Background The aim of this study was to explore the association between tumor necrosis factor superfamily number 4 (TNFSF4) rs1234315, rs2205960 polymorphisms and systemic lupus erythematosus (SLE) susceptibility. Methods A meta-analysis was performed on the association between rs1234315 and rs2205960 polymorphisms and SLE by allelic contrast, additive model, recessive model and dominant model. Results Regarding rs1234315 polymorphism, a total of five studies were included (6575 cases, 14,798 controls). Meta-analysis showed significant associations between the T allele and SLE in overall subjects and Asians (OR = 1.310, 95%CI: 1.104–1.553, p = 0.002; OR = 1.458, 95%CI: 1.328–1.602, p < 0.001). With respect to the rs2205960 polymorphism, significant associations between the T allele and SLE were found in all subjects (OR = 1.333, 95%CI: 1.254–1.418, p < 0.001), Asians (OR = 1.407, 95%CI: 1.345–1.471, p < 0.001) and Europeans (OR = 1.254, 95%CI: 1.185–1.328, p < 0.001). Results also showed significant associations between the additive model and SLE in all subjects and Asians (OR = 1.934, 95%CI: 1.500–2.494, p < 0.001; OR = 1.882, 95%CI: 1.318–2.689, p = 0.001). Furthermore, we detected significant associations between the dominant model and SLE in all subjects and Asians (OR = 1.421, 95%CI: 1.239–1.629, p < 0.001; OR = 1.297, 95%CI: 1.083–1.555, p = 0.005). Significant associations were found between the recessive model and SLE in overall subjects and Asians (OR = 1.677, 95%CI: 1.312–2.144, p < 0.001; OR = 1.751, 95%CI: 1.235–2.483, p = 0.002). Conclusion The present study suggested that TNFSF4 rs1234315 and rs2205960 polymorphisms were associated with SLE susceptibility.