brand logo

Am Fam Physician. 2000;62(11):2453-2462

See editorial on page 2402.

Heart failure, the only cardiovascular disease with an increasing incidence, is associated with significant mortality and poses a considerable economic burden. Traditionally, beta blockers have been considered to be contraindicated in patients with heart failure. Recently, however, several large randomized, controlled mortality trials have been stopped early because of significant improvement in mortality rates in patients with heart failure who were given beta blockers in addition to angiotensin-converting enzyme inhibitors, diuretics and, sometimes, digoxin. Beta blockers should now be considered standard therapy in patients with New York Heart Association class II or class III heart failure who are hemodynamically stable, who do not have dyspnea at rest and who have no other contraindications to the use of these agents.

Congestive heart failure is a significant public health problem that affects an estimated 4.7 million Americans and 15 million people worldwide.1,2 Data from the Framingham Heart Study indicate that as many as 465,000 new cases of this disease are identified each year in the United States.3

The increasing incidence of congestive heart failure is in contrast to the decreasing incidence of other cardiovascular disorders.4 Mortality and hospitalization rates for patients with the disease are high and continue to rise.5 The Framingham study found mortality rates of 17 percent at one year, 30 percent at two years and 56 percent at five years.6 The economic burden of heart failure is also substantial. In 1991, Medicare spent 5.5 billion dollars for heart failure hospitalizations, double the amount spent on all cancer hospitalizations.7

Despite the magnitude of the problem, treatment of congestive heart failure is often inadequate. Most patients with heart failure are cared for by primary care physicians.8 Yet a recent survey found that angiotensin-converting enzyme (ACE) inhibitors were being administered to only 80 percent of eligible patients by cardiologists, 71 percent of eligible patients by internists and 60 percent of eligible patients by primary care physicians.9 Data supporting the use of ACE inhibitors in patients with heart failure are compelling and have been well-publicized for a decade.

Recently, three large, randomized, controlled trials demonstrated a substantial mortality benefit for beta-blocker therapy in patients with heart failure.1012 In addition, new data have shown a significant mortality benefit for the use of spironolactone in patients with severe heart failure.13 Other neurohormonal interventions are also being investigated. As the number of effective interventions increases, it has become increasingly important for primary care physicians to stay abreast of developments in heart failure treatment.

Definition of Heart Failure

“Congestive heart failure” is a broad term used to describe a clinical condition resulting from the inability of the heart to adequately pump blood and causing symptoms such as orthopnea, dyspnea on exertion and edema. In practice, “congestive heart failure” is used to describe these symptoms whether they result from diastolic dysfunction, which is a condition of impaired ventricular filling, or systolic dysfunction, which is a condition of impaired ventricular emptying.

As the understanding of heart failure has advanced, it has become important to separate systolic dysfunction from diastolic dysfunction because the etiology, treatment and prognosis of these disorders is quite different. The discussion in this article is limited to heart failure resulting from left ventricular systolic dysfunction, defined as an ejection fraction of less than 40 percent.

Systolic Dysfunction in Heart Failure: Neurohormonal Theory

Left ventricular systolic dysfunction is a mechanical defect (i.e., decreased ejection fraction) that results in the heart's inability to pump oxygenated blood to peripheral tissues. The impact of this dysfunction is manifested in hemodynamic and physiologic abnormalities, including decreased cardiac output, elevated pulmonary capillary wedge pressure and decreased exercise tolerance. Historically, interventions to treat patients with heart failure have been directed at improving these hemodynamic end points. However, clinical trials found a lack of correlation between hemodynamic improvement and clinical improvement, and new strategies were sought.

Recent studies have revealed a correlation between prognosis in heart failure and plasma levels of such neurohormones as endothelin,14 norepinephrine15 and renin,16 among others. Neurohormones have also been shown to stimulate necrosis, fibrosis and apoptosis in the heart. As a result, a neurohormonal hypothesis has evolved to describe how activation of these neurohormonal systems may cause progression of heart failure.16,17

Elevated levels of norepinephrine may affect individual myocytes, heart rate, peripheral resistance and, perhaps, other neurohormonal systems. Blocking the impact of norepinephrine by blocking beta receptors would, in theory, block the adverse affects of norepinephrine. The clinical trials described in this article provide evidence that currently available beta-blocking agents intended to block this neurohormonal process have a significant positive clinical impact in patients with heart failure.

Review of Clinical Trials


Bisoprolol (Zebeta) is a selective beta1 antagonist without significant intrinsic sympathetic activity or vasodilating properties (Table 1). Two randomized, controlled mortality trials of bisoprolol have been completed in patients with heart failure (Table 2.)10,18

Beta blockerReceptorsVasodilating properties
Bisoprolol (Zebeta)Beta1-selectiveNo
Metoprolol tartrate (Lopressor) and metoprolol succinate (Toprol XL)Beta1-selectiveNo
Carvedilol (Coreg)Alpha1, beta1, beta2Yes
Bucindolol (Bextra)*Beta1, beta2Yes (mild)
Features of the trialsCIBIS (1994)CIBIS II (1999)
Number of patients6412,647
Mean age60 years61 years
GenderMen: 83 percentMen: 80 percent
Women: 17 percentWomen: 20 percent
RaceNot reportedNot reported
Ejection fractionLess than 40 percentLess than 35 percent
New York Heart Association classificationClass II: noneClass II: none
Class III: 95 percentClass III: 83 percent
Class IV: 5 percentClass IV: 17 percent
Etiology of heart failureIdiopathic and ischemicIdiopathic and ischemic
Starting dosage1.25 mg per day1.25 mg per day
Target dosage5 mg per day10 mg per day
Average duration of study23 months16 months
Primary end pointMortalityMortality
ResultsAbsolute mortality rate: placebo, 20.9 percent; bisoprolol, 16.6 percentAll-cause mortality: placebo, 17.3 percent; bisoprolol, 11.8 percent
CommentsMortality benefit was not statistically significant.Greatest mortality effect was on sudden death rate.
Placebo mortality was lower than expected and decreased the statistical power of the study.No significant difference was found in the death rate from progressive pump failure.

The Cardiac Insufficiency Bisoprolol Study (CIBIS)18 was the first true mortality trial of beta antagonists. The absolute mortality rate at the end of the approximately 23-month surveillance period was 16.6 percent in the patients who were treated with bisoprolol compared with 20.9 percent in the study subjects who received placebo. However, bisoprolol was titrated up to the target dosage of 5 mg per day in only one half of treated patients; this may have affected the outcome of the study. Nonetheless, CIBIS demonstrated the safety of bisoprolol, along with a significant improvement in New York Heart Association (NYHA) class (Table 3) and a decrease in hospitalizations in the patients who received bisoprolol.

CIBIS-II was similar to CIBIS but had greater statistical power to detect a mortality benefit for bisoprolol therapy. CIBIS-II also had approximately four times as many patients as CIBIS, included more patients with NYHA class IV symptoms and used a higher dosage of bisoprolol.10 The data safety monitoring board stopped CIBIS-II early because of significant improvement, over about 16 months, in a number of parameters among patients receiving bisoprolol compared with those who were given placebo: all-cause mortality, 11.8 percent for the bisoprolol group versus 17.3 percent for the placebo group (P < 0.0001); cardiovascular deaths, 9 percent versus 12 percent (P = 0.0049); and all-cause hospitalizations, 33 percent versus 39 percent (P = 0.0006). The benefits applied irrespective of the etiology of the heart failure. There were 42 percent fewer sudden deaths in the bisoprolol-treated patients. This improvement was statistically significant and accounted for almost all of the mortality benefit. The decreased death rates from other causes were not significant.

NYHA class IAsymptomatic
NYHA class IISymptomatic with moderate exertion
NYHA class IIISymptomatic with minimal exertion
NYHA class IVSymptomatic at rest


Like bisoprolol, metoprolol tartrate (Lopressor) and metoprolol succinate (Toprol XL) are beta1-selective blockers without significant intrinsic sympathetic activity or vasodilating properties. Reports have been published from two mortality trials on the effectiveness of metoprolol in patients with heart failure (Table 4).11,19

Features of the trialsMDC (1993)MERIT-HF (1999)
Number of patients3833,991
Mean age49 years64 years
GenderMen: 73 percentMen: 78 percent
Women: 25 percentWomen: 22 percent
RaceNot reportedWhite: 94 percent
Black: 5 percent
Ejection fractionLess than 40 percentLess than 40 percent
New York Heart Association classificationClass II: 47 percentClass II: 41 percent
Class III: 47 percentClass III: 55 percent
Class IV: 4 percentClass IV: 4 percent
Etiology of heart failureIdiopathicIschemic and nonischemic
Starting dosageMetoprolol tartrate, 10 mg per dayMetoprolol succinate, 12.5 to 25 mg per day
Target dosage100 to 150 mg per day in divided doses twice daily200 mg per day
Average duration of study12 to 18 months12 months
Primary end pointMortality plus need for transplantationMortality
ResultsRisk of reaching end point: placebo, 20.1 percent; metoprolol tartrate, 12.9 percentAbsolute mortality: placebo, 11.0 percent per patient-year of follow-up; metoprolol succinate, 7.2 percent per patient-year of follow-up
CommentsTreatment improved clinical status, reduced the need for heart transplantation and was safe.Study was stopped early. Treatment reduced sudden deaths and deaths from progressive pump failure.
No improvement in mortality.

The Metoprolol in Dilated Cardiomyopathy (MDC) trial19 was a multicenter, randomized, controlled trial designed to study the impact of metoprolol tartrate on a combined end point of death plus the need for heart transplantation in patients with idiopathic heart failure and NYHA class II and class III symptoms. The risk of reaching this end point over 12 to 18 months was 12.9 percent for patients in the treatment group versus 20.1 percent for those in the placebo group, but this difference was not statistically significant. At 12 months, the patients who received metoprolol tartrate had significant improvement in quality of life, ejection fraction and exercise capacity. The two groups had no difference in all-cause mortality. All of the benefit in the primary end point came from a decrease in the need for heart transplantation.

Like the MDC, the Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF)11 investigated the impact of metoprolol on mortality in patients with heart failure. However, MERIT-HF was larger, included patients with ischemic as well as idiopathic heart failure, used metoprolol succinate (controlled-release formulation) and structured the end point to be just mortality instead of the combined end point of mortality plus need for transplantation. MERIT-HF was stopped early because the absolute mortality rate was 7.2 percent per patient-year of follow-up in the treatment group compared with 11.0 percent per patient-year of follow-up in the placebo group. The difference was statistically significant. The improvement in mortality applied to sudden deaths and deaths from progressive pump failure. Only 27 persons with congestive heart failure would need to be treated with metoprolol succinate for one year to prevent one death.


Carvedilol (Coreg) is a novel agent with antagonist activity against alpha1, beta1 and beta2 receptors, as well as some antioxidant activity. It is the only beta blocker labeled by the U.S. Food and Drug Administration (FDA) for the treatment of heart failure. Carvedilol was tested in the U.S. Carvedilol Heart Failure Program, which consisted of four trials in patients stratified into mild, moderate or severe heart failure groups based on exercise capacity12,2023 (Table 512).

Features of the trialsU.S. Carvedilol Heart Failure Program
Number of patients1,094
Mean age58 years
GenderMen: 77 percent
Women: 23 percent
RaceNot reported
Ejection fractionLess than 35 percent
New York Heart Association classificationClass II: 53 percent
Class III: 44 percent
Class IV: 3 percent
Etiology of heart failureIdiopathic and ischemic
Starting dosage3.125 or 6.25 mg twice daily
Target dosage25 to 50 mg twice daily
Average duration of studySix and one-half months
Primary end points*Exercise tolerance
Quality of life
Progression of disease
ResultsNo improvement in exercise tolerance
No improvement in quality of life
48 percent reduction in progression of disease (P = 0.008)
Mortality rate: placebo, 7.8 percent; carvedilol, 3.2 percent
CommentsThe trials were stopped early because of outcome.
Benefit was across all subgroups.

Only one of the carvedilol trials, the study of 366 patients with mild heart failure, evaluated mortality as a primary end point.20 The other trials defined primary end points of exercise capacity and quality of life.2123 Mortality was not a predefined primary end point for the combined trials, but it was defined prospectively and was monitored by the data safety monitoring board, which terminated the trials after an average of only six and one-half months because of a large, statistically significant reduction in mortality risk associated with the use of carvedilol. The absolute mortality rate was 3.2 percent in the treatment group compared with 7.8 percent in the placebo group. Although the magnitude of the benefit was large, the number of deaths was small and the follow-up was short term. The mortality benefit occurred in all subgroups. Treated patients also had a significant reduction in risk of hospitalization for a cardiovascular cause.

Carvedilol was also tested in the Australia/New Zealand Heart Failure Research Collaborative Trial.24 This trial of carvedilol in 415 patients was not a predefined mortality trial, and it showed no significant difference in risk of death between treated patients and those who received placebo.

Finally, carvedilol was recently tested against placebo in 2,200 patients with severe heart failure in the Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS) Trial. The data safety monitoring committee stopped the trial early because of a large mortality benefit in the treatment group.25 The details of this study will not be released until later in 2000.


Bucindolol (Bextra) is a nonselective beta-blocking agent with mild vasodilating properties. It was recently tested in patients with heart failure as part of the Beta Blocker Evaluation Survival Trial (BEST). The results of this trial are not yet available.

BEST was designed to evaluate the mortality impact of bucindolol in patients with severe heart failure (NYHA class III or IV) resulting from idiopathic or ischemic cardiomyopathy.26 However, the data safety monitoring board stopped the study, and only preliminary results are available.27 The available data reveal that the study included a relatively low percentage of patients with NYHA class IV symptoms (8 percent) and found no overall statistically significant difference in mortality between the treatment group (14.9 percent) and the placebo group (16.6 percent). There also appears to have been a differential race effect, with treatment benefit occurring in nonblacks but no benefit and even potential harm occurring in black patients. Bucindolol is not yet available for use in the United States.


A meta-analysis of 15 smaller trials plus the MDC, CIBIS and carvedilol trials included 3,023 patients.28 This analysis revealed that beta blockers reduced all-cause mortality by 32 percent (P = 0.003), reduced the combined risk of death or hospitalization because of heart failure by 37 percent (P < 0.001) and increased the ejection fraction by 29 percent (P < 10−9).


This article reviews the results of all major mortality trials published to date. When the patients in MERIT-HF and CIBIS-II are added to those in the trials covered in the meta-analysis, approximately 10,000 patients with heart failure have been enrolled in trials involving beta blockers. The safety of beta blockers has been confirmed in all of the trials. The three large trials (MERIT-HF, CIBIS-II and the U.S. Carvedilol Heart Failure Program) all demonstrated significant reductions in mortality, of a magnitude far exceeding that of even ACE inhibitors.

To date, all published mortality trials have been placebo controlled. A major comparative trial of carvedilol versus metoprolol is currently being conducted.

Using Beta Blockers in Patients with Heart Failure

Compelling evidence now exists to support the safety and efficacy of beta-blocker therapy in patients with heart failure. Guidance on the implementation of beta-blocker therapy in these patients is provided in the following sections.


Beta blocker therapy is appropriate in patients with NYHA class II or class III symptoms resulting from left ventricular systolic dysfunction. Unless contraindicated, beta blockers should be considered a mainstay of therapy in these patients to improve symptoms and mortality and to decrease hospitalizations.

None of the trials described in this article tested the use of beta blockers in patients with NYHA class I disease. Many asymptomatic patients with left ventricular dysfunction have had a myocardial infarction sometime in the past. Data on the administration of beta blockers subsequent to myocardial infarction are compelling enough to justify the use of these agents, if tolerated, in patients with NYHA class I disease. Many of these patients have other comorbid conditions, such as hypertension or angina, for which beta blockers may also be indicated.

None of the published trials have included sufficient numbers of patients with NYHA class IV symptoms to justify the safety or efficacy of beta blockers in this group. BEST was designed to better address this issue, but the trial included relatively few patients with NYHA class IV symptoms and was stopped early. The COPERNICUS trial may provide guidance once data become available. Pending release of the COPERNICUS trial results or other data, beta blockers should not be used in patients with NYHA class IV symptoms; however, these agents may be started or resumed once NYHA class IV symptoms resolve and patients are hemodynamically stable.

When beta blockers are used in patients with heart failure, they should be considered disease-modifying agents rather than “rescue” agents. Their utility lies in blocking the neurohormonal cascade that leads to progression of the disease, not in providing immediate symptomatic relief. Thus, patients should be hemodynamically stable when beta-blocker therapy is initiated. This approach provides certain pragmatic difficulties in dealing with patients who may already be taking a number of medications and may be resistant to the thought of adding another medication at a time when they seem well. In this setting, however, the risks of polypharmacy seem to be justified when compared to the mortality benefit of beta-blocker therapy.

Beta blockers appear to be effective irrespective of the etiology of the heart failure. Similarly, no age-specific phenomena have been reported. Preliminary data from BEST imply that racial differences may exist in the response of patients with heart failure to beta blockers; however, conclusions about this issue must await release of the trial results.


Currently, no data indicate the superiority of one beta blocker over another in the treatment of heart failure. Comparative trials are pending.

Carvedilol is the only agent labeled by the FDA for use in patients with heart failure. It is also the only agent that is available in the appropriate starting dosage (3.125 mg twice daily). The starting dosages for metoprolol tartrate, metoprolol succinate and bisoprolol require that the tablet in the smallest available dose size be split into fourths, which may be a cumbersome task for some patients. In addition, dividing the metoprolol succinate tablet into fourths may disrupt the delivery system, although it is not known if tablet division would have an adverse clinical impact. Metoprolol tartrate and bisoprolol are the least expensive of these agents (Table 6).

Once the patient has tolerated the starting dosage of the selected beta blocker, the dosage should be doubled every two to four weeks as tolerated. While the dosage is being titrated, the patient should be monitored for signs of worsening heart failure, hypotension or bradycardia. If symptoms develop, the dosage may need to be held at the current level or decreased; in some patients, the drug may need to be stopped. Otherwise, the dosage should be increased until the target dosage is achieved or the patient is receiving the maximal tolerated dosage, if below the target level. Once the desired dosage has been reached, no further adjustments need to be made. Even if the patient's symptoms stabilize or the ejection fraction normalizes, most experts recommend continuing beta-blocker therapy indefinitely.

DrugStarting dosage*Target dosageCost (generic)
Bisoprolol (Zebeta)1.25 mg per day10 mg per day$ 35
Metoprolol tartrate (Lopressor)12.5 mg per day50 to 75 mg twice daily42 (26)
Metoprolol succinate (Toprol XL)12.5 mg per day200 mg per day52
Carvedilol (Coreg)3.125 mg twice daily25 to 50 mg twice daily93


In all of the published mortality trials, beta blockers were added to background therapy with ACE inhibitors, diuretics and, sometimes, digoxin (Lanoxin). All patients for whom beta-blocker therapy is indicated should also be taking an ACE inhibitor as tolerated, unless contraindications exist. Diuretics should be titrated as needed for symptoms of volume overload. Digoxin may be used to improve symptoms, but it has never been shown to improve mortality. Spironolactone (Aldactone) has recently been shown to improve mortality in heart failure and its use is indicated in patients with systolic dysfunction who have symptoms at rest or a recent history of symptoms at rest.13

The long list of potential medications poses challenges for both the physician and the patient. Close surveillance for volume, electrolyte and hemodynamic changes is essential. The frequency of office monitoring is based on the characteristics and needs of the individual patient.


Beta blockers should not be administered to patients with heart failure who have bradycardia, heart block or hemodynamic instability. Patients hospitalized for heart failure may receive beta blockers only after they have been stabilized. Patients with severe asthma should not be given beta blockers, although those with milder symptoms may be able to tolerate these medications.

Continue Reading

More in AFP

More in Pubmed

Copyright © 2000 by the American Academy of Family Physicians.

This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP.  See permissions for copyright questions and/or permission requests.