Cardiovascular disease (CVD) is highly prevalent in patients living with type 2 diabetes (T2DM). There is a strong association between hyperglycaemia, poor blood glucose control and the development of heart failure (HF). Literature suggests that for each 1% increase in mean glycaemic levels, as referenced by HbA1c (glycosylated haemoglobin), there is an 8% increased risk of HF.¹ Equally, insulin resistance (IR) is a strong predictor of HF and was identified as one of the main drivers in the pathogenesis of HF and diabetic cardiomyopathy (DMCM).^2,3,4 In patients with diabetes or IR the cardiac energy metabolism is significantly altered. Toxic lipid and glucose intermediates trigger several adverse signalling pathways, mitochondrial dysfunction and tissue inflammation.⁵ This subsequently induces the typical features of diabetic and/or IR cardiomyopathy, which are left ventricular hypertrophy and diastolic dysfunction.⁶ Heart failure and T2DM have a synergistic effect, which is demonstrated in increased symptom burden, hospital admissions and mortality in affected patients.⁵

The predominant symptoms of HF are dyspnoea, fatigue and other manifestations of poor cardiac output and congestion. Thirst is a common feature among patients living with HF. It can be exacerbated by the clinical management and self-care directed at alleviating the symptoms of systemic congestion.^7,8

Diet has been identified as a modifiable risk factor and a cornerstone for the treatment of both diabetes and HF.⁹ Standard care as outlined in the existing global guidelines for the primary and secondary prevention of CVD, includes restrictions in calories, salt, saturated and total fat and focuses on increasing the intake of vegetables, fruit, whole grains as well as foods rich in mono- and poly-unsaturated fats.^10,11,12

A separate body of evidence suggests that therapeutic carbohydrate restriction may offer benefits over low-fat (LF) diets particularly in patients with underlying glucose metabolism disorders such as diabetes.¹³

Multiple systematic reviews and meta-analyses indicate that low-carbohydrate (LC) diets are an important addition to existing heart healthy dietary patterns and are efficacious in improving glycaemic control and cardiovascular health in patients living with diabetes.^14,15,16,17 Generally, LC diets reduce overall dietary carbohydrate (CHO) intake to less than 130 g/day. Depending on the style (e.g. Atkins, Paleo) or goal (weight loss, diabetes remission) of the LC diet, very LC ketogenic diets may recommend less than 30 g of dietary CHO per day.^18,19 Specifically, LC diets (compared with LF diets) have demonstrated greater reductions in glycated haemoglobin (HbA1c) and sustained improvements in diabetic dyslipidaemia as defined by elevated triglyceride and small dense low density lipoprotein (LDL) particles and decreased high density lipoprotein (HDL).^14,16,20

However, research investigating the effects of a LC diet in patients with DMCM or HF is scarce. In one 5-week crossover study, Von Bibra et al.²¹ compare the effects of a LC and a LF diet on diastolic cardiac function and markers of IR. While weight loss was similar in the LC and the LF groups, mean systolic blood pressure, fasting and postprandial triglyceride and insulin levels only improved significantly on the LC diet (p < 0.05). In addition, the authors described a significant improvement in maximal exercise capacity (2 ± 7-watt, p < 0.05) and markers of IR (Homeostatic Model Assessment for IR [HOMA-IR]) in the LC but not in the LF diet group (LC −0.9 ± 2.4, p < 0.05; LF −0.1 ± 1.4, p > 0.05). On the other hand, data from animal studies have repeatedly demonstrated that LC diets improve cardiac remodelling and halt HF development.^22,23,24 This success appears to be in part because of diet-induced ketogenesis and the protective role of ketone bodies against oxidative stress and myocardial damage.²⁵ The enhanced fatty acid metabolism driven by the LC diet and the dietary CHO restriction itself has also shown to prevent and reverse HF in mice.²⁶ As a result of the scarcity of evidence regarding an optimal dietary pattern for patients with DMCM, the authors propose to undertake a randomised controlled trial (RCT) to determine the effects of a LC diet in patients with DMCM.

The specific aims for the RCT are as follows:

The authors hypothesise that a LC diet would improve HF symptoms through two main pathways: (1) IR is known to be a major culprit for the progression of HF.²⁷ Dietary CHO restriction reduces blood glucose and insulin levels, hence manages IR. It is plausible that a LC diet improves myocardial function and energy supply in the failing heart. (2) Reducing blood glucose and insulin levels through a LC diet promotes diuresis and sodium excretion by decreasing the expression of sodium glucose transporter 2 (SGLT2) receptors in the renal tubules.²⁸ Consequently, this may lower blood pressure levels and alleviate oedema in patients living with DMCM. It is further hypothesised that these effects of the LC diet will have a flow-on effect on the well-being and QoL of patients with DMCM.

Ethical approval was sought and granted from the study hospital and the University Human Research Ethics Committee (ethical approval no. ND 69645/2020).

All procedures performed in this study are in accordance with the ethical standards of the National Research Committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Baseline collection will consist of the following demographical and biometric data: (1) age, (2) gender, (3) weight and (4) blood pressure (Table 2). In addition, the medical history of each participant will be recorded including food allergies/intolerances, current medications and relevant pathology including HbA1c. Moreover, at baseline the participants current New York Heart Association class (NYHA), is assessed. Lastly, QoL and thirst will be evaluated. Measures will be collected as described next.

Low levels of carbohydrate intake (< 50 g per day) lead to a change in metabolism, where ketone bodies and fatty acids (as compared with glucose) become the primary energy source in the body. This state, also referred to as nutritional ketosis, is a benign process and is characterised by the presence of ketones in the blood (usually < 3 mmol/L), normal blood glucose levels, low insulin levels and an associated breakdown of adipose tissue.¹⁸

In contrast, the medical emergency DKA occurs most commonly in people with frank insulin deficiency such as T1DM and insulin dependent T2DM and may occur in patients taking SGLT2i. Both populations are excluded from this RCT.

Low-carbohydrate diets are considered safe; however, literature indicates that LC diets lower blood sugar levels in patients with metabolic disease and antihyperglycaemic medication frequently needs adjustment.^42,43 The authors will therefore coach participants to gradually decrease the CHO amount in their meals over the initial four weeks of the intervention to ensure more stable blood sugar levels and reduce the risk of hypoglycaemia. Participants with T2DM will be required to maintain a blood glucose diary for the duration of the study and will be asked to perform regular blood glucose measurements (at least twice per day) and anytime, if required. In addition to this, participants will be re-educated on how to recognise hypoglycaemic events and how to manage lower than normal blood sugar levels. Should participants report a hypoglycaemic event or the review of the blood glucose diary reveals consistent borderline low blood glucose levels, a researcher will liaise with the participants own health practitioner to discuss management and medication adjustment if necessary.

Low-carbohydrate diets lead to lowered plasma insulin levels and blood glucose levels, which naturally decreases SGLT2 receptor expression in the kidneys.²⁸ This well-known diuretic effect of the LC diet may lead to a significant decrease in blood pressure levels in study participants.^28,44 As part of the weekly education sessions, participants will be assessed for signs of orthostatic hypotension such as lightheadedness and dizziness when standing up quickly. Should participants report such symptoms, a blood pressure (BP) measurement is taken. Furthermore, participants are referred to their usual HF team for review and medication adjustment if needed.⁴⁵

All participants will attend their regular scheduled HF specialist and other health appointments throughout the study period.

Descriptive statistics for continuous variables will be used to describe the mean (and standard deviation) or median (25th and 75th percentiles; interquartile range [IQR]) values from outcomes. Frequencies and chi-square will be used to describe categorical variables. Data will be analysed on an intention to treat basis with the last observation carried forward for analyses of all study participants. Missing data will be followed up with the participant after each visit. Any remaining missing data will be handled using multiple imputation. For categorical variables, independent T-test statistics will be used to compare independent samples. Comparisons of intervention and control group participants will use logistic regression models (with odds ratio [OR] and 95% CIs for binary variables), logistic regression for categorical variables and linear regression for approximately normally distributed continuous variables. Statistical significance will be defined as a p-value less than 0.05 (2-sided). All analyses will use Statistical Package for Social Sciences (SPSS) version 26 (SPSS Inc., Chicago, IL, United States).

Nutrition and diet play a key role in the management of both HF and diabetes. Current dietary guidelines for patients with HF appear to be lacking effectiveness. Alternative diets, including LC diets should be rigorously researched in the context of HF (with and without diabetes). Low-carbohydrate eating patterns and diets consistently improve glycaemic control and have shown to effectively address dyslipidaemia in over 30 clinical trials.¹³ However, limited research is available that investigates the effects of a LC diet with focus on patients with diabetes and HF.

This study aims to explore the effects of a LC diet compared with a LF diet (UG) on HF symptoms and QoL in patients with HF and T2DM or IR from a single centre quaternary metropolitan hospital in Victoria, Australia. The outcomes of this RCT will provide new insight into a dietary approach for HF with T2DM or IR and may be of interest to researchers wanting to pursue further studies in this field. In addition, the results of this RCT may be used to inform dietary guidelines for patients living with both HF and diabetes.

Some aspects of this study need to be highlighted. To assess and increase diet adherence the 3-day food diaries and FFQ have been included. All care will be taken that participants’ resources are easy to use, adaptable and based on the needs of HF patients. The food lists will provide a simple and user friendly tool for study participants. However, as participants plan their meals according to the food lists, the macronutrient composition, especially CHO consumption per day may vary significantly. However, the authors believe that by selecting food items from the always/every day (green list), sometimes (orange list) and avoid (red list), the total CHO intake will be not more than 130 g/d, hence it is in line with the definition of what is considered a LC diet.¹⁸ On the other hand, some patients may eat very low amounts of CHO throughout the study period (< 50 g/d). This may have greater therapeutic impact than a diet where an amount close to the upper limit of 130 g/d is consumed.⁴⁶

Another potential limitation is the single centre design of the study as samples taken from multiple sites are more representative. Based on power calculations, it is believed that sample (n = 80) is large enough to give an accurate estimate of the intervention effect of this diet in RCT.

Lastly, because of recommendations in the most recent 2021 European Society of Cardiology (ECS) Guidelines for the diagnosis and management of acute and chronic HF, SGLT2i are now advised for all HF patients, regardless of diabetes status (Class I, Level A recommendation).⁴⁷ Thus, recruitment of patients with DMCM who are not currently taking SGLT2i proves to be challenging and as such this population will likely be underrepresented. This may need to be considered when interpreting the results.

Insulin resistance is one of the key drivers in the development and progression of DMCM. The authors’ protocol describes a dietary approach targeting IR by means of therapeutic CHO restriction. This study represents a step in the evaluation of the efficacy of a LC diet intervention for patients with DMCM and may provide valuable insight into evidence-based nutrition care for this priority patient group. This study is novel and innovative as it investigates the effects of a LC diet on HF symptoms and QoL. There is paucity of research in this area. The results of this pilot RCT may be used to design larger multicentre RCTs. Future studies will be able to evaluate long-term outcomes and challenges associated with long-term LC diet adherence in patients with DMCM: a question which might not be addressed in this work.