Abstract
While human milk composition is characterised by marked dynamicity, we are far from having a clear picture of what factors drive this variation. Hormones in human milk are known to vary according to specific maternal phenotypes, but limited evidence shows the infant also has a role in determining milk composition. The present study aimed to investigate the interplay between maternal and infant characteristics in relation to human milk hormonal profile. In total, 501 human milk samples from mothers recruited in the Finnish STEPS cohort study (Steps to the healthy development) were analysed. Pre-pregnancy and pregnancy maternal data, socioeconomic status and infant characteristics at birth were collated. Leptin, adiponectin, insulin-like growth factor-1 and cyclic Glycine-Proline in milk were measured. Multivariate analysis of variance (MANOVA) and linear regression were utilised for statistical analysis. Sex-specific interactions with maternal factors were observed, as the infant sex mediated associations between gestational diabetes and milk adiponectin (p = 0.031), birth-mode and total protein (p = 0.003), maternal education and insulin-like growth factor-1: cyclic Glycine-Proline ratio (p = 0.035). Our results suggest that changes in human milk composition are associated with interactions between maternal and infant characteristics and pathophysiological factors. Future work should expand on these findings and further explore the link between hormonal profiles in human milk and infant outcomes.