A Bidirectional Mendelian Randomization Study of Gut Microbiota and Cerebral Small Vessel Disease

The causal nature of gut microbiota and cerebral small vessel disease (CSVD) is still obscure regardless of evidence supporting their observational correlations.

The primary objective of this research is to investigate the potentially pathogenic or protective causal impacts of specific gut microbiota on various neuroimaging subtypes of CSVD.

We obtained the latest summary-level genome-wide databases for gut microbiota and 9 CSVD traits. The univariable and multivariable Mendelian randomization (MR) studies were conducted to examine the possible causal link between exposure and outcome. Meanwhile, we conducted sensitivity analyses sequentially, containing the heterogeneity, pleiotropy, and leave-one-out analysis. Additionally, to clarify the potential bidirectional causality, the causality from CSVD traits to the identified gut microbiota was implemented through reverse MR analysis.

The univariable MR analysis identified 22 genetically predicted bacterial abundances that were correlated with CSVD traits. Although conditioning on macronutrient dietary compositions, 2 suggestive relationships were retained using the multivariable MR analysis. Specifically, the class Negativicutes and order Selenomonadales exhibited a negative causal association with strictly lobar cerebral microbleeds, one neuroimaging trait of CSVD. There is insufficient evidence indicating the presence of heterogeneity and horizontal pleiotropy. Furthermore, the identified causal relationship was not driven by any single nucleotide polymorphism. The results of the reverse MR analysis did not reveal any statistically significant causality from CSVD traits to the identified gut microbiota.

Our study indicated several suggestive causal effects from gut microbiota to different neuroimaging subtypes of CSVD. These findings provided a latent understanding of the pathogenesis of CSVD from the perspective of the gut-brain axis.