Protein intake in type 1 diabetes

Several epidemiological studies have demonstrated that type 1 diabetes is associated with an up to 10-fold increase in the risk of cardiovascular (CV) disease [1–4]. The predominant and independent risk factors for CV events in patients with type 1 diabetes are the duration of disease, glycemic control, hypertension, and the presence of nephropathy [5–16]. The results of a cross-sectional survey in 3250 patients enrolled in the EURODIAB Complications Study [17] showed that 24% of type 1 diabetic patients had hypertension, among whom fewer than one-half (48.5%) were aware of a previous diagnosis, a similar proportion (42.2%) were on treatment, but very few 11.3% were both treated and controlled. Incidentally, in the EURODIAB follow-up study, among those who were treated, a modest increase in the proportion of those controlled for hypertension was found (41 vs. 32%, P = 0.048) which, however, disappeared after adjustment for age [12]. Around one-third of patients newly diagnosed with type 1 diabetes develop persistent microalbuminuria or overt diabetic nephropathy within 20 years since the onset of diabetes [18]. Given the high cost of treatment of these chronic complications, nonpharmacological strategies, such as dietary interventions, have been encouraged to reduce the burden of CV disease [19,20]. The effect of a reduced protein intake has reached interest as a strategy for the prevention of hypertension and renal disease in type 1 diabetes [21]. Although the progress in this field is exciting, clinicians need to consider that experimental and clinical studies generated controversial results and, thus, some effects of dietary protein intake in type 1 diabetes are still unaddressed. In this context, the new analysis by EURODIAB investigators, published in the present issue of the Journal [22], adds new insights on the protein controversy in type 1 diabetes. Although the pathophysiological basis of diabetic complications has yet to be fully clarified, hyperglycemia appears to play a pivotal role. The development of hypertension in type 1 diabetes is associated with poor glycemic control during the first decade of diabetes, and the development of hypertension in turn appears to increase susceptibility for renal disease [15]. An excessive protein intake might, through gluconeogenesis, increase glucose production. Gannon and Nuttall [23] reported that in 1915, Janney calculated that 3.5 g of glucose could be produced for every gram of nitrogen excreted in the urine as the result of a beef protein meal. Beef protein is 16% nitrogen, thus, 1 g of nitrogen is excreted for every 6.25 g of protein. In other words, 56% of ingested beef protein can be converted to glucose [23,24]. This has been the basis of the common statements that a proportion of ingested protein is converted to glucose and that blood glucose levels are directly affected by protein intake. However, this belief has been challenged. After Gannon and Nuttall's report [23], some researchers showed that the ingestion of protein by individuals with and without diabetes did not result in an increase in blood glucose levels. Westphal et al.[25] treated seven individuals with 50 g protein, 50 g glucose, and 10, 30, or 50 g protein along with 50 g glucose in random order. Ingestion of 50 g protein alone did not change the plasma glucose concentration. The various amounts of protein ingested with 50 g glucose also did not affect glucose response as compared with that observed with 50 g glucose alone [25]. Consistent findings were obtained also by Nuttal et al.[26], Khan et al.[27] and Gannon et al.[28]. Recently, the picture was complicated by the potential role of insulin availability on the protein metabolism in patients with type 1 diabetes [29,30]. With insulin deficiency, gluconeogenesis is increased and protein catabolism augmented. The accompanying rise in glucose levels is, therefore, because of an increased conversion of ingested protein into glucose and to a decreased glucose removal rate (Fig. 1). In individuals with diabetes who had insulin withheld for 24 h, there was a three-fold to four-fold increase in liver glucose output after protein ingestion [31]. However, in the presence of insulin, alanine uptake by the liver is virtually zero, and hepatic glucose production falls by 85% [30]. In other words, in type 1 diabetic patients with adequate insulin replacement and well controlled blood glucose levels, minimal amounts of hepatic glucose are expected to be released into the general circulation after protein ingestion.FIGURE 1: Influence of insulin availability on the relationship between protein intake and blood glucose levels. The metabolism of glucose and dietary proteins in healthy subjects is depicted in the left panel. In patients with insulin deficiency (right panel), rise in glucose levels is due to an increased conversion of ingested protein into glucose and to a decreased glucose removal rate.Available data are scanty to explore the potential role of glycemic control on the relationship between protein intake and development of type 1 diabetes complications. In the new analysis of the EURODIAB Prospective Complications Study (PCS), Altorf-van der Kuil et al.[22] observed that the risk for incident hypertension and microalbuminuria by tertiles of protein intake seems to be unaffected by glycemic control, as expressed by levels of HbA1c. However, the researchers of EURODIAB PCS [22] did not plan a formal investigation to test whether glycemic control is a potential mediator of the relationship between protein intake and incident comorbidities, leaving the issue unresolved. Diet is a main determinant of blood pressure (BP) and certain foods seem to have a direct role in the reduction of the risk of incident hypertension. Dietary interventions to prevent hypertension include low alcohol and salt intake and low-fat dairy products [32,33]. In the last few years, interest has grown into dietary patterns and macronutrient intakes, including protein. Whether or not protein content in the diet is relevant for BP control remains to be established. A recent systematic review [34] including 15 observational studies, 13 biomarker studies and 20 trials, suggested a small beneficial effect of protein on BP, especially for vegetal protein. Trial data suggest that increased intake of protein may be beneficial to BP, although no clear dose–response association had been identified [34]. Similarly, a pooled analysis of 40 trials including 3277 participants in total [35], showed that compared with carbohydrate, dietary protein intake was associated with significant changes in mean SBP and DBP. Both vegetal and animal protein was associated with significant BP changes when compared with carbohydrate while BP reduction was not significantly different when plant protein was compared directly with animal protein. Unfortunately, data on type 1 diabetes are scanty and the results reported by the researchers of the EURODIAB PCS are certainly welcomed [22]. From their findings it is clear that intake of total, animal and plant protein was not significantly related to incident hypertension. After adjustment for several confounders (including age, sex, diabetes duration, total fat, carbohydrates and alcohol), the odds ratio (OR) and 95% confidence intervals (CI) for hypertension were 0.86 (0.60–1.25) and 0.91 (0.59–1.43) in the middle and upper tertile of total protein. Similar results were obtained for animal and plant protein [22]. Additional analyses also suggested that protein intake was not significantly associated with mean SBP during follow-up [22]. Some renal hemodynamic changes, including increases in transcapillary pressure gradients and increased glomerular filtration rate (GFR) and renal growth, occurs during the early stages of insulin-deficient diabetes in humans [36–38] and have been proposed as the main factors in the progression of diabetic glomerulosclerosis [39–41]. Some experimental models demonstrated that continued high protein intake or aminoacid administration are associated with renal hypertrophy and increasing GFR [42,43] and seem to aggravate the extent and progression of glomerular damage [40,43,44]. It has been proposed, therefore, that further enhancement of diabetic hyperfiltration induced by dietary protein may result in acceleration of the progression of diabetic glomerulosclerosis [39]. Just recently, Meek et al.[45] observed that high-protein diets increase glomerular cell death and inflammation, exacerbating early indicators of diabetic kidney disease. They documented that the negative response to high-protein diet (Fig. 2) was mediated by the receptors for advanced glycation end-products (AGEs).FIGURE 2: Possible mechanisms involved in glomerular cell death and inflammation. As documented by Meek et al. [44], the effects of high-protein diet on the progression of diabetic kidney disease seem to be mediated by enhancement of formation of advanced glycation end-products and overexpression of their receptors. AGE, advanced glycation end-product.Despite the experimental suggestions that protein ingestion negatively affects renal function, data from epidemiological studies are controversial. A meta-analysis of five studies [46] showed that the use of a low-protein diet delayed the progression of renal disease independent of blood glucose or BP levels. Conversely, Watts et al.[47] reported that dietary protein intake was similar in type 1 diabetic patients with or without nephropathy [47]. Moreover, a relationship between the amount of protein ingested and the progression of nephropathy was not observed by Nyberg et al.[48]. Notably, in all of the above studies, protein intake was in the range of usual dietary intake and rarely exceeded 20% of the calories. Only a previous analysis of the EURODIAB Complications Study was specifically conducted among type 1 diabetic patients who reported protein consumption of more than 20% of total calories. It showed that prevalence of normoalbuminuria, microalbuminuria and macroalbuminuria was 20, 23 and 32%, respectively [49]. However, this trend was more pronounced in patients with hypertension and/or poor glycemic control [49]. Interestingly, the new analysis of EURODIAB PCS data [22] was restricted to normotensive type 1 diabetic patients. After a mean follow-up of 7 years, 135 incident cases of microalbuminuria were recorded among 1045 participants. After adjustment for potential confounders (including HbA1c as an indirect measure of glycemic control), intake of total, animal and plant protein was not significantly related to incident microalbuminuria [22]. Thus, protein commonly consumed by European normotensive patients with type 1 diabetes seems to marginally affect the risk of renal disease [22]. Concluding our reflections on the effects of dietary protein intake in type 1 diabetes, another aspect of the analysis by Altorf-van der Kuil et al.[22] needs to be mentioned. Understanding the potential mechanism of action for protein intake in type 1 diabetes has proved to be a challenging area of research, and considerable work is still required to prove the potential impact of different protein sources (i.e. fish, dairy, whey, grain and soy) in these patients. The results of the new analysis of the EURODIABB PCS [22] do not support the suggestion that animal or plant protein dietary intake habits significantly affect the risk for hypertension and microalbuminuria. Nonetheless, different dietary proteins may exert different effects on BP and the kidney. For example, whey proteins seem to improve metabolic and BP control and to reduce the expression of both inflammatory and oxidative stress markers [50]. Moreover, a diet including a high amount of fish protein significantly reduced the risk for microalbuminuria in young type 1 diabetic patients [51]. ACKNOWLEDGEMENTS This study was funded in part by the Fondazione Umbra Cuore e Ipertensione, Onlus, Perugia, Italy. Conflicts of interest None of the authors of this study has financial or other reasons that could lead to a conflict of interest.