Studies have shown that uremia, renal failure and heart failure (HF) are closely related. However, whether this association reflects a causal effect is still unclear. The aim of this study was to evaluate the causal effect of uremic metabolites or toxins on HF. Mendelian randomization (MR) analysis was conducted to evaluate the causal effect of 11 uremia-related metabolites on HF risk using single-nucleotide polymorphisms (SNPs) from a genome-wide association study. A protein–protein interaction network was constructed to study the function of SNPs corresponding to HF-related factors. Univariate and multivariate MR analyses demonstrated that lipoprotein A and apolipoprotein B were positively correlated with HF. The SNPs corresponding to these key factors were related mainly to MAP kinase activity and lipid metabolic processes. Overall, we identified 2 uremia-related exposure factors (lipoprotein A and apolipoprotein B) closely related to HF, laying a theoretical foundation for the treatment of HF with renal failure or uremia.