Background: The relationship between antisense non-coding RNA (ncRNA) in the INK4 locus (ANRIL) polymorphisms and coronary artery disease (CAD) remains inconclusive. Thus, we conducted this meta-analysis to better evaluate the roles of ANRIL polymorphisms in CAD. Methods: Systematic literature search of PubMed, Medline, and Embase was performed to identify potential relevant articles. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to estimate the strength of association. Results: Fifteen studies were finally enrolled for analyses. Overall analyses suggested that rs1333040 (dominant model: P<0.0001; recessive model: P<0.0001; allele model: P<0.0001), rs1333049 (dominant model: P=0.02; allele model: P=0.02) and rs2383207 (additive model: P=0.004) polymorphisms were significantly associated with the risk of CAD. Further subgroup analyses showed that rs1333040, rs1333049, and rs2383207 polymorphisms were significantly correlated with the risk of CAD in East Asians, rs2383206 and rs10757274 polymorphisms were significantly correlated with the risk of CAD in West Asians, while rs2383206, rs10757274, and rs10757278 polymorphisms were significantly correlated with the risk of CAD in Caucasians. Conclusion: Our findings indicated that rs1333040, rs1333049, rs2383206, rs2383207, rs10757274, and rs10757278 polymorphisms might serve as genetic biomarkers of CAD in certain ethnicities.