RESUMO
BACKGROUND: Analysis of serum 25-hydroxyvitamin D (s-25(OH)D) may be complicated by the less active or in-active vitamin D metabolite C3-epi-25(OH)D3 (C3-epimer). We aimed to explore the relationship between s-C3-epimer and s-25(OH)D and other determinants and describe the longitudinal course of the C3-epimer fraction in paired mother-child samples. METHOD: S-25(OH)D and s-C3-epimer were estimated by liquid chromatography mass spectrometry in 290 mother-infant pairs from the population-based Odense Child Cohort. Longitudinal analyses were feasible in two subcohorts; B) early and late pregnancy, cord, three and 18months (n=132); and C) early and late pregnancy, delivery and cord (n=105). RESULTS: Mean s-25(OH)D was 50.6-110.4nmol/L at the six time points. The mean C3-epimer fraction was 10.1% at three months, 1.1%-3.0% at the other time points. In multivariate analyses, the s-C3-epimer correlated with s-25(OH)D (all time points, p<0.001), and season, maternal and infant age and maternal vitamin D supplementation at some time points. The C3-epimer fraction fluctuated between adjacent time points. By cosinor analyses, a season-dependent sinusoidal pattern for s-25(OH)D and C3-epimer fraction was found and changes between adjacent time points depended on season (p<0.007 or trend). In early infancy, subtraction of the C3-epi-25(OH)D3 from total s-25(OH)D resulted in reclassification of 8% of the children by use of the 75nmol/L cut off for s-25(OH)D. CONLCUSION: The s-C3-epimer was independently correlated to s-25(OH)D, season, maternal vitamin D supplementation, maternal and infant age. The C3-epimer fraction was only of clinical importance in early infancy, where it could lead to misclassification of the vitamin D status.