Electrical Conduction Between the Right Atrium and the Left Atrium via the Musculature of the Coronary Sinus
Background—The purpose of this study was to determine whether the coronary sinus (CS) musculature has electrical connections to the right atrium (RA) and left atrium (LA) and forms an RA-LA connection.
Methods and Results—Six excised dog hearts were perfused in a Langendorff preparation. A 20-electrode catheter (2-4-2-mm spacing center to center) was placed along the CS. Excision of the pulmonary veins provided access to the LA, and a second 20-electrode catheter was placed along the LA endocardium opposite the CS catheter. An incision opened the CS longitudinally, and microelectrodes were inserted into the CS musculature and adjacent LA myocardium. Continuous CS musculature was visible along a 35±9-mm length of the CS beginning at the ostium. During lateral LA pacing, CS electrodes recorded double potentials, a rounded, low-frequency potential followed by a sharp potential. The rounded initial potential propagated in the lateral-to-septal direction and represented “far-field” LA activation (timing coincided with adjacent LA potentials and with action potentials recorded from microelectrodes in adjacent LA cells). The sharp potential represented CS activation (timing coincided with action potentials recorded from CS musculature). A distal LA-CS connection (earliest sharp potential in the CS during lateral LA pacing) was located 26±7 mm from the ostium. During RA pacing posterior to the CS ostium, CS electrodes recorded septal-to-lateral activation of the high-frequency potential, with slightly later activation of the rounded potential (LA activation). Incisions surrounding the CS ostium isolating the ostium from the RA had no effect on the CS musculature and LA potentials during RA pacing within the isolated segment containing the CS ostium. RA pacing outside the isolated segment delayed activation of the CS musculature until after LA activation, confirming that the RA-CS connection was located in the region of the CS ostium as well as confirming the presence of the LA-CS connection.
Conclusions—In canine hearts, the CS musculature is electrically connected to the RA and the LA and forms an RA-LA connection.
Several anatomic studies1 2 3 4 5 6 7 in human hearts and canine hearts have shown that muscle morphologically identical to atrial myocardium consistently surrounds the coronary sinus and that the coronary sinus musculature is continuous with left atrial myocardium in the proximal portion of the coronary sinus and with right atrial myocardium at the coronary sinus ostium (Figure 1⇓). The present study was undertaken to determine in Langendorff-perfused canine hearts whether the coronary sinus musculature forms an electrical connection between the right atrium and the left atrium.
Studies were performed according to the guidelines for humane care and treatment of animals established by the National Institutes of Health and were locally monitored by the Animal Studies Subcommittee of the Department of Veterans Affairs Medical Center, Oklahoma City, Okla.
Six mongrel dogs weighing 16 to 20 kg were anesthetized with sodium pentobarbital 30 mg/kg IV and mechanically ventilated with room air. A right thoracotomy was performed, and the heart was exposed. After administration of heparin 3000 U IV, the hearts were excised and immediately submerged in iced, oxygenated (95% O2/5% CO2) modified Tyrode’s solution with the following composition8 (mmol/L): NaCl 117, KCl 4.1, MgCl2 0.5, CaCl2 1.35, NaHCO3 24, NaH2PO4 1.8, dextrose 5.5, and sodium pyruvate 2, plus 0.6 μmol/L albumin (to increase the stability of the Langendorff preparation9) and 15 mmol/L butanedione monoxime to markedly reduce myocardial contractility.10 11 12
The ascending aorta of the excised canine heart was cannulated and perfused with oxygenated, warmed (37±0.5°C) modified Tyrode’s solution of the same composition as described above (pH 7.40±0.05). Perfusion was performed at a constant pressure of 90 to 100 mm Hg, and flow was kept stable at a rate of 200 to 250 mL/min. After the washout of blood, the heart was submerged in a tissue bath with warmed modified Tyrode’s solution. A previous study has shown that atrial and ventricular refractory periods and AV nodal conduction remain stable in this preparation for ≥3.5 hours.8
A 20×10-mm segment of the lateral right atrial wall was excised to provide access to the right atrial endocardium. A 20-electrode catheter (2-4-2-mm spacing center to center) was placed into the coronary sinus and great cardiac vein such that the proximal pair of electrodes was positioned at the coronary sinus ostium and the distal pair of electrodes was located at the lateral aspect of the great cardiac vein (Figure 2⇓). The left atrium surrounding the pulmonary veins was excised (area ≈20×20 mm), providing access to the left atrial endocardium. A second 20-electrode catheter was positioned along the left atrial endocardium parallel to and ≈10 to 15 mm from the mitral annulus, directly opposite the coronary sinus catheter13 (Figure 2⇓).
A bipolar electrode consisting of 2 Teflon-coated silver wires was used for pacing the lateral left atrial epicardium at a cycle length of 400 ms. Simultaneous bipolar recordings were obtained from the left atrial and coronary sinus catheter electrodes to differentiate between components of the atrial potentials generated by the left atrial myocardium and those originating from the coronary sinus musculature. Afterward, the coronary sinus was opened longitudinally from the posterior region to the lateral great cardiac vein with an epicardial incision to visualize the location of the coronary sinus electrodes and to facilitate adjacent microelectrode insertion. Lateral left atrial pacing (cycle length, 400 ms) was repeated to verify the absence of effects of the incision on the coronary sinus catheter electrograms. Microelectrodes were inserted into the coronary sinus musculature and the left atrial myocardium adjacent to the catheter electrodes to correlate local cellular activation with the potentials recorded from catheter electrodes (Figure 2⇑).
The coronary sinus ostium was then isolated from the right atrium in 2 steps (Figure 3⇓). Step 1 consisted of 2 incisions from the tricuspid annulus to the anterior margin of the coronary sinus ostium (incision 1A in Figure 3⇓) and from the posterior margin of the coronary sinus ostium to the inferior vena cava (incision 1B in Figure 3⇓), which isolated the coronary sinus ostium from the right atrium superior to the tendon of Todaro, including the interatrial septum. Simultaneous recordings from the coronary sinus and left atrial catheter electrodes and from a bipolar catheter electrode (2-mm spacing center to center) positioned against the apex of the triangle of Koch (septal recording site at the muscular atrioventricular septum14) were obtained during right atrial pacing posterior to the coronary sinus ostium (pacing site A in Figure 3⇓) at a cycle length of 400 ms both before and after incisions 1A and 1B. A marked delay in the timing of atrial activation at the apex of the triangle of Koch confirmed completion of the isolating incision. In step 2, the coronary sinus ostium was isolated from the remainder of the right atrium by an incision from the tricuspid annulus to the inferior vena cava (incision 2 in Figure 3⇓). Simultaneous bipolar recordings were obtained from the coronary sinus, left atrial, and septal right atrial catheter electrodes during right atrial pacing (cycle length of 400 ms) outside the isolated segment (right atrial pacing site A in Figure 3⇓) and during pacing inside the isolated segment (pacing site B in Figure 3⇓).
After completion of the study, gross examination was used to measure the length of the continuous musculature within the coronary sinus and the distance between the coronary sinus ostium and the distal left atrium–coronary sinus (LA-CS) connection, defined as the location of the bipolar electrode on the coronary sinus catheter that recorded the earliest sharp potential during lateral left atrial pacing.
Transmembrane action potentials were obtained with standard glass microelectrodes filled with 3 mol/L KCl connected to a high-impedance amplifier (model FD 223, World Precision Instruments, Inc) and filtered at 0.01 to 2000 Hz. Bipolar electrograms were filtered at 30 to 500 Hz. All signals were displayed and recorded on a computerized acquisition and analysis system (Bard Electrophysiology).
Evidence for an LA-CS Connection
During lateral left atrial pacing, the coronary sinus catheter electrodes recorded double “atrial” potentials. The first potential was recorded sequentially from the distal to the proximal catheter electrodes, indicating a wave front propagating in the lateral to septal direction, and was low in frequency (far field), suggesting that it represents activation of the left atrium adjacent to the mitral annulus (Figure 4⇓). The second potential was high in frequency and was recorded from only the region of the coronary sinus in which continuous muscular tissue was visible macroscopically (35±9 mm in length, beginning at the coronary sinus ostium), suggesting activation of the coronary sinus musculature (Figure 4⇓). The earliest second potential was recorded 26±7 mm from the coronary sinus ostium, with later activation occurring septally and laterally, suggesting activation of the coronary sinus musculature by the left atrium at this site.
The electrograms recorded from the coronary sinus catheter during lateral left atrial pacing did not change after the longitudinal coronary sinus incision (Figure 5⇓), indicating that the incision was not responsible for the pattern of double potentials recorded during lateral left atrial pacing.
Further evidence regarding the origin of the low-frequency and high-frequency potentials recorded from the coronary sinus catheter was provided by simultaneous recordings from the 20-electrode catheter in the left atrium and from microelectrode recordings. The initial, low-frequency (far-field) potentials recorded in the coronary sinus electrograms during lateral left atrial pacing corresponded in timing with left atrial activation recorded from left atrial catheter electrodes located directly opposite the coronary sinus electrodes (Figure 6⇓). Transmembrane action potentials recorded from microelectrodes inserted into the left atrial myocardium coincided in timing with the adjacent bipolar left atrial potential and with the rounded initial potential in the coronary sinus electrogram (Figure 7⇓). Action potentials recorded from microelectrodes inserted into the coronary sinus musculature coincided in timing with the second sharp potential in the adjacent coronary sinus electrogram (Figure 7⇓). These observations suggest that the pattern of double potentials in the coronary sinus electrogram recorded during lateral left atrial pacing is explained by initial activation of the left atrial myocardium and secondary activation of the coronary sinus musculature via an LA-CS connection.
Evidence for a Right Atrium–Coronary Sinus Connection
When the right atrium was paced just posterior to the coronary sinus ostium, the coronary sinus electrodes recorded activation of the high-frequency potential (coronary sinus musculature) beginning proximally and propagating distally (Figure 8B⇓). The high-frequency potential was consistently recorded before the low-frequency potential representing left atrial activation (Figure 8B⇓). These observations indicate activation of the coronary sinus musculature through a right atrium–coronary sinus (RA-CS) connection at the coronary sinus ostium, with later activation of the left atrium.
The first pair of incisions (tricuspid annulus to the anterior margin of the coronary sinus ostium and posterior margin of the coronary sinus to the inferior vena cava, incisions 1A and 1B in Figure 3⇑) was designed to isolate the coronary sinus ostium from the right side of the interatrial septum. Evidence for this separation was obtained by pacing the right atrium below the incisions (pacing site A in Figure 3⇑), which showed a delay in the timing of left atrial activation adjacent to the septum (tracing LA 9 in Figure 9B⇓) and even greater delay in the timing of atrial activation along the septum on the right (tracing RAS in Figure 9B⇓), suggesting reversal of transseptal conduction from the right-to-left direction to the left-to-right direction. This delay in timing of left atrial activation was not associated with a change in the timing or activation sequence of the high-frequency potential in the coronary sinus electrograms (Figure 9B⇓), indicating activation of the coronary sinus musculature from the right atrium at the coronary sinus ostium. Left atrial activation in the region of the posterior mitral annulus followed activation in the region of the coronary sinus musculature, suggesting that the left atrium was activated via a CS-LA connection (arrow, Figure 9B⇓).
The second incision (tricuspid annulus to the inferior vena cava) isolated the coronary sinus ostium from the remainder of the right atrium. Right atrial pacing outside the isolated segment (pacing site A in Figure 3⇑) was associated with a marked further delay in activation of the septal right atrium (tracing RAS in Figure 9C⇑), suggesting that the atrial impulse propagated around the tricuspid annulus before reaching the septal right atrium at the apex of the triangle of Koch. There was an even greater delay in the timing of left atrial activation close to the septum (tracing LA 9 in Figure 9C⇑), suggesting right-to-left conduction across the interatrial septum and Bachmann’s bundle. The sharp potential in the coronary sinus electrograms followed the low-frequency potential (Figure 9C⇑), indicating activation of the coronary sinus musculature by an LA-CS connection. The delay in activation of the coronary sinus musculature by this incision, preventing the paced right atrial wave front from reaching the coronary sinus ostium, confirms the location of the RA-CS connection in the region of the coronary sinus ostium. Pacing the right atrium within the isolated segment containing the coronary sinus ostium (pacing site B in Figure 3⇑) resulted in unchanged activation of the coronary sinus musculature and secondary activation of the left atrium (Figure 10⇓), confirming the RA-CS connection in the region of the coronary sinus ostium as well as an overall RA-CS-LA connection, because the pacing site does not directly activate any other part of the right atrium to reach a septal RA-LA connection and/or Bachmann’s bundle.
Recent studies suggesting that propagation of impulses between the left and right atria plays an important role in sustaining atrial fibrillation combined with the development of atrial compartmentalization procedures to eliminate atrial fibrillation15 16 17 18 have led to a renewed interest in identifying RA-LA connections. Bachmann’s bundle and the interatrial septum (region of the fossa ovalis) are well-established interatrial electrical connections. The possibility that the coronary sinus musculature forms another RA-LA connection is supported by surgical reports showing that complete isolation of the left atrium in dogs and successful elimination of atrial fibrillation by the maze procedure required cryoablation of the muscular fibers of the coronary sinus.16 17 18 Electrical conduction from the right atrium into the coronary sinus as far as the junction between the coronary sinus and the great cardiac vein was suggested previously in experiments using excised and superfused canine tissue consisting of the coronary sinus along with several millimeters of the posterior right atrial wall surrounding the orifice of the coronary sinus.19 Several anatomic studies1 2 3 4 5 6 7 in human and canine hearts have shown that the coronary sinus musculature is anatomically continuous with left atrial myocardium in the proximal portion of the coronary sinus and with right atrial myocardium at the ostium. The present study confirmed the presence of electrical connections between the coronary sinus musculature and the right and left atrial myocardium, forming an RA-LA connection.
Embryologically, the coronary sinus develops from the sinus venosus.20 21 22 During fetal development, the sinus venosus becomes absorbed into the right atrium, and its muscle becomes continuous with that of the atrium. The right horn of the primitive sinus venosus becomes the cardiac end of the superior vena cava, and the left horn becomes the coronary sinus. The proximal portion of the coronary sinus is surrounded with a spiral sheath of myocardium that is a remnant of sinus venosus musculature.23 This musculature stops abruptly at or shortly beyond the orifices of the entering coronary veins, including the great cardiac vein, and is continuous with that of the morphological right atrium.23
The electrical activity of the coronary sinus musculature and its connection with the right and left atria may have multiple implications for the generation of atrial arrhythmias. Already mentioned is the potential role of the additional RA-LA connection in the maintenance of atrial fibrillation and the requirement to interrupt this connection in the surgical maze procedure.17 18 The coronary sinus musculature may also form part of the atrial end of the slow AV nodal pathway in some patients with AV nodal reentrant tachycardia. We studied 12 patients in whom slow pathway conduction was attenuated or eliminated by ablation of the posterior or posterolateral mitral annulus (4 patients)24 or by ablation within the coronary sinus >10 mm from the ostium (8 patients).25 In patients with so-called epicardial “posteroseptal” accessory pathways requiring ablation from the proximal portion of the coronary sinus or the orifice of the middle cardiac vein, the coronary sinus musculature may form the atrial connection of these pathways.5 6 7 Finally, we have seen a small number of patients in whom focal atrial tachycardia appeared to originate within the coronary sinus musculature or within the proximal portion of the middle cardiac vein. This study also has implications regarding the interpretation of electrograms recorded from the coronary sinus. Potentials from both the coronary sinus musculature and the adjacent left atrium are recorded and may affect the morphology and the timing of activation in patients with so-called “posteroseptal” accessory pathways and may simulate accessory pathway potentials.
Tissue swelling in the Langendorff preparation may have led to a slight overestimation of the length of the visible coronary sinus musculature and of the distance between the coronary sinus ostium and the distal LA-CS connection identified during lateral left atrial pacing. In addition, tissue swelling caused an amplitude decrease of the recorded bipolar potentials. The latter was observed primarily in potentials recorded from the 20-electrode left atrial catheter, because the swelling also pushed the catheter away from the left atrial myocardium, reducing the catheter-wall contact at selected catheter electrodes. This study was performed in canine hearts, because only an ex vivo experiment such as the canine Langendorff preparation allowed the use of microelectrode recordings to validate the catheter recordings and the use of incisions surrounding the coronary sinus ostium to study the electrical connection between the right atrium and the coronary sinus musculature. Although morphological and histological studies suggest that the coronary sinus musculature and its left atrial and right atrial connections appear to be similar in canine and human hearts,1 26 further studies are warranted to confirm the electrophysiological findings in the human heart in vivo.
In canine hearts, electrical activation can propagate along coronary sinus musculature, extending 35±9 mm from the coronary sinus ostium. The coronary sinus musculature is electrically connected to the right atrium (via the coronary sinus ostium) as well as to the left atrium (distal LA-CS connection located 26±7 mm from the ostium), forming an electrical RA-LA connection. This RA-LA electrical bridge may form part of the reentrant circuit in so-called epicardial “posteroseptal” accessory pathways and in AV nodal reentrant tachycardia and may help to sustain atrial fibrillation.
Dr Antz was supported by a fellowship grant from the Deutsche Forschungsgemeinschaft, Bonn, Germany. Dr Tondo was the recipient of a Bristol-Myers Squibb Cardiovascular International Fellowship Award. The authors want to thank Jonathan Bussey for his technical assistance.
- Received February 5, 1998.
- Revision received May 14, 1998.
- Accepted May 27, 1998.
- Copyright © 1998 by American Heart Association
Arruda M, Widman L, Antz M, Otomo K, Wang X, Nakagawa H, Imai S, Scherlag B, Pitha J, Lazzara R, Jackman W. Myocardial continuity between right atrium and left atrium via the coronary sinus. J Am Coll Cardiol. 1997;29(suppl A):358A. Abstract.
von Lüdinghausen M, Ohmachi N, Boot C. Myocardial coverage of the coronary sinus and related veins. Clin Anat. 1992;5:1–15.
McAlpine WA. Heart and Coronary Arteries. New York, NY: Springer; 1975:206–207.
McGuire MA, de Bakker JMT, Vermeulen JT, Opthof T, Becker AE, Janse MJ. Origin and significance of double potentials near the atrioventricular node: correlation of extracellular potentials, intracellular potentials, and histology. Circulation. 1994;89:2351–2360.
Li T, Sperelakis N, Teneick RE, Solaro RJ. Effects of diacetyl monoxime on cardiac excitation-contraction coupling. J Pharmacol Exp Ther. 1985;232:688–695.
Kumagai K, Khrestian C, Waldo AL. Simultaneous multisite mapping studies during induced atrial fibrillation in the sterile pericarditis model: insights into the mechanism of its maintenance. Circulation. 1997;95:511–521.
Elvan A, Pride HP, Eble JN, Zipes DP. Radiofrequency catheter ablation of the atria reduces inducibility and duration of atrial fibrillation in dogs. Circulation. 1995;91:2235–2244.
Wit AL, Cranefield PF. Triggered and automatic activity in the canine coronary sinus. Circ Res. 1977;41:435–445.
Anderson HR, Becker AE, Wenink ACG, Janse MJ. The development of the cardiac specialized tissue. In: Wellens HJJ, Lie KI, Janse MJ, eds. The Conduction System of the Heart: Structure, Function and Clinical Implications. Leiden, Netherlands: HE Stenfert Kroese BV; 1976:3–28.
Moore KL. The Developing Human: Clinically Oriented Embryology. 4th ed. Philadelphia, Pa: WB Saunders; 1988:286–333.
Clark EB, Van Mierop LHS. Development of the cardiovascular system. In: Adams FH, Emmanouilides GC, Riemenschneider TA, eds. Heart Disease in Infants, Children, and Adolescents. 4th ed. Baltimore, Md: Williams & Wilkins; 1989:2–15.
Gerlis LM, Davies MJ, Boyle R, Williams G, Scott H. Pre-excitation due to accessory sinoventricular connexions associated with coronary sinus aneurysms. Br Heart J. 1985;53:314–322.
Tondo C, Otomo K, McClelland J, Beckman K, Gonzalez M, Widman L, Arruda M, Antz M, Nakagawa H, Lazzara R, Jackman W. Atrioventricular nodal reentrant tachycardia: is the reentrant circuit always confined in the right atrium? J Am Coll Cardiol. 1996;27:159A. Abstract.
Tondo C, Beckman KJ, McClelland J, Otomo K, Antz M, Arruda M, Gonzalez M, Nakagawa H, Lazzara R, Jackman W. Response to radiofrequency catheter ablation suggests that the coronary sinus forms part of the reentrant circuit in some patients with atrioventricular nodal reentrant tachycardia. Circulation. 1996;94(suppl I):I-380. Abstract.
Maric I, Bobinac D, Ostojic L, Petkovic M, Dujmovic M. Tributaries of the human and canine coronary sinus. Acta Anat. 1995;156:61–69.