Carbohydrate restriction as the default treatment for type 2 diabetes and metabolic syndrome

Abstract

Dietary carbohydrate restriction in the treatment of diabetes and metabolic syndrome is based on an underlying principle of control of insulin secretion and the theory that insulin resistance is a response to chronic hyperglycemia and hyperinsulinemia. As such, the theory is intuitive and has substantial experimental support. It has generally been opposed by health agencies because of concern that carbohydrate will be replaced by fat, particularly saturated fat, thereby increasing the risk of cardiovascular disease as dictated by the so-called diet-heart hypothesis. Here we summarize recent data showing that, in fact, substitution of fat for carbohydrate generally improves cardiovascular risk factors. Removing the barrier of concern about dietary fat makes carbohydrate restriction a reasonable, if not the preferred method for treating type 2 diabetes and metabolic syndrome. We emphasize the ability of low carbohydrate diets to improve glycemic control, hemoglobin A1C and to reduce medication. We review evidence that such diets are effective even in the absence of weight loss.

• Diabetes
• carbohydrate restriction
• metabolic syndrome

The forward advance of medicine and science is marked by several detours during which an established orthodoxy resists a minority viewpoint even in the face of compelling evidence. The resistance of the medical establishment to carbohydrate restriction as a strategy for treating diabetes is particularly unusual in that 1) the treatment is intuitive, based on the mechanism of the disorder, 2) has substantial experimental support and 3) has a historical underpinning as the traditional treatment before the discovery of insulin. In addition, and most relevant to the current symposium honoring Uffe Ravnskov, is that the primary justification for not reducing carbohydrate as the first line of attack in diabetes is the concern that the dietary carbohydrate that is removed will be replaced by fat and particularly saturated fat and that this will increase the risk of cardiovascular disease. Dr. Ravnskov has been in the forefront of the movement to bring to light the limitations of the low fat-cholesterol paradigm and the general failure of clinical trials to fulfill its promise 1, 2. In fact, recent studies show that replacement of carbohydrate with fat, even saturated fat reliably improves cardiovascular risk. More generally, carbohydrate restriction, long considered a stratagem for weight loss, is beneficial for health and specifically targets those physiologic markers that define the metabolic syndrome (MetS) 3–5.

In practice, the choice of therapy must be decided by the patient and physician. We present here some of the scientific evidence that makes carbohydrate restriction a reasonable candidate as the default nutritional approach to metabolic syndrome and diabetes, that is, the one to try first.

The epidemic of diabetes

The most salient feature of the epidemic of obesity and diabetes is the increase in carbohydrate and the decrease in fat consumption. Figure 1A and 1B show the parallel increases in diabetes incidence and carbohydrate consumption 6. For men, the absolute amount of fat was reduced and the absolute amount of saturated fat was similarly decreased significantly. A presumed decrease in exercise is also frequently cited, but why a decrease in exercise should cause specific increase in carbohydrate consumption or vice-versa is secondary to the prima facie association in Figure 1. This association points to the hypothesis that over consumption of carbohydrate is connected to continued high levels of glucose and insulin, which may cause, or at least exacerbate an insulin resistant state which is the characteristic of metabolic syndrome and frank diabetes.

Figure 1.  Nutrient consumption in the United States. National Health and Nutrition Education Study (NHANES). Data from reference 6. B. Incidence of diabetes in the United States. Data from Centers for Disease Control and Prevention (http://www.cdc.gov/diabetes/statistics/incidence/fig2.htm)

Diabetes and metabolic syndrome

Glucose is the major stimulus for insulin secretion. Generally anabolic in its effects, insulin enhances protein synthesis and biases organisms towards storage of fat and carbohydrate while simultaneously inhibiting catabolism of nutrient stores. Fat accumulation is initially brought about by inhibition of hormone-sensitive lipase and long-term, by the increased expression of lipoprotein lipase, providing substrate for TG synthesis. Insulin increases recruitment of GLUT4 (glucose) receptors and inhibits gluconeogenesis and glycogenolysis and promotes hepatic glycogen storage. States of insulin resistance lead to disruptions in these processes, notably persistent gluconeogenesis and enhanced lipolysis leading to hyperglycemia and increased plasma fatty acids that are not oxidized which, in turn, may exacerbate the insulin resistant state. In this sense, diabetes is sensibly described as a disease of lipids as much as carbohydrate intolerance. The increase in fatty acids leads to disruptions in TG synthesis and processing and the abnormalities collectively referred to as atherogenic dyslipidemia (high TG, low HDL, small dense LDL, increased apoB/apoA1 ratio (Reviews: 7, 8). The reduced effectiveness if insulin signaling in the resistant state leads to compensatory overproduction, so that hyperinsulinemia occurs until the pancreas is no longer able to overcome insulin resistance in peripheral tissues.

Carbohydrate restriction and diabetes

Figure 1B shows the effect of five weeks on a low carbohydrate diet (% CHO:fat:protein = 20:50:30) in patients with type 2 diabetes. A control diet provided 55% carbohydrate (55:30:15) and there is obvious benefit in glycemic control in the low carbohydrate intervention 9. For comparison, the effect of fibre 10, more commonly recommended by health agencies is shown in Figure 1A. It is important to note that the experiments in Figure 1A were carried out under eucaloric conditions, that is, benefit to glycemic control in people with diabetes does not require weight loss or a restriction in energy intake.

That these changes can be maintained was shown by Dashti et al. who demonstrated a dramatic reduction in blood glucose in patients with diabetes on a ketogenic diet over the course of 56 weeks 11. Nielsen et al. reported that a 20% carbohydrate diet was significantly superior to a 55–60% carbohydrate diet with regard to bodyweight, glycemic control and reduction in HbA_1c. Insulin and oral hypoglycemic agents were reduced or discontinued. In a subsequent publication reporting 22 months of follow-up, the HbA_1c remained improved at 6.9% 12. Notably, after the first six month period, two thirds of patients in the high-carbohydrate changed to a low carbohydrate diet and showed corresponding improvement in bodyweight and glycemic control. The benefits of carbohydrate reduction on glycemic control have been widely reported and it is recognized by the American Diabetes Association (ADA) that “dietary carbohydrate is the major determinant of postprandial glucose levels 13.” The ADA, nonetheless counsels against low carbohydrate diets (< 130 g/d) on the grounds that “that they eliminate many foods that are important sources of energy, fiber, vitamins, and minerals and are important in dietary palatability.” Energy, has increased during the diabetes epidemic (Figure 1A) and is just what should be reduced. Few would consider that having to take a vitamin supplement has the same qualitative character as diabetic medications. An on-line low-carbohydrate support group has recently passed the 100 000 member mark, and a survey of this group found few complaints about palatability 14.

Perhaps most telling in considering low-carbohydrate diets is the reduction in medication that is, in fact, considered, a pre-requisite. In a study of carbohydrate restriction, Yancy et al. 15 found that 17 of 21 patients with type 2 diabetes reduced or discontinued diabetes medication. Similar results have been reported by Boden 16 and Nielsen 12.

Carbohydrate restriction and metabolic syndrome

There is a continuum in the severity and time course for the appearance of effects of insulin resistance. The metabolic syndrome (MetS) which is clinically defined as any of three of the markers overweight, hyperglycemia, hyperinsulinemia, atherogenic dyslipidemia, high blood pressure and an expanding number of other states 17, indicates that, depending on the K_m of individual processes and tissues, disparate symptoms may appear at different times. The association, however, is assumed to reflect a common cause, generally believed to be disruption in insulin control and some researchers consider that that insulin resistance syndrome is a more appropriate name.

An extensive review of the literature brought out that the markers of MetS are precisely those targeted by diets that restrict carbohydrate 5. Again, the underlying mechanism, that reduction in chronic high glucose returns glucose-insulin cycles to normal, is manifest in many different processes. Table I shows that, compared to a low-fat diet as a control, low-carbohydrate diets are better at ameliorating the markers of MetS. Notable in the Table is that benefits were seen in interventions designed to maintain body weight, although small amounts of weight were lost anyway in one of the low carbohydrate arm, a phenomenon reported by others 18. The hypothesis, then, was that the response to carbohydrate restriction was a virtual definition of MetS.

Table I.  Comparison of low CHO vs. LF diets on markers for Metabolic Syndrome Data shown in bold indicate low CHO (or mod-PROT) shows greater improvement in marker than LF; plain, LF is better. Original references in Volek & Feinman 5. Data below the double line are from experiments carried out under conditions designed to maintain body weight.

				CHANGE
Reference	Subjects	Duration	Diet	CHO (g/d)	weight (%)	HDL (%)	TAG (%)	TAG/HDL (%)	Glucose (%)	Insulin (%)	DBP (mm)
Brehm et al. 2003	Obese Women	6 mo	Low-CHO	41–97	−9.3	13.4	−23.4	−32.4	−9.1	−14.8	−5
			LF	163–169	−4.2	8.4	1.6	−6.3	−4.0	−23.0	−1
Sondike et al. 2003	Overweight Adolescents	12 wk	Low-CHO	37	−10.7	8.7	−40.5	−45.2
			LF	154	−4.1	4.2	−5.4	−9.2
Samaha et al. 2003	Obese Men/Women	6 mo	Low-CHO	150	−4.5	0.0	−20.2	−20.2	−8.6	−27.3
			LF	201	−1.4	−2.4	−4	−1.6	−1.6	5.6
Foster et al. 2003	Obese Men/Women	1 yr	Low-CHO	ad lib	−7.3	18.2	−28.1	−29.5
			LF	ad lib	−4.5	1.4	0.7	−2.6
Volek et al. 2004	Overweight Women	4 wk	Low-CHO	29	−3.9	1.3	−23	−28.3	−3.8	−8.8
			LF	186	−1.4	−8.6	−11.2	−4.2	1.3	23.2
Sharman et al. 2004	Overweight Men	6 wk	Low-CHO	36	−5.6	−3.3	−44.1	−42.3	−5.8	−41.5
			LF	224	−3.6	−6.6	−15	−8.3	−5.2	−28.1
Brehm et al. 2004	Obese Women	4 mo	Low-CHO	69	−10.8	16.3	−37.3	−46.1			−9
			LF	174	−6.8	4.5	−10.3	−14.2			−3
Meckling et al. 2004	Obese Men/Women	10 wk	Low-CHO	59	−7.7	12.2	−29.4	−37.1	−8.0	−28.7	−6.1
			LF	225	−7.4	−15.4	−25.4	−11.8	−10.2	−3.3	−5
Stern et al. 2004	Obese Men/Women	1 yr	Low-CHO	120	−3.9	−2.8	−28.6	−26.8
			LF	230	−2.3	−12.3	2.7	29.6
Yancy et al. 2004	Obese Men/Women	24 wk	Low-CHO	30	−12.3	9.8	−47.2	−51.8			−6
			LF	198	−6.7	−2.9	−14.4	−12.1			−5.2
Aude et al. 2004	Obese Men/Women	12 wk	Low-CHO	ad lib	−6.2	−2.6	−23.2	−21.1
			LF	ad lib	−3.4	−7	−10.5	−3.8
Seshadri et al. 2004	Obese Men/Women	6 mo	Low-CHO	113	(−8.5)	−2.4	−7.4			−40
			LF	198	(−3.5 kg)	−2.4	−2.3			11.2
McAuley et al. 2004	Obese Women	8 wk	Low-CHO	41	−6.9	0.9	−38.8	−44.2	−5.9	−39.3
			LF	172	−4.4	−6	−17.5	−15.1	−0.1	−28.4
			Mod-PRO	130	−5.8	−4.1	−33.9	−31.8	−3.9	−24.4
Sharman et al. 2002	Normal Weight Men	6 wk	Ketogenic	46	−2.8	11.5	−33.0	−39.9	−0.2	−34.2
		6 wk	Low-Fat	271	0.5	0.0	−5.3	−5.3	1.8	13.0
Volek et al. 2003	Normal Weight Women	4 wk	Ketogenic	43	−2.0	32.0	−30.2	−47.2	−1.9	11.6
		4 wk	Low-Fat	249	−1.3	−7.7	3.8	0.5	−5.3	18.7
Allick et al. 2004	Type 2 Diabetics	2 wk	Ketogenic	0	0	23.5	−43.9	−55	−16.9	−16.7
		2 wk	Low-Fat	775

A 12 week prospective trial from Volek's laboratory tested this hypothesis 3, 4: 40 overweight men women with MetS were randomly assigned to an ad libitum very low carbohydrate ketogenic diet (VLCKD) (1510 kcal:%CHO:fat:protein = 13:59:28) or a low fat diet (LFD) (1521 kcal:%CHO:fat:protein = 56:24:20). Subjects following the VLCKD significantly reduced body weight and adiposity, improved glycemic control and insulin sensitivity and had consistently more favorable triglycerides (TG), HDL-C and total cholesterol/HDL-C ratio responses compared to the LFD. In addition to these markers for MetS, the VLCKD subjects showed greater leptin sensitivity and more favorable responses to the markers for atherogenic dyslipidemia: postprandial lipemia, ape B, apo A-1, and the apo B/Apo A-1 ratio. Subjects following the VLCKD showed enhanced endothelial function and reduced vascular inflammation indicated by improved postabsorptive and postprandial flow-mediated dilation and greater reductions in circulating chemokines and adhesion molecules.

The mechanism of the effects carbohydrate restriction

The rationale of carbohydrate reduction as a treatment for diabetes and metabolic syndrome is the idea that dietary carbohydrate, beyond its role as a source of energy, serves as a control element, either directly via glucose or fructose or indirectly through the effects of insulin and other hormones. In this view, dietary fat plays a passive role, and the insulin-resistant state, at least in part, represents down-regulation due to hyperinsulinemia as in other hormonal systems. Reduction in dietary carbohydrate and associated reduction in carbohydrate-induced insulin is then proposed to return the impaired regulation described above to normal levels.

The clear mechanistic link between dietary carbohydrate and insulin effects is strengthened by the emerging understanding of the downstream effects of both glucose and insulin. The lipid effects of the carbohydrate-insulin axis are coordinated, in part through interacting pathways involving the liver X receptor (LXR), the glucose response element binding protein, ChREBP and a sterol response element 1c (SREBP-1c) (see, e.g.19). Similarly, forkhead transcription factor FoxO1 is a down-stream target of the insulin-mediated regulation of gluconeogenesis and other components of carbohydrate metabolism 20 and interacts with SREBP-1c pathways.

Weight loss

The literature records numerous studies of the effectiveness of diets based on carbohydrate restriction for weight loss in normal subjects, and patients with metabolic syndrome or diabetes and such interventions are at least as effective as low-fat diets and generally substantially better (Reviews: 5, 21, 22); in studies where there is good compliance and accurate determination of actual intake, low-carbohydrates do drastically better. Several studies have shown that there is a spontaneous decrease in caloric intake when carbohydrates are restricted. More controversial is the idea that such diets are also more metabolically inefficient, that is, provide greater weight loss, calorie-for-calorie. An exhaustive study of the literature concludes that such effects have been adequately demonstrated 21 and an explanation of how it could happen theoretically has been proposed 23.

Inflammation

The breakdown in cellular function that leads to overproduction of free radicals and microvascular damage is of mechanistic importance in understanding the role of carbohydrate restriction in the symptoms of diabetes, retinopathy, renal failure and peripheral neuropathies. Brownlee has conducted pioneering work leading to a comprehensive theory whereby high glucose, by increasing flux through the electron transport chain, leads to generation of superoxide from molecular oxygen at the level of complex III-coenzyme Q. Mitochondrial overproduction of superoxide leads to increased activity in the polyol pathway, intracellular production of glycation end-product precursors, protein kinase C activation, and increased hexosamine pathway activation 24. Figure 3 shows the effects of carbohydrate restriction on several inflammatory markers in Volek's study described above 3. In addition to those shown in the figure, the carbohydrate-restricted diet produced significant reductions in TNF-α, IL-6, and PAI-1 while the low fat group showed little change.

Figure 2.  Effect of diet on plasma glucose. Meal points are Breakfast (B), lunch (L), dinner (D, snacks (S, S1, S2). A. Mean plasma glucose concentrations for 13 patients with type 2 diabetes followed two diets for six weeks: a diet containing moderate amounts of fiber (total, 24 g; 8 g of soluble fiber and 16 g of insoluble fiber), as recommended by the American Diabetes Association (ADA), and a high-fiber diet (total, 50 g; 25 g of soluble fiber and 25 g of insoluble fiber), containing foods not fortified with fiber (unfortified foods). Figure redrawn from reference 10. B. Mean plasma glucose concentration before (triangles) and after 5 weeks on control diet (light circles: (CHO:fat:protein = 55:30:15)) or 5 weeks on lower carbohydrate diet (dark circles: (20:50:30)). Data from reference 9.

Figure 3.  Changes in markers of inflammation in people with metabolic syndrome with diet. Forty overweight men women with MetS were randomly assigned to an ad libitum very low carbohydrate ketogenic diet (VLCKD) (1510 kcal:%CHO:fat:protein = 13:59:28) or a low fat diet (LFD) (1521 kcal:%CHO:fat:protein = 56:24:20). Data from reference 3F.

Benefits of carbohydrate restriction do not require weight loss

Whereas diabetes is frequently associated with obesity, a causal link has not been demonstrated. Nonetheless, obesity is commonly cited as a cause of insulin resistance and, in fact, weight loss is as much a target of standard therapy as glycemic control 13. Obesity is a response, however, and although nutrient excess per se can be a primary cause, any mechanism for obesity must posit a role for insulin. The question then is, given how hard it is to lose weight by any methods, is carbohydrate restriction effective for treatment of diabetes and MetS in the absence of weight loss? Gannon and Nuttal have been instrumental in showing that this is so (Figure 2; reference 9). Even under hypocaloric conditions, individual changes in markers for MetS did not correlate with individual weight loss 3, 4.

In addition, the effects of obesity on insulin resistance and the associated increased fatty acids are downstream from the primary impact of diet and, given the mechanistic arguments above, emphasis of treatment on glycemic control rather than weight loss would seem reasonable. More generally, the pervasive effects of carbohydrate restriction shown in Figure 1 support a hypothesis whereby insulin resistance represents a down-regulation of hormonal response as a result of persistent high levels of insulin, a feature common to other hormonal systems and one in which diabetes, obesity and the components of MetS can be seen as parallel effects of hyperinsulinemia and/or hyperglycemia (Figure 4).

Figure 4.  Theories of mechanisms of insulin resistance.

Can we move forward against the epidemic of obesity and diabetes?

Science is an inherently cooperative venture but like all human activities it is not free from controversies and disagreements. Historical examples exist in which conflicts cannot be resolved by the usual processes and an entrenched establishment can repress minority opinion. It is argued here that the pervasive emphasis on fat reduction for diabetes and metabolic syndrome or, more precisely, the refusal of government and health agencies to incorporate data from carbohydrate restriction studies into current thinking, has not been productive; if it has not caused the diabetes epidemic, it has done little to stop it. There are many explanations for this state of affairs: personal hostility towards Dr. Atkins and, not least, the persistence of the diet heart-hypothesis. The deconstruction of this hypothesis by Uffe Ravnskov and others thus has radiating impact.

Carbohydrate restriction is offered as an alternative to pharmacologic and traditional nutritional approaches. Where the latter are successful, they are to be applauded but the obesity and the diabetes epidemic is considered to still be with us and in need of new approaches if we are to move forward. Moreover, thirty years of “concerns” about low carbohydrate diets from the medical community have produced an adversarial atmosphere with little evidence of risk. There has been a degree of contention on both sides but, in the end, what we know about diet and health is far exceeded by what we don't know and nobody has a corner on the truth in this field. Dialogue, however, must come from the majority and there are ample grounds for its initiation.