Therapeutic carbohydrate restriction and sodium-glucose transporter 2 inhibitors in chronic kidney disease: A potentially powerful combination

The introduction of the sodium-glucose transporter 2 inhibitors (SGLT2i) into clinical practice has had a profound impact on the management of individuals with kidney disease and have quickly been incorporated into clinical guidelines. Indeed, SGLT2i are considered foundational therapy for people with chronic kidney disease (CKD) across a wide range of glomerular filtration rate and proteinuria levels, regardless of diabetes status or CKD aetiology.1 These drugs inhibit the SGLT2 receptor in the proximal tubule of the kidney resulting in glycosuria of ~80 g/day – the sugar content of two cans of Coca-Cola! The drugs in clinical use are not selective for the SGLT2 receptor with additional activity at the SGLT1 receptor, expressed in the intestine and in the distal part of the renal proximal tubule. SGLT2i induce weight loss (2-3 kg), reduce systolic blood pressure (2-3 mmHg), shift substrate utilization from carbohydrates to lipids generating ketones enhancing organ energetics,2 decrease triglycerides; however, increase levels of low-density lipoprotein (LDL)-cholesterol.3 In the kidney, SGLT2i lower glomerular pressure, induce autophagy and increase urate excretion.4 The SGLT2i empagliflozin increased the richness and diversity of the gut microbiota in parallel to beneficial changes in the metabolome.5 Lifestyle advice is the first pillar in the management of CKD. Low carbohydrate diets have become increasingly popular and have been endorsed as an initial therapeutic strategy for the management of type 2 diabetes (T2D). The definition of a low carbohydrate diet differs across studies. For the purposes of this commentary, we will align definitions as follows: ketogenic very low carbohydrate <50 g of carbohydrates per day; and liberal low carbohydrate 50-130 g per day.6 Low carbohydrate diets reduce body weight (7-9 kg) and improve cardiovascular risk factors such as systolic (4.8 mmHg) and diastolic blood pressure, plasma levels of glucose, insulin, triglyceride, high-density lipoprotein and C reactive protein.7 An increase in LDL-cholesterol and total cholesterol has been noted.8 In the kidney, a ketogenic diet inhibits oxidative stress and fibrosis, promotes autophagy and downregulates the NLRP3 inflammasome improving mitochondrial function.9 Enrichment of the gut microbiome was observed after 12 weeks of a low carbohydrate diet.10 The similarities between the effects of a low carbohydrate dietary approach and SGLT2i are striking. In this commentary we discuss the evidence for this dietary approach in the management of CKD and the potential risks and benefits when combined with SGLT2i. Poor metabolic health is almost universal in those with CKD: T2D is the major driver of CKD and in those without T2D, insulin resistance is an early metabolic change. Advanced CKD is characterized by a pro-atherogenic lipid profile of increased apolipoprotein B/apolipoprotein A1 ratio, small dense triglyceride-rich lipoproteins and reduced high-density lipoprotein cholesterol.11 LDL-C does not differ between those with early or advanced CKD.11 Based on this pathophysiology, a low carbohydrate approach may be advantageous in improving kidney health. By default, a diet low in carbohydrate must be enriched in protein and/or fat to maintain macronutrient composition. Three components require consideration in the development of low carbohydrate interventions: the amount of carbohydrate, types of foods and mode of delivery.6 There is no single carbohydrate amount that has been shown to be optimum highlighting the importance of tailoring the prescription to individual requirements accounting for prevailing insulin resistance, metabolic flexibility, physical activity among other factors. Most low carbohydrate dietary interventions have prescribed whole foods, ad libitum energy in combination with high or unrestricted amounts of fat and/or protein. Finally, in terms of delivery this can be fixed or adaptive. In a fixed mode, the amount of carbohydrate remains the same over time, whereas in an adaptive programme there may be liberalization of the carbohydrate prescription after an initial phase. In the setting of CKD, dietary protein has been purported to potentiate progression leading to the 2020 Kidney Disease Quality Outcomes Initiative (KDOQI) guidelines advocating protein restriction.12 This guideline is based on opinion level evidence and is not supported by a recent Cochrane review of protein restriction (0.6-0.8 g/kg/day) compared with usual or unrestricted protein restriction (≥1 g/kg/day) in diabetic kidney disease.13 Indeed, in individuals with T2D and Stage 2-3 kidney disease allocated to a carbohydrate-restricted (35%), low-iron available, polyphenol-enriched (CR-LIPE) diet or control, the CR-LIPE diet was 40-50% more effective than standard protein restriction in improving kidney and overall survival rates over 4 years.14 Further, in a retrospective cohort study, Mitchell et al15 modelled kidney function on 2004 individuals (15% with CKD Stage 3, half of whom had T2D) who had been treated with a low carbohydrate diet. For those with CKD Stage 3 without T2D and 5% weight loss, the estimated glomerular filtration rate was modelled to improve by about 1-3 ml/min/1.73 m2. For those with CKD Stage 3 and T2D, the estimated glomerular filtration rate was essentially unchanged across all weight change categories. Finally, for participants in the Dietary Intervention Randomized Controlled Trial (DIRECT), a low carbohydrate diet improved kidney function and reduced albuminuria over 2 years probably mediated by weight loss induced improvements in insulin sensitivity and systolic blood pressure.16 An important question to consider is whether combining the therapies of a low carbohydrate dietary approach with SGLT2i would have additive or synergistic beneficial effects or alternatively be detrimental. At least in small animals, the combination of low carbohydrate diet and SGLT2i appears to be beneficial in reducing glycaemia, without an additive increase in ketosis. SGLT2i and low carbohydrate diets have differential effects on body composition and metabolic profiles and, in the setting of SGLT2i and calorie restriction, had a syngergistic effect improving insulin resistance, creatinine clearance and blood pressure.17 The principal concern with such an approach is of the risk of ketoacidosis. The risk of ketoacidosis with SGLT2i use is infrequent in the setting of kidney disease without diabetes (one case from 4894 participants); in the presence of diabetes the relative risk doubled, albeit with low absolute numbers (46 vs. 18 cases from over 20 000 participants).1 Three enabling factors have been identified for ketoacidosis: (i) reduction in insulin/glucagon ratio; (ii) accumulation of ketone bodies; and (iii) hypovolaemia as may occur in the settings of prolonged fast, surgical intervention (mainly immediate postoperative), acute infection, relative insulinopenia and volume depletion.18 These data align with findings from a rat model in which ketoacidosis was induced when insulinopenia occurred together with volume depletion, with neither alone sufficient.19 These data underpin the importance of a sick day management plan when prescribed SGLT2i therapy. Specifically, SGLT2i should be ceased if an acute stressor develops (for example: infection, surgery, vomiting and/or diarrhoea, or poor oral intake); fluid intake should be maintained with water and blood ketone levels monitored. Unrecognized type 1 diabetes has been identified in some who develop euglycaemic ketoacidosis and as such remains a contraindication to the prescription of SGLT2i. This latter point is critical as ~10% of people diagnosed with T2D have latent autoimmune diabetes of adults, which poses a greater risk of euglycaemic ketoacidosis because of insulin deficiency. It is worth noting that initial studies of SGLT2i therapy in people with type 1 diabetes have shown benefits in glycaemic control, weight control and cardiovascular risk parameters20 and as such remains an area of research. SGLT2is are indicated for the management of proteinuric kidney disease irrespective of diabetes status and have clear benefits on morbidity and mortality.1 The risk of ketoacidosis is low particularly in the absence of diabetes,1 which may be in part explained by metabolic dysfunction evident early in CKD and the prominence of hyperinsulinaemia. A low carbohydrate diet is a therapeutic option in the management of T2D and obesity and appears at least safe and potentially beneficial in the management of CKD. The effect of dual therapy, while in small animals appears promising with potentially synergistic effects, is yet to be determined in humans. We propose the following algorithm (Figure 1) for the use of carbohydrate reduction in conjunction with SGLT2i therapy in people with CKD. In the absence of T2D, therapeutic carbohydrate restriction is a potential viable initial option. The degree of carbohydrate restriction needs to be tailored to the individual needs and concomitant protein and fat composition varied according to CKD stage. Relative hypotension and diuresis may occur with the reduction of carbohydrates, prompting the pre-emptive deprescription of diuretics and/or antihypertensives. For the individual with CKD in the setting of T2D, unrecognized latent autoimmune diabetes of adults should be excluded by screening for anti-GAD, IA2 and zinc transporter antibodies and measurement of paired C peptide and glucose levels. The introduction of a low carbohydrate diet will reduce glycaemia necessitating the pre-emptive reduction of hypoglycaemic agents: insulin generally should be reduced and sulphonylureas ceased. Antihypertensive medications and/or diuretics may also require adjustment. For both groups after stabilization, SGLT2i may be introduced with, for safety reasons, adaption of the dietary carbohydrate content away from a very low ketogenic approach. A sick day management plan should be agreed upon with an emphasis on avoiding prolonged fasting and volume depletion. We believe the combination of a low carbohydrate diet and SGLT2i has the potential to be a powerful therapeutic intervention in delaying progression of CKD because of their complementary mechanisms of action. Clinicians must use caution and monitor patients closely as we await definitive clinical trials. Open access publishing facilitated by Deakin University, as part of the Wiley - Deakin University agreement via the Council of Australian University Librarians. KMD has received honorarium from AstraZeneca, Boehringer Ingelheim, Bayer. KMD has received grant funding from Servier. AG & PF: No COI to declare. The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/dom.15195. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.