In developed countries, heart failure affects 1–2% of the population; but in those aged >70 years, this figure rises to >10% (Ponikowski et al, 2016). Heart failure confers substantial morbidity (Mogensen et al, 2011; Mentz et al, 2014; Baron-Franco et al, 2017) and reduced life expectancy (Mogensen et al, 2011; Taylor et al, 2019), as well as significant cost to health services (National Clinical Guideline Centre, 2010; Braunschweig et al, 2011).
Type 2 diabetes mellitus is a common comorbidity seen in patients with heart failure with a prevalence of around 30% in chronic heart failure (Seferovic´ et al, 2018). Regardless of the type of heart failure they have, people with diabetes will likely have a worse New York Heart Association (NYHA) class, experience more heart failure symptoms and have poorer quality of life than the equivalent patient who does not have diabetes (Suskin et al, 2000; Kristensen et al, 2016, 2017), and have a higher mortality rate (Dauriz et al, 2017; Seferovic et al, 2018).
In the UK, many patients with heart failure have access to heart failure specialist teams, with both acute and community services realising benefits in terms of improved therapy, outcomes and cost-effective care (Stewart and Blue, 2004; Thompson et al, 2005; National Cardiac Audit Programme, 2016). Heart failure specialist nurses are central to these teams, enabling many benefits including the review and modification of the vast array of treatments in line with national and international guidelines. For heart failure patients with reduced ejection fraction, this will include the introduction and titration of medications known to improve prognosis, symptoms and quality of life, as well as reduce hospitalisations for heart failure, comprising angiotensin converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), angiotensin receptor neprilysin inhibitors (ARNI), beta blockers, mineralocorticoid receptor antagonists (MRAs) and, when indicated, ivabradine. The use of hydralazine and nitrate in combination, and digoxin may also be used (Ponikowski et al, 2016; National Institute for Health and Care Excellence (NICE), 2018). For heart failure with preserved ejection fraction, there are no prognostically beneficial medications; however, good management of hypertension and rate control in atrial fibrillation are key to successful treatment (Ponikowski et al, 2016).
While the roles and services provided by heart failure specialist teams may vary, most are prescribers, independently altering medication, who improve the timeliness and efficiency of care (Pumping Marvellous, 2018). UK nurses have the most extensive prescribing privileges in the world (Courtenay, 2018), with a range of benefits reported by patients, nurses and health professionals (Cope et al, 2016). This is not the situation in other countries because of differing legislation.
There is also a vast array of treatments available for people with diabetes including insulins, biguanides, sulfonylureas, glucagon-like peptide 1 receptor agonists, dipeptidyl peptidase 4 inhibitors, thiazolidinediones and sodium glucose cotransporter-2 (SGLT2) inhibitors. Although there have been doubts raised (Boussageon et al, 2016), metformin remains the cornerstone treatment (NICE, 2015; Wise, 2016).
The main treatment target of diabetes is to reduce glycosylated haemoglobin levels and minimise micro- and macro-vascular complications. In meta-analysis, while tighter control of glycosylated haemoglobin appears to lessen microvascular complications, but with a risk of more hypoglycaemic events, it does not lead to reduced cardiovascular morbidity, mortality or reduced risk of heart failure.
This is not the situation in other countries because of differing legislation. As the risk of cardiovascular complications is high, the medicines regulatory authorities (i e in the UK, the Medicines and Healthcare products Regulatory Agency (MHRA)) have set a secondary requirement that new agents must also demonstrate reduction in major adverse cardiovascular events, or at least not increase their risk (McMurray et al, 2014). Indeed, some diabetic drugs have been shown to have deleterious effects in respect of heart failure, including the thiazolidinediones, and dipeptidyl peptidase 4 inhibitors (Fonarow, 2014).
A more recent review of diabetes trials notes that while some hypoglycaemic medicines have been shown not to increase major adverse cardiovascular events (glargine and degludec (insulins), sitagliptin, alogliptin and saxagliptin (DPP4 inhibitors), lixisenatide and once-weekly exenatide (GLP-1)), they can reduce microvascular complications (Cefalu et al, 2018). However, there are now medications that can significantly reduce major adverse cardiovascular events: empagliflozin, canagliflozin (SGLT2 inhibitors) and liraglutide (GLP-1) (Fitchett et al, 2016; Marso et al, 2016; Neal et al, 2017), and a more recently published trial by Wiviott et al (2019), all evidenced a significant reduction in heart failure hospitalisations.
While disease-modifying medications are the cornerstone of treatment for heart failure and diabetes, many patients with heart failure also have a degree of congestion. Normally, heart failure specialist nurses will use various diuretics, alone or in combination, to treat congestion (Ponikowski et al, 2016). For those patients with type 2 diabetes mellitus, another option is now available in that diuretic armamentarium – the SGLT2 inhibitor.
SGLT2 inhibitors for heart failure
The action of SGLT2 inhibitors is to reduce renal glucose and sodium reabsorption in the proximal tubule of the nephron resulting in increased urinary glucose excretion with osmotic and natriuretic diuresis (although it is likely that the latter that accounts for most of the diuretic effect (Kimura, 2016)). The glucose loss brings about the lowering of glycosylated haemoglobin.
The increased urinary glucose excretion results in many benefits for the patient and their management from both a diabetic and heart failure perspective
Reduction in glucose reabsorption results in caloric loss and contributes to caloric weight reduction, while reduction in plasma volume results in decongestion and fluid weight loss (Ribola et al, 2017), along with a likely afterload and preload reduction (Yudkin et al, 2010). Average weight loss in trials was 2–3 kg (Van Baar et al, 2018). The caloric weight loss and diuretic effect leads to the lowering of blood pressure in the order of 4–5/1–2 mmHg (Abdul-Ghani et al, 2015). While it may be of concern to further lower blood pressure in some patients with heart failure who already have low blood pressure, it is likely that the SGLT2 inhibitor will replace some diuretic in congested patients, which would reduce the risk of blood pressure falling too low. For those with higher blood pressure, it should not be forgotten that blood pressure control for people with diabetes is a key target of care, as reduction in overall cardiovascular risk is better gained in people with diabetes by improved blood pressure control compared with cholesterol lowering or intensive glucose lowering (Yudkin et al, 2010). Unlike some antihypertensive and diabetic medications that can result in a rise in heart rate, SGLT2 inhibitors do not (Cefalu et al, 2018). As such, this avoids a negative impact on prognosis in patients with heart failure with reduced ejection fraction and sinus rhythm, where the target heart rate is 60 beats per minute (Swedberg and Komajda, 2012).
It is common that drugs used in heart failuire such as ACE inhibitors, ARBs and MRAs impact either afferent or efferent renal arteriole blood flow and so affect renal function (Damman and McMurray, 2014). Despite a reduction in renal function in some patients, the benefit of these medications is not reduced and they protect renal function in the long term (Clark et al, 2019). Likewise, the SGLT2 inhibitors reduce the glomerular filtration rate in the early weeks of treatment by roughly 5 ml/min −1/1.73 m 2. Thereafter, there is a small improvement followed by a period of stabilisation resulting in preservation of renal function in the long term (Van Bommel et al, 2017).
Additionally, SGLT2 inhibitors reduce uric acid levels (Heerspink et al, 2016). Elevated uric acid negatively impacts prognosis in heart failure (Huang et al, 2014) and is associated with a higher incidence of gout (Pittman and Bross, 1999). However, gout is not uncommonly experienced by patients with heart failure, partly because of the relationship between uric acid levels and the risk of developing cardiovascular disease (Huang et al, 2014) and partly as a side effect of diuretics (Salem et al, 2017), thus SGLT2 inhibitors may offer a diuretic effect while reducing the probability of this unwanted debilitating condition.
While it is not clear whether these medications are more effective in heart failure with reduced ejection fraction than heart failure with preserved ejection fraction, there are indications that the vascular benefits are only realised in those patients with cardiovascular disease, while reduction in heart failure hospitalisations was present across all groups (Verma and McMurray, 2019).
SGLT2 inhibitors for heart failure and type 2 diabetes
Tolerated metformin remains the first-line single agent in the treatment of type 2 diabetes mellitus (NICE, 2015). Based on the proven benefits of the SGLT2 inhibitor studies, and the fact that reduction in the risk of hospitalisation for heart failure occurs rapidly after starting these medications (Fitchett et al, 2016), there are situations where it would be reasonable that patients with heart failure and type 2 diabetes are prescribed SGLT2 inhibitors by heart failure specialist nurses:
- In patients with congestion whose glycosylated haemoglobin is elevated, despite optimum doses of metformin alone (or those patients who are intolerant to metformin)
- In patients with and without congestion but needing a diuretic to remain congestion free, whose glycosylated haemoglobin is elevated, despite optimum doses of metformin plus additional glucose lowering medications (such as SUs, GLP1, DPP4 inhibitors, TZDs and insulin). In this case, consideration needs to be given as to whether the SGLT2 inhibitors should be added to the treatment or whether one of the other medications should be swapped
- In those patients with congestion, or those patients without congestion who need a diuretic to remain congestion free, whose glycosylated haemoglobin is within range but need other medications in addition to the optimum doses of metformin to achieve this (such as the addition of SUs, GLP1, DPP4 inhibitors, TZDs and insulin). In this case, consideration needs to be given as to whether one of any additional agents should be swapped for a SGLT2 inhibitors.
In any situation where more than metformin is being used, those agents that are in use can be potentially deleterious to patients with heart failure and should be stopped/swapped first. This would include thiazolidinediones and saxagliptin, which increase the risk of heart failure (Lago et al, 2007; Scirica et al, 2014). The risk of hypoglycaemic events is higher in individuals taking insulins or sulphonylureas (Maruthur et al, 2016) but the merits of all medications in use need to be considered. SGLT2 inhibitors have been safely combined with metformin, insulin, pioglitazone, sitagliptin (Van Baar et al, 2018). From this perspective, close working with the patient's diabetic team (whether based in primary or secondary care) is paramount to ensure a full review of the risks and benefits of any combination of medications. For the more advanced and complex decisions for the management of heart failure, NICE recommends a discussion within the heart failure multidisciplinary team (NICE, 2018). It may be prudent for a diabetic specialist to be present at a meeting to discuss the management of an individual patient.
A summary of effects of various classes of diabetic medications is shown in Table 1.
Table 1. Overview of common effects of differing classes of hypoglycaemic medications
| Class | Insulins | SUs | Biguanides | GLP-1 | DPP4i | TZDs | SGLT2i |
|---|---|---|---|---|---|---|---|
| Effect | |||||||
| Weight gain | ✓ | ✓ | ✓ | ||||
| Weight loss | ✓ | ✓ | ✓ | ✓ | |||
| Nausea | ✓ | ✓ | ✓ | ||||
| Diarrhoea | ✓ | ✓ | ✓ | ||||
| ‘Hypos’ | ✓ | ✓ | |||||
| Fluid retention | (✓)* | ✓ | |||||
| Other | Lactic acidosis | Fractures | Ketoacidosis genito-urinary infections | ||||
| Reduction in MACE | ?^ | Liraglutide only | ✓** | ||||
| Reduction in HHF | ✓** | ||||||
| Reduction in mortality | ?^ | Liraglutide only | ✓** | ||||
| Reduction in nephropathy | ✓** | ||||||
| Reduction in microvascular events | Glargine and degludec | ✓ | ✓ |
Based on meta-analyses of SGLT2 inhibitor trials (Zhang et al, 2018; Zelniker et al, 2019).
^Conflicting evidence on benefits (Kooy et al, 2009; Lamanna et al, 2011).
NB: SUs: Sulphonylureas; GLP-1: Glucagon-like peptide 1 receptor agonists; DPP4i: dipeptidyl peptidase 4 inhibitors; TZDs: thiazolidinediones; SGLT2i: Sodium glucose co-transporter 2 inhibitors; MACE; major adverse cardiovascular event; HFF: Hospitalisation for heart failure
Prescribing and patient education
Before starting SGLT2 inhibitors, it is important to rule out the presence of any absolute contraindications and consider any relative contraindications. The British National Formulary lists the only absolute contraindication as diabetic ketoacidosis (NICE, 2019b) but includes safety information on this, lower limb amputation and Fournier's gangrene. However, as estimated glomerular filtration rate falls closer to 45 ml/min/1.73m 2, the glucose lowering effect is lost; thus estimated glomerular filtration rate should be >60 ml/min/1.73m 2 before starting. It should be noted that the reductions in major adverse cardiovascular events and heart failure are not lost when estimated glomerular filtration rate falls (Zelniker et al, 2019). Patients should be <75 years of age for canagliflozin and dapagliflozin and <85 years of age for empaglfilozin due to limited therapeutic experience in older patients.
Part of any prescribing decision ensures the patient's understanding of the potential benefits and risks of starting any medication (Royal Pharmaceutical Society, 2016). With regards to SGLT2 inhibitors, their benefits have been outlined in this article.
Potential detriments will include, for some patients, the opposite of some of the benefits, as might be expected. Increased diuresis may compound incontinence and may, due to blood pressure lowering, result in dizziness. It should be noted that the risk of falls with any medication that lowers blood pressure is increased within the first 2 weeks (Shimbo et al, 2016). Meta-analysis of the SGLT2 inhibitor studies showed an increased risk of genital and urinary infections (Zhang et al, 2018). These are easy to treat, and rarely lead to discontinuation of the drug. Post marketing surveillance has identified a potential small increased risk of diabetic ketoacidosis with SGLT2 inhibitors (Zhang et al, 2018), although this is more common in people who are on insulin. Others have concluded that the risk is small and is no greater than when using other antihyperglycaemic agents (Wang et al, 2017). While extremely rare, a very serious complication is Fournier's gangrene, identified in post marketing reports (MHRA, 2019). With canagliflozin, there is also a potential risk of lower limb amputation (this was not seen with other SGLT2 inhibitors) (Neal et al, 2017).
As SGLT2 inhibitors are diuretics, they should be suspended along with ACE inhibitors, ARBs, ARNI, MRAs and metformin should dehydration occur (eg with diarrhoea) due to the potential for kidney injury (Think Kidneys, 2018). A review of fluid intake should be undertaken before prescribing to euvolaemic patients. Because of the glycosuric effect, patients should be reminded that a urine dip will be positive for glucose.
Role of heart failure specialist nurses
Patients who have diabetes in primary care should have their glycosylated haemoglobin reviewed every 6 months, as a minimum (NICE, 2015). As such, a borderline glycosylated haemoglobin may not necessarily prompt immediate intervention if the decision is to ‘watch and wait’. Patients with diabetes as well as heart failure under the care of a heart failure specialist nurse should have their glycosylated haemoglobin repeated, if not recently done, to assess whether the patient should be considered for these more novel agents in respect of the fact that they may help in the treatment of their heart failure and diabetes.
Following the schema in Figure 1 can help with the following of a systematic approach. Because of the diuretic effect of SGLT2 inhibitors, concomitant increase of other medications that have a diuretic effect, including any other diuretics and potentially sacubitril/valsartan, should be avoided (Hormann et al, 2018; Beck-da-Silva et al, 2019; Vardeny et al, 2019). As per the schema notes, a concomitant dose reduction in diuretic should be done in euvolaemic patients to minimise the risks of dehydration and hypotension
If the patient is keen to proceed, close monitoring of the patient's diabetes by the primary or secondary care team is required, so they can highlight any concerns regarding the patient's condition and are updated with any changes in treatment.
Within 2 weeks of starting, the patient must be checked to ensure they are tolerating the medication. A renal function test should be repeated at around 4 weeks, with the expectation that it will fall (up to 6% eGFR) but, thereafter, should improve or stabilise. Nurses should also be prepared to monitor more frequently, if needed. Glycosylated haemoglobin should be repeated at around 3 months. At this time, it should be considered whether the dose should be increased or whether ongoing care regarding the patient's diabetes can be handed back to primary care.
Future implications
The use of SGLT2 inhibitors by cardiologists has been reported to be low (Vaduganathan et al, 2018), so it is likely that its use by heart failure specialist nurses will also be low. Personal communication to local heart failure services has identified that very few nurses are routinely considering or prescribing these agents. By initiating SGLT2 inhibitors, heart failure specialist nurses will help to raise the profile of the benefits of these medications when used in treating patients with heart failure.
Whether SGLT2 inhibitors are only effective in heart failure with reduced ejection fraction, heart failure with preserved ejection fraction, or both, is uncertain; in the initial studies (Fitchett et al, 2016; Neal et al, 2017; Wiviott et al 2019), the type of heart failure was not recorded. Trials are, however, underway to assess the impact of these medications in patients both with and without diabetes and both with heart failure with reduced ejection fraction and heart failure with preserved ejection fraction (Cefalu et al, 2018) – so they may become mainstream very soon, for a much larger group of patients living with heart failure. It would therefore be prudent for heart failure specialist nurses to gain competence in a small number of patients, where the impact is more typical and when the needed support and expertise of diabetic nurse specialists and primary care colleagues is easier to find for a small number of patients.
There is no clear evidence as to what the equivalent diuretic dose would be for each equivalent dose of SGLT2 inhibitors. It is also unknown what degree of weight loss noted in the trials was from fluid or caloric loss and, thus, whether congested patients may lose more weight than non-congested patients. The RECEDE CHF trial may provide information about this (Mordi et al, 2017).
Whether SGLT2 inhibitors can be used as an additive ‘nephron blockade’, as done when using loop plus thiazide (plus MRA), is not reported on. Bearing in mind that eGFR needs to be >60 ml/min/1.73 m 2 before starting, and the use of ACE inhibitors/ARB/ARNI/MRA and other diuretics will in most patients have reduced eGFR, it may not be possible.
Finally, where SGLT2 inhibitors will fit into heart failure treatment algorithms in the future is not clear. Should they be used first as diuretics in patients with diabetes alongside loop diuretics early in the titration of ACE inhibitors/ARBs, or only used as an add-on after optimization of ACE inhibitors/ARBs/MRAs is also not certain. However, this may become clearer from future heart failure trials.
Conclusion
SGLT2 inhibitors offer a host of important symptom and outcome benefits to patients with heart failure who have diabetes. Use of these agents in this group of patients is currently less than it should be. Horizon scanning suggests that these medications may well become part of the treatment of more patients with heart failure (with or without diabetes). Heart failure specialist nurses are ideally placed to review the complex treatment of patients with diabetes as well as heart failure, with whom they are involved. Heart failure specialist nurses, who are prescribers, should act now to update their knowledge and skills to prescribe these medications and work closely with the patient's usual diabetic team, so that opportunities to optimise the treatment of these high-risk patients are not missed.
Key Points
- SGLT2 inhibitors improve outcomes in patients with diabetes, reduce glycosylated haemoglobin, the mortality rate and the frequency of major adverse cardiovascular outcomes
- SGLT2 inhibitors act as a diuretic and reduce hospital admissions with heart failure, so symptom control is improved as well as the quality of life in patients with heart failure
- Heart failure specialist nurses are ideally positioned to prescribe SGLT2 inhibitors to carefully selected patients with diabetes and heart failure, who could benefit from these medications
- Education specific to SGLT2 inhibitors should include how to deal with thrush and the recognition of urinary tract infection, diabetic ketoacidosis and Fournier's gangrene
- Ongoing trials of SGLT2 inhibitors in patients with heart failure both with and without diabetes may demonstrate a broader use of these medications; heart failure specialist nurses should grow their competence in prescribing these medications in order to prepare for a potential increased future use of these medications in the management of heart failure.
CPD reflective questions
- Do you know the relevant criteria that need to be met, and can you identify which patients on your caseload would benefit from having SGLT2 inhibitors?
- What are the important benefits and risks you would need to convey to your patient when offering an SGLT2 inhibitor?
- What barriers might be stopping you from prescribing SGLT2 inhibitors to relevant patients on your caseload and where could you get support locally for starting SGLT2 inhibitors for your patients?