An investigation of association between human milk mineral patterns and infant growth.

Abstract

Human milk is widely acknowledged as the optimal food for infant aged 0 ~ 6 months. While there has been extensive documentation on the mineral and trace element composition of human milk, results on the relationship between mineral content and infant growth remain mixed. This cross-sectional study aims to explore human milk mineral patterns and to investigate associations between human milk mineral patterns, human milk metabolomic profile and infant growth. A total of 200 breast milk samples from seven cities in China was included. Human milk mineral and trace elements was detected by inductively coupled plasma mass spectrometer (ICP-MS). K-means cluster analysis was utilized to derived human milk mineral patterns. Untargeted human milk metabolomic profiles was determined using high performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS). Differences of infant growth rate and metabolomic profiles were then compared across patterns identified. Three human milk mineral patterns were identified. Cluster I was characterized as the highest levels of potassium, magnesium and calcium, while the lowest levels of copper, zinc, manganese and selenium. Cluster II showed the most abundant sodium, iron, zinc, manganese and selenium. Cluster III had the lowest levels of sodium, potassium, magnesium, iron and calcium. Infants of cluster I showed significantly higher length-for-age z score (0.60 ± 2.03, p = 0.03). Compared with other clusters, samples of cluster I showed lower expression of metabolites of arachidonic acid (ARA) and nicotinate and nicotinamide metabolism pathway. A human milk mineral pattern was identified which is related to increased infant growth rate and altered metabolic signature. Future work is needed to understand these human milk patterns in terms of biologic mechanisms and generalization to other populations.