Sorry Dr Lustig: A calorie is still a calorie

In his book: “Fat chance: The bitter truth about sugar”, Dr Robert Lustig argues that a calorie is not a calorie or basically, not all calories from different foods are equal. He makes three arguments to support his idea. This part of his argument is central to the subsequent claim he makes that sugar and in particular fructose is the villain of obesity.

He begins by pointing out that weight loss frequently reaches a plateau because as we lose weight our resting energy expenditure falls. This resting energy expenditure is the energy required to keep our heart beating, our kidney’s filtering, our lungs breathing, our brain thinking and so on. It’s the calories you burn when you are asleep. The fall in resting energy expenditure is one of several adaptions the body makes when energy intake is restricted. Another adaptation is that the brain agrees to reduce its insistence on glucose as its sole fuel and agrees to start burning fats for fuel. If it did not do this, then the body would have to make glucose from amino acids, which would deplete body protein stores. So, several adaptations are made when energy is restricted but this has absolutely nothing to do with the calorific value of the fuels used by the body. The calorific value of amino acids, glucose, fats and ethanol are determined by metabolic pathways that allow energy to be extracted from these metabolites and these pathways are not amenable to change. Adaptations do occur but the calorific value of nutrients remains absolutely constant. A major US trial of diet composition on diet-induced reduction in resting energy expenditure found no evidence that this drop in resting energy expenditure was related to variation in dietary composition[1]

The second argument made is that within the categories of nutrients (carbohydrate, fat and protein) there is considerable variation. Thus Lustig points out that there are good fats and bad fats, that proteins vary in their quality or nutritional value and that carbohydrates range from complex molecules such as starch to simple molecules such as sugar. So lets consider fats. A typical dietary fat is made up of one molecule of glycerol (a sugar alcohol) and three fatty acids. Each fatty acid can vary in length typically from12 to 22 carbons long and the amount of hydrogen attached to each carbon can vary between 1 and 2. So, of course there is a wide variety of fats but that has no bearing on their calorific value. Since hydrogen is the atom that is central to the extraction of biological energy, we can predict exactly what the energy value of a fat can be based on its hydrogen count. A gram of fat can be made up of a lot of small chain fats or a a lesser amount of long chain fats. The calorific content will not change. Some fats are primarily designed to contribute to the structure of the body and their energy potential is not their primary function. Thus the fatty acid arachidonic acid (which the body synthesizes from fats found in vegetable oils) plays a major role in cell wall architecture and in the regulation of blood clotting and inflammation. The fatty acid eicospentaenoic acid (EPA derived from fatty fish) plays a role in the structure of nerves and in the transmission of nerve signals. Sure, fats vary in their structure and function but this has nothing to do with the theory that not all calories are the same.

The third argument he makes is that that our diet quality has changed and that we have reduced our fat intake and increased our sugar intake. However, the calorific value of fats and sugars remain constant. A calorie is still a calorie. Indeed in a major dietary intervention study of weight-reducing diets involving 811 obese subjects using the following 4 radically diets found no significant difference in weight loss.

Diet	% calories from
	Fat	Protein	Carbohydrate
1	20	15	65
2	20	25	55
3	40	15	45
4	40	25	35

 So its calories that count and a calorie really is just that irrespective of the nutrient or food it comes from.[2]

---

[1] De Jonge L et al (2012) Obesity,20(12):2384-9. Effect of diet composition and weight loss on resting energy expenditure in the POUNDS LOST study

[2] de Souza RJ et al (2012) Am J Clin Nutr. (2012) 95(3):614-25 Effects of 4 weight-loss diets differing in fat, protein, and carbohydrate on fat mass, lean mass, visceral adipose tissue, and hepatic fat: results from the POUNDS LOST trial.

Oh Sugar! Wrong about fructose

Of late, the words “toxic”, “sugar” and “fructose” have been widely used together, implying a most dangerous aspect of sugar on human metabolism. The irony is that in Greek mythology, Cronus, the Titan leader was fed so much honey that he fell into a deep sleep during which time his son Zeus killed him. The original words of the Greek legend refer to the “intoxicating effect” of the large intake of honey on Cronus. Just as we are amused but not troubled by the language or beliefs of Greek mythology, we should not be so troubled by the same nonsense reformulated in modern Californian mythology.  Honey was always held as a truly prized food: hard to harvest, made by bees through some mysterious process foreign to all other plant and animal foods, golden in colour and above all, sweet as nothing else ever known to man. The sweetness of honey was down to a combination of two simple sugars, fructose and glucose present at 55% and 45% respectively. Sugar, as we know it today, is also an ancient food but newer, relatively speaking, than honey. It is plant-derived and the very first commercial facility for the extraction of sugar in crystalline form from sugar cane or sugar beet was located on the Island of Crete.  The Arabian merchants who funded this production facility had another name for Crete, which they called Qandi. Hence the term “candy”, used today mainly in the US for sugar confectionary products. The main component of this sugar derived from cane or beet is the sugar “sucrose” which is a couplet of two sugars joined together, fructose and glucose.

Honey and particularly sugar, dominated the sweetness aspect of the human diet. That was to change in the 1980’s with the advent of high fructose corn syrup (HFCS) production, driven by simple economics. In the period up to the early 1980’s, US and global sugar prices were pretty identical and highly subject to wild fluctuations in market supply. Thus, in 1974 and 1979-1980, US and global sugar prices soared 5 fold in two separate market peaks. The advent of a new technology that could replace sugar with an identical alternative at a stable low price became a simple no-brainer. Sugar was priced out of the US markets with strict import quotas introduced in the early 1980s to maintain very high domestic sugar prices, double the global price.  HFCS was to almost completely replace sugar in the US diet. The manufacture of HFCS is technically simple. Starch, which is a polymer of glucose units, is extracted from corn and enzymes are used to first break down the starch to glucose. Half the glucose is converted to fructose, again using a simple enzyme system.  The glucose and fructose can now be blended together and the most popular blend with consumers was 55% fructose and 45% glucose, an identical blend to that found in honey.  HFCS intakes soared 8-fold in the US from 1975 to the 80s-90s. However, in recent years HFCS intake has fallen in the US and is now back to values in 1980. During the surge in the use of HFCS, that of sugar fell pro rata.

In 2004, some leading US obesity researchers published data to show that the epidemic of obesity in the US coincided with the surge in HFCS use in the food chain. Whilst most scientific commentators have dismissed this putative link, the debate rages on with thousands of doom-laden Internet postings fuelled by a handful of media-friendly scientists. The term “high fructose corn syrup” was in hindsight a foolish name to introduce since HFCS is quite simply not high in fructose, equal in fact to the level found in honey and almost equal to the level found in sugar. Fructose is the element of HFCS that has been singled out as the bad part and the research in this area leaves a lot to be desired. To begin with, humans don’t and never have consumed fructose in isolation. It is always consumed with glucose and thus experiments in humans or animals using diets with high fructose levels with no accompanying glucose are basically unrealistic. They may show what is possible but they have no bearing on what is probable. In a paper presented to the US Experimental Biology conference in 2012[1], the levels of fructose used in these diets was compared to the average daily intake of fructose by US adults. In every one of the 37 human studies and every one of the 21 animal studies, the level of fructose used exceeded the US average intake value (9% of calories). Of course the average hides high consumers so this paper also looked at the fructose intake of the top 5% of fructose consumption (15% of calories). Only 3 human and 1 animal study were at or below this very high level of intake. The majority of animal studies used as much as 55% of calories from fructose, a situation, which is impossible to envisage in the human diet except maybe in the make-believe land of milk and honey.

None of these studies needed to be funded since a natural experiment was being acted out on both sides of the Atlantic. Just as the US jacked up sugar prices to promote HFCS usage, in the EU sugar beet farmers were protected under the CAP limiting the use of HFCS to 5% of total supply. Thus beverages in the US contain HFCS whilst beverages in the EU do not. Nonetheless, obesity levels have grown dramatically either side of the ocean.  While the debate on HFCS rages on the Internet, two key organisations have pinned their colours to the mast. Both the American Medical Association[2] and the American Dietetic Association[3] have issued position statements dismissing any claim that HFCS use contributes to obesity or associated biochemical abnormalities of blood lipids or blood glucose.

---

[1] White JS (2013) Adv Nutr, 4, 246-256

[2] http://www.ama-assn.org/resources/doc/csaph/csaph3a08-summary.pdf

[3] Journal of the American Dietetic Association, (2004) 104, 255-275

Nutrition in the first 1,000 days: How strong is the data?

The first 1,000 days represents the development of a child from conception through to 2 years of age. Maternal and infant nutrition during this period has become the corner stone of many international programmes to combat malnutrition. The message relayed within this area is simple: Optimal height for age and optimal cognitive function are largely determined during the first 1,000 days. If a child suffers poor nutrition during this period, then there is permanent reduction in stature and a permanent loss of cognitive function. The UN initiative “Scaling Up Nutrition” (SUN) which has now been adopted by 45 countries has the first 1,000 days and maternal-infant nutrition as its core. However, a recent review and analysis published in the American Journal of Clinical Nutrition and led by Andrew Prentice of the London School of Hygiene and Tropical Medicine[1], would certainly demand a more rigorous review of a policy which effectively espouses the view that the first 1,000 days is the make-or-break period for physical development.

The paper begins by describing the data upon which the first 1,000 days theory is based. These data show that in 54 countries with low incomes, children are born with heights below the WHO growth standards and that this height deficit deteriorates over the first 2 years of life and then remains stable for the remainder of the study, which lasted 5 years. The authors point out that these data come from an “amalgamation of large-scale nationally representative data sets that were not collected for research purposes”.  They also point out that the original data from Africa does show some, albeit modest recovery in height between 24 and 48 months.

The second point made by the authors is that whereas most brain and neuronal development takes place in the first 1,000 days, most other tissues show significant growth after this period, all of which are driven by hormonal development, differing for males and females. The peak growth of lymphoid tissue occurs between 5 and 10 years of age while muscles, bones and reproductive organs show a surge in growth in the early to mid teens. If different organs grow at different rates at different ages, then it is logical to assume that sub-optimal nutrition can modify this growth well outside the first 1,000 days. The authors present data from Brazil, Guatemala, The Philippines and South Africa, which clearly shows recovery in height after the first 1,000 days and that this recovery is not based on any special nutrition intervention. India is exceptional in not showing any post 2-year height recovery. The research base of Andrew Prentice is in rural Gambia and over 6 decades, the growth of children from local subsistence farming villages has been recorded. The data show the expected fall in height in the first 1,000 days of these poor children. However, it shows very good recovery thereafter. Then as growth demands are increased in puberty, there is a temporary fall in height for age, which again shows recovery and plateaus in the second decade of life.  All of these data challenge the concept that the first 1,000 days is the only critical period of growth and that interventions outside that period are unlikely to have any effect.

The authors now move on to look at the actual evidence of the effects of nutritional intervention during pregnancy and early childhood. As regards pregnancy, the authors cite the Cochrane Review of 23 protein-energy supplement trials reached the following conclusion[2]: “Dietary advice appears effective in increasing pregnant women's energy and protein intakes but is unlikely to confer major benefits on infant or maternal health. Balanced energy/protein supplementation improves fetal growth and may reduce the risk of fetal and neonatal death. High-protein or balanced protein supplementation alone is not beneficial and may be harmful to the infant. Protein/energy restriction of pregnant women who are overweight or exhibit high weight gain is unlikely to be beneficial and may be harmful to the infant.” The authors also cite meta-analyses of pre-natal trials involving 17 with zinc supplementation and 49 with iron and folic acid supplementation. The outcome of the meta-analyses was that these nutritional supplements produced non-significant effects on birth outcome. As the authors point out, these trials cannot be dismissed and furthermore cannot be considered to be flawed by design. They then cite a meta-analysis of 42 trials, which involved nutritional intervention in childhood involving complimentary feeding. Whereas some benefits were seen such as reduced rates of anaemia and improved micronutrient status, the authors argue “ in the context of the current discussion their analysis underscores the fact that the range of interventions before 24 months reported to date could only make a limited contribution to reducing stunting in poor populations”. Disappointing as they may be, again these studies cannot be dismissed. It may well be, as the authors argue, that a combination of poor hygiene, infection and infestation may negate any nutritional impact and they point out that trials combining both dietary and hygiene interventions are underway.

This is a very important paper and one that will trouble the first 1000-day proponents. It is also a very thoughtful paper because it emphasizes that that the first 1,000 days remains a very important period for potential life long impacts on growth. However, it is a paper that challenges both the strength of evidence of the first 1000 days and the concept that other critical periods of growth are of lesser importance. The authors did not consider cognition as an outcome of the first 1,000-day nutritional intervention. However, they do cite 2 papers, which challenge the view that cognitive development is again, confined to the first 1,000 days.

---

[1] Prentice AM et al (2013) AJCN, 97:911-918

[2] http://www.ncbi.nlm.nih.gov/pubmed/14583907