Abstract Background and Aims Patients with end-stage kidney disease (ESKD) have an extremely high incidence of cardiovascular (CV) diseases, partly driven by insufficient clearance of uremic toxins. ESKD patients have a characteristically adverse lipid profile, however data investigating the relationship between uremic toxins and lipid profile, potentially contributing to increased CV risk, is scarce. To determine if uremic retention solutes (URS) associate with an adverse lipid profile in ESKD, we studied a large, trinational cohort with a detailed lipid profile, as well as a comprehensive panel of uremic toxins. Method Total, high density lipoprotein (HDL), non-HDL, low density lipoprotein (LDL), and remnant cholesterol, as well as triglyceride, levels were associated with a panel of 15 uremic retention solutes in a combined cohort of 591 European, adult patients with advanced chronic kidney disease (CKD) from UZ Leuven, Belgium (n=150), Karolinska Hospital, Stockholm, Sweden (n=235) or University of Leipzig Medical Center, Leipzig, Germany (n=226). Total and HDL cholesterol, as well as triglycerides, were quantified at each study center, whereas non-HDL cholesterol, LDL cholesterol, and remnant cholesterol were calculated. In all subjects of this trinational study, a selected panel of solutes, including CMPF, TMAO, aromatic amino acids and corresponding end-products of endogenous and microbial metabolism, was centrally quantified in a single lab by liquid chromatography - tandem mass spectrometry. Univariate correlations were assessed using non-parametric Spearman’s rank correlation method. To identify independent associations between solutes and lipid profile, multivariate linear regression models were used with adjustment for age, sex, as well as markers of inflammation, protein energy wasting, renal function, diabetes and dialysis. Results In total, 189 patients in CKD stage 3-5 not on renal replacement therapy (RRT), as well as 402 subjects on RRT, were included. All URS except phenylalanine significantly differed between patients on RRT vs. not on RTT. In univariate analyses, URS negatively correlated with most lipid markers, including LDL and HDL cholesterol. In contrast, the amino acids tryptophan, phenylalanine, and tyrosine were positively correlated with a large majority of lipid markers. After combining URS concentrations based on molecule size, similar associations were observed for the respective groups, i.e. small water-soluble molecules, protein-bound molecules, and amino acids. After adjustment for age, sex, presence of diabetes, dialysis treatment, inflammation, protein energy wasting, and renal function, significant associations were lost for URS and total cholesterol or HDL cholesterol, excluding total cholesterol and phenylacetyl glutamine. However, high triglyceride levels were independently predicted by p-cresyl sulphate, tryptophan, indole-3 acetic acid, phenylalanine, TMAO, small water-soluble molecules combined, and protein-bound molecules combined. Non-HDL cholesterol was independently predicted by phenyl glucuronide, TMAO, phenylacetyl glutamine and small water-soluble molecules combined, while remnant cholesterol was independently associated with 10 out of the 15 URS, as well as small water-soluble molecules combined and amino acids combined. Furthermore, LDL cholesterol independently associated with tryptophan, TMAO, phenylacetyl glutamine and protein-bound molecules combined. Conclusion Significant inverse associations between lipid profile and small water-soluble or protein-bound uremic toxins in advanced CKD highlight the complexity of the uremic environment. Our data suggest that not all URS interactions with conventional CV risk markers may be pathogenic.