Discussion
Maternal supplementation with 1000 IU cholecalciferol daily from 14 weeks’ gestation until delivery resulted in a reduced odds of infant atopic eczema at age 12 months. The effects of supplementation, however, were non-significant at ages 24 and 48 months. Interaction terms between supplementation during pregnancy and breastfeeding duration were not statistically significant, but sensitivity analysis showed that the protective effect of maternal cholecalciferol supplementation on infantile eczema was only significant in offspring breastfed for more than one month.
Current evidence relating to maternal vitamin D status and its effect on offspring atopic eczema is inconsistent, and evidence from supplementation trials is sparse. A U-shaped association between maternal vitamin D supply and status and offspring atopic eczema is biologically plausible, whereby both low and high intakes and 25(OH)D insufficiency and high 25(OH)D concentrations might be associated with increased risk of atopic eczema (see below for discussion of potential mechanisms).6 We would note that mothers with serum 25(OH)D levels above 100 mmol/L at baseline were excluded from MAVIDOS, but found no evidence for an increased risk of atopic eczema with 1000 IU/d cholecalciferol supplementation. A reduced risk of wheeze and eczema has been reported in children of mothers consumed 174 IU/d or more of dietary vitamin D during pregnancy15, and infants with cord blood 25(OH)D levels ≥ 75 nmol/L were found to have a lower risk of eczema in infancy when compared to those with cord blood levels of < 50 nmol/L.16 There are also reports of no association between maternal or cord serum 25(OH)D concentrations and atopic eczema.17 Others studies have reported no clear associations between maternal vitamin D status in late pregnancy and asthma, wheeze or skin sensitisation at age 1, 3 or 6 years.18
In the VDAART randomised controlled trial in women at high risk of having children with asthma, the prevalence of offspring asthma at age 6 years was similar in those whose mothers received antenatal supplementation with 4400 vs 400 IU/d vitamin D, as were the secondary outcomes of eczema and total IgE levels; a between-group reduction in asthma and recurrent wheeze was, however, suggested at early time points through the age of 3 years.7, 19 Our study examined the effect of gestational supplementation on offspring eczema at ages from 12 to 48 months, also finding an effect at early age, 12 months. We found no effect on eczema in the past year at ages 24 and 48 months suggesting that other, post-natal influences might become important at older ages in affecting the risk of atopic eczema beyond infancy. Conceivably, supplementation during the postnatal period may be needed for a sustained effect. There is evidence supportive of postnatal vitamin D supplementation, with a meta-analysis of 11 intervention studies in children with atopic eczema reporting a reduction in eczema severity.20
Vitamin D has immunomodulatory effects on innate and adaptive responses.21 The vast majority of cells of the adaptive immune system express the vitamin D receptor and CYP27B1, enabling the production of the active metabolite 1,25(OH)2 D, thought to act predominantly in an auto-and paracrine fashion. Evidence fromin vitro and in vivo studies22 has demonstrated that vitamin D supplementation inhibits expression of Th2 response cytokines, the predominant immune response seen acutely in atopic eczema and other allergic disease. 23Vitamin D deficiency in utero and early life has been linked with increased Th2 lymphocytes and reduced T regulatory cells and interleukin IL-10, leading to macrophage and dendritic cells producing pro-inflammatory cytokines.24 However, contrary to this, there is evidence that 1,25(OH)2 D promotes Th2 responses, with inhibition of IFN-γ and promotion of IL-4, IL-5, and IL-10 production.25 Skin barrier function is important in the pathogenesis of atopic eczema; vitamin D and its metabolites can impact this through involvement in the synthesis of proteins such as filaggrin, and through stratum corneum formation, keratinocyte formation and differentiation and production and the regulation of skin antimicrobial peptides. 26
Our data suggest that the effect of vitamin D supplementation on offspring eczema risk may be seen soon after pregnancy, but weaken as children grow older, where other risk factors can be influential. We speculate that during infancy there may be a role of breast milk vitamin D content. Evidence from MAVIDOS has demonstrated an increase in maternal serum levels with 1000 IU/d cholecalciferol supplementation,12 but in line with the Southampton Women’s Survey observational study,18 our data showed no association between maternal serum 25(OH)D in late pregnancy and offspring atopic eczema. Previous studies have shown that gestational vitamin D supplementation increases breast milk vitamin D content,8, 9 and in MAVIDOS the vitamin D content of breast milk is likely to have been higher in the supplemented group, influenced by mobilisation from maternal fat and muscle tissue. This may explain our finding of a protective effect only in children that were breastfed for more than 1 month. Heterogeneity in the aetiology and pathogenesis of atopic eczema in early childhood is increasingly recognised,27 and an alternative possibility is that vitamin D supplementation may only have an effect on particular atopic eczema phenotypes.
In MAVIDOS, examination of genetic variants in genes related to the vitamin D pathway has shown that rs12785878 (DHCR7) was associated with baseline 25(OH)D, likely influencing cutaneous synthesis; achieved 25(OH)D status following supplementation was associated with rs10741657 (CYP2R1, which determines the efficiency of vitamin D to 25(OH)D conversion), whereas rs12785878 and rs6013897 (CYP24A1) were not.10 There were weak trends for higher ORs of atopic eczema with rs10741657 (CYP2R1) at ages 12 and 24 months but no associations for other SNPs examined. In a case-control study of Chinese children, rs4674343 on CYP27A1 (27-hydroxylase, an enzyme converting the pre-vitamin D3 metabolite lumisterol into further downstream metabolites with biological activity in skin cells28) was reported to be protective against atopic eczema and CYP2R1 and VDR haplotypes also influenced atopic eczema risk and eosinophil count.29
A strength of this study is analysis of data from a placebo‐controlled, double‐blind, randomised trial. Atopic eczema was not the primary outcome in MAVIDOS, but the data collected enabled ascertainment of offspring atopic eczema using the UKWPDC well-recognised criteria for the diagnosis of atopic eczema. Furthermore, these criteria were determined by trained research nurses who examined the offspring. While some participants were taking vitamin D in addition to the intervention/placebo provided, supplement use at interview did not differ between the intervention and placebo groups. A limitation of this study is that cord blood and offspring 25(OH)D levels were not measured, precluding examination of these in relation to atopic eczema. Possible effects of UVB and any interaction with supplementation could not be investigated because no UVB exposure data were available. Our study population did not include many women who were not white and we could not include participants with baseline 25(OH)D concentrations of less than 25 nmol/L as a result of stipulations made during the ethics approval process; these reduce the generalisability of our findings.
In conclusion, in a randomised controlled trial maternal supplementation with 1000 IU/d cholecalciferol from 14 weeks’ gestation to delivery led to a reduced incidence of atopic eczema in the first year of life. Many international and national guidelines recommend 400-600 IU/d (10-15 micrograms) cholecalciferol throughout pregnancy, with the strongest evidence for the prevention of neonatal hypocalcaemia and emerging evidence for effects on other health outcomes affecting skeletal, respiratory and immune systems.30 The current findings inform understanding of the early life influences on infantile eczema and support recommendations for routine vitamin D supplementation during pregnancy.