Abstract
Background
Whether low plasma 25-hydroxyvitamin D concentrations cause osteoporotic fractures is unclear. We tested the hypothesis that low plasma 25-hydroxyvitamin D concentrations are associated with increased risk of osteoporotic fractures using a Mendelian randomization analysis.
Methods
We genotyped 116 335 randomly chosen white Danish persons aged 20–100 years in 2 population-based cohort studies for plasma 25-hydroxyvitamin D decreasing genotypes in CYP2R1 (rs117913124 and rs12794714), DHCR7 (rs7944926 and rs11234027), GEMIN2 (rs2277458), and HAL (rs3819817); 35 833 had information on plasma 25-hydroxyvitamin D. We assessed risk of total, osteoporotic, and anatomically localized fractures from 1981 through 2017. Information on fractures and vital status was obtained from nationwide registries.
Results
During up to 36 years of follow-up, we observed 17 820 total fractures, 10 861 osteoporotic fractures, and 3472 fractures of hip or femur. Compared with individuals with 25-hydroxyvitamin D ≥ 50nmol/L, multivariable adjusted hazard ratios (95% CIs) for total fractures were 1.03 (0.97–1.09) for individuals with 25–49.9 nmol/L, 1.19 (1.10–1.28) for individuals with 12.5–24.9 nmol/L, and 1.39 (1.21–1.60) for individuals with 25-hydroxyvitamin D < 12.5 nmol/L. Corresponding hazard ratios were 1.07 (1.00–1.15), 1.25 (1.13–1.37), and 1.49 (1.25–1.77) for osteoporotic fractures and 1.09 (0.98–1.22), 1.37 (1.18–1.57), and 1.41 (1.09–1.81) for fractures of hip or femur, respectively. Hazard ratios per 1 increase in vitamin D allele score, corresponding to 3.0% (approximately 1.6 nmol/L) lower 25-hydroxyvitamin D concentrations, were 0.99 (0.98–1.00) for total fractures, 0.99 (0.97–1.00) for osteoporotic fractures, and 0.98 (0.95–1.00) for fractures of hip or femur.
Conclusions
Low plasma 25-hydroxyvitamin D concentrations were associated with osteoporotic fractures; however, Mendelian randomization analysis provided no evidence supporting a causal role for vitamin D in the risk for osteoporotic fractures.
Causal Relationship between Obesity and Vitamin D Status: Bi-Directional Mendelian Randomization Analysis of Multiple Cohorts