BACKGROUND

The bidirectional causal relationship between ischemic stroke (IS) and inflammatory bowel disease (IBD) remains unclear, prompting us to propose a bidirectional Mendelian randomization (MR) study to investigate this relationship further.

METHODS

We obtained IS data from the MEGASTROKE consortium and IBD data, including its subtypes ulcerative colitis (UC) and Crohn's disease (CD), from the International Inflammatory Bowel Disease Genetics Consortium. In this study, we utilized IBD and its subtypes as exposure variables and IS as the outcome variable, and vice versa, to explore the bidirectional relationship between them. We used the IBD genetic data from the FinnGen database as replication data to further explore the causality. In this study, we employed the inverse variance weighting method as our primary approach. For sensitivity analyses, we utilized additional methods including MR-Egger regression, weighted median estimation, MR pleiotropy residual sum and outlier (MR-PRESSO), and MR-Robust adjusted profile score. Furthermore, we conducted a random effects meta-analysis to combine the causal relationships derived from both the International Inflammatory Bowel Disease Genetics Consortium and FinnGen datasets, aiming to ascertain more robust causal associations.

RESULTS

The initial phase of the bidirectional MR study revealed a causal relationship between IS and the risk of CD (odds ratio [OR] = 1.56, 95% confidence interval [CI]: 1.20-2.02, P = 0.0008) and UC (OR = 1.33, 95% CI: 1.05-1.69, P = 0.0179), but did not find a causal relationship between IBD as a whole and the risk of IS, nor between IBD subtypes and the risk of IS. During the replication phase, the FinnGen database did not reveal any significant correlation between IS and the risk of IBD, including its subtypes CD and UC. However, additional meta-analysis of the combined data from both databases indicated that IS is significantly associated with an increased risk of CD (OR inverse-variance weighted  = 1.38, 95% CI: 1.07-1.69, P < 0.05) and UC (OR = 1.27, 95% CI: 1.04-1.50, P < 0.05), but not with the overall risk of IBD (OR = 1.05, 95% CI: 0.87-1.16, P > 0.05). No significant effects were observed between IBD and IS risk, nor were there significant effects between IS and the risks of IBD, CD, or UC. To ensure the robustness of these findings, heterogeneity and pleiotropy tests were conducted.

CONCLUSIONS

IBD and its subtypes were not found to be causally associated with the risk of IS, whereas IS was found to be causally associated with the risk of CD and UC. This suggests that the risks of CD and UC should be closely monitored in patients with IS.