INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

THE ATRIOVENTRICULAR (AV) conduction axis is in a strategic position for the study of cardiac electrophysiology. Various perturbations of this electrically insulated conduit between the atrium and ventricle can be evaluated by the monitoring of standard electrophysiological parameters. Experimental heart block or the application of test substances to the conduction tissue has been used to gain a better understanding of not only physiology of the AV node and bundle but also automaticity of the myocardium in general, coronary vascular reactivity, and cardiovascular hemodynamics (8, 16, 18, 19). Thus access to the AV conduction axis is critical in many studies of cardiovascular function.

Key to producing complete heart block is an understanding of the anatomic basis of AV conduction in the heart. Ever since Tawara (26) described the morphology of the AV node and AV conduction axis in the hearts of several species, the critical importance of this tiny isthmus of specialized cardiac muscle between the atria and ventricles has been recognized. Located within a triangular region bounded superiorly by the tendon of Todaro, apically by the septal attachment of the tricuspid valve, and posteriorly by the coronary sinus orifice (10), the compact AV node consists of a distinct cluster of small-diameter muscle fibers adjacent to the central fibrous body. This specialized cardiac muscle projects anteriorly, becoming continuous with a bundle of similar muscle that penetrates the septum at the point at which the tendon and valve annulus come together ("apex" of the triangle of Koch). It is important to emphasize that, on penetration of the septum, the bundle of specialized muscle is encased in the dense fibrous tissue of the central fibrous body. Because atrial and ventricular muscle are separated from each other by collagen and adipose tissue elsewhere in the normal heart, this isthmus of specialized muscle is the only point at which an electrical impulse can be conducted between atrial and ventricular muscle. Still more anteriorly, the AV bundle coursing along the crest of the ventricular septum projects specialized muscle apically in the subendocardium on either side of the muscular septum dividing the left and right ventricles (the left and right bundle branches). The basic features of this anatomy are similar in many species (2, 26).

It is this anatomic course that has served as the basis for models of AV block. Investigators have employed many techniques to interrupt conduction by damaging specialized muscle along the AV conduction axis. The majority of complete heart block models have been created in large animals. In early studies, complete heart block was produced surgically (20, 22, 24, 25). Subsequently, percutaneous vascular access has enabled closed-chest techniques for creation of complete heart block (5, 7, 9). Catheter-directed instillation of necrosis-inducing substances, such as ethanol, has also been reported (27).

In the present age, space and cost considerations often necessitate use of smaller, less expensive animal models (3, 21, 28), and, for these reasons, the rat is a popular species. To our knowledge, rats have been used only infrequently as models of heart block. A preliminary study of catheter-induced electrical injury resulting in heart block in rats was reported many years ago (1). This report, however, lacked a detailed description of the technique, histological verification of the site, and extent of injury and success rates. The objective of the present study was to devise a convenient and reproducible model of complete heart block in the rat, a species that is still relatively inexpensive. Although the thorax must be opened for the epicardial approach, morbidity and mortality can be minimized, and the success rate is very high. This technique can be not only used in physiological investigations for production of heart block but also adapted to deliver biologically active substances into the AV conduction axis.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Surgical technique. The study protocol was approved by the Committee for Animal Research of the University of California at San Francisco. The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication No. 85-23, revised 1985).

View larger version (84K): [in this window] [in a new window]   Fig. 1.   Epicardial fat pad marks approach to atrioventricular (AV) node. A: right atrial appendage (Raa) is reflected laterally with forceps to expose epicardial fat pad (arrowhead) between aortic root (Ao) and medial wall of right atrium. Bent portion of needle, positioned parallel to aorta, will penetrate epicardial surface at a point 1 mm lateral and 1 mm posterior to fat pad. B: fat pad (fp) is identified adjacent to noncoronary (nc) and right coronary (rc) sinuses of aortic valve. a, Atrium; v, ventricle. Magnification ×25; stained with Masson's trichrome.

Hemodynamic recordings. Additional animals were used to determine the hemodynamic effects of complete heart block. The rats were divided into three groups: control (baseline animals, n = 7), acute (1 h after the production of heart block, n = 6), and chronic (5-7 days after the production of heart block, n = 4). The rats were anesthetized by pentobarbital sodium (40-50 mg/kg ip), positioned supine on an animal surgery table, and held in a stable position by paw restraints. The right internal jugular vein was cannulated with a 22-gauge angiocath needle and connected to a pressure transducer to record central venous pressure. The right carotid artery was cannulated with a 22-gauge angiocath needle and connected to a pressure transducer to record systemic blood pressure. The maximum level of left ventricular pressure was obtained by direct left ventricular puncture with a 19-gauge needle connected to a pressure transducer. Heart rate was calculated from the ECG. Hemodynamic monitoring was performed over a 2-min period for each group, to ensure stable hemodynamic recordings. Measurements were the average of three separate cardiac cycles. Hemodynamic measurements were obtained from animals with a narrow QRS complex. The transducer measurements were recorded with the analog-to-digital MacLab chart recorder. ANOVA was used to determine whether the means of control, acute, and chronic conditions were significantly different.

Histology. All rats were killed 1 wk after surgery by an overdose of pentobarbital sodium. After thoracotomy, the hearts were rapidly excised and rinsed in cold saline. The apical half of the ventricles of each heart was cut away and discarded. The remainder was then immersed in 4% buffered paraformaldehyde (pH 7.4) for 1 or 2 days. Under a dissecting microscope, excess tissue at the base of each heart was trimmed. To orient the specimen during embedding, a frontal cross section of the base of the heart was made in a plane roughly parallel to the right ventricular outflow tract; sectioning commenced at this flat surface. The specimen was dehydrated in graded alcohols and xylene before paraffin embedding. Serial 10-µm sections were cut and laid out on a tray. Every 20th section was stained with hematoxylin and eosin for orientation, and, from those containing specialized muscle, every 20th section (staggered by 10 from the hematoxylin and eosin sections) was stained with Masson's trichrome. The result was a composite view of the AV conduction axis of the rat heart cut in the classical transverse plane of section (10).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Surgical results. Representative surface ECGs from before and 1 h after 70% ethanol or saline injections into the AV node area are shown in Fig. 2. Injection of ethanol into the AV node created heart block, whereas saline produced no noticeable difference in the surface electrogram, compared with the baseline state.

View larger version (35K): [in this window] [in a new window]   View larger version (36K): [in this window] [in a new window]   Fig. 2.   Comparisons of surface electrocardiogram before and 1 h after ethanol (A) or saline injections (B) into AV node area. P, representative P waves.

Hemodynamic response. Acute and chronic hemodynamic changes after the production of heart block are presented in Table 1. With onset of complete heart block, there was an increase in central venous pressure (P = 0.015) and a nonsignificant decrease in systolic blood pressure (P = 0.032) but a significant decrease in left ventricular pressure (P < 0.001). Later, the central venous pressure returned to baseline, but there was a persistent decrease in systolic blood pressure and a significant decrease in left ventricular pressure (P < 0.001).

Normal anatomy of the AV conduction axis in the rat heart. Three noninjected rat hearts were dissected and serially sectioned to investigate the normal anatomy of the AV conduction axis. In sections made in the transverse plane of section, the specialized cardiac muscle could be clearly identified by its histological features (Fig. 3). The approximate dimensions of the AV node were 0.7 × 0.6 × 0.1 mm. The compact portion of the node abutted the central fibrous body on its right side. Because of offsetting of the tricuspid valve, however, the AV node was physically much closer to the mitral valve, which attached to the fibrous skeleton at this point (Fig. 3). The specialized muscle of the compact node was continuous with that of the AV bundle in more anterior sections. The bundle was a distinctive aggregate of specialized muscle completely surrounded by collagen (Fig. 3). The branching bundle was a triangular structure at the crest of the ventricular septum.

View larger version (166K): [in this window] [in a new window]   Fig. 3.   Microscopic anatomy of normal AV conduction axis. In classic transverse plane of section at low magnification (A), AV node (*) is a discrete collection of lighter staining muscle fibers in septal area where right atrial wall (a) overlaps muscular ventricular septum (v). Because of offsetting of AV valves, specialized muscle appears in closer proximity to mitral valve (mv) than to tricuspid valve (tv). Higher magnification (B) shows specialized muscle between atrium and ventricle. More anteriorly (C), AV bundle (arrowheads) is composed of a larger collection of specialized muscle surrounded by dense collagen of central fibrous body. High magnification (D) shows lighter staining specialized muscle. A and C: magnification ×30; B and D: magnification ×200. All stained with Masson's trichrome.

Localization of lesions in the ethanol- and saline-injected hearts. The histopathology of the AV node area was assessed in 12 ethanol- and 5 saline-injected hearts. Ethanol-induced myocardial damage appeared as necrotic muscle in various phases of inflammation and repair, depending on time elapsed since injury (Fig. 4). Coagulative necrosis of muscle, mild hemorrhage, and polymorphous inflammation were observed in the acute stage (<4 days, n = 2), whereas the chronic stage (4 days to 1 wk, n = 10) was characterized by granulation tissue replacing specialized muscle. Among the ethanol-injected hearts, seven were from animals with permanent, complete heart block. The extent of the damage in these seven hearts was variable. Five hearts had focal lesions involving essentially only the site of the specialized muscle. The ECG of these five animals revealed a narrower QRS complex than that of the other two animals, which had more widespread granulation tissue involving the atrial septum, crest of the ventricular septum, and the AV bundle. Thus, in the hearts from the animals with permanent, complete AV dissociation and a wide QRS complex, there was necrosis of, or reparative tissue replacing, the AV node and AV bundle. Five of the ethanol-injected rats submitted for histological examination demonstrated only short-term heart block (<30 min). Histological examination of the hearts from these rats demonstrated perinodal injury involving working muscle in the interatrial septum or the ventricular septum, sparing the specialized muscle comprising the AV node and bundle.

View larger version (87K): [in this window] [in a new window]   Fig. 4.   Damage in AV conduction axis from a rat with AV block. A: granulation tissue (*) is present at expected location of AV node in a rat heart injected 7 days previously with 70% ethanol. There is virtually no damage in working muscle of either a or v. B: high magnification shows that there is no remaining specialized muscle in AV junction. A: magnification ×30; B: magnification ×200. Both stained with Masson's trichrome.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

This investigation demonstrates a straightforward and reproducible method of injecting the AV conduction axis in rats. The accuracy of the injection site was validated by the production of AV block. A new finding from this study is that an epicardial fat pad, corresponding to the position of the commissure between the right and noncoronary leaflets of the aortic valve, can be used as a landmark to guide placement of the needle used to inject the AV node area. Starting at a point 1 mm posterior and 1 mm lateral to the fat pad and in a line parallel to the ascending aorta (i.e., directed toward the apex of the heart), the needle tip can easily be made to penetrate the AV node 3 mm below the epicardial surface.

In designing our method, we adapted a surgical approach used in large animals (25). This simple procedure does not require special equipment. Because the injection is through the medial wall of the right atrium, there is minimal effect on morbidity or mortality due to the needle puncture. Furthermore, the open-chest model facilitates electrophysiological experiments such as those employing programmed electrical stimulation. Finally, by allowing animals to recover, we have shown that this method can be useful in chronic preparations. In our series of rats, onset of complete heart block was associated with an increase in central venous pressure, which returned to normal over time, and a more permanent reduction in systolic blood pressure and left ventricular pressure.

Previous studies of the rat AV conduction axis have focused on the histology and ultrastructure of the specialized muscle and have not correlated the gross anatomy with the microscopic anatomy and electrophysiology (2, 12-14, 17). In the present study, we documented histologically the extent of damage to the specialized muscle and correlated these findings with the electrophysiological data. Our structural data from normal hearts corroborate those from other studies defining the normal microscopic anatomy of the conduction axis (2, 10, 11, 26). Each heart in which permanent, complete heart block had been achieved contained a lesion obliterating either the AV node or AV bundle. In many hearts, the limited extent of the lesion underscored the accuracy of the method of localization. A few hearts showed a larger lesion, which extended into the specialized muscle. In any case, the interruption of the specialized muscle was the structural basis for nonconduction of the electrical impulse through the AV conduction axis.

This simple surgical method in rats provides an inexpensive model with which to conduct sophisticated experiments in cardiac physiology and molecular biology. The recent advances in the field of cardiovascular gene transfer have inspired visions of repair and even regeneration of damaged myocardium, possibly as an adjunct to conventional medical therapy or alternative to cardiac transplantation (15, 23). Thus far, studies of gene therapy directed at myocardial cells have focused on the restoration of contractility, and little attention has been paid to the potential of gene therapy in modulating cardiac conduction. The ability to reliably perform AV nodal injections facilitates the study of the effects of drugs and biological substances on cardiac conduction, as measured by changes in the P-R interval of the surface electrogram. In addition, the open-chest model allows programmed electrical stimulation of the atrium to be performed during pharmacological, autonomic blockade to determine the AV node block cycle length and the AV node effective refractory period (6), thus increasing the sensitivity of detection of a change in electrical conduction through the AV conduction axis.

In conclusion, we describe a reliable technique for producing complete heart block in rats. The ability to successfully inject substances into the AV conduction axis provides an easy and precise physiological means of assessing cardiac conduction.