Clinical Review Regular review

Drug treatment in heart failure

BMJ 2000; 320 doi: (Published 29 April 2000) Cite this as: BMJ 2000;320:1188
  1. Eva Lonn, associate professor of medicine (lonnem{at},
  2. Robert McKelvie
  1. Hamilton Health Sciences Corporation, General Site, McMaster Clinic, Hamilton, Ontario L8L 2X2, Canada
  1. Correspondence to: E Lonn

    Heart failure represents a complex clinical syndrome characterised by abnormalities of left ventricular function and neurohormonal regulation, exercise intolerance, shortness of breath, fluid retention, and reduced longevity.1 Despite improvements in treatment the prognosis for patients with heart failure remains poor: the risk of death annually is 5%-10% in patients with mild symptoms and 30%-40% in those with advanced disease.2 3 This condition is also associated with major morbidity and healthcare expenditure, being responsible for about 5% of hospital admissions in the United Kingdom.4

    Box 1: Treatment of heart failure


    • Relief of symptoms

    • Improvement in quality of life

    • Prevention of admissions to hospital, recurrent ischaemic events, and further deterioration in left ventricular function

    • Reduction in mortality

    Treatment modalities

    • Non-pharmacological treatments, such as exercise rehabilitation, dietary and psychological interventions, and patient education including self care strategies

    • Non-pharmacological treatments, such as exercise rehabilitation, dietary and psychological interventions, and patient education including self care strategies

    • Surgical and device based interventions

    Mitral valve surgery

    Coronary revascularisation

    Surgical ventricular remodelling procedures


    Dual chamber pacing

    Implantable cardioverter defibrillator treatment

    Ventricular assist devices

    Artificial heart

    Heart transplantation

    This review deals only with pharmacological treatments in chronic heart failure. Non-pharmacological measures apply to all patients, whereas surgical and device treatments (many still experimental) apply only to specific patient subsets. Patients with clinical symptoms of heart failure but normal or near normal left ventricular systolic function often have impaired left ventricular diastolic function. This heterogeneous group has been generally excluded from heart failure trials. We do not discuss the treatment of diastolic left ventricular dysfunction or acute heart failure syndromes: more comprehensive reviews are available.5

    Summary points

    The prognosis for patients with heart failure remains poor

    Drugs clearly shown to improve survival in patients with heart failure are ACE inhibitors and βblockers

    These drugs should be used in most patients with heart failure but require careful titration and monitoring

    Diuretics and digoxin frequently lead to improvement of symptoms

    Class I antiarrhythmic agents, non-glycoside positive inotropic drugs, and calcium channel blockers with negative inotropic actions are associated with increased risk of death, and they should be avoided in patients with heart failure


    This review is based primarily on randomised clinical trials of drug treatments in chronic heart failure. We selected original articles from Medline (1966-99) published in quality journals using the keywords heart failure, congestive, therapy, and randomised controlled trials, and we studied meta-analyses and major reviews in heart failure.

    Pharmacological treatments

    Drugs for all patients with heart failure without contraindications


    Diuretics are essential for the symptomatic management of patients with lung congestion or peripheral oedema from fluid overload. Generally, they have not been tested in long term trials evaluating survival, mainly because of difficulties in enrolling patients with symptoms who would benefit from these agents. Many short term trials have shown the ability of diuretics to decrease the physical signs of fluid retention, and a few short term studies have shown improvements in cardiac function, symptoms, and exercise tolerance. In severe heart failure thiazides as well as metolazone have a synergistic effect with loop diuretics and may be used in combination. Diuretics are generally prescribed for all patients with heart failure who have evidence of fluid retention, such as pulmonary congestion, raised jugular venous pressure, peripheral oedema, or ascites. Treatment is generally initiated with low doses of a diuretic, and the dose is increased until urine output increases and weight decreases. In patients who fail to respond readily even to high doses of these drugs intravenous administration or the combination of two or more diuretics, or both, should be considered. Once fluid retention has resolved, treatment with diuretics is generally maintained to prevent recurrent fluid retention, and this may require frequent adjustments to the dose. Keeping a daily weight record is useful in allowing the doctor, trained nurses, or patients to make such adjustments and to prevent severe states of volume overload. In using diuretics, specific attention needs to be directed towards prevention of side effects including electrolyte imbalances, activation of neurohormonal systems, hypotension, and azotaemia, which may occur with excessive use of these agents, and interactions with other drugs (for example, hyperkalaemia) when using potassium sparing diuretics in combination with ACE inhibitors. Diuretics should not be used as monotherapy even if symptoms are controlled. Instead diuretics should generally be combined with an angiotensin converting enzyme inhibitor and often a β blocker.

    Box 2: Examples of diuretics commonly used in heart failure

    • Loop diuretics



      Etacrinic acid

    • Thiazides




    • Potassium sparing agents




    Angiotensin converting enzyme inhibitors

    Angiotensin converting enzyme (ACE) inhibitors prevent the conversion of angiotensin I to angiotensin II and cause the degradation of bradykinin. Although it is uncertain how these actions contribution towards the benefits observed with ACE inhibitors, it is believed that modulation of the renin-angiotensin system and the potentiating effects of bradykinin are important. ACE inhibitors therefore alter the balance between the vasoconstrictive, salt retentive, and hypertrophic properties of angiotensin II and the vasodilatory and natriuretic properties of bradykinin.

    In large randomised clinical trials ACE inhibitors have been shown to prolong survival and reduce the need for hospital admissions in patients with heart failure. In many trials treatment with these agents was also associated with relief of dyspnoea and improved exercise tolerance, ventricular function, and the need for emergency care for worsening heart failure. The landmark cooperative north Scandinavian enalapril survival study (CONSENSUS-I) showed a 40% reduction in mortality and reductions in heart size and the need for diuretics in patients with severe heart failure (New York Heart Association functional class IV) treated with enalapril (up to 40 mg/day) for an average of six months.6 Subsequent trials conducted in patients with less severe symptoms of chronic heart failure reported as well significant survival benefits and improvement in symptoms in patients treated with ACE inhibitors. The largest of these studies was the study of left ventricular dysfunction (SOLVD) treatment, which showed clear benefits of enalapril (up to 20 mg/day) in patients with mild-moderate symptoms of heart failure and a 35% or less left ventricular ejection fraction. A meta-analysis of 32 randomised clinical trials, which collectively enrolled 7105 patients with heart failure showed a 23% (95% confidence interval 12% to 23%; P=<0.001) reduction in total mortality and a 35% reduction in the combined end point of mortality or admission to hospital for heart failure (26% to 42%; P 0.001) in those treated with ACE inhibitors versus placebo.7 8 Clinical trials in survivors of recent myocardial infarction with low left ventricular ejection fraction with or without overt symptoms of heart failure, including the survival and ventricular enlargement (SAVE) trial,9 the acute infarction ramipril efficacy (AIRE) study,10 and the trandolapril cardiac evaluation (TRACE) study,11 showed also important reductions in mortality and decreased risk of developing severe heart failure in patients treated with ACE inhibitors. Some of these studies also showed reductions in subsequent ischaemic events, acute myocardial infarction, and admission to hospital for unstable angina associated with treatment with ACE inhibitors.12 Asymptomatic patients with significant impairment in left ventricular systolic function benefit as well from long term treatment with ACE inhibitors. In the SOLVD study enalapril given to patients with 35% or less left ventricular ejection fraction who were largely asymptomatic reduced the development of symptoms of heart failure and related admissions to hospital compared with those treated with placebo, although the study did not show a major effect on mortality.13

    These data provide persuasive evidence for the routine use of ACE inhibitors in all patients with symptoms of heart failure due to left ventricular systolic dysfunction in the absence of contraindications.

    Contraindications to the use of these inhibitors include major adverse effects such as angio-oedema and anuric renal failure during prior exposure, pregnancy, very low systolic blood pressure, profoundly increased serum creatinine or potassium concentrations, and bilateral renal artery stenosis.

    Major adverse effects of angiotensin converting inhibitors, such as hypotension, syncope, renal insufficiency, hyper kalaemia, and angio-oedema are fairly infrequent and can generally be avoided by careful initiation of treatment and subsequent regular monitoring of blood pressure, electrolytes, and renal function. Avoidance of the aggressive use of diuretics, particularly potassium sparing diuretics, during initiation of treatment with ACE inhibitors is generally recommended. Among other side effects, cough occurs in 5%-15% of treated patients and is the most common reason for drug withdrawal. To minimise the potential for side effects treatment should be initiated at low doses (for example, captopril 6.25mg three times daily, enalapril 2.5 mg twice daily, and lisinopril 2.5 mg once daily) followed by gradual increments in dose with monitoring of potassium concentrations, renal function, and blood pressure. The choice of the specific ACE inhibitor and the dose targeted should generally be guided by the large scale clinical trials. Thus, specific agents and doses used in clinical trials in heart failure include captopril 150 mg daily, enalapril 20-40 mg daily, and lisinopril 20-40 mg daily. Other agents approved for the treatment of heart failure include fosinopril and ramipril (the latter for heart failure after acute myocardial infarction). The question of low versus high dose treatment with ACE inhibitors was recently addressed in the assessment of treatment with lisinopril and survival (ATLAS) trial in over 3000 patients with New York Heart Association class II, III, and IV symptoms and a 30% or less left ventricular ejection fraction. In this study high dose lisinopril (32.5-35 mg daily) was superior to low dose lisinopril (2.5-5 mg daily) in decreasing the combined endpoint of death and need for hospital admission (risk reduction 14%; P=0.002), although there was no statistically significant reduction in all cause mortality and cardiovascular mortality. The high dose lisinopril regimen was generally equally well tolerated as the low dose treatment.14

    In summary, ACE inhibitors have been evaluated extensively in randomised controlled trials in a large number of patients with heart failure and were found to reduce mortality and morbidity. Benefits extend to different patient groups, such as elderly patients, women, patients with or without coronary artery disease, with different degrees of functional impairment and with a history of use of diuretics and digitalis. In the absence of clear contraindications these drugs should be used therefore as first line agents in all patients with left ventricular dysfunction who do or do not have symptoms.

    Embedded Image



    βBlockers act by inhibiting the activation of β adrenoreceptors, and their beneficial effects in heart failure are likely related to the prevention of the deleterious actions of chronic increased adrenergic stimulation on the failing heart. These agents have negative inotropism and have been traditionally contraindicated in heart failure. None the less, their use in the treatment of selected patients with heart failure has been advocated for more than two decades. More recently, randomised controlled trials have shown the benefits of β blockers on cardiac function, symptoms, exercise performance, and survival. The metoprolol in dilated cardiomyopathy (MDC) trial reported important improvements in left ventricular ejection fraction and exercise tolerance and a trend towards a decrease in deaths or the need for cardiac transplantation in individuals with idiopathic dilated cardiomyopathy treated with the selective β1receptor inhibitor metoprolol.15 The first cardiac insufficiency bisoprolol study (CIBIS-I) found a 34% reduction in hospital admission rates for patients with heart failure treated with bisoprolol, a highly selective antagonist ofβ1adrenoreceptors, and major improvements in functional status.16 The larger CIBIS-II trial enrolled over 2600 patients with New York Heart Association class III and IV symptoms of heart failure and a 35% or less left ventricular ejection fraction treated with diuretics and angiotension converting enzyme inhibitors and showed a noticeable 34% reduction in mortality (P<0.0001) and a 20% decrease in the risk of admissions to hospital for any reason (P=0.006) in those treated with bisoprolol for a mean of 1.3 years.17 Preliminary analyses of the metoprolol CR/XL randomized intervention trial in heart failure (MERIT-HF) in almost 4000 patients with dominantly mild and moderate symptoms of heart failure (New York Heart Association functional class II-III) of ischaemic or non-ischaemic cause indicate similar mortality benefits in patients treated with sustained release metoprolol up to 200 mg/day (SS Gottlieb, personal communication). Carvedilol, a non-selective β blocker without intrinsic sympathomimetic activity and possessing peripheral vasodilating activity owing to γ1 adrenergic blockade, as well as antioxidant effects, was evaluated in the carvedilol heart failure studies in the United States, the first to show a mortality advantage with β blockers given in chronic heart failure.18 This study represented the pooling of four trials in a prospectively planned survival analysis, including a total of 1094 patients primarily with New York Heart Association functional class II and III symptoms and a 35% or less left ventricular ejection fraction. The results of the carvedilol studies are further strengthened by similar findings in the Australia and New Zealand Heart Failure Research Collaborative Group study of carvedilol in over 400 patients.19 Although this study was not powered to show clear mortality benefits, it did show improvements in cardiac function and decreased rates of admissions to hospital in patients taking carvedilol. Several meta-analyses of trials of β blockers in heart failure also showed improvement in survival, measures of cardiac function, and symptoms in patients treated with β blockers.20

    In summary, β blockers have been evaluated in nearly 10 000 patients with chronic heart failure in over 20 randomised clinical trials and have been shown to decrease the risk of death and the need for hospital admission and to improve patients' clinical status. These benefits were attained in patients already taking ACE inhibitors and diuretics (with or without digitalis), with ischaemic and non-ischaemic disease and with different degrees of functional impairment, although clinical trials have generally carefully selected stable patients. Therefore all stable patients with New York Heart Association class II and III symptoms and mild-moderate impairment in left ventricular function should be considered for β blocker therapy in the absence of contraindications and in general after therapy with diuretics and ACE inhibitors has been optimised. Treatment should not be initiated in patients with congestive symptoms and clinical evidence of fluid overload and is also contraindicated in the presence of bronchospastic disease, symptomatic bradycardia or advanced heart block, and very low blood pressure. Low dose treatment should be given initially (for example, carvedilol 3.125 mg twice daily, metoprolol 6.25-12.5 mg twice daily, or once daily for the sustained release preparation, bisoprolol 1.25 mg daily)followed by slow 2-4 weekly doubling of the dose until the target or highest tolerated dose is attained. Target doses include those used in clinical trials, such as carvedilol 25 mg twice daily, metoprolol 150 mg/day in divided doses, or metoprolol sustained release 200 mg/day, or bisoprolol 10 mg/day. Even in carefully selected and treated patients there is often an initial deterioration in symptoms, followed frequently by subsequent improvement. Major side effects, such as hypotension, fluid retention, worsening heart failure, and bradyarrhythmias, are often avoided by careful patient selection, dose titration, and monitoring. Owing to the complexities of initiating, titrating, and monitoring treatment with β blockers in heart failure, this treatment should generally be given only by experienced doctors or in specialised clinics. The choice of β blocker (selective versus non-selective versus carvedilol) remains controversial and is currently under investigation in randomised trials.


    Digitalis glycosides are mild positive inotropic agents and they act also by increasing the delivery of sodium to the distal tubules leading to the suppression of renin secretion from the kidneys and by parasympathetic activation with resultant increased vagal tone. Although widely prescribed in the treatment of heart failure for more than two centuries, until recently controversy has surrounded the use of these agents, particularly in the absence of atrial fibrillation. When atrial fibrillation is present in patients with heart failure cardiac glycosides are clearly indicated to lower ventricular response rates. In the absence of atrial fibrillation the use of digoxin and digitoxin has been more controversial and until recently it was evaluated in randomised controlled trials with major methodological flaws. Some of these studies, such as the milrinone-digoxin study, the prospective randomised study of ventricular failure and the efficacy of digoxin (PROVED) study, and the randomised assessment of digoxin on inhibitors of the angiotensin converting enzyme (RADIANCE)trial suggested that digoxin may prevent clinical worsening and improve symptoms and left ventricular function.21 22 23 These studies, however, were all short term and importantly utilised a withdrawal design in evaluating digoxin. More recently the Digoxin Investigators Group (DIG) study has evaluated the use of digoxin in heart failure in the absence of atrial fibrillation in a randomised controlled trial that is adequately designed and powered to provide data on symptoms and long term survival.24 In this study over 6800 patients with heart failure were randomised and followed for a mean of 37 months. Treatment with digoxin did not result in a decrease in all cause mortality. There was, however, a a significant 28% reduction in admissions to hospital for worsening heart failure, although total hospital admissions were reduced only by 6%. Overall, it seems reasonable to continue to use digoxin to improve the clinical status of patients with heart failure, especially in those whose symptoms persist after receiving drugs with proved efficacy in reducing the risk of death and admissions to hospital—namely, ACE inhibitors and β blockers. Treatment with digoxin is commonly initiated and maintained at 0.25 mg/day, although lower doses may be appropriate in elderly patients and those with impaired renal function. Side effects such as cardiac arrhythmias, anorexia, nausea and vomiting, and neurological complaints are commonly associated with serum digoxin concentrations of more than 2 ng/ml and are more likely to occur in the presence of electrolyte abnormalities, hypothyroidism, or concomitant use of other cardiac drugs including amiodarone, verapamil, spironolactone, and quinidine.

    Embedded Image


    Preventive treatments

    Aggressive treatment of hypertension and hyperlipidaemia can be effective in the prevention of heart failure and should be implemented in all patients with left ventricular dysfunction.25 26

    Several drugs (or combinations of drugs) can be used in selected patients with heart failure. Thus, angiotensin II receptor blockers or combined treatment with hydralazine and isosorbide dinitrate can be considered for patients who cannot tolerate or have contraindications to ACE inhibitors. Aldosterone was recently shown to improve considerably the outcomes of patients with severe heart failure. Amiodarone and chronic anticoagulation may also be of use in selected patients with heart failure.

    Although the consistent use of drugs proved to be beneficial can improve patient outcomes, it is equally important to carefully avoid the use of pharmacological interventions associated with increased risk of death in patients with heart failure. Such interventions include non-glycoside positive inotropic drugs, class I antiarrhythmic agents, and calcium channel antagonists with potent negative inotropic properties (for example, nifedipine, diltiazem, and verapamil).


    • Competing interests None declared.

    • Embedded Image Drugs recommended for use in selected patient groups appear on the BMJ' swebsite


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.
    21. 21.
    22. 22.
    23. 23.
    24. 24.
    25. 25.
    26. 26.