Previous observational studies have identified various risk factors associated with the development of osteoporosis, including sex hormone-binding globulin (SHBG). The aim of this study was to determine the potential causal effects of circulating SHBG concentrations on bone mineral density (BMD). Two-sample Mendelian randomization (MR) approach was applied in analyses. From summary-level data of genome-wide association studies (GWAS), we selected 11 single-nucleotide polymorphisms (SNPs) associated with SHBG levels as instrumental variable, and used summary statistics for BMD at forearm (FA) (n = 8143), femoral neck (FN) (n = 32,735), lumbar spine (LS) (n = 28,498) and heel (HL) (n = 394,929), and total-body BMD of different age-stages (15 or less, 15-30, 30-45, 45-60, 60 or more years old) (n = 67,358). Inverse causal associations was observed between SHBG levels and FA BMD (Effect = - 0.26; 95% CI - 0.49 to - 0.04; P = 0.022), HL eBMD (Effect = - 0.09; 95% CI - 0.12 to - 0.06; P = 3.19 × 10-9), and total-body BMD in people aged 45-60 years (Effect = - 0.16; 95% CI - 0.31 to - 2.4 × 10-3; P = 0.047) and over 60 years (Effect = - 0.19; 95% CI - 0.33 to - 0.05; P = 0.006). Our study demonstrates that circulating SHBG concentrations are inversely associated with FA and HL eBMD, and total-body BMD in people aged over 45 years, suggesting that the role of SHBG in the development of osteoporosis might be affected by chronological age of patients and skeletal sites.
Keywords: Bone mineral density; Causal effect; Mendelian randomization; Osteoporosis; Sex hormone-binding globulin.