Thursday, June 25, 2009

High Carbohydrate Foods Can Cause Heart Attacks!


It isn't often that you see headlines like that in the mainstream.....but I saw it today:

In a landmark study, new research from Tel Aviv University now shows exactly how these high carb foods increase the risk for heart problems.
Enormous peaks indicating arterial stress were found in the high glycemic index groups: the cornflakes and sugar group. "We knew high glycemic foods were bad for the heart. Now we have a mechanism that shows how," says Dr. Shechter. "Foods like cornflakes, white bread, french fries, and sweetened soda all put undue stress on our arteries. We've explained for the first time how high glycemic carbs can affect the progression of heart disease." During the consumption of foods high in sugar, there appears to be a temporary and sudden dysfunction in the endothelial walls of the arteries.
Endothelial health can be traced back to almost every disorder and disease in the body. It is "the riskiest of the risk factors," says Dr. Shechter, who practices at the Chaim Sheba Medical Center — Tel Hashomer Hospital. There he offers a treatment that can show patients — in real time — if they have a high risk for heart attacks. "Medical tourists" from America regularly visit to take the heart test.
The take-away message? Dr. Shechter says to stick to foods like oatmeal, fruits and vegetables, legumes and nuts, which have a low glycemic index. Exercising every day for at least 30 minutes, he adds, is an extra heart-smart action to take.

Of course another take home message might be to meat, eggs, cheese, cream etc....no GI worries there. Or just just listen to Kurt....or Mark....or Richard.....or Peter.....or Stephan....

43 comments:

Asclepius said...

What made me laugh about that article was the link to the side headlined "'Bad Carbs' Not The Enemy, Professor Finds" (http://www.sciencedaily.com/releases/2007/09/070928160756.htm), which went on to say "Eating sandwiches with white bread, or an occasional doughnut, isn't going to kill you, or necessarily even lead to obesity".

With such conflicting advice how the hell are laymen/women going to make informed decisions about nutrition?

I personally think that the low fat complex carb ideologists are dying a death from a thousand cuts!

:E said...

Asclepius,

What about the first sentence of the Science Daily article:

"Doctors have known for decades that foods like white bread and corn flakes aren't good for cardiac health."

Yet they keep right on preaching the same low-fat dogma they have since Ancel Keys. Egads.

Nice find, Chris.

Anonymous said...

Chris Wrote:
Of course another take home message might be to meat, eggs, cheese, cream etc....no GI worries there.

Reply:
No so fast Chris. There recent well done experimental data that shows saturated fats degrade endothelial healh.

In this study (1) 40 subjects were fed iso-calorie calorie diets for 3weeks weeks that were either low fat or atleast 25 grams of Pufa, Mufa, or Sat. All subjects were crossed over to each of the diets after an month wash out period. Only the saturated fat diet resulted in a degradation of endothelial function.

A simular protocal was applied in this study (2) comparing Atkins, South Beech and Ornish (high carb, low fat). Once again only the Atkin diet resulted in degraded endothelial health.

Finally this study (3) showed that an adlib (eat as much as you want) high carb, low fat high fiber diet resulted in much lower inflamatory indicators without much weight. Insulin levels were decreased and insulin sensitivity was in increased.

Unlike other studies often sited to support high saturated fat diets, these studies controlled for calories and weight loss.

Regards
Randy

1.
Flow-Mediated Dilatation Is Impaired by a High–Saturated Fat Diet but Not by a High-Carbohydrate Diet

2.
Comparative effects of three popular diets on lipids, endothelial function, and C-reactive protein during weight maintenance.
Miller M, Beach V, Sorkin JD, Mangano C, Dobmeier C, Novacic D, Rhyne J, Vogel RA.
Divisionof Cardiology, University of Maryland Hospital, Room S3B06, 22 S. Greene St, Baltimore, MD 21201, USA. of variance).


3.
Effect of a short-term diet and exercise intervention on oxidative stress, inflammation, MMP-9, and monocyte chemotactic activity in men with metabolic syndrome factors

Asclepius said...

randy - another one to look at: http://www.proteinpower.com/drmike/cardiovascular-disease/fast-food-and-endothelial-dysfunction/

Asclepius said...

And this one: http://www.proteinpower.com/drmike/lipid-hypothesis/does-the-atkins-diet-damage-blood-vessels/

Chris said...

Ascelpius - thanks for stepping in with those links.

I've been up north for a couple of days - see above - so am just catching up

Anonymous said...

Asclepius Wrote:
randy - another one to look at: http://www.proteinpower.com/drmike/cardiovascular-disease/fast-food-and-endothelial-dysfunction/

Reply:
This study compared the effect FMD after only ONE meal. Additionallly as Eades points out the meals differed in other ways.

The study(1) I provided tested subjects for 3 weeks worth of meals and the meals were Identical except for different levels of saturated fat. All subjects were crossed over to each diet after 1 washout. Only the saturated fat diet degraded endothelial health.

Please note that these researches have prevously published papers on the benefits of Low Carb diets, but they have shown that saturated fats does harm to blood vessels.

1.
Flow-Mediated Dilatation Is Impaired by a High–Saturated Fat Diet but Not by a High-Carbohydrate Diet

Asclepius said...

Anonymous, a link to the full paper would be useful to make sure I and others look at the specific paper you refer to. I found one from 2005, but am not sure it is the one you mean.

Anyways, Dr John Briffa has blogged about this self same topic recently. He references a paper entitled 'A Systematic Review of the Evidence Supporting a Causal Link Between Dietary Factors and Coronary Heart Disease' from which he states "[the] notion that saturated fat causes cardiovascular diseases like heart disease is not supported by the science". The full post is below.

http://www.drbriffa.com/blog/2009/06/26/high-gi-carbs-again-implicated-in-cardiovascular-disease/

I am not saying you are wrong, it is just that (once again), it seems we can all pick our own expert to substantiate our prejudice no matter how opposing our views! :)

Anonymous said...

Here are links to the 3 papers mentioned in my first post.

http://atvb.ahajournals.org/cgi/reprint/25/6/1274 Full Paper

http://www.ncbi.nlm.nih.gov/pubmed/19328268 (Abstract)

http://jap.physiology.org/cgi/reprint/100/5/1657 (Full Paper)


What makes these studies powerful in my view are:
1. The studies were done under strick control. We knew exactly what these folks ate.

2. They were 3-4 weeks long. Not just one meal as the study the Eades reported.

3. Weight loss was removed as a confounder

Regards
Randy

Anonymous said...

Asclepius Wrote:
I am not saying you are wrong, it is just that (once again), it seems we can all pick our own expert to substantiate our prejudice no matter how opposing our views! :)

Reply:
The topic I'm responding to is endothelial health. Chris suggested that saturated fats would not damage vessel endotheliam health. I just presented data to show it does.

I have not seen any evidence to the contrary. That not to say it doesn't exist or there are problems with studies I presented. I just have not seen any evidence to the contrary and the studies seem very solid.

I like to keep things specific, in this case the subject is satuated fat intake and endothelial health.

Regards
Randy

Chris said...

Charles has a good post on this here

Anonymous said...

Thanks Chris.

Here's a post I just put up on the site you mentioned.

//*************
Hello,

Chris from the "Conditioning Research" blog directed me to your site after I posted some objections to the same assertions you are making. I would like to point out that there is good strong data that contradicts some your clams.

Firstly, I agree that there is solid data that shows the Hi Glycemic foods induce wild swings in blood glucose levels and measures of endothelial health degrade. No disaggrement there.

BUT, there is also strong recent evidence that High Carb Low glycemic, Low fat High Fiber (>50 grams) diets Improve endothelial health while high saturated fat diets do just the opposite (1,2). Also, when an Atkins ketogenic diet is compared to either an Ornish (High (low) Carb, High Fiber diet) or the South Beach diet, endothelial health is impaired (3).

While I do agree that a high carb, hi gi, low fiber diet can often cause weight gain, studies of adlib High Carb, low gi, high fiber diets generally cause weight loss. Let me know if you want the references and I will provide them.

1.
Effect of a short-term diet and exercise intervention on oxidative stress, inflammation, MMP-9, and monocyte chemotactic activity in men with metabolic syndrome factors
http://jap.physiology.org/cgi/reprint/100/5/1657

2.
Flow-Mediated Dilatation Is Impaired by a High–Saturated Fat Diet but Not by a High-Carbohydrate Diet
http://atvb.ahajournals.org/cgi/reprint/25/6/1274

3.
Comparative effects of three popular diets on lipids, endothelial function, and C-reactive protein during weight maintenance.
http://www.ncbi.nlm.nih.gov/pubmed/19328268

//****************************

Chris said...

Randy

I think we have been through similar discussions before. I am not going to stop drawing attention to bits and pieces that confirm my belief that a lower carb, higher fat diet built around real food is healthier than the low fat one you seem to like.

Get used to the idea - if you read the stuff here you will disagree with it. So be it

Erik said...

I took a look at the study that compared high-SFA diets to high-carb diets. The high-SFA diet was not low-carb; rather, they got around 40% of their calories from carbohydrate!

This only tells us that saturated fat, when coupled with high levels of carbohydrate intake, might be harmful. What about a high fat, low carb diet? That wasn't tested.

Anonymous said...

Erik Wrote:

I took a look at the study that compared high-SFA diets to high-carb diets. The high-SFA diet was not low-carb; rather, they got around 40% of their calories from carbohydrate!

This only tells us that saturated fat, when coupled with high levels of carbohydrate intake, might be harmful. What about a high fat, low carb diet? That wasn't tested.

Reply:
Your quite correct.
But this study (1), which I also posted, did look at a low carb (ketogenic) diet and found it degraded endothelial health compared to higher carb diets with less saturated fat.

1.
Comparative effects of three popular diets on lipids, endothelial function, and C-reactive protein during weight maintenance.
http://www.ncbi.nlm.nih.gov/pubmed/19328268

Chris said...

Randy

two studies that show that low carb diets do not have a negative impact on endothelial function:

http://wholehealthsource.blogspot.com/2008/03/bigger-than-yer-britches.html

http://tinyurl.com/2ktdov

Erik said...

I'm not seeing any hard evidence that it was indeed a ketogenic diet. They merely describe it as "Atkins." As we've seen before, that can mean many things to different people.

Do you have the actual data for the study?

I also find it curious that they only looked at "maintenance levels." Why not study subjects as they lost weight?

This study did (http://www.ncbi.nlm.nih.gov/pubmed/16685042?dopt=Abstract) and found that reduced carbohydrate and increased saturated fat intakes improved lipid markers during weight loss (also increased LDL particle size and reduced triglycerides). And that wasn't even true low carb (still 26% calories as carb).

Anonymous said...

Chris Wrote:
I am not going to stop drawing attention to bits and pieces that confirm my belief that a lower carb, higher fat diet built around real food is healthier than the low fat one you seem to like.

Reply:
My objections (based on data) was not against all low carb, higher fat diets, but those that include alot of saturated fat and its effect on endothelial health.

My other point is that high carb, high fiber diets have been shown to have positive effects on endothelial health and metabolic functioning (insulin sensitivity).
Unfortunately the distinction is often not made between the type of carb involved. I base my statements on evidence not philosphy.

Regards
Randy

Anonymous said...

Erik Wrote:
I'm not seeing any hard evidence that it was indeed a ketogenic diet. They merely describe it as "Atkins." As we've seen before, that can mean many things to different people.

Do you have the actual data for the study?

Reply:
The full paper used to be online and does include the details of each diet. I'll try to find it and send it to you via private email. Please give me a day or two.

Eric Wrote:
I also find it curious that they only looked at "maintenance levels." Why not study subjects as they lost weight?

Reply:
This was done on purpose to remove the confounder of weight loss on the results. All sorts of good things happen with weight loss no matter what kind of food is eaten. They intentionally kept calories the same on all diets to remove this confounder. That's what makes this study so powerfull.

Chris said...

Randy... the study I posted?

Anonymous said...

Hi Chris,

Chris Wrote:
Randy... the study I posted?

Reply:
The study you posted included weight loss. These parameters almost always improve with weight loss.

The studies I provided were intentionally designed for no weight loss. That way only the effect of diet could be studied while removing the weight loss confounder.

Regards
Randy

Chris said...

You said My other point is that high carb, high fiber diets have been shown to have positive effects on endothelial health and metabolic functioning (insulin sensitivity).

Of course a high fibre diet can bring its own problems

Chris said...

Randy the other thing about the study you posted was that it seemed to be based on food diaries - notoriously unreliable as a method of working out what people are eating. Plus as Erik said, "Atkins" can mean anything.

Anonymous said...

Chris Wrote:
Randy the other thing about the study you posted was that it seemed to be based on food diaries - notoriously unreliable as a method of working out what people are eating.

Reply:
This is no different that simular comparison studies you have posted here previously. Why doesn't the same critism apply to those studies.

The researchers took pains in instructing the subject to properly record and weight their food intake.(as reported in the paper)

Additionally all subjects followed all diets (sat, poly, mufa, and carb) and only the sat degraded endothelail health. It would be doubt that all the subjects lied about their food intake to make sat fats look worse.

Its important to keep in mind that these researchers have published many positive benefits of low carb diet previously and were expecting the high carb arm of the study to degrade blood vessel health but found and reported that the saturated fat diet caused problems

Regards
Randy

Anonymous said...

Chris Wrote:
Of course a high fibre diet can bring its own problems

Reply:
Chris, the link was an opinion piece. I think its much more productive to provides links to actuall research.

If you got references to research on the dangers of higher fiber diet (50/50 soluable/insoluable)I'll be happy to respond.

BTW, I have no ideological stake in these aruguments. I eat a moderate carb, moderate fat diet, but data is data.

Regards
Randy

Stephan Guyenet said...

Let's not get too carried away by a 3-week study. Although FMD is apparently impaired on that timescale, it doesn't lead to heart attacks or any long-term functional impairment I'm aware of. That has been demonstrated in a number of years-long controlled trials reducing total fat or substituting PUFA for saturated fat.

I question whether FMD on a 3-week timescale (and especially after a single meal) has any relationship to cardiac outcomes at all.

Furthermore, the effect of saturated fat on cholesterol seem to be limited largely to a short timescale as well. In MRFIT, the Helsinki study and the Lyon study, changing fat composition did not affect total cholesterol in the long term.

Overall, the data on glycemic index do not suggest that high-glycemic foods impair insulin sensitivity or cause weight gain in the long term (2 months or longer). That is a myth that has originated due to the overemphasis of short-term results. High GI foods also do not cause increased food intake in the long term.

Stephan Guyenet said...

My point above is that short-term effects don't always persist in the long term. The body adapts.

Anonymous said...

Stephan Wrote:
My point above is that short-term effects don't always persist in the long term. The body adapts.

Reply:
Ofcourse your correct, but the human evidence we currently have shows that FMD is degraded on higher sat intakes at 3 to 4 week intervals. I know of know evidence to show that at longer intervals the trend reverses.

We can only make educated quesses on the evidence that's available.

Regards
Randy

Stephan Guyenet said...

We have evidence from years-long controlled trials in which they replaced saturated fat with PUFA or reduced total fat. Overall, those studies don't support the hypothesis that total or saturated fat causes cardiac events or increases mortality. So either FMD goes back to normal in the long term, or it's irrelevant to cardiac outcomes.

Anonymous said...

Stephan Wrote:
We have evidence from years-long controlled trials in which they replaced saturated fat with PUFA or reduced total fat. Overall, those studies don't support the hypothesis that total or saturated fat causes cardiac events or increases mortality.

Reply:
Well that's a very, very involved contention. You present it as if it's a closed case.

I've found the Finnish Mental Health study very convincing as far as humnan studies go. That together with the animal studies where conditions can be tightly controlled make it doubtfull (for me) to think that saturated fat is not involved in CHD. Ofcourse that not to say it's the only thing involved

Regards
Randy

Stephan Guyenet said...

Randy,

In my opinion, the Finnish mental hospital trial suffers from methodological flaws that are fatal to its use as scientific evidence. The finding has also never been replicated. The only other trial that came close was the VA trial, but that had significantly more heavy smokers in the control group.

Two trials that were published around the same time as the Finnish trial, the Rose et al corn oil trial and the Anti-Coronary club trial, found the opposite effect-- that replacing saturated fat with PUFA vegetable oils dramatically increased CHD death and overall mortality. The anti-coronary club trial wasn't super well controlled either, but given the authors' and participants' exuberance, you would expect any bias to favor the experimental group. The Rose trial was better conducted.

You can't extrapolate from rodent studies or studies in herbivorous/frugivorous/insectivorous monkeys to humans. How an animal responds to a diet depends on what that animal is adapted to. You can't give a dog elevated blood cholesterol or atherosclerosis by feeding it saturated fat or cholesterol. Rabbits don't thrive on steak and dogs don't thrive on grass.

Given the fact that we've been eating large mammals, just like dogs have, since before H. sapiens even evolved, I would expect dogs to be a better model than herbivores, frugivores and insectivores. Neanderthals, our closest non-human relatives, were carnivores.

Anonymous said...

Hi Stepah,
Alot to responde to.

Stehpan Wrote:
In my opinion, the Finnish mental hospital trial suffers from methodological flaws that are fatal to its use as scientific evidence

Reply:
I disagree as does all the peer reviewed evaluation of the study.
Here's a brief preview. Two mental hopitals in Finnland fed subjects identical diets except polyunsaturated fats were exchanged for saturated fats. There were about 300 - 400 folks in each group. After 6 years the group eating the sat fats had 50% more cardica events.

For the next 6 years the diets were reversed and the trend was reversed. Now the folks that had been eating sat fats were eating polyunsaturated and visa/versa. After another 6 years the trends reversed. The folks that had previously been eating the sat fats but started eating polys reduced their cardiac events by 50%

This study is unique in that it was a cross over type. The same folks were exposed to both diets for a long period of time and results only degraded on the saturated fat diet.

The critism you refer is that some folks were allowed to leave and come back to the study. The Authors, Statistians, and peer considered that this did not sckew the results. I've seen the claim made, but i've never seeen the numbers cruched to verify that this destroyed the validity of the study.

This finding has been verified in other studis in this time period.
See:
The Bierenbaum St. Vincent’s Hospital Study, 1967.

The Paul Leren Oslo Study, 1966

The Wadsworth Veterans Administration Hospital Study, 1969Heart Attacks and Stroke

Stephan Wrote:
"and the Anti-Coronary club trial, found the opposite effect"

Reply:
Quite the contrary, this study for a 60% reduction rate in cardiac events after 4 - 5 years with a p<001.

There were problems with the selection of the control group, but a strong positive was found.

Stephan Wrote:
"You can't give a dog elevated blood cholesterol or atherosclerosis by feeding it saturated fat or cholesterol"

Reply:
That because Dogs have very robust bile systems ldl receptors. All you have to do is decrease their ldl receptors effectiveness and their ldl levle increase and they get heart disease just like Humans.
This can be done with most all animals.
See (1), (2) (3)

The ldl recptors in Monkeys (like humans) cna be turned down by just feeding sat fats. When done in the lab this creates Heart disease in primate just like humans

Regards
Randy

1.
Steiner, A., and F. E. Kendall. 1946. Atherosclerosis and arteriosclerosis
in dogs following ingestion of cholesterol and thiouracil.
Arch. Pathol.
42:
433–444.

2.
Clarkson, T. B. 1972. Animal models of atherosclerosis.
Adv. Vet.
Sci. Comp. Med.
16:
151–173.

3.
Arterioscler. Thromb. Vasc. Biol. 1994;14;587-597
MS Wolfe, JK Sawyer, TM Morgan, BC Bullock and LL Rudel
pediatric-aged population of African green monkeys
Dietary polyunsaturated fat decreases coronary artery atherosclerosis in a

Stephan Guyenet said...

Randy,

You said regarding my statement that the Finnish trial was poorly conducted "I disagree as does all the peer reviewed evaluation of the study."

Actually, the Finnish trial has been extensively criticized in the peer-reviewed literature, including by the NHLBI itself in the Lancet (2(7782):835-8. 1972). Dean Ornish can reference that study but no one really takes it (or him) seriously.

The design of the Finnish trial was so poor they would be laughed out of the room by a modern funding agency. They included people in their analysis that were at the institution for as little as a month. People came and went, yet they included them. Furthermore, the biggest increase in mortality was in the group that was eating a normal diet that had previously eaten the PUFA for 6 years. Was the increase in mortality due to the saturated fat, or the PUFA that they had stuffed themselves with for the previous 6 years? We don't know, because of the study design.

Again, the finding was never replicated by properly controlled trials. It was excluded from the Cochrane collaboration meta-analysis (yes, the one that found no effect on mortality or CVD mortality from reducing saturated fat) because it didn't fit modern inclusion criteria.

The Birenbaum et al study you referenced does not support your hypothesis. It found no difference in mortality between a coconut oil or a PUFA diet. In fact, there was a trend toward increased mortality in the PUFA group.

In the Anti-Coronary club trial, there were 27 deaths in the experimental group and 6 in the control group. 9 in the experimental group died of CHD and zero in the control group. I'd hardly call that a success.

Anonymous said...

Hello Stephan,

Stephan Wrote:
Actually, the Finnish trial has been extensively criticized in the peer-reviewed literature, including by the NHLBI itself in the Lancet (2(7782):835-8. 1972). Dean Ornish can reference that study but no one really takes it (or him) seriously.

Reply:
Your way off the mark here Stephan.

Your quote from Lancet (which I can't find) is from 1972, the Finnish study I'm referencing was published in 1979 (1). It couldn't reference the data I'm citing.

Also if the data so "laughable" in 1972 as you claim why would The "International Journal of Epidemiology" publish a paper based on the same data set in 1979?

The paper I've reference specifically deal with possible confounders showing influences would be small in the worse case to alter the conclusions.

Stephan Wrote:
The Birenbaum et al study you referenced does not support your hypothesis. It found no difference in mortality between a coconut oil or a PUFA diet. In fact, there was a trend toward increased mortality in the PUFA group.

Reply:
No it didn't. The results weren't large but it showed signifcantly increased fatal CVD events at ages below 45 and more events total in those on the coconut diet. (2)

Finally, Loren Cordain, (listed on your Nutrition Hall of Fame),has published extensively in peer reviewed pubs presenting good solid arguments supporting: (3)(4)
(5)

a. "Hunter-gatherer populations,
following their indigenous lifestyles, have shown
no evidence of atherosclerosis on the basis of clinical data
and/or autopsy studies.5 These hunter-gatherers had total
cholesterol levels of approximately 100 to 140 mg/dl, corresponding
to LDL cholesterol levels of about 50 to 70"

b. Thes population consumed low saturated fat diets

c. These very low levels of total and ldl cholesterol have been shown to drastically reduce CVD to almost non-existent levels

d. The basic science supporting the above is well established.

Regards
Randy

1. http://ije.oxfordjournals.org/cgi/content/abstract/8/2/99

2.
http://circ.ahajournals.org/cgi/reprint/42/5/943

3.
http://www.thepaleodiet.com/published_research/

4.
http://www.thepaleodiet.com/articles/Am%20J%20Cardiol%202006.pdf

5.
http://www.thepaleodiet.com/articles/JACC%20LDL%20Final.pdf

6.
http://www.thepaleodiet.com/articles/Hunter%20Gatherer%20Mayo%20Letter.pdf

Stephan Guyenet said...

Sorry Randy, you're the one who's mixed up here. The Finnish trial ended in 1971, and the article I referenced from 1972 was in fact a criticism from the U.S. NHLBI. The article is titled "Effect of a cholesterol-lowering diet on mortality from coronary heart disease and other causes. A twelve-year clinical trial in men and women." Hmm, which study could it have been referring to?

The reason the Finnish trial has been cited by certain sectors of the research world is not because it's a good study, it's because it's the ONLY controlled trial with mortality as a primary endpoint that confirms their anti-sat fat stance. Again, the finding was never replicated by properly conducted trials. And the Finnish trial was excluded from the Cochrane collaboration meta-analysis because it doesn't hold up to modern scientific standards.

As for Loren Cordain, he recently sent out a Paleo Diet newsletter update stating that the normal saturated fat range for human hunter-gatherers is approximately 10-15%. His position has shifted to a more logical one relatively recently. Considering mean intake in the U.S. is 11% saturated fat right now, he feels we can increase our intake by almost 50% and still be within the evolutionary norm, i.e. not suffer adverse effects. You would have to receive his e-mail updates to know about this recent development.

As far as the Birenbaum study, 8 died in the coconut oil group and 10 in the PUFA experimental group over 5 years. Mortality is a hard endpoint so it's the most reliable indicator. Don't tell me you're interpreting that result in favor of avoiding saturated fat?

Anonymous said...

Hello Stephan

Stephan Wrote:
Sorry Randy, you're the one who's mixed up here. The Finnish trial ended in 1971, and the article I referenced from 1972 was in fact a criticism from the U.S. NHLBI. The article is titled "Effect of a cholesterol-lowering diet on mortality from coronary heart disease and other causes. A twelve-year clinical trial in men and women." Hmm, which study could it have been referring to?

Reply:
The study your referencing is this one:

Effect of a cholesterol-lowering diet on mortality from coronary heart disease and other causes.
Lancet. 1972 Oct 21;2(7782):835-8.

The study I'm referring to is this one:

http://ije.oxfordjournals.org/cgi/content/abstract/8/2/99International Journal of Epidemiology

The raw data for both studies came from the same data set, but 1979 study, amoung other things, address some of the confounders you mentioned.

Stephan Wrote:
Again, the finding was never replicated by properly conducted trials

Reply:
That's a misleading statement.
The study hasn't been replicated because theres never been a study of this power done before.

This is the only study using a cross-over design where you feed two different diets at six year intervals. So everbody is exposed to indentical diets with sat fat being the only variable (dietary).

There's never been a cross-over on this subject at this level.

On Louis Cordain:
Are you saying the Cordain no longer belives that very low levels of LDL beneficial?

Are you saying that Cordain would not now recommend lowering one sat fat intake to lower LDL?

Or is he saying its OK to increase your sat fat intake if you LDL stays low?

Are you saying he longer thinks LDL levels below 80 are optimal?

I would find it hard to believe that he no longer believe one's must be cautious in sat fat intake with all the published work supporting the lipid theory (increase blood LDL causes heart disease).

1.
http://www.thepaleodiet.com/published_research/

2.
http://www.thepaleodiet.com/articles/Am%20J%20Cardiol%202006.pdf

3.
http://www.thepaleodiet.com/articles/JACC%20LDL%20Final.pdf

3.
http://www.thepaleodiet.com/articles/Hunter%20Gatherer%20Mayo%20Letter.pdf

Stephan Guyenet said...

Randy,

Here are Dr. Cordain's words:

"This data suggests that the normal dietary intake of saturated fatty acids that conditioned our species genome likely fell between 10 to 15% of total energy, and that values lower than 10% or higher than 15% would have been the exception...

Consequently, population-wide recommendations to lower dietary saturated fats below 10% to reduce the risk of CAD have little or no evolutionary foundation in pre-agricultural Homo sapiens.

...there is a lack of knowledge regarding how low saturated fat intake can be without the risk of potentially deleterious health consequences.

So we do not need to restrict ourselves to only tuna and turkey breast, avoiding every last gram of saturated fat."

Dr. Cordain understands that saying saturated fat = high LDL = heart attack is an oversimplification. Atherosclerosis is probably not caused by normal LDL, but by oxidized LDL. The amount of oxLDL depends on the total amount of LDL, the plasma residence time, the antioxidant content of LDL (tocopherols and coQ10), the PUFA content of the LDL and the size of the LDL particle. Saturated fat increases LDL size and decreases LDL PUFA, both factors that decrease oxLDL, so the effect of sat fat on atherosclerosis are not straightforward. Then there's the fact that atherosclerosis isn't generally enough to cause a heart attack by itself, it requires thrombosis and sometimes fibrillation. The only way to know the actual effect of saturated fat on heart attacks is to test it directly in humans, which has been done numerous times.

The Finnish trial is one of 12 that reduced saturated fat and had total mortality as a primary endpoint. I can cite three trials that found the opposite effect on total mortality: Rose et al, the Anti-Coronary Club trial, and the Sydney Diet-Heart trial. The Finnish trial result is obviously not a question of statistical power, since these other trials had enough power to detect the opposite effect. Most of the trials found no significant effect, and some of them were highly powered. So the Finnish trial remains an outlier that is most logically explained by its poor design.

You said "1979 study, among other things, address some of the confounders you mentioned." But the raw data didn't change. You can't make up for a poor study design with a good discussion section. Skeptical epidemiologists don't take the Finnish trial seriously, and it gets excluded from self-respecting meta-analyses.

The overall weight of the controlled trials examining the effect of reducing sat fat on health says clearly that reducing saturated fat is an ineffective intervention for avoiding CHD and avoiding death in general. In fact, if you consider the prospective observational studies, they also overwhelmingly exonerate saturated fat. To claim otherwise is to selectively cite the evidence.

The other thing to keep in mind is that saturated fat only elevates LDL in the short term. Long term studies including MRFIT, WHI dietary modification trial, Helsinki businessmen trial, and Lyon diet-heart trial, show that reducing saturated fat and/or increasing PUFA has little or no effect on LDL level in the long term. That's the same thing you see in dogs, which makes sense since we're both adapted to animal fat. The Framingham trial noted no connection between a) saturated fat, and b) cholesterol or heart attacks. Same with the Tecumseh study. It blows my mind that people are still saying saturated fat increases LDL in the long term. I'll give them points for a healthy imagination.

Continued below...

Stephan Guyenet said...

And don't forget that LDL particle size associates with heart attacks better than total LDL concentration (bigger is better). What increases LDL particle size??? You guessed it, saturated fat. At least in the short term; I'm not aware of any long-term data. So looking at LDL concentration alone is only a small piece of the puzzle, and relying on short-term studies is misleading. That's why we can't just make assumptions: we have to test it directly in humans, which has already been done numerous times.

As a parting point, I'll mention the Tokelauans, who get nearly 50% of their calories from saturated fat (coconut). This is the highest sat fat intake in the world that I'm aware of. By all indications (ECG included), they have a vanishingly low rate of heart attack. Men get a moderate amount of low-intensity exercise and women mostly do low-intensity domestic tasks. If saturated fat were a relevant factor, you'd expect them to be dropping like flies.

I'm going to bow out of this conversation, but I'll read your reply if you choose to post one.

Sincerely,
Stephan

Anonymous said...

Hello Stehpan,

Stephan Wrote:
Quoting Dr. Cordain:

"This data suggests that the normal dietary intake of saturated fatty acids that conditioned our species genome likely fell between 10 to 15% of total energy, and that values lower than 10% or higher than 15% would have been the exception...

Reply:
That's still not a reversal of his view that SFA are arthogenic.

Here's more of Cordain words:

"Furthermore, little or no
objective data support the assertions that dietary SFA is
nonatherogenic under eucaloric conditions or that SFA represents
an effective satiating macronutrient."

He also seems to support the lipid hypothesis and believes there is epidemiological evidence to back it up.

Here's a futher quote:

"We do not recommend consuming a high-SFA diet because
SFA down-regulates the low-density lipoprotein receptor,
14 thereby elevating total plasma cholesterol and lowdensity
lipoprotein cholesterol concentrations even in normal-
weight individuals. Numerous epidemiological studies
have shown that elevated plasma cholesterol concentrations
increase the risk for coronary heart disease.15,16 A necessary
caveat to this statement is that dietary SFAs elicit this effect
only under chronic hypercaloric or eucaloric conditions.17"

Stephan Wrote:
That's the same thing you see in dogs, which makes sense since we're both adapted to animal fat. The Framingham trial noted no connection between a) saturated fat, and b) cholesterol or heart attacks

Reply:
I've posted before (with citations) that dogs respond to saturated fats differently than humans because of differences in their biles systems and LDL receptors. Turn down dog's LDL receptors and you increase their LDL level just like humans.

BTW, the basic science about LDL receptors and consequencely how saturated fats increase these levels was recognized with a Nobel prize in 1984.

Regards
Randy

Anonymous said...

Hello Stephan,

Stephan Wrote:
And don't forget that LDL particle size associates with heart attacks better than total LDL concentration (bigger is better).

Reply:
The very lattest analysis has cast doubt on the notion that small LDL particle size is an independent predicator of CVD whereas LDL particle number irregardless of particle size is.

When multivariate analysis is done on LDL paricle size and confounders are properly dealt with the association with LDL particle size disappears. (1) (2).

Reference 1 is the full paper that provides an overview of the issue.
I highly recommend that anyone interested in the science behind this argurments read this paper.
Its a great example of how good science being used come up with a fair conclusion. Even if you don't agree with the results you can see how real scientist make object efforts to be fair.

Finally,
I enjoy this site and all the interesting stuff Chris continually digs up. Its obviously a labor of love.

I just posted some research that was at odds with some statement made. It wasn't intended to attack the general philosphy. I had not idea it would get so involved.

Kind Regards to anyone that's been following this thread.

Thanks,
Randy

1.
Low-Density Lipoprotein Size and Cardiovascular
Disease: A Reappraisal
FRANK M. SACKS AND HANNIA CAMPOS
Conclusions
The burden of proof for any newly proposed risk factor is
that it must add significantly to risk assessment by existing
measurements, or that it is equivalent but more economical.
LDL subtyping does not meet either of these expectations.
Metabolic studies demonstrate that large and small LDL
subtypes are atherogenic. In as much as any type of LDL is
contained in the plasma total LDL concentration, the standard
clinical measurement of risk, all LDL types should be
viewed as harmful. The best indicator of response to lipid
therapy is a reduction in the plasma concentration of atherogenic
lipoproteins, as conventionally measured by LDL and
triglycerides, but alternatively by non-HDL cholesterol or
apo B (91).
Acknowledgments

http://jcem.endojournals.org/cgi/reprint/88/10/4525.pdf

2.
Systematic review: association of low-density lipoprotein subfractions with cardiovascular outcomes
http://www.ncbi.nlm.nih.gov/pubmed/19349632?dopt=Citation

Sarah said...

http://conditioningresearch.blogspot.com/2009/06/high-carbohydrate-foods-can-cause-heart.html

Sarah said...

At any rate, I liked some of the researcher cartoons on VADLO search engine!

Heart Attack said...

Depression is increasing day by day because of global economy meltdown these days. A lot of work load to the corporate has led them to be unhealthy. The over work load is disrupting their health schedules, and hence their health concerns are creating a problem. Many corporate employees are reporting to have a severe headache, and heart pains due to increased tension and stress being levied on them. Being one of them, I was having a severe pain on heart and head because of bundles of work being thrust on us. Many of the co-employees were experiencing the same, and many have already gone sick because of unscheduled work, high stress environment round the clock, no exercise, improper food. This was giving a distinguishable impact on the companies work efficiency. And at that time, a corporate wellness program served by a medical service provider elite health was initiated by the company. It was quite an appreciable step to hire a medical service to monitor the health of employees. This decision really ameliorated the working efficiency of the company, which was a direct advantage to the company.