formula

Chole Island (Creative Commons)

Yesterday I read an interesting paper in Human Brain Evolution: The Influence of Freshwater and Marine Food Resources. I have some issues with this book, which is a collection of papers, but there is some great information. One of the interesting chapters is Lessons from Shore-Based Hunter-Gatherer Diets in East Africa. Some of it is available as this paper Milk in the island of Chole [Tanzania] is high in lauric, myristic, arachidonic and docosahexaenoic acids, and low in linoleic acid reconstructed diet of infants born to our ancestors living in tropical coastal regions.

Chole is an island in Tanzania, home to a population that is a mixture of various peoples from the African inlands, the legacy of the Arab slave trade. The paper describes their diet as being coconut, marine fish (which they boil), vegetables, fruits (oranges, mango, and banana), and an occasional flying fox. I do not believe this description is completely accurate. The researchers were looking for a culture that eats close to the "paleolithic diet" as described by Cordain: lean meat, fruits, and vegetables. Unfortunately, this culture does not exist, which leads to the bizarre paradox of using cultures that eat either high-carb, high-fat, or both to bolster the idea that this diet is the best for humans. Later in the paper they use this hypothesis and the data from the people of Chole, to estimate a paleo diet that is nothing like the diet of the people of Chole. They published a separate paper on this, which Don at Primal Wisdom has blogged about. I am more skeptical than Don, as I don't believe the diet we evolved on would be close to the upper limit of the % calories of protein that is the estimate for the max capacity of the liver to convert excess nitrogen to urea (35%)*.  Their estimated ratios are suspiciously close to the zone diet...

I'm not a fan of the method of deciding what is healthy and then trying to fit the ancestral evidence into it, which seems to be their main method. They repeatedly say that staple carbohydrates weren't part of Paleolithic diets, only citing Cordain, who has no evidence for this. I notice they don't hawk low-fat much explicitly, despite their estimated paleolithic diets, since they are working with data from people eating high-fat. But I digress, because I really just wanted to talk about breast milk here and their breast milk data is great. They have data from the people of Chole, three groups of fish-eating controls (Kerewe, Nyakius, and Nyiramba), four groups from the inland (Hadza- who are foragers, Maasai, Songo, and Iraqiw), plus they presented historical data from Dar Es Salaam and several Western countries. 

Here we can see the people from Chole very high amounts of two particular saturated fatty acids: lauric and myristic. The Kerewe have similar levels of myristic and the people of Dar es Salaam have similar levels of lauric. Chole and Dar es Salaam are located in a costal region where palm and coconut trees are abundant. Other places where coconut is eaten frequently like Dominica and Surinam, also had high levels of lauric acid. What about myristic? The authors explain that the Kerewe do not consume coconut, but have a high carbohydrate intake from ugali (a corn/wheat porridge) and muhoho (cassava). They do not explain why other cultures eating a high-carbohydrate diet don't have similar levels or why the levels in the Chole are so high. 

Despite it not being mentioned in the paper, the Chole do eat plenty of carbohydrates (though in what amounts remains to be studied). This book mentions that they grow potatoes, corn, millet, squash, cassava, and rice. This ethnography on storytelling also mentions these crops. Here is a woman in New Scientist, pounding rice: 

Did their culture change all the sudden? Why aren't the authors mentioning these foods? Out of the blue they say that "carbohydrates cause the highest increase in total cholesterol/HDL cholesterol (Mensink et al., 2003), suggesting an atherosclerosis-promoting effect of the carbohydrate-rich diet of Kerewe." I can't find any evidence that the Kerewe suffer from this condition and considering how similar their diet is to the Kitavans, I suspect that they don't have it. 

Apparently you can get lauric and myristic acid from coconut, but there is evidence that carbohydrate-rich diets raise the milk content of medium chain fatty acids as well by de novo synthesis from glucose. I will have to look at the papers cited, but perhaps this mechanism is suppressed somehow in people with excessive linoleic acid in their diets, which would explain why the people of Palestine have low levels, for example. The authors not that in their data set, lauric acid correlates inversely with linoleic acid and positively with DHA and AA. 

As for why this carb to medium chain fatty acid mechanism exists, we get no speculation, but another citation to Cordain for the idea that carbohydrate-rich diets were not part of our dietary habits.

No matter how they got there, medium chain fatty acids in breastmilk seem to be beneficial.  They are easily absorbable as energy (especially since babies are in ketosis no matter what their mother's diet is), and they have antiviral and antimicrobial properties. 

When were coconuts introduced into the human diet? The fact that much of the Paleolithic coastline is underwater and decomposition tends to be rapid in hot humid climates means the fossil evidence is scant. But a silicified coconut fruit was found in the Chinchilla sands in Southern Queensland from 2 million years ago, which suggests that they were widely dispersed even then, since the current origin of the coconut based on genetic studies seems to be East Africa (where humans may have evolved into our modern form) or the American West coast. 

The breastmilk of Chole is very low in linoleic acid and pretty high in DHA, though not as high as in the Kerewe. This is not surprising considered their diet, which is rich in seafood.

How much exogenous DHA is needed for infants for optimum brain and eye development is currently under debate. The authors of this book believe that the DHA needs in infants require the mother to consume seafood, or at least large amounts of DHA-rich organs like brain (though insects also are a source of DHA too). I'm not sure this is true myself, but would be curious to see the Hadza (inland foragers) compared to the people of Chole.

Hilariously, the breast milk of Chole VIOLATES formula recommendations of the Commission Directive of 1991 (I hope the recent ones have been revised), which recommend that lauric and myristic fatty acids be no more than 15%. 90% of Chole samples violate this recommendation. They also are too high in Arachidonic acid, which has a bad rap, but it is important for infant brain development and there is evidence its negative effects only occur when omega-3 is low. There is also evidence that infants cannot create enough of the needed AA from precursors. 

The unfortunate fact is that often guidelines for synthetic substitutes like formulas are based on "normal" women. And considering the health of the normal people in the US, normal might be a bad thing. This paper points out that current data is taken from populations with high levels of degenerative diseases of civilization. According to nutritionism-ists like Marion Nestle, you need a bunch of studies to show DHA is needed in formula. Studies aren't the be-all and end-all, particularly if they are based on populations that are not living optimally. If I were forced to use formula, I'd rather have it be based on the breastmilk of a healthy population than wait decades for a bunch of studies and continue basing it on a bad dataset. 

While I disagree with some assertions in this paper, as they seem to be bent by preconceived notions about macronutrients and the Paleolithic, it is very interesting and points to the need for more studies on populations like that on Chole before vegetable oil is introduced. 

*I am working on a different post on this issue.