Is the Sun Setting on Vitamin D?

Osteoporotic fractures and low vitamin D concentrations are prevalent, especially among older adults. One in 2 women and 1 in 5 men in the United States aged ≥50 years will have an osteoporotic fracture (1). According to the National Health and Nutrition Examination Survey (NHANES) 2011–2014, 24% of women, 23% of men, and 15% of adults aged ≥60 years in the United States have low 25-hydroxy vitamin D (25-OHD) concentrations (i.e., 25-OHD < 20 ng/mL [50 nmol/L]) (2). Several observational studies, including the Cardiovascular Health Study and NHANES III, have found an association between low serum 25-OHD concentrations and higher risk of fractures, at least in these predominantly white populations (3, 4). These observational studies raise the possibility that low 25-OHD concentrations may be a risk factor for fracture but, given the potential for uncontrolled confounding, cannot prove a cause–effect relationship.

We depend on randomized controlled trials (RCTs) to provide the strongest evidence of causation and, perhaps more important, to determine whether correction of the modifiable risk factor improves clinical outcomes. Most recent systematic reviews and meta-analyses of RCTs have found that vitamin D alone does not prevent fractures (5, 6). In contrast, these reviews have suggested that combination vitamin D plus calcium supplementation may modestly reduce fracture risk (5, 7). Interpretation of meta-analyses, however, can be complicated because they collate RCTs of very different designs. The few RCTs of supplemental vitamin D alone on fractures in general populations have had conflicting results that have been attributed to differences in baseline 25-OHD concentrations, differences in dosing (bolus vs daily), and other factors. In a Dutch study, vitamin D supplementation of 400 IU/day in elderly adults had no effect on hip or peripheral fractures compared with placebo (8), whereas a British study found that supplemental vitamin D₃ of 100 000 IU every 4 months versus placebo resulted in 33% reduction in the relative risk of first hip, wrist/forearm, or spine fracture (9). However, a trial in Australia by Sanders et al. (10) suggested that very high bolus doses of 500 000 IU of vitamin D per year increased fracture risk in older women. The most recent Vitamin D Assessment (VIDA) RCT in New Zealand indicated that high-dose bolus vitamin D supplementation (initial dose of 200 000 IU followed by 100 000 IU monthly) had a neutral effect on fracture risk (11).

Based largely on observational studies, and despite a lack of evidence from interventional studies to support the role of supplemental vitamin D alone in reducing fractures in the general population, vitamin D supplement use among US adults aged ≥70 years has increased almost 100-fold from 0.4% in 2000 to 38.5% in 2014 (12).

In the context of such inconsistencies and controversies, Mendelian randomization affords an alternative strategy for assessing causality. Mendelian randomization studies are less susceptible to confounding and reverse causation bias than observational studies by testing genetic variants that affect biomarkers or exposures and evaluating the relationship of those gene variants with health outcomes. In the case of vitamin D, gene variants that decrease serum 25-OHD concentrations can be examined in relation to fracture risk. Because these gene variants are essentially “randomly assigned” at conception before disease onset and reflect long-term concentrations relatively unaffected by lifestyle or behavioral factors, they more directly assess causal relationships with chronic disease outcomes than a 25-OHD concentration measured at a selected point in time.

In the present study published in Clinical Chemistry (13), Çolak et al. performed Mendelian randomization analysis to explore whether low plasma 25-OHD concentrations may cause osteoporotic fractures. Restricting to white Danish population cohorts, the authors found that low 25-OHD concentrations were associated with osteoporotic fractures. They then carefully selected genetic variants that affect synthesis of, or conversion to, 25-OHD and did not include gene variants that modify vitamin D binding protein (which may affect concentrations of 25-OHD but not the biological activity). These genetic variants were then used as proxies of low 25-OHD concentrations. Of note, although all selected genotypes were associated with lower plasma 25-OHD concentrations, the combined vitamin D allele score explained only 1.2% of the variation in the biomarker. Compare this with a Mendelian randomization analysis using KIV-2 repeats, which explained 21% to 27% of all variation in plasma lipoprotein(a) concentrations, to support a causal association of increased lipoprotein(a) values with increased myocardial infarction risk (14).

Using Mendelian randomization analysis, Çolak et al. did not find any evidence to suggest that low plasma 25-OHD concentrations cause osteoporotic fractures. Instead, there was a suggestion that these genetic variants associated with low vitamin D concentrations may, paradoxically, reduce risk of fractures. The conflicting findings between the observational analyses and the Mendelian randomization analyses suggest that there were unaccountable confounding factors or residual pleiotropy, and/or these genetic variants were not suitable alternatives for 25-OHD concentrations. The alignment of the null Mendelian randomization findings of Çolak et al. with the results of systematic reviews and meta-analyses of RCTs of vitamin D alone is reassuring (5, 6).

Results from the bone-related ancillary studies to the Vitamin D and Omega-3 Trial (VITAL) will provide more data. Recent findings from the VITAL: Effects on Bone Structure and Architecture study showed that vitamin D₃ supplementation of 2000 IU/day vs placebo for 2 years did not improve bone mineral density or structure in a subcohort of 771 participants (15). The VITAL: Effects on Fracture study will determine whether supplemental vitamin D reduces incidence of fractures in the overall VITAL cohort of 25 871 participants enrolled from 50 states and followed for a median of 5.3 years. These analyses of fracture outcomes are ongoing. Furthermore, VITAL investigators will be testing whether genetic variation(s) relevant to vitamin D metabolism, absorption, and/or receptor function modify effects of vitamin D₃ supplementation on bone density and fractures. The Women’s Health Initiative RCT showed that although calcium with vitamin D supplementation vs placebo (in the intention-to-treat analysis) did not reduce total fracture incidence, a significant risk reduction with supplemental calcium and vitamin D was observed for women with low genetic predisposition to low bone mineral density (16). Collectively, these studies may help advance personalized medicine.

For now, despite results from observational studies showing a relationship between low vitamin D status and fracture risk, the results of Mendelian randomization studies and metaanalyses of RCTs do not suggest a causal relationship and do not support recommendation for vitamin D supplementation alone for fracture risk reduction in the general population. These findings should not be applied to patients with osteoporosis, particularly as the vast majority of RCTs evaluating osteoporosis medications are performed with concurrent vitamin D and calcium supplementation. Many systematic metaanalyses do suggest a modest benefit of cosupplementation of calcium and vitamin D for fracture risk (5, 7). The vitamin D findings also do not apply to individuals with very low 25-OHD concentrations that are associated with osteomalacia (i.e., 25-OHD concentrations <12 ng/mL [30 nmol/L]), which account for only 2.9% of the population aged ≥60 years in the United States (2). The research thus far suggests, however, that treatment with vitamin D supplementation does not prevent fractures in the general population.

Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 4 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved.

Authors’ Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:

Employment or Leadership: None declared.

Consultant or Advisory Role: None declared.

Stock Ownership: None declared.

Honoraria: None declared.

Research Funding: National Institutes of Health sponsored the VITAL trial. Pharmavite donated study pills for the VITAL trial.

Expert Testimony: None declared.

Patents: None declared.