Section 1 – Mark

Mark, a 59-year-old with type 2 diabetes for 7 years, came to see his GP reporting increased breathlessness on exercise, ankle swelling and general fatigue (but no chest pain) over the previous 2 months. Examination revealed pulse 78 beats/minute regular, blood pressure 132/67 mmHg, heart sounds normal with no added sounds or murmurs, bilateral inspiratory crackles on the lung bases, and bilateral ankle oedema.

What do you think is the most likely explanation for Mark’s symptoms?

Section 2

Mark has clinical features consistent with heart failure (see Table 1). Whilst there are various underlying causes of heart failure, diabetic cardiomyopathy (secondary to hypertension, dyslipidaemia and hyperglycaemia) must be a major consideration. People with diabetes are around 2–3 times more likely to develop heart failure than those without diabetes and, furthermore, this carries a worse prognosis (Lehrke and Marx, 2017; Pop-Busui et al, 2022).

Recent investigations: HbA_1c 67 mmol/mol, eGFR 72 mL/min/1.73 m² with electrolytes in the normal range, total cholesterol 3.9 mmol/L, non-HDL cholesterol 2.3 mmol/L, normal LFTs and TFTs, ACR 2.3 mg/mmol. Random fingerprick blood glucose readings in the range of 6–14 mmol/L.

Current medications: Metformin 1 g twice daily, gliclazide 160 mg twice daily, atorvastatin 20 mg once daily, ramipril 10 mg once daily.

Mark is known to have background diabetic retinopathy. He works as a car mechanic and is a driver himself. He is an ex-smoker and consumes around 28 units of alcohol per week.
 
What further investigations available to primary care might be helpful in establishing a diagnosis of heart failure and underlying causes of this pathology?

Section 3

A key diagnostic investigation is measurement of N-terminal-pro-B-type natriuretic peptide (NT-proBNP). Raised levels of this polypeptide are a biomarker for heart failure and cardiac dysfunction, whilst levels in the normal range make the diagnosis of heart failure very unlikely and another cause of symptoms should be sought (Cao et al, 2019; Pop-Busui et al, 2022). The greater the value of NT-proBNP, the more likely heart failure is to be present and the more severe it is likely to be.

An ECG and chest x-ray are important core investigations given Mark’s clinical picture.
 
How might you help improve Mark’s symptoms whilst awaiting the results of investigations?

Section 4

It would be reasonable to commence a loop diuretic to deal with Mark’s symptoms and signs of fluid overload. The importance of lifestyle factors in type 2 diabetes – diet (including salt restriction), moderation of alcohol consumption, exercise and weight loss – could be usefully revisited (NICE, 2015). Mark’s hypertension and dyslipidaemia appear to be satisfactorily controlled.

Mark was commenced on furosemide 40 mg once daily.

Subsequently results showed: moderately raised NT-proBNP levels (991 ng/L); normal renal function and electrolytes; ECG: sinus rhythm with left ventricular hypertrophy; and pulmonary congestion on chest x-ray.
 
How would you interpret these results?

Section 5

The raised NT-proBNP level, ECG and chest x-ray results support the diagnosis of heart failure (and exclude certain causes such as cardiac arrhythmia and severe lung disease).
 
What action would you now take to enable a specific diagnosis of heart failure?

Section 6

At this stage, Mark should be referred to the cardiologist for further assessment. Transthoracic echocardiography is a key investigation and, depending on local arrangements, it may be possible to simultaneously request this. The aim here is to ascertain the degree and type of heart failure, including any underlying causes, which will direct management.

It is useful to divide heart failure into three categories: heart failure with reduced ejection fraction (HFrEF), in which the left ventricular ejection fraction (LVEF: the percentage of blood pumped out of the left ventricle per beat) is <40%; heart failure with mildly reduced ejection fraction (HFmrEF), in which LVEF is between 41% and 49%; and heart failure with preserved ejection fraction (HFpEF; sometimes referred to as diastolic heart failure), in which the LVEF is ≥50%. Evidence-based treatments for HFrEF are well established but, until recently, this had not been the case for HFpEF. Nearly half of people with heart failure have preserved LVEF on echocardiography (NICE, 2018).
 
Can you suggest a treatment that would improve glycaemic control and be supportive of the clinical diagnosis of heart failure whilst Mark awaits his appointment with the cardiologist?

Section 7

There is now strong evidence that the SGLT2 inhibitors dapagliflozin and empagliflozin can improve prognosis (preventing worsening of heart failure and cardiovascular death) of both HFrEF and HFpEF. This is true both for people with and without type 2 diabetes. It is also important to realise that, whilst the improvements in glycaemic control and weight with SGLT2 inhibitors fall off with declining renal function, the heart failure benefits persist at low eGFR (Morris, 2022; Fernando, 2022). If you would like more details and references regarding this evidence, please see Section 8.

The positioning of SGLT2 inhibitors in people with type 2 diabetes and heart failure is now embedded in both NICE and ADA/EASD guidelines (NICE, 2015; Davies et al, 2022). Dapagliflozin and empagliflozin hold a license for HFrEF and HFpEF irrespective of diabetes status. Dapagliflozin can be commenced for heart failure treatment with an eGFR of 15 mL/min/1.73 m² or above, with no lower eGFR limit for continuation. Empagliflozin is licensed for use in heart failure at an eGFR of 20 mL/min/1.73 m² or above (stop treatment if eGFR falls below this value).

A summary of the SGLT2 inhibitor licences and indications can be found in our Need to know guide.

Mark was commenced on dapagliflozin 10 mg once daily. As a precaution against developing hypoglycaemia, Mark’s dose of gliclazide was reduced to 80 mg twice daily and he was advised to monitor his glucose levels more regularly. The furosemide had significantly reduced Mark’s ankle swelling and, because of the diuretic properties of dapagliflozin, the dose of furosemide was reduced to 20 mg once daily to counter the risk of volume depletion (if ankle swelling had persisted then the dose of furosemide probably would not have been reduced), and Mark was warned to be alert for symptoms of postural hypotension.

Section 8: Evidence supporting the use of SGLT2 inhibitors in heart failure

Improvement in heart failure outcomes was a consistent finding in the cardiovascular outcome trials of the SGLT2 inhibitors, although these were generally secondary outcomes. More recently, trials have been directed towards heart failure as the primary outcome.

DAPA-HF was a randomised controlled study comparing dapagliflozin to placebo in individuals with HFrEF in people with and without type 2 diabetes (McMurray et al, 2019). The primary outcome was a composite of cardiovascular death and worsening heart failure (defined as hospitalisation or an urgent visit requiring intravenous therapy for heart failure). Over a median of 18 months, individuals receiving dapagliflozin had a 4.9% absolute risk reduction in the primary outcome compared with placebo, equating to a 26% relative risk reduction. These benefits extended to people irrespective of their diabetes status, and whether or not baseline eGFR was above or below 60 mL/min/1.73 m². Importantly, the treatment group was already receiving largely optimised medication for heart failure (ACE inhibitors, ARBs, beta-blockers, mineralocorticoid receptor antagonists [MRAs] and diuretics) at baseline.

The EMPEROR-Reduced study similarly demonstrated a significant reduction in the primary outcome of hospitalisation for heart failure (HHF) and cardiovascular death for empagliflozin versus placebo in people with HFrEF, with or without diabetes. Again, the positive outcome was largely driven by the reduction in HHF (Packer et al, 2020).

The EMPEROR-Preserved trial investigated the effect of empagliflozin in individuals with HFpEF. Compared to placebo, empagliflozin significantly reduced the composite primary outcome of HHF and cardiovascular death, irrespective of diabetes status (Anker et al, 2021).

In the DELIVER study of people with mildly reduced or preserved ejection fraction, dapagliflozin also improved the same primary outcome, principally due to a reduction in worsening heart failure. Again, benefit was seen whether or not the subjects had type 2 diabetes or not (Solomon et al, 2022).

Section 9

Mark’s echocardiogram identified a reduced left ventricular ejection fraction consistent with HFrEF.
 
What medications other than SGLT2 inhibitors could Mark take to improve prognosis in HFrEF?

Section 10

ACE inhibitors (e.g. ramipril) or ARBs (e.g. losartan) are well established as treatments that improve prognosis in HFrEF. They are also preferred treatments for hypertension in people with diabetes, and should be titrated to the highest tolerated dose (NICE, 2018). This can be escalated to sacubitril/valsartan (a combined neprilysin inhibitor and ARB), which has demonstrated superiority over angiotensin-based treatment alone in reducing HHF/cardiovascular death, with or without diabetes (Pop-Busui et al, 2022; Brown, 2022). NICE has advised that sacubitril/valsartan should be initiated by a consultant cardiologist (NICE, 2016).

Robust evidence supports the use of beta-blockers (e.g. bisoprolol, metoprolol, carvedilol) in those with HFrEF, and they should be routinely employed in this condition (NICE, 2018).

A further evidence-based option that improves prognosis in HFrEF, if required, is a steroidal mineralocorticoid receptor antagonist (MRA), such as spironolactone or eplerenone (NICE, 2018).

Mark was already taking ramipril and dapagliflozin, which will support his HFrEF. Subsequently, bisoprolol was added to his regimen by the cardiologists, with the future possibility of adding an MRA.
 
Aside from SGLT2 inhibitors, what do you think about the safety of other glucose-lowering medications for use in people with type 2 diabetes who have heart failure?

Section 11

Pioglitazone is contraindicated in heart failure and should be used cautiously in those at risk of the condition, including the elderly (Dormandy et al, 2005). The combination of pioglitazone and insulin is more likely to lead to fluid retention and precipitate heart failure (Electronic Medicines Compendium, 2023).

The cardiovascular outcomes trial (CVOT) of the DPP-4 inhibitor saxagliptin (Scirica et al, 2013) showed a significantly increased risk of heart failure, so this agent should be avoided. Similarly, as a non-significant increase in heart failure was seen in the alogliptin CVOT (White et al, 2013), it would be prudent to also avoid this agent in people with heart failure. In general, although signals for heart failure were not identified in the sitagliptin and linagliptin CVOTs, it would be prudent to use other DPP-4 inhibitors cautiously in people with heart failure. The ADA Consensus Report does not recommend using DPP-4 inhibitors in people with heart failure (Pop-Busui et al, 2022).

Both metformin and GLP-1 receptor agonists (GLP-1 RAs) are safe to use in heart failure. Although they do not appear to offer any specific benefit to heart failure, they can provide useful cardiovascular protection (and in the case of GLP-1 RAs, substantial weight loss), and thus are attractive agents for improving glycaemic control in those with heart failure.

Sulfonylureas and insulin can be used to improve glycaemic control but bring no additional benefit to people with heart failure, and their properties of fluid retention and weight gain are undesirable.

Section 12 – Jane

At her diabetes review appointment with the practice nurse, Jane, a 66-year-old with type 2 diabetes for 9 years, reported gradually increasing shortness of breath and fatigue when exercising, such that she was struggling to keep up with her walking group.

Background medical history: Hypertension, dyslipidaemia, background diabetic retinopathy, diabetic neuropathy, chronic kidney disease (G3aA1).

Medication: Metformin 1 g twice daily, pioglitazone 30 mg once daily, subcutaneous semaglutide 1 mg once weekly, atorvastatin 20 mg once daily, ezetimibe 10 mg once daily, losartan 100 mg once daily, amlodipine 5 mg once daily.

Social history: Non-smoker, alcohol 0–10 units/week, part-time secretary.

Recent investigations: HbA_1c 57 mmol/mol; eGFR 62 mL/min/1.73 m²; total cholesterol 4.1 mmol/L, non-HDL cholesterol 2.4 mmol/L; ACR 1.8 mg/mmol.

Examination findings: Pulse rate 84 beats/minute regular, blood pressure 137/82 mmHg, BMI 31.5 kg/m². Cardiovascular and respiratory systems were unremarkable, other than mild bilateral ankle oedema (no calf swelling or tenderness).

Subsequent investigations: ECG: left ventricular hypertrophy but no signs of ischaemia; chest x-ray: interstitial oedema, cardiomegaly; D-dimer level: normal range. Raised NT-proBNP level (1355 ng/L). Spirometry within normal limits for age.
 
How would you interpret Jane’s symptoms and what would be your next step in management?

Section 13

Jane’s symptomatology could be respiratory or cardiological in origin, but the investigations direct towards a cardiac cause of breathlessness and likely heart failure. She was referred to the cardiologist for further assessment.
 
Are there any changes in medication you would consider making at this stage?

Section 14

Given the findings, it would be appropriate to stop the pioglitazone, which could exacerbate symptoms of heart failure (and indeed is contraindicated in this condition).

As explained in Section 8, there is now robust data supporting the use of SGLT2 inhibitors in heart failure. Thus, it would be a good idea to initiate an SGLT2 inhibitor for Jane’s presumptive heart failure, which would also improve glycaemic control (given the discontinuation of pioglitazone) and facilitate weight loss.

Jane was commenced on empagliflozin 10 mg once daily, and her pioglitazone was stopped.

Section 15

Jane’s echocardiogram revealed normal left ventricular function (with a normal ejection fraction of 54%) but reduced diastolic filling. This picture is consistent with heart failure with preserved ejection fraction (HFpEF). In this situation, the left ventricle contracts satisfactorily in systole but is too stiff to fill adequately in diastole, limiting the amount of blood that can then be pumped out.

HFpEF is more common in people with diabetes than without diabetes (Brown, 2022) and is more likely to affect women. Overall, it has a similar prevalence as HFrEF (Gevaert et al, 2022). Clinical features of HFpEF are similar to those found in HFrEF: shortness of breath on exercise, reduced exercise tolerance, fatigue, peripheral oedema and orthopnoea.
 
Do you know any effective treatments for HFpEF?

Section 16

Whilst loop diuretics may be used to deal with symptoms associated with fluid overload, until recently there were no treatments that significantly improved prognosis in HFpEF (in contrast to HFrEF). However, the evidence presented in Section 8 conclusively demonstrates that empagliflozin and dapagliflozin improve outcomes in HFpEF and so should be a standard treatment in this condition, irrespective of the presence of diabetes (Anker et al, 2021; Solomon et al, 2022).

In the TOPCAT randomised controlled trial, in people with a left ventricular fraction of ≥45%, spironolactone significantly reduced rates of hospitalisation for heart failure (a component of the primary outcome) compared with placebo (Pitt et al, 2014). In the PARAGON-HF trial, looking at a similar subset of subjects, there were signals that sacubitril/valsartan reduced the severity of heart failure compared to valsartan alone (Solomon et al, 2019). Whilst neither of these trials achieved the primary composite outcome of reduction in HHF and cardiovascular death, the ADA Consensus Report concluded it was reasonable to offer spironolactone or sacubitril/valsartan in HFpEF (Pop-Busui et al, 2022).