Menu Close

TV twins pitch sugar against fat but the combination is a killer

Yummy. BBC/Joanna Barwick

This could have been such a yawn. Is sugar the “toxic drug” or fat “the deadly ingredient”? Is it yet more popularised quasi-science, claiming life-threatening consequences from foods we eat every day?

But Sugar v Fat, a BBC Horizon programme that took two identical twin doctors and put them on extreme low-fat and low-sugar diets for a month, was certainly different – possibly even a first for a diet programme. It was quick-moving, highly entertaining, and its conclusions were largely (but not completely) correct.

The big problem with attempts to show science on human subjects on television is that for science to show anything beyond the blindingly obvious or totally banal normally requires studies of large numbers of individuals. People can be enormously different, so we often need huge numbers to be sure that a result is true and a treatment or diet really does work. That’s even before we begin to consider statistics.

So choosing a pair of identical twins – Xand and Chris Van Tulleken – to compare the effects of low-carb and low-fat diets was inspired, because they are a genetic and metabolic match. And following two prevailing threads of dietary advice in the US and the UK, Xand took on a low-carb diet while Chris got low-fat.

Following any kind of rigid dietary advice for a month is very tough, which is why good dietitians don’t tend to give out inflexible dietary prescriptions, but the twins really got stuck in. As participants they were pretty average, just a little overweight perhaps, but nothing serious. Their body fat contents, estimated by the slightly erratic “BodPod” whole-body plethysmograph (which measured changes in body volume) were 22% and 26%, which is a bit high for men.

Mental agility

From early on, it was apparent that Chris was enjoying his low-fat diet, with its reciprocal high carbohydrate content and the energy it gave him. But while Xand started by enjoying the Atkins-style meats, cheeses, creams and other high-fat foods, he was soon missing his carbs. He felt sleepy and lacked energy. His restricted carbohydrate supply limited glucose supply to his brain and he fared much less well in a test of mental agility and memory, playing the New York Stock Exchange on-screen.

So round one was a convincing victory for the low-fat/high-carb UK-style diet (Chis made pots of money on the stock exchange, while Xand lost track and struggled badly). This shouldn’t be a surprise to anyone with knowledge of nutrition: the brain functions almost exclusively on glucose as its fuel.

Physical performance

Similarly, when it came to physical performance, a fortnight into the diets, Xand soon flagged in a cycling trial up a hill (even with a Tour de France trainer) while Chris stormed on to the top. The high-carb/low-fat UK-style diet gave Chris tremendous energy. Any athlete or exercise-lover knows this: high intensity exertion and untrained muscles need glucose. This can only really come from dietary carbohydrate.

The full story is a little more complicated than it was possible to show in the programme, as muscles can be trained to burn fat for endurance activity. But even Tour de France cyclists, we learned, eat huge amounts of the ultimate energy food – porridge. It releases energy more slowly and consistently. So round two also went convincingly to the low-fat/high-carb UK-style diet.


Round three was the denouement, the metabolic round, with a series of physiological tests at the beginning and the end of the month.

The backdrop to this was an interview with the ubiquitous Professor Robert Lustig, the latest in a very long series of high-profile individuals who have attacked sugar as the root of all health-evils. He recently generated publicity from a European survey which purported to show that consuming more sugary drinks caused diabetes. This old chestnut has been debated for years. But Lustig’s study, like many before, failed to measure body fat, which is the true culprit. It was pointed out on the programme that there was no clear published evidence that sugar has any uniquely damaging role, and Diabetes UK recently released a paper to that effect.

So how did the twins do in the metabolic round? A slightly clumsy eat-till-you-are-full laboratory experiment appeared to show that Chris would eat more calories from his low-fat/high-carb diet than Xand from his low-cab/high-fat diet. But this wasn’t a well-designed experiment. An alternative interpretation, not discussed, was that Chris simply enjoyed his high-carb foods more. A better design would be to have offered them both the same buffet selection to choose from. An important point was made though – fat gives you twice as many calories per mouthful as sugar.

There was no difference in their blood cholesterols, but worryingly Xand’s blood glucose had actually risen, not fallen, on the low-carb diet, and his insulin levels were paradoxically high. Essentially, he became “insulin-resistant” – the first step towards Type 2 diabetes. This is exactly the result experienced nutritional scientists would expect, even if someone doesn’t gain weight. High fat diets promote diabetes because insulin stops working well.

The weigh-off

The twins’ weight changes on their different diets were not as well explained. Chris remained within 1kg or so of his starting weight, despite eating as much high-carb food as he liked for a month, while Xand lost 3.5kg on th low carb diet. A rather complicated explanation was offered, based again on the BodPod measurements, to suggest that Xand had lost 2kg of muscle. That would be really worrying, but improbable.

The true explanation for his weight loss is much simpler: an extreme low-carb diet, as favoured by Lustig and before him Atkins, completely depletes the body of carbohydrate stores over a few days. Carbohydrate is stored in the liver and in muscle in the form of glycogen, where each gram of carbohydrate is stored with about 4g of water. On average, people have about 400g of glycogen stores, so its total weight is about 2kg. A tallish biggish man like Xand is likely to have nearer 3kg in total, and all that is lost rapidly on a low-carb diet.

Large studies comparing low-fat and low-carb diets, aminly in women, have consistently shown about 2kg greater weight loss, at least initially, with the low-carb diet. It has been known for 40 years that this difference is water. A good trick if your name is Atkins and you can persuade people to pay for it. I wish the Horizon programme had explained that.

The conclusion for metabolic risks and weight gain, shown nicely by some very persistent rats and a sad-looking cheesecake in Paul Kenny’s lab at Mount Sinai Hospital, is that the bogey man is not an overall high-fat diet, nor an overall high-sugar or high-carbohydrate diet.

The problem is our extraordinary weakness for foods which present us with a 50:50 mix of the two. That type of food does not exist in nature or in our evolutionary history. But it is sitting in a fridge in almost every American home as a grazing food. On both sides of the pond, the most popular doughnut is the plain, sugar-glazed type, which is almost exactly 50:50 fat:carbohydrate.

Takeaway anyone?

There is much more dietary tweaking if you seek perfection, for example the hazards from trans-fatty acids, now reduced by trade agreements but still not banned in UK so likely to creep back into our foods, and high intakes of fructose readily converted into liver fat.

This is one thing Lustig is right about: we can’t let fructose intakes in the UK climb to reach the levels in US. And then there’s the small matter of some 27 essential nutrients needed in all human diets that is mostly ignored by manufacturers and caterers.

So it’s a simple take-home message then: avoid manufactured foods which present carbohydrate and fat in equal proportions, and if you want to stay awake, active and smart, go for a UK-style higher carbohydrate, lower fat diet. Well done, Horizon! Two stars and a wish!

Want to write?

Write an article and join a growing community of more than 139,400 academics and researchers from 4,242 institutions.

Register now