Premium
Human Milk Oligosaccharides and Food Sensitization at 12 months in the CHILD Cohort
Author(s) -
Azad Meghan B,
Robertson Bianca,
Sharma Atul K,
Subbarao Padmaja,
Becker Allan B,
Mandhane Piushkumar J,
Turvey Stuart E,
Lefebvre Diana L,
Sears Malcolm R,
Bode Lars
Publication year - 2017
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.31.1_supplement.448.2
Subject(s) - medicine , interquartile range , sensitization , cohort , confounding , infant formula , physiology , immunology , pediatrics
Rationale Human milk oligosaccharides (HMOs) provide selective substrates for infant gut microbiota and influence immune system maturation. HMOs appear to modify allergic disease development in rodent models, but this has not been confirmed in humans. Methods We studied a representative subset of 421 mother‐infant dyads from the Canadian Healthy Infant Longitudinal Development (CHILD) birth cohort. Milk samples were collected at 4 months postpartum (median 16 weeks, interquartile range 14 – 19) and analyzed by rapid high‐throughput HPLC to quantify the 20 most abundant HMOs. Skin prick testing was performed at 12 months to determine sensitization (≥2mm wheal) to egg, milk, soy and peanut. Associations were examined for total and individual HMO levels (univariate analysis: Wilcoxon rank‐sum test) and overall HMO profiles (multivariable classifier: partial least squares discriminant analysis tested for significance by leave‐one‐out‐cross‐validation). Discriminant scores were examined in regression models with adjustment for potential confounders and known allergy risk factors. Results Overall, 59/421 infants (14.0%) were sensitized to one or more food allergen by 12 months. Total HMO levels were similar in milk consumed by sensitized vs. non‐sensitized infants (14,010 ± 3,510 vs. 13,937 ± 4,085 nmol/mL; p=0.96) and individual HMO levels were not significantly associated with food sensitization (all p>0.09). However, overall HMO profiles differed significantly (mean discriminant score: 0.129 vs. 0.209, p<0.001; area‐under‐the‐curve: 0.73, 95%CI 0.66–0.79). This association was robust to cross‐validation and remained significant in adjusted models, where each 0.1‐unit increase in HMO discriminant score was associated with a 2.5‐fold increased odds of food sensitization (adjusted OR 2.46, 95%CI 1.78–3.41, p<0.001), independent of maternal food allergy, ethnicity, household pets, and lactation time postpartum. HMO profiles associated with higher odds of food sensitization were characterized by lower concentrations of fucosyl‐disialyllacto‐N‐hexaose (FDSLNH), lacto‐N‐fucopentaose‐I (LNFPI), LNFPII, and higher concentrations of lacto‐N‐hexaose (LNH) and lacto‐N‐tetraose (LNT). Conclusion HMO composition is associated with food sensitization in the first year of life. Ongoing research in the CHILD study will seek to confirm these findings in a larger sample of mother‐infant dyads, and evaluate associations with confirmed food allergy at 3 and 5 years of age. Support or Funding Information This study was funded by Research Manitoba and supported by the Canadian Institutes of Health Research and the Allergy, Genes and Environment Network of Centres of Excellence.