The role of dietary protein in the development of diabetes-related kidney disease: Is a high protein diet safe?
Kidney disease is a major complication of suboptimally managed diabetes, second only to cardiovascular disease in its contribution to the increased mortality of people with diabetes.1-3 Diabetic kidney disease (DKD) , or diabetic nephropathy, is characterised by glomerulosclerosis,4 which starts as glomerular basement membrane thickening , possibly caused by the endothelial dysfunction that occurs in diabetes.5 This leads to leakage of molecules such as protein through the glomerular capillaries7 and progresses to end stage renal failure (ESRF).7 Hyperglycaemia and hypertension both accelerate the progression of DKD.1-8 Although treatment of chronic kidney disease (CKD) has improved vastly, postponing progression to ESRF, kidney disease is still a threat to quality of life and costly to treat. Therefore it is logical to seek strategies to prevent it.9-10
A reduced consumption of dietary protein is one strategy recommended to prevent the development and progression of DKD.1, 11-13 This contradicts popular recommendations to reduce carbohydrate intake and increase dietary protein in order to promote weight loss and manage blood glucose levels (BGLs) in people with diabetes.13, 14 While short-term reduction of BGLs is desirable, in the long term the reduction of complications is the primary goal of diabetes management and other risk factors as well as BGLs must be managed to achieve this.
Dietary protein requirements, recommendations and intakes
In Australia, the Estimated Average Requirement (EAR) of protein for women is 0.6g/day/kg bodyweight (bw) and for men is 0.7g/day/kg bw, while the Recommended Dietary Intake (RDI) for women is 0.75g/day/kg bw and for men is 0.84g/day/kg bw.15 The protein intakes of most Australians exceed these recommendations considerably, with the Australian Health Survey of 2011-12 showing that 19-30 year olds consumed an average of 113g/day, 31-50 year olds consumed an average of 108g/day and 51-70 year olds consumed an average 98g/day.16
Stages in the development of kidney disease
In the first stage of CKD renal function, as measured by Glomerular Filtration Rate (GFR), is normal or high, while stages two to four of CKD are characterised by proteinuria and increasingly reduced GFR. Stage five, ESRF, is marked by uraemia.17
Diabetic nephropathy progresses through five slightly different stages.9, 18 The first stage, characterised by hyperfiltration and renal hypertrophy, can be found at diagnosis of even type 1 diabetes,19 possibly related to the endothelial dysfunction that occurs in diabetes.5 This can progress to a silent stage two in which the GFR continues to rise and the glomerular basement membrane thickens. Stage three, incipient nephropathy, may develop after several years of diabetes, with persistent and increasing microalbuminuria, rising blood pressure and continuing supranormal GFR. Stage four, overt diabetic nephropathy, is indicated by clinical proteinuria and declining GFR. Stage five, also ESRF, is again marked by uraemia.9, 18
The role of diet in managing CKD
If introduced earlier during declining GFR and clinical proteinuria, stages two to four of CKD, a diet lower in protein can slow the decline in GFR and the progression to renal failure.12, 22, 24-28 The National Kidney Foundation therefore recommends a protein intake of 0.8g /kg bw/day in adults with moderately or severely decreased GFR. They also recommend that these individuals avoid a protein intake of more than 1.3g/kg bw/day.1
As studies have confirmed the benefit of protein restriction in people with overt diabetic nephropathy in slowing the decline in renal function,28-35 the American Diabetes Association therefore also recommended a protein intake of 0.8g/kg bw/day in people with diabetes who have albuminuria and/or reduced GFR.36
To maintain adequate energy consumption, lowering the protein intake usually means increasing carbohydrate and/or fat consumption. Ideally this is achieved with a diet relatively high in carbohydrates from whole-grains, fruit and vegetables, high in unsaturated fats and low in animal proteins.2, 37, 38 However there is evidence that these guidelines are not usually followed by people with CKD.39
The effect of dietary protein on renal function in healthy people and those with early kidney disease
Dietary protein has a significant effect on kidney function, workload and structure. Protein consumption increases urea production, which requires an increase in urine volume to excrete.12, 40 This in turn stimulates renal blood flow and raises intraglomerular pressure, leading to higher GFR and more efficient excretion of protein-derived nitrogenous waste products.12
GFR is positively associated with protein intake.41 In healthy young men the GFR, blood urea, serum uric acid, urinary albumin and urea excretion were significantly higher in those on a high protein diet compared with those on a normal-protein diet.42 A meta-analysis of randomised controlled or cross-over trials studying the effect of high protein diets, used mainly for weight reduction in people without kidney disease, showed that high protein diets were associated with raised GFR, serum urea, urinary calcium excretion and serum uric acid.43 In people with mild renal insufficiency a diet high in protein was associated with an increase in GFR.44
In some studies this rise in GFR in response to a high protein diet in people without kidney disease or in early kidney disease has been reported as beneficial45 and a fall as harmful,46 but evidence contradicts this interpretation. In two population-based studies of apparently healthy adults with normal renal function, the prevalence of renal hyperfiltration was highest in those with the highest protein intake, but over time they experienced a greater decline in GFR.47, 48 Among people who had experienced myocardial infarction, those consuming higher protein intakes had a higher rate of decline in kidney function, with a two-fold difference between those consuming more than 1.2g/kg ideal bw/day compared with those consuming less than 0.8/kg ideal bw/day and a three-fold stronger effect in those who had diabetes compared with those who did not.28
Some studies, including the Modification of Diet in Renal Disease (MDRD) study, did not show a significant change in renal function when protein intake was modified.25, 49, 50 In the MDRD study an initial rise in GFR was followed by a slower decline, resulting in a non-significant change that was interpreted as evidence of no benefit in restricting protein intake. It was later suggested that a longer time would have shown a significant benefit.12, 26
People whose kidneys have sufficient reserve to accommodate the rise in GFR do not develop pathology. However if a proportion of nephrons are already damaged, as occurs in many people who are obese41 or have diabetes, the glomerular capillary plasma flow rate and mean glomerular capillary hydraulic pressure increase in the remaining nephrons. The glomerular hyperfiltration and hypertension cause further damage to the glomerular basement membrane in the glomerular filtration barrier.51 This progresses to glomerular sclerosis and eventually renal failure.7, 11
Since reducing abnormally high GFR is desirable in order to protect the kidneys from further damage 48, it is not appropriate to use the rate of decline in GFR as an undesirable end-point of an intervention such as dietary modification; an alternative measurement is required.4, 50 Microalbuminuria is an early indicator of renal disease 8, 31, 52-54 and at this stage is a better predictor of kidney health than declining GFR.31 A high protein intake is associated with increased microalbuminuria.55-57
The role of protein restriction on the risk of progression to diabetic nephropathy
In people with diabetes in the earlier stages of nephropathy, increased renal volume, glomerular hyperfiltration and glomerular hypertrophy promote glomerulosclerosis.9 An 8-year study of adolescents with type 1 diabetes (T1D) showed that increased GFR precedes persistent microalbuminuria and predicts the development of nephropathy.54
In young adults with T1D, protein intake was correlated with GFR, and when the protein intake was reduced from 2.0g/kg bw/day to 1.0g/kg bw/day, the GFR dropped,58 especially in individuals with hyperfiltration.54 In adults with T1D, a protein-restricted diet over two years reduced albuminuria and GFR.59 In those without albuminuria, reducing the protein intake from 1.45g/kg bw/day to 0.76g/kg bw/day was accompanied by decreases in glomerular hyper-filtration and renal plasma flow (RPF)60, while in those with microalbuminuria, reducing protein intake from 1.25g/kg bw/day to 0.63g/kg bw/day resulted in decreases in glomerular hyperfiltration and albumin excretion.61
Similarly, when people with type 2 diabetes (T2D) reduced their protein intake from an average 1.43g/kg bw/day to 0.66g/kg bw/day, their GFR decreased.62 In those with microalbuminuria, reducing protein intake from 2.0g/kg ideal bw/day to a moderate 0.8g/day/kg ideal bw/day was accompanied by a 29% reduction in GFR and 65% reduction in albumin excretion.63
Although some studies of people with diabetes14, 35, 50, 64 did not show a significant change in renal function when protein intake was modified, the National Kidney Foundation and the American Diabetes Association both considered the evidence sufficient to recommend a protein intake of 0.8g/kg bw/day across the first three stages of diabetic nephropathy as well as the fourth stage.1, 36
The effect of source of dietary protein on renal function
Not only the amount of protein consumed, but also the source of the protein, affect kidney function.
Studies have shown that intake of animal protein is positively associated with renal hyper-perfusion, with a higher risk of CKD and a higher level of mortality in people with CKD, while intake of vegetable protein is negatively associated with renal hyperperfusion,65 the risk of CKD66 and mortality in people with CKD.67
After three weeks on a diet high in animal protein, healthy volunteers had higher GFR and renal perfusion rate (RPF) than those consuming only vegetable proteins.68 Adding animal protein increased GFR and RPR, but adding vegetable protein did not.69
In the Nurses’ Health Study, those who consumed two or more servings of red and processed meat were more likely to have microalbuminuria.70 In a population-based study, red meat consumption was positively correlated to the risk of end stage renal disease.71
In vegans, GFR has been found to be lower in than in lactovegetarians, while the GFR was lower in lactovegetarians than omnivores.72 After a meal containing soy protein the increment in GFR was lower than after fish and poultry, while the GFR after a meal containing red meat was highest.73 Replacing red meat with chicken reduced the GFR and albumin excretion.62 When compared to the effect of red meat, consumption of milk protein had little effect on kidney function.74
The ACR (Urinary Albumin-to-Creatinine Ratio, a spot check for microalbuminuria and risk factor for CVD) increased when non-dairy animal products were consumed, but whole grains, fruit and low fat dairy products were associated with a lower ACR.75
People with T2D who consumed soy protein in place of some animal protein had lower levels of proteinuria at the end of the four-year trial.76 In a cross-sectional study of people with T2D and declining GFR, the total protein intake was inversely associated with renal impairment, but a higher intake of vegetable protein had the opposite effect.77 Those consuming a high meat diet had higher GFR and higher creatinine excretion compared with those consuming a high fish and vegetable diet.78
A review of the benefits of increased intake of plant-based foods and decreased consumption of animal-based foods for renal health showed that animal protein, especially red meat and processed meat, increased the risk of CKD.79 Similarly, a review of the effect of red meat on CKD found that red meat has negative effects on the kidneys and recommends that people with CKD obtain their protein from legumes, nuts, dairy products, poultry and fish.80 Chaveau et al81 recommend a Mediterranean-style diet for those at risk of CKD because this diet is low in red meat and processed meat, and moderate in protein from vegetables, eggs, fish, poultry and dairy food.
A number of mechanisms have been proposed to explain the harmful effect of protein, especially animal protein, on kidney health.
One explanation is that a high protein meal leads to a surge of amino acids that stimulates dilation of the afferent arteriole, increased GFR and increased intraglomerular pressure11 which damages the glomeruli.83 ACE inhibitors, which protect kidneys, reduce intraglomerular pressure.32
A high intake of animal proteins is also normally associated with low potassium intake, which in turn is associated with increased hypertension, a known risk factor for renal damage,82 while the bioavailability of phosphate , also a risk factor for kidney damage, is lower in plant foods.70
The increase in inflammation may be mediated by the gut microbiota, possibly via toxins metabolised from animal proteins, especially red meat.80 Microbiota-derived uremic retention solutes, such as trimethylamine N-oxide (TMAO) from animal proteins and Advanced Glycation End (AGE) products from reactions between sugars and amino groups, have been shown to have detrimental effects on the kidney.12, 87 Lowering protein intake, using a vegetarian diet and increasing intake of resistant starch can reduce the pathobionts and increase the symbionts, which in turn slows CKD progression.88, 89
Animal proteins increase acid load65, 79, 90, 91 and acid levels, which in vulnerable individuals with reduced renal capacity can lead to acidosis.40 In particular the sulphur-containing amino acids induce a higher acid load12 In people with T2D, the proportion of energy from animal protein was correlated with renal acid load.91
A low protein diet (LPD) can reduce the formation of oxidation products that destroy the integrity of the glomerular basement membrane, allowing it to filter toxic products and leak protein.22
It has also been suggested that plant foods contribute to reduced risk of CKD through a number of risk factors, including reduced hypertension, uraemic toxins, inflammation and oxidative stress.82
Currently there is much enthusiasm and research supporting the use of diets low in carbohydrates and high in protein to lower BGLs, HbA1c and bodyweight. Short-term benefits have been demonstrated, but not sustained in the longer term.92
In many studies, only decline in GFR has been used to indicate an effect of protein intake on renal pathology70 and the increase in GFR during early diabetic nephropathy has been regarded as a benefit of increasing protein intake.45 Therefore an initial rise in GFR followed by a slower decline resulted in a non-significant change that was interpreted as evidence of no benefit in restricting protein intake.12
Studies into restricting protein intake are fraught with challenges and many do not show a significant benefit.35, 46, 50, 64 Non-significant differences can result from difficulties in accurately monitoring food intake and therefore consumption of specific nutrients,33, 92 non-adherence to the prescribed diet by some participants,23, 31, 33, 35, 59, 92, 93 limited duration,35 small participant numbers33, 35, 93 and deterioration in compliance in long-term dietary interventions.64, 92
Perhaps the efficacy of protein restriction is not the true barrier to its implementation. According to Ko et al12, “one of the main obstacles to the implementation of LPD is the large gap in protein intake between the amount recommended from guidelines and what is consumed contemporarily”.
In people with healthy kidneys, a high protein diet affects kidney function but does not lead to kidney pathology. However in people with vulnerable or damaged kidneys, a high protein diet can increase the rate of progression to kidney pathology. Initially, the high protein diet stimulates the GFR, but as the kidney damage progresses, the GFR declines more quickly to pathological levels.
The source of protein is important, with animal protein more harmful than vegetable protein. Within animal proteins, red meat is more harmful than fish and chicken which are in turn more harmful than dairy protein.
High protein, low carbohydrate diets have been given much publicity and credibility. However, despite a short-term benefit on BGLs, the evidence indicates that a diet high in protein, especially animal protein, increases the risk of kidney disease in people with diabetes. People with diabetes should be encouraged to reduce protein serves to the sizes advised in the Australian Guide to Healthy Eating 95 and to reduce red meat and processed meat in favour of protein from vegetables, eggs, fish, poultry and dairy food,82. Further, they should be discouraged from adopting a high protein diet. Rather than following the advice of celebrities advocating high protein low carbohydrate diets, individuals should be referred to an Accredited Practising Dietitians (APD’-s) for Medical Nutrition Therapy through which these recommendations are tailored to individual requirements.
Dr Kate Marsh has kindly provided encouragement, proof-reading and advice that were invaluable in writing this article. Her support was much appreciated.
Dr Mark Thomas, Nephrologist at Royal Perth Hospital, provided encouragement and advice that guided the writing of the original literature review thirty years ago.
National Kidney Foundation. KDOQI Clinical Practice Guideline for Diabetes and CKD: 2012 update. American Journal of Kidney Diseases. 2013;60(5):36.2.
Ko GJ, Kalantar-Zadeh K, Goldstein-Fuchs J, Rhee CM. Dietary Approaches in the Management of Diabetic Patients with Kidney Disease. Nutrients. 2017;9(8).3.
Australian Institute of Health and Welfare. Deaths among people with diabetes in Australia, 2009â2014. Canberra; 2017. Report No.: Cat. no. CVD 79.4.
National Kidney Foundation. KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for Diabetes and Chronic Kidney Disease. American Journal of Kidney Diseases. 2007;49(2, Supplement 2):S12-S154.5.
Shestakova MV, Jarek-Martynowa IR, Ivanishina NS, Kuharenko SS, Yadrihinskaya MN, Aleksandrov AA, et al. Role of endothelial dysfunction in the development of cardiorenal syndrome in patients with type 1 diabetes mellitus. Diabetes research and clinical practice. 2005;68 suppl1:S65-S72.6.
Andersen AR, Christiansen JS, Andersen JK, Kreiner S, Deckert T. Diabetic nephropathy in type 1 (insulin-dependent) diabetes: An epidemiological study. Diabetologia. 1983;25(6):496-501.7.
Brenner BM, Lawler E, V. , Mackenzie H, S. . The hyperfiltration theory: A paradigm shift in nephrology. Kidney International. 1996;49(6):1774.8.
American Diabetes Association. Diabetic Nephropathy. Diabetes Care. 2002;25(suppl 1):s85.9.
Mogensen CE, Christensen CK, Vittinghus E. The stages in diabetic renal disease. With emphasis on the stage of incipient diabetic nephropathy. Diabetes. 1983;32(2):64-78.
White S, Chadban S. Diabetic kidney disease in Australia: Current burden and future projections. 2014;19(8):450-8.