Introducing Allergenic Foods in Breastfed Infants: Does Timing Matter?

William T. Basco, Jr, MD, MS


May 02, 2016

Randomized Trial of Introduction of Allergenic Food in Breast-Fed Infants

Perkin MR, Logan K, Tseng A, et al; EAT Study Team
N Engl J Med. 2016 Mar 4. [Epub ahead of print]

Study Summary

This randomized trial, which took place from 2009 to 2012 at a single site in the United Kingdom, evaluated early introduction of multiple potentially allergenic foods in infancy. Exclusively breastfed infants were recruited at age 3 months from a general population (not necessarily at higher risk for allergy). The standard-introduction group was exclusively breast-fed until age 6 months, after which the parents could introduce all foods, including potential allergens.

In the early-introduction group, parents introduced six allergenic foods at age 3 months—beginning with cow's milk, then followed by five other allergenic foods (peanut, egg, sesame, whitefish, and eventually wheat). If the child passed an initial oral challenge without a reaction, the parents were instructed to continue giving the child at least 2 g of each allergenic food twice weekly. If a child reacted to the oral challenge, the parents were instructed to avoid that specific food but continue introduction of the other foods to which the child did not react.

To assess exposure to allergenic foods, the parents completed monthly food diaries for the first 12 months, then quarterly food diaries until the children were 3 years old. The main outcome of interest was whether the children developed an allergy to one of the six allergenic foods at any time between age 12 and 36 months, as evidenced by a positive response to an oral food challenge.

In general, the trial failed to show a protective effect of early introduction of allergenic foods. For the primary outcome of challenge-proven allergy, 5.6% of the early-introduction group had a food allergy at 36 months, compared with 7.1% of the standard-introduction group (relative risk, 0.80; 95% confidence interval, 0.51-1.25)—a nonsignificant difference.

With respect to specific foods, peanut allergy was documented in 1.2% of children in the early-introduction group vs 2.5% of children in the standard-introduction group—again failing to reach statistical significance. Egg allergy occurred in 3.7% of the early-introduction group compared with 5.4% of the standard-introduction group, also not significantly different. The frequency of allergies to the other foods were all less than 1% and did not differ between the two groups.

The investigators also conducted per-protocol analyses on only the infants who adhered to their group assignment for at least 5 months. Across multiple outcomes, the per-protocol analysis suggested that early introduction of allergenic foods was protective, and these differences did reach statistical significance. Additional analyses looking at the amount of foods consumed generally demonstrated that increasing consumption of allergenic foods correlated with lower risk of developing food allergy at either 12 or 36 months.

The investigators concluded that early introduction of multiple allergenic foods did not protect against development of food allergies in the intention-to-treat analyses, but there appeared to be an inverse dose-response in the secondary analyses.


Although the results of this trial are negative, as pointed out in an accompanying editorial,[1] several findings are worth noting and suggest that further research should be done.

First, even though the differences in the primary outcome and secondary outcomes in the intention-to-treat analyses were not statistically significant, the allergy frequencies were consistently lower among the early intervention group, regardless of food or outcome evaluated (oral challenge results or immunoglobulin E levels, for example).

Second, this trial was conducted in a general population, whereas the LEAP trial (which had such promising results with early introduction of peanut) enrolled children from high-risk families. It's possible that the hint of protective effect shown in this study might be magnified in a population at greater risk.

Third, the duration of feeding of the allergenic foods was relatively short—from age 6 months to 1 year. The LEAP study suggests that longer administration of allergens may be more likely to lead to favorable outcomes.

Finally, the demonstration of an inverse relationship between the amount of allergen consumed and risk of developing an allergy later on is compelling and suggests that adherence to protocol might have produced better effects.

In summary, although this was a well-designed trial that had negative results, I'm not sure that it yet closes the door on similar investigation in either larger or higher-risk populations.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: