A single nucleotide polymorphism (SNP) of the protein tyrosine phosphatase nonreceptor gene (PTPN22) confers susceptibility to rheumatoid arthritis (RA) and certain other classical autoimmune diseases. The association between PTPN22 1858C/T polymorphism and the risk of RA is still controversial and ambiguous; therefore, we performed this meta-analysis to confirm some relationships. We conducted a search in the PubMed database without a language limitation, covering all papers published until June 20, 2011. Overall, 19 case-control studies with 11,727 cases and 12,640 controls were retrieved based on the search criteria for RA susceptibility related to the 1858C/T polymorphism. Odds ratios (OR) and 95% confidence intervals (CI) were used to assess the strength of this association. Publication bias was assessed with Eggers test. We found that PTPN22 1858C/T polymorphism could increase RA risk in overall genetic models in Europeans (T-allele vs. C-allele, OR= 1.54, 95% CI= 1.47-1.62, P(heterogeneity)= 0.143; TT vs. CC, OR= 2.86, 95% CI= 2.29-3.57, P(heterogeneity)= 0.302; TC vs. CC, OR= 1.45, 95% CI= 1.38-1.53, P(heterogeneity)= 0.273; TT+ TC vs. CC, OR= 1.49, 95% CI= 1.42-1.56, P(heterogeneity)= 0.208; TT vs. TC+ CC, OR= 2.52, 95% CI= 1.95-3.25, P(heterogeneity)= 0.296). Furthermore, significant relationships were detected among PTPN22 1858C/T polymorphism and RF(+) or RF(-) RA risk. No obvious evidence of publication bias was detected in the overall analysis. Our study indicated that PTPN22 1858T allele was significantly associated with increased RA risk.