Bone Abstracts

Evidence is unclear on the effect of vitamin D intake on vitamin D status in children.

Objective: In a meta-analysis, investigate the effect of vitamin D supplements and/or fortified foods on vitamin D status, using the biomarker 25-hydroxyvitamin D (25(OH)D) in children 2–18 years.

Methods: Eligible studies were randomized placebo-controlled trials, published in English, in children 2–18 years that compared vitamin D supplements or fortified foods. Using PRISMA guidelines, literature searches of Ovid MEDLINE, PubMed, CINAHL, Embase, and Cochrane Central Register of Controlled Trials were conducted up to December 2016. The Cochrane qualitative bias tool and the Jadad scale assessed evidence strength and I² assessed heterogeneity. Subgroups included age (2–8, 9–18 years), baseline 25(OH)D (<30, 30–49.9, ≥50 nmol/l), latitude (≥40° N or S, <40° N or S) and daily supplements, fortified foods or high dose injections.

Results: We included 29 trials (4972 children) with interventions (10 using fortified foods, 17 using supplements, 2 using bolus injections) from 2.5 to 100 μg/d vitamin D equivalent over 4 weak to 2 years. Due to the variation in design, heterogeneity was high (I²=73%). Once adjusted for dose, heterogeneity was low (I²=0%). Study designs were qualitatively high and 97% had Jadad scores ≥4. The 25(OH)D weighted mean difference (26.5 nmol/l, 95% CI 22.8–30.2 nmol/l) was greater with mean baseline 25(OH)D <30 nmol/l, compared to higher status categories (P<0.05). The 25(OH)D increase per μg/d of vitamin D (2.3 nmol/l, 95% CI 2.1–2.5 nmol/l) in trials using fortified food was greater than daily supplements (P=0.02), but not bolus injections (P=0.20). Interventions of < 10 μg/d had greater 25(OH)D increase per μg than those of ≥25 μg/d (P=0.03), but not 10–24.9 μg/d (P=0.08). Using a segmented-plateau quadratic regression, the 25(OH)D change per μg of vitamin D plateaued at 0.5 nmol/l when the dose reached 33 μg/d.

Conclusion: To the best of our knowledge, this is the first vitamin D intake and status meta-analysis specific to children. The 25(OH)D response to vitamin D intake appears to differ based on baseline status and delivery mode, but not age, sex or latitude. (Funding: Canada Research Chairs).

Disclosure: The authors declared no competing interests.