The Architecture of the Atrioventricular Conduction Axis in Dog Compared to Man:

Abstract

AV Node in Dog and Man. Advances in treating patients with dual atrioventricular nodal pathways have called for a better understanding of the morphology of the approaches to the atrioventricular node. In this respect, it has recently been suggested that, in dog, anatomically discrete muscle bundles originating from the sinus node represent the substrate of the dual pathways recognized electrophysiologically in patients with atrioventricular nodal reentrant tachycardia. This concept is at odds with most anatomic studies of the human specialized atrioventricular junctional area. In this study, therefore, we studied histologically the junctional area in dog hearts, comparing them with our own findings in human heart and the descriptions of the earliest investigators. Five dog and six human hearts were prepared for histology and sectioned serially in different planes. Reconstructions were then made from each of three dog and two human hearts sectioned in orthogonal planes. Gross differences in the anatomy of the atrioventricular junctional area and in the structure of the conduction system were obvious between dog and human hearts. The penetrating portion of the conduction axis was longer in the dog, being much more extensively embedded in the central fibrous body. The atrioventricular node, in both dog and man, was composed of a zone of transitional cells overlying a compact region. The zone of transitional cells in the dog was more extensive posteriorly than anteriorly. No bundles insulated anatomically by fibrous tissue were found either in the internodal atrial myocardium or in the approaches to the atrioventricular node. Our findings in both dog and man are comparable with the initial descriptions of the atrioventricular junctional area. Although the disposition of the conduction system in dog and man is basically similar, there are important differences which relate to the gross anatomy. The anatomic substrate for functional duality of the inputs to the atrioventricular node remains unclear, since our study confirms that the concept of insulated atrionodal tract has no morphologic basis.