Comparative Efficacy of Nebivolol and Metoprolol to Prevent Tachycardia-Induced Cardiomyopathy in a Porcine Model

As their names imply, pacing- and tachycardia-induced cardiomyopathies (PTICMs) result from ongoing right ventricular (RV) pacing and from ventricular or supraventricular tachyarrhythmia, respectively. These cardiomyopathies manifest themselves as ventricular systolic dilation and dysfunction and result in heart failure-related symptoms, most of which can be reversed by normalizing the heart rate. Pacing of the RV has led to left ventricular (LV) dysfunction and progression of congestive heart failure (CHF). Investigators have shown that rate control with negative chronotropic agents significantly improves systolic function.^1,2

Patients with chronic tachycardia are often prescribed β-blockers to slow the heart rate and prevent tachycardia-induced cardiomyopathy. Long-term β-blocker use has improved symptoms and increased survival rates in patients with CHF³; however, it is not clear whether β-blockers benefit such patients by modulating chronotropy, inotropy, or both—or by some other mechanism.

Nebivolol (Bystolic^®; Forest Laboratories, LLC, an affiliate of Allergan, Inc.; New York, NY), a 3rd-generation cardioselective β-blocker, induces peripheral vasodilation by increasing the production of endothelial nitric oxide (NO).* Nitric oxide has been shown to augment the parasympathetic effects of acetylcholine, both by increasing its release and by augmenting its stimulatory effect on the production of guanylyl.⁴

L-arginine is a precursor of NO and, therefore, also promotes vasodilation. In patients with heart failure, dysfunction in the L-arginine–NO pathway leads to decreased myocardial perfusion and increased endothelial dysfunction, which partly explains the reduced exercise capacity of patients with CHF. Doutreleau and colleagues⁵ showed that L-arginine supplementation improved endurance and exercise tolerance in patients with stable CHF: compared with a placebo group, patients who took L-arginine had a significant decrease in their average heart rate throughout exercise and the recovery period.

We speculated that the effects of this NO release, combined with the established therapeutic effects of β-blockade on chronic tachycardia, might make nebivolol a promising pharmacologic tool for treating patients with CHF. Therefore, we tested the relative efficacy of nebivolol and a pure β-blocker, metoprolol, in a porcine model, hypothesizing that nebivolol would be superior in preventing tachycardia-induced cardiomyopathy.

Materials and Methods

Fifteen healthy Yucatan pigs underwent implantation of dual-chamber pacemakers (Medtronic, Inc.; Minneapolis, Minn). The pigs were then randomly assigned to groups of 6, 6, and 3, respectively, to be given nebivolol, metoprolol, or placebo once daily, beginning 2 days after pacemaker implantation. Our objective was to lower each pig's heart rate by 10% to 15% from baseline (the mean heart rate on the 2nd day after pacemaker implantation). We began with a dose of 10 mg of nebivolol or 25 mg of metoprolol, with the intent to increase the dose if the target heart rate was not achieved. However, in all cases, these doses were sufficient to achieve the targeted heart rate and appeared to be safe.

To simulate tachycardia, the pacemaker was set at atrioventricular pace (rate, 170 beats/min) on day 7 after implantation, and the pigs were observed for another 7 weeks. The intention was to sustain a high ventricular rate without causing hemodynamic compromise due to atrioventricular dyssynchrony, and to test the effect of the β-blocker in reducing the catecholamine surge caused by sustained tachycardia. Two-dimensional and Doppler echocardiographic studies were performed at baseline and at the end of week 8. We measured serum basic metabolic values, as well as serum levels of brain natriuretic peptide, fibrinogen, epinephrine, norepinephrine, angiotensin-converting enzyme, renin, and C-reactive protein, at baseline and at the end of the study. At the end of week 8, the pigs were euthanized, and complete histopathologic studies were performed. The study was conducted in accordance with the “Guide for the Care and Use of Laboratory Animals.”

The collected data were analyzed with use of SAS statistical software version 9.2 (SAS Institute, Inc.; Cary, NC). Two-way analysis of variance was used to compare the changes in the outcome variables among the 3 groups. Our primary endpoint was the degree of decline in LV ejection fraction (LVEF) from baseline to week 8. A P value <0.05 was considered to be significant.

Results

All the pigs remained hemodynamically stable and survived to the end of the study. All developed LV cardiomyopathy. The mean LVEF decreased from baseline by 34%, 20%, and 20% in the nebivolol, metoprolol, and placebo groups, respectively. These changes did not differ significantly among the 3 groups (P=0.51). Echocardiographic and chemistry findings (Table I) showed changes from baseline in inflammatory- and autonomic-response blood markers in each group, but these changes were not significantly different among the groups. There was mild LV perivascular fibrosis with cardiomyocyte hypertrophy in 14 of the 15 pigs (6 nebivolol, 5 metoprolol, and 3 placebo). The remaining pig, which was in the metoprolol group, had severe multifocal inflammatory changes, including extensive myocardial fibrosis and necrosis in the LV.

TABLE I. Relevant Echocardiographic, Inflammatory, and Autonomic Response Findings in the 3 Groups (N=15)

View Fullscreen

Discussion

The chief finding of the current study is that, outside of any direct effects on heart rate, β-blockers might not be protective against PTICM. This finding suggests that the salutary effects of β-blockers on PTICM are probably not substantial enough to manifest themselves as a reduction in LVEF. This is in direct contradistinction to ischemic and nonischemic cardiomyopathies, against which these drugs are clearly effective.

The reasons for this finding are not obvious. Like other forms of systolic heart failure, PTICM involves a downregulation of β-receptors.⁶ Thus, one would expect that, like other nonischemic cardiomyopathies, PTICM would respond to β-blockade with a rebound in β-receptor density. Although we did not analyze β-receptor density, the absence of LVEF improvement and the trend toward higher plasma levels of adrenergic neurohormonal markers in the β-blockade groups than in the placebo group suggest that administering metoprolol or nebivolol did not lead to improvement in β-receptor density in the treatment groups. If anything, there was a trend toward less susceptibility to pacing-induced diminution of LVEF in the placebo group. Thus, our negative findings cannot be dismissed as an artifact of small sample size; rather, they suggest that administering β-blockers while rapid pacing is underway worsens hemodynamic values, as has been noted in clinical settings.

The PTICMs are heralded by alterations of cellular architecture, including fibrosis.⁷ In our study, no salutary effects of β-blockers on fibrosis were noted. Our study results support other studies of RV pacing that have shown fibrosis in PTICM.⁸ In one such study, the same rate of 170 beats/min was used to induce RV pacing-induced cardiomyopathy.⁹ In our study, β-blockers did not reduce the extent of myocardial fibrosis; rather, the amount and uniformity of the fibrosis was what one would expect to see in this model of heart failure. This finding might have been due to the continued insult of rapid pacing and tachycardia or to the relatively brief duration of the study. In this context, it is important to point out that some of the strongest trends that we found were those pointing to decreased systemic inflammation (as indicated by C-reactive protein and creatine kinase levels) in the β-blockade groups in comparison with the placebo group at 8 weeks.

It should be noted that both metoprolol and nebivolol administration resulted in a nonsignificant attenuation of the renin surge normally associated with cardiomyopathy. Thus, although some of the neurohormonal effects of β-blockade were manifest, the salutary effects of such blockade were not apparent in any of the echocardiographic variables that we studied. This finding seems to support the hypothesis that intrinsic β-receptor and second-messenger dysfunction (and not simply downregulation) at the level of the myocardium promotes the development of PTICM.¹⁰ It might also explain why other study results have shown improved cardiac function with the use of inotropic agents in rapid RV pacing-induced cardiomyopathy.^11,12

The levels of epinephrine and norepinephrine trended higher in both treatment groups than in the placebo group, probably indicating a response to long-term receptor blockade. This suggests that, although systemic activation of the sympathetic system is increased—probably through a renal trigger—during PTICM, the effect of β-blockade on the myocardium in animals undergoing active, rapid pacing is minimal. The reasons for this paucity of effect are unknown, but again, myocardial adrenergic receptor dysfunction seems a plausible explanation.

The response of serum epinephrine and norepinephrine levels to pacing was mixed. Our study showed an increase in epinephrine levels across all groups, whereas norepinephrine levels decreased in the metoprolol-treated pigs and, particularly, in the placebo-treated pigs. The significance and cause of these changes are uncertain.

This study has a few notable limitations. The greatest of these is that because it was an animal study, the number of data points was limited in comparison with those in larger clinical studies that established the efficacy of β-blockers for treating nonischemic cardiomyopathies. In addition, we did not study the effects of higher doses of nebivolol or metoprolol on preventing PTICM; because 10 mg of nebivolol and 25 mg of metoprolol sufficiently reduced heart rate by 10% to 15% in all the pigs, we did not use higher doses in any of them. Another limitation is that we did not study the course of PTICM recovery after pacing was discontinued. We did, however, initiate β-blockade one week before starting pacing. It appears that, in our limited sample, a one-week period of premedication was not sufficient to protect against PTICM. An additional limitation of this study was the titration of the medication, which would have better shown the drugs' clinical effects had it been done over 8 weeks instead of 1 week. Finally, we induced cardiomyopathy with a combination of RV pacing and tachycardia, which are both independent risk factors for cardiomyopathy. We chose this model to maximize the chance of inducing cardiomyopathy in each pig, thereby avoiding having to run more experiments than necessary (the 3-armed study design already necessitated a relatively large sample size), and because in many cases (such as premature ventricular contractions) the foci of ectopy may be associated with dyssynchronous ventricular contraction. However, the combined use of both pacing and tachycardia might have overwhelmed any protective effects specific to increased levels of NO.

Our findings indicate that neither nebivolol nor metoprolol was effective in mitigating the effects of PTICM. The overall outcomes in the placebo group were better than those in both medication groups, which raises the question of whether the side effects of these β-blockers affected the outcomes. In addition, these results suggest that, once the rate-controlling effects of nebivolol and metoprolol are negated, neither has significant effects on myocardial remodeling. This in turn supports the theory that the pathophysiology of PTICM involves fibrosis (which was equally prevalent among our 3 groups) and β-receptor and second-messenger dysfunction, as has been suggested in previous reports. Nonetheless, because this is a pilot study, our findings should be interpreted with caution regarding their clinical implications. Further research is warranted to evaluate the additional effect of NO enhancement of β-blockers in the prevention of tachycardia-induced cardiomyopathy. We plan to perform additional studies that will focus specifically on tachycardia (using atrial pacing only, so that ventricular activation occurs through the normal His-outline conduction system) and, separately, on ventricular pacing mechanisms (in the absence of tachycardia).