For a long time, there has been a theory in mainstream nutrition that saturated fat causes the body to increase the production of cholesterol which, in turn, increases the risk of heart disease. This theory has not been without its detractors.

A large meta-analysis from the UK released this week caused quite a splash because it found that saturated fat was not significantly associated with heart disease (1). It wasn’t a complete surprise, as a 2010 meta-analysis of prospective observational studies had already produced similar findings (2).

The 2014 meta-analysis produced results for three different types of studies:

1. 32 prospective cohort studies analyzing self-reported dietary intake of fatty acids.

2. 19 prospective cohort studies analyzing blood levels of fatty acids.

3. 27 randomized, controlled trials of various fatty acid supplementation regimens.

The results were fairly consistent in that very few associations were statistically significant.

Among the cohort studies analyzing intakes, total saturated fat had a slight trend towards more cardiovascular disease. The only statistically significant finding was for trans fats increasing the risk of heart disease, while long chain omega-3s (EPA and DHA) were almost significantly associated with a lower risk.

Among the cohort studies looking at blood levels, total saturated fat again had a slight trend towards more cardiovascular disease. When looking at individual types of saturated fats, the common saturated fats found in animal products, palmitic acid and stearic acid, were more strongly associated with heart disease, though still not statistically significant.

Interestingly, the only fatty acids in the blood that were significantly associated with heart disease (all inversely) were margaric acid (a saturated fat found primarily in dairy products), the long chain omega-3 fatty acids (EPA, DPA, and DHA), and, more surprisingly, the long chain omega-6 fatty acid, arachidonic acid.

Arachidonic acid had previously been thought to be a cause of inflammation, and therefore heart disease (though other research has countered that idea, see Omega-6s: Not So Bad?).

In terms of clinical trials, only supplementation with omega-3s and omega-6s were analyzed and didn’t find any statistically significant associations, though EPA and DHA came close to being associated with a lower risk.

There were some errors in the version of the paper I have, but these errors did not alter the conclusion according to the attached notice.

So what should someone think about all of this?

One of my regular readers suggested that if you take a bunch of studies with measurement error and throw them all together, you shouldn’t be surprised that you don’t find anything. Perhaps – I don’t know enough about biostatistcs to be able to assess that sort of thing.

Here is what I believe: The primary saturated fats found in animal products, palmitic and stearic acid, most likely contribute to an increase in cholesterol and an increase in the risk of heart disease for people who have high cholesterol. But what is probably more important is not eating too many calories. Fiber is also probably as important as saturated fat, if not more so (3), because it can transport cholesterol out of your system.

Speaking of saturated fat, Dr. Michael Greger’s latest video on low-carb diets might be relevant, Low Carb Diets and Coronary Blood Flow.

References

1. Rajiv Chowdhury, Samantha Warnakula, Setor Kunutsor, Francesca Crowe, Heather A. Ward, Laura Johnson, Oscar H. Franco, Adam S. Butterworth, Nita G. Forouhi, Simon G. Thompson, Kay-Tee Khaw, Dariush Mozaffarian, John Danesh, Emanuele Di Angelantonio; Association of Dietary, Circulating, and Supplement Fatty Acids With Coronary RiskA Systematic Review and Meta-analysis. Annals of Internal Medicine. 2014 Mar;160(6):398-406. | link

2. Siri-Tarino PW, Sun Q, Hu FB, Krauss RM. Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease. Am J Clin Nutr. 2010 Mar;91(3):535-46. | link

3. Threapleton DE, Greenwood DC, Evans CE, Cleghorn CL, Nykjaer C, Woodhead C, Cade JE, Gale CP, Burley VJ. Dietary fibre intake and risk of cardiovascular disease: systematic review and meta-analysis. BMJ. 2013 Dec 19;347:f6879. doi: 10.1136/bmj.f6879. | link

Here’s something I’ve been mentioning in my talks lately. A 2000 study measured the percentage of calories as fat in the diet as well as cholesterol levels in a subset of EPIC-Oxford vegan and non-vegan men. Here is what they found:

Vegans ate 30% of their calories as fat compared to 34% for meat-eaters. Not a huge difference and many people would be horrified at such a high fat intake on behalf of the vegans. Yet, their cholesterol levels were well below what is commonly considered the danger zone.

What accounts for this? The vegans’ much lower intake of saturated fat probably explains a lot. But the vegans’ 20% fewer calories also probably accounts for much of their lower cholesterol levels. Addendum 1/27/12: Additionally, higher fiber intake and zero or near-zero cholesterol intake all likely contribute to the vegans’ lower cholesterol levels. Clarification 1/30/12: People who exercise a lot can eat more calories without cholesterol levels increasing as long as they are not eating so much that they gain body fat (thanks, Ginny).

Some people might point out that ideal cholesterol levels are actually less than 150 mg/dl, so 158 mg/dl is too high. While many clinical trials in people with heart disease (and normally on cholesterol-lowering medication) show a benefit to getting levels below 150 mg/dl, I have not seen evidence that this is ideal, or even desirable in people without diagnosed heart disease or normally high cholesterol. Instead, the observational studies I’ve seen measuring cholesterol levels and mortality have not shown a benefit from cholesterol levels less than 160 mg/dl.

I would not completely rule out the idea that studies have not shown reduced mortality in people with cholesterol levels less than 160 mg/dl because they have not included enough people with cholesterol levels that low and who do not have such low levels due to undiagnosed disease. But “not completely ruling out something” is a far cry from saying there is good evidence that it is true.

The reason I think this is particularly important is anecdotal evidence that people on long-term, low-fat diets can find them hard to stick with. I know there are some exceptions – people who find them easy to stick with, but I sense that there are more who find it difficult. When people crave meat, they tend to think they are craving the protein. But meat is also about 50% fat on average and it would not surprise me if such people are often craving fat as much or more than protein. Eating a diet closer to 30% fat might prevent such cravings.

Yes, lots of qualifiers above that I’m not 100% certain of everything I’m saying. But I think there is enough evidence that I should share it with readers rather than just keeping it to myself until “further studies” are done.

Reference

Allen NE, Appleby PN, Davey GK, Key TJ. Hormones and diet: low insulin-like growth factor-I but normal bioavailable androgens in vegan men. Br J Cancer. 2000 Jul;83(1):95-7. Link

My October 22 post about Doug Graham’s B12 claims garnered a lot of comments. Among them was one suggesting that I am alarmist at times. So, it is with hesitation that I report the following.

Background: If you are not familiar with omega-3 fatty acids, some of the conversation below might not make much sense. See Omega-3 Fatty Acid Recommendations for Vegetarians for background.

DHA in Elderly Vegan Men

I have been in dialogue with Dr. William Harris about DHA. Dr. Harris has been vegan for many decades and will be 80 years old this December. He has been concerned about making sure he has enough DHA, but in the past when he took DHA, he started bruising very easily. A more recent report from him is that he has been trying DHA again and the bruising has not reappeared.

Dr. Harris cc’d me on a discussion he was having with Dr. Joel Fuhrman and this led me to find out from Dr. Fuhrman that he has been seeing numerous elderly vegans with severe DHA deficiency, and he believes it may have exacerbated Parkinson’s disease and tremors in some of his patients. Upon more questioning, Dr. Furhman had the following to say:

“I have seen thousands of vegan patients, raw foodists, natural hygienists, McDougall and Ornish participants, as well as my own ‘nutritarian clients’ over the last 20 years. I test B12 on everyone, of course we are not talking about B12 [deficiency in regards to the patients with Parkinson’s and tremors], these individuals were well-educated about B12. I have seen some paralysis and other major B12 problems in hygienists and vegan raw foodists. Some that even died from hyperhomocysteine resulting from severe B12 deficiency. I have also seen vegans with balance and ambulation issues with B12 deficiency, unable to walk. One raw foodist who came to see me with this problem, who could not walk, made almost a complete recovery after B12 supplements and then he announced on his radio show that he recovered from M.S. with a raw food diet. ”

“Many of the visits were initiated by complaints. Many people who started or adopted vegan diets went back to eating meat after suffering from fatty acid deficiency symptoms from not eating sufficient seeds and nuts. I have performed fatty acid tests, B12, MMA, amino acid profiles and others on many people. I have seen significant DHA and EPA deficiencies even in middle aged women, but the most predictable pattern is the dramatically low levels in elderly vegan men. I do feel to err on the side of caution, either a blood test to confirm adequacy or a low dose of DHA is indicated, and, as was discussed, you do not need very much [200 – 300 mg DHA per day for one month] to fix the blood test findings.”

Because of the above conversation, I have tweaked my DHA recommendations for vegans, emphasizing that elderly vegans need to take more:

    Under 60 years old: 200 – 300 mg every 2-3 days
    60+ years old, pregnancy, or breastfeeding: 200 – 300 mg per day

This amount may be somewhat more than necessary, but until we know what level can sustain DHA levels long term, it seems like the most prudent amount. This is based both on what Dr. Fuhrman says above, as well as a 2003 study that showed blood levels of DHA to increase 48% in vegans taking 200 mg per day for 3 months (1).

Vegans Convert DHA Better than Fish Eaters

In other DHA news, a study from EPIC-Norfolk recently came out showing that while vegans have lower levels of DHA in their blood, they are more efficient at converting ALA to DHA than people who eat fish (2). This is not surprising, as an abstract by the same lead author was published in 2008 finding the same thing. You can see the EPA and DHA levels in Table 4 of Omega-3 Fatty Acid Recommendations for Vegetarians.

There were only 5 vegan men and 5 vegan women in this study. Despite the higher conversion rate, the vegan men still had significantly lower DHA levels than the fish-eaters. However, the vegan women actually had the highest DHA levels of any diet group (although the standard deviations was quite large indicating that some of the women had very high levels and some had very low). The authors did not address this unusual finding.

Omega-3 Lab Tests

If you are interested in getting your DHA levels tested, Dr. Harris has compiled a list of three labs he was able to find that test them. Dr. Harris was only completely confident in the results from Mayo Clinic.

1. Mayo Labs – $394.60 for 29 different fatty acids including LA, AA, ALA, EPA, and DHA

2. MetaMetrix – $206 for 7 fatty acids

3. Genova – $188.65 for 4 Omega-3 and 6 Omega-6 fatty acids

I am not suggesting that all vegans need to get their DHA levels tested and I do not know anything further about these tests. I am just providing them for people who might be interested.

References

1. Lloyd-Wright Z, Preston R, Gray R, Key TJA, Sanders TAB. Randomized placebo controlled trial of a daily intake of 200 mg docosahexaenoic acid in vegans. Abstracts of Original Communications. Proceedings of the Nutrition Society 2003:42a. (No link available.)

2. Welch AA, Shakya-Shrestha S, Lentjes MA, Wareham NJ, Khaw KT. Dietary intake and status of n-3 polyunsaturated fatty acids in a population of fish-eating and non-fish-eating meat-eaters, vegetarians, and vegans and the precursor-product ratio of alpha-linolenic acid to long-chain n-3 polyunsaturated fatty acids: results from the EPIC-Norfolk cohort. Am J Clin Nutr. 2010 Nov;92(5):1040-51. Link

Every once in awhile I check in with the trans fats research, to see if they are living up to their bad reputation.

Some background:

Fats contain long chains of carbon molecules. Each carbon molecule has positions (or bonds) at which a hydrogen can be attached. If all the potential hydrogen bonds on all the carbons are filled, the fat is saturated. Monounsaturated fats have one open bond, while polyunsaturated fats have two or more. The more open bonds, the less stable (and more liquid) a fat is.

Hydrogenation is the process of adding hydrogen molecules to an unsaturated fat’s open bonds. If you take a batch of unsaturated fats and add hydrogen molecules to much of it, it will make the batch more stable and improve some of the cooking qualities.

If you completely hydrogenate a batch of oil you end up with saturated fat. But if you only partially hydrogenate the batch, you will produce some saturated fats as well as some unsaturated fats with an unusual shape, called trans.

An unsaturated fat normally has a curved, or cis structure (unlike a saturated fat which is straight). But a trans fat is an unsaturated with a structure that is more straight.

When reading food labels, if an ingredient is called “hydrogenated oil,” then it is fully hydrogenated and will contain practically no trans fats. If an oil is listed as “partially hydrogenated” then it will contain trans fats.

The research is clear that trans fats increase LDL cholesterol and lower HDL. They also increase inflammation and harm the walls of blood vessels. The American Heart Association recommends that trans fats be less than 1% of energy intake.

In 2006 a meta-analysis by Mozaffarian et al. (1) was published in the New England Journal of Medicine. It included four prospective studies that calculated the increased risk for coronary heart disease based on substituting 2% of carbohydrate calories with trans fats:

A. Nurse’s Health Study (2005) – 33%
B. Health Professionals Follow-up Study (2005) – 26%
C. Alpha-Tocopherol Beta-Carotene Cancer Prevention Study (1997) – 14%
D. Zutphen Elderly Study (2001) – 28%

A & D were statistically significant; C & D were close to statistical significance. The combined results showed an increased risk of 23% (1.23, 1.11 – 1.37).

They also looked at three retrospective studies in which those in the top quarter of trans fat levels in their fat tissues were compared to those in the bottom quarter to measure an increased risk of heart disease:

E. EURAMIC (1995) – minus 3%
F. Costa Rica (2003) – 194%
G. Australia (2004) – 150%

Study E was not even close to statistically significant, while F and G were highly significant.

The pooled risk in all seven studies showed a 29% increase in risk (1.29, 1.11 – 1.49).

Thus, the findings have been pretty consistent (especially for nutrition studies on fatty acids). I looked for other, more recent studies such as these and did not find any.

Interestingly, the only place where trans fats are found naturally are animal products, where they are produced by bacteria in the gut of ruminant animals. In fact, a trans fats in dairy products, conjugated linoleic acid, is touted for its health benefits and sometimes pointed to by people arguing against veganism as necessary for optimal health. Here is what Mozaffarian et al. say about trans fats from ruminant animals:

[D]ietary trials indicate that consumption of conjugated linolenic* acid reduces insulin sensitivity, increases lipid peroxidation, and has mixed effects on markers of inflammation and immune function. Of four prospective studies evaluating the relation between the intake of trans fatty acids from ruminants and the risk of CHD, none identified a significant positive association, whereas three identified nonsignificant trends toward an inverse association….[T]he sum of the current evidence suggests that the public health implications of consuming trans fats from ruminant products are relatively limited.

*This appears to be a typing error and should say “linoleic” (not “linolenic”). The studies they cite are done on conjugated linoleic acid.

In the USA, a food label is allowed to list the trans fat level as zero if it contains less than .5 g per serving, and apparently many foods listed as zero have close to .5 g. So, it’s best to rely on the ingredients list and avoid foods with “partially hydrogenated oil.”

Reference:

1. Mozaffarian D, Katan MB, Ascherio A, Stampfer MJ, Willett WC. N Engl J Med. 2006 Apr 13;354(15):1601-13.

Moloney F, Yeow TP, Mullen A, Nolan JJ, Roche HM. Conjugated linoleic acid supplementation, insulin sensitivity, and lipoprotein metabolism in patients with type 2 diabetes mellitus. Am J Clin Nutr. 2004 Oct;80(4):887-95.

Remig V, Franklin B, Margolis S, Kostas G, Nece T, Street JC. Trans fats in America: a review of their use, consumption, health implications, and regulation. J Am Diet Assoc. 2010 Apr;110(4):585-92.

Risérus U, Smedman A, Basu S, Vessby B. Metabolic effects of conjugated linoleic acid in humans: the Swedish experience.Am J Clin Nutr. 2004 Jun;79(6 Suppl):1146S-1148S.

The Vegan Dietitian is at it again, stirring up controversy with two articles on olive oil:

Olive oil is a healthy addition to vegan diets [article has been removed]

Olive Oil, Health and Advocacy

Another good post by Ginny Messina on low-fat diets. Link.

By Ginny Messina: Fat in Vegan Diets: How Low Should You Go?