目的：探讨静脉麻醉暴露的遗传易感性是否影响术后认知功能障碍（postoperative cognitive dysfunc
tion，POCD）风险，以及脑脊液N-乙酰精氨酸是否在此关联中起中介作用。方法：采用主要来自欧洲血统队
列的 GWAS（Genome-Wide Association Study）汇总统计量进行了两样本、两步孟德尔随机化（Two-Step
Mendelian Randomization，TSMR）分析。首先检测脑脊液代谢物对静脉麻醉药物的遗传易感性，其次评估
这些代谢物对认知结果的影响，作为 POCD 风险的替代指标。主要估计值通过逆方差加权Mr获得，并辅以
加权中位数、Mr-Egger和基于众数的方法。稳健性通过Cochran’s Q、Mr-Egger截距、Mr-PRESSO（全局/异
常值）和Steiger方向性进行评估。中介效应通过系数乘积测量，不确定性通过delta方法/自助法估计，并在
代谢组中进行多重检验校正。探索性生物信息学将差异表达基因与蛋白质-蛋白质相互作用/基因本体论/
京都基因与基因组百科全书富集分析相结合。结果：静脉麻醉药物的遗传易感性与脑脊液N-乙酰精氨酸
水平的变化相关，后者与 POCD 风险增加相关。这种通过 N-乙酰精氨酸介导的效应占总效应的显著部
分。共定位和多变量孟德尔随机化证实了代谢物特异性信号，富集结果表明一氧化氮信号传导及与Notch
通路的相互作用参与其中。结论：TSMR显示脑脊液N-乙酰精氨酸部分介导了静脉麻醉剂易感性与POCD
风险之间的联系。
To investigate whether genetic susceptibility to intravenous anesthesia exposure affects
the risk of postoperative cognitive dysfunction (POCD) and whether cerebrospinal fluid N-acetylarginine plays a
mediating role in this association. Methods: Two-sample, two-step Mendelian randomization (TSMR) analysis
was conducted using genome-wide association study (GWAS) summary statistics primarily from cohorts of
European ancestry. First, we examined the genetic susceptibility of cerebrospinal fluid metabolites to intravenous
anesthetic agents. Then, we evaluated the impact of these metabolites on cognitive outcomes as a surrogate
marker for POCD risk. The primary estimates were obtained through inverse variance-weighted MR,
supplemented by weighted median, MR-Egger, and mode-based methods. Robustness was assessed using
Cochran’s Q, MR-Egger intercept, MR-PRESSO (global/outlier), and Steiger directionality tests. The mediating
effect was measured by the product of coefficients, with uncertainty estimated via the delta method/
bootstrapping and multiple testing correction applied within the metabolome. Exploratory bioinformatics
integrated differentially expressed genes with protein-protein interaction (PPI)/gene ontology (GO)/Kyoto
Encyclopedia of Genes and Genomes (KEGG) enrichment analyses. Results: Genetic susceptibility to
intravenous anesthetic agents was associated with changes in cerebrospinal fluid N-acetylarginine levels, which,
in turn, were linked to an increased risk of POCD. This N-acetylarginine-mediated effect accounted for a
significant portion of the total effect. Colocalization and multivariable Mendelian randomization (MVMR)
confirmed metabolite-specific signals, and enrichment results indicated involvement of nitric oxide signaling and
interactions with the Notch pathway. Conclusion: TSMR revealed that cerebrospinal fluid N-acetylarginine
partially mediates the association between susceptibility to intravenous anesthetic agents and POCD risk.
PDF(4486 KB)
