Background: The effect of mode of infant feeding on adiposity deposition is not fully understood.
Objective: The objective was to test the hypothesis that differences in total and regional adipose tissue content and intrahepatocellular lipid (IHCL) arise in early infancy between breast- and formula-fed infants and to describe longitudinal changes.
Design: This prospective longitudinal cohort study was performed in 2 hospitals in the United Kingdom. Healthy, full-term, appropriate weight-for-gestational age infants were recruited; adipose tissue volume and distribution were directly quantified by using whole-body magnetic resonance imaging; IHCL was assessed by in vivo proton magnetic resonance spectroscopy. Measurements were performed after birth (median age: 13 d) and at 6-12 wk of age. Method of infant feeding was recorded prospectively by using maternally completed feeding diaries. Breastfed was defined as >80% of feeds consisting of breast milk at both points; formula-fed was defined as >80% of feeds consisting of formula milk at both points.
Results: Longitudinal results were obtained from 70 infants (36 breastfed, 9 mixed-fed, and 25 formula-fed). No differences were found in total or regional adipose tissue or IHCL between breastfed and formula-fed infants. In pooled analyses including all feeding groups, IHCL and total adipose tissue approximately doubled between birth and 6-12 wk: IHCL after birth (median: 0.949; IQR: 0.521-1.711) and at 6-12 wk (1.828; 1.376-2.697; P < 0.001) and total adipose tissue after birth (0.749 L; 0.620-0.928 L) and at 6-12 wk (1.547 L; 1.332-1.790 L; P < 0.001). Increasing adiposity was characterized by greater relative increases in subcutaneous than in internal adipose tissue depots.
Conclusions: No differences were detectable in adipose tissue or IHCL accretion between breastfed and formula-fed infants up to 2 mo. The substantial increase in IHCL seen over this period in both breastfed and formula-fed infants is a novel observation, which suggests that hepatic storage of lipids may be physiologic up to 2 mo. This trial was registered at www.clinicaltrials.gov as NCT02033005.