Prevalence and incidence of heart failure

Studies including objective assessment of left ventricular (LV) function, usually echocardiography, indicate a prevalence of left ventricular systolic dysfunction (LVSD) of 2.9% in patients under 754 and up to 7.5% in 75–84 year olds.^w3 In the largest recent prospective evaluation of heart failure in the community, definite heart failure was found in 2.3% (95% confidence interval (CI) 1.9% to 2.8%) of the population, with left ventricular ejection fraction (LVEF) <40% in 41% of cases. However, if the LVEF cut-off was set at under 50% rather than 40%, now advocated in some guidelines, 3.1% (95% CI 2.6% to 3.7%) of people aged 45 or over were defined as having heart failure.5

Estimates on heart failure incidence are less available, and vary from 0.9–2.26 per 1000 females (age 45–74) per year and 1.6–4.6 men per 1000 population (age 45–74) per year. Incidence rises rapidly in the elderly, with 1% of men per year developing heart failure after 75 and almost 2% per year in the over 85s.

The typical primary care physician, caring for 2000 patients, is likely to have 40–50 patients with heart failure (more if the population is elderly), and to have five or so new cases of heart failure diagnosed each year, with the diagnosis being suspected in perhaps three times as many. The syndrome is, therefore, relatively common but not so frequent that the individual practitioner will become expert in diagnosis. Clear guidance is needed on which clinical features and combination of investigations is best to rule out or rule in heart failure in patients with new symptoms.

Burden of heart failure on healthcare systems

Chronic heart failure remains one of the most costly conditions to manage: the syndrome is common, it frequently results in hospital admission, admissions are long, and re-admission is frequent, particularly where chronic disease management programmes are not running (nearly 25% of patients are readmitted within 12 weeks of discharge).7 As a consequence, heart failure accounts for around 2% of total healthcare expenditure in most developed countries.

Good diagnosis is the cornerstone of good management

There is a large evidence base for treatments that improve morbidity, global quality of life, and mortality in all grades of heart failure due to LVSD,8 9 and can delay LVSD progression to symptomatic heart failure.^w4 Despite this extensive evidence, heart failure remains suboptimally diagnosed and treated in many countries.^w5 An essential element for treatment success is the reliable and precise diagnosis of heart failure.

Diagnostic issues in heart failure

The evaluation of patients with suspected heart failure entails determining more than just whether the syndrome is present or not. It is important to identify the underlying cardiac pathologies, exacerbating or precipitating factors, and the whole body's response to the cardiac dysfunction—including particularly renal function. Guidelines for the evaluation and management of heart failure have been published by professional societies in North America (American College of Cardiology/American Heart Association (ACC/AHA)10 and Consensus Recommendations^w6) and Europe (European Society of Cardiology).11 The latter state that the diagnosis of heart failure is justified when there are typical signs and symptoms of heart failure and myocardial dysfunction, confirmed by the objective evidence of cardiac dysfunction at rest. In the case of diagnostic uncertainty, a clinical response to treatment directed at heart failure is helpful in establishing the diagnosis (box 1). Simple and reliable diagnostic procedures are very important for primary care physicians, who are responsible for the early diagnosis of heart failure and implementation of adequate treatment in most healthcare systems.

Current NICE guidance on diagnosing heart failure

The National Institute for Health and Clinical Excellence (NICE) guideline on chronic heart failure for primary and secondary care in England recommends that patients with suspected heart failure should have an electrocardiograph (ECG) and/or B-type natriuretic peptide (BNP or N terminal pro-BNP (NT-proBNP)) assay performed.12 If both are normal, then heart failure is unlikely, and an alternative diagnosis to explain the symptoms should be considered. If either is abnormal, then the patient should have an echocardiogram, or other imaging test, to confirm underlying cardiac dysfunction (figure 1). This guidance was based on the high sensitivity of both plasma BNP and the ECG, and the result of a health economic analysis that demonstrated that the cost per life year gained through echocardiography is dependent upon the proportion of patients referred for echocardiography in whom the diagnosis of heart failure is confirmed.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Diagnostic algorithm recommended for use in primary care by the National Institute for Health and Clinical Excellence in England, 2003.12 BNP, B-type natriuretic peptide; FBC, full blood count; LFTs, liver function tests; NTproBNP, N terminal pro-BNP; TFTs, thyroid function tests; U&Es, urea and electrolytes.

However, the current diagnosis of heart failure in primary care is often inaccurate. In one recent UK study, only 34% of patients with an existing clinical label of heart failure in routine general practice records had this diagnosis confirmed at echocardiography and review.13

There have been seven recent systematic reviews relevant to the diagnosis of heart failure in primary care. Three covered all symptoms, signs and diagnostic tests,12 14 ^w7 three concerned BNP alone,15 ^{w8 w9} and one was a review of the accuracy of the 12 lead ECG.16

Symptoms and signs

Individual symptoms (such as breathlessness, fatigue, exercise intolerance and fluid retention) and signs (such as resting tachycardia, raised jugular venous pressure (JVP), displaced apex beat, third heart sound) are generally weak predictors of heart failure, with poor reliability and little agreement between clinicians on their presence or absence, even among specialists.

In a recent systematic review, a number of symptoms were reasonably specific, including a history of myocardial infarction (89% specificity), orthopnoea (89%), cardiomegaly (85%), added heart sounds (99%), lung crepitations (81%), and hepatomegaly (97%) (table 1). However, in primary care, the most useful symptoms/signs in diagnosis have high sensitivity, since this might enable the clinician to rule out heart failure if the symptom/sign was absent, without the need to refer for further investigation. Dyspnoea is the only clinical feature that comes close to this ideal with a sensitivity of 87%. In practice this symptom is present in the majority of patients in whom heart failure is suspected, with a frequency as high as 95% in one of the datasets. Nevertheless, a sensitivity of 87% is not high enough on its own to rule out heart failure if dyspnoea is absent.

Investigations

Both ECG16 ^{w10 w11} and BNP15 ^{w7 w8} have high sensitivity for heart failure, and so are good tests at ruling out the diagnosis (table 2). UK based studies restricted to use of ECG in primary care, however, give a more mixed picture on the value of ECG, with sensitivity in one study as low as 73%.16 These differences may relate to both differences in population characteristics, and to the skill of the practitioner interpreting the ECG. Similarly, while the chest x-ray may show evidence of heart failure (eg, cardiomegaly, pulmonary vascular congestion), it is not a good independent predictor, and is of most value in identifying alternative causes of symptoms such as lung disease or pleural effusion.

In a review of four studies that evaluated the diagnostic accuracy of the resting ECG in the specific context of referral from primary care to echocardiographic services, sensitivity of the ECG varied from 73–91%, and the authors concluded that the ECG was an inadequate screening tool.16 A review of the diagnostic accuracy of natriuretic peptides and ECG in the diagnosis of LVSD found similar diagnostic accuracy between ECG, BNP and NT-proBNP and no value from combining BNP with ECG.17 A recent systematic review confirmed that adding the ECG to clinical features plus the plasma BNP result did not improve the accuracy of diagnosis.14

It is likely that BNP is a more accurate test for heart failure than it is for LVSD. Indeed, a recent systematic review of BNP studies concluded that while BNP is useful for excluding heart failure, it is more limited for ruling out systolic dysfunction, with an area under the curve (AUC) of 0.93 for heart failure but only 0.75 for systolic dysfunction.18 There is no evidence of any significant differences in test performance between BNP and NT-proBNP.17 18 Other recent reviews of BNP have confirmed its value as a ‘rule out’ test for heart failure.^w12–w14

Echocardiography is the most readily available ‘reference standard’ investigation for LVSD and valve disease. Assessment of diastolic dysfunction can also be made on echocardiography,19 but the interpretation of the findings may be difficult, particularly in the elderly and in patients with atrial fibrillation (up to 30% of new cases of heart failure in most series). In practice today, ‘normal ejection fraction’ (‘preserved systolic function’, ‘diastolic’ or ‘non-systolic’) heart failure often remains a diagnosis of exclusion. Echocardiography after a plasma BNP test has been performed is an attractive option, particularly if the supply of echocardiography is limited or access is slow, though appropriate cut points for BNP (or NT-proBNP) are then needed.

Role of clinical decision rules in the work up of patients with suspected heart failure

There are several well developed heart failure diagnostic tools which combine the results of different symptoms/signs and investigations. Mosterd et al applied criteria from six established heart failure scores including Framingham, Walma, and Boston to a sample of 54 participants in the population based Rotterdam study.20 Most showed high utility to detect definite heart failure with areas under the receiver operated characteristic (ROC) curve (AUC) ranging between 0.89–0.96. However, use of these would be impractical in routine primary care because of the substantial number of variables in several scores; and also because many of the clinical signs have considerable inter-observer variation even among specialists (raised jugular venous pressure (JVP), third heart sound, hepatojugular reflux).^{w15 w16} Furthermore, four of the scores include specific chest x-ray parameters difficult to apply in general practice.

A recent meta-analysis using individual patient data identified a simple new diagnostic score. This rule suggests that in a patient presenting with new symptoms such as breathlessness in whom heart failure is suspected, if the patient has any one of (a) history of myocardial infarction or (b) basal crepitations or (c) is a male with ankle oedema (so called MICE score), then it is appropriate to refer straight for echocardiography as the probability of heart failure is high. Otherwise, a plasma BNP test should be performed, and referral to echocardio-graphy should be organised depending on the concentration of plasma BNP (box 2).14

Limitations of the evidence base

There is no single ideal reference standard for heart failure, since there is no single cardiac disorder that accounts for the syndrome. The underlying cardiac disorders can be classified in different ways. The standard approach is to divide heart failure into low ejection fraction and normal ejection fraction heart failure, if significant valve disease has been excluded as the primary cause of the heart failure. Echocardiography is a suitable reference standard for low ejection fraction heart failure, but is not as robust for normal ejection fraction heart failure. The definitive tests to diagnose normal ejection fraction heart failure (pressure–volume measurements at cardiac catheterisation) are often not carried out, so the diagnosis usually relies upon clinical judgement and supportive evidence.

Studies that have tested the value of the ECG in the diagnosis of heart failure have used different criteria with which to define abnormality, and there has been variation in the experience and expertise of the person reading the ECG. Many general practitioners are unable to interpret ECGs accurately.

For natriuretic peptides, most of the existing research has been done on secondary care populations, or in the context of screening studies that identify prevalent cases of heart failure, or include patients with existing diagnoses of heart failure. This may introduce significant spectrum bias and affect whether the cut-off values are reliable for the target population, namely symptomatic patients presenting in primary care.