胃食管反流病增加慢性阻塞性肺疾病的患病风险：一项两样本双向孟德尔随机化研究

Gastroesophageal reflux disease causes an increased risk of chronic obstructive pulmonary disease: a two-sample bidirectional Mendelian randomization study

摘要:

目的 采用两样本双向孟德尔随机化（Mendelian randomization，MR）方法研究胃食管反流病（gastroesophageal reflux disease，GERD）与慢性阻塞性肺疾病（chronic obstructive pulmonary disease，COPD）的因果关系。方法 基于大规模全基因组关联研究（genome-wide association studies，GWAS）汇总数据，选择相互独立且与胃食管反流病及COPD高度相关的单核苷酸多态性（single nucleotide polymorphism，SNP）作为工具变量。采用逆方差加权法固定效应模型（inverse variance weighted-fixed effects，IVW-FE）为主要分析方法，简单中位数法、加权中位数法、MR Egger回归法进行结果验证和稳定性检验；采用F统计量、Cochran Q检验、MR Egger截距测试、孟德尔随机多态性残差和离群值（mendelian randomization pleiotropy residual sum and outlier，MR-PRESSO）检验、“留一法”进行敏感性分析。以比值比（odds ratio，OR）及95%置信区间（confidence interval，CI）作为效应指标对胃食管反流病及COPD的双向因果关系进行探讨。结果 IVW-FE分析表明胃食管反流病可增加COPD的患病风险（OR=1.757，95%CI=1.425~2.166，P=0.000），且在简单中位数法、加权中位数法中被证实。然而MR Egger回归法未显示出二者之间的因果关联。在反向MR分析中，IVW-FE分析显示没有证据表明COPD增加胃食管反流病的发病风险（OR=0.999，95%CI=0.962~1.037，P=0.962），同样简单中位数法、加权中位数法、MR Egger回归法未发现二者存在因果关联。工具变量的F值均大于10，不存在弱工具变量；Cochran Q检验、MR Egger截距测试、MR-PRESSO检验表明工具变量间不存在异质性或水平多效性；“留一法”分析显示没有单个SNP对整体结果有较大影响。结论 胃食管反流病与COPD的风险增加有显著的因果关系，但是没有发现COPD导致胃食管反流病风险增加的证据。

Abstract:

Objective To investigate the causal relationship between gastroesophageal reflux disease（GERD） and chronic obstructive pulmonary disease（COPD） using two-sample bidirectional Mendelian randomization（MR）.Methods Based on the pooled data from large-scale genome-wide association studies（GWAS），single nucleotide polymorphisms（SNPs） which were independent of each other and highly correlated with GERD and COPD were selected as instrumental variables. The inverse-variance weighted fixed-effects（IVW-FE） model was used as the main analysis method，while simple median，weighted median，and MR-Egger regression methods were used to validate the results and test the stability. The F-statistic，Cochran’s Q test，MR-Egger intercept test，Mendelian randomization pleiotropy residual sum and outlier（MR-PRESSO） test，and leave-one-out method were used for sensitivity analysis. The odds ratio（OR） and 95% confidence interval（CI） were used as effect sizes for the bidirectional causal relationship between GERD and COPD.Results The IVW-FE analysis showed that GERD increased the risk of COPD（OR=1.757，95%CI=1.425-2.166，P=0.000），which was confirmed by the simple median and weighted median methods. However，no causal relationship between the two factors was shown by the MR-Egger method. In the inverse MR analysis，there was no evidence of increased risk of GERD caused by COPD using IVW-FE（OR=0.999，95%CI=0.962-1.037，P=0.962）. Similarly，no association between the two factors was found using the simple median，weighted median，and MR-Egger regression methods. The F-statistic of the instrumental variable was greater than 10，which showed that there was no weak instrumental variable. The Cochran’s Q test，MR-Egger intercept test，and MR-PRESSO test did not show heterogeneity or horizontal pleiotropy between instrumental variables. The leave-one-out analyses indicated that no single SNP had a significant effect on the overall outcome.Conclusion GERD is significantly and causally associated with an increased risk of COPD. However，there is no evidence suggesting that COPD causes an increased risk of GERD.