牙周炎和糖尿病视网膜病变的因果关系:一项双向两样本孟德尔随机化分析

The causal association between periodontitis and diabetic retinopathy:A bidirectional two-sample mendelian randomization analysis

目的利用双向孟德尔随机化(MR)研究方法探索牙周炎与糖尿病视网膜病变(DR)之间的因果关联。方法从来源于欧洲人群的全基因组关联研究(GWAS)数据库GLIDE和FinnGen中分别筛选与牙周炎和DR密切相关且相互独立的单核苷酸多态性(SNP)位点作为工具变量。采用逆方差加权(IVW)、加权中位数(WM)和MR-Egger回归分析方法进行两样本MR分析,以比值比(OR)评价牙周炎和DR疾病风险之间的因果关联。采用Cochran′s Q检验分析工具变量的异质性,孟德尔随机多态性残差和离群值(MR-PRESSO)法检测SNP异常值,MR-Egger回归截距项分析基因的水平多效性,留一法敏感性检测分析单个SNP对结果的影响,从而验证结果的稳定性和可靠性。结果IVW分析结果不支持牙周炎与DR及其不同分期背景型糖尿病视网膜病变(BDR)和增殖型糖尿病视网膜病变(PDR)之间存在因果关联[OR_(DR)=1.006,95%CI_(DR)(0.945,1.071),P_(DR)=0.852;OR_(BDR)=0.978,95%CI_(BDR)(0.862,1.109),P_(BDR)=0.730;OR_(PDR)=0.996,95%CI_(PDR)(0.948,1.046),P_(PDR)=0.869]。WM和MR-Egger分析所得结果与IVW法一致。采用Cochran′s Q和MR-PRESSO检验表明纳入的工具变量SNP之间无显著异质性(P>0.05);MR-Egger回归截距项显示纳入的SNP不存在多效性(P>0.05);留一法分析未发现显著影响结果的SNP。结论基于欧洲人群的MR分析结果不支持牙周炎与DR之间存在双向因果关联。

Objective To evaluate the causal relationship between periodontitis and diabetic retinopathy(DR)with Mendelian randomization(MR)analysis.Methods Mutually independent SNP associated with periodontitis and DR were screened out separately from two Genome-wide association study(GWAS)databases(GLIDE and FinnGen)and selected as instrumental variables.A bidirectional two-sample mendelian randomization analysis was conducted with methods of inverse variance weighted(IVW),MR-Egger regression and weighted median(WM).The causal association between periodontitis and DR was estimated with odds ratios(OR).Cochran′s Q test was employed to assess the heterogeneity between individual single nucleotide polymorphisms(SNP).Additionally,the Mendelian randomization pleiotropy residual sum and outlier(MR-PRESSO)test was used to detect SNP outliers.The MR-Egger intercept test was used to assess the horizontal pleiotropy of SNP,and a leave-one-out sensitivity analysis was performed to examine the influence of individual SNP on the overall results.Results IVW did not support a causal relationship between periodontitis and DR and its different stages background diabetic retinopathy(PDR)and proliferative diabetic retinopathy(PDR)[OR_(DR)=1.006,95%CI_(DR)(0.945,1.071),P_(DR)=0.852;OR_(BDR)=0.978,95%CI_(BDR)(0.862,1.109),P_(BDR)=0.730;OR_(PDR)=0.996,95%CI_(PDR)(0.948,1.046),P_(PDR)=0.869].The results of Cochran′s Q test,MR-PRESSO,and the MR-Egger intercept test revealed no heterogeneity or horizontal pleiotropy among instrumental variables(P>0.05).Furthermore,the leave-one-out sensitivity analysis confirmed that individual SNP had no significant influence on the overall results.Conclusion This MR study did not support a bidirectional causal association between periodontitis and DR in European populations.