Addressing heart failure in primary care:
Part 1 - Identifying and diagnosing heart failure

Key practice points:

• The presentation of people with heart failure in primary care can vary substantially, ranging from mild and non-specific symptoms (e.g. a reduced exercise capacity and malaise), through to those with classical key features (e.g. ankle swelling, shortness of breath, orthopnoea)
• Clinical examination can help to identify more specific signs of heart failure, e.g. elevated jugular venous pressure (JVP), positive abdominojugular reflux, S₃ (gallop rhythm) and a laterally displaced apical impulse; however, the absence of these features does not exclude the possibility of heart failure
• The patient’s history should be reviewed to:
  • Identify whether they have made behavioural changes to compensate for symptoms, e.g. reducing physical activity in response to shortness of breath
  • Assess for other factors that may be an underlying cause or exacerbate symptoms, e.g. co-morbidities (such as arrhythmias or ischaemia) or concomitant medicine use
• If heart failure is still suspected, perform an ECG to check for any obvious underlying abnormalities (e.g. atrial fibrillation, ischaemia) and request a brain natriuretic peptide (BNP) level. Additional investigations can then be considered depending on patient-specific factors, e.g. chest X-ray may be useful to confirm fluid overload and to exclude other causes of dyspnoea.
• A clinical diagnosis can be made if patients have symptoms/signs of heart failure and are not “ruled-out” based on having a low BNP result. Referral for an echocardiogram helps to refine long-term treatment decisions, however, it is not required to make an initial diagnosis.
• Echocardiography can distinguish the type of heart failure, and is particularly useful for identifying associated cause(s); this in turn can guide some treatments (e.g. medical, surgical and devices). The three classifications of heart failure are:
  • Heart failure with reduced ejection fraction (HFrEF) – patients with reduced left ventricular ejection fraction (LVEF) of ≤ 40%
  • Heart failure with mildly reduced (or “mid-range”) ejection fraction (HFmrEF) – patients with a LVEF of 41 – 49%
  • Heart failure with preserved ejection fraction (HFpEF) – patients with a LVEF of ≥ 50%, and evidence of relevant structural heart disease, and/or diastolic dysfunction with a high filling pressure
• If the patient has significant or distressing symptoms at presentation, or an acute cardiac condition is suspected to be causing heart failure (e.g. acute coronary syndrome), consider urgent secondary care referral where further management decisions will be made. Treatment for patients initially managed in primary care is guided by the severity of symptoms, the presence and type of co-morbidities and relevant laboratory investigations.

The burden of heart failure in New Zealand

Heart failure affects 2 – 3% of all adults in New Zealand, and ≥ 10% of those aged over 70 years.^{1, 2} The associated burden is projected to increase over time due to an ageing population as well as increased prevalence of risk factors, e.g. obesity, sedentary lifestyles, type 2 diabetes and ischaemic heart disease.³ An estimated one in five residents in aged residential care facilities have heart failure,⁴ but it is also becoming more common in younger age groups.¹

Māori and Pacific peoples in particular are disproportionately affected by heart failure. Onset typically occurs at a younger age and rates of hospitalisation and mortality are substantially increased in Māori compared with non-Māori (at least four- and two-fold higher, respectively).^{5, 6} These disparities have widened over time. Between 2006 – 2018, hospitalisation rates for heart failure decreased among older Europeans, but rates remained unchanged among Māori and Pacific peoples.⁵ Māori and Pacific peoples aged < 50 years are six times more likely to be hospitalised for heart failure than Europeans in the same age group.⁵

Most patients with heart failure require secondary care input at some point. The five-year survival rate is < 50% following diagnosis,³ and < 37% following first hospitalisation.⁷ This prognostic outlook is improving with more widespread use of guideline-directed medical therapy (see: “Part 2 - Initiating and optimising treatment for heart failure”), but remains worse than for many common cancers. As such, there is a clear need for clinicians to identify potential cases as early as possible, so that effective pharmacological treatment can be initiated. However, making a diagnosis of heart failure in primary care can be challenging; it is largely a clinical diagnosis relying on presenting features in the context of the patient’s history, initially supported by laboratory markers (see: “Request brain natriuretic peptide (BNP) level”), and later by echocardiography findings (see: “Refer for an echocardiogram but do not delay treatment”).

Identifying patients with potential heart failure

Identifying heart failure in primary care can be difficult as presentation varies between patients due to differences in the underlying pathology and compensatory mechanisms (Table 1).^{3, 8} In addition, symptoms and signs may only be mild at first, have a gradual onset or be difficult to interpret, e.g. in patients with multiple co-morbidities. As a result, recognising heart failure often relies on the clinician determining whether the combination of symptoms and signs a patient is presenting with warrants further investigation based on their clinical history (see: “Piecing together the puzzle: patient history”).

Non-specific symptoms such as dyspnoea, fatigue or reduced exercise tolerance may not immediately raise suspicion of heart failure as these can be associated with other conditions, e.g. COPD. Heart failure becomes a more likely diagnosis if these symptoms and signs occur with other features of fluid overload, such as lower limb swelling, rapid weight gain or specific types of dyspnoea, e.g. in a recumbent position (orthopnoea) or that awakens a patient from sleep (paroxysmal nocturnal dyspnoea).^{8, 9} More specific signs may be identified during clinical examination, e.g. elevated jugular venous pressure (JVP), positive abdominojugular reflux, S₃ (gallop rhythm) and a laterally displaced apical impulse.^{8, 9} However, these features can be difficult to detect accurately, and their absence does not exclude the possibility of heart failure.⁹

Table 1. Symptoms and signs of heart failure.^{3, 8, 9}

Piecing together the puzzle: patient history

Considering the patient’s history provides context for the presenting features if heart failure is suspected and clues regarding the likely cause(s).

Ask about behavioural or lifestyle changes that have been made to avoid symptoms. Emerging features of heart failure can be subtle or unintentionally disguised by patients making behavioural or lifestyle changes to compensate, e.g. limiting physical activity to avoid shortness of breath and therefore dyspnoea is not reported as a symptom.⁸ Patients may not be aware that they have made these changes until they are specifically asked about them.

Identify contributing factors and co-morbidities that are inadequately controlled. In some cases, co-morbidities may be the underlying cause of heart failure; their presence can add to clinical confidence regarding the diagnosis, and if their management can be optimised then the risk of acute decompensation (new or worsening symptoms and signs) may reduce.⁸ For example, at least 90% of people with heart failure have a history of hypertension and approximately 40% have diabetes.³

Contributing factors and conditions associated with heart failure include:^{3, 8, 10}

• Hypertension
• Ischaemic heart disease
• Valvular disease
• Cardiomyopathies, e.g. diabetic, hypertrophic, alcohol-related, genetic*
• Arrhythmias, e.g. atrial fibrillation
• Obesity
• Thyrotoxicosis (causing high output failure)
• Chronic kidney disease or worsening renal function
• COPD
• Obstructive sleep apnoea
• Anaemia
• Smoking
• Infection (leading to myocarditis or cardiomyopathy)
• Drug-induced, e.g. prior use of cardiotoxic chemotherapy, methamphetamine

* An estimated 25 – 40% of people with dilated cardiomyopathy and a positive family history have an underlying genetic cause¹⁰

Consider current medicine use and whether adjustments can be made. Given that most patients with heart failure have co-morbidities, it is likely they are already taking other medicines. Polypharmacy is common in this context, and is associated with increased mortality risk.¹¹ Numerous medicines can influence cardiac functioning – these may increase the risk of heart failure developing, or worsen outcomes for patients with established disease.⁸ Examples include:^{8, 10, 12}

• NSAIDs – can cause renal impairment and sodium and water retention, increased vasoconstriction, and an impaired response to diuretics/angiotensin-converting enzyme (ACE) inhibitors
• Corticosteroids – can cause sodium and water retention
• Thiazolidinediones, e.g. pioglitazone – can cause dose related fluid retention; avoid use in patients with heart failure
• Most calcium channel blockers (with the exception of amlodipine and felodipine) – can cause negative inotropic effects such as weakening cardiac muscle contraction and slowing heart rate
• Antiarrhythmic medicines, e.g. sotalol and disopyramide can prolong the QT interval, flecainide can cause negative inotropic effects
• Some antidepressants, e.g. tricyclic antidepressants, citalopram, venlafaxine – can cause a range of adverse cardiovascular effects, including bradycardia, tachycardia, hypertension

Red flags for more urgent referral

As a chronic and progressive condition, most patients with heart failure are seen in secondary care at some stage – whether it be for echocardiography to refine primary care management or for cardiologist review. However, people presenting to primary care with the following features should be more urgently referred (or immediately sent to the emergency department):^{8, 9}

• Significant or distressing symptoms of orthopnoea, paroxysmal nocturnal dyspnoea, syncope, ischaemic chest pain
• An acute cardiac condition is suspected to be the cause of heart failure, including fast atrial fibrillation or other sustained arrythmias (e.g. tachycardia, bradycardia), a significant new heart murmur or acute coronary syndrome
• Newly diagnosed or worsening heart failure during pregnancy

Delving deeper to make a clinical diagnosis

Perform an ECG

When heart failure is suspected, perform an ECG to identify any underlying cardiac abnormalities (e.g. left bundle block, poor R wave progression, ischaemic changes, atrial fibrillation) or other rhythm disturbances as these are common causes of heart failure.^{8, 10} Ventricular arrhythmias are responsible for a high proportion of sudden deaths in people with heart failure and the presence of significant dysfunction may influence the urgency of secondary care involvement.¹⁰

Request brain natriuretic peptide (BNP) level

Laboratory assessment of BNP levels should be requested for any patient with a pattern of symptoms and signs indicating possible heart failure. BNP is a hormone produced by cardiomyocytes in response to increased ventricle wall tension, which is associated with natriuretic, diuretic and vasodilatory effects (thereby countering the effects of sympathetic nervous system and RAAS activation).^{15, 16} Following synthesis, the BNP prohormone is cleaved and secreted into the blood in equimolar amounts (1:1 ratio), including:^{15, 16}

1. The biologically active BNP-32 hormone; and
2. The inactive NT-proBNP portion

Both BNP biomarkers can be used as a proxy for how “stressed” the heart is; the higher the levels, the more likely a diagnosis of heart failure.⁸ However, NT-proBNP testing has several advantages over BNP-32, including a longer half-life, greater stability and levels that are not influenced by heart failure medicines (e.g. the neprilysin inhibitor sacubitril prevents BNP-32 degradation but does not significantly influence NT-proBNP).¹⁶ In most cases, clinicians in New Zealand cannot specify the type of BNP test on a laboratory form, and there is regional variation in which test will be done (i.e. BNP-32 or NT-proBNP).

Interpreting BNP results

BNP testing is considered primarily to be a “rule out” test, as other conditions can also affect levels.^{8, 10} For example, increased BNP levels may be present in people with atrial fibrillation, COPD, severe renal impairment, post-myocardial infarction, or in those with left ventricular hypertrophy.^{8, 10, 15} Decreased BNP levels may occur in people with obesity, hypothyroidism or who are taking certain medicines (such as diuretics, vasodilators or ACE inhibitors).^{8, 10, 15} N.B. A decreased BNP in patients with heart failure taking diuretics or other heart failure medicines usually correlates with a favourable change in volume status. However, serial BNP monitoring is not routinely recommended post-diagnosis unless it will affect management decisions (see: “Part 2 - Initiating and optimising treatment for heart failure”).

For patients with “intermediate” BNP readings, consider these potentially confounding factors within the context of the full clinical picture to determine whether a diagnosis of heart failure should be made.

Only at very high levels is BNP considered to be a “rule in” test for heart failure and for NT-proBNP this threshold is age dependent. Table 2 outlines the diagnostic thresholds used in 2018 Australasian guidelines, however, “normal ranges” are generally provided alongside laboratory results. Many laboratories in New Zealand report BNP results in units of pmol/L, but pg/mL (or ng/L) is more commonly used in international literature.

Table 2. General BNP diagnostic thresholds for heart failure based on Australasian guidelines and used in most New Zealand laboratories.^*† N.B. Diagnostic thresholds may differ between different international guidelines and laboratories.

* The type of test available and the reference range for BNP may differ between laboratories or regions. If the BNP reference range is not provided by your local laboratory, consider discussing the findings with a clinical biochemist.

† To covert BNP from pmol/L to pg/mL, multiply by 3.47; to convert NT-ProBNP from pmol/L to pg/mL, multiply by 8.46

Additional tests to consider

BNP is the primary biomarker to assess the likelihood of heart failure, however, additional laboratory tests should also be requested to help clarify the diagnostic picture and guide treatment decisions, e.g. full blood count, electrolytes, renal and liver function testing.^{8, 10} In many cases, it is also reasonable to assess HbA_1c and lipid levels as part of a routine cardiovascular disease risk evaluation.^{8, 10}

Depending on the specific patient, other tests can also be considered ^8–10

Refer for an echocardiogram but do not delay treatment

Echocardiography is considered the “gold standard” test for supporting a heart failure diagnosis. However, wait times vary substantially across New Zealand, and early intervention is essential to improve prognostic outcomes.⁸

Therefore, heart failure should be diagnosed clinically first (i.e. based on the history, physical examination, ECG and BNP findings), and treatment initiated immediately under the assumption that the patient has HFrEF (see: “Defining heart failure” and “Part 2 - Initiating and optimising treatment for heart failure“). Long-term management decisions can then be refined once the echocardiogram result is available, if necessary. For example, the echocardiogram findings may reveal abnormalities in the LVEF, left ventricular hypertrophy, systolic and diastolic function, elevated right heart pressures, or evidence of ischaemic damage.^{9, 20}

Keep reading: Part 2 – Initiating and optimising treatment for heart failure