Transitional Cells

The transitional cells extend anteriorly from the anterior limbus of fossa ovalis, located just above the noncoronary cusp of the aortic valve, and cover the middle and anterior part of the triangle of Koch after crossing the tendon of Todaro, before connecting to the proximal portion of the compact AVN. Such a distribution of transitional cells enables the electrical impulse to spread inside the triangle of Koch without conduction block by the tendon of Todaro.10–12 The area of transitional cells distributed at the right anterior interatrial septum just outside the tendon of Todaro and behind the bundle of His corresponds approximately to the fast pathway site that was targeted during early attempts to cure AVNRT using radiofrequency ablation.^6,13 Transitional cells have also been observed in the left atrial septum,¹⁴ supporting clinical observations of earliest atrial activation at that site during retrograde fast pathway conduction. Although transitional cells have traditionally been considered to constitute the “fast pathway,” experimental observations do not confirm fast conduction velocity in this pathway.^15–17 A shorter conduction delay in this pathway is primarily caused by anatomically shorter distance compared with the “slow pathway.”

The inferior nodal extension (INE) lies parallel to the superior portion of the tricuspid annulus within the triangle of Koch, extends towards the coronary sinus, and then merges with the compact AVN (see Figure 28-1, B). The distal portion of the INE contacts atrial myocytes directly. In the middle of the triangle of Koch, loosely packed transitional cells and the INE converge, and transitional cells overlie the INE at the anterior part of the triangle of the Koch.^14,18 The anatomic relationship between the transitional cells and the INE within the triangle of Koch may explain the two different types of double potentials observed on intracardiac electrograms while targeting the slow pathway potential in the ablation of AVNRT. The site where the double potential consists of a low-frequency deflection (nodal component) followed by high-frequency deflection (atrial component), as described by Jackman and colleagues,¹⁹ corresponds approximately to the distal part of the INE beneath the orifice of the coronary sinus.²⁰ The double potential of high-frequency followed low-frequency deflections described by Haissaguerre et al.²¹ may be generated by the two populations of cells in the middle and anterior parts of the triangle of Koch.²⁰

There are important species differences in the anatomy of the AVJ. In contrast to rabbits, which have only one INE, humans typically have two INEs,4,8 which are referred to as the rightward and leftward extensions. As shown in a study of the anatomy of the human AVJ, there is significant anatomic variability in human.⁸ The majority of humans have two INEs (13 of 21 hearts), some have only the rightward INE (7 of 21 hearts), and, rarely, others have only the leftward INE.⁸ The leftward INE in humans extends from the compact AVN and lies more superiorly to and is usually shorter than the rightward extension (see Figure 28-1, B).⁸ Interestingly, three-dimensional reconstruction of the human AVJ shows age-related changes of the INEs and the compact AVN (see Figure 28-1, C). Compared with the left extension, the length of the right extension increases with age, accompanied by a widening of transitional cell zone.²² Evidence of age-related change of INEs may explain the high incidence of AVNRT within an older aged population compared with children (i.e., >10 years of age compared with <5 years of age). The rightward INE in humans has been regarded to constitute the slow pathway, yet the existence of the leftward extension should also be noted, because it is associated with the atypical left variant of AVNRT.²³

Molecular Characteristics of Transitional Cells and Inferior Nodal Extension Cells

Histologically, the transitional cells and cells of the INE are relatively small and dispersed among connective tissues, and although these two populations of cells share some similar features, there are important molecular and electrophysiologic differences between these cells. It has recently been demonstrated in human tissue that the transitional cells and the INE cells both express the intermediate levels of Tbx3, a transcription factor that regulates the development of the cardiac conduction system compared with the compact AVN.14,24 However, in rabbit, transitional cells stain negative for neurofilament, and INE cells are neurofilament positive, suggesting that these two cell populations have different embryologic origins.^14,16 An important electrophysiologic difference is that, compared with atrial cells, transitional cells express connexin43 (Cx43), an important cardiac gap junction protein, at the intermediate level, whereas the cells of the compact AVN and the leftward INE scantly express Cx43. However, detailed histologic and molecular discrimination between the rightward and the leftward INE in the human AVJ is currently lacking.

Another important difference is that pacemaker activity is observed in INE cells but not in transitional cells in rabbit studies.²⁵ This is supported by gene expression data from humans showing that messenger RNA (mRNA) levels for HCN4, responsible for the hyperpolarization-activated “funny” (I_f) current, and Cav1.3, an alternative L-type calcium-channel isoform, are highly abundant in the INE, whereas expression of SCN5A, the gene encoding for the cardiac sodium channel protein Nav 1.5, is low.²⁴ In contrast, the expression of HCN4 and Cav1.3 is lower in transitional cells, and the expression of SCN5A is high.²⁴ The INE cells have slower Ca²⁺-dependent action potential upstrokes, like the nodal cells, whereas transitional cells have fast sodium current–dependent upstrokes. This may account for relatively slow conduction across the INE. These differences in gene expression may explain why sodium-channel blocking Class IA and IC antiarrhythmic drugs are effective in blocking the fast pathway, whereas L-type calcium-channel blockers such as verapamil are effective in blocking the slow pathway.²⁶ The presence of nodal-like cells in the INE could also explain the occurrence of junctional rhythms during radiofrequency application at the inferior triangle of Koch.²⁷

Anatomy of the Compact Atrioventricular Node and the Bundle of His

The AVN and the His bundle are demarcated by the central fibrous body, as initially described by Tawara.¹ Morphologically¹⁸ and based on Cx43 expression,^28,29 the AVN can be subdivided into the lower nodal bundle and the compact AVN. The compact node is the superior portion of the AVN connected to transitional tissue and the INEs, which was previously referred to as the “open node” by earlier investigators.¹⁸ The lower nodal bundle indicates the inferior portion that is in some cases enveloped by connective tissue, which was previously referred to as the “closed node.” Cells of the lower nodal bundle are longer and arranged more parallel to each other than in the compact AVN and possess an intermediate functional phenotype between that of the compact AVN and the His bundle. Some investigators have recently described the lower nodal bundle as the penetrating bundle of His because it is enclosed by connective tissue.⁵ The lower nodal bundle has been shown to connect to the rightward INE, whereas the leftward INE and the compact AVN are a continuous structure (see Figure 28-1, C).²⁸ Regarding age-related change, the compact AVN is known to change from an oval shape to a spindle shape as age increases (see Figure 28-1, C).²²