The jury is still out on possible links between cows’ milk and type 1 diabetes

Abstract

There is an ongoing global epidemic of type 1 diabetes, and it has been increasing by 3–6% per year in a large number of countries (1). In spite of rather stable genetic background, the increasing incidence, as well as several other facts (migrant studies, great difference in incidence of type 1 diabetes between genetically very similar populations etc) show that environmental factors have a crucial role, but the causes of type 1 diabetes remains a puzzle (2). Several facts suggest that infections are involved, either directly or via their influence on, for example, the immune system and gut microbiome (3). However, other mechanisms are plausible, such as rapid growth (4), psychological stress (5) and any factors causing beta cell stress (6). Not least, the parents of diabetic children often think back to try and find explanations for why just their child has got type 1 diabetes, especially in the 85–90% of cases where there is no family history of the condition. One area that tends to lead to feelings of guilt and anxiety is early nutrition, as parents, especially mothers, may feel that it is their fault that their child has got diabetes. At a Nordic diabetes symposium I arranged in Vadstena, Sweden, in 1981, I showed a slide that suggested that breastfeeding was an early part of the process that finally ended in diabetes. Shortly afterwards, some of the participants carried out studies that suggested that breastfeeding might protect against the development of type 1 diabetes (7). These were followed by a large number of studies in different countries. The results were very different, and when we carried out a meta-analysis of 43 observational studies, we found very weak support for the theory that breastfeeding provided a protective effect (8). In fact, breastfeeding had increased considerably from the early 1970s to the end of the 1990s, especially in Sweden and Finland, and this was paralleled by a steep increase in the incidence of type 1 diabetes. A hypothesis gathered strength that suggested links between type 1 diabetes and the early introduction of cows’ milk proteins to babies, although it was difficult to differentiate it from short-term exclusive breastfeeding. The hypothesis was supported by a number of findings, including studies that showed an increased prevalence of antibodies against cows’ milk proteins in newly diagnosed diabetic children and that it was possible to prevent autoimmune diabetes in experimental animals by avoiding cows’ milk proteins. Epidemiological associations between cows’ milk consumption and the incidence of type 1 diabetes have also been reported. In addition, studies have found similarities in the amino acid sequence of cows’ milk proteins and parts of the human leucocyte antigen chains associated with an increased risk of type 1 diabetes. These findings prompted comprehensive studies that aimed to prevent type 1 diabetes by giving babies formula that contained hydrolysed cows’ milk rather than untreated cows’ milk when their mothers were no longer breastfeeding. The first trials looked promising (9), and there were a number of studies that showed how early diet could influence diabetes-related autoimmunity. An extensive international study, the Trial to Reduce Insulin dependent diabetes in the Genetically at Risk, was performed in Europe, North America and Australia. More than 2000 newborn infants with first-degree relatives with type 1 diabetes and an increased genetic risk for type 1 diabetes were included in a randomised double-blind placebocontrolled trial, which had a 10-year follow-up period. One group of children received hydrolysed formula after the end of breastfeeding, and others were given formula based on cows’ milk, but with the addition of some hydrolysed formula to make it indistinguishable from the other formula in terms of smell and taste. In parallel with this trial, there were also studies based on the hypothesis that the bovine insulin in cows’ milk could stimulate antibodies against insulin. The results of a Finnish study provided some support for this suggestion (10). After the main TRIGR trial was followed up for six years, it was surprising to learn that there was no difference in diabetes-related autoantibodies between the two groups. These findings were in contrast to the results produced by the pilot trial (9). And after 25 years of intensive research based on the hypothesis, which was presented in the early 90s, the hypothesis was almost killed off in 2018 when the TRIGR trial results were published that showed no difference at all between the two groups (11). Avoiding cows’ milk early in life did not prevent type 1 diabetes. This was a big disappointment from both the clinical and research point of view, but it provided reassurance for the parents of children who had already developed type 1 diabetes, because it meant that they were not to blame for giving their infants cows’ milk-based formula when their infants were too young.