Several studies have suggested the association between ADAM 12 polymorphisms and the risk of osteoarthritis (OA), but the results remained controversial. Therefore, we designed a meta-analysis to systematically evaluate the association on this issue. A literature search for eligible studies was conducted in PubMed, Web of Science and Google Scholar databases. The association between ADAM 12 polymorphisms and knee OA risk was calculated by odds ratios (ORs) and 95% confidence intervals (CIs). Study heterogeneity, sensitivity and publication bias analyses were also conducted. Ten articles covering 5048 cases and 6848 controls met our criteria for the final analysis. We found that the rs1871054 was significantly associated with the risk of knee OA (allele model OR 1.72, 95% CI 1.43-2.07, P<0.001; additive model: OR 2.06, 95% CI 1.19-3.56, P=0.010; dominant model: OR 2.45, 95% CI 1.85-3.25, P<0.001; recessive model: OR 1.54, 95% CI 1.13-2.10, P=0.007). rs1044122 was significantly associated with knee OA susceptibility in recessive model (OR 1.45, 95% CI 1.03-2.04, P=0.031). For rs3740199 and rs1278279, no significant associations with knee OA were found. In the stratified analysis by gender, significant association was identified with the risk of knee OA for rs3740199 in men in allele model (OR 2.41, 95% CI 1.51-3.84, P<0.001), dominant model (OR 2.68, 95% CI 1.17-6.14, P=0.02) and recessive model (OR 3.51, 95% CI 1.68-7.36, P=0.001), but not for additive model (OR 1.30, 95% CI 0.81-2.08, P=0.28). This meta-analysis suggests that the ADAM 12 genetic polymorphisms rs1871054 and rs1044122 might be associated with risk of knee OA; rs3740199 might be associated with risk of knee OA in men. Further well-designed and large scale studies are warranted to validate these associations.