Vitamin D – a systematic literature review for the 5th edition of the Nordic Nutrition Recommendations

Background: The present literature review is part of the NNR5 project with the aim of reviewing and updating the scientific basis of the 4th edition of the Nordic Nutrition Recommendations (NNR) issued in 2004. Objectives: The overall aim was to review recent scientific data on the requirements and health effects of vitamin D and to report it to the NNR5 Working Group, who is responsible for updating the current dietary reference values valid in the Nordic countries. Methods: The electronic databases MEDLINE and Swemed were searched. We formulated eight questions which were used for the search. The search terms related to vitamin D status and intake and different health outcomes as well as to the effect of different vitamin D sources on vitamin D status. The search was done in two batches, the first covering January 2000–March 2010 and the second March 2009–February 2011. In the first search, we focused only on systematic literature reviews (SLRs) and in the second on SLRs and randomized control trials (RCTs) published after March 2009. Furthermore, we used snowballing for SLRs and IRCTs published between February 2011 and May 2012. The abstracts as well as the selected full-text papers were evaluated in pairs. Results: We found 1,706 studies in the two searches of which 28 studies were included in our review. We found 7 more by snowballing, thus 35 papers were included in total. Of these studies, 31 were SLRs and 4 were RCTs. The SLRs were generally of good or fair quality, whereas that of the included studies varied from good to poor. The heterogeneity of the studies included in the SLRs was large which made it difficult to interpret the results and provide single summary statements. One factor increasing the heterogeneity is the large variation in the assays used for assessing 25-hydroxyvitamin D concentration [25(OH)D], the marker of vitamin D status. The SLRs we have reviewed conclude that the evidence for a protective effect of vitamin D is only conclusive concerning bone health, total mortality and the risk of falling. Moreover, the effect was often only seen in persons with low basal 25(OH)D concentrations. In addition, most intervention studies leading to these conclusions report that intervention with vitamin D combined with calcium and not vitamin D alone gives these benefits. It was difficult to establish an optimal 25(OH)D concentration or vitamin D intake based on the SLRs, but there are evidence that a concentration of ≥50 nmol/l could be optimal. The dose–response studies relating vitamin D intake (fortification and supplementation) to S-25(OH)D suggested that an intake of 1–2.5 µg/day will increase the serum concentration by 1–2 nmol/l but this is dependent on the basal concentration with a response being greater when the basal concentration is low. Conclusion: Data show that a S-25(OH)D concentration of 50 nmol/l would reflect a sufficient vitamin D status. Results from this review support that the recommendation in NNR 2004 needs to be re-evaluated and increased for all age groups beyond 2 years of age. We refer to the total intake from food as well as supplements, given minimal sun exposure. Limited sunshine, however, does not reflect the situation for the majority of the Nordic population in the summertime. It should also be emphasized that there are large differences in results depending on assay methods and laboratories measuring 25(OH)D, adding to the uncertainty of determining an appropriate target concentration. Moreover, the dose–response of vitamin D on serum 25(OH)D-concentrations is not well established and is dependent on the basal concentrations, sunshine exposure and dietary intake. We advise that these uncertainties should be taken into account when setting the final Nordic recommendations