基于两样本孟德尔随机化的甲状腺功能减退症与骨质疏松症因果关系研究

The causal relationship between hypothyroidism and osteoporosis by two sample Mendelian randomization

目的运用孟德尔随机化(mendelian randomization,MR)方法探究甲状腺功能减退症(hypo-thyroidism,HT)与骨质疏松症(osteoporosis,OP)的因果关系。方法本研究基于全基因组关联研究(genome-wide association study,GWAS)的汇总数据进行二次数据分析,分别从多数据库筛选出与HT和OP密切相关的单核苷酸多态性位点(single nucleotide polymorphisms,SNPs)作为工具变量,采用逆方差加权法(inverse variance weighted,IVW)、MR-Egger法、加权中位数估计(weighted median,WM)法、Simple Mode法、Weighted Mode法进行双向两样本MR分析,并使用Meta分析对多组MR结果合并进行总风险评估。结果进行异质性、发表偏倚及敏感性分析。MR结果以OR值和95%置信区间表示,以OR值评估HT与OP的因果关系。结果以IVW法为主要MR方法,合并进行Meta分析,结果显示遗传基因决定的HT患者发生OP的风险增加(OR=1.0016,95%CI:1.0008~1.0025,P<0.01),且未观察到OP与HT的发生风险相关(OR=1.0816,95%CI:0.8252~1.4175,P>0.05)。结果不存在异质性及发表偏倚(P>0.05),敏感性分析均显示结果稳健。结论MR联合Meta分析发现遗传基因决定的HT与OP的发生存在因果关联,可能会增加OP的患病风险。在HT患者中积极进行OP筛查可有利于早期发现与及时干预。

Objective To explore the causal relationship between hypothyroidism(HT)and osteoporosis(OP)using mendelian randomization(MR).Methods A secondary analysis of genome-wide association study(GWAS)was conducted.The single nucleotide polymorphisms(SNPs)closely related to HT and OP were screened from multiple databases as instrumental variables.Inverse variance weighted(IVW)method,MR-Egger method,weighted median(WM)method,Simple Mode method and Weighted Mode method were used for two-way two-sample MR analysis.Meta-analysis was used to assess total risk across multiple sets of MR.The results were tested by heterogeneity analysis,publication bias and sensitivity analysis.The MR results were represented by OR values and 95%confidence intervals.The causal relationship between HT and OP was evaluated through OR value.Results IVW as the main MR method combined with Meta-analysis showed that the risk of OP increased in genetically determined HT patients(OR=1.0016,95%CI:1.0008-1.0025,P<0.01),and no correlation between OP and the risk of HT was observed(OR=1.0816,95%CI:0.8252-1.4175,P>0.05).There was no heterogeneity and publication bias in the analysis(P>0.05).Sensitivity analysis showed that the results were robust.Conclusion MR combined with Meta-analysis method reveals that genetically determined HT has causal relationship with OP and may increase the risk of OP.Active screening for OP in HT patients is beneficial to early detection and timely intervention.