The association of the CYP2J2 G-50T polymorphism with coronary artery disease has been explored, but the results remain controversial. Thus, a meta-analysis was conducted to provide a comprehensive estimate of this association. We selected ten articles encompassing 12 independent case-control studies with 7063 cases and 10,453 controls for this meta-analysis. Overall, we found significant associations between the CYP2J2 G-50T polymorphism and coronary artery disease risk in three genetic models (allele model: odds ratio (OR) = 1.19, 95% confidence interval (CI) = 1.05–1.34; homozygote model: OR = 2.25, 95% CI = 1.27–4.01; recessive model: OR = 2.17, 95% CI = 1.22–3.86). In these three genetic models, a significant association was observed in Caucasians but not in Asians when the data were stratified by ethnicity. However, no significant associations were found between the CYP2J2 polymorphism G-50T and coronary artery disease risk in heterozygote model and dominant model. In conclusion, our meta-analysis suggested that the CYP2J2 G-50T polymorphism was associated with coronary artery disease risk in the allele, homozygote and recessive models in Caucasians.