Dietary Carbohydrate Essential For Human Nutrition?

By Eric C Westman M.D.

I read with interest the article by Dewailly et al (1) regarding diet and cardiovascular disease in the Inuit of Nunavik, but I was disappointed that no information regarding macronutrient intake was presented or considered in the estimation of cardiovascular risk. The traditional Inuit diet consists primarily of protein and fat, somewhat similar to the low-carbohydrate diets promoted in popular weight-reducing diets (2). These diets have caused concern among nutritionists because of the metabolic changes and health risks associated with limited carbohydrate consumption (3). However, in exploring the risks and benefits of carbohydrate restriction, I was surprised to find little evidence that exogenous carbohydrate is needed for human function.

The currently established human essential nutrients are water, energy, amino acids (histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, and valine), essential fatty acids (linoleic and α-linolenic acids), vitamins (ascorbic acid, vitamin A, vitamin D, vitamin E, vitamin K, thiamine, riboflavin, niacin, vitamin B-6, pantothenic acid, folic acid, biotin, and vitamin B-12), minerals (calcium, phosphorus, magnesium, and iron), trace minerals (zinc, copper, manganese, iodine, selenium, molybdenum, and chromium), electrolytes (sodium, potassium, and chloride), and ultratrace minerals (4). (Note the absence of specific carbohydrates from this list.)

Although one current recommended dietary carbohydrate intake for adults is 150 g/d, it is interesting to examine how this recommendation was determined at a recent international conference (5):

“The theoretical minimal level of carbohydrate (CHO) intake is zero, but CHO is a universal fuel for all cells, the cheapest source of dietary energy, and also the source of plant fiber. In addition, the complete absence of dietary CHO entails the breakdown of fat to supply energy [glycerol as a gluconeogenic substrate, and ketone bodies as an alternative fuel for the central nervous system (CNS)], resulting in symptomatic ketosis. Data in childhood are unavailable, but ketosis in adults can be prevented by a daily CHO intake of about 50 g. This value appears to approximate the quantity of glucose required to satisfy minimal glucose needs of the CNS and during starvation. The Group therefore concluded that the theoretical minimum intake of zero should not be recommended as a practical minimum.…about 100 g of glucose/d are irreversibly oxidized by the brain from the age of 3–4 y onward. However, this excludes recycled carbon, gluconeogenic carbon, for example from glycerol, and it does not account for glucose used by other non-CNS tissues. For example, in the adult, muscle and other non-CNS account for an additional 20–30 g of glucose daily. For this reason a safety margin of 50 g/d is arbitrarily added to the value of 100 g/d and the practical minimal CHO intake set at 150 g/d beyond the ages of 3–4 y.”

Thus, although carbohydrate could theoretically be eliminated from the diet, the recommended intake of 150 g/d ensures an adequate supply of glucose for the CNS. However, it appears that during starvation (a condition in which the intakes of carbohydrate, protein, and fat are eliminated), an adequate amount of substrate for the CNS is provided through gluconeogenesis and ketogenesis (6). The elimination of dietary carbohydrate did not diminish the energy supply to the CNS under the conditions of these experiments. Second, carbohydrate is recommended to avert symptomatic ketosis. In the largest published series on carbohydrate-restricted diets, ketosis was not typically symptomatic (7).

The most direct way to determine whether carbohydrate is an essential nutrient is to eliminate it from the diet in controlled laboratory studies. In studies involving rats and chicks, the elimination of dietary carbohydrate caused no obvious problems (8–,12). It was only when carbohydrate restriction was combined with glycerol restriction (by substituting fatty acids for triacylglycerol) that chicks did not develop normally (13). Thus, it appears that some minimum amount of a gluconeogenic precursor is essential—for example, glycerol obtained from fat (triacylglycerol) consumption. More subtle abnormalities from carbohydrate elimination might not have been observed in these studies. In addition, the essentiality of some nutrients is species-specific; therefore, these studies do not provide convincing evidence that elimination of dietary carbohydrate is safe in humans (4).

The usual way to discover the essentiality of nutrients is through the identification of specific deficiency syndromes (4). I found no evidence of a carbohydrate deficiency syndrome in humans. Protein deprivation leads to kwashiorkor, and energy deprivation leads to marasmus; however, there is no specific carbohydrate deficiency syndrome. Few contemporary human cultures eat low-carbohydrate diets, but the traditional Eskimo diet is very low (≈50 g/d) in carbohydrate (2). It is possible that if more humans consumed diets severely restricted in carbohydrate, a carbohydrate deficiency syndrome might become apparent.

When carbohydrates are eliminated from the diet, there is a risk that intakes of vitamins, minerals, and perhaps yet unidentified beneficial nutrients provided by carbohydrate-rich foodstuffs (eg, fiber) will be inadequate. There are case reports of extreme dieters who probably developed deficiencies. One dieter who only ate cheese, meat, and eggs (no vegetables) was reported to have developed thiamine-deficient optic neuropathy (14). Another dieter may have developed a relapse of acute variegate porphyria (15). However, most of the current low-carbohydrate, weight-reducing diets advocate the consumption of low-carbohydrate vegetables and vitamin supplements.

Although there is certainly no evidence from which to conclude that extreme restriction of dietary carbohydrate is harmless, I was surprised to find that there is similarly little evidence to conclude that extreme restriction of carbohydrate is harmful. In fact, the consequential breakdown of fat as a result of carbohydrate restriction may be beneficial in the treatment of obesity (7). Perhaps it is time to carefully examine the issue of whether carbohydrate is an essential component of human nutrition.


1) Dewailly E, Blanchet C, Lemieux S, et al. n−3 Fatty acids and cardiovascular disease risk factors among the Inuit of Nunavik. Am J Clin Nutr 2001;74:464–73. Abstract/FREE Full Text
2)  Shaffer PA. Antiketogenesis. II. The ketogenic antiketogenic balance in man. J Biol Chem 1921;47:463–73.
3) Westman EC. A review of very low carbohydrate diets for weight loss. J Clin Outcomes Manage 1999;6:36–40.
4) Harper AE. Defining the essentiality of nutrients. In: Shils MD, Olson JA, Shihe M, Ross AC, eds. Modern nutrition in health and disease. 9th ed. Boston: William and Wilkins, 1999:3–10.
5) Bier DM, Brosnan JT, Flatt JP, et al. Report of the IDECG Working Group on lower and upper limits of carbohydrate and fat intake. Eur J Clin Nutr 1999;53(suppl):S177–8.
6) Cahill GF. Starvation in man. N Engl J Med 1970;282:668–75.
7) Palgi A, Read JL, Greenberg I, Hoefer MA, Bistrian BR, Blackburn GL. Multidisciplinary treatment of obesity with a protein-sparing modified fast: results in 668 outpatients. Am J Public Health 1985;75:1190–4. Medline
8) Follis RH, Straight WM. The effect of a purified diet deficient in carbohydrate on the rat. Bull Johns Hopkins Hosp 1943;72:39–41.
9) Renner R, Elcombe AM. Metabolic effects of feeding “carbohydrate-free” diets to chicks. J Nutr 1967;93:31–6.
10) Renner R, Elcombe AM. Protein as a carbohydrate precursor in the chick. J Nutr 1967;93:25–30.
11) Renner R. Effectiveness of various sources of nonessential nitrogen in promoting growth of chicks fed carbohydrate-containing and “carbohydrate-free” diets. J Nutr 1968;98:297–302.
12) Renner R. Factors affecting the utilization of “carbohydrate-free” diets by the chick. I. Level of protein. J Nutr 1964;84:322–6. Medline
13) Renner R, Elcombe AM. Factors affecting the utilization of “carbohydrate-free” diets by the chick. II. Level of glycerol. J Nutr 1964;84:327–30. Medline
14) Hoyt CS, Billson FA. Low-carbohydrate diet optic neuropathy. Med J Aust 1977;1:65–6. Medline
15) Quiroz-Kendall E, Wilson FA, King LE Jr. Acute variegate porphyria following a Scarsdale Gourmet Diet. J Am Acad Dermatol 1983;8:46–9. Medline