Hi All and possibly Brian,
I'd just like to direct anyone who is remotely interested in a diametrically opposite view to my own to the comment down on this post.
This is an essential read. I think we all have to realise that there MUST be two sides to a given controversy and, while I wouldn't want to make a habit of it, both sides of the argument need an airing.
For Brian's benefit,
Yes, my diet is exactly as described, no joking.
You ask:
"Are all these experts wrong, as well as the expert advisory panels on cardiovascular disease?"
Yes!
A much better question:
Which hormone converts a vascular smooth muscle cell to an osteogenic cell (calcium phosphate secreting) in the vascular media?
Answer: Insulin
The NCEP answer: It's a statin deficiency! (what was the question? Oh never mind)
Thanks for the list of references, but I already feel there is far too much cholesterol hypothesis bashing on my blog. To go through them one at a time would be tedious and far better commentators than I have done so many times.
I note in passing that you cite the PROSPER study. Which particular cancer would you prefer to trade for your heart attack? Or perhaps you'd like to choose one for a grandparent?
Your concerns for my well being are touching, but I'll stick as I am thanks.
Peter
Subscribe to:
Post Comments (Atom)
56 comments:
He puts his point of view well and in a reasonable fashion, even if most people on this site don't subscribe to his point of view. Peter, when are you going to post an overall summary of your beliefs, splitting them into a) what you feel sure of, and b) what you believe intuitively, but don't yet have enough evidence to back up?
Remember - Karl Popper always believed in taking on the enemy on their strongest points!
Blair
PS Enjoying the site a lot - keep it up.
Hi Blair,
Yes, the problem with a blog is that it gets ideas and snippets as they come to me or based on what comes through the news. Currently almost all of my reading around the blog is about IHD, particularly the causes. The basic injury to the structure of the artery and how the repair goes wrong to give atheroma is fascinating. I don't have it well enough referenced to put a hard overview together but it's coming along slowly.
The more I look at Dr Davis' approach the more I divide it up in to those effects which improve insulin sensitivity (matter a lot, low carb, low "sugar", D3, omega threes), those which affect inflammation (sticking plasters like statins) and those which look like red herrings to improve lipid numbers (fibrates, vegetarianism).
My assumption about people like Ornish and Esseltyn is that they use hunger to improve insulin sensitivity and control glycaemia, which seem to be what matters. This is wrapped up in mumbo jumbo about the miracles of plants.
It's very obvious that there is a huge range of macronutrient ratios at which humans can live healthily, provided they maintain insulin sensitivity and avoid grains. I happen to think LC is the easiest.
As I say, I'm looking at what goes wrong, within the limits of the research which is published. A lot of it is in papers which are hard to get at without a medical library, to which I don't have easy access!
Peter
Your blog is must read Peter.
wccaguy/acipher
Peter....but regarding fidh oils have you read this:
http://raypeat.com/articles/articles/fishoil.shtml
Let us know what you think about his explanations on why fish oils are not exactly a good thing.
Thanks
It is funny - these discussions take on a "theological"/sectarian feel sometimes. The commitment that people feel to their position and the desire to "save" heretics lie us is quite startling. It also becomes a clash of presuppositions. Even in the face of contrary data people cling to the "fat is bad" dogma and will interpret the world according to that paradigm.
I came to the same conclusion: it mostly comes down to insulin. Diets that reduce insulin such as vegans (due to low absorption rates of non-starchy plant food), hypocaloric diets or low carb diets are protective against arteriosclerosis. Those that lead to hyperinsulinemia such as vegetarian diet high in refined starch or sugar (e.g. India) or medium fat high carb diet, lead to worse arteriosclerosis and higher heart risk.
There is a recent study on t2 diabetes where the goal was set at maintaining a normal blood glucose range, at the cost of maintaining a higher insulin level than in the control, and the outcome was worse!
Stan (Heretic)
Peter,
The study you linked
http://www.ncbi.nlm.nih.gov/pubmed/17606264?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_RVDocSum
says, quote:
"Emerging data suggest that the two pathways may converge on the regulation of phosphate transport and extracellular inorganic phosphate (Pi) concentrations. Two antagonist enzymes governing Pi metabolism are alkaline phosphatase (ALP) and the ectonucleotide pyrophosphatase/phosphodiesterase-1 (also known as PC-1): while ALP is up-regulated in calcified diabetic arteries, PC-1 is also implicated in the genesis of insulin resistance. Therefore, we suggest that the functional interactions between ALP and PC-1 may link insulin resistance to vascular calcification."
Since my knowledge of biochemistry ended somewhere halfway through the above passage, I would like you to comment on that. I find it curious that Dr. Kwasniewski has always been claiming that arteriosclerosis is caused very specifically by the excessive "Pentose Shunt" metabolic mode, or (synonymously) "Pentose Phosphate Pathway" (PPP). I wonder if that "regulation of phosphate transport" may have something to do with it?
Regards,
Stan(Heretic)
Peter, "ideas and snippets" as they come along is the point of a Blog, as far as I believe. The aim is to motivate thought and discussion, not present full papers. So keep on doing what you're doing, it's great stuff.
And as one commenter pointed it, it does at times seem like more of a theological discussion than a rational, scientific one. And I imagine it will become even more so as the theoretical underpinnings of the faddish, "fat is bad, carbs are great" hypothesis increasingly crumble with the wash of new evidence to the contrary.
Misty,
re Ray Peat's ideas, particularly trying to see what is what in the text you mention:
I have a number of problems with this document. The first is that the references in the text are in the format (Author/s year). In the reference section they are listed alphabetically by the title of the journal in which they appeared. This makes following them a nightmare.
I checked up on three of them, hitting fairly at random on those supporting this section which purports that fish oil is toxic to the liver:
"Unfortunately for that argument, it's now known that the triglycerides in the blood are decreased because of the fish oil's toxic effects on the liver (Hagve and Christophersen, 1988; Ritskes-Hoitinga, et al., 1998). In experiments with rats, EPA and DHA lowered blood lipids only when given to rats that had been fed, in which case the fats were incorporated into tissues, and suppressed mitochondrial respiration (Osmundsen, et al., 1998)."
Hagve and Christophersen, 1988
This one actually says they encourage ketogenesis and stop hepatic steatosis (fatty liver). Sounds OK to me.
Ritskes-Hoitinga, et al., 1998
This one fed rabbits 40% of calories as fat and up to 20% of calories as fish oil. Rabbits eat grass.
Osmundsen et al 1998
As far as I can make out this last paper is looking at the metabolism features of omega 3 fatty acids, which are clearly specific to omega three fatty acids, and different from other fatty acids. I'm happy with that. I don't see any mention of toxicity. This paper is available full text and I've been through the introduction and discussion. These discuss metabolism without mentioning toxicity. There are undoubted differences in fatty acid metabolism between different fats. So be it. Toxicity??? I'm not convinced.
Too much of a slog to go through the whole document. My take on Alzheimers is that it is carbohydrate poisoning. Re BSE, the prion hypothesis is absolutely that, hypothesis. Researches in this field who give small group seminars to the scientists, my wife included, at her institute are VERY careful to stress that the prion hypothesis is not remotely proven. Not what you would think from the BBC reports, or DEFRA. The cause of BSE is currently unknown.
Grass has omega 3 fats, cows put omega three fats to their muscles. I eat those muscles. This sort of "bovine" level seems fine to me. A few grams extra to compensate for grain feeding to beef also seems fine. I'd not go for 30ml a day from a bottle of cod liver oil myself, but I worry that Peat goes over the top. I would agree that anyone getting 20% of their calories from flax oil is in trouble!
Peter
PS, oops
Not that you can "prove" an hypothesis. Perhaps a better phrase would be that it is currently inadequately tested to be accepted as a good approximation to reality...
Peter
Hi Stan,
I've been through the pentose shunt and I don't really see that it frees up anything in the way of excess inorganic phosphate. But then I don't see anything there as to why Dr K dislikes this pathway so much, unless it is the generation NADPH which supplies energy for lipid generation, when we should be beta oxidising lipid without having to detoxify glucose to palmitate! Nice idea though...
Peter
I took a look at Brian's comment, and also his blog. There's a lot to slog through, and I may at some point, actually do that, but I doubt it. Short of time, you know.
And I was stopped short by something right up front in his "about me" statement:
"In 2008 I am 67 and was diagnosed with CHD in 2000. Prior to that I had a low-fat diet, and was vegetarian/fish for 15 years, running for 30 years and lean and fit. I hadn’t smoked for 30 years. Cholesterol levels over a 20 year period averaged 5.2 with HDL always high around 2.0. LDL varied between 2.5 and 3.9. TriG always low at 0.8 approx. My father and his brothers had all died too early of heart disease. This prompted me to have a calcium scan at age 60 and sure enough I found I had CHD."
So he followed all the conventional recommendations, hit all the right lab values (as far as I can see with my lousy conversion skills to US values) and still got CHD. Doesn't sound like a great lifestyle intervention recommendation to me.
I have more confidence in my low carb, higher fat diet - my BGs stay normal (& would be prediabetic/diabetic on more carbs), and my insulin levels stay low (instead of elevated as they are with high carb). From what I have seen, that is my best strategy for avoiding frank diabetes and far more than CVD/CHD. And it is a heck of a lot more tolerable than fanatic exercising and a low fat, low cholesterol strict vegetarian diet.
annaandgabriel - superb comment!
annaandgabriel,
What you wrote quoting from Brian's bio:
"In 2008 I am 67 and was diagnosed with CHD in 2000. Prior to that I had a low-fat diet, and was vegetarian/fish for 15 years, running for 30 years and lean and fit. I hadn’t smoked for 30 years...."
- seems very typical of vegetarian/vegan community. I have always been fascinated by the blind singleminded tennacity whith which they tend to defend their faith inspite the obvious contradictions including contradictions with their own experience. I hope I won't get like that with my high fat diet. I have been studying this phenomenon on the webmd forum. There are people with CHD who have been low fat vegetarians for 30 years yet they would argue with you to death defending their theories that meat and animal fat are supposed to be causing heart disease and all sorts of ills. It's so weird, almost like a cult!
Stan(Heretic)
I skimmed through the studies Brian referenced. It looks like most of them either included exercise or HDL increases. Since statins don't affect HDL, how does that prove his point?
No one's saying exercise isn't good for you. And as far as HDL, saturated fat is a great way to raise it.
His argument does have one strong point though. Those healthy hunter-gatherers with no cavities and no degenerative disease? Everyone knows they all ate low-fat vegan diets.
Well guys, I’ve picked up all your comments. I’m afraid I’m unimpressed by the emotion/evangelical feel of these, with a few exceptions. Peter dismisses 50 years of research and all the experts with just one word – ‘yes’ ( they’re wrong) but without quoting me an equivalent number of studies (or any studies) showing they’re wrong. In particular, that a high fat diet either resulted in less CHD, or cured it. Furthermore, he says what I said is ‘an essential read’, but with regard to the studies I referenced, ‘to go through them one at a time would be tedious’. Too tedious Peter, or too much in danger in having your theory decimated?
Where I came in, Peter was saying ‘ I am always looking for evidence that I may be wrong to eat against most current medical advice’ Peter is now saying it’s too tedious to look into the 80 studies showing what he was after. That tells me Peter isn’t after knowledge at all, even if it could save his life! Peter clearly is defending a theory, dismissing 50 years of experts without any alternative evidence, and unwilling to look at my evidence. How would you describe that attitude?
Do you guys think I wasted 7 years looking for one side of the story only? Unlike Peter, I don’t have a pet theory I am pushing. I have searched everything I can find on heart disease all that time - trying to find the ‘truth’ . This research is what I have presented. I must have missed all the (unnamed and unreferenced) studies that prove Peter’s philosophy. I have only ever come across half a dozen in this vein, and they are completely outweighed by the hundreds that demonstrate that high saturated fat diets greatly contribute to CHD and diabetes.
I’m afraid the complete lack of reference by Peter (or any you) to a body of evidence – not just a few studies – we all know you can prove anything with just a few – doesn’t indicate or prove anything about Peter’s/your beliefs.
So, to summarise my position, I’m not peddling a position, I’m stating what I’ve discovered. If any of you guys can show me lots of studies that say the cure for CHD is eating truckloads of saturated fat.......please fire away. If you want to ignore what I’ve said, we can all compare notes in the Afterlife, well, I can send messages to you via a medium when you’re already there and I am still going at 120!
To the person who assumed(too many assumptions guys) that Esseltyn(Ornish....Gould?) achieved their reversal of CHD through calorie restriction and therefore reduction of insulin levels, this is a theory looking for evidence. Have a look at their published research – there isn’t anything about calorie restriction that I can recall. If the calories were less, it was simply because they were satiated on high fibre plant foods/low fat diet with a reasonable amount of protein. The point is their CHD was cured on a 10% fat diet with evidence from angiograms, and also the fact they kept on living when the surgeons had abandoned them to die. Have you got high fat studies with the same effect? Please tell me.
To Peter who remembered something to his benefit about cancer in the PROSPER study, do us all a favour and get professional - I sent you about a dozen references to how lowering cholesterol does not cause cancer, and how the effect, (say the medical experts) if ever seen, is because people in the studies were early cancer sufferers, and that cancer, together with many other diseases, lowers cholesterol. It’s not that low cholesterol causes cancer, as the studies I referenced show. Please take the time to enlighten yourself Peter. To help you, I’ve now provided summaries of some of these studies below.
As to my background not being a good advertisement for avoiding CHD – firstly, I did say vegetarian plus fish – not vegetarian. Secondly most of my carbs are low GI, and I don’t eat much refined food of any kind – certainly not sugar. I eat oily fish about 3 times/week as well as almonds. It is too convenient to blame my recent past lifestyle for my CHD. The other facts are that earlier in my life, my diet was exactly what you guys are proposing – high saturated fat! This continued up to being about 35-40. It’s my contention that I have plaque – now calcified, basically as a result of this lifestyle you are recommending!! However, to assume my average TC of 5.2 for 20 years is OK is invalid in at least my case – and many others. The advice of many experts that have done all the research in their long careers that Peter hasn’t, 5.2 (approx 202) is too high. I can argue with lots of evidence that my plaque was formed during the first 35-40 years due to a high fat diet, and I continued to grow plaque even with TC at 5.2, when, for me, with my genes, 3.5 might have guaranteed no new plaque.
You should also have a think about the native lifestyles of people in undeveloped countries many years ago – mainly vegetarian with little meat/fish. CHD was practically unknown. These were in my references. The same as in prehistoric man – which someone picked up from one of my references. That study by the way also said that their cholesterol was extremely low, and I can’t see that applying to you guys on a high saturated fat diet. They also had to run 10km to get the next bison, which would have helped.
I really urge you to check the evidence I presented. Peter we know won’t but others might get benefit. One example. David Blankenhorn is a respected heart disease researcher. In his 1990 study in JAMA(my ref 41), he states that for people in his study, for every quartile of fat in their diet, there was an increased risk in developing new lesions in the coronary arteries. Conversely, those people in the study who reduced their intake of fat did not develop lesions.
Here are more fat-related studies, the references refer to my list in my original post.
• In a study by Sinnett(4) of 800 New Guineans, fat was measured at only 3% in their diet. There was no cardiovascular disease, and blood pressure remained constant with age.
• Hellerstein(5) showed that during World War 2, death rates from CHD fell from 50% to 20% in countries which were rationed, and in which fat consumption dropped to less than 25%. Prior to 1939, these populations had consumed 40% of their diet as fat. The situation reverted after rationing was lifted and CHD again increased when fat consumption did. (OK I agree you could equally argue this result was something to do with reduced calories, but the researchers don’t make an insulin connection)
• The analysis of 6 dietary trials by Gordon(7) included over 6,000 person years of follow up. It showed that by lowering TC levels by decreasing intakes of saturated fat significantly reduced CHD by 24% ,with trends towards lower CHD deaths(21%) and lower total mortality(6%). The extent of TC lowering which can be expected from recommending lipid lowering diets they said is likely to be small and the effect on clinical events has been disappointing due to lack of compliance.
• The Finnish Hospital study(8) was a controlled intervention trial carried out in Helsinki in 1959-71. Subjects were in mental institutions and were given either a low saturated fat diet or the normal hospital diet. It was considered that the reduction in TC exerted a significant downward effect on CHD
• A paper by Connor and Connor(9) states that the association between fat intake, TC and CHD has been recognised since 1907 in studies by de Langen in Java. They continue that Keys found that mortality from CHD was directly related to both the amount of dietary fat intake and TC. Years later, angiographic evidence of improvement in lesions was found in persons treated with low fat diets alone or diet plus TC-lowering drugs. They say there is universal agreement that saturated fat intake should be lowered.
• The Pritikin Program which began in the early 1980’s has had many thousands of people through its Longevity Centre in the USA. The regime is a very low fat diet(<10% of calories) with an emphasis on plant foods, low-GI carbohydrates, small amounts (15%) of protein and exercise. In as little as 3 weeks on the program, blood pressure falls, lipids reduce, CRP reduces and people lose weight – all these improvement being in risk factors for CHD . A study showed improvements in angina in patients on the Pritikin Program(10), and there are individually documented cases of regression of CHD checked with before and after angiograms . The Program’s TC levels are recommended to be below 4.0
• A 1985 paper by a famous researcher – Virgil Brown(11) examines diet and population studies and makes observations which include the Tarahumara Indians of Mexico, whose TC levels are around 3.5, in whom CHD was virtually unknown. Overall, he states that then, over two thirds of the world derived 70% or more of its calories from cereal grains and legumes and that such a diet may be associated with general good health and the virtual absence of CHD.
• The Western Electric Study(12) and the Ireland-Boston Study(13) both showed positive correlations between intake of fats and cholesterol and 19/20 year mortality rates from CHD
Here are some more titbits for you, again referring to my original list of references, this time to do with the role of statins.
TC Lowering With Statins – Stroke and Total Mortality(27)
This famous analysis by Herbert et al looked at all of the trials of statins between 1985 and 1995 and published its report in 1997. There were some 29,000 patients included with follow up an average of 3.3 years. Trials included EXCEL,SSSS, MARS, CCAIT, PLAC-1, PLAC-11, MAAS, ACAPS, REGRESS, KAPS, WOSCOP ,CARE and others. It pooled both primary and secondary trials. In the primary trials, reductions in CHD risk for treated patients was 37%, with reduction in overall mortality in the primary trials of 26%..The higher the dose of statin, the more favourable was the risk of overall and CHD mortality. The researchers concluded ‘With respect to whether TC lowering benefits individuals at lower as well as higher risk, our overview demonstrates significant reductions in total mortality in the primary as well as the secondary prevention trials’
WOSCOPS(20)
This primary trial was published in 1995 after running some 5 years. It included some 6500 men with high TC without CHD history.
It reported a 22% reduction in all cause mortality, 31% reduction in cardiovascular events, 32% reduction in deaths from CAD and non fatal MI, and a 10% reduction in strokes.
Aronow 2001(28)(AFCAPS Trial)
This paper analysed data from the AFCAPS (21) trial. AFCAPS was a primary trial(US Airforce) of some 6600 men without CHD and with average TC (mean 5.7) who were treated with a statin. The following reductions amongst the treated group (all significant) are noted
• Fatal or non fatal MI, angina, sudden cardiac death – 37%
• Fatal and non fatal MI – 40%
• Fatal and non fatal cardiovascular events – 25%
• Fatal and non fatal coronary events – 25%
The trial was called off early due to the benefit derived by the treated group
There was no difference in overall mortality between treated and non treated groups, but significant improvement in CHD mortality and events. However, according to the Joint Task Force and Other Societies on Coronary Prevention(29), the trial was not powered to detect a change in total mortality. The only reason the total mortality statistic was calculated was under a safety heading, and as a tertiary objective. What the researchers said they were looking for in this context was ‘ Whether long term treatment with Lovastatin compared to placebo would result in similar rates of total mortality…’ The researchers concluded that the statin reduced the risk of a first coronary event. They also reported no differences in total mortality, and no differences in the incidence of cancer and traumatic deaths between the two groups. They commented that this provided reassurance about the safety of long-term treatment with the statin. This was the first primary trial to show reductions in risks from lipid lowering in people with average TC levels and without evidence of CHD..
The LIPID Research Clinics Coronary Primary Prevention Trial(30)
This primary trial was published in 1984 of men with high TC. The treated group, as opposed to the placebo group, achieved a 19% reduction in risk of CHD overall, with a 50% reduction in risk for men who achieved a TC reduction of 25%. The conclusion of the researchers was that a decrement of 10% in LDL is associated with a reduction of 19% in CHD risk and that their results were of great clinical significance
Low Cholesterol and Mortality – 39 Year Follow Up(31)
This study, published in 2004 examined some 3200 originally healthy men aged between 30 and 45. After 39 years, the survivors and health records were checked. There was a graded association between baseline TC levels and mortality risk. Those whose TC at baseline was below 5 had a highly significant(25%) reduction in total mortality. Additionally they had better quality of life as measured using the RAND 36 item survey. The researchers said there was not any association between low baseline TC and mental health. The authors were cautious in relating the results to TC- lowering interventions since the people with low baseline TC were likely to have had low TC throughout their life.
Gordon Meta Analysis – 6 Dietary Trials (7)
This meta analysis is mentioned by the NCEP It analysed 6 robust dietary trials with some 6,000 person-years of follow up. It showed that lowering TC by reducing the intake of saturated fat significantly reduced CHD by 24%, with trends towards reduced CHD mortality(21%) and reduced total mortality(6%).There was not an increase in non-CHD mortality.
Heart Protection study(32)
This study included some 20,500 UK adults, of which some 7000 did not have a history of CHD and was designed to check on the protection in strokes of lowering TC via a statin. The trial reported
• all cause mortality was significantly reduced by 12% in the treated group
• CHD mortality rate dropped 18% (highly significant) in the treated group
• highly significant reduction of approx 25% in the first event rate for non fatal MI
• Stroke incidence reduced in the treated group by a highly significant 25%.
There was not any increase in cancer or any non-vascular disease. .Separate figures are not given for the group of 7000 without a CHD history
Yusuf Analysis of 22 Trials(33)
Twenty two trials of the effects of TC-lowering involving some 40,000 patients were analysed by Yusuf et al. In the 5 primary trials, reductions of 20%, 23%, 19% 47% and 34% in cardiovascular events was noted for treated patients. They calculated that a 10% reduction in TC resulted in a 20% reduction in CHD deaths
MRFIT(35)
This trial was classified as primary prevention by the researchers. It looked at some 316,000 men beginning in 1975, with an average follow-up period of 12 years. The trial was designed to assess the effects of smoking, TC levels and BP on CHD mortality. Total mortality was not measured. The men were free of CHD at baseline. Findings were that there was a strong graded relationship between TC levels above 4.65 and CHD mortality.
The Holme Analysis(36)
The aim of this study was to evaluate the relationship between TC reduction and total mortality and CHD. A total of 19 trials were analysed for total mortality, of which 16 were analysed for CHD. Primary trials were 7, with 12 secondary. The researcher, Ingar Holme, PhD notes that the distinction between primary and secondary trials is somewhat artificial since the average degree of CHD was probably not much different for patients in the two types of trials. Holme goes on to explain that patients in secondary trials have a built-in advantage with regard to overall mortality, Since they have a higher chance of dying than primary patients – some of whom are CHD-free, the effect of TC lowering on mortality is seen more in them. However Holme comments that in the data examined, there was no difference between primary and secondary trials with respect to the efficacy of TC-lowering on CHD . For every 1% of TC reduction there was a 2.5% reduction in CHD risk. TC lowering of at least 8-9% was needed to affect total mortality
French Analysis Primary Trials(37)
This study analysed 7 primary prevention trials of TC-lowering. There was a 24% reduction in fatal and non-fatal CHD events(significant), a 14% reduction in fatal coronary events(significant) and a 13% reduction in total mortality(non-significant). Nevertheless, CHD mortality was improved.
ASAP Trial(38)
This 2001 double blind study of 325 hypercholesterolaemic patients, 69% of whom did not have CHD, showed that aggressive LDL lowering(45%) resulted in regression of their carotid intima thickness, whereas conventional lipid lowering did not
Here are some more, this time, trials of cholesterol - lowering
• A total of 22 trials of TC lowering showed that statins reduced CHD mortality by 25%(22)
• An analysis of 32 randomised TC-lowering trials including primary and secondary trials involved over 42,000 patients(46). This showed that TC-lowering reduced CHD deaths related to the strength of the intervention – that is, amount by which TC was lowered.
• Lawrence Gould is a famous CHD researcher. His meta analysis(47) published in 1995 looked at 35 trials of TC –lowering, of which 5 were primary. For every 10% of TC lowering CHD mortality was reduced by 13% and total mortality by 10%.The benefit was directly related to the reduction in TC
• One of the largest analyses I have found is by Law, Wald and Rudnicka(48). This was published in the BMJ in 2003. They examined an enormous amount of data to see whether statins reduced heart disease and stoke. Totals of 164 short term statin trials,and 58 trials of TC lowering by any means were included. Their conclusion was that statins can lower LDL by an average of 1.8 mmol/l which reduces the risk of CHD events by 60% and stroke by 17%. The 58 trials had 49% of primary patients .There was no excess risk in people allocated to statins
Here are some experts who disagree with Peter and the high-fat/cholesterol doesn’t matter philosophy
• Aronow(28) “Statins are also effective in reducing cardiovascular events in older persons with hypercholesterolemia but without cardiovascular disease.”
• La Rosa (49) states that both primary and secondary studies of TC-lowering have demonstrated prevention of CHD morbidity and mortality
• The NCEP III Guidelines Executive Summary(50) state that primary prevention trials show that LDL reduction reduces the risk for coronary events and coronary death even in the short term
• Bonow(51) issues a call to the medical profession for action in primary prevention of CHD by lowering TC amongst other methods. He states that in apparently healthy young people without CHD, their risk factors predispose them to death and the risk factors often go untreated . High TC is a risk factor for CHD death. He goes on to state the NCEP III guidelines, which include lowering TC in anyone with risk factors, will save lives.
• In September 2002, the editor of the American Journal of Cardiology –Dr W. C. Roberts wrote an editorial in this journal (52). He strongly advocated a preventative approach to CHD. He argued that CHD essentially does not occur when TC is below 3.8. He advocates – for everybody, levels of TC less than 3.8 and LDL less than 2.6 . he says CHD should be viewed as a potentially preventable disease. He argues that this can be achieved by a vegetarian-fruit diet, or by lipid lowering agents, and that payment for these could be afforded by reducing by-passes, angioplasty and expensive diagnostic tests.
• La Rosa(39) says that statins have enabled investigators to achieve enough TC-lowering to demonstrate unequivocally that coronary events can be prevented at no cost in increased mortality from non-cardiovascular events. These benefits extend to people without manifestation of CHD
• Silverstein(53) states in his 2003 letter to The American Journal of Cardiology that the new(2004) NCEP guidelines are too complex. He says that societies where TC is between 2.3 and 3.3, CHD is virtually absent, and there is a vastly decreased incidence of arthritis, cancer of the colon/prostate/uterus/breast, plus more. He says 35% of coronary events occur with TC levels between 3.8 and 5.1 and that the Framingham Study found that CHD essentially ceases to exist with TC levels below 3.8. In his own practice, virtually none of his patients have developed CHD if their TC levels are kept below 3.8. He advocates a very high fibre diet of whole, unprocessed foods with fish twice a week. He concludes that we should not aim for a decrease in CHD, we should strive to absolutely prevent it. He proposes the NCEP guidelines be changed to simple markers of TC below 3.8 for everyone free of disease, and below 3.3 for those with CHD and /or diabetes.
• Hedblad et al(49) say statins reduce CHD events in both primary and secondary studies
• Levine et al(33) say that TC reduction has been established as reducing mortality from CHD
• Corti et al(55) say ‘Lipid lowering with statins can significantly decrease cardiovascular morbidity and mortality when used in primary and secondary prevention’
• Superko(56) and Krauss stated in 1994 that in the past two decades large trials had convincingly shown that reductions in TC of 10% average, with increases in HDL to varying degrees had significantly reduced CHD morbidity and mortality
• Peter Libby, who writes textbooks on cardiology states(57)that lipid lowering reduces MI, stoke and prolongs life
• Ballantyne(58) states there is abundant evidence that lowering LDL decreases the risk of CHD. He says unfortunately many high-risk people are not receiving either diet or drug therapy.
• Dr William Castelli ran the famous Framingham study for over 35 years and is widely regarded as one of the world’s experts on CHD. In his 1998 paper(59), he says that 70% of heart attacks came from fat-rich plaques which rupture and which do not show up on angiograms. He says the higher the TC, the higher the risk of CHD and that people whose TC is low – about 3.2 to 3.5 – do not get CHD. The more TC is lowered he says, the more subsequent CHD is lowered. He also advocates increasing HDL.
Peter’s comment about the PROSPER study implied that lowering cholesterol might cause increased risk of cancer. He hasn’t bothered to check the references I gave him on this. Here’s why he’s wrong. Again the references are in my original list
Research has shown that the early TC-lowering drugs – fibrates and hormones caused increases in mortality which had nothing to do with the degree of TC-lowering – it was caused by the drugs themselves. There are several references to this.(39). These dangers have been overcome by the discontinuation of these drugs and their replacement by statins, as the examples below show.
• Muldoon et al published a study in the BMJ in 2001(60). Its aim was to look at the association between TC lowering and deaths from suicide, accident or trauma. A total of 19 trials were examined, which covered some 42,000 participants in primary trials and 28,000 in secondary trials – about 338,000 patient years of data. In the 13 trials which used dietary or non-statin treatment, there was a non-significant increase in death from suicides and trauma in the treated group. The overall conclusion was that deaths from accidents and violence were not significantly increased amongst TC-lowered patients, this was true for both primary and secondary studies. Additionally there was no association between the size of TC reduction and likelihood of death in these categories. Muldoon went on to say TC lowering by statins does not increase new psychiatric diagnoses, hospital admissions or non-fatal trauma.
• Several studies have shown that the reason diseases show up in people with low TC is almost entirely due to the fact that people with low TC include a disproportionate number whose TC has been reduced by illness. Early cancer, respiratory disease, gastrointestinal disease, and alcoholism are examples, plus others.(22)
• In 1990, an expert panel assembled by the USA National Heart Lung and Blood Institute published its findings on the association between low TC and mortality(46). The panel examined the findings from 19 trials, in which a total of over 68,000 deaths occurred. Its conclusions were that most of the experts considered it likely that many of the statistical associations of low or lowered TC with disease were explained by confounding in one form or another. Additional points made were that endemic hepatitis in Asian populations is the true cause of excess mortality at low TC levels. Lifetime alcohol exposure can cause low TC as well as liver diseases. Excess cancer death at low TC levels was restricted to smoking-related cancers. In cancer generally, low TC levels are almost certainly a consequence of the disease. Chronic inflammatory diseases cause TC to lower. Anemia and leukemia can cause lowered TC whereas in leukemia patients in remission, TC levels rise significantly.
• A 1994 analysis((61) included data from studies of deaths from causes other than CHD. Ten of the largest studies were included, as well as 2 international studies, and 28 randomised trials. The conclusion was there was no evidence that low or reduced TC increases mortality from any cause except haemorrhagic stroke. This stroke association was only in people with very low TC concentrations, and was outweighed by the benefits of reduced CHD risk. The researchers also note that low TC in people with certain types of cancer, suicide, chronic bronchitis, chronic liver and bowel diseases were satisfactorily explained as being caused by the diseases themselves. Depression was also noted as causing low TC.(Which might explain some suicides in low-TC people)
• The Heart Protection Study(32) found no increase in cancer or non vascular disease in its trial, which lowered TC and reduced overall mortality
• A 2002 analysis(62) of a number of TC-reducing trials(WOSCOPS, CARE, LIPID) looked into an accumulated 112,000 person years of statin exposure . The researchers noted no differences in non – cardiovascular serious adverse effects between treated and placebo patients. With over 243,000 blood samples, both treated and non-treated patients had similar liver function results
• An analysis of all the statin trials between 1985 and 1995(27) found no increase in non-CHD mortality or of cancer mortality and incidence and concerns that TC lowering may increase other forms of death was not supported by the data.
• Another meta analysis (63) of all statin trials between 1966 and 1999 found 5 trials where cancer had been studied. The conclusion was there was not any association between statin use over a 5 years period and cancers . Limitations were the limited follow-up periods of the trials.
• An analysis of 35 trials(47) states that non-CHD mortality is not significantly related to TC reduction. The researchers go on to say that their findings do not support the notion that reducing TC increases the risk of mortality from non-CHD causes, as suggested by others.
• Aronow(64) states that low TC may result from malnutrition and chronic diseases with an inflammatory component
• The 2003 meta analysis by Law et al(48) states that concerns over hazards of TC reduction has been resolved by earlier studies, and the data from the 58 randomised trials of TC reduction they analysed confirms this
The conclusion from this body of evidence and expert opinion is that TC-lowering is not harmful
-----------------------------------------------------------------------------------------------------------------------------
So, in summary, I really would like to learn what I have missed – just in case I have got it all wrong. Peter, please put some effort in and send me all the studies that show a high saturated fat diet avoids or cures CHD. Unlike you I will find time to get to the local Uni medical library and check these out and report back. I notice one of your bloggers asks you a similar question about where is all the evidence.
Brian - you might be interested in this one - just published:
http://www.nutritionandmetabolism.com/content/5/1/6
Carbohydrate restricted diets (CRD) consistently lower glucose and insulin levels and improve atherogenic dyslipidemia [decreasing triglycerides and increasing HDL cholesterol (HDL-C)].
Chris
www.conditioningresearch.com
sorry - try this link
http://tinyurl.com/3y2n3r
Brian, I am jealous of the time you have for this. I hope this doesn't offend you, but your argument reminds me of the Jehovah Witnesses that come to my door, nice as can be, but glued to the front porch, determined not leave until I see "their" truth. I'm always polite to these people (I know sometimes they get doors slammed in their faces, are called them rude names, and otherwise treated shabbily or even threatened). But I try not to become engaged in conversation because a) it is pointless and b) it is unwelcome and assumes that what I think is absolutely wrong, wrong, wrong. It's hard to have an interesting conversation in such a situation. We see and seek our truth in such different places and ways.
At the risk of becoming engaged in this debate, it looks to me that no one here is going to take the time to go through each of your points and counter them, either due to lack of time or lack of interest. I took a quick scan and I've seen most of this before and it wasn't convincing then so I doubt it will be more convincing now. Whole books have already been written about why the "evidence" supporting the saturated fat/cholesterol lipid hypothesis is so shaky, picking apart the studies to date that were devised to show that saturated fat and cholesterol cause CVD. There is no point rewriting them, which seems to be what you are asking someone to do. Check out Malcolm Kendrick's Great Cholesterol Con, Uffe Ravnskov's Cholesterol Myths, and Gary Taubes' Diet Delusion (UK title)/Good Calories, Bad Calories (US title) for a start. There are many more, too.
And I suspect there is just nothing in it for the readers here to argue point-by-point. You may feel differently, but I think you might have to find someone else to argue against you.
Which reminds me, I have some nice pork fat rendering that needs my attention...
Re: "41. Blankenhorn DH, Johnson RL, Mack WJ, El Zein HA, Vail LI. The Influence of Diet in the Appearance of New Lesions in Human Coronary Arteries. JAMA 1990;263:1646-52"
Brian,
This study (by the abstract, I have no access to the full text) compared a low fat high carbohydrate diet with a medium fat high carbohydrate diet. Consuming large amounts of carbohydrates with some fat (any fat) induces temporary insulin resistance and worsens glucose control in patients with metabolic syndrome, as compared to consumming the same amount of carbohydrates without fat. Fat is a glycemic response modifier not the main culprit. Especially if your vegetarian diet may have been too much skewed towards omega-6 from uhealthy veg oils, that was probably another negative factor. You will find many studies on that too. Please read them.
If you look thru the refs on all sides not only those glorifying vegetarianism, you may find many leads pointing towards excessive insulin and excessive glucose as as one of the main factors (see for example R.W. Stout 1968, 1969, Lancet papers). There are no contradictions with what we are saying: both low fat low caloric vegan and high fat low carb can lower serum insulin. By blaming fish and butter you may be rationalizing the failure and fighting your friends rather than the real perpetrators of your disease. Fish is an unlikely culprit because every single study I have ever seen was positive for fish! See recent Peter's refs, for example Lyon study, Bantu study etc; they all show only cardioprotective effects.
Regards,
Stan(Heretic)
Brian,
With a few possible exceptions, most hunter-gatherers ate plenty of animal fat. This is particularly true of our paleolithic ancestors, who survived primarily on large game. This game was not lean, particularly because they valued the fatty organ meats. There is a strong archaeological consensus on this.
There is a fairly large body of evidence showing that many pre-agricultural cultures eating diets high in animal protein and fat, were (or are) very robust and did not have CVD. Some of this evidence is summarized in Taubes' "Good Calories, Bad Calories", and there is plenty more in "Nutrition and Physical Degeneration" (although the latter doesn't relate specifically to CVD).
Some Pacific island cultures were eating 70+% calories as fat, from coconut. As I'm sure you know, coconut fat is almost all saturated. Did they get heart disease? Not until they adopted a Western diet full of grains and sugar, and stopped eating coconut. I don't have time to dig up the references but they're in "Good Calories, Bad Calories" if you're interested.
I'd also point you to studies of the Kuna indians in Mexico, who drink high-saturated fat chocolate all day and have 10% the heart disease of their city-dwelling brethren, and do not have blood pressure increases with age. This study was age-matched. Chocolate is rich in stearic acid, the same fat that's found in beef.
From an Occam's Razor and obvious logic perspective, it seems difficult to imagine that our primitive ancestors would have trimmed the fat off anything (and we certainly don't observe it in other animals). Also, fat has more than twice the energy of both carbs and protein, and our bodies evolved to store 2-3 months worth of energy from tissue that provides no other discernible role (unlike lean tissue, which can also be broken down). All the while, you can completely deplete your "stored carbs" in a matter of hours.
How much evolutionary sense does that make? That carbs would be our primary energy source? Say what.
Here was my breakfast. 97% fat cals.
http://www.honestylog.com/root/2008/02/liquid-fat-bomb.html
Hi all,
Not much time for blogging, just starting a four day work stint. It's impossible to know where to start and Brian is too sad to make fun of.
But for a typical cholesterophile view, look at this quote from Brian:
"MRFIT(35)
This trial was classified as primary prevention by the researchers. It looked at some 316,000 men beginning in 1975, with an average follow-up period of 12 years. The trial was designed to assess the effects of smoking, TC levels and BP on CHD mortality. Total mortality was not measured. The men were free of CHD at baseline. Findings were that there was a strong graded relationship between TC levels above 4.65 and CHD mortality."
Then go to the actual abstract:
MRFIT
Just try and reconcile the two. I never thought I'd ever see anyone ever use MRFIT to justify cholesterol lowering, but that's just what happens.
Chainey, enjoy your reading. Have you hit page 63 yet? Anyone looking for the effects of WW2 and rationing on heart disease really needs to see this graph. Rationing started in the UK in 1941 and went on for 12 years. Severe saturated fat reduction, more fruit and fish consumed. Heart disease deaths went from about 180 deaths/100,000 at the start of rationing to about 450 deaths/100,000 at the time it was stopped (early 1950s).
I get depressed seeing work cited as supportive of the cholesterol hypothesis when the opposite is true.
Anna, I know exactly how you feel re doorstep religion. Religion is fine. It does appear to be very very important to humans, even if I personally am a total failure. I always try to be nice to Jehovah Witnesses, but as quickly as possible, especially if there is beef suet in the kitchen which needs rendering.
As Brian knows, I'm doomed. He's here to save you. I think you're all grown ups. And can make your own decisions.
Peter
Re: WOSCOPS (Pravastatin study)
None of the mortality reductions were statistically significant! Even if we accepted non-significant result as possible indication of the trend it tells us nothing about diet and it does not prove that the effect was due to serum cholesterol lowering rather than the statin itself.
Re: Blankenhorn and Ornish studies
There is another aspect: angigraphic measurement of vascular diameter do not take into account effects of dillation or constrictions caused by other factors. For example: "a hand grip sustained for a few minutes was follwed by a 35% decrease in vessel diameter" (Ravnskov's book "The Cholesterol Myths" page 190).
Angigraphic measurements of arterial diameters are surogate markers which may or may not be indicative of arterial disease. They may also be indicative of patient's stress level at the time of the test.
I think people who put too much confidence in some of the studies that registered only the low single digit % relative and ~0.05 mm absolute changes in the arterial lumen are standing on very thin ice.
Regards,
Stan(Heretic)
Hi, Brian quoted • Another meta analysis (63) of all statin trials between 1966 and 1999 found 5 trials where cancer had been studied.
Ah, I can remember all the statin trials from the 60's 70's and 80's with crystal clear perfection.
Those magic mushrooms !!
Back to reality, as you say Peter, using MRFIT as an example validating the diet heart hypothesis is a new one on me too!
• Another meta analysis (63) of all statin trials between 1966 and 1999 found 5 trials where cancer had been studied.
Where did this come from ? Why would anyone who knows anything at all about statins use 1966 as part of a search parameter?
To those trying to be helpful, my thanks. In particular the following. Chris – yes I’m aware that carbohydrate restricted diets can lower glucose/insulin etc, but two points on this. First, my glucose is normal, so is my insulin on a glucose tolerance test. My TriG couldn’t be much lower - 0.8 – and my HDL is very high at 2.0, yet my diet on last check was about 60% carbs. Since these measurements are OK, I haven’t got insulin resistance nor the beginnings of metabolic syndrome, and since they are the result of my diet/lifestyle it’s safe to say this is in control as far as CHD goes. Fish/fish oil helps bring down the TriG and exercise helps also – as with glucose and insulin levels. Thanks for the reference, but it doesn’t show what I keep asking you guys for – evidence that a high fat diet will avoid or fix heart disease.
To Stan, on Blankenhorn’s 1990 study. The patients were 82 of the placebo group from the CLAS study which tested an early cholesterol lowering agent called colestipol. So the placebo group were not on this drug. Eighteen of the placebo group had developed new lesions (blinded angiographs before and at the end), over the course of the CLAS study which was over a 2 year period and it was the whole of the placebo group that Blankenhorn looked at, to try and determine what aspects of their diet might have caused the lesions. Of the two subsets of the placebo group, 64 did not have lesions, 18 did. The percentage of carbs in the diets of both those with lesions and those without was nearly the same – 52.3% in those that didn’t, 50.6% in those that did. Protein was 17.4/15.0, total fat 27.5/34.1.
In other words, the lowest carbs group developed the lesions – the highest carbs group did not have any lesions. As I mentioned before, in the lesions group, the risk of lesions increased for each quartile of fat ingested. Stan’s comment that high carbs(50% isn’t high to me but it might be to him) induces temporary insulin resistance is surely governed by the type of carbs – high or low GI – about which, in this study, we know nothing, so I think he’s assuming without facts. Disregarding this, Stan’s assertion that the carbs are the culprit because in the presence of fat they induce insulin resistance(which presumably leads to the lesions), just doesn’t fit the facts. For carbs to be the culprit, BOTH groups would have had to develop lesions because they both ate similar carbs. In fact, the group without lesions ate more carbs than the lesion group, but did not develop lesions. Blankenhorn blamed the higher level of fat in the lesion group. The odds ratios for lesions were as follows –protein 0.72, carbs 0.98, fat 1.11, sat fat 1.13, polyunsaturated fat 1.28. An odds ratio of 0.89 for carbs does not support an implication they were the culprit.
Blankenhorn’s summary of the study was that increased fat consumption was associated with significant increased risk of developing lesions. To be developing lesions on a 34% fat diet would be a big concern for me – more so I imagine for people like Peter, who claim an 80% fat diet. Has he had a calcium heart scan? Angiogram? Do we know Peter is clear of heart disease? Do any of you have this certainty as you cook up your bacon and eggs breakfasts? I am intrigued how confident you are that your coronary arteries are OK on your high fat diet.
I do not take issue with Stan’s comment that excessive insulin and glucose are risk factors for CHD, but they don’t apply to me, and I haven’t yet seen the link to a high fat diet being beneficial for avoiding or fixing CHD. Insulin and glucose can be kept adequately under control with a sensible low GI diet with appropriate good fats – nothing remotely like 80%.
Sasquatch offers comments on ancient man, but doesn’t support these with evidence from studies. The one I cited(my ref 79) was by O’keefe et al ‘Optimum LDL is 50-70 mg/dl. Lower is better and phisiologically normal.’ JAMA 2004 43 2142-2146. These researchers say that ancient hunter-gatherers had TC levels of 2.5-3.8, with estimated LDL of 1.3-1.9 and an absence of CHD even in their 70’s and 80’s. This compares with an average US LDL level of 3.3. They claim CHD is endemic because commonly LDL is twice the normal physiological level. The show the LDL figures for hunter gatherer populations such as the Hazda/Inuit/Ikung/Pygmy and San – all of which are in the extremely low range above. They say extremely low LDL may prevent CHD, regardless of other risk factors and quote numerous studies showing lowering LDL by any means reduces heart disease and mortality. There is no comment about the typical diet of ancient man, but if it was high in fat as Sasquatch and Mikoley claim, maybe this was intermittent. Maybe they only caught a bison once a month, gorged on it, fat and all, then between ran hundreds of clicks to catch the next one and survived on roots etc between ? The bottom line is that if their LDL levels were as low as the above suggests, it doesn’t matter what they ate – we just need to get ours down to that level. (Now referred to as the regression level). Again my question always is, how can a high fat diet get LDL to this low level? I haven’t seen any of you parading your LDL levels – are they at this hunter gatherer level on your high fat diets?
As to the South Pacific cultures Mikoley mentioned who ate 70% of calories as fat from coconuts, I haven’t come across this study but would be interested in it and knowing how this affected their mortality/cancer rates. Without looking into the details I don’t have an opinion. Yes I have the studies on the people who ingest a lot of cocoa and have normal BPs etc and agree that the studies show this can have a beneficial effect on heart disease risk factors. However, I have not been able to find any study that showed high cocoa consumption reduced heart disease itself, or cured anyone of it, and I have approx 10 cocoa studies.
Anna wasn’t very helpful – likening me to the Jehovah’s Witnesses. She –like Peter, just dismissed all the studies I cited out of hand saying someone has already done this before, in a book – somewhere - so there’s no point anyone else putting effort in......etc. I’m afraid that’s just a cop-out to me. Unlike the JW’s, I am not hammering on any of your doorsteps trying to convert you to my religion. I just happened to have found your blog by accident and was horrified by what I read – starting with the gleeful way Peter tried to use one somewhat strange study to justify a whole theory. I then started to wonder whether I had missed something important in my years of research. Each time I respond I ask, but never get evidence that your high fat diet is good – ie that it cures CHD. The reason I’ve provided all the detail on the opposite view is that I am seriously concerned you are on the wrong track, and since I have all this information to hand, am happy to share it with you. Anna indicates she’s envious of my free time, yet In the time it took Anna to give me a bit of a serve, she could have been more positive and given me a couple of studies from the book she recommends I buy that show high fat diets are good for CHD. I’m also concerned that Anna sounds like she believes in some sort of conspiracy theory on cholesterol. ‘Whole books have already been written about why the ‘evidence’ supporting the saturated fat/cholesterol hypothesis is so shaky, picking apart the studies to date (my emphasis) that were designed to show that saturated fat and cholesterol cause CVD’ Oh dear. Like Peter, 50 years of thousands of sets of independent academic research, long before cholesterol was popularly in the firing line, dumped – wrong, and designed with a common goal – to dupe us all into an incorrect cholesterol hypothesis? And someone said so in a book. How can she get so brainwashed?!
Picking up Peter’s recent comments on the MRFIT study, I don’t follow his sarcastic objection. The study, as I said, was to assess smoking, BP and cholesterol. As we are all concentrating on cholesterol I quoted its results on this. They are not at all good. Quote from the same abstract Peter is reading “Strong graded relationships between serum cholesterol levels above 4.65 (180) .....and mortality due to CHD were evident “ Perhaps Peter is misreading the words about BP and smoking signified by my dots as being in combination with the cholesterol finding. They are not if he reads the study. There was an independent strong association between levels of cholesterol 4.65 and above with increased risk of death. The actual risk ratios were TC less than 4.7 1.0, TC to 5.2 1.4, to 5.7 1.7, to 6.3 2.1 and above 6.3(245) 2.9
Clearly this demonstrates that higher TC is bad news for heart disease. The other point of this study is that the level above which risk was calculated is an ‘OK’ level in the eyes of many doctors. My earlier points about my own history, and the many studies that support this, was that having OK cholesterol – eg 5.0 is probably far too high for many individuals. MRFIT shows this to be so.
With regard to Peter’s claim that WW2 rationing in the UK caused an increase in heart disease, (not referenced), is in conflict with the studies I found accidentally some time ago(my refs 5 and 6). In particular Hellerstein found that during WW2, death rates from CHD fell from 50% to 20% in countries which were rationed, and in which fat consumption fell to less than 25%. Prior to 1939, these populations had consumed 40% of calories as fat. The CHD figures reverted after rationing ended, and CHD deaths rose again as fat consumption increased.
Turning to WOSCOPS and Stan’s comment that none of the mortality reductions were significant, I will leave him to argue with the lead researcher – Dr Shepherd. Deaths from all causes were 22% less in the Pravastatin group, risk of a heart attack or death from CHD was 31% less in the P group, and Dr Shepherd says “ Treatment with Pravastatin significantly reduced the incidence of myocardial infarction and death from cardiovascular diseases without affecting the risk of death from non cardiovascular disease.....Further he went on to say “Thus our results indicate that reducing cholesterol levels with pravastatin reduces the risk of coronary events in asymptomatic subjects with hypercholesterolemia”. Shepherd, unlike Eric does not link the improved mortality to other possible affects of statins.
Again to Stan – I can’t believe the appearance of lesions can be induced by something the patient does when an angiogram is being taken. Where is the evidence that lesions have so appeared? It’s very strange that all the people who had lower fat diets all did the same thing when tested to skew their angiogram and show no lesions, whereas, all the people on the higher fat diets did exactly the opposite thing on the test to make lesions appear on their angiograms. I’m not trying to be sarcastic, I just think Stan is struggling a bit. In any case, some of the studies of regression I cited used IVUS which is reportedly more accurate than angiograms.
Anyway, good to see someone has looked at some of the studies – about 76 to go approximately.
Getting to the Bottom Line
In my first blog comment, I gave you all some 12 references to studies which demonstrated regression of CHD – mostly by lowering LDL.
I am really disappointed that no one – especially Peter who apparently started this blog – has risen to this challenge and given me even a list of studies that shows that a high fat(or high LDL) diet will regress CHD. Why is this so difficult ? I can only assume the answer is these don’t exist. I haven’t found them in 7 years, and I’ve looked at anything and everything I could.
Unfortunately you have all failed this simple challenge which would enable you perhaps to claim a convert.
I challenge you again – to produce such studies which surely must be in the books you keep quoting at me because you are so adamant – Peter certainly – that high fat is good for CHD and cholesterol doesn’t matter.
Peter says he doesn’t know where to start with my comments – start here Peter – a simple request.
Stan: I don't think insulin is the boogey man. I would say that it's a matter of deficiencies and toxins. Eating refined carbs will not cause disease, unless you displace other foods (like animal protein). If you eat like the Japanese (white rice, seafood, vegetables) or like people in Kitavan (coconut, tubers, fish), you won't have diseases. I disagree with your comparison between vegan and low-carb diets. Vegan diets are NOT healthy long-term, period.
Peter: Ray Peat's point is that we should avoid all PUFAs equally, not add omega-3 to balance the omega-6, but limit both omega-6 and omega-3.
Brian: "hundreds that demonstrate that high saturated fat diets greatly contribute to CHD and diabetes."
All of which fail to isolate those saturated fats from carbs (esp the refined ones), PUFA vegetable oils, processed trans fats, reduced fat dairy, lean meat, etc. The reality is that very few people are eating "high saturated fat" diets, Brian. That is the biggest myth there is. People are drinking skim milk and cutting the fat off of their meat. Peter has mentioned how the ground beef he was getting was reduced in fat recently. You ignore all these factors and others.
You are just regurgitating all the stale arguments of Fuhrman, Ornish, Esselstyn, McDougall, Pritikin, T. Colin Campbell, Barnard, and so on. This is vegan propaganda. They are destroyed by Uffe Ravnskov, Anthony Colpo, Malcolm Kendrick, and other writers. Not to mention all of the "paradoxical" findings like people in the Gascony region of France.
http://tinyurl.com/2xy9gb
Chris Masterjohn has torn the China Study apart, showing the real truth that limiting PUFAs and eating more saturated fat widens arteries, not the low-fat program. Low-fat diets are low-PUFA and high-SFA diets in disguise. The body makes saturated fat very easily from carbs. All the claims about the benefits of vegan or low-fat diets are lies.
http://www.cholesterol-and-health.com/China-Study.html
http://www.cholesterol-and-health.com/Campbell-Masterjohn.html
Hi All,
I began my reading with John Yudkin, many years ago. It was brought up to date by Uffe Ravnskov. Obviously Malcolm Kendrick is a great recent read. I even bought Anthony Colpo's book. I'm about a third of my way through Gary Taubes book but keep getting distracted. I haven't started my copy of Barry Groves' latest book but have his older book. Somewhere along the line I picked up Russell Smith and Edward Pinckney's book. Obviously there is a lighter text from Wolfgang Lutz and the Freemans' work on the ketogenic diet. There has been enough debunking.
This is the wheel. It works. I just use it, I've no need to reinvent it. I certainly have no interest in going through these excellent books, enjoyable though they are, to help occupy Brian's excessive free time. I have no interest in changing his mind.
If anyone here feels a burning desire to reinvent the wheel, please feel free. But don't waste too much of your time on it, I suspect he has certainty.
Peter
PS to Bruce, careful of your time, arguing religion is never very productive!
Brian: "In other words, the lowest carbs group developed the lesions – the highest carbs group did not have any lesions. As I mentioned before, in the lesions group, the risk of lesions increased for each quartile of fat ingested."
This does not prove anything. What fats were used? Were they high in PUFAs or trans fats? The standard American Diet is about 10% soybean oil by calories, according to CNN (America's Killer Diet). Saturated fat gets blamed, but PUFAs are the ones doing all the damage. Studies have found that adding PUFAs to a high-fat diet causes cancer, which does not occur with saturated fat like coconut oil, butter, and beef fat. Here is one study with a lot of references to other studies that show the same thing. Taubes noted this briefly in his book.
http://cancerres.aacrjournals.org/cgi/reprint/45/5/1997
I would like you to consider for a second, Brian, that the reason the low-fat diets MIGHT provide benefit is that they are also low in PUFAs and trans fats. NOT because they're low in SFAs and MUFAs (the healthy fats IMO). If you add PUFAs to beef tallow or coconut oil, it may cause a variety of pathologies. I am not convinced by the low-carb rhetoric completely, but the low-fat dogma is far less credible.
Here's another idiotic study where PUFAs (safflower oil) were added to SFAs (beef tallow). They blame the saturated fat and ignore the PUFAs. They even call it a high saturated fat diet, with over 40% PUFAs. What a joke. Unless you control for the PUFAs, then saying that a high-fat diet is worse proves nothing. But I agree that the fats most people eat today are unhealthy. Things like canola oil, corn oil, soybean oil, safflower oil, hydrogenated oil...
http://ajpendo.physiology.org/cgi/content/full/282/6/E1352
If you want to prove that the high fat diet is inherently bad, though, you have to control PUFAs. You have not done that, so your argument is bogus. Show us a study of high-fat diet that was limited to 4% PUFAs, or less, by calories. Then, we can talk. Otherwise, you're comparing high-PUFA diets to low-PUFA diets.
Brian: "I have not been able to find any study that showed high cocoa consumption reduced heart disease itself, or cured anyone of it, and I have approx 10 cocoa studies."
It probably wouldn't, unless people were using 85% dark chocolate or at least 70%. Most chocolate is 40-60% sugar. I use Lindt's 85% chocolate, with no soy lecithin. A 40g serving (4 squares) has 5 grams of sugar. I think it would be ideal to get 100% baking chocolate or unrefined cocoa butter, so there's no sugar.
"Hellerstein found that during WW2, death rates from CHD fell from 50% to 20% in countries which were rationed, and in which fat consumption fell to less than 25%."
Sugar was also rationed, as Taubes pointed out in his book. Brian has committed the same fallacy as Keys did in arguing that fat caused the heart disease. Another point, it's silly to just look at CHD. I think we need to look at total mortality and Peter would agree with me here. It's no good to have less CHD, but more cancer, stroke, suicide, etc.
"Prior to 1939, these populations had consumed 40% of calories as fat. The CHD figures reverted after rationing ended, and CHD deaths rose again as fat consumption increased."
Prior to 1939, they had most likely consumed 40% of calories as refined sugar and white flour. Those foods, to use the term 'food' lightly, may also have been rationed, along with fat. After rationing ended, the use of refined carbs increased. Taubes pointed all of this out in his book and I suggest you read it. Or just use your brain. I think the people during WW2 had difficulty obtaining doughnuts, carbonated sodas, sweet chocolate, ice cream, etc.
So, was it the reduction of fats or the reduction of carbs that brought the alleged benefits? Or was it the reduced calories in general?
Brian,
I didn't have the full Blankenhorn’s study, just the abstract. Since you described that the control group diet was so similiar to the study group then the difference in lesions are probaly due to other factors, like colestipol drug, due to chance or some instrumental errors. The numbers you quoted (risk factors) are also so close to 1 that I am suspicious that it may have been all statistically marginal anyway.
I maintain my view that WOSCOPS study is not only statistically no-significant (yes that ~20-30% relative differences in mortalities), but also deals with a stain drug and is thus irrelevant to our discussion.
I do not strive to convert you or anyone. Why would I? Feel free to eat what you want.
I am not impressed by the "thousands of papers and 50 years of research", including those you referenced.
I favor quality over quantity. I find the existing scientific literature in medicine and nutrition to be generally unreliable and their scientific standards appaling. The lack of good direct data forces me to rely on indirect studies, metabolic theories and my personal experiences and observations.
If all those dire warnings against animal fat and cholesterol comming out of your references and your own writing were to a some degree valid I would be dead 8 years ago. You can say that I am a living proof that not all you believe in may be factual.
Regards,
Stan (Heretic)
Bruce,
You may be right about PUFA. Vegetable oils now form a majority of fats served in restaurants and fast food. Based a personal observation most people around us (Canada) rarely consume any natural saturated fats since everything is low fat or fat free, and the cooking fats are all veg oils. Probably most saturated fat they ever get is from hydrogenation.
You are of course correct that the first thing that got rationed in Europe during WWII was sugar! You could still buy some black market pork (it was strictly controlled under occupation in Poland) and butter - it was extremely expensive but there was no sugar. Heart disease came down indeed but infectious diseases like TB, typhoid etc skyrocketed.
Stan (Heretic)
Not just restaurants and fast food. All the junk foods (just about) are made with PUFA oils: corn, soybean, canola, rapeseed, safflower, cotton seed, sunflower, etc. Almost all of the prepared foods and frozen food. Mayonnaise, salad dressings, and so forth. Even things like salsa often have PUFA oils added. It has become pervasive. It takes a lot of effort to avoid these oils, because you're getting them in restaurants and all kinds of prepared foods.
Very few people eat "high saturated fat" diets any more, because it's a full time job to avoid PUFAs. Peter is doing pretty good, IMO, but it's getting harder every day. Everybody wants low-fat dairy and lean meats, so who is eating high saturated fat diets? They're eating fat, but they got rid all the saturated fats, and replaced them with PUFAs.
This is one of the reasons there is so much disease, I think. Back when food was made saturated fats (suet, coconut oil, butter, palm oil, palm kernel oil), people were healthier. Vegetable oils are sabotaging human health on a massive scale.
Brian,
Brainwashed, huh? My husband got a laugh out of that because he thinks I can't possibly "just accept" anything without looking for more info. What about the low-fat brainwashing? Most of the people who follow the low-fat theory haven't questioned it at all. And Peter must be brainwashed, too. His recently mentioned book list is nearly exactly like mine, though I have finished the Taubes book and didn't finish the Colpo book due to the tedious faint, small type.
You could be right about me taking too much time to respond, though. But slow cooking a nice fatty cut of chuck roast allows for that, unlike tasteless, dry skinless, boneless chicken cutlets.
And I'll cop to the "proof" copout. Of course I wouldn't bother trying to convince you saturated fat doesn't cause CVD any more than I would try to convince a Jehovah Witness to stop their door-to-door conversions. Changing a mindset has to come from within a person, not an external challenge. I started questioning the low fat dogma because it did look like brainwashing to me, and certainly I wasn't seeing healthy outcomes from it, not for me, nor for my family.
You don't really sound open to new ideas at all; you sound like you are extremely opposed to what is presented and discussed and bandied about on this blog and you want to convert us to your view, one that the vast majority of this blog's readers (the vocal ones anyway) have either begun to question or have already dismissed. And the comments to this particular post, while remaining cordial, don't have their usual tone: the friendly exchange of a variety of ideas (discussion) has taken on a more debate-like exchange.
If one doesn't need/want to get into a debate, then there is no point responding to a challenge of "proofs". For those who want to debate, great. I'm learning from that, too. But it doesn't mean I want to debate.
Best wishes in your quest to correct the heretics,
Anna
"Thanks for the list of references, but I already feel there is far too much cholesterol hypothesis bashing on my blog."
I don't understand this sentence at all. Most of us here agree with you that cholesterol and animal fat has been wrongly blamed. We are bashing the cholesterol hypothesis.
Maybe you meant to say "cholesterol bashing"? If so, then only Brian is bashing that, AFAIK. Please clarify what you mean here. Thanks.
Hi Bruce,
I bash the cholesterol hypothesis! That is I put up info that shows it to be the garbage it is. It never occurred to me that the term "bash" could be used as meaning promoting!!!!!! But yes, I agree it can and is used this way. As you say, Brian is the main "basher" in these terms.
So to re phrase, I and most commentators spend too much time criticising the cholesterol hypothesis.
That is, none of us believe it (I hope) and wasting our time on it is pointless. The more amusing papers which come to hand I'm happy to post and discuss, but if any of us sat down and trawled through all of Brian's references and tried to explain to him the difference between an anti inflammatory like lovastatin and a simple toxin like cholestyramine we'd be there all day for weeks and who want's to read this again and again?
So yes, not the clearest phrase!
Peter
I didn't mean that "bash" was slang for "promote." Is it? I just didn't get why you said there was "enough" cholesterol hypothesis bashing like that bothered you. There will never be enough! Until Dean Ornish admits he was wrong, or dies from his poor diet. Reading it again, your remark makes more sense, but thank you for clarifying your meaning.
I have been absent for some time investigating your gurus – at least the names you gave me – more on these later. Turning to your comments on Blankenhorn, the study people were, as I said earlier, all in the placebo group and therefore did not take any Colestipol and so this cannot be a cause of anything. My apologies on the OR’s. There were a number of OR tables. The odds ratios I gave you from this study were correct, but were per 1% increase in the quoted dietary source. The actual overall OR of developing lesions over 2 years from nothing were as below, for each quartile of each type of fat:
Total fat; 1.0/5.0/6.67/12.31
Sat Fat; 1.0/1.06/1.87/3.23
PUFA; 1.0/3.53/8.57/12.31
Mono; 1.0/2.0/0.3/4.91
Remembering that carbs were nearly the same in both the lesion and non-lesion groups, (and in fact the non-lesion group were a bit higher on carbs), it is clear that total fat, saturated fat , PUFA and monounsaturated fat were each independent risks for lesion development. I started off quoting this study as one showing that a high fat diet caused lesion development. I still maintain this, and for those of you who like to blame PUFA’s, here is some evidence, but we aren’t told what the split of PUFA’s between natural and unnatural is. Higher saturated fats also had twice or three times the risk of lower or no saturated fats. Higher monosaturated fats had a fivefold increase in risk over low monosaturated. I realise that acceptance that saturated fat is in any way to blame will be difficult for you to accept, and that this is why you have been trying hard to wish away this study blaming first carbs, possibly Colestipol and then PUFA.
I have been urged by some of you to look at your gurus as quoted to me – Colpo, Ravnskov, Kendrick , Taubes and Masterjohn. My local bookshops did not have any of their books so I searched on Pub Med for studies by them, then got the full text.
Colpo, Masterjohn and Kendrick
I can’t find any research on Pub Med by A Colpo. Masterjohn seems to have published only two papers– one on safflower oil and the other on Vit D. Kendrick has published a number of papers but out of the 50 I looked at going back 7 years I could only find 3 with relevance to fats/cholesterol. One was whether women should be offered statins, another on stroke in the Japanese and the third about insulin resistance and its connection to CHD. Nothing controversial or new in them. Amongst this trio, no new studies about fats/cholesterol, and nothing that hinted that high fat diets of any kind are good, or that the diet/heart association is wrong.
Ravnskov
Most of Ravnskov’s papers seem to be complaints to medical journals about cholesterol/fats papers they have published. He seems to be an independent researcher whose area may be glomerulonephritis – about which he has published a few papers. Going back over 50 of his papers, there is a scarcity of any study on fats/cholesterol he has done himself or with others. The main thrust of his argument against the diet/heart association seems to be a statistical review he did in 1998 entitled The Questionable Role of Saturated fat and Polyunstaurated Fatty Acids in Cardiovascular Disease. This was published in the Journal of Clinical Epidemiology. It is difficult to summarise as it is 15 pages long with 7 tables of statistics and 112 references. In a nutshell, he looks at mostly old studies and national food consumption data and challenges their support of conventional thinking that saturated fatty acids(SFA) and/or PUFAs are an important cause of CHD. My concerns with this paper are listed below.
• The data often combines SFA with PUFA and this disguises any individual effect
• Other types of fats are not mentioned – eg trans fats
• I was concerned that unpublished data was used – what data we aren’t told, or how it affected anything he did
• In the first set of studies – observational, 5 were in support of conventional wisdom, 3 were not.
• In the second set, he attempts to correlate changes in food consumption data in many countries and mortality changes. However, there seems to be no logic between the years of one against the other. For example, consumption changes for 1947-1962 are compared with mortality changes for 1955-1965 in 23 countries. Since there is no group of people being followed, I cannot see the connection. It is impossible to draw inference from this because any consumption change would need to be followed up for many years to see a difference in mortality. His ‘follow up’ period is almost identical to the consumption period, lagging by only 3 years. This same method is used for all the countries and periods. The results are that some changes in fat consumption seem to support conventional thinking, some do not. I question the basis of this methodology
• The third analysis is of racial differences. There are some very small numbers in the 38 studies he looks at - <100 people is common. He performs a number of mini meta analyses by grouping say 5 or 6 studies of the same race. Whether these meta analyses are valid or not I can’t judge. I counted 17 studies in support of convention, 21 against,
• A fourth analysis looks at cohort studies – 28 groups with follow ups over 4-23 years. In 3 groups, high fat was associated with CHD, in 1, the opposite, and in 24 no difference. The difficulty with this is the data presented doesn’t show the source of fat - hydrogenated etc and so I can’t comment on why there were apparently no major differences in fat consumption between those who got CHD and those who didn’t. Cholesterol levels including LDL are not stated.
• The fifth table looks at CHD on autopsy in 4 studies. Its major problem is that diet/fat information was, typically, a one-shot look 12-17 years prior to death and so tells us nothing about diet in the intervening period. In my view this is worthless.
• Table 6 is on case control studies, 6 groups, mainly small groups of around 150 people. There were no significant differences between these
• Table 7 shows experimental studies – 9 in all,2 primary in which fat was reduced. One study found lowered CHD in the lower fat group, the others, no difference. The so-called reduced fat levels were far too high – between 20-38% to expect any substantial benefit in CHD/mortality. Another worthless comparison.
So, out of this, many studies in support of the diet/heart concept, some not, some irrelevant and a good number neutral. This analysis, if we are to believe its accuracy, does not demolish anything – least of all the diet/heart connection. It raises some inconsistencies, but that is all. It also does not support the view that a high fat diet is beneficial.
A critique of this study was also published in the same journal. (Dr Golomb). Whilst acknowledging Ravnskov brought to light some important limitations in these mostly early studies, she makes the following critical comments:
• A rather less extreme conclusion would be warranted in her view
• Ravnskov gives insufficient consideration to the impact of diet as a whole on CHD
• He does not consider the separate impact of SFA’s and PUFA’s, or monounsaturated and trans fats
• Studies he cites seem more consistent with a relation of SFA rather than PUFA to CHD
• Other important covariates that may modify risk of CHD are inadequately considered
• Some of his own analysis serve to support rather than discredit a relation of SFA to coronary mortality
• He has a tendency to label as contradictory evidence that is at least partially supportive.
My impression is that he has carefully cherry picked studies to find those with inconsistencies in the fat/cholesterol/CHD argument and used these. He did not find a universal connection damning this argument – there were many studies which supported it. Additionally some of his analysis did not prove anything as I’ve mentioned above. A common problem in dietary interventions is that, typically through lack of compliance, they do not reduce fat consumption by enough(eg down to levels around 10%). It is typically only at this sort of level that benefits of improved mortality are seen. (This is also a common problem in some statin trials, where diet was not altered – people simply took a pill. They did not achieve a substantial enough cholesterol reduction nor alter their postprandial blood fat levels and thus the mortality benefits were lower than could be achieved with dietary improvements as well as the statins).
My biggest criticism is the concentration on very old studies, whose authors are generally not around to explain/defend, and his omission of pertinent studies which would have shown the opposite of what he is claiming. A Pub Med search on ‘Saturated Fat’ revealed 6047 papers – Ravnskov looked at some 60,deliberately chosen to help his pitch.
The second paper by Ravnskov that I obtained was published in the same journal as that above, 2002 55 1057-1063. (A Hypothesis Out of Date: The Heart Diet Idea). The first part of this is a repeat of the data from his 1998 study above. He commits the crime of exaggeration fairly early when he says that the epidemiological and experimental studies he reviewed in 1998 were ‘almost all inconclusive or, indeed, flatly contradictory’. Thus is untrue. Many studies were supportive in the 1998 paper, depending on what categories he wants to count in epidemiological – as this is not a category in the 1998 paper. The same omissions apply here as to the 1998 paper, but more so. In 2002 there would have been more studies to consider, yet he does not. He rests on those he included in 1998, yet even these were not a reasonable cross section.
He claims (exaggeration again) that ‘almost all’ postmortem studies show the degree of atherosclerosis to be independent of cholesterol levels, then cites only two postmortem studies.(One of which is 1936). See later comments on these.
He makes the same point as in 1998 that the dietary intervention trials showed little effect on mortality. I have dealt with this above as being largely because such interventions fail, through lack of compliance, to reduce fat and cholesterol by enough to show up in mortality. This is another example of omission. Why hasn’t he reported this well-known aspect? Because it isn’t convenient to his theory.
On the statin trials, he makes the comment that these have been successful, but this success was due to their pleiotropic effects and had nothing to do with cholesterol lowering. A 1998 reference is given (55)but this does not deal with pleiotropic effects. I have seen many papers on these effects and none of them that I can recall got close to saying this explained ALL of the benefit of statins. He floats this contentious theory without foundation.
He argues that an excess of saturated fat may raise cholesterol in a laboratory but then invents a new theory why this should not apply in humans. He contends that high cholesterol may be a marker, not a cause. A marker for something else that causes CHD. What is this secret, I eagerly read? His answer, smoking, lack of exercise and mental stress. These he claims are probably the most important elements in causing CHD. They all raise cholesterol he says, and that is the only connection of cholesterol. THE EVDENCE FOR THIS EARTH-SHATTERING THEORY COMES FROM JUST 3 STUDIES. Wow! There you have it folks. Everything everybody has missed/falsified for 50-100 years is undone by these three studies. He had over 60 studies in his 1998 paper trying to tear down conventional wisdom, but only 3 to support his theory. I got the full text of two of these, thinking they must be very important.
Let’s agree immediately that smoking is a definite CHD risk – though I have no idea whether it raises cholesterol. But lack of exercise and stress as the other major contributors? What happened to the other possible major culprits such as hypertension and diabetes/refined carbs. Are we to believe these have gone away? If we don’t smoke, meditate and exercise that’s us all fixed? What happened to the hundreds of thousands of people who never smoked but died from CHD? Jim Fixx ran a lot and also died from CHD. I ran thousands of kms and didn’t smoke for long, but I got it. What happened to the importance of genetic tendencies such as hypercholesterolemia and others which hasten CHD? I found it hard to give this theory any time, but I read on hoping to find something useful.
Sure, lack of exercise is an issue, and I have dozens of studies showing adding exercise will reduce the risk of CHD- mainly through it raising HDL and assisting lowering inflammatory conditions. However I have never found any study that said it would protect us from or fix heart disease and his single reference study from 1999 doesn’t say this either – see later.
On stress reduction, his 1995 study examined the effects of short term frustration test on blood lipids etc. A stress effect has not figured as a main culprit in any study I have seen, and stress as a major cause of CHD is certainly not supported by this one – see later.
I’m afraid his theory just ignores many studies showing the conventional risk factors, has no evidence to support it, and, importantly for me, no advice on a lifestyle approach that would fix my CHD.
The final paper that I looked at by Ravnskov was ‘Should We Lower Cholesterol As Much As Possible? BMJ 2006 1330-1332. Here he criticises NCEP guidelines published in 2001 for advocating lowering LDL in those with CHD to less than 1.8(70). As is now his pattern, he omits much evidence that lowering LDL is beneficial in saving cardiovascular events and improving mortality and cherry picks studies which don’t support this. He also raises FUD about very low cholesterol and implies a cancer risk which isn’t there. (An earlier blog of mine gave a number of references to papers explaining the association of low cholesterol to cancer is not causal.) Again this shows he’s not trying to be impartial and is selling a particular line. A correspondent claims that a Q10 study he references does not support his point; something I have fund occurs too often with him.
Omissions, Misrepresentations and Spin
Unfortunately Ravnskov is guilty of what can only be misrepresentation – mainly by omission. I did not have time to check all his referenced studies but I found serious misrepresentation in some of the ones I did with omissions enabling him to portray a study as having a completely different result. This cannot have been accidental. On this showing, I would be confident of finding more – why would I have, by chance, found all of them? Here are some examples.
His 1998 paper(and also the 2002 paper) makes a big splash about an Indian study from 1967 in which apparently CHD in railway workers in the South was 7 times higher than in the North, despite the fact that those in the North ate up to 19 times the fat of people in the South. Inexplicable on the surface and a great advertisement for Ravnskov’s arguments. However, this Indian study is full of surprises which are completely at odds with conventional thinking – and in fact Ravnskov’s own new theory of CHD. The same study found that high sugar consumption resulted in less CHD, higher smoking also. In addition, CHD was lower in sedentary workers against those in high physical exercise jobs. Who would believe all this? Yet Ravnskov did not report these anomalies. Nor did he report that diet was measured in three ways – and how these were combined we aren’t told; interviews with ‘a number’ of railway workers and their wives, ‘several’ roadside eating houses and diet surveys from the Indian Research council. Though the study author states the higher fat consumption in the North, the diet graph does not show fats. There is a much higher milk consumption in the North, and higher oils, but this would not explain a statement like between 8 and 19 times the fat. Where are these data? Certainly not published in the study. Finally, in an attempt to explain the apparent contradiction of CHD versus fat, the author says at the end of the paper that food preparation methods, North and South, were markedly different, resulting in a change to the nature of the fats such that in the North, fats would have been mostly short chain fatty acids, whereas in the South, mostly long chain. This, combined with the bile flow effect these fats have, could have resulted in less damage from the greater fats consumed in the North. Ravnskov prefers you do not know this, since it spoils his story.
In his 2002 paper he describes data from the Framingham study as showing that a decreasing cholesterol level predicted an increased risk of CHD and total mortality. He quotes a 1mg/dL drop in TC as generating an 11% increase in risk in both types of mortality. The actual figure is 1mg.dL per year. However, he omits to tell us that this effect is only seen for the first 14 years of follow up, and the authors’ explanation is that spontaneously falling TC levels(there being no treatments for reducing TC in the 1950’s) indicates the onset of diseases which can reduce TC levels,(eg cancer) and leading to death. Another big omission is the major conclusion of the study – that for people under 50 years at follow up, there was a definite association between higher TC levels and death from CHD/all causes with respective increased risks of 9% and 5% for each 10mg/dL increase in TC. (Over 50 years old this increased risk was not evident). This omission enables Ravnskov to completely misrepresent the study’s findings.
His 2002 paper refers to two papers – one 1936 and one in 1982 which looked at severely narrowed arteries – 5mm segments after autopsy and their correlation with cholesterol levels. He uses this study to justify the words ‘ In almost all post mortem studies the degree of atherosclerosis was independent of blood cholesterol’. Firstly, ‘almost all’ is a huge exaggeration. There were only 40 people in the 1982 study and there would have been many more post mortem studies in the 66 years between his paper and the 1936 study which of course are not included. Secondly, the degree of atherosclerosis WAS correlated significantly with cholesterol. Cholesterol correlated significantly with the number of severely narrowed arteries per person (but not the overall % of severely narrowed segments). Secondly, triglceride level correlated with the overall % of severely narrowed arteries. The 1982 authors also mention the results of 6 other similar studies, of which 4 found associations between cholesterol/triglyceride levels and narrowed arteries, but Ravnskov doesn’t tell us any of this. For completeness, I should note that 5 people had severely narrowed arteries despite the fact they had low cholesterol – an anomaly that Ravnskov should have picked up since it helps his argument.
His 2002 paper references a 1998 study(55) with the claim (about statins); ‘...the effect on CHD reduction has been achieved independently of the initial cholesterol level or the degree of cholesterol lowering’. Once again this is totally misleading. This paper actually says ‘ The rate of coronary events was associated strongly with the plasma LDL concentration during treatment...’ His comment that CHD reduction was not associated with initial cholesterol level is correct, but so what?
In Ravnskov’s 2002 paper, he claims one of the main causes of CHD is lack of exercise and quotes a 1999 study to support this. The problem is this paper does not support the proposition at all. The 1999 study was a meta analysis of 33 randomised trials looking at the effect of exercise on cholesterol levels. These trials did not even look at CHD, and so Ravnskov’s use of this study to support lack of exercise as a major factor in CHD is invalid. In fact the study only showed that over all trials, exercise had a modest and beneficial effect on cholesterol levels. The authors did not in any way suggest exercise will help avoid a heart attack. They even urge caution in drawing conclusions because of the significant heterogeneity in the 33 trials, and also poor quality in some of them.
Another study Ravnskov cites to underpin his new theory of CHD is a 1995 study on stress. Remembering that he is claiming a new theory, one would expect a little more than one stress study, and more than 41 people being frustrated in a 20 minute test. Their cholesterol levels rose(as well as BP and heart rate), but, to the delight of the Hyperlipids, both LDL and HDL rose – an effect I believe some use to justify eating saturated fat. I think we all need a little more than this to believe that stress is a major factor in CHD, although it may well be a contributor.
The problem for Ravnskov is that he makes himself out to be the white knight of truth – writing to journals railing about biased studies with omitted and misrepresented data yet commits the same crime himself and ruins his credibility. He has cherry-picked studies that show anomalies he can manipulate rather than look more closely for their possible cause. Instead he invents a new theory of CHD which is not based on anything, and fails to support this with substantial data. For his theory of CHD to hold water, he has to explain away the scores of studies showing the association between fat/cholesterol and CHD not just some old studies which are no longer even relied upon for this evidence. Applying his own logic about theories to his theory, if there is evidence that supports another explanation(plenty of studies showing a fat/cholesterol/heart link) he must abandon it and look for another.
Taubes
I found Taubes had 114 articles published. In the first 50, all looked to be in the Science magazine on anything from magnets for cancer to black holes. There were 4 touching on cholesterol/saturated fat. I couldn’t see any new research he’d done in these areas. I found 2001 articles described below where he argues against conventional wisdom on heart disease.
I read both 2001 articles – What If Americans Ate Less Saturated Fat, and the Soft Science of Dietary Fat. The first is completely irrelevant in my view. It seems to be a ‘conspiracy’ article about the burying of three computer analyses that modelled life expectancy in the USA after reducing saturated fats. I lost interest as soon as I saw computer modelling, and also that the ‘reduced’ level of fats was 30%. The fact that little extra life expectancy was forecast at this level of fats does not surprise me, as it has been shown many times that dietary fat must be much lower for any mortality improvements. In any case, computer models are not studies of real people with measured outcomes.
The lengthy article on the so-called Soft Science of Dietary fat I found boring in that it concentrates on how American Health decisions were made(if correct), all the while trying to build a picture of incompetence and political/commercial/personal influence. He also tries to paint the picture of poor science.
However, ignoring the history lesson, there is s startling lack of references to studies to support the bad science claim. I could only find 16 studies on CHD or related topics mentioned. This is totally inadequate to cover what he describes as a 50 year old history. As does Ravnskov, Taubes omits many studies supporting the diet-heart connection, since these are inconvenient for the line he is selling. The same applies to the experts who are quoted. The full picture of pros and cons is not presented. I understand some of the experts complained that what Taubes wrote did not represent what they had told him.
He agrees that LDL can lead to plaque –in contrast to Ravnskov who seems to be saying cholesterol doesn’t matter. I had to laugh at the fallacy that rasing both LDL and HDL through eating saturated fat is somehow OK. It isn’t! The HDL won’t protect you enough. A much healthier way to raise HDL and lower LDL is through exercise. He touches on the old potato of low cholesterol causing cancer, but at least reports the other view. The Lyon Diet Heart Study is mentioned with two groups - one on a Mediterranean-style diet. Despite similar cholesterol fractions in the two groups, the Mediterranean group had less CHD. What Taubes doesn’t say about the Lyon Study is that the Mediterranean group ate significantly less saturated fat than the control group, instead, olive oil. There was a trend towards reduced LDL and increased HDL in the M group and, CHD risk was reduced in this group by 78% at the two year point, and 65% at the 4 year point. This emphasises do not eat saturated fats –replace them with healthier oils.
There are various criticisms of his article by Scott Grundy and Hegsted. The former makes the point that Taubes has underestimated the bad effect of SFA’s which have a bad effect on LDL and then CHD. Grundy says there are many high quality studies to demonstrate this. I tend to agree with him as mentioned above, and it is remiss of Taubes to have carefully avoided these. Hedsted says when he was in the health system loop, he recommended reduced salt and sugar and increased unsaturated fats, whole grains and fruit and vegetables in place of SFA’s. He also makes the point I’ve been making in a different way. A 30% fat diet, (dangerously low for brain function says Taubes), cannot be regarded as low in terms of effect on the brain, and Hegsted says he hasn’t noticed too many brain dead Japanese or Chinese whose traditional diet is lower in fat. Someone called Anne Astrup also argues with Taubes and emphasises that low-fat diets high in fruit, vegetables and fish have other beneficial effects through pathways such as blood pressure, thrombotic, fibrinolytic and arrhythmic factors. I see also that Ravnskov gets in on the act and argues with Grundy about the correctness of his references. I haven’t had time to follow these up.
In a nutshell, Taubes tells a good political conspiracy story and chooses studies and people to complement this, but does not portray a useful overview of both sides of the diet-heart debate. In the context of you Hyperlipids, and my search for what benefits are in high fat diets for me, he does not advocate, nor do any of the people or research he mentions, a high fat diet as being beneficial for anything, certainly not CHD.
How Do My Gurus Stack Up?
I won’t bore you with dozens of these, and remember they are picked not because I am coming from a direction, with a placard, having had my training(eg vegetarianism). I found these by following many trails on heart disease – particularly regression.
These people, in contrast to the Hyperlipids’ gurus, are experts in heart disease. They have studied it, treated it and sometimes regressed it. Esseltyn’s published studies showed he had regressed the heart disease of people given up by their cardiologists who then lived on for many years, but you say this is vegan propaganda, destroyed by others. I don’t think his thankful survivors would agree with you. Gould’s 700 page textbook on heart disease and regressing it is used world wide. I have it and it has over 500 referenced studies to back up what he says, apart from his own published regression research. There is a chapter dedicated to regressing heart disease, and some 85 clinical case studies with positron emission tomography scans showing before and after blood flows to the hearts. Amongst the many other experts I originally cited, these are people who have cured heart disease apart from researching it for 30 years. Against them we have the Hyperlipid gurus above who I suspect have never cured anyone of heart disease, nor done any new research in it themselves and whose published papers contain misrepresentations. How credible can I regard these?
More on Diet/Fat/Heart
Over the past week, and in the absence of any studies from the Hyperlipids about high fat diets regressing CHD, I did a Pub Med search just to make sure I haven’t missed anything vital in my earlier research going back some years. I entered ‘Saturated fat and CHD’ and got 216 hits. I looked at the abstracts of the first 100 of these. Excluding about 20 as being not relevant (eg educating people/ gender and ethnic differences....etc), some 76 said that saturated fat was associated with CHD and many advocated replacing most of this with PUFA (n-3 type).
These studies are additional to my list of 80 I gave you a couple of weeks ago, yet are of the same mind. There were only 4 in this 100 that had different views. The 100 went back to 2000.
This new Pub Med search didn’t reveal any advice about a high fat diet being good for CHD(or anything), and study after study stressed the connection between saturated fat and CHD. I hasten to emphasise I did this search in order to satisfy myself I had not missed something useful about high fat diets that you were not telling me, not to convert anyone. However, as usual, I share this information with you. Out of the first 100, here are just a few, which do not rely on the old studies that Ravsnkov tried to discredit.
1. Lipoproteins nutrition and heart disease ( American Journal of Clinical Nutrition 2002 75 191-212 Schaefer EJ).
This 20 page review of what is known about these topics does not rely on the studies Ravnskov tried to discredit in 1998. I can only find about 6 of these out of 219 references. Instead it uses much later(eg 1990) studies on fat metabolism. The paper is a comprehensive background(given as a lecture) on all the major contributors to heart disease and the mechanisms by which these work. It covers all the cholesterol fractions, lipoprotein(a), particle size, saturated fat(giving stearic acid a tick), PUFA, mono, trans....genetic factors....Dietary fat is a major contributor to CHD it says. The conclusion lists smoking, high glucose, high blood pressure, high LDL and obesity as risks for CHD and recommends decreasing saturated fats, total fats, sugar, trans fats and increasing fruits, vegetables and grains, coupled with exercise.
2. Diet and Coronary Heart Disease: Clinical Trials (Current Atherosclerosis Reports 2000 2 387-493. Brousseau ME et al)
This deals with many dietary intervention trials and the effects of fats. Of interest is the number of very early studies showing beneficial effects of reducing fats from the 1950/60’s which I had never heard of and which are not those Ravnskov tried to discredit. No wonder he didn’t mention these. The review also looked at Vit E supplementation and concluded there was little evidence of benefit. The conclusion was that restrictions on total and saturated fat significantly lowers CHD risk, whereas a relative increase in n-3 fatty acids is protective. Smoking cessation and an active lifestyle are recommended.
3. Dietary Fat Intake And The Risk of Coronary Heart Disease : The Strong Study.(American Journal of Clinical Nutrition 2006 84 894-902. Jiaqiong Xu, et al)
This starts off with words to gladden the hearts of the Hyperlipids – listing some studies which showed a diet/heart association, some that did not, and talking about inconsistencies and deciding to do a study of their own – of about 3000 North American Indians. The people were free of CHD at baseline and second examination and were followed up for 7 years. The conclusions were that total fat, saturated fat and monounsaturated fat were strong predictors of CHD mortality. The hazard ratios for people in the highest quartile of each type of fat intake were; total fat 3.57, saturated fat 5.17, mono 3.57, all after controlling for confounders. These associations were in people aged 47-59 and were not seen in people 60-79. The authors propose this may be because the 60-79 group might be a specialsed group in that existing CHD people were excluded at the outset, and these would have been predominantly in this age group.
Which Theory Wins?
It is true there are unexplained anomalies in the diet/heart approach, (eg the Nurses Study which I haven’t had tome to look into), but it’s also true that the ‘high fat/cholesterol doesn’t matter’ idea is not supported by many numbers of robust studies – rather mostly by exploiting the anomalies above. It’s also true that people on both sides of the debate are guilty of omission and misrepresentation to sell their stance. However, I am not persuaded by the evidence from Ravnskov and Taubes which seems to be trying hard to find facts to fit a theory, with both of them omitting and misrepresenting data. The number of studies that show the fat/cholesterol connection is overwhelming, particularly those which show the regression of CHD by extreme low fat diets/lowering LDL/exercise....fibre, fruit and vegetables etc. I keep on saying that after all this searching I have still not seen a study that shows such regression on a high fat/high cholesterol regime.
Summary
Some have complained about my excessive free time. I am spending too much time at the moment explaining what is in studies in answer to various questions. I have also spent a large amount of time looking into the gurus you put up. This is becoming a one way street. I originally listed various studies supporting the diet/heart concept which were howled down without being looked at. Ravnskov, I was told had demolished all these except he hasn’t, and his credibility is lacking as shown above. The recent Pub Med search above showed over 70 more in support of the diet/fat/heart concept. I doubt many of you would have checked ‘my’ studies or even Ravnskov’s. I get the impression you are getting your ‘facts’ from books with controversial slants, and don’t seem to go to the actual studies upon which the books are based and question these. Dissident medical theories seem to be readily-accepted, whether there are substantial studies to support them or not. In contrast, I have been slogging it out at the medical library at the University for the past 7 years getting and reading the full studies in nearly all cases. I’ve answered Peter’s objections to the MR FIT study but he hasn’t had the good grace to help me with some vital information. One of you raised the theory that all the successes of low fat regimes are to do with weight loss. I don’t have time to list for you the ‘same weight’ studies that have been done or keep on answering your questions and looking into your unproven theories, especially when the information I asked you to volunteer has not been forthcoming.
You know my background, I have heart disease. I am on an extreme low fat/low GI carbs/high fibre/vegetarian plus fish/exercise program that I think will either regress it, or arrest it. In an earlier blog I presented some 12 regression studies which supported this approach. I my recent trawls I found another one I had forgotten about – Morrison 1946, JAMA 1955 159 1425 and JAMA 1960 173 884) in which people who had survived heart attacks were split into two groups - an extreme low fat group and a control group. At the 12 year follow up the low fat group had much less CHD mortality. I was astounded by the philosophy your blog expounded, I picked Peter up on his invitation to learn about where he may be going wrong, then from his responses I realised this was facetious. From the strength of the responses, I was intrigued that perhaps I had missed something vital. I can’t find it. You seem to think my program is wrong, yet I cannot get one of you one to contribute any studies which show that a high fat diet, versus what I am doing, will help me regress my heart disease. Despite Anna trying to make me into a disciple of some sort, I am, to the contrary, actually still asking for this assistance. All I get is read this that or the other book. Books by people I’ve now investigated and who do not have heart disease credentials, and without any hint that these books refer to such studies. Are you, as a group, really this uncharitable in the way of information? The other things I’ve asked for include whether any of you have had tests to see whether you are clear of heart disease on your high fat diets, and what your LDL levels are – again nothing forthcoming though I have told you mine. These are reasonable questions that anyone coming to this blog might ask to try and determine the soundness of your beliefs.
The gurus you mention seem lightweight to me and not experts in the CHD field. Ravnskov’s new CHD theory is unsupported. None of the gurus you’ve given to me advocate a high fat diet in any case.
Once again – let’s by-pass the debate entirely and cut to the bottom line as I’ve asked before – are there any studies supporting a high fat diet with regression or heart disease that might question my program? And – how are you people travelling in terms of heart disease yourselves –do you know the state of your coronary arteries, what are your LDL’s?
I could raise all sorts of other matters with you which might be at issue with your chosen diet – well at least the one Peter says he’s on – like the beneficial associations between cereals, fruit and vegetables, exercise, fibre and CHD. Also the increased risk of certain cancers from meat/meat products, but I see no point in opening up the debate into these areas, because to me, you seem, as a group, encouraged by Peter, closed-minded, and overly defensive if anyone questions your beliefs on high fat consumption.
So, there seems to be nothing much in this exercise for me, if no one is willing to volunteer this simple information. I have asked for this now I think three times. You leave me with no other impression that as far as the studies go, they just doesn’t exist and that the case for a high fat diet being safe for CHD or helping fix it has no evidence.
I think by the way that the comment by Bruce K about Ornish dying from his poor diet is a pretty pathetic level of comment which he ought to delete.
Brian
you are obviously doing a lot of work on this, which is to be commended.
One point - there is more to look at than is on pubmed. Kendrick and Colpo have written books on this and Masterohn has a website:
Kendrick 's book and Colpo's book are both called "The Great Cholesterol Con" of the two I'd recommend Kendrick's.
There is a series of videos on YouTube of a talk Kendrick did to doctors in Leeds explaining his position.
This is also a good point from Kendrick re MONICA:
http://tinyurl.com/2c3yj8
Here is another Kendrick article for you on Statins:
http://tinyurl.com/2a3ar5
Bill Sardi reports that Coronary Calcification Predicts Future Heart Attacks and Coronary Death. Cholesterol Not Found To Be A Significant Risk Factor.
He cites this new study: Coronary Calcium as a Predictor of Coronary Events in Four Racial or Ethnic Groups in The New England Journal of Medicine, v358 n13, March 27, 2008, pages 1336-1345.
Nice link, thanks. I tend to keep Bill Sardi at arms length, but he does come up with some good stuff occasionally. This is one! Pity my Athens account doesn't get me access to NEJM.
Peter
The problem with the current theory of heart disease is that we have a lot of data that can used to both prove and disprove that cholesterol is the main factor. There are many papers sited by Brian that strongly suggest that an elevated LDL be a major cause of CHD. Unfortunately there is not a doctor or scientist on the planet that can definitively explain the mechanism by which LDL causes atherosclerosis. Then, there are the anomalies that go against the current thinking. The people (like my aging dad and his sister) who both have high cholesterol (7 and upwards) but do not have any sign of heart disease. We have whole countries that go against the grain. France and Switzerland both have high saturated fat intakes and have average cholesterol levels in the high 5’s. Yet their heart disease rates are one quarter of those in the UK. Then we have fit athletes that have low cholesterol but still go on to develop heart disease.
The main problem I have with all this is that you would think after all these years (about 50 years have passed since Ancel Keys released his 7 countries study) that somebody somewhere would have nailed the cause of the biggest killer in the Western world. Yet, here we are, still arguing what we should do about CVD.
For the moment, the best I can say on the subject is that raised LDL is one risk factor amongst many (such as diabetes, raised blood pressure, sedentary lifestyle, family history... etc). But to call it THE cause of heart disease is unsupported.
Hi pc,
I never could be bothered to slog through Brian's posts. Once people start confusing the cholesterol lowering effects of statins with their antithrombotic effects my eyes glaze over and I'm asleep within 15 seconds. Whatever else they come up with is dwarfed by this mistake...
Peter
You are not alone, Peter. I stopped reading Brian when he said my claim about Dean Ornish having a bad diet was a "pathetic" comment and should be deleted. I've read reports about how Dean eats in public: white rice and steamed vegetables with no fat. Maybe some egg white omelettes. His diet is absurd and he is very pudgy and unhealthy looking. I don't feel it's pathetic to point out that Dr. Ornish looks pasty and overweight.
Brian
You are obviously not aware of the 18 randomized clinical dietary intervention trials to date that have NEVER shown saturated fat restriction to lower CAD mortality, CAD incidence or total mortality.
You obviously have NOT read the LRC-CPPT or you would know thew difference between the treatment group taking cholestyramine ands the controls was 30 vs 38 CAD deaths Total mortaluty was 68 vs 71 This trial WAS NOT supportive of cholesterol lowering.
Nice try though using the "relative risk"......
You are obviously not aware that statins have at least 11 other strong pleotropic effects besides mere cholesterol reduction. The PROSPER trial numbers, you ask ?
participants treatment controls
2891/2913
treatment controls CAD deaths 94/122
298/306
total mortality
Lastly you are obviously not aware of Ancel Keys chioncanerty in the Seven Countries Study, deliberately omitting data available at the time that did nto fit his false idea. And even with this trickery CAD rates varied 2-6 fold WITHIN nations.
All the info needed to DEBUNK Brian's nonsense is in Anthony Colpo's The Great Cholesterol Con.
I suggest you strat reading the studies, Brian, FULL TEXT and VERY CAREFULLY next time. NO MISLEADING SUMMARIES
Thank you
P.S. Brian is ALSO unaware that Dean Ornish's Lifestyle trial was MULTI FACTORIAL. He violated the control you variables rule. Participants meditated, exercised, reduced stress, included fruits vegetables ( rich in anti- oxidants ) ALL these things have been shown in clinical trials to reduce CAD mortality
Saturated fat restriction has NEVER been shown.
NO ONE can make the claim low fat diets reverse heart disease UNTIL Ornish has low fat group and saturated fat rich group and EVERYTHING ELSE is the SAME that they do.
LOW FAT DIETS SEVERELY LOWER ANTI OXIDANT status and REDUCE absorption of EPA and DHA and nutrients from fruits and vegetables.
This ancient post needs an update.
Peter's Agatston (CAC) score is 7. <10 =minimal risk of CHD or MI.
/discussion
I have been on a very low sugar/grain/PUFA diet for a couple of months, primarily influenced by some paleo bloggers and Taubes's book. Via a link, I came to hyperlipid and then to Brian's original long comment (that Peter highlighted in a subsequent post).
In the debate here, Brian is the one who has shown a civil and scientific attitude. Most others, including Peter, have not.
Somewhat separately, for Brian, in case he's still around: Taubes is not a researcher in the sense of one who does studies; he is a journalist and researcher in the sense of one who *reads* studies. His book, "Good Calories, Bad Calories" cities hundreds of references and would, I think, be of interest to you.
Hi Brec,
The world is full of people like Brian who can copy paste a list of studies showing associations, statin studies and citations of the NCEP. Gary Taube's book is excellent but I doubt it would comfort a lipophobe. I have no perosnal interest in trying to save Brian. He has his own deep belief structure. He assures me that I have got it all wrong. That's his problem. I actaully wish him luck. He's going to need it.
Peter
Peter, I know you don't have the time to re-hash all of the de-bunking you have done on this blog just to try to convince Brian of something which is obviously tabu in his religion. But a lot more people with (fat)fears and doubts read your blog. I would just want to plead with you that you copy+paste the top 3-4 blog post links to what you have already written on the subject, that might be a good introduction into the other side of the lipid hypothesis.
I see that Emily (Evolutionary Psychiatry) has a similar idea, with some core posts linked to from the top of her blog. No idea which ones I could use and my ideas do change slowly over the years.....
Peter
Post a Comment