| Objective:To assess the relationship between three polymorphisms of eNOS gene, G894T, T786C and 4b/a, with the risk of diabetic nephropathy (DN) via meta-analysis.Methods:Pubmed, Embase, Medline, the China National Knowledge Infrastructure, WanFang database were searched for articles published from January 2000 to September 2015. According to the inclusion and exclusion criterias, we chose the case-control studies which were associated with eNOS gene polymorphisms (G894T, T786C or 4b/a) and DN. The distribution of genotype in the control groups were compatible with Hardy-weinberg equilibrium. We used odds ratio (OR) and its 95% confidence interval (95%CI) to evaluate the strength of association between three polymorphisms and risk of DN by STATA12.0. Publication bias is testing through the Begger and Egger regression. The study includes five genetic models, for example allele model (T vs.G, C vs.T, a vs. b), additive model (TT vs. GG, CC vs.TT, aa vs. bb), dominant model (TT+GT vs. GG, CC+TC vs. TT, aa+ab vs. bb), recessive model (TT vs. GT+GG, CC vs.TC+TT, aa vs. ab+bb), codominance model (GT vs. GG+TT, TC vs. TT+CC, ab vs. aa+bb).Results:1、Thirty-three studies were eligible for the meta-analysis (G894T:19 studies,4b/a:20 studies, T786C:8studies).2、Considering G894T polymorphism, an association with the risk of DN was observed in all genetic models (allele model T vs. G:OR=1.518, P< 0.001; additive model TT vs. GG:OR=1.486, P< 0.05; dominant model TT+GT vs. GG:OR=1.510, P< 0.05; recessive model TT vs. GT+GG: OR=1.381, P< 0.001; codominance model GT vs. GG+TT:OR=1.323, P< 0.05). Subgroup analysis revealed a significant association between G894T polymorphism and DN risk in Asian populations (P< 0.05), but not in Caucasian population (P> 0.05). At the same time, the results showed that the G894T polymorphism could significantly increase the incidence of DN in type 2 diabetes patients (P< 0.01).3、Meanwhile, T786C polymorphism showed association with all models except for the codominance model (allele model C vs. T:OR=1.184, P< 0.01; additive model CC vs. TT: OR=1.380, P< 0.05; dominant model CC+TC vs. TT:OR=1.201, P< 0.05; recessive model CC vs. TC+CC:OR=1.307, P< 0.05; codominance model TC vs. CC+TT:OR=1.082, P> 0.05). However, for the region subgroup, there were no obvious association with DN risk (P> 0.05).4、As for the 4b/a polymorphism, a significant association with DN risk was observed except for dominant and codominance model, (allele model a vs. b:OR=1.194, P< 0.05; additive model aa. vs. bb:OR=1.780, P< 0.05; recessive model aa. vs. ab+bb:OR=1.795, P< 0.05; dominant model aa+ab vs. bb:OR=1.112, P> 0.05; codominance model ab vs. aa+bb:OR=0.992, P> 0.05), but not observed in African and Caucasia populations (P> 0.05).Conclusions:1. In the present study, the evidence suggested that G894T polymorphism in the eNOS gene were strongly associated with susceptibility of DN, especially in Asian populations, and it was also typically found in patients with type 2 diabetes mellitus.2. And for the T786C polymorphism, we found that the individuals containing the CC genotype and the C allele were more likely to suffer from DN.3. As for the 4b/a polymorphism, the aa genotype and the a allele were the risk factors of DN.4. The data suggested that the eNOS gene polymorphisms could play a significant role in the development of DN. |
PDF Full Text Request