Author + information
- Received January 26, 2017
- Revision received March 7, 2017
- Accepted March 10, 2017
- Published online May 15, 2017.
- Ankeet S. Bhatt, MD, MBAa,
- Adam D. DeVore, MD, MHSa,b,c,
- Tracy A. DeWald, PharmD, MHSa,d,
- Karl Swedberg, MD, PhDe,f and
- Robert J. Mentz, MDa,b,c,∗ ()
- aDepartment of Medicine, Duke University Medical Center, Durham, North Carolina
- bDivision of Cardiology, Duke University Medical Center, Durham, North Carolina
- cDuke Clinical Research Institute, Durham, North Carolina
- dDivision of Pharmacology, Duke University Medical Center, Durham, North Carolina
- eSahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden
- fNational Heart and Lung Institute, Imperial College London, London, United Kingdom
- ↵∗Address for correspondence:
Dr. Robert J. Mentz, Duke Clinical Research Institute, PO Box 17969, Durham, North Carolina 27715.
Heart failure (HF) is associated with significant morbidity and mortality. Although initially thought to be harmful in HF, beta-adrenergic blockers (β-blockers) have consistently been shown to reduce mortality and HF hospitalization in chronic HF with reduced ejection fraction. Proposed mechanisms include neurohormonal blockade and heart rate reduction. A new therapeutic agent now exists to target further heart rate lowering in patients who have been stable on a “maximally tolerated β-blocker dose,” but this definition and how to achieve it are incompletely understood. In this review, the authors summarize published reports on the mechanisms by which β-blockers improve clinical outcomes. The authors describe differences in doses achieved in landmark clinical trials and those observed in routine clinical practice. They further discuss reasons for intolerance and the evidence behind using β-blocker dose and heart rate as therapeutic targets. Finally, the authors offer recommendations for clinicians actively initiating and up-titrating β-blockers that may aid in achieving maximally tolerated doses.
Beta-adrenergic blockade has been a mainstay of therapy in chronic heart failure (HF) with reduced ejection fraction (EF) for more than 2 decades. Despite once being thought too dangerous for use in HF due to negative inotropic effects, β-blocker therapy has consistently been shown to reduce mortality and HF-related hospitalizations (1–6). β-blocker therapy is strongly supported across major consensus recommendation statements in patients with reduced EF (7–10). Recent guidelines recommend consideration of adding a new heart rate–lowering agent, ivabradine, in patients with stable chronic HF with EF ≤35% and sinus rhythm with resting heart rate ≥70 beats/min on guideline-directed medical therapy, “including a beta-blocker at maximally tolerated dose” (11). These guidelines recommend initiation and up-titration of β-blockers to “target doses, as tolerated” before consideration of ivabradine. Unlike β-blockers, ivabradine has not been shown to confer a reduction in all-cause mortality, but does reduce HF hospitalization and HF-related deaths as compared with placebo (12). In an era with a new therapeutic option for heart rate reduction, the question of how to achieve a maximally tolerated β-blocker dose is exceedingly relevant for clinicians considering active titration of β-blockers and/or initiation of ivabradine. Despite decades of experience with the use of β-blockers and randomized clinical trials (RCTs) enrolling more than 10,000 patients, optimally defining and achieving a maximally tolerated β-blocker dose remains a clinical challenge.
In this review, we aim to: 1) provide background on proposed β-blocker mechanisms of benefit; 2) describe the current β-blocker doses achieved in practice and compare them with those achieved in landmark β-blocker trials; 3) summarize the evidence supporting β-blocker dose versus heart rate reduction as therapeutic targets; and 4) offer an algorithm for clinicians regarding up-titration of β-blocker therapy.
Overview of Data Sources
To identify relevant articles, we searched MEDLINE (via PubMed) for articles from January 1996 to September 2014. We used Medical Subject Headings (MeSH) and key words, focusing on the most relevant terms. The following search terms were used: (beta blockers[tiab] OR “adrenergic beta-antagonists” [pharmacological action] AND “adrenergic beta-antagonists”[Mesh]) AND (“HF”[Mesh] OR “HF”[tiab] OR congestive HF) AND (dose OR dosing OR heart rate OR “dose-response relationship, drug”[Mesh]). We manually searched for pertinent reviews and studies to find additional relevant citations missed in our original search. We imported all citations into an EndNote X7 (Clarivate Analytics, Philadelphia, Pennsylvania) database.
Mechanisms of Action and Pathophysiology of HF
HF with reduced EF is a progressive, heterogeneous disorder with a complex pathophysiology. Existing evidence on the pathophysiology of HF has been reviewed previously (13). In brief, the HF phenotype may, in part, involve interaction between myocardial injury and left ventricular dysfunction, and compensatory hemodynamic and neurohormonal mechanisms aimed at maintaining cardiac output (13–15). Compensatory mechanisms include activation of the sympathetic nervous system (SNS), renin-angiotensin-aldosterone system (RAAS), and vasodilatory molecules (e.g., natriuretic peptides, prostaglandins, nitric oxide) (14,15). Chronic activation of adrenergic signaling can lead to adverse biological effects, accelerated cardiovascular pathology, and disease progression (14,16). Sympathetic activation may be central to the progression of HF. Current guideline-directed medical therapy targets these compensatory pathways and aims to interfere with the neuroendocrine consequences that develop from their chronic activation (11). Interference with these pathways is postulated as one mechanism of the observed benefits of β-blockers (16).
In general, the pharmacological action of β-blockers is to attenuate SNS activity. Possible mechanisms by which β-blockers improve outcomes include antiarrhythmic effects, slowing detrimental remodeling, decreased myocyte death from catecholamine-induced necrosis, and/or prevention of other detrimental effects of chronic SNS activation, such as increased heart rate (16,17). If general SNS blockade is the key mechanism by which β-blockers provide clinical benefit in HF, it would be expected that a class effect would emerge. However, although selected β-blockers have established a clear mortality benefit in HF with reduced EF (1–6), others demonstrated equivocal results (18,19). Principle characteristics of major β-blocker clinical trials are listed in Table 1.
Notable differences in β-blockers clinically used in HF include differences in beta-1 receptor selectivity and/or affinity, presence of alpha-1 receptor antagonism, antioxidant properties, and vasodilating effects (Table 2). Genetic variants associated with variability in β-blocker response include polymorphisms in beta-adrenergic receptors, alpha-adrenergic receptors, cytochrome P450 2C6, and norepinephrine transporter (NET) (20,21).
Individual variations in β-blocker pharmacology combined with genetic variance may assist in understanding the variable effectiveness of different β-blockers, and may ultimately enhance our understanding of both dose-related clinical benefit and adverse effects.
What Is a Maximally Tolerated Dose?
Across major β-blocker trials in chronic HF with reduced EF, doses achieved in trial participants generally approached target doses. In the CIBIS (Cardiac Insufficiency Bisoprolol Study) II, bisoprolol dosing was progressively increased from a 1.25-mg starting dose to a 10.00-mg daily target dose (1). Target dose was achieved in 42% of patients, with >50% of patients receiving at least 75% of the target dose during the maintenance phase. In the USCS (U.S. Carvedilol HF Study), study participants were initiated on carvedilol 6.25 mg or 12.5 mg twice daily (2). Mean total daily dose was 45 ± 27 mg of carvedilol, with 80% of patients receiving the target daily dose of 50 mg. An initial run-in period may have excluded patients demonstrating early intolerance. Similar results were seen in other major carvedilol trials (3–5). In MERIT-HF (Metoprolol CR/XL Randomized Intervention Trial in Congestive Heart Failure), dose titration started at 12.5 mg or 25 mg daily of metoprolol succinate, with titration over 8 weeks (6). The mean daily dose of the study drug was 159 mg daily, with 87% of patients receiving >100 mg daily and 64% receiving the target dose of 200 mg daily.
Despite achieving high rates of target doses in early, landmark β-blocker clinical trials, the rates of target dose achievement in clinical practice and in subsequent trials are lower (Figure 1). Analysis of the OPTIMIZE-HF (Organized Program to Initiate Lifesaving Treatment in Hospitalized HF Patients) registry found that the mean β-blocker daily dose was <50% of target doses, with greater than two-thirds of patients receiving no up-titration 90 days post-hospitalization (22). Less than 10% of patients achieved target doses of β-blocker at discharge (23). In the COHERE (Coreg [carvedilol] Heart Failure Registry), 41% of patients reached 25 mg of carvedilol twice daily, although more than one-half of these patients were taking <25 mg twice daily at the end of titration (24). Contemporary clinical trial data suggest a similar trend, with ≤50%, ≤30%, and ≤25% of patients at target β-blocker doses in, respectively, the HF-ACTION (Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training) (25), SHIFT (Systolic Heart failure treatment with the If inhibitor ivabradine Trial) (12), and CIBIS-ELD (Cardiac Insufficiency Bisoprolol Study in Elderly) trials (26). Heterogeneous studies have shown the routine failure to achieve target doses in the usual care setting.
Reasons for Delayed Initiation/Up-Titration or Discontinuation/Down-Titration
Reasons behind the discrepancy in dosages achieved in landmark β-blocker trials and later clinical trials or registries are multifactorial. “Intolerance” may involve the selective nature of trial enrollment as well as provider aversion, therapeutic inertia, real or perceived undesired side effects, and clinically significant adverse effects. These factors may result in reluctance to initiate β-blocker therapy, slowed or early-terminated up-titration, and down-titration or permanent discontinuation.
Analyses have shown lower adherence to evidence-based HF therapies in women and older adults (27,28). Perceptions of risks and the burden of multiple medications may lead to provider aversion. Underuse in older adults is seen even after adjustment for comorbidities, EF, and other relevant indicators of risk (29). Providers may also be reluctant to prescribe β-blocker therapy in chronic obstructive pulmonary disease (COPD), for fear of worsening pulmonary status. In a registry of more than 20,000 HF patients, β-blockers were prescribed to 66% of patients at hospital discharge with COPD and at higher rates (75%) in patients without COPD (30). Of note, this survey was taken before widespread adoption of β-blockers, and more current evidence is needed. These data suggest that HF patients at the highest cardiac risk may consistently receive lower doses (28). In addition, fill patterns in HF have not been evaluated on a large scale, but experience from the post-myocardial infarction population suggests a discrepancy between prescribing and fill rates. In a study of 846 myocardial infarction patients prescribed β-blockers, 85% of survivors had filled a prescription by 30 days post-discharge, but 63% and 61% were users at 180 and 365 days, respectively (31).
Trial population selection may also partially explain the observed discrepancy. Landmark β-blocker trials generally enrolled a younger cohort (mean age 60 to 65 years) compared with subsequent analyses (mean age 70 to 75 years) (26). Earlier trial populations also generally had a higher mean resting heart rate, representing greater room for up-titration without inducing clinically significant and dose-limiting bradycardia.
Reasons for intolerability have varied greatly across prior analyses, but include: 1) worsening HF symptoms; 2) bradycardia; 3) hypotension and/or orthostasis; and 4) fatigue. Dizziness and bradycardia are the most commonly cited. In 1 small study of 87 patients with chronic HF, 40% of patients could not tolerate targeted doses of β-blockers using a titration scheme similar to that used in the COPERNICUS (Carvedilol Prospective Randomized Cumulative Survival) trial. Dizziness was the most common reason for discontinuation (41%), with bradycardia/arrhythmia second (16%) (32).
Tolerability may be affected by the specific β-blocker chosen. In the CIBIS-ELD trial, the principal reason for restricted titration was an undesirable reduction in heart rate ≤60 beats/min, seen at higher rates in bisoprolol versus carvedilol (26). The same study found higher rates of reductions in spirometry parameters in patients receiving carvedilol. Cardioselective β-blockers may be preferable in patients with reactive airway disease of chronic obstruction (33), although data are conflicting (30).
Intolerance is influenced by implementation practices. Patients who had specifically trained nurses actively managing up-titration reached higher doses with lower intolerance (34). This strategy simulated the controlled up-titration environments of landmark β-blocker trials. Similar success rates were not seen when patients were simply sent clinical reminders advocating β-blocker use (34). Furthermore, intolerance may not be a class effect. In a retrospective analysis by Butler et al. (35), 80% of patients intolerant to one β-blocker were successfully treated with another. In-class switching resulted in a final β-blocker tolerance rate of 90%.
Predictors for intolerability vary across trials. Large retrospective data suggest higher serum creatinine and more severe chronic HF (lower EF, higher N-terminal pro–B-type natriuretic peptide) may predict intolerance (29,34). No specific predictors emerged after multivariate analysis in these studies. In another small prospective analysis, serum creatinine was the only significant factor predicting β-blocker intolerance (32).
The development of new therapeutic targets, particularly the angiotensin receptor–neprilysin inhibitor (ARNI) sacubitril/valsartan, may also affect the ability to successfully up-titrate β-blockers. The use of ARNI resulted in a 3.2 ± 0.4 mm Hg reduction in mean systolic blood pressure as compared with enalapril in the PARADIGM-HF (Prospective Comparison of ARNI With ACEI to Determine Impact on Global Mortality and Morbidity in Heart Failure) clinical trial (36). The mean daily dose of β-blockers at follow-up in the treatment versus active comparator arms has not been published; therefore, it is currently difficult to quantify the impact that switching from angiotensin-converting enzyme inhibitor/angiotensin receptor blocker therapy to ARNI may have on the maximally tolerated β-blocker dose.
A Dose Relationship or a Heart Rate Reduction Relationship?
Evidence for a dose relationship
Conflicting evidence exists on whether clinicians should target β-blocker dose, heart rate reduction, or both, in chronic HF. Analyses examining β-blocker dose and clinical outcomes are listed in Table 3. The MOCHA (Multicenter Oral Carvedilol HF Assessment) study was the first trial of its kind to demonstrate an overall positive dose-response of carvedilol on left ventricular EF and mortality rates (37). A dose-response relationship was found exclusively in nonischemic cardiomyopathy. Although there was an overall dose response, crude mortality rates were actually higher in the 12.5-mg twice-daily group as compared with the 6.25-mg twice-daily group. Large reductions in mortality rates were only observed when the dose reached ≥25 mg twice daily. Low event rates and a small sample size limit the generalizability of these conclusions.
Recent evidence from the chronic HF trial HF-ACTION (Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training) showed a dose response with respect to all-cause death or hospitalization, but not cardiovascular-specific endpoints (25). The dose-response relationship was seen until 50 mg of carvedilol daily equivalents, after which increasing dose no longer correlated with improved outcomes. Of note, there was no β-blocker randomization in this clinical trial, placing the observed results at risk for bias, confounding, and statistical chance. A follow-up analysis showed that although higher β-blocker dose and lower heart rate were both associated with lower CV mortality in unadjusted analysis, only β-blocker dose was associated with lower all-cause death or hospitalization after multivariate adjustment (38). The effect of β-blocker dose on outcomes was not modified by heart rate. Similar results were seen with bisoprolol (39). A CIBIS II trial analysis found the effect of permanent treatment withdrawal of bisoprolol on outcomes may also be dose-dependent. Higher mortality rates were seen after withdrawal in the high-dose group, despite patients being younger and with fewer comorbidities than those on lower bisoprolol doses (40).
Overall, there is emerging evidence for a relationship between β-blocker dose and clinical outcomes in major clinical trials. The effect of dose on outcomes is not modified by heart rate in these analyses.
Evidence for a heart rate reduction relationship
Epidemiological and clinical studies suggest an association between high resting heart rate and increased risk of cardiovascular events and hospitalization (12,41–44). Secondary analyses suggest the magnitude of heart rate reduction is an important indicator of effect on left ventricular function and mortality (39,45). However, other studies have not confirmed this relationship (46,47). A meta-analysis of 23 RCTs showed an 18% reduction in death for every heart rate reduction of 5 beats/min (48). No relationship between β-blocker dose and all-cause mortality was observed. Another analysis of 654 ambulatory HF patients found β-blocker use and heart rate both predicted mortality, although β-blocker dose did not (49). Interestingly, a mortality rate nadir was seen in the group with heart rates between 58 and 64 beats/min, with a trend toward harm when the heart rate was reduced significantly lower. None of these analyses included patients with HF device therapies, which is particularly relevant because pacing at lower rates may have clinical advantages in HF (as long as consistent right ventricle pacing is avoided), although this has not been validated in large trials (50).
A Revised Model for Initiation and Up-Titration
In accordance with guideline recommendations, in most cases, β-blockers should be up-titrated to reach maximally tolerated doses whenever possible. There are specific specialized populations, including some patients with Stage D HF, where uncertainty still exists regarding the role of β-blockers and the desirability and/or efficacy of up-titration; despite this, the overall evidence suggests that patients who tolerate any dose of a β-blocker are likely to receive overall benefit as compared with nonuse.
Achieving a maximally tolerated dose will require aggressive up-titration in the appropriate clinical setting with accompanying strategies to overcome barriers to up-titration. It will further require a balancing of the known benefits of β-blockers with unintended consequences, which include the need to reduce doses of other vasoactive peptides and the development of unwanted or intolerable adverse effects. The European Society of Cardiology provides guidelines for the initiation and up-titration of β-blockers in HF (10), suggesting low initial doses with dose doubling no earlier than every 2 weeks. These guidelines also suggest strategies for intolerance, such as decreasing the diuretic agent dose in symptomatic hypotension or reviewing the need for other atrioventricular nodal blockers in bradycardia.
On the basis of existing published reports, with the need for further validation, additional strategies may aid clinicians in reaching target doses. A clinical algorithm for achieving a maximally tolerated β-blocker dose is presented in the Central Illustration.
β-Blocker selection may be an important initial consideration. Patients with reactive airway disease may benefit from the use of cardioselective β-blockers. Consequently, patients at risk for dose-limiting bradycardia may benefit from agents other than bisoprolol. HF patients with comorbid peripheral vascular disease or Raynaud disease may benefit from carvedilol, given its positive effects on vascular tone.
In cases of intolerance refractory to dose reduction, efforts should be made to switch to another evidence-based β-blocker, given the discussed improvement in adherence rates with in-class switching (35). Other crossover studies, including a secondary analysis from COMET (Carvedilol or Metoprolol European Trial), showed similar findings. When in-class switching occurred, initial halving of the new β-blocker dose led to lower rates of side effects (51). Precision medicine initiatives may ultimately allow for more personalized β-blocker choice on the basis of individual genetic variations that may predict superior responses.
Active up-titration should be done under the supervision of a nurse facilitator or clinical provider. A RCT of 169 patients found that nurse facilitators allowed for maintenance of β-blocker therapy in 67% of patients, far higher than in the group receiving provider education and computerized reminders and/or patient letters (34). Similar results were seen in a European study group using a clinic staffed by specialized nurses and pharmacists, increasing the proportion of patients on ≥50% of target β-blocker dose from 18% to 57% over a median follow-up of 112 days (52). Understanding that cost considerations may limit the feasibility of this intervention across all HF patients, specialized facilitators should be considered in high-risk HF populations wherever possible, as these groups tend to have lower β-blocker utilization rates.
Further investigation should evaluate implementation models allowing for shared decision-making between providers and patients and/or families. These models may have favorable cost-effectiveness profiles and allow for more rapid up-titration than can be achieved by physicians alone. Self-monitoring and self-titration schemes, including telemonitoring, have been effective and safe in other chronic cardiovascular diseases, such as hypertension and diabetes (53–55). Such strategies should be evaluated in chronic HF and may be particularly useful in patients with transportation barriers.
Dose differences observed in earlier landmark β-blocker RCTs and newer clinical trials and registry data are significant. Our review finds that the observed discrepancy is likely a combination of clinical aversion, differences in the enrollment and titration strategies, and dose-limiting adverse effects. Defining a maximally tolerated β-blocker dose continues to remain a challenge for clinicians. The data we have presented seek to summarize the complex interactions between heart rate, β-blocker dose, and clinical outcomes. Our review offers strategies for achieving a maximally tolerated β-blocker dose and suggestions for new implementation approaches that have been successful in other chronic diseases.
The development of new pharmacological targets for patients with HF and high resting heart rate has increased the relevance of this topic to clinicians and researchers. It is particularly relevant to those considering up-titration of β-blockers and/or the proper time to initiate ivabradine. Implementing high-quality HF care requires aggressive medication titration and optimization of all interactions with the medical system. Ultimately, a maximally tolerated β-blocker dose will be determined through a longitudinal patient-physician interaction. In this case, there is still room for doctoring.
Dr. Devore has received research support from the American Heart Association, Amgen, and Novartis; and has served on an advisory board for Novartis. Dr. Swedberg has received research support from Servier; and has received honoraria from and consulted for Amgen, AstraZeneca, Novartis, and Servier. Dr. Mentz has received research support from the National Institutes of Health, Amgen, AstraZeneca, Bristol-Myers Squibb, GlaxoSmithKline, Gilead, Medtronic, Novartis, Otsuka, and ResMed; has received honoraria from HeartWare, Janssen, Luitpold Pharmaceuticals, Novartis, ResMed, and Thoratec/St. Jude; and has served on an advisory board for Luitpold Pharmaceuticals and Boehringer Ingelheim. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- angiotensin receptor–neprilysin inhibitor
- chronic obstructive pulmonary disease
- ejection fraction
- heart failure
- randomized clinical trial
- sympathetic nervous system
- Received January 26, 2017.
- Revision received March 7, 2017.
- Accepted March 10, 2017.
- 2017 American College of Cardiology Foundation
- Poole-Wilson P.A.,
- Swedberg K.,
- Cleland J.G.F.,
- et al.,
- COMET Investigators
- Packer M.,
- Fowler M.B.,
- Roecker E.B.,
- et al.,
- Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS) Study Group
- Yancy C.W.,
- Jessup M.,
- Bozkurt B.,
- et al.
- McMurray J.J.V.,
- Adamopoulos S.,
- Anker S.D.,
- et al.
- Ponikowski P.,
- Voors A.A.,
- Anker S.D.,
- et al.
- Yancy C.W.,
- Jessup M.,
- Bozkurt B.,
- et al.
- Mann D.L.,
- Bristow M.R.
- Lymperopoulos A.,
- Rengo G.,
- Koch W.J.
- Silke B.
- Bristow M.R.
- Flather M.D.,
- Shibata M.C.,
- Coats A.J.,
- et al.,
- SENIORS Investigators
- DeVore A.D.,
- Mi X.,
- Mentz R.J.,
- et al.
- Fiuzat M.,
- Wojdyla D.,
- Kitzman D.,
- et al.
- Düngen H.D.,
- Apostolovic S.,
- Inkrot S.,
- et al.,
- CIBIS-ELD Investigators and Project Multicentre Trials in the Competence Network Heart Failure
- Butler J.,
- Arbogast P.G.,
- BeLue R.,
- et al.
- Ansari M.,
- Shlipak M.G.,
- Heidenreich P.A.,
- et al.
- Bristow M.R.,
- Gilbert E.M.,
- Abraham W.T.,
- et al.
- Fiuzat M.,
- Wojdyla D.,
- Pina I.,
- Adams K.,
- Whellan D.,
- O’Connor C.M.
- Lechat P.,
- Hulot J.S.,
- Escolano S.,
- et al.
- Simon T.,
- Mary-Krause M.,
- Funck-Brentano C.,
- Lechat P.,
- Jaillon P.
- Laskey W.K.,
- Alomari I.,
- Cox M.,
- et al.,
- AHA Get With The Guidelines-Heart Failure Program
- Kristal-Boneh E.,
- Silber H.,
- Harari G.,
- Froom P.
- Poole-Wilson P.A.,
- Uretsky B.F.,
- Thygesen K.,
- Cleland J.G.,
- Massie B.M.,
- Ryden L.
- Fox K.,
- Ford I.,
- Steg P.G.,
- Tendera M.,
- Robertson M.,
- Ferrari R.,
- BEAUTIFUL Investigators
- Metra M.,
- Torp-Pedersen C.,
- Swedberg K.,
- et al.
- Arnold R.H.,
- Kotlyar E.,
- Hayward C.,
- Keogh A.M.,
- Macdonald P.S.
- Gullestad L.,
- Wikstrand J.,
- Deedwania P.,
- et al.,
- MERIT-HF Study Group
- Kennedy L.,
- Herman W.H.,
- Strange P.,
- Harris A.,
- GOAL A1C Team
- Central Illustration
- Overview of Data Sources
- Mechanisms of Action and Pathophysiology of HF
- What Is a Maximally Tolerated Dose?
- Reasons for Delayed Initiation/Up-Titration or Discontinuation/Down-Titration
- A Dose Relationship or a Heart Rate Reduction Relationship?
- A Revised Model for Initiation and Up-Titration