The Relationship Between Vitamin D and Brain Morphology in Children

Post by Kasey Hemington

What's the science?

Maternal vitamin D deficiencies during gestation have been associated with cognitive or neuropsychiatric disorders in offspring, such as autism, lower IQ, and schizophrenia. However, the neurobiological basis of these links (for example, between autism and low vitamin D) remain unclear. Further, our understanding of how gestational vitamin D affects brain morphology during development is limited. In animal studies lower gestational vitamin D has been associated with smaller brain volumes, however, there have been no investigations of the effects of vitamin D on brain morphology in human children. This week in NeuroImage, Zou and colleagues used different structural magnetic resonance imaging (MRI) techniques to examine the relationship between gestational vitamin D levels in human mothers and the brain morphology of their offspring.

How did they do it?

As part of the prospective cohort Generation R study (Rotterdam, the Netherlands), the authors included data from 2597 mother-child dyads, where gestational vitamin D concentration information and structural MRI brain scans of the children (aged between 9-11 years) were available. Vitamin D concentration information was obtained via maternal blood-samples at mid-pregnancy, and at birth from the umbilical cord. Some dyads only had vitamin D concentration data at one of the time points and 1536 dyads had data at both time points. In their statistical analysis of the relationship between vitamin D concentration and different measures of brain morphology, the authors also included data on factors that could potentially confound the relationship: maternal age, ethnicity, socioeconomic factors, alcohol and drug use, vitamin supplement use, and the season at which the vitamin D concentration was sampled. Vitamin D concentration was studied both as a continuous variable and as a categorical variable, where categories were defined as ‘deficient’ (<25 nmol/L) ‘insufficient’ (25-50 nmol/L) and ‘sufficient’ (>50 nmol/L).

The authors assessed 1) The relationship between gestational vitamin D concentrations at mid-pregnancy and the child’s total brain volume, cortical grey matter volume, cortical white matter volume and cerebellar volume using multiple linear regression, 2) Whether having sufficient vitamin D concentration at only one time point or both time points was associated with different brain volumes, 3) The relationship between gestational vitamin D concentration and the brain volume of subcortical structures and brain ventricles, and 4) The relationship between vitamin D concentration and surface-based brain metrics: cortical thickness, surface area, and gyrification (the curvature or folding of the brain’s cortical surface). For each regression analysis, the authors created one model (‘Model 1’) with age at MRI scan and the child’s sex as covariates, and a second model (‘Model 2’) which included the other potential confounders described above.

What did they find?

Vitamin D concentration at mid-pregnancy and at birth was positively associated with the children’s total brain volume, total grey matter, and total white matter in Model 1, but not when other potential confounders were included (Model 2). Vitamin D concentration was positively associated with cerebellar volume in both models 1 and 2, but this relationship did not survive when the authors corrected their statistical analysis for multiple statistical comparisons. There was no association between vitamin D concentration category (deficient, insufficient, sufficient) and any measures of brain volume in model 1 or 2. However, smaller cerebral grey matter volumes were found in children who were ‘consistently insufficient’ or ‘consistently deficient’ (not sufficient at either time point) versus children who were ‘consistently sufficient’ children after correction for multiple statistical comparisons. Smaller total brain cerebral white matter volumes were also seen in ‘consistently deficient’ children compared to the ‘consistently sufficient’ group. No relationships were found between subcortical or ventricle volume and vitamin D concentration. Finally, when the authors assessed cortical thickness, surface area, and gyrification in other groups compared to the ‘consistently sufficient’ group, they found smaller surface area of temporal region of the right hemisphere of the brain in children with ‘consistently insufficient’ vitamin D concentrations and smaller frontal and occipital surface area in the right hemisphere and less gyrification in the left hemisphere in those ‘consistently deficient’ in models 1 and 2.

What's the impact?

This study, which is the first longitudinal study to assess the relationship between brain morphology in children and gestational vitamin D levels, found that consistently low vitamin D levels were associated with smaller brain volumes and different brain surface area and gyrification in children. Studies such as this could help to provide insight on the link between vitamin D deficiencies and neurodevelopment or neuropsychiatric abnormalities. 

Zou et al. A prospective population-based study of gestational vitamin D status and brain morphology in preadolescents. NeuroImage (2020). Access the original scientific publication here.