Objectives
Mechanistic data have suggested a role of vitamin A in the celiac disease immunopathology1. We aimed to investigate if higher dietary vitamin A in pregnancy or early life associate with the risk of celiac disease.
Methods
In 85,244 children from the Norwegian Mother, Father, and Child Cohort study, we estimated total dietary vitamin A (retinol and beta-carotene from food and supplements) in pregnant women and from supplements at age 6 and 18 months, from questionnaires. Celiac disease diagnosis was collected from the Norwegian Patient Registry. SNPs tagging for HLA and involved in vitamin A metabolism were available in about 65,000 children (76%).
Results
We identified 1,362 (1.6%, 60.5% females) children with a celiac disease diagnosis, at mean follow-up 16.0 years (range 12.5-19.8). There was no association of total vitamin A intake in pregnancy (mean 1668 µg/day [SD 927], aOR=1.00 [0.99-1.01], per 100ug), retinol (aOR=1.00 [0.99-1.01]) or beta-carotene (aOR=1.00 [1.00-1.00]) with the child’s risk of celiac disease (adjusted for maternal celiac disease, education level, smoking, pregnancy gluten intake and diet quality). Vitamin A from supplements in pregnancy (reported in 75.8%) was associated with a borderline higher risk of child celiac disease (aOR=1.01 1.00-1.01], per 100µg/day, use versus no use aOR=1.18 [1.02-1.35]). Results were similar in children carrying HLA DQ2.5 and/or DQ8 haplotypes. Vitamin A from supplements at age 6 months (reported in 57.6%, use versus no use aOR=0.96 [0.85-1.08]) or at age 18 months (reported in 73.4%, aOR=1.04 [0.90-1.20]) was not associated with the risk of celiac disease (additionally adjusted for the child’s sex, gluten intake, infectious episodes). Results for associations with SNPs related to vitamin A status will be presented at the ICDS congress.
Conclusion
Overall, our prospective large-scale study did not support an association between higher dietary vitamin A in pregnancy or early life, and the subsequent risk of celiac disease.