维生素D含量水平与除肺外其他器官结核病的因果关联——一项孟德尔随机化研究
Causal Relationship between Vitamin D Content Levels and Tuberculosis in Organs Other Than the Lungs—A Mendelian Randomization Study

目的：探讨维生素D含量水平与除肺外其他器官结核病的关联。方法：使用“TwoSampleMR”包进行2SMR分析，首先，以r² = 0.001，kb = 10,000，选择欧洲人群，进行连锁不平衡处理，确保工具变量符合独立性。以逆方差加权法IVW为评估的最主要方法，采用了逆方差加权法(IVW)作为主要评估手段，并辅以加权中位数法(WM)、加权模式法(weighted method)以及MR-Egger回归等多种方法，以增强因果推断的稳健性和可靠性。我们设定P值小于0.05作为判断2SMR研究结果是否具有统计显著性的标准。为了检测暴露因素和结局变量之间是否存在水平基因多效性，应用了MR-PRESSO检验，还采用了Leave-one-out法进行敏感度分析，以识别并评估单个SNP对MR估计值是否有显著影响。使用Cochran Q检验评价其异质性。结果：结果显示，维生素D含量水平与除肺外其他器官结核病的发病呈现负相关，存在显著因果关系，MR-PRESSO检验表明本次研究所纳入的SNP不存在水平多效性，Cochran Q检验结果显示，均不存在显著异质性。以单核苷酸多态(SNPs)作为工具变量，根据SNP的森林图可以得知，维生素D含量水平的增加可以减少除肺外其他器官结核病的发生。进一步分析散点图发现，不同算法的线总体是斜向下的，这就意味着随着维生素D含量水平表达的升高，除肺外其他器官结核病呈下降趋势。漏斗图显示两侧散点大致均匀，异质性较小。Leave-one-out法敏感性分析结果未发现对结果显著影响的SNP。结论：维生素D含量水平与除肺外其他器官结核病的发病呈现负相关，存在显著因果关联。
Objective: To explore the correlation between vitamin D content level and tuberculosis in other organs except lung. Methods: The 2SMR analysis was performed with “TwoSampleMR” package. First, r² = 0.001, kb = 10,000, select a European population and perform linkage disequilibrium treatment to ensure that the instrumental variables are independent. The main evaluation method is the inverse variance weighted method (IVW), which is used as the main evaluation tool, supplemented by various methods such as weighted median method (WM), weighted method, and MR Egger regression to enhance the robustness and reliability of causal inference. We set a P-value less than 0.05 as the criterion for determining whether the 2SMR study results have statistical significance. To detect whether there is horizontal gene pleiotropy between exposure factors and outcome variables, MR-PRESSO test was applied, and the Leave-one-out method was used for sensitivity analysis to identify and evaluate whether a single SNP has a significant impact on MR estimation values. Use Cochran Q test to evaluate its heterogeneity. Result: The results showed that the level of vitamin D content was negatively correlated with the incidence of tuberculosis in other organs except the lung, and there was a significant causal relationship. MR-PRESSO test showed that the SNPs included in this study did not have level pleiotropy, and Cochran Q test showed that there was no significant heterogeneity. With single nucleotide polymorphisms (SNPs) as the tool variable, according to the forest map of SNPs, the increase of vitamin D content can reduce the occurrence of tuberculosis