Table1_Association between the SLC6A11 rs2304725 and GABRG2 rs211037 polymorphisms and drug-resistant epilepsy: a meta-analysis.docx

Background: Previous studies have shown that SLC6A11 and GABRG2 are linked to drug-resistant epilepsy (DRE), although there have been conflicting results in the literature. In this study, we systematically assessed the relationship between DRE and these two genes.
Methods: We systematically searched the PubMed, Embase, Cochrane Library, Web of Science, Google Scholar, Wanfang Data, CNKI, and VIP databases. To clarify whether heterogeneity existed between studies, tools such as the Q-test and I2 statistic were selected. According to study heterogeneity, we chose fixed- or random-effects models for analysis. We then used the chi-squared ratio to evaluate any bias of the experimental data.
Results: In total, 11 trials and 3,813 patients were selected. To investigate the relationship with DRE, we performed model tests on the two genes separately. The results showed that SLC6A11 rs2304725 had no significant correlation with DRE risk in the allele, dominant, recessive, and additive models in a pooled population. However, for the over-dominant model, DRE was correlated with rs2304725 (OR = 1.08, 95% CI: 0.92–1.27, p = 0.33) in a pooled population. Similarly, rs211037 was weakly significantly correlated with DRE for the dominant, recessive, over-dominant, and additive models in a pooled population. The subgroup analysis results showed that rs211037 expressed a genetic risk of DRE in allele (OR = 1.01, 95% CI: 0.76–1.35, p = 0.94), dominant (OR = 1.08, 95% CI: 0.77–1.50, p = 0.65), and additive models (OR = 1.14, 95% CI: 0.62–2.09, p = 0.67) in an Asian population.
Conclusion: In this meta-analysis, our results showed that SLC6A11 rs2304725 and GABRG2 rs211037 are not significantly correlated with DRE. However, in the over-dominant model, rs2304725 was significantly correlated with DRE. Likewise, rs211037 conveyed a genetic risk for DRE in an Asian population in the allele, dominant, and additive models.