失眠与2型糖尿病的因果关系:一项双向双样本孟德尔随机化研究
Association between insomnia and type 2 diabetes:a bidirectional two-sample Mendelian randomization study
黄飘 1艾思志 2王昆莹 3赖水青 1关海霞1
作者信息
- 1. 广东省心血管病研究所广东省人民医院广东省医学科学院,广东 广州 510080
- 2. 广州医科大学附属脑科医院睡眠与节律医学中心,广东 广州 510370
- 3. 福建省南安市医院内分泌科,福建 南安 362300
- 折叠
摘要
目的 应用孟德尔随机化(MR)方法,探索失眠与2型糖尿病(T2DM)的因果关系.方法 利用大规模人群全基因组关联研究(GWAS)的汇总数据,使用随机效应逆方差加权(IVW)法作为两样本MR分析的主要分析方法,并采用敏感性分析[包括加权中位数法(WM)、MR-Egger回归分析法和逐个剔除检验等]评估MR分析结果的稳健性.结果 IVW法分析结果显示,失眠与T2DM发病风险增加存在因果关联(IVW:OR 1.10,95％CI 1.06～1.14,P＜0.001).加权中位数方法和MR-Egger回归均支持该结论(WM:OR 1.09,95％CI1.06～1.13,P＜0.001;MR-Egger:OR 1.19,95％CI 1.03～1.37,P＜0.05).反向MR分析提示T2DM与失眠风险增加之间无因果关系(IVW:OR 1.01,95％CI1.00～1.03,P=0.058).研究的工具变量均不存在水平多效性,敏感性分析提示MR结果稳定.结论 研究结果支持失眠为T2DM的风险因素,而T2DM对失眠风险无显著影响.
Abstract
Objective To explore the association between insommia and type 2 diabetes(T2DM)by using Mendelian randomization(MR).Methods Summary data from large-scale population-based genome-wide association studies(GWAS)was used.The inverse variance weighted(IVW)method was used as the primary analysis approach to evaluate the causal effects.Sensitivity analyses,including weighted median(WM)method,MR-Egger regression method and leave-one-out method et al,were used to verify the reliability of the MR analysis results.Results The analysis of IVW demonstrated that insomnia could increase the risk of T2DM(IVW:OR 1.10,95％CI 1.06-1.14,P＜0.001).The results were supported by the WM and MR-Egger regression method(WM:OR 1.09,95％CI 1.06-1.13,P＜0.001;MR-Egger:OR1.19,95％CI 1.03-1.37,P＜0.05).Reverse MR analysis showed no relationship between T2DM and increased risk of insomnia(IVW:OR 1.01,95％CI 1.00-1.03,P=0.058).There was no horizontal pleiotropy in instrumental variables(IV).Sensitivity analysis suggested that MR analysis results were stable.Conclusion The results demonstrate that insomnia is a risk factor for T2DM.while T2DM has no significant effect on the risk of insomnia.
关键词
失眠/2型糖尿病/孟德尔随机化/因果关系Key words
insomnia/type 2 diabetes/Mendelian randomization/causality引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据