Statistics from Altmetric.com
Editor,—The study by de Jong et al tries to demonstrate that the administration of three or more cows’ milk supplements to healthy full term newborn infants during the first three days of life does not increase the risk of atopy in the first two years.1 In our opinion, the study does not demonstrate this hypothesis.
Allergy to cows’ milk can occur as an IgE mediated anaphylactic reaction. This is well documented in infants given cows’ milk in the first months of life after being sensitised in the first few days.2 3 Anaphylactic reactions are rare and limited to infants at risk, and we think that this study does not have the power to rule out this negative effect of cows’ milk. Delayed allergy is more common, but it probably depends on a larger and sustained exposure to the antigens of cows’ milk. The three feeds given to the study infants may not be sufficient to trigger this mechanism and increase the risk of atopic symptoms. The studies reporting an increased incidence of atopy were generally based on the sustained administration of cows’ milk in the first few months of life.4 In addition, de Jong et al do not consider the incidence of atopy in a group of exclusively breast fed infants, which might be as low as 0.5% per year3
their study subjects were, by definition, not exclusively breast fed by WHO criteria5
they do not say what they mean by exclusive breast feeding after the end of the intervention
in the absence of a clear definition, they do not report how the 46% of infants classified as not exclusively breast fed to six weeks of age were distributed in the two groups.
The potential bias introduced by an unequal distribution of this ill defined variable in the two groups might outweigh the effect of the brief earlier intervention. For all these reasons, we are not surprised by the findings and we think that this trial will keep the more than 50 year old controversy unresolved.6
Suppose the conclusion of the authors is true—that is, that early and brief exposure to cows’ milk in breast fed children does not increase the risk of atopic disease. This may be enough to save the lives of midwives who (inadvertently?) administer formula supplements to healthy, full term, newborn babies,7 but it is certainly not enough to justify such a common non-evidence-based practice. The advantages of exclusive breast feeding are so many, well beyond the postulated effect on atopy, and the role of early supplements in decreasing the duration of exclusive breast feeding is so clear that prescribing supplements without indication should still be considered inappropriate.8 In fact, we wonder whether it is ethical, despite the approval of an ethics committee, to give supplements without indication as done in this trial. Finally, we believe that it would have been preferable that this study be carried out without the support of a party (a baby food company) with obvious interests in the use of formula supplements.
Dr de Jong et al comment:
Cows’ milk allergy is a problem, even in exclusively breast fed infants; therefore, investigation of possibilities to decrease the incidence is unlikely to be questioned. Induction of immunological tolerance by early oral antigenic stimulation is a well established concept in mucosal immunology and supported by data from Lindfors and Enocksson.1-1 However, such an approach is in marked contrast to the “dangerous bottle” hypothesis and therefore requires careful attention to its possible risks.1-2
Our results are disappointing to some extent, in the sense that the protocol we used did not induce specific tolerance. On the other hand, our results are positive in that we did not find support for the “dangerous bottle”. This result is relevant both for a number of practical issues as well as from the conceptual point of view: it facilitates the testing of other tolerance induction protocols—of course, any such an investigation should not be embarked on lightly.
Our scientific approach vis a vis the dangerous bottle hypothesis is fully in line with the principles of evidence-based practice: a double blind, placebo controlled study to substantiate a notion based on anecdotal information. The power of the study is indicated by the 95% confidence intervals; even if the analysis is restricted to the per protocol analysis, the power is sufficient to conclude that the relative risk is less than 1.7.
Tamburlini et al seem to be disappointed by our results, because they feel it weakens the case for breast feeding. However, as they justly point out, the scientific case in favour of breast feeding is strong, so there is no need to be sorry for the loss of a non-evidence-based argument.
We accept that the introduction of food or drink other than breast milk should be discouraged as a general practice, but in our study 90% of the mothers who intended to breast feed actually did at two weeks after delivery.
We did not investigate the effect of prolonged exposure to cows’ milk and thus did not address the controversy discussed in Kramer’s famous review.1-3
We are currently involved in a follow up of the cohort, so we hope to be able to report on the effects (if any) of brief early exposure to cows’ milk at the more informative age of 6 years.