A vitamin D nutritional cornucopia: new insights concerning the serum 25-hydroxyvitamin D status of the US population

Abstract

It is generally agreed that the serum concentration of 25hydroxyvitamin D [25(OH)D] in healthy persons is the best indicator of the vitamin D status of patients with vitamin D–related disease states (1). The report by Looker et al (2) in this issue of the Journal provides a cornucopia of new insights into the vitamin D nutritional status of US citizens from 1988 through 2004. In particular, this report compares serum 25(OH)D concentrations of 20 289 participants in the National Health and Nutrition Examination Survey (NHANES) 2000–2004 with those of 18 158 participants in NHANES III (1988–1994). For the first time, results were reported for children 1–11 y old, pregnant women, and the Mexican American population. The most important finding in the report is that the mean serum 25(OH)D concentrations of persons 12 y old fell by 5–20 nmol/L between 1988–1994 and 2000–2004. Assay changes, which were unrelated to changes in 25(OH)D status, accounted for a portion of this apparent difference between the 2 study groups; nevertheless, at the end of the complex analysis, there still was a significant (7.1 nmol/L) reduction in 25(OH)D concentrations in males (except Mexican Americans) but not in most female groups. This result is potentially disturbing if it represents the beginning of the downward trend in serum 25(OH)D concentrations that has been seen over the past 10–15 y. It will therefore be essential to repeat this study after another decade. The vitamin D cornucopia that Looker et al provided contains a variety of other novel and important observations. These include differences in 25(OH)D status 1) by ethnicity and sex in nonHispanic whites, non-Hispanic blacks, and Mexican Americans; 2) by age group (1–5, 6–11, 12–19, 20–49, 50–69, or 70 y old); 3) by season (November–March or April–October); and 4) by pregnancy status. In addition, a second vitamin D cornucopia can be found under “Supplementary data” in the current online issue. This material contains extensive data on 4 topics: 1) a comparison of 25(OH)D assay methods; 2) seasonal variations in serum 25(OH)D by race-ethnicity in persons 12 y old; 3) the prevalence of serum 25(OH)D concentrations 25 nmol/L (an indication of nutritional vitamin D deficiency); and 4) the prevalence of serum 25(OH)D3 concentrations below selected thresholds by age, sex, and raceethnicity. Some readers may be troubled by 2 technical aspects of the study by Looker et al. A potentially confounding problem is that there was anevolution in25(OH)Dassaymethodologybetween the first study in 1988–1994 and the second study in 2000–2004. Possible assay differences were assessed by repeating measurements of the 25(OH)D concentrations in 150 stored serum samples from NHANES III with the samples from the 2000–2004 samples. Looker et al used sophisticated statistical techniques to tease out the shift in the mean serum 25(OH)D concentrations between the 2 decades that was due to changes in assay methods. This problem was recently discussed in detail (3). However, there still is a need for significant improvements in 25(OH)D assays so that consistent and precise assay results may be obtained over multiyear periods. The second potential concern relates to the protocol used in NHANES; that is, the serum samples were collected in the US southern latitudes ( 35 °N) only in November–March and in the US northern latitudes ( 35 °N) only in April–October. Thus, the magnitude of the seasonal fall in 25(OH)D concentrations in the winter in the northern latitudes is underestimated, as is the increase in concentrations in the summer in the southern latitudes. Ideally, in a study focusing on serum 25(OH)D concentrations, there should be no bias of the time of the year for the latitude at which the serum samples were collected. These same concerns also apply to the involvement of the vitamin D endocrine system with the disease process of cancer. Epidemiologic studies have suggested a direct link between greater sunlight (ultraviolet B light) production of vitamin D3 in populations living at lower latitudes and lower incidences of breast, colon, prostate, and ovarian cancers and nonHodgkin lymphoma(4).Other studieshaveshownthatpersonswith low serum concentrations of 25(OH)D (ie, 8 nmol/L) had a higher incidence of cancer (5), which reinforces the need to improve vitamin D nutritional status. The study by Looker et al is particularly important, given the current concerns about the appropriate recommendations for the daily intake of vitamin D3 (6, 7). The current adequate intake allowance of vitamin D, recommended in 1997, is considered by many scientists tobe too lowand tobe focusedonlyonvitaminD’sactions on calcium and bone issues (1). However, over the past decade, new evidence has shown that there are 5 additional physiologic systems in which the vitamin D receptor and its cognate steroid hormone, 1 ,25-dihydroxyvitamin D3 [1 ,25(OH)2D3], generate biological responses (8). These are the immune, pancreas, heartcardiovascular, muscle, and brain systems; the control of the cell