内科

目的采用两样本孟德尔随机化（MR）方法，探讨ω-3多不饱和脂肪酸（PUFAs）与银屑病之间的因果关系。方法基于公开发布的全基因组关联研究（GWAS）汇总数据，以ω-3 PUFAs为暴露因素、银屑病为结局变量，筛选与ω-3 PUFAs强相关且满足独立性假设的的单核苷酸多态性（SNP）作为工具变量。主要采用逆方差加权（IVW）法进行因果效应估计，并辅以加权中位数估计（WME）法、MR-Egger回归法、简单模式法和加权模式法作为补充和验证。采用MR-Egger截距检验评估水平多效性，采用Cochran Q检验评估异质性，并通过留一法进行敏感性分析。结果最终筛选出37个SNP作为有效的工具变量进行MR分析。IVW法（核心分析方法）结果显示，ω-3 PUFAs与银屑病之间无统计学意义上的因果关系（OR=0.998 6，95%CI：0.996 7～1.000 5，P=0.155 0）；WME法、简单模式法及加权模式法的结果均与IVW法一致（均P>0.05）。MR-Egger回归分析显示，二者之间的因果关系具有边缘统计学意义（OR=0.996 1，95%CI：0.992 7～0.999 4，P=0.028 7），但该效应值极接近无效值1，结合MR-Egger截距检验结果提示无显著水平多效性（P>0.05），推测该边缘显著性并非真实因果效应。Cochran Q检验结果显示，各SNP间无统计学意义上的异质性（P>0.05）。留一法敏感性分析结果显示，逐一剔除任一SNP后，合并效应估计值均保持稳定。结论本研究未发现ω-3 PUFAs与银屑病之间存在遗传层面的因果关联。

Objective To investigate the causal association between ω-3 polyunsaturated fatty acids (PUFAs) and psoriasis using a two-sample Mendelian randomization (MR) approach. Methods Based on publicly released genome-wide association study (GWAS) summary data, with ω-3 PUFAs as the exposure factor and psoriasis as the outcome variable, single nucleotide polymorphisms (SNPs) strongly associated with ω-3 PUFAs and satisfying the independence assumption were selected as instrumental variables. The inverse-variance weighted (IVW) method was the primary approach for estimating causal effects, supplemented and validated by the weighted median estimator (WME) method, MR-Egger regression method, simple mode method, and weighted mode method. Horizontal pleiotropy was assessed using the MR-Egger intercept test, heterogeneity was evaluated using Cochran's Q test, and sensitivity analysis was performed using the leave-one-out method. Results Thirty-seven SNPs were ultimately selected as valid instrumental variables for MR analysis. The results from the IVW method (the core analytical method) indicated no statistically significant causal association between ω-3 PUFAs and psoriasis (OR=0.998 6, 95% CI: 0.996 7-1.000 5, P=0.155 0), which were consistent with results from the WME, simple mode, and weighted mode methods (all P>0.05). MR-Egger regression analysis showed a causal association with marginal statistical significance between the two (OR=0.996 1, 95% CI: 0.992 7-0.999 4, P=0.028 7), however, the effect value was extremely close to 1, the null value; combined with the MR-Egger intercept test result indicating no significant horizontal pleiotropy (P>0.05), this marginal significance was unlikely to represent a true causal effect. Cochran's Q test results showed no statistically significant heterogeneity among the SNPs (P>0.05). Leave-one-out sensitivity analysis demonstrated that the combined effect estimates remained stable after sequentially removing each SNP. Conclusion No genetic-level causal association between ω-3 PUFAs and psoriasis was found in this study.