Breast milk is certainly a hot topic these days. Allergies in childhood are almost equally hot in the media as food allergies seem to be on the rise (not my specialty by a long shot) as well as rates of other atopic illness. Given what is known about the modifiable risks in terms of a number of conditions such as NEC and late onset sepsis in preterm infants it wouldn’t be a stretch to wonder what impact avoidance of cow’s milk exposure could have in the term newborn.
A Landmark Japanese Study
Urashima et al just published in JAMA Pediatrics the following paper Primary Prevention of Cow’s Milk Sensitization and Food Allergy by Avoiding Supplementation With Cow’s Milk Formula at Birth: A Randomized Clinical Trial . This paper looked at 312 infants (≥ 36 weeks at birth) who were randomized to either receive breastfeeding plus an elemental formula if needed vs breastfeeding plus intact protein cow’s milk formula with a volume of at least 5 mL/kg per day. In order to have a group of infants truly at risk of atopic disease, all infants had to have at least one immediate relative with atopic disease. In each arm of the study, infants were followed with blood IgE levels at 5 and 24 months of age to detect a level of CM-IgE ≥ allergen units/mL. This was the primary outcome on which the power calculation was based for the study. Using an estimated incidence of 10% in the breastmilk group vs 25% in the exposed group the authors needed 300 patients to detect a difference. Secondary outcomes included detection of other allergens aside from allergy to cow’s milk.
Given that I called this a landmark study it might not be surprising to know that they found a difference favoring protection with human milk.
Also curious is the relationship to vitamin D levels. Previous research has documented an inverse relationship between vitamin D levels in children and risk of atopy. Why only the middle tertile in this study but not the higher tertile had less IgE response is unknown.
Perhaps even more surprising (at least to me) was that the risk of allergy at age 2 for other allergens was also lower.
Included in this lower risk was food allergy in general, risk of anaphylaxis and cow’s milk allergy that I presume manifested as rectal bleeding.
What Impact Could This Have
It is important to point out here that all these infants were ≥ 36 weeks so although I would love to infer that this strategy would have a huge impact on our preterm population I can’t say that yet (until a study is done). We certainly do see a fair bit of cow’s milk protein intolerance though that often leads to infants being placed NPO and on occasion worked up for NEC with a week or so of antibiotics. If this study is to be trusted, the rate of cow’s milk allergy was reduced from 6.6% to 0.7% in at risk infants (based on an immediate relative with atopy) and I would expect the risk in those without relatives to be less.
What might the impact be if we were to supplement with donor breast milk all term newborns who didn’t have enough maternal milk and take the elemental formula out of the equation entirely? If a 4 kg infant exclusively breastfed on day 1 and was give a couple ounces of supplement followed by full supplementation to 80 mL/kg/d on day 2 and then 100 mL/kg/d on day 3 that would total 26 ounces of donor milk in a worst case scenario assuming no maternal milk production during that time. At $4 per ounce we are looking at a cost to the system of about $100 a baby. Multiple that by the number of term infants in your centre to get an overall cost. In my own centre with about 12000 term deliveries a year that would come to 1.2 million dollars a year (again assuming no maternal milk at all). Is it worth the expense? I am not a health economist but I suspect if you were to add up the costs of workups/office visits etc for rectal bleeding, ED visits for asthma and anaphylaxis and the cost to families for food alleriges (let alone all the epi pens that need to be bought) it is worth it.
At the very least it does raise the question on post partum wards everywhere as to whether provision of cow’s milk formula should be one that someone has to consent to. With the publication of this study it certainly seems that it should be!
When I was growing up it was quite the rare thing to hear that someone had a food allergy. I knew of a few children with eczema or asthma but these days one can’t help but notice the warnings everywhere at schools, camps etc to avoid bringing high risk allergens into close proximity with potentially susceptible children. As a Neonatologist I don’t profess to be an expert in the area of Atopic conditions but I do believe as others have suggested that society’s obsession with antibacterial soaps and not playing the sandbox so to speak has led to an over development of the allergic response of our immune system due to lack of stimulation on the “infection fighting side” during our infant and toddler years. As a Pediatric resident I recall treating many children in clinic for atopic conditions and certainly atopic dermatitis was extremely common and incredibly frustrating for parents not to mention the children.
In October of 2014, in the Journal of Allergy and Clinical Immunology Simpson et al reported a pilot study that could change the approach from treatment to prevention for many children if the results hold true in larger studies. The article abstract can be found here with the full article being free as well: http://1.usa.gov/1BR1typ
In this pilot randomized control trial emolients were used in the treatment arm once a day for the first 6 months of life and had to be started within three weeks of life. All of these children were deemed to be high risk by having a first degree relative with allergic disease. The emolients used in the US group were sunflower oil, cetaphil cream and aquafor ointment and were applied to all skin surface except for the head. The theory was that skin breaks in newborns and infants allow allergens to reach the subcutaneous tissue and elicit an immune response thus creating allergic skin disease no different than an exposure to peanuts, tree nuts, dust mites etc can later trigger the immune system to cause allergic rhinitis or food allergy (Fig 1 from quoted paper below). Previous observational studies had yielded some support to this theory so the RCT was an attempt to answer the question as to whether keeping the skin smooth without breaks could prevent atopic eczema from happening.
The study was not powered to show a difference but the results of the study make this comment irrelevant. If a study is underpowered and no difference is seen between two groups one cannot say with certainty if no effect is seen that there was in fact no effect. If a difference is seen in even the smallest study between two groups in the primary outcome this is in fact significant as it was in this study.
The study found that the treatment group incidence of atopic dermatitis at 6 months was 22% vs a 43% incidence in controls. This corresponds to a 50% reduction in risk with a number needed to treat to prevent 1 case of atopic dermatitis of about 6 but with wide confidence intervals due to the small size of the study (3-1138). This speaks to the issue with small studies. Although the results seem impressive at first when you examine how confident you are with the estimate of the benefit it may be much less than originally thought. One thing to note though is that about 20% of families did not adhere to the daily application of emolient so one has to wonder if the results would have been even more impressive had the use been more consistent.
That all being said, applying a cream or ointment once a day to keep your newborns’ skin smooth and free of cracks seems to me to be something that many parents (especially those of us who live in the North) do anyway. Atopic dermatitis is a terrible condition that causes a lot of stress and discomfort for children and their families. If applying a cream or ointment once a day for the first 6 months of life was able to achieve such dramatic results and avoid such a condition I see very little harm in trying. Of course a larger study will now need to be done to confirm the cited paper here but I think this is a potentially exciting simple treatment to prevent a terrible condition!
I am a fan of simple treatments and even more so of simple interventions to prevent children from ever experiencing a condition at all. Might early use of emolients be one of these strategies?