Growth to age 18 months following prenatal supplementation with docosahexaenoic acid differs by maternal gravidity in Mexico

Abstract

Little is known about the long-term effects of DHA intake during pregnancy. Offspring of primagravid Mexican women who received 400 mg/d DHA from wk 20 of gestation through delivery were heavier and had larger head circumferences at birth than children whose mothers received placebo; no effect was observed in offspring of multigravidae. We have followed these children (n = 739; 76% of the birth cohort), measuring length, weight, and head circumference at 1, 3, 6, 9, 12, and 18 mo. At 18 mo, intent-to-treat differences between placebo and DHA, adjusted for maternal height and child sex and age at measurement, were: length, -0.21 cm (95% CI = -0.58, 0.15); weight, -0.03 kg (95% CI =-0.19, 0.13); and head circumference, 0.02 cm (95% CI = -0.18, 0.21) (all P > 0.05). There was heterogeneity of associations by maternal gravidity for weight (P < 0.08), length (P < 0.02), and head circumference (P < 0.05). Among offspring of primagravid women, length at 18 mo was increased by 0.72 cm (95% CI = 0.11, 1.33) following DHA supplementation, representing 0.26 length-for-age Z-score units; among offspring of multigravidae, the estimate was -0.13 cm (95% CI = -0.59, 0.32) (P > 0.5). Maternal DHA supplementation during the second half of gestation may enhance growth through 18 mo of children born to primagravid women.

Trial registration: ClinicalTrials.gov NCT00646360.

Conflict of interest statement

Author disclosures: A. Stein, M. Wang, R. Martorell, L. Neufeld, R. Flores-Ayala, J. Rivera, and U. Ramakrishnan, no conflicts of interest.

Figures

FIGURE 1

Difference in length among 739 children born to women who participated in a trial of DHA in pregnancy and who were followed through 18 mo of age, by treatment assignment and maternal gravidity. Data are differences attributable to DHA and 95% CI and are adjusted for maternal height and offspring sex and age at assessment. *P < 0.05; **P < 0.01 for heterogeneity of estimates by parity.