Vitamin D supplementation: bones of contention.

The discovery of vitamin D as an essential nutrient for skeletal development a century ago was a major public health victory. Supplementation, whether solar or dietary, prevented the devastating eff ects of rickets in chil dren. Five decades later, the molecular mechanisms of the vitamin’s active form (1,25-dihydroxyvitamin D) and its receptor (vitamin D receptor [VDR]), were elucidated, and subsequently clinical investigators linked vitamin D defi ciency or insuffi ciency with osteoporosis. This fi nding seemed logical because osteomalacia (ie, the softening of bone in adults due to impaired mineralisation) can cause fractures and often coexists histologically with osteoporosis. Slowly, vitamin D supplementation became established for prevention of osteoporosis. But, as shown in a meta-analysis in The Lancet, the story is more complex, both from an epidemiological and mechanistic perspective. Ian Reid and colleagues did a systematic review of the eff ects of vitamin D supplements on bone mineral density that included 23 randomised controlled trials encompassing more than 4000 participants, with a mean age of 59 years. The authors found that vitamin D supplementation for 2 years resulted in no change in bone mineral density at four major skeletal sites (spine, total hip, radius, and total body), with a signifi cant increase (0·8%, 95% CI 0·2–1·4) only at the femoral neck. Surprisingly, any benefi t reported in bone mineral density was independent of calcium supplementation, baseline concentration of 25-hydroxyvitamin D, duration of treatment, or age. The investigators conclude that widespread vitamin D prophylaxis in healthy community dwelling adults to prevent osteoporosis is unwarranted. How can these surprising fi ndings be reconciled with clinical practice and public health strategies to prevent osteoporosis? First, bone mineral density was the primary outcome in the present analysis and is widely used as a surrogate measure of fracture risk. However, changes of bone mineral density in this age group are a modest predictor of subsequent fractures. Even with bisphosphonate treatment in high-risk elderly patients (older than 70 years), the bone mineral density increase with bisphosphonates accounts for less than 50% of the eff ect on fractures. Thus, the absence of a positive relation between vitamin D supplementation and change in bone mineral density could be dismissed as the fi ndings having few clinical implications. However, the results are consistent with those of two recent meta-analyses of randomised trials with vitamin D supplementation alone that recorded no effi cacy in fracture prevention, nor in another metaanalysis of vitamin D with an intention-to-treat design. Second, it is diffi cult to distinguish between the eff ects of calcium versus those of vitamin D on skeletal integrity, because the main mechanism of action for vitamin D is promotion of calcium absorption in the gut and not direct incorporation of calcium in bone. In the present meta-analysis, only half of the trials used both calcium and vitamin D supplementation. In trials in which vitamin D was given simultaneously with calcium, a signifi cant reduction of 11% in hip fractures and a very modest increase in hip bone mineral density was reported. This fi nding formed the basis of the recommendation by the US Institute of Medicine that 1200 mg of calcium and 800 IU of vitamin D were optimum intakes for skeletal health in elderly people. The inclusion of more studies with calcium plus vitamin D in the present report could have resulted in greater increases in bone mineral density, but confounding by calcium supplementation would not have clarifi ed the role of vitamin D alone in supporting bone mass. Third, mechanisms of vitamin D action on the skeleton have recently been re-examined leading to a new appreciation of the vitamin’s biological role; these fi ndings also Published Online October 11, 2013 http://dx.doi.org/10.1016/ S0140-6736(13)61721-3