First published Winter 2018


Maternal gluten intake associated with type 1 diabetes risk in children, but researchers caution against changing dietary guidelines


Researchers have found that women who eat high levels of gluten during pregnancy may have children with a higher risk of developing the autoimmune condition type 1 diabetes.

Gluten is a name for proteins found in wheat, rye, and barley. Analysing data from 63,529 women, scientists at the Bartholin Institute, and the Statens Serum Institut, Denmark, found that for every additional 10 g of gluten ingested during pregnancy, the risk for type 1 diabetes in children increased by a factor of 1.31. At the upper and lower ends of the scale, however, children born to women who ate high amounts of gluten (20 g/day or more) had double the risk of going on to develop type 1 diabetes, compared with those born to women who ate the lowest amount of gluten (less than 7 g/day).

Senior researcher Knud Josefsen, PhD, explained what this might mean in terms of typical foods. “I calculated that there is approximately three grams of gluten in a slice of white bread and approximately five to 10 grams of gluten in a large serving of pasta (150 g),” he said. “So 20 grams/day is a fair amount, but not unrealistic.”

Previous studies on rats and mice have found that a gluten-free diet during pregnancy almost completely prevents type 1 diabetes in offspring. To date, however, no such controlled intervention study has been undertaken in pregnant women. These new findings followed an observational study that analysed data from women who had been enrolled into the Danish National Birth Cohort, looking at a food frequency questionnaire the women had completed at week 25 of pregnancy. Information on type 1 diabetes in their children was then obtained over a mean follow-up period of 15.6 years through the Danish Registry of Childhood and Adolescent Diabetes.

Despite their findings and the authors stating that this was a large, high-quality study, changing dietary recommendations for gluten during pregnancy was not recommended.

“We would still say that — at least in relation to gluten — [pregnant women] should not change their diet until we fully know the consequences of a gluten-free diet,” said Josefsen. “We primarily consider the study as idea-generating. We think it is interesting that environmentally-controlled diseases can be determined already in the womb, and that gluten was identified as a possible risk in the study.”

He cautioned that the study did not prove cause and effect. “It is absolutely possible that the effect that we see is caused by something else than gluten,” he said. “In principle, another component that is present in equivalent amounts with gluten could also be responsible.”

However, it may be a long time before it is possible to confirm the role of gluten one way or the other, because of the difficulty in conducting a clinical trial.

“An intervention study is unfortunately unpractical,” said Josefsen.

The number of children expected to develop type 1 diabetes is low (one per cent) and so to have enough “statistical power” any intervention study would need a population as large as that used in the observational study (around 63,000).

“Further, ideally, intervention studies should be blinded, which is of course not possible in this instance,” he said. A statistical study was the best the team could currently do, but he and his colleagues would like to see the study replicated in other populations.

... the study did not prove cause and effect. “It is absolutely possible that the effect that we see is caused by something else than gluten”

Meanwhile, the association between gluten consumption during pregnancy and type 1 diabetes in offspring remains unexplained. But if gluten is a trigger, it has been suggested that increased inflammation or increased gut permeability, often called ‘leaky’ gut, may be involved. Josefsen describes gluten as “a pretty strong immune system activator”, as experiments have shown gluten-fed animals to have a more pro-inflammatory profile in their immune system when compared with animals fed a gluten-free diet. But he added that it is not known whether removing gluten — or at least the grains that carry gluten — could negatively impact health in other ways.

“For instance, it was demonstrated recently that there is an inverse relationship between type 2 diabetes and gluten intake [higher risk of type 2 diabetes with lower consumption of gluten] — although not in relation to pregnancy. So we need to make sure that we do not affect the frequency of type 2 diabetes by giving a gluten-free diet.”


Read more articles and recipes  



  1. BMJ 2018;362:k3547