PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of bmjclinevidLink to Publisher's site
 
BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.
PMCID: PMC3275321

Heart failure

Robert Samuel McKelvie, BSc MSc MD PhD FRCPC, Professor of Medicine

Abstract

Introduction

Heart failure occurs in 3% to 4% of adults aged over 65 years, usually as a consequence of coronary artery disease or hypertension, and causes breathlessness, effort intolerance, fluid retention, and increased mortality. The 5-year mortality in people with systolic heart failure ranges from 25% to 75%, often owing to sudden death following ventricular arrhythmia. Risks of cardiovascular events are increased in people with left ventricular systolic dysfunction (LVSD) or heart failure.

Methods and outcomes

We conducted a systematic review and aimed to answer the following clinical questions: What are the effects of multidisciplinary interventions for heart failure? What are the effects of exercise in people with heart failure? What are the effects of drug treatments for heart failure? What are the effects of devices for treatment of heart failure? What are the effects of coronary revascularisation for treatment of heart failure? What are the effects of drug treatments in people at high risk of heart failure? What are the effects of treatments for diastolic heart failure? We searched: Medline, Embase, The Cochrane Library, and other important databases up to August 2010 (Clinical Evidence reviews are updated periodically; please check our website for the most up-to-date version of this review). We included harms alerts from relevant organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA).

Results

We found 80 systematic reviews, RCTs, or observational studies that met our inclusion criteria. We performed a GRADE evaluation of the quality of evidence for interventions.

Conclusions

In this systematic review we present information relating to the effectiveness and safety of the following interventions: aldosterone receptor antagonists, amiodarone, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, anticoagulation, antiplatelet agents, beta-blockers, calcium channel blockers, cardiac resynchronisation therapy, coronary revascularisation, digoxin (in people already receiving diuretics and angiotensin-converting enzyme inhibitors), exercise, hydralazine plus isosorbide dinitrate, implantable cardiac defibrillators, multidisciplinary interventions, non-amiodarone antiarrhythmic drugs, and positive inotropes (other than digoxin).

Key Points

Heart failure occurs in 3% to 4% of adults aged over 65 years, usually as a consequence of coronary artery disease or hypertension, and causes breathlessness, effort intolerance, fluid retention, and increased mortality.

  • The 5-year mortality in people with systolic heart failure ranges from 25% to 75%, often owing to sudden death following ventricular arrhythmia. Risks of cardiovascular events are increased in people with left ventricular systolic dysfunction (LVSD) or heart failure.

Multidisciplinary interventions may reduce admissions to hospital and mortality in people with heart failure compared with usual care. Exercise may reduce admissions to hospital due to heart failure compared with usual care. However, long-term benefits of these interventions remain unclear.

ACE inhibitors, angiotensin II receptor blockers, and beta-blockers reduce mortality and hospital admissions from heart failure compared with placebo, with greater absolute benefits seen in people with more severe heart failure.

  • Combined treatment with angiotensin II receptor blockers and ACE inhibitors may lead to a greater reduction in hospital admission for heart failure compared with ACE inhibitor treatment alone.

Aldosterone receptor antagonists (spironolactone, eplerenone, and canrenoate) may reduce all-cause mortality in people with heart failure, but increase the risk of hyperkalaemia.

Digoxin slows the progression of heart failure compared with placebo, but may not reduce mortality.

Hydralazine plus isosorbide dinitrate may improve survival and quality-of-life scores compared with placebo in people with chronic congestive heart failure.

We don't know whether amiodarone, anticoagulants, or antiplatelets are effective at reducing mortality or hospital re-admission rates.

CAUTION: Positive inotropic agents (other than digoxin), calcium channel blockers, and antiarrhythmic drugs (other than amiodarone and beta-blockers) may all increase mortality and should be used with caution, if at all, in people with systolic heart failure.

Implantable cardiac defibrillators and cardiac resynchronisation therapy can reduce mortality in people with heart failure who are at high risk of ventricular arrhythmias. However, studies evaluating cardiac resynchronisation therapy were performed in centres with considerable experience, which may have overestimated the benefits.

We don't know how coronary revascularisation and drug treatment compare for reducing mortality in people with heart failure and left ventricular dysfunction because all the trials assessing this comparison were conducted before ACE inhibitors, aspirin, beta-blockers, and statins were in routine use, thus limiting their applicability to current clinical practice.

ACE inhibitors delay the onset of symptomatic heart failure, reduce cardiovascular events, and improve long-term survival in people with asymptomatic LVSD compared with placebo.

  • Angiotensin II receptor blockers and ACE inhibitors seem equally effective at reducing all-cause mortality and cardiovascular mortality in people at high risk of heart failure.
  • The combination of angiotensin II receptor blockers and ACE inhibitors seems no more effective than ACE inhibitors alone and causes more adverse effects.

ACE inhibitors or angiotensin II receptor blockers seem no more effective at reducing mortality or rate of hospital admissions for cardiovascular events in people with diastolic heart failure compared with placebo.

About this condition

Definition

Heart failure occurs when abnormal cardiac function causes failure of the heart to pump blood at a rate sufficient for metabolic requirements under normal filling pressure. It is characterised clinically by breathlessness, effort intolerance, fluid retention, and poor survival. Fluid retention and the congestion related to this can often be relieved with diuretic therapy. However, diuretic therapy should generally not be used alone and, if required, should be combined with the pharmacological treatments outlined in this review. Heart failure can be caused by systolic or diastolic dysfunction, and is associated with neurohormonal changes. Left ventricular systolic dysfunction (LVSD) is defined as a left ventricular ejection fraction (LVEF) <0.40. It may be symptomatic or asymptomatic. Defining and diagnosing diastolic heart failure can be difficult. Proposed criteria include: (1) clinical evidence of heart failure; (2) normal or mildly abnormal left ventricular systolic function; (3) evidence of abnormal left ventricular relaxation, filling, diastolic distensibility, or diastolic stiffness; and (4) evidence of elevated N-terminal-probrain natriuretic peptide. However, assessment of some of these criteria is not standardised.

Incidence/ Prevalence

Both incidence and prevalence of heart failure increase with age. Studies of heart failure in the US and UK found annual incidence in people 45 years or over to be between 29 and 32 cases/1000 people/year, and, in those over 85 years of age, incidence was considerably higher, at 45 to 90 cases/1000 people/year. The study carried out in the US reported a decline in incidence of heart failure (all age groups) over a 10-year period, with incidence falling from 32.2 cases/1000 people/year in 1994 to 29.1 cases/1000 people/year in 2003. However, analysis of those aged 65 years or over indicated an increase in prevalence of heart failure (from 89.9 cases/1000 people in 1994 to 121 cases/1000 people in 2003). Prevalence of heart failure was higher in men (130 cases/1000 men) compared with women (115 cases/1000 women). In older people (65 years or over), incidence of heart failure after a myocardial infarction (MI) is on the rise, with one study finding an increase of 25.1% in in-hospital heart failure from 1994 through to 2000 (from 31.4% to 39.3%, P = 0.001). Furthermore, the study noted that 76% of people who survived MI had developed heart failure at 5 years' follow-up. Prevalence of asymptomatic LVSD is 3% in the general population, and the mean age of people with asymptomatic LVSD is lower than that of symptomatic individuals. Both heart failure and asymptomatic LVSD are more common in men. Prevalence of diastolic heart failure in the community is unknown. Prevalence of heart failure with preserved systolic function in people in hospital with clinical heart failure varies from 13% to 74%. Less than 15% of people with heart failure under 65 years of age have normal systolic function, whereas prevalence is about 40% in people over 65 years of age.

Aetiology/ Risk factors

Coronary artery disease is the most common cause of heart failure. Other common causes include hypertension and idiopathic dilated congestive cardiomyopathy. After adjustment for hypertension, the presence of left ventricular hypertrophy remains a risk factor for the development of heart failure. Other risk factors include cigarette smoking, hyperlipidaemia, and diabetes mellitus. The common causes of left ventricular diastolic dysfunction are coronary artery disease and systemic hypertension. Other causes are hypertrophic cardiomyopathy, restrictive or infiltrative cardiomyopathies, and valvular heart disease.

Prognosis

The prognosis of heart failure is poor, with 5-year mortality ranging from 26% to 75%. Up to 16% of people are re-admitted with heart failure within 6 months of first admission. In the US, heart failure is the leading cause of hospital admission among people over 65 years of age. In people with heart failure, a new MI increases the risk of death (RR 7.8, 95% CI 6.9 to 8.8). About one third of all deaths in people with heart failure are preceded by a major ischaemic event. Sudden death, mainly caused by ventricular arrhythmia, is responsible for 25% to 50% of all deaths, and is the most common cause of death in people with heart failure. Women with heart failure have a 15% to 20% lower risk of total and cardiovascular mortality compared with men with heart failure (risk after adjustment for demographic and social economic characteristics, comorbidities, cardiovascular treatments, and LVEF). The presence of asymptomatic LVSD increases an individual's risk of having a cardiovascular event. One large prevention trial found that the risk of heart failure, admission for heart failure, and death increased linearly as ejection fraction fell (for each 5% reduction in ejection fraction: RR for mortality 1.20, 95% CI 1.13 to 1.29; RR for hospital admission 1.28, 95% CI 1.18 to 1.38; RR for heart failure 1.20, 95% CI 1.13 to 1.26). The annual mortality for people with diastolic heart failure varies in observational studies (1–18%). Reasons for this variation include age, presence of coronary artery disease, and variation in the partition value used to define abnormal ventricular systolic function. The annual mortality for left ventricular diastolic dysfunction is lower than that found in people with systolic dysfunction.

Aims of intervention

To relieve symptoms; to improve quality of life; and to reduce morbidity and mortality with minimum adverse effects.

Outcomes

Effects of treatments in people with heart failure: mortality; functional capacity (assessed by the New York Heart Association functional classification or more objectively by using standardised exercise testing or the 6-minute walk test); hospital admission rates; quality of life (assessed with questionnaires); adverse effects of treatment. Effects of treatments in people at high risk of heart failure: mortality; cardiovascular events (including non-fatal MI and the composite outcomes of cardiovascular mortality, MI, stroke, or hospital admission); hospital admission rates; adverse effects of treatment. Proxy measures of clinical outcome (e.g., LVEF) are used only when clinical outcomes are unavailable. Where a study reported only the composite outcome of death or hospital admission, we have reported this under hospital admission.

Methods

Clinical Evidence search and appraisal August 2010. The following databases were used to identify studies for this systematic review: Medline 1966 to August 2010, Embase 1980 to August 2010, and The Cochrane Database of Systematic Reviews, August 2010 (online; 1966 to date of issue). An additional search within The Cochrane Library was carried out for the Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessment (HTA). We also searched for retractions of studies included in the review. Abstracts of the studies retrieved from the initial search were assessed by an information specialist. Selected studies were then sent to the contributor for additional assessment, using predetermined criteria to identify relevant studies. Study design criteria for inclusion in this review were: published systematic reviews of RCTs and RCTs in any language, at least single blinded, and containing >100 individuals of whom >80% were followed up. Generally, RCTs with <500 people have been excluded because of the number of large RCTs available. If, for any comparison, large RCTs or systematic reviews were found, then smaller RCTs have been excluded, even if they include >500 people. There was no minimum length of follow-up required to include studies. We excluded all studies described as "open", "open label", or not blinded unless blinding was impossible. We included systematic reviews of RCTs and RCTs where harms of an included intervention were studied applying the same study design criteria for inclusion as we did for benefits. In addition we use a regular surveillance protocol to capture harms alerts from organisations such as the FDA and the MHRA, which are added to the reviews as required. To aid readability of the numerical data in our reviews, we round many percentages to the nearest whole number. Readers should be aware of this when relating percentages to summary statistics such as relative risks (RRs) and odds ratios (ORs). We have performed a GRADE evaluation of the quality of evidence for interventions included in this review (see table). The categorisation of the quality of the evidence (high, moderate, low, or very low) reflects the quality of evidence available for our chosen outcomes in our defined populations of interest. These categorisations are not necessarily a reflection of the overall methodological quality of any individual study, because the Clinical Evidence population and outcome of choice may represent only a small subset of the total outcomes reported, and population included, in any individual trial. For further details of how we perform the GRADE evaluation and the scoring system we use, please see our website (www.clinicalevidence.com).

Table
GRADE Evaluation of interventions for Heart failure.

Glossary

High-quality evidence
Further research is very unlikely to change our confidence in the estimate of effect.
Low-quality evidence
Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Minnesota Living with Heart Failure Questionnaire
Scores range from 1 to 105, with higher scores reflecting a lower quality of life.
Moderate-quality evidence
Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
New York Heart Association functional classification
Classification of severity by symptoms. Class I: no limitation of physical activity; ordinary physical activity does not cause undue fatigue or dyspnoea. Class II: slight limitation of physical activity; comfortable at rest, but ordinary physical activity results in fatigue or dyspnoea. Class III: limitation of physical activity; comfortable at rest, but less than ordinary activity causes fatigue or dyspnoea. Class IV: unable to carry out any physical activity without symptoms; symptoms are present even at rest; if any physical activity is undertaken, symptoms are increased.
Usual or conventional care
describes the comparator arm of some controlled trials. It refers to appropriate drug and non-drug treatment, in the absence of the intervention being examined in the active treatment arm of the trial.
Very low-quality evidence
Any estimate of effect is very uncertain.

Notes

Disclaimer

The information contained in this publication is intended for medical professionals. Categories presented in Clinical Evidence indicate a judgement about the strength of the evidence available to our contributors prior to publication and the relevant importance of benefit and harms. We rely on our contributors to confirm the accuracy of the information presented and to adhere to describe accepted practices. Readers should be aware that professionals in the field may have different opinions. Because of this and regular advances in medical research we strongly recommend that readers' independently verify specified treatments and drugs including manufacturers' guidance. Also, the categories do not indicate whether a particular treatment is generally appropriate or whether it is suitable for a particular individual. Ultimately it is the readers' responsibility to make their own professional judgements, so to appropriately advise and treat their patients. To the fullest extent permitted by law, BMJ Publishing Group Limited and its editors are not responsible for any losses, injury or damage caused to any person or property (including under contract, by negligence, products liability or otherwise) whether they be direct or indirect, special, incidental or consequential, resulting from the application of the information in this publication.

References

1. Poole-Wilson PA. History, definition, and classification of heart failure. In: Poole-Wilson PA, Colucci WS, Massie BM, et al, eds. Heart failure. Scientific principles and clinical practice. Churchill Livingston, 1997:269–277.
2. Paulus WJ, Tschope C, Sanderson JE, et al. How to diagnose diastolic heart failure: a consensus statement on the diagnosis of heart failure with normal left ventricular ejection fraction by the Heart Failure and Echocardiography Associations of the European Society of Cardiology. Eur Heart J2007;28:2539−2550. [PubMed]
3. de Giuli F, Khaw KT, Cowie MR, et al. Incidence and outcome of persons with a clinical diagnosis of heart failure in a general practice population of 696,884 in the United Kingdom. Eur J Heart Failure2005;7:295−302. [PubMed]
4. Curtis LH, Whellan DJ, Hammill BG, et al. Incidence and prevalence of heart failure in elderly persons, 1994-2003. Arch Intern Med2008;168:418−424. [PubMed]
5. Ezekowitz JA, Kaul P, Bakal JA, et al. Declining in-hospital mortality and increasing heart failure incidence in elderly patients with first myocardial infarction. J Am Coll Cardiol2009;53:13−20. [PubMed]
6. McKelvie RS, Benedict CR, Yusuf S. Prevention of congestive heart failure and management of asymptomatic left ventricular dysfunction. BMJ1999;318:1400–1402. [PMC free article] [PubMed]
7. Vasan RS, Benjamin EJ, Levy D. Congestive heart failure with normal left ventricular systolic function. Arch Intern Med1996;156:146–157. [PubMed]
8. Davie AP, Francis CM, Caruana L, et al. The prevalence of left ventricular diastolic filling abnormalities in patients with suspected heart failure. Eur Heart J1997;18:981–984. [PubMed]
9. Cowie MR, Mosterd A, Wood DA, et al. The epidemiology of heart failure. Eur Heart J1997;18:208–225. [PubMed]
10. Yusuf S, Pepine CJ, Garces C, et al. Effect of enalapril on myocardial infarction and unstable angina in patients with low ejection fractions. Lancet1992;340:1173–1178. [PubMed]
11. Parashar S, Katz R, Smith NL, et al. Race, gender, and mortality in adults > or =65 years of age with incident heart failure (from the Cardiovascular Health Study). Am J Cardiol2009;103:1120−1127. [PMC free article] [PubMed]
12. McKelvie RS, Benedict CR, Yusuf S. Prevention of congestive heart failure and treatment of asymptomatic left ventricular dysfunction. In: Yusuf S, Cairns JA, Camm AJ, et al, eds. Evidenced based cardiology. BMJ, 2003:643–658.
13. Bittner V, Weiner DH, Yusuf S, et al, for the SOLVD Investigators. Prediction of mortality and morbidity with a 6-minute walk test in patients with left ventricular dysfunction. JAMA1993;270:1702–1707. [PubMed]
14. Rogers WJ, Johnstone DE, Yusuf S, et al, for the SOLVD Investigators. Quality of life among 5025 patients with left ventricular dysfunction randomized between placebo and enalapril. The studies of left ventricular dysfunction. J Am Coll Cardiol1994;23:393–400. [PubMed]
15. Roccaforte R, Demers C, Baldassarre F, et al. Effectiveness of comprehensive disease management programmes in improving clinical outcomes in heart failure patients. A meta-analysis. Eur J Heart Fail2005;7:1133–1144. [PubMed]
16. Gohler A, Januzzi JL, Worrell SS, et al. A systematic meta-analysis of the efficacy and heterogeneity of disease management programs in congestive heart failure. J Card Fail2006;12:554–567. [PubMed]
17. Holland R, Battersby J, Harvey I, et al. Systematic review of multidisciplinary interventions in heart failure. Heart2005;91:899−906. [PMC free article] [PubMed]
18. Kozak AT, Rucker-Whitaker C, Basu S, et al. Elements of nonpharmacologic interventions that prevent progression of heart failure: a meta-analysis. Congestive Heart Failure2007;13:280−287. [PubMed]
19. Koshman SL, Charrois TL, Simpson SH, et al. Pharmacist care of patients with heart failure: a systematic review of randomized trials. Arch Intern Med2008;168:687−694. [PubMed]
20. Klersy C, De Silvestri A, Gabutti G, et al. A meta-analysis of remote monitoring of heart failure patients. J Am Coll Cardiol2009;54:1683−1694. [PubMed]
21. Inglis SC, Clark RA, McAlister FA, et al. Structured telephone support or telemonitoring programmes for patients with chronic heart failure. In: The Cochrane Library, Issue 2, 2010. Chichester, UK: John Wiley & Sons, Ltd. [PubMed]
22. Jaarsma T, van der Wal MH, Lesman-Leegte I, et al. Effect of moderate or intensive disease management program on outcome in patients with heart failure: Coordinating Study Evaluating Outcomes of Advising and Counseling in Heart Failure (COACH). Arch Intern Med2008;168:316−324. [PubMed]
23. Ferrante D, Varini S, Macchia A, et al. Long-term results after a telephone intervention in chronic heart failure: DIAL (Randomized Trial of Phone Intervention in Chronic Heart Failure) follow-up. J Am Coll Cardiol2010;56:372−378. [PubMed]
24. Smart N, Marwick TH. Exercise training for patients with heart failure: a systematic review of factors that improve mortality and morbidity. Am J Med2004;116:693–706. Search date 2003. [PubMed]
25. Davies EJ, Moxham T, Rees K, et al. Exercise based rehabilitation for heart failure. In: The Cochrane Library, Issue 2, 2010; Chichester, UK: John Wiley & Sons, Ltd.
26. Jolly K, Taylor RS, Lip GY, et al. A randomized trial of the addition of home-based exercise to specialist heart failure nurse care: the Birmingham Rehabilitation Uptake Maximisation study for patients with Congestive Heart Failure (BRUM-CHF) study. Eur J Heart Failure2009;11:205−213. [PMC free article] [PubMed]
27. Dracup K, Evangelista LS, Hamilton MA, et al. Effects of a home-based exercise program on clinical outcomes in heart failure. Am Heart J2007;154:877−883. [PubMed]
28. O'Connor CM, Whellan DJ, Lee KL, et al. Efficacy and safety of exercise training in patients with chronic heart failure: HF-ACTION randomized controlled trial. JAMA2009;301:1439−1450. [PMC free article] [PubMed]
29. Garg R, Yusuf S, for the Collaborative Group on ACE Inhibitor Trials. Overview of randomized trials of angiotensin-converting enzyme inhibitors on mortality and morbidity in patients with heart failure. JAMA1995;273:1450–1456. Search date 1994. [PubMed]
30. Flather M, Yusuf S, Kober L, et al, for the ACE-Inhibitor Myocardial Infarction Collaborative Group. Long-term ACE-inhibitor therapy in patients with heart failure or left-ventricular dysfunction: a systematic overview of data from individual patients. Lancet2000;355:1575–1581. Search date not reported. [PubMed]
31. Agusti A, Bonet S, Arnau JM, et al. Adverse effects of ACE inhibitors in patients with chronic heart failure and/or ventricular dysfunction: meta-analysis of randomized clinical trials. Drug Saf2003;26:895–908. Search date 1999. [PubMed]
32. Packer M, Poole-Wilson PA, Armstrong PW, et al, on behalf of the ATLAS Study Group. Comparative effects of low and high doses of the angiotensin-converting enzyme inhibitor, lisinopril, on morbidity and mortality in chronic heart failure. Circulation1999;100:2312–2318. [PubMed]
33. Lakhdar R, Al-Mallah MH, Lanfear DE, et al. Safety and tolerability of angiotensin-converting enzyme inhibitor versus the combination of angiotensin-converting enzyme inhibitor and angiotensin receptor blocker in patients with left ventricular dysfunction: a systematic review and meta-analysis of randomized controlled trials. J Card Fail2008;14:181–188. [PubMed]
34. Lee VC, Rhew DC, Dylan M, et al. Meta-analysis: angiotensin-receptor blockers in chronic heart failure and high risk acute myocardial infarction. Ann Intern Med2004;141:693–704. Search date 2003. [PubMed]
35. Shibata MC, Tsuyuki RT, Wiebe N. The effects of angiotensin-receptor blockers on mortality and morbidity in heart failure: a systematic review. Int J Clin Pract2008;62:1397−1402. [PubMed]
36. Dimopoulos K, Sulukhe TV, Coats AJS, et al. Meta-analyses of mortality and morbidity effects of an angiotensin receptor blocker in patients with chronic heart failure already receiving an ACE inhibitor (alone or with a beta-blocker). Int J Cardiol2004;93:105–111. Search date 2003. [PubMed]
37. Brophy JM, Joseph L, Rouleau JL. Beta-blockers in congestive heart failure: a Bayesian meta-analysis. Ann Intern Med2001;134:550–560. Search date 2000. [PubMed]
38. Abdulla J, Kober L, Christensen E, et al. Effect of beta-blocker therapy on functional status in patients with heart failure - a meta-analysis. Eur J Heart Failure2006;8:522−531. [PubMed]
39. McAlister FA, Wiebe N, Ezekowitz JA, et al. Meta-analysis: beta-blocker dose, heart rate reduction, and death in patients with heart failure. Ann Intern Med2009;150:784−794. [PubMed]
40. Ko DT, Hebert PR, Coffey CS, et al. Adverse effects of beta-blocker therapy for patients with heart failure. A quantitative overview of randomised trials. Arch Intern Med2004;164:1389–1394. Search date 2002. [PubMed]
41. Flather MD, Shibata MC, Coats AJS, et al. Randomized trial to determine the effect of nebivolol on mortality and cardiovascular hospital admission in elderly patients with heart failure (SENIORS). Eur Heart J2005;26:215–225. [PubMed]
42. Whorlow SL, Krum H. Meta-analysis of effect of beta-blocker therapy on mortality in patients with New York Heart Association class IV chronic congestive heart failure. Am J Cardiol2000;86:886–889. Search date not reported. [PubMed]
43. Krum H, Haas SJ, Eichhorn E, et al. Prognostic benefit of beta-blockers in patients not receiving ACE-inhibitors. Eur Heart J2005;26:2154–2158. [PubMed]
44. Willenheimer R, van Veldhuisen DJ, Silke B, et al. Effect on survival and hospitalization of initiating treatment for chronic heart failure with bisoprolol followed by enalapril, as compared with the opposite sequence: results of the Randomized Cardiac Insufficiency Bisoprolol study (CIBIS) III. Circulation2005;112:2426–2435. [PubMed]
45. The CAPRICORN Investigators. Effect of carvedilol on outcome after myocardial infarction in patients with left ventricular dysfunction: the CAPRICORN randomized trial. Lancet2001;357:1385–1390. [PubMed]
46. de Peuter OR, Lussana F, Peters RJ, et al. A systematic review of selective and non-selective beta blockers for prevention of vascular events in patients with acute coronary syndrome or heart failure. Neth J Med2009;67:284–294. [PubMed]
47. Poole-Wilson PA, Swedberg K, Cleland JG, et al. Comparison of carvedilol and metoprolol on clinical outcomes in patients with chronic heart failure in the Carvedilol Or Metoprolol European Trial (COMET): randomised controlled trial. Lancet2003;362:7–13. [PubMed]
48. The Beta-Blocker Evaluation of Survival Trial Investigators. A trial of the beta-blocker bucindolol in patients with advanced chronic heart failure. N Engl J Med2001;344:1659–1667. [PubMed]
49. Haas SJ, Vos T, Gilbert RE, et al. Are beta-blockers as efficacious in patients with diabetes mellitus as in patients without diabetes mellitus who have chronic heart failure? A meta-analysis of large-scale clinical trials. Am Heart J2003;146:848–853. [PubMed]
50. Bell DSH, Lukas MA, Holdbrook FK, et al. The effect of carvedilol on mortality risk in heart failure patients with diabetes: results of a meta-analysis. Curr Med Res Opin2006;22:287−296. [PubMed]
51. Fauchier L, Pierre B, de Labriolle A, et al. Comparison of the beneficial effect of beta-blockers on mortality in patients with ischaemic or non-ischaemic systolic heart failure: a meta-analysis of randomised controlled trials. Eur J Heart Failure2007;9:1136−1139. [PubMed]
52. Dulin BR, Haas SJ, Abraham WT, et al. Do elderly systolic heart failure patients benefit from beta blockers to the same extent as the non-elderly? Meta-analysis of more than 12,000 patients in large-scale clinical trials. Am J Cardiol2005;95:896–898. [PubMed]
53. Hood WB, Dans AL, Guyatt GH, et al. Digitalis for treatment of congestive heart failure in patients in sinus rhythm: a systematic review and meta-analysis. J Card Fail2004;10:155–164. Search date 2003. [PubMed]
54. Digitalis Investigation Group. The effect of digoxin on mortality and morbidity in patients with heart failure. N Engl J Med1997;336:525–533. [PubMed]
55. Thackray S, Easthaugh J, Freemantle N, et al. The effectiveness and relative effectiveness of intravenous inotropic drugs acting through the adrenergic pathway in patients with heart failure – a meta-regression analysis. Eur J Heart Fail2002;4:515–529. Search date 2000. [PubMed]
56. Amsallem E, Kasparian C, Haddour G, et al. Phosphodiesterase III inhibitors for heart failure. In: The Cochrane Library, Issue 2, 2010. Chichester, UK: John Wiley & Sons, Ltd. Search date 2004. [PubMed]
57. Cohn J, Goldstein S, Greenberg B, et al. A dose-dependent increase in mortality with vesnarinone among patients with severe heart failure. N Engl J Med1998;339:1810–1816. [PubMed]
58. Hampton JR, van Veldhuisen DJ, Kleber FX, et al. Randomised study of effect of ibopamine on survival in patients with advanced severe heart failure. Lancet1997;349:971–977. [PubMed]
59. Ezekowitz JA, McAlister FA. Aldosterone blockade and left ventricular dysfunction: a systematic review of randomized clinical trials. Eur Heart J2009;30:469−477. [PubMed]
60. Pitt B, Zannad F, Remme WJ, et al, for the Randomized Aldactone Evaluation Study Investigators. The effects of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med1999;341:709–717. [PubMed]
61. Pitt B, Willem R, Zannad F, et al. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 2003;348:1309–1321. [Erratum in: N Engl J Med2003;348:2271] [PubMed]
62. Juurlink DN, Mamdani MM, Lee DS, et al. Rates of hyperkalemia after publication of the Randomized Aldactone Evaluation Study. N Engl J Med2004;351:543–551. [PubMed]
63. Piepoli M, Villani GQ, Ponikowski P, et al. Overview and meta-analysis of randomised trials of amiodarone in chronic heart failure. Int J Cardiol1998;66:1–10. Search date 1997. [PubMed]
64. Amiodarone Trials Meta-analysis Investigators. Effect of prophylactic amiodarone on mortality after acute myocardial infarction and in congestive heart failure: meta-analysis of individual data from 6500 patients in randomised trials. Lancet1997;350:1417–1424. Search date not reported. [PubMed]
65. Lip GYH, Gibbs CR. Anticoagulation for heart failure in sinus rhythm. In: The Cochrane Library, Issue 2, 2010. Chichester, UK: John Wiley & Sons, Ltd. Search date 2005. [PubMed]
66. Cleland JGF, Findlay I, Jafri S, et al. The Warfarin/Aspirin Study in Heart Failure (WASH): a randomised trial comparing antithrombotic strategies for patients with heart failure. Am Heart J2004;148:157–164. [PubMed]
67. Cokkinos DV, Haralabopoulos GC, Kostis JB, et al. Efficacy of antithrombotic therapy in chronic heart failure: the HELAS study. Eur J Heart Fail2006;8:428–432. [PubMed]
68. Dunkman WB, Johnson GR, Carson PE, et al, for the V-HeFT Cooperative Studies Group. Incidence of thromboembolic events in congestive heart failure. Circulation1993;87:94–101. [PubMed]
69. Al-Khadra AS, Salem DN, Rand WM, et al. Warfarin anticoagulation and survival: a cohort analysis from the studies of left ventricular dysfunction. J Am Coll Cardiol1998;31:749–753. [PubMed]
70. Lip GYH, Gibbs CR. Antiplatelet agents versus control or anticoagulation for heart failure in sinus rhythm. In: The Cochrane Library, Issue 2, 2010. Chichester, UK: John Wiley & Sons, Ltd. Search date 2005. [PubMed]
71. Massie BM, Collins JF, Ammon SE, et al. Randomized trial of warfarin, aspirin, and clopidogrel in patients with chronic heart failure: the Warfarin and Antiplatelet Therapy in Chronic Heart Failure (WATCH) trial. Circulation2009;119:1616−1624. [PubMed]
72. Al-Khadra AS, Salem DN, Rand WM, et al. Antiplatelet agents and survival: a cohort analysis from the Studies of Left Ventricular Dysfunction (SOLVD) Trial. J Am Coll Cardiol1998;31:419–425. [PubMed]
73. Latini R, Tognoni G, Maggioni AP, et al, on behalf of the Angiotensin-converting Enzyme Inhibitor Myocardial Infarction Collaborative Group. Clinical effects of early angiotensin-converting enzyme inhibitor treatment for acute myocardial infarction are similar in the presence and absence of aspirin. Systematic overview of individual data from 96 712 randomized patients. J Am Coll Cardiol2000;35:1801–1807. [PubMed]
74. Cleophas T, van Marum R. Meta-analysis of efficacy and safety of second-generation dihydropyridine calcium channel blockers in heart failure. Am J Cardiol2001;87:487–490. Search date not reported. [PubMed]
75. Gheorghiade M, Benatar D, Konstam MA, et al. Pharmacotherapy for systolic dysfunction: a review of randomized clinical trials. Am J Cardiol1997;80(8B):14H–27H. [PubMed]
76. Levine TB, Bernink P, Caspi A, et al. Effect of mibefradil, a T-type calcium channel blocker, on morbidity and mortality in moderate to severe congestive heart failure. The MACH-1 study. Circulation2000;101:758–764. [PubMed]
77. Packer M, O'Connor CM, Ghali JK, et al, for the Prospective Randomized Amlodipine Survival Evaluation Study Group. Effect of amlodipine on morbidity and mortality in severe chronic heart failure. N Engl J Med1996;335:1107–1114. [PubMed]
78. Cohn JN, Archibald DG, Ziesche S, et al. Effect of vasodilator therapy on mortality in chronic congestive heart failure. Results of a Veterans Administration Cooperative Study. N Engl J Med1986;314:1547–1552. [PubMed]
79. Taylor AL, Ziesche S, Yancy C, et al. Combination of isosorbide dinitrate and hydralazine in blacks with heart failure. N Engl J Med 2004;351:2049–2057. [Erratum in: N Engl J Med2005;352:1276] [PubMed]
80. Finks SW, Finks AL, Self TH, et al. Hydralazine-induced lupus: maintaining vigilance with increased use in patients with heart failure. South Med J2006;99:18–22. [PubMed]
81. Ezekowitz JA, Armstrong PW, McAlister FA. Implantable cardioverter defibrillators in primary and secondary prevention: a systematic review of randomised, controlled trials. Ann Intern Med2003;138:445–452. Search date 2002. [PubMed]
82. Desai AS, Fang JC, Maisel WH, et al. Implantable defibrillators for the prevention of mortality in patients with nonischemic cardiomyopathy. A meta-analysis of randomised controlled trials. JAMA2004;292:2874–2879. Search date 2004. [PubMed]
83. Ghanbari H, Dalloul G, Hasan R, et al. Effectiveness of implantable cardioverter-defibrillators for the primary prevention of sudden cardiac death in women with advanced heart failure: a meta-analysis of randomized controlled trials. Arch Intern Med2009;169:1500−1506. [PubMed]
84. Connolly SJ, Dorian P, Roberts RS, et al. Comparison of beta-blockers, amiodarone plus beta-blockers, or sotalol for prevention of shocks from implantable cardioverter defibrillators - The OPTIC study: a randomized trial. JAMA2006;295:165−171. [PubMed]
85. Lubitz SAL. Effectiveness of cardiac resynchronization therapy in mild congestive heart failure: systematic review and meta-analysis of randomized trials. Eur J Heart Fail2010;12:360–366. [PMC free article] [PubMed]
86. McAlister FA, Ezekowitz JA, Wiebe N, et al. Systematic review: cardiac resynchronization in patients with symptomatic heart failure. Ann Intern Med2004;141:381–390. Search date 2003. [PubMed]
87. Freemantle N. Cardiac resynchronisation for patients with heart failure due to left ventricular systolic dysfunction: a systematic review and meta-analysis. Eur J Heart Failure2006;8:433−440. [PubMed]
88. Lam SK, Owen A. Combined resynchronisation and implantable defibrillator therapy in left ventricular dysfunction: Bayesian network meta-analysis of randomised controlled trials. BMJ2007;335:925. [PMC free article] [PubMed]
89. Lemos Junior HP, Atallah AN. Cardiac resynchronization therapy in patients with heart failure: systematic review. Sao Paulo Med J2009;127:40–45. [PubMed]
90. Leon AR, Abraham WT, Curtis AB, et al. Safety of transvenous cardiac resynchronization system implantation in patients with chronic heart failure: combined results of over 2,000 patients from a multicenter study program. J Am Coll Cardiol2005;46:2348–2356. [PubMed]
91. Yusuf S, Zucker D, Peduzzi P, et al. Effect of coronary artery bypass graft surgery on survival: overview of 10-year results from randomised trials by the Coronary Artery Bypass Graft Trialists Collaboration. Lancet1994;344:563–570. [PubMed]
92. Allman KC, Shaw LJ, Hachamovitch R, Udelson JE. Myocardial viability testing and impact of revascularisation on prognosis in patients with coronary artery disease and left ventricular dysfunction. J Am Coll Cardiol2002;39:1151–1158. [PubMed]
93. Bourque JM, Hasselblad V, Velazquez EJ, et al. Revascularization in patients with coronary artery disease, left ventricular dysfunction, and viability: a meta analysis. Am Heart J2010;146:621–627. [PubMed]
94. Sackett DL, Rosenberg WM. On the need for evidence-based medicine. J Public Health Med1995;17:330–334. [PubMed]
95. Verhagen AP, de Vet HC, de Bie, et al. The Delphi list: a criteria list for quality assessment of randomized clinical trials for conducting systematic reviews developed by Delphi consensus. J Clin Epidemiol1998;51:1235–1241. [PubMed]
96. Velazquez EJ, Lee KL, O'Connor CM, et al. The rationale and design of the Surgical Treatment for Ischemic Heart Failure (STICH) trial. J Thorac Cardiovasc Surg2007;134:1540–1547. [PMC free article] [PubMed]
97. Dagenais GR, Pogue J, Fox K, et al. Angiotensin-converting-enzyme inhibitors in stable vascular disease without left ventricular systolic dysfunction or heart failure: a combined analysis of three trials. Lancet2006;368:581–588. [PubMed]
98. Baker WL, Coleman CI, Kluger J, et al. Systematic review: comparative effectiveness of angiotensin-converting enzyme inhibitors or angiotensin II-receptor blockers for ischemic heart disease. Ann Intern Med2009;151:861–871. [PubMed]
99. Rutherford JD, Pfeffer MA, Moyé LA, et al. Effects of captopril on ischaemic events after myocardial infarction. Circulation1994;90:1731–1738. [PubMed]
100. Jong P, Yusuf S, Rousseau MF, et al. Effect of enalapril on 12-year survival and life expectancy in patients with left ventricular systolic dysfunction: a follow-up study. Lancet2003;361:1843–1848. [PubMed]
101. Buch P, Rasmussen S, Abildstrom SZ, et al. The long-term impact of the angiotensin-converting enzyme inhibitor trandolapril on mortality and hospital admissions in patients with left ventricular dysfunction after a myocardial infarction: follow-up to 12 years. Eur Heart J2005;26:145–152. [PubMed]
102. SOLVD Investigators. Effect of enalapril on mortality and the development of heart failure in asymptomatic patients with reduced left ventricular ejection fractions. N Engl J Med1992;327:685–691. [PubMed]
103. Pfeffer MA, McMurray JJV, Velazquez EJ, et al. Valsartan, captopril, or both in myocardial infarction complicated by heart failure, left ventricular dysfunction, or both. N Engl J Med2003;349:1893–1906. [PubMed]
104. ONTARGET Investigators, Yusuf S, Teo KK, et al. Telmisartan, ramipril, or both in patients at high risk for vascular events. N Engl J Med2008;358:1547–1559. [PubMed]
105. TRANSCEND Investigators, Yusuf S, Teo K, et al. Effects of the angiotensin-receptor blocker telmisartan on cardiovascular events in high-risk patients intolerant to angiotensin-converting enzyme inhibitors: a randomised controlled trial. Lancet2008;372:1174–1183. [PubMed]
106. Yusuf S, Diener HC, Sacco RL, et al. Telmisartan to prevent recurrent stroke and cardiovascular events. N Engl J Med2008;359:1225–1237. [PMC free article] [PubMed]
107. Shah RV, Desai AS, Givertz MM. The effect of renin-angiotensin system inhibitors on mortality and heart failure hospitalization in patients with heart failure and preserved ejection fraction: a systematic review and meta-analysis. J Card Fail2010;16:260–267. [PubMed]
108. Yusuf S, Pfeffer MA, Swedberg K, et al. (CHARM Investigators and Committees). Effects of candesartan in patients with chronic heart failure and preserved left-ventricular ejection fraction: the CHARM-Preserved Trial. Lancet2003;362:777–781. [PubMed]
109. Massie BM, Carson PE, McMurray JJ, et al. Irbesartan in patients with heart failure and preserved ejection fraction. N Engl J Med2008;359:2456−2467. [PubMed]
110. Cleland JGF, Tendera M, Adamus J, et al. The perindopril in elderly people with chronic heart failure (PEP-CHF) study. Eur Heart J2006:2338–2345. [PubMed]
111. Hogg K, McMurray J, Hogg Karen, et al. The treatment of heart failure with preserved ejection fraction ("diastolic heart failure"). Heart Failure Rev2006;11:141−146. [PubMed]
112. Heckman GA, McKelvie RS. Diagnosis and management of heart failure with preserved ejection fraction in older adults. Geriatr Aging2009;12:93–96.
113. Ahmed A, Rich MW, Fleg JL, et al. Effects of digoxin on morbidity and mortality in diastolic heart failure: the ancillary digitalis investigation group trial. Circulation2006;114:397–403. [PMC free article] [PubMed]
BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Multidisciplinary interventions

Summary

Multidisciplinary interventions may reduce admissions to hospital and mortality in people with heart failure compared with usual care, although long-term benefits remain unclear.

Benefits and harms

Multidisciplinary interventions versus usual care:

We found 7 systematic reviews (search date 2004, 33 RCTs, 7387 people; search date 2005, 36 RCTs, 8341 people; search date 2004, 30 RCTs, 7532 people; search date 2006, 26 RCTs, 4671 people; search date 2007, 12 RCTs, 2060 people; search date 2008, 20 RCTs, 6258 people, 12 cohorts, 2354 people; and search date 2008, 25 RCTs, 8323 people) and two subsequent RCTs. The reviews identified some of the same RCTs, however, they included different combinations of RCTs in their meta-analyses, and analysed different aspects of multidisciplinary programmes, and so we report data from all reviews.

Mortality

Compared with usual care Multidisciplinary programmes are more effective at reducing all-cause mortality (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
5308 people
28 RCTs in this analysis
Rate of all-cause mortality
389/2587 (15%) with disease management programme
492/2721 (18%) with usual care

OR 0.80
95% CI 0.69 to 0.93
P = 0.003
RCTs of multidisciplinary treatment were generally small, involving highly selected patient populations (see further information on studies for more details)
Small effect sizedisease management programme

Systematic review
Number of people in analysis not reported
30 RCTs in this analysis
Rate of all-cause mortality
with disease management programme
with usual care
Absolute results not reported

ARR –3%
95% CI –5% to –1%
P <0.01
Benefit of the intervention was dependent on age, severity of disease, guideline-based treatment at baseline, and disease management programme modalities
Effect size not calculateddisease management programme

Systematic review
7532 people
27 RCTs in this analysis
Rate of all-cause mortality
613/3867 (16%) with multidisciplinary programme
661/3580 (18%) with control (not further defined)

RR 0.79
95% CI 0.69 to 0.92
P = 0.002
There was significant heterogeneity among RCTs (P = 0.04); the review identified 2 RCTs that were outliers as potential sources of heterogeneity
Small effect sizemultidisciplinary programme

Systematic review
7213 people
26 RCTs in this analysis
Sensitivity analysis
Rate of all-cause mortality
with multidisciplinary programme
with control (not further defined)
Absolute results not reported

RR 0.83
95% CI 0.73 to 0.95
P = 0.002
Sensitivity analysis excluding 1 outlier removed heterogeneity with only a small reduction in results for effectiveness
Small effect sizemultidisciplinary programme

Systematic review
553 people
3 RCTs in this analysis
Subgroup analysis
Rate of all-cause mortality
35/316 (11%) with televideo or remote monitoring-based programme
51/237 (22%) with control (not further defined)

RR 0.49
95% CI 0.33 to 0.73
P = 0.0004
Moderate effect sizetelevideo or remote monitoring-based programme

Systematic review
3384 people
11 RCTs in this analysis
Subgroup analysis
Rate of all-cause mortality
220/1679 (13%) with programmes incorporating contact by telephone or mail
279/1705 (16%) with control (not further defined)

RR 0.70
95% CI 0.53 to 0.94
P = 0.02
Small effect sizeprogrammes incorporating contact by telephone or mail

Systematic review
1811 people
11 RCTs in this analysis
Subgroup analysis
Rate of all-cause mortality
149/890 (17%) with programmes consisting of home visits
183/921 (20%) with control (not further defined)

RR 0.87
95% CI 0.72 to 1.06
P = 0.17
Not significant

Systematic review
1784 people
3 RCTs in this analysis
Subgroup analysis
Rate of all-cause mortality
209/982 (21.3%) with programmes delivered in hospital, clinic, or general practice
170/802 (21.2%) with control (not further defined)

RR 1.00
95% CI 0.84 to 1.20
P = 0.98
Not significant

Systematic review
3918 people
22 RCTs in this analysis
Rate of all-cause mortality
with multidisciplinary programmes
with control (predominantly usual care)
Absolute results not reported

OR 0.69
95% CI 0.56 to 0.85
Small effect sizemultidisciplinary programme

Systematic review
Number of RCTs and people included in subgroup analysis not reported
Subgroup analysis
Rate of all-cause mortality
with programmes involving face-to-face contact
with control (predominantly usual care)
Absolute results not reported

OR 0.63
95% CI 0.44 to 0.91
Small effect sizeprogrammes involving face-to-face contact

Systematic review
Number of RCTs and people included in subgroup analysis not reported
Subgroup analysis
Rate of all-cause mortality
with programmes involving face-to-face contact plus telephone contact
with control (predominantly usual care)
Absolute results not reported

OR 0.68
95% CI 0.44 to 1.06
Not significant

Systematic review
Number of RCTs and people included in subgroup analysis not reported
Subgroup analysis
Rate of all-cause mortality
with programmes involving telephone (non face-to-face) management
with control (predominantly usual care)
Absolute results not reported

OR 0.82
95% CI 0.48 to 1.40
Not significant

Systematic review
2060 people
12 RCTs in this analysis
Rate of all-cause mortality
117/1001 (12%) with pharmacist care
136/1059 (13%) with no pharmacist care

OR 0.84
95% CI 0.61 to 1.15
Not significant

Systematic review
6133 people
19 RCTs in this analysis
All-cause mortality
390/3320 (12%) with remote patient monitoring
397/2813 (14%) with usual care

RR 0.83
95% CI 0.73 to 0.95
P = 0.006
Small effect sizeremote patient monitoring

Systematic review
5563 people
15 RCTs in this analysis
All-cause mortality
332/2948 (11%) with structured telephone support
332/2615 (13%) with usual care

RR 0.88
95% CI 0.76 to 1.01
P = 0.08
Not significant

Systematic review
2710 people
11 RCTs in this analysis
All-cause mortality
147/1410 (10%) with telemonitoring
200/1300 (15%) with usual care

RR 0.66
95% CI 0.54 to 0.81
P = 0.00005
Small effect sizetelemonitoring

RCT
3-armed trial
1049 people hospitalised because of heart failure, New York Heart Association (NYHA) functional class II to IV Rate of all-cause mortality 18 months
83/344 (24%) with intensive disease management programme
99/339 (29%) with usual care

Significance not assessed
The RCT was not powered to assess mortality alone
The results from this large RCT do not correlate with the results of the 6 systematic reviews reported (see further information on studies for details)

RCT
1518 people having outpatient care for stable chronic heart failure with mainly NYHA class II or III symptoms All-cause mortality up to 1 year after completion of the trial
189/760 (25%) with previous telephone monitoring by specialised nurses
197/758 (26%) with previous usual care

RR 0.94
95% CI 0.77 to 1.16
P = 0.59
Not significant

RCT
1518 people having outpatient care for stable chronic heart failure with mainly NYHA class II or III symptoms All-cause mortality up to 3 years after completion of the trial
326/760 (43%) with previous telephone monitoring by specialised nurses
308/758 (41%) with previous usual care

RR 1.02
95% CI 0.87 to 1.20
P = 0.73
Not significant

Admission to hospital

Compared with usual care Multidisciplinary programmes are more effective at reducing all-cause hospital admissions, and hospital admissions for heart failure (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause hospital re-admission

Systematic review
7387 people
32 RCTs in this analysis
Proportion of people admitted to hospital for any cause
with disease management programme
with usual care
Absolute results not reported

OR 0.76
95% CI 0.69 to 0.94
P <0.00001
RCTs of multidisciplinary treatment were generally small, involving highly selected patient populations (see further information on studies for more details)
Small effect sizedisease management programme

Systematic review
6569 people
21 RCTs in this analysis
Proportion of people admitted to hospital for any cause
1332/3331 (40%) with multidisciplinary programme
1442/3238 (45%) with control (not further defined)

RR 0.87
95% CI 0.79 to 0.95
P = 0.002
There was significant heterogeneity among RCTs (P = 0.04)
Small effect sizemultidisciplinary programme

Systematic review
2026 people
11 RCTs in this analysis
Proportion of people admitted to hospital for any cause
363/984 (37%) with pharmacist care
449/1042 (43%) with no pharmacist care

OR 0.71
95% CI 0.54 to 0.94
Small effect sizepharmacist care

Systematic review
4122 people
11 RCTs in this analysis
All-cause hospital admissions
918/2137 (43%) with remote patient monitoring
901/1985 (45%) with usual care

RR 0.93
95% CI 0.87 to 0.99
P = 0.03
Small effect sizeremote patient monitoring

Systematic review
4295 people
11 RCTs in this analysis
All-cause hospital admissions
822/2140 (38%) with structured telephone support
888/2155 (41%) with usual care

RR 0.92
95% CI 0.85 to 0.99
P = 0.024
Small effect sizestructured telephone support

Systematic review
2343 people
8 RCTs in this analysis
All-cause hospital admissions
582/1232 (47%) with telemonitoring
579/1111 (52%) with usual care

RR 0.91
95% CI 0.84 to 0.99
P = 0.022
Small effect sizetelemonitoring
Heart failure-specific hospital re-admission

Systematic review
3817 people
20 RCTs in this analysis
Proportion of people admitted to hospital for heart failure-specific causes
with disease management programme
with usual care
Absolute results not reported

OR 0.58
95% CI 0.50 to 0.67
P <0.00001
RCTs of multidisciplinary treatment were generally small, involving highly selected patient populations (see further information on studies for more details)
Small effect sizedisease management programme

Systematic review
Number of people in analysis not clear
16 RCTs in this analysis
Proportion of people admitted to hospital for heart failure-specific causes
with multidisciplinary programme
with control (not further defined)
Absolute results not reported

RR 0.70
95% CI 0.61 to 0.81
P <0.0001
There was significant heterogeneity among RCTs (P = 0.04)
Small effect sizemultidisciplinary programme

Systematic review
3844 people
21 RCTs in this analysis
Proportion of people admitted to hospital for heart failure-specific causes
with multidisciplinary programmes
with control (predominantly usual care)
Absolute results not reported

OR 0.41
95% CI 0.30 to 0.56
Moderate effect sizemultidisciplinary programme

Systematic review
Number of RCTs and people included in subgroup analysis not reported
Subgroup analysis
Proportion of people admitted to hospital for heart failure
with programmes involving face-to-face contact
with control (predominantly usual care)
Absolute results not reported

OR 0.42
95% CI 0.22 to 0.81
Moderate effect sizeprogrammes involving face-to-face contact

Systematic review
Number of RCTs and people included in subgroup analysis not reported
Subgroup analysis
Proportion of people admitted to hospital for heart failure
with programmes involving face-to-face contact plus telephone contact
with control (predominantly usual care)
Absolute results not reported

OR 0.37
95% CI 0.21 to 0.64
Moderate effect sizeprogrammes involving face-to-face contact plus telephone contact

Systematic review
Number of RCTs and people included in subgroup analysis not reported
Subgroup analysis
Proportion of people admitted to hospital for heart failure
with programmes involving telephone (non face-to-face) management
with control (predominantly usual care)
Absolute results not reported

OR 0.67
95% CI 0.36 to 1.26
Not significant

Systematic review
1977 people
11 RCTs in this analysis
Proportion of people admitted to hospital for heart failure
183/959 (19%) with pharmacist care
238/1018 (23%) with no pharmacist care

OR 0.69
95% CI 0.51 to 0.94
Small effect sizepharmacist care

Systematic review
4310 people
13 RCTs in this analysis
Hospital admissions for heart failure
424/2231 (19%) with remote patient monitoring
546/2079 (26%) with usual care

RR 0.71
95% CI 0.64 to 0.80
P <0.001
Small effect sizeremote patient monitoring

Systematic review
4269 people
13 RCTs in this analysis
Hospital admissions for heart failure
346/2102 (16%) with structured telephone support
462/2167 (21%) with usual care

RR 0.77
95% CI 0.68 to 0.87
P = 0.00004
Small effect sizestructured telephone support

Systematic review
1570 people
4 RCTs in this analysis
Hospital admissions for heart failure
189/844 (22%) with telemonitoring
207/726 (28%) with usual care

RR 0.79
95% CI 0.67 to 0.94
P = 0.008
Small effect sizetelemonitoring

RCT
3-armed trial
1049 people hospitalised because of heart failure, New York Heart Association (NYHA) functional class II to IV Rate of composite outcome of all-cause mortality or hospital re-admission as a result of heart failure 18 months
132/344 (38%) with intensive disease management programme
141/339 (42%) with usual care

HR 0.93 (intensive disease management v usual care)
95% CI 0.73 to 1.17
The results from this large RCT do not correlate with the results of the 6 systematic reviews reported (see further information on studies for details)
Not significant

RCT
3-armed trial
1049 people hospitalised because of heart failure, NYHA functional class II to IV Proportion of people admitted to hospital for chronic heart failure 18 months
92/344 (27%) with intensive disease management programme
84/339 (25%) with usual care

Significance not assessed
The results from this large RCT do not correlate with the results of the 6 systematic reviews reported (see further information on studies for details)

RCT
1518 people attending outpatients for stable chronic heart failure with mainly NYHA class II to III symptoms Hospital admission for heart failure up to 1 year after completion of the trial
174/760 (23%) with previous telephone monitoring
220/758 (29%) with previous usual care

RR 0.73
95% CI 0.60 to 0.90
P = 0.002
Small effect sizeprevious telephone monitoring

RCT
1518 people attending outpatients for stable chronic heart failure with mainly NYHA class II to III symptoms Hospital admission for heart failure up to 3 years after completion of the trial
217/760 (29%) with previous telephone monitoring
266/758 (35%) with previous usual care

RR 0.72
95% CI 0.60 to 0.87
P = 0.0004
Small effect sizeprevious telephone monitoring

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

No data from the following reference on this outcome.

Further information on studies

The RCTs of multidisciplinary treatment were generally small, involving highly selected patient populations. Many lasted <6 months and were usually carried out in academic centres, and so the results may not generalise to longer-term outcomes based in smaller community centres.

A multidisciplinary intervention was defined as one in which heart failure management was the responsibility of a team incorporating medical input and input from one or more other areas (specialist nurse, pharmacist, dietitian, or social worker).

The systematic review also identified 6 cohort studies (1925 people). It found that remote patient monitoring significantly reduced all-cause mortality compared with usual care over the duration of the included studies (67/980 [7%] with remote patient monitoring v 123/945 [13%] with usual care; RR 0.53, 95% CI 0.29 to 0.96; P <0.001). It also found that remote patient monitoring significantly reduced all-cause hospital admissions (3 cohort studies, 819 people: 84/420 [20%] with remote patient monitoring v 153/399 [38%] with usual care; RR 0.52, 95% CI 0.28 to 0.96; P <0.001).

One possible reason for the lack of observed benefit for the intensive disease management programme is that, during the course of the study, people in the usual-care group had a closer follow-up by the cardiologist than was anticipated before starting the study.

Comment

Clinical guide:

The multiple systematic reviews identified have suggested that disease management programmes may reduce mortality, all-cause hospital admissions, and hospital admission for heart failure. It is reassuring that, despite the lack of specific analysis of adverse effects, all-cause mortality and all-cause hospital re-admissions were reduced, suggesting that multidisciplinary treatment overall is beneficial and not associated with any clinically important adverse effects. The data at this time are supportive of the use of disease management programmes to treat people with heart failure, with the expectation that there will be a reduction in mortality and morbidity compared with usual care.

Substantive changes

Multidisciplinary interventions New evidence added. Categorisation unchanged (Beneficial).

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Exercise

Summary

Exercise may reduce admissions to hospital due to heart failure in people with heart failure compared with usual care, although long-term benefits remain unclear.

Benefits and harms

Exercise versus usual care:

We found two systematic reviews (search date 2003, 30 parallel-group RCTs plus 9 crossover RCTs, 2387 people;and search date 2008, 19 RCTs, 3647 people) and one subsequent RCT.The reviews identified 6 RCTs in common, however they presented different meta-analyses and so we report both reviews here.

Mortality

Compared with usual care Exercise seems no more effective at reducing mortality (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
1197 people
14 RCTs in this analysis
Rate of all-cause mortality mean 5.9 months' follow-up
26/622 (4%) with exercise
41/575 (7%) with control

OR 0.71
95% CI 0.37 to 1.02
Follow-up among RCTs ranged from 4 weeks to 192 weeks; about half the RCTs included in the review had a follow-up of 3 months or less
Not significant

Systematic review
962 people
13 RCTs in this analysis
Rate of all-cause mortality up to 12 months' follow-up
42/480 (8.8%) with exercise
41/482 (8.5%) with control

RR 1.02
95% CI 0.70 to 1.51
P = 0.90
Follow-up among RCTs ranged from 6 months to 12 months
Not significant

Systematic review
2658 people
4 RCTs in this analysis
Rate of all-cause mortality >12 months' follow-up
238/1319 (18%) with exercise
268/1339 (20%) with control

RR 0.88
95% CI 0.73 to 1.07
P = 0.21
Follow-up among RCTs ranged from 26 months to 75 months
Not significant

No data from the following reference on this outcome.

Admission to hospital

Compared with usual care Exercise may be more effective than usual care at reducing hospital admissions caused by heart failure at up to 30 months' follow-up. However, we don't know whether exercise is more effective at reducing all-cause hospital admissions or composite outcomes including all-cause hospital admission (other outcomes in composites include emergency department admission, urgent transplant, and death) (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Death or hospital admission

Systematic review
1197 people
14 RCTs in this analysis
Rate of events (including hospital admission causing temporary or permanent withdrawal from exercise) mean 5.9 months' follow-up
30/622 (5%) with exercise
34/575 (6%) with control

OR 0.83
95% CI 0.50 to 1.39
Follow-up among RCTs ranged from 4 weeks to 192 weeks; about half the RCTs included in the review had a follow-up of 3 months or less
Not significant

Systematic review
659 people
8 RCTs in this analysis
Rate of all-cause hospital admissions up to 12 months' follow-up
47/329 (14%) with exercise
61/330 (18%) with control

RR 0.79
95% CI 0.58 to 1.07
P = 0.13
Follow-up among RCTs ranged from 6 months to 12 months
Not significant

Systematic review
2658 people
4 RCTs in this analysis
Rate of all-cause hospital admissions >12 months' follow-up
764/1319 (58%) with exercise
810/1339 (60%) with control

RR 0.96
95% CI 0.90 to 1.02
P = 0.15
Follow-up among RCTs ranged from 26 months to 75 months
Not significant

Systematic review
569 people
7 RCTs in this analysis
Rate of hospital admissions for heart failure up to 30 months' follow-up
44/286 (15%) with exercise
61/283 (22%) with control

RR 0.72
95% CI 0.52 to 0.99
P = 0.042
Small effect sizeexercise

No data from the following reference on this outcome.

Functional improvement

Compared with usual care Exercise may be more effective at increasing exercise time at 3 and 12 months, and at increasing distance walked on the 6-minute walk at 3 months, but we don't know whether it is more effective at increasing distance walked on the 6-minute walk at 6 or 12 months or at improving performance on the incremental shuttle walking test at 6 months (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Walking distance

RCT
173 people with heart failure and left ventricular ejection fraction (LVEF) 40% or less, New York Heart Association (NYHA) functional class II to IV
In review
Change in distance on the 6-minute walk (change from baseline) 6 months
From 1350 to 1422 with home-based exercise programme (a combination of aerobic and resistance exercise training)
From 1324 to 1385 with usual care

P = 0.275
The method of randomisation of the RCT is unclear
Not significant

RCT
169 people with LVEF 40% or less and NYHA functional class II or more Change in the incremental shuttle walking test 6 months
with home-based exercise training plus specialist heart failure nurse care
with usual care (specialist heart failure nurse care alone)
Absolute results not reported

Mean difference: +14.98 metres
95% CI –11.89 metres to +41.86 metres
P = 0.1
Not significant

RCT
1835 people with medically stable heart failure and LVEF <35%
In review
Change in distance on the 6-minute walk (change from baseline) 3 months
Median increase +20 metres (interquartile range [IQR] –15 metres to +57 metres) with exercise training
Median increase +5 metres (IQR –28 to +37 metres) with usual care

P <0.001
Effect size not calculatedexercise training

RCT
1914 people with medically stable heart failure and LVEF <35%
In review
Change in exercise time (change from baseline) at 3 months
Median increase 1.5 minutes (IQR 0.3 minutes to 3.0 minutes) with exercise training
Median increase +0.3 minutes (IQR –0.6 to +1.4 minutes) with usual care
Absolute results not reported

P <0.001
Effect size not calculatedexercise training

RCT
1444 people with medically stable heart failure and LVEF <35%
In review
Change in distance on the 6-minute walk (change from baseline) 12 months
Median increase +13 metres (IQR –28 metres to +61 metres) with exercise training
Median increase +12 metres (IQR –30 metres to +55 metres) with usual care

P = 0.26
Not significant

RCT
1476 people with medically stable heart failure and LVEF <35%
In review
Change in exercise time (change from baseline) at 12 months
Median increase 1.5 minutes (IQR 0 minutes to 3.2 minutes) with exercise training
Median increase +0.2 minutes (IQR –1.0 minutes to +1.7 minutes) with usual care

P <0.001
Effect size not calculatedexercise

No data from the following reference on this outcome.

Quality of life

Compared with usual care We don't know how exercise impacts quality of life (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Quality of life

RCT
169 people with left ventricular ejection fraction (LVEF) 40% or less and New York Heart Association (NYHA) functional class II or more Change in the Minnesota Living with Heart Failure Questionnaire (MLHFQ) 6 months
with home-based exercise training plus specialist heart failure nurse care
with usual care (specialist heart failure nurse care alone)
Absolute results not reported

Mean difference –2.53
95% CI –7.87 to +2.80
P = 0.3
Not significant

RCT
169 people with LVEF 40% or less and NYHA functional class II or more Change in the MLHFQ 12 months
with home-based exercise training plus specialist heart failure nurse care
with usual care (specialist heart failure nurse care alone)
Absolute results not reported

Mean difference –0.55
95% CI –5.87 to +4.76
P = 0.8
Not significant

Systematic review
700 people
6 RCTs in this analysis
Health-related quality of life with MLHFQ up to 60 months' follow-up
with exercise
with usual care
Absolute results not reported

SMD –10.33
95% CI –15.89 to –4.77
P = 0.00027
Effect size not calculatedexercise

Systematic review
3109 people
10 RCTs in this analysis
Health-related quality of life with all questionnaires up to 75 months' follow-up
with exercise
with usual care
Absolute results not reported

SMD –0.56
95% CI –0.82 to –0.30
P = 0.00002
Effect size not calculatedexercise

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Adverse effects

Systematic review
Number of people not clear Exercise-related mortality
with exercise
with control

RCT
2331 people with medically stable heart failure and left ventricular ejection fraction of 35% or less
In review
Proportion of people with at least 1 hospital admission as a result of an event during or within 3 hours after exercise
37/1159 (3%) with exercise training
22/1171 (2%) with usual care

Significance not assessed

No data from the following reference on this outcome.

Further information on studies

None.

Comment

Clinical guide:

The specific form of exercise training varied among trials, and the relative merits of each strategy are unknown. Adherence to home-based exercise programmes is typically low, which could result in underestimation of the beneficial effects of exercise training. The most recent RCT (HF-ACTION RCT) identified by review is the largest (2331 people) study identified to date. Results from this multicentre, international study may be more appropriately generalised to smaller community centres. The findings from HF-ACTION and overall by the second systematic review support a prescribed exercise training programme for patients with heart failure in addition to other evidence-based treatments.

Substantive changes

Exercise New evidence added. Categorisation unchanged (Likely to be beneficial).

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

ACE inhibitors for treating heart failure

Summary

ACE inhibitors reduce mortality and hospital admissions from heart failure compared with placebo, with greater absolute benefits seen in people with more severe heart failure.

Combined treatment with angiotensin II receptor blockers and ACE inhibitors may lead to a greater reduction in hospital admission for heart failure compared with ACE inhibitor treatment alone.

Benefits and harms

ACE inhibitors versus placebo:

We found two systematic reviews (search dates 1994 and not reported) of ACE inhibitors versus placebo in heart failure. The second review analysed long-term results from large RCTs comparing ACE inhibitors versus placebo. Three RCTs identified by the review examined the effects of ACE inhibitors in people for 1 year after MI. We found one systematic review (search date 1999) that specifically examined the adverse effects of ACE inhibitors in people with heart failure.

Mortality

Compared with placebo ACE inhibitors are more effective at reducing mortality (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
7105 people, New York Heart Association (NYHA) functional class III or IV
32 RCTs in this analysis
Rate of all-cause mortality
611/3870 (16%) with ACE inhibitors
709/3235 (22%) with placebo

ARR 6%
95% CI 4% to 8%
OR 0.77
95% CI 0.67 to 0.88
Small effect sizeACE inhibitors

Systematic review
12,763 people with left ventricular dysfunction or heart failure of mean duration 35 months
5 RCTs in this analysis
Rate of all-cause mortality
1467/6391 (23%) with ACE inhibitors
1710/6372 (27%) with placebo

OR 0.80
95% CI 0.74 to 0.87
See further information on studies for details of relative benefits of ACE inhibitors
Small effect sizeACE inhibitors

Systematic review
5966 people at 1 year post-infarction
3 RCTs in this analysis
Rate of all-cause mortality
702/2995 (23%) with ACE inhibitors
866/2971 (29%) with placebo

OR 0.74
95% CI 0.66 to 0.83
See further information on studies for details of relative benefits of ACE inhibitors
Small effect sizeACE inhibitors

Admission to hospital

Compared with placebo ACE inhibitors are more effective at reducing hospital admissions for heart failure (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Heart failure-specific re-admission to hospital

Systematic review
12,763 people with left ventricular dysfunction or heart failure of mean duration 35 months
5 RCTs in this analysis
Proportion of people admitted to hospital for heart failure causes
876/6391 (14%) with ACE inhibitors
1202/6372 (19%) with placebo

OR 0.67
95% CI 0.61 to 0.74
See further information on studies for details of relative benefits of ACE inhibitors
Small effect sizeACE inhibitors

Systematic review
5966 people at 1 year post-infarction
3 RCTs in this analysis
Proportion of people admitted to hospital for heart failure causes
355/2995 (12%) with ACE inhibitors
460/2971 (16%) with placebo

OR 0.73
95% CI 0.63 to 0.85
See further information on studies for details of relative benefits of ACE inhibitors
Small effect sizeACE inhibitors

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Withdrawal owing to adverse effects

Systematic review
18,234 people
22 RCTs in this analysis
Proportion of people withdrawing because of an adverse effect about 2 years
1035/7487 (14%) with ACE inhibitor
661/7025 (9%) with control (placebo or non-ACE inhibitor treatments)

RR 1.54
95% CI 1.30 to 1.83
Small effect sizecontrol
Adverse effects

Systematic review
11,989 people Risk of cough
123/6191 (2%) with ACE inhibitor
34/5798 (1%) with control (placebo or non-ACE inhibitor treatments)

RR 3.19
95% CI 2.22 to 4.57
Moderate effect sizecontrol

Systematic review
11,989 people Risk of hypotension
102/6191 (2%) with ACE inhibitor
45/5798 (1%) with control (placebo or non-ACE inhibitor treatments)

RR 1.95
95% CI 1.39 to 2.74
Small effect sizecontrol

Systematic review
11,989 people Risk of renal dysfunction
59/6191 (0.9%) with ACE inhibitor
31/5798 (0.5%) with control (placebo or non-ACE inhibitor treatments)

RR 1.84
95% CI 1.20 to 2.81
Small effect sizecontrol

Systematic review
11,989 people Risk of dizziness
92/6191 (1.4%) with ACE inhibitor
56/5798 (0.9%) with control (placebo or non-ACE inhibitor treatments)

RR 1.60
95% CI 1.15 to 2.23
Small effect sizecontrol

Systematic review
11,989 people Risk of impotence
10/6191 (0.16%) with ACE inhibitor
1/5798 (0.02%) with control (placebo or non-ACE inhibitor treatments)

RR 6.46
95% CI 1.14 to 36.58
Large effect sizecontrol

No data from the following reference on this outcome.

Difference doses of ACE inhibitors versus each other:

We found one large RCT comparing low-dose lisinopril versus high-dose lisinopril.

Mortality

Different doses of ACE inhibitors compared with each other Low-dose and high-dose lisinopril seem equally effective at reducing mortality (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

RCT
3164 people with New York Heart Association functional class II to IV heart failure Rate of mortality
717/1596 (45%) with low-dose lisinopril (2.5 or 5.0 mg/day)
666/1568 (43%) with high-dose lisinopril (32.5 or 35.0 mg/day)

ARR 2.4%
CI not reported
HR 0.92
95% CI 0.80 to 1.03
P = 0.128
Not significant

Admission to hospital

Different doses of ACE inhibitors compared with each other Low-dose lisinopril seems less effective than high-dose lisinopril at reducing admissions for heart failure (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Death or hospital admission

RCT
3164 people with New York Heart Association (NYHA) functional class II to IV heart failure Rate of combined outcome of death or hospital admission for any reason
1338/1596 (84%) with low-dose lisinopril (2.5 or 5.0 mg/day)
1250/1568 (80%) with high-dose lisinopril (32.5 or 35.0 mg/day)

ARR 4.1%
CI not reported
HR 0.88
95% CI 0.82 to 0.96
Small effect sizehigh-dose lisinopril
Heart failure-specific re-admission to hospital

RCT
3164 people with NYHA functional class II to IV heart failure Proportion of people admitted to hospital for heart failure
1576/1596 (99%) with low-dose lisinopril (2.5 or 5.0 mg/day)
1199/1568 (77%) with high-dose lisinopril (32.5 or 35.0 mg/day)

ARR 22.2%
CI not reported
P = 0.002
Effect size not calculatedhigh-dose lisinopril

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Withdrawal

RCT
3164 people with New York Heart Association (NYHA) functional class II to IV heart failure Proportion of people withdrawing from RCT
18% with low-dose lisinopril (2.5 or 5.0 mg/day)
17% with high-dose lisinopril (32.5 or 35.0 mg/day)
Absolute numbers not reported

Significance not assessed
Adverse effects

RCT
3164 people with NYHA functional class II to IV heart failure Proportion of people with dizziness
12% with low-dose lisinopril (2.5 or 5.0 mg/day)
19% with high-dose lisinopril (32.5 or 35.0 mg/day)
Absolute numbers not reported

Significance not assessed

RCT
3164 people with NYHA functional class II to IV heart failure Proportion of people with hypotension
7% with low-dose lisinopril (2.5 or 5.0 mg/day)
11% with high-dose lisinopril (32.5 or 35.0 mg/day)
Absolute numbers not reported

Significance not assessed

RCT
3164 people with NYHA functional class II to IV heart failure Proportion of people with worsening renal function
7% with low-dose lisinopril (2.5 or 5.0 mg/day)
10% with high-dose lisinopril (32.5 or 35.0 mg/day)
Absolute numbers not reported

Significance not assessed

RCT
3164 people with NYHA functional class II to IV heart failure Proportion of people with significant change in serum potassium concentration
7% with low-dose lisinopril (2.5 or 5.0 mg/day)
7% with high-dose lisinopril (32.5 or 35.0 mg/day)
Absolute numbers not reported

Significance not assessed

RCT
3164 people with NYHA functional class II to IV heart failure Proportion of people with cough
13% with low-dose lisinopril (2.5 or 5.0 mg/day)
11% with high-dose lisinopril (32.5 or 35.0 mg/day)
Absolute numbers not reported

Significance not assessed

ACE inhibitors versus angiotensin II receptor blockers:

See option on angiotensin II receptor blockers.

ACE inhibitors alone versus ACE inhibitors plus angiotensin II receptor blockers:

See option on angiotensin II receptor blockers.

ACE inhibitors versus beta-blockers:

See option on beta-blockers.

Further information on studies

The relative benefits of treatment with ACE inhibitors began soon after the start of treatment, persisted in the long term, and were independent of age, sex, and baseline use of diuretics, aspirin, and beta-blockers. Although there was a trend towards greater relative reduction in mortality or re-admission for heart failure in people with lower ejection fraction, benefit was apparent over the range examined.

Comment

The first systematic review found similar benefits with different ACE inhibitors.

Clinical guide:

The relative benefits of ACE inhibitors were similar in different subgroups of people with heart failure. Most RCTs evaluated left ventricular function by assessing left ventricular ejection fraction, but some studies defined heart failure clinically, without measurement of left ventricular function in people at high risk of developing heart failure (soon after MI). It is unclear whether there are additional benefits from adding an ACE inhibitor to antiplatelet treatment in people with heart failure (see antiplatelet agents).

Substantive changes

No new evidence

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Angiotensin II receptor blockers for treating heart failure

Summary

Angiotensin II receptor blockers reduce mortality and hospital admissions from heart failure compared with placebo, with greater absolute benefits seen in people with more severe heart failure.

Combined treatment with angiotensin II receptor blockers and ACE inhibitors may lead to a greater reduction in hospital admission for heart failure compared with ACE inhibitor treatment alone.

Benefits and harms

Angiotensin II receptor blockers versus placebo:

We found one systematic review (search date 2003, 24 RCTs, 38,080 people with New York Heart Association (NYHA) functional class II–IV, follow-up 4 weeks to 2.7 years).

Mortality

Compared with placebo Angiotensin II receptor blockers are more effective at reducing all-cause mortality at 4 weeks to 2.7 years (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
4623 people
9 RCTs in this analysis
Rate of mortality
299/2821 (11%) with angiotensin II receptor blockers (ARBs)
319/1802 (18%) with placebo

OR 0.83
95% CI 0.69 to 1.00
Small effect sizeARBs

Admission to hospital

Compared with placebo Angiotensin II receptor blockers are more effective at reducing hospital admissions for heart failure (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Heart failure-specific re-admission to hospital

Systematic review
2590 people
3 RCTs in this analysis
Proportion of people admitted to hospital for heart failure
230/1340 (17%) with angiotensin II receptor blockers (ARBs)
314/1250 (25%) with placebo

OR 0.64
95% CI 0.53 to 0.78
Small effect sizeARBs

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

No data from the following reference on this outcome.

Angiotensin II receptor blockers versus ACE inhibitors:

We found two systematic reviews (search date 2003, 10 RCTs, 25,739 people with New York Heart Association functional class II–IV, follow-up 4 weeks to 2.7 years; and search date 2007, 5 RCTs [all of which were identified by the first review], 24,822 people). The reviews differed in their inclusion criteria in minimum number of people enrolled and length of follow-up. The first review included RCTs of any size with a minimum length of follow-up of 4 weeks, whereas the second review specified a minimum number of 500 people and at least 6 months' follow-up. The reviews therefore included different RCTs in their analysis of all-cause mortality and so we report both reviews here.

Mortality

Angiotensin II receptor blockers compared with ACE inhibitors Angiotensin II receptor blockers and ACE inhibitors are equally effective at reducing all-cause mortality at 4 weeks to 2.7 years (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
5201 people
8 RCTs in this analysis
Rate of mortality
331/2889 (11%) with angiotensin II receptor blockers (ARBs)
295/2312 (13%) with ACE inhibitors

OR 1.06
95% CI 0.90 to 1.26
Not significant

Systematic review
4310 people
3 RCTs in this analysis
Rate of mortality
317/2257 (14.0%) with ARBs
286/2053 (13.9%) with ACE inhibitors

RR 1.06
95% CI 0.56 to 1.62
Not significant

Admission to hospital

Angiotensin II receptor blockers compared with ACE inhibitors Angiotensin II receptor blockers and ACE inhibitors are equally effective at 4 weeks to 2.7 years at reducing hospital admissions for heart failure (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Heart failure-specific re-admission to hospital

Systematic review
4310 people
3 RCTs in this analysis
Proportion of people admitted to hospital with heart failure
333/2257 (15%) with angiotensin II receptor blockers (ARBs)
321/2053 (16%) with ACE inhibitors

OR 0.95
95% CI 0.80 to 1.13
Not significant

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

No data from the following reference on this outcome.

Angiotensin II receptor blockers plus ACE inhibitors versus ACE inhibitors alone:

We found three systematic reviews (search date 2003, 7 RCTs, 8260 people with New York Heart Association functional class II–IV heart failure; search date 2003, 4 RCTs; and search date 2007, 3 RCTs, 7999 people). All RCTs identified by the second and third reviews were identified by the first review. However, there was variation among the reviews in their inclusion criteria and outcomes assessed and so we report all three reviews here. The first review included RCTs of any size with a minimum length of follow-up of 4 weeks, whereas the second and third reviews specified at least 6 months' follow-up. The second review (4 RCTs identified by the first systematic review, 7666 people) primarily assessed the effects of angiotensin II receptor blockers (ARBs) plus ACE inhibitors versus ACE inhibitors alone with and without beta-blockers. The third review specified inclusion criteria of 500 or more people and 6 months' or longer follow-up. The review found the same results as the other two reviews. We also found a fourth systematic review (search date 2006) that assessed only adverse effects.

Mortality

Angiotensin II receptor blockers plus ACE inhibitors compared with ACE inhibitors alone We don't know whether angiotensin II receptor blockers plus ACE inhibitors are more effective than ACE inhibitors alone at reducing mortality or a composite outcome of mortality plus morbidity independent of whether people are taking beta-blockers (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
8260 people
7 RCTs in this analysis
Rate of all-cause mortality
903/4265 (21%) with angiotensin II receptor blocker (ARB) plus ACE inhibitor
901/3995 (23%) with ACE inhibitor alone

OR 0.97
95% CI 0.87 to 1.08
Not significant

Systematic review
Number of people not clear
2 RCTs in this analysis
Subgroup analysis
Rate of mortality
with ARB plus ACE inhibitor (without beta-blockers)
with ACE inhibitor alone
Absolute results not reported

OR 0.93
95% CI 0.81 to 1.06
Not significant

Systematic review
Number of people not clear
2 RCTs in this analysis
Subgroup analysis
Rate of mortality
with ARB plus ACE inhibitor (with beta-blockers)
with ACE inhibitor alone
Absolute results not reported

OR 1.08
95% CI 0.90 to 1.29
There was statistical heterogeneity between the RCTs (P <0.05)
Not significant

Systematic review
7999 people
3 RCTs in this analysis
Rate of mortality
901/4119 (22%) with ARB plus ACE inhibitor
900/3980 (23%) with ACE inhibitor alone

RR 0.98
95% CI 0.84 to 1.15
Not significant
Mortality and morbidity

Systematic review
7666 people
4 RCTs in this analysis
Rate of a composite outcome of mortality and morbidity
with ARB plus ACE inhibitor (with or without beta-blockers)
with ACE inhibitor alone
Absolute results not reported

OR 0.89
95% CI 0.81 to 0.98
Small effect sizeARB plus ACE inhibitor

Systematic review
Number of people not clear
2 RCTs in this analysis
Subgroup analysis
Rate of a composite outcome of mortality and morbidity
with ARB plus ACE inhibitor (without beta-blockers)
with ACE inhibitor alone
Absolute results not reported

OR 0.83
95% CI 0.73 to 0.94
Small effect sizeARB plus ACE inhibitor

Systematic review
Number of people not clear
2 RCTs in this analysis
Subgroup analysis
Rate of a composite outcome of mortality and morbidity
with ARB plus ACE inhibitor (with beta-blockers)
with ACE inhibitor alone
Absolute results not reported

OR 0.94
95% CI 0.80 to 1.10
There was statistical heterogeneity between the RCTs (P <0.05)
Not significant

Admission to hospital

Angiotensin II receptor blockers plus ACE inhibitors compared with ACE inhibitors alone Angiotensin II receptor blockers plus ACE inhibitors are more effective at reducing hospital admissions for heart failure (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Heart failure-specific re-admission to hospital

Systematic review
8108 people
4 RCTs in this analysis
Proportion of people admitted to hospital with heart failure
688/4176 (16%) with angiotensin II receptor blocker (ARB) plus ACE inhibitor
819/3932 (21%) with ACE inhibitor alone

OR 0.77
95% CI 0.69 to 0.87
Small effect sizeARB plus ACE inhibitor

Systematic review
7999 people
3 RCTs in this analysis
Proportion of people admitted to hospital with heart failure
686/4119 (17%) with ARB plus ACE inhibitor
818/3980 (21%) with ACE inhibitor alone

RR 0.83
95% CI 0.71 to 0.97
Small effect sizeARB plus ACE inhibitor

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Any adverse event

Systematic review
18,160 people with heart failure or left ventricular dysfunction
9 RCTs in this analysis
Proportion of people experiencing any adverse event
11% with angiotensin II receptor blocker (ARB) plus ACE inhibitor
9% with ACE inhibitor alone
Absolute numbers not reported

RR 1.27
95% CI 1.15 to 1.40
P <0.00001
See further information on studies for methodological issues highlighted by the review
Small effect sizeACE inhibitor alone
Worsening renal function

Systematic review
18,160 people with heart failure or left ventricular dysfunction
9 RCTs in this analysis
Proportion of people with worsening renal function
2% with ARB plus ACE inhibitor
1% with ACE inhibitor alone
Absolute numbers not reported

RR 2.12
95% CI 1.30 to 3.46
Moderate effect sizeACE inhibitor alone
Hypotension

Systematic review
18,160 people with heart failure or left ventricular dysfunction
9 RCTs in this analysis
Hypotension
2% with ARB plus ACE inhibitor
1% with ACE inhibitor alone
Absolute numbers not reported

RR 1.91
95% CI 1.37 to 2.66
P = 0.0002
Small effect sizeACE inhibitor alone
Hyperkalaemia

Systematic review
18,160 people with heart failure or left ventricular dysfunction Proportion of people with hyperkalaemia
0.87% with ARB plus ACE inhibitor
0.20% with ACE inhibitor alone
Absolute numbers not reported

RR 4.17
95% CI 2.31 to 7.53
P <0.00001
Moderate effect sizeACE inhibitor alone

No data from the following reference on this outcome.

Further information on studies

The review identified heterogeneity among studies and therefore conducted subgroup analyses using a statistical test of interaction. Subgroups analysed included the proportion of people with diabetes, sex, follow-up duration, and diuretic use. The only end point that showed significant interaction by subgroup was shorter trial duration leading to higher estimates of RR of renal failure for ARB plus ACE inhibitor relative to ACE inhibitor alone. This review also reported several other limitations. One limitation was that many included studies did not explain how adverse effects were defined; another limitation was that the initial search revealed and excluded some trials in which no adverse events were reported, and it is not clear whether there were actually no adverse events or simply that the investigators did not report them; finally the authors advise caution in interpreting meta-analytic results when the number of events per study are small, as is the case for many studies of adverse events.

Comment

Clinical guide:

Evidence suggests that, in people intolerant of ACE inhibitors, an ARB would be as effective at reducing mortality and morbidity. Furthermore, the evidence suggests that, for people with NYHA functional class II–IV, an ARB should be added to treatment after ACE inhibition and beta-blocker treatment have been optimised, to further reduce both mortality and morbidity.

Substantive changes

Angiotensin II receptor blockers for treating heart failure New evidence added. Categorisation unchanged (Beneficial).

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Beta-blockers

Summary

Beta-blockers reduce mortality and hospital admissions from heart failure compared with placebo, with greater absolute benefits seen in people with more severe heart failure.

Benefits and harms

Beta-blockers versus placebo:

We found three systematic reviews (search date 2000, 22 RCTs, 10,315 people; search date 2004, 28 RCTs, 7637 people; and search date 2008, 23 RCTs, 19,209 people) in people with any severity of heart failure. The reviews identified many RCTs in common, however they applied different inclusion criteria (e.g., in minimum number of people enrolled and length of follow-up) and reported on different outcomes. We found one systematic review (search date 2002, 9 RCTs, 14,594 people followed up for 6–24 months) that assessed adverse effects of beta-blockers in people with heart failure.

Mortality

Compared with placebo (in people with any severity of heart failure) Beta-blockers seem more effective at reducing mortality in people with heart failure of any severity also receiving triple therapy, and in particular ACE inhibitors. (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
10,315 people with heart failure, most receiving triple therapy, and in particular ACE inhibitors
22 RCTs in this analysis
Rate of mortality
444/5273 (8%) with beta-blockers
624/4862 (13%) with placebo

OR 0.65
95% CI 0.53 to 0.80
This is equivalent to 3 fewer deaths per 100 people treated for 1 year
The results were consistent for selective and non-selective beta-blockers
Sensitivity analysis and funnel plots found that publication bias was unlikely
Small effect sizebeta-blockers

Systematic review
19,209 people with heart failure, most receiving ACE inhibitors and digoxin
23 RCTs in this analysis
Rate of mortality
1205/9820 (12%) with beta-blockers
1515/9389 (16%) with placebo

RR 0.76
95% CI 0.68 to 0.84
There was significant statistical heterogeneity among RCTs (P = 0.09)
Small effect sizebeta-blockers

No data from the following reference on this outcome.

Admission to hospital

Compared with placebo (in people with any severity of heart failure) Beta-blockers may be more effective at reducing hospital admissions in people with heart failure of any severity also receiving triple therapy, and in particular ACE inhibitors, and may be more effective at reducing a composite outcome of mortality and hospital admissions in older people (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause hospital admissions

Systematic review
10,315 people with heart failure, most receiving triple therapy, and in particular ACE inhibitors
22 RCTs in this analysis
Proportion of people admitted to hospital for any cause
540/5244 (10%) with beta-blockers
754/4832 (16%) with placebo

OR 0.64
95% CI 0.53 to 0.79
This is equivalent to 4 fewer hospital admissions per 100 people treated for 1 year
The results were consistent for selective and non-selective beta-blockers
Sensitivity analysis and funnel plots found that publication bias was unlikely
Small effect sizebeta-blockers
Death or hospital admission

RCT
2128 older people with heart failure, mean age 76 years, mean left ventricular ejection fraction (LVEF) 36%, 35% of people had LVEF >35%
In review
Rate of composite end point of all-cause mortality or cardiovascular hospital admission
332/1067 (31%) with nebivolol
375/1061 (35%) with placebo

HR 0.86
95% CI 0.74 to 0.99
Small effect sizebeta-blockers

No data from the following reference on this outcome.

Functional improvement

Compared with placebo (in people with any severity of heart failure) Beta-blockers seem more effective at increasing the proportion of people with an improvement in function (New York Heart Association functional classification) by at least one class, and at improving exercise time (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Change in NYHA class

Systematic review
7511 people with heart failure
25 RCTs in this analysis
Proportion of people who improved NYHA class by at least 1 class
with beta-blockers
with placebo
Absolute results not reported

OR 1.80
95% CI 1.33 to 2.43
P <0.0001
Small effect sizebeta-blockers
Exercise duration

Systematic review
1120 people with heart failure
10 RCTs in this analysis
Proportion of people with improved exercise time
with beta-blockers
with placebo
Absolute results not reported

Mean difference 44.19 seconds
95% CI 6.62 seconds to 81.75 seconds
P = 0.021
Effect size not calculatedbeta-blockers

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Withdrawal for any cause

Systematic review
14,594 people
9 RCTs in this analysis
Proportion of people withdrawing from RCT for any cause
1195/7458 (16%) with beta-blockers
1287/7136 (18%) with placebo

RR 0.89
95% CI 0.81 to 0.98
Small effect sizebeta-blockers
Cardiac adverse effects

RCT
2128 older people with heart failure, mean age 76 years, mean left ventricular ejection fraction (LVEF) 36%, 35% of people had LVEF <35%
In review
Proportion of people experiencing aggravated cardiac failure
256/1067 (24%) with nebivolol
265/1061 (25%) with placebo

Significance not assessed

Systematic review
4439 people
4 RCTs in this analysis
Proportion of people with worsening heart failure
625/2379 (26%) with beta-blockers
691/2060 (34%) with placebo

RR 0.83
95% CI 0.71 to 0.98
Small effect sizebeta-blockers

Systematic review
13,796 people
7 RCTs in this analysis
Proportion of people with bradycardia
400/7057 (6%) with beta-blockers
118/6739 (2%) with placebo

RR 3.62
95% CI 2.48 to 5.28
Moderate effect sizeplacebo
Dizziness

Systematic review
10,082 people
4 RCTs in this analysis
Proportion of people with dizziness
1117/5196 (22%) with beta-blockers
810/4886 (17%) with placebo

RR 1.37
95% CI 1.09 to 1.71
Small effect sizeplacebo
Hypotension

Systematic review
13,796 people
7 RCTs in this analysis
Proportion of people with hypotension
535/7057 (8%) with beta-blockers
409/6739 (6%) with placebo

RR 1.41
95% CI 0.96 to 2.06
Not significant
Fatigue

Systematic review
7793 people
3 RCTs in this analysis
Proportion of people with fatigue
953/4040 (24%) with beta-blockers
840/3753 (22%) with placebo

RR 1.04
95% CI 0.97 to 1.11
Not significant

No data from the following reference on this outcome.

Beta-blockers versus placebo (in people with severe heart failure):

We found two systematic reviews (search dates not reported) assessing the effects of beta-blockers in people with severe heart failure, which identified 7 RCTs between them. The second systematic review (6 RCTs, 13,370 people with chronic heart failure [people with New York Heart Association functional class III or IV heart failure]) assessed the effects of beta-blockers in people with and without ACE inhibitors or angiotensin II receptor blockers (ARBs) at baseline. We found one systematic review (search date 2002, 9 RCTs, 14,594 people followed up for 6 to 24 months) that assessed adverse effects of beta-blockers in people with heart failure.

Mortality

Compared with placebo (in people with severe heart failure) Beta-blockers seem more effective at reducing mortality in people with severe heart failure who are also taking ACE inhibitors and diuretics with or without digitalis, but may be no more effective in those not taking ACE inhibitors or angiotensin II receptor blockers at baseline (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
635 people with New York Heart Association (NYHA) functional class IV heart failure, on ACE inhibitor and diuretic with or without digitalis
4 RCTs in this analysis
Rate of mortality
56/313 (18%) with beta-blockers
81/322 (25%) with placebo

RR 0.71
95% CI 0.52 to 0.96
Small effect sizebeta-blockers

Systematic review
12,728 people with chronic heart failure (people with NYHA functional class III or IV heart failure)
6 RCTs in this analysis
Subgroup analysis
Rate of mortality
867/6496 (13%) with beta-blockers
1120/6232 (18%) with placebo

RR 0.76
95% CI 0.71 to 0.83
Small effect sizebeta-blockers

Systematic review
642 people with chronic heart failure (people with NYHA functional class III or IV heart failure)
6 RCTs in this analysis
Subgroup analysis
Rate of mortality
50/347 (14%) with beta-blockers
62/295 (21%) with placebo

RR 0.73
95% CI 0.53 to 1.02
Not significant

Admission to hospital

Compared with placebo (in people with severe heart failure) Beta-blockers may be more effective at reducing the combined outcome of death and hospital admissions independent of whether people are taking ACE inhibitors or angiotensin II receptor blockers at baseline (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Death or hospital admission

Systematic review
Number of people in analysis not clear (8988 people in 3 RCTs)
3 RCTs in this analysis
Subgroup analysis
Rate of composite outcome of death or hospital admission for heart failure
26% with beta-blockers
33% with placebo
Absolute numbers not reported

RR 0.78
95% CI 0.74 to 0.83
Small effect sizebeta-blockers

Systematic review
Number of people in analysis not clear (8988 people in 3 RCTs)
3 RCTs in this analysis
Subgroup analysis
Rate of composite outcome of death or hospital admission for heart failure
28% with beta-blockers
35% with placebo
Absolute numbers not reported

RR 0.81
95% CI 0.61 to 1.08
Not significant

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Withdrawal for any cause

Systematic review
14,594 people
9 RCTs in this analysis
Proportion of people withdrawing from RCT for any cause
1195/7458 (16%) with beta-blockers
1287/7136 (18%) with placebo

RR 0.89
95% CI 0.81 to 0.98
Small effect sizebeta-blockers
Cardiac adverse effects

Systematic review
4439 people
4 RCTs in this analysis
Proportion of people with worsening heart failure
625/2379 (26%) with beta-blockers
691/2060 (34%) with placebo

RR 0.83
95% CI 0.71 to 0.98
Small effect sizebeta-blockers

Systematic review
13,796 people
7 RCTs in this analysis
Proportion of people with bradycardia
400/7057 (6%) with beta-blockers
118/6739 (2%) with placebo

RR 3.62
95% CI 2.48 to 5.28
Moderate effect sizeplacebo
Dizziness

Systematic review
10,082 people
4 RCTs in this analysis
Proportion of people with dizziness
1117/5196 (22%) with beta-blockers
810/4886 (17%) with placebo

RR 1.37
95% CI 1.09 to 1.71
Small effect sizeplacebo
Hypotension

Systematic review
13,796 people
7 RCTs in this analysis
Proportion of people with hypotension
535/7057 (8%) with beta-blockers
409/6739 (6%) with placebo

RR 1.41
95% CI 0.96 to 2.06
Not significant
Fatigue

Systematic review
7793 people
3 RCTs in this analysis
Proportion of people with fatigue
953/4040 (24%) with beta-blockers
840/3753 (22%) with placebo

RR 1.04
95% CI 0.97 to 1.11
Not significant

No data from the following reference on this outcome.

Beta-blockers versus ACE inhibitors:

We found one RCT.

Admission to hospital

Compared with ACE inhibitors The beta-blocker bisoprolol and the ACE inhibitor enalapril may be equally effective at reducing the composite outcome of all-cause mortality or hospital admission in people with heart failure (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Death or hospital admission

RCT
1010 people with left ventricular ejection fraction (LVEF) 35% or less and not receiving ACE inhibitors, beta-blockers, or angiotensin II receptor blockers (ARBs) Rate of composite outcome of all-cause mortality or hospital admission 6 months
178/505 (35%) with bisoprolol (10 mg daily)
186/505 (37%) with enalapril (10 mg twice daily)

HR 0.94
95% CI 0.77 to 1.16
In the intention-to-treat analysis, bisoprolol was found to be non-inferior compared with enalapril
These data suggest that bisoprolol may be as safe and efficacious as enalapril for treating heart failure
Not significant

Mortality

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

No data from the following reference on this outcome.

Further information on studies

None.

Comment

Clinical guide

Fears that beta-blockers may cause excessive problems with worsening heart failure, bradyarrhythmia, or hypotension have not been confirmed. We found good evidence for beta-blockers in people with moderate symptoms (NYHA functional class II or III) receiving standard treatment, including ACE inhibitors. Data suggest that the magnitude of the prognostic benefit conferred by beta-blockers in the absence of ACE inhibitors or ARBs is similar to that of ACE inhibitors. Therefore, evidence suggests that either ACE inhibitors or beta-blockers could be used as first-line treatment in systolic heart failure. The value of beta-blockers is uncertain in heart failure with preserved ejection fraction and in asymptomatic left ventricular systolic dysfunction. One RCT (1959 people) found that carvedilol reduced all-cause mortality compared with placebo in people with MI and LVEF 40% or less (AR for death: 12% with carvedilol v 15% with placebo; HR 0.77, 95% CI 0.60 to 0.98).

Effect of magnitude of heart rate reduction:

One systematic review (search date 2008) examined the relationship between degree of heart rate reduction and death in people with heart failure treated with beta-blockers. The review performed a meta-regression analysis of data from 17 RCTs (17,831 people) and found that, for every 5 beats/minute reduction in heart rate with beta-blocker treatment, the relative risk for death decreased by 18% (95% CI 6% to 29%). These data suggest that the magnitude of the heart rate reduction is significantly associated with the survival benefit, further supporting evidence that both non-selective and selective beta-blockers are effective for heart failure.

Effect of different doses of beta-blocker:

The review (search date 2008) also compared the effects of different dosing schedules of beta-blockers on mortality. It compared results from higher dose trials (trials where people received a target dose of 50% or more of the dose recommended in guidelines) with results from lower dose trials. The review found no significant difference between dosing schedule and reduction in all-cause mortality (17 RCTs, 17,660 people; RR 1.02, 95% CI 0.93 to 1.10 per dose increment; P = 0.69). The review reported that the RR for death was 0.74 (95% CI 0.64 to 0.86) in 15 trials in which people received high-dose beta-blockers, and the RR was 0.78 (95% CI 0.63 to 0.96) in 7 trials in which people received low-dose beta-blockers.

Effects of different beta-blockers:

One systematic review (search date 2009, 4 RCTs, 3501 people) compared non-selective versus selective beta-blockers in people with heart failure. It found that non-selective beta-blockers significantly reduced all-cause mortality compared with selective beta-blockers (538/1754 [31%] with non-selective v 625/1747 [36%] with selective; RR 0.86, 95% CI 0.78 to 0.94). It should be noted that one trial (the COMET trial) made up 96% of these results. The analysis of indirect comparisons of non-selective beta-blockers and selective beta-blockers found similar effects on mortality with non-selective (RR 0.75, 95% CI 0.61 to 0.92; P = 0.005) and selective beta-blockers (RR 0.76, 95% CI 0.68 to 0.87). Similarly, the analysis of indirect comparisons found similar rates of vascular events (fatal and non-fatal strokes, fatal and non-fatal MI, fatal pulmonary embolisms, other venous thromboembolic events) with both non-selective (RR 0.88, 95% CI 0.64 to 1.00) and selective (RR 1.33, 95% CI 0.86 to 2.04) beta-blockers; neither class of beta-blocker significantly reduced vascular events. It must be noted that the vascular event rate was relatively low, especially in people receiving selective beta-blockers (84 events in total; only 3 RCTs included in the analysis). The non-selective (RR 0.82, 95% CI 0.70 to 0.95) and selective (RR 0.77, 95% CI 0.61 to 0.97) beta-blockers equally decreased the events of fatal or non-fatal worsening heart failure. Thus, the data would suggest that non-selective and selective beta-blockers that have been used in clinical trials are equally effective for the management of heart failure.

One RCT suggested that results for non-black people were consistent between the non-selective beta-blockers bucindolol and carvedilol.

Effects in different populations:

The lack of observed benefit for black people in one RCT raises the possibility that there may be race-specific responses to pharmacological treatment for cardiovascular disease. There may also be different responses in people with diabetes mellitus. A meta-analysis (6 RCTs, 13,129 people) examined whether beta-blockers in people with heart failure are as efficacious in those with as without diabetes mellitus. It found that overall mortality was significantly increased in people with diabetes mellitus compared with people without diabetes mellitus, regardless of treatment (RR 1.25, 95% CI 1.15 to 1.36). Carvedilol has also been assessed in people with diabetes in a meta-analysis because it is believed that carvedilol has unique characteristics compared with other beta-blockers. In this meta-analysis, 7 RCTs were examined (5757 people, 25% with diabetes mellitus) to determine whether the effects of carvedilol were similar in people with and without diabetes mellitus. There was no significant difference in mortality or the number needed to treat (NNT) to prevent one death for 1 year for people with or without diabetes (mortality in people with diabetes: carvedilol v placebo: RRR 28%, 95% CI 3% to 46%; P = 0.03; people without diabetes: RRR 37%, 95% CI 22% to 48%; P <0.001; difference between 2 groups reported as not significant; P value not reported; NNT 25, 95% CI 14 to 118 for people with diabetes mellitus v NNT 23, 95% CI 17 to 37 for people without diabetes mellitus). Although beta-blockers significantly reduced mortality compared with placebo in people with diabetes mellitus (RR 0.84, 95% CI 0.73 to 0.96), the magnitude of benefit was significantly lower than that in people who did not have diabetes mellitus (P = 0.023).

One systematic review (search date 2007) identified assessed whether the magnitude of the benefit of beta-blockers differs in ischaemic and non-ischaemic heart failure. The review searched for RCTs that reported mortality data for people with ischaemic or non-ischaemic heart failure separately. In the RCTs identified by the review (4 RCTs, 7250 people), heart failure was associated with ischaemic aetiology in 4746 (65%) people included in the analysis and with non-ischaemic aetiology in 2504 (35%) people. The review found the risk reduction in mortality for non-ischaemic heart failure (75/1335 [6%] with beta-blocker v 108/1169 [9%] with placebo; RR 0.62, 95% CI 0.45 to 0.84; P = 0.002) to be similar to that of ischaemic heart failure (226/2457 [9%] with beta-blocker v 324/2289 [14%] with placebo; RR 0.62, 95% CI 0.52 to 0.75; P <0.00001).

We found one systematic review (search date not reported, 5 RCTs, 17,346 people) investigating whether beta-blockers are as effective in older people as in non-elderly people for chronic heart failure. The cut-off points for older age ranges varied across trials (59–71 years). The review found that beta-blocker treatment significantly reduced all-cause mortality for non-elderly people (RR 0.66, 95% CI 0.52 to 0.85; P = 0.001; absolute numbers not reported) and for older people (RR 0.76, 95% CI 0.64 to 0.90; P = 0.002; absolute numbers not reported), without a statistically significant difference in mortality reduction between the two groups (P = 0.38).

Substantive changes

Beta-blockers New evidence added. Categorisation unchanged (Beneficial).

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Digoxin

Summary

Digoxin slows the progression of heart failure compared with placebo, but may not reduce mortality.

Benefits and harms

Digoxin versus placebo:

We found one systematic review (search date 2003, 13 RCTs with >7 weeks' follow-up, 7896 people in sinus rhythm).

Mortality

Compared with placebo Digoxin seems no more effective at reducing mortality in people in sinus rhythm (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
7756 people
8 RCTs in this analysis
Rate of mortality
with digoxin
with placebo
Absolute results not reported

OR 0.98
95% CI 0.89 to 1.09
See further information on studies for separate reporting of data from largest RCT in meta-analysis (6800 people)
Not significant

Admission to hospital

Compared with placebo Digoxin seems more effective at reducing all-cause hospital admissions in people receiving ACE inhibitors and diuretics (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Hospital admission

Systematic review
7262 people
4 RCTs in this analysis
Proportion of people admitted to hospital for any cause
with digoxin
with placebo
Absolute results not reported

OR 0.68
95% CI 0.61 to 0.75
See further information on studies for separate reporting of data from largest RCT in meta-analysis (6800 people)
Small effect sizedigoxin

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Digoxin toxicity

RCT
6800 people
In review
Proportion of people with digoxin toxicity
12% with digoxin
8% with placebo
Absolute numbers not reported

ARI 4%
95% CI 2% to 6%
RR 1.50
95% CI 1.30 to 1.73
Small effect sizeplacebo
Cardiac adverse effects

RCT
6800 people
In review
Proportion of people with ventricular fibrillation or tachycardia
37/3397 (1.1%) with digoxin
27/3403 (0.8%) with placebo

ARI +0.3%
95% CI –0.1% to +1.0%
RR 1.37
95% CI 0.84 to 2.24
Not significant

RCT
6800 people
In review
Proportion of people with supraventricular arrhythmia
3% with digoxin
1% with placebo
Absolute numbers not reported

ARI 1.3%
95% CI 0.5% to 2.4%
RR 2.08
95% CI 1.44 to 2.99
Moderate effect sizeplacebo

RCT
6800 people
In review
Proportion of people with second- or third-degree atrioventricular block
1.2% with digoxin
0.4% with placebo
Absolute numbers not reported

ARI 0.8%
95% CI 0.2% to 1.8%
RR 2.93
95% CI 1.61 to 5.34
Moderate effect sizeplacebo

No data from the following reference on this outcome.

Further information on studies

The largest RCT in the review, which dominated the meta-analysis (6800 people, 88% male, mean age 64 years, New York Heart Association [NYHA] functional class I to III, 94% already taking ACE inhibitors, 82% taking diuretics), compared blinded additional treatment with either digoxin or placebo for a mean of 37 months. It found no significant difference between digoxin and placebo in all-cause mortality (1181/3397 [34.8%] with digoxin v 1194/3403 [35.1%] with placebo; ARR +0.3%, 95% CI –2.0% to +2.6%; RR 0.99, 95% CI 0.93 to 1.06). It found that digoxin significantly reduced admission rates for heart failure over 37 months compared with placebo and reduced the combined outcome of death or hospital admission caused by worsening heart failure (heart failure admissions: 910/3397 [27%] with digoxin v 1180/3403 [35%] with placebo; ARR 8%, 95% CI 6% to 10%; RR 0.77, 95% CI 0.72 to 0.83; NNT 13, 95% CI 10 to 17; death or hospital admission: 1041/3397 [31%] with digoxin v 1291/3403 [38%] with placebo; ARR 7%, 95% CI 5% to 9%; RR 0.81, 95% CI 0.75 to 0.87).

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Positive inotropes other than digoxin

Summary

Positive inotropic agents (other than digoxin) may increase mortality and should be used with caution, if at all, in people with systolic heart failure.

Benefits and harms

Positive inotropes (other than digoxin) versus placebo:

We found two systematic reviews and two additional RCTs on inotropic agents. The first systematic review (search date 2000, 21 RCTs, 632 people) assessed the effects of intravenous inotropic agents that act through the adrenergic pathway (beta-agonists and phosphodiesterase inhibitors) in people with heart failure. The review identified 11 RCTs comparing inotropic agents (including dobutamine, dopexamine, toborinone, and milrinone) versus placebo or control. The second review (search date 2004, 21 RCTs, 8408 people) assessed the effects of phosphodiesterase inhibitors.

Mortality

Compared with placebo Positive inotropic drugs other than digoxin (including intravenous inotropes acting through the adrenergic pathway, and phosphodiesterase III inhibitors) seem less effective at reducing mortality at 6 to 11 months (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Mortality

Systematic review
Number of people not clear
11 RCTs in this analysis
Rate of mortality
with intravenous inotropes that act through the adrenergic pathway
with placebo or control
Absolute results not reported

OR 1.50
95% CI 0.51 to 3.92
See further information on studies for authors' conclusions
Not significant

Systematic review
8408 people
21 RCTs in this analysis
Rate of mortality
897/5138 (17%) with phosphodiesterase III inhibitors (PDIs)
478/3270 (15%) with placebo or control

RR 1.17
95% CI 1.06 to 1.30
See further information on studies for discussion of effects of PDIs
Small effect sizeplacebo or control

RCT
3833 people with heart failure Rate of mortality 9 months
292/1275 (23%) with vesnarinone (a PDI) 60 mg daily
242/1280 (19%) with placebo

ARI 4%
95% CI 1% to 8%
RR 1.21
95% CI 1.04 to 1.40
Small effect sizeplacebo

RCT
1906 people with heart failure Rate of mortality 11 months
232/953 (24%) with ibopamine
193/953 (20%) with placebo

RR 1.26
95% CI 1.04 to 1.53
Small effect sizeplacebo

Admission to hospital

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

No data from the following reference on this outcome.

Further information on studies

The authors of the review concluded that "intravenous inotropic agents acting through the adrenergic pathway are often used in people with worsening heart failure to achieve arbitrary haemodynamic targets. Our analyses show that there is very little evidence that such treatment improves symptoms or patient outcomes, and may not be safe". Of the 21 RCTs identified, 16 RCTs (474 people) were acute invasive haemodynamic studies of symptomatically severe heart failure, and 5 RCTs (158 people) were based on intermittent inotropic treatment in an outpatient setting. Included RCTs were often small.

Considering mortality from all causes, the deleterious effects of PDIs were consistent, regardless of the severity of heart failure, use of background treatment, or type of PDI.

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Aldosterone receptor antagonists

Summary

Aldosterone receptor antagonists (spironolactone, eplerenone, and canrenoate) may reduce all-cause mortality in people with heart failure, but increase the risk of hyperkalaemia.

Benefits and harms

Aldosterone receptor antagonists versus placebo:

We found one systematic review (search date 2008, 19 RCTs, 10,807 people) assessing the effectiveness of aldosterone receptor antagonists (spironolactone, eplerenone, and canrenoate) in people with symptomatic and asymptomatic left ventricular dysfunction, including heart failure and post MI with heart failure. The review did not specify left ventricular ejection fraction (LVEF) for inclusion: two identified RCTs (58 people) recruited people with an LVEF >45%, and 5 RCTs (883 people) did not report LVEF of people included. Of the 19 RCTs identified, 15 RCTs included people with chronic heart failure (3395 people), and 4 RCTs included people who had previous MI and had heart failure (7412 people). Two RCTs (134 people) compared aldosterone receptor antagonists versus usual care rather than versus placebo and one RCT (105 people) assessed an active comparator (metoprolol) plus usual care.

Mortality

Compared with placebo Aldosterone receptor antagonists (spironolactone, eplerenone, and canrenoate) seem more effective at reducing all-cause mortality in people with heart failure and in people with heart failure after an MI (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
10,807 people
19 RCTs in this analysis
Rate of mortality
804/5565 (14%) with aldosterone receptor antagonists
994/5200 (19%) with control

RR 0.80
95% CI 0.74 to 0.87
The results of the review may not be generalisable to all people with heart failure (see further information on studies for more details)
Small effect sizealdosterone receptor antagonists

Systematic review
3353 people with heart failure
15 RCTs in this analysis
Subgroup analysis
Rate of mortality
303/1858 (16%) with aldosterone receptor antagonists
404/1495 (27%) with control

RR 0.75
95% CI 0.67 to 0.84
The results of the review may not be generalisable to all people with heart failure (see further information on studies for more details)
Small effect sizealdosterone receptor antagonists

Systematic review
7412 people with heart failure after an MI
4 RCTs in this analysis
Subgroup analysis
Rate of mortality
501/3707 (14%) with aldosterone receptor antagonists
590/3705 (16%) with control

RR 0.85
95% CI 0.76 to 0.95
The results of the review may not be generalisable to all people with heart failure (see further information on studies for more details)
Small effect sizealdosterone receptor antagonists

Admission to hospital

Compared with placebo Aldosterone receptor antagonists (spironolactone, eplerenone, and canrenoate) seem more effective at reducing rate of hospital admission in people with heart failure (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Hospital admission for any cause

Systematic review
8699 people
9 RCTs in this analysis
Rate of all-cause hospital re-admission
with aldosterone receptor antagonists
with placebo
Absolute results not reported

RR 0.77
95% CI 0.68 to 0.87
The results of the review may not be generalisable to all people with heart failure (see further information on studies for more details)
Small effect sizealdosterone receptor antagonists

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Hyperkalaemia

Systematic review
10,261 people
17 RCTs in this analysis
Proportion of people with hyperkalaemia
315/5314 (6%) with aldosterone receptor antagonists
148/4947 (3%) with control

Significance not assessed
The results of the review may not be generalisable to all people with heart failure (see further information on studies for more details)
Worsening renal failure

Systematic review
1613 people
11 RCTs in this analysis
Proportion of people with worsening renal failure
86/959 (9%) with aldosterone receptor antagonists
11/654 (2%) with control

Significance not assessed
The results of the review may not be generalisable to all people with heart failure (see further information on studies for more details)
Gynaecomastia

Systematic review
10,213 people
16 RCTs in this analysis
Proportion of people with gynaecomastia
88/5291 (2%) with aldosterone receptor antagonists
26/4922 (1%) with control

Significance not assessed
The results of the review may not be generalisable to all people with heart failure (see further information on studies for more details)

Further information on studies

The two largest studies identified by the review (one assessing spironolactone including 1663 people and one assessing eplerenone including 6632 people) included people with only New York Heart Association (NYHA) functional class III or IV; therefore, these results cannot necessarily be generalised to people with milder heart failure. The RCT of eplerenone was limited to people who were post-MI with heart failure and therefore these results cannot necessarily be generalised to people with stable heart failure — that is, those without a recent MI and who have milder symptoms of heart failure. The contribution of these two large RCTs, which represent 76% (8295 people) of the people included in the analysis carried out by the review, should be considered when interpreting the results of the review.

Comment

A population-based time series analysis examined the trends in the rate of spironolactone prescriptions and the rate of hospital admissions for hyperkalaemia in ambulatory patients before and after the publication of an RCT that demonstrated the benefits of spironolactone. The spironolactone prescription rate significantly increased after publication of the RCT (rising from 34/1000 people to 149/1000 people; P <0.001). There was also a significant increase in the rate of hospital admission for hyperkalaemia (from 2.4/1000 people to 11.0/1000 people; P <0.001) and associated mortality (from 0.3/1000 people to 2.0/1000 people; P <0.001). The results of the study are important because they emphasise the need for appropriate monitoring of people treated with spironolactone.

Substantive changes

No new evidence

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Amiodarone

Summary

We don't know whether amiodarone is effective at reducing mortality.

Benefits and harms

Amiodarone versus placebo or conventional care:

We found two systematic reviews comparing amiodarone versus placebo in heart failure. The most recent review (search date 1997, 10 RCTs, 4766 people) included people with a wide range of conditions (symptomatic and asymptomatic heart failure, ventricular arrhythmia, recent MI, and recent cardiac arrest). The earlier systematic review (search date not reported) found 8 RCTs (5101 people after MI) comparing prophylactic amiodarone versus placebo or usual care, and 5 RCTs (1452 people) in people with heart failure.

Mortality

Compared with placebo or conventional treatment Amiodarone may be more effective at 3 to 24 months at reducing all-cause mortality (in people with a wide range of heart conditions such as symptomatic and asymptomatic heart failure, ventricular arrhythmia, recent MI, and recent cardiac arrest) and at reducing arrhythmic death or sudden death (very low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
4525 people
8 RCTs in this analysis
Rate of mortality 3 to 24 months
436/2262 (19%) with amiodarone
507/2263 (22%) with placebo or conventional treatment

ARR 3.0%
95% CI 0.8% to 5.3%
RR 0.86
95% CI 0.76 to 0.96
Small effect sizeamiodarone

Systematic review
6553 people
13 RCTs in this analysis
Rate of annual mortality mean follow-up of 16 months
11% with amiodarone
12% with placebo or conventional treatment
Absolute numbers not reported

Fixed effects model
OR 0.87
95% CI 0.78 to 0.99
Random effects model
OR 0.85
95% CI 0.71 to 1.02
The effect of amiodarone was significantly greater in RCTs comparing amiodarone versus usual care than in placebo-controlled RCTs

Systematic review
6553 people
13 RCTs in this analysis
Rate of arrhythmic death or sudden death mean follow-up of 16 months
with amiodarone
with placebo or conventional treatment
Absolute results not reported

OR 0.71
95% CI 0.59 to 0.85
The effect of amiodarone was significantly greater in RCTs comparing amiodarone versus usual care than in placebo-controlled RCTs
Small effect sizeamiodarone

Systematic review
1452 people with heart failure
5 RCTs in this analysis
Subgroup analysis
Rate of annual mortality mean follow-up of 16 months
20% with amiodarone
24% with placebo or conventional treatment
Absolute numbers not reported

OR 0.83
95% CI 0.70 to 0.99
The effect of amiodarone was significantly greater in RCTs comparing amiodarone versus usual care than in placebo-controlled RCTs
Small effect sizeamiodarone

Admission to hospital

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Withdrawal because of adverse effects

Systematic review
Number of people and RCTs in analysis not clear Proportion of people permanently discontinuing study medication, primarily owing to adverse effects 2 years
41% with amiodarone
27% with placebo
Absolute numbers not reported
Adverse effects

Systematic review
Number of people in analysis not clear
10 RCTs in this analysis
Odds of reporting adverse drug reactions
with amiodarone
with placebo
Absolute results not reported

OR 2.22
95% CI 1.83 to 2.68
Moderate effect sizeplacebo

Systematic review
Number of people and RCTs in analysis not clear Proportion of people reporting hypothyroidism
7% with amiodarone
1% with placebo
Absolute numbers not reported

Systematic review
Number of people and RCTs in analysis not clear Proportion of people with hyperthyroidism
1.4% with amiodarone
0.5% with placebo
Absolute numbers not reported

Systematic review
Number of people and RCTs in analysis not clear Proportion of people with peripheral neuropathy
0.5% with amiodarone
0.2% with placebo
Absolute numbers not reported

Systematic review
Number of people and RCTs in analysis not clear Proportion of people with lung infiltrates
2% with amiodarone
1% with placebo
Absolute numbers not reported

Systematic review
Number of people and RCTs in analysis not clear Proportion of people with bradycardia
2% with amiodarone
1% with placebo
Absolute numbers not reported

Systematic review
Number of people and RCTs in analysis not clear Proportion of people with liver dysfunction
1% with amiodarone
0.4% with placebo
Absolute numbers not reported

No data from the following reference on this outcome.

Further information on studies

None.

Comment

Clinical guide:

RCTs of amiodarone versus usual treatment found larger effects than placebo-controlled trials. These findings suggest bias; unblinded follow-up may be associated with reduced usual care or improved adherence with amiodarone. Further studies are required to assess the effects of amiodarone treatment on mortality and morbidity in people with heart failure.

Substantive changes

No new evidence

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Antiarrhythmics other than amiodarone

Summary

Evidence extrapolated from studies in people treated after an MI suggest that other antiarrhythmic drugs (apart from beta-blockers) may be associated with increased mortality in people with heart failure.

Benefits and harms

Antiarrhythmics other than amiodarone:

We found no systematic review or RCTs. Apart from beta-blockers, other antiarrhythmic drugs increase mortality in people at high risk (see class I antiarrhythmic agents [quinidine, procainamide, disopyramide, encainide, flecainide, and moracizine] in review on secondary prevention of ischaemic cardiac events).

Further information on studies

None.

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Anticoagulation

Summary

We don't know whether anticoagulants are effective at reducing mortality.

Benefits and harms

Anticoagulation versus placebo or no antithrombotic treatment:

We found one systematic review (search date 2005, 1 RCT), and one subsequent RCT.

Mortality

Warfarin compared with placebo We don't know whether warfarin is more effective at 27 months than no antithrombotic treatment at reducing a combined outcome of death, MI, and stroke (very-low quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Death, MI, or stroke

RCT
3-armed trial
279 people, 70% with New York Heart Association (NYHA) functional class III
In review
Rate of combined outcome of death, MI, and stroke mean follow-up of 27 months
26% with warfarin (international normalised ratio [INR] 2.5)
27% with no antithrombotic treatment
Absolute numbers not reported

Reported as not significant (warfarin v no antithrombotic treatment)
P value not reported
Not significant

RCT
197 people aged 20 to 75 years with NYHA class II to IV caused by either previous MI or idiopathic dilated cardiomyopathy Incidence of primary outcome (composite of non-fatal stroke, peripheral or pulmonary embolism, MI, hospital admission, exacerbation of heart failure, or death from any cause) mean follow-up of 27 months
8.9/100 patient-years with warfarin
14.8/100 patient-years with placebo

Significance not assessed

Admission to hospital

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Haemorrhagic events

RCT
3-armed trial
279 people, 70% with New York Heart Association (NYHA) functional class III
In review
Rate of haemorrhagic events mean follow-up of 27 months
with warfarin (international normalised ratio [INR] 2.5)
with no antithrombotic treatment

RCT
197 people aged 20 to 75 years with NYHA class II to IV caused by either previous MI or idiopathic dilated cardiomyopathy Haemorrhagic event rate mean follow-up of 27 months
4.6/100 patient-years with warfarin
0/100 patient-years with placebo

Significance not assessed

Anticoagulation versus antiplatelet agents:

See option on antiplatelet agents.

Further information on studies

None.

Comment

The systematic review (search date 2005) found three additional non-randomised trials. Meta-analysis of these trials and the RCT found that anticoagulant significantly reduced death from all causes and cardiovascular event rates compared with control (death from all causes: 1087 people; OR 0.64, 95% CI 0.45 to 0.90; cardiovascular event rates: 1130 people; OR 0.26, 95% CI 0.16 to 0.43). Meta-analysis of two non-randomised trials (645 people) found no significant difference in bleeding complications between warfarin and no warfarin (OR 1.52, 95% CI 0.56 to 4.10). The non-randomised controlled studies were performed in the early 1950s in hospitalised people with a high prevalence of rheumatic heart disease and atrial fibrillation, and the methods used may be considered unreliable today.

One retrospective analysis assessed the effect of anticoagulants used at the discretion of individual investigators in RCTs on the incidence of stroke, peripheral arterial embolism, and pulmonary embolism. The first cohort was from one RCT (642 men with chronic heart failure) comparing hydralazine plus isosorbide dinitrate versus prazosin versus placebo. The second cohort was from another RCT (804 men with chronic heart failure) comparing enalapril versus hydralazine plus isosorbide dinitrate. All people were given digoxin and diuretics. The retrospective analysis found that, without treatment, the incidence of all thromboembolic events was low (2.7/100 patient-years in the first RCT; 2.1/100 patient-years in the second RCT) and that anticoagulation did not reduce the incidence of thromboembolic events (2.9/100 patient-years in the first RCT; 4.8/100 patient-years in the second RCT). In this group, atrial fibrillation was not associated with a higher risk of thromboembolic events.

A second retrospective analysis was from two large RCTs (2569 people with symptomatic and asymptomatic left ventricular dysfunction) comparing enalapril versus placebo. The analysis found that people treated with warfarin at baseline had a significantly lower risk of death during follow-up (HR adjusted for baseline differences 0.76, 95% CI 0.65 to 0.89). Warfarin use was associated with a reduction in the combined outcome of death plus hospital admission for heart failure (adjusted HR 0.82, 95% CI 0.72 to 0.93). The benefit with warfarin use was not significantly influenced by the presence of symptoms, randomisation to enalapril or placebo, sex, presence of atrial fibrillation, age, ejection fraction, NYHA functional class, or cause of heart failure. Warfarin reduced cardiac mortality, specifically deaths that were sudden or associated with either heart failure or MI.

Neither of the retrospective studies was designed to determine the incidence of thromboembolic events in heart failure or the effects of treatment. Neither study included information about the intensity of anticoagulation or warfarin use. We found several additional cohort studies showing a reduction in thromboembolic events with anticoagulation, but they all reported on too few people to provide useful results. The two RCTs are of inadequate size to definitively conclude whether anticoagulation is of benefit in people with heart failure who are in sinus rhythm.

An RCT is still needed to compare anticoagulation versus no anticoagulation in people with heart failure.

Substantive changes

No new evidence

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Antiplatelet agents

Summary

We don't know whether antiplatelets are effective at reducing mortality or hospital re-admission rates.

Benefits and harms

Antiplatelet agents versus no treatment:

We found two systematic reviews (search dates 2005), both of which identified the same three-arm RCT.

Mortality

Aspirin compared with no treatment We don't know whether aspirin is more effective at 27 months than no antithrombotic treatment at reducing the combined outcome of death, MI, and stroke (very low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Death, MI, and stroke

RCT
3-armed trial
279 people, 70% with New York Heart Association (NYHA) functional class III
In review
Rate of combined outcome of death, MI, and stroke mean follow-up of 27 months
29/91 (32%) with aspirin (300 mg/day)
26/99 (27%) with no antithrombotic treatment

Reported as not significant (aspirin v no antithrombotic treatment)
P value not reported
Not significant

Admission to hospital

Aspirin compared with no treatment Aspirin may be less effective at reducing all-cause hospital re-admission rates at 27 months (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause hospital admissions

RCT
3-armed trial
279 people, 70% with New York Heart Association (NYHA) functional class III
In review
Rate of all-cause hospital admissions mean follow-up of 27 months
with aspirin (300 mg/day)
with no antithrombotic treatment
Absolute results not reported

P <0.05 (aspirin v no antithrombotic treatment)
Effect size not calculatedno antithrombotic treatment

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Haemorrhagic events

RCT
3-armed trial
279 people, 70% with New York Heart Association (NYHA) functional class III
In review
Rate of haemorrhagic events mean follow-up of 27 months
with aspirin (300 mg/day)
with no antithrombotic treatment
Absolute results not reported

Significance not assessed
Serious adverse effects (unspecified)

Systematic review
3-armed trial
279 people, 70% with NYHA functional class III
Data from 1 RCT
Total number of serious adverse effects
198 with aspirin
163 with warfarin
178 with no antithrombotic treatment

P = 0.08 (among group difference)

No data from the following reference on this outcome.

Antiplatelet agents versus warfarin:

We found two RCTs comparing aspirin versus warfarin, and one RCT comparing aspirin versus clopidogrel versus warfarin.

Mortality

Antiplatelet agents compared with warfarin We don't know whether antiplatelet agents (aspirin and clopidogrel) are more effective than warfarin at reducing mortality at 21 months or composite outcomes that include mortality (other outcomes in composite include MI, stroke, exacerbation of heart failure, and peripheral or pulmonary embolism) (very low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

RCT
3-armed trial
1587 people with New York Heart Association (NYHA) class II to IV Rate of mortality median follow-up of 21 months
94/523 (19%) with aspirin
92/540 (17%) with warfarin

HR 0.98 (warfarin v aspirin)
95% CI 0.85 to 1.13
P = 0.75
The RCT was terminated early because of slow enrolment (1587 people randomised rather than the planned 4500 people) and may have been underpowered to detect a clinically important difference
Results should be interpreted with caution
Not significant

RCT
3-armed trial
1587 people with NYHA class II to IV Rate of mortality median follow-up of 21 months
96/524 (18%) with clopidogrel
92/540 (17%) with warfarin

HR 0.92 (warfarin v clopidogrel)
95% CI 0.69 to 1.23
P = 0.58
The RCT was terminated early because of slow enrolment (1587 people randomised rather than the planned 4500 people) and may have been underpowered to detect a clinically important difference
Results should be interpreted with caution
Not significant
Death, MI, and stroke

RCT
3-armed trial
279 people, 70% with NYHA functional class III Rate of combined outcome of death, MI, and stroke mean follow-up of 27 months
29/91 (32%) with aspirin (300 mg/day)
23/89 (26%) with warfarin

Reported as not significant (aspirin v warfarin)
P value not reported
Not significant
Composite outcome including all-cause mortality

RCT
3-armed trial
1587 people with NYHA class II to IV Rate of composite outcome of death, non-fatal MI, or non-fatal stroke
108/523 (21%) with aspirin
106/540 (20%) with warfarin

HR (warfarin v aspirin) 0.98
95% CI 0.86 to 1.12
P = 0.77
Not significant

RCT
3-armed trial
1587 people with NYHA class II to IV Rate of composite outcome of death, non-fatal MI, or non-fatal stroke
113/524 (22%) with clopidogrel
106/540 (20%) with warfarin

HR (warfarin v clopidogrel) 0.89
95% CI 0.68 to 1.16
P = 0.39
Not significant

RCT
197 people aged 20 to 75 years with NYHA class II to IV caused by either previous MI or idiopathic dilated cardiomyopathy Time to the primary combined outcome of non-fatal stroke, peripheral or pulmonary embolism, MI, hospital admission, exacerbation of heart failure, or death from any cause
14.9/100 patient-years with aspirin
15.7/100 patient-years with warfarin

Significance not assessed

Admission to hospital

Antiplatelet agents compared with warfarin Aspirin may be less effective at reducing all-cause hospital re-admission rates at 21 to 27 months, but we don't know whether clopidogrel is more effective than warfarin at reducing all-cause hospital re-admission rates at 21 months (very low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause hospital admissions

RCT
3-armed trial
279 people, 70% with New York Heart Association (NYHA) functional class III Rate of all-cause hospital admissions mean follow-up of 27 months
with aspirin (300 mg/day)
with warfarin
Absolute results not reported

P <0.05 (aspirin v warfarin)
Effect size not calculatedwarfarin
Admission to hospital for heart failure-specific causes

RCT
3-armed trial
1587 people with NYHA class II to IV Proportion of people admitted to hospital with worsening heart failure median follow-up of 21 months
116/523 (22%) with aspirin
89/540 (16%) with warfarin

P <0.02 (aspirin v warfarin)
The RCT was terminated early because of slow enrolment (1587 people randomised rather than the planned 4500 people) and may have been underpowered to detect a clinically important difference
Results should be interpreted with caution
Effect size not calculatedwarfarin

RCT
3-armed trial
1587 people with NYHA class II to IV Proportion of people admitted to hospital with worsening heart failure median follow-up of 21 months
97/524 (19%) with clopidogrel
89/540 (16%) with warfarin

P = 0.38 (clopidogrel v warfarin)
The RCT was terminated early because of slow enrolment (1587 people randomised rather than the planned 4500 people) and may have been underpowered to detect a clinically important difference
Results should be interpreted with caution
Not significant

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Haemorrhagic events

RCT
3-armed trial
279 people, 70% with New York Heart Association (NYHA) functional class III
In review
Rate of haemorrhagic events mean follow-up of 27 months
with aspirin (300 mg/day)
with warfarin
Absolute results not reported

Significance not assessed

RCT
197 people age 20 to 75 years with NYHA class II to IV caused by either previous MI or idiopathic dilated cardiomyopathy Haemorrhagic event rate
0/100 patient-years with aspirin
4.6/100 patient-years with warfarin

Significance not assessed

RCT
3-armed trial
1587 people with NYHA class II to IV Proportion of people with major haemorrhage median follow-up of 21 months
19/523 (4%) with aspirin
28/540 (5%) with warfarin

P = 0.2184 (aspirin v warfarin)
The RCT was terminated early because of slow enrolment (1587 people randomised rather than the planned 4500 people) and may have been underpowered to detect a clinically important difference
Results should be interpreted with caution
Not significant

RCT
3-armed trial
1587 people with NYHA class II to IV Proportion of people with minor bleed median follow-up of 21 months
123/523 (24%) with aspirin
155/540 (29%) with warfarin

P = 0.0544 (aspirin v warfarin)
The RCT was terminated early because of slow enrolment (1587 people randomised rather than the planned 4500 people) and may have been underpowered to detect a clinically important difference
Results should be interpreted with caution
Not significant

RCT
3-armed trial
1587 people with NYHA class II to IV Proportion of people with major haemorrhage median follow-up of 21 months
11/524 (2%) with clopidogrel
28/540 (5%) with warfarin

P <0.01 (clopidogrel v warfarin)
The RCT was terminated early because of slow enrolment (1587 people randomised rather than the planned 4500 people) and may have been underpowered to detect a clinically important difference
Results should be interpreted with caution
Effect size not calculatedclopidogrel

RCT
3-armed trial
1587 people with NYHA class II to IV Proportion of people with minor bleed median follow-up of 21 months
119/524 (23%) with clopidogrel
155/540 (29%) with warfarin

P = 0.0254 (clopidogrel v warfarin)
The RCT was terminated early because of slow enrolment (1587 people randomised rather than the planned 4500 people) and may have been underpowered to detect a clinically important difference
Results should be interpreted with caution
Effect size not calculatedclopidogrel

Further information on studies

None.

Comment

In people not taking ACE inhibitors:

We found no systematic review or RCTs. We found one retrospective cohort analysis within one RCT in 642 men with heart failure. The RCT compared hydralazine plus isosorbide dinitrate versus prazosin versus placebo in men receiving digoxin and diuretics. Aspirin or dipyridamole, or both, were used at the discretion of the investigators. The number of thromboembolic events was low in both groups (1 stroke, and 0 peripheral and 0 pulmonary emboli in 184 person-years of treatment with antiplatelet agents v 21 strokes, and 4 peripheral and 4 pulmonary emboli in 1068 person-years of treatment without antiplatelet agents; 0.5 events/100 person-years with antiplatelet agents v 2.0 events/100 person-years without antiplatelet agents; P = 0.07).

In people taking ACE inhibitors:

We found no RCTs. We found two large retrospective cohort studies. The first retrospective analysis assessed the effect of antiplatelet agents used at the discretion of individual investigators on the incidence of stroke, peripheral arterial embolism, and pulmonary embolism within one RCT. The RCT (804 men with chronic heart failure) compared enalapril versus hydralazine plus isosorbide dinitrate. It found that the incidence of all thromboembolic events was low without antiplatelet treatment and found no significant difference between groups (1.6 events/100 person-years with antiplatelet treatment v 2.1 events/100 person-years with no antiplatelet treatment; P = 0.48).

The second cohort analysis was from two large RCTs comparing enalapril versus placebo (2569 people with symptomatic and asymptomatic left ventricular dysfunction). It found that people treated with antiplatelet agents at baseline had a significantly lower risk of death (HR adjusted for baseline differences 0.82, 95% CI 0.73 to 0.92). Subgroup analysis suggested that antiplatelet agents might have an effect in people randomised to placebo (mortality HR [for antiplatelet treatment at baseline v no antiplatelet treatment at baseline] 0.68, 95% CI 0.58 to 0.80), but not in people randomised to enalapril (mortality HR [for antiplatelet treatment v no antiplatelet treatment] 1.00, 95% CI 0.85 to 1.17). Both retrospective studies have important limitations common to studies with a retrospective cohort design. One study did not report on the proportions of people taking aspirin and other antiplatelet agents. The other study noted that >95% of people took aspirin, but the dose and consistency of antiplatelet use was not recorded. One retrospective non-systematic review (4 RCTs, 96,712 people) provided additional evidence about the effect of aspirin on the benefits of early ACE inhibitors in heart failure. It found a similar reduction in 30-day mortality with ACE inhibitors versus control for those people not taking aspirin compared with those taking aspirin (aspirin: OR 0.94, 95% CI 0.89 to 0.99; no aspirin: OR 0.90, 95% CI 0.81 to 1.01). However, the analysis may not be valid because the people who did not receive aspirin were older and had a worse baseline prognosis than those taking aspirin. The effects of antiplatelet treatment in combination with ACE inhibitors in people with heart failure require further research.

Substantive changes

No new evidence

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Calcium channel blockers

Summary

Calcium channel blockers may increase mortality and should be used with caution, if at all, in people with systolic heart failure.

Calcium channel blockers have been found to exacerbate symptoms of heart failure and increase mortality after MI in people who also have pulmonary congestion or left ventricular dysfunction.

Benefits and harms

Calcium channel blockers (for heart failure other than MI) versus placebo:

We found one systematic review (search date not reported, 18 RCTs, 3128 people with moderate to advanced heart failure for >2 months) of second-generation dihydropyridine calcium channel blockers, one non-systematic review of all calcium channel blockers (3 RCTs, 1790 people with heart failure), and one subsequent RCT. For effects of calcium channel blockers after MI, see calcium channel blockers in review on myocardial infarction (ST-elevation).

Mortality

Compared with placebo Calcium channel blockers may be no more effective at reducing mortality (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
1603 people
2 RCTs in this analysis
Risk of mortality
with dihydropyridine calcium channel blockers
with placebo
Absolute results not reported

OR 0.94
95% CI 0.79 to 1.12
Not significant

RCT
421 people with primary cardiomyopathy
In review
Subgroup analysis
Rate of mortality 14 months
45/209 (22%) with amlodipine
74/212 (35%) with placebo

ARR 13%
95% CI 5% to 20%
RR 0.62
95% CI 0.43 to 0.85
Small effect sizeamlodipine

RCT
People with heart failure caused by coronary artery disease (number of people not clear)
In review
Subgroup analysis
Rate of mortality 14 months
with amlodipine
with placebo
Absolute results not reported

Reported as not significant
P value not reported
Not significant

RCT
2590 people with New York Heart Association (NYHA) functional class II to IV heart failure Rate of mortality
350/1295 (27%) with mibefradil
319/1295 (25%) with placebo

RR 1.10
95% CI 0.96 to 1.25
Not significant

Admission to hospital

Compared with placebo Calcium channel blockers seem no more effective at reducing the composite outcome of all-cause mortality and hospital admission for cardiovascular events (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Death or hospital admission

RCT
1153 people with New York Heart Association (NYHA) functional class III or IV, left ventricular ejection fraction (LVEF) <0.30, using diuretics, digoxin, and ACE inhibitors
In review
Proportion of people with combined end point of all-cause mortality and hospital admission for cardiovascular events 14 months
222/571 (39%) with amlodipine
246/582 (42%) with placebo

ARR +3.4%
95% CI –2.3% to +8.8%
RR 0.92
95% CI 0.79 to 1.06
Not significant

Systematic review
186 people, idiopathic dilated cardiomyopathy, NYHA functional class I to III
Data from 1 RCT
Transplant-fee listing survival
with diltiazem
with placebo
Absolute results not reported

Significance not assessed

Systematic review
451 people with mild heart failure, NYHA functional class II or III
Data from 1 RCT
Mortality
with felodipine
with placebo
Absolute results not reported

Significance not assessed

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Adverse effects

Systematic review
Adverse effects
with dihydropyridine calcium channel blockers
with placebo

Systematic review
Adverse effects
with diltiazem
with placebo

RCT
2590 people with New York Heart Association (NYHA) functional class II to IV heart failure Rate of mortality in people taking digoxin, class I or II antiarrhythmics, amiodarone, or drugs associated with torsade de pointes
with mibefradil
with placebo
Absolute results reported graphically

Reported as significant
P value not reported
Effect size not calculatedplacebo

Further information on studies

None.

Comment

Many of the RCTs were underpowered and had wide confidence intervals. One RCT of amlodipine in people with primary dilated cardiomyopathy is in progress.

Substantive changes

No new evidence

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Hydralazine

Summary

Hydralazine plus isosorbide dinitrate may improve survival and quality-of-life scores compared with placebo in people with chronic congestive heart failure.

We found no direct information from RCTs about the effects of hydralazine alone in the treatment of people with heart failure.

Benefits and harms

Hydralazine versus placebo:

We found no systematic review or RCTs.

Hydralazine plus isosorbide dinitrate versus placebo:

We found no systematic review but found two RCTs.

Mortality

Compared with placebo Hydralazine plus isosorbide dinitrate may be more effective at reducing cumulative mortality at 2 years and at reducing all-cause mortality at 6 months in people with heart failure also receiving standard treatment (very-low quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

RCT
3-armed trial
642 men aged 18 to 75 years with stable chronic congestive heart failure (all participants were already receiving diuretics and digoxin) Rate of cumulative mortality 2 years
78/186 (42%) with hydralazine plus isosorbide dinitrate
139/273 (51%) with placebo

Estimated cumulative reduction in mortality risk 34% (hydralazine plus isosorbide dinitrate v placebo)
95% CI 4% to 54%
P <0.03 (hydralazine plus isosorbide dinitrate v placebo)
Effect size not calculatedhydralazine plus isosorbide dinitrate

RCT
1050 African-Americans with New York Heart Association (NYHA) class III or IV heart failure with dilated ventricles Rate of mortality 6 months
6% with hydralazine plus isosorbide dinitrate
10% with placebo
Absolute numbers not reported

P = 0.02
The RCT was terminated early (after 10 months instead of 18 months) because mortality was higher with placebo
Effect size not calculatedhydralazine plus isosorbide dinitrate

Quality of life

Compared with placebo Hydralazine plus isosorbide dinitrate may be more effective at 6 months at improving quality of life in people with heart failure also receiving standard treatment (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Quality of life

RCT
1050 African-Americans with New York Heart Association (NYHA) class III or IV heart failure with dilated ventricles Quality-of-life score (measured on a scale where lower scores indicate better quality of life) 6 months
–5.6 with hydralazine plus isosorbide dinitrate
–2.7 with placebo

P = 0.02 (hydralazine plus isosorbide dinitrate v placebo)
The RCT was terminated early (after 10 months instead of 18 months) because mortality was higher with placebo
Effect size not calculatedhydralazine plus isosorbide dinitrate

No data from the following reference on this outcome.

Admission to hospital

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Headache

RCT
3-armed trial
642 men aged 18 to 75 years with stable chronic congestive heart failure (all participants were already receiving diuretics and digoxin) Proportion of people with headache
23/186 (12%) with hydralazine plus isosorbide dinitrate
1/273 (0.4%) with placebo

Significance not assessed

RCT
1050 African-Americans with New York Heart Association (NYHA) class III or IV heart failure with dilated ventricles Proportion of people with headache 6 months
48% with hydralazine plus isosorbide dinitrate
19% with placebo
Absolute numbers not reported

P <0.001
The RCT was terminated early (after 10 months instead of 18 months) because mortality was higher with placebo
Effect size not calculatedplacebo
Dizziness

RCT
3-armed trial
642 men aged 18 to 75 years with stable chronic congestive heart failure (all participants were already receiving diuretics and digoxin) Proportion of people with dizziness
12/186 (6%) with hydralazine plus isosorbide dinitrate
5/273 (2%) with placebo

Significance not assessed

RCT
1050 African-Americans with NYHA class III or IV heart failure with dilated ventricles Proportion of people with dizziness 6 months
29% with hydralazine plus isosorbide dinitrate
12% with placebo
Absolute numbers not reported

P <0.001
The RCT was terminated early (after 10 months instead of 18 months) because mortality was higher with placebo
Effect size not calculatedplacebo

Further information on studies

The primary end point was a composite score made up of weighted values for mortality from any cause, a first hospital admission for heart failure within 18 months, and change in quality of life at 6 months. The score could range from –6 to +2 with higher scores indicating improved outcomes. The RCT found that hydralazine plus isosorbide dinitrate significantly improved the composite end point compared with placebo (–0.1 with hydralazine-isosorbide dinitrate v –0.5 with placebo; P = 0.01).

Comment

Clinical guide:

One systematic review has highlighted the potential risk of developing hydralazine-induced systemic lupus erythematous (SLE). Although the risk is small because of lower doses used, people taking hydralazine should be monitored at each visit for signs and symptoms of SLE. A baseline antinuclear antibody (ANA) level should be determined before initiating hydralazine. However, it is not recommended to regularly check ANA levels. If any symptoms or signs of SLE develop, hydralazine treatment should be discontinued immediately because complications from the syndrome can be potentially fatal.

Hydralazine plus isosorbide dinitrate could be used in combination with other medications for heart failure and in people intolerant to ACE inhibitors or angiotensin receptor blockers. The combination of hydralazine and isosorbide dinitrate would not be considered first-line treatment for heart failure.

Substantive changes

No new evidence

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Implantable cardiac defibrillators

Summary

Implantable cardiac defibrillators can reduce mortality in people with heart failure who are at high risk of ventricular arrhythmias.

People with an implantable cardiac defibrillator are at risk of shocks from the device.

Benefits and harms

Implantable cardiac defibrillators versus usual care:

We found three systematic reviews. The first systematic review (search date 2002, 8 RCTs, 4909 people at risk for sudden cardiac death or ventricular arrhythmia) compared implantable cardiac defibrillator (ICD) treatment versus usual care in the primary prevention (people at risk for sudden cardiac death or ventricular arrhythmia who had evidence of heart failure or coronary artery disease) or secondary prevention population (people who had survived sudden cardiac death or had unstable ventricular rhythm) of life-threatening arrhythmias and sudden cardiac death. The second systematic review (search date 2004, 7 RCTs, 2110 people) compared ICD treatment versus usual care in people with heart failure caused by non-ischaemic cardiomyopathy, and analysed results separately for primary and secondary prevention RCTs. The third systematic review (search date 2008, 5 RCTs, all included in the previous reviews) performed a subgroup analysis of women from the trials comparing ICD versus usual care (934 women with heart failure and reduced left ejection fraction).

Mortality

Compared with usual care Implantable cardiac defibrillators seem more effective at reducing all-cause mortality and sudden cardiac death in people with heart failure when both men and women are analysed together. However, when women with heart failure are analysed alone, effects on mortality in this subgroup are unclear (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
4909 people
8 RCTs in this analysis
Rate of mortality
459/2428 (19%) with implantable cardiac defibrillator (ICD)
695/2481 (28%) with usual care

RR 0.74
95% CI 0.67 to 0.82
Small effect sizeICD

Systematic review
2946 people with evidence of heart failure or coronary artery disease (primary prevention)
5 RCTs in this analysis
Rate of mortality
260/1494 (17%) with ICD
391/1452 (27%) with usual care

RR 0.72
95% CI 0.63 to 0.84
The magnitude of absolute-mortality benefit increased with increasing baseline risk of sudden cardiac death
There was significant heterogeneity among RCTs because 3 RCTs were in people at high risk of heart failure and 2 RCTs were in people at moderate risk of heart failure
Small effect sizeICD

Systematic review
1963 people who had survived sudden cardiac death or had unstable ventricular rhythm (secondary prevention)
3 RCTs in this analysis
Rate of mortality
199/934 (21%) with ICD
304/1029 (30%) with usual care

RR 0.76
95% CI 0.65 to 0.89
Small effect sizeICD

Systematic review
2110 people with heart failure caused by non-ischaemic cardiomyopathy
7 RCTs in this analysis
Rate of mortality
with ICD
with usual care
Absolute results not reported

RR 0.69
95% CI 0.56 to 0.86
Small effect sizeICD

Systematic review
1457 people with heart failure caused by non-ischaemic cardiomyopathy (primary prevention)
4 RCTs in this analysis
Rate of mortality
with ICD
with usual care
Absolute results not reported

RR 0.74
95% CI 0.58 to 0.96
Small effect sizeICD

Systematic review
256 people with previous resuscitated cardiac arrest or symptomatic ventricular tachycardia (secondary prevention)
2 RCTs in this analysis
Rate of mortality
with ICD
with usual care
Absolute results not reported

RR 0.69
95% CI 0.39 to 1.24
The number analysed may have been too small to detect a significant difference
Not significant

Systematic review
934 women with heart failure and reduced left ventricular ejection fraction
5 RCTs in this analysis
Mortality
with ICD
with usual care
Absolute results reported graphically

HR 1.01
95% CI 0.76 to 1.33
P = 0.95
Not significant
Cardiac mortality

Systematic review
4909 people
8 RCTs in this analysis
Rate of cardiac mortality
124/2428 (5%) with ICD
339/2481 (14%) with usual care

RR 0.43
95% CI 0.35 to 0.53
Moderate effect sizeICD

Systematic review
2946 people with evidence of heart failure or coronary artery disease (primary prevention)
5 RCTs in this analysis
Rate of cardiac mortality
57/1494 (4%) with ICD
177/1452 (12%) with usual care

RR 0.37
95% CI 0.27 to 0.50
The magnitude of absolute-mortality benefit increased with increasing baseline risk of sudden cardiac death
There was significant heterogeneity among RCTs because 3 RCTs were in people at high risk of heart failure and 2 RCTs were in people at moderate risk of heart failure
Moderate effect sizeICD

Systematic review
1963 people who had survived sudden cardiac death or had unstable ventricular rhythm (secondary prevention)
3 RCTs in this analysis
Rate of cardiac mortality
67/934 (7%) with ICD
162/1029 (16%) with usual care

RR 0.50
95% CI 0.38 to 0.66
Moderate effect sizeICD

Admission to hospital

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Adverse effects

Systematic review
Adverse effects
with implantable cardiac defibrillator (ICD)
with usual care
Absolute results not reported

No data from the following reference on this outcome.

ICDs plus cardiac resynchronisation therapy versus usual care:

See option on cardiac resynchronisation therapy.

ICDs plus cardiac resynchronisation therapy versus either device alone:

See option on cardiac resynchronisation therapy.

Further information on studies

None.

Comment

Clinical guide:

The systematic reviews suggest that ICDs are more beneficial than drug treatment for secondary prevention of sudden cardiac death, and for primary prevention in certain high-risk groups. The third review would suggest that women may not derive the same benefit from ICD treatment as do men. The decreased overall rate of sudden cardiac death with an increased rate of other competing causes of death leads to a smaller net benefit from ICDs in women with advanced heart failure and reduced left ventricular ejection fraction. There have been <1000 women studied in RCTs of ICD treatment and based on the event rate for women in these studies, >4000 women would need to be randomised to ICD or placebo to more definitively assess the benefit of ICD treatment. ICD treatment is expensive and must be used appropriately in people in whom indications for treatment clearly exist. Further research is required to develop accurate risk-stratification tools, to determine the impact of ICD treatment in different subgroups of people, and to evaluate quality-of-life issues.

People with ICDs are at risk of shocks from the device and this can adversely affect quality of life. An RCT has demonstrated that a combination of amiodarone plus a beta-blocker may be better to reduce the risk of shock compared with either sotalol (HR 0.43, 95% CI 0.22 to 0.85; P = 0.02) or beta-blocker alone (HR 0.27, 95% CI 0.14 to 0.52; P <0.001). There was a trend for sotalol to reduce shocks compared with beta-blockers alone (HR 0.61, 95% CI 0.37 to 1.01; P = 0.05). There was a slightly greater incidence of adverse pulmonary and thyroid events, and of symptomatic bradycardia in people receiving amiodarone. As people with an ICD require some form of treatment to reduce the potential for shocks, therapeutic decisions should be individualised. The type of treatment used must take into consideration the possible improvements in quality of life, and small but increased risks of drug-related adverse effects.

Substantive changes

Implantable cardiac defibrillators New evidence added. Categorisation unchanged (Beneficial).

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Cardiac resynchronisation therapy

Summary

Cardiac resynchronisation therapy can reduce mortality in people with heart failure who are at high risk of ventricular arrhythmias.

However, studies evaluating cardiac resynchronisation therapy were performed in centres with considerable experience, which may have overestimated the benefits.

Benefits and harms

Cardiac resynchronisation therapy (CRT) alone versus usual care/control:

We found 4 systematic reviews (search date 2003, 9 RCTs, 3216 people, 85% with New York Heart Association [NYHA] functional class III or IV symptoms; search date 2005, 8 RCTs, 3380 people; search date 2006, 7 RCTs, 3889 people; and search date 2006, 7 RCTs, 3164 people). The reviews included different RCTs in their meta-analyses and so we report data from all 4 reviews. We also found one non-systematic review that combined the results from the three studies in the MIRACLE implantable cardioverter-defibrillator (ICD) programme (2078 people) to evaluate the safety of CRT implantation.

Mortality

Compared with usual care Cardiac resynchronisation therapy may be more effective at reducing all-cause mortality and death from progressive heart failure and reducing the proportion of people classed as "worsened" on the heart failure clinical composite response (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
3203 people
8 RCTs in this analysis
Rate of mortality
with cardiac resynchronisation therapy (CRT)
with usual care
Absolute results not reported

RR 0.79
95% CI 0.66 to 0.96
Some RCTs included in the analysis compared CRT plus implantable cardiac defibrillators (ICDs) versus ICDs alone
Small effect sizeCRT

Systematic review
3380 people
8 RCTs in this analysis
Rate of mortality
264/1847 (14%) with CRT
260/1533 (17%) with control (not further defined)

OR 0.72
95% CI 0.59 to 0.88
The review included RCTs in which ICDs were used in all people randomised
Small effect sizeCRT

Systematic review
2249 people
4 RCTs in this analysis
Rate of mortality
245/1283 (19%) with CRT
247/966 (26%) with usual care (medical treatment)

OR 0.67
95% CI 0.50 to 0.90
This analysis included those RCTs in which neither arm was treated with an ICD
Small effect sizeCRT

Systematic review
3164 people with heart failure
5 RCTs in this analysis
All-cause mortality
224/1664 (14%) with CRT
244/1364 (18%) with usual care

RR 0.70
95% CI 0.60 to 0.83
Small effect sizeCRT
Cardiac mortality

Systematic review
1647 people
7 RCTs in this analysis
Rate of death from progressive heart failure
with CRT
with usual care
Absolute results not reported

RR 0.60
95% CI 0.36 to 1.01
Some RCTs included in the analysis compared CRT plus ICDs versus ICDs alone
Not significant

Systematic review
1716 people with heart failure
2 RCTs in this analysis
Death from congestive heart failure
98/1004 (9%) with CRT
92/712 (13%) with usual care

RR 0.79
95% CI 0.60 to 1.03
Not significant

Systematic review
2085 people with heart failure
3 RCTs in this analysis
Sudden cardiac death
49/1191 (4%) with CRT
59/894 (7%) with usual care

RR 0.67
95% CI 0.46 to 0.96
P = 0.03
Small effect sizeCRT

Admission to hospital

Compared with usual care Cardiac resynchronisation therapy may be more effective at reducing hospital admissions for heart failure or major cardiovascular events (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Admission to hospital for heart failure-specific causes

Systematic review
1642 people
6 RCTs in this analysis
Rate of hospital admission for heart failure
with cardiac resynchronisation therapy (CRT)
with usual care
Absolute results not reported

RR 0.68
95% CI 0.41 to 1.12
Some RCTs included in the analysis compared CRT plus implantable cardiac defibrillators (ICDs) versus ICDs alone
Not significant

Systematic review
People with New York Heart Association (NYHA) class III or IV symptoms (number of people in analysis not reported)
6 RCTs in this analysis
Rate of hospital admission for heart failure
with CRT
with usual care
Absolute results not reported

RR 0.65
95% CI 0.48 to 0.88
Some RCTs included in the analysis compared CRT plus ICDs versus ICDs alone
Small effect sizeCRT

Systematic review
2455 people
7 RCTs in this analysis
Proportion of people admitted to hospital for worsening heart failure
174/1230 (14%) with CRT
282/1225 (23%) with control (not further defined)

OR 0.55
95% CI 0.44 to 0.68
The review included RCTs in which ICDs were used in all people randomised
Small effect sizeCRT

Systematic review
1892 people with heart failure
5 RCTs in this analysis
Rate of hospital admission for heart failure
181/950 (19%) with CRT
277/942 (29%) with usual care

RR 0.64
95% CI 0.50 to 0.80
Small effect sizeCRT

No data from the following reference on this outcome.

Functional improvement

Compared with usual care Cardiac resynchronisation therapy may be more effective at improving function (New York Heart Association functional classification) by at least one functional class (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Improvement in functional class

Systematic review
Number of people included in analysis not reported
4 RCTs in this analysis
Proportion of people whose function improved by at least 1 New York Heart Association (NYHA) functional class
58% with cardiac resynchronisation therapy (CRT)
37% with usual care
Absolute numbers not reported

RR 1.6
95% CI 1.3 to 1.9
Some RCTs included in the analysis compared CRT plus implantable cardiac defibrillators (ICDs) versus ICDs alone
Small effect sizeCRT

No data from the following reference on this outcome.

Quality of life

Compared with usual care Cardiac resynchronisation therapy seems more effective at improving quality-of-life scores as assessed by the Minnesota Living with Heart Failure Questionnaire (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Quality of life

Systematic review
2472 people
7 RCTs in this analysis
Quality-of-life score on the Minnesota Living with Heart Failure Questionnaire (MLHFQ)
with cardiac resynchronisation therapy (CRT)
with usual care
Absolute results not reported

Weighted mean reduction: 7.6 points
95% CI 3.8 points to 11.5 points
Some RCTs included in the analysis compared CRT plus implantable cardiac defibrillators (ICDs) versus ICDs alone
Effect size not calculatedCRT

Systematic review
3380 people
8 RCTs in this analysis
Improved quality of life (assessed by the MLHFQ)
with CRT
with control (not further defined)
Absolute results not reported

WMD –7.1
95% CI –11.4 to –2.9
The review included RCTs in which ICDs were used in all people randomised
Effect size not calculatedCRT

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Implant safety

Non-systematic review
3078 people
3 RCTs in this analysis
Implant safety and success
with cardiac resynchronisation therapy (CRT)
with control
Absolute results not reported

Systematic review
3203 people
8 RCTs in this analysis
Implant safety and success
with CRT
with usual care
Absolute results not reported

Some RCTs included in the analysis compared CRT plus ICDs versus ICDs alone

No data from the following reference on this outcome.

CRT plus implantable cardiac defibrillator (ICD) versus usual care:

We found one systematic review (search date 2006, 7 RCTs, 3889 people), which assessed the effects of CRT plus ICD.

Mortality

Cardiac resynchronisation therapy plus implantable cardiac defibrillator compared with usual care Cardiac resynchronisation therapy plus implantable cardiac defibrillator is more effective than medical therapy at reducing all-cause mortality (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
903 people
Data from 1 RCT
Rate of mortality
105/595 (18%) with cardiac resynchronisation therapy (CRT) plus implantable cardiac defibrillator (ICD)
77/308 (25%) with usual care (medical treatment)

OR 0.64
95% CI 0.46 to 0.90
Small effect sizeCRT plus ICD

Admission to hospital

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

No data from the following reference on this outcome.

CRT plus ICD versus ICD alone:

We found two systematic reviews (search date 2006, 7 RCTs, 3889 people; and search date 2009, 2 RCTs, 2430 people), which assessed the effects of CRT plus ICD versus ICD alone.

Mortality

Cardiac resynchronisation therapy plus implantable cardiac defibrillator compared with implantable cardiac defibrillator alone Cardiac resynchronisation therapy plus implantable cardiac defibrillator may be no more effective than implantable cardiac defibrillator alone at reducing all-cause mortality (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
1045 people
3 RCTs in this analysis
Rate of mortality
27/517 (5%) with cardiac resynchronisation therapy (CRT) plus implantable cardiac defibrillator (ICD)
33/528 (6%) with ICD alone

OR 0.81
95% CI 0.48 to 1.37
Not significant

Systematic review
2430 people with heart failure
2 RCTs in this analysis
All-cause mortality
with CRT plus ICD
with ICD alone
Absolute results not reported

OR 0.96
95% CI 0.67 to 1.37
Not significant

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Admission to hospital

Cardiac resynchronisation therapy plus implantable cardiac defibrillator compared with implantable cardiac defibrillator alone Cardiac resynchronisation therapy plus implantable cardiac defibrillator may be more effective than implantable cardiac defibrillator alone at reducing admissions to hospital (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Admission to hospital

Systematic review
2430 people with heart failure
2 RCTs in this analysis
Heart failure events including hospital admission
with cardiac resynchronisation therapy (CRT) plus implantable cardiac defibrillator (ICD)
with ICD alone
Absolute results not reported

OR 0.57
95% CI 0.46 to 0.70
Small effect sizeCRT plus ICD

No data from the following reference on this outcome.

Adverse effects

No data from the following reference on this outcome.

CRT plus ICD versus CRT alone:

We found one systematic review (search date 2006, 7 RCTs, 3889 people), which assessed the effects of CRT plus ICD versus CRT alone.

Mortality

Cardiac resynchronisation therapy plus implantable cardiac defibrillator compared with cardiac resynchronisation therapy alone Cardiac resynchronisation therapy plus implantable cardiac defibrillator is no more effective than cardiac resynchronisation therapy alone at reducing all-cause mortality (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
1212 people
Data from 1 RCT
Rate of mortality
105/595 (18%) with cardiac resynchronisation therapy (CRT) plus implantable cardiac defibrillator (ICD)
131/617 (21%) with CRT alone

OR 0.79
95% CI 0.60 to 1.06
Not significant

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Admission to hospital

No data from the following reference on this outcome.

Adverse effects

No data from the following reference on this outcome.

Further information on studies

The review reported that the success rate for implantation was between 97% and 99%. It examined the complications related to placement of CRT devices in the two RCTs identified, but did not directly compare complication rates with combined CRT plus ICD versus ICD alone. Peri-implantation mechanical complications, including pneumothorax, coronary dissection, and pericardial tamponade occurred with a frequency of 1% to 2% in people receiving CRT. Left ventricular lead problems following implantation were reported in 4% of participants by 30 days and 7% of participants at 12 months of follow-up. Device-related infections occurred in 1% of participants within 30 days of implantation.

Comment

Clinical guide:

The results presented in the systematic reviews indicate beneficial effects with CRT. People deriving benefit are those with the more severe symptoms of heart failure, although one systematic review suggests that cardiac resynchronisation therapy in people with milder symptoms of heart failure may have fewer heart failure events and reduction of cardiac remodelling. Most people included in the studies were well selected, and procedures were performed in centres with experience. However, because in almost all RCTs people were randomly assigned to different modes of operation after placement of the pacemaker, the results may over-estimate the potential benefits of CRT. Furthermore, meta-analysis of RCTs comparing combination of CRT plus ICD versus either device alone found that the combination does not seem more effective than either ICD or CRT alone in reducing mortality.

Substantive changes

Cardiac resynchronisation therapy New evidence added. Categorisation unchanged (Likely to be beneficial).

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Coronary revascularisation

Summary

Coronary revascularisation may reduce mortality in people with heart failure and left ventricular dysfunction.

However, very few people in the RCTs had clinical evidence of heart failure and the trials we found comparing coronary revascularisation with drug treatment were all conducted before ACE inhibitors, aspirin, beta-blockers, and statins were in routine use. Thus, the clinical relevance of the evidence to current clinical practice is unclear.

Benefits and harms

Coronary revascularisation versus drug treatment:

We found one systematic review (search date not reported), which performed a meta-analysis of individual patient data from 7 RCTs (2649 people with coronary artery disease, most with stable angina, 20% with left ventricular dysfunction [ejection fraction <49], 4% with clinical evidence of heart failure) comparing revascularisation (CABG surgery) versus drug treatment. We also found two systematic reviews of observational studies, which did not meet Clinical Evidence inclusion criteria (see comments).

Mortality

Compared with drug treatment We don't know how coronary revascularisation and drug treatment compare with each other, at reducing all-cause mortality (very low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Mortality

Systematic review
549 people with left ventricular ejection fraction (LVEF) <50%, coronary artery disease, and deemed eligible for either CABG or drug treatment
7 RCTs in this analysis
Subgroup analysis
Mortality 5 years
Not reported with CABG
25% with drug treatment
Absolute numbers not reported

OR 0.59
95% CI 0.39 to 0.91
P = 0.02
Small effect sizeCABG

Admission to hospital

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

No data from the following reference on this outcome.

Coronary revascularisation versus non-drug treatment:

We found no systematic review or RCTs.

Coronary revascularisation versus devices:

We found no systematic review or RCTs.

Further information on studies

None.

Comment

We found two systematic reviews (search date 1999, 24 observational studies, 3088 people with coronary artery disease and left ventricular dysfunction, mean New York Heart Association (NYHA) functional class 2.8; and search date not reported, 9 observational studies, all identified by the first review, 1244 people), both of which evaluated the effects of myocardial viability on mortality. All the studies identified by the reviews were completed before ACE inhibitors, aspirin, beta-blockers, and statins were in routine use. Neither review assessed outcomes other than mortality.

The first review found that, in people with myocardial viability, annual mortality was 16% in those who received drug treatment compared with 3% in those treated with revascularisation (P <0.0001). Retrospective, within-group meta-analysis found that annual mortality was 3% in people with myocardial viability who were revascularised compared with 8% in those without viability. For people with myocardial viability who received drug treatment, annual mortality was 16%, compared with 6% in those without viability. These findings suggest that the presence of viability is important when considering revascularisation of patients with coronary artery disease who have left ventricular dysfunction and heart failure.

The second review further supports this conclusion. This review examined the same group of studies as the first review, but only included studies that could contribute data from all 4 relevant parameters: presence or absence of viability, drug treatment, or revascularisation. In their meta-analysis the authors examined the interaction between myocardial viability and treatment allocation. The combined estimated interaction ratio for all 9 observational studies suggested that people who had left ventricular dysfunction, heart failure, and viable myocardium had a better result from revascularisation therapy than from drug treatment.

Clinical guide:

Although these reviews suggest that revascularisation of people with coronary artery disease, left ventricular dysfunction, and heart failure (especially in those with demonstrated myocardial viability) is better than drug treatment, there are several important limitations. The RCTs and observational studies examining this question were all conducted before treatments such as ACE inhibitors, aspirin, beta-blockers, and statins were routinely used. We found no RCTs solely in people with heart failure; all the evidence comes from subgroup analyses that represent a relatively small number of people. The other studies included in the meta-analyses were observational. Although meta-analysis is useful to increase the statistical power of small studies by pooling the data, there are inherent flaws of this technique that can be amplified by deficiencies within the primary resources. Thus, it cannot be definitively concluded that revascularisation is of benefit in our population of interest. There is currently an ongoing RCT that is designed to more definitively answer this question, the Surgical Treatment for Ischemic Heart Failure (STICH) trial.

Substantive changes

Coronary revascularisation New option added. Categorised as Unknown effectiveness because evidence is insufficient to judge this intervention in heart failure.

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

ACE inhibitors in people at high risk of heart failure

Summary

ACE inhibitors delay the onset of symptomatic heart failure, reduce cardiovascular events, and improve long-term survival in people with asymptomatic left ventricular systolic dysfunction compared with placebo.

Benefits and harms

ACE inhibitors versus placebo:

We found two systematic reviews, and three additional RCTs, one of which reported the 12-year follow-up of one of the RCTs identified by the first review. The first review (search date not reported) identified three RCTs of people with vascular disease, but no heart failure or left ventricular systolic dysfunction (LVSD) (29,805 people), and 5 RCTs of people with LVSD or heart failure (12,763 people). The second review (search date 2009) identified 6 RCTs, three identified by the previous review, of people with vascular disease, but no heart failure (32,210 people). The reviews performed different analyses so we report both here.

Mortality

Compared with placebo ACE inhibitors seem more effective at reducing all-cause mortality and cardiovascular mortality in people with asymptomatic left ventricular systolic dysfunction, and in people with vascular disease without known evidence of left ventricular dysfunction or heart failure, and at reducing fatal MI in people with left ventricular systolic dysfunction (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
29,805 people with vascular disease
3 RCTs in this analysis
Rate of mortality
8% with ACE inhibitors
9% with control
Absolute numbers not reported

OR 0.86
95% CI 0.79 to 0.94
P = 0.0004
Small effect sizeACE inhibitors

Systematic review
12,763 people with left ventricular systolic dysfunction (LVSD) or heart failure
5 RCTs in this analysis
Rate of mortality
23% with ACE inhibitors
27% with control
Absolute numbers not reported

OR 0.80
95% CI 0.74 to 0.87
P <0.0001
Small effect sizeACE inhibitors

Systematic review
42,568 people
8 RCTs in this analysis
Rate of mortality
12% with ACE inhibitors
14% with placebo
Absolute numbers not reported

OR 0.83
95% CI 0.79 to 0.88
P <0.0001
Small effect sizeACE inhibitors

Systematic review
32,210 people with vascular disease
6 RCTs in this analysis
All-cause mortality
1188/16,123 (7%) with ACE inhibitors
1365/16,087 (8%) with placebo

RR 0.87
95% CI 0.81 to 0.94
Small effect sizeACE inhibitors

Systematic review
31,750 people with vascular disease
5 RCTs in this analysis
Cardiovascular mortality
656/15,894 (4%) with ACE inhibitors
798/15,856 (5%) with placebo

RR 0.83
95% CI 0.70 to 0.98
Small effect sizeACE inhibitors

RCT
5165 people followed up
Further report of reference
Rate of mortality 12 years
1074/2111 (51%) with enalapril given for 3 to 4 years
1195/2117 (56%) with placebo

HR 0.86
95% CI 0.79 to 0.93
Small effect sizeACE inhibitors

RCT
1749 people with MI and LVSD, ejection fraction 35% or less Rate of mortality 12 years
with trandolapril given 3 to 7 days after MI
with placebo
Absolute results not reported

RR 0.89
95% CI 0.80 to 0.99
P = 0.03
Small effect sizeACE inhibitors
Cardiac mortality

RCT
2231 asymptomatic people after MI with documented LVSD Proportion of people with fatal MI
56/1115 (5%) with captopril
80/1116 (7%) with placebo

RR 0.68
95% CI 0.49 to 0.96
Small effect sizeACE inhibitors

RCT
5165 people followed up
Further report of reference
Rate of cardiac mortality 12 years
736/2111 (35%) with enalapril given for 3 to 4 years
826/2117 (39%) with placebo

HR 0.85
95% CI 0.77 to 0.94
Small effect sizeACE inhibitors

Admission to hospital

Compared with placebo ACE inhibitors seem more effective at reducing all-cause hospital admissions, cardiovascular hospital admissions, and heart-failure hospital admissions in people with heart failure, asymptomatic left ventricular dysfunction, or other risk factors for heart failure (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Admission to hospital for any cause

RCT
1749 people with MI and left ventricular systolic dysfunction (LVSD), ejection fraction 35% or less Rate of hospital admission for any cause 12 years
with trandolapril given 3 to 7 days after MI
with placebo
Absolute results not reported

RR 0.92
95% CI 0.88 to 0.96
P <0.001
Small effect sizeACE inhibitors
Admission to hospital for heart failure-specific causes

Systematic review
29,805 people with vascular disease
3 RCTs in this analysis
Rate of hospital admission for heart failure
2% with ACE inhibitors
3% with control
Absolute numbers not reported

OR 0.77
95% CI 0.67 to 0.90
P = 0.0007
Small effect sizeACE inhibitors

Systematic review
12,763 people with LVSD or heart failure
5 RCTs in this analysis
Rate of hospital admission for heart failure
14% with ACE inhibitors
19% with control
Absolute numbers not reported

OR 0.66
95% CI 0.60 to 0.74
P <0.0001
Small effect sizeACE inhibitors

Systematic review
42,568 people
8 RCTs in this analysis
Proportion of people admitted to hospital for heart failure
5% with ACE inhibitors
7% with control
Absolute numbers not reported

OR 0.70
95% CI 0.64 to 0.76
P <0.0001
Small effect sizeACE inhibitors

RCT
1749 people with MI and LVSD, ejection fraction 35% or less Rate of cardiovascular hospital admissions 12 years
with trandolapril given 3 to 7 days after MI
with placebo
Absolute results not reported

RR 0.95
95% CI 0.91 to 1.00
P = 0.047
Small effect sizeACE inhibitors

RCT
1749 people with MI and LVSD, ejection fraction 35% or less Rate of hospital admission for heart failure 12 years
with trandolapril given 3 to 7 days after MI
with placebo
Absolute results not reported

RR 0.85
95% CI 0.77 to 0.93
P <0.001
Small effect sizeACE inhibitors

No data from the following reference on this outcome.

Cardiovascular events

Compared with placebo ACE inhibitors seem more effective at reducing non-fatal MIs in people at high risk of heart failure (people with asymptomatic left ventricular systolic dysfunction, and people with vascular disease without known evidence of left ventricular dysfunction or heart failure) (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Non-fatal MI

Systematic review
42,568 people
8 RCTs in this analysis
Proportion of people with non-fatal MI
6% with ACE inhibitors
7% with control
Absolute numbers not reported

OR 0.80
95% CI 0.74 to 0.87
P <0.0001
Small effect sizeACE inhibitors

Systematic review
32,210 people with vascular disease
6 RCTs in this analysis
Non-fatal MI
813/16,123 (5%) with ACE inhibitors
981/16,087 (6%) with placebo

RR 0.83
95% CI 0.73 to 0.94
Small effect sizeACE inhibitors

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Adverse effects

RCT
5165 people followed up
Further report of reference
Proportion of people reporting an adverse effect 40 months
76% with enalapril given for 3 to 4 years
72% with placebo
Absolute numbers not reported

Significance not assessed

RCT
5165 people followed up
Further report of reference
Proportion of people with dizziness or fainting 40 months
46% with enalapril given for 3 to 4 years
33% with placebo
Absolute numbers not reported

Significance not assessed

RCT
5165 people followed up
Further report of reference
Proportion of people with cough 40 months
34% with enalapril given for 3 to 4 years
27% with placebo
Absolute numbers not reported

Significance not assessed

RCT
5165 people followed up
Further report of reference
Proportion of people with angio-oedema 40 months
1% with enalapril given for 3 to 4 years
1% with placebo
Absolute numbers not reported

Significance not assessed

Systematic review
10,235 people with vascular disease
3 RCTs in this analysis
Proportion of people who withdrew because of adverse effects
732/5139 (14%) with ACE inhibitor
343/5096 (7%) with placebo

RR 2.30
95% CI 1.34 to 3.95
Small effect sizeplacebo

Systematic review
10,974 people with vascular disease
3 RCTs in this analysis
Proportion of people with hypotension
38/5490 (0.7%) with ACE inhibitors
26/5484 (0.5%) with placebo

RR 1.79
95% CI 0.68 to 4.71
Not significant

Systematic review
17,587 people with vascular disease
2 RCTs in this analysis
Proportion of people with syncope
203/8803 (2.3%) with ACE inhibitors
162/8784 (1.8%) with placebo

RR 1.24
95% CI 1.02 to 1.52
Small effect sizeplacebo

Systematic review
18,915 people with vascular disease
3 RCTs in this analysis
Proportion of people with cough
1726/9476 (18%) with ACE inhibitors
1183/9439 (12%) with placebo

RR 1.67
95% CI 1.22 to 2.29
Small effect sizeplacebo
Discontinuation of treatment

RCT
5165 people followed up
Further report of reference
Proportion of people permanently discontinuing treatment 40 months
8% with enalapril given for 3 to 4 years
5% with placebo
Absolute numbers not reported

Significance not assessed

No data from the following reference on this outcome.

ACE inhibitors versus angiotensin II receptor blockers:

See option on angiotensin II receptor blockers.

Angiotensin II receptor blockers plus ACE inhibitors versus ACE inhibitors alone:

See option on angiotensin II receptor blockers.

Further information on studies

None.

Comment

Asymptomatic LVSD is prognostically important, but we found no prospective studies that evaluated screening to detect its presence.

Substantive changes

ACE inhibitors in people at high risk of heart failure New evidence added. Categorisation unchanged (Beneficial).

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Angiotensin II receptor blockers in people at high risk of heart failure

Summary

Angiotensin II receptor blockers and ACE inhibitors seem equally effective at reducing all-cause mortality and cardiovascular mortality in people at high risk of heart failure.

The combination of angiotensin II receptor blockers and ACE inhibitors seems no more effective than ACE inhibitors alone and causes more adverse effects.

We don't know how angiotensin II receptor blockers as a class compare with placebo, as the evidence available assesses only telmisartan.

Benefits and harms

Angiotensin II receptor blockers versus placebo:

We found two RCTs comparing angiotensin II receptor blockers versus placebo in people with vascular disease at high risk of developing heart failure. In one of the RCTs the people were intolerant of ACE inhibitors. One of the papers included a prespecified pooling of the data from both trials for two composite outcomes.

Mortality

Compared with placebo We don't know how effective angiotensin II receptor blockers are at reducing mortality in high risk people (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Mortality

RCT
5926 people with vascular disease who were intolerant to ACE inhibitors Cardiovascular mortality
227/2954 (7.7%) with telmisartan
223/2972 (7.5%) with placebo

HR 1.03
95% CI 0.85 to 1.24
P = 0.78
Not significant

No data from the following reference on this outcome.

Admission to hospital

Compared with placebo We don't know how effective angiotensin II receptor blockers are at reducing admission to hospital (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Admission to hospital

RCT
5926 people with vascular disease who were intolerant to ACE inhibitors Hospital admission for heart failure
134/2954 (4.5%) with telmisartan
129/2972 (4.3%) with placebo

HR 1.05
95% CI 0.82 to 1.34
P = 0.69
Not significant

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Cardiovascular events

Compared with placebo We don't know how angiotensin II receptor blockers and placebo compare at reducing the composite outcome of cardiovascular mortality, MI, stroke, or hospital admission for heart failure in high risk people (very-low quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Cardiovascular events

RCT
5926 people with vascular disease who were intolerant to ACE inhibitors Cardiovascular mortality, MI, stroke, or hospital admission for heart failure
465/2954 (16%) with telmisartan
504/2972 (17%) with placebo

HR 0.92
95% CI 0.81 to 1.05
P = 0.22
Not significant

RCT
5926 people with vascular disease who were intolerent to ACE inhibitors Cardiovascular mortality, MI, or stroke
384/2954 (13%) with telmisartan
440/2972 (15%) with placebo

HR 0.86
95% CI 0.76 to 1.00
P = 0.045
Small effect sizetelmisartan

RCT
20,332 people with previous stroke and risk factors for vascular disease Cardiovascular mortality, MI, stroke, or worsening or new heart failure
1367/10,146 (13.5%) with telmisartan
1463/10,186 (14.4%) with placebo

HR 0.94
95% CI 0.87 to 1.01
Not significant

RCT
20,332 people with previous stroke and risk factors for vascular disease Cardiovascular mortality, MI, stroke, or worsening or new heart failure within 6 months of randomisation
474/10,146 (5%) with telmisartan
433/10,986 (4%) with placebo

HR 1.10
95% CI 0.97 to 1.26
Not significant

RCT
20,332 people with previous stroke and risk factors for vascular disease Cardiovascular mortality, MI, stroke, or worsening or new heart failure >6 months after randomisation
893/10,146 (9%) with telmisartan
1030/10,186 (10%) with placebo

HR 0.87
95% CI 0.80 to 0.95
Small effect sizetelmisartan

RCT
26,258 people with vascular disease
2 RCTs in this analysis
Cardiovascular mortality, MI, stroke, or hospital admission for heart failure
1832/13,100 (14%) with telmisartan
1967/13,158 (15%) with placebo

OR 0.93
95% CI 0.86 to 0.99
P = 0.03
Small effect sizetelmisartan

RCT
26,258 people with vascular disease
2 RCTs in this analysis
Cardiovascular mortality, MI, stroke, or hospital admission for heart failure within 6 months of randomisation
546/13,100 (4.2%) with telmisartan
492/13,158 (3.7%) with placebo

OR 1.12
95% CI 0.99 to 1.27
P = 0.075
Not significant

RCT
26,258 people with vascular disease Cardiovascular mortality, MI, stroke, or hospital admission for heart failure >6 months after randomisation
1286/12,484 (10%) with telmisartan
1475/12,575 (12%) with placebo

OR 0.86
95% CI 0.80 to 0.94
P <0.001
Small effect sizetelmisartan

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Adverse effects

RCT
5926 people with vascular disease who were intolerant to ACE inhibitors Withdrawal for any cause
1090/2954 (37%) with telmisartan
1143/2972 (39%) with placebo

P = 0.22
Not significant

RCT
5926 people with vascular disease who were intolerant to ACE inhibitors Hypotensive symptoms
29/2954 (1.0%) with telmisartan
16/2972 (0.5%) with placebo

P = 0.05
Not significant

RCT
5926 people with vascular disease who were intolerant to ACE inhibitors Doubling of serum creatinine
60/2954 (2.0%) with telmisartan
42/2972 (1.4%) with placebo

Significance not assessed

RCT
5926 people with vascular disease who were intolerant to ACE inhibitors Hyperkalaemia (potassium >5.5 mmol/L)
111/2954 (4%) with telmisartan
49/2972 (2%) with placebo

Significance not assessed

RCT
20,332 people with previous stroke and risk factors for vascular disease Withdrawal because of adverse effects
1450/10,146 (14%) with telmisartan
1127/10,186 (11%) with placebo

P <0.001
Effect size not calculatedplacebo

RCT
20,332 people with previous stroke and risk factors for vascular disease Withdrawal because of hypotension
393/10,146 (4%) with telmisartan
186/10,186 (2%) with placebo

P <0.001
Effect size not calculatedplacebo

Angiotensin II receptor blockers versus ACE inhibitors:

We found two RCTs comparing angiotensin II receptor blockers versus ACE inhibitors in people with vascular disease at high risk of developing heart failure. Both RCTs tested a non-inferiority hypothesis for angiotensin II receptor blockers compared with ACE inhibitors. The primary outcome for one of the RCTs was all-cause mortality. The primary outcome for the other RCT was a composite of cardiovascular death, MI, stroke, or hospital admission for heart failure.

Mortality

Angiotensin II receptor blockers compared with ACE inhibitors Angiotensin II receptor blockers and ACE inhibitors seem equally effective at reducing all-cause mortality and cardiovascular mortality (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Mortality

RCT
3-armed trial
9818 people with MI, clinical heart failure, left ventricular ejection fraction (LVEF) <0.40, around 15% with diagnosed heart failure All-cause mortality
979/4909 (19.9%) with valsartan
958/4909 (19.5%) with captopril

HR 1.00
97.5% CI 0.90 to 1.11
Non-inferiority satisfied (see further information on studies)
P = 0.98
Not significant

RCT
3-armed trial
17,118 people with vascular disease All-cause mortality
989/8542 (11.6%) with telmisartan
1014/8576 (11.8%) with ramipril

RR 0.98
95% CI 0.90 to 1.07
Non-inferiority satisfied (see further information on studies)
Not significant

RCT
3-armed trial
9818 people with MI, LVEF <0.40, or clinical heart failure Cardiovascular mortality
827/4909 (16.8%) with valsartan
830/4909 (16.9%) with captopril

HR 0.98
97.5% CI 0.87 to 1.09
Non-inferiority satisfied (see further information on studies)
P = 0.62
Not significant

RCT
3-armed trial
17,118 people with vascular disease Cardiovascular mortality
598/8542 (7.0%) with telmisartan
603/8576 (7.0%) with ramipril

RR 1.00
95% CI 0.89 to 1.12
Non-inferiority satisfied (see further information on studies)
Not significant

Admission to hospital

Angiotensin II receptor blockers compared with ACE inhibitors Angiotensin II receptor blockers seem less effective than ACE inhibitors at reducing admission to hospital (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Admission to hospital

RCT
3-armed trial
17,118 people with vascular disease Hospital admission for heart failure
394/8542 (5%) with telmisartan
354/8576 (4%) with ramipril

RR 1.12
95% CI 0.97 to 1.29
Non-inferiority not satisfied (see further information on studies)
Not significant

No data from the following reference on this outcome.

Cardiovascular events

Angiotensin II receptor blockers compared with ACE inhibitors Angiotensin II receptor blockers and ACE inhibitors seem equally effective at reducing cardiovascular events (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Cardiovascular events

RCT
3-armed trial
9818 people with MI, left ventricular ejection fraction (LVEF) <0.40, or clinical heart failure Cardiovascular mortality or heart failure
1326/4909 (27.0%) with valsartan
1335/4909 (27.2%) with captopril

HR 0.97
97.5% CI 0.90 to 1.05
Non-inferiority satisfied (see further information on studies)
P = 0.51
Not significant

RCT
3-armed trial
17,118 people with vascular disease Cardiovascular mortality, MI, stroke, or hospital admission for heart failure
1423/8542 (16.7%) with telmisartan
1412/8576 (16.5%) with ramipril

RR 1.01
95% CI 0.94 to 1.09
Non-inferiority satisfied (see further information on studies)
Not significant

RCT
3-armed trial
17,118 people with vascular disease Fatal and non-fatal MI
440/8542 (5.2%) with telmisartan
413/8576 (4.8%) with ramipril

RR 1.07
95% CI 0.94 to 1.22
Non-inferiority not satisfied (see further information on studies)
Not significant

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Adverse effects

RCT
3-armed trial
9818 people with MI, left ventricular ejection fraction (LVEF) <0.40, or clinical heart failure Withdrawal because of adverse effects
282/4885 (5.8%) with valsartan
375/4879 (7.7%) with captopril

P <0.05
Effect size not calculatedvalsartan

RCT
3-armed trial
9818 people with MI, LVEF <0.40, or clinical heart failure Withdrawal because of hypotension
70/4885 (1.4%) with valsartan
41/4879 (0.8%) with captopril

P <0.05
Effect size not calculatedvalsartan

RCT
3-armed trial
17,118 people with vascular disease Withdrawal because of hypotension
229/8542 (3%) with telmisartan
149/8576 (2%) with ramipril

P <0.001
Small effect sizeramipril

RCT
3-armed trial
9818 people with MI, LVEF <0.40, or clinical heart failure Withdrawal because of renal impairment
53/4885 (1.1%) with valsartan
40/4879 (0.8%) with captopril

Reported as not significant
P value not reported
Not significant

RCT
3-armed trial
17,118 people with vascular disease Withdrawal because of renal impairment
68/8542 (0.8%) with telmisartan
60/8576 (0.7%) with ramipril

P = 0.46
Not significant

RCT
3-armed trial
9818 people with MI, LVEF <0.40, or clinical heart failure Withdrawal because of hyperkalaemia
7/4885 (0.1%) with valsartan
4/4879 (0.1%) with captopril

Reported as not significant
P value not reported
Not significant

Angiotensin II receptor blockers plus ACE inhibitors versus ACE inhibitors alone:

We found two RCTs comparing angiotensin II receptor blockers plus ACE inhibitors versus ACE inhibitors alone in people with vascular disease at high risk of developing heart failure. Both RCTs tested a non-inferiority hypothesis for the combination of angiotensin II receptor blockers plus ACE inhibitors versus ACE inhibitors alone. The primary outcome for one of RCTs was all-cause mortality. The primary outcome for the other RCT was a composite of cardiovascular mortality, MI, stroke, or hospital admission for heart failure.

Mortality

Angiotensin II receptor blockers plus ACE inhibitors compared with ACE inhibitors alone Angiotensin II receptor blockers plus ACE inhibitors seem no more effective at reducing all-cause mortality and cardiovascular mortality (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Mortality

RCT
3-armed trial
9794 people with MI, left ventricular ejection fraction (LVEF) <0.40, or clinical heart failure All-cause mortality
941/4885 (19.3%) with valsartan plus captopril
958/4909 (19.5%) with captopril alone

HR 0.98
97.5% CI 0.89 to 1.09
Non-inferiority satisfied (see further information about studies)
P = 0.73
Not significant

RCT
3-armed trial
17,078 people with vascular disease All-cause mortality
1065/8502 (12.5%) with telmisartan plus ramipril
1014/8576 (11.8%) with ramipril alone

RR 1.07
95% CI 0.98 to 1.16
Not significant

RCT
3-armed trial
9794 people with MI, LVEF <0.40, or clinical heart failure Cardiovascular mortality
827/4885 (16.9%) with valsartan plus captopril
830/4909 (16.9%) with captopril alone

HR 1.00
97.5% CI 0.89 to 1.11
P = 0.95
Not significant

RCT
3-armed trial
17,078 people with vascular disease Cardiovascular mortality
620/8502 (7.3%) with telmisartan plus ramipril
603/8576 (7.0%) with ramipril alone

RR 1.04
95% CI 0.93 to 1.17
Not significant

Admission to hospital

Angiotensin II receptor blockers plus ACE inhibitors compared with ACE inhibitors alone We don't know how effective angiotensin II receptor blockers plus ACE inhibitors are compared with ACE inhibitors alone in reducing admission to hospital (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Admission to hospital

RCT
3-armed trial
17,118 people with vascular disease Hospital admission for heart failure
332/8502 (3.9%) with telmisartan plus ramipril
354/8576 (4.1%) with ramipril alone

RR 0.95
95% CI 0.82 to 1.10
Not significant

No data from the following reference on this outcome.

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Cardiovascular events

Angiotensin II receptor blockers plus ACE inhibitors compared with ACE inhibitors alone Angiotensin II receptor blockers plus ACE inhibitors seem no more effective at reducing cardiovascular events (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Cardiovascular events

RCT
3-armed trial
9794 people with MI, left ventricular ejection fraction (LVEF) <0.40, or clinical heart failure Cardiovascular mortality or heart failure
1331/4885 (27.2%) with valsartan plus captopril
1335/4909 (27.2%) with captopril alone

HR 1.00
97.5% CI 0.92 to 1.09
P = 0.94
Not significant

RCT
3-armed trial
17,078 people with vascular disease Cardiovascular mortality, MI, stroke, or hospital admission for heart failure
1386/8502 (16.3%) with telmisartan plus ramipril
1412/8576 (16.5%) with ramipril alone

RR 0.99
95% CI 0.92 to 1.07
Not significant

RCT
3-armed trial
17,078 people with vascular disease Fatal and non-fatal MI
438/8502 (5.2%) with telmisartan plus ramipril
413/8576 (4.8%) with ramipril alone

RR 1.08
95% CI 0.94 to 1.23
Not significant
Adverse events

RCT
3-armed trial
9794 people with MI, LVEF <0.40, or clinical heart failure Withdrawal because of adverse effects
438/4862 (9%) with valsartan plus captopril
375/4879 (8%) with captopril alone

P <0.05
Effect size not calculatedcaptopril alone

RCT
3-armed trial
9794 people with MI, LVEF <0.40, or clinical heart failure Withdrawal because of hypotension
90/4862 (2%) with valsartan plus captopril
41/4879 (1%) with captopril alone

P <0.05
Effect size not calculatedcaptopril alone

RCT
3-armed trial
17,118 people with vascular disease Withdrawal because of hypotension
406/8502 (5%) with telmisartan plus ramipril
149/8576 (2%) with ramipril alone

P <0.001
Moderate effect sizeramipril alone

RCT
3-armed trial
9794 people with MI, LVEF <0.40, or clinical heart failure Withdrawal because of hyperkalaemia
12/4862 (0.2%) with valsartan plus captopril
4/4879 (0.1%) with captopril alone
Not significant

RCT
3-armed trial
17,118 people with vascular disease Hyperkalaemia (potassium >5.6 mmol/L)
480/8502 (6%) with telmisartan plus ramipril
283/8576 (3%) with ramipril alone

P <0.001
Moderate effect sizeramipril alone

Further information on studies

Both trials calculated that for non-inferiority to be satisfied, the upper confidence limit of the hazard ratio must be <1.13.

Comment

Clinical guide:

An angiotensin II receptor blocker seems as effective as an ACE inhibitor for reducing cardiovascular events in people with vascular disease who are at high risk of developing heart failure. Interestingly, the data for angiotensin II receptor blocker compared with placebo are not as robust with regards to showing benefit in patients with vascular disease. There has been no mortality benefit demonstrated in people with vascular disease. However, there have been benefits demonstrated for the composite cardiovascular outcome of cardiovascular death, MI, or stroke. The less robust findings in studies comparing angiotensin II receptor blocker versus placebo may have been in part due to reduced power in the studies because of event rates that were less than was expected at the start of the study and follow-up may have been too short. In fact, pooling of the data shows significant benefit for the composite outcome of cardiovascular death, MI, stroke, and hospital admission for heart failure. Furthermore, there is an interaction with time as demonstrated by a significant reduction in the composite outcome of cardiovascular death, MI, stroke, and hospital admission for heart failure occurring >6 months after randomisation compared with within 6 months of randomisation. In these studies, the finding of the lower rate of treatment discontinuation for the angiotensin II receptor blocker group is important with regards to life-long adherence to treatment. The combination of angiotensin II receptor blocker plus ACE inhibitor was not found to be superior to ACE inhibitor alone and there was a greater rate of adverse events for the combination treatment. Therefore, combination treatment would not be advisable in people with vascular disease in the absence of chronic symptomatic heart failure.

Substantive changes

Angiotensin II receptor blockers in people at high risk of heart failure New option added. Categorised as Likely to be beneficial.

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

ACE inhibitors or angiotensin II receptor blockers for diastolic heart failure

Summary

ACE inhibitors or angiotensin II receptor blockers seem no more effective at reducing mortality or rate of hospital admissions for cardiovascular events in people with diastolic heart failure compared with placebo.

Benefits and harms

ACE inhibitors or angiotensin II receptor blockers versus placebo:

We found one systematic review (search date 2008) comparing renin-angiotensin inhibitors versus placebo, which identified three RCTs (8021 people) assessing the ACE inhibitor perindopril or the angiotensin II receptor blockers candesartan or irbesartan in people with diastolic heart failure. The review did not assess adverse effects so we report the results from the individual trials it identified.

Mortality

ACE inhibitors or angiotensin II receptor blockers compared with placebo ACE inhibitors or angiotensin II receptor blockers are no more effective at reducing all-cause mortality (high-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

Systematic review
8021 people with diastolic heart failure
3 RCTs in this analysis
All-cause mortality
745/4005 (19%) with ACE inhibitors or angiotensin II receptor blockers
726/3996 (18%) with placebo

OR 1.03
95% CI 0.92 to 1.15
P = 0.62
Not significant

Admission to hospital

ACE inhibitors or angiotensin II receptor blockers compared with placebo ACE inhibitors or angiotensin II receptor blockers seem no more effective at reducing hospital admissions for heart failure (moderate-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Admission to hospital for heart failure-specific causes

Systematic review
8021 people with diastolic heart failure
3 RCTs in this analysis
Hospital admissions for heart failure
with ACE inhibitors or angiotensin II receptor blockers
with placebo
Absolute results not reported

OR 0.90
95% CI 0.80 to 1.02
P = 0.09
Not significant

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Discontinuation of treatment because of adverse effects

RCT
3023 people with New York Heart Association (NYHA) functional class II to IV heart failure and left ventricular ejection fraction (LVEF) >40% Proportion of people permanently discontinuing treatment caused by an adverse effect (hypotension, hyperkalaemia, and increase in plasma creatinine) or an abnormal laboratory value median follow-up of 36.6 months
270/1514 (18%) with candesartan
204/1509 (14%) with placebo

P = 0.001
Effect size not calculatedplacebo

RCT
4128 people with NYHA II to IV heart failure symptoms and LVEF 45% or greater Proportion of people withdrawing because of an adverse effect mean follow-up of 49.5 months
331/2067 (16%) with irbesartan
288/2061 (14%) with placebo

P = 0.07
Not significant
Adverse effects

RCT
4128 people with NYHA II to IV heart failure symptoms and LVEF 45% or greater Proportion of people with hypotension mean follow-up of 49.5 months
60/2067 (2.9%) with irbesartan
62/2061 (3.0%) with placebo

P = 0.84
Not significant

RCT
4128 people with NYHA II to IV heart failure symptoms and LVEF 45% or greater Proportion of people with renal dysfunction mean follow-up of 49.5 months
69/2067 (3.3%) with irbesartan
57/2061 (2.8%) with placebo

P = 0.29
Not significant

RCT
4128 people with NYHA II to IV heart failure symptoms and LVEF 45% or greater Proportion of people with renal hyperkalaemia mean follow-up of 49.5 months
12/2067 (0.6%) with irbesartan
9/2061 (0.4%) with placebo

P = 0.34
Not significant

RCT
850 people with NYHA functional class I to IV heart failure and LVEF >40% Proportion of people with a serious adverse effect (including oedema and hypotension)
9/424 (2%) with perindopril 4 mg daily
4/426 (1%) with placebo

Significance not assessed

Further information on studies

Given the higher symptom severity in one of the trials (IPRESERVE) included in the review and the possibility that renin-angiotensin inhibition may not be as effective at reducing hospital admission for heart failure in more advanced disease, the meta-analysis was repeated excluding the IPRESERVE data. Excluding the results of IPRESERVE did not change the outcome for all-cause mortality (OR 1.04, 95% CI 0.87 to 1.24; P = 0.69) and resulted in only a non-significant trend towards reduced hospital admissions for heart failure (OR 0.85, 95% CI 0.72 to 1.00; P = 0.06).

Comment

Clinical guide:

The causes of diastolic dysfunction vary among people with diastolic heart failure. Current treatment is largely based on the results of small clinical studies and consists of treating the underlying cause and coexistent conditions with interventions optimised for individuals. The findings from this systematic review would not support the routine use of renin-angiotensin antagonists to reduce cardiovascular mortality or morbidity in this population.

Substantive changes

ACE inhibitors or angiotensin II receptor blockers for diastolic heart failure New evidence added. Categorisation changed from Unknown effectiveness to Unlikely to be beneficial.

BMJ Clin Evid. 2011; 2011: 0204.
Published online 2011 August 30.

Treatments other than angiotensin II receptor blockers for diastolic heart failure

Summary

We don't know whether treatments other than angiotensin II receptor blockers are beneficial in reducing mortality in people with diastolic heart failure as we found only one trial.

Benefits and harms

Treatments other than angiotensin II receptor blockers versus placebo:

We found no systematic review but found one RCT.

Mortality

Treatments other than angiotensin II receptor blockers compared with placebo We don't know whether digoxin is more effective at reducing all-cause or cardiovascular mortality at a mean follow-up of 37 months (low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
All-cause mortality

RCT
988 people with New York Heart Association (NYHA) functional class I to IV heart failure and left ventricular ejection fraction (LVEF) >45%
Subgroup analysis
Rate of mortality mean follow-up of 37 months
with digoxin
with placebo
Absolute results not reported

HR 0.99
95% CI 0.76 to 1.28
P = 0.92
Not significant
Cardiovascular mortality

RCT
988 people with NYHA functional class I to IV heart failure and LVEF >45%
Subgroup analysis
Rate of cardiovascular mortality mean follow-up of 37 months
with digoxin
with placebo
Absolute results not reported

HR 1.00
95% CI 0.73 to 1.36
P = 0.98
Not significant

Admission to hospital

Treatments other than angiotensin II receptor blockers compared with placebo We don't know whether digoxin is more effective at reducing the combined outcome of all-cause mortality and unplanned heart failure-related hospital admissions (very low-quality evidence).

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Death or hospital admission

RCT
988 people with New York Heart Association (NYHA) functional class I to IV heart failure and left ventricular ejection fraction (LVEF) >45%
Subgroup analysis
Rate of combined primary outcome of hospital admission for heart failure, or heart failure mortality mean follow-up of 37 months
with digoxin
with placebo
Absolute results not reported

HR 0.82
95% CI 0.63 to 1.07
P = 0.136
Not significant

Functional improvement

No data from the following reference on this outcome.

Quality of life

No data from the following reference on this outcome.

Adverse effects

Ref (type)PopulationOutcome, InterventionsResults and statistical analysisEffect sizeFavours
Adverse effects

RCT
988 people with New York Heart Association (NYHA) functional class I to IV heart failure and left ventricular ejection fraction (LVEF) >45%
Subgroup analysis
Rate of digoxin toxicity
10% with digoxin
4% with placebo
Absolute numbers not reported

P <0.001
Effect size not calculatedplacebo

RCT
988 people with NYHA functional class I to IV heart failure and LVEF >45%
Subgroup analysis
Proportion of people admitted to hospital with unstable angina
with digoxin
with placebo
Absolute results not reported

HR 1.37
95% CI 0.99 to 1.91
P = 0.06
Result was of borderline significance
Not significant

Further information on studies

None.

Comment

Clinical guide:

The causes of diastolic dysfunction vary among people with diastolic heart failure. Current treatment is largely based on the results of small clinical studies and consists of treating the underlying cause and coexistent conditions with interventions optimised for individuals. Further RCTs with clinically relevant outcome measures are needed to determine the benefits and harms of treatments in diastolic heart failure.

Substantive changes

No new evidence


Articles from BMJ Clinical Evidence are provided here courtesy of BMJ Publishing Group