PDF(Pubmed)
Abstract：
The causal association between vitamin E status and osteoarthritis (OA) remains controversial in previous epidemiological studies. We employed a Mendelian randomization (MR) analysis to explore the causal relationship between circulating alpha-tocopherol levels (main forms of vitamin E in our body) and OA. The instrumental variables (IVs) of circulating alpha-tocopherol levels were obtained from a Genome-wide association study (GWAS) dataset of 7781 individuals of European descent. The outcome of OA was derived from the UK biobank. Two-sample MR analysis was used to estimate the causal relationship between circulating alpha-tocopherol levels and OA. The inverse-variance weighted (IVW) method was the primary analysis in this analysis. We used the MR-Egger method to determine horizontal pleiotropic in this work. The heterogeneity effect of instrumental IVs was detected by MR-Egger and IVW analyses. Sensitivity analysis was performed by removing single nucleotide polymorphism (SNP) one by one. Three SNPs (rs964184, rs2108622, and rs11057830) (P < 5E-8) strongly associated with circulating alpha-tocopherol levels were used in this analysis. The IVW-random effect indicated no causal relationship between circulating alpha-tocopherol levels and clinically diagnosed OA (OR = 0.880, 95% CI 0.626, 1.236, P = 0.461). Similarly, IVW analysis showed no causal association between circulating alpha-tocopherol levels and self-reported OA (OR = 0.980, 95% CI 0.954, 1.006, P = 0.139). Other methods of MR analyses and sensitivity analyses revealed consistent findings. MR-Egger and IVW methods indicated no significant heterogeneity between IVs. The MR-Egger intercept showed no horizontal pleiotropic. The results of this linear Mendelian randomization study indicate no causal association between genetically predicted alpha-tocopherol levels and the progression of OA. Alpha-tocopherol may not provide beneficial and more favorable outcomes for the progression of OA. Further MR analysis based on updated GWASs with more IVs is required to verify the results of our study.
摘要：
在先前的流行病学研究中，维生素E状态与骨关节炎（OA）之间的因果关系仍然存在争议。我们采用孟德尔随机化（MR）分析来探索循环α-生育酚水平（体内维生素E的主要形式）与OA之间的因果关系。循环α-生育酚水平的工具变量（IV）是从7781名欧洲血统个体的全基因组关联研究（GWAS）数据集中获得的。OA的结果来自英国生物银行。使用双样本MR分析来估计循环α-生育酚水平与OA之间的因果关系。逆方差加权(IVW)方法是该分析中的主要分析。在这项工作中，我们使用了MR-Egger方法来确定水平多效性。通过MR-Egger和IVW分析检测到仪器IVs的异质性效应。通过逐一去除单核苷酸多态性（SNP）进行敏感性分析。在该分析中使用与循环α-生育酚水平强相关的三个SNP(rs964184、rs2108622和rs11057830)(P<5E-8)。IVW随机效应表明循环α-生育酚水平与临床诊断的OA之间没有因果关系（OR=0.880，95％CI0.626，1.236，P=0.461）。同样，IVW分析显示，循环α-生育酚水平与自我报告的OA之间没有因果关系（OR=0.980，95％CI0.954，1.006，P=0.139）。其他MR分析和敏感性分析的方法显示了一致的结果。MR-Egger和IVW方法表明IVs之间没有明显的异质性。MR-Egger截距显示无水平多效性。这项线性孟德尔随机化研究的结果表明，遗传预测的α-生育酚水平与OA的进展之间没有因果关系。α-生育酚可能不会为OA的进展提供有益和更有利的结果。需要基于具有更多IVs的更新GWAS的进一步MR分析来验证我们的研究结果。
