Soy, Re-dun‏ Part 2: The Nitty-Gritty

This is Part 2 of my two-part post about the article The Soy Ploy, where we get into my defense of the contents of the article. Click here to see my apology (part 1).

“The Soy Ploy” begins by listing the health-problems that soy has been linked to: “malnutrition, digestive problems, thyroid dysfunction, cognitive decline, reproductive disorders, immune system breakdown, and even heart disease and cancer”, saying that this linkage was established in “hundreds of studies”. This is essentially the whole claim, the whole point of the article. It goes on to explain the proposed mechanism for a few of the problems listed, in detail, and why some cultures (like the Japanese), who have consumed soy for a long time don’t seem to have these problems in as great a quantity as you’d expect, but in essence these 7 or 8 claims are the heart of the issue. Let’s address them, and one other point the article raises that seemed objectionable to some, and call it a day.


The claim that soy has been linked to malnutrition is a fairly easy one to support: studies like J Nutr, 49, 527-39 and J Am Vet Med Assoc, 123, 38-9 , (in animal models); JOACS, 1974 Jan;51:161A-170A and Pediatrics 1981;68(3):394-6 (in humans) make pretty clear a pattern of ‘failure to thrive’ and micro-nutrient insufficiency caused by soy. In studies like these, interventions (where one element in particular is changed and everything else is left the same) isolating the effects of any particular element becomes much easier, so the results are more convincing than the often-seen epidemiological studies.

Further, the logic behind this claim is pretty easy to see: as The Soy Ploy says, soy contains phytates and digestion inhibitors (, and also Journal of Food Science. January/February 1984;49(1):199-201 and Journal of Food Composition and Analysis 1989;2:67-68, which I could not find an online source for, sorry), materials that bind to nutrients in food, causing them to be unavailable for digestion. I cite sources… there really is no need though: the presence of these substances (“toxins” is what the article’s author calls them) is established fact, essentially undisputed as far as I can tell.

Digestive Problems

The logic is pretty easy to understand here too: proteins in many beans, including soy, are irritating in the intestines, causing poor digestion (and perhaps some of the malnutrition we just spoke of). It may or may not be of interest to the reader that this effect is quite similar to the effect that the consumption of grains has.

Medicina (B Aires) 1999;59:747-752 (spanish!). Rats fed soy-based chow had reduced growth and an increase in gastrointestinal problems compared to controls.

J Nutr2000;130:2292-2298. Dietary mixtures for pigs, which are carefully formulated to promote reproduction and growth, allow approximately 1 percent of the ration as soy in a diet based on grains and supplements. (Pigs have a digestive system similar to humans.) The Central Soya Company, Inc. website gives a range of 2.5 percent to 17.5 percent soy in the diet of pigs, citing a number of anti-nutritional components that “have been documented to cause gastrointestinal disturbance, intestinal damage, increased disease susceptibility and reduced performance in pigs.”

Food Chem Toxic. 25 (10), 739-45 (link unavailable). Ingestion of soy produced enlarged pancreases in mice and “pancreatic adenoma and carcinoma in the rat.”

(animal models)

J Pediatr Gastroenterol Nutr 1983 May;2(2):271-87 . Soy feeding caused damage to small bowel mucosa in 2 infants. The damage was similar to that of celiac disease and consistent with a lectin-induced toxicity.

Pediatr Res 1981 Sep;15(9):1240-1244. Soy formula fed to premature babies caused in increase in digestive enzymes compared to milk-fed babies, indicating low digestibility of soy formula.

Soy Protein and Human Nutrition, Harold L Wilcke and others, eds, Academic Press, New York, 1979 (It’s a book, publishing the findings of a conference). A group of Central American children suffering from malnutrition was first stabilized and brought into better health by feeding them native foods, including meat and dairy products. Then for a two-week period these traditional foods were replaced by a drink made of soy protein isolate and sugar. All nitrogen taken in and all nitrogen excreted were measured. The researchers found that the children retained nitrogen and that their growth was “adequate,” so the experiment was declared a success. However, the researchers noted that the children vomited “occasionally,” usually after finishing a meal; over half suffered from periods of moderate diarrhea; some had upper respiratory infections; and others suffered from rash and fever. It should be noted that the researchers did not dare to use soy products to help children recover from malnutrition, and were obliged to supplement the soy sugar mixture with nutrients largely absent in soy products, notably vitamins A, D, B12, iron, iodine and zinc.

Thyroid Dysfunction

Perhaps the best supported of all of the claims about soy is that soy causes thyroid problems, especially the auto-immune form of hypothyroidism, known as Hashimoto’s. In this experiment, published in 1939, the researchers found Goiter (thyroid dysfunction) in rats after feeding them soy flour. The rats’ thyroids grew to 4 or 5 times their usual size in seven weeks (they also required iodine supplementation, a deficiency (see “malnutrition”) created by the soy itself).

J Nutr 17 (Jun), 545-55 . That, of course, is just the tip of the iceberg: J Nutr, 22, 43-52, Gann 1976, 67:763-765 and others show other animal models of the same phenomenon. Pediatrics 1959;24;854 (link unavailable), Pediatrics, 24, 752-60, J Am Coll Nutr 1990;9:164-167 , Archives of Disease in Childhood 2004 Jan;89(1):37-40, Biochem Pharmacol 1997 Nov 15; 54:1087-96 , all show very clear links between soy consumption and thyroid dis-regulation of some kind in humans, be it soy formula increasing the likelihood of thyroid problems later in life, or soy eaten later in life inducing thyroid problems or making them worse. I see no need to belabor the point.

Cognitive Decline

Problems with cognition are harder to explain, perhaps, than goiter, reproductive problems or malnourishment. I’m certainly not Trainer enough to claim that I really understand it in detail. I do find it interesting, however, the degree to which this claim falls in line with both personal experience and widely held stereotype. It is perhaps not for nothing that hippies and vegetarians are often depicted as dopy and forgetful.

Brain Res 2000 Mar 17;859(1):123-31. Animals fed diets containing phytoestrogens for 5 weeks had elevated levels of phytoestrogens in the brain and a decrease of brain calcium-binding proteins. Calcium-binding proteins are associated with protection against neurodegenerative diseases.

Plenary Session #8: Cognitive Function, The Third International Soy Symposium, Program, November 1999, page 26 . An ongoing study of Japanese Americans living in Hawaii found a significant statistical relationship between two or more servings of tofu per week and “accelerated brain aging.” Those participants who consumed tofu in mid life had lower cognitive function in late life and a greater incidence of Alzheimer’s and dementia.

Dementia and Geriatric Cognitive Disorders 2008;26(1):50-7 . The study found that those who ate tofu regularly had worse memory than those who did not. The study also found that tempeh consumption increased memory, possibly due to its high levels of folate caused by fermentation.

J Am Geriatr Soc 1998 Jul;46(7):918-20. Women in the higher estrone quartiles had lower performance on two cognitive tests.

Another part of the picture is probably zinc deficiency, which soy induces (also shown in Pediatrics 1981;68(3):394-6, which is quoted above), and the increased uptake of aluminum by the brain that results:

Brain Res 1983;288:393-395 . Zinc deficiency will cause more aluminum to be absorbed into the body in general, and into the brain in particular. Aluminum will be absorbed by competing for binding sites on a zinc-containing ligand. Fluoride and phytates in soy formula will induce zinc deficiency.

Reproductive Disorders

Reproductive disorders are another piece of low-hanging fruit: phytoestrogens are known to interact with steroidal hormone receptors, especially the sex hormone receptors. It’s only logical that these sorts of interactions have consequences for reproductive health.

J Nutr 1954;55:639. Exposure to the phytoestrogen genistein caused significant advancement of the vaginal opening and a decrease in the number of litters born to mice.

BJU Int 2000 Jan;85(1):107-113. Vegetarian women are more likely consume more soy than the general population. Incidence of hypospadias was twice as great in vegetarian mothers than in non-vegetarian mothers. Hypospadias is a birth defect due to interrupted development of the penis.

Canadian Medical Association Journal 1983, May 15;128(10):1197-8. Male children exposed during gestation to diethylstilbesterol (DES), a synthetic estrogen that has effects on animals similar to those of phytoestrogens from soy, had testes smaller than normal on maturation.

Am J Clin Nutr 1994 Sep;60(3):333-340. Six women with regular menstrual cycles were given 60 grams soy protein containing 45 mg isoflavones daily. After one month, all experienced delayed menstruation. Luteinizing hormone and follicle-stimulating hormone were significantly suppressed. The effects were similar to those of tamoxifen, an antiestrogen drug. Regular menstruation did not resume until 3 months following the cessation of soy protein consumption.

Proc Natl Acad Sci USA 2000 Apr 25;97(9):4790-5. Researchers found that flavonoids, especially genistein, can cross the placenta and induce cell changes that lead to infant leukemia.

JAMA 2001 Nov 21;286(19):2402-3 (sorry, no abstract). Although reported in the media as a vindication of soy infant formula, the study actually found that soy-fed infants had more reproductive problems and more asthma as adults.

Immune System Breakdowns

This is another one that I don’t fully understand, but it does seem to have some traction. Apparently the isoflavones genistein and daizedin, found in soy, are suspected of blocking immune system function.

Transplantation 1991 Feb;51(2):448-50. Genistein blocks the production of T cells needed for the immune system. The authors conclude: ” . . . that genistein is a powerful immunosuppressive agent. . .” and suggest that it has a potential use in the treatment of allograft rejection.

Cancer Res 1992 Nov 15;52(22):6200-8. Effects of genistein on the growth and cell cycle progression of normal human lymphocytes and human leukemic MOLT-4 and HL-60 cells…. The results suggest that genistein “is expected to be a strong immunosuppressant.”

Heart Disease and Cancer

The seemingly most controversial claims of the “anti-soy” camp, that soy consumption is potentially a factor in heart disease and cancer are just that: controversial.

I personally remain undecided about soy even as a contributing cause for heart disease: some of the confounding factors I would like to exclude in a study of the relationship – gluten, excessive fructose consumption, other legumes, industrial seed oils, etc – are all endemic in populations that consume soy.

Luckily neither I nor Nourishing Our Children are making causal claims in regards to cancer, heart disease and soy.

First, as to Heart Disease:

Am J Clin Nutr 1999;69:419-25. Many studies have shown that soy consumption can lower serum cholesterol levels. These studies have led to claims that soy can prevent heart disease. However, the theory that high cholesterol levels cause heart disease is becoming more and more untenable. Cholesterol levels are not a good marker for proneness to heart disease. However Lipoprotein(a) or Lp(a), does serve as a good marker for heart disease. This study indicates that soy raises Lp(a), meaning that it is likely to contribute to heart disease.

… that’s all I could find (easily). While I agree with the statement “lipoprotein(a) serves as a good marker for heart disease” and that the “cholesterol causes heart disease” theory is dead as dead can be, I still find the “soy contributes to heart disease” assertion tenuous.

Aaaand Cancer:

Cancer Epidemiol Biomarkers Prev. 2002 Dec;11(12):1674-7. People who consumed 92.5 grams of soy per 1000 Kcal were found to be 2.3 times more likely to be at risk for bladder cancer. The results were calculated to factor in levels of education and cigarette consumption in study participants.

Endocrinology 1978 Nov;103(5):1860-7 . Phytoestrogens (“coumestrol, genistein, and formononetin and the mycotoxins, zearalenone and its reduced derivative, zearalenol”) “translocate the cytoplasmic estrogen receptor and bind to unfilled nuclear estrogen receptors in whole cells. Bound nuclear receptors are then processed in a manner similar to estradiol in a step, which rapidly decreases total cellular estrogen receptors. The phytoestrogens are also biologically active; they can markedly enhance tumor cell proliferation.”

Cancer Epidemiol Biomarkers Prev 1996 Oct;5(10):785-794. Twenty-four normal pre- and postmenopausal white women, ages 30 to 58 were studied for one year. During months 4-9, the women ingested 38 g soy protein isolate containing 38 mg genistein. Seven of the 24 women developed epithelial hyperplasia during the period of soy feeding, a condition that presages breast cancer. The authors noted that “the findings did not support our a priori hypothesis” that soy protected Asian women against breast cancer. “Instead, this pilot study indicates that prolonged consumption of soy protein isolate has a stimulatory effect on the pre-menopausal female breast, characterized by increased secretion of breast fluid, the appearance of hyperplastic epithelial cells and elevated levels of plasma estradiol. These findings are suggestive of an estrogenic stimulus from the isoflavones genistein and diadzein contained in soy protein isolate.”

Environ Health Perspect1997 Apr;105 (Suppl. 3):633-636. Dietary estrogens were found to increase enzymatic activity leading to breast cancer. “Our findings are consistent with a conclusion that dietary estrogens at low concentrations do not act as anti-estrogens, but act like DDT and estradiol to stimulate human breast cancer cells to enter the cell cycle.”

Br J Nutr 2000 Oct;84(4):557-63. Men consuming tofu instead of meat for 4 weeks had lower testosterone-oestradiol ratios as well as changes in other hormone levels. “Thus, replacement of meat protein with soyabean protein, as tofu, may have a minor effect on biologically-active sex hormones which could influence prostate cancer risk.”

American Journal of Clinical Nutrition 2010 Feb;91(2):440-8. “Dietary phytoestrogens may contribute to the risk of colorectal cancer among women and prostate cancer among men.”

In an unexpected twist, soy and cancer are linked in tons and tons of studies, and while I’m still sorta’ unclear about the mechanism here, the link is strong enough, to say the least.


There was another claim in the article that garnered some attention: Isoflavone (phytoestrogen) content in soy and its hormonal effects. The article claims:

10 mg provides the estrogenic equivalent of a contraceptive pill. Thus, the average amount of soy-based formula taken in by a child provides the estrogenic equivalent of at least four birth control pills. Because babies are more vulnerable than adults to the effects of dietary estrogens, the effects could actually be much greater than that of four birth control pills.

Hence the statement, “Babies on soy formula receive the estrogenic equivalent of at least five birth control pills per day.”

The doubts about this claim come from the math, specifically the conversion of the one (isoflavones) into the other (estrogen) in the specified amounts: How did we get to the equation “10mg of isoflavones (in an infant) = 1 contraceptive pill”? Is this, in fact, “black magic argumentation”?

The answer is SCIENCE. Science is how we got there.

You tell ’em, Robb! Yeah!

While I’ll admit that we are talking about estimates (though thoroughly credible ones), it sounds like it was just simple extrapolation upon known quantities:

Bulletin de L’Office Federal de la Santé Publique, No 28, July 20, 1992 (I worked hard trying to find a link to this. I… I just don’t speak Swiss-German. The Journal exists, and it’s online. It existed in 1992… but I couldn’t make heads or tails of it, is really all I can tell you. If you’ve got a little more confidence with your French or German (or your Swiss-German!), you could have a go yourself and help me out). This study is quoted in a lot of places, and referenced in the article itself. The Weston A. Price Foundation summarizes: “The Swiss health service estimates that 100 grams of soy protein provides the estrogenic equivalent of the contraceptive pill.”

Proc Nutr Soc of NZ 1995;20:35-42 (this one doesn’t appear to be online, sorry). “Concerns have been expressed about possible adverse effects, particularly to the foetal-neonatal nervous and reproductive system. Adverse effects may occur by inhibition of the enzyme which converts the relatively impotent estrone to the much more potent oestradiol and by occupying the estrogen receptor resulting in antagonism of the naturally produced oestradiol. Adequate oestradiol is necessary for the imprinting and development of many physical, physiological and behavioral characteristics during the neonatal period and infancy. Infants on soy-based formula have been identified as a high-risk group because the formula is the main source of nutrient, and because of their small size and developmental phase. Infants absorb phytoestrogens and have a calculated daily dietary intake (per kg) 3-6 times that shown to have physiological effects on women. . .”

Of course, this study based its findings on serum measurements of isoflavones, so I may have spoken too soon about “estimates”:

Lancet1997;3530(9070):23-27 (YES! GOT ONE!). “The daily exposure of infants to isoflavones in soy infant formula is 6-11 fold higher on a body weight basis than the dose that has hormonal effects in adults consuming soy foods. Circulating concentrations of isoflavones in the seven infants fed soy-based formula were 12,000-22,000 times higher than plasma oestradiol concentrations in early life, and may be sufficient to exert biological effects, whereas the contribution of isoflavones from breast-milk and cow-milk is negligible.”

Supporting the Lancet finding (if disagreeing about the conclusion): Am J Clin Nutr 1998 Dec;68(6 Suppl):1453S-1461S. Noting the results of an earlier study which found that plasma isoflavone levels in infants fed soy-based formula were 13,000-22,000 higher than concentrations found in fed breast milk or milk-based formula, the authors explain these high levels as due to “. . . reduced intestinal biotransformation, as evidenced by low or undetectable concentrations of equol and other metabolites, and is maintained by constant daily exposure from frequent feeding.” The authors assert that these unnaturally high levels of isoflavones in the bloodstreams of soy-fed children “may have long-term health benefits for hormone-dependent diseases.”

If you’re interested in the subject and willing to put forth the effort you can have a look at the Weston A. Price Foundation’s website: They have a pretty exhaustive collection of references with regards to soy and its health consequences, which was invaluable in writing this post. There’s plenty more there for the interested (or obsessive). Have fun.

So there you have it citizens, Soy Re-Dun. You can be fairly sure that I will not be addressing this again, unless some powerful new information comes to light.

Yoroshu ni



Leave a comment

Filed under Food, Meta

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s