Advanced Mediterranean Diet

Did you know that atherosclerosis - hardening of the arteries - starts in childhood?

The American Journal of Medicine published a supplement in January this year (volume 122, number 1A) entitled “Management of Atherosclerosis: A Pratical Guide in 2008.”

William Insull, Jr., M.D., wrote a chapter on the development and natural history of atherosclerosis.  I summarized it at the Heart Health Blog at NutritionData.com.

The heart attack that seems to strike out of the blue in the 57-year-old has actually been developing for years.  Modification of atherosclerosis risk factors, if maintained for years, prevents heart attacks and strokes, and prolongs life.

Steve Parker, M.D.

On April 24, 2009, NutritionData.com published a blog post of mine discussing heart failure.  If interested, please click here to see the post. 

Today, NutritionData published my post on causes and prevention of heart failure.

Steve Parker, M.D.

Do you know of a website, book, or other generally available resource devoted to adaptation of the Mediterranean diet for use by people with type 2 diabetes?

In the Winter of 2009, I was looking for step-by-step instructions for implementing a “diabetic Mediterranean diet,” for my patients with diabetes.  Why?  People with type 2 diabetes are prone to premature heart attacks, strokes, dementia, and death. The traditional Mediterranean diet is associated with lower rates of those diseases, along with increased lifespan.  So a combination “diabetic” and “Mediterranean” diet seems like a great idea.

[“Diet” in this context refers to the usual food and drink of a person, not a weight-loss program.  Nevertheless, most people with type 2 diabetes are carrying excess weight.]

Dietitians know “diabetic diet” and “Mediterranean diet” but can’t necessarily combine the two easily.  It’s a new concept within the scientific literature.  I need educational resources for my patients now, not five years from now.

Searching “mediterranean and diabetes” books at Amazon.com got me 170 hits, but nothing pertinent. Googling “Mediterranean diet and diabetes” yielded 201,000 hits. I found nothing pertinent in the first 50 hits.

Since I couldn’t find any readily available resources devoted to a diabetic Mediterranean diet, I decided to investigate on my own whether such a program could and should be devised.  The potential benefits to my patients and others are enormous.

Hence, the birth of a new blog: the Diabetic Mediterranean Diet Blog.

Over the coming months, I will be reviewing the scientific literature that might support a diabetic Mediterranean diet, as well as getting input from dietitians, nutritionists, primary care physicians, endocrinologists, and people with type 2 diabetes.  I’ll report my musings, theories, and conclusions at the new blog, and to a lesser extent here.

I expect readership at the new blog will be mostly people with type 2 diabetes, dietitians, nutritionists, primary care physicians, and endocrinologists.

Let’s figure this thing out together!

Steve Parker, M.D.

Update May 27, 2010:  The initial version of the Diabetic Mediterranean Diet is now available at the Diabetic Mediterranean Diet Blog.

I am pleased and honored to be blogging for the next three months with the merry band of bloggers at NutritionData.com.  I’ve been visiting and recommending ND for years.

I will be writing in the NutritionData Heart Health Blog.

My compadres at ND are

• Monica Reinagel, M.S., L.D.N., C.N.S.; ND’s chief nutritionist and a prolific author
• Dana Lilienthal, M.S., R.D.
• Stephen Cabral, C.S.C.S, C.P.T., N.S.
• Elaine Murphy, B.A., C.N.C.

[Gee, now I feel like I need more letters after my name!]

Please take time to visit the NutritionData website, or see my review of it here.  The site is frequently updated and improved, so my review may be outdated.

I’ll still be posting here at my usual frequency, about twice weekly.

Steve Parker, M.D.

I recently blogged about glycemic index (GI), glycemic load (GL), and glycemic diets in preparation for today’s post.

The concept of glycemic index was introduced by Jenkins et al in 1981 at the University of Toronto.

Studies investigating the association between disease risk and GI/GL have been inconsistent.  By “inconsistent,” I mean some studies have made an association in one direction or the other, and other studies have not.  Diseases possibly associated with high-glycemic diets have included diabetes, cardiovascular disease, cancer, gallbladder disease, and eye disease.

“Diet” in this post refers to a habitual way of eating, not a weight loss program.

Researchers with the University of Sydney (Sydney, Australia) identified the best-designed published research reports investigating the relationship between certain chronic diseases and glycemic index and load.  The studied diseases were type 2 diabetes, coronary heart disease, stroke, breast cancer, colorectal cancer, pancreatic cancer, endometrial cancer, ovarian cancer, gallbladder disease, and eye disease.

Methodology

Literature databases were searched for articles published between 1981 and March, 2007.  The researchers found 37 studies that enrolled 1,950,198 participants ranging in age from 24 to 76, with BMI’s averaging 23.5 to 29.  These were human prospective cohort studies with a final outcome being occurrence of a chronic disease (not its risk factors).  Twenty-five of the studies were conducted in the U.S., five in Canada, five Europe, and two in Australia.  Ninety percent of participants were women [for reasons not discussed].  Food frequency questionnaires were used in nearly all the studies.  Individual studies generated between 4 to 20 years of follow-up, and 40,129 new cases of target diseases were identified.

Associations between GI, GL, and risk of developing a chronic disease were measured as rate ratios comparing the highest with the lowest quantiles.  For example, GI and GL were measured in the study population.  The population was then divided into four groups (quartiles), reflecting lowest GI/GL to medium to highest GI/GL diets.  The lowest GI/GL quartile was compared with the highest quartile to see if disease occurrence was different between the groups.  Some studies broke the populations into tertiles, quintiles, deciles, etc.

Findings

Comparing the highest with the lowest quantiles, studies with a high GI or GL independently

• increased the risk of type 2 diabetes by 27 (GL) or 40% (GI)

• increased the risk of coronary heart disease by 25% (GI)

• increased the risk of gallbladder disease by 26% (GI) or 41% (GL) [gallstones and biliary colic, I assume, but the authors don’t specify]

• increased the risk of breast cancer by 8% (GI)

• increased risk of all studied diseases (11) combined by 14% (GI) or 9% (GL)

Overall, high GI was more strongly associated with chronic disease than was high GL
So low-GI diets may offer greater protection against disease than low-GL diets.

Comments from the Researchers

They speculate that low-GI diets may be more protective than low-GL because the latter can include low-carb foods such as cheese and meat, and low-GI, high-carb foods.  Both eating styles will reduce glucose levels after meals while having very different effects in other areas such as pancreas beta cell function, free fatty acid levels, triglyceride levels, and effects on satiety.

High GI and high GL diets, independently of known confounders, modestly increase the risk of chronic lifestyle-related diseases, with more pronounced effects for type 2 diabetes, coronary heart disease, and gallbladder disease.

Direct quotes:

. . . 90% of participants were female; therefore, the findings may not be generalizable to men.

There are plausible mechanism linking the development of certain chronic diseases with high-GI diets.  Specifically, 2 major pathways have been proposed to explain the association with type 2 diabetes risk.  First the same amount of carbohydrate from high-GI food produces higher blood glucose concentrations and a greater demand for insulin.  The chronically increased insulin demand may eventually result in pancreatic beta cell failure, and, as a consequence, impaired glucose tolerance.  Second, there is evidence that high-GI diets may directly increase insulin resistance through their effect on glycemia, free fatty acids, and counter-regulatory hormone secretion.  High glucose and insulin concentrations are associated with increased risk profiles for cardiovascular disease, including decreased concentrations of HDL cholesterol, increased glycosylated protein, oxidative status, hemostatic variables, and poor endothelial function

Low-GI and/or low-GL diets are independently associated with a reduced risk of certain chronic diseases.  In diabetes and heart disease, the protection is comparable with that seen for whole grain and high fiber intakes.  The findings support the hypothesis that higher postprandial glycemia is a universal mechanism for disease progression.

My Comments

Studies like this tend to accentuate the differences in eating styles since they compare the highest with the lowest post-prandial (after meal) glucose levels.  Most people are closer to the middle of the pack, so a person there has potentially less to gain by moving to a low-GI diet.  But still some to gain, on average, particularly in regards to avoiding type 2 diabetes and coronary heart disease.

[To be fair, many population-based studies use this same quantile technique.  It increases the odds of finding a statistically significant difference.]

Only two of the 37 studies examined coronary heart disease, the cause of heart attacks.  One study was the massive Nurses’ Health Study database with 75,521 women.  The other was the Zutphen (Netherlands) Elderly Study which examined men 64 and older.  Here’s the primary conclusion of the Zutphen authors verbatim:

Our findings do not support the hypothesis that a high-glycemic index diet unfavorably affects metabolic risk factors or increases risk for CHD [coronary heart disease] in elderly men without a history of diabetes or CHD.

So there’s nothing in the meta-analysis at hand to suggest that high-GI/GL diets promote heart disease in males in the general population.

However, the recent Canadian study in Archives of Internal Medicine found strong evidence linking CHD with high-glycemic index diets.  Although not mentioned in the text of that article, Table 3 on page 664 shows that the association is much stonger in women than in men.  Relative risk for women on a high-glycemic index/load diet was 1.5 (95% confidence interval = 1.29-1.71), and for men the relative risk was 1.06 (95% confidence interval = 0.91-1.20).  See reference below.

Nine of the 37 studies examined the occurrence of type 2 diabetes.  Only one of these studied men only - 42,759 men: the abstract is not available online and the Sydney group does not mention if high-GI or high-GL was positively associated with onset of diabetes in this cohort.  Two of the diabetes studies included both men and women, but the abstracts don’t break down the findings by sex.  [I’m trying to deduce if the major overall findings of this meta-analysis apply to men or not.]

I don’t know anybody willing to change their diet just to avoid the risk of gallstones.  It’s only after they develop symptomatic gallstones that they ask me what they can do about them.  The usual answer is surgery.

The report is well-done and seems free of commercial bias, even though several of the researchers are authors or co-authors of popular books on low-GI eating.

Steve Parker, M.D.

References:

Barclay, Alan W.; Petocz, Peter; McMillan-Price, Joanna; Flood, Victoria M.; Prvan, Tania; Mitchell, Paul; and Brand-Miller, Jennie C.  Glycemic index, glycemic load, and chronic disease risk - a meta-analysis of observational studies [of mostly women].  American Journal of Clinical Nutrition, 87 (2008): 627-637.

Brand-Miller, Jennie, et al.  “The New Glucose Revolution: The Authoritative Guide to the Glycemic Index - The Dietary Solution for Lifelong Health.”  Da Capo Press, 2006.

Mente, Andrew, et al.  A Systematic Review of the Evidence Supporting a Causal Link Between Dietary Factors and Coronary Heart Disease.  Archives of Internal Medicine, 169 (2009): 659-669.

Major media outlets in the last 48 hours have reported a new Canadian study that confirms the heart-healthy characteristics of the Mediterranean diet.  Examples are Forbes.com (Review Confirms Links Between Diet, Heart Health, April 12, 2009), UPI.com (Vegetables, nuts, Med Diet good for heart, April 14, 2009), and the Wall Street Journal (Heart Health: Mediterranean Diet Good, Trans Fats Bad, April 13, 2009).

Canadian researchers sought to systematically evaluate the strength of the evidence supporting links between dietary factors and coronary heart disease.  Coronary heart disease (CHD) is the number one cause of death in Western societies.

It’s important to understand the two types of studies meta-analyzed by the Canadians.

Prospective cohort studies can be used to identify a dietary factor that may be associated with a disease or outcome.  For example, researchers could study the health of 20,000 people over the course of 10 years, giving them questionnaires to find out what foods they typically ate.  They might find after 10 years that the people who ate the most saturated fat tended to die earlier and had more heart attacks and strokes compared to the people who ate the least saturated fat.  This would establish an association between dietary saturated fat and premature death, heart attacks, and strokes.  It does not prove that saturated fats cause those outcomes, it’s just an association.

Randomized trials, often called randomized controlled trials (RCTs), typically take two groups of people and apply an intervention to one group but not the other.  The groups are followed over time to see if there is a difference in outcome.  For example, take another group of 20,000 people.  Randomly assign 10,000 of them to eat more-than-usual saturated fat.  The other 10,000 similar people serve as the control group, eating their usual amount of saturated fat.  Follow these 20,000 people over 10 years, then compare their health outcomes: death, heart attacks, strokes.  If the high-saturated-fat group has worse outcomes, you are much closer to proving that dietary saturated fat causes premature death, heart attacks, and strokes.

Methodology 

The scientists located and analyzed every English-language prospective cohort study (146 studies) or randomized trial (43) investigating food intake and coronary heart disease (CHD), from 1950 through June, 2007.  They wrote:

We used the Bradford Hill guidelines to derive a causation score based on four criteria (strength, consistency, temporality, and coherence) for each dietary exposure in cohort studies and examined for consistency with the findings of randomized trials.

The different dietary patterns evaluated in studies were noted.  The “Mediterranean” dietary pattern emphasizes a higher intake of vegetables, legumes, fruits, nuts, whole grains, cheese or yogurt, fish, and monounsaturated relative to saturated fatty acids.  The “prudent” dietary pattern is characterized by a high intake of vegetables, fruit, legumes whole grains, and fish and other seafood.  The “western” pattern is characterized by a high intake of processed meat, red meat, butter, high-fat dairy products, eggs, and refined grains.

Results

Strong evidence (four Bradford Hill criteria satisfied) supported protection against CHD with consumption of:

• vegetables
• nuts
• monounsaturated fatty acids [prominent in olive oil, for example]
• Mediterranean diet
• prudent diet

Modertately strong evidence (three criteria satisfied) supported protection against CHD with consumption of:

• fish
• marine omega-3 fatty acids
• folate
• whole grains
• dietary vitamins E and C (as opposed to vitamin supplements)
• beta carotene
• alcohol
• fruit
• fiber

Strong evidence supported the following as harmful dietary factors, in terms of CHD:

• trans-fatty acids
• foods with a high-glycemic index or load
• western diet

Researchers found insufficient evidence (two or less criteria) to support an association between CHD and:

• total fat
• saturated and polyunsaturated fatty acids
• eggs
• meat
• milk
• vitamin supplements E and C
• alpha-linolenic acid

Selected Comments of the Researchers  [my comments in brackets]

Cohort studies provide abundant evidence of an association with total mortality for many dietary exposures.  Randomized controlled trials corroborate these associations for the consumption of omega-3 fatty acids and a Mediterraneandiet because most of the other dietyary factors have not been evaluated to date.

Among the dietary exposures with strong evidence of causation from cohort studies, only a Mediterranean dietary pattern is related to CHD in randomized trials.  [The association is inverse: Higher adherence to the Mediterranean diet leads to lower rates of CHD.]

A wealth of epidemiologic studies have evaluated associations between dietary exposures and CHD.  The general consensus from the evidence currently available is that a reduced consumption of saturated and trans-fatty acids and a higher intake of fruits and vegetables, polyunsaturated fatty acids including omega-3 fatty acids, and whole grains are likely beneficial.  This is reflected in the revised Dietary Guidelines for Americans 2005 from the US Departments of Health and Human Services and Agriculture.  However, little direct evidence from RCTs supports these recommendations.  [Emphasis added.]  In some cases, RCTs have not been conducted, and RCTs that have been conducted have generally not been adequately powered or have evaluated surrogate end points rather than clinical outcomes.

Single-nutrient RCTs have yet to evaluate whether reducing saturated fatty acid intake lowers the risk of CHD events.

More recently, the lack of benefit of diets of reduced total fat has been established [in women only? (reference below)], and the evidence supporting the adverse effect of trans-fatty acids on cholesterol levels and CHD has increased, which is reflected in our findings.  [This is the only mention of cholesterol in the report.]

My Comments

I wonder about vegetarian/vegan diets.  Have they been tested for efficacy against CHD?  What about Dr. Dean Ornish’s program?

Although not mentioned in the text of the article, Table 3 on page 664 shows that the positive association between CHD and high-glycemic index/load is much stonger in women than in men.  Relative risk for women on a high-glycemic index/load diet was 1.5 (95% confidence interval = 1.29-1.71), and for men the relative risk was 1.06 (95% confidence interval = 0.91-1.20).  I question whether the association for men is statistically significant. 

Why wasn’t there discussion of dietary cholesterol?  The public and physicians have been told for years that dietary cholesterol causes or aggravates coronary heart disease.  In a near-future post, I quote one expert stating that people at risk for coronary heart disease (aren’t we all?) should limit cholesterol to 200 mg/day.  Is there no evidence?

I’m a little surprised, but some of you won’t be, that the researchers found no clear association between CHD and saturated and total fat intake.  In traditional medical circles, these findings are considered sacrilegious!

Medical schools and cardiologists have been teaching for thirty or more years that they are related positively.  “Positively” means the more saturated and total fat in your diet, the more likely you are to develop atherosclerosis, which in the heart is coronary heart disease.  Dietary cholesterol is often thrown into the equation.  The is the dogmatic Diet-Heart Hypothesis.

Steve Parker, M.D.

References:

Mente, Andrew, et al.  A Systematic Review of the Evidence Supporting a Causal Link Between Dietary Factors and Coronary Heart Disease.  Archives of Internal Medicine, 169 (2009): 659-669.

van Dam, R.M., et al.  Dietary glycemic index in relation to metabolic risk factors and incidence of coronary heart disease: the Zutphen Elderly Study.  European Journal of Clinical Nutrition, 54  (2000): 726-731.

Howard, B.V., et al.  Low-fat dietary pattern and risk of cardiovascular disease: the Women’s Health Initiative Randomized Controlled Dietary Modification Trial.  Journal of the American Medical Association, 295 (2006): 655-666.

This is an April 16 update to my April 14 post which was based solely on the journal article abstract.  I have reviewed the full article. 

Any diet that contains carbohydrates can be ranked as being either low-, medium-, or high-glycemic, referring to the glycemic index or load.  This ranking system would apply to both weight-loss diets and habitual ways of eating.

Specific foods with significant amounts of carbohydrate have been tested for their ability to rapidly raise blood sugar levels as compared to eating pure glucose, a type of sugar.  The number generated by the test is the glycemic index (GI) and ranges between 0 and 100.  For details, see my blog post of April 11, 2009.

Here are some GI values from Wikipedia’s entry on glycemic index:

Low GI (55 or less)

• most fruit and vegetables (except potatoes, watermelon), grainy breads [made of or resembling grain?], pasta, legumes/pulses, milk, products extremely low in carbohydrates (fish, eggs, meat, nuts, oils), brown rice

Medium GI (56 - 69)

• whole wheat products, basmati rice, orange, sweet potato, table sugar, most white rices (eg, jasmine),

High GI (70 and above)

• corn flakes, baked potato, watermelon, croissant, white bread, extruded cereals (eg, Rice Krispies), straight glucose (100)

You figured it out, didn’t you, Spanky?

Yes, low-GI diets encourage the low-GI foods.  High-GI diets favor the high-GI foods.  Etc.

The Human Nutrition Unit at the University of Sydney (Australia) maintains a free online database of glycemic index values of various foods.  For example, the GI of Doritos is 42 (corn chips, plain, salted, 1998).

A near-future post will review some of the chronic diseases that might be associated with or caused by high-GI eating by the general public.

Steve Parker, M.D.

Additional Resources

GI News: The Official Glycemic Index Newsletter

“The Official Website of the Glycemic Index and GI Database”

And why should you care?

Because these concepts are related to some common chronic diseases.  Diets - i.e., habitual ways of eating - with a high glycemic index or glycemic load increase the risk of type 2 diabetes, coronary heart disease, gallbladder disease, and breast cancer.  At least in women.

Glycemic index is a measure of how much a specific food is likely to influence blood sugar (glucose) levels.  Carbohydrates we eat, except for fiber, are usually converted by the process of digestion into glucose which we use as fuel.  Any glucose not needed immediately for energy is converted into a storage form called glycogen, for use later.  We have the capacity to store only a half days’ worth of energy in the form of glycogen.  Carbohydrates are converted to fat when eaten in excess of what we can use immediately or store as glycogen.

The standard for glycemic index is set by eating 50 grams of pure glucose, a type  of sugar.  The pattern of bloodstream glucose levels over the next two hours is given a rank of 100.  Oral glucose leads to a more rapid and higher peak in blood sugar compared to nearly all (or all?) other carbohydrates.  All other foods containing carbs (carbohydrates) can be ranked in relation to glucose, on a scale of 0 to 100.  Test subjects are given specific foods in whatever serving size contains 50 grams of available carbohydrate, and eaten without other foods.  Blood glucose levels are measured repeatedly over the next two hours, and compared to the pattern observed for ingested glucose.  The comparison yields a number, the gylcemic index.

Even when they have the same grams of carbohydrate, some foods cause a higher rise in blood sugar.  That is, they have a higher glycemic index.

Cashews have a glycemic index (GI) of 22, which means they don’t raise blood sugar nearly as quickly or as much as glucose.  An overripe banana has a GI of 52, so it would tend to raise your blood glucose more than cashews.  Watermelon’s GI is 72.  Doesn’t eating watermelon remind you of drinking flavored sugar water?

With higher GI foods, your body digests and absorbs the foods’ carbs faster, leading to greater release of insulin by the pancreas to reduce the blood sugar levels back to normal.

Some of you have already figured out the the actual rise in blood sugar will depend on other factors, such as how much of the food you eat!  To account for the amount of food eaten, the Glycemic Load concept was devised (at Harvard?).  You calculate the glycemic load by taking the grams of carbohydrate in the serving size, multiply by the food’s glycemic index, then divide by 100.  For example, a cup of white rice has a glycemic load of 33 points; a cup of brown rice has a glycemic load of 23 points.

Remember, the glycemic index is based on the observed blood sugar rise after eating the serving of a food that contains 50 grams of carbohydrate.  For example, corn’s glycemic index is 53.  One half cup of canned yellow corn has 15.2 grams of carbohydrate.  So to get the full glycemic effect, you’d have to eat over one and a half cups of corn.  Most people just eat a half cup as a serving.  The glycemic load of a half cup of corn is 8 points (15.2 x 53, then divide by 100).  Carrots are a classic example of a healthy food with a high glycemic index, but a good/low glycemic load of 2.8 points per half cup (slices, frozen, boiled).

[I’m ignoring for now the grams of nondigestible carbohydrate (fiber) in the corn and carrots.]

Doesn’t this sound just like something you’d like to do at every meal?

Let not your hearts be troubled.  You can get most of the pertinent GI’s and GL’s free on the Internet or in various books or pamphlets.

Look for more on the health effects of GI and GL here soon.

Now put away your books and take out a clean sheet of paper and a pencil for a pop quiz . . .

Steve Parker, M.D.

[Blood sugar and blood glucose are identical.  Glucose is also a simple sugar you can eat.  I’ve eschewed the term “blood glucose” today to prevent confusion.]

Additional resources:

Glycemic Index entry at Wikipedia.

For additional information, see Laura Dolson’s good work at About.com.

WebMD.com has an article on factors that affect glycemic index and glycemic load.

Jenkins, D.J., et al.  Glycemic index of foods: a physiological basis for carbohydrate exchange.  American Journal of Clinical Nutrition, 34 (1981): 362-366.  This paper from the University of Toronto introduced the concept of glycemic index to the world.

In short, the answer is not clear.  But a recent study suggests it might.

The Atkins diet is higher in saturated fat (and probably cholesterol, too) than standard American fare.  It’s the classic low-carbohydrate diet.  Current dietary dogma is that high intake of saturated fat promotes atherosclerosis - hardening of the arteries - which in turn leads to premature heart attacks, strokes, and death.  Whether related or not, it’s pretty clear that higher saturated fat intake leads to higher blood cholesterol levels.  This is a major concern about the Atkins diet and other low-carb plans among traditional medical practitioners and dietitians.

The postulated relationship between high dietary saturated fat (and cholesterol) and higher rates of atherosclerotic complications is often called the Diet-Heart Hypothesis.

The aforementioned study compared effects of the Atkins diet, South Beach diet, and Ornish diet in healthy people.  Fat content of the diets was 50% (in Atkins), 30% (South Beach), and 10% (Ornish).  Researchers wanted some idea of how the diets affect specifice cardiovascular risk factors in the maintenance phase of these diets.

Methodology

The 18 healthy study participants were in the normal weight range (average body mass index 22.6). Participants tried each of the three diets for four weeks.  So they served as their own controls. 

In between diet runs, participants went back to their usual dietary patterns for four weeks.  Weight loss was not a goal; adherence to the specified diets was.

At the start and end of each four-week diet, researchers measured cholesterol levels, C-reactive protein, and a test of blood vessel functioning (brachial artery testing of flow-mediated vasodilation). Three-day food records were examined at the end of each diet phase to check for compliance.

Results

Participants ate about 30 grams of saturated fat daily on the Atkins diet, about 14 grams on South Beach, and 3 grams on Ornish.

The 8% increase in LDL cholesterol seen with Atkins was not statistically significant.  [This means there is a good chance the 8% increase was due to chance alone rather than a true effect of the Atkins diet.]

LDL cholesterol decreased by 12% on South Beach, and by 17% on Ornish.  These were statistically significant.

As saturated fat intake increased, flow-mediated vasodilation decreased (p=0.016, meaning statistically significant). Compared with Atkins, the other groups showed statisically significant decreases in apolipoprotein B.

Conclusion of the Researchers

These data suggest that during weight maintenance, less favorable biological effects are observed during a simulated, high-fat Atkins diet when compared to the South Beach and Ornish diet. The findings support additional study in subjects with visceral obesity and the metabolic syndrome, in whom an increased risk of coronary disease at baseline may be accentuated with chronic consumption of a diet that exhibits unfavorable effects on lipids and endothelial function.

Comments from Atkins Nutritionals (as reported by HealthDay on April 1, 2009)

Representatives from Atkins Nutritionals took exception with the study, however. In a statement, Colette Heimowitz, vice president of nutrition and education at Atkins Nutritionals, noted that, “given the short duration of his study, the very small sample size and the weak correlations, drawing conclusions about possible long-term health risks tied to fat consumption in the maintenance phase of any weight control program is not good science.” She also questioned whether the participants were actually following a true Atkins diet.

My Comments

I regret that I only have access to the article abstract, not the entire article.  C-reactive protein results were not reported in the abstract, nor were HDL and total cholesterol levels. 

The observed blood lipid changes would tend to favor the South Beach and Ornish diets over Atkins, in terms of future cardiovascular health.  But Heimowitz is right: this was a very small, short-term study, so it’s difficult to apply the results to common clinical situations.  There is ligitimate cause for concern, however.

If the Diet-Heart Hypothesis is wrong, then the observed lipid changes don’t matter.

I’m not very familiar with brachial artery testing of flow-mediated vasodilation.  It would make sense that “healthy vessel dilation is important to proper blood flow” (from the HealthDay article).  But that doesn’t mean it is so.

South Beach’s maintenance phase is similar to the traditional Mediterranean diet, although lower in carbohydrates and higher in protein.  [If interested in my review of South Beach, click here.]

Last time I checked, the Ornish diet was vegetarian.

Hundreds of thousands of people have been eating low-carb/Atkins-style for years.  What we need is a study comparing their long-term health and longevity outcomes with a group of people following a standard American diet.

The United States’ National Institutes of Health should fund it.

I don’t care if low-carb eating outperforms the Mediterranean, DASH, Ornish, paleo, vegetarian, Dr. Siegal’s Cookie Diet, or any other diet.  Actually, I do care; I want to know.  My ego can easily handle it if Atkins beats Mediterranean.

Just show me the data.

Steve Parker, M.D.

References:

Miller, Michael, et al.  Comparative Effects of Three Popular Diets on Lipids, Endothelial Function, and C-Reactive Protein During Weight Maintenance.  Journal of the American Dietetic Association, 109 (2009): 713-717.

Moens, A.L., et al.  Flow-mediated vasodilation: a diagnostic instrument, or an experimental tool?  Chest, 127 (2005): 2,254-2,263.

Diet fads come and go, by definition.

Out of curiosity, I searched “detox weight loss” at Amazon.com and found the following book titles:

• The Weekend Weight-Loss Plan: Detox, Shape Up and Feel Fabulous
• Dr. Deal’s Delicious Detox Diet Weight Loss Wellness Lifestyle
• Daniel’s Diet: The 10-day Detox and Weight Loss Plan
• The 21-Day Detox Weight-Loss Plan: Lose Weight and Restore Your Health the Safe and Effective Way!
• Rice Diet Solution - World-famous Low-sodium, Good-carb, Detox Diet For Quick And Lasting Weight
• Collins Gem 48-Hour Detox: Quick-Fix Plans for Healthy Weight Loss
• The Great American Detox Diet: 8 Weeks to Weight Loss and Well-Being
• The Raw Food Detox Diet: The Five-Step Plan for Vibrant Health and Maximum Weight Loss
• The Maker’s Diet for Weight Loss: 16-week strategy for burning fat, cleansing toxins, and living a healthier life!
• The Fat Elimination and Detox Program (FED): A Holistic Approach to Disease Prevention and Weight Loss
• The Great American Detox Diet: The Proven 8-week Programme for Weight Loss, Good Health and Well Being - As Featured in the Hit Movie “Super Size Me”
• DETOX ! The Ultimate Guide to Detoxification! Fact, Fiction and Everything You Should Know to Detoxify Your Body the Right Way for Maximum Energy, Health and Weight Loss!
• The Fast Track Detox Diet: Boost metabolism, get rid of fattening toxins, jump-start weight loss and keep the pounds off for good
• 21 Pounds in 21 Days: The Martha’s Vineyard Diet Detox
• Food Combining 2-Day Detox: Beat Weight Gain & Protect Your Health the All Natural Way
• The Fast Track One-Day Detox Diet: Boost metabolism, get rid of fattening toxins, lose up to 8 pounds overnight and keep it off for good

Sixteen titles, four exclamation marks.  And that’s probably not all.

“Detox” peaked in 2006.

I give it another year, which is a good run for a gimmick.  Just wish I could predict the next blockbuster buzzword.

Steve Parker, M.D.

Update April 20, 2009

One more published this month:  Detox 4 Women: An All New Approach for a Sleek Body and Radiant Health in 4 Weeks!