Intended for healthcare professionals


Phytoestrogens and soy based infant formula

BMJ 1996; 313 doi: (Published 31 August 1996) Cite this as: BMJ 1996;313:507
  1. Charles Essex
  1. Consultant community paediatrician Child Development Unit, Gulson Hospital, Coventry CV1 2HR

    Risks remain theoretical

    Soy protein is one of the cheapest sources of protein and has been used as a substitute for cow's milk since the turn of the century. Soy based infant formulas have been available in Britain for over 20 years and account for about 7% of infant formula sales (compared with 13% in New Zealand (C Wham, personal communication) and 10-20% in the USA).1 This represents gross sales in Britain of £10m annually. However, with rates of initial breastfeeding of only about 63%,2 many infants will be fed a soy based formula at some time in their first year of life.

    Plant protein sources such as soy are quite complex and very different from milk proteins found in most infant formulas. Soy is a rich source of phytoestrogens, non-steroidal oestrogens of the isoflavone class.3 These compounds are structurally similar to oestrogens; they bind to oestrogen receptor sites and behave as partial oestrogen agonists and antagonists.4 It is unclear whether these effects are beneficial or detrimental to health, and there are virtually no data on their oestrogenic effects in children. The safety of these formulas has recently been questioned, and the chief medical officer has written to all doctors advising them on the issue.5

    Epidemiological studies of populations whose diets contain high levels of soy show that they have a lower incidence of and mortality from hormone dependent cancers such as cancer of the breast and prostate.6 7 In vitro studies have shown that genistein and diadzein, two isoflavones found in soy, can inhibit the growth of breast cancer8 and prostate cancer tissue.9 Conversely, dietary oestrogens from soybean products have been implicated as a possible cause of infertility and liver disease in some animal species, although these effects seem to be species specific.10

    In adults faecal excretion of isoflavones is only 1-2% of the amount ingested, implying there is a significant absorption of ingested isoflavones.11 A diet with 60 g of soy protein a day, which contains 45 g of isoflavones, affected the menstrual cycle and levels of luteinising hormone and follicle stimulating hormone in adult premenstrual women.12 On a weight for weight basis, neonates fed recommended amounts of soy based formula would be consuming between three and five times that amount of isoflavones. These formulas are usually their sole source of nutrition for the first three to six months of life until other foods are introduced, yet paediatricians and paediatric endocrinologists do not see large numbers of infants with feminisation. The hypothalamic-pituitary-gonadal axis is much more active in neonates than in older children and adults, which may limit the neonatal response to these apparently high levels of oestrogen-like compounds. However, the long term effects are unknown.

    In the meantime how should doctors and other health professionals advise parents? Obviously breastfeeding is best for babies. If mothers do not breastfeed their babies, they should use a recognised cow's milk based formula unless there are valid reasons not to do so. Since the carbohydrate in soy based formulas comprises sucrose or glucose polymers rather than lactose, it would be appropriate to use these formulas for galactosaemia and lactose intolerance (either primary or secondary). Parents who are vegans may choose to use a soy based formula for their infants as it contains no animal products.

    Indiscriminate swapping between formulas, often on the advice of health professionals, should be avoided, as should spurious recommendations to use a soy based formula for vague symptoms and signs.13 These include normal crying-fussing behaviour of young infants, colic, and rashes, any of which may be ascribed to cow's milk protein intolerance. Casual treatment in this manner is undesirable because it leads to overdiagnosis of food allergy, with possible long term effects on children's dietary habits and calcium intake. The diagnosis of gastrointestinal cow's milk protein intolerance should not be made without careful evaluation by an expert in the field. When it has been proved, infants should be fed formulas containing protein hydrolysates rather than soy based formula, as soy protein is also a potential allergen.14 Soy based formula should not be given routinely as prophylaxis to infants thought to be at risk of developing allergy or atopy. The evidence to support this practice is conflicting.15

    In the short term the theoretical hazards for infants of consuming phytoestrogens in soy based formulas have not been recognised clinically. More research is needed into both the immediate and long term effects of soy based formulas. However, in the meantime, parents whose babies are satisfied and thriving on a soy based formula should not change to another formula.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    View Abstract