Dear Editor, Thank you for the opportunity to respond to the comments of Karras et al. on our study of a cohort of Danish women in early pregnancy. 1 The authors raise several important questions regarding the analysis method of serum 25(OH)D and the prevalence of vitamin D insufficiency.We measured serum 25(OH)D 2+3 by the LC-MS/MS method validated according to the American National Institute of Standards and Technology (NIST). We chose to report the combined 25(OH)D 2 + 25(OH)D 3 values only. In the vast majority of participants (n = 1319), the level of 25(OH)D 2 was below the detection threshold (<6Á5 nmol/l). Detectable, but low, levels of D 2 were seen in 29 women (2Á2%), and for these women, the level of the reported 25(OH)D was the sum of D 2 and D 3 .Vitamin D 3 is of animal origin and known to have a higher biological activity than the plant-derived vitamin D 2 . In Denmark, vitamin D supplementation during pregnancy is recommended at a dose of 10 lg daily, and most products consist of D 3 . Further dietary vitamin D sources in pregnant Danish women are predominantly of animal origin, especially fish, sea food, low-fat dairy products and eggs. 2 We agree that in other settings with a higher vitamin D 2 intake, or specific vitamin D 2 supplementation, separate 25(OH)D 2 and 25(OH)D 3 data could be potentially useful.We did not report on the less active C3-and C2-epimers, which in the relatively small (n = 60) study from Karras et al. 3 represented almost 25% of the total 25(OH)D in pregnant women at term and 22% in their offspring. In our study, the C2-epimer was expected to be negligible given the low 25(OH) D 2 levels. Our laboratory has separately examined the concentration of 3-epi-25(OH)D in serum from 400 randomly selected patients who needed assessment of serum 25(OH)D. In these samples, none had a content of C3-epimer above the detection limit (1Á5 nmol/l), despite full detection rate for the C3-epimer in the set-up (unpublished data). This lack of C3-detection may be due to selection of patients with low total 25(OH)D levels. Engelman recently reported a mean C3-epimer concentration of 0Á5 ng/ml (1Á25 nmol/l; below our detection limit) in adults with a total level of 25(OH)D below 62Á5 nmol/l. 4 In various studies, the C3-epimer represented 3Á9-6% of the total 25(OH)D in children and adults, but 11-21% in neonates and infants. 4-6 It could be hypothesized that the C3-epimer percentage is especially high in late pregnancy and neonates, in keeping with other alterations during pregnancy including the oestrogen-induced rise in vitamin D binding globulin, a twofold increase in 1,25 (OH) 2 D and a rise in the free fraction of 1,25(OH) 2 D. In early pregnancy, as in our study, a much lower C3-epimer fraction compared with late pregnancy could be expected. More data during early and late pregnancy including a higher number of pregnant women should be published, before conclusions on a substantially high C3-epimer fraction during late pregnancy eventually can be drawn. Future studies should, furtherm...