The originally published version of this paper was incorrect. In the table on page 1816, the Class of recommendation and Level of evidence for ‘The patient is pacemaker dependent as a result of AV nodal ablation’ should have read ‘IIa’ and ‘B’ respectively.
Appendix: six tables (3,10,11,12,13,15) are available on the ESC Website only at www.escardio.org/guidelines-surveys/esc-guidelines/Pages/acute-chronic-heart-failure.aspx and labelled as ‘Web Tables’ throughout the document.
Abbreviations and acronyms
adult congenital heart disease
Atrial Fibrillation and Congestive Heart Failure
acute heart failure
Acute Infarction Ramipril Efficacy
angiotensin receptor blocker
absolute risk reduction
Assessment of Treatment with Lisinopril And Survival
MorBidity-mortality EvAlUaTion of the If inhibitor ivabradine in patients with coronary disease and left ventricULar dysfunction
Beta-Blocker Evaluation of Survival Trial
bi-ventricular assist device
B-type natriuretic peptide
beats per minute
bridge to candidacy
bridge to decision
bridge to recovery
bridge to transplantation
coronary artery bypass graft
coronary artery disease
Cardiac Resynchronization in Heart Failure Study
Cardiac failure, Hypertension, Age ≥75 (Doubled), Diabetes, Stroke (Doubled)-Vascular disease, Age 65–74 and Sex category (Female)
Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity
- CIBIS II
Cardiac Insufficiency Bisoprolol Study II
cardiac magnetic resonance
Carvedilol or Metoprolol European Trial
Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure
Cooperative North Scandinavian Enalapril Survival Study
chronic obstructive pulmonary disease
Carvedilol Prospective Randomized Cumulative Survival
Controlled Rosuvastatin Multinational Trial in Heart Failure
continuous positive airway pressure
cardiac resynchronization therapy
cardiac resynchronization therapy-defibrillator
cardiac resynchronization therapy-pacemaker
Defibrillators in Non-ischemic Cardiomyopathy Treatment Evaluation
Digitalis Investigation Group
extracorporeal membrane oxygenation
estimated glomerular filtration rate
- ELITE II
Second Evaluation of Losartan in the Elderly Trial
Eplerenone in Mild Patients Hospitalization and Survival Study in Heart Failure
glomerular filtration rate
Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto miocardico-heart failure
hydralazine and isosorbide dinitrate
Hypertension, Abnormal renal/liver function (1 point each), Stroke, Bleeding history or predisposition, Labile INR, Elderly (>65), Drugs/alcohol concomitantly (1 point each)
Heart failure Endpoint evaluation of Angiotensin II Antagonist Losartan
Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training
heart failure with ‘preserved’ ejection fraction
heart failure with reduced ejection fraction
Irbesartan in heart failure with preserved systolic function
intra-aortic balloon pump
left bundle branch block
left ventricular assist device
left ventricular ejection fraction
Multicenter Automatic Defibrillator Implantation Trial II
mechanical circulatory support
multi-detector computed tomography
Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure
mineralocorticoid receptor antagonist
mid-regional atrial (or A-type) natriuretic peptide
Multisite Stimulation in Cardiomyopathies
non-invasive positive pressure ventilation
number needed to treat
non-steroidal anti-inflammatory drug
New York Heart Association
Optimal Therapy in Myocardial infarction with the Angiotensin II Antagonist Losartan
Perindopril for Elderly People with Chronic Heart failure
positron emission tomography
polyunsaturated fatty acid
Resynchronization/Defibrillation for Ambulatory Heart Failure Trial
Randomised Aldactone Evaluation Study
randomized controlled trial
relative risk reduction
Survival and Ventricular Enlargement
Sudden Cardiac Death in Heart Failure Trial
Study of Effects of Nebivolol Intervention on Outcomes and Rehospitalization in Seniors With Heart Failure
Systolic Heart failure treatment with the If inhibitor ivabradine Trial
Studies of Left Ventricular Dysfunction
single-photon emission computed tomography
Surgical Treatment for Ischemic Heart Failure
tricuspid annular plane systolic excursion
tissue Doppler imaging
TRAndolapril Cardiac Evaluation
Valsartan Heart Failure Trial
Valsartan In Acute myocardial infarction
maximal oxygen consumption
Guidelines summarize and evaluate all available evidence at the time of the writing process, on a particular issue with the aim of assisting physicians in selecting the best management strategies for an individual patient, with a given condition, taking into account the impact on outcome, as well as the risk–benefit ratio of particular diagnostic or therapeutic means. Guidelines are no substitutes, but are complements, for textbooks and cover the European Society of Cardiology (ESC) Core Curriculum topics. Guidelines and recommendations should help physicians to make decisions in their daily practice. However, the final decisions concerning an individual patient must be made by the responsible physician(s).
A large number of Guidelines have been issued in recent years by the ESC as well as by other societies and organizations. Because of the impact on clinical practice, quality criteria for the development of guidelines have been established in order to make all decisions transparent to the user. The recommendations for formulating and issuing ESC Guidelines can be found on the ESC website (http://www.escardio.org/guidelines-surveys/esc-guidelines/about/Pages/rules-writing.aspx). ESC Guidelines represent the official position of the ESC on a given topic and are regularly updated.
Members of this Task Force were selected by the ESC to represent professionals involved with the medical care of patients with this pathology. Selected experts in the field undertook a comprehensive review of the published evidence for diagnosis, management, and/or prevention of a given condition according to ESC Committee for Practice Guidelines (CPG) policy. A critical evaluation of diagnostic and therapeutic procedures was performed including assessment of the risk–benefit ratio. Estimates of expected health outcomes for larger populations were included, where data exist. The level of evidence and the strength of recommendation of particular treatment options were weighed and graded according to pre-defined scales, as outlined in Tables A and B.
The experts of the writing and reviewing panels filled in declarations of interest forms of all relationships which might be perceived as real or potential sources of conflicts of interest. These forms were compiled into one file and can be found on the ESC website (http://www.escardio.org/guidelines). Any changes in declarations of interest that arise during the writing period must be notified to the ESC and updated. The Task Force received its entire financial support from the ESC without any involvement from the healthcare industry.
The ESC CPG supervises and coordinates the preparation of new Guidelines produced by Task Forces, expert groups, or consensus panels. The Committee is also responsible for the endorsement process of these Guidelines. The ESC Guidelines undergo extensive review by the CPG and external experts. After appropriate revisions, it is approved by all the experts involved in the Task Force. The finalized document is approved by the CPG for publication in the European Heart Journal.
The task of developing ESC Guidelines covers not only the integration of the most recent research, but also the creation of educational tools and implementation programmes for the recommendations. To implement the guidelines, condensed pocket guidelines versions, summary slides, booklets with essential messages, and an electronic version for digital applications (smartphones, etc.) are produced. These versions are abridged and, thus, if needed, one should always refer to the full text version which is freely available on the ESC website. The National Societies of the ESC are encouraged to endorse, translate, and implement the ESC Guidelines. Implementation programmes are needed because it has been shown that the outcome of disease may be favourably influenced by the thorough application of clinical recommendations.
Surveys and registries are needed to verify that real-life daily practice is in keeping with what is recommended in the guidelines, thus completing the loop between clinical research, writing of guidelines, and implementing them into clinical practice.
The guidelines do not, however, override the individual responsibility of health professionals to make appropriate decisions in the circumstances of the individual patients, in consultation with that patient, and, where appropriate and necessary, the patient's guardian or carer. It is also the health professional's responsibility to verify the rules and regulations applicable to drugs and devices at the time of prescription3031.
The aim of this document is to provide practical, evidence-based guidelines for the diagnosis and treatment of heart failure (HF). The principal changes from the 2008 guidelines1 relate to: There are also changes to the structure and format of the guidelines. Therapeutic recommendations now state the treatment effect supported by the class and level of recommendation in tabular format; in the case of chronic heart failure due to left ventricular (LV) systolic dysfunction, the recommendations focus on mortality and morbidity outcomes. Detailed summaries of the key evidence supporting generally recommended treatments have been provided. Practical guidance is provided for the use of the more important disease-modifying drugs and diuretics. When possible, other relevant guidelines, consensus statements, and position papers have been cited to avoid unduly lengthy text. All tables should be read in conjunction with their accompanying text and not read in isolation.
an expansion of the indication for mineralocorticoid (aldosterone) receptor antagonists (MRAs);
a new indication for the sinus node inhibitor ivabradine;
an expanded indication for cardiac resynchronization therapy (CRT);
new information on the role of coronary revascularization in HF;
recognition of the growing use of ventricular assist devices; and
the emergence of transcatheter valve interventions.
Definition and diagnosis
Definition of heart failure
Heart failure can be defined as an abnormality of cardiac structure or function leading to failure of the heart to deliver oxygen at a rate commensurate with the requirements of the metabolizing tissues, despite normal filling pressures (or only at the expense of increased filling pressures).1 For the purposes of these guidelines, HF is defined, clinically, as a syndrome in which patients have typical symptoms (e.g. breathlessness, ankle swelling, and fatigue) and signs (e.g. elevated jugular venous pressure, pulmonary crackles, and displaced apex beat) resulting from an abnormality of cardiac structure or function. The diagnosis of HF can be difficult (see Section 3.6). Many of the symptoms of HF are non-discriminating and, therefore, of limited diagnostic value.2–6 Many of the signs of HF result from sodium and water retention and resolve quickly with diuretic therapy, i.e. may be absent in patients receiving such treatment. Demonstration of an underlying cardiac cause is therefore central to the diagnosis of HF (see Section 3.6). This is usually myocardial disease causing systolic ventricular dysfunction. However, abnormalities of ventricular diastolic function or of the valves, pericardium, endocardium, heart rhythm, and conduction can also cause HF (and more than one abnormality can be present) (see Section 3.5). Identification of the underlying cardiac problem is also crucial for therapeutic reasons, as the precise pathology determines the specific treatment used (e.g. valve surgery for valvular disease, specific pharmacological therapy for LV systolic dysfunction, etc.).
Terminology related to left ventricular ejection fraction
The main terminology used to describe HF is historical and is based on measurement of LV ejection fraction (EF). Mathematically, EF is the stroke volume (which is the end-diastolic volume minus the end-systolic volume) divided by the end-diastolic volume. In patients with reduced contraction and emptying of the left ventricle (i.e. systolic dysfunction), stroke volume is maintained by an increase in end-diastolic volume (because the left ventricle dilates), i.e. the heart ejects a smaller fraction of a larger volume. The more severe the systolic dysfunction, the more the EF is reduced from normal and, generally, the greater the end-diastolic and end-systolic volumes.
The EF is considered important in HF, not only because of its prognostic importance (the lower the EF the poorer the survival) but also because most clinical trials selected patients based upon EF (usually measured using a radionuclide technique or echocardiography). The major trials in patients with HF and a reduced EF (HF-REF), or ‘systolic HF’, mainly enrolled patients with an EF ≤35%, and it is only in these patients that effective therapies have been demonstrated to date.
Other, more recent, trials enrolled patients with HF and an EF >40–45% and no other causal cardiac abnormality (such as valvular or pericardial disease). Some of these patients did not have an entirely normal EF (generally considered to be >50%) but also did not have a major reduction in systolic function either. Because of this, the term HF with ‘preserved’ EF (HF-PEF) was created to describe these patients. Patients with an EF in the range 35–50% therefore represent a ‘grey area’ and most probably have primarily mild systolic dysfunction. The diagnosis of HF-PEF is more difficult than the diagnosis of HF-REF because it is largely one of exclusion, i.e. potential non-cardiac causes of the patient's symptoms (such as anaemia or chronic lung disease) must first be discounted (Table 1).7,8 Usually these patients do not have a dilated heart and many have an increase in LV wall thickness and increased left atrial (LA) size. Most have evidence of diastolic dysfunction (see Section 4.1.2), which is generally accepted as the likely cause of HF in these patients (hence the term ‘diastolic HF’).7,8
HF = heart failure; HF-PEF = heart failure with ‘preserved’ ejection fraction; HF-REF = heart failure and a reduced ejection fraction; LA = left atrial; LV = left ventricular; LVEF = left ventricular ejection fraction.
aSigns may not be present in the early stages of HF (especially in HF-PEF) and in patients treated with diuretics (see Section 3.6).
It is important to note that EF values and normal ranges are dependent on the imaging technique employed, method of analysis, and operator. Other, more sensitive measures of systolic function may show abnormalities in patients with a preserved or even normal EF (see Section 4.1.1), hence the preference for stating preserved or reduced EF over preserved or reduced ‘systolic function’.9,10
Terminology related to the time-course of heart failure
The terms used to describe different types of HF can be confusing. As described above, in these guidelines the term HF is used to describe the symptomatic syndrome, graded according to the New York Heart Association (NYHA) functional classification (see Section 3.4 and Table 2), although a patient can be rendered asymptomatic by treatment. In these guidelines, a patient who has never exhibited the typical signs or symptoms of HF is described as having asymptomatic LV systolic dysfunction (or whatever the underlying cardiac abnormality is). Patients who have had HF for some time are often said to have ‘chronic HF’. A treated patient with symptoms and signs, which have remained generally unchanged for at least a month, is said to be ‘stable’. If chronic stable HF deteriorates, the patient may be described as ‘decompensated’ and this may happen suddenly, i.e. ‘acutely’, usually leading to hospital admission, an event of considerable prognostic importance. New (‘de novo’) HF may present acutely, for example as a consequence of acute myocardial infarction or in a subacute (gradual) fashion, for example in a patient who has had asymptomatic cardiac dysfunction, often for an indeterminate period, and may persist or resolve (patients may become ‘compensated’). Although symptoms and signs may resolve in the latter patients, their underlying cardiac dysfunction may not, and they remain at risk of recurrent ‘decompensation’. Occasionally, however, a patient may have HF due to a problem that resolves completely (e.g. acute viral myopericarditis). Some other patients, particularly those with ‘idiopathic’ dilated cardiomyopathy, may also show substantial or even complete recovery of LV systolic function with modern disease-modifying therapy [including an angiotensin-converting enzyme (ACE) inhibitor, beta-blocker, and mineralocorticoid receptor antagonist (MRA)]. ‘Congestive HF’ is a term that is sometimes still used, particularly in the USA, and may describe acute or chronic HF with evidence of congestion (i.e. sodium and water retention). Congestion, though not other symptoms of HF (e.g. fatigue), may resolve with diuretic treatment. Many or all of these terms may be accurately applied to the same patient at different times, depending upon their stage of illness.
Terminology related to the symptomatic severity of heart failure
The NYHA functional classification (Table 2) has been used to select patients in almost all randomized treatment trials in HF and, therefore, to describe which patients benefit from effective therapies. Patients in NYHA class I have no symptoms attributable to heart disease; those in NYHA classes II, III or IV are sometimes said to have mild, moderate or severe symptoms, respectively.
It is important to note, however, that symptom severity correlates poorly with ventricular function, and that although there is a clear relationship between severity of symptoms and survival, patients with mild symptoms may still have a relatively high absolute risk of hospitalization and death.11–13 Symptoms can also change rapidly; for example, a stable patient with mild symptoms can become suddenly breathless at rest with the onset of an arrhythmia, and an acutely unwell patient with pulmonary oedema and NYHA class IV symptoms may improve rapidly with the administration of a diuretic. Deterioration in symptoms indicates heightened risk of hospitalization and death, and is an indication to seek prompt medical attention and treatment. Obviously, improvement in symptoms (preferably to the point of the patient becoming asymptomatic) is one of the two major goals of treatment of HF (the other being to reduce morbidity, including hospital admissions, and mortality).
The Killip classification may be used to describe the severity of the patient's condition in the acute setting after myocardial infarction.14
Epidemiology, aetiology, pathophysiology, and natural history of heart failure
Approximately 1–2% of the adult population in developed countries has HF, with the prevalence rising to ≥10% among persons 70 years of age or older.15 There are many causes of HF, and these vary in different parts of the world (Web Table 3). At least half of patients with HF have a low EF (i.e. HF-REF). HF-REF is the best understood type of HF in terms of pathophysiology and treatment, and is the focus of these guidelines. Coronary artery disease (CAD) is the cause of approximately two-thirds of cases of systolic HF, although hypertension and diabetes are probable contributing factors in many cases. There are many other causes of systolic HF (Web Table 3), which include previous viral infection (recognized or unrecognized), alcohol abuse, chemotherapy (e.g. doxorubicin or trastuzumab), and ‘idiopathic’ dilated cardiomyopathy (although the cause is thought to be unknown, some of these cases may have a genetic basis).16
HF-PEF seems to have a different epidemiological and aetiological profile from HF-REF.17,18 Patients with HF-PEF are older and more often female and obese than those with HF-REF. They are less likely to have coronary heart disease and more likely to have hypertension and atrial fibrillation (AF). Patients with HF-PEF have a better prognosis than those with HF-REF (see below).19
In patients with LV systolic dysfunction, the maladaptive changes occurring in surviving myocytes and extracellular matrix after myocardial injury (e.g. myocardial infarction) lead to pathological ‘remodelling’ of the ventricle with dilatation and impaired contractility, one measure of which is a reduced EF.11,20 What characterizes untreated systolic dysfunction is progressive worsening of these changes over time, with increasing enlargement of the left ventricle and decline in EF, even though the patient may be symptomless initially. Two mechanisms are thought to account for this progression. The first is occurrence of further events leading to additional myocyte death (e.g. recurrent myocardial infarction). The other is the systemic responses induced by the decline in systolic function, particularly neurohumoral activation. Two key neurohumoral systems activated in HF are the renin–angiotensin–aldosterone system and sympathetic nervous system. In addition to causing further myocardial injury, these systemic responses have detrimental effects on the blood vessels, kidneys, muscles, bone marrow, lungs, and liver, and create a pathophysiological ‘vicious cycle’, accounting for many of the clinical features of the HF syndrome, including myocardial electrical instability. Interruption of these two key processes is the basis of much of the effective treatment of HF.11,20
Clinically, the aforementioned changes are associated with the development of symptoms and worsening of these over time, leading to diminished quality of life, declining functional capacity, episodes of frank decompensation leading to hospital admission (which is often recurrent and costly to health services), and premature death, usually due to pump failure or a ventricular arrhythmia. The limited cardiac reserve of such patients is also dependent on atrial contraction, synchronized contraction of the left ventricle, and a normal interaction between the right and left ventricles. Intercurrent events affecting any of these [e.g. the development of AF or conduction abnormalities, such as left bundle branch block (LBBB)] or imposing an additional haemodynamic load on the failing heart (e.g. anaemia) can lead to acute decompensation.
Before 1990, the modern era of treatment, 60–70% of patients died within 5 years of diagnosis, and admission to hospital with worsening symptoms was frequent and recurrent, leading to an epidemic of hospitalization for HF in many countries.21–23 Effective treatment has improved both of these outcomes, with a relative reduction in hospitalization in recent years of 30–50% and smaller but significant decreases in mortality.21–23
Diagnosis of heart failure
Symptoms and signs
The diagnosis of HF can be difficult, especially in the early stages. Although symptoms bring patients to medical attention, many of the symptoms of HF (Table 4) are non-specific and do not, therefore, help discriminate between HF and other problems. Symptoms that are more specific (i.e. orthopnoea and paroxysmal nocturnal dyspnoea) are less common, especially in patients with milder symptoms, and are, therefore, insensitive.2–6
Many of the signs of HF result from sodium and water retention, and are, therefore, also not specific. Peripheral oedema has other causes as well, and is particularly non-specific. Signs resulting from sodium and water retention (e.g. peripheral oedema) resolve quickly with diuretic therapy (i.e. may be absent in patients receiving such treatment, making it more difficult to assess patients already treated in this way). More specific signs, such as elevated jugular venous pressure and displacement of the apical impulse, are harder to detect and, therefore, less reproducible (i.e. agreement between different doctors examining the same patient may be poor).2–6
Symptoms and signs may be particularly difficult to identify and interpret in obese individuals, in the elderly, and in patients with chronic lung disease.24–26
The patient's medical history is also important. HF is unusual in an individual with no relevant medical history (e.g. a potential cause of cardiac damage), whereas certain features, particularly previous myocardial infarction, greatly increase the likelihood of HF in a patient with appropriate symptoms and signs.2–5 These points highlight the need to obtain objective evidence of a structural or functional cardiac abnormality that is thought to account for the patient's symptoms and signs, to secure the diagnosis of HF (see below).
Once the diagnosis of HF has been made, it is important to establish the cause, particularly specific correctable causes (Web Table 3). Symptoms and signs are important in monitoring a patient's response to treatment and stability over time. Persistence of symptoms despite treatment usually indicates the need for additional therapy, and worsening of symptoms is a serious development (placing the patient at risk of urgent hospital admission and death) and merits prompt medical attention.
General diagnostic tests in patients with suspected heart failure
In view of the difficulty in grading the evidence for diagnostic tests, all diagnostic recommendations have been given an arbitrary evidence level of C32.
BNP = B-type natriuretic peptide; CAD = coronary artery disease; CMR = cardiac magnetic resonance; COPD = chronic obstructive pulmonary disease; ECG = electrocardiogram; HF = heart failure; LV = left ventricular; LVEF = left ventricular ejection fraction; MR-proANP = mid-regional pro atrial natriuretic peptide; NT-proBNP = N-terminal pro B-type natriuretic peptide; PET = positron emission tomography; SPECT = single photon emission computed tomography; TIBC = total iron-binding capacity.
aClass of recommendation.
bLevel of evidence.
cThis list is not exhaustive and other investigations are discussed in the text. Additional investigations may be indicated in patients with suspected acute HF in the emergency department/hospital, including troponins and D-dimer measurement and right heart catheterization.
Essential initial investigations: echocardiogram, electrocardiogram, and laboratory tests
The echocardiogram and electrocardiogram (ECG) are the most useful tests in patients with suspected HF. The echocardiogram provides immediate information on chamber volumes, ventricular systolic and diastolic function, wall thickness, and valve function.7–10,27–34 This information is crucial in determining appropriate treatment (e.g. an ACE inhibitor and beta-blocker for systolic dysfunction or surgery for aortic stenosis). Echocardiography is discussed in detail later (see Section 4). The ECG shows the heart rhythm and electrical conduction, i.e. whether there is sinoatrial disease, atrioventricular (AV) block, or abnormal intraventricular conduction (see Table 5). These findings are also important for decisions about treatment (e.g. rate control and anticoagulation for AF, pacing for bradycardia, or CRT if the patient has LBBB) (see Section 9.2 on treatment). The ECG may also show evidence of LV hypertrophy or Q waves (indicating loss of viable myocardium), giving a possible clue to the aetiology of HF. HF is very unlikely (likelihood <2%) in patients presenting acutely and with a completely normal ECG.2,3,35–38 In patients with a non-acute presentation, a normal ECG has a somewhat lower negative predictive value (likelihood <10–14%).
AV = atrioventricular; CMR = cardiac magnetic resonance; CRT-P = cardiac resynchronization therapy pacemaker; CRT-D = cardiac resynchronization therapy defibrillator; ECG = electrocardiogram; HF = heart failure; ICD = implantable cardioverter-defibrillator; LBBB = left bundle branch block; LV = left ventricular. 99mTc-DPD = technetium-99m 3,3-diphosphono-1,2-propanodicarboxylic acid.
The information provided by these two tests will permit an initial working diagnosis and treatment plan in the majority of patients. Routine biochemical and haematological investigations are also important, partly to determine whether renin–angiotensin–aldosterone blockade can be initiated safely (renal function and potassium) and to exclude anaemia (which can mimic or aggravate HF) and because they provide other, useful information (see Section 3.6.6).
Other tests are generally only required if the diagnosis remains unclear (e.g. if echocardiographic images are suboptimal or if an unusual cardiac cause, or a non-cardiac cause, of the patient's condition is suspected) or if further evaluation of the underlying cause of the patient's cardiac problem is indicated (e.g. perfusion imaging or angiography in suspected CAD or endomyocardial biopsy in certain infiltrating diseases of the myocardium). Special tests are discussed in more detail in Sections 4 and 5.
Because the signs and symptoms of HF are so non-specific, many patients with suspected HF referred for echocardiography are not found to have an important cardiac abnormality. Where the availability of echocardiography is limited, an alternative approach to diagnosis is to measure the blood concentration of a natriuretic peptide, a family of hormones secreted in increased amounts when the heart is diseased or the load on any chamber is increased (e.g. by AF, pulmonary embolism, and some non-cardiovascular conditions, including renal failure).39–42 Natriuretic peptide levels also increase with age, but may be reduced in obese patients.26 A normal natriuretic peptide level in an untreated patient virtually excludes significant cardiac disease, making an echocardiogram unnecessary (investigation for a non-cardiac cause of the patient's problems is likely to be more productive in such patients).39,42 The use of natriuretic peptides as a ‘rule-out’ test in the diagnosis of HF is discussed in detail elsewhere.39–50 Multiple studies have examined the threshold concentration that excludes HF for the two most commonly used natriuretic peptides, B-type natriuretic peptide (BNP) and N-terminal pro B-type natriuretic peptide (NT-proBNP).43–50 The exclusion threshold differs for patients presenting with acute onset or worsening of symptoms (e.g. to a hospital emergency department) and those presenting with a more gradual onset of symptoms.
For patients presenting with acute onset or worsening of symptoms, the optimal exclusion cut-off point is 300 pg/mL for NT-proBNP and 100 pg/mL for BNP. In one other study, mid-regional atrial (or A-type) natriuretic peptide (MR-proANP), at a cut-off point of 120 pmol/L, was shown to be non-inferior to these thresholds for BNP and NT-proBNP in the acute setting.51
For patients presenting in a non-acute way, the optimum exclusion cut-off point is 125 pg/mL for NT-proBNP and 35 pg/mL for BNP. The sensitivity and specificity of BNP and NT-proBNP for the diagnosis of HF are lower in non-acute patients.43–50
A chest X-ray is of limited use in the diagnostic work-up of patients with suspected HF. It is probably most useful in identifying an alternative, pulmonary explanation for a patient's symptoms and signs. It may, however, show pulmonary venous congestion or oedema in a patient with HF. It is important to note that significant LV systolic dysfunction may be present without cardiomegaly on the chest X-ray.
Routine laboratory tests
In addition to standard biochemical [sodium, potassium, creatinine/estimated glomerular filtration rate (eGFR)] and haematological tests (haemoglobin, haematocrit, ferritin, leucocytes, and platelets), it is useful to measure thyroid-stimulating hormone (thyrotropin) as thyroid disease can mimic or aggravate HF (Table 6). Blood glucose is also worth measuring as undiagnosed diabetes is common in patients with HF. Liver enzymes may also be abnormal in HF (important if considering amiodarone or warfarin).
ACE = angiotensin-converting enzyme; ARB = angiotensin receptor blocker; AVP = arginine vasopressin; BNP = B-type natriuretic peptide; BUN = blood urea nitrogen; CRP = C-reactive protein; eGFR = estimated glomerular filtration rate; HF = heart failure; MRA = mineralocorticoid receptor antagonist; NSAID = non-steroidal anti-inflammatory drug.
As well as a pre-treatment check, biochemical monitoring is important after the initiation of renin–angiotensin system blockers, while the dose is being up-titrated (see Section 7.2) and during longer term follow-up, especially if an intercurrent illness leading to sodium and water loss occurs (e.g. diarrhoea and vomiting) or another drug that affects sodium and water homeostasis or renal function is started or the dose altered [e.g. non-steroidal anti-inflammatory drugs (NSAIDs) or diuretics]. Many routine laboratory tests provide valuable prognostic information (see Section 6).
Algorithm for the diagnosis of heart failure
An algorithm for the diagnosis of HF or LV dysfunction is shown in Figure 1.
In patients presenting to hospital as an emergency with suspected HF and acute onset of symptoms, early echocardiography is recommended (and immediate echocardiography in shocked or severely haemodynamically compromised patients). If a natriuretic peptide is measured, a high exclusion cut-off point should be used.39–50 In patients presenting non-emergently in primary care, or to a hospital outpatient clinic, with slow onset of symptoms (and signs) suggestive of HF, an ECG and natriuretic peptide measurement may be used as a means of identifying patients who most need echocardiography (an echocardiogram is indicated if the natriuretic peptide level is above the exclusion threshold/ECG is abnormal). In these patients, a lower exclusion natriuretic peptide cut-off point should be used to prevent a ‘false-negative’ diagnosis of HF.39–50 Patients with a high pre-test likelihood of HF, such as those with a history of myocardial infarction, may be referred directly for echocardiography.
The role of cardiac imaging in the evaluation of patients with suspected or confirmed heart failure
Imaging plays a central role in the diagnosis of HF and in guiding treatment. Of the several imaging modalities available, echocardiography is the method of choice in patients with suspected HF for reasons of accuracy, availability (including portability), safety, and cost.27–34 It may be complemented by other modalities, chosen according to their ability to answer specific clinical questions and taking account of contraindications to, and risks of, specific tests (see Table 7).9,10,52–60 All imaging examinations, regardless of type, should be performed only by individuals competent and experienced in the specific technique.32
Selection of a test in daily practice should consider availability, local expertise, advantages/disadvantages, and, in the case of several questions to address, which test could best answer several of them.
ARVC = arrhythmogenic right ventricular cardiomyopathy; CAD = coronary artery disease; Cath = cardiac catheterization; CMP = cardiomyopathy; CMR = cardiac magnetic resonance; EDV = end-diastolic volume; EF = ejection fraction; ESV = end-systolic volume; HCM = hypertrophic cardiomyopathy; LV = left ventricular; MDCT = multidetector computed tomography; PET = positron emission tomography; RV = right ventricular; SPECT = single photon emission computed tomography.
aStress (dobutamine) imaging.
bFractional flow reserve or ‘Doppler’ flow reserve measurements.
cIncluding measurements of aortic annulus for transcatheter aortic valve implantation.
eHaemodynamic evaluation (constriction).
fDescribes disease activity by contrast-enhanced CMR.
hGood quality irrespective of patient habitus.
iExcellent attenuation correction.
kForeign metallic bodies in specific locations (e.g. in the eye) and electronic devices (some pacemakers are MR-compatible); relative contraindication: claustrophobia.
Echocardiography is a term used here to refer to all cardiac ultrasound imaging techniques, including two-dimensional/three-dimensional echocardiography, pulsed and continuous wave Doppler, colour flow Doppler, and tissue Doppler imaging (TDI).8,27–34,61–64 Echocardiography provides information about cardiac anatomy (e.g. volumes, geometry, mass) and function (e.g. LV function and wall motion, valvular function, right ventricular function, pulmonary artery pressure, pericardium).
Assessment of left ventricular systolic dysfunction
LVEF is not an index of contractility as it depends on volumes, preload, afterload, heart rate, and valvular function, and is not the same as stroke volume. Stroke volume may be maintained by LV dilation in a patient with HF-REF, whereas it may be reduced in patients with HF-PEF and concentric LV hypertrophy. EF may also be preserved (and stroke volume reduced) in patients with significant mitral regurgitation. Thus EF must be interpreted in its clinical context.
The recommended echocardiographic method for measurement of EF is the apical biplane method of discs (the modified Simpson's rule).8,27–34,61 However, because this method relies on accurate tracing of the endocardial border, use of a contrast agent to better delineate the endocardial border is recommended when image quality is suboptimal (i.e. where <80% of the endocardial border is adequately visualized).61 The Teichholz and Quinones methods of calculating EF from linear dimensions may result in inaccuracies, particularly in patients with regional LV dysfunction; the same is true for another technique for assessing LV systolic function—fractional shortening. These and visual assessment of EF (‘eye-balling’) are not recommended.61 Three-dimensional echocardiography of adequate quality further improves the quantification of ventricular volumes and EF calculation.62 The LV wall motion score index may be an acceptable alternative to EF but is not widely used. Other indices of LV systolic function include AV plane systolic excursion, systolic tissue Doppler velocities, and measurements of deformation (strain and strain rate). Deformation imaging is more sensitive than EF in detecting minor changes in LV systolic function. However, issues of reproducibility and standardization currently limit the routine clinical use of deformation imaging. Stroke volume and cardiac output can also be calculated by measuring the velocity time integral at the LV outflow tract area.
The most common echocardiographic abnormalities seen in patients with HF and their clinical significance are presented in Table 8.
E/e' = ratio of the mitral inflow E wave to the tissue Doppler e' wave; HF= heart failure; LV = left ventricular; RV = right ventricular; TAPSE = tricuspid annular plane systolic excursion.
Assessment of left ventricular diastolic dysfunction
LV diastolic dysfunction is thought to be the underlying pathophysiological abnormality in patients with HF-PEF, and thus its identification is fundamental to the diagnosis of this type of HF (Table 9).7,8,27–34,63,64 The Doppler echocardiographic diastolic indices commonly measured in patients with HF are shown in Table 9. Of note, normal values for functional echocardiographic indices of LV diastolic dysfunction may also depend on age, heart rate, and body size.63,64 Importantly, no single echocardiographic parameter is sufficiently accurate and reproducible to be used in isolation to make a diagnosis of LV diastolic dysfunction. Therefore, a comprehensive echocardiographic examination incorporating all relevant two-dimensional and Doppler data is recommended.8,63,64 This should include the evaluation of both structural (LV hypertrophy, LA dilation) and functional abnormalities (Table 1). Tissue Doppler imaging-derived early diastolic myocardial velocities (e'), measured at the mitral annulus, allow the assessment of myocardial relaxation. A normal e' (>8 cm/s septal, >10 cm/s lateral, or >9 cm/s average, measured using real-time pulsed TDI) is very unusual in a patient with HF. The E/e' ratio correlates with LV filling pressure.63,64 (Table 9). Thus, echocardiographic evidence of LV diastolic dysfunction may consist of a reduced e' (e' average <9 cm/s) or an increased E/e' ratio (>15), or a combination of these parameters (Table 9). The presence of at least two abnormal measurements and/or AF increases the likelihood of the diagnosis.
A pulm–A mitral = time difference between pulmonary vein flow A-wave duration and mitral flow A-wave duration; E/A = ratio of early to late diastolic mitral inflow waves; e' = early diastolic velocity of mitral annulus; E/e' = ratio of the mitral inflow E wave to the tissue Doppler e' wave; HF = heart failure; LV = left ventricular.
bHighly variable and unsuitable for diagnosis on its own; largely depending on loading conditions; age-corrected normal values exist.63
Transoesophageal echocardiography (TOE) is not needed in routine diagnostic assessment unless the transthoracic ultrasound window is inadequate (e.g. because of obesity, chronic lung disease, ventilated patients) and an alternative modality [e.g. cardiac magnetic resonance (CMR) imaging] is not available or applicable.
TOE is, however, valuable in patients with complex valvular disease (especially mitral disease and prosthetic valves), suspected endocarditis, and in selected patients with congenital heart disease. TOE is also used to check for thrombus in the left atrial appendage of patients with AF.
Exercise or pharmacological stress echocardiography may be used to identify the presence and extent of inducible ischaemia and to determine whether non-contracting myocardium is viable (see Section 13).34 This technique may also be useful in evaluating patients with suspected severe aortic stenosis, reduced EF, and a low transvalvular gradient (see Section 13.3.1). Diastolic stress testing is an emerging procedure to identify HF-PEF in patients with HF symptoms during physical activity, normal EF, and inconclusive diastolic function parameters at rest.63
Cardiac magnetic resonance
CMR is a non-invasive technique that provides most of the anatomical and functional information available from echocardiography, including evaluation of ischaemia and viability, as well as additional assessments.52,57,65 CMR is regarded as the gold standard with respect to accuracy and reproducibility of volumes, mass, and wall motion. Because CMR yields good image quality in most patients, it is the best alternative imaging modality in patients with non-diagnostic echocardiographic studies.
CMR is particularly valuable in identifying inflammatory and infiltrative conditions, and in predicting prognosis in patients with these (Table 7).65 CMR is also useful in the work-up of patients with suspected cardiomyopathy, arrhythmias, suspected cardiac tumours (or cardiac involvement by tumour), or pericardial diseases, and is the imaging method of choice in patients with complex congenital heart disease.66
Limitations include lack of availability, inability to image patients with certain metallic implants (including many, but not all, cardiac devices), and cost. Also, the accuracy of functional analysis is limited in patients with atrial arrhythmias. Some patients cannot tolerate the procedure, often because of claustrophobia. Linear gadolinium chelates are contraindicated in individuals with a GFR <30 mL/min/m2 because they cause the rare condition known as nephrogenic systemic fibrosis (this may be less of a concern with newer macrocyclic gadolinium chelates).67,68
Single-photon emission computed tomography and radionuclide ventriculography
Single-photon emission computed tomography (SPECT) may be useful in assessing ischaemia and viability if CAD is suspected, and provides prognostic as well as diagnostic information (Table 7).54 Gated SPECT can also yield information on ventricular volumes and function, but exposes the patient to ionizing radiation.
Positron emission tomography imaging
Positron emission tomography (PET) [alone or with computed tomography (CT)] may be used to assess ischaemia and viability, but the flow tracers (N-13 ammonia or O-15 water) require an on-site cyclotron.58,60,69 Rubidium is an alternative tracer for ischaemia testing with PET, which can be produced locally at relatively low cost (Table 7). Lack of availability, radiation exposure, and cost are the main limitations.
Coronary angiography should be considered in patients with angina pectoris or a history of cardiac arrest if the patient is otherwise suitable for coronary revascularization. Angiography should also be considered in patients with evidence of reversible myocardial ischaemia on non-invasive testing, especially if the EF is reduced (because coronary artery bypass surgery may be beneficial) (Section 13). Non-invasive assessment of myocardial viability may also be carried out before angiography as some observational data show that coronary angiography may be of little, if any, benefit and may confer considerable risk, in the absence of significant viability. In cases where ischaemia information is lacking, fractional flow reserve gives information about the haemodynamic relevance of lesions.70
Coronary angiography may be required, urgently, in selected patients with acute HF (AHF) (shock or acute pulmonary oedema), particularly those with an associated acute coronary syndrome (see Section 12.7.1 and revascularization guidelines71). Coronary angiography may also be indicated in patients with valve disease when surgical correction is planned.
Cardiac computed tomography
The main use of CT in patients with HF is a non-invasive means to visualize the coronary anatomy.59 The risk vs. benefit of this procedure should be considered as discussed above, under coronary angiography (Section 4.7).
Cardiac catheterization and endomyocardial biopsy
In patients with suspected constrictive or restrictive cardiomyopathy, cardiac catheterization used in combination with other non-invasive imaging techniques may help to establish the correct diagnosis (see Table 7). In patients with suspected myocarditis and infiltrative diseases (e.g. amyloidosis, see Table 7), endomyocardial biopsy may be needed to confirm the diagnosis. The use of this procedure is described in detail in other guidelines.72
Exercise testing allows objective evaluation of exercise capacity and exertional symptoms, such as dyspnoea and fatigue.73 The 6-min walk test and a variety of treadmill and bicycle protocols are available. Gas exchange analysis helps differentiate between cardiac and respiratory causes of dyspnoea, shows whether the anaerobic threshold has been reached, and provides prognostic information (peak oxygen consumption is often measured as part of the assessment of candidates for heart transplantation). A normal exercise capacity in a patient not receiving treatment effectively excludes the diagnosis of symptomatic HF, although it must be remembered that there is a poor correlation between exercise capacity and resting haemodynamic measures, including EF.
The emerging role of genetic testing in ‘idiopathic’ dilated and hypertrophic cardiomyopathy is described in detail elsewhere.16 Currently this is recommended in patients with dilated cardiomyopathy and AV block or a family history of premature unexpected sudden death, as a prophylactic implantable cardioverter-defibrillator (ICD) may be indicated.
Ambulatory electrocardiographic monitoring
Ambulatory ECG monitoring is valuable in the assessment of patients with symptoms suggestive of an arrhythmia or bradycardia (e.g. palpitations or syncope) and in monitoring ventricular rate control in patients with AF. It is useful for identifying the type, frequency, and duration of atrial and ventricular arrhythmias, silent episodes of ischaemia and bradycardia, and conduction disturbances, which may cause or exacerbate HF.
Many variables provide prognostic information (Web Table 10), although most of this can be obtained from readily available data such as age, aetiology, NYHA class, EF, key co-morbidities (renal dysfunction, diabetes, anaemia, hyperuricaemia), and plasma natriuretic peptide concentration.74–80 Clearly these variables change over time, as does prognosis. Assessment of prognosis is particularly important when counselling patients about devices and surgery (including transplantation) and in planning end-of-life care with patients, their family, and caregivers.
Pharmacological treatment of heart failure with reduced ejection fraction (systolic heart failure)
Objectives in the management of heart failure
The goals of treatment in patients with established HF are to relieve symptoms and signs (e.g. oedema), prevent hospital admission, and improve survival. Although the focus of clinical trials was previously mortality, it is now recognized that preventing HF hospitalization is important for patients and healthcare systems.81 Reductions in mortality and hospital admission rates both reflect the ability of effective treatments to slow or prevent progressive worsening of HF. This is often accompanied by reverse LV remodelling and a reduction in circulating natriuretic peptide concentrations.82,83
The relief of symptoms, improvement in quality of life, and increase in functional capacity are also of the utmost importance to patients, but they have not been the primary outcome in most trials.84 This is in part because they are difficult to measure and partly because some treatments previously shown to improve these outcomes also decreased survival.85,86 However, effective pharmacological therapies and CRT improve these outcomes, as well as mortality and hospitalization.
Figure 2 shows a treatment strategy for the use of drugs (and devices) in patients with HF-REF; the recommendations for each treatment are summarized below. Three neurohumoral antagonists—an ACE inhibitor [or angiotensin receptor blocker (ARB)], a beta-blocker, and an MRA—are fundamentally important in modifying the course of systolic HF and should at least be considered in every patient. They are commonly used in conjunction with a diuretic given to relieve the symptoms and signs of congestion. The following text summarizes the evidence supporting the recommendations in this section, in Web Tables 11–13 and in Figure 2. The recommended doses of these disease-modifying medications are given in Table 14. The recommendations given in Section 7.4 summarize drugs that should be avoided in patients with HF-REF.
ACE = angiotensin-converting enzyme; ARB = angiotensin receptor blocker; b.i.d. = bis in die (twice daily); MRA = mineralocorticoid receptor antagonist; o.d. = omni die (once every day); t.i.d. = ter in die (three times daily).
aIndicates an ACE inhibitor where the dosing target is derived from post-myocardial infarction trials.
bIndicates drugs where a higher dose has been shown to reduce morbidity–mortality compared with a lower dose of the same drug, but there is no substantive placebo-controlled randomized controlled trial and the optimum dose is uncertain.
cIndicates a treatment not shown to reduce cardiovascular or all-cause mortality in patients with heart failure or after acute mycocardial infarction (or shown to be non-inferior to a treatment that does).
Treatments recommended in potentially all patients with systolic heart failure
Angiotensin-converting enzyme inhibitors and beta-blockers
The pivotal trials with beta-blockers were conducted in patients with continuing symptoms and a persistently low EF, despite treatment with an ACE inhibitor and, in most cases, a diuretic. Despite this, there is consensus that these treatments are complementary and that a beta-blocker and an ACE inhibitor should both be started as soon as possible after diagnosis of HF-REF. This is in part because ACE inhibitors have a modest effect on LV remodelling whereas beta-blockers often lead to a substantial improvement in EF. Furthermore, beta-blockers are anti-ischaemic, are probably more effective in reducing the risk of sudden cardiac death, and lead to a striking and early reduction in overall mortality33.
ACE = angiotensin-converting enzyme; ARB = angiotensin receptor blocker; EF = ejection fraction; HF = heart failure; MRA = mineralocorticoid receptor antagonist; NYHA = New York Heart Association.
aClass of recommendation.
bLevel of evidence.
Key evidence supporting the use of angiotensing-converting enzyme inhibitors
Two key randomized controlled trials [Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS)87 and Studies of Left Ventricular Dysfunction (SOLVD)-Treatment]88 assigned ∼2800 patients with mild to severely symptomatic HF to placebo or enalapril. Most were also treated with a diuretic and digoxin, but <10% of patients in each trial were treated with a beta-blocker. In CONSENSUS, which enrolled patients with severe HF, 53% of patients were treated with spironolactone.
Both of these two RCTs showed that ACE inhibitor treatment reduced mortality [relative risk reduction (RRR) 27% in CONSENSUS and 16% in SOLVD-Treatment]. In SOLVD-Treatment there was also an RRR of 26% in HF hospitalization. These benefits were additional to those gained with conventional treatment at that time (i.e. a diuretic, digoxin, and spironolactone).
The absolute risk reduction (ARR) in mortality in patients with mild or moderate HF (SOLVD-Treatment) was 4.5%, equating to a number needed to treat (NNT) of 22 to postpone one death (over an average of 41 months). The equivalent figures for severe HF (CONSENSUS) were 14.6% for ARR and 7 for NNT (over an average of 6 months).
These findings are supported by a meta-analysis of smaller, short-term, placebo-controlled randomized controlled trials (RCTs), which showed a clear reduction in mortality within only 3 months.89 These RCTs also showed that ACE inhibitors improve symptoms, exercise tolerance, quality of life, and exercise performance.
In the Assessment of Treatment with Lisinopril And Survival (ATLAS) trial,90 3164 patients with mainly moderate to severe HF were randomized to low- or high-dose lisinopril. There was an RRR of 15% in the risk of death or HF hospitalization in the high-dose lisinopril group compared with the low-dose lisinopril group.
Additional support for the use of ACE inhibitors comes from an RCT in patients with a low EF but no symptoms of HF (‘asymptomatic LV systolic dysfunction’) and three large (5966 patients in total) placebo-controlled, randomized, outcome trials in patients with HF, LV systolic dysfunction, or both after acute myocardial infarction.91 In the SOLVD-Prevention trial (which randomized 4228 patients with asymptomatic LV systolic dysfunction), there was a 20% RRR in death or HF hospitalization. In the myocardial infarction trials, which used captopril [Survival and Ventricular Enlargement (SAVE)], ramipril [Acute Infarction Ramipril Efficacy (AIRE)], and trandolapril [TRAndolapril Cardiac Evaluation (TRACE)], there was a 26% RRR in death and a 27% RRR in death or HF hospitalization.101
ACE inhibitors occasionally cause worsening of renal function, hyperkalaemia, symptomatic hypotension, cough, and, rarely, angioedema. An ACE inhibitor should only be used in patients with adequate renal function (creatinine ≤221 µmol/L or ≤2.5 mg/dL or eGFR ≥30 mL/min/1.73 m2) and a normal serum potassium level (see Web Table 11).
Practical guidance on how to use ACE inhibitors is given in Web Table 11.102
Key evidence supporting the use of beta-blockers
More RCTs have been undertaken with beta-blockers than with ACE inhibitors in patients with HF.
Three key trials [Cardiac Insufficiency Bisoprolol Study II (CIBIS II), Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS), and Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF)] randomized nearly 9000 patients with mild to severely symptomatic HF to placebo or a beta-blocker (bisoprolol, carvedilol, or metoprolol succinate CR/XL).92–96 More than 90% of the patients were on an ACE inhibitor or ARB.
Each of these three trials showed that beta-blocker treatment reduced mortality (RRR ∼34% in each trial) and HF hospitalization (RRR 28–36%) within ∼1 year of starting treatment. There was also an also an improvement in self-reported patient well-being in COPERNICUS and MERIT-HF. These benefits were additional to those gained with conventional treatment, including an ACE inhibitor.
The ARR in mortality (after 1 year of treatment) in patients with mild to moderate HF (CIBIS II and MERIT-HF combined) was 4.3%, equating to an NNT (for 1 year to postpone one death) of 23. The equivalent figures for severe HF (COPERNICUS) were ARR 7.1% and NNT 14.
These findings are supported by another placebo-controlled RCT [Study of Effects of Nebivolol Intervention on Outcomes and Rehospitalization in Seniors With Heart Failure (SENIORS)] in 2128 elderly (≥70 years) patients, 36% of whom had an LVEF >35%. Treatment with nebivolol resulted in an RRR of 14% in the primary composite endpoint of death or cardiovascular hospitalization, but did not reduce mortality.97
The findings of these trials were also supported by an earlier programme of studies with carvedilol (US carvedilol studies), a meta-analysis of other small beta-blocker trials, and a placebo-controlled RCT in 1959 patients with an LVEF ≤0.40 after acute myocardial infarction in which the RRR in mortality with carvedilol was 23% during a mean follow-up of 1.3 years.98
One large RCT [Beta-Blocker Evaluation of Survival Trial (BEST)] with bucindolol, a beta-blocker with partial agonist properties, did not show a significant reduction in mortality, though its findings were generally consistent with the above studies.103
Another RCT [Carvedilol or Metoprolol European Trial (COMET)] showed that carvedilol increased survival compared with short-acting metoprolol tartrate (different from the long-acting succinate formulation used in MERIT-HF).104
Beta-blockers should usually be initiated in stable patients, and used only with caution in recently decompensated patients (and only initiated in hospital in these patients). Recently decompensated patients were, however, safely initiated on beta-blocker treatment in COPERNICUS.105
Continuation of beta-blocker treatment during an episode of decompensation has been shown in an RCT to be safe, although dose reduction may be necessary.106 Temporary discontinuation is advised in shocked or severely hypoperfused patients. Re-institution of treatment should be attempted before discharge.
Practical guidance on how to use beta-blockers is given in Web Table 12.102
Mineralocorticoid/aldosterone receptor antagonists
Spironolactone and eplerenone block receptors that bind aldosterone and other corticosteroids, and are best characterized as MRAs. Although patients in the Eplerenone in Mild Patients Hospitalization and Survival Study in Heart Failure (EMPHASIS-HF)100 were required to have additional features elevating risk (recent cardiovascular hospitalization or elevated natriuretic peptide concentration), the benefits of MRAs probably extend to all patients with systolic HF, particularly as the two RCTs in chronic HF are supported by an additional RCT in patients with acute myocardial infarction.99,100,107
Key evidence supporting the use of mineralocorticoid receptor antagonists
The Randomized Aldactone Evaluation Study (RALES) trial99 was undertaken with the MRA spironolactone in patients with severe HF.
In RALES, 1663 patients with an EF ≤35% and in NYHA functional class III (having been in class IV within the past 6 months) were randomized to placebo or spironolactone 25–50 mg once daily added to conventional treatment. At the time this trial was conducted, beta-blockers were not widely used to treat HF, and only 11% were treated with a beta-blocker.
Treatment with spironolactone led to an RRR in death of 30% and an RRR in HF hospitalization of 35% within an average of 2 years of starting treatment. These benefits were additional to those gained with conventional treatment, including an ACE inhibitor.
The ARR in mortality (after a mean of 2 years of treatment) in patients with severe HF was 11.4%, equating to an NNT (for 2 years to postpone one death) of 9.
More recently the EMPHASIS-HF trial100 was undertaken in patients with systolic HF and mild symptoms.
In EMPHASIS-HF, 2737 patients aged ≥55 years with NYHA functional class II symptoms and an EF ≤30% (≤35% if the QRS duration was >130 ms) were enrolled. Patients had to have either experienced a cardiovascular hospitalization within the previous 6 months or have an elevated plasma natriuretic peptide concentration and be treated with an ACE inhibitor, ARB, or both, and a beta-blocker.
Treatment with eplerenone (up to 50 mg once daily) led to an RRR of 37% in cardiovascular death or HF hospitalization. Reductions were also seen in rates of death from any cause (24%), cardiovascular death (24%), hospitalization for any reason (23%), and HF hospitalization (42%). These benefits were obtained within an average of 21 months of starting treatment and were additional to those gained with conventional treatment, including an ACE inhibitor and beta-blocker.
The ARR in the primary composite mortality–morbidity endpoint in patients with mild symptoms was 7.7%, equating to an NNT (for an average of 21 months to postpone one event) of 13. The ARR in mortality was 3%, equating to an NNT of 33.
These findings are supported by another RCT [Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS)], which enrolled 6632 patients 3–14 days after acute myocardial infarction with an EF ≤40% and HF or diabetes.107 Patients were randomized to placebo or eplerenone 25–50 mg once daily added to conventional treatment including an ACE inhibitor/ARB (87%) and a beta-blocker (75%). Treatment with eplerenone led to an RRR in death of 15%.
Spironolactone and eplerenone can cause hyperkalaemia and worsening renal function, which were uncommon in the RCTs, but may occur more frequently in ordinary clinical practice, especially in the elderly. Both should only be used in patients with adequate renal function and a normal serum potassium concentration; if either is used, serial monitoring of serum electrolytes and renal function is mandatory.
Spironolactone can also cause breast discomfort and enlargement in men (10% compared with 1% on placebo, in RALES99); this side effect is infrequent with eplerenone.
Practical guidance on how to use MRAs is given in Web Table 13.102
7.2.3 Other treatments recommended in selected patients with systolic heart failure
This section describes other treatments that are valuable in patients with systolic HF. They have not, however, been shown clearly to reduce all-cause mortality [or in the case of hydralazine and isosorbide dinitrate (H-ISDN), this has only been clearly shown in African-Americans]. Most of these drugs have shown convincing benefits in terms of symptom reduction, HF hospitalization, or both, and are useful alternative or additional treatments in patients with HF34.
ACE = angiotensin-converting enzyme; ARB = angiotensin receptor blocker; CHARM-Added = Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity-Added; EF = ejection fraction; HF = heart failure; H-ISDN = hydralazine and isosorbide dinitrate; MRA = mineralocorticoid receptor antagonist; NYHA = New York Heart Association; PUFA = polyunsaturated fatty acid.
aClass of recommendation.
bLevel of evidence.
dIn the CHARM-Added trial, candesartan also reduced cardiovascular mortality.
eEuropean Medecines Agency has approved ivabradine for use in patients with a heart rate ≥75 b.p.m.
fPreparation studied in cited trial; the GISSI-HF trial had no EF limit.
Angiotensin receptor blockers
ARBs remain recommended as an alternative in patients intolerant of an ACE inhibitor.109,109 However, ARBs are no longer the first-choice recommendation in patients with HF and an EF ≤40% who remain symptomatic despite optimal treatment with an ACE inhibitor and beta-blocker. This is because in EMPHASIS-HF, eplerenone led to a larger reduction in morbidity–mortality than seen in the ARB ‘add-on’ trials discussed below, and because in both the Randomized Aldactone Evaluation Study (RALES) and EMPHASIS-HF, MRA treatment reduced all-cause mortality, whereas ARB ‘add-on’ treatment did not.
Two key placebo-controlled RCTs [Valsartan Heart Failure Trial (Val-HeFT) and CHARM-Added] randomized ∼7600 patients with mild to severely symptomatic HF to placebo or an ARB (valsartan and candesartan), added to an ACE inhibitor (in 93% of patients in Val-HeFT and all patients in CHARM-Added)110,111 In addition, 35% of patients in Val-HeFT and 55% in CHARM-Added were treated with a beta-blocker.
Each of these two trials showed that ARB treatment reduced the risk of HF hospitalization (RRR 24% in Val-HeFT and 17% in CHARM-Added) but not all-cause hospitalization. There was a 16% RRR in the risk of cardiovascular death with candesartan in CHARM-Added. These benefits were additional to those gained with conventional treatment, including a diuretic, digoxin, an ACE inhibitor, and a beta-blocker (but few patients were taking an MRA).
The ARR in the primary composite mortality–morbidity endpoint in patients with mild to moderate HF was 4.4%, equating to an NNT (for an average of 41 months to postpone one event) of 23 in CHARM-Added. The equivalent figures for Val-HeFT were ARR 3.3% and NNT 30 (over an average of 23 months).
The CHARM trials and Val-HeFT also showed that ARBs improve symptoms and quality of life. Other trials showed that these agents improve exercise capacity.
CHARM-Alternative was a placebo-controlled RCT with candesartan in 2028 patients with an LVEF ≤40%, intolerant of an ACE inhibitor. Treatment with candesartan resulted in an RRR of cardiovascular or HF hospitalization of 23% (ARR 7%, NNT 14, over 34 months of follow-up).108 Valsartan was also beneficial in the subset of patients in Val-HeFT not treated with an ACE inhibitor.109
Another trial [Evaluation of Losartan In The Elderly (ELITE) II118] failed to show that losartan 50 mg daily was as effective as captopril 50 mg three times daily. However, a subsequent RCT [Heart failure Endpoint evaluation of Angiotensin II Antagonist Losartan (HEAAL)119] showed that 150 mg daily of losartan was superior to 50 mg daily, supporting the similar findings of the Assessment of Treatment with Lisinopril And Survival (ATLAS) trial with the ACE inhibitor lisinopril—see above. In HEAAL there was an RRR of 10% in death or HF hospitalization in the high-dose losartan group (P = 0.027) over a median follow-up of 4.7 years. The results from these two trials, ATLAS90 and HEAAL,119 indicate that more benefit is obtained from using higher doses of renin–angiotensin system blockers and underscore the importance of attaining, if possible, the target doses proven to be of benefit in the key RCTs.
Additional support for the use of ARBs comes from the Valsartan In Acute myocardial infarction trial (VALIANT),120 an RCT in which 14 703 patients with HF, LV systolic dysfunction, or both after acute myocardial infarction were assigned to treatment with captopril, valsartan, or the combination. Valsartan was found to be non-inferior to captopril. In a similar trial [Optimal Therapy in Myocardial infarction with the Angiotensin II Antagonist Losartan (OPTIMAAL)121], losartan 50 mg once daily did not demonstrate non-inferiority when compared with captopril.
Practical guidance on how to use an ARB is given in Web Table 11.102
Ivabradine is a drug that inhibits the If channel in the sinus node. Its only known pharmacological effect is to slow the heart rate in patients in sinus rhythm (it does not slow the ventricular rate in AF).
The Systolic Heart failure treatment with the If inhibitor ivabradine Trial (SHIFT) enrolled 6588 patients in NYHA functional class II–IV, sinus rhythm with a rate of ≥70 b.p.m., and an EF ≤35%.112 Patients were also required to have had a HF hospitalization in the previous 12 months. They were randomized to ivabradine (up-titrated to a maximal dosage of 7.5 mg twice daily) or placebo, added to a diuretic (in 84%), digoxin (22%), an ACE inhibitor (79%), an ARB (14%), a beta-blocker (90%), and an MRA (60%). Only 26% of patients were, however, on full-dose beta-blocker. The median follow-up was 23 months. The RRR in the primary composite outcome of cardiovascular death or HF hospitalization was 18% (P < 0.0001); the reduction in cardiovascular death (or all-cause death) was not significant, but the RRR in HF hospitalization was 26%. The ARR in the primary composite mortality–morbidity endpoint was 4.2%, equating to an NNT (for an average of 23 months to postpone one event) of 24. Ivabradine also improved LV function and quality of life.
Five per cent of patients on ivabradine had symptomatic bradycardia compared with 1% of the placebo group (P < 0.0001). Visual side effects (phosphenes) were reported by 3% of patients on ivabradine and 1% on placebo (P < 0.0001).
Additional safety evidence for ivabradine comes from the MorBidity-mortality EvAlUaTion of the If inhibitor ivabradine in patients with coronary disease and left ventricULar dysfunction (BEAUTIFUL) trial, an RCT in which 10 917 patients with coronary heart disease and an EF <40% were assigned to treatment with ivabradine 7.5 mg twice daily or placebo and followed for a median of 19 months. Although ivabradine did not reduce the primary outcome of cardiovascular death, myocardial infarction, or HF hospitalization, it was well tolerated.122
Digoxin and other digitalis glycosides
In patients with symptomatic HF and AF, digoxin may be used to slow a rapid ventricular rate, although other treatments are preferred (see Section 10.1).
Digoxin may also be used in patients in sinus rhythm with symptomatic HF and an LVEF ≤40% as recommended below, based on the evidence summarized below.113
A single large morbidity–mortality RCT [Digitalis Investigation Group (DIG)] has been undertaken with digoxin in patients with symptomatic HF and a low EF.113
In the DIG trial, 6800 patients with an EF ≤45% and in NYHA functional class II–IV were randomized to placebo or digoxin (0.25 mg once daily), added to a diuretic and an ACE inhibitor. This trial was performed before beta-blockers were widely used for HF.113
Treatment with digoxin did not alter all-cause mortality but did lead to an RRR for hospital admission for worsening HF of 28% within an average of 3 years of starting treatment. The absolute ARR was 7.9%, equating to an NNT (for 3 years to postpone one patient admission) of 13.
These findings are supported by a meta-analysis of smaller trials suggesting that digoxin can improve symptoms and prevent deterioration.123
Digoxin can cause atrial and ventricular arrhythmias, particularly in the context of hypokalaemia, and serial monitoring of serum electrolytes and renal function is mandatory.
The efficacy and safety of other digitalis glycosides such as digitoxin have not been studied properly in heart failure.
Combination of hydralazine and isosorbide dinitrate
In one relatively small RCT conducted exclusively in men (and before ACE inhibitor or beta-blockers were used to treat HF), this vasodilator combination led to a borderline reduction in mortality when compared with placebo.114–116 In a subsequent RCT, the addition of H-ISDN to conventional therapy (ACE inhibitor, beta-blocker, and MRA) reduced morbidity and mortality (and improved symptoms) in African-Americans with HF.116 The selected patient population studied, relatively small RCT size, and early termination (for mortality benefit) have left uncertainty about the real value of this combination therapy, especially in non-black patients.
There are two placebo-controlled (V-HeFT-I and A-HeFT) RCTs and one active-controlled (V-HeFT-II) RCT with H-ISDN.114–116
In V-HeFT-I, 642 men were randomized to placebo, prazosin, or H-ISDN added to a diuretic and digoxin.114 No patients were treated with a beta-blocker or an ACE inhibitor (and the use of MRAs was not documented). Mortality rates were not different in the placebo and prazosin groups. With H-ISDN, there was a trend to a reduction in all-cause mortality during the overall period of follow-up (mean 2.3 years): RRR 22%; ARR 5.3%; NNT 19. H-ISDN increased exercise capacity and LVEF compared with placebo.
In A-HeFT, 1050 African-American men and women in NYHA class III or IV were randomized to placebo or H-ISDN, added to a diuretic (in 90%), digoxin (60%), an ACE inhibitor (70%), an ARB (17%), a beta-blocker (74%), and spironolactone (39%).116 The initial dose of treatment was 20 mg ISDN/37.5 mg hydralazine thrice daily, increasing to a target of 40 mg/75 mg thrice daily. The trial was discontinued prematurely, after a median follow-up of 10 months, because of a significant reduction in mortality (RRR 43%; ARR 4.0%; NNT 25). H-ISDN also reduced the risk of HF hospitalization (RRR 33%) and improved quality of life.
In V-HeFT-II, 804 men, mainly in NYHA class II or III, were randomized to enalapril or H-ISDN, added to a diuretic and digoxin.115 No patients were treated with a beta-blocker. There was a trend in the H-ISDN group to an increase in all-cause mortality during the overall period of follow-up (mean 2.5 years): relative increase in risk was 28%.
The most common adverse effects with H-ISDN in these trials were headache, dizziness/hypotension, and nausea. Arthralgia leading to discontinuation or reduction in dose of H-ISDN occurred in 5–10% of patients in V-HeFT I and II and a sustained increase in antinuclear antibody in 2–3% of patients (but lupus-like syndrome was rare).
Omega-3 polyunsaturated fatty acids
The small treatment effect of n-3 polyunsaturated fatty acids (PUFAs) in the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto miocardico-heart failure (GISSI-HF) trial was only detected after covariate adjustment in the statistical analysis and there was no effect on HF hospitalization.117 The effect of n-3 PUFAs after myocardial infarction is uncertain.
In the GISSI-HF PUFA trial, 6975 patients with NYHA class II–IV symptoms and an EF ≤40% (or if >40%, HF hospitalization in the previous year) were randomized to placebo or 1 g daily of an n-3 PUFA preparation in addition to standard therapy including an ACE inhibitor/ARB in 94%, beta-blocker in 65%, and spironolactone in 39%.117 The median follow-up was 3.9 years. n-3 PUFA treatment led to an RRR of 8% in the co-primary composite outcome of death or cardiovascular hospitalization in an adjusted analysis (adjusted P = 0.009). There was no reduction in HF hospitalization, but there was a 10% RRR in cardiovascular mortality (adjusted P = 0.045) and 7% RRR in cardiovascular hospitalization (adjusted P = 0.026).
These findings are supported by one post-myocardial infarction RCT (GISSI-Prevenzione124) but not by another (OMEGA125). In GISSI-Prevenzione, involving 11 324 patients enrolled after a recent (≤3 months) myocardial infarction, patients received placebo or 1 g daily of n-3 PUFA. n-3 PUFA treatment led to an RRR of 10% in the primary composite outcome of death, myocardial infarction, or stroke (largely driven by a reduction in cardiovascular death).
OMEGA randomized 3851 patients 3–14 days after acute myocardial infarction to placebo or 1 g n-3 PUFA daily for 1 year. Outcomes did not differ between treatment groups.
n-3 PUFA preparations differ in composition and the dose may be important.
The main adverse effects of n-3 PUFAs reported in these trials were nausea and other minor gastrointestinal disturbances.
Treatments not recommended (unproven benefit)
3-Hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors (‘statins’)
Although there is a wealth of robust evidence supporting the value of statins in patients with atherosclerotic (arterial) disease, most trials excluded patients with HF (because it was uncertain that they would benefit126). Two recent trials studied statin treatment specifically in patients with chronic HF and did not demonstrate convincing evidence of benefit (although there was little evidence of harm).127,128 Despite the evidence in other areas of cardiovascular medicine, the evidence does not therefore support the initiation of statins in most patients with chronic HF.
CORONA enrolled 5011 older patients (≥60 years) with symptomatic HF (NYHA class II–IV) of ischaemic aetiology with an EF ≤40%, felt by the investigator not to require cholesterol-lowering therapy. Rosuvastatin did not reduce the primary endpoint (cardiovascular death, myocardial infarction, or stroke) or all-cause mortality.127
The GISSI-HF statin trial enrolled 4574 patients with symptomatic HF (NYHA class II–IV) of ischaemic and non-ischaemic aetiology. Patients had an EF ≤40% (or if >40%, HF hospitalization in the previous year) and were randomized to placebo or rosuvastatin 10 mg daily, in addition to standard therapy including an ACE inhibitor/ARB in 94%, beta-blocker in 63% and spironolactone in 40%. The median follow-up was 3.9 years. The co-primary endpoints of all-cause mortality and the composite of all-cause death or cardiovascular hospitalization were not reduced by rosuvastatin.
Other than in patients with AF (both HF-REF and HF-PEF), there is no evidence that an oral anticoagulant reduces mortality–morbidity compared with placebo or aspirin (see Section 10.1).130a
Treatments not recommended (believed to cause harm)
ACE = angiotensin-converting enzyme; ARB = angiotensin receptor blocker; CCB = calcium-channel blocker; COX = cyclo-oxygenase; EF = ejection fraction; HF = heart failure; NSAID = non-steroidal anti-inflammatory drug; NYHA = New York Heart Association.
aClass of recommendation.
bLevel of evidence.
The effects of diuretics on mortality and morbidity have not been studied in patients with HF, unlike ACE inhibitors, beta-blockers, and MRAs (and other treatments). However, diuretics relieve dyspnoea and oedema and are recommended for this reason in patients with signs and symptoms of congestion, irrespective of EF.
Loop diuretics produce a more intense and shorter diuresis than thiazides, which cause a more gentle and prolonged diuresis. Thiazides may be less effective in patients with reduced kidney function. Loop diuretics are usually preferred to thiazides in HF-REF although they act synergistically and the combination may be used (usually on a temporary basis) to treat resistant oedema.
The aim of using diuretics is to achieve and maintain euvolaemia (the patient's ‘dry weight’) with the lowest achievable dose. This means that the dose must be adjusted, particularly after restoration of dry body weight, to avoid the risk of dehydration leading to hypotension and renal dysfunction. This may reduce cardiac output in patients with HF-PEF and often needlessly prevents the use of (or achievement of the target dose of) other disease-modifying therapies such as ACE inhibitors (or ARBs) and MRAs in patients with HF-REF. Many patients can be trained to self-adjust their diuretic dose, based on monitoring of symptoms/signs of congestion and daily weight measurements.
Practical guidance on the use of diuretics is given in Web Table 15 and the doses of commonly used diuretics are shown in Table 16.
ACEi = angiontensin-converting enzyme inhibitor; ARB = angiotensin receptor blocker.
aOral or intravenous; dose might need to be adjusted according to volume status/weight; excessive doses may cause renal impairment and ototoxicity.
bDo not use thiazides if estimated glomerular filtration rate <30 mL/min, except when prescribed synergistically with loop diuretics.
cIndapamide is a non-thiazide sulfonamide.
dA mineralocorticoid antagonist (MRA) i.e. spironolactone/eplerenone is always preferred. Amiloride and triamterene should not be combined with an MRA.
Use of potassium-sparing diuretics and potassium supplements
If a potassium-losing diuretic is used with the combination of an ACE inhibitor and an MRA (or ARB), potassium replacement is usually not required.
Serious hyperkalaemia may occur if potassium-sparing diuretics or supplements are taken in addition to the combination of an ACE inhibitor (or ARB) and MRA.
The use of all three of an ACE inhibitor, MRA and ARB is not recommended.
Pharmacological treatment of heart failure with ‘preserved’ ejection fraction (diastolic heart failure)
No treatment has yet been shown, convincingly, to reduce morbidity and mortality in patients with HF-PEF. Diuretics are used to control sodium and water retention and relieve breathlessness and oedema as in HF-REF. Adequate treatment of hypertension and myocardial ischaemia is also considered to be important, as is control of the ventricular rate in patients with AF (see Section 11). Two very small studies (<30 patients each) have shown that the heart rate-limiting calcium-channel blocker (CCB) verapamil may improve exercise capacity and symptoms in these patients.137,138 Rate-limiting CCBs may also be useful for ventricular rate control in patients with AF and in the treatment of hypertension and myocardial ischaemia (which is not the case in patients with HF-REF where their negative inotropic action can be dangerous). Beta-blockers may also be used to control the ventricular rate in patients with HF-PEF and AF.
The drugs that should be avoided in HF-REF (see Section 7.4) should also be avoided in HF-PEF, with the exception of CCBs.
The key mortality–morbidity trials to date are:
The 3023-patient Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM)-Preserved trial, which showed no reduction in the primary composite endpoint (cardiovascular death or HF hospitalization).139
The 850-patient Perindopril for Elderly People with Chronic Heart failure trial (PEP-CHF), which showed no reduction in the primary composite endpoint of death or HF hospitalization.140
The 4128 patient Irbesartan in heart failure with preserved systolic function trial (I-Preserve) which showed no reduction in the primary composite outcome of death or cardiovascular hospitalization (specifically, HF, myocardial infarction, unstable angina, arrhythmia, or stroke).141
Non-surgical device treatment of heart failure with reduced ejection fraction (systolic heart failure)
This section discusses the use of ICDs and CRT. While no new ICD RCT has completed since publication of the 2008 guidelines,1 there have been several important RCTs using CRT that have changed the recommendations (see below). Other technologies including a wearable defibrillator vest142 and implantable monitors (either ‘stand-alone’ or incorporated into other devices) are of research interest, but do not yet have enough evidence behind them to support guideline recommendations.
Approximately half of the deaths in patients with HF, especially in those with milder symptoms, occur suddenly and unexpectedly, and many, if not most, of these are related to ventricular arrhythmias (whereas others may be related to bradycardia and asystole). Prevention of sudden death is therefore an important goal in HF. While the key disease-modifying neurohumoral antagonists mentioned earlier reduce the risk of sudden death, they do not abort it. Specific antiarrhythmic drugs do not decrease this risk (and may even increase it).143 For this reason, ICDs have an important role to play in reducing the risk of death from ventricular arrhythmias36.
HF = heart failure; ICD = implantable cardioverter defibrillator; NYHA = New York Heart Association.
aClass of recommendation.
bLevel of evidence.
Secondary prevention of sudden cardiac death
ICDs reduce mortality in survivors of cardiac arrest and in patients with sustained symptomatic ventricular arrhythmias. Consequently, an ICD is recommended in such patients, irrespective of EF, with good functional status, a life expectancy of >1 year, and where the intent is to increase survival.144,147
Primary prevention of sudden cardiac death
The Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) enrolled 2521 patients with non-ischaemic dilated cardiomyopathy or ischaemic HF, no prior symptomatic ventricular arrhythmia, and an EF ≤35% who were in NYHA functional class II or III. These patients were randomized to placebo, amiodarone, or an ICD, in addition to conventional treatment including an ACE inhibitor or ARB (96%) and a beta-blocker (69%); MRA use was not reported.149
ICD treatment led to an RRR in death of 23% (P = 0.007) over a median follow-up of 45.5 months. This benefit was additional to that gained with conventional treatment, including an ACE inhibitor and a beta-blocker. Amiodarone did not reduce mortality.
The ARR in mortality with an ICD was 6.9%, equating to an NNT (for 45.5 months to postpone one death) of 14.
Additional support for the use of ICDs comes from the Multicenter Automatic Defibrillator Implantation Trial II (MADIT-II),148 an RCT in which patients with a prior myocardial infarction and an EF ≤30% (59% of which were in NYHA class II or III) were assigned to receive either conventional treatment or conventional treatment plus an ICD. Use of an ICD led to a 31% RRR in mortality. Two other RCTs showed no benefit in patients treated with an ICD early (≤40 days) after myocardial infarction.150,151 This is why ICD use in patients with coronary heart disease receives level of evidence A, but only in patients >40 days after acute myocardial infarction.
There is less evidence in patients with non-ischaemic HF, with one moderate sized trial [Defibrillators in Non-ischemic Cardiomyopathy Treatment Evaluation (DEFINITE), n = 458] showing only a non-significant trend to a reduction in mortality; hence the evidence level of B.152
ICD implantation should be considered only after a sufficient period of optimization of medical therapy (at least 3 months) and only if the EF remains persistently low.
ICD therapy is not indicated in patients in NYHA class IV with severe, drug-refractory, symptoms who are not candidates for CRT, a ventricular assist device, or cardiac transplantation (because such patients have a very limited life expectancy and are more likely to die from pump failure).
Patients should be counselled as to the purpose of an ICD and the complications related to its use (predominantly inappropriate shocks).153
If HF deteriorates, deactivation of a patient's ICD may be considered after appropriate discussion with the patient and caregiver(s).
Cardiac resynchronization therapy
Two large RCTs have shown that CRT is of benefit in patients with mild (NYHA class II) symptoms154,155 as well as in those who are more severely symptomatic.156,157 There is little doubt that patients expected to survive with good functional status for >1 year should receive CRT if they are in sinus rhythm, their LVEF is low (≤30%), QRS duration is markedly prolonged (≥150 ms), and an ECG shows a left bundle branch morphology, irrespective of symptom severity. There is less consensus about patients with right bundle branch block or interventricular conduction delay (based on subgroup analyses) and those in AF (because most trials excluded these patients and because a high ventricular rate will prevent resychronization). Another area of debate is what to do in an HF-REF patient without an indication for CRT who needs a conventional pacemaker.158 The possibility that patients with a QRS duration of <120 ms may have ‘mechanical dyssynchrony’ (detectable by imaging) and might benefit from CRT is another area of research interest but remains to be proven.159,1603738
CRT-D = cardiac resynchronization therapy defibrillator; CRT-P = cardiac resynchronization therapy pacemaker; EF = ejection fraction; HF = heart failure; LBBB = left bundle branch block; NYHA = New York Heart Association.
aClass of recommendation.
bLevel of evidence.
CRT-D = cardiac resynchronization therapy defibrillator; EF = ejection fraction; HF = heart failure; LBBB = left bundle branch block; NYHA = New York Heart Association.
aClass of recommendation.
bLevel of evidence.
Recommendations for cardiac resynchronization therapy where the evidence is certain
Key evidence supporting the use of cardiac resynchronization therapy
Moderate to severely symptomatic heart failure Mild to moderately symptomatic HF
Two key placebo-controlled RCTs [Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) and Cardiac Resynchronization in Heart Failure Study (CARE-HF)] randomized 2333 patients with moderate to severely symptomatic HF (NYHA class III or IV) to either optimal medical therapy or optimal medical therapy plus CRT.156,157 Patients in COMPANION were required to be in sinus rhythm, to have an EF ≤35% and a QRS duration of at least 120 ms, and a HF hospitalization or equivalent in the preceding year. Patients in CARE-HF were required to be in sinus rhythm and to have an EF ≤35%, a QRS duration ≥120 ms (if the QRS duration was 120–149 ms other echocardiographic criteria for dyssynchrony had to be met), and an LV end-diastolic dimension of at least 30 mm (indexed to height).
Each of these two trials showed that CRT reduced the risk of death from any cause and hospital admission for worsening HF [RRR in death of 24% with a CRT-pacemaker (CRT-P) and of 36% with CRT-defibrillator (CRT-D) in COMPANION and of 36% with CRT-P in CARE-HF]. In CARE-HF, the RRR in HF hospitalization with CRT-P was 52%. These benefits were additional to those gained with conventional treatment, including a diuretic, digoxin, an ACE inhibitor, a beta-blocker, and an MRA.
The ARR with CRT-D in the composite outcome of cardiovascular death or cardiovascular hospitalization in COMPANION was 8.6%, equating to an NNT (over a median duration of follow-up of ∼16 months) to postpone one event of 12. The corresponding figures for CRT-P in CARE-HF (over a mean follow-up of 29 months) were an ARR of 16.6% and an NNT of 6.
These trials also showed that CRT improves symptoms, quality of life, and ventricular function. Other trials showed that these agents improve exercise capacity.
Because these severely symptomatic patients have much to gain and because there was no subgroup of patients that clearly did not benefit from CRT, individuals in NYHA functional class III and IV have been given the broadest indication for CRT.
Two key placebo-controlled RCTs randomized 3618 patients with mild (MADIT-CRT, 15% NYHA class I and 85% NYHA class II) to moderately [Resynchronization/Defibrillation for Ambulatory Heart Failure Trial (RAFT), 80% NYHA class II and 20% NYHA class III] symptomatic HF to either optimal medical therapy plus an ICD or optimal medical therapy plus a CRT-D.154,155 Patients in MADIT-CRT were required to have an EF ≤30%, a QRS duration ≥130 ms, and to be in sinus rhythm. Patients in RAFT were required to have an EF ≤30% and a QRS duration ≥120 ms (13% of enrolled patients had AF with a well-controlled ventricular rate).
Each of these two trials showed that CRT reduced the risk of the primary composite endpoint of death or HF hospitalization (HF event in MADIT-CRT) (RRR of 34% in MADIT-CRT and 25% in RAFT). There was a 25% reduction in all-cause mortality in RAFT (P = 0.003), but mortality was not reduced in MADIT-CRT. These benefits were additional to those gained with conventional treatment, including a diuretic, digoxin, an ACE inhibitor, a beta-blocker, an MRA, and an ICD.
The ARR in the primary composite mortality–morbidity endpoint in MADIT-CRT was 8.1%, equating to an NNT (for an average of 2.4 years to postpone one event) of 12. The equivalent figures for RAFT were ARR 7.1% and NNT 14 (over an average of 40 months).
These trials also showed that CRT improves symptoms, quality of life, and ventricular function. Other trials showed that these agents improve exercise capacity.
Both MADIT-CRT and RAFT showed a significant treatment-by-subgroup interaction whereby QRS duration modified the treatment effect (CRT appeared more effective in patients with a QRS ≥150 ms) and patients with LBBB also seemed to obtain more benefit than those with right bundle branch block or an interventricular conduction defect (these groups overlap considerably, as patients with LBBB are more likely to have a QRS duration ≥150 ms). These findings are supported by echocardiographic analyses.161 For these reasons, in patients with milder symptoms, CRT is recommended only in those with either a QRS duration ≥150 ms or ≥130 ms plus an LBBB pattern.
Recommendations for cardiac resynchronization therapy where the evidence is uncertain
Two commonly encountered clinical situations where there is little robust evidence for (or against) CRT are AF and when a patient with a reduced EF has an indication for conventional pacing and no other indication for CRT.
One small, single-blind study [Multisite Stimulation in Cardiomyopathies (MUSTIC)] included 59 HF-REF patients with persistent/permanent AF, a slow ventricular rate necessitating permanent ventricular pacing, and a paced QRS duration ≥200 ms.162 The study had a crossover design (3 months conventional pacing vs. 3 months CRT). There was a high drop-out rate (42%) and there was no difference in the primary endpoint of 6-min walk distance. The key large RCTs of CRT all excluded patients in AF, with the exception of RAFT.158 RAFT included 229 patients with permanent AF or flutter either with a controlled ventricular rate (≤60 b.p.m. at rest and ≤90 b.p.m. during a 6-min walk test) or with planned AV junction ablation. Further analysis did not show a significant interaction between baseline rhythm and treatment effect, but this subgroup represented only a small proportion of the overall population. Other data suggesting that patients with AF (without AV nodal ablation) may benefit from CRT are limited by being observational in nature.163
Patients with an indication for conventional pacing
All the major RCTs of CRT, with the exception of RAFT, excluded patients with a conventional indication for pacing. RAFT included 135 patients with a paced QRS duration ≥200 ms, a subgroup too small for meaningful analysis.155 Conventional right ventricular pacing, however, alters the normal sequence of cardiac activation in a similar way to LBBB, and experimental and observational data suggest that this may lead to deterioration in LV systolic function.164,165 It is on this basis that CRT is recommended as an alternative to conventional right ventricular pacing in patients with HF-REF who have a standard indication for pacing or who require a generator change or revision of a conventional pacemaker39.
CRT-D = cardiac resynchronization therapy defibrillator; CRT-P = cardiac resynchronization therapy pacemaker; EF = ejection fraction; HF = heart failure; NYHA = New York Heart Association.
aClass of recommendation.
bLevel of evidence.
Arrhythmias, bradycardia, and atrioventricular block in patients with heart failure with reduced ejection fraction and heart failure with preserved ejection fraction
AF is the most common arrhythmia in HF; it increases the risk of thrombo-embolic complications (particularly stroke) and may lead to worsening of symptoms. Whether AF is an independent predictor of mortality is less certain, as is whether it can cause systolic HF (‘tachycardiomyopathy’).
AF should be classified and managed according to the current AF guidelines (i.e. first episode, paroxysmal, persistent, long-standing persistent, or permanent), recognizing the uncertainty about the actual duration of the episode and about previous undetected episodes.166
The following issues need to be considered in patients with HF and AF, especially a first episode of AF or paroxysmal AF:
Identification of correctable causes (e.g. hyperthyroidism, electrolyte disorders, uncontrolled hypertension, mitral valve disease).
Identification of potential precipitating factors (e.g. recent surgery, chest infection or exacerbation of chronic pulmonary disease/asthma, acute myocardial ischaemia, alcohol binge) as this may determine whether a rhythm-control strategy is preferred to a rate-control strategy.
Assessment for thromboembolism prophylaxis.
AF = atrial fibrillation; AV = atrioventricular; EF = ejection fraction; HF = heart failure; LV = left ventricular; NYHA = New York Heart Association.
aClass of recommendation.
bLevel of evidence.
For rate control in patients with HF-REF, a beta-blocker is preferred over digoxin as the latter does not provide rate control during exercise.167 Furthermore, beta-blockers have favourable effects on mortality and morbidity in systolic HF per se (see above). The combination of digoxin and a beta-blocker is more effective than a beta-blocker alone in controlling the ventricular rate at rest.168
In patients with HF-PEF, rate-limiting CCBs (verapamil and diltiazem) are an effective alternative to a beta-blocker (but their use is not recommended in patients with HF-REF as their negative inotropic action may further depresses LV systolic function).134,167 The combination of digoxin and a rate-limiting CCB is more effective than a CCB alone in controlling the ventricular rate at rest.169
Assessment of control of the ventricular rate on exertion requires either ambulatory ECG monitoring or measurement of the rate during moderate exercise. The optimum ventricular rate in patients with HF and AF is uncertain because the one RCT comparing strict with lenient rate control included very few patients with HF.170 In the Atrial Fibrillation and Congestive Heart Failure (AF-CHF) study (which showed similar outcomes for a rate-control compared with a rhythm-control strategy) the target rate was <80 b.p.m. at rest and <110 b.p.m. during a 6-min walk test.171
In extreme cases, AV node ablation and pacing may be required; in this situation in patients with systolic HF, CRT may be considered instead of conventional pacing (see Section 9.2).164
In patients with chronic HF, a rhythm-control strategy (including pharmacological or electrical cardioversion) has not been demonstrated to be superior to a rate-control strategy in reducing mortality or morbidity.171 This strategy is probably best reserved for patients with a reversible secondary cause of AF (e.g. hyperthyroidism) or an obvious precipitant (e.g. recent pneumonia) and in patients who cannot tolerate AF after optimization of rate control and HF therapy. Amiodarone is the only antiarrhythmic that should be used in patients with systolic HF.172,173 The role of catheter ablation as a rhythm control strategy in HF is at present uncertain.174,175
In patients with AHF, emergency cardioversion may be required to correct profound haemodynamic instability (see Section 12.2)41.
AF = atrial fibrillation; EF = ejection fraction; HF = heart failure; LV = left ventricular; NYHA = New York Heart Association.
aClass of recommendation.
bLevel of evidence.
Thrombo-embolism prophylaxis in patients with HF and AF should be based on the Cardiac failure, Hypertension, Age ≥75 (Doubled), Diabetes, Stroke (Doubled)-Vascular disease, Age 65–74 and Sex category (Female) (CHA2DS2-VASc) score (see Table 17), in keeping with the 2010 ESC AF guidelines.166,179 Most patients with systolic HF will have a risk score consistent with a firm indication for (score ≥2), or preference for, an oral anticoagulant (score = 1), although bleeding risk must also be considered (see below).
The Hypertension, Abnormal renal/liver function (1 point each), Stroke, Bleeding history or predisposition, Labile INR, Elderly (>65), Drugs/alcohol concomitantly (1 point each) (HAS-BLED score) (Table 18) is recommended to assess bleeding risk, in keeping with the 2010 ESC AF guidelines.166,180 A substantial proportion of patients with HF will have a score ≥3, indicating that careful consideration should be given before prescribing an oral anticoagulant and that regular review is needed (and correctable risk factors addressed) if an oral anticoagulant is given.
Some new anticoagulant drugs such as the oral direct thrombin inhibitors and oral factor Xa inhibitors are contraindicated in severe renal impairment (creatinine clearance <30 mL/min).181–183 This is clearly a concern in many patients with HF and, if these drugs are used, serial monitoring of renal function is required. There is no known way to reverse the anticoagulant action of these new drugs.42
AF = atrial fibrillation; CHA2DS2-VASc = Cardiac failure, Hypertension, Age ≥75 (Doubled), Diabetes, Stroke (Doubled), Vascular disease, Age 65–74 and Sex category (Female); EF = ejection fraction; HAS-BLED = Hypertension, Abnormal renal/liver function (1 point each), Stroke, Bleeding history or predisposition, Labile international normalized ratio, Elderly (>65), Drugs/alcohol concomitantly (1 point each); HF = heart failure; i.v. = intravenous; LMWH = low molecular weight heparin; LV = left ventricular; NYHA = New York Heart Association; TOE = transoesophageal echocardiography.
aClass of recommendation.
bLevel of evidence.
CHA2DS2-VASc = Cardiac failure, Hypertension, Age ≥75 (Doubled), Diabetes, Stroke (Doubled), Vascular disease, Age 65–74, and Sex category (Female); HF = heart failure; LVEF = left ventricular ejection fraction.
HAS-BLED = Hypertension, Abnormal renal/liver function (1 point each), Stroke, Bleeding history or predisposition, Labile international normalized ratio, Elderly (>65), Drugs/alcohol concomitantly (1 point each); NSAID = non-steroidal anti-inflammatory drug.
Ventricular arrhythmias are frequent in HF patients, particularly in those with a dilated left ventricle and reduced EF. Ambulatory ECG recording detects premature ventricular complexes in virtually all HF patients, and episodes of asymptomatic, non-sustained ventricular tachycardia are common.143 Historical studies have suggested that ‘complex ventricular arrhythmias’ (frequent premature ventricular complexes and non-sustained ventricular tachycardia) are associated with a poor outcome in HF. Certain recommendations from the American College of Cardiology/American Heart Association/ESC guidelines on the management of ventricular arrhythmias and sudden death, which may be particularly relevant to patients with HF, are summarized below. The role of catheter ablation in patients with HF other than as an adjunct in the treatment of refractory ventricular arrhythmias is uncertain.186 The reader is also referred to the section on ICDs (Section 9.1)43.
ACE = angiotensin-converting enzyme; ARB = angiotensin receptor blocker; HF = heart failure; ICD = implantable cardioverter-defibrillator; MRA = mineralocorticoid receptor antagonist.
aClass of recommendation.
bLevel of evidence.
Symptomatic bradycardia and atrioventricular block
Although the indications for pacing in patients with HF are similar to those in other patients, as described in the ESC guidelines on pacing,165 there are issues specific to HF, including:
Before implanting a conventional pacemaker in a patient with HF-REF, consider whether there is an indication for an ICD, CRT-P, or CRT-D (see Sections 9.1 and 9.2).
Because right ventricular pacing may induce dyssynchrony and worsen symptoms, CRT should be considered instead of conventional pacing in patients with HF-REF (see Section 9.2).
Physiological pacing to maintain an adequate chronotropic response and maintain atrial–ventricular coordination with a DDD system is preferable to VVI pacing in patients with both HF-REF and HF-PEF.165
Pacing solely in order to permit initiation or titration of beta-blocker therapy in the absence of a conventional indication is not recommended.
Importance and management of other co-morbidity in heart failure with reduced ejection fraction and heart failure with preserved ejection fraction
Heart failure and co-morbidities
Co-morbidities are important in patients with HF for four main reasons. First, co-morbidities may affect the use of treatments for HF (e.g. it may not be possible to use renin–angiotensin system inhibitors is some patients with renal dysfunction) (see Section 7.2). Secondly, the drugs used to treat co-morbidities may cause worsening of HF (e.g. NSAIDs given for arthritis) (see Section 7.4). Thirdly, the drugs used to treat HF and those used to treat co-morbidities may also interact with one another [e.g. beta-blockers and beta-agonists for chronic obstructive pulmonary disease (COPD) and asthma] and reduce patient adherence. Lastly, most co-morbidities are associated with worse clinical status and are predictors of poor prognosis in HF (e.g. diabetes). This has led to some co-morbidities themselves becoming targets for treatment (e.g. anaemia).187
Management of co-morbidities is a key component of the holistic care of patients with HF (see Section 14).
Anaemia (defined as a haemoglobin concentration <13 g/dL in men and <12 g/dL in women) is common in HF, particularly in hospitalized patients. It is more frequent in women, the elderly, and in patients with renal impairment. Anaemia is associated with more symptoms, worse functional status, greater risk of HF hospitalization, and reduced survival. A standard diagnostic work-up should be undertaken in anaemic patients. Correctable causes should be treated in the usual way, although no definite aetiology is identified in many patients. Correction of iron deficiency using i.v. iron has been specifically studied in patients with HF (see Section 11.14). The value of erythropoietin-stimulating agents as a treatment for anaemia of unknown aetiology is unknown but is currently being tested in a large mortality–morbidity RCT.187
Beta-blockers are effective agents for angina as well as an essential treatment for systolic HF. Certain other effective antianginal drugs have been studied in large numbers of patients with systolic HF and shown to be safe (e.g. amlodipine,188,189 ivabradine,112,122 and nitrates114–116). The safety of other antianginal agents such as nicorandil and ranolazine is uncertain, while other drugs, specifically dilatiazem and verapamil, are thought to be unsafe in patients with HF-REF (although they may be used in HF-PEF).134 Percutaneous and surgical revascularization are alternative approaches to the treatment of angina (see Section 13). Coronary artery bypass graft surgery may reduce morbidity and mortality in patients with HF-REF44.
EF = ejection fraction; HF = heart failure; LV, left ventricular; NYHA = New York Heart Association.
aClass of recommendation.
bLevel of evidence.
Asthma: see chronic obstructive pulmonary disease
See Section 11.7.
A generalized process, wasting all body compartments [i.e. lean tissue (skeletal muscle), fat tissue (energy reserves), and bone tissue (osteoporosis)], may occur in 10–15% of patients with HF, especially those with HF-REF. This serious complication is associated with worse symptoms and functional capacity, more frequent hospitalization, and decreased survival. Cachexia is specifically defined as involuntary non-oedematous weight loss ≥6% of total body weight within the previous 6–12 months.192 The causes are uncertain, but may include poor nutrition, malabsorption, impaired calorie and protein balance, hormone resistance, pro-inflammatory immune activation, neurohormonal derangements, and reduced anabolic drive. Potential treatments include appetite stimulants, exercise training, and anabolic agents (insulin, anabolic steroids) in combination with the application of nutritional supplements, although none is of proven benefit and their safety is unknown.
Certain chemotherapeutic agents can cause (or aggravate) LV systolic dysfunction and HF. The best recognized of these are the anthracyclines (e.g. doxorubicin) and trastuzumab.193,194 Dexrazoxane may confer some cardioprotection in patients receiving anthracyclines. Pre- and post-evaluation of EF is essential in patients receiving cardiotoxic chemotherapy, as detailed elsewhere.193,194 Patients developing LV systolic dysfunction should not receive further chemotherapy and should receive standard treatment for HF-REF. Mediastinal irradiation can also lead to a variety of long-term cardiac complications, although the less frequent use of high-dose, wide-field radiotherapy has led to a decline in these problems.
Chronic obstructive pulmonary disease
COPD and asthma may cause diagnostic difficulties, especially in HF-PEF.24,25 These conditions are associated with worse functional status and a worse prognosis. Beta-blockers are contraindicated in asthma but not in COPD, although a selective beta-1 adrenoceptor antagonist (i.e. bisoprolol, metoprolol succinate, or nebivolol) is preferred.195 Oral corticosteroids cause sodium and water retention, potentially leading to worsening of HF, but this is not believed to be a problem with inhaled corticosteroids. COPD is an independent predictor of worse outcomes in HF.
Depression is common and is associated with worse clinical status and a poor prognosis in HF. It may also contribute to poor adherence and social isolation. A high index of suspicion is needed to make the diagnosis, especially in the elderly. Routine screening using a validated questionnaire is good practice. Psychosocial intervention and pharmacological treatment are helpful. Selective serotonin reuptake inhibitors are thought to be safe, whereas tricyclic antidepressants are not because they may cause hypotension, worsening HF, and arrhythmias.196
Dysglycaemia and diabetes are very common in HF, and diabetes is associated with poorer functional status and worse prognosis. Diabetes may be prevented by treatment with ARBs and possibly ACE inhibitors.197 Beta-blockers are not contraindicated in diabetes and are as effective in improving outcome in diabetic patients as in non-diabetic individuals, although different beta-blockers may have different effects on glycaemic indices.198 Thiazolidinediones (glitazones) cause sodium and water retention and increased risk of worsening HF and hospitalization, and should be avoided (see recommendations, Section 7.4).131–133 Metformin is not recommended in patients with severe renal or hepatic impairment because of the risk of lactic acidosis, but is widely (and apparently safely) used in other patients with HF.199 The safety of newer antidiabetic drugs in HF is unknown.
Erectile dysfunction should be treated in the usual way; phosphodiesterase V inhibitors are not contraindicated other than in patients taking nitrates. Indeed short-term studies have shown that these agents have favourable haemodynamic and other effects in patients with HF-REF.200 There are, however, reports of phosphodiesterase V inhibitors causing worsening LV outflow tract obstruction in patients with hypertrophic cardiomyopathy, which may be a concern in some patients with HF-PEF.
Hyperuricaemia and gout are common in HF and may be caused or aggravated by diuretic treatment. Hyperuricaemia is associated with a worse prognosis in HF-REF.80 Xanthine oxidase inhibitors (allopurinol, oxypurinol) may be used to prevent gout, although their safety in HF-REF is uncertain.201 Gout attacks are better treated by colchicine than with NSAIDs (although colchicine should not be used in patients with very severe renal dysfunction and may cause diarrhoea). Intra-articular corticosteroids are an alternative for monoarticular gout, but systemic corticosteroids cause sodium and water retention.
Elevated low-density lipoprotein cholesterol is uncommon in HF-REF; patients with advanced HF-REF often have low concentrations of low-density lipoprotein, which is associated with a worse prognosis. Rosuvastatin did not reduce the primary composite mortality–morbidity endpoints in two large RCTs in HF.127,128
Hypertension is associated with an increased risk of developing HF; antihypertensive therapy markedly reduces the incidence of HF (with an exception of alpha-adrenoceptor blockers, which are less effective than other antihypertensives in preventing HF).202 Negatively inotropic CCBs (i.e. diltiazem and verapamil) should not be used to treat hypertension in patients with HF-REF (but are believed to be safe in HF-PEF), and moxonidine should also be avoided in patients with HF-REF as it increased mortality in patients in one RCT.203 If blood pressure is not controlled with an ACE inhibitor (or ARB), a beta-blocker, MRA, and diuretic, hydralazine and amlodipine (or felodipine204), are additional blood pressure-lowering agents shown to be safe in systolic HF. The blood pressure targets recommended in hypertension guidelines205 are applicable to HF.
In patients with AHF, i.v. nitrates (or sodium nitroprusside) are recommended to lower blood pressure (see Section 12).45
ACE = angiotensin-converting enzyme; ARB = angiotensin receptor blocker; HF = heart failure; LV = left ventricular; LVEF = left ventricular ejection fraction; MRA = mineralocorticoid receptor antagonist; NYHA = New York Heart Association.
aClass of recommendation.
bLevel of evidence.
Iron deficiency may contribute to muscle dysfunction in HF and causes anaemia. In a single RCT, 459 patients with NYHA class II or III systolic HF, a haemoglobin concentration between 9.5 and 13.5 g/dL, and iron deficiency (see below) were randomized 2:1 to i.v. ferric carboxymaltose or saline. In this trial, iron deficiency was diagnosed when serum ferritin was <100 µ/L or when the ferritin concentration was between 100 and 299 µg/L and transferrin saturation was <20%.208 Over 6 months of treatment, iron therapy improved self-reported patient global assessment and NYHA class (as well as 6-min walk distance and health-related quality of life) and may be considered as a treatment for these patients. The effect of treating iron deficiency in HF-PEF and the long-term safety of iron therapy in HF is unknown.
Kidney dysfunction and cardiorenal syndrome
The GFR is reduced in most patients with HF, especially if advanced, and renal function is a powerful independent predictor of prognosis in HF. Renin–angiotensin–aldosterone blockers (ACE inhibitors, renin inhibitors, ARBs, and MRAs) frequently cause a fall in GFR, although any reduction is usually small and should not lead to treatment discontinuation unless marked (see Web Table 11). Conversely, an immediate and large fall in GFR should raise the suspicion of renal artery stenosis. Sodium and water depletion (due to the excessive diuresis or fluid loss due to vomiting or diarrhoea) and hypotension are well recognized causes of renal dysfunction, but less well known is that volume overload, right heart failure, and renal venous congestion may also cause renal dysfunction. Other causes of kidney dysfunction are prostatic obstruction and nephrotoxic drugs such as NSAIDs and certain antibiotics (e.g. trimethoprim and gentamicin), all of which should be considered (and corrected or avoided) in HF patients with worsening renal function. Thiazide diuretics may be less effective in patients with a very low eGFR, and certain renally excreted drugs (e.g. digoxin, insulin, and low molecular weight heparin) may accumulate in patients with renal impairment. Sometimes the term ‘cardiorenal syndrome’ is used to describe concurrent heart and renal failure (and ‘cardiorenal–anaemia syndrome’ if there is concomitant anaemia).209
Chronic or acute renal dysfunction is a particular problem in patients with AHF, and is discussed further in that section (see Section 12).
Obesity is a risk factor for HF and complicates its diagnosis because it causes dyspnoea, effort intolerance, and ankle swelling, and may result in poor-quality echocardiographic images. Obese individuals also have reduced natriuretic peptide levels. Obesity is more common in HF-PEF than in HF-REF, although it is possible that misdiagnosis may explain at least some of this difference in prevalence. Obesity should be managed as recommended in other guidelines.210
Alpha-adrenoceptor blockers cause hypotension, and sodium and water retention, and may not be safe in systolic HF (see Section 11.13).202,206,207 For these reasons, 5-alpha reductase inhibitors are generally preferred. Prostatic obstruction should be ruled out in men with deteriorating renal function.
See Section 11.15.
Sleep disturbance and sleep-disordered breathing
Patients with HF frequently have sleep disturbance; the causes are many, including pulmonary congestion (leading to orthopnea and paroxysmal nocturnal dyspnoea) and diuretic therapy causing nocturnal diuresis. Anxiety and other psychological problems can also lead to insomnia, and reviewing sleep history is part of the holistic care of patients with HF (see Section 14). Up to one-third of patients with HF have sleep-disordered breathing.211,212 Sleep apnoea is of concern in patients with HF because it leads to intermittent hypoxaemia, hypercapnia, and sympathetic excitation. Obstructive sleep apnoea also causes recurrent episodes of negative intrathoracic pressure and increases in LV afterload. It is more common in patients who are obese and whose sleeping partners report that the patient snores or exhibits daytime somnolence (the patient may not be aware of these). However, not all patients with obstructive sleep apnoea are obese. The prevalence of central sleep apnoea (including Cheyne–Stokes respiration) in HF is uncertain and may have declined since the widespread use of beta-blockers and CRT. Screening for and the diagnosis and treatment of sleep apnoea is discussed in detail elsewhere.211,212 Diagnosis currently requires overnight polysomnography. Nocturnal oxygen supplementation, continuous positive airway pressure, bi-level positive airway pressure, and adaptive servo-ventilation may be used to treat nocturnal hypoxaemia.
Acute heart failure
Acute heart failure (AHF) is the term used to describe the rapid onset of, or change in, symptoms and signs of HF. It is a life-threatening condition that requires immediate medical attention and usually leads to urgent admission to hospital. In most cases, AHF arises as a result of deterioration in patients with a previous diagnosis of HF (either HF-REF or HF-PEF), and all of the aspects of chronic management described in these guidelines apply fully to these patients. AHF may also be the first presentation of HF (‘de novo’ AHF). AHF may be caused by an abnormality of any aspect of cardiac function (Web Table 3). In patients with pre-existing HF there is often a clear precipitant or trigger (e.g. an arrhythmia or discontinuation of diuretic therapy in a patient with HF-REF and volume overload or severe hypertension in patients with HF-PEF) (Table 19). The ‘acuteness’ may vary, with many patients describing a period of days or even weeks of deterioration (e.g. increasing breathlessness or oedema) but others developing HF within hours to minutes (e.g. in association with an acute myocardial infarction). Patients may present with a spectrum of conditions ranging from life-threatening pulmonary oedema or cardiogenic shock to a condition characterized, predominantly, by worsening peripheral oedema.
Diagnosis and treatment are usually carried out in parallel, especially in patients who are particularly unwell, and management must be initiated promptly. Close monitoring of the patient's vital functions is essential during the initial evaluation and treatment (see Sections 12.3 and 12.4) and some patients are best managed in an intensive or coronary care unit. Although the immediate goals of treatment are to improve symptoms and stabilize the patient's haemodynamic condition, longer term management, including post-discharge care, is also particularly important to prevent recurrences and improve prognosis in HF-REF. Pre- and post-discharge care should follow the recommendations outlined elsewhere in these guidelines, where applicable.
AHF = acute heart failure; COPD = chronic obstructive pulmonary disease; NSAID = non-steroidal anti-inflammatory drug.
Initial assessment and monitoring of patients
Three parallel assessments must be made during the initial evaluation of the patient, aided by the investigations listed in Figure 4.
Does the patient have HF or is there an alternative cause for their symptoms and signs (e.g. chronic lung disease, anaemia, kidney failure, or pulmonary embolism)?
If the patient does have HF, is there a precipitant and does it require immediate treatment or correction (e.g. an arrhythmia or acute coronary syndrome)?
Is the patient's condition immediately life-threatening because of hypoxaemia or hypotension leading to underperfusion of the vital organs (heart, kidneys, and brain)?
Treatment of acute heart failure
Often treatment must be administered in parallel with the diagnostic work-up (see treatment algorithm, Figure 5). Although not ‘evidence based’ in the same way as treatments for chronic HF, the key drugs are oxygen, diuretics, and vasodilators. Opiates and inotropes are used more selectively, and mechanical support of the circulation is required only rarely. Non-invasive ventilation is used commonly in many centres, but invasive ventilation is required in only a minority of patients.
Systolic blood pressure, heart rhythm and rate, saturation of peripheral oxygen (SpO2) using a pulse oximeter, and urine output should be monitored on a regular and frequent basis until the patient is stabilized (see also Sections 12.3 and 12.4)460.
ACE = angiotensin-converting enzyme; ACS = acute coronary syndrome; AF = atrial fibrillation; ARB = angiotensin receptor blocker; CABG = coronary artery bypass graft; CPAP = continuous positive airway pressure; ECG = electrocardiogram; EF = ejection fraction; HF = heart failure; i.v. = intravenous; LBBB = left bundle branch block; LMWH = low molecular weight heparin; LV = left ventricular; PaO2 = partial pressure of oxygen; PCI = percutaneous coronary intervention; TOE = transoesophageal echocardiography.
aClass of recommendation.
bLevel of evidence.
Oxygen may be given to treat hypoxaemia (SpO2 <90%), which is associated with an increased risk of short-term mortality. Oxygen should not be used routinely in non-hypoxaemic patients as it causes vasoconstriction and a reduction in cardiac output.224
Most patients with dyspnoea caused by pulmonary oedema obtain rapid symptomatic relief from administration of an i.v. diuretic, as a result of both an immediate venodilator action and subsequent removal of fluid. The optimum dose and route of administration (bolus or continuous infusion) are uncertain. A recent, small, prospective RCT compared 12-hourly bolus injection with continuous infusion and low-dose (equal to pre-existing oral dose) with high-dose (×2.5 times previous oral dose) using a 2 × 2 factorial design.213 There was no difference between either of the treatment comparisons for the co-primary endpoints (patient global assessment of symptoms and change in serum creatinine). Compared with the low-dose strategy, the high-dose strategy was, however, associated with greater improvement in a number of secondary outcomes (including dyspnoea) but at the expense of more transient worsening of renal function.
In patients with resistant peripheral oedema (and ascites), a combination of a loop and a thiazide (e.g. bendroflumethiazide) or thiazide-like diuretic (metolazone) may be needed to achieve an adequate diuresis (see Web Table 15).225,226 This potent combination is usually only needed for a few days and requires careful monitoring to avoid hypokalaemia, renal dysfunction, and hypovolaemia.
Opiates such as morphine may be useful in some patients with acute pulmonary oedema as they reduce anxiety and relieve distress associated with dyspnoea. Opiates are also thought to be venodilators, reducing preload, and may also reduce sympathetic drive. Conversely, opiates induce nausea (necessitating the concomitant administration of an antiemetic, one of which, cyclizine,227 has vasoconstrictor activity) and depress respiratory drive, potentially increasing the need for invasive ventilation.
Although vasodilators such as nitroglycerine (Table 20) reduce preload and afterload and increase stroke volume, there is no robust evidence that they relieve dyspnoea or improve other clinical outcomes.218,220 Vasodilators are probably most useful in patients with hypertension and should be avoided in patients with a systolic blood pressure <110 mmHg. Excessive falls in blood pressure should also be avoided because hypotension is associated with higher mortality in patients with AHF. Vasodilators should be used with caution in patients with significant mitral or aortic stenosis.
aNot available in many European Society of Cardiology countries.
Nesiritide—a human BNP that acts mainly as a vasodilator—was recently shown to reduce dyspnoea by a small but statistically significant amount when added to conventional treatment (mainly diuretic).228
Use of an inotrope such as dobutamine (Table 21) should usually be reserved for patients with such severe reduction in cardiac output that vital organ perfusion is compromised. Such patients are almost always hypotensive (‘shocked’). Inotropes cause sinus tachycardia and may induce myocardial ischaemia and arrhythmias. There is long-standing concern that they may increase mortality. There is pharmacological rationale to use levosimendan (or a phosphodiesterase III inhibitor such as milrinone) if it is felt necessary to counteract the effect of a beta-blocker.
aAlso a vasodilator.
bBolus not recommended in hypotensive patients (systolic blood pressure <90 mmHg).
α = alpha adrenoceptor; β = beta adrenoceptor; δ = dopamine receptor.
Drugs with prominent peripheral arterial vasoconstrictor action such as norepinephrine (Table 21) are sometimes given to severely ill patients with marked hypotension. These agents are given to raise blood pressure and redistribute cardiac output from the extremities to the vital organs. However, this is at the expense of an increase in LV afterload, and these agents have adverse effects similar to those of inotropes (and the most commonly used of these agents, norepinephrine and epinephrine, have inotropic activity). Their use should be restricted to patients with persistent hypoperfusion despite adequate cardiac filling pressures.
In large doses (>5 µg/kg/min) dopamine has inotropic and vasoconstrictor activity. At lower doses (<3 µg/kg/min) dopamine may have a selective renal arterial vasodilator activity and promote natriuresis, although this is uncertain. Dopamine may cause hypoxaemia.229 Arterial oxygen saturation should be monitored, and supplemental oxygen administrated as required.
Other pharmacological therapy
Thrombo-embolism prophylaxis with heparin or another anticoagulant should be used, unless contraindicated or unnecessary (because of existing treatment with oral anticoagulants).214–216 Tolvaptan (a vasopressin V2-receptor antagonist) may be used to treat patients with resistant hyponatraemia (thirst and dehydration are recognized adverse effects).230
Angiotensin-converting enzyme inhibitor/angiotensin receptor blocker
In patients with reduced EF not already receiving an ACE inhibitor (or ARB), this treatment should be started as soon as possible, blood pressure and renal function permitting (see recommendations in Section 7.2.1 and Web Table 11). The dose should be up-titrated as far as possible before discharge, and a plan made to complete dose up-titration after discharge.
In patients with reduced EF not already receiving a beta-blocker, this treatment should be started as soon as possible after stabilization, blood pressure and heart rate permitting (see recommendations in Section 7.1 and Web Table 12). The dose should be up-titrated as far as possible before discharge, and a plan made to complete dose up-titration after discharge. It has been shown that beta-blocker treatment may be continued in many patients during an episode of decompensation and started safely before discharge after an episode of decompensation.
Mineralocorticoid (aldosterone) receptor antagonist
In patients with reduced EF not already receiving an MRA, this treatment should be started as soon as possible, renal function and potassium permitting (see recommendations in Section 7.2 and Web Table 13). As the dose of MRA used to treat HF has a minimal effect on blood pressure, even relatively hypotensive patients may be started on this therapy during admission. The dose should be up-titrated as far as possible before discharge, and a plan made to complete dose up-titration after discharge.
In patients with reduced EF, digoxin may be used to control the ventricular rate in AF, especially if it has not been possible to up-titrate the dose of beta-blocker. Digoxin may also provide symptom benefit and reduce the risk of HF hospitalization in patients with severe systolic HF (see recommendations in Section 7.2.6).
It is common to restrict sodium intake to <2 g/day and fluid intake to <1.5–2.0 L/day, especially (the latter in hyponatraemic patients) during the initial management of an acute episode of HF associated with volume overload, although there is no firm evidence to support this practice.
Continuous positive airway pressure (CPAP) and non-invasive positive pressure ventilation (NIPPV) relieve dyspnoea and improve certain physiological measures (e.g. oxygen saturation) in patients with acute pulmonary oedema. However, a recent large RCT showed that neither type of non-invasive ventilation reduced mortality or the rate of endotracheal intubation when compared with standard therapy, including nitrates (in 90% of patients) and opiates (in 51% of patients).217 This result is in contrast to the findings of meta-analyses of earlier, smaller studies.
Non-invasive ventilation may be used as adjunctive therapy to relieve symptoms in patients with pulmonary oedema and severe respiratory distress or who fail to improve with pharmacological therapy. Contraindications include hypotension, vomiting, possible pneumothorax, and depressed consciousness.
Endotracheal intubation and invasive ventilation
The primary indication for endotracheal intubation and invasive ventilation is respiratory failure leading to hypoxaemia, hypercapnia, and acidosis. Physical exhaustion, diminished consciousness, and inability to maintain or protect the airway are other reasons to consider intubation and ventilation.
Mechanical circulatory support
Intra-aortic balloon pump
The conventional indications for an intra-aortic balloon pump (IABP) are to support the circulation before surgical correction of specific acute mechanical problems (e.g. interventricular septal rupture and acute mitral regurgitation), during severe acute myocarditis and in selected patients with acute myocardial ischaemia or infarction before, during, and after percutaneous or surgical revascularization. There is no good evidence that an IABP is of benefit in other causes of cardiogenic shock.231 More recently, balloon pumps (and other types of short-term, temporary circulatory support) have been used to bridge patients until implantation of a ventricular assist device or heart transplantation (see Section 13.5).
Ventricular assist devices
Ventricular assist devices and other forms of mechanical circulatory support (MCS) may be used as a ‘bridge to decision’ or longer term in selected patients (see Section 13.5).
Venovenous isolated ultrafiltration is sometimes used to remove fluid in patients with HF,232 although is usually reserved for those unresponsive or resistant to diuretics.
Insertion of an intra-arterial line should only be considered in patients with persistent HF and a low systolic blood pressure despite treatment.
Pulmonary artery catheterization
Right heart catheterization does not have a general role in the management of AHF, but may help in the treatment of a minority of selected patients with acute (and chronic) HF.233 Pulmonary artery catheterization should only be considered in patients: (i) who are refractory to pharmacological treatment; (ii) who are persistently hypotensive; (iii) in whom LV filling pressure is uncertain; or (iv) who are being considered for cardiac surgery. A primary concern is to ensure that hypotension (and worsening renal function) is not due to inadequate LV filling pressure, in which case diuretic and vasodilator therapy should be reduced (and volume replacement may be required). Conversely, a high LV filling pressure and/or systemic vascular resistance may suggest an alternative pharmacological strategy (e.g. inotropic or vasodilator therapy), depending on blood pressure. Measurement of pulmonary vascular resistance (and its reversibility) is a routine part of the surgical work-up before cardiac transplantation.
Monitoring after stabilization
Heart rate, rhythm, blood pressure, and oxygen saturation should be monitored continuously for at least the first 24 h of admission, and frequently thereafter. Symptoms relevant to HF (e.g. dyspnoea) and related to the adverse effects of treatments used (e.g. dizziness) should be assessed at least daily. Fluid intake and output, weight, and the jugular venous pressure and extent of pulmonary and peripheral oedema (and ascites if present) should be measured daily to evaluate the correction of volume overload. Blood urea nitrogen, creatinine, potassium, and sodium should be monitored daily during i.v. therapy and when renin–angiotensin–aldosterone system antagonists are being initiated or if the dose of any of these drugs is changed.
Other in-patient assessments
After initial treatment of the acute episode, every patient should be assessed for possible causes of HF (if the HF is new) and precipitants of worsening (if the HF has previously been diagnosed). The focus is detection of reversible or treatable causes (Table 19).
Readiness for discharge
Before discharge is contemplated, the acute episode of HF should have resolved and, in particular, congestion should be absent and a stable oral diuretic regimen established for at least 48 h.234–236 Long-term disease-modifying therapy (including a beta-blocker) should be optimized as much as possible and appropriate education provided to the patient and family/caregivers. Pre- and post-discharge management should follow the standards of care laid out by the Heart Failure Association.236 The goals of treatment during the different stages of management of patients with HF are summarized in Table 22.
CCU = coronary care unit; ED = emergency department; ICU = intensive care unit.
Special patient populations
Patients with a concomitant acute coronary syndrome
Patients with a concomitant acute coronary syndrome should be assessed and treated according to the current acute coronary syndrome guidelines.237,238 They should undergo coronary angiography and revascularization as appropriate. This should be undertaken as an urgent procedure in patients with haemodynamic instability and as an emergency procedure in those in cardiogenic shock. If haemodynamic instability persists despite optimal medical treatment, an IABP should be inserted before coronary angiography and revascularization. Persistent haemodynamic instability may also be caused by mechanical complications of infarction (e.g. mitral valve papillary muscle rupture), which may be identified using echocardiography and may require urgent corrective surgery.
Isolated right ventricular failure
New-onset isolated right ventricular failure may occur secondarily to an acute coronary syndrome (and is managed as described above) and following massive pulmonary embolism (see pulmonary embolism guidelines239). In both situations, diuretics and vasodilators should be used cautiously or avoided so as not to reduce right ventricular filling.
Progressive isolated right ventricular failure may occur in patients with pulmonary hypertension. Type V phosphodiesterase inhibitors, endothelin antagonists, and prostacyclin analogues may help by decreasing pulmonary arterial resistance (see guidelines240).
12.7.3 Acute heart failure with ‘cardiorenal syndrome’
Acutely worsening HF, or its treatment, or both may cause acute worsening of renal function (the so-called ‘type 1 cardiorenal syndrome’) in up to one-third of patients, and is associated with worse survival and prolonged hospitalization.209 An acute renocardiac syndrome (the so-called ‘type 3 cardiorenal syndrome’), characterized by worsening cardiac function secondary to volume overload resulting from acute kidney injury, may also occur, but is less common. The main management issues with these patients are that renal dysfunction may limit the use of renin–angiotensin–aldosterone system blockers and that progressive uraemia and volume overload may necessitate renal replacement therapy. Often these patients are best cared for jointly with a nephrologist.
Perioperative acute heart failure
AHF may occur in patients before (e.g. because of pre-operative infarction), during (‘failure to wean’), and after (mechanical complications and pericardial tamponade must be excluded) cardiac surgery. The specialized management of this group of patients is described in detail elsewhere241 and may involve use of mechanical support, including extracorporeal membrane oxygenation (ECMO).
Adult congenital heart disease
Patients with adult congenital heart disease (ACHD) are a very heterogeneous patient population. The diagnosis and management of HF in these patients can be very complex, and close collaboration with a tertiary referral centre is mandatory.
Patients with ACHD may present with HF due to a reduced systemic LVEF, reduced systemic right ventricular EF, or isolated subpulmonary right ventricular failure (see Section 12.7.2). Patients with univentricular hearts, either unoperated or palliated by a Fontan procedure, are particularly difficult to evaluate and treat. CMR and cardiopulmonary exercise testing are especially valuable in their assessment, but the acquisition and interpretation of data require special expertise.66,244
There is a lack of multicentre RCTs to guide the treatment of HF in patients with ACHD. There are, however, a number of general empirical principles of management: (i) residual (post-repair) or new haemodynamic lesions should always be sought first; (ii) the value of ACE inhibitors, ARBs, and beta-blockers in ACHD is controversial and these drugs may even be harmful in certain patients, e.g. those with a Fontan circulation (see ESC guidelines245); (iii) pulmonary arterial vasodilators may be useful in certain patients with pulmonary hypertension (see ESC guidelines240); (iv) the role of CRT is unknown; and (v) heart transplantation is an option but may be precluded by factors such as complex cardiovascular anatomy, and renal and hepatic dysfunction.
Coronary revascularization and surgery, including valve surgery, ventricular assist devices, and transplantation
Surgical (and percutaneous) coronary revascularization is indicated for the relief of angina pectoris in patients with either HF-REF or HF-PEF, and surgical coronary revascularization is indicated for ‘prognostic’ reasons in other patients with severe CAD, particularly those with three-vessel disease or left-main stenosis. The detailed indications for coronary revascularization are covered elsewhere.71
This section focuses on recent developments relevant to HF. The Surgical Treatment for Ischemic Heart Failure (STICH) trial addressed the broader role of surgical revascularization in patients with HF-REF and less severe CAD.191 Patients with an EF ≤35% and CAD who were suitable for surgery were randomized to coronary artery bypass graft (CABG) plus medical therapy or medical therapy alone. The patients enrolled were young (average age 60 years), predominantly male (88%), and were in NYHA class I (11%), II (52%), or III (34%). Their Canadian Cardiovascular Society angina class was 0 in 36%, I in 16%, II in 43%, III in 4%, and IV in 1%. Most patients had two-vessel (31%) or three-vessel (60%) CAD, and 68% had a severe proximal left anterior descending stenosis; very few (2%) had a left-main stenosis. The primary outcome (all-cause death) was not reduced by CABG. CABG did, however, reduce the secondary outcomes of cardiovascular death (RRR 19%) and death from any cause or cardiovascular hospitalization (RRR 26%). This trial may therefore extend the indication for CABG to ‘STICH-like’ patients with two-vessel CAD, including a left anterior descending stenosis, who are otherwise suitable for surgery and expected to survive >1 year with good functional status.
The benefit–risk balance for CABG in patients without angina/ischaemia or without viable myocardium remains uncertain. Patients with >10% of dysfunctional but viable LV myocardium may be more likely to benefit from myocardial revascularization (and those with ≤10% less likely to benefit) although this approach to patient selection for revascularization is unproven. Several non-invasive techniques can be used to assess myocardial viability (Table 7). Nuclear imaging has a high sensitivity, whereas techniques evaluating contractile reserve have lower sensitivity but higher specificity. CMR is excellent for assessing the transmural extent of scar, but is not better at detecting viability or predicting recovery of wall motion.
The choice between percutaneous coronary intervention and CABG should be made by the Heart Team, including a HF specialist, and be based on the extent of CAD, expected completeness of revascularization, associated valvular disease, and the presence of co-morbidities47.
CABG = coronary artery bypass graft; EF = ejection fraction; HF = heart failure; LV = left ventricular; PCI = percutaneous coronary intervention.
aClass of recommendation.
bLevel of evidence.
The value of surgical ventricular reconstruction during which scar tissue is removed from the LV wall, with the aim of restoring a more physiological LV volume and shape, is uncertain and was not shown to be of benefit in STICH.246 This technique is not recommended for routine use and is discussed further in the revascularization guidelines.71 External containment devices are not recommended.
Valvular heart disease may cause or aggravate HF. This section briefly addresses problems particularly relevant to HF, and the reader is referred to the recent ESC/European Association for Cardio-Thoracic Surgery guidelines on valvular disease for more information.247
The main concern in patients with LV systolic dysfunction is the entity of ‘low-flow, low-gradient’ aortic stenosis (valve area <1 cm2, EF <40%, mean gradient <40 mmHg) because some may have severe aortic stenosis and others ‘pseudo-aortic stenosis’ (i.e. where the low flow across the aortic valve is not caused by a severe fixed obstruction but by low stroke volume). In such individuals, low-dose dobutamine stress echocardiography may help differentiate between these two types of patient and provide information about contractile reserve which is of prognostic importance. In patients with severe aortic stenosis and a low EF, individuals with contractile reserve have a lower operative mortality and better long-term prognosis.
If the mean gradient is >40 mmHg, there is theoretically no lower EF limit for aortic valve replacement in symptomatic patients with severe aortic stenosis. However, substantial recovery of LV function is only likely when the reduced EF is caused by excessive afterload and is not due to scar.
Medical treatment should be optimized, although vasodilators (ACE inhibitors, ARBs, renin inhibitors, CCBs, hydralazine, and nitrates) may cause substantial hypotension in patients with severe aortic stenosis and should only be used with great caution. Optimization of treatment should not delay surgical decision-making. In patients not medically fit for surgery (e.g. because of severe pulmonary disease), transcatheter aortic valve replacement should be considered.248,249
Aortic valve repair or replacement is recommended in all symptomatic patients and in asymptomatic patients with severe aortic regurgitation and an EF <50%, who are otherwise fit for surgery. Surgery should also be considered in patients with severe aortic regurgitation and an LV end-diastolic diameter >70 mm or end-systolic diameter >50 mm (or >25 mm/m2 body surface area if small stature).31 Surgery is indicated to reduce the risk of death, and HF and LV function usually improve after aortic valve repair.
It is important not to confuse mild to moderate aortic incompetence secondary to LV dilatation with LV dilatation and systolic dysfunction due to primary severe aortic regurgitation.
Assessment of mitral regurgitation is complex, particularly in patients with systolic dysfunction (and assessment of systolic function is complicated in the presence of mitral regurgitation—see Section 4.1). Differentiating between primary and secondary mitral regurgitation is crucial (see below).
The decision to recommend surgery should take account of symptoms, age, concurrent AF, reduced LV systolic function, pulmonary hypertension, and the suitability of the valve for repair, which are the most important predictors of post-operative outcome.
Primary (organic) mitral regurgitation
In primary mitral regurgitation due to flail leaflets, an LV end-systolic diameter ≥40 mm is associated with increased mortality whether the patient is treated medically or surgically. When the EF is <30%, a durable surgical repair may improve symptoms, although its effect on survival is unknown. In this situation, the decision to operate should take account of response to medical therapy, co-morbidity, and the likelihood that the valve can be repaired (rather than replaced).
Secondary mitral regurgitation
This occurs because LV enlargement and remodelling lead to reduced leaflet closing. Effective medical therapy leading to reverse remodelling of the LV may reduce functional mitral regurgitation, and every effort should be made to optimize medical treatment in these patients.
Ischaemic mitral regurgitation is a particular type of secondary mitral regurgitation that may be more suitable for surgical repair. As it is often a dynamic condition, stress testing is important in its evaluation. An exercise-induced increase of effective regurgitant orifice (≥13 mm2) is associated with a worse prognosis. Combined valve and coronary surgery should be considered in symptomatic patients with LV systolic dysfunction, coronary arteries suitable for revascularization, and evidence of viability. Predictors of late failure of valve repair include large interpapillary muscle distance, severe posterior mitral leaflet tethering, and marked LV dilatation (LV end-diastolic diameter >65 mm). In these patients, mitral valve replacement, rather than repair, may be advisable. In the presence of AF, atrial ablation and left atrial appendage closure may be considered at the time of mitral valve surgery.
The role of isolated mitral valve surgery in patients with severe functional mitral regurgitation and severe LV systolic dysfunction who cannot be revascularized or have non-ischaemic cardiomyopathy is questionable, and in most patients conventional medical and device therapy are preferred. In selected cases, repair may be considered in order to avoid or postpone transplantation.
In patients with an indication for valve repair but judged inoperable or at unacceptably high surgical risk, percutaneous edge-to-edge repair may be considered in order to improve symptoms.250
Heart transplantation is an accepted treatment for end-stage HF.251,252 Although controlled trials have never been conducted, there is consensus that transplantation—provided that proper selection criteria are applied—significantly increases survival, exercise capacity, quality of life, and return to work compared with conventional treatment.
Apart from the shortage of donor hearts, the main challenges in transplantation are the consequences of the limited effectiveness and complications of immunosuppressive therapy in the long term (i.e. antibody-mediated rejection, infection, hypertension, renal failure, malignancy, and coronary artery vasculopathy). The indications for and contraindications to heart transplantation are summarized in Table 23.
HF = heart failure.
Mechanical circulatory support
MCS is an umbrella term describing a number of different technologies used to provide both short- and longer term assistance in patients with either chronic HF or AHF. A variety of terms have been used to describe the use of these technologies (Table 24).211,253 The most experience is with MCS in end-stage HF, initially as bridge to transplantation (BTT), but more recently as destination therapy (DT).
MCS = mechanical circulatory support.
End-stage heart failure
For selected patients with end-stage HF, transplantation remains the gold-standard treatment, with good long-term survival. However, because of the increasing numbers of patients with end-stage HF, limited organ donation, and technological advances, MCS with an LV assist device (LVAD) or bi-ventricular assist device (BiVAD) is increasingly seen as an alternative for some of these individuals. Initially MCS was used as a short-term BTT treatment (Table 24), but is now being used long-term, as so-called ‘destination therapy (DT)’, in patients not eligible for transplantation. Ventricular assist devices may ultimately become a more general alternative to transplantation, as current 2- to 3-year survival rates in carefully selected patients receiving the latest continuous flow devices are much better than with medical therapy only.254,255 Patients receiving these devices also have a post-transplant survival rate similar to those not requiring bridging. However, despite technological improvements, bleeding, thrombo-embolism (both of which can cause stroke), infection, and device failure remain significant problems; these issues, plus the high cost of devices and implantation, have limited their wider use. It is recommended that such devices are only implanted and managed at tertiary heart failure centres with appropriately trained, specialist HF physicians and surgeons. Ideally these centres should also undertake transplantation.
In some patients, LV reverse remodelling and functional improvement during MCS permit removal of the ventricular assist devices (‘bridge-to-recovery, BTR)’. This outcome may occur in some patients with non-ischaemic cardiomyopathy, but is more likely in patients with an acute fulminant, but reversible, cause of HF such as acute myocarditis.256 Another concept is using MCS to permit recovery of end-organ dysfunction, so-called ‘bridge to candidacy (BTC)’, which may allow ineligible patients to become eligible for transplantation. The difficult decision to withdraw MCS may need to be made if the patient does not become eligible and DT is not possible.
Typically, patients with end-stage HF considered for MCS are on continuous inotropic support (Table 25).211,253,257 Evaluation of right ventricular function is crucial as post-operative right ventricular failure greatly increases perioperative mortality and reduces survival to, and after, transplantation. Consequently, BiVAD, rather than LVAD, support should be considered for BTT in patients with biventricular failure or at high risk of developing right ventricular failure after LVAD implantation. Referral before right ventricular failure develops is preferable. Indeed, earlier ventricular assist device implantation in less severely ill patients (e.g. with an EF <25%, peak oxygen consumption <12 mL/kg/min, and only requiring intermittent inotropic support), and before right ventricular or multiorgan failure develops, leads to better surgical outcomes.
CI = cardiac index; HF = heart failure; i.v. = intravenous; LVEF = left ventricular ejection fraction; PCWP = pulmonary capillary wedge pressure; SBP = systolic blood pressure.
Patients with active infection, severe renal, pulmonary, or hepatic dysfunction, or uncertain neurological status after cardiac arrest or due to cardiogenic shock are not usually candidates for BTT or DT, but may be candidates for BTC48.
BiVAD = bi-ventricular assist device; HF = heart failure; LVAD = left ventricular assist device.
aClass of recommendation.
bLevel of evidence.
dSee text and Table 25.
Acute heart failure
In addition to ventricular assist devices, other forms of short-term, temporary MCS may be used in selected patients with AHF, including intra-aortic balloon counterpulsation, other percutaneous cardiac support, and ECMO. In addition to the uses described above, MCS, particularly ECMO, can be used as a ‘bridge to decision (BTD)’ in patients with acute and rapidly deteriorating HF where full evaluation has not been possible and in whom death will occur without MCS. However, the difficult decision to withdraw MCS may need to be made if the patient is not eligible for conventional corrective surgery or longer term MCS.
Holistic management, including exercise training and multidisciplinary management programmes, patient monitoring, and palliative care
Non-pharmacological non-device/surgical interventions used in the management of HF (both HF-REF and HF-PEF) are summarized in Tables 26 and 27, and detailed practical recommendations on their use have been published by the Heart Failure Association.259 There is no evidence that most of these improve mortality or morbidity, and some long-cherished approaches may not be beneficial, e.g. advice to restrict sodium intake and self-management counselling.260,261 For this reason, these interventions have not been given a recommendation with an evidence level. The exceptions are implementation of care in a multidisciplinary framework and exercise training, both of which are discussed further below49.
aClass of recommendation.
bLevel of evidence.
Several systematic reviews and meta-analyses of small studies have shown that physical conditioning by exercise training improves exercise tolerance, health-related quality of life, and HF hospitalization rates in patients with HF. Recently, a single large RCT [Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION)] investigated the effects of exercise training in 2331 relatively young (mean age 59 years) medically stable patients with mild to moderately severe symptoms (NYHA class II 63% and class III 35%) and an EF ≤35%.262 The intervention comprised 36 supervised sessions in the initial 3 months followed by home-based training. The median follow-up was 30 months. In an adjusted analysis, exercise training led to an 11% reduction in the primary composite outcome of all-cause mortality or all-cause hospitalization (unadjusted P = 0.13; adjusted P = 0.03). There was also a 15% RRR in a secondary composite outcome of cardiovascular death or HF hospitalization (unadjusted P = 0.06; adjusted P = 0.03). There was no reduction in mortality, and no safety concerns were raised. Adherence to exercise declined substantially after the period of supervised training.
Collectively, the evidence suggests that physical training is beneficial in HF, although typical elderly patients were not enrolled in many studies and the optimum exercise ‘prescription’ is uncertain. Furthermore, the single large trial showed a borderline treatment effect that was only obtained with a very intensive intervention that may not be practical to deliver in every centre. Exercise training is discussed in more detail in a recent Heart Failure Association consensus paper.263
Organization of care and multidisciplinary management programmes
The goal of management of HF is to provide a ‘seamless’ system of care, embracing both the community and hospital, to ensure that the management of every patient is optimal, from the beginning to the end of their healthcare journey. The standards of care that patients with HF should expect have been published by the Heart Failure Association.236 To achieve this goal, other services, such as cardiac rehabilitation and palliative care, must be integrated into the overall provision for patients with HF. Fundamental to the delivery of this complete package of care are multidisciplinary management programmes designed to improve outcomes through structured follow-up with patient education, optimization of medical treatment, psychosocial support, and improved access to care.264 Key to the success of these programmes is coordination of care along the continuum of HF and throughout the chain-of-care delivered by the various services within the healthcare system. This necessitates close collaboration between HF practitioners (cardiologists and HF nurses) and experts in allied health professions, including pharmacists, dieticians, physiotherapists, psychologists, primary care providers, and social workers. Although the content and structure of HF management programmes may vary in different countries and healthcare settings, the components shown in Tables 26 and 27 are recommended.
Serial natriuretic peptide measurement
High natriuretic peptide concentrations are associated with a poor prognosis, and a fall in peptide levels correlates with a better prognosis. However, several RCTs that evaluated natriuretic peptide-guided treatment (intensifying treatment in order to lower peptide levels) have given conflicting results.265 It is uncertain whether outcome is better using this approach than by simply optimizing treatment (combinations and doses of drugs, devices) according to guidelines.
Remote monitoring (using an implanted device)
Management adapted in response to monitoring thoracic impedance (as an indirect measure of intrathoracic fluid) has not been shown to improve outcomes.266 Treatment adjusted in response to pulmonary artery pressure measured using an implanted monitor did reduce hospital admission for HF in one RCT,267 but the general applicability of this approach is uncertain and a guideline recommendation is not yet possible.
Remote monitoring (no implanted device)
The optimum approach to non-invasive remote monitoring is uncertain, and RCTs performed to date have given inconsistent results and do not yet support a guideline recommendation.268
Structured telephone support
Although a meta-analysis of RCTs suggests that structured telephone support in addition to conventional care may reduce the risk of hospitalization in patients with HF, few individual RCTs showed this benefit, and the evidence is not robust enough to support a guideline recommendation.268,269
HF has an unpredictable disease trajectory and it is often difficult to identify a specific time point to consider palliative care. Features that should trigger consideration of palliative care are listed in Tables 28 and 29. At this point in a patient's disease trajectory, the focus should be on improvement in quality of life, control of symptoms, early detection, and treatment of episodes of deterioration, and on pursuing a holistic approach to patient care, encompassing physical, psychological, social, and spiritual well-being. Liaison between the specialist palliative care service and the HF team and/or the primary care physician, using a shared-care approach, is required in order to address and coordinate the patients' care optimally. Palliative care has been discussed in detail in a position paper from the Heart Failure Association.270
NYHA = New York Heart Association.
HF = heart failure; ICD = implantable cardioverter-defibrillator.
Gaps in evidence
Clinicians responsible for managing patients with HF must frequently make treatment decisions without adequate evidence or a consensus of expert opinion. The following is a shortlist of selected, common issues that deserve to be addressed in future clinical research.
The diagnosis of HF-PEF remains a particular challenge, and the optimum approach incorporating symptoms, signs, imaging, biomarkers, and other investigations is uncertain.
Strain/speckle imaging—value in diagnostic and prognostic assessment of both HF-REF and HF-PEF?
Diastolic stress test—value in diagnosis of HF-PEF?
The long-term safety and efficacy of many treatments for co-morbidities are unknown, but are of great interest and importance.
Anaemia—erythropoiesis-stimulating agents, iron?
Depression—selective serotonin reuptake inhibitors, cognitive therapy?
Diabetes—metformin, GLP-1 agonists/analogues, DPP IV inhibitors, SGLT-2 inhibitors?
Sleep-disordered breathing—positive airways pressure therapies?
Non-pharmacological, non-interventional therapy
Salt restriction—is it effective and safe?
Cardiac cachexia—is there an effective and safe treatment?
Digoxin—efficacy and safety in modern era of pharmacological and device therapy?
Hydralazine and ISDN—efficacy and safety in non-black patients?
Renin inhibition—is it an effective and safe alternative to/addition to ACE inhibition?
New oral anticoagulants—efficacy and safety compared with aspirin in patients in sinus rhythm?
Clopidogrel and other novel antiplatelet agents—efficacy and safety compared with aspirin in patients in sinus rhythm?
Dual neprilysin/angiotensin receptor inhibitors—efficacy and safety compared with an ACE inhibitor?
CRT—the efficacy and safety of CRT remains unknown in certain groups of patients.
patients with a normal QRS duration but echocardiographic dyssynchrony?
patients with RBBB and IVCD?
patients in AF?
LVADs—the long-term efficacy and safety of LVADs as an alternative to heart transplantation or medical therapy remains uncertain
Remote monitoring—the long-term efficacy and safety of the various remote monitoring strategies available remain uncertain
Acute heart failure
The treatment of acute heart failure remains largely opinion-based with little good evidence to guide therapy.
Intravenous nitrates—efficacy and safety still uncertain.
Levosimendan—efficacy and safety still uncertain.
Omecamtiv mecarbil—is it effective and safe?
Ultrafiltration—efficacy and safety unknown.
What is the optimum palliative care package?
When should palliative care be started?