634 - Understanding the Influence of Maternal Health on the Human Milk Metabolome: Preliminary Results from the MOM2CHild Study

Publication Number: 634.6847

Abigail M. Galyon, University of California, Davis, Davis, CA, United States; Xuan He, University of California, Davis, Davis, CA, United States; Shannon C. Conrey, University of Cincinnati College of Medicine, Cincinnati, OH, United States; Laurie Nommsen-Rivers, University of California Davis, Davis, CA, United States; Allison R. Burrell, Cincinnati Children's Medical Center Hospital, CINCINNATI, OH, United States; Mary Allen Staat, CCHMC, Cincinnati, OH, United States; Ardythe L. Morrow, University of Cincinnati College of Medicine, Cincinnati, OH, United States; Carolyn M. Slupsky, University of California, Davis, Davis, CA, United States

Background: Human milk (HM) contains low molecular weight metabolites that may benefit the infant and provide insight into the metabolic state of the mother. Maternal obesity, diabetes, and hypertension have been associated with variations in these HM metabolites. However, few studies have examined these relationships concurrently in a large and diverse cohort.

Objective: To investigate the associations between maternal obesity, diabetes, and hypertension and HM metabolites using samples and data from the MOM2CHild study.

Design/Methods: HM samples (n=302) were collected at 2 weeks postpartum and subjected to metabolomics analysis using 1H-nuclear magnetic resonance spectroscopy. A total of 59 metabolites were identified. Metabolite concentrations were expressed in micromolar (uM) and log10-transformed. Maternal BMI was calculated using self-reported pre-pregnancy weight and height and categorized as obesity (BMI > 30 kg/m2) or no obesity (BMI < 30 kg/m2). Maternal diabetes and hypertension were classified as none (0), gestational (1), or pre-existing (2). The Wilcoxon rank-sum test, corrected for multiple comparisons using the Benjamini–Hochberg procedure, was used to analyze the relationship between maternal obesity and individual HM metabolites, with a false discovery rate (FDR) corrected p-value < 0.05 considered statistically significant. Dunn’s tests examined relationships between categories of maternal diabetes or hypertension and individual HM metabolites.

Results: Of the 302 mothers, 36% had obesity, 15% had diabetes (4% pre-existing, 11% gestational), and 18% had hypertension (9.5% pre-existing and 8.5% gestational). Our preliminary findings indicate that maternal obesity was associated with significant differences in 7 HM metabolites, maternal diabetes with 1 HM metabolite, and maternal hypertension with 15 HM metabolites (all p < 0.05). These metabolites included derivatives of energy and fat metabolism, free amino acids, and microbial fermentation products.

Conclusion(s): Maternal obesity, diabetes, and hypertension may give rise to distinct metabolic signatures in HM. Further research will work to understand how these signatures may be related to breastfeeding and infant outcomes.