INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

COMPLETE ATRIOVENTRICULAR (AV) conduction block [complete heart block (CHB)] is a useful model for investigation of hemodynamic and electrophysiological abnormalities arising from ventricular bradycardia, in general, and from loss of AV synchrony, in particular. Most previously described CHB models have been surgically created in large animals, such as pigs (4) and dogs (11, 12), and rely primarily on intrinsic ventricular escape mechanisms for maintenance of heart rate and cardiac output. We have developed a closed-chest rabbit CHB model based on transcatheter radiofrequency (RF) AV node ablation and permanent transvenous ventricular pacemaker implantation. Our model is unique, because it minimizes surgical trauma and recovery time and enables chronic as well as acute ventricular rate control at reduced expense relative to large animal surgical CHB models.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This investigation conformed with the Guide to the Care and Use of Experimental Animals published by the Canadian Council on Animal Care (2nd edition, 1993) and was approved by the Hospital for Sick Children Animal Care Committee.

Pacemaker insertion and AV node ablation. Healthy young adult male New Zealand White rabbits (3.5-4.0 kg) were orotracheally intubated and inhalationally anesthetized with a mixture of O₂ (1 l/min), nitrous oxide (1 l/min), and halothane (1-1.5%) delivered in tidal volumes of 90-100 ml/breath at rates of 20-25 breaths/min. A six-lead surface electrocardiogram was recorded by direct acquisition to a 386-MHz personal computer running Axotape 1.1 software (Axon Instruments, Foster City, CA) via a custom-modified analog amplification system. Filters were set at 1 and 50 Hz. This system was also used to acquire and record intracardiac signals.

View larger version (175K): [in this window] [in a new window]   Fig. 1.   Anteroposterior fluoroscopic image of a rabbit thorax demonstrating catheter course through the right internal jugular vein to the right side of the heart. A 5-F quadripolar electrophysiology [radiofrequency (RF)] catheter is positioned with its tip at the crux of the heart, a location at which favorable intracardiac signals have typically been observed, with a high probability of successful RF energy application. A 7-F passive fixation endocardial pacing lead (PACE) is positioned with its tip in the right ventricular apex.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Surface (top trace) and intracardiac (bottom trace) electrocardiographic (ECG) signals recorded before successful induction of complete heart block (CHB) by local application of RF energy. Atrial (A) and ventricular (V) signals are similar in amplitude, and a small His bundle potential (H) is consistently present. P, P wave.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Surface (top trace) and intracardiac (bottom trace) ECG signals recorded after successful RF induction of CHB and before initiation of ventricular pacing. Atrial activation (A, corresponding to P waves on top trace) continues at a typical sinus rate of 250-300/min, with ventricular escape beats (V) observed at an approximate rate of 75/min.

Cardiac performance. Cardiac performance before and after AV node ablation was assessed using M-mode echocardiography and hemodynamic measurements. Central venous and left carotid arterial pressures were measured directly before and after AV node ablation in a representative group of animals (n = 8). Central venous pressure was obtained by transducing the venous sheath. A second catheter was placed in the left carotid artery through a small incision in the vessel wall and used to obtain arterial pressure recordings and blood samples (~0.5 ml) for automated blood gas and hemoglobin measurements (model ABL510, Radiometer Medical, Copenhagen, Denmark). Central venous blood gases were obtained from the superior vena cava. An assumed oxygen consumption of 10 ml · kg¹ · min¹ (7) was used to calculate cardiac output based on the Fick principle (6).

Pathological examination. Hearts were submitted for pathological assessment after 8 days of chronic ventricular pacing at 140/min in animals with CHB (n = 4). Rabbits were euthanized with intravenous pentobarbital (65 mg/kg). After rapid cardiectomy through a midline sternotomy, the aorta was cannulated and perfused retrograde with Ca²⁺-containing HEPES-buffered saline solution (in mM: 135 NaCl, 5.4 KCl, 1.0 CaCl₂, 1.0 MgCl₂, 10.0 HEPES, 10 D-glucose, pH 7.4, with NaOH at 37°C) to wash out the blood. The heart was immersed in 10% buffered paraformaldehyde (pH 7.4). It was subsequently opened through a linear incision from the superior vena cava across the tricuspid valve to the apex of the right ventricle and another incision from the right ventricular apex through the right ventricular outflow tract across the pulmonic valve. This allowed visualization of the landmarks for obtaining a tissue block inclusive of the AV node and adjacent interatrial septum and interventricular septum. Because of the relatively small heart size, it was possible to include most of the interatrial and interventricular septa in a single block, which was routinely processed for histology, embedded in paraffin, sectioned at 5-µm intervals on a rotary microtome, and stained with elastic-trichrome to demonstrate myocardium in red and collagen in blue.

Data analysis. Mean ventricular-to-atrial ratios for successful vs. unsuccessful RF lesions were compared using the unpaired Student's t-test. Changes in cardiac performance variables were assessed using one-way repeated-measures analysis of variance followed by pairwise comparison using Tukey's test. P < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Feasibility. Permanent CHB was achieved in 34 of 38 attempts overall. All 4 failures occurred in the first 6 attempts, with 100% success in the latter 32 preparations in the series. Acute procedural mortality occurred in nine animals, including three in which CHB was not successfully induced. These nine deaths occurred mainly during the early part of the experience and were attributed to cardiac perforation and tamponade (n = 2), airway complications (n = 2), and unknown causes (n = 5). Among 28 survivors with CHB, 3 experienced late mortality (24 h after the procedure) due to postoperative infection (n = 1) or ongoing respiratory issues culminating in euthanasia (n = 2). CHB survivors were maintained for 8 (n = 18), 15 (n = 5), or 22 (n = 2) days after the procedure, during which spontaneous recovery of AV conduction was never observed. The number of RF lesions required to achieve CHB progressively declined to 2.5 ± 1.3 (SD), with single-plane fluoroscopic exposure time down to 3.2 ± 2.1 min in the last 17 preparations in the series. Ventricular pacing thresholds were consistently low, permitting the use of generator pulse widths of <0.1 ms at 5 V in all instances. Rabbits were not continuously monitored after pacemaker implantation, but there were no obvious cases of capture failure as evidenced by surface electrocardiograms obtained 8 days after the procedure on all animals. Moreover, each of the three animals experiencing late death was evaluated on several occasions preterminally, and all had heart rates consistent with their pacemaker programming.

Electrophysiological findings. We retrospectively assessed electrophysiological predictors of ablation success, including the presence or absence of a discrete His bundle electrogram and the ratio of His catheter ventricular-to-atrial electrogram amplitude. RF applications resulting in CHB were associated with a ventricular-to-atrial signal ratio of 1.03 ± 0.35 in the last 17 preparations, whereas failed applications were associated with a ratio of 3.32 ± 2.61 (P < 0.001). Discrete His potentials were observed on 13 occasions, were associated with a ventricular-to-atrial ratio of 2.33 ± 1.62, and were 62% predictive of success in CHB induction. The combination of a discrete His potential and a ventricular-to-atrial ratio in the range 0.7-1.5 was 100% predictive of success in CHB induction (Fig. 4).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Relationship of ventricular (V) to atrial (A) intracardiac signal amplitude and likelihood of successful CHB induction. Each vertical tick on the horizontal axis represents a single animal; each data point represents an RF energy application. The presence of a His potential with balanced A and V signals was most highly predictive of successful CHB induction.

Hemodynamic findings. Before ablation, rabbits in sinus rhythm at 305 ± 33 beats/min had mean arterial and central venous pressures of 51.1 ± 7.1 and 5.3 ± 1.4 mmHg, respectively. After AV node ablation and with ventricular pacing at 140/min, the mean arterial and central venous pressures changed to 40.3 ± 6.1 (P = 0.013) and 7.1 ± 2.5 mmHg (P = 0.052), respectively. Interestingly, with an increase in the ventricular pacing rate to 280 beats/min, mean arterial pressure fully recovered to 50.8 ± 12.5 mmHg (P = 0.041 vs. pacing at 140/min, not significantly different from sinus rhythm), but central venous pressure remained elevated at 7.3 ± 2.1 mmHg. This likely reflects improvement in cardiac output without recovery of AV synchrony. Indeed, cardiac output declined acutely from 2.2 ± 0.8 l/min (n = 7) in sinus rhythm to 1.1 ± 0.2 l/min with ventricular pacing at 140 beats/min after AV node ablation (P = 0.008), partially recovering to 1.7 ± 0.4 l/min with an increase in pacing rate to 280 beats/min (Fig. 5).

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Hemodynamic changes after induction of CHB with application of endocardial pacing from the right ventricular apex. A: Fick-based cardiac output declined significantly during pacing at 140/min but was closer to normal with pacing at 280/min, which approximates the typical heart rate in sinus rhythm. B: mean arterial pressure was slightly but significantly depressed during pacing at 140/min but was essentially normal with pacing at 280/min. C: central venous pressure (CVP) increased after CHB induction and remained elevated irrespective of pacing rate, suggesting a possible effect of loss of atrioventricular synchrony on cardiac filling. *P < 0.05

Echocardiographic findings. Right parasternal short-axis M-mode measurements are provided in Table 1. None of these parameters changed significantly with AV node ablation and right ventricular pacing at 140 or 280 beats/min, as might be anticipated in the acute setting and in keeping with hemodynamic changes described above.

Pathological findings. The histopathology consistently showed a lesion in the interatrial septum, immediately superior to the AV node, consisting of an admixture of necrotic myocardium and early fibrosis with deposition of collagen around fibroblasts together with some macrophages among necrotic myocytes, consistent with thermal injury sustained 8 days before explantation. The AV node was partially or completely intact and demarcated from the adjacent interventricular septum by the central fibrous body. The connection between atrial myocardium and the AV node was interrupted by early fibrotic replacement of myocardium, providing histopathological corroboration of in vivo CHB (Fig. 6).

View larger version (131K): [in this window] [in a new window]   Fig. 6.   Left: interatrial septum (top), atrioventricular node (AVN), central fibrous body (CFB), and superior portion of the interventricular septum (bottom). Trichrome staining shows muscle in red and collagen in blue; ×20. Right: inset magnified ×100. Thermal injury sustained 8 days earlier resulted in substantial necrosis to the atrial myocardium and subsequent replacement by early fibrosis (blue area) with numerous fibroblasts evident. Arrow, interruption in the bridge of atrial myocardium to the AVN, accounting for the CHB observed electrophysiologically.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CHB in humans is most often an acquired condition. It typically complicates inferior wall myocardial infarction in older adults (1), surgical repair of congenital heart lesions in children (2), or alloimmune injury in the fetus (3). The intrinsic response to CHB, irrespective of its cause, is activation of a hierarchy of subsidiary pacemakers in the proximal His-Purkinje conduction system or in the ventricles themselves, usually resulting in a ventricular rate substantially slower than the sinus node or atrial rate. Inadequate or nonsustained escape mechanism activation can result in severe ventricular bradycardia or asystole, culminating in congestive heart failure, various arrhythmias, or sudden cardiac death. Consequently, CHB constitutes a prime clinical indication for permanent artificial pacemaker implantation.

CHB models can provide valuable insights into the hemodynamic and electrophysiological consequences of ventricular bradycardia, congestive heart failure, and loss of synchrony between atrial and ventricular activation. These models can also be used to assess the risks and benefits of various acute and chronic pacing strategies. Recognition of these advantages has led to development of several CHB models, most of them in larger animals such as dogs (11, 12) or pigs (4). Lee and colleagues (5) recently described CHB induction in rats on the basis of ethanol injection into the AV node area.

Vos and colleagues (9-12) reported extensively on studies utilizing their surgical canine CHB model, which is based on an earlier description by Steiner and Kovalik (8). In all cases, dogs have been left in their intrinsic escape rhythm after CHB induction by injection of 37% ethanol directly into the AV junction via a right thoracotomy. These studies have yielded important insights into structural and electrophysiological remodeling processes that occur in response to acquired CHB. Nevertheless, dogs are relatively expensive to acquire and maintain, and there can be significant societal and institutional impediments to their routine use in biomedical research.

Responding primarily to cost issues inherent in the use of larger animals, Lee et al. (5) recently described a rat CHB model that essentially "miniaturizes" the basic canine model of Steiner and Kovalik (8). Approaching the heart through a midline sternotomy, they injected the AV junction with 70% ethanol under direct visualization of epicardial and vascular landmarks. Rat mortality, CHB induction success, and complication rates were similar to those we experienced with our rabbit model. Postoperative recovery time was not specifically addressed in their report. In the rat model, there is obligatory reliance on intrinsic escape mechanisms, inasmuch as permanent pacing would undoubtedly pose a formidable technical challenge. The use of a transvascular approach to provide AV conduction injury and permanent ventricular pacing has inherent advantages in experimental models of bradycardia. Avoidance of thoracotomy and provision of ventricular rate support should ultimately enhance survival.

The electrophysiological predictors of success in CHB induction provide guidance for optimal catheter placement and refinement of our model. A low ventricular-to-atrial electrogram ratio suggests that the site of successful conduction system interruption is rather high on the AV septum in relation to the AV groove, as confirmed by histopathological evidence that successful RF lesions were actually located slightly proximal to the true AV node. A His bundle signal, if present, likely demonstrates an appropriately anteroseptal catheter position along the AV ring.

The novel CHB model described here offers several potentially important advantages over preexisting models. Rabbits are relatively inexpensive and easy to procure and maintain compared with larger animals such as dogs. However, the techniques reported here should be readily transferable to larger species if rabbits are considered unsuitable in any specific investigational circumstances. Transcatheter RF AV node ablation allows for more rapid postprocedure recovery associated with less discomfort than does surgical thoracotomy; consequently, data gathering in the critical few days immediately after the procedure should be less difficult and more reliable. Postablation ventricular pacing prevents acute CHB-related death and congestive heart failure. Ventricular pacing also enables assessment of chronic rate-dependent electrical remodeling, repolarization changes attributable specifically to loss of AV conduction, and repolarization response to acute ventricular rate changes under various chronic conditions. Finally, our model allows for detailed investigation of morphological, hemodynamic, and electrophysiological effects of various pacing strategies in CHB at levels ranging from the intact organism to subcellular gene expression.