Subgroup analyses of a randomized trial of DHA supplementation for infants born preterm with assessments of cognitive development up to 7-years of age: What happens in infants born <29 weeks' gestation?

A recent trial showed that high-dose docosahexaenoic acid (high-DHA) supplementation of infants born <29 weeks' gestation improves intelligence quotient (IQ) at five years' corrected age. However, this finding has not been detected by other trials of DHA, which either did not measure IQ or included more mature infants. We analyzed the subgroup of 204 infants born <29 weeks' from our earlier randomized trial of high-DHA (∼1 % total fatty acids) or standard-DHA (∼ 0.3 % total fatty acids). Participants were assessed for cognition at 18 months, and IQ and behavior at seven years' corrected age. No group differences were detected for mean cognitive, IQ or behavior scores. At 18 months, 18.8 % of children in the high-DHA group had a cognitive score <85, compared with 31.1 % of children in the standard-DHA group, but at seven years there was no difference. Although an underpowered post-hoc subgroup analysis, this study provides limited support to recommendations that infants born <29 weeks' gestation require supplemental DHA.

The influence of DHA supplementation during pregnancy on language development across childhood: Follow-up of a randomised controlled trial.

Numerous randomised controlled trials have explored the effect of docosahexaenoic acid (DHA) supplementation in early life on neurodevelopment, with some suggested positive effects on language. Australian women with a singleton pregnancy <21 weeks' gestation were randomised to receive 800 mg DHA/day or a placebo until birth. A sample of 726 children (all n=96 born preterm, random sample of n=630 born at term) were invited to undergo assessments of language, academic, and language-based cognitive abilities at 1.5, four and seven years of age. No group differences were detected for any group comparison. Exploratory analyses for sex by treatment interactions revealed a possible adverse effect of DHA supplementation on the language of females at 1.5 years but no effects on outcomes at four or seven years. Taken as a whole, evidence of an effect of prenatal DHA supplementation on language abilities across childhood is negligible and could be a chance finding.

Maternal characteristics influence response to DHA during pregnancy.

We explored the degree to which maternal and offspring outcomes resulting from consuming prenatal docosahexaenoic acid (DHA, 800mg/day) in a clinical trial were influenced by maternal characteristics. Among non-smokers, women who received DHA had heavier babies (adjusted mean difference (MD)=99g 95% CI 45-153, p<0.01; interaction p=0.01) and fewer low birth weight babies than control women (adjusted relative risk=0.43 95% CI 0.25-0.74, p<0.01; interaction p=0.01). From women who had not completed further education, children in the DHA group had higher cognitive scores at 18 months compared with control children (adjusted MD=3.15 95% CI 0.93-5.37, p=0.01; interaction p<0.01). Conversely, the children of women who completed further education in the DHA group had lower language scores than control children (adjusted MD -2.82 95% CI -4.90 to -0.73, p=0.01; interaction p=0.04). Our results support the notion that responsiveness to prenatal DHA may depend on the characteristics of specific population subgroups.