However, just this week, Dr. Michael Greger M.D. of NutritionFacts.org released a video review of research on this topic that provides evidence that humans do require some dietary DHA in order to maintain brain structure and function with aging.
These are the sources Greger cites for this video:
- J S O'Brien, E L Sampson. Lipid composition of the normal human brain: gray matter, white matter, and myelin. J Lipid Res. 1965 Oct;6(4):537-44.
- Z S Tan, W S Harris, A S Beiser, R Au, J J Himali, S Debette, A Pikula, C Decarli, P A Wolf, R S Vasan, S J Robins, S Seshadri. Red blood cell ω-3 fatty acid levels and markers of accelerated brain aging. Neurology. 2012 Feb 28;78(9):658-64.
- J V Pottala, K Yaffe, J G Robinson, M A Espeland, R Wallace, W S Harris. Higher RBC EPA + DHA corresponds with larger total brain and hippocampal volumes: WHIMS-MRI study. Neurology. 2014 Feb 4;82(5):435-42.
- E Sydenham, A D Dangour, W S Lim. Omega 3 fatty acid for the prevention of cognitive decline and dementia. Cochrane Database Syst Rev. 2012 Jun 13;(6):CD005379.
- A V Witte, L Kerti, H M Hermannstädter, J B Fiebach, S J Schreiber, J P Schuchardt, A Hahn, A Flöel. Long-chain omega-3 fatty acids improve brain function and structure in older adults. Cereb Cortex. 2014 Nov;24(11):3059-68.
- B Sarter, K S Kelsey, T A Schwartz, W S Harris. Blood docosahexaenoic acid and eicosapentaenoic acid in vegans: Associations with age and gender and effects of an algal-derived omega-3 fatty acid supplement. Clin Nutr. 2015 Apr;34(2):212-8.
- J Fuhrman. Dietary Protocols to Maximize Disease Reversal and Long Term Safety. American Journal of Lifestyle Medicine May 4, 2015.
- P Y Lin, C C Chiu, S Y Huang, K P Su. A meta-analytic review of polyunsaturated fatty acid compositions in dementia. J Clin Psychiatry. 2012 Sep;73(9):1245-54.
- G L Bowman, H H Dodge, N Mattek, A K Barbey, L C Silbert, L Shinto, D B Howieson, J A Kaye, J F Quinn. Plasma omega-3 PUFA and white matter mediated executive decline in older adults. Front Aging Neurosci. 2013 Dec 16;5:92.
- J K Virtanen, D S Siscovick, R N Lemaitre, W T Longstreth, D Spiegelman, E B Rimm, I B King, D Mozaffarian. Circulating omega-3 polyunsaturated fatty acids and subclinical brain abnormalities on MRI in older adults: the Cardiovascular Health Study. J Am Heart Assoc. 2013 Oct 10;2(5):e000305.
- S C Cunnane, J A Schneider, C Tangney, J Tremblay-Mercier, M Fortier, D A Bennett, M C Morris. Plasma and brain fatty acid profiles in mild cognitive impairment and Alzheimer's disease. J Alzheimers Dis. 2012;29(3):691-7.
- E Courchesne, H J Chisum, J Townsend, A Cowles, J Covington, B Egaas, M Harwood, S Hinds, G A Press. Normal brain development and aging: quantitative analysis at in vivo MR imaging in healthy volunteers. Radiology. 2000 Sep;216(3):672-82.
- F A Muskiet, M R Fokkema, A Schaafsma, E R Boersma, M A Crawford. Is docosahexaenoic acid (DHA) essential? Lessons from DHA status regulation, our ancient diet, epidemiology and randomized controlled trials. J Nutr. 2004 Jan;134(1):183-6.
- W S Harris, J V Pottala, S A Varvel, J J Borowski, J N Ward, J P McConnell. Erythrocyte omega-3 fatty acids increase and linoleic acid decreases with age: observations from 160,000 patients. Prostaglandins Leukot Essent Fatty Acids. 2013 Apr;88(4):257-63.
Reference #1 is basic science on human brain lipid content.
Reference #2 reported on a 2012 study of participants in the Framingham Study that found that people with lower DHA levels and red blood cell (RBC) ω-3 index (red blood cell DHA+EPA as percent of total fats) also had lower scores on tests of visual memory, executive function, and abstract thinking: "....lower levels of RBC DHA and EPA in late middle age were associated with markers of accelerated structural and cognitive aging. The MRI finding of lower brain volume represents a change equivalent to approximately 2 years of structural brain aging."
Reference #3 reported in 2014 that women who had a higher ω-3 index at baseline had larger brain volumes 8 years later. Reduced brain volume is an important feature of dementia.
Reference #4 is from the Cochrane Collaboration, which in 2012 reported: "Direct evidence on the effect of omega-3 PUFA on incident dementia is lacking. The available trials showed no benefit of omega-3 PUFA supplementation on cognitive function in cognitively healthy older people." This was published before I published Powered By Plants, and was part of the evidence that led me to conclude that humans have no requirement for dietary EPA or DHA.
Reference #5 reports a 2014 double-blind, placebo-controlled intervention trial that found that supplementing older adults with 2.2 g daily of long-chain omega-3 fats from fish oil improved executive function and "exerted beneficial effects on white matter microstructural integrity and gray matter volume in frontal, temporal, parietal, and limbic areas primarily of the left hemisphere, and on carotid intima media thickness and diastolic blood pressure." In other words, the oils improved brain structure.
Reference #6 found an average red blood cell omega-3 index of 3.7% in vegans. The index was higher in females (ave. 3.9%) than males (ave. 3.5%), consistent with the research I cited in Powered By Plants showing that women convert linolenic acid to EPA and DHA much more efficiently than men. Omega-3 status also improved with age: "The omega-3 index was directly and linearly related with age, with means of 3.5%, 3.6%, 4.0%, and 4.0% for the youngest to the oldest age groups..."
Vegan DHA levels and omega-3 index were non-significantly lower than a group of omnivores, but the vegans were on average 7 years older. Given the observation of age-related increase in omega-3 index in both these vegans and omnivores (see discussion of reference 14 below), this may indicate that vegans had a reduced tissue-accumulation of omega-3s compared to the omnivores who also did not have optimum omega-3 indices. It would be expected that had the omnivores been age-matched (i.e. 7 years older than they were on average) they would have had higher accumulation of omega-3s in their RBCs. Thus, it is possible to interpret this as suggesting that vegans were about 7 years behind the omnivores in tissue accumulation of omega-3s.
Of the vegans, 64% had an RBC omega-3 index below 4%, and 27% were below 3%. Two vegan subjects had an index of greater than 8%, the putative cardioprotective level. The authors attributed this to "the marked metabolic variability possible between individuals." Although the study excluded vegans taking supplements, and included only those vegan for at least one year, the report did not provide data about the range of durations of vegan adherence nor specifics for these two individuals. Perhaps these two individuals with very high n-3 indices were among those with the least duration of vegan experience, the most intake of linolenic acid (e.g. flax oil, but see discussion of reference #11 below), or simply genetically unique in their ability to convert linolenic acid to long-chain omega-3s and perhaps hence more likely to thrive long-term on the diet.
Since in this study, the omega-3 index was not correlated with dietary linolenic acid (ALA) intake, the authors believe that their data suggests "that post-intake processes (e.g., absorption, synthetic enzyme activities, clearance, etc.) are more determinants of omega-3 index than dietary ALA." Put otherwise, these authors are suggesting that genetic factors may likely influence omega-3 index more than intake of plant-source linolenic acid (ALA). In this case, the two individuals with exceptionally high omega-3 indices were just genetically unusual, and the evolved norm was indicated by the 64% who had low indices in the absence of preformed dietary EPA and DHA.
Regarding comparison of the vegans and omnivores, the authors wrote:
Thus the evidence generally indicates that vegans have significantly lower omega-3 levels (50-60% lower) than omnivores. This again tends to support a conclusion that humans may have a dietary requirement for long-chain omega-3s.
In phase 2 of this study, vegans took 243 mg of algae-sourced EPA+DHA. The mean absolute increase in omega-3 index was 1.7% over 4 months. This was comparable to findings from other studies of omega-3 supplementation in omnivores, vegetarians, and vegans.
It is interesting to note that in this study, 243 mg/d of algal EPA+DHA only raised the vegans' omega-3 index to ~5.4%, while in the German study cited, 940 mg/d was required to raise the level above the putatively cardioprotective level of 8%. If an n-3 index of 8% is required for cardioprotection or neuroprotection, this would suggest that the actual requirement for EPA+DHA is more like 1000 mg per day.
The authors suggest that a consistency of epidemiological evidence indicating that low DHA status puts people at higher risk of cognitive decline "suggests that a lifetime DHA insufficiency may put vegans at increased risk for cognitive dysfunction":
They conclude that "a majority of long-term vegans appear to be relatively deficient in DHA and EPA, but whether this leads to adverse health consequences is unclear."
Here I would like to point to data from the Adventist Health Study 2:
The post-2013 research on omega-3 status of vegans suggests that omega-3 deficiencies also contribute to the higher all-case, IHD, cardiovascular, and cancer mortality among female Adventist vegans compared to pesco-vegetarians.
Back to Greger's references, #7 is an article by Joel Fuhrman M.D.. It is behind a paywall of $40 I don't want to expend. According to the abstract of this clinical report, Fuhrman discusses "clinical observations over the last 25 years treating people with a nutritarian diet, which could be either vegan or flexitarian" and reports on "health concerns observed from an unsupplemented vegan diet, specifically the risk of depression and later life dementia in individuals with enhanced need for long-chain omega-3 fatty acids."
Reference #8 is a meta-analysis of 10 articles including 2,280 subjects investigating the role of EPA and DHA in dementia. The reviewers concluded that the evidence supports concluding that n-3 PUFAs play an important role in the pathophysiology of dementia, and that low EPA status may be an early detectible risk factor for cognitive impairment.
Reference #9 reports a study of 86 non-demented older adults (mean age 86 y at baseline) followed over 4 years after measurement of plasma omega-3 fatty acids with annual evaluations of cognitive function. A subset of the subjects also had brain MRI to measure white matter hyperintensity volume. The researchers reported that every 100 µg/ml increase in plasma omega-3 FA was associated with 4 s less change in executive decline per year of aging. "The calculated estimate indicates a 1-year delay in age-dependent executive decline per 100 μg/ml increase in plasma O3PUFA at baseline."
Reference #10 comes from the Cardiovascular Health Study, in which 3660 subjects aged 65 years or more had brain MRIs and blood omega-3 levels measured in 1992-94, and 2313 were rescanned 5 years later. The study found that those in the highest quartile of omega-3 levels had a 40% reduced risk of subclinical brain infarcts. Higher long-chain omega-3 FA content (EPA+DHA) was associated with a better white matter grade, but linolenic acid was "associated only with modestly better sulcal and ventricular grades." The results "support the beneficial effects of fish consumption, the major source of long-chain omega-3 PUFAs, on brain health in later life."
Reference #11 reports a study of plasma and brain tissue samples from subjects with Alzheimer's disease (AD), mild cognitive impairment (MCI), and no cognitive impairment (NCI). Findings include:
"In plasma total phospholipids, the concentration (mg/dl) of palmitate (16 : 0), eicos-apentaenoic acid (20 : 5n-3) and DHA were 30–50% lower while α-linolenate (18 : 3n-3) was 8-fold higher in AD compared to NCI."The alarming finding here is that AD patients had 8 times more plant-source alpha-linolenic acid (ALA) in plasma compared to those with no cognitive impairment. It is unclear whether this is a consequence of the dementia process impairing conversion of ALA to longer-chain omega-3s (disease process causes the accumulation of ALA), or the accumulation of ALA promotes the disease. Perhaps AD patients are individuals who have a more genetically limited ability to convert ALA to longer-chain omega-3s, in which case high dietary ALA might be neurologically toxic to these individuals? The answer is unknown to me at this time.
In brain tissue, "The main difference was a statistically significantly lower % composition of DHA in the AD group, specifically in phosphatidylserine of the mid-frontal (−14%) and superior temporal cortex (−12%)."
Reference #12 reports an evaluation of the utility of magnetic resonance imaging (MRI) in depicting age-related changes in the brain. The authors reported that brain gray matter increases up to the 4th decade of life, then declines gradually afterwards.
Reference #13 is the 2004 paper by Muskiet et al which argued that DHA is essential for humans, based on evidence for weak ability to synthesize it from precursors; epidemiological links to cardiovascular, inflammatory, suboptimal neurodevelopment, and neurological diseases; and randomized trials showing reduced CVD mortality, improved neonatal neurodevelopment, and lower blood pressure with aging. When I wrote Powered By Plants, I relied on evidence indicating that although human synthesis of DHA is weak, it seemed adequate, and I found evidence for suboptimal neurodevelopment to be relatively weak and contradicted by large populations apparently developing normal brains with diets having little or no preformed DHA. At the same time, rates of CVD, inflammatory diseases, neurological diseases, and hypertension appeared lowest among vegans and vegetarians who had little or no dietary DHA. The case for DHA being essential seemed quite weak.
Reference #14 reports a study that sought to establish normal levels of red blood cell omega-3 fats. The authors found that the median omega-3 index was 4.5%, and "across the decades it increased by about 1.5 percentage points' but stabilized after the age of 70. This indicates that it is natural for EPA+DHA to accumulate in RBCs with age.
Note then first that in reference #6, the vegans had an omega-3 index below the median, and, as I mentioned above, despite being on average 7 years (almost a decade) older than the omnivores to which they were compared, their omega-3 index was similar to the omnivores. This suggests the possibility that the vegans were actually depleting EPA+DHA from RBCs after starting their unsupplemented vegan diets, whereas the younger omnivores were still increasing levels.
An Explanation for Vegetarian Adherence and Attrition Rates?
According to Haddad and Tanzman, 65% of self-reported vegetarians actually report eating 38-80 g/d meat, poultry, or fish on dietary recalls, and self-reported vegetarians who do not eat meat of any type probably form only 0.9% of the U.S. population. The Humane Research Council reports that 84% of people who adopt a vegetarian or vegan diet eventually return to eating meat. People often make multiple attempts at this. Why?
According to The Humane Research Council report, only 29% of ex-vegans or vegetarians report experiencing specific health problems on the meat-free diets. Often ex-vegans or ex-vegetarians write books about their experiences. While they often experience some health improvements initially, some also describe a gradual onset of a vague general malaise without any specific health problems.
In Powered By Plants I attributed this attrition to the social conflicts people experience as well as the general tendency for people to believe that meat-eating is essential (part of what Melanie Joy called carnism) so that there exists a tendency for friends, family, physicians, and the vegan/vegetarian him- or her- self to attribute any minor health issue that arises to the absence of meat in the diet.
In other words, I was believing that any perceived need for meat is just a socially constructed ideology, not biology. This is a common undercurrent in so-called "social justice" movements these days: they are founded on the Marxist-Leninist idea that all human behavior is "socially constructed" with no biological basis. I only recently realized that the idea that so-called carnism is a "socially constructed ideology" with no biological basis is also a manifestation of cultural Marxism. I will have more to say about this elsewhere.
This research that indicates depletion of omega-3s in vegans suggests to me that the low rate of adherence to vegan and vegetarian diets, and the high rate of turnover among vegetarians, has this biological basis: slight, gradual depletion of long-chain omega-3 fats from the nervous system leads to a general malaise, just not feeling right and vital, in other words what may be called depression.
This leads me to an evolutionary explanation for the fact that many ex-vegans report a type of euphoria on resuming eating animal products.
If human biology has any evolutionary logic to it, it would make sense that people suffering from depression induced by DHA-deficiency would start craving foods that would provide the missing nutrient: fats, fish, and meats. It also makes sense that these people would feel much, much better upon eating those foods.
The effect could even be an immediate euphoria. Since our ancestors lacked blood tests and intellectual knowledge of nutritional composition of foods, Nature would necessarily have favored the survival of those individuals who could immediately recognize and immensely enjoy a food that supplied a nutrient that they had become deficient in. Those who had nervous systems that could immediately recognize and enjoy the food most needed would be more successful in meeting nutritional needs, and hence in having successful progeny, than those who could not.
This would explain why some ex-vegans and ex-vegetarians report feeling euphoric immediately upon ingesting some type of animal food that would be rich in DHA if it was from wild game (that would include fatty fish, bird eggs, and wild game flesh).
I hypothesize that vegans who take DHA supplements would be more likely to adhere to the diet long-term, and less likely to crave animal products.
Does DHA Deficiency Explain Some Vegan Behavior?
Some individuals may do fine on unsupplemented vegan diets because they have an extraordinary ability to convert plant-source ALA to EPA and DHA. This is part of genetic variation in the human species. These people may not be able to understand anyone who does not thrive on the meat-free diet, and will often use their own experience as "proof" that their version of a meat-free diet is adequate and supplements are unnecessary.
Some of these individuals become angry and aggressive toward anyone who reports not thriving on a vegan or vegetarian diet. Ironically, it is possible that an angry and aggressive attitude is also a sign of essential fatty acid deficiency developing in the brain; speaking from personal experience with individuals suffering from Alzheimer's dementia, delusion, anger and aggression are common symptoms in people who have this disorder.
As indicated above, the depletion of neural n-3 may lead also to declines in cognitive and executive functions in long-term vegans. This could lead to poor judgement, including poor judgement in choosing an alternative to a vegan or vegetarian diet. One may be easily convinced that all fats are good in unlimited amounts, and go from unsupplemented low-fat vegan (frying pan) to a diet high in saturated fat (fire).
These are just some speculations based on my own experience and observations from both sides of the dietary fence (vegan vs. meat-eater).
Evolution of a Dietary Requirement for DHA
In any case, based on the research reviewed above, I am barring further evidence inclined to accept that humans do require a direct dietary DHA source, and advise that vegans take a daily supplement of a minimum of 250 mg of algal-source DHA.
Some people have suggested that if vegans eat a diet with a high n-3:n-6 ratio, they would have no problem producing adequate DHA. I have not seen any controlled studies confirming this hypothesis. In addition, since few terrestrial plants have a high n-3:n-6 ratio, and the vast majority have a very low ratio, in the light of the data from reference #6 above, I find it very implausible that human DHA requirements reflect an ancestry of specific adaptation to some combination of unusual plant foods that would supply a high n-3:n-6 ratio.
Terrestrial plant foods lack long-chain omega-3 fats EPA and DHA. Seaweeds have very little of these, but very high amounts of iodine, so anyone relying only on seaweeds for a minimum of 250 mg daily of DHA would be exposed to toxic levels of iodine.
Hence, if, as it seems from the evidence discussed above, most humans require a direct dietary source of at least 250 mg DHA daily, this suggests that humans do have at least one dietary requirement that indicates specific adaptation to consumption of wild animal foods, perhaps particularly seafoods.
If, as mentioned above, the EPA+DHA requirement is more like 1000 mg daily–the amount needed, according to clinical trials, to raise the n-3 index to the putative cardioprotective level of at least 8%–the case is even stronger.
The data on mortality rates from the Adventist Health Study 2 (discussed above) seems also to support this conclusion, particularly suggesting that, probably due to large drains on EPA+DHA stores during pregnancy and lactation, women have the highest need for direct dietary sources of these nutrients, despite their having a significantly greater ability to convert ALA to DHA.
In Powered By Plants, I noted that an organism will evolve means to increase production, conservation, and retention of a nutrient when its habitual diet is a marginal source of that nutrient. If humans need 250-1000 mg daily of EPA+DHA not only for cardioprotection but also to support the n-3 drain of pregnancy and lactation, it seems likely that women have the ability to synthesize DHA at much higher rates than men, not because this ability in itself is sufficient to meet their DHA needs for pregnancy and lactation, but because ancestral diets did not themselves provide sufficient DHA directly to support women's needs during pregnancy and lactation.
For any non-vegan or macrobiotic dieter reading this blog, Sot et al. have performed a risk-benefit analysis of seafood intake to determine optimal consumption. They report that the optimal seafood consumption to balance nutrient benefits (omega-3, selenium, iodine) with risks (mercury, arsenic, pollutants, etc.) appears to be about 200 g of fatty fish (specifically, swordfish, herring, halibut, salmon, mackerel, sardines) and 50 g lean fish per week.
The decision as to whether to take supplements or eat fish (or other DHA-rich animal products) is a personal one and I recognize individual genetic variations will determine how well any individual does with any diet.
The Ethical Question
The Adventist Health Study 2 indicates that presumably unsupplemented vegan women suffer significantly higher mortality rates than pesco-vegetarian women. Hence, it raises the question of conflict between the interests of all humans, and the interests of fish. I specify all humans, because health of women is a concern of all humans who want to see the human species continue.
Some would call anyone who considers human interests more important than the interests of fish specieists. The implication is that to be a moral person, one must be willing to sacrifice the health or existence of oneself or one's species to "save" members of the other species.
Is that true? Or is anti-speciesist a code word for anti-human, or an ideology that aims at the extermination of the human species because (it is believed) humans are bad ("speciesist"), and all other species are good?
One may argue that there is no ethical question because people can use supplements to obtain the same results as eating fish. That is possible, although not proven; it is an hypothesis, not a fact. We don't have any large scale studies showing that vegan Adventist women who take DHA supplements have the same low mortality and disease rates as Adventist pesco-vegetarian women.
Hence, for some people there will remain what they will consider an ethical dilemma: altruism vs. speciesism.
In The Intelligence Paradox: Why The Intelligent Choice Isn't Always the Smart One, evolutionary psychologist Satoshi Kanazawa notes that dedication to vegetarian diet is without doubt evolutionarily novel:
"Humans are naturally omnivorous, and anyone who eschewed animal protein and ate only vegetables in the ancestral environment, in the face of food scarcity and precariousness of its supply, was not likely to have survived long enough and stayed healthy enough to have left many offspring. So such a person is not likely to have become our ancestor. Anyone who preferentially ate animal protein and fat in the ancestral environment would have been much more likely to live longer and stay healthier. They are therefore much more likely to have become our ancestors." [p. 192-193]In other words, even if (as I argued in Powered By Plants) our ancestors primarily ate and adapted to a plant-based diet, since (as I acknowledged in Powered By Plants) they also hunted animals, those who ate animal products would have had an advantage in survival and reproduction by also eating animal products. I think this is from a biological standpoint indisputable.
Yet, as Kanazawa proves, more intelligent individuals (higher IQ) are more likely to choose to become dedicated vegetarians than less intelligent (lower IQ) individuals. On average those who eat vegetarian diets have a 10 point higher IQ than non-vegetarians. In the UK, one standard deviation (15 point) increase in childhood IQ increases the odds of adult vegetarianism by 37% among women and 48% among men.
Why do people with higher IQs have a higher propensity to try to be dedicated vegetarians? In Kanazawa's view, it is precisely because vegetarianism is evolutionarily novel. People who have high IQs appear to have a fatal attraction to evolutionarily novel behaviors. In his book he provides evidence that people who have high IQs are also more likely to fail to reproduce. He comments:
"And intelligent people–especially intelligent women–have fewer children and are more likely to remain childless for life than less intelligent people. Once again, whether or not to have children is a matter of personal choice, at least in western liberal societies, and it is neither better nor worse to have children than not to have children. Strictly from the perspective of your genes, however, not having children, or having fewer children than you can safely raise to sexual maturity, is the worst thing you can possibly do in your life. You are failing at the most important task in life, the one thing–the most important thing–that you are evolutionarily designed to do...And this is why the intelligent choice isn't always the smart one.
"Reproductive success is the ultimate goal of all living organisms, including all humans. That is what humans are evolutionarily designed to do. It is the meaning of life itself. Voluntary childlessness is therefore the greatest crime against nature, which is why intelligent people do it." [p. 208]