Thursday, March 11, 2010

Paleo Diet pH: Does It Matter? Part V

Researchers who have discovered the unusual early onset and accelerated progression of bone loss in Eskimos have explored various possible explanations for it.  Basically, we can simply look at all the factors we know may affect bone mass accrual or release and evaluate the Eskimo diet and lifestyle to see if any one or several of those factors stands out as an influence strong enough to produce the observed bone loss.


Bones lose mineral content when not subjected to loads.  Primitive Eskimos led vigorous lives of hunting, kayaking, walking through snow, carrying loads of food, fuel, or children, so it seems unlikely that inactivity caused bone losses found in precontact Eskimos.  

Eskimos compared to Pueblos had more evidence of bone demineralization activity.  It seems unlikely that the Pueblo life in the high desert had higher bone loads than Eskimos.  Walking through the desert is less strenuous than walking in snow with snow shoes and heavy clothing, carrying loads of meat probably is more strenuous than picking corn cobs.

In the 1970s, modern Eskimos getting about 50% of subsistence from hunted wild foods had more demineralization than modern Wisconsin whites who got most if not all of their foods by driving automobiles to supermarkets.  Very unlikely that the whites had higher intensity activity overall.


Vitamin A – Modern population data indicates that excessive vitamin A intake relative to vitamin D and K levels may promote osteoporosis.  Eskimos who ate animal livers frequently, particularly liver from species high on the food chain (e.g. polar bears), may have ingested levels of vitamin A that would promote osteoporosis.

Vitamin D – Eskimos had extensive summer sun exposure with sufficient skin exposure to generate some vitamin D.  Those that ate marine animals would also have gotten large doses of vitamin D from fish and sea mammals.   Weston Price reported Eskimos getting at least 10 times the vitamin D found in modern diets.  Lack of vitamin D would have caused rickets and poor craniofacial development, neither of which appear to have occurred in either precontact Eskimos or those Weston Price found still eating native diets.  The Wainwright Eskimos studied by Mazess got about 50% of their energy from wild foods including fish supplying vitamin D, whereas the whites having higher bone mineral content ate none of those foods.  Vitamin D deficiency seems an unlikely cause of Eskimo age-related bone loss.

Vitamin K – Inadequate vitamin K, particularly K2, results in loss of bone matrix.  Eskimos who ate ruminant animal fats and livers from caribou would very likely have gotten adequate vitamin K2.  This would seem confirmed by the excellent craniofacial development of Eskimos studied by Weston Price, since lack of vitamin K2 would have resulted in poor craniofacial development.

Vitamin B-12 – Deficiency of vitamin B-12 in pregnancy and childhood appears to impair bone mineralization in youth.  Eskimos ate plenty of meat and fish rich in vitamin B-12.

Vitamin C – Deficiency of vitamin C would impair formation and repair of the cartilage bone matrix.  Eskimos apparently did not suffer from vitamin C deficiency severe enough to impair bone growth during development since most reports including Weston Price have found Eskimos have normal bone growth and development in youth.  However, their vitamin C intake fell far short of the 400-500 mg estimated daily intake of an equatorial hunter-gatherer,  which is only a fraction of the the intake a comparably sized non-human primate would consume from a wild diet.  I consider it possible that their vitamin C intake was suboptimal for maintaining bone matrix during aging.  

Mineral Intake

Boron – Modern studies have found higher resistance associated with lower risk of osteoporotic bone fractures among people consuming higher amounts of boron.  Fruits, vegetables, and nuts provide most of the boron in modern diets. I have not yet found information of the boron content of fish and meat, the main mineral sources in the Eskimo diet.

Calcium – Eskimos consumed calcium in meat and bones, primarily fish bones.  They had an estimated variable intake ranging from 500 mg to 2000 mg daily.   Since in most studies including the reports of Weston Price it appears that they developed and maintained normal bones and bone density comparable to non-Eskimos up to the third decade of life, it does not seem likely that calcium deficiency caused their early onset osteoporosis.

Copper – Copper supports collagen formation for bone structure and deficiency results in decreased bone formation and bone deformities, and increases loss of calcium from bone.   In the Eskimo diet, meat, particularly organ meats, would have supplied copper.  Since Price found isolated Eskimos had no bone deformities, precontact Eskimos probably had adequate copper intake for development. 

Magnesium – The bones contain 60 percent of the magnesium in the body.  Rude et al found that dietary magnesium deficiency induces bone loss, decrease in osteoblasts, and an increase in osteoclasts in rats maintained at levels of 10%, 25%, and 50% of recommended intakes.  They also found that magnesium restriction in humans induces changes that would promote osteoporosis.  Of foods eaten by Eskimos, meat does not supply much magnesium but seafoods and kelp do.  Nuts and green leafy vegetables supply high amounts.  

Manganese – Manganese plays a role in collagen formation and is required for normal bone formation and development.  

Phosphorus – Calcium phosphate forms the primary component of bones and teeth. Phosphate deficiency would result in failure to form normal bones.   Humans rarely experience phosphorus deficiency, and since meat and fish supply plenty of phosphorus, Eskimos would not have experienced phosphate deficiency unless during starvation. 

On the other hand,  studies on humans consuming diets with a low calcium:phosphorus ratio produce elevations of parathyroid hormone and urinary calcium (see for example Kemi et al  or Calvo).
An Eskimo Analogue Diet

To get some idea of the nutrient profile of a precontact Eskimo diet, I created an Eskimo analogue diet and subjected it to nutrition analysis.  I composed it of sardines with edible bones, venison (as an alternative to caribou, not in the database), kelp (8 ounces) and animal fat.  About 50% of the protein comes from the sardines, and the rest from venison.  The diet supplies 3000 kcal, with 21% of calories from protein and 77% from fat, approximately the ratio recommended by Stefansson.  

This slide displays the macronutrient values and ratio of this food selection:

This slide gives the micronutrient analysis in tabular form:

And this slide gives the micronutrient analysis in graphic form:

 Using this selection of foods, this diet has adequate calcium but shows significant deficiencies of vitamin A, vitamin C, manganese, potassium, and thiamin.  I know that the Eskimos met their vitamin A needs by eating liver regularly.  They obtained vitamin C from adrenal glands, and the need for vitamin C is probably reduced by low carbohydrate intake.  The manganese content only 36% of recommended levels.  As noted above, this deficiencies could have an adverse effect on bones.  

Absent the kelp, this diet provides only 16% of the recommended intake of manganese.  This means that increased kelp consumption would provide the most efficient way to increase manganese intake without increasing protein intake.  Since one-half pound of kelp provides 20% of the daily requirement for manganese, to reach 80% of the recommended level would require adding more than one pound of kelp to the diet, for a total of more than 1.5 pounds of kelp daily.  That seems possible and would also increase intake of other minerals including calcium.  It would fit with the mineral intake of Eskimos reported by Weston Price.

Including the half-pound of kelp, the diet also provides early 1500 mg of calcium and about 2100 mg of phosphorus.  Assuming precontact Eskimos ate like this, the diet does not have a low calcium to phosphorus ratio, so this would probably exclude a low calcium:phosphorus ratio in the precontact Eskimo diet as a factor to their bone condition.  


Thus it seems unlikely to me that isolated coastal Eskimos who ate kelp suffered from any activity or mineral deficiency that could account for their unusual bone metabolism and mineral loss with age.  Inland Eskimos who ate less fish and kelp may have had a more difficult time getting adequate calcium and magnesium,  since without the kelp this diet would have only 42% of the recommended level of magnesium and about 400 mg less calcium, and without the sardines the calcium intake would fall very low.

It does seem possible that excessive vitamin A could have played a role, at least in some cases where people consumed liver from certain species.  Subacute deficiency of vitamin C may also have played a role.   On page 72 of NPD, Dr. Price summarized the omnivorous Eskimo diet:  "Eskimos were able to provide their bodies with all the mineral and vitamin requirements from sea foods, stored greens and berries and plants from the sea."  I don't know what other greens or berries the Eskimos ate, but they would have increased the vitamin C content as well.  Thus, I do not feel certain that either of these can completely account for the degree of bone mineral loss observed in Eskimos.

I do want to emphasize the importance of kelp in this menu.  If you take it out of the outlined diet, this makes the diet deficient in vitamin E, magnesium, manganese, potassium, and thiamin.  It appears that by Weston Price's report, high nutrient density plant foods (seaweeds, greens, and berries) played an important nutritive role in the Eskimo diet.

Addendum 3/13/2010:

Shortly after I posted this I noticed that I did not feel confident in my supposition that Eskimos ate the amount of kelp I proposed in my analogue diet, but I did not have time to edit it.  This reminded me that when I posted on the Masai Use of Herbs, in response to a comment I stated that I would post on plant foods used by Eskimos.  Then I received comments from Stephan and Tom (below) that expressed the same doubt.  I got ahold of two old reports on plants used as food by Eskimos, and from these two papers it appears that coastal Eskimos did not use kelp in anywhere near the quantity I thought possible.  It appears that they actually used more of land plants than seaweeds.

So in my next post on this subject I will present a revised Eskimo analogue diet.  Suffice it to say the without kelp, the Eskimo diet has multiple mineral deficiencies that could promote osteoporosis.


Greg said...

Great analysis - what tool are you using for that and generating the graphs?

Stephan said...

Hi Don,

Hmm, I skeptical that the Inuit would have eaten half a pound of kelp a day. It seems like it would be difficult to procure that when the ocean is frozen over for most of the year. Unless they dried it, which is plausible. I don't recall Stefansson mentioning seaweed in his accounts. It also seems like that amount of kelp would contain an excessive level of iodine.

I wanted to mention a minor matter of interpretation with Price's findings. He stated that the Inuit diet contained 10X the fat-soluble and more than twice the minerals of the US food of his day. But it's important to keep in mind that the typical US diet in the 1920s-1930s was really bad, in some ways worse than it is today. The poor in particular ate practically nothing but refined foods, with very few sources of fat-soluble vitamins.

I'm in the process of thinking through the acid-base hypothesis again due to your posts. I may post on it eventually.

Tom Garnett said...

Dear Don,
thank you so much for going into this level of detail on topic that has assumed almost iconic status in parts of the paleo community. Epidemiological studies can provide suggestive evidence but are not conclusive. Your detailed study behind some assumptions serves a good corrective function.
I do question one of your assumptions, however. You write, “would require adding more than one pound of kelp to the diet, for a total of more than 1.5 pounds of kelp daily.” I have been eating various sea vegetables for decades. These days I use kombu in soup stock and stew like dishes. 1.5 pounds of kelp is an enormous amount. I assume the 1.5 pounds refers to cooked or wet kelp because 1.5 pounds of dried kelp would yield 8-9 pounds of wet kelp. If you eat even close to 1.5 pounds of kelp or any other sea vegetable in one day or two or three you will probably have severe diarrhea. It will also fill your stomach to such a degree that I doubt you would be able to eat the amounts of meat and fat your analysis indicates they ate. Alginate is used commercially as an appetite suppressant. Also note research that links kelp consumption far less than the amounts you cite cause thyroid suppression “

Don said...

Stephan and Tom,

Yes, soon after I posted this I realized that I did not feel confident that Eskimos would have eaten 8 ounces of kelp daily.

As I pointed out recently I have first-hand experience harvesting seaweeds in the Atlantic. Working with Larch, we harvested hundreds of pounds of seaweed in just a couple of months. Dried out it came to more than 1000 pounds. We did it all by hand, using small water craft, so I think Eskimos could have harvested large amounts of seaweed for year-round use.

My analysis used eight ounces of fresh kelp, which would dry out to about 1 ounce, or ~30 g. Since kelp contains 80+% water, eight ounces fresh would measure about 1 cup, which seems like a possible daily intake. I know that I have eaten ½ cup of seaweed at a meal in certain preparations. It did not prevent me from eating other foods, nor did it give me diarrhea. 30 g dried would be a cinch to eat.

Tom, in the study you linked to, they found TSH elevated by ingestion of 15g or 30g kelp each day supplying 35mg or 70mg iodide. However, they say it exceeded normal limits only in some subjects. Further, they say that T3 and T4 “were slightly decreased within the normal limits. During long term daily ingestion of 15 g of Kombu (55-87 days), the TSH levels were elevated and sustained while the FT(4) and FT(3) levels were almost unchanged.” They don’t report the long-term effect of 30g daily. Since the T3 and T4 remained in normal limits, and and TSH only exceeded normal limits in some subjects, and Eskimos had no awareness of either T3, T4, or TSH levels, it seems unlikely that Eskimos would have noted any effect to dissuade them from consuming up to 30g kelp each day. I can’t know from this whether even greater doses would produce any noticeable effect. I will note also that some reports find the opposite, namely that high kelp intake stimulates thyroid function.

But both of you make a reasonable point about iodide that I overlooked, probably because the USDA database incredibly does not report the iodide content of kelp, and FitDay does not account it! Consulting several sources, I find that eight ounces of fresh seaweed could provide as much as 240-718 mg of iodide. Price claimed that the Eskimo diet provided 49 times the iodide found in deficient modern diets.

If I assume that the deficient diet supplied only 10% of the 150 mcg RDA for iodide, i.e. 15 mcg, then 49 times that would be 735 mcg. Eight ounces of kelp would exceed this by nearly a factor of 1000.

If I assume the deficient diet supplied 50% of the RDA for iodide, i.e. 75mcg, then 49 times that would give 3.7mg iodide. Most likely this indicates that Eskimos did not consume eight ounces of kelp daily.

I wish Price had given absolute quantities, instead of comparative values, then I wouldn’t have to speculate here. If Eskimos got 735 mcg iodide daily, this would fall below the IOM maximal tolerable intake level of 1000 mcg and be 49 times a very deficient level of 10% of the RDA. If we use 2mg iodide/g kelp (as reported in the study Tom linked to), this would mean Eskimos ate less than one gram of kelp daily.

For simple sake of argument I will assume that Eskimos consumed enough kelp daily to provide a similar intake of iodide to that found in modern Japanese, namely 12mg. This would mean a daily fresh kelp intake of 6g.

With this assumption, the Eskimo diet will appear quite deficient in magnesium etc., leading me to conclude that nutrient deficiencies likely played a role in their bone loss.

I also forgot to comment on boron. I doubt Eskimo diets supplied adequate boron.

Stephan, I'm going through all this to evaluate the acid-base hypothesis also.

Don said...


I use FitDay.

Paul Bergner said...

Nice work. Thanks. We still have the possibility that regular eating of liver provided overly high amounts of vitamin A, low grade chronic excess of dietary retinol has been found possibly associated with osteoporosis. Fatal acute vitamin A toxicity was known to the Inuit, polar bear liver and sled dog liver being the culprits. they would not have identified a slower chronic vitamin A effect on the bones. I remain unconvinced that acid load is associated with illness, probably until we see something happen with measurements after intervention trial. Epidemiology never demonstrates causality. I also question the research that 240+ grams of protein becomes toxic. It contradicts anthropological reports of contemporary hunter gatherer consumption of 2 lbs salmon per capita per day (Fraser River Indians) or 2+ Kg per day of hunted meat (Australian Aborigines. ). Thanks again.

Don said...


Thanks. Re VT-A, see part VI of this series. Also magnesium deficiency seems likely.

Do you have references for those levels of meat intake? Wondering the durations of these intakes? That would be >288g/d for the salmon and >492g/d protein for the Aborigines. I would agree that anyone can tolerate >240g Protein/d for some time (don't know how long), but suspect like anything else this has a limit. Stefansson seemed to strongly believe from experience that Eskimos did not eat that much protein regularly.

Greg said...

Paul, do you have references for fatal vit A consumption in the Inuit? Wikipedia lists only one known case- that of an Antarctic explorer eating the liver of his sled dog in an effort to survive.

Don said...


At least one intervention trial has shown that treatment with potassium bicarbonate (richly supplied by vegetables and fruits) reduces calciuria.

Full text:

Paul Bergner said...

Yes you more than covered the issue in the next post. thanks.

O’Dea K. Marked improvement in carbohydrate
and lipid metabolism in diabetic Australian Aborigines
after temporary reversion to traditional
lifestyle. Diabetes 1984;33:596-603

followed aborigines hunting and gathering, and found that they ate in excess of 2lb of meat (up to 4 lb?) daily.

I had personal communication with a PhD in Seattle who is a salmon fisherman in season, and a Phd anthropologist otherwise, his thesis was on salmon consumption in the Fraser River Indians. He said their consumption was on average 2lb per day throughout the year.

The wikipedia entry on arctic vitamin a poisoning: The liver of certain animals — including the polar bear, seal, walrus[5], and husky — is unsafe to eat because it is extraordinarily high in vitamin A. This danger has long been known to the Inuit and has been recognized by Europeans since at least 1597 when Gerrit de Veer wrote in his diary that, while taking refuge in the winter in Nova Zemlya, he and his men became severely ill after eating polar bear liver.[6] In 1913, Antarctic explorers Douglas Mawson and Xavier Mertz were both poisoned (and Mertz died) from eating the liver of their sled dogs.[7]

I am only assuming that the surviving Inuit found this out the hard way.

I appreciate the work of Fresetto, Morris, Sebestian, but I am unconvinced by their conclusions. They found increased calcuria with acid load, but did not find elevated parathyroid hormone. In a very unscientific statement say the calcium "presumably comes from bone". Calcium metabolism 101: increased calcurea does not = bone loss or bone turnover. See the work of JE Kerstetter over the last ten years. Demonstrates that the hypercalcurea of high protein diet results from increased intestinal absorption, and in studies with radiolabeled calcium show that the amount of calcium in urine from bone source actually decreases as protein in the diet increases. The most recent shows both increased calcium abosrption and retention with high protein.

Thanks again.

Don said...

Hi Paul,

If you read the post I put up today, I found (or rather, Greg directed me to) a paper with some interesting data suggesting traditional Eskimo protein intakes also exceeding 40% of calories.

I question the long-term applicability of paper you cite as indicative of traditional Aborigine practice since it was a temporary intervention returning diabetic (ie. westernized) Aborigines to the bush. They may not have had the traditional knowledge required for finding more fat or CHO rich foods.

Not sure if the figures on the Fraser River natives are for gross or net weight of salmon consumed. Some carcass loss. Nevertheless, the Eskimo data I found does suggest a higher chronic tolerance level for protein.

I'm aware that Kerstetter et al have shown that higher protein than RDA levels of protein increase calcium absorption and don't increase bone resorption, but I have not seen one yet that tested levels above 3 times the protein RDA, whereas it seems Eskimos could have consumed 6-10 times the RDA. Effects could be different. I would expect to find a sweet spot for protein, which from what I have read so far I think might lie in the 3-4 times the RDA range, then diminishing benefit and finally at the very high end detriment, just as with every other nutrient to my knowledge. The question is where does the ceiling (toxic level) lie?

Paul Bergner said...

All those are good points. I agree the "sweet spot" is about 3x the RDA, and a minimum for optimal health is about 2x. As far as acid-alkaline and the health demerits of an acid-load diet, I simply don't see the evidence. The most I can find in the literature is Frasetto, Sebastian, et al, but their hypothesis is only that the aging kidney loses its ability to compensate for acid load. The health effect of acid load from eating meat seems entirely theoretical to me.
We did an experiment in my teaching clinic. A fellow on an intermittent ketogenic diet measured urine acidity for 50 successive urinations are various times of day in a man with controlled insulin resistance. The average urinary pH was almost exactly 6.0. It moved in the direction of alkalinity under 3 situations. 1) During the "alkaline" tide period after a meal, when bicarbonate is dumped to compensate for the H+ ions in the secreted stomach acid (strong effect); after light to moderate aerobic exercise (moderate alkalinizing effect); and the two most alkaline levels were first thing in the AM on night when he spent the night in his lovers bed (but did not engage in sex.) Heavy exercise overwhelmed the respiratory buffer and acidified the urine. There is clearly something happening with acid-alkaline more than the effects of food, and possible compensatory mechanisms we have not considered. These folks walked or jogged around 6-9 miles a day, and may have had enhanced respiratory buffering that no one has considered in their dietary studies. Subsequently in my own measurements, 30-90 of low aerobic level hiking invariably alkalinizes my own urine.
As for kelp, Ryan Drum,PhD, and expert in this field, recommends daily intakes of 1-3 grams.

Don said...


Thanks for sharing the results of your experimentation. I agree that factors other than diet must affect urinary pH. This study found low urinary pH as a feature of metabolic syndrome, and did not find low pH linked to dietary potassium or sulfate, suggesting that metabolic syndrome generates metabolic acidity:

Regarding the Frasseto experiment I linked to

it has a major problem, that by using K-bicarbonate, they could not determine whether the reduction of calciuria resulted from bicarbonate (alkalizing) or potassium. But this study

did dissociate bicarbonate from potassium and did find reduced calciuria and markers of bone resorption, as well as reduction of cortisol levels. PTH/1,25-D didn't change.

Anonymous said...

This is another great article with very useful tips about. I am hoping to pick up more blogging work, too! I like to review my back pain