Association between serum iron status and the risk of five bone and joint-related diseases: a Mendelian randomization analysis

Abstract

Background: According to reports, iron status has been associated with the risk of bone and joint-related diseases. However, the exact role of iron status in the development of these conditions remains uncertain.

Method: We obtained genetic data on iron status, specifically serum iron, ferritin, transferrin saturation (TSAT), and transferrin, as well as data on five common bone and joint-related diseases (osteoarthritis, osteoporosis, rheumatoid arthritis [RA], ankylosing spondylitis [AS], and gout) from independent genome-wide association studies involving individuals of European ancestry. Our primary approach for causal estimation utilized the inverse variance weighted (IVW) method. To ensure the reliability of our findings, we applied complementary sensitivity analysis and conducted reverse causal analysis.

Result: Using the IVW method, we revealed a positive causal relationship between ferritin levels and the risk of osteoarthritis (OR [95% CI], 1.0114 [1.0021-1.0207]). Besides, we identified a protective causal relationship between serum iron levels and TSAT levels in the risk of RA (OR [95% CI] values of serum iron and TSAT were 0.9987 [0.9973-0.9999] and 0.9977 [0.9966-0.9987], respectively). Furthermore, we found a positive causal relationship between serum iron levels and the risk of AS (OR [95% CI], 1.0015 [1.0005-1.0026]). Regarding gout, both serum iron and TSAT showed a positive causal relationship (OR [95% CI] values of 1.3357 [1.0915-1.6345] and 1.2316 [1.0666-1.4221] for serum iron and TSAT, respectively), while transferrin exhibited a protective causal relationship (OR [95% CI], 0.8563 [0.7802-0.9399]). Additionally, our reverse causal analysis revealed a negative correlation between RA and ferritin and TSAT levels (OR [95% CI] values of serum iron and TSAT were 0.0407 [0.0034-0.4814] and 0.0049 [0.0002-0.1454], respectively), along with a positive correlation with transferrin (OR [95% CI], 853.7592 [20.7108-35194.4325]). To ensure the validity of our findings, we replicated the results through sensitivity analysis during the validation process.

Conclusion: Our study demonstrated a significant correlation between iron status and bone and joint-related diseases.

Keywords: Mendelian randomization; bone and joint-related diseases; genetics; genome wide association studies; iron status.

Figure 1

Directed acyclic graph of this bidirectional Mendelian randomization (MR) study. TSAT, transferrin saturation; RA, rheumatoid arthritis; AS, ankylosing spondylitis.

Figure 2

The association of iron status in the risk of five bone and joint-related diseases. SNPs, single nucleotide polymorphisms; I², degree of heterogeneity; CI, confidence interval; IVW, inverse variance weighting; WM, weighted median; TSAT, transferrin saturation; RA, rheumatoid arthritis; AS, ankylosing spondylitis.

Figure 3

MR leave-one-out sensitivity analysis of serum iron status on bone and joint-related diseases. Circles indicate the results of MR analysis of remaining SNPs on serum iron status on bone and joint-related diseases after omitting each SNP in turn. Bars indicate CI. (A) Leave-one-out plot from TSAT on AS. (B) Leave-one-out plot from transferrin on osteoporosis. (C) Leave-one-out plot from transferrin on RA. (D) Leave-one-out plot from transferrin on AS. MR, Mendelian randomization; SNPs, single nucleotide polymorphisms; CI, confidence interval; TSAT, transferrin saturation; RA, rheumatoid arthritis; AS, ankylosing spondylitis.

Figure 4

The association of bone and joint-related diseases in iron status. SNPs, single nucleotide polymorphisms; I², degree of heterogeneity; CI, confidence interval; IVW, inverse variance weighting; WM, weighted median; TSAT, transferrin saturation; RA, rheumatoid arthritis; AS, ankylosing spondylitis.

Figure 5

MR leave-one-out sensitivity analysis of bone and joint-related diseases on serum iron status. Circles indicate the results of MR analysis of remaining SNPs on bone and joint-related diseases on serum iron status after omitting each SNP in turn. Bars indicate CI. (A) Leave-one-out plot from RA on serum iron. (B) Leave-one-out plot from RA on TSAT. (C) Leave-one-out plot from RA on transferrin. (D) Leave-one-out plot from AS on serum iron. (E) Leave-one-out plot from AS on ferritin. (F) Leave-one-out plot from AS on TSAT. (G) Leave-one-out plot from AS on transferrin. (H) Leave-one-out plot from AS on TSAT after detecting three SNPs. (I) Leave-one-out plot from AS on transferrin after detecting three SNPs. MR, Mendelian randomization; SNPs, single nucleotide polymorphisms; CI, confidence interval; TSAT, transferrin saturation; RA, rheumatoid arthritis; AS, ankylosing spondylitis.