Published data on the association between tumor necrosis factor-alpha (TNF-α) promoter-308 A/G polymorphism and systemic lupus erythematosus (SLE) risk are inconclusive. To derive a more precise estimation of the relationship, a meta-analysis was performed. A total of 28 studies including 2,992 cases and 4,326 controls (5,924 cases and 8,484 controls in A versus G comparison) were involved in this meta-analysis. Meta-analysis was performed for genotypes A/A (recessive effect), A/A+A/G (dominant effect), and A allele in fixed or random effects models. In addition, we also performed a "model-free" analysis by considering the G/G genotype as the reference and estimated the OR for the A/A versus G/G and A/G versus G/G genotype. Overall, an association of TNF-α promoter-308 A/G polymorphism with SLE was found (A versus G: OR = 1.686, 95% CI = 1.400-2.032, P < 0.001; A/A versus A/G+G/G: OR = 3.043, 95% CI = 2.185-4.238, P < 0.001; A/A+A/G versus G/G: OR = 1.822, 95% CI = 1.379-2.407, P < 0.001; A/A versus G/G: OR = 3.686, 95% CI = 2.628-5.172, P < 0.001; A/G versus G/G: OR = 1.691, 95% CI = 1.291-2.215, P < 0.001). However, stratification by ethnicity indicated that the risk A allele was not associated with SLE in Asian (A versus G: OR = 1.207, 95% CI = 0.856-1.702, P = 0.283) and African population (A versus G: OR = 1.225, 95% CI = 0.597-2.516, P = 0.580). In summary, this meta-analysis indicated that TNF-α promoter-308-A/G polymorphism is associated with susceptibility to SLE.