BACKGROUND The association between PTPN22 R620W polymorphism and risk of myasthenia gravis (MG) remains controversial. Therefore, we did this meta-analysis to investigate this association. METHODS We did a comprehensive search in PubMed, Medline, Embase, CNKI (China National Knowledge Infrastructure), and Wanfang electronic databases to retrieve relevant articles. The overall effect was measured by odds ratios (ORs) with its 95% confidence intervals (CIs). Statistical analyses were conducted with STATA software. RESULTS Overall, a total of 7 case-control studies with 2802 cases and 3730 controls were finally included in this review. PTPN22 R620W polymorphism was significantly associated with an increased risk of MG (OR=1.57; 95% CI, 1.34-1.82; I(2)=31%). In the subgroup analysis, thymoma patients were significantly associated with risk of MG (OR=1.59; 95% CI, 1.28-1.98; I(2)=0%). However, non-thymoma patients with this polymorphism did not have increased MG risk (OR=1.36; 95% CI, 0.86-2.15; I(2)=77%). In addition, PTPN22 R620W polymorphism showed increased early-onset myasthenia gravis (EOMG) risk (OR=2.38; 95% CI, 1.52-3.71; I(2)=0%). CONCLUSIONS This meta-analysis shows a significant association between PTPN22 R620W polymorphism and MG risk.