In the present study, no evidence of significant publication bias was found in either allelic or genotypic analysis (P values > 0.05 in both Egger's regression and Begg’s rank correlation (Kendall's tau) (Begg and Mazumdar 1994) tests).