Effect of Short‐Term Maternal Supplementation with Small Amounts of Vitamin A or Beta‐Carotene on Breast Milk Retinol Concentrations among Lactating Filipino Women

Abstract

This study assessed the effect of daily supplementation with low doses of vitamin A (VA, as retinyl palmitate) or β‐carotene, in amounts that could be provided by fortified or biofortified foods, on breast milk and plasma retinol concentrations after 1 d or 3 wk of supplementation, and after 1 wk of discontinuing supplementation.Lactating women (18–45 y; n=85) were randomly assigned to receive capsules containing 600 μg RAE/d as either retinol (VA group) or β‐carotene (BC group) in corn oil, or 0 μg RAE/d (oil only; control). The daily dose was delivered as 2 capsules/d along with low‐VA meals, 6 d/week for 3 wk. Fasting blood samples were collected at baseline (d 0) and after 3‐wk of supplementation (d 22). Fasting breast milk samples were collected on days 0, 1, 2, 22 and 29, using both “full” and “casual” milk collection methods. Milk was also collected before and after a single meal (d1AM and d1PM) to assess the effect of one dose of VA or BC on milk VA. Plasma and milk retinol and BC concentrations were measured by HPLC, inflammatory markers by ELISA, milk fat by creamatocrit, and β‐carotene‐15,15″‐mono‐oxygenase (BCO1) polymorphisms by PCR.At baseline, the prevalence of low plasma retinol (22% <1.05 μmol/L) and low milk retinol (64% (full) and 53% (casual) <28 nmol/g fat) did not differ by group. After 3 wk of supplementation, controlling for baseline, mean plasma retinol concentration was greater in the VA group (inflammation‐adjusted mean±SE 1.56 ± 0.29 μmol/L) relative to the control group (1.31 ± 0.44 μmol/L; P<0.01), whereas that of the BC group (1.48 ± 0.33 μmol/L) did not differ from either the VA or control groups (P>0.1). Controlling for baseline, mean milk retinol concentration at d 22 was greater in the VA group relative to the control group (full: 33.8 ± 18.4 vs 24.0 ± 7.3 nmol/g fat; casual: 32.0 ± 13.4 vs 24.1 ± 9.4 nmol/g fat; P<0.01); this difference persisted 1 wk after discontinuing supplementation (day 29; P<0.05). In full milk samples, the mean milk retinol concentration in the BC group (d 22: 27.8 ± 10.1 nmol/g fat) did not differ from that of the control group at d 22 or d 29. In casual samples, the mean milk retinol concentration was greater in the BC group (d22: 29.2 ± 9.4 nmol/g fat) vs the control group on day 22 (P<0.05) but not d 29, and did not differ from the VA group at either time. Mean milk retinol concentrations did not increase in response to a single VA or BC capsule (d1AM vs d1PM), or to a single day of VA or BC capsule receipt (d1AM vs d2AM) relative to the control group. BCO1 variants were prevalent but did not modify the effect of BC capsules on plasma or milk retinol (although power to detect interactions was limited).In summary, an effect of supplementation with small daily doses of retinol was detected in both plasma and breast milk after only 3 wk of supplementation, suggesting that breast milk is useful for evaluating food‐based interventions with retinol, such as fortification. The effect of BC was less consistent, so evaluation of the response to increased provitamin A intake may require a larger sample size or longer duration of supplementation.Support or Funding InformationThis study was supported by the Bill & Melinda Gates Foundation.