点击显示 收起
1 Pathology and Laboratory Medicine CMC-V, 1st floor, Suite Y1–147 PO Box 30.001 University Hospital Groningen 9700 RB Groningen Netherlands E-mail: f.a.j.muskiet{at}lab.azg.nl
2 Department of Research and Development Friesland Coberco Dairy Foods Leeuwarden Netherlands
Dear Sir:
Vieth et al (1) regard a daily intake of 100 µg (4000 IU) vitamin D3 necessary "to ensure desirable 25(OH)D [25-hydroxyvitamin D] concentrations." In their study, Vieth et al administered 100 µg vitamin D3/d for 5 mo to 18–56-y-old healthy subjects living at high latitudes. They found that the mean serum 25(OH)D concentration plateaued at 96 nmol/L (with an extreme of almost 140 nmol/L) without concomitant changes in serum calcium or the urinary calcium-to-creatinine ratio, which were used as indexes of hypervitaminosis D. Vieth et al concluded that "consumption of vitamin D3 at intakes 100 µg/d causes no harm" and that the lowest adverse effect level is higher than the current estimate of 95 µg/d, as recognized by the Food and Nutrition Board.
We question whether there is at present any evidence-based medicine showing that a vitamin D intake of 100 µg/d is necessary to prevent disease either in the short term or in the long run. There is evidence that lower doses (10–20 µg) improve bone mineral density (2–4) and reduce fracture incidence (3, 4). The quoted inverse relation between 25(OH)D and parathyroid hormone is well taken, but cannot be considered other than as an experimental target that may set the stage for an as yet to be conducted randomized controlled trial with hard endpoints, such as fracture incidence. The question is, who would need as much as 100 µg/d, because the physiologic plateau of 100 nmol 25(OH)D/L encountered by Vieth et al is surprisingly comparable with that obtained with much lower dosages in different settings (for review see reference 5). Dawson-Hughes et al (3) supplemented 70-y-old subjects with 17.5 µg vitamin D for 3 y to reach 25(OH)D concentrations of 112 nmol/L, and Chapuy et al (4) reached 104 nmol/L in 84-y-old women with a 20-µg supplement. Another example is the 108-nmol/L mean 25(OH)D concentration that was reached with 20 µg/d in 74-y-old women who lived in the tropics and had high baseline 25(OH)D concentrations (6).
These findings raise questions regarding the fate of any excess vitamin D that becomes unconverted to 25(OH)D in persons who have been treated with 100 µg vitamin D/d. One option is accumulation in adipose tissue: rats treated with supraphysiologic vitamin D dosages show linear accumulation of vitamin D in their fatty tissue with time (7). Storage in adipose tissue is likely to be at the basis of the 100-nmol/L physiologic 25(OH)D plateau; this pool can apparently be mobilized because rats deprived of food had higher 25(OH)D concentrations than did fed controls. These data suggest that daily intakes of 100 µg may be safe in the short run, but that prolonged consumption may, depending on baseline values, constitute a potential time bomb with an as yet poorly understood detonator.
We nevertheless tend to agree with Vieth et al that the dangers of hypervitaminosis D at moderate vitamin D doses are generally exaggerated. To remove this fear, fat-soluble vitamin toxicity studies may have to be based on long-term observations that also consider the tendency of these compounds to accumulate in the body. Any discussion of the results of such investigations should stress the many limitations of the study design, such as age group, supplement duration, background dietary intake, and background sunlight exposure, to prevent unintended extrapolation to other populations. It is all right to aim at those with the lowest initial status, but it is also right to remain inhibited by those with the highest.
REFERENCES