Interleukin-6 (IL-6), as a pleiotropic cytokine, has been demonstrated to be closely associated with the pathogenisis of rheumatoid arthritis (RA). However, whether this association is causal or not remains unclear, because of the multifactorial role of IL-6 and related confounding factors. We aimed to evaluate the causal relevance between circulating IL-6 levels and the risk of RA through meta-analytical Mendelian randomization approach. IL-6 gene -174G/C variant was selected as an instrument in this Mendelian randomization meta-analysis. Article identification and data collection were conducted in duplicate and independently by 2 authors. The STATA software was used for data analysis. In total, 15 and 5 articles on the association of the -174G/C variant with RA risk and circulating IL-6 level, respectively, were included. The overall analysis showed that C allelic and GC+CC genotype were significantly with 1.59-fold (95% CI: 1.19-2.14) and 1.63-fold (95% CI: 1.17-2.26) increased risk of developing RA, respectively. Asian populations showed stronger association with 4.55-fold (95% CI: 1.62-12.75), 1.84-fold (95% CI: 1.13-2.99), and 4.69-fold (95% CI: 1.68-13.14) increased RA risk in carriers of -174C allelic, CC, and GC+CC genotype, respectively. Carriers of GC+CC genotype showed significant reduction in the circulating IL-6 level compared with GG carriers (WMD = -0.77; 95% CI: -1.16 to -0.38; P = 0.000) in overall populations. Mendelian randomization presented 6% and 22% increased risk of RA with 0.1 pg/mL reduction of circulating IL-6 level in overall and Asian populations, respectively. This Mendelian randomization meta-analysis demonstrated that the long-term genetically reduced circulating IL-6 level might be causally related to a higher risk of RA, especially in Asian populations.