Purpose: Multiple risk factors have been implicated in the development of osteoporosis. This study examined potential associations between serum nutritional factors and bone mineral density (BMD).
Methods: Six nutritional factors were selected as exposures. Outcomes included total body BMD (n = 66 945); BMD at the forearm (FA), femoral neck (FN) and lumbar spine (LS) (n = 8143, n = 32 735, and n = 28 498, respectively); estimated heel BMD (HL eBMD) (n = 394 929); and HL eBMD stratified by sex (n = 206 496). A 2-sample Mendelian randomization approach was adopted to estimate the association between serum nutritional factors and BMD. The threshold for adjusted P value was 1.39 × 10-3.
Results: Serum calcium levels were inversely associated with LS BMD (effect = -0.55; 95% CI, -0.86 to -0.24; P = 0.001), whereas serum selenium levels were positively correlated with HL eBMD (effect = 0.22; 95% CI, 0.10 to 0.33; P = 1.70 × 10-4). Regarding nominal significance, there was a positive association between serum selenium levels and FA BMD. Nominally significant results were also obtained for serum retinol as well as vitamin E levels and HL eBMD. Moreover, sex-specific effects of serum retinol and vitamin E levels on BMD were observed in men.
Conclusion: Serum calcium and selenium levels influence BMD at specific skeletal sites. This implies that these nutritional factors play crucial roles in bone metabolism.
Keywords: Bone mineral density; Mendelian randomization; calcium; nutritional factor; selenium.
© The Author(s) 2021. Published by Oxford University Press on behalf of the Endocrine Society. All rights reserved. For permissions, please e-mail: [email protected].