Abstract
AIMS/INTRODUCTION: Studies have been carried out to evaluate the correlation between TCF7L2 genetic polymorphisms and gestational diabetes mellitus (GDM) risk. However, the conclusions from these studies are incomplete, because partial single nucleotide polymorphisms (SNPs) were analyzed. We carried out a meta-analysis aimed to systematically evaluate TCF7L2 gene polymorphisms and GDM susceptibility in all population and racial/ethnic subgroups to afford a foundation for future research. MATERIALS AND METHODS: Published studies censoring TCF7L2 variants and GDM risk were captured from the EMBASE, PubMed, CNKI and Wanfang databases. The meta-analysis was processed using software of RevMan 5.2 and Stata13. The relationship between TCF7L2 polymorphism and GDM occurrence was evaluated by pooled odds ratios. Stratified analysis based on race/ethnicity was also carried out. The allele-specific odds ratios and 95% confidence intervals were counted, and based on homogeneity evaluated using the I(2) -test, fixed- or random-effects pooled measures were selected. RESULTS: A total of 22 studies were covered, capturing eight TCF7L2 SNPs and involving 5,573 cases and 13,266 controls. Six of eight SNPs showed significant relationships with GDM occurrence, of which the SNPs rs7903146, rs12255372 and rs7901695 were the most powerful. Stratified analysis by race/ethnicity showed discrepant results in these three SNPs. In Caucasians and other races, all these SNPs were found to have a significant association with GDM risk, but in Asians, only SNP rs7903146 showed a significant association. CONCLUSIONS: Six of eight SNPs were found to have significant associations between TCF7L2 variants and GDM risk in the overall population, with the most powerful in SNPs being rs7903146, rs12255372 and rs7901695, but the contribution of these SNPs to GDM risk were variable among different racial/ethnic groups.