Background <p>The incidence of rheumatoid arthritis (RA) is rising. However, its pathogenesis has not been fully understood, and the current therapeutic regimens are still limited. The aim of this study was to investigate the causal effect of plasma proteins on RA using Mendelian randomization (MR) analysis.</p> Methods <p>We performed MR analysis with 4907 plasma protein genetic associations used for exposure and RA genome-wide association data used as outcomes. The method was dominated by Inverse Variance Weighting, in addition to MR-Egger and Weighted Median. Meanwhile, further external validation and reverse MR analysis were conducted to systematically assess the causal relationship between plasma proteins and RA.</p> Result <p>Preliminary MR analysis identified two proteins (SPAG11B and DEFB135) associated with RA, and elevated plasma levels of both proteins would reduce the risk of RA (for SPAG11B, OR = 0.49, 95% CI = 0.40–0.61, <i>p</i> = 1.19 × 10^− 10; for DEFB135, OR = 0.28, 95% CI = 0.15–0.52, <i>p</i> = 4.51 × 10^− 5, using the IVW method). In the external validation phase, the results were reproducible for SPAG11B, but not for DEFB135. Reverse MR analysis pointed out that RA exhibited reverse causality for plasma levels of SPAG11B (OR = 0.93, 95% CI = 0.89–0.98, <i>p</i> = 0.004), but not for DEFB135 (<i>p</i> = 0.93).</p> Conclusion <p>The results of MR analysis in this study supported that SPAG11B as a novel biomarker for RA was worthy of further investigation.</p>