A meta-analysis of the relationship between MTHFR gene A1298C polymorphism and the risk of adult stroke
Ming-Jie Zhang,Zi-Cheng Hu,Yan-Wei Yin,Bing-Hu Li,Yun Liu,Shao-Qiong Liao,Chang-Yue Gao,Jing-Cheng Li,Li-Li Zhang
DOI: https://doi.org/10.1159/000369122
Abstract:Background: The association between methylenetetrahydrofolate reductase (MTHFR) gene A1298C polymorphism and adult stroke remains controversial. The present article was designed to clarify this relationship through pooled analysis of the numerous epidemiological studies focusing on this association. Methods: We comprehensively searched all published papers in electronic database including PubMed, Embase, Web of Science, Chinese Biomedical Literature on disc (CBMdisc) and China National Knowledge Infrastructure (CNKI) up to 2013. The pooled odds ratios (ORs) with 95% confidence intervals (CIs) for allelic (C allele vs. A allele), additive (CC vs. AA), dominant (CC+AC vs. AA), and recessive (CC vs. AA+AC) models were calculated. Subgroup and sensitivity analyses were performed to detect the heterogeneity and examine the reliability of results, respectively. Begg's funnel plots and Egger's regression test were used to assess the potential publication bias. Results: A total of fifteen studies containing 2,361 cases and 2,653 controls were included in the final meta-analysis. The combined results of overall analysis showed that there was significant association between MTHFR gene A1298C polymorphism and adult stroke (allelic model: OR=1.36, 95% CI=1.11-1.67; additive model: OR=1.88, 95% CI=1.12-3.18; dominant model: OR=1.33, 95% CI=1.08-1.65 and recessive model: OR=1.77, 95% CI=1.07-2.94, respectively). On subgroup analysis by ethnicity of study population, significant association was shown in meta-analysis based on Asian population (allelic model: OR=1.40, 95% CI=1.19-1.65; additive model: OR=2.58, 95% CI=1.34-4.96; dominant model: OR=1.44, 95% CI=1.20-1.73 and recessive model: OR=2.12, 95% CI=1.20-3.76, respectively), but not in Caucasian population (allelic model: OR=1.30, 95% CI=0.93-1.82; additive model: OR=1.65, 95% CI=0.81-3.33; dominant model: OR=1.17, 95% CI=0.86-1.61 and recessive model: OR=1.70, 95% CI=0.83-3.50, respectively). In addition, the heterogeneity was effectively removed or decreased by limiting the included studies with population of Asian ethnicity. Furthermore, the corresponding pooled ORs were not materially changed in all genetic models of meta-analysis after limiting the included studies with population-based controls. However, except the recessive model, publication bias presented in the allelic, additive, dominant models identified by the Begg's funnel plots and Egger's regression test. Conclusions: In conclusion, the overall analysis suggests that MTHFR gene A1298C polymorphism plays an important role in the development of adult stroke. Genotype CC of MTHFR-1298A/C could increase the risk of stroke and may act as a predictor for clinical evaluation, especially in the Asian population. More studies with large-scale and different ethnicities are required to further confirm our findings.