Author + information
- Received November 9, 2009
- Revision received January 8, 2010
- Accepted January 11, 2010
- Published online April 27, 2010.
- Andrew Jabbour, MBBS*,†∥,
- Peter S. Macdonald, MD, PhD*,†∥,
- Anne M. Keogh, MD*,†∥,
- Eugene Kotlyar, MD*,
- Soren Mellemkjaer, MD, PhD‡,
- Cathie F. Coleman, MBBS*,
- Maros Elsik, MBBS§,
- Henry Krum, MBBS, PhD§ and
- Christopher S. Hayward, MD*,†,* ()
- ↵*Reprint requests and correspondence:
Assoc. Prof. Christopher Hayward, Cardiology Department, St. Vincent's Hospital, Liverpool Street, Sydney, New South Wales 2010, Australia
Objectives The purpose of this study was to determine the respiratory, hemodynamic, and clinical effects of switching between β1-selective and nonselective beta-blockers in patients with chronic heart failure (CHF) and chronic obstructive pulmonary disease (COPD).
Background Carvedilol, metoprolol succinate, and bisoprolol are established beta-blockers for treating CHF. Whether differences in beta-receptor specificities affect lung or vascular function in CHF patients, particularly those with coexistent COPD, remains incompletely characterized.
Methods A randomized, open label, triple-crossover trial involving 51 subjects receiving optimal therapy for CHF was conducted in 2 Australian teaching hospitals. Subjects received each beta-blocker, dose-matched, for 6 weeks before resuming their original beta-blocker. Echocardiography, N-terminal pro-hormone brain natriuretic peptide, central augmented pressure from pulse waveform analysis, respiratory function testing, 6-min walk distance, and New York Heart Association (NYHA) functional class were assessed at each visit.
Results Of 51 subjects with a mean age of 66 ± 12 years, NYHA functional class I (n = 6), II (n = 29), or III (n = 16), and left ventricular ejection fraction mean of 37 ± 10%, 35 had coexistent COPD. N-terminal pro-hormone brain natriuretic peptide was significantly lower with carvedilol than with metoprolol or bisoprolol (mean: carvedilol 1,001 [95% confidence interval (CI): 633 to 1,367] ng/l; metoprolol 1,371 [95% CI: 778 to 1,964] ng/l; bisoprolol 1,349 [95% CI: 782 to 1,916] ng/l; p < 0.01), and returned to baseline level on resumption of the initial beta-blocker. Central augmented pressure, a measure of pulsatile afterload, was lowest with carvedilol (carvedilol 9.9 [95% CI: 7.7 to 12.2] mm Hg; metoprolol 11.5 [95% CI: 9.3 to 13.8] mm Hg; bisoprolol 12.2 [95% CI: 9.6 to 14.7] mm Hg; p < 0.05). In subjects with COPD, forced expiratory volume in 1 s was lowest with carvedilol and highest with bisoprolol (carvedilol 1.85 [95% CI: 1.67 to 2.03] l/s; metoprolol 1.94 [95% CI: 1.73 to 2.14] l/s; bisoprolol 2.0 [95% CI: 1.79 to 2.22] l/s; p < 0.001). The NYHA functional class, 6-min walk distance, and left ventricular ejection fraction did not change. The beta-blocker switches were well tolerated.
Conclusions Switching between β1-selective beta-blockers and the nonselective beta-blocker carvedilol is well tolerated but results in demonstrable changes in airway function, most marked in patients with COPD. Switching from β1-selective beta-blockers to carvedilol causes short-term reduction of central augmented pressure and N-terminal pro-hormone brain natriuretic peptide. (Comparison of Nonselective and Beta1-Selective Beta-Blockers on Respiratory and Arterial Function and Cardiac Chamber Dynamics in Patients With Chronic Stable Congestive Cardiac Failure; Australian New Zealand Clinical Trials Registry, ACTRN12605000504617)
Randomized controlled trials have demonstrated that carvedilol (1), metoprolol succinate (2), and bisoprolol fumarate (3) all improve survival in patients with chronic heart failure (CHF). These beta-blockers differ, however, in their pharmacological characteristics. The nonselective beta-blocker carvedilol has similar affinity for both β1- and β2-receptors as well as α-blocking capacity, whereas metoprolol and bisoprolol have greater selective affinity for β1- than β2-receptors, which predominate in the lungs. These beta-blockers, particularly the nonselective beta-blockers, are still underprescribed to patients with coexistent chronic obstructive pulmonary disease (COPD) and CHF, largely because of concern about precipitating respiratory deterioration. This is an important consideration as the prevalence of COPD in patients with CHF ranges between 20% and 32% (4). Evidence confirms that the benefits of β1-selective beta-blockers (B1B) in these patients outweigh the risks (5), with data also for carvedilol (6). Limited information, however, exists regarding the safety of switching between these beta-blockers. This study assessed the respiratory, hemodynamic, and clinical effects of switching between B1B and nonselective beta-blockers in patients with CHF and COPD.
Design, settings, and participants
The open-label, randomized, triple-crossover study was conducted in the heart failure clinics of St. Vincent's Hospital (Sydney, New South Wales, Australia) and The Alfred Hospital (Melbourne, Victoria, Australia). In all, 51 CHF patients with stable New York Heart Association (NYHA) functional classes I through III symptoms, already receiving a beta-blocker in addition to standard CHF therapy including angiotensin-converting enzyme inhibitors and/or angiotensin-receptor antagonists, spironolactone, and digoxin were enrolled. A subgroup of 35 (69%) patients with coexistent COPD (meeting GOLD criteria ) was enrolled. Patients with a recent or unstable heart failure diagnosis, recent myocardial infarction, biventricular pacemaker implantation, revascularization, or history of significant intolerance to any of the 3 CHF beta-blockers were excluded. Demographics between the CHF patients with and without COPD were not significantly different (Table 1).
The study was approved by local ethics committees and registered with the Australian New Zealand Clinical Trials Registry (ACTRN12605000504617). All patients gave written informed consent.
Randomization and interventions
Following baseline investigations, patients were randomly switched (8) to a dose-matched beta-blocker (Table 2).The second beta-blocker, which had to be different from the subject's baseline treatment, was administered for 6 weeks before reassessment of outcome measures and crossover to the third beta-blocker for another 6 weeks and reassessment. Subjects then resumed their original beta-blocker and were reassessed 4 weeks later for comparison with baseline (Fig 1).
Outcomes and follow-up
The primary outcome measure was post-bronchodilator forced expiratory volume in 1 s (FEV1). Secondary outcome measures were changes in bronchodilator response, large arterial and cardiac chamber function as measured by pulse waveform analysis and echocardiography, 6-min walk distance (6MWD), and biochemical markers of heart failure severity. Outcomes were measured at baseline and at 6, 12, and 16 weeks.
Spirometry and 6MWD were performed in accordance with American Thoracic Society (ATS) guidelines. The radial arterial pressure waveform was recorded noninvasively by applanation tonometry at the wrist (SphygmoCor Px Version 7.1, ATCOR Medical P/L, West Ryde, Australia). Standard transthoracic echocardiograms were performed as per American Society of Echocardiography guidelines and reported by a blinded cardiologist. Blood samples were collected for NT-proBNP, creatinine, sodium, and bilirubin. Supine, brachial blood pressure was measured using an automatic sphygmomanometer (Welch Allyn Spot Vital Signs, Model 5200–103Z, Welch Allyn, Skaneateles Falls, New York). Three consecutive readings were obtained, and the average of the last 2 recorded. This was repeated after 2 min standing. Weight and resting supine heart rate were also recorded.
Sample size was based on >80% power to detect an absolute change of 150 ml in post-bronchodilator FEV1, assuming 2-sided tests and an alpha value of 0.05 (9). Results are reported as mean with 95% confidence intervals (CIs). Outcomes were analyzed by repeated measures analysis of variance with Bonferroni correction. Friedman's test with post-hoc Dunn's multiple comparison test was performed for nonparametric N-terminal pro-hormone brain natriuretic peptide (NT-proBNP) subgroup data analysis. Statistical analyses were performed on SPSS for Windows version 17.0 (SPSS Inc., Chicago, Illinois) and Prism version 5.0 (GraphPad Software, Inc., San Diego, California) for NT-proBNP graphic and analysis.
The FEV1was significantly higher among patients receiving bisoprolol compared with carvedilol, both in subjects with coexistent COPD (150 ml, 95% CI: 40 to 260 ml; p < 0.01) and without (120 ml, 95% CI: 20 to 220 ml; p = 0.02) (Table 3).A similar, but slightly less marked difference was noted between carvedilol and metoprolol (80 ml, 95% CI: 10 to 150 ml; p = 0.04). No difference in FEV1was noted between B1Bs. The FVC remained unchanged. There was also no significant change in bronchodilator responsiveness between treatments.
Brachial blood pressure was unchanged between treatments. However, central augmented pressure was lowest on carvedilol, with a reduction of 2.3 mm Hg (p = 0.03) compared with bisoprolol and 1.6 mm Hg (p = 0.05) compared with metoprolol (Table 4).A reduction in heart rate of 4.0 beats/min (p = 0.03) was also noted between carvedilol and bisoprolol. Notably, NT-proBNP levels were lowest in subjects when receiving carvedilol (Table 5).
In the subgroup of patients who commenced the study on carvedilol, NT-proBNP and respiratory function tests were performed 4 weeks after resuming carvedilol, at the final visit. Figure 2demonstrates NT-proBNP returning to baseline levels in these patients after rising significantly during the study (p = 1.0). Similarly, Figure 3demonstrates FEV1returning to baseline levels on resumption of carvedilol.
A total of 7 patients (14%) withdrew from the study. No withdrawals occurred when patients were switched from a B1B to carvedilol. Two patients (4%) withdrew after switching between B1Bs, 1 for social reasons and the other for dyspnea. Five patients (10%) withdrew after switching from carvedilol to a B1B (3 carvedilol to metoprolol [with dyspnea]; 2 carvedilol to bisoprolol [1 pancreatic cancer, 1 dyspnea and angina]. Most patients described mild lethargy and dyspnea immediately after switching beta-blockers that generally subsided within a week. Full report of adverse events is outlined in Table 6.
This is the first study to demonstrate significant differences between carvedilol, metoprolol succinate, and bisoprolol in patients with stable CHF and coexistent COPD tolerating beta-blocker therapy. Specifically, switching between B1Bs and carvedilol results in demonstrable changes in airway function, and switching from B1Bs to carvedilol causes short-term reduction of central augmented pressure and NT-proBNP.
Studies have shown that FEV1is not significantly worsened by β1-selective beta-blockade in patients with COPD (5), and although few data exist regarding the effects of carvedilol on FEV1in COPD, a significant difference in FEV1between B1Bs and carvedilol was observed in our study. Further, the magnitude of difference was similar to observed FEV1improvements in studies of tiotropium in COPD (10). Despite this, no excess intolerance was observed changing from bisoprolol to carvedilol, likely related to the lack of post-bronchodilator reversibility of large airway function—as would be expected in a stable COPD cohort.
NT-proBNP is a powerful prognostic marker in CHF (11). In our study, median NT-proBNP levels were lowest on carvedilol compared with bisoprolol and metoprolol. Analysis of the COMET (Carvedilol Or Metoprolol European Trial) data concluded that the achieved median NT-proBNP concentration after beta-blocker treatment is a predictor of mortality (11). Bettencourt et al. (12) showed that during hospitalization for acute decompensated heart failure, lack of a significant (>30%) decrease in NT-proBNP was associated with adverse outcome when compared with patients in whom NT-proBNP decreased significantly (>30%). In our study, 50% of patients had a >30% increase and only 9% had a >30% decrease of NT-proBNP after switching from carvedilol to a B1B. By contrast, switching from metoprolol to bisoprolol yielded mixed findings, with only 21% experiencing a >30% increase and 14% a >30% decrease in NT-proBNP. This is the first description of changes in NT-proBNP levels in a matched cohort of patients with CHF receiving carvedilol, metoprolol succinate, and bisoprolol.
We observed no change in brachial artery pressure; however, central augmented pressure, measured by noninvasive tonometric pressure wave recordings, was lowest on carvedilol, consistent with decreased wave reflection due to the α-blocking effect of the drug. The lower central pulsatile afterload with carvedilol demonstrated in this study may further provide a hemodynamic explanation for the lower NT-proBNP with this drug.
The 6MWD assesses functional capacity and is of prognostic value for patients with CHF and COPD (13). There was no clinically significant difference between beta-blockers in our study. This further validates findings by Metra et al. (14), who showed similar improvements in 6MWD between metoprolol tartrate and carvedilol in a heterogeneous group of patients with CHF. No changes in left ventricular dimensions or ejection fraction were detected, also consistent with current research in CHF (15).
The tolerability of beta-blockers in COPD is of importance as underprescription has been attributed to physicians' perception of patient intolerance (4). We have previously shown that patients with CHF and coexistent COPD without reversible airflow obstruction tolerate carvedilol well (6). In 1 study, 76 of 89 such patients (85%) tolerated carvedilol (16). A randomized trial of 87 CHF patients demonstrated no difference in the tolerability of carvedilol and bisoprolol, although the incidence of COPD in their cohort was not specified (17). Metoprolol was also well tolerated in a small study of patients with ischemic heart disease and COPD (18).
There were 5 (10%) drug-related adverse events necessitating withdrawal from our study, 4 (8%) occurred after carvedilol was switched to a B1B, and 1 (2%) occurred when switching from metoprolol to bisoprolol. No withdrawals occurred after switching from a B1B to carvedilol. Our study provides additional evidence for the tolerability of carvedilol by patients with CHF and coexistent COPD and also suggests that switching from B1Bs to carvedilol is safe. Our data are in agreement with results from the post-study phase of the COMET study, which revealed CHF-related event rates of 4.7% for patients switching from carvedilol to metoprolol and 1.5% for patients switching from metoprolol to carvedilol (19).
Switches between beta-blockers were undertaken without temporary dose reduction to minimize potential risk of worsening CHF and arrhythmias after withdrawal of beta-blockade (20). Although switches were generally safe, most patients described mild worsening of their heart failure symptoms during the first week on their new beta-blocker. Improved tolerability has been reported when the initial dose of the second beta-blocker is reduced (15); this may be a prudent way to exchange beta-blockers in the clinical setting.
The reduction in resting heart rate of 4 beats/min between carvedilol and bisoprolol raises the issue of beta-blocker dose equipotency. Receptor kinetics and differing beta-blocker pharmacokinetic and pharmacodynamic properties may give rise to differences in the extent and duration of beta- and alpha-blockade. It is, however, reassuring that no difference in brachial blood pressure was detected, as was observed in the COMET (11). Limited clinical data are available to guide dose matching (15,19,21), which is further complicated by extrapolation of clinical trial data on short-acting metoprolol tartrate to metoprolol succinate (21). Our dose-matching protocol was based on doses used in landmark clinical trials, all of which showed similar reductions in mortality (1–3). From a practical perspective, these are the doses used in up-titration protocols, irrespective of perceived relative equipotency.
Other limitations include the relatively small cohort and short treatment periods. It should be noted, however, that the study was sufficiently powered to demonstrate changes in outcomes. Additionally, given the pharmacokinetics of the beta-blockers examined, a carryover effect from each drug is unlikely despite potential alterations in receptor density and sensitivity caused by their usual beta-blocker, a notion supported by the observed reduction of NT-proBNP and FEV1back to baseline levels at 4 weeks after recommencement of the original carvedilol in a large subgroup. The majority of patients recruited to the study were on carvedilol at the onset. This introduces selection bias toward subjects with COPD who are able to tolerate nonselective beta-blockade. Patients with a previous severe intolerance to any of the beta-blockers were also excluded for safety reasons. Although the study was open label, the key outcome measures of respiratory function and NT-proBNP were acquired and reported by scientists blinded to the study.
The present study demonstrates that in a cohort of patients with CHF with or without coexistent COPD who are able to tolerate beta-blockers, switching between B1Bs and carvedilol results in demonstrable changes in airway function. Switching from B1Bs to carvedilol causes short-term reduction of central augmented pressure and NT-proBNP, a powerful prognostic marker in chronic heart failure.
The authors wish to thank Drs. Sharon Chih, Ping Nee Lee, Dipak Kotecha, Michele McGrady, Susan Wright, Chick Foo, Queenie Lo, Graham Jones, Deborah Yates, Gary Gazibarich, and Alan Glanville for their academic input into the study. They also acknowledge Mss. Claire Coates, Sharon Stuart, Robyn Funston, Marina Skiba, and Mr. Marcus Juul for their assistance in data acquisition and input.
This study was supported by an unrestricted education grant (“Cardiovascular Lipid” Grant) from Pfizer Australia(West Ryde, New South Wales, Australia) to the principal investigator Assoc. Prof. Hayward. The funding source had no role in the design, conduct, data analysis, or reporting of this study or in the decision to submit the manuscript for publication. Dr. Jabbour has received minor honoraria and educational financial support from AstraZeneca, Roche, and Alpha-Pharma. Dr. Macdonald has received speaker fees from AstraZeneca, Actelion Janssen-Cilag, Novartis, and Pfizer. Dr. Keogh has received trial funding from Actelion, Myogen, Gilead, Pfizer, GlaxoSmithKline, Venragr, Heartware, Novartis, Roche, Bayer, and Schering. Dr. Kotlyar has received honoraria for speaking engagements from Alpha-Pharma and AstraZeneca. Dr. Krum has received grant funding from Roche, Alpha-Pharma, and Merck CSL. Assoc. Prof. Hayward has received grant funding from Pfizer, and honoraria/travel grants from AstraZeneca, Roche, and Alpha-Pharma, CSL Biotherapies, and Merck.
- Abbreviations and Acronyms
- β1-selective beta-blockers
- chronic heart failure
- confidence interval
- chronic obstructive pulmonary disease
- forced expiratory volume in 1 s
- forced vital capacity
- N-terminal pro-hormone brain natriuretic peptide
- New York Heart Association
- 6-min walk distance
- Received November 9, 2009.
- Revision received January 8, 2010.
- Accepted January 11, 2010.
- American College of Cardiology Foundation
- Le Jemtel T.H.,
- Padeletti M.,
- Jelic S.
- Salpeter S.,
- Ormiston T.,
- Salpeter E.
- Macdonald P.S.,
- Keogh A.M.,
- Aboyoun C.L.,
- Lund M.,
- Amor R.,
- McCaffrey D.J.
- Urbaniak G.C.,
- Plous S.
- van Noord J.A.,
- Bantje T.A.,
- Eland M.E.,
- Korducki L.,
- Cornelissen P.J.
- Casaburi R.,
- Mahler D.A.,
- Jones P.W.,
- et al.
- Bettencourt P.,
- Azevedo A.,
- Pimenta J.,
- Frioes F.,
- Ferreira S.,
- Ferreira A.
- Zugck C.,
- Kruger C.,
- Durr S.,
- et al.
- Metra M.,
- Giubbini R.,
- Nodari S.,
- Boldi E.,
- Modena M.G.,
- Dei Cas L.
- Maack C.,
- Elter T.,
- Nickenig G.,
- et al.
- Krum H.,
- Ninio D.,
- MacDonald P.