Résumé
A comprehensive review of known uremic retention molecules goes back to more than 10 years ago and did not consider metabolomic analyses. The present analysis searches for as of yet unclassified solutes retained in chronic kidney disease (CKD) by analyzing metabolites associated with relevant outcomes of CKD. This untargeted metabolomics-based approach is compared with a conventional targeted literature search. For the selected molecules, the literature was screened for arguments regarding toxic (harmful), beneficial, or neutral effects in experimental or clinical studies. Findings were independently crosschecked. In total, 103 molecules were selected. No literature on any effect was found for 55 substances, 3 molecules had no significant effect, and 13 others showed beneficial effects. For the remaining 32 compounds, we found at least one report of a toxic effect. Whereas 62.5% of the compounds with at least one study on a toxic effect was retrieved via the bottom-up approach, 69.2% of the substances originating from metabolomics-based approaches showed a beneficial effect. Our results suggest that untargeted metabolomics offer a more balanced view of uremic retention than the targeted approaches, with higher chances of revealing the beneficial potential of some of the metabolites.