OBJECTIVE. Our aim was to investigate the role of measles vaccination and measles infection in the development of allergic disease and atopic sensitization.
METHODS. A total of 14 893 children were included from the cross-sectional, multicenter Prevention of Allergy–Risk Factors for Sensitization in Children Related to Farming and Anthroposophic Lifestyle study, conducted in 5 European countries (Austria, Germany, the Netherlands, Sweden, and Switzerland). The children were between 5 and 13 years of age and represented farm children, Steiner-school children, and 2 reference groups. Children attending Steiner schools often have an anthroposophic (holistic) lifestyle in which some immunizations are avoided or postponed. Parental questionnaires provided information on exposure and lifestyle factors as well as symptoms and diagnoses in the children. A sample of the children was invited for additional tests, and 4049 children provided a blood sample for immunoglobulin E analyses. Only children with complete information on measles vaccination and infection were included in the analyses (84%).
RESULTS. In the whole group of children, atopic sensitization was inversely associated with measles infection, and a similar tendency was seen for measles vaccination. To reduce risks of disease-related modification of exposure, children who reported symptoms of wheezing and/or eczema debuting during first year of life were excluded from some analyses. After this exclusion, inverse associations were observed between measles infection and “any allergic symptom” and “any diagnosis of allergy by a physician.” However, no associations were found between measles vaccination and allergic disease.
CONCLUSION. Our data suggest that measles infection may protect against allergic disease in children.
Dear Editor:
We commend the authors of the March 2009 Pediatrics article titled “Allergic Disease and Atopic Sensitization in Children in Relation to Measles Vaccination and Measles Infection” for conducting an interesting study and highlighting numerous potential confounders in their report. However, we wish to highlight several issues that warrant further discussion.
The first issue is the disconnect between subjective outcome measures “allergic symptoms” and “diagnosis of allergy by a physician” reported by parents versus the objective measure of atopic sensitization. While the authors point out that excluding children with history of wheezing and/or eczema during the first year of life revealed an inverse association between measles infection and subjective outcomes allergic symptoms/diagnosis of allergy, they do not point out in their conclusions the absence of association with respect to the objective outcome of atopic sensitization. It is hard to attribute this to the decreased sample size, as the odds ratios are close to 1 even in the higher IgE level group.
Additionally, the authors do a nice job of pointing out that the prevalence of allergic disease was higher among Steiner children who provided blood specimens versus Steiner children who did not provide blood specimens, and adjust the analysis to determine the magnitude of this bias (results not presented). We are concerned about another possible bias, that due to the reference group and the group with measles being composed mostly of Steiner-school children. Study group was included in the models developed, but there is no mention of possible interaction between study group and allergy symptoms or diagnosis, or whether interactions were assessed in the models. The authors note a lack of heterogeneity in results when the study groups were analyzed separately, but they present no data. It would be helpful to the reader, in assessing the extent to which Steiner-school children dominated the analysis, to have results (including raw data) for symptoms/diagnosis and atopic sensitization presented separately by study group.
Finally, we believe that the authors neglected to discuss the critical issues of plausible mechanism and risk/benefit analysis of the associations reported. If the underlying immune mechanism can be simplified to a Th1 versus Th2 skew, it is plausible that measles infection and to a lesser extent [live-attenuated] measles vaccination protects from atopic disease, as suggested by the analysis of all children with larger sample size. Regardless of whether measles vaccination is less protective than measles infection or if measles infection alone is protective against atopy, the risk of having measles infection versus the risk of having asthma, eczema, and rhinoconjuntivitis should be considered and discussed.
Sincerely,
Jakub Simon, M.D., M.S. Pediatric Infectious Disease Specialist Center for Vaccine Development University of Maryland, Baltimore
William Blackwelder, Ph.D. Biostatistician Center for Vaccine Development University of Maryland, Baltimore
Conflict of Interest:
None declared