We read with great interest the observation of eminent colleague Chen et al showing that patients with IBD had a higher incidence of dementia, particularly Alzheimer’s disease (AD, HR=6.19).1 Due to various confounding factors (age of onset, treatment, smoking, body mass index, etc), it is usually hard to evaluate the causal link between IBD and AD using observational studies. Two-sample Mendelian randomisation (MR) treated genetic variants as instrumental variables (IVs) for exposures, effectively minimising the impact of confounding factors. Using two-sample MR analysis, we adopt the publicly available summary statistics from large-scale genome-wide association studies (GWAS) to investigate the causal effect of IBD (Ncases=12 882; Ncontrols=21 770),2 including UC and Crohn’s disease (CD), on AD (Ncases=21 982, Ncontrols=41 944) of European descent.3 Briefly, single-nucleotide polymorphisms (SNPs) with p values less than 5E–08 for IBD were selected as IVs and were clumped based on the linkage disequilibrium (LD) structure (R2 <0.0001 within 10 000 kb, the 1000 Genomes Project European panel). In …

中文翻译：

---

致编辑的信：基因决定的 IBD 与阿尔茨海默病风险降低有关：孟德尔随机化研究

我们非常感兴趣地阅读了著名同事 Chen 等人的观察结果，该观察表明 IBD 患者的痴呆症发病率较高，尤其是阿尔茨海默病 (AD, HR=6.19)。1 由于各种混杂因素（发病年龄、治疗、吸烟，体重指数等），通常很难使用观察性研究来评估 IBD 和 AD 之间的因果关系。两样本孟德尔随机化 (MR) 将遗传变异视为暴露的工具变量 (IV)，有效地减少了混杂因素的影响。使用两样本 MR 分析，我们采用来自大规模全基因组关联研究 (GWAS) 的公开汇总统计数据来研究 IBD 的因果效应（Ncases=12 882；Ncontrols=21 770），2 包括 UC 和克罗恩病疾病 (CD)，在 AD (Ncases=21 982, Ncontrols=41 944) 的欧洲血统。3 简而言之，IBD 的 p 值小于 5E-08 的单核苷酸多态性 (SNP) 被选为 IV，并根据连锁不平衡 (LD) 结构 (R2 <0.0001 在10 000 kb，千人基因组计划欧洲小组）。在 …