Literature-based human milk nutrient composition values for use in North American food composition databases

BACKGROUND: The profile for human milk (HM) nutrient composition jointly used by Canada and the United States [Standard Reference (SR), legacy] is largely based on >40-y-old studies. In 2018, it was deemed unsuitable by the United States Department of Agriculture for estimating current nutrient exposures. OBJECTIVES: To review data available in the literature to develop interim HM nutrient profile data values. METHODS: Two reviewers screened and extracted data from 3 recent systematic reviews covering the period from 1980‒2022 and 1 large Canadian biomonitoring study. Eligible studies: reported quantitative concentrations of ≥1 component of interest (energy, macronutrients, vitamins, minerals, or fatty acids), included data specific to United States or Canadian participants, reported on mature milk (>21 d postpartum) for term-born infants (>37 wk of gestation), had samples collected within the first 6 mo of lactation, and used appropriate methods for milk collection and nutrient analysis. Studies were combined as weighted means and pooled standard deviations and compared with the existing SR and international literature. RESULTS: Updated data were identified for >40 HM components, including some that had previously been assigned 0 in SR legacy (e.g., DHA), or estimated from other foods (e.g., vitamin K). Concentrations for interim HM nutrient profile data differed by >20% from values in the SR legacy for total fat, iron, manganese, and most vitamins and fatty acids. Eligible data were lacking for niacin, vitamin B-12, vitamin C, and total vitamin D, and were only available from single studies for manganese, β-carotene, thiamin, riboflavin, pantothenic acid, vitamin B6, total choline, 25-hydroxyvitamin D, vitamin E, cholesterol, and some fatty acids. CONCLUSIONS: These findings summarize data available to develop an interim nutrient profile for North American HM that could be used until an empirically measured profile can be developed. They also highlight gaps in the literature to be addressed by future studies.