Horizon: Sugar vs. Fat

by Zoë Harcombe

The big news story last week was an experiment done with identical twins on the BBC Horizon programme. A friend of mine, Kate, found an interesting twist – this was not a new experiment for Chris and Alexander (Xand) Van Tulleken.

A 2008 Channel 4 programme followed the brothers, both doctors, travelling to Chukotka (the farthest northeast region of Russia) to do some experiments. Chukotka is an icy peninsula, washed by the Bering Sea, a section of the Pacific Ocean, and part of the Arctic Ocean. It covers a vast area, but with a population of 55,000. It is one of the increasingly rare parts of the world where people can still be found living a primitive lifestyle.

Episode 3 of series 1 shows Chris following the local diet of whale, walrus & seal (with a bit of reindeer for variety!), while Xand follows a processed food diet from the only store selling ‘western’ food in the region. The last four minutes of the programme descends into the astonishing medical ignorance, shared by 99% of doctors, that HDL and LDL are good and bad cholesterol respectively (they are not even cholesterol). The twins get terribly excited that Chris’s ratio of ‘good’ to ‘bad’ cholesterol gets 30% better, while Xand’s gets 30% worse. Relative risk errors, causation vs. association thus compound not knowing the difference between lipoproteins and cholesterol, but the conclusion is that Chris’s diet was the best. Chris’s diet was high in fat and protein and devoid of carbohydrate.

Horizon on Wednesday 29th January 2014, was billed as a unique experiment to test whether or not sugar or fat is worse for us. Arguably the original experiment was best – real food vs. processed food. High fat vs. high carb. How we used to eat vs. how we eat now. It had all the right ingredients, but let’s look at the recent Horizon programme.

The Experiment

The idea was very simple. Chris was given a virtually fat free/high carb/sugar diet for one month, while his identical twin, Xand, was given a virtually carb free/high fat diet for the same period of time. It was also pitched a bit as America vs. the UK, as America is starting to think that sugar is the baddie and the UK still thinks that fat is the baddie.

Enter Amanda Ursell, nutritionist, who lays out visually on a table what the two brothers can eat. I was highly amused to see Amanda lay out Chris’s table – explaining that all of this food “bread, bagels, pasta, rice, potatoes, any description of breakfast cereals and unlimited fruit and veg ultimately breaks down into blood sugar“.  And she’s right. And that’s the main point that public health advisors just don’t get. They are telling us to cut down on sugar, while advising us to eat more carbohydrate. Carbohydrate is, or breaks down into, sugar.

With the jelly/sweets and other fat free junk Chris was allowed, his table looked remarkably like the government eat badly plate: 33% starchy foods; 33% fruit & veg; 8% junk and he just needed beans/pulses as non dairy protein and low fat dairy and he’d be on the perfect public health diet.

Xand’s side of the table featured cheese, meat, butter, burgers, chicken with the skin on, double cream, mayonnaise, etc. Immediately we see a problem stemming from nutritional ignorance. Xand is not allowed vegetables, but he’s allowed unlimited dairy – which, as rule of thumb, is approximately 5% carbohydrate (hard cheeses and fluid dairy defining the extremes). Burgers invariably have wheat, starch, rusk and/or sugar. Xand is seen later in the programme having meat slices, which invariably contain dextrose/sugar and mayonnaise invariably contains sugar, so Xand could also have been having sugar/starch inadvertently.

Performance

A couple of extreme experiments are done to test brain and body function in the twins. Stock broking is used to simulate a brain test and the brothers go out with the Team Sky cycling coach, Nigel Mitchell, to test the body. There are two fundamental problems:

1) The brain will fuel on glucose or ketones, if glucose is not available. The body will fuel on glucose or fat (dietary or body) if glucose is not available. Xand is unlikely to have been in ketosis/fuelling from fat if he is having carbohydrate in dairy products and processed meats/sauces. Xand does well considering that his body is not being given a fair chance to compete on an alternative to glucose.

We’ll come on to the results soon, but the fact that Xand started at a body fat percentage of 26.7% and only lost 3.5kg tells me that he was not in ketosis. I would have expected Xand to lose that weight in one week, not four, if he were genuinely on a zero carbohydrate, ketogenic, diet.

2) It takes time for the body to adapt to ketones/fat when glucose has been readily available for 35 years. (The Daily Mail article has their age). Expecting Xand to perform as well on new fuel, as his brother does staying on his life time favoured fuel is unreasonable.

Performance, however, was not what worried the general audience. The two issues that worried people on twitter, and that I received a number of queries about during and following the programme, were:

1) Muscle mass;

2) Insulin production and type 2 diabetes.

Let’s look at each:

The results – muscle mass

Xand started off at 26.7% body fat and Chris at 22.6% body fat. Both pretty flabby – as the mug shot photos confirmed. Approximately 36 minutes into the programme, the results are shared:

Xand is told that he lost 3.5kg (approximately 8lb) and this is said to be 1.5kg of fat and 2kg of muscle. Chris is told he has lost 1kg – also claimed to be half fat (0.5kg) and half muscle (0.5kg).

The BodPod measures fat mass and fat-free mass. More accurately, it measures air displacement and fat mass and fat-free mass are estimated from prediction equations (thank you Dr Chris Easton – @easto82).

I have written to Horizon asking them the following:

Dr Richard Mackenzie says to Xand “You’ve lost 2kg of muscle mass and that isn’t so healthy.”

Please can you help me understand how the statement to Xand can be made? My understanding of the BodPod is that it can measure fat mass and fat-free mass. Thus the programme can estimate (within the accuracy of the BodPod) how much fat has been lost and how much fat-free mass has been lost but the latter is not all muscle. The latter will include water and it is virtually guaranteed that water will be lost on a low carb diet, as glycogen will be depleted and water is stored at approximately four parts to every one of glycogen.

I look forward to your explanation

I’ll let you know the reply.

The results – blood glucose

The programme repeats the blood tests done at the start of the experiment – first testing cholesterol. Chris narrates “We thought that, because Xand was eating so much fat on his diet, his levels would be so much higher. What is amazing is that they were nearly exactly the same as they were at the start of our diets. In fact, there was little, or not change, for either of us.”

What is amazing is how quickly they glossed over cholesterol! No difference so, instead of exploring this, they just ignored it. I would have expected measurements to be different simply because of the known margin for error in the cholesterol test (c. 15%). Chris (if not both brothers) had a prejudice (literally to pre-judge) that Xand’s cholesterol would “be so much higher” and yet it wasn’t – explanation please?!

The final test was insulin – a hormone that regulates blood glucose levels. The programme didn’t explain what was happening at this point but it looked like the brothers were doing a glucose tolerance test. This involves the ingestion of a glucose solution and then blood tests measure subsequent blood glucose and insulin levels. Dr Richard Mackenzie said to Chris (the carb twin) “Your body’s ability to produce insulin improved.” The doctor brothers found this counter intuitive. So, Mackenzie went on to say “Your body has probably just got used to dealing with the sugar, the glucose intake and therefore responding by producing insulin.” Chris’s take on this is “Because I’ve been eating loads of sugar I’ve become better at managing it.” Mackenzie corrects with “You’ve become better at producing insulin.”

Xand asks if this is good or not and Mackenzie says in the short term it is good, but in the long term it might produce a problem. You’re not kidding! The long term likelihood is type 2 diabetes. When the body says ‘enough is enough’ – I cannot cope with this intake of carbohydrate/sugar anymore. I cannot continue to produce enough insulin to bring my blood glucose back into the normal range and I cannot do it with the frequency with which you are chucking carbs into me.’

Then we turn to Xand and he is told that “Your body is not responding to insulin as well as it did. If you eat too much fat, that can stop your body responding to insulin [how?!] and it can also tell your body to produce more glucose” [again how?! Is Mackenzie confusing the presence of fat with the absence of carbohydrate?] Mackenzie continues his warning [in a very worried tone] “your blood glucose has climbed from 5.1, which it was before the diet, to 5.9. Now you’re only 0.2 away from being pre-diabetic.”

I have three points:

1) There is a known 20% margin for error in blood glucose tests. Xand’s first test of 5.1, with a 20% margin of error, could have been anything between 4.1 and 6.1 and hence his second reading is well within the known margin of error. His fasting blood glucose levels could have gone down over the month and we would not know this because of measurement margins for error.

2) Xand has had essentially no glucose for one month. He’s had small amounts of lactose in dairy products and some starch/sugar in processed meats/mayonnaise etc, but he’s had very little glucose to deal with. He’s then been subjected to a glucose tolerance test, which would give him a massive dose of glucose in one hit. The carb twin, Chris, is not surprisingly able to cope with this (produce insulin) because that’s what he’s been doing several times a day for one month. Xand’s body is going to need time to adapt back to glucose just as it needed time to adapt away from it.

3) Xand is not pre-diabetic at 5.9, notwithstanding margins for error etc. Normal blood glucose levels have been redefined, just as normal cholesterol levels have been redefined, just as normal blood pressure levels have been redefined. Healthy people are non-profitable. Sick people are profitable. The more people who can be placed in a ‘sick’ category the better.

In 1999, the World Health Organisation announced a “major change” in the diagnosis of diabetes. The diagnostic level of the fasting plasma glucose concentration would be lowered from 7.8 mmol/l (140 mg dl) to 7.0 mmol/l (126mg dl) and the whole blood level benchmark would be lowered from 6.7 mmol/l (120 mg dl) to 6.1 mmol/l (110 mg dl). Over one million Americans became ‘diabetic‘ overnight.

“The meeting was made possible by generous financial support from Bayer, UK; Bayer, Germany; Novo Nordisk, Copenhagen, Denmark; and The Institute for Diabetes Discovery, New Haven, USA”. I bet it was!

The conclusion

Xand concludes “Basically what I get out of this is – I have to avoid the processed food: the doughnuts; the ice cream; the cheesecake. That sort of 50:50 fat and sugar mixture I cannot stop eating and that’s the problem.”

Clearly I like any conclusion that says don’t eat processed food. Spelling out – “Eat Real Food!” would have been even better. Xand needs to stop at the first part of his conclusion as well – avoid processed food full stop. Not just the 50:50 fat and sugar mixtures, but the breads, cereals, bagels and starchy foods that adorn the government’s ‘eatwell plate’.

It would have been better still if Xand, as a 10 year doctor/6 years in medical school, could have explained why fat and sucrose combined are so bad for humans. It’s because real food doesn’t contain both – with one exception – avocado. Nature provides carb/proteins (any real foods that Chris was eating) and fat/proteins (any real foods that Xand was eating). It’s food manufacturers who have worked out that the sucrose/fat combo is irresistible, moreish and fattening – to rats or humans. That’s why fake food needs to be avoided.

Chris’s conclusion was “Where I end up is all faddish diets – all faddish diets – are wrong and misguided. And doing exercise is really important.” [how did an exercise conclusion follow from this experiment?!]

The “don’t eat processed food” message had already been lost with 15 minutes of the programme left.

The bias

The BBC is supposed to be balanced. This programme was not balanced and the bias against fat is so ingrained that the BBC may not even have realised this.

The mind (stock broking) experiment was presided over by Professor Robin Kanarek. Her views were very clear – “glucose is the best fuel for the brain“; “memory will be significantly compromised without enough carbs in the diet“; “a high carb diet will facilitate memory” and so on. Why was Dr Emily Deans not the expert on hand? “Ketosis for the body means fat-burning (hip hip hooray!). For the brain, it means a lower seizure risk and a better environment for neuronal recovery and repair.” Or even – why not have both Kanarek and Dean to provide balance?

The body (cycling) experiment was presided over by Nigel Mitchell from Team Sky cycling. He favours porridge for breakfast and states “your body needs the sugar. It needs the carbohydrates“. Why not balance him with Peter Brukner, the low carb/high fat coach celebrating a 5-0 Ashes victory with the Australian cricket team? Or Djokovic’s gluten-free coach?

Dr Robert Lustig, the current lead global expert on sucrose and fructose (who pays tribute to Professor John Yudkin from decades earlier) was interviewed. I was surprised and disappointed at how quickly Chris dismissed Lustig’s input. This did not display a doctor who had gone into this experiment with an open mind. Chris dismissed the fructose studies claiming they had unrealistic intakes of fructose and tossed away “The insulin hypothesis” as untested.

I had the privilege of seeing Dr Richard Johnson  present on the topic of fructose in 2009. On his opening slide he showed a typical continental breakfast tray with 40 grams of fructose in the cereal, juice, fruit, jam, croissant and coffee. And that’s just breakfast. The points being made by sucrose & fructose experts (not infectious disease – Chris), such as Johnson and Lustig is that normal sugar consumption is now abnormal and humans are paying the price with obesity, type 2 diabetes, cancer, heart disease, Alzheimers and so on. As for the insulin ‘hypothesis’ – the lipid hypothesis is also unproven. Indeed the COMA report openly admits: “There has been no controlled clinical trial of the effect of decreasing dietary intake of saturated fatty acids on the incidence of coronary heart disease nor is it likely that such a trial will be undertaken.” (COMA, 1984).

When Professor Susan Jebb was interviewed, however, there was no subsequent dismissal. Indeed the opposite happened. Xand reinforced Jebb’s views “And we’re in no doubt about this link – that being fat makes a lot of people ill.” “Absolutely“, says Jebb” it increases your risk of heart disease and cancer and of diabetes.” Jebb then describes her experiments that have looked at what Xand helpfully calls good fats (monounsaturated fats) and bad fats (saturated fats) [Puh-lease!] The final conclusion? Changes found were “modest”, “surprisingly small” when trying to look at the impact of different macronutrients or types of fat. This may explain why these experiments were useless.

I loved an insightful comment on Dr John Briffa’s excellent post on this Horizon programme. Check out ‘Jennifer’ who observed that “Lustig’s contribution was trivialised…..located in a fun fair with silly music (so he must be barmy)” while “Jebb was shown as a white-coated academic in a university setting (so she must be sensible).”

Lustig was the only anti sugar voice on the programme and the only expert dismissed before the viewer could digest his points. Amanda Ursell cautioned Xand that he would get bad breath and be constipated, but gave Chris no warnings about his sugar diet. Professor Kanarek was pro-carb, with no dismissal. Nigel Mitchell was pro-carb, with no dismissal. Richard Mackenzie issued severe cautions to Xand, but not to Chris.

Even the title of the programme showed bias. Instead of putting the foods in alphabetical order, the programme was Sugar vs. Fat. The people vs. Larry Flynt. The Crown vs. Ronnie Biggs. The party on trial is named second – that would be fat.

But then, the programme opened with Chris setting out the current belief “When I trained as a doctor it was clear that fat was the enemy because it raises your cholesterol, then blocks up your arteries causing strokes and heart disease” and that is one heck of a position to move away from. Sadly Chris represents 99% of the doctors in the UK – similarly entrenched in a dangerous and non evidence based mindset.

Comments:

This is a very thought provoking article. It is a shame the experiment did not run a little longer and there was a stronger bias on sugar.

The blood glucose test is not very good for checking the actual level of insulin in the blood. The reason is because someone who is in ketosis or has limited net carbs and protein does not produce large amounts of insulin. If they are suddenly given a large amount of glucose the body is not used to it and so the response is not as strong or as fast as someone who regularly consumes large amounts of net carbs. (The test for insulin should have been to measure insulin directly with antibodies rather then test how 2 subjects reacted to a glucose test. This just adds an assumption and is poor science.)

(The brain can not run just on ketones because it always requires a very small amount of glucose. The glucose requirement for the brain is much lower during ketosis. This small amount can be provided by protein or amino acid conversion to glucose. The rest of the fuel can be ketones.)

Bayer and Novo Nordisk are both companies that make insulin or diabetic related products. It is financially better for them to have more customers by lowering the standard by which diabetes is judged.

Ketogenic diets in the wild are quite different to the ones people choose to create themselves so the micronutrients are likely to be different as well. People who have existed for centuries on a mainly ketogenic diet will also have the sense to combine what meats, organs and other foods that are needed to prevent disease. (For instance, in places where there is not much sun light people should all be vitamin D deficient but they might then consume the liver oil of fish that contain vitamin D etc. The skin of certain animals contains vitamin C, where there may not be vegetable/fruit containing sources.)

Endurance athletes often burn nearly all their glycogen stores because the body has pathetically low stores of carbs. In tests of physical fitness an endurance athlete in ketosis burns glycogen much more slowly and has easy access to their fat stores. (People can carry a hundred thousand calories as fat stores and perhaps only 2000C worth of glycogen.)

Just deciding fat or sugar is not enough for health because you have to know what comes with the sugar or the fat as well - these are star players as well. Whole foods are always a mixture of carbs, fats and proteins. Refined products can be pure.

Combining refined products together can make very tasty foods that have the right sweetness, fat content, saltiness, texture and flavours. These are usually much better to eat than whole foods - the brain stimulation is much more intense in the pleasure detecting areas. They are addictive and repeated stimulation keeps on reinforcing the want to have them. (It is quite hard to go back to whole foods because they are normally quite bland by comparison.)