吸烟、饮酒与偏头痛关系的多变量孟德尔随机化研究  

Multivariate Mendelian randomization study on the relationship between smoking,alcohol consumption and migraine

作　　者：冯昌强 林燕妮 梁霞 童玲玲 蒋小英 朱健清 FENG Changqiang;LIN Yanni;LIANG Xia;TONG Lingling;JIANG Xiaoying;ZHU Jianqing(Department of Neurology,the Sixth Affiliated Hospital of Guangxi Medical University,Guangxi Zhuang Autonomous Region,Yulin537000,China)

机构地区：[1]广西医科大学第六附属医院神经内科,广西玉林537000

出　　处：《中国医药导报》2024年第26期104-110,共7页China Medical Herald

基　　金：广西壮族自治区玉林市科学研究与技术开发计划项目(玉市科202235028)。

摘　　要：目的探讨吸烟、饮酒与偏头痛的因果关系。方法利用两样本多变量孟德尔随机化(MR)研究,以单核苷酸多态性(SNP)作为工具变量,吸烟、饮酒为暴露因素,偏头痛为结局。采用逆方差加权(IVW)法、加权中位数法、MR-Egger回归法、加权模式法和简单模式法评估吸烟、饮酒对偏头痛的影响;同时进行反向MR分析探索反向因果关系。使用Cochran’s Q检验进行异质性分析,MR-Egger截距法分析多效性,留一法检验是否存在某个特定SNP引起的偏倚,漏斗图评估因果关联的可靠性。结果正向分析显示,IVW法中OR值为1.39(95%CI:1.06~1.83),差异有统计学意义(P<0.05),同时加权中位数法OR值与其方向相同;多变量分析在调整饮酒的影响后,IVW法中OR值为1.71(95%CI:1.53~1.89),差异有统计学意义(P<0.05)。反向分析显示,偏头痛和吸烟不存在因果关联。正向分析显示,IVW法中OR值为0.58(95%CI:0.43~0.78),结果有统计学意义(P<0.05),同时加权中位数法和加权模式法OR值与其方向相同;多变量分析在调整吸烟的影响后,IVW法中OR值为0.66(95%CI:0.40~0.92),差异有统计学意义(P<0.05),且多变量MR-Egger回归法OR值方向同IVW法。反向分析显示,IVW法中OR值为0.94(95%CI:0.91~0.96),差异有统计学意义(P<0.05)。正向分析显示,吸烟、饮酒与偏头痛存在异质性、不存在多效性;反向分析显示,偏头痛与吸烟存在异质性、不存在多效性。留一法显示,没有单个SNP对结果造成明显影响。漏斗图显示,工具变量对称分布,结果无偏倚,较为稳健。结论本研究提供了遗传证据,支持吸烟是偏头痛的有害因素,而饮酒与偏头痛风险降低的关系可能部分归因于反向因果关系。Objective To explore the causal relationship between smoking,alcohol consumption and migraine.Methods A two-sample multivariate Mendelian randomized(MR)study was conducted with single nucleotide polymorphism(SNP)as instrumental variable,smoking and alcohol consumption as exposure factors and migraine as outcome.Inverse variance weighting(IVW)method,weighted median method,MR-Egger regression method,weighted model method,and simple model method were used to evaluate the effects of smoking and alcohol consumption on migraine;at the same time,reverse MR analysis was performed to explore reverse causality.Cochran’s Q test was used for heterogeneity analysis,MR-Egger intercept method was used to analyze pleiotropy,and remain-one method was used to test whether there was bias caused by a specific SNP,and funnel plot was used to evaluate the reliability of causal association.Results Forward analysis showed that the OR value of IVW method was 1.39(95%CI:1.06-1.83),the difference was statistically significant(P<0.05),and the OR value of weighted median method was the same as its direction;after adjusting for the influence of alcohol consumption in multivariate analysis,the OR value of IVW method was 1.71(95%CI:1.53-1.89),and the difference was statistically significant(P<0.05).Reverse analysis showed no causal link between migraines and smoking.Forward analysis showed that the OR value of IVW method was 0.58(95%CI:0.43-0.78),and the difference was statistically significant(P<0.05),the OR value of weighted median method and weighted model method was the same as its direction;after adjusting the influence of smoking in multivariate analysis,the OR value of IVW method was 0.66(95%CI:0.40-0.92),and the difference was statistically significant(P<0.05),and the direction of OR value of multivariate MR-Egger regression method was the same as that of IVW method.Reverse analysis showed that the OR value of IVW method was 0.94(95%CI:0.91-0.96),and the difference was statistically significant(P<0.05).The forward analysis showed that

关 键 词：吸烟 饮酒 偏头痛 孟德尔随机化 

分 类 号：R747.2[医药卫生—神经病学与精神病学]