Add Boston Globe to the list of heart scan blunders

17. September 2008 William Davis (15)
Yet another piece of mass media misinformation hit the airwaves today. This time it's not from the New York Times or the LA Times, both of which have previously mangled the issues surrounding heart scans. This time it's from the Boston Globe.

In an article titled What is a calcium scan for heart disease, and who should undergo the test?, the report states:

". . . calcium scans may not be a good idea, or prove terribly useful, for most people. For one thing, the scans expose a patient to significant radiation - equivalent to roughly 50 chest X-rays" said Dr. Warren Manning, chief of noninvasive cardiac imaging at Beth Israel Deaconess Medical Center."

As many before him, Dr. Manning is confusing two tests: CT coronary angiography and CT heart scanning. Perhaps we can't blame him: This technology has had its weakest following in the northeast, for reasons not entirely clear to me. (In fact, Track Your Plaque followers have had the greatest struggle obtaining heart scans in that part of the country.) Nonetheless, you'd think he'd have his simple facts straight before talking to the press. Unfortunately, hospital public relations departments will usually just grab whoever they can willing to talk to the press--regardless of their expertise or lack of.


The story goes on to say:

. . ." it's not clear what to do with the results from a calcium scan. If you have diabetes, high cholesterol, high blood pressure, or a family history of heart disease, you already know - or should know - that you are at increased risk of heart problems and should lower these risk factors. So, a calcium scan provides little additional information," Manning said.

"Moreover, even a high score doesn't necessarily mean that the calcified plaque in your arteries is obstructing blood flow, said Dr. Adolph Hutter, a cardiologist at Massachusetts General Hospital."

"The vast majority of people with high calcium tests don't have obstructions and they do fine long-term. So you'd have to test lots and lots of people to prevent one heart attack or sudden death," said Manning.

And if you get a low calcium score, a sign of little or no calcification of plaques, that's not very useful, either, because it could be wrong, or it could be right but lull you into believing you do not have to exercise and watch your diet, cholesterol, and blood pressure levels. "You can still be at risk even if your calcium test is negative," Hutter said.


It is truly shocking how little many (not all, thank goodness) of my colleagues really know about 1) heart scans, 2) coronary disease prevention, and 3) prevention in general. These same "experts" likely advocate high-dose statin drugs and low-fat diets for people at risk. They likely refer patients to the American Heart Association for diet advice and themselves obtain a lot of information from the pharmaceutical industry. The notion of identification, tracking, and purposeful reversal of coronary plaque is entirely foreign to this bunch.

"The vast majority of people with high calcium tests don't have obstructions and they do fine long-term. So you'd have to test lots and lots of people to prevent one heart attack or sudden death." Well, take a look at a graph from a database of 25,000 people undergoing heart scans then observed for several years afterwards:




You can see quite clearly from the curves that heart scan scores very clearly predict your future (if no preventive action is taken). The higher the score, the greater the likelihood of heart attack and death. How much clearer can it get?

The most recent addition to this literature is the PREDICT study which concluded:

Hazard ratios relative to CACS [coronary artery calcium scores] in the range 0-10 Agatston units (AU) were: CACS 11-100 AU, 5.4 (P = 0.02); 101-400 AU 10.5 (P = 0.001); 401-1000 AU, 11.9 (P = 0.001), and >1000 AU, 19.8 (P < 0.001).

In other words, a heart scan score of >1000 is associated with a 20-fold increased risk of cardiovascular events (without preventive efforts). That kind of predictive power and quantitative confidence simply cannot be squeezed out of blood pressure and cholesterol values.

How about the 2008 University of California-Irvine study from the New England Journal of Medicine (do the northeast docs even pay attention to something that is published in their own neighborhood?) that reported:

There were 162 coronary events, of which 89 were major events (myocardial infarction or death from coronary heart disease). In comparison with participants with no coronary calcium, the adjusted risk of a coronary event was increased by a factor of 7.73 among participants with coronary calcium scores between 101 and 300 and by a factor of 9.67 among participants with scores above 300 (P<0.001 for both comparisons). Among the four racial and ethnic groups, a doubling of the calcium score increased the risk of a major coronary event by 15 to 35% and the risk of any coronary event by 18 to 39%.

How about the Prospective Army Coronary Calcium (PACC) project (men average age 43 years):

"In these men, coronary calcium was associated with an 11.8-fold increased risk for incident coronary heart disease (CHD) (p = 0.002) in a Cox model controlling for the Framingham risk score. Among those with coronary artery calcification, the risk of coronary events increased incrementally across tertiles of coronary calcium severity (hazard ratio 4.3 per tertile)."

Calcium score provided additional information even after factoring in the Framingham risk score.

That's just a sample of the studies. There are a number more.

Add to these conversations the fact that, unlike reducing blood pressure or LDL cholesterol, the heart scan score is a quantification of the disease itself. It can also be tracked over time to gauge the success or failure of prevention efforts. To believe that blood pressure reduction or LDL cholesterol reduction is sufficient to eliminate risk is something only a fool would believe.



Contary to the above statements, the data are clear:

--The higher the heart scan score, the greater the risk. This has been demonstrated beyond any shadow of a doubt in at least a dozen published studies. In fact, heart scan scores outshine lipid/cholesterol values several-fold.

--A person with a zero score has a nearly zero risk for cardiovascular events over a 5-year timeline.

--Heart scans are the only quantitative test available of coronary atherosclerotic plaque. This means that they can be repeated to gauge progression or regression. Cholesterol does not do that. Stress tests do not do that.

--Heart scans are not the same as CT coronary angiography.

--The lack of "need" for a procedure does not equate to the absence of disease.

The power of heart scans is that they can uncover evidence for coronary atherosclerotic plaque 10 years before a cardiac disaster strikes. Witness Tim Russert's heart scan score of 210 in 1998 at age 48. 10 years later, you know what happened.

Beware the camipaign of misinformation and ignorance that continues that is hell-bent on maintaining the procedural status quo or locking us into a "drugs for all" mentality.

What's worse than sugar?

16. September 2008 William Davis (11)
There are a number of ways to view the blood sugar-raising or insulin-provoking effect of foods.

One way is glycemic index (GI), simply a measure of how high blood sugar is raised by a standard quantity of a food compared to table sugar. Another is glycemic load (GL), a combination (multiplied) of glycemic index and carbohydrate content per serving.

Table sugar has a GI of 65, a GL of 65.

Obviously, table sugar is not good for you. The content of white table sugar in the American diet has exploded over the last 100 years, totaling over 150 lb per year for the average person. (Humans are not meant to consume any.)

What is the GI of Rice Krispies cereal, organic or not? GI = 82-- higher than table sugar. GL is 72, also higher than table sugar.

How about Corn Flakes? GI 81, GL 70--also both higher than sugar.

How about those rice cakes that many dieters will use to quell hunger? GI 78, GL 64.

How about Shredded Wheat cereal? GI 75, GL 62.

All of the above foods with GI's and GL's that match or exceed that of table sugar are made of wheat and cornstarch. Some, like Shredded Wheat cereal and rice cakes, don't even have any added sugar.

Stay clear of these foods if you have low HDL, high triglycerides, high blood sugar, or small LDL. Or, for that matter, if you are human.

Keep the eloquent words of New York University nutritionist, Marion Nestle, author of the book, Food Politics, in mind:

“Food companies—just like companies that sell cigarettes, pharmaceuticals, or any other commodity—routinely place the needs of stock holders over considerations of public health. Food companies will make and market any product that sells, regardless of its nutritional value or its effect on health. In this regard, food companies hardly differ from cigarette companies. They lobby Congress to eliminate regulations perceived as unfavorable; they press federal regulatory agencies not to enforce such regulations; and when they don’t like regulatory decisions, they file lawsuits. Like cigarette companies, food companies co-opt food and nutrition experts by supporting professional organizations and research, and they expand sales by marketing directly to children, members of minority groups, and people in develop countries—whether or not the products are likely to improve people’s diets.” ??

Are sterols the new trans fat?

13. September 2008 William Davis (16)
By now, I'm sure you're well-acquainted with the hydrogenated, trans fat issue.

Hydrogenation of polyunsaturated oils was a popular practice (and still is) since the 1960s, as food manufacturers sought a substitute for saturated fat. Bubbling high-pressure hydrogen through oils like cottonseed, soybean, and corn generates trans fatty acids. These man-made fatty acids, while safe in initial safety testing, proved to be among the biggest nutritional mistakes of the 20th century.

Trans fatty acids have been associated with increased LDL cholesterol, reduction in HDL, oxidative reactions, abnormal rigidity when incorporated into cell membranes, and cancer. Trans fats still dominate many processed foods like chips, cookies, non-dairy creamers, food mixes, and thousands of others. They're also found prominently in fast foods.

Fast forward to today, and most Americans have become aware of the dangers of trans fats and many try to avoid them.

But I worry there is yet another substance that has worked its way into the American processed food cornucopia that has some potential for repeating the trans fat debacle: sterol esters.


Sterols are naturally-occurring oils found in vegetables, nuts, and numerous other foods in small quantities. Most of us take in 200-400 milligrams per day just by eating plant-sourced foods.

Curiously, the chemical structure of sterols are very similar to human cholesterol (differing at one carbon atom). Sterols, by not fully understood means, block the intestinal absorption of cholesterol. Thus, sterol esters, as well as the similar stanol esters, have been used to reduce blood levels of total and LDL cholesterol.

So far, so good.

The initial commercial products, released in the late 1990s, were Take Control (sterol) and Benecol (stanol), both of which were marketed to reduce cholesterol when 2-3 tbsp are used daily, providing 3400 – 5100 mg of sterol or stanol esters, about 10- to 20-fold more than we normally obtain from foods. Several clinical trials have conclusively confirmed that these products reduce cholesterol levels.

They do indeed perform as advertised. Add either product to your daily diet and LDL cholesterol is reduced by about 10-15%. In fact, in the original Track Your Plaque book, these products were advocated as a supplemental means of reducing LDL when other methods fell short.

In 2008, there are now hundreds of products that have additional quantities of sterol esters in them, such as orange juice, mayonnaise, yogurt, breakfast cereals, even nutritional supplements. Most of these products proudly bear claims like "heart healthy." Stanol esters have not enjoyed the same widespread application. (I believe there may be patent issues or other considerations. However, it's the sterols that are the principal topic here, not stanols.)

Now, here's where it gets a bit tricky. There is a rare (1 per million) disease called sitosterolemia, a genetic disorder that permits the afflicted to absorb more than the usual quantity of sterols from the intestine. While you and I obtain some amount of sterols from plant-based foods, absorption is poor, and we absorb <10% of sterols ingested. However, people with sitosterolemia absorb sterols far more efficiently, resulting in high blood levels of sterols that result in coronary disease and aortic valve disease, with heart attacks occurring as young as late teens or 20s. Treatment to block sterol absorption are used to treat these people.

There are also a larger number, though still uncommon (1/500) of people who have only one of the two genes that young people with sitosterolemia have. These people may have an intermediate capacity for sterol absorption.

Okay, so what does this have to do with you? Well, if you and I now take in 10-20 times greater amounts of sterol esters, do our blood levels of sterols increase?

Several studies now suggest that, yes, sterol blood levels increase with sterol ingestion. One study from Finland, the STRIP Study, showed that children who had double usual sterol intake increased blood levels by around 50%.

Similarly, a Johns Hopkins study in adults with only one of the genes ("heterozygotes") for sitosterolemia increased sterol blood levels by between 54-116% by ingesting 2200 mg of sterols added per day, despite reduction of LDL cholesterol levels.

Even people with neither gene for sitosterol hyperabsorption can increase their blood levels of sterols. But the crucial question: Do the blood levels of sterols that occur in unaffected people or in heterozygotes increase the risk of coronary heart disease? The answer is not known.

Despite the several clinical trials performed with sterol esters, all of them have examined LDL and total cholesterol reduction as endpoints, not cardiovascular events. It is conceivable that, while sterol esters reduce cholesterol, risk for heart disease is increased due to higher blood levels of sterols.

The question is not settled. For now, it is just a suspicion. But that's enough for me to steer clear of processed foods supplemented with these uncertain sterol esters. My previous recommendations for sterol ester products will be removed with the next edition of Track Your Plaque. Until we have solid evidence that there are no adverse cardiovascular effects of sterol esters, in my view they should not be part of anyone's heart-disease prevention program.

(The same argument does not seem to apply to stanol esters, such as that contained in butter-substitute Benecol, since stanol esters are not absorbed at all and remain confined to the intestine.)

The Diabetes Gold Rush

11. September 2008 William Davis (9)
Lou came into the office. Clearly, his program had gone sour.

Lou had initially obtained wonderful control over his heart scan score of 1114, having reversed modestly in his first three years of effort through correction of his multiple causes (including low HDL, severe small LDL, Lp(a), and a diabetic tendency).

But Lou now came into the office red-faced and sporting a big bulging abdomen. Blood sugar? Now in the overtly diabetic range. Lou said that his primary care doctor had suggested that he start on three new medications (glucophage, injectable Byetta, and Actos) to control his blood sugar. His doctor also told him to increase his intake of fibers by eating more "healthy" breakfast cereals like Cheerios.

Lou had apparently done just that (added "healthy" fiber-rich foods) even before his doctor had suggested it. (Lou failed to remember the several conversations we'd had about healthy eating.) Unfortunately, Lou also failed to connect his increased intake of "healthy fiber-rich foods" and his growing abdominal girth (his "wheat belly").

Here's the dirty little secret: Much of the world wants you to be diabetic. It is the health gold rush of this century. "Go West, young man!"




To find out what I mean, you need only ask: Who profits when people become diabetic? That's easy:

The pharmaceutical industry--Diabetes is a booming growth industry, a source of tens of billions of dollars of revenue, poised for enormous growth as the population ages and gets fatter. It is common for a newly-diagnosed diabetic to be given new prescriptions for two or three drugs with a monthly cost of $300. Of course, the chronic nature of the disease make this far more profitable than, say, a two week course of antibiotics. Presently, 70 new drugs are under development.

Diabetes drug maker Novo Nordisk reported a 25% increase in revenues in 2007 from diabetic agents in the North American market, along with near $2 billion increase in profit for the year. Merck's recently-released DPP-4 inhibitor, Januvia, has already sold $668 million in 2007 and is growing rapidly.

The medical device and supply industry. Take a look at the Medtronic quarterly earnings report, detailing the breakdown of their record-setting quarterly revenue of $3.7 billion:

Diabetes revenue of $269 million grew 12 percent driven by sales
of consumables, the accessories required by insulin pump users, and
continuous glucose monitoring products. Revenue from international
sales grew 31 percent over the same quarter last year.

That's what I call a growth industry.

The processed food industry. The food industry is as big or bigger than the drug industry. ADM, Kraft, General Mills all have annual revenues in the $12-50 billion range. There are plenty of others.

When we're told, for instance, that Cheerios reduces cholesterol, we're not told that it skyrockets blood sugar or triggers small LDL. When we're sold whole wheat crackers, Cocoa Puffs (which the American Heart Asscociation says is heart-healthy), or granola bars, hunger is stimulated, impulse to eat more grows, blood sugar escalates, we get fat, we get diabetic. It's a simple formula.

So be aware that there is little incentive among corporate giants in the food, medical device, or drug industries to encourage behaviors that decrease the incidence of diabetes. In fact, there is enormous financial incentive to make sure that diabetes continues to grow at the startling rate it has over the last decade.

To be sure, the drug and medical device industry will also develop better tools to deal with diabetes and its complications. But the very best way to deal with diabetes is to not develop it in the first place.

What else is there?

7. September 2008 William Davis (11)
This question comes up frequently:

Aren't there any alternatives to heart scans performed on a CT or EBT device?

Yes, there are.

First of all, heart scans are performed best on an electron-beam CT device (EBT) or a 64-slice multi-detector CT (MDCT) device. (While they are also obtainable through less-than-64 slice CT devices (e.g., 16 slices and less), I would advise against it because of the excessive radiation exposure and poor accuracy.) CT heart scans are not to be confused with now more popular CT coronary angiograms, which are performed on the same devices but require intravenous x-ray dye and many times more radiation.(See CT scans and radiation exposure and Heart scan frustration.) Heart scans currently form the basis for the Track Your Plaque program, a program of tracking plaque in the hopes of stopping or reversing the otherwise inevitable 30% per year increase.

Let's confine our discussion to people without symptoms, meaning people like you and me sitting at home, not in an emergency room having chest pain or other similar acute symptomatic presentation.

Among the other ways to uncover hidden coronary plaque:

--Heart catheterization--to yield a coronary angiogram. Yes, this does tell us whether coronary plaque is present. However, it is invasive, expensive, and crude. (I've performed 5000 over my career; they are crude, though useful, tools in acute settings like unstable symptoms or heart attack, a different situation.) Coronary angiography is also non-quantitative. While they provide a value like "40% blockage mid-way in right coronary" or "90% blockage in left anterior descending" they do not provide a trackable lengthwise index of total plaque volume. Identifying severe blockages in people with symptoms leads to stents, bypass surgery and the like, but it is not practical nor of long-term usefulness in apparently, healthy people without symptoms.

--Carotid ultrasound--Here's is where a lot of confusion comes from. Standard carotid ultrasound (U/S) performed in virtually every hospital and many clinics will yield crude qualitative results, e.g., "16-49% stenosis (blockage) in right internal carotid artery". The crude value range is because much of carotid U/S is based on flow velocities, not just direct visualization of the plaque itself ("2-D imaging). However, if carotid stenosis of any degree is identified, the likelihood of silent coronary plaque is much greater.

Limitations: The qualitative, non-quantitative nature of carotid U/S make it difficult to follow long-term in a precise way. Also, this is carotid plaque, not coronary plaque. It makes it very difficult to follow carotid plaque as an indirect means of tracking coronary plaque. The two arterial territories, carotid and coronary, do not track together: there are divergences in many people, with carotid plaque absent in some people with advanced coronary plaque, carotid plaque more susceptible to different risk factors than coronary. So carotid U/S is helpful for its own purposes, but not terribly helpful for coronary tracking.

How about carotid intimal-medial thickness (CIMT) obtained also with carotid U/S? CIMT is a useful index of bodywide atherosclerosis. CIMT is simply a measure not of plaque (and is measured in regions of the carotid artery away from plaque), but of the thickness of the lining of the carotid arteries. Everybody has a measurable CIMT, but it thickens as atherosclerosis grows. CIMT is a radiation-free test that takes several minutes.

Limitations: Hardly anybody does it outside of research protocols. I know of no hospital or clinic in my area that performs CIMT, though it is slowly being adopted in some centers. It is also difficult to rely on repeated tests, because there is substantial variation when one technologist or another performs it. CIMT is also a flawed index of coronary plaque. When CIMT is compared to heart scan scores, CT coronary angiography, or conventional coronary angiography, CIMT correlates about 60-70% with the degree of coronary atherosclerosis.

CIMT is therefore a useful test for research, but a distant 2nd choice--if you can obtain it.

--Ankle-brachial index (ABI)--ABI is a crude measure, simply a comparison of the blood pressure (obtained with a blood pressure cuff) in the legs divided by blood pressure in the arms. The ratio is called ABI. Any ABI <1.0, meaning less pressure in the legs compared to the arms, is indirectly indicative of advanced coronary disease. ABI is, in fact, a very powerful predictor of cardiovascular events. If ABI is <1.0, your future risk for heart attack is very high, even in the absence of symptoms.

Limitations: The vast majority of people with heart disease, even those having undergone stents or bypass surgery, have normal ABI's. Virtually all people with high heart scan scores have normal ABI's. In other words, ABI is a measure of very advanced atherosclerosis only.

--Stress tests--I lump all stress tests together in their various forms, e.g., stress thallium, stress Cardiolite, stress Myoview, persantine/adenosine Cardiolite, dobutamine echocardiography, etc. Stress tests are tests of coronary blood flow, not of plaque. Stress tests are useful in people with symptoms, like chest pain or breathlessness, since stress tests are provocative tests that can help determine whether reduced coronary blood flow is the cause behind a symptom, or whether hiatal hernia, esophagitis, gallstones, pleurisy, musculoskeletal causes, or some other process is behind symptoms.

Limitations: Stress test are virtually useless in people without symptoms. This is why people like Tim Russert and Bill Clinton, both without symptoms, underwent several (Russert 3, Clinton 5) nuclear stress tests---all normal. You know what happened to them. Stress tests do not reliably uncover hidden coronary plaque in people without symptoms. Stress tests are, like coronary angiograms, non-quantitative. They are normal or abnormal.


Outside of experimental settings, that's it.

You can probably see why I advocate CT heart scans for tracking plaque. I do not advocate heart scans because I sell them (I don't), because scan centers pay me to say these things (they don't, and in fact my relationship with my usual heart scan centers has become deeply contentious, though I still endorse the technology). I say that heart scans are superior because they are, in 2008, the only way to 1) identify and 2) track coronary plaque that is easy, safe, low-radiation, and reasonably priced (<$200 in Milwaukee at 5 centers).

The need for a technology that allows tracking of plaque, not just initial identification, is also an important distinction. People who've had some measure of atherosclerosis all catch on to this eventually. "Can I reverse it?" is an inevitable question once the disease is identified in some way. So a tool for tracking over time to gauge the success or failure of a program of prevention can be assessed.

Perhaps in 10 years, another technology will emerge as the preferred means to do the same, but better. If that proves true, we will convert to that technology. But today heart scans performed on CT heart scans are the only rational way to both detect, then track, coronary atherosclerotic plaque.

Let's gamble with your health

5. September 2008 William Davis (4)
Let's play a game.

I'm going to list some lipid patterns and you tell me whether or not the person with these values has heart disease.

Patient 1

Total cholesterol 150 mg/dl
LDL cholesterol 75 mg/dl
HDL 50 mg/dl
Triglycerides 125 mg/dl


Patient 2

Total cholesterol 300 mg/dl
LDL cholesterol 200 mg/dl
HDL cholesterol 35 mg/dl
Triglycerides 325


Patient 3

Total cholesterol 300 mg/dl
LDL cholesterol 100 mg/dl
HDL cholesterol 25 mg/dl
Triglycerides 875 mg/dl



Let's say that any one of these profiles is yours. Should you be getting your affairs in order, preparing for your cardiac catastrophe? Should you demand a stress test from your doctor, hoping that it will shed some light on your dilemma? Should you go ahead and go to the all-you-can-eat rib restaurant, content that you will be attending your granddaughger's wedding in 2020 in full health?

If you can tell, you're a lot better at this than I am.

I provide consultation to other physicians and patients on complex hyperlipidemias in my area. In other words, if someone has a difficulty to manage lipid disorder, the doctor sends the patient to me.

Managing these wildly variable values is the easy part. Deciding whether or not heart disease is concealed within the patient . . . well, that's the hard part.

Let's take it a step further: Suppose all three profiles also have 50% of all LDL particles as the abnormal small particles. And they all have a lipoprotein(a) level of 50 mg/dl, an abnormally high level.

How about now: Can you tell whether any or all of these people have hidden heart disease?

What if they are 20 years old? Does that make a difference?

What if they are all females over 65 years--how about now?

If the only tool you have to divine the presence of hidden heart disease is a lipid panel, or even a lipoprotein panel, then the best you can manage is to hazard a guess based on statistical probability. You also assume that this "snapshot" represents the sorts of values someone has had for their entire lives. You cannot factor in the fact that the first person gained 60 lbs in the last three years since completing menopause. You can't factor in that patient 2 smoked two packs of cigarettes a day for 25 years, but quit 10 years ago.

It's also foolhardy to believe that every known cause of heart disease is currently identifiable and revealed by modern-day blood testing.

A heart scan is simply a means to quantify the sum-total of risk factors--causes--that have exerted an effect up until the moment of your scan. It will reveal the quantity of coronary atherosclerotic plaque present, regardless of whether you stopped smoking 20 years ago or lost 30 lbs last year.

For these reasons, nothing can replace the value of quantifying plaque: not cholesterol, not the Framingham risk calculation, not measures of small LDL or lipoprotein(a), not the presence or absence of symptoms. In 2008, the method of choice for measuring plaque remains a CT heart scan. Perhaps in 10 years it will be some other method.

As always, let me remind Heart Scan Blog viewers that I make this point NOT to sell heart scans, which I have no reason whatsoever to do. I say this because we require a tool to track this potentially fatal disease. We require a yardstick for tracking progression or regression. The only tool that suits these purposes in 2008 is a CT heart scan.

Who knows what

4. September 2008 William Davis (4)
You know that cynical old saying:


It’s not what you know, it’s who you know.

In other words, knowing the right person provides you strategic advantage in business, social advancement, etc.

In health, it was often true. Knowing who the better doctors were, for instance, in your city might provide you with access to better care.

Enter the Information Age. You now have access to medical information equal to that of your doctor. You now have access to patient discussions about doctors, their practices, their performance records. There is now a depth and breadth of information on health that was never available before.

I’d therefore turn the old saying into the new Health 2.0 version:


It’s not who you know, it’s what you know.


In health, information now reigns supreme, not knowing somebody else who has the right connections.

Positive: Everybody now theoretically has access to an equal amount of information, since you can access information on any topic just as easily as I can.

Negative: It puts more of the burden on you. If you screw up in health, perhaps you didn’t try to get the best information hard enough.

I love this new development, this emergence of empowerment in health. I call it self-directed health, the individual capacity to exert enormous influence over the quality of your healthcare.

This is obviously a work in progress. All the answers and tools for self-directed care, self-empowerment are not yet available, some haven’t even yet been imagined.

But they are coming.

“Too many false positives”

4. September 2008 William Davis (6)
“Do you really think I need a heart scan?” asked Terry.

“My doctor said that heart scans show too many false positives. He says that many people end up getting unnecessary heart catheterizations because of them.”

At age 56, Terry was becoming increasingly frightened. His father had suffered his first heart attack at age 53, Terry’s paternal uncle had a heart attack at age 56, his paternal grandfather a heart attack at age 50.

Is this true? Do heart scans yield too many false positives, meaning abnormal results when there really is no abnormality?

No, it is not. What Terry’s doctor is referring to is the fact that, in the decades-long process that leads to heart attack, heart scans have the ability to detect early phases of developing coronary atherosclerotic plaque.

Let’s take Terry’s case, for example. Given his family history, it is quite likely that he does indeed have coronary atherosclerotic plaque. Will it be detectable by performing a stress test? Probably not. In fact, Terry jogs and feels well while doing so. While a stress test abnormality that fails to reach conscious perception is possible, it’s fairly unlikely given his exercise routine.

Will Terry’s coronary atherosclerotic plaque be detectable by heart catheterization? Very likely. But why perform an invasive hospital procedure just as a screening test? Should a woman wishing to undergo a screening test for breast cancer undergo breast removal? Of course not.

Is waiting for symptoms a rational way to approach diagnosis of heart disease? Well, when symptoms appear, it means that coronary blood flow is reduced. Stents and bypass surgery may be indicated. The risk of heart attack and death skyrocket. Sudden death becomes a real possibility.

In the 30 or so years required to establish sufficient coronary plaque to permit the appearance of symptoms or the development of an abnormality detectable by stress testing, there were many years when the disease was early--too early to generate symptoms, too early to be detectable by stress testing.

That’s when heart scans uncover evidence for silent coronary atherosclerotic plaque.

Should we call this a “false positive” just because it doesn’t also correlate with “need” for a catheterization, stent, bypass operation or result in heart attack within the next few weeks?

The detection of early plaque is just that: early disease detection.

Imagine, for instance, that the breast cancer that will grow into a palpable nodule or mass detectable by mammogram is detectable by a special breast scan 15 years before it becomes a full-blown tumor, metastasizing to other organs. What if effective means to halt that earliest evidence of cancer could put a stop to this devastating disease decades ahead of danger? Is this a “false positive” too?

In my view, this is the knuckleheaded thinking of the conventional practitioner: “Don’t bother me until you’re really sick.” Prevention is a practice that has become fashionable only because of the push of the drug industry. Nutrition is an afterthought, a message conceived through consensus of “experts” with suspect motivations and allegiances.

So, no, heart scans do not uncover “false positives.” They uncover early disease--true positives--years before it is detectable by standard tests or by the appearance of catastrophe. But that is the whole point: Early detection means getting a head start on prevention.

Do heart scans lead to unnecessary heart catheterizations? Yes, sadly they do. But not because heart scans are false positive. It happens because of unscrupulous or ignorant cardiologists who use the information wrongly. In my view, heart scans should NEVER lead directly to heart catheterization in an asymptomatic patient. Heart scans, as helpful as they are, do not modify the standard reasons for performing heart procedures.

If a car mechanic is dishonest and fixes a carburetor that didn't need fixing, should we condemn all car mechanics? No, of course not. We only need to develop the means to weed out the bad apples. The same applies to heart scans.

Triglycerides divided by five

3. September 2008 William Davis (8)
Here's a bit of lipid tedium that might nonetheless help you one day decipher the meaning of shifts in your cholesterol panel.

Recall from prior discussions that conventional LDL cholesterol is a calculated value. Contrary to popular opinion, LDL is usually not measured, but calculated from the Friedewald equation:

LDL cholesterol = Total cholesterol - HDL cholesterol - triglycerides/5

For the sake of simplicity, let's call total cholesterol TC; HDL cholesterol HDL, and triglycerides TG.

We've also talked in past how a low HDL makes calculated LDL inaccurate, sometimes wildly so. (See Low HDL makes Dr. Friedewald a liar.)

Here's yet another source of inaccuracy of the Friedewald-calculated LDL: any increase in triglycerides.

Let's say, for instance, that starting lipid panel shows:

TC 170 mg/dl
LDL 100 mg/dl
HDL 50 mg/dl
TG 100 mg/dl


You're advised to follow a standard low-fat, whole grain-rich diet advocated by "official" agencies (the diet I bash as knuckleheaded). Another panel a few months later shows:

TC 230 mg/dl
LDL 140 mg/dl
HDL 50 mg/dl
TG 200 mg/dl


(Obviously, I've oversimplified the response for the sake of argument. HDL would likely go down, LDL would change more depending on body weight, small LDL tendencies, and other factors. You'd also likely get fat.)

Now your doctor declares that your LDL has gone up and you "need" a statin agent.

Nonsense, absolute nonsense.

What has really happened is that the increased dietary intake of wheat and other "healthy whole-grain foods" has caused triglycerides to skyrocket. LDL increases, in turn, by a factor of TG/5, or 40 mg/dl. Thus, LDL has been inflated by the triglyceride-raising effect of whole grains.

This is yet another reason why the standard lipid panel, full of hazards and landmines, needs to be abandoned. But calculated LDL in particular is an exercise in frustration.

Though the example used is hypothetical, I've witnessed this effect thousands of times. I've also seen many people placed on statin drugs unnecessarily, due to the appearance of a high LDL cholesterol that really represented increased TG/5, usually induced by an excessive carbohydrate intake, including those commonly misrepresented as healthy such as whole grains.

Who reads The Heart Scan Blog?

2. September 2008 William Davis (0)
In the Heart Scan Blog, I am often guilty of speaking out loud of my varied thoughts on this crazy thing that we've created called the cardiovascular healthcare machine. But I discuss it in the context of asking "How could this be done better--better outcomes, more patient-friendly, more accessible . . . more do-it-yourself?

The last part is the part that throws most people. Do-it-yourself? My colleagues would claim I'm nuts, suggesting that coronary heart disease is something manageable by yourself. In the conventional pathway, after all, coronary disease is that unpredictable, poorly detected by standard tests, condition that then leads to heart catheterization, stents, bypass , and the like.

Several factors distinguish the readers of The Heart Scan Blog that surprised me:

--Nearly 60% are women
--There are a disproportionate number of Asian people. (Can someone explain this to me?)
--A great number have graduate degrees

I believe this tells me that The Heart Scan Blog appeals to a somewhat more sophisticated audience. This, to some degree, warms my heart, since it means that I've captured the attention of some people who may be more discriminating and thoughtful in their Internet surfing.

However, I also lament the fact that these conversations are not achieving the mainstream. After all,
Butter: Just because it's low-carb doesn't mean it's good

Butter: Just because it's low-carb doesn't mean it's good

21. October 2010 William Davis (59)
The diet I advocate in the Track Your Plaque program to gain control over the factors that lead us to coronary plaque and heart attack is a low-carbohydrate diet. We begin with elimination of wheat, cornstarch, oats, and sugars in the context of an overall carbohydrate-reduced diet. We refine the program by monitoring postprandial (after-meal) glucoses.

But not everything low-carb is good for you. Fried sausages, for instance, are exceptionally unhealthy, despite having little to no carbohydrates.

An emerging but potentially very powerful issue is that of Advanced Glycation End-products, or AGEs. There are two general varieties of AGEs: endogenous (formed within the body) and exogenous (formed in food that is consumed).

Endogenous AGEs form in the body as a result of high blood glucose, i.e., glycation. When exposed to any blood glucose level of 100 mg/dl or greater, some measure of glycation will develop due to a reaction between glucose and various proteins, e.g., proteins in the lens of the eye, forming cataracts over time.

Exogenous AGEs form in food, generally as a result of heating to high-temperature. (AGEs is really a catch-all term; there are actually a number of reactions that occur in foods, not all of them involving sugars. However, the "AGE" label is used to signify all the various related compounds. The values quoted here are from Dr. Helen Vlassara's Mt. Sinai Hospital laboratory; reference below.)

Beef cooked to high-temperature yields plentiful AGEs. One gram of roast beef, for instance, contains 306,238 units. This means that an 8-oz serving yields 13.8 million units AGEs. Compare this to a boiled egg with 573 units per gram, raw tomato with 234 units per gram.

Butter contains an impressive 264,873 units AGEs per gram, the highest content per gram in the entire list of 250 foods tested in the Mt. Sinai study. A couple pats of butter (10 g) therefore contains 2.64 million units. A stick of butter that you might add to cake batter to make a cake therefore yields 30 million units of AGEs.

So there's nothing wrong with the fat of butter. It's AGEs that appear to be responsible for the endothelial dysfunction/artery-constricting, insulin-blocking, oxidation and inflammation reactions that are triggered. Among all of our food choices, butter is among the worst from this viewpoint.

Throw in the peculiar "insulinotrophic" effect of butter, and you have potent distortion of metabolic pathways, courtesy of the butter on your lobster.

(AGE data from Goldberg 2004. In this analysis, carboxymethyllysine was the marker used for AGE content.)

Incidentally, the new Track Your Plaque diet will soon be released as chapter 9 of the new Track Your Plaque book on the website.
Tags :

Comments (59) -

  • rhc

    10/20/2010 10:15:00 PM |

    Are you talking about cold butter consumed without heating?

  • GK

    10/20/2010 10:20:53 PM |

    And do exogenous AGEs make it into systemic circulation, or are they broken down into simpler forms on digestion?  That would be the crucial thing to know.

  • Anonymous

    10/20/2010 10:28:55 PM |

    food gone and water gone... we are to survive on air? no wait thats polluted too..

  • Anonymous

    10/20/2010 10:34:19 PM |

    Is there a way to mitigate potential damage caused by exogenous AGEs?

  • Tuck

    10/20/2010 11:20:34 PM |

    "The results indicate that diet can be a significant environmental source of AGEs, which may constitute a chronic risk factor for cardiovascular and kidney damage."

    I'll start worrying when they can do a little better than "may".

    We're back to the "Eating fat makes you fat" mindset here...

  • Cameron

    10/20/2010 11:29:46 PM |

    I'd echo the question about whether or not this issue is limited to over-heated butter or butter in general.

    Also, is there enough information in the source data to indicate whether or not clarifying the butter into ghee would offer any improvement?

  • Bill

    10/20/2010 11:50:56 PM |

    Funny.
    You promote soy, which is known to be bad for you, but dump on butter which is known to be good for you....
    Strange?

  • Anonymous

    10/21/2010 12:22:02 AM |

    From the article:
    "...(AGEs), the derivatives of glucose-protein or glucose-lipid interactions"

    Can anyone explain the glucose-lipid interactions in ...butter?! Sheesh! Talk about bad science, those people did not follow the DEFINITION, never mind the protocols!

  • Daniel

    10/21/2010 12:55:44 AM |

    Exogenous AGEs are handily dealt with my people with healthy metabolims.  

    I know that's not many of your patients, so if you consider this a patient blog, ignore my comment.  

    Many people think of this blog as a "paleo blog" or a "low-carb blog" but in recent months, you've been basing many of your posts (and thinking) on the metabolically impaired.

    I can eat a plain mashed potato for breafast without seeing my blood glucose go over 100.  Are potatoes bad for me?  I really don't think so.  2 million years of evolution suggests otherwise.  Are potatoes bad for your patients that have been poisoned by years of fructose and PUFA induced metabolic carnage?  Yes.

    Same for butter.  It's a convenient and healthy source of good quality fat.  It has a lot of AGEs, but you have presented ZERO evidence that dietary AGES are unhealthy for otherwise healthy PEOPLE.    In fact, such evidence doesn't exist.  

    So, Doctor, are you treating sick patients or trying to remain a figure in the world of the super healthy?

  • Jared M Johnson

    10/21/2010 1:25:41 AM |

    Is the high level of AGEs in butter due to pasteurization?

  • Anonymous

    10/21/2010 3:15:43 AM |

    not buyin' it

  • Robin

    10/21/2010 4:02:41 AM |

    You are slowly hacking away at all I hold dear. Sausages! Butter! Sigh.

  • Joel

    10/21/2010 4:30:49 AM |

    Dr. Eades addressed this issue in 2008 and came to a different conclusion:

    http://www.proteinpower.com/drmike/low-carb-library/low-carb-diets-reduce-oxidative-stress/

    He specifically addresses the Goldberg 2004 study in the first comment:

    "I agree that there are vastly more AGEs in cooked foods, especially meats. What I’m not so sure about is whether or not the AGEs we eat end up as AGEs in us. The transit through the extreme acidity of the stomach would, I imagine, reduce the AGEs to their components, which we would absorb. The healthy human GI tract doesn’t have the ability to absorb large molecules. Even diglycerides (sugars composed of two other sugars, sucrose, for example) must be broken down to monoglycerides before being absorbed, so I seriously doubt that complex molecules such as AGEs could be absorbed in there native state. As a consequence, I’m not particularly worried about the AGEs I eat – I much more worried about the AGEs I create within."

    He also cites studies indicating that ketogenic diets reduce oxidative stress, despite butter and fried sausage being very common components of a ketogenic diet.

  • Joel

    10/21/2010 4:41:56 AM |

    Another one showing how vegetarians have higher levels of AGEs than omnivores:

    http://www.proteinpower.com/drmike/sugar-and-sweeteners/vegetarians-age-faster-2/

    Most likely due to a high fructose intake.

  • Anonymous

    10/21/2010 6:20:16 AM |

    What about butter from grass-few cows, ghee, goat's butter, or high vitamin butter oil? Do you relate to them in the same way?

  • Hans Keer

    10/21/2010 7:17:02 AM |

    Are we talking about heated butter here? Dietary AGEs should not be a problem; unless you have a leaky gut, they don't make it into the bloodstream. The problem with butter is that it, like all dairy, raises insulin and it still contains growth hormones and dangerous proteins.

  • D.M.

    10/21/2010 7:53:04 AM |

    Couple of points.
    First, that very paper says that only about 10% of exogenous AGEs actually make it into circulation, so that automatically takes butter down to 26.5KU/g. Of course if a patient has advanced kidney failure then worry about exogenous AGEs should be a concern, but so should protein, potassium etc etc.

    Secondly, the focus on exogenous AGEs in this table is obviously one-sided. Saying that butter contains more AGEs than a bowl of fructose, ignores the fact that once inside the body, the carbohydrate will cause immeasurably more glycation than the fat. These researchers are quite obviously pushing an lipophobic agenda here and I wouldn't fall for it.

    Third, it's not just butter apparently, but olive oil is also 120KU/ml or about 900 times more than an apple. But it would surely be absurb to think that apples will glycate less then olive oil?

    Fourtly, there something extremely suspect about the fact that whole milk contains 5300 times less AGE than butter. This should make us think twice before thinking that there's something uniquely bad about dairy fat that this study has discovered.

  • medeldist

    10/21/2010 8:03:21 AM |

    I find it hard to believe that butter (you do mean butter made from cow-milk, not margarine?) and red meat, two natural products, could be unhealthy for you. Anecdotal evidence says otherwise.

  • JLL

    10/21/2010 9:20:08 AM |

    The studies on AGEs are most often done on animals that have problems to begin with (e.g. diabetes). It's not clear at all whether consuming (a reasonable amount of) AGEs is harmful for healthy individuals.

    I've also reported about the AGE content of butter (see the list of AGEs in various foods) and I don't quite understand how they got such a high reading for butter. Did they heat it up? The processing of butter doesn't seem like it should result in much AGEs since milk is pretty low in AGEs.

    Like most commenters, I'm more worried about AGEs produced inside the body than AGEs from foods. And I'm even more worried about ALEs (Advanced Lipid peroxidation End-products) than AGEs.

    See my blog for more posts on glycation and lipid peroxidation (and how to avoid them).

  • Greensmu

    10/21/2010 12:05:28 PM |

    With the combination of A1 beta casein and AGEs in typical butter I think clarified butter/ghee with the cholesterol, lactose, and casein removed should be an improvement.

    But I second D.M. on the milk/butter thing, even though (like everyone else apparently =p) I have not checked the study referenced. It would follow that if they are both pasteurized they should be similarly high in AGEs.

  • Peter

    10/21/2010 12:06:22 PM |

    How do we know that eating more AGE's damages our cardiovascular system?

  • Stephen

    10/21/2010 1:08:00 PM |

    This sounds rather similar to "eating cholesterol results in an increase in cholesterol in the blood which causes heart disease and thus death."

    And butter is bad while soy is good? I'm not buying it.

    As others have mentioned - what population are we talking about here?

  • Alfredo E.

    10/21/2010 1:39:40 PM |

    Very illuminating post. I had no idea that butter had all those AGEs, I use it liberally in my cooking. I wonder what to use now instead of butter, lard?

    It would be very illustrative to educate us in ways to cook meat at low temperature.

    Thanks for the wonderful information.

  • Anna Delin

    10/21/2010 2:02:45 PM |

    Would a measurement of CRP reveal the inflammation potentially caused by the AGEs i eat? If I maintain an ideal CRP for years on a butter-rich diet, should I still worry?

  • Anand Srivastava

    10/21/2010 2:54:41 PM |

    I wonder why we love the taste of roasted meat when it is supposedly so unhealthy.

    It makes sense that the AGEs will not reach the blood stream if you have a good digestive system. If not well everything is a poison.

    Still Meat and Fat would be less of a poison than lectins from grains and legumes or even vegetables.

  • Martin Levac

    10/21/2010 3:34:39 PM |

    Dr. Davis, I'm confused. It's all your fault. If I just stick to low carb, it's all fine. But as soon as you start blaming butter, this low carb idea stops making any sense. Why would a low carb diet return me to good health when this very same low carb diet is blamed for disease?

    Clean slate. Start over. Fact, a  low carb diet returns me, and pretty much everybody else, to good health. Fact, a low carb diet contains lots of fat especially saturated animal fat. Fact, butter is one such fat and now we find that it contains lots of AGEs. Fact, in spite of this butter returns me to good health because it's part of a low carb diet. Logical conclusion, whatever I find in butter must be why I am now in good health.

    So why are you saying that butter is now bad for me?

  • Diana

    10/21/2010 4:41:17 PM |

    WoW great blog good to know since i love butter... but i totaly dont understand the whole Can anyone explain the glucose-lipid interaction thing.... thanks!

  • zach

    10/21/2010 6:13:00 PM |

    Butter is better for normal humans under normal circumstances than any plant food in existence. Butter: Food of the gods.

  • Eric

    10/21/2010 8:13:50 PM |

    I would also wonder if it's due to pasteurization.

  • Jack

    10/21/2010 8:38:18 PM |

    well dr davis, clearly you are ruffling the feathers of your readers with this one. nothing wrong with that in particular, except for when, as in this case, the information presented ruffles feathers because we all know it's just not possible. people have been eating (and studying the effects of) butter for a reaaallly long time. pretty much all whole food health gurus (meaning the awesome new wave of nutrionist/doctor bloggers that has sprung up this past decade) agree that full fat butter is very healthy to consume even in fairly substantial amounts. in fact, they ARE consuming it, and living very well while doing so. grass fed butter in particular, as you are well aware, has been tested and studied extensively, and the fat soluable vitamins and nutrients are so rich its astounding.

    just because something is found to have high AGEs before consumption, doesn't mean that particular item is causing the problems that you blame butter for here. be careful not to attack one of the most hallowed health foods unless you have have absolutely rock solid information that people can stand on.

    i only say this because i know you run a well articulated blog here and your name gets around on many other similar minded blog sites. i have read many of your articles, but reading articles like this make me (and many of your other 'faithfuls') cringe, because we really cannot agree with this.

  • Dr. William Davis

    10/21/2010 10:37:41 PM |

    Unfortunately, the data do not specify how or what was done to the butter, if anything. I suspect it was just off-the-shelf butter.

  • Dr. William Davis

    10/21/2010 10:44:27 PM |

    There seems to be a lot of misunderstandings here about what Vlassara et al's data are showing. This one perspective reported here does not do justice to this fascinating topic, which is clearly worth pursuing further.

    It's not my role to indulge anyone's low-carb fantasies. I am trying to interpret observations and data to employ in as effective a diet approach as possible.

    The data stand: Butter has some problems, despite fitting into most people's conception of low-carb.

  • Anonymous

    10/21/2010 11:31:39 PM |

    This can be interesting news, apparently not all paleo people had a paleodiet
    http://www.dailymail.co.uk/sciencetech/article-1321844/Stone-Age-man-ate-bread-just-meat.html

  • Joel

    10/22/2010 12:56:48 AM |

    Somebody correct me if I'm wrong, but every study on AGEs I've managed to dig up involves feeding humans or rats a lab "preparation" of AGEs, rather than actual real food.

    Some of the earliest arguments against a high protein diet came from  experiments with feedings of pure casein or liquid protein powders. When these experiments are repeated with whole food, the results are markedly different.

    "It's not my role to indulge anyone's low-carb fantasies."

    You're shunning of butter seems to follow this chain of association:

    1) Certain AGEs in the body are  bad.
    2) Butter contains significant AGEs (type of butter? type of AGEs?).
    3) Feeding pure AGE solutions to humans increases AGEs in the body.
    4) Ergo, eating butter increases AGEs in the body.

    However, certain AGEs such as pyrraline (commonly found in milk products) have been shown NOT to be metabolized in the body:

    http://www.biochemsoctrans.org/bst/031/1383/0311383.pdf

    Are we getting the full picture here? Until a study shows that feeding butter significantly increases AGEs in the body, I think we're in the land of speculation.

  • Martin Levac

    10/22/2010 2:09:15 AM |

    Dr. Davis, the kind of data you presented in your "case against butter" is merely the sort that explains how it works and what it's made of, not the sort that tells us whether butter is good or bad. We can figure out if something's good or bad without knowing how it works, we just feed it to somebody and wait for a result. We can also learn how it works without knowing if it's good or bad. We just feed it to somebody and draw some blood.

    The data you rely on here is the latter kind. It doesn't tell us whether butter is good or bad, it merely tells us how butter works and what it's made of. Now you believe that some of what it's made of, and some ways it works, is bad for us and you conclude that because of this butter is also bad for us. But in order to fully believe this you must also ignore the data that says that butter is good for us.

    Dr. Davis, you of all people should know health is not merely a measure of what's in the blood, let alone the measure of a single blood parameter.

    What we should conclude instead is that our understanding of the data regarding butter has problems.

  • Anonymous

    10/22/2010 3:20:12 AM |

    diglycerides (sugars composed of two other sugars, sucrose, for example)

    Eades really wrote that??? LOL. He should go back and study some Biochem 101 to find out the difference between diglycerides and disaccharides.

  • escee

    10/22/2010 3:30:15 AM |

    I might have seen this article referenced at this site previously, but I think it is worth revisiting in view of the topic.

    Food Choices and Coronary Heart Disease: A Population Based Cohort Study of Rural Swedish Men with 12 Years of Follow-up

    Abstract: Coronary heart disease is associated with diet. Nutritional recommendations are frequently provided, but few long term studies on the effect of food choices on heart disease are available. We followed coronary heart disease morbidity and mortality in a cohort of rural men (N = 1,752) participating in a prospective observational study. Dietary choices were assessed at baseline with a 15-item food questionnaire. 138 men were hospitalized or deceased owing to coronary heart disease during the 12 year follow-up. Daily intake of fruit and vegetables was associated with a lower risk of coronary heart disease when combined with a high dairy fat consumption (odds ratio 0.39, 95% CI 0.21-0.73), but not when combined with a low dairy fat consumption (odds ratio 1.70, 95% CI 0.97-2.98). Choosing wholemeal bread or eating fish at least twice a week showed no association with the outcome.
    Int. J. Environ. Res. Public Health 2009, 6, 2626-2638;

  • greensmu

    10/22/2010 2:11:29 PM |

    @Martin levac

    It doesn't tell us what butter is made of because we don't know if it was pasteurized or heated/cooked. My guess would be heated since that's what the study in question is looking at, heated foods. It's also known that butter has a very low smoking point.

  • Anonymous

    10/22/2010 3:46:00 PM |

    This is very interseting about butter. I have avoided butter because it is a non paleo food. It always seems that there are problems with these "new foods"

    Some things that I wonder though, are has this AGE content be measured accurately? Are there other studies that confirm this high level of AGEs in butter? Could butter from  pasteurized milk be higher in AGEs? Also could the level of freshness and the time it was frozen have some impact? These are some of the questions to consider.

    So far as the contention by some here that these chemicals don't pass into your system through your digestive system. The literature that I have seen clearly shows that they do pass through into your system.

  • Chuck

    10/22/2010 6:00:48 PM |

    questions about butter.  first as many have asked, was the butter heated for patuerization? my guess is yes.  second, what were the cows feed?  standard grain feed would probably lead to ore endogenous AGE in cows compared to a diet of grass.  as for now, i am sticking with my grass fed, non pasteurized butter.

  • Anonymous

    10/24/2010 7:21:18 AM |

    Nothing wrong with saying "Whoops.  My bad.  Thanks for correcting me with your comments guys and gals".

  • Anonymous

    10/24/2010 6:10:30 PM |

    Sorry Doc,

    This has been one of your least helpful, and nearly destructive blogs, I've ever seen. If you truly believe butter is not good, why not research how it could be 'better', such as clarifying it into ghee, or buying only grass-fed butter.

    So then what do YOU suggest instead as the best possible source of dietary fat???

    You must realize that the majority of people buy that horrible slow-poison known as margarine, because it has been billed as 'healthier', and your blog will only strengthen that perception.

    It seems like occasionally you go on vacation, and let the TYP committee post an article for you. This one stunk.

    The 6-year old study you quoted sounds like it was paid for by the vegetable oil industry.
    Anything we swallow gets nearly destroyed by our stomach acids, and who says that carboxymethyllysine (prior to digestion) is a proper marker for eventual AGE cell damage? Wouldn't Uric Acid have an even greater role? OR Hydrogen Peroxide induced in the blood or tissues? Doesn't Glucose, by far, cause the greatest destruction? Remind me what the G in AGE stands for?

    Weakly researched or justified blogs like this one make us lose faith in you as an expert.

  • Dr. William Davis

    10/25/2010 2:45:23 AM |

    No apologies from me.

    Just because you wish it weren't true, or that the data should be better sorted out, doesn't make it so.

    Until we obtain more clarification, butter remains on my list of "watch out."

    Wheat is unquestionably bad. Some foods, like spinach and kale, are unquestionably good. Other foods, like butter and other dairy products, have mixed effects.

    I'm talking butter here. I'm not insulting your aunt.

  • Anonymous

    10/25/2010 7:54:06 AM |

    I'm not that much of a fan of butter since I've got an autoimmune disorder which seems to get slightly worse with dairy, but, wouldn't ghee/clarified butter remove all/most of the AGEs throught seperation and physical removal of the sugars and proteins, leaving only the pure fat?
    Even AGEs from super-heated pasturized butter would be removed...
    Unless the fat itself gets glycated
    (this is the first time I've heard of this but it seems plausible, and ghee won't get rid of oxidized unsaturated fats from pasturized butter)

    Here's something else I don't understand: what makes butter so special in regard to external A.G.E.s as opposed to other low-carb, high-fat foods that it would warrant special attention?
    If butter can be filled with A.G.E.s, wouldn't a bunch of other low-carb foods considered healthy now become suspect?
    Or is the heating process itself that makes the pasteurized butter they likely tested on the culprit?
    (In the same way canola and soybean oils are hot-pressed to reduce toxins and therefore are highly oxidized)

  • Stephen

    10/25/2010 7:51:35 PM |

    I thought that the butter used in that study was whipped butter. If so, the measured AGE content might be drastically different from normal butter.

  • travis t

    10/26/2010 7:37:57 PM |

    Am I missing something, I thought AGEs were a combination of sugars and proteins. The label of my butter says zero carbs and zero protein. So what is glycated ?

  • Jack

    10/27/2010 4:59:46 PM |

    "No apologies from me."

    “It's not my role to indulge anyone's low-carb fantasies.”

    “I'm not insulting your aunt.”

    interesting attitude. i'm not real certain that an apology is in order specifically for your article, but perhaps a more in depth look at the 'data' is. the type of people who come here have a veracious appetite to find the real truth, and you are ignoring a host of excellent replies that directly negate the 'data' and 'facts' that you are standing on.

    i am not seeing "i love justifying my high fat foods because i am hopelessly addicted to butter" kind of replies here. i am seeing well researched, well articulated points about why the 'data' you presented here (and in your other previous article where you do state as a fact that "butter makes you fat") are not holding up well. And therefore, the quotes from you that I point out above do actually seem to be a bit insulting to your readers. your reply is quite pompous as well.

    please keep in mind that we (meaning the collective group of caring folk who frequent your blog) are only making noise on this one for everyone’s good. you may not want to be so hasty in shunning good responses that question your findings, but, uh, it's your call doc, and your reputation.

    as always, i appreciate the work you do. even with my disagreement about an article like this, i believe you do a great service to the health community and i sincerely thank you for it.

  • Sebastien

    10/28/2010 9:34:50 AM |

    It's funny you mentioned that spinach and kale are unquestionably good. I can easily find plenty of bad in those two vegetables. High levels of oxalates is one. Kale is also highly goitrogenic. Those two vegetables are also some the most pesticide laden. On top of the pesticides, spinach is often irradiated.

    I'll stick with occasional greens and frequent butter consumption.

  • Olga

    10/28/2010 5:32:11 PM |

    Hi Dr. Davis:

    Please take a look at the daily lipid's post from today, on AGE's.  Here is the link:
    http://blog.cholesterol-and-health.com/2010/10/is-butter-high-in-ages.html

  • blogblog

    10/31/2010 12:59:32 AM |

    To paraphrase Henry Ford "nutrition is bunk". No statistically valid long term dietary clinical trial has ever been performed on humans. So we have no statistically valid evidence-based science on what constitutes a healthy diet. In particular the recommendations for eating fruit and vegetables is totally irrational. All vegetables are full of toxins and contain large quantities of known carcinogens. In fact the EPA would be required by law to ban the consumption and sale all vegetables if they were man made.

    Nutrition 'science' consists entirely of extremely dubious experiments on rats, meaningless population studies and irrelevant test tube experiments.

  • Anonymous

    11/3/2010 9:23:11 PM |

    @blogblog

    What you say is ridicolous.
    Consumption of vegetables has always been found to have nothing but extremely positive effects and not even one negative effect, except for people with Chrons.

    Not even one evidence of cangerous or toxic effect.

  • Ed

    11/16/2010 5:23:50 AM |

    The source of the butter data is this paper: "Advanced Glycoxidation End Products in Commonly Consumed Foods" (2004, Journal of the American Dietetic Association, via Google Scholar cache).

    Here are some numbers from Table 1:

    Milk, cow, whole .... 0.05 kU/mL
    Butter .............. 265 kU/g

    The table caption refers to "foods prepared by standard cooking methods" (these include frying). Expecting high AGEs in uncooked butter -- over 5000 times the level in milk! -- would make little sense. There's every reason to think that this butter had been exposed to high temperatures.

  • Jack

    11/17/2010 6:27:41 PM |

    @anonymous (Nov 3 comment)
    Actually, what you say is ridiculous too. I'd be careful not to make blanket statements like that. Built-in defense mechanisms are not reserved for Venus Fly-traps only. Vegetables, like many other plants, have them too.

    PLANTS BITE BACK

  • Joe

    12/7/2010 1:22:29 PM |

    What do you think about this from Dr Mercola?

    Good-old-fashioned butter, when made from grass-fed cows, is a rich in a substance called conjugated linoleic acid (CLA). CLA is not only known to help fight cancer and diabetes, it may even help you to lose weight, which cannot be said for its trans-fat substitutes.

    http://articles.mercola.com/sites/articles/archive/2010/12/07/why-is-butter-better.aspx

  • Anonymous

    12/7/2010 6:32:42 PM |

    According to the chart, a frankfurter or a serving of roast beef is quite a bit worse than a serving of butter.
    http://inhumanexperiment.blogspot.com/2009/09/age-content-of-foods.html

  • jpatti

    6/18/2011 9:42:06 PM |

    Butter is not good because it's low carb.  Butter is good because it's butter.  

    Before I ever heard of low-carb, or vitamins or minerals or any of that, when ALL I knew about nutrition was that sugar was bad and veggies good cause mom said so, butter was good.  Butter made me WANT to eat an artichoke.  And... it still works today!

    If there were no other benefit to butter than it made vegetables palatable, butter would be an unqualified good.  I would not eat 1/10th the veggies I do if not for butter.  

    Since I am stubbornly of the opinion that eating at least half the diet (by volume) as nonstarchy vegetables is the main thing anyone can do for health, butter is an unqualified good in my world.  

    If it makes people voluntarily eat their veggies, it's good.  

    *********************************************************************

    While just the veggie intake with butter in the diet is a HUGE good; butter is better than just the vegetables that go with it.  

    Butter is the number one source of butyric acid, a fatty acid that is a major constituent of the GI tract and often deficient in folks with GI disturbances like celiac and Chron's and systemic Candida.  IMO, the number one thing anyone with GI issues can do is eat lots of butter.  If you want to heal even faster, don't just eat it, but take it in both ends, so to speak.  

    Butyric acid also counteracts inflammation, the main underlying issue with heart disease as I understand, and the apparent underlying issue with the epidemic of autoimmune disorders we're seeing.

    My grandmother's generation ate GOBS of bread, wheat was a mainstay of their diet.  But they didn't have all the gluten-intolerance this generation has.  IMO, the reason is cause they slathered butter on their bread.  

    Anyways, she lived to 102, so must've done SOMeTHING right.  And she never believed the hype about margarine, always overate butter like crazy.

    Butyric acid has other interesting effects... it lowers total cholesterol 25%, serum triglycerides 50%, fasting insulin 25%, and increases insulin sensitivity 300% - there's a bunch of pubmed references listed here: http://wholehealthsource.blogspot.com/2009/12/butyric-acid-ancient-controller-of.html

    Note that "metabolic syndrome," the precursor to T2 diabetes, is pretty much insulin resistance and high triglycerides.  When metabolic syndrome is the question, apparently, butter is the answer.

    *********************************************************************

    Butter is particularly good from pasture-raised animals, which maximizes the vitamins A, D3 and K2 in it.  

    Very few of us get enough vitamin A.  Many of us, diabetics being an example I'm terrifically familair with, do not convert beta-carotene to vitamin A well at all.  In general, omnivores and carnivores don't do this efficiently, even the healthy ones with good genes.  

    Herbivores do it wonderfully.  All the gorgeous colors of the pasture convert into lots of real vitamin A for us to eat.  You can take nasty cod liver oil, or you can just melt yummy butter on your veggies.

    I do not spend 16 hours in the sun in summer.  But I rent a small house on a farm and am surrounded by cattle, and they do.  They walk about, eating pasture, chewing cud and the calves frolicking across the fields, in the sunshine all day, where they also are making loads of vitamin D3 - the real stuff, not the crappy D2 they "fortify" factory farmed milk with.

    Butter from cows eating rapidly growing grass is also the best known source of K2 other than natto.  Just like Vitamin A, we are not good at making K2, but cows are.

    *********************************************************************

    IMO, butter is a near-miraculous food, one of the true health foods.  

    I buy from a farm that makes butter from cream from cows on pasture, with no ingredients except cream.  When the beta-carotene content is highest, it turns darker, which is also when the vitamin A, D3 and K2 is highest.  When it gets like that, I buy 40 lbs and stick it in my freezer for consumption over the next year.  When I run out, I just buy it weekly again until it gets dark again.

    I eat between 1/2 - 1 lb butter every week. It's yummy.  As noted, it's wonderful on vegetables.  But it's also nice just melted over some over-easy eggs, or a pat melted on a burger or steak.  

    Also, pasture-raised butter tastes better.  The stuff I buy comes in tubs, not sticks, but hubby being a truck driver finds sticks more convenient.  He buttered a dish with his butter recently before he served it to me and... well, I added the real butter.  His butter just wasn't... buttery enough.  

    Butter is... just awesome stuff.  And for those who REALLY disagree, my advice is to heed Julia Child who said, "If you're afraid of butter, use cream."

  • Florent Berthet

    2/7/2012 6:04:57 PM |

    Like Olga, I''d be very interested to hear your opinion on this daily lipid''s post:
    http://blog.cholesterol-and-health.com/2010/10/is-butter-high-in-ages.html

    Also, what about ghee?

  • Alex Tahti

    11/5/2012 7:21:42 PM |

    Apparently the AGEs in the study cited by Dr. Davis were measured using anti-body immunoassay which is an indirect method that is susceptible to distortions.   A mass spectrometer, a direct measurement, was used to analysis AGE in butter in this study http://biomedgerontology.oxfordjournals.org/content/65A/9/963.full and found: "The CML concentrations of various foods vary widely from about 0.35–0.37 mg CML/kg food for pasteurized skimmed milk and butter to about 11 mg CML/kg food for fried minced beef and 37 mg CML/kg food for white bread crust".

    So wheat in the form of white bread crust is a factor of 100 more than butter in CML AGE.

Add comment

Loading