[2017] Lore of Nutrition: Challenging Conventional Dietary Beliefs

by Tim Noakes

  It is time to admit that the theory has failed. We need to adopt an open mind if we are ever to discover the real cause(s) of the current global epidemic of obesity, diabetes and coronary heart disease, all of which are likely caused by the same factors.

  Four days later, on 21 September 2012, what appeared to be a follow-up attack found its way into the Mail & Guardian under the headline ‘Is Tim Noakes the Malema of medicine?’26

  Earlier that month, Dr Martinique Stilwell had interviewed me for the article she was writing for the weekly newspaper. Probably best known for her memoir about growing up at sea, Thinking up a Hurricane, Stilwell is a medical graduate of the University of the Witwatersrand and currently works as an anaesthetist. She has no more formal training in nutrition than any other South African doctor.

  Stilwell began her article by genuflecting to the complaints of the UCT cardiologists. The gist of her argument was that while I had successfully exposed the non-science behind the sports-drink scam and had developed the Central Governor Model of Exercise Regulation, the sole evidence I had for promoting the LCHF eating plan were my personal experiences, specifically my marked weight loss and the improvement in my running. ‘Noakes said he experiments on himself and judges the results by how he feels,’ Stilwell wrote. ‘He encourages lay people to do the same, which doctors say is irresponsible.’

  A more impartial journalist would have reported that my dietary conversion occurred only after I had read the remarkable book by doctors Westman, Phinney and Volek. The New Atkins for a New You was full of science about the LCHF diet, much of it derived from a series of world-class, peer-reviewed scientific papers published by the authors over the past 20 years. My personal experience after my December 2010 Damascene moment had simply confirmed the hard science contained in their book. Those experiences also contradicted much of what I had been taught by the anointed professors that Stilwell seemed so keen to champion.

  Returning to the cardiologists’ complaints, Stilwell said that since I began to question the benefits of statins, there had been ‘murmurs among cardiologists that Noakes is the Malema of medicine, a man with a hoarde [sic] of followers and considerable media sway, who is capable of producing charismatic, easy to hear and probably irresponsible solutions to very complex problems’.

  The fact is that there is nothing complex about the diet-heart and lipid hypotheses promoted by these cardiologists and accepted by Stilwell. They simply boil down to: eating saturated fat raises blood cholesterol concentrations, which clog coronary arteries and lead to death from heart disease. Nothing particularly complex here, except the difficult question that cardiologists choose to ignore: the role of sudden plaque rupture as the decisive event causing heart attack and sudden death in those with even a tiny amount of ‘clogging’ in their coronary arteries.27

  Stilwell interviewed a signatory of the Commerford letter, Dr Dirk Blom, who is described as ‘an academic at UCT’s lipid clinic’. Blom confirmed my contention that most people who suffer heart attacks have normal blood cholesterol concentrations. How do we explain this inconvenient finding if an elevated blood cholesterol concentration is the single most important factor causing heart attack, as Blom and his colleagues endlessly repeat? How can cholesterol cause heart attacks when present in the bloodstream in ‘normal’ amounts that are too low to justify treatment with cholesterol-lowering statin drugs? The inability to grasp this paradox is known as cognitive dissonance.

  In fact, UCT lipid expert Professor David Marais, another signatory of the Commerford letter, was the co-author of a large study of blood cholesterol concentrations and other risk factors in South African men and women living in Soweto. The so-called Heart of Soweto study found that the majority of black South Africans who have had heart attacks have low blood cholesterol concentrations with an average value of only 4.1 mmol/L (see Figure 3.1).28 Others have reported essentially identical findings in North Americans hospitalised with acute or prior heart attacks.29

  Figure 3.1

  Frequency distributions of total plasma cholesterol, plasma LDL, plasma HDL and plasma triglyceride concentrations (mmol/L) in Sowetan residents of (a) African, (b) White European, (c) mixed and (d) Indian descent treated for heart attack in the Heart of Soweto study. Note that only in the Indian group is the average total plasma cholesterol concentration greater than the value of 5 mmol/L that is considered ‘dangerous’, requiring the prescription of statin therapy. But in all groups, plasma HDL cholesterol concentrations are low and triglyceride concentrations are elevated, indicating the presence of insulin resistance

  These values are too low to qualify for treatment with cholesterol-lowering statin drugs that supposedly reduce heart-attack risk. But if these patients’ blood cholesterol levels are too low to justify treatment to prevent heart attacks, what caused their heart attacks? The obvious conclusion is that the cholesterol theory is false. I will discuss this further in Chapter 17.

  Using industry-speak designed to obfuscate, Blom attempted to explain: ‘If cholesterol is not markedly elevated it is not a sole predictor of heart disease, but it is part of the risk calculation.’ Which means what, exactly? In the first place, these patients do not have blood cholesterol levels that are ‘not markedly elevated’; they have values that are too low to warrant treatment according to the internationally accepted guidelines developed by Blom and others.

  Surely Stilwell should have interrogated his evasion more critically if she were truly intent on providing her readers with intelligently moderated information? She should also have asked Blom to declare any conflicts of interest that might influence his apparent religious devotion to the promotion of this form of drug treatment. My own simple investigation found that since he first became interested in cholesterol and heart disease in 2002, Blom has published 33 articles on the topic, including nine reports on large-scale trials of different statins to which he personally contributed. The results of most of these specific trials were published in world-leading medical journals, indicating that the studies were very well funded. It is fair to say that Blom’s research career is dependent on the largesse of the pharmaceutical industry. There is a saying that you do not bite the hand that feeds you. Of course, the UCT Faculty of Health Sciences also benefits financially from Blom’s tight relationship with the makers of statin drugs.

  Interestingly, Blom has co-authored seven articles on the cholesterol-lowering drug evolocumab, produced by Amgen. A one-month supply of the drug costs around $1 200, and the FOURIER study found that 74 people would need to take evolocumab for two years for just one to show benefit in terms of avoiding a cardiovascular event. Clearly this is not a drug that has much relevance to the greater South African public.

  Blom’s special academic interest is the condition called familial hypercholestrolaemia (FH), a genetic ‘disease’, the key feature of which is an elevated blood cholesterol concentration on a familial (hereditary) basis. It is the disease that cardiologists use as the absolute proof that ‘bad’ cholesterol clogs the coronary arteries, leading to CHD. In Chapters 7 and 17, I present evidence showing that ‘bad’ cholesterol probably has little, if anything, to do with the coronary artery clogging found in some, but certainly not all, people with FH. The scientific evidence strongly suggests that people with FH develop arterial ‘clogging’ for exactly the same reasons as everyone else (without FH) – as a result of those with IR eating diets high in carbohydrates, sugar and processed foods. If true, this suggests that cardiologists like Blom should perhaps prescribe low-carbohydrate diets rather than statins for their patients with FH. Not that I suspect the pharmaceutical industry would be too enthusiastic to fund any studies testing this possibility.

  In the Stilwell article, Blom continued by saying that statins ‘unequivocally’ save lives and that those who do not benefit are simply low-risk individuals who should not have been on the drugs in the first place. I have learnt that scientists only use words like ‘unequivocal’ when the evidence is the opposite – that is,
when it is entirely equivocal. As already mentioned, statin drugs produce miniscule benefits with the risk of significant side effects in perhaps as many as 20 per cent of subjects. The drug on which Blom has focused most of his attention is hardly a lifesaver and is not likely to be without worrying side effects.

  The benefits for women seem to be even less, perhaps non-existent. Cardiologist Dr Barbara Roberts writes in her book The Truth about Statins: ‘In women under the age of sixty-five who don’t have established vascular disease, we have zero evidence that statin treatment lowers their risk for having a cardiac event … And women experience more side effects from statins than men do.’30 As a female doctor interested in health promotion, Stilwell should have interrogated her male colleague on this topic. Instead, she allowed him to mislead her into believing that what applies to men must equally apply to women. With regard to statin therapy, this is certainly not the case.

  Stilwell went on to quote a cardiologist who wished to remain anonymous: ‘I feel sorry for Tim. He likes to take on a challenge, but when it comes to statins I think he’s painted himself into a corner.’

  I have no doubt that sometime in the not-too-distant future those of us who have warned that the diet-heart and lipid hypotheses are the greatest scams in the history of modern medicine, and that statins are among the most ineffective drugs ever produced by the pharmaceutical industry, will be vindicated.

  Stilwell concluded her article with a clarion call to safe orthodoxy:

  Noakes has swung from high carbs to no carbs [this is not true]. The most widely accepted research suggests something in between. Avoid transfats, sugar and refined carbohydrates, eat a calorie-restricted, balanced diet with whole grains, protein and healthy fats, exercise in moderation and, if you are thirsty, drink water.

  I seriously question the merits of Stilwell’s dietary advice, taken, as it is, from the classic 1977 US dietary guidelines.

  How, for example, does one exist forever on a calorie-restricted diet? By definition, a calorie-restricted diet will cause progressive weight loss leading to relentless hunger and, ultimately, death. The exception is if one is eating LCHF, which in 1970 Professor John Yudkin described as being ‘in reality a low-calorie diet’.31 In Chapter 4, I discuss Ancel Keys’s Minnesota Starvation Experiment and what it reveals about the psychological and behavioural effects of trying to live on a low-calorie, low-fat diet of the kind that Stilwell proposes.

  The dietary recommendations that Stilwell promotes fail to take into account the role of the brain controller (the appestat) in the regulation of hunger. Carbohydrates stimulate hunger, whereas fat and protein satiate it. After 40 years of practice, I think we can finally recognise the traditional dietary guidelines for what they are: a complete failure. As the saying goes, ‘however beautiful the strategy, you should occasionally look at the results’.

  My suspicion is that Stilwell’s article formed part of a game plan directed by, among others, her acquaintance Jacques Rousseau. As described in Chapter 1, Rousseau was using his blog to show me up as a scientist who had ‘lost his way’. Between the release of the updated edition of Challenging Beliefs in March 2012 and the first pot shots in the press, on 21 July 2012 I was invited by Professor Diane McIntyre from the School of Public Health and Family Medicine in the UCT Faculty of Health Sciences to speak to faculty members about cholesterol and heart disease. McIntyre has an international profile and has since been appointed executive director of the International Health Economics Association. She had personally adopted the LCHF diet for a time and wanted to give me the opportunity to speak on the scientific reasons why I was no longer advocating a low-fat diet.

  I presented a standard, hour-long talk explaining why the epidemiological (associational) evidence linking elevated blood cholesterol concentrations to increased risk of heart disease was, at best, weak; that associational studies linking dietary saturated-fat intake to higher rates of heart disease provided equally weak evidence; and that any associational link was, in any case, disproved by a number of prospective RCTs, all of which showed that low-fat diets do not reduce the risk for developing heart disease. And finally, as fully discussed in Chapter 17, that there was no evidence from autopsy studies that pre-morbid blood cholesterol concentrations predict the extent of coronary artery disease present at death. This material would form the basis for the 25 000-word chapter I would write for The Real Meal Revolution a year later, in July 2013.

  After the talk, the questions were fairly banal and predictable.

  An example: How can you say that blood cholesterol concentrations do not predict heart-attack risk when ‘we’ (the anointed) know that cholesterol causes heart disease?

  As I presented in my talk and as discussed in Chapter 17, markers of IR/T2DM are much better predictors of future heart-attack risk. In any case, T2DM and its associated medical complications, and not heart attacks, are the illnesses that we must prevent if we wish to improve human health and save global medical services from bankruptcy and collapse. However, because the low-fat diet is the most likely cause of the obesity/T2DM epidemic, it is difficult, if not impossible, for those who have advocated this fallacy for the past 40 years to suddenly find the courage to acknowledge and apologise for their gross error.

  Another example: How can you prescribe a high-fat diet knowing that it will cause blood cholesterol levels to rise and must therefore increase heart-attack rates?

  There is actually no evidence that cholesterol causes heart disease. Instead, as I argue in Chapter 17, the introduction of population-based screening for blood cholesterol concentrations is what leads to arterial disease, because it teaches doctors and dietitians to prescribe high-carbohydrate diets (and statins) to people with normal (i.e. below 7.5 mmol/L) blood total cholesterol concentrations but who are insulin resistant. Prescribing high-carbohydrate diets for those with IR can produce only one result: an epidemic increase in T2DM, with its associated scourge – disseminated obstructive arterial disease of all the key blood vessels in the body.

  Another question came from an industry-funded hypertension expert: How can you say that statin drugs are so ineffective when ‘we know’ that lowering blood cholesterol concentrations with statin drugs reduces heart-attack risk by at least 30 per cent?

  My counter-questioning made me realise that the expert had not been taught the difference between the relative (30 per cent) and absolute (~1 per cent) benefits supposedly provided by these drugs. He also did not understand, because he had not been taught to think that way, that both values are meaningless unless one understands an individual’s baseline risk for developing the disease of interest. He also did not appear to understand that the only drug effect of real value is whether a specific drug makes people live longer by reducing all-cause mortality (i.e. reduces the risk of dying prematurely from all diseases considered together, not just as a result of suffering a heart attack).

  There is no value in taking a drug that reduces the risk of suffering a heart attack if it simply causes people to die at a younger age from something else, such as cancer. A drug that causes death from cancer in all who take it will certainly prevent all deaths from heart attack. But that might not be the best reason to justify prescribing that drug. Unfortunately, this is often the reasoning cardiologists use to prescribe statins – that these drugs marginally reduce one’s risk of suffering a heart attack. The key question is whether or not these drugs also reduce all-cause mortality, and it is now abundantly clear that most do not.32

  Importantly, a 100 per cent increase in baseline risk for developing a particular disease is irrelevant if the risk is trivial. Certainly, buying two Lotto tickets increases the chance of winning the jackpot by 100 per cent. But the probability of winning even with two tickets is still only one in tens of millions. Similarly, if anyone has a one in 1 000 chance of suffering a heart attack in the next five years, using a drug that (allegedly) reduces that risk by 100 per cent (to one in 500) still means that 499 of 500 people will receive no benefi
t from the use of that drug. Yet all 500 will be exposed to the detrimental effects of that drug, whatever those might be.

  Typically, the pharmaceutical industry lavishes money on its industry-friendly key opinion leaders, of which the hypertension expert was one, to promote the ‘benefits’ of their drugs by using relative benefits (the larger number). When talking about the risks of these drugs, however, they either ignore the topic or use absolute risk (the much smaller number). And no one ever talks about baseline risk, because for all diseases – even lung cancer in smokers – for any single individual, the risk is usually quite low.

  For example, people considered at highest risk for a future heart attack experience only nine events per 10 years per 100 subjects, and those at lowest risk just two per 10 years.33 Which means that after 10 years, 91 of the 100 people considered to be at highest risk for a cardiac event will not have experienced any such event. And 98 of those 100 considered to be the healthiest will also not have experienced any cardiac problems.

  Perhaps predictably, predictive models used by cardiologists to estimate future risk overestimate risk by as much as 150 per cent in men, but less in women (up to 67 per cent).34 Naturally, the effect of such overestimation will be twofold: more healthy patients will be terrified that they are about to drop dead in the next few days because their blood cholesterol concentrations have suddenly risen above 5.01 mmol/L; and the over-prescription of a lifetime of statin drugs to healthy people, the majority of whom will not ever suffer cardiac events.

  I came away from the lecture satisfied that I had made my points and that no one in the audience had punched any holes in my arguments. The reception was not overtly hostile and was certainly not a portent of things to come, when four months later I would lock horns with Professor Jacques Rossouw in the hubristically termed ‘Centenary Debate’ organised by the same faculty, in front of an audience oozing animosity.