Vitamin D in the Infant: Requirements for Safety

Article Summary

  • Infants should receive at least 400 IU of vitamin D per day from all sources to protect against overt vitamin D deficiency.
  • Doses of 2,000 IU per day or more from all sources may protect against type 1 diabetes and other autoimmune disorders.
  • Infants can obtain vitamin D from breast milk, sunshine and supplements such as cod liver oil.
  • Breast milk can be a poor source of vitamin D or an excellent source of vitamin D, depending on the mother’s intake.
  • Mothers who are nursing infants should obtain 4,000 IU of vitamin D per day from all sources.
  • Vitamin D raises the requirement for vitamins A and K. Vitamin D should be taken by mothers or given to infants in the form of cod liver oil, and mothers should consume a diet rich in grass-fed butterfat, cheese, fermented foods, bone broths and grass-fed organ meats to supply vitamin D in a way that is safest and most effective.

Weston A. Price emphasized the special importance of nutrition surrounding the events of marriage and childbirth. The isolated pre-modern groups that Price found to be immune to dental caries and degenerative diseases all prescribed special diets to women–and in some cases men–prior to marriage or conception, to expectant mothers during pregnancy, and to mothers with nursing infants. Price’s studies led him to believe that nutrition during the prenatal, perinatal and postnatal periods was paramount to the formation of children possessing the physical excellence characteristic of the groups he studied.

While much attention is currently focused on research showing that the needs of adults for vitamin D are much greater than once thought, very little information on vitamin D is available to those planning a pregnancy. While official recommendations for both pregnant women and nursing infants are unlikely to protect either the mother or the infant even from overt deficiency, research suggests that infants, like adults, may benefit from intakes of vitamin D much greater than the official recommendations. Accurate information about the role of vitamin D in pregnancy and lactation is therefore critically needed if we are to provide the next generation with the health its members deserve.

Table of Contents
Official Recommendations
The Important Question
Sources of Vitamin D for the Nursing Infant
Vitamin D Requirements of the Nursing Infant
Vitamin D Toxicity in Infants: Linear Growth
Vitamin D Toxicity in Infants: Hypercalcemia
Vitamin D, Atopy, Allergic Rhinitis and Asthma
A Sensible Approach

Rickets & Vegetarianism
Bad Advice
Vitamin D and Type 1 Diabetes
The Vitamin D Study of Wayne Brehm
Lowest Observed Adverse Effect?

Official Recommendations

Expectant mothers-to-be nourish inside them growing, living beings that begin developing their skeletons just eight weeks after conception–a process dependent on vitamin D, calcium and phosphorus supplied by the mothers’ diets. Maternal intake of vitamin D is necessary for proper fetal growth and bone mineralization, and supplies a growing store of vitamin D to the fetus that he or she will depend on as a newborn.1 An adequate store of vitamin D will protect the newborn against tetany, convulsions and heart failure.2

Although the fetal skeleton begins developing early in gestation, it is late in the third trimester when the bulk of fetal skeletal development takes place. The skeleton of an infant born six weeks prematurely contains only half as much calcium as that of a newborn carried to term, which is why infants born prematurely require more calcium than breast milk alone can supply.3

This rapid skeletal development late in pregnancy also requires phosphorus and vitamin D; because of the increased demand of the fetus for vitamin D, the mother’s own stores of vitamin D are depleted over the course of the third trimester.2 For this reason, the American Academy of Pediatrics stated officially in 1963 that “amounts of vitamin D greater than those needed by the normal adult are required for optimal nutrition during the last trimester of pregnancy,” while also lamenting the lack of data about exactly how much more vitamin D is required.3 Likewise, Nicholas Bishop of the University of Sheffield’s Academic Department of Child Health cited evidence in the 2005 edition of the authoritative textbook, Vitamin D, that vitamin D levels are depleted during the last trimester of pregnancy.2

Yet somehow all this knowledge was lost on the U.S. Institute of Medicine’s Food and Nutrition Board in 1997 when it officially declared that the transfer of vitamin D from the mother to the fetus is so small that the mother’s vitamin D status is not affected. Citing a 1978 study showing that the average vitamin D level of pregnant women consuming small amounts of vitamin D at high latitudes was 9.1 ng/mL (just under the level required to protect against overt deficiency) the Institute concluded that “there is no additional need to increase the vitamin D age-related AI [adequate intake] during pregnancy above that required for non-pregnant women.”4 This conclusion is strange, not only because many of the mothers in this study must have had vitamin D levels below the average, but because the average level itself was already deficient.

The Institute did not cite a 1980 study that had been conducted just two years later showing that of 115 Asian women living in London and of 50 of their newborn infants, 36 percent of the women and 32 percent of the infants had no detectable vitamin D in their blood at all.5 Several years after the Institute issued its policy statement, Nicholas Bishop’s research team showed that over 60 percent of infants born to mostly white mothers in the spring and early summer in Sheffield had umbilical cord blood levels of vitamin D under 8 ng/mL.2

It is on rather dubious grounds, then, that the Institute of Medicine recommends a daily intake of 200 IU for pregnant women, which it supposes “may actually represent an overestimate of true biological need.”4 Until the 1997 publication of that recommendation, the Committee on Nutrition of the American Academy of Pediatrics recommended a daily intake of 400 IU, which they considered “adequate to provide for added demands during the second half of pregnancy and during lactation.”3 In 2003, however, the Academy’s Committee on Nutrition issued a joint statement with the Academy’s Section on Breastfeeding and with Dr. Lawrence Gartner and Dr. Frank Greer, in which they overturned the 40-year position of the Academy in favor of adopting the lower so-called “overestimate” of the Institute of Medicine.6 Nicholas Bishop, on the other hand, recommended 400 IU per day during pregnancy in the 2005 edition of Vitamin D.2

All of these recommendations are based on our understanding of vitamin D’s action on the skeletal system. Yet we know from research on adults that when we consider the many other roles of vitamin D, we find substantial evidence that optimal nutrition requires much more vitamin D than that which is required simply to maintain skeletal health.7

The Important Question

The question is, are these increased requirements also true for the developing fetus? Researchers from Norway conducted a small and relatively crude study in 2000 suggesting that maternal use of cod liver oil during pregnancy was associated with a 70 percent reduced risk of the child being diagnosed with type 1 diabetes by the age of 15.8 Unfortunately, the researchers neither took the dose of cod liver oil into account nor had any way of differentiating the effect of vitamin D from the other constituents of cod liver oil. Nevertheless, the study does provide preliminary evidence suggesting that vitamin D may play a much larger role in fetal nutrition than simply providing for growth and skeletal development. Future research may show that the developing fetus, like the adult, needs more vitamin D than previously thought.

Evidence suggests that an adult’s requirement for vitamin D is 10 to 20 times the daily dose recommended by the Institute of Medicine for pregnant women,7,9 and there is certainly no reason to believe that a woman’s need for vitamin D declines when she becomes pregnant. As we will see in the next section, vitamin D nutrition is at least as important for the nursing infant as it is for the developing fetus, and extra vitamin D during pregnancy provides newborn infants with essential stores of the vitamin that they will use in their first weeks of life. But stored vitamin D is not enough for infants–these delicate new beings must also be nourished by high-quality, vitamin D-rich breast milk from well-fed mothers and exposed to adequate sunshine as they face a new world in which their quality of life will be powerfully shaped by the nutritional experience of their first and most important years.

Sources of Vitamin D for the Nursing Infant

Immediately after birth, the direct transfer of the mother’s own nutrition across the placental wall to her baby is no longer an option. The use of stored vitamin D, exposure to the UV-B portion of sunshine when available, and breast milk are the newborn’s three natural methods of obtaining vitamin D. Supplementing with cod liver oil, artificial supplements, or vitamin D-fortified formula can also provide additional sources of vitamin D.

At birth, the vitamin D levels in the mother’s blood are correlated closely with those in the blood of the baby’s umbilical cord.10,11 The mother must therefore maintain optimal levels of vitamin D during pregnancy in order for the newborn to begin life with optimal stores of vitamin D.

These stores, however, are used up quickly. In 1989, researchers followed 39 pregnant women from the beginning of their third trimester until 24 weeks after birth, measuring the vitamin D levels of both the mothers and their infants.12 At birth, the average maternal vitamin D level was 34 ng/mL and the vitamin D levels of the infants were closely correlated with those of their mothers. During the 24 weeks after birth, the infants were exclusively breast fed and did not receive vitamin D supplements. Over the course of the first eight weeks, infant vitamin D levels declined substantially, after which they remained constant for the duration of the study. This suggests that the stores of vitamin D that infants accumulate as fetuses are used up within the first eight weeks of life, after which their vitamin D levels reach equilibrium with what they continue to acquire from sunshine, breast milk or supplements.

Infants, like adults, can only obtain vitamin D from sunshine when there is adequate UV-B available, which varies according to season and latitude. Also like adults, they obtain vitamin D from the sun in amounts proportionate to the surface of skin they have exposed to it. At most latitudes, however, there will be times of the year during which infants will not be able to obtain vitamin D from sunlight. For this reason, the vitamin D levels of infants who were studied in Cincinnati, Ohio, reached optimal levels of 41-45 ng/mL in the summer, but sank to 21 ng/mL in the winter. Infants born during the winter had an average level of only 16 ng/mL one month after birth.13

The Institute of Medicine and the American Academy of Pediatrics consider human milk to be a poor source of vitamin D.4,6 The Institute of Medicine cites research showing that the vitamin D content of a mother’s milk has no effect on the nursing infant’s vitamin D status because it is overwhelmed by the effect of sunlight. A closer look at these studies, however, suggests that the vitamin D content of human milk varies widely with the mother’s diet, and, as one would logically expect, that the relative contributions of sunlight and breast milk to the infant’s vitamin D status vary according to the availability of sunlight and the vitamin D content of the mother’s milk.

On one end of the spectrum, nearly all reported cases of rickets have occurred in infants who were exclusively or primarily breast fed for considerable lengths of time.14,15,16,17,18 In rare and tragic occurrences, the consequences of vitamin D deficiency can be even more extreme: In 1992, medical doctors at Johns Hopkins Hospital in Baltimore, Maryland reported three cases of vitamin D deficiency in exclusively breastfed infants, which led to grand mal seizures, and a fourth in which the infant suffered from fever, cough and vomiting for a single day before dying of cardio-respiratory arrest.17

In most of these severe cases, exclusive breastfeeding of vitamin D-poor milk has not been the only contributing factor leading to disease. In nearly all cases, the infants have been black or dark-skinned.14,15,16,17,18 In a large number of cases, the infants have belonged to religious groups requiring the use of restrictive clothing14,16 or to families practicing vegetarian, vegan and even fruitarian diets,14,15,16,18 and in one case the infant was never taken outside because the mother could not afford to buy him warm clothes.15

While increased skin pigmentation, decreased exposure to sun, religious norms, restrictive diets and poverty have all contributed to the severe cases of infant vitamin D deficiency reported in exclusively breastfed infants, these cases also show that what should be nature’s perfect food for the infant is not always sufficient to supply the infant’s full nutritional needs. Nevertheless, the vitamin D content of human milk is widely variable between individuals, and can be a very important source of vitamin D for the infant if the mother is consuming a vitamin D-rich diet.

In 25 nursing mothers studied in Cincinnati, Ohio who consumed between 0 IU and 700 IU per day of vitamin D, the vitamin D activity of the milk ranged from 30 IU/L to 100 IU/L, correlating with the maternal vitamin D intake.19 In 45 nursing mothers studied in Tampere, Finland, who supplemented with either 0 IU, 1000 IU or 2000 IU per day of vitamin D3, the vitamin D activity of the milk ranged from 14 IU/L to just under 500 IU/L, depending on the vitamin D intake of the mother, the available sunshine, and the individual response of the mother to vitamin D supplementation.20

In the Cincinnati study, the majority of mothers were consuming vitamin D2, the vegetarian form of vitamin D, from prenatal vitamins. It is possible that the vitamin D concentrations of the milk would have been higher if the mothers had been consuming only vitamin D3, the animal form of vitamin D. Vitamin D is carried into breast milk attached to the vitamin D-binding protein;21 since one study found vitamin D2 to have a lower affinity than vitamin D3 for the vitamin D-binding protein, it may be that vitamin D3 is much more effective than vitamin D2 at raising the levels of vitamin D in milk. The group that published the Finnish study, in fact, uncovered preliminary evidence indicating that this is, in fact, the case.22

Interestingly, vitamin D from breast milk may act more effectively than vitamin D from foods and supplements. The primary form of vitamin D in milk is the semi-activated 25(OH)D, and there is twice the proportion of the fully activated calcitriol in milk than is usually found in the blood. When infants nurse on fresh milk, they receive vitamin D that is bound to water-soluble proteins that may enhance its absorption. It is only after milk settles for several days that its vitamin D transfers to the cream.21

The relative contributions of sunlight and breast milk to the infant’s vitamin D status depend on the availability of sunlight and the vitamin D content of the mother’s milk. Researchers studying a group of 61 mothers and their exclusively breastfed infants over the course of a year in Cincinnati, Ohio, found that sunshine was highly correlated to the infant’s vitamin D status. When the mothers’ vitamin D levels were higher than 35 ng/ml, however, the correlation was substantially reduced, and sunlight could account for only half of the infant’s vitamin D status. In this study, the mothers were consuming an average of 500 IU per day of vitamin D, ranging from 0 IU to 1000 IU.23

A Finnish study of 49 mothers and their breastfed infants showed that between January and April, when sunlight is very limited in Finland, supplementing the mother with 1000 IU of vitamin D was not able to raise the vitamin D levels of the nursing infant. Supplementing the mother with 2000 IU of vitamin D, on the other hand, was just as effective at raising the vitamin D levels of the infant as was supplementing the infant directly with the recommended 400 IU.11

Clearly, the conclusion that breast milk is not an important source of vitamin D can only be derived from the fact that most mothers in these studies are not consuming enough vitamin D to produce vitamin D-rich breast milk. Just as we seek to drink milk from cows grazing on pasture grown on rich soil because it is more nutritious, so should we seek to provide nursing mothers with the education and resources needed to nurture their children with the same type of high-quality milk.

A final source of vitamin D for the newborn is the various forms of vitamin D supplements. The American Academy of Pediatrics recommends that all infants receive a vitamin D supplement of 200 IU per day. Since all formulas available in the United States are supplemented with vitamin D, this recommendation in practical terms applies only to breastfed infants.6 Supplementation of infants with vitamin D is not an exclusively modern concept. In Europe, supplementation of infants with a teaspoon of cod liver oil was a traditional folk remedy used since the 1700s, which was meant to insure the infant would thrive.9 Such supplementation is essential if there is any doubt at all about the mother’s vitamin D status.

Vitamin D Requirements of the Nursing Infant

The Institute of Medicine’s recommendation of 200 IU per day is based on the lowest dietary intake of vitamin D associated with an average serum vitamin D level of 11 ng/mL, which is sufficient to protect the infant from severe disorders of the skeletal system such as rickets. The Institute claims that this figure assumes that the infant receives no vitamin D from sunlight and is “multiplied by a safety factor of 100 percent” (presumably meaning that it is increased by 100 percent) so that it “may actually represent an overestimate of true biological need.”

There are several problems with the Institute’s perspective. First, if the intake from which the recommendation is derived results in an average vitamin D level at the exact border of deficiency, by definition some infants with this intake will fall below the average and therefore within the range of deficiency. Second, the studies showing the lowest dietary intake necessary to guarantee minimum sufficiency are likely to be the ones in which exposure to sunlight was most likely to be a confounding variable. For example, if one study shows 400 IU to be necessary to achieve minimum sufficiency and another study at a more southern latitude shows 100 IU to be necessary to achieve minimum sufficiency, it is reasonable to suspect that sunlight supplied an additional 300 IU to infants in the latter study that it did not supply to infants in the former study.

The Institute cites several studies conducted in northern and southern China that illustrate this principle. In northern China, 30 percent of infants supplemented with 200 IU per day had deficient vitamin D status, while only 6 percent of those supplemented with 400 IU had deficient vitamin D status. In southern China, by contrast, 100 IU was sufficient to protect all infants from overt deficiency. Thus, the Institute concludes that 100 IU, being the lowest level associated with sufficiency, increased by a safety factor to 200 IU represents “an overestimate of true biological need,” even though its own data show that in some regions it is insufficient to protect 30 percent of the infant population from deficiency, and that even double its recommendation is incapable of guaranteeing sufficiency for everyone.

It is therefore nothing less than perplexing that the American Academy of Pediatrics chose to replace its long-standing recommendation of 400 IU, meant to provide a margin of protection for uncertainty in requirements, with the Institute of Medicine’s recommendation of 200 IU, which the Institute’s own data shows to be insufficient to protect large numbers of infants from overt deficiency. The Academy’s reduction in the recommendation along with a simultaneous directive to mothers to keep their infants out of sunlight, dress them in protective clothing and cover them in sunscreen is certain to put the majority of infants whose parents follow these recommendations at risk of deficiency.

Vitamin D Toxicity in Infants: Linear Growth

Although emerging evidence suggests that the optimal dose of vitamin D for an infant may be higher than 2000 IU per day, the Institute of Medicine’s upper limit remains set at 1000 IU per day for infants under one year of age,4 which may be less than half the optimal dose (see sidebar “Bad Advice” below). The upper limit was achieved by dividing the “lowest observed adverse effect level” of 1800 IU per day by an “uncertainty factor” of 1.8. In turn, the basis of this lowest observed adverse effect level was a 1938 paper published by P. C. Jeans and Genevieve Stearns of the Iowa State University Department of Pediatrics purporting to find that doses of 1800 IU of vitamin D per day or higher decreased linear growth in infants during the first year of life,25 which had been rightly dismissed as inconclusive by the American Academy of Pediatrics decades before the Institute created its upper limit.3

Jeans and Stearns conducted a small, uncontrolled study of nine infants whose vitamin D status they were unable to quantify. In one of two infants fed 1800 IU per day of vitamin D, the authors observed a slight decrease in growth that could not be distinguished from a random fluctuation and attributed it without evidence to the high dose of vitamin D that the infant was given.

The highly questionable interpretation that Jeans and Stearns applied to their study was contradicted by a similarly uncontrolled but much larger study published the same year by Peatman and Higgons in which 1152 infants were fed between 1600 IU and 9600 IU per day without any adverse effect on growth.26 Jeans and Stearns argued that their own study of nine infants was superior because it was an inpatient study, and bizarrely concluded from this fact that the Peatman and Higgons study “admirably support[ed]” their own results.

In addition to the Jeans and Stearns study, the Institute of Medicine cited a second study that compared 13 infants fed between 1400 IU and 2200 IU per day of vitamin D in the form of evaporated milk to 11 infants receiving between 350 IU and 550 IU per day in the same form. The infants were followed for six months and no difference in linear growth was found between the two groups.

The Institute’s protocol for developing a “no observed adverse effect level” would ordinarily require using the upper limit in this last study of 2200 IU per day, since it was shown to result in no adverse effect. It chose instead to deviate from this protocol and use the average intake of the high-vitamin D group as a whole, 1800 IU per day, because it conveniently matched the lowest adverse effect purportedly observed by Jeans and Stearns. Thus the Institute considers 1800 IU per day the point above which we can expect adverse effects and below which we can expect no adverse effects. This is then divided by an “uncertainty factor” of 1.8 to yield a tolerable upper limit of 1000 IU per day for infants younger than one year of age.

The recently published study showing a vitamin D intake of 2000 IU per day in the first year of life to be associated with a greatly reduced risk of type 1 diabetes included over 10,000 subjects who were followed from infancy through adulthood. In this large cohort, neither vitamin D deficiency nor regular supplementation with vitamin D at any dose was associated with any adverse effect on linear growth.28 There is, then, a complete absence of quality evidence supporting the concern that vitamin D adversely affects linear growth in infants.

Vitamin D Toxicity in Infants: Hypercalcemia

In 1963, the American Academy of Pediatrics suggested that the hypercalcemic response of infants to extreme doses of vitamin D is probably similar to that of adults. Reasoning that 20 percent of normal adults who receive 100,000 IU per day of vitamin D2 for several weeks or months develop hypercalcemia, the Academy adjusted this figure for bodyweight to the equivalent of 10,000 IU to 30,000 IU per day for a 1-year-old infant.3

It is possible that in rare cases infants may have a hypersensitivity to vitamin D. In the 1950s, British physicians identified a condition they named “idiopathic hypercalcemia,” which was associated with increased blood levels of calcium, anorexia, vomiting, wasting, constipation, thirst, excessive urination and, in some cases, kidney stones.29 At its peak, the syndrome was diagnosed at a rate of 100 cases per year. Although the hypercalcemia was never conclusively linked to vitamin D, the liberal enrichment of dried milk powder and infant cereals with crystalline vitamin D2, and the common practice of giving Viosterol (vitamin D2) drops to infants, made it readily possible for an infant to consume 4000 IU per day of vitamin D at that time.30 The rate at which hypercalcemia was diagnosed declined after the revision of regulations on the fortification of milk products, and most infants with the syndrome recovered after a period of restricting vitamin D and calcium.32

Most researchers currently believe that there were two forms of idiopathic hypercalcemia: a mild form involving a hypersensitivity to vitamin D, and a severe form characterized by Williams Syndrome, a disorder believed to be genetic in origin, the manifestation of which may or may not have been related to excess vitamin D.32

In 1990, Norwegian physicians reported several cases of calcification of the kidneys in ten children aged 1.5 to 14 years old.33 All of them had during infancy received four separate doses of 600,000 IU of vitamin D2. While still infants, five of them reacted with anorexia, vomiting, constipation, fever and dehydration, and four others showed failure to thrive and had recurrent fever and vomiting. One other patient had recurring kidney stones after six years of age. The reporting physicians noted that this is an extremely rare reaction to the treatment, although they admitted that a full third of infants react to the treatment with a transient state of hypercalcemia. While there is no evidence that this transient hypercalcemia results in any harm, it clearly demonstrates the unnatural nature of the treatment.

Unfortunately, there is insufficient evidence to judge whether such hypersensitive individuals would demonstrate the same type of hypersensitivity toward vitamin D3, or whether this hypersensitivity would be ameliorated by administering vitamin D within the protective and synergistic context of the full complement of fat-soluble vitamins.

Vitamin D, Atopy, Allergic Rhinitis and Asthma

In the same Finnish cohort that found doses of vitamin D in excess of 2,000 IU per day during infancy to powerfully protect against type 1 diabetes (see sidebar “Vitamin D and Type 1 Diabetes below), regular supplementation with vitamin D was associated with a 33 percent increased risk of atopy and allergic rhinitis compared to irregular or no supplementation; among those who supplemented regularly, the data suggested that supplementation with 2000 IU or more per day may increase the risk of asthma by as much as four times compared to regular supplementation with lower doses, although the study lacked the statistical power to determine whether or not this apparent effect was due to chance.34

These data provide us with more questions than answers. Further study will be needed to determine whether the apparent association with asthma is indeed real, and to determine what the causal relationship is, if one exists, between vitamin D, atopy and allergic rhinitis.

According to one theoretical perspective, allergies and autoimmune diseases like diabetes occupy two opposite poles in the spectrum of immune function. Within this perspective, vitamin D could be seen as moving one away from diabetes and toward allergies.

It is also possible that the phenomenon, if real, could result from an imbalance of the fat-soluble vitamins. Vitamin D increases the need for vitamins A and K,35 and may even exert toxicity by the direct mechanism of depleting the body of vitamin K.36 In fact, research shows that vitamins A and D protect against the allergic symptoms associated with vitamin D in the Finnish cohort.

Children with asthma have lower vitamin A levels than children without asthma, and the degree of vitamin A deficiency they exhibit is directly proportional to the severity of their asthma.37,38 In cell experiments, vitamin A eliminates the response of bronchial smooth muscle cells to growth factors that characterizes the asthmatic reaction39 and suppresses the activity of mast cells, which are involved in asthmatic or other reactions mediated by histamine or other inflammatory chemical messengers called leukotrienes.40 Consistent with studies in isolated cells, vitamin A deficiency causes asthmatic bronchial hyper-reactivity in live rats.41

Vitamin A also regulates the differentiation of bone marrow cells. Asthma, atopy and allergic rhinitis are characterized by the differentiation of bone marrow cells into eosinophil precursor cells; allergic reactions induce these precursor cells to travel from the bone marrow to the site of the allergic reaction, while inducing those precursor cells already present to finish differentiating into eosinophils; the eosinophils then engage in the inflammatory process and cause the symptoms associated with the reaction. Vitamin A suppresses the production of these inflammatory cells in bone marrow and further suppresses the chemical messages that cause their differentiation at the site of the reaction.42

Although there is little research on the mechanism by which vitamin K may be involved in protecting against allergic responses, one double-blind placebo-controlled study published in 1975 found treatment with vitamin K2 supplementation to relieve asthmatic symptoms in 90 percent of mild cases, 87 percent of moderate cases, and 73 percent of severe cases.43

There is sufficient evidence, then, to consider the hypothesis that vitamin D may raise the risk for allergies only when provided without the support of ample amounts of vitamins A and K. This is the most hopeful hypothesis, because it suggests that sufficient and properly administered vitamin D in infant nutrition may be able to nearly eradicate type 1 diabetes and perhaps other autoimmune disorders without the risk of adverse effects.

A Sensible Approach

Although one published study offers preliminary data suggesting that vitamin D supplementation in infancy may aggravate the risk of allergies–while simultaneously offering protection of greater magnitude against type 1 diabetes–little evidence exists to indicate that vitamin D causes other adverse effects. No quality evidence supports concerns over linear growth, and hypercalcemia has thus far only been shown to be a rare reaction to very large doses of vitamin D2. As research on the benefits of vitamin D during infancy progresses, we can expect to obtain a clearer understanding of what true risks, if any, vitamin D carries for the infant, and whether or not these risks, if they exist, can be obviated by supplying infants with vitamin D along with a rich supply of the other fat-soluble vitamins.

In the meantime, a sensible way to approach the issue is for mothers to maintain their own vitamin D status at optimal levels through diet and sunshine, thereby optimizing the vitamin D content of their milk, allow growing infants plenty of time outdoors, and to introduce foods rich in vitamin D and the other fat-soluble activators, such as cod liver oil, egg yolks and liver, as the baby’s first foods. If sun exposure for the mother is limited or impossible, supplementation is essential, in which case a high-dose cod liver oil is the perfect choice. Adequate nutrition during the formative years may well make a lasting and even permanent contribution to well being and quality of life, and we owe it to all newborns to ensure for them their right of birth–vibrant, radiant health.


Rickets and Vegetarianism

While the majority of infants who develop vitamin D-deficiency rickets are exclusively breastfed, this is never the only contributing factor. In addition to skin pigmentation, the use of restrictive religious clothing and poverty, vegetarian or vegan diets have frequently been reported to contribute to the vitamin D deficiency found in rickets.

Among 23 of the 24 cases of rickets documented at the Children’s Hospital of Philadelphia between 1974 and 1978 for which the dietary habits of the families were reported, 17 occurred in families practicing vegetarianism (74 percent) and 11 occurred in families practicing veganism (48 percent).14

Of 10 cases of rickets reported in 1978 by physicians from the University of Chicago’s Michael Reese Hospital and Medical Center, even occurred in families practicing veganism (70 percent) and one occurred in a family practicing fruitarianism (10 percent).18

Of 62 cases of rickets published between 1980 and 1990 in which the dietary preferences of the families are noted, 56 occurred in families practicing vegetarianism or veganism (90 percent).15

Of three cases of rickets reported in 1991 by Lyndon Key of the Bowman Gray School of Medicine in Winston Salem, North Carolina, one occurred in a family practicing vegetarianism for religious reasons, and one occurred in an infant whose mother could not afford meat.15

Most vegetarians, like most breastfed infants, do not develop rickets. Although many infants obtain vitamin D from sunlight or supplements, the association often found between vegetarianism and rickets reflects the fact that true vitamin D3 is only found in animal foods.

Bad Advice

In 1963, the American Academy of Pediatrics issued its official recommendation that infants should receive 400 IU per day of vitamin D from all sources. The Academy’s recommendation was based on the amount of vitamin D found necessary to protect against rickets, maximize intestinal absorption of calcium and maximize linear growth.3

This recommendation stood for 40 years until 2003, when the Academy lowered its recommendation to 200 IU in order to conform to the recommendation made in 1997 by the Institute of Medicine’s Food and Nutrition Board.6 The Academy’s revision was issued in a report barely longer than a single page and which accepted the Institute’s recommendation at face value without addressing any of the research on which it was based.

In fact, nearly a third of the report was dedicated to the reiteration of the Academy’s 1999 proclamation entitled, “Ultraviolet light: a hazard to children,” calling for parents to keep infants younger than six months out of direct sunlight, select activities for them that minimize exposure to sunlight and cover them with protective clothing and sunscreen. The final third of the report recommended that all infants who are either exclusively breastfed or receive less than 500 mL of infant formula should receive a supplement of 200 IU per day of vitamin D. The Academy dismissed the “typical” breast milk containing 25 IU/L as an insufficient source to meet the needs of the infant but provided no discussion of how a mother could increase the vitamin D content of her milk.

Vitamin D and Type 1 Diabetes

While the official recommendations for vitamin D continue to focus on the classically understood functions of vitamin D in supporting the skeletal system, recent research on more recently understood effects of vitamin D suggests that infants, like adults, have requirements for vitamin D that are far higher than those needed to protect against severe deficiency diseases like rickets and osteomalacia.

A group of researchers led by Elina Hypponen published a landmark study in The Lancet in 2001 suggesting that intakes of vitamin D over 2000 IU per day in infancy may be able to nearly eradicate type 1 diabetes.24 The study began in Finland in 1966 when over 10,000 infants were enrolled and researchers recorded whether they were supplemented with the then-official recommendation of 2000 IU per day of vitamin D for the first year of life, more than this amount, less than this amount, or were not supplemented at all. They then followed the study participants for over 30 years until 1997, recording which of the participants were diagnosed with type 1 diabetes by that time and which were not.

The benefit associated with vitamin D occurred at the lowest and highest levels of vitamin D intake and at all points measured in between. Children who developed rickets were 2.6 times as likely to develop type 1 diabetes compared to children who did not develop rickets. Compared to infants who did not receive a supplement of vitamin D at all, those who supplemented regularly, most but not all of whom supplemented with 2000 IU per day or more, had an 88 percent reduced risk of type 1 diabetes. Out of those who supplemented regularly, those who received the dose of 2000 IU per day had a further reduced risk of 78 percent compared to those who received a lower dose. Those who received a dose higher than 2000 IU per day had an 86 percent reduced risk compared to those who received a lower dose. Thus, vitamin D administered during infancy appears to exert a powerful protection against type 1 diabetes in amounts that extend from those just sufficient to prevent rickets to those exceeding 2000 IU.

Since the study was conducted in Finland where sunlight is very limited, we can be reasonably sure that the infants were not receiving very much vitamin D from that source. We unfortunately do not know what the mothers’ intakes of vitamin D were, or how much vitamin D they provided to the infants through breast milk. We therefore cannot say with any certainty exactly how much vitamin D the infants were obtaining from all sources combined.

The study does provide clear evidence, however, that the true vitamin D requirements of an infant may be far higher than those that are officially recommended and far higher than those indicated by the use of the traditional criteria of overt deficiencies like rickets.

The Vitamin D Study of Wayne Brehm

In 1937, Wayne Brehm presented before the Ohio State Medical Association the results of an experiment comparing the effects of the administration of cod liver oil with that of vitamin D2 to over 500 pregnant women. Dr. Weston Price describes this study in his book Nutrition and Physical Degeneration.

Vitamin D2, especially in conjunction with calcium, produced extensive abnormal calcification of the placenta, in one case extending into the uterine wall, and in three cases produced kidney stones within the developing fetus; cod liver oil, by contrast, produced no more tissue calcification than seen in controls. Brehm could not demonstrate whether the results of his experiment were attributable to the difference between vitamins D2 and D3, to a protective effect of vitamin A, to a protective effect of other constituents of cod liver oil, or to some combination thereof. His results make it clear, however, that cod liver oil is a safe and health-promoting source of vitamin D for the expectant mother which does not carry the risk of adverse effects carried by synthetic supplements.

Source: Brehm, W. “Potential dangers of viosterol during pregnancy with observations of calcification of placentae,” Ohio State Medical Journal, 1937; 33(9): 989-993.

Lowest Observed Adverse Effect?

The Institute of Medicine’s lowest observed adverse effect level of 1800 IU of vitamin D per day for infants less than one year of age is based on a paper published in 1938 by Jeans and Stearns of the Iowa State University Department of Pediatrics.

Jeans and Stearns carried out a small, uncontrolled study25 in which they fed three different preparations of vitamin D to nine infants: four of them received 4600 IU per day of vitamin D2 (viosterol) without any vitamin A; three of them received 2200 IU per day of vitamin D from a cod liver oil concentrate emulsified in cream, which also contained 1400 IU per day of vitamin A; finally, two of them received 1800 IU per day of vitamin D as cod liver oil, which also contained 6300 IU per day of vitamin A. The sample size from which the Institute of Medicine draws its lowest observed adverse effect level of 1800 IU per day, then, is a total of two infants. In substitution for a control group of infants receiving less vitamin D, the authors compared the growth curves of the nine infants to a “standard” curve showing the average growth of infants receiving 300 to 400 IU per day of vitamin D.

Jeans and Stearns claimed that the excellent degree of dietary control they were able to maintain compensated for the small sample size used in their study. Yet they also wrote that they carried out the study “without regard to season,” and brought the infants out into the sun whenever weather permitted. In reality, then, they had no idea what dose of vitamin D the infants actually received, and it was certainly higher than that which they recorded.

Additionally, they had no way of controlling for the confounding effect of vitamin A. The preparation with the highest amount of vitamin D did not contain any vitamin A while the preparation with the lowest amount of vitamin D contained the highest amount of vitamin A.

As one would logically expect, the growth curves for each infant never perfectly matched the average, but fluctuated between periods of above-average and below-average growth. Where Jeans and Stearns observed a small increase of growth, they made no correlations; where they observed a small decrease of growth, they attributed it to the dose of vitamin D and either immediately or within several weeks lowered the dose without waiting to see whether the growth rate would soon change again. Whether the next increase of growth came soon after or weeks later, the authors claimed that it was a direct result of the reduction in the dose.

The first infant who was fed 1800 IU of vitamin D from cod liver oil exhibited growth that roughly matched the standard curve for the first six months, with a spurt of rapid growth from week 20 to week 25. She then “showed evidence of decreased growth” for the next 11 weeks until she was discharged from the hospital with instructions to take only one teaspoon of cod liver oil. At her point of discharge, she was only one centimeter shorter than average.

On her first birthday, the baby fed cod liver oil was considerably taller than another baby who had received 4600 IU per day of vitamin D2 for the first 40 weeks after birth and 600 IU per day of vitamin D from cod liver oil thereafter. In a moment of prescient scientific insight, Jeans and Stearns concluded that this demonstrated the rapid growth that followed the reduction of the first baby’s dose of cod liver oil.

The second infant given 1800 IU per day of vitamin D as cod liver oil was studied for a mere for 18 weeks. His growth was slightly above average, and the authors noted only that his appetite was excellent throughout the study.

These are the two infants fed 1800 IU per day who form the entire basis of the Institute of Medicine’s upper limit of 1000 IU per day. Neither of them exhibited any substantial deviation from the average rate of growth. The results obtained from feeding larger doses of vitamin D were equally inconclusive. While Jeans and Stearns were conducting their uncontrolled study of nine infants, Peatman and Higgons26 published a report of 1152 infants fed high doses of vitamin D. Peatman and Higgons administered “cod liver oil or one of its substitutes” beginning at four weeks. They administered 1600 IU per day or more of vitamin D by two months, 3200 IU per day at four months, and between 3200 IU and 9600 IU after eight months, depending on the individual needs of the child. The growth rate of the boys receiving these large quantities of vitamin D was “upper average” and the growth rate of the girls was “exactly average” according to the same standard curve that Jeans and Stearns had used in their own study.

Although Jeans and Stearns could not quantify the dose that their own inpatients received because they did not control for exposure to sunlight, they insisted that their own study of nine infants was more reliable than the Peatman and Higgons study because the latter was of outpatients and therefore had less stringent control over the dose. From this observation, Jeans and Stearns came to the untenable conclusion that “the data presented by Peatman and Higgons thus support admirably the data presented in this [their own] paper.”

Thank you to Elina Hyppönen of the University College London’s Institute of Child Health for reviewing this article and offering constructive criticisms.


  1. Kalkwarf, H. Specker, B.L. “Vitamin D Metabolism in Pregnancy and Lactation.” In Feldman, D. Pike, J.W. Glorieux, F.H. eds., Vitamin D: Second Edition, Burlington: Elsevier Academic Press, 2005; 839-850.
  2. Bishop, N. “Perinatal Vitamin D Actions.” In Feldman, D. Pike, J.W. Glorieux, F.H. eds., Vitamin D: Second Edition, Burlington: Elsevier Academic Press, 2005; 803-810.
  3. American Academy of Pediatrics, Committee on Nutrition. “The prophylactic requirement and the toxicity of vitamin D,” Pediatrics, March 1963; 512-525.
  4. Institute of Medicine, Food and Nutrition Board, Standing Committee on the Scientific Evaluation of Dietary Reference Intakes. “Vitamin D.” In: Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride, Washington, DC: National Academy Press, 1997; 250-287.
  5. Brooke, O.G. Brown, I.R.F. Cleeve, H.J.W. Sood, A. “Observations on the vitamin D state of pregnant Asian women in London,” British Journal of Obstetrics Gynaecology., 1980; 88: 18-26. As cited in Bishop, N. “Perinatal Vitamin D Actions.” In Feldman, D. Pike, J.W. Glorieux, F.H. eds., Vitamin D: Second Edition, Burlington: Elsevier Academic Press, 2005; 803-810.
  6. Gartner, L.M. Greer, F.R. Section on Breastfeeding. Committee on Nutrition. American Academy of Pediatrics Clinical Report. “Prevention of Rickets and Vitamin D Deficiency: New Guidelines for Vitamin D Intake,” Pediatrics, 2003; 111(4): 908-910.
  7. Heaney, Robert P. “The Vitamin D requirement in health and disease,” Journal of Steroid Biochemistry and Molecular Biology, 2005; 97: 13-19.
  8. Stene, L.C. Ulriksen, J. Magnus, P. Joner, G. “Use of cod liver oil during pregnancy associated with lower risk of Type 1 diabetes in the offspring,” Diabetologia, 2000; 43: 1093-1098.
  9. Vieth, R. “The Pharmacology of Vitamin D, Including Fortification Strategies.” In Feldman, D. Pike, J. W. Glorieux, F. H., eds., Vitamin D: Second Edition, Burlington: Elsevier Academic Press (2005) 995-1015
  10. Hoogenboezem, T. Degenhart, H. J. De Muinck Keizer-Schrama, et al., “Vitamin D Metabolism in Breast-Fed Infants and their Mothers,” Pediatric Research, 1989; 25: 623-628.
  11. Ala-Houhala, M. Koskinen, T. Terho, A. Koivula, T. Visakorpi, J. “Maternal compared with infant vitamin D supplementation,” Archives of Disease in Childhood, 1986; 61: 1159-1163.
  12. Hoogenboezem, op cit, 2000.
  13. Specker, B. L. Tsang, R. C. “Cyclical serum 25-hydroxyvitamin D concentrations paralleling sunshine exposure in exclusively breast-fed infants,” The Journal of Pediatrics, 1987; 110(5): 744-747.
  14. Bachrach, S. Fisher, J. Parks, J. S. “An Outbreak of Vitamin D Deficiency Rickets in a Susceptible Population,” Pediatrics, 1979; 64(6): 871-877.
  15. Key, L. L. “Nutritional Rickets,” Trends in Endocrinology and Metabolism, 1991; 2: 81-85.
  16. Kruger, D.M. Lyne, E.D. Kleerekoper, M.K. “Vitamin D Deficiency Rickets: A Report on Three Cases,” Clinical Orthopaedics and Related Research, 1987; 224: 277-283.
  17. Chang, Y. T. Germain-Lee, E. L. Doran, T. F. Migeon, C. J. Levine, M. A. Berkovitz, G. D. “Hypocalcemia in Nonwhite Breast-Fed Infants: Vitamin D Deficiency Revisited,” Clinical Pediatrics, 1992; 31(11): 695-698.
  18. Edidin, D. V. Levitsky, L. L. Schey, W. Dumbovic, N. Campos, A. “Resurgence of Nutritional Rickets Associated with Breast-Feeding and Special Dietary Practices,” Pediatrics, 1980; 65(2): 232-235.
  19. Specker, B. L. Tsang, R. C. Hollis, B. W. “Effect of Race and Diet on Human-Milk Vitamin D and 25-Hydroxyvitamin D,” American Journal of Diseases of Children, 1985; 139: 1134-1137.
  20. Ala-Houhala, M. Koskinem, T. Parviainen, M. T. Visakorpi, J. K. “25-hydroxy-vitamin D and vitamin D in human milk: effects of supplementation and season,” American Journal of Clinical Nutrition,1988; 48: 1057-60.
  21. Hollis, B. W. Roos, B. A. Draper, H. H. Lambert, P. W. “Vitamin D and Its Metabolites in Human and Bovine Milk,” Journal of Nutrition, 1981; 111: 1240-1248.
  22. Ala-Houhala, M. Koskinen, T. Terho, A. Koivula, T. Visakorpi, J. “Maternal compared with infant vitamin D supplementation,” Archives of Disease in Childhood, 1986; 61: 1159-1163.
  23. Specker, B. L. Valanis, B. Hertzberg, V. Edwards, N. Tsang, R. C. “Sunshine exposure and serum 25-hydroxyvitamin D concentrations in exclusively breast-fed infants,” Journal of Pediatrics, 1985; 107: 372-376. Note: The vitamin D status of the infants was not correlated with the maternal dietary intake of vitamin D, the vitamin D content of breast milk was not reported, and both dietary intake of vitamin D and sunlight exposure would contribute to the content of breast milk (see Ala-Houhala, et al., 1988). Maternal sunlight exposure could therefore confound the effect of maternal vitamin D intake.
  24. Hypponen, E. Laara, E. Reunanen, A. Jarvelin, M. R. Virtanen, S. M. “Intake of vitamin D and risk of type 1 diabetes: a birth-cohort study,” The Lancet, 2001; 358: 1500-03.
  25. Jeans, P. C. Stearns, G. “The effect of vitamin D on linear growth in infancy: II. The effect of intakes above 1,800 U.S.P. Units Daily,” Journal of Pediatrics, 1938; 13: 730-740. Note: The Institute of Medicine describes this study as consisting of 35 infants rather than 9. I have contacted the IOM to clarify this discrepancy but did not receive a response by the time of the publication of this article.
  26. Peatman, J. G. Higgons, R. A. American Journal of Diseases of Childhood. 1938; 55: 1233. As cited in Jeans and Stearns, 1938, op cit.
  27. Fomon S. J. Younoszai M. K. Thomas LN. “Influence of vitamin D on linear growth of normal full-term infants,” Journal of Nutrition, 1966; 88:345-50. As cited in Institute of Medicine, 1997, op cit.
  28. Hypponen, E. Fararouiei, M. Robertson, C. Jarvelin, M. R. “Vitamin D and linear growth: cohort study on long-term effects of rickets and large dose supplementation,” Nutrition Society, 2006; (submitted).
  29. Creery, R. D. G. “Idiopathic Hypercalcaemia of Infants,” The Lancet, 1953; 2:17-19.
  30. British Pediatric Association, Committee on Hypercalcaemia, “Hypercalcaemia in Infants and Vitamin D,” British Medical Journal, 1956; 2: 149. The vitamin D used for fortification is termed “crystalline calciferol,” which was the name by which vitamin D2 was referred at the time. See, for example, reference 31.
  31. Leading Article, “Vitamin D as a Public Health Problem,” British Medical Journal [Clin Res], 1964; 1: 1654-1655.
  32. Jones, K. L. “Williams Syndrome: An Historical Perspective of Its Evolution, Natural History, and Etiology,” American Journal of Medical Genetics, Supplement, 1990; 6: 89-96.
  33. Misselwitz, J. Hesse, V. Markestad, T. “Nephrocalcinosis, Hypercalciuria and Elevated Serum Levels of 1,25-Dihydroxyvitamin D in Children,” Acta Paediatr Scand, 1990; 79: 637-643.
  34. Hypponen, E. Sovio, U. Wjst, M. Patel, S. Pekkanen, J. Hartikainen, A-L. Jarvelin, M-R. “Infant Vitamin D Supplementation and Allergic Conditions in Adulthood,” Annals of the New York Academy of Science, 2004; 1037: 84-95.
  35. Masterjohn, C. “From Seafood to Sunshine: A New Understanding of Vitamin D Safety,” Wise Traditions, Fall 2006; 14-33.
  36. Masterjohn, C. “Vitamin D toxicity redefined: Vitamin K and the molecular mechanism,” Medical Hypotheses, 2006; doi: 10.1016/j.mehy.2006.09.051.
  37. Mizuno, Y. Furusho, T. Yoshida, A. Nakamura, H. Matsuura, T. Yoshikatsu, E. “Serum vitamin A concentrations in asthmatic children in Japan.” Pediatrics International, 2006; 48: 261-264.
  38. Arora, P. Kumar, V. Batra, S. “Vitamin A status in children with asthma,” Pediatric Allergy Immunology, 2002; 13: 223-226.
  39. Day, R. M. Lee, Y. H. Park, A-M. Suzuki, Y. J. “Retinoic Acid Inhibits Smooth Muscle Cell Migration,” American Journal Respiratory Cell Mollecular Biology, 2006; 34: 695-703.
  40. Ko, J. Yun, C-Y. Lee, J-S. Kim, D-H. Yuk, J. E. Kim, I. S. “Differential regulation of CC chemokine receptors by 9-cis-retinoic acid in the human mast cell line, HMC-1.” Life Sciences, 2006; 79: 1293-1300.
  41. McGowan, S. E. Smith, J. Holmes, A. J. Smith, L. A. Businga, T. R. Madsen, M. T. Kopp, U. C. Kline, J. N. “Vitamin A deficiency promotes bronchial hyperreactivity in rats by altering muscarinic M2 receptor function,” American Journal of Physiology, Lung Cell Mollecular Physiology, 2002; 282: L1031-L1039.
  42. Cyr, M. M. Denburg, J. A. “Systemic aspects of allergic disease: the role of the bone marrow,” Current Opinion in Immunology, 2001; 13: 727-732.
  43. Kimur, I. Tanizaki, Y. Sato, S. Saito, K. Takahashi, K. “Menaquinone (vitamin K2) therapy for bronchial asthma. II. Clinical effect of menaquinone on bronchial asthma.” Acta Med Okayama,1975; 29(2): 127-35.

This article appeared in Wise Traditions in Food, Farming and the Healing Arts, the quarterly magazine of the Weston A. Price Foundation, Winter 2006.

Leave a reply

© 2015 The Weston A. Price Foundation for Wise Traditions in Food, Farming, and the Healing Arts.