Can breastfeeding prevent celiac disease?

The recent rise of celiac disease reflects more than just increased awareness. Many scientists think that modern lifestyles or something in the environment is triggering celiac, which causes the body to have an autoimmune reaction to the protein gluten, found in grains like wheat, barley and rye. For years, physicians believed that breastfeeding and introducing gluten into a baby’s diet at just the right time could prevent celiac disease, and research seemed to back this up. But two studies published in the New England Journal of Medicine in 2014 cast doubt on the standard thinking, showing that neither breastfeeding nor timing of introduction of gluten in a baby’s diet had any effect in preventing celiac.

Those two studies led a group of gastroenterologists from the board of the North American Society for the Study of Celiac Disease, including Stefano Guandalini, MD, medical director of the University of Chicago Celiac Disease Center, to write an article in the American Journal of Gastroenterology this summer about the state of research on celiac disease prevention.

Science Life spoke to Guandalini about the research that led to this point, and where he and his colleagues are turning next to find new ways to prevent celiac.

What was the standard thinking about timing of breastfeeding or introducing gluten to a baby’s diet before these papers came out?

There have been studies trying to understand the factors in the environment that might increase the risk of developing celiac disease. Once the genetic components of it were identified, we started to look at what is in the environment that makes those who have the genes more or less predisposed to develop celiac. Over the years, several factors were identified. From the mid- to late-70s researchers had been focusing on the protective role of breastfeeding, because conceptually, there are many reasons to believe it might help. It helps in preventing food allergies; it helps in preventing other auto-immune conditions such as type 1 diabetes and multiple sclerosis, so the data were pretty hard.

Stefano Guandalini, MD

Initial observational studies have actually suggested that this was also the case for celiac disease, to the point that in 2006, an analysis by a group in England who reviewed all the papers published at the time that had to do with the protective role of breastfeeding and celiac disease concluded that there is a significant protective role. In fact, we thought the issue was basically closed, and we kept teaching to everybody that, yes, breastfeeding indeed has a protective role. There were two components to it. One was the duration of breastfeeding: the longer the better, up to a year. The second component was having a baby breastfed at the time of gluten introduction. That also made a lot of sense, because in breast milk, there are active immune components, such as white blood cells and immunoglobulin. It just made sense to think that if you had gluten presented for the first time in this milieu, that would favor the onset of tolerance rather than intolerance. The 2006 analysis concluded both that duration and breastfeeding at the time of gluten introduction was protective.

Some observational studies have also shown that very early introduction of gluten—before the age of four months—was detrimental. In several studies, celiac patients actually had a history of receiving gluten a month or two before their matched controls who did not develop celiac disease. The theory was developed that since breastfeeding was protective at the time of gluten introduction, gluten should be introduced early in small amounts. This evidence came from epidemiological studies in Sweden, where there was a large increase in prevalence of celiac disease in children below the age of two that lasted three to four years, and then declined back to previous levels. Investigators linked this spurt to different infant feeding practices that resulted in higher amounts of gluten introduced, because gluten was added to some formulas that were widespread in Sweden. When this was realized, conditions were changed, infant feeding practices were changed back and the evidence of celiac disease also decreased. So everybody thought that’s it, evidently gluten is also harmful if given in larger amounts.

And now these new studies come along and cast doubt on all of this this evidence. What happened?

There is one important caveat to bear in mind: All the studies that showed the protective effect of breastfeeding and the role of timing of gluten were done on unselected populations. So that means you have celiac patients on one side, and matched controls [patients without celiac] in the general population. These two new studies focus only on children that had the genetic risk for celiac disease, so that is an important difference.

The two new studies were independently conducted. One was an Italian study that had 553 infants who were followed from birth. They also belonged to the at risk population, so they actually had to be born in families where there was at least one relative with celiac disease. One group of these children received no gluten until the age of 12 months, and then they received normal amounts of gluten. The other group received gluten as people usually would do in Italy and many other countries, that is, at around 6 months. Then they were followed for up to 10 years to see if they developed celiac disease.

They were actually expecting that to happen. I know the principal investigator of the study very well, and he was almost sure that delaying the introduction of gluten beyond the first year of life would actually result in a lower prevalence of celiac disease. But the results didn’t show that—the prevalence was the same. Yes, they started later, as you would expect, but basically they caught up, so that by five years of age the prevalence of celiac disease was the same, and on up to 10 years of age. So when focusing on a population that is predisposed to celiac, delaying the introduction of gluten beyond the first year of life has no preventive effect.

The other study was a multi-center European study conducted by the European Society for Pediatric Gastroenterology, with almost 1,000 patients born in families at risk for celiac disease. The intervention this time was different. The idea was that giving small amounts of gluten between four and six months would be preventative because most babies at this age were still being breastfed and it would induce tolerance. So they had two groups: in one they added 200mg of gluten a day to their diets at the age of four month, which is a very small amount, and then at six months they liberalized the diet with normal amounts of gluten. The other group had no gluten until six months, and then they had normal amounts, which would be the normal way of weaning a child. Again, the expectation was that the group that received minimal amounts of gluten between four and six months would actually show, with time, lower prevalence of celiac disease. But that didn’t prove to be true. These two groups developed celiac disease at the same rate. There was no preventative effect of giving small amounts of gluten.

Does the new research say anything about the preventative effect of breastfeeding?

Neither of the new studies focused on the role of breastfeeding. They did no interventions; they just observed what happened in the groups. But it turns out when examining the results, there was no difference between those who had been breastfed longer and those who had been breastfed for a shorter period of time. This is true in both studies. They could not conclude any evidence of breastfeeding being protective for celiac.

Your article’s subtitle is “Back to the Drawing Board.” Does this mean we’re back to the beginning in understanding how to prevent celiac disease?

It means we’ve learned the hard way. Rather than relying a few observational studies and making this a general paradigm, we need to be more careful. In reality, we can’t document that breastfeeding has a protective effect. On gluten introduction, the one thing that is sure is that waiting doesn’t help. I would still be skeptical about introducing gluten too soon. This is against Mother Nature—babies have no teeth until about six months, so why would you give them cereal? So there are many reasons to continue recommending not introducing gluten too soon. But “back to the drawing board” means, essentially, because we see this huge rise in prevalence of celiac disease, and genes do not change overnight or even 1,000 years, what is going on? What is causing this prevalence? Let’s try to look at other factors. Let’s abandon the idea of breastfeeding or gluten introduction being so crucial, and look at something else.

What would that something else be?

Everybody seems to agree right now, that it’s the microbiome. We are profoundly changing the microbiome of our babies by a number of interventions: C-section, use of antibiotics and so on. Now preventing infections is a good thing, but Mother Nature had our system wired in such a way that we were expected to receive contaminants from the environment, and react to that by building a specific kind of immune response. This is now profoundly being changed by lifestyles, overuse of antibiotics, and so forth.

The problem is there is a huge amount of work that has to be done, and it won’t be solved in the next two or even five years. We have to tease out the individual components. It’s not simply a matter of finding the good bacteria and the bad bacteria. There is much more to that. The interaction of the microbiome and the host has so many implications that people need to go back to the drawing board and try to design studies to understand what is wrong and what can be done. Once we have the same genes and the same risk in a family, why would some children become celiac and some not? Is the microbiome different? Which way is it different? What can we do to make up for the difference and restore the microbiota to defend against celiac?

Is it more realistic to think that we can prevent it from developing, or treat it better after it develops?

It’s probably more feasible to identify a preventive strategy, because once we reach a full understanding about the microbiota composition, when you detect that in a baby you might have already developed a plan of intervention to restore it. It could be the use of probiotics. It could be something that has to do with feeding. Restoring the lost tolerance might be more difficult, though, because the current teaching is that celiac disease is permanent once it develops. That would require a lot of investigation on what leads to the loss of tolerance, which is not entirely clear.

I may be proved wrong, but my bet would be on the preventative strategy to be more feasible. It might be probiotics, something that is genetically engineered that we can use to personalize the treatment. If there is a child who develops celiac based mostly on the environmental factors and some erratic composition of the microbiota, can we change that to prevent celiac? Probably so. But it could be years to come.

About Matt Wood (479 Articles)
Matt Wood is a senior science writer for the University of Chicago Medicine and editor of the Science Life blog.
%d bloggers like this: