Serum biochemical markers and rheumatoid arthritis: Insights from observational study and Mendelian randomization analysis

Abstract Objectives: To identify the specific serum biochemical markers that exhibit causal associations with the risk rheumatoid arthritis (RA) Methods: A cross-sectional study based on the NHANES database was conducted to investigate which serum biochemical markers were significantly associated with RA. Two-sample Mendelian randomization (MR) and Generalized summary data-based Mendelian randomization (GSMR) analysis were employed to analyze the causal relationships between biomarkers exhibiting differential expression identified within the NHANES database. The primary MR analysis utilized the inverse variance weighted (IVW) method. Restricted cubic spline (RCS) regression was conducted to explore non-linear associations between variables. Results: The NHANES cohort included 1,416 participants with clinically diagnosed RA and 3,744 control participants without RA. GSMR analysis revealed a significant causal relationship between serum albumin and RA risk (OR = 0.80, 95% CI: 0.72-0.89, P = 3.67×10−5). The IVW meta-analysis also confirmed that decreased albumin expression significantly associated with decreased risk of RA (P = 0.002). A four-node RCS model demonstrated non-linear correlations between albumin, uric acid, and RA (all P non-linear < 0.05), while calcium, globulin, and lactate dehydrogenase (LDH) exhibited linear relationships (all P non-linear > 0.05). However, multiple MR methods did not show any significant causal association between calcium, LDH, urate, and RA risk. Conclusion: This study found a linear association between calcium, globulin, LDH, and RA risk, while a U-shaped non-linear relationship between albumin, uric and RA was observed. MR analysis revealed a significant causal association between albumin and RA. However, the molecular mechanisms by which albumin affects the risk of RA remain to be further elucidated.