失眠与骨质疏松因果关联的孟德尔随机化研究

A Mendelian randomization study on the relationship between insomnia and osteoporosis

目的探讨失眠与骨质疏松的因果关系。方法孟德尔随机化(MR)研究。从全基因组关联分析研究数据中筛选与失眠相关的单核苷酸多态性(SNP)作为工具变量,用逆方差加权(IVW)、MR-Egger回归、加权中位数法、最大似然估计、线性中位数加权及孟德尔随机化稳健调整轮廓分数法(MR-RAPS)来确定失眠与骨质疏松之间是否存在因果关系,以比值比(OR)值及95%置信区间(95%CI)评价失眠与骨质疏松之间的关联强度。采用Cochran′s Q法检验SNP的异质性,采用MR-Egger回归检验水平多效性,采用留一法检验敏感性。采用MR多效性残差和与异质性检验法(MR-PRESSO)与径向孟德尔随机化(radial MR)检测纳入的SNP是否有离群值。结果以MR三大假设为前提设置筛选标准,最终31个SNP纳入MR分析。以IVW为主要分析方法,因果关联效应分析显示,失眠导致骨质疏松的发生风险上升约0.7%(OR=1.007,95%CI 1.001~1.014,P=0.044)。异质性检验显示SNP之间存在异质性(Q=57.91,P<0.001)。MR-Egger截距测试显示本研究没有水平多效性(截距值=3.807×10^(-5),P=0.888)。留一法检验显示没有单个SNP对整体的结果有较大影响。MR-PRESSO结果显示未检测到异常的SNP(P=0.059),radial MR未发现离群值。结论通过孟德尔随机化分析得出失眠可增加骨质疏松的发病风险。

ObjectiveTo explore the relationship between insomnia and osteoporosis.MethodsMendelian randomization(MR)analysis were used in this study.The single nucleotide polymorphisms(SNPs)related to insomnia from genome-wide association analysis research data were selected as the instrumental variables by using inverse variance weighted(IVW),MR-Egger regression,weighted median method,maximum likelihood,penalized weighted median estimator,and Mendelian randomization robust adjusted profile score(MR-RAPS)to determine the causal relationship between insomnia and osteoporosis.Odds ratio(OR)and 95%confidence interval(CI)values were used to evaluate the association between insomnia and osteoporosis.Cochran′s Q-test was used to detect heterogeneity of SNPs,MR-Egger regression was used to test for level pleiotropy,and the leave-one-out method was used to test sensitivity,MR pleiotropy residual sum and outlier(MR-PRESSO)method and radial MR were used to detect erroneous outliers.ResultsThe screening criteria were set based on the three major assumptions of MR;finally,31 SNPs were included in the MR analysis.The results of MR causal effect analysis using the IVW method showed that insomnia increased the risk of osteoporosis by about 0.7%(OR=1.007,95%CI 1.001-1.014,P=0.044);heterogeneity testing showed heterogeneity between SNPs(Q=57.91,P<0.001);and the MR-Egger intercept test did not indicate horizontal pleiotropy in this study(intercept value=3.807×10^(-5),P=0.888).Leave-one-out method showed that no single SNP had a significant impact on the overall results.No abnormal SNP was detected according to the MR-PRESSO results(P=0.059),and radial MR did not detect any outliers.ConclusionMendelian randomization analysis showed that insomnia can increase the risk of osteoporosis.