Most cardiac surgeons do not learn the anatomy that is pertinent to arrhythmia surgery even though cardiac surgical training programs are traditionally steeped in the teaching of anatomy. Indeed, the first surgical procedure designed specifically to treat a clinical cardiac arrhythmia (Wolff-Parkinson-White [WPW] syndrome) demanded that the surgeon perform extensive dissection within the atrioventricular (AV) groove, an anatomic space that cardiac surgeons had traditionally avoided. In addition, the incomplete understanding of cardiac anatomy that is characteristic of many nonsurgical physicians has been considered relatively unimportant in the past, even to interventional electrophysiologists. However, with the recent advances in cardiac imaging and the development of new medical devices and catheter delivery systems to treat patients with atrial fibrillation (AF), both arrhythmia surgeons and interventional electrophysiologists have been forced to learn the cardiac anatomy that is pertinent to cardiac arrhythmias.
Fibrous Skeleton of the Heart
For any muscle to perform work, it is necessary for the muscle to have a solid support against which to contract. For example, if a skeletal muscle such as the biceps were excised, placed on a table, and stimulated with an electrical current, it would contract. Because it would have no solid attachments, however, its contraction would not result in any work. This is why skeletal muscles are attached to the bony skeleton. The ability to perform work is no different for cardiac muscle; that is, it must have a relatively solid support against which to contract, or it cannot perform any work. Although the skeletal muscles have the bony skeleton for support, the heart’s support is made of cartilage rather than bone and is called the fibrous skeleton of the heart ( Fig. 3.1 ). The fibrous skeleton of the heart consists primarily of the mitral valve annulus but also includes the tricuspid valve annulus. Where the mitral and tricuspid annuli become contiguous, the fibrous skeleton is slightly thicker and is called the central fibrous body .
Fibrous skeleton of the heart (dashed line) . The components of the fibrous skeleton include the mitral valve annulus, tricuspid valve annulus, central fibrous body, and right and left fibrous trigones.
The anterior portion of the central fibrous body adjacent to the aortic valve has been designated the right fibrous trigone . Some authors in the past have erroneously used the terms central fibrous body and right fibrous trigone interchangeably, but the latter is actually only the anterior portion of the central fibrous body where the anterior leaflet of the mitral annulus is contiguous with the aortic valve. Where these two valves converge laterally, the fibrous skeleton thickens and is called the left fibrous trigone .
The mitral annulus is slightly “bowed” between the right and left fibrous trigones, and in many patients it is somewhat thicker than its free-wall component. The cusps of the aortic valve are thickened at their attachments to the aorta, and together they form a scalloped ridge of tissue inside the aortic root that is commonly referred to as the aortic valve annulus . However, histologically, there is no evidence of fibrous tissue in the aortic root comparable to that seen in the mitral or tricuspid valve annuli, so technically there is no true “aortic valve annulus.”
Atrial Septum
To understand the anatomy of the atrial septum, it is necessary to have a clear three-dimensional (3D) concept of the center of the heart where the fibrous skeleton, both ventricles, and both atria intersect. This 3D picture begins with the understanding that the plane of the tricuspid valve annulus is not in the same plane as the mitral valve annulus because the tricuspid valve is slightly closer to the apex of the heart than the mitral annulus ( Fig. 3.2 ). This slightly more apical plane of the tricuspid valve relative to the mitral valve results in the membranous portion of the ventricular septum extending above the level of the tricuspid annulus medially to become the membranous portion of the atrial septum ( Fig. 3.3 ). Thus, if one were to pass a probe through the membranous portion of the atrial septum from the right atrial side, the tip of the probe would pass below the level of the mitral valve annulus into the left ventricular cavity. This is why infections in this area can result in right atrial-to-left ventricular fistulas. This difference in the horizontal planes of the mitral and tricuspid valves also contributes to the coronary sinus emptying into the right atrium (RA) well above the level of the tricuspid valve annulus ( Fig. 3.4 ).
The mitral valve annulus and tricuspid valve annulus are not in the same plane. The tricuspid valve is located in a more apical position than the mitral valve. (A) Posterior view of the heart with designation of the levels of the mitral and tricuspid valves showing that the tricuspid valve is more apically located than is the mitral valve. (B) If one placed a pane of glass across the heart at the level of the tricuspid valve annulus, everything else in this plane of the heart would be above the pane of glass, and the tricuspid valve would be below it.
The disparity in the planes of the mitral and tricuspid valves results in the superior tip of the membranous ventricular septum extending slightly above the level of the tricuspid annulus to become the membranous portion of the atrial septum. This membrane actually separates the right atrium from the left ventricle.
The disparity in the planes of the mitral and tricuspid valves also contributes to the coronary sinus emptying into the right atrium well above the level of the tricuspid valve annulus.
The most prominent anatomic feature of the atrial septum is the fossa ovalis, which is bordered superiorly by a thick bundle of muscle called the anterior limbus of the fossa ovalis ( Fig. 3.5 ). Posterior to the fossa ovalis, the two medial free walls of the right and left atria may be adjacent and adherent, but there is a potential space between them ( Fig. 3.6 ). This is the potential plane called Waterston’s groove that surgeons develop in preparation for performing mitral valve surgery or before encircling the pulmonary veins. The anatomic fusion between the two atrial walls anteriorly forms the posterior boundary of the fossa ovalis. Dissection in Waterston’s groove is safe until the fossa ovalis is reached. As most cardiac surgeons know from experience, further dissection in this plane will result in the inadvertent entrance into both atria.
Surgeon’s view of the internal right atrium. IVC, Inferior vena cava; FO, Fossa Ovalis ; RAA, right atrial appendage; RV, right ventricle; SVC, superior vena cava.
Three-dimensional diagram of the relationship between the medial walls of the right and left atria to form the atrial septum.
The isthmus of the right atrial wall between the IVC orifice and the tricuspid annulus is called the cavotricuspid isthmus (CTI) ( Fig. 3.7 ). Classic atrial flutter can be caused by one of three different macro-reentrant circuits, and all three of them use the CTI. Therefore placing a lesion across this CTI (“flutter line”) is highly successful for the treatment of classic atrial flutter.
The cavo-tricuspid isthmus (CTI) is the area of the right atrium (RA) between the IVC orifice and the tricuspid annulus in the RA. Classic right atrial flutter uses the CTI in virtually all patients, and a lesion (yellow dashed line) placed across the CTI cures most classic atrial flutter. This lesion must be placed from inside the RA because it cannot be exposed from the epicardial side without extensive surgical dissection. The successful catheter-based treatment of atrial flutter has resulted in this lesion being called the “flutter line.” However, the CTI can support other right atrial reentrant circuits that can contribute to atrial fibrillation (AF). Therefore, under certain circumstances, a CTI lesion can improve the results of ablation procedures for AF as well as for atrial flutter.
Specialized Conduction System
Sinoatrial Node
The anatomic sinus node is commonly located near the junction of the superior vena cava (SVC) and RA in the superior lateral portion of the right atrial free wall ( Fig. 3.8 ). The specialized cells in the sinus node are capable of generating an electrical impulse spontaneously however often one is needed to maintain an optimal heart rate. The generation of these impulses results from a natural phenomenon called spontaneous phase 4 depolarization , which is discussed in Chapter 4 .
The muscle fibers of the right atrial free-wall are arranged diagonally as highlighted in this illustration. The most common location of the anatomic sinoatrial (SA) node is depicted. SVC, Superior vena cava.
For decades it was assumed that all sinus rhythm beats originated from this anatomic sinus node, which is a definitive anatomic structure that can be identified histologically. However, electrophysiological studies performed over the past several years documented that sinus impulses can originate from disparate regions of the RA depending on the heart rate. This electrically active region has been called the atrial pacemaker complex ( Fig. 3.9 ), and sinus rhythm beats normally originate from one of three different regions within this pacemaker complex.
The atrial pacemaker complex consists of three areas in the right atrium (RA). The lateral right atrial free wall includes the anatomic sinoatrial (SA) node and generates normal sinus rhythm beats at a rate of 60 to 100 beats/min. The small area at the junction of the superior vena cava and anterior RA is called the sinus tachycardia site and generates sinus impulses at a rate greater than 100 beats/min. The sinus bradycardia site is in the lower RA near the inferior vena cava and coronary sinus and it generates sinus impulses at a rate below 60 beats/min.
(Reproduced from Cox JL, Ad N, Churyla A, et al. The maze procedure and postoperative pacemakers. Ann Thorac Surg. 2018;106(5):1561–1569.)
During sinus rhythm at normal heart rates of 60 to 100 beats/min, the sinus impulses originate from within or near the anatomic sinus node in the midportion of the atrial pacemaker complex. However, during sinus tachycardia, regardless of whether it is neurally or humorally mediated, the sinus impulses consistently originate from a small area located in the upper right atrial free wall immediately anterior to the junction of the SVC and RA. The importance of this so-called sinus tachycardia site is that if it is damaged during surgery, patients will be unable to generate an appropriate chronotropic response to exercise postoperatively. Injury of this particular portion of the atrial pacemaker complex most commonly occurs as a result of an atrial incision or ablative lesion being placed immediately anterior to the orifice of the SVC. As will be discussed later, this was one of the major problems with the pattern of lesions in the original Maze-I procedure, which had two incisions directly across the sinus tachycardia site, because at the time, the atrial pacemaker complex had not been described in humans.
During sinus bradycardia, the sinus impulses usually originate from the lower portion of the atrial pacemaker complex. Thus, if one administers propranolol to decrease the heart rate, the resultant sinus rhythm beats will originate near the inferior vena cava on the lateral right atrial free wall. The same is true when the normal sinus rhythm rate slows naturally during sleep.
One of the frequently quoted myths in electrophysiology and cardiac surgery is that injury of the sinoatrial (SA) node artery causes sinus node dysfunction. The legendary cardiologist and cardiac pathologist Thomas James once told the author the following:
“The SA node artery, which has few if any branches to the anatomic AV node, reminds me of those interstate highways that pass through small towns and have no exit-ramps. I have never seen any evidence that the so-called “SA node artery” has anything to do with SA node function” ( Fig. 3.10 ).
James’ observation is supported by the fact that neither the presence nor the severity of coronary artery disease in the right coronary artery system correlates with SA node dysfunction. Furthermore, a meta-analysis of 66 studies of the anatomy of the so-called SA node artery in 21,455 human hearts documented that it most commonly originates as a single branch from the right coronary artery and takes a retrocaval course to reach the SA node in 47% of patients, but its location is highly variable ( Fig. 3.11 ).
