Causal relationship between bone mineral density and intervertebral disc degeneration: a univariate and multivariable mendelian randomization study

Abstract

Background: Although previous studies have suggested a possible association between bone mineral density (BMD) and intervertebral disc degeneration (IDD), the causal relationship between them remains unclear. Evidence from accumulating studies indicates that they might mutually influence one another. However, observational studies may be affected by potential confounders. Meanwhile, Mendelian randomization (MR) study can overcome these confounders to assess causality.

Objectives: This Mendelian randomization (MR) study aimed to explore the causal effect of bone mineral density (BMD) on intervertebral disc degeneration (IDD).

Methods: Summary data from genome-wide association studies of bone mineral density (BMD) and IDD (the FinnGen biobank) have been acquired. The inverse variance weighted (IVW) method was utilized as the primary MR analysis approach. Weighted median, MR-Egger regression, weighted mode, and simple mode were used as supplements. The Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) and MR-Egger regression were performed to assess horizontal pleiotropy. Cochran's Q test evaluated heterogeneity. Leave-one-out sensitivity analysis was further conducted to determine the reliability of the causal relationship. Multivariate MR (MVMR) analyses used multivariable inverse variance-weighted methods to individually and jointly adjust for four potential confounders, body mass index (BMI), Type2 diabetes, hyperthyroidism and smoking. A reverse MR analysis was conducted to assess potential reverse causation.

Results: In the univariate MR analysis, femoral neck bone mineral density (FNBMD), heel bone mineral density (eBMD), lumbar spine bone mineral density (LSBMD), and total body bone mineral density (TB BMD) had a direct causal effect on intervertebral disc degeneration (IDD) [FNBMD-related analysis: OR(95%CI) = 1.17 (1.04 to 1.31), p = 0.008, eBMD-related analysis: OR(95%CI) = 1.06 (1.01 to 1.12), p = 0.028, LSBMD-related analysis: OR(95%CI) = 1.20 (1.10 to 1.31), p = 3.38E-7,TB BMD-related analysis: OR(95%CI) = 1.20 (1.12 to 1.29), p = 1.0E-8]. In the MVMR analysis, it was revealed that, even after controlling for confounding factors, heel bone mineral density (eBMD), lumbar spine bone mineral density (LSBMD), and total body bone mineral density (TB BMD) still maintained an independent and significant causal association with IDD(Adjusting for heel bone mineral density: beta = 0.073, OR95% CI = 1.08(1.02 to 1.14), P = 0.013; Adjusting for lumbar spine bone mineral density: beta = 0.11, OR(95%CI) = 1.12(1.02 to 1.23), P = 0.03; Adjusting for total body bone mineral density: beta = 0.139, OR95% CI = 1.15(1.06 to 1.24), P = 5.53E - 5). In the reverse analysis, no evidence was found to suggest that IDD has an impact on BMD.

Conclusions: The findings from our univariate and multivariable Mendelian randomization analysis establish a substantial positive causal association between BMD and IDD, indicating that higher bone mineral density may be a significant risk factor for intervertebral disc degeneration. Notably, no causal effect of IDD on these four measures of bone mineral density was observed. Further research is required to elucidate the underlying mechanisms governing this causal relationship.

Keywords: Bone mineral density; GWAS data; Intervertebral disc degeneration; Mendelian randomization (MR); Osteoporosis.

Fig. 1

Overall flow chart of the univariable MR study design

Fig. 2

Overview of the multivariable MR study design

Fig. 3

Univariable Mendelian randomization (MR) results exhibit diverse associations with intervertebral disc degeneration (IDD) across the four types of bone mineral density (BMD). BMI, body mass index (BMI); FNBMD, femoral neck bone mineral density; eBMD, Heel bone mineral density; LSBMD, lumbar spine bone mineral density; TBBMD, Total body bone mineral density

Fig. 4

Forest plots of multivariate Mendelian randomization analyses examined causal associations between BMD and IDD, adjusted for, type 2 diabetes, BMI, smoking, and hyperthyroidism. Estimates of causality were expressed as odds ratios (ORs) and 95% confidence intervals (CIs). FNBMD, femoral neck bone mineral density; eBMD, Heel bone mineral density; LSBMD, lumbar spine bone mineral density; TBBMD, Total body bone mineral density

Fig. 5

Reverse Mendelian randomization (MR) results