The relationship of TNF-alpha promoter polymorphisms and ankylosing spondylitis (AS) has been reported with conflicting results. We perform this meta-analysis to collect all the relevant studies up to date to further clarify the association of TNF-alpha promoter polymorphisms with AS. A review was conducted of studies reporting on the association between TNF-alpha promoter polymorphisms and AS susceptibility in Medline, Pubmed, Embase, and Web of Science. The numbers of individuals with various genotypes and alleles in both the case and control groups were extracted from relevant studies. Odds ratios (ORs) with 95% confidence interval (CI) were used to estimate the association. Fourteen eligible studies, contributing data on 3,880 subjects (1,766 patients; 2,114 controls), were included in this meta-analysis. The ORs of various comparisons indicated that there was no association between TNF-alpha 238, 308 polymorphisms, and AS susceptibility in the overall population. For HLA-B27+ population, although the frequency of 308 A allele decreased in AS patients (OR = 0.721; 95%CI = 0.522-0.995), the result was no longer statistically significant after excluding the Hardy-Weinberg equilibrium violation studies (OR = 1.150; 95%CI = 0.568-2.310). No relationship was found between TNF-alpha promoter 238 polymorphisms and AS in HLA-b27+ population. No association was found between TNF-alpha promoter 238/308 polymorphisms and ankylosing spondylitis susceptibility in both the overall and HLA-B27+ population.