brand logo

Several large clinical trials conducted over the past decade have shown that pharmacologic interventions can dramatically reduce the morbidity and mortality associated with heart failure. These trials have modified and enhanced the therapeutic paradigm for heart failure and extended treatment goals beyond limiting congestive symptoms of volume overload. Part II of this two-part article presents treatment recommendations for patients with left ventricular systolic dysfunction. The authors recommend that, if tolerated and not contraindicated, the following agents be used in patients with left ventricular systolic dysfunction: an angiotensin-converting enzyme inhibitor in all patients; a beta blocker in all patients except those who have symptoms at rest; and spironolactone in patients who have symptoms at rest or who have had such symptoms within the past six months. Diuretics and digoxin should be reserved, as needed, for symptomatic management of heart failure. Other treatments or treatment programs may be necessary in individual patients.

Part I of this guideline detailed how the therapeutic paradigm of heart failure has changed over the past decade.1 This new understanding of heart failure has led to new treatment approaches. The end point of treatment is no longer merely the limiting of congestive symptoms. Rather, the new standard for treatment also includes decreasing mortality.

Part II presents evidence-based treatment options that are practical for use by primary care physicians. The treatment guideline is based on the symptomatic classification scheme described in the first part of this article.1 The methods for guideline development and the evidence weighting for the recommendations are also detailed in part I.1

Pharmacologic Therapy: Primary Drugs

In this guideline, drugs specifically indicated for use in patients with heart failure are referred to as “primary drugs.” Use of these drugs according to the new classification scheme is presented in Table 1.2 Dosing and cost information are provided in Table 2,2 and “clinical pearls” are given in Table 3.2 A comparison of the guideline’s recommendations with those from other recent guidelines is presented in Table 4.35 Common drug interactions are listed in Table 5.2


Angiotensin-converting enzyme (ACE) inhibitors are indicated in the treatment of all patients with systolic heart failure. A number of landmark randomized, controlled trials68 have demonstrated the effectiveness of these drugs in reducing morbidity and mortality in asymptomatic and symptomatic patients. Unless absolutely contraindicated, ACE inhibitor therapy should be considered a priority (level of evidence: A—see Table 1 in part I).

Because of perceived risks and contraindications, ACE inhibitors are often avoided in patients with heart failure. Renal dysfunction and cough should not be considered absolute contraindications to the use of these agents. When the systolic blood pressure is less than 100 mm Hg or the creatinine level is elevated, careful monitoring is warranted during the initiation of ACE inhibitor therapy.

The dosing of ACE inhibitors is controversial. Target dosages that reflect those used in mortality trials are provided in Table 2.2 The table includes only agents that have been tested in mortality trials. The Assessment of Treatment with Lisinopril and Survival (ATLAS) Study9 sought to explore whether higher dosages of ACE inhibitors would be more effective. However, the ATLAS study compared high dosages to low dosages, and there was no comparison to the target dosages that had been shown to be effective in mortality trials.10 At present, we cannot support the routine use of high dosages of ACE inhibitors as being more effective than the target dosages listed in Table 22 (level of evidence: D).


Beta-blocker therapy is indicated in all patients with systolic heart failure except those with dyspnea at rest, those who are unable to tolerate beta blockers and those who are hemodynamically unstable (level of evidence: A). Randomized, controlled trials1113 have shown that treatment with carvedilol, metoprolol succinate (controlled release) or bisoprolol decreases mortality by at least 34 percent in patients with heart failure. These agents differ pharmacologically; however, they all have demonstrated benefit, and any of the three can be used. Comparative trials of various beta blockers are currently being conducted.

Asymptomatic*ACE inhibitor
Beta blocker
Symptomatic†ACE inhibitor
Beta blocker
If symptoms persist: digoxin (Lanoxin)
Symptomatic with recent history of dyspnea at rest‡Diuretic
ACE inhibitor
Spironolactone (Aldactone)
Beta blocker
Symptomatic with dyspnea at rest§Diuretic
ACE inhibitor

Although beta blockers must be administered cautiously to patients with heart failure, primary care physicians can prescribe them safely (Table 3).2 The initial dosage should be doubled every two to four weeks until the patient is unable to tolerate higher levels or the target dosage is reached (Table 2).2 Patients who develop hypotension, symptoms of hypotension, increasing dyspnea or worsening heart failure should be evaluated. It may be necessary to increase the diuretic dosage, decrease the beta-blocker dosage or discontinue the beta blocker.

Beta blockers should only be added when patients are clinically stable. The purpose of beta-blocker therapy is to slow the progression of the disease. Beta blockers are not to be added as rescue drugs in patients who are decompensating.

Most patients with known asymptomatic left ventricular dysfunction have also had a myocardial infarction. The benefits of beta blockers in such patients have been well described, and these agents should be administered14 (level of evidence: A). Many other patients in this group may have hypertension or other indications for beta-blocker therapy. No comparable data are available on asymptomatic patients with idiopathic heart failure. We recommend at least consideration of beta-blocker therapy in asymptomatic patients (level of evidence: D).

Some uncertainty still exists about the safety and efficacy of beta blockers in patients with dyspnea at rest. Two large trials—the Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS) trial15 and the Beta Blocker Evaluation Survival Trial (BEST)16—were designed to address this issue.

DrugStarting dosageTarget dosage or common dosageCost for brand name (generic)*
Decrease mortality and improve symptoms
ACE inhibitors‡
Captopril (Capoten)6.25 mg three times daily (one-half tablet§)12.5 to 50 mg three times daily$ 82 (52 to 58)
Enalapril (Vasotec)2.5 mg twice daily10 mg twice daily68 (46 to 48)
Lisinopril (Zestril)5 mg daily10 to 20 mg daily29
Ramipril (Altace)1.25 mg twice daily5 mg twice daily60
Trandolapril (Mavik)1 mg daily4 mg daily22
Aldosterone antagonist‡
Spironolactone (Aldactone)25 mg daily25 mg daily14 (12 to 13)
Beta blockers‡
Bisoprolol (Zebeta)1.25 mg daily (one-fourth tablet§)10 mg daily36
Carvedilol (Coreg)3.125 mg twice daily25 to 50 mg twice daily95
Metoprolol tartrate (Lopressor)12.5 mg twice daily (one-fourth tablet§)50 to 75 mg twice daily42 (26 to 31)
Metoprolol succinate (Toprol-XL)12.5 mg daily (one-half tablet§)200 mg daily52
Treat symptoms‖
Thiazide diuretics¶
Hydrochlorothiazide (Esidrex)25 mg daily25 to 100 mg daily5 (1 to 3)
Metolazone (Zaroxolyn)2.5 mg daily2.5 to 10 mg daily20
Loop diuretics¶
Bumetanide (Bumex)1 mg daily1 to 10 mg once to three times daily14 (12 to 13)
Ethacrynic acid (Edecrin)25 mg daily25 to 200 mg once or twice daily10
Furosemide (Lasix)40 mg daily40 to 400 mg once to three times daily8 (4 to 5)
Torsemide (Demadex)20 mg daily20 to 200 mg once or twice daily23
Digoxin (Lanoxin#)0.125 mg daily0.125 to 0.375 mg daily6 (3 to 5)

The data safety monitoring board stopped the COPERNICUS trial early because of significant improvements in mortality.15 Over-all, carvedilol reduced mortality by 35 percent over placebo. However, these data have not yet been published, and some controversy exists concerning the exclusion criteria used in patient selection. Patients who had overt volume overload and patients who recently required intravenous inotropic agents were among those who were excluded from the COPERNICUS trial.

Drug categoryClinical pearls
ACE InhibitorsACE inhibitors appear to exhibit a class effect. All members of this class may be equally effective.
The ideal dosage of ACE inhibitors is controversial.*
Beta blockersBeta blockers should not be given to patients with absolute contraindications to their use, such as heart block or bradycardia.
Beta blockers should not be used in patients who have dyspnea at rest or who are hemodynamically unstable, or as initial therapy in patients who are hospitalized for heart failure. Once these issues have resolved, beta blockers may be added to the chronic treatment regimen. Beta blockers may be added when patients are stable. They may be used in patients with heart failure caused by systolic dysfunction who do not have contraindications to their use.
Beta blockers are to be added to arrest the progression of the disease; they are not to be added as rescue therapy for patients who are decompensating.
Beta blockers should be initiated in a low dosage; the dosage should then be doubled every 2 to 4 weeks as tolerated until the target dosage is reached (see Table 2).
No clinical differences have been demonstrated among beta blockers that have a proven mortality benefit in the treatment of heart failure.
DiureticsDiuretics are included as background therapy. Although diuretics have not been specifically tested in clinical trials, they should still be used as needed for the treatment of volume overload. Diuretics have consistently been part of background therapy in all published placebo-controlled mortality trials of ACE inhibitors, beta blockers and spironolactone (Aldactone) in the treatment of symptomatic patients.
ACE inhibitors and beta blockers may reduce the need for diuretic therapy.
Combining drugsMany patients with heart failure may benefit from both beta blockers and ACE inhibitors. Which to start first, or which to use if the patient’s blood pressure will not tolerate both, is controversial.†
The drug therapies listed in Table 1 are the desired end points for patients in the symptomatic classes. No data are available to indicate how best to introduce all of these medications. In all major trials, a beta blocker or spironolactone was added to background therapy of ACE inhibitors, diuretics and, sometimes, digoxin (Lanoxin).
ACE inhibitors, beta blockers and spironolactone should not be stopped if symptoms improve, because these medications slow disease progression and decrease mortality.
ReferralConsider referral in the following clinical situations:
Diagnostic procedure
Revascularization procedure
Worsening or refractory heart failure
Ventricular arrhythmia
Valvular heart disease
Consideration of transplantation

BEST was stopped early without demonstration of a significant mortality benefit.16 The results of this study also remain unpublished, and information is available only from early data releases. The BEST study investigated the beta blocker bucindolol, and some researchers believe that the results of the trial may be specific to this drug. Racial differences were also discovered in BEST, and it is unclear whether the high number of black patients, whose outcomes were the worst in the trial, may have affected the overall outcome. Also, it is possible that the negative result came from the inclusion of a relatively high number of patients with dyspnea at rest.

Pending complete review of the data, we cannot recommend the use of beta blockers in all patients with dyspnea at rest (level of evidence: D). There may be instances in which, with appropriate surveillance, beta blockers may be used safely in these patients.

Beta blockers should not be given if their use is absolutely contraindicated because of conditions such as bradycardia, heart block or severe bronchospastic disease. Diabetes and mild asthma should not be considered absolute contraindications to the use of these drugs. Patients with these comorbid conditions should be given a trial of beta blockers.


Spironolactone is a potassium-sparing diuretic and an aldosterone antagonist. The Randomized Aldactone Survival Study (RALES)5 recently demonstrated that aldosterone antagonism with spironolactone lowers mortality and hospitalization rates in patients with systolic dysfunction who have dyspnea at rest or a history of dyspnea at rest within the past six months. Significant hyperkalemia was rare (incidence: 0.5 percent), even though patients were also receiving ACE inhibitors. Asymptomatic gynecomastia was also relatively uncommon (7 percent) when the drug was given in an average dosage of 25 mg daily. Patients with heart failure, dyspnea at rest (current or recent) and no significant renal impairment (serum creatinine level of less than 2.5 mg per dL [221 mmol per L]), should be treated with spironolactone in a dosage of 25 mg daily (level of evidence: A).

Drug classUniversity of MichiganCONSENSUS–HF3 Heart Failure Society of America4
ACE inhibitorsAll patients with systolic dysfunctionAll patients with systolic dysfunctionAll patients with systolic dysfunction
Beta blockers†All patients except those with dyspnea at restAll patients in NYHA class II or IIIAll patients in NYHA class II and III
Await additional data on NYHA class IVInsufficient evidence to recommend for patients in NYHA class IV
No comment made on NYHA class IConsidered for patients in NYHA class I
Spironolactone (Aldactone)Patients with dyspnea at rest or recent history of dyspnea at rest“Consideration in patients with recent or current class IV symptoms”3(p27A)“Considered for patients receiving standard therapy who have severe heart failure (recent or recurrent class IV)”4(p374)
DiureticsAll symptomatic patients, dosed as necessary to control symptomsAll symptomatic patients, dosed as necessary to control symptomsAll symptomatic patients, dosed as necessary to control symptoms
Digoxin (Lanoxin)As needed in patients with persistent symptoms or recurrent hospitalizations; used, as indicated, in conjunction with ACE inhibitors, beta blockers, diuretics and aldosterone antagonists“To improve the clinical status of patients with heart failure in conjunction with diuretics, an ACE inhibitor, and a beta blocker”3(p23A) “Considered for patients who have symptoms of heart failure and NYHA class IV…while receiving standard therapy”4(p366–8)

To minimize the possibility of hyperkalemia, primary care physicians should be cognizant of the risk factors for this disorder, which include decreased renal function, potassium supplementation and the use of ACE inhibitors or adrenergic receptor blockers. As a diuretic, spironolactone may also affect fluid balance, which should be monitored.

Some experts believe that spironolactone also may be beneficial in patients with less severe heart failure and in those with diastolic disease. However, no large-scale clinical trials have addressed the safety or efficacy of spironolactone therapy in these patients.


Although no large, controlled clinical studies have been conducted on the use of diuretics in the treatment of heart failure, these drugs have been given to most patients as part of baseline therapy in trials of ACE inhibitors, beta blockers, spironolactone (Aldactone) and digoxin (Lanoxin).

Drug or drug classInteracting drugsEffect
ACE inhibitorsAntacidsDecreased drug absorption
LithiumIncreased lithium levels
NSAIDMay decrease renal function
Spironolactone (Aldactone)With co-administration, may result in elevated potassium levels, especially in the elderly and in patients with renal dysfunction
Amiodarone (Cordarone)Beta blockersDecreased heart rate and atrioventricular node conduction
Calcium channel blockers (e.g., diltiazem [Cardizem] and verapamil [Calan])Decreased heart rate and atrioventricular node conduction
Digoxin (Lanoxin)Increased digoxin concentration, decreased heart rate and atrioventricular node conduction
QuinidineIncreased quinidine concentration
Phenytoin (Dilantin)Increased phenytoin concentration, decreased amiodarone concentration
Procainamide (Pronestyl)Increased procainamide concentration
TheophyllineIncreased theophylline concentration
Warfarin (Coumadin)Increased INR
Beta blockersAmiodarone, diltiazem, verapamil, propafenone (Rythmol), sotalol (Betapace)Decreased heart rate and atrioventricular node conduction
DigoxinAmiodaroneIncreased digoxin concentration, decreased heart rate and atrioventricular node conduction
AntacidsDecreased digoxin absorption (space administration of drugs at least 2 hours apart)
Beta blockersCarvedilol (Coreg) may increase digoxin concentration; decreased heart rate and atrioventricular node conduction
Cholestyramine (Questran), colestipol (Colestid)Decreased digoxin absorption
Diltiazem, verapamilIncreased digoxin concentration, decreased heart rate and atrioventricular node conduction
Omeprazole (Prilosec)Increased digoxin concentration
PropafenoneIncreased digoxin concentration, decreased heart rate and atrioventricular node conduction
QuinidineIncreased digoxin concentration
Rifampin (Rifadin)Decreased digoxin concentration
SotalolDecreased heart rate and atrioventricular node conduction
SpironolactoneIncreased digoxin concentration; interferes with some digoxin assays, yielding falsely elevated digoxin concentrations
WarfarinAmiodarone, antibiotics (including trimethoprim-sulfamethoxazole [Bactrim, Septra] and erythromycin), antidepressants, beta blockers, cimetidine (Tagamet), fluconazole (Diflucan), itraconazole (Sporanox), ketoconazole (Nizoral), lovastatin (Mevacor), omeprazole, oral diabetic agents, phenytoin, propafenone, quinidine, quinine, simvastatin (Zocor)Increased INR
NSAIDs, aspirin, ticlopidine (Ticlid), clopidogrel (Plavix)Increased risk of bleeding because of effect on platelet function
Phenobarbital, rifampin, cholestyramine, carbamazepine (Tegretol), phenytoin, spironolactone, sucralfate (Carafate)Decreased INR

Loop diuretics are the most potent agents in the diuretic class, but they are associated with acute and chronic distal tubular compensation. Combining a loop diuretic with a thiazide diuretic increases diuretic potency by minimizing distal tubular compensation. Both loop and thiazide diuretics produce urinary potassium and magnesium wasting.

Diuretics should be used as needed to treat volume overload. Diuretic therapy may be needed acutely and chronically (level of evidence: D).


Digoxin is indicated in the treatment of patients with heart failure who also have atrial fibrillation (level of evidence: A). It is also indicated to improve symptoms and decrease hospitalization rates in patients with symptomatic heart failure (level of evidence: A). The appropriate dosage is 0.25 mg daily; the dosage can be adjusted as needed, based on symptoms, other drugs or renal impairment (Table 2).2

Digoxin has been shown to have an impact on symptoms and hospitalization rates, but not on mortality.17 The number of drugs that have a beneficial impact on mortality in heart failure is expanding. Because taking an increasing number of medications can become a barrier to compliance, the role that digoxin will ultimately play in the treatment of heart failure is unclear.

Currently, we recommend the use of digoxin in patients who are symptomatic despite treatment with diuretics, ACE inhibitors and beta blockers. We also recommend digoxin therapy for patients with dyspnea at rest or a recent history of dyspnea at rest.

Pharmacologic Therapy: Secondary Drugs

Drugs occasionally used in patients with heart failure but not specifically recommended for use in this setting are referred to as “secondary drugs” in this guideline.


In combination, the direct-acting vasodilators isosorbide dinitrate (Isordil) and hydralazine (Apresoline) were the first medications shown to improve survival in heart failure.18 Subsequently, randomized, controlled trials demonstrated that ACE inhibitors were superior to these agents.7 Guidelines from the Agency for Health Care Policy and Research19 indicate that direct-acting vasodilators can be considered a potential alternative if ACE inhibitors are poorly tolerated in patients with heart failure (level of evidence: A). A retrospective analysis20 of these clinical trials suggests that blacks may gain greater benefit from isosorbide dinitrate and hydralazine than nonblacks.


Calcium channels blockers have no direct role in the treatment of heart failure resulting from systolic dysfunction. The first-generation agents diltiazem and nifedipine were shown to have adverse outcomes in patients with systolic dysfunction who had had a myocardial infarction21,22 (level of evidence: A). Subsequent studies23,24 using the dihydropyridine calcium channel blockers (i.e., amlodipine and felodipine) in patients with New York Heart Association (NYHA) III/IV symptoms suggested that the drugs were safe but did not demonstrate their efficacy (level of evidence: A).

A subgroup analysis23 of a randomized, controlled trial involving amlodipine suggested a mortality benefit in patients with nonischemic heart failure. However, a larger subsequent study25 found no benefit in administering amlodipine to patients with nonischemic cardiomyopathy. Currently, no evidence supports the use of calcium channel blockers in patients with heart failure (level of evidence: A).


Intravenous inotropic therapy with the sympathomimetics (dobutamine [Dobutrex] or dopamine [Intropin]) or the phosphodiesterase inhibitors (milrinone [Primacor] or amrinone [Inocor]) is reserved for use in patients hospitalized for acutely decompensated heart failure who do not respond adequately or in a timely manner to diuretic therapy, although mortality data are lacking.26 Inotropic agents may increase cardiac output and decrease systemic and pulmonary vascular resistance. Intermittent bolus treatment or continuous home infusion therapy with dobutamine or milrinone is not currently indicated for the routine management of heart failure (level of evidence: C).


Anticoagulation therapy is indicated in patients with heart failure who are at risk for thromboembolism. Included in this group are patients with atrial fibrillation, demonstrated left ventricular thrombus or a history of embolic stroke with the likely source being a dilated left ventricle27,28 (level of evidence: A).

Anticoagulation therapy has also been prescribed for patients with a low ejection fraction or an intracardiac thrombus (level of evidence: C). However, data supporting the use of anticoagulant drugs for this indication are limited and controversial. The majority of studies did not correct for the presence of well-established risk factors for thrombus formation and did not control for the level of anticoagulation or the initiation of anticoagulation. If anticoagulation is needed, the appropriate dose of warfarin (Coumadin) is determined by the patient’s International Normalized Ratio (INR). The target INR will depend on the clinical condition necessitating therapy.


Angiotensin receptor blockers are safe29 but have never been shown to be more effective30 than ACE inhibitors in mortality trials of patients with heart failure. In a recently completed study,31 no improvement in all-cause mortality occurred when angiotensin receptor blockers were added to background therapy consisting of ACE inhibitors and, sometimes, beta blockers. Furthermore, a subgroup analysis31 showed a 10 percent increase in the composite end point of mortality plus morbidity among patients receiving valsartan whose baseline therapy included a beta blocker.

We recommend the use of angiotensin receptor blockers only in patients who are intolerant of ACE inhibitors. Even in this setting, we recommend that these agents be used with caution in patients who are also taking beta blockers (level of evidence: A).


Evidence is inconsistent regarding the benefit of antiarrhythmic therapy in patients with heart failure.32,33 At this time, it does not appear that substantial benefit is derived from the application of antiarrhythmic therapy in patients who have heart failure and an asymptomatic rhythm disturbance. There are benefits from use of implantable cardioverter defibrillators in selected patients. This issue is discussed in part I of this article.1

Arrhythmias such as atrial fibrillation, ventricular tachycardia and bradyarrhythmias are common in patients with heart failure. Therefore, individualized management is recommended (level of evidence: D).

Some frequently used antiarrhythmics, including calcium channel blockers, beta blockers and digoxin, are addressed in other sections of this article. Amiodarone (Cordarone), a common antiarrhythmic, has a number of drug interactions that should be noted (Table 5).2


Aspirin is used in many patients with heart failure caused by ischemic cardiomyopathy. Although patients with heart failure who are already taking aspirin may be maintained on the drug, heart failure itself is not an indication to start aspirin therapy (level of evidence: D). Potential adverse affects of aspirin on gastric mucosa and on renal function should be considered. Patients with heart failure caused by coronary artery disease should be maintained on aspirin (level of evidence: A).

There are theoretic concerns about aspirin negating some of the beneficial impact of ACE inhibitors. However, no clinical data have demonstrated a harmful effect from combining these agents.

Nonpharmacologic Treatment


Many small and sometimes randomized clinical trials have examined the benefits of exercise training in patients with heart failure.34 Different exercise formulas have been tested in men and women. The clinical trials have measured intermediate outcomes such as ventilatory capacity, maximum oxygen consumption, skeletal muscle parameters and neurohormonal levels; a few have measured exercise capacity and quality of life.

Exercise may be recommended based on clinical trials that have shown exercise to be safe under study conditions (level of evidence: D). As yet, however, no major randomized, controlled trial has demonstrated an improvement in clinical outcomes such as medication reduction, decreased hospital days or mortality.


We are unaware of any controlled clinical trials of salt or fluid restriction in the treatment of heart failure. However, dietary sodium restriction may reduce the need for diuretics in patients with congestion (level of evidence: D). The most commonly recommended limit is 2,000 mg of sodium daily. Restricting fluid intake to 2 L or less daily may be useful in patients with hyponatremia.


Some causes of heart failure are potentially reversible. Surgery should be considered in patients with heart failure resulting from aortic stenosis, but the management of this condition is beyond the scope of this article. Similarly, revascularization, if indicated, should be considered in patients with coronary artery disease (level of evidence: D).

Continue Reading

More in AFP

Copyright © 2001 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.