Causal relationships between the gut microbiome, blood lipids, and heart failure: a Mendelian randomization analysis

Abstract Aims Studies have linked gut microbiome and heart failure (HF). However, their causal relationships and potential mediating factors have not been well defined. To investigate the causal relationships between the gut microbiome and HF and the mediating effect of potential blood lipids by using genetics. Methods and results We performed a bidirectional and mediation Mendelian randomization (MR) study using summary statistics from the genome-wide association studies of gut microbial taxa (Dutch Microbiome Project, n = 7738), blood lipids (UK Biobank, n = 115 078), and a meta-analysis of HF (115 150 cases and 1550 331 controls). We applied the inverse–variance weighted estimation method as the primary method, with several other estimators as complementary methods. The multivariable MR approach based on Bayesian model averaging (MR-BMA) was used to prioritize the most likely causal lipids. Six microbial taxa are suggestively associated with HF causally. The most significant taxon was the species Bacteroides dorei [odds ratio = 1.059, 95% confidence interval (CI) = 1.022–1.097, P-value = 0.0017]. The MR-BMA analysis showed that apolipoprotein B (ApoB) was the most likely causal lipid for HF (the marginal inclusion probability = 0.717, P-value = 0.005). The mediation MR analysis showed that ApoB mediated the causal effects of species B. dorei on HF (proportion mediated = 10.1%, 95% CI = 0.2–21.6%, P-value = 0.031). Conclusion The study suggested a causal relationship between specific gut microbial taxa and HF and that ApoB might mediate this relationship as the primary lipid determinant of HF.