儿童期体质指数与妊娠期高血压的潜在因果关系:孟德尔随机化研究

Potential Causal Relationship between Childhood BMI and Gestational Hypertension:a Mendelian Randomization Study

目的采用孟德尔随机化(MR)方法探讨儿童期体质指数(BMI)与妊娠期高血压(GH)的潜在因果关系。方法从全基因组关联研究(GWAS)数据库获取儿童期BMI数据集(ID:ebi-a-GCST90002409)和GH数据集(ID:finn-b-O15_HYPTENSPREG),其中儿童期BMI数据集的样本量为39620例,包括8173382个单核苷酸多态性(SNP);GH数据集的样本量为123579例,包括16379784个SNP。本研究采用逆方差加权法(IVW)、MR-Egger回归、加权中位数法、简单模式法、加权模式法分析儿童期BMI与GH的潜在因果关系。采用CochranQ检验评估各SNP间的统计学异质性,采用MR-Egger回归的截距项、MR-PRESSO检验及漏斗图分析SNP的水平多效性,采用留一法评估单个SNP对IVW分析结果的影响。结果本研究共筛选出16个与儿童期BMI高度相关的SNP,剔除rs2076308(回文SNP)、rs17817449(离群值)、rs61765651(离群值)后,最终纳入13个与儿童期BMI高度相关的SNP。IVW分析结果显示,儿童期BMI升高是GH的危险因素[OR=1.29,95%CI(1.09~1.53),P<0.01];且MR-Egger回归、加权中位数法、简单模式法、加权模式法分析的β值与IVW分析的β值方向一致。CochranQ检验结果显示,与儿童期BMI高度相关的SNP间不存在统计学异质性(P=0.47)。MR-Egger回归的截距项、MR-PRESSO检验、漏斗图分析结果均显示,与儿童期BMI高度相关的SNP不存在水平多效性。留一法分析结果显示,剔除单个SNP后,MR分析结果无明显改变。结论儿童期BMI升高是GH的危险因素。

Objective To analyze the potential causal relationship between childhood body mass index(BMI)and gestational hypertension(GH)through Mendelian randomization(MR)method.Methods The childhood BMI dataset(ID:ebi-a-GCST90002409)and GH dataset(ID:finn-b-O15_HYPTENSPREG)were obtained from the genome-wide association study(GWAS)database.The sample size of the childhood BMI dataset was 39620,including 8173382 single nucleotide polymorphisms(SNPs).The sample size of GH dataset was 123579,including 16379784 SNPs.In this study,inverse variance weighting(IVW),MR-Egger regression,weighted median method,simple model method and weighted model method were used to analyze the potential causal relationship between childhood BMI and GH.CochranQtest was used to evaluate the statistical heterogeneity among SNPs.The intercept term of MR-Egger regression,MR-PRESSO test and funnel plot were used to analyze the level pleiotropy of SNPs.The effect of single SNPs on IVW analysis results was evaluated by leave-one-out method.Results A total of 16 SNPs highly correlated with childhood BMI were screened in this study.After excluding rs2076308(palindrome SNP),rs17817449(outlier)and rs61765651(outlier),13 SNPs highly correlated with childhood BMI were finally included.IVW analysis results showed that increased childhood BMI was a risk factor for GH[OR=1.29,95%CI(1.09-1.53),P<0.01].Theβvalues of MR-Egger regression,weighted median method,simple mode method and weighted mode method were consistent with theβvalue of IVW.The results of CochranQtest showed that there was no statistical heterogeneity among SNPs highly correlated with childhood BMI(P=0.47).The results of intercept term of MR-Egger regression,MR-PRESSO test and funnel plot analysis showed that there was no level pleiotropy in SNPs highly correlated with childhood BMI.The results of leave-one-out analysis showed that there was no significant change in MR analysis results after removing a single SNP.Conclusion Increased childhood BMI is a risk factor for GH.