BACKGROUND

Observational studies have shown gut microbiomes were associated with cardiovascular diseases (CVDs), but their roles remain controversial, and these associations have not yet been established causally.

METHODS

Two-sample Mendelian randomization (MR) was used to investigate whether gut microbiome had a causal effect on the risk of CVDs. To obtain comprehensive results, we performed two sets of MR analyses, one with single nucleotide polymorphisms (SNPs) that smaller than the genome-wide statistical significance threshold (5 × 10) as instrumental variables, and the other with SNPs that lower than the locus-wide significance level (1 × 10). Summary-level statistics for CVDs, including coronary artery disease (CAD), myocardial infarction, heart failure, atrial fibrillation, stroke and its subtypes were collected. The ME estimation was performed using the inverse-variance weighted and Wald ratio methods. Sensitivity analysis was performed using the weighted median, MR-Egger, leave-one-out analysis, MR pleiotropy residual sum and outlier and MR Steiger.

RESULTS

Based on the locus-wide significance level, genetically predicted genus was positively associated with the risk of CAD (odds ratio (OR) = 1.06, 95% confidence interval (CI), 1.03 - 1.10, = 1.67 × 10), family was negatively correlated with stroke risk (OR = 0.83,95% CI, 0.75-0.93, = 7.76 × 10) and ischemic stroke risk (OR = 0.823,95% CI, 0.74-0.92, = 4.15 × 10). There was no causal relationship between other genetically predicted gut microbiome components and CVDs risk. Based on the genome-wide statistical significance threshold, the results showed that the gut microbiome had no causal relationship with CVDs risk.

CONCLUSION

Our findings reveal that there are beneficial or adverse causal effects of gut microbiome components on CVDs risk and provide novel insights into strategies for the prevention and management of CVDs through the gut microbiome.