Abstract
PURPOSE: Paraoxonase 1 (PON1) polymorphisms have been implicated as risk factors for coronary artery disease, but the results of genetic association studies on the related phenotype of ischemic stroke are inconclusive. We performed a meta-analysis of published studies investigating the association between ischemic stroke and two nonsynonymous PON1 polymorphisms, rs662 (p.Q192R) and rs854560 (p.L55M) in humans. METHODS: We searched multiple electronic databases through June 30, 2009 for eligible studies. In main analyses, we calculated allele-based odds ratios with random effects models. In secondary analyses, we examined dominant and recessive genetic models as well, and performed subgroup and sensitivity analyses. RESULTS: Regarding rs662, we identified 22 eligible studies (total of 7384 cases/11,074 controls), yielding a summary odds ratio of 1.10 per G allele (95% confidence interval, 1.04-1.17) with no evidence of between-study heterogeneity. For rs854560, 16 eligible studies (total of 5518 cases/8951 controls) yielded a summary odds ratio of 0.97 per T allele (95% confidence interval, 0.90-1.04), again with no evidence of between-study heterogeneity. For both polymorphisms, analyses with dominant and recessive genetic models yielded the same inferences as allele-based comparisons. Subgroup and sensitivity analyses showed similar results. CONCLUSION: In agreement with observations in coronary artery disease, PON1 rs662 appears to be associated with a small increase in the risk of ischemic stroke.