Cow's milk and other early nutritional influences

There is evidence dating back to the 1980s that prolonged breast-feeding offers modest protection against later development of type 1 diabetes. This could mean either that breast milk is protective or that constituents of cow's milk products or other factors in infant feeds might in some way predispose to diabetes. The focus on cow's milk was supported by animal studies in which diabetes-prone rodents fed with hydrolysed milk (hydrolysis disrupts its constituent proteins) had a lower risk of diabetes. These observations prompted a major intervention study called TRIGR in which neonates have been randomized to standard or modified feeds upon weaning. The trial has completed recruitment, and has reported lower levels of diabetes-related autoantibodies in children who avoided cow's milk, but it has yet to be seen whether they will have a lower rate of diabetes.

Milk and other early nutritional influences

Coeliac disease and type 1 diabetes have common features, including overlap between HLA haplotypes conferring susceptibility. The environmental agent responsible for coeliac disease is well known, but the condition may nonetheless develop years or even decades following first exposure to dietary gluten. Could dietary constituents encountered early in life (and the immune responses these provoke) play a similar role in the pathogenesis of type 1 diabetes? There has been controversy for 15 years as to whether early exposure to cow’s milk predisposes to childhood diabetes; similar arguments have taken place in other diseases, including asthma and multiple sclerosis, but remain equally inconclusive.

Animal studies have shown that elimination of milk proteins (present in standard laboratory chow) from the early diet greatly reduces the risk of diabetes in the BB rat, with similar results in the NOD mouse. Other dietary constituents, including wheat and soya beans have also been implicated, however, and cow’s milk is only one among a number of possible dietary candidates.[1] The only systematic review of the epidemiological evidence in humans found a weak (odds ratio 1.5)[2] but detectable positive effect of early exposure to cow’s milk proteins, but subsequent reports have yielded conflicting results and the area remains controversial.[3][4]

Serological responses to milk constituents have also been studied in infants, but once again no consistent pattern has emerged; as with analysis of antibody responses to viruses, case–control studies need to be corrected for HLA susceptibility. One study did however suggest that some species of milk antibody are more common in cases than in siblings sharing the same HLA alleles.[5] As with virus exposure, prospective analysis of high risk populations from birth has great advantages over other means of analysis, and will allow exposure to milk and other dietary constituents to be correlated with formation of islet autoantibodies. Improved and standardised methods of analysis of cellular immune responses to dietary constituents are also badly needed, and should also be corrected for HLA type,[3] but the issue will only be resolved beyond doubt by intervention studies.


TRIGR (Trial to Reduce Diabetes in the Genetically at Risk) is a randomized controlled trial taking place on three continents in which 2160 eligible infants have been randomized to test or control formulas when mothers decide to wean from exclusive breastfeeding. Recruitment is now completed and the participants will be monitored up to the age of 10 years for the appearance of diabetes-predictive autoantibodies and clinical type 1 diabetes. More information is available at


Nobody has commented on this article

Commenting is only available for registered Diapedia users. Please log in or register first.