Maternal 25-hydroxyvitamin D and its association with childhood atopic outcomes and lung function.
Bottom Line: Sixty-eight per cent of mothers had sufficient (> 50 nmol/L) concentrations of 25(OH)D, 27% were insufficient (27.5-49.99 nmol/L) and 5% were deficient (< 27.5 nmol/L).These findings remained after adjustment for season of measurement and for potential confounders.We found no evidence that maternal blood 25(OH)D concentration in pregnancy is associated with childhood atopic or respiratory outcomes.
Affiliation: MRC Centre for Causal Analyses in Translational Epidemiology, University of Bristol, Bristol, UK; School of Social and Community Medicine, University of Bristol, Bristol, UK.Show MeSH
Related in: MedlinePlus
Mentions: Figure 3 shows the associations of maternal 25(OH)D with each of the lung function outcomes (these are also tabulated in Online Repository Table S3). In unadjusted models, there was a weak suggestion of a threshold association between maternal 25(OH)D and FVC, FEV1 and BHR – offspring of mothers in the lowest quintile of 25(OH)D (< 38 nmol/L) tended to have poorer lung function and a higher odds of BHR compared with those from mothers in the higher quintiles. This pattern was attenuated after adjusting for potential maternal confounders and in particular for seasonality and the potential mediating effect of offspring growth. Furthermore, while this appeared to a be a qualitative dose–response relationship, there was actually no statistical evidence to support this overall relationship, and reanalysing 25(OH)D as categories of deficient, insufficient and sufficient 23 and using a threshold of 75 nmol/L also showed results for a test of overall association (see Figures S5, S7 & S8 in online repository). As a post hoc analysis to further examine the suggestive trends in Fig. 3, we fitted equivalent models using age, height and sex standardized lung function outcomes at 15 years and there was no evidence that the patterns described above existed at this later age (see Figure S10 in Online Repository). There was a suggestion of an overall association between seasonally adjusted maternal 25(OH)D and FEV1 and FEF25-75 at 15 years although the pattern across quintiles was inconsistent, for example, compared with quintile 1, quintiles 2 and 4 had positive associations with FEV1 and FEF25-75, whereas quintiles 3 and 5 showed no association.
Affiliation: MRC Centre for Causal Analyses in Translational Epidemiology, University of Bristol, Bristol, UK; School of Social and Community Medicine, University of Bristol, Bristol, UK.