 2A⇑). A negative flutter wave in ECG lead II (Figure 3B⇓) was associated with early activation of the posterior-inferior left atrium compared with Bachmann’s bundle, and activation of a considerable portion of the left atrium was in an inferior-to-superior direction (Figure 3B⇓).
The Atrial Flutter Reentrant Circuit
These data expand our knowledge of atrial flutter in the sterile pericarditis model and have important implications for understanding atrial flutter in humans. First, the atrial flutter reentrant circuit is more extensive than previously realized. It is not simply a functionally determined reentrant circuit contained in the right atrial free wall.1 2 3 4 5 6 7 8 9 12 Rather, the reentrant circuit in the right atrial free wall, when present, is part of a figure-of-eight reentrant configuration, with 1 complete circuit involving the atrial septum, Bachmann’s bundle, and the pectinate muscle region of the right atrial free wall, and the other complete circuit involving only the right atrial free wall. These 2 reentrant circuits share a pathway in common in the pectinate muscle region of the right atrial free wall. In addition, a single reentrant circuit that includes the atrial septum, Bachmann’s bundle, and the right atrial free wall may also generate atrial flutter. It seems that this single reentrant circuit occurs when the line of functional block that forms in the right atrium is more posterior, with the inferior aspect of the line of block ending at the inferior vena cava. This prevents an inferior pivot point from developing in the right atrial free wall. Although other mapping studies6 7 8 9 of the atrial flutter reentrant circuit in this same model did not include mapping of the atrial septum, reinterpretation of several of their maps is completely consistent with a reentrant circuit involving the atrial septum.
The demonstration that a reentrant circuit involving the atrial septum is always present during atrial flutter in the sterile pericarditis model and that septal activation may be either inferior-to-superior or superior-to-inferior in direction suggests a similarity with the atrial flutter reentrant circuit in humans. Importantly, it also suggests the possibility that a free-wall reentrant circuit may also be present in some patients with atrial flutter, but it simply has not yet been recognized due to limitations of the mapping resolution using available techniques.
Role of the Line of Functional Block
Lewis and colleagues13 recognized the difficulty of inducing atrial flutter in the normal canine heart. In fact, as previously summarized,14 all reliable models of atrial flutter have had to alter the underlying atrial substrate, either by administering exogenous substrates15 16 or by creating selected lesions.17 18 19 20 21 Nevertheless, Lewis et al13 mapped a few selected sites during induced atrial flutter in 4 canine atria and showed that the reentrant circuit traveled either up (inferior to superior) or down (superior to inferior) the right atrial free wall. The remainder of the reentrant circuit was assumed to include left atrial activation and activation around one or both of the great veins (superior and inferior vena cavae). However, Lewis’ group never mapped the atrial septum, and the only sites on the left side that they mapped were the left atrial appendage and a site on Bachmann’s bundle.13
Later, Rosenblueth and Garcia-Ramos17 postulated that the infrequency of atrial flutter induction resulted because “…a bridge of conduction tissue between the 2 vessels [superior and inferior vena cava]…” existed. They created a crush lesion between the cavae and, subsequently, could reliably induce atrial flutter. When they created temporary block between the cavae using cocaine, atrial flutter was initially inducible, but it lasted only for the duration of the functional block between the cavae. Implicit in their study is that the difficulty Lewis et al13 had in inducing atrial flutter in normal canine atria was the absence of stable block between the vena cavae. Five subsequent limited mapping studies15 18 19 20 21 in which either a crush lesion or an incisional lesion was made between the cavae also permitted reliable induction of atrial flutter, again suggesting the importance of a line of block between the cavae to achieve stable atrial flutter. Of note, in one of these studies,18 incomplete intercaval block in 1 dog resulted in failure of atrial flutter induction.
In the canine sterile pericarditis model, a line of functional block in the right atrial free wall was always present during atrial flutter. Of interest, when the line of block was posterior, with the inferior portion ending on the inferior vena cava, only a single reentrant circuit was present, but when the line of block was more anterior, pivoting around either end of the line of block was possible, and figure-of-eight reentry was present. In both instances, this line of functional block provided an important, in fact, apparently vital boundary that, like the above described intercaval block studies, ensured the integrity of the atrial flutter reentrant circuit. Again, because of apparent similarities only now appreciated between the canine sterile pericarditis model of atrial flutter and atrial flutter in humans, it is reasonable to suggest, if not expect, that a similar observation will be made in patients. Furthermore, the presence of a functional line of block and its variability in location and extent support the notion that our findings are likely true in humans as well.
Role of Bachmann’s Bundle in the Atrial Flutter Reentrant Circuit
This study provides new insights into the superior turn-around or pivot point of the atrial flutter reentrant circuit. For the right atrial free wall reentrant circuit, when it is present, the superior pivot point is as previously described.1 2 3 For the reentrant circuit that includes activation of the atrial septum, activation of Bachmann’s bundle plays an important role. When atrial septal activation is in an inferior-to-superior direction, as it is in most patients with so-called common or typical type I atrial flutter, epicardial breakthrough is at Bachmann’s bundle; superior-to-inferior activation of the right atrial free wall follows. When atrial septal activation is superior to inferior, reentry into the atrial septum of the reentrant wave front that has travelled in an inferior-to-superior direction in the right atrial free wall is via Bachmann’s bundle. These observations are likely to be relevant to atrial flutter in patients, as the nature of the superior aspect of the atrial flutter reentrant circuit in patients has not been well worked out.22
Flutter Wave Polarity
As summarized previously,23 we have shown that the polarity of the P wave in ECG leads II, III, and aVF depends principally on the direction of the activation wave front(s) in the left atrium. A prior study from our laboratory on the polarity of the flutter wave in the ECG during induced atrial flutter in the canine sterile pericarditis model was completely consistent with that concept, but presumably during figure-of-eight–type reentry. The present study confirmed this finding and extended it to atrial flutter with only a single reentrant circuit.
A limitation of this study was that activation of the atrial septum was determined from only one multipolar catheter electrode. Clearly, better resolution of septal activation is desirable. However, because of the excellent electrode density and, therefore, resolution provided by the epicardial electrode arrays, correlation with the data from the septal electrodes permitted adequate understanding of atrial septal activation. Nevertheless, during atrial flutter, it would have been desirable to map more of the atrial septum, particularly the endocardial aspect of the posterior-inferior region of the right atrium between the inferior vena cava, the coronary sinus ostium, and the tricuspid valve annulus and compare the data with what is known of activation of this area during atrial flutter in patients. With the techniques used in this study, that was not possible.
This study was supported in part by grant HL-38408 from the National Institutes of Health, National Heart, Lung, and Blood Institute, and a grant from the Wuliger Foundation.
- Received December 31, 1998.
- Revision received May 19, 1999.
- Accepted May 19, 1999.
- Copyright © 1999 by American Heart Association
Shimizu A, Nozaki A, Rudy Y, Waldo AL. Multiplexing studies of effects of rapid atrial pacing on the area of slow conduction during atrial flutter in the canine sterile pericarditis model. Circulation. 1991;83:983–994.
Niwano S, Ortiz J, Abe H, Gonzalez X, Rudy Y, Waldo AL. Characterization of the excitable gap in a functionally determined reentrant circuit: studies in the sterile pericarditis model of atrial flutter. Circulation. 1994;90:1997–2014.
Ortiz J, Nozaki A, Shimizu A, Khrestian C, Waldo AL. Mechanism of interruption of atrial flutter by morcizine: electrophysiological and multiplexing studies in the canine sterile pericarditis model of atrial flutter. Circulation. 1994;89:2860–2869.
Schoels W, Gough WB, Restivo M, El-Sherif N. Circus movement atrial flutter in the canine sterile pericarditis model: activation patterns during initiation, termination, and sustained reentry in vivo. Circ Res. 1990;67:35–50.
Pagé PL, Hassanalizadeh H, Cardinal R. Transitions among atrial fibrillation, atrial flutter, and sinus rhythm during procainamide infusion and vagal stimulation in dogs with sterile pericarditis. Can J Physiol Pharmacol. 1990;69:15–24.
Olshansky B, Euler D, Wang Y, Kall JG, Hariman RJ. Autonomic innervation is present and affects atrial flutter in canine sterile pericarditis model. Circulation. 1991;84:II-266. Abstract.
Kumagai K, Khrestian C, Waldo AL. Simultaneous multisite mapping studies during induced atrial fibrillation in the canine sterile pericarditis model: insights into the mechanism of its maintenance. Circulation. 1997;95:511–521.
Lewis T, Feil HS, Stroud WD. Observations upon flutter and fibrillation: part II: the influence of artificial obstacles on experimental auricular flutter. Heart. 1920;7:191–233.
Waldo AL. Mechanism of atrial fibrillation, atrial flutter, and ectopic atrial tachycardia: a brief review. Circulation. 1987;75:III-37–III40.
Allessie MA, Lammers WJEP, Bonke IM, Hollen J. Intraatrial reentry as a mechanism for atrial flutter induced by acetylcholine and rapid pacing in the dog. Circ Res. 1984;70:123–133.
Hayden WG, Hurley EJ, Rytand DA. The mechanism of canine atrial flutter. Circ Res. 1967;20:496–505.
Waldo AL, Hoffman BF, James TN. The relationship of atrial activation to P wave polarity and morphology. In: Little RC, ed. Physiology of Atrial Pacemakers and Conductive Tissues. Mt. Kisco, NY: Futura Publishing Company; 1980:261–289.