基于两样本孟德尔随机化的身高和冠心病风险之间因果关系  

Causal association between height and risk of coronary heart disease: a two-sample Mendelian randomization analysis

作　　者：李云霞 李洪凯[1,2] 马韫韬 于媛媛 孙晓茹 刘新辉[1,2] 司书成 侯蕾 袁同慧 刘璐 李文超[1,2] 薛付忠 刘言训 LI Yunxia;LI Hongkai;MA Yuntao;YU Yuanyuan;SUN Xiaoru;LIU Xinhui;SI Shucheng;HOU Lei;YUAN Tonghui;LIU Lu;LI Wenchao;XUE Fuzhong;LIU Yanxun(Department of Biostatistics,School of Public Health,Cheeloo College of Medicine,Shandong University,Jinan 250012,Shandong,China;Institute for Medical Dataology,Shandong University,Jinan 250002,Shandong,China;Lanzhou University Second Hospital,Lanzhou 730030,Gansu,China)

机构地区：[1]山东大学齐鲁医学院公共卫生学院生物统计学系,山东济南250012 [2]山东大学健康医疗大数据研究院,山东济南250002 [3]兰州大学第二临床医学院,甘肃兰州730030

出　　处：《山东大学学报（医学版）》2020年第5期107-114,共8页Journal of Shandong University:Health Sciences

摘　　要：目的采用孟德尔随机化方法探讨身高和冠心病之间的因果关系。方法对基于全基因关联研究(GWAS)的Meta分析数据进行二次数据分析。利用大样本GWAS汇总数据,选择与身高密切相关的遗传位点作为工具变量,分别用逆方差加权法、加权中位数法和MR-Egger回归作两样本孟德尔随机化分析,以OR值评价身高与冠心病之间的因果关系。结果共纳入108个SNP位点作为工具变量,随机效应逆方差加权法和加权中位数法估计得OR值分别为0.760 5(0.644 8~0.896 9)和0.737 9(0.589 9~0.923 1),两种方法结果相近,均支持身高与冠心病之间存在因果关系。MR-Egger回归方程截距项为-0.003 1(-0.013 8~0.007 6),表明因果估计结果受基因多效性影响的可能性较小。身高每增加一倍标准差(SD=0.0699m),冠心病风险降低约25%。结论身高与冠心病之间存在负向因果关联。Objective To explore the causal relationship between height and cardiovascular heart disease(CHD) using the Mendelian randomization(MR) analysis. Methods This study was a secondary data analysis based on Meta-analysis data of Genome-Wide Association Study(GWAS). Genetic variants which were closely related to height in large sample GWAS summary data were identified as instrumental variables(IV). Using odds ratio(OR) as outcome indicator, the causal relationship between height and CHD was analyzed by two-sample MR methods, including inverse variance weighted(IVW) method, weighted median estimator(WME) method and MR-Egger regression. Results A total of 108 SNPs were chosen as IV. The OR and 95%CI between height and CHD were estimated to be 0.760 5(0.644 8-0.896 9) and 0.737 9(0.589 9-0.923 1) by IVW and WME methods, respectively. The results were similar, and suggested a causal relationship between height and CHD. The MR-Egger regression results suggested that the genetic pleiotropy was unlikely to bias our results(the intercept was-0.003 1, 95%CI:-0.013 8-0.0076). The risk of CHD decreased by about 25% when height increase one standard deviation(SD=0.0699 m). Conclusion There is a negative causal relationship between height and CHD.

关 键 词：孟德尔随机化 身高 冠心病 因果推断 

分 类 号：R541.4[医药卫生—心血管疾病]