Protein-bound uremic retention solutes constitute a group whose common characteristic is their difficult removal by dialysis. In 2003, the EUTox group described 25 protein-bound solutes. They comprised six advanced glycation end products (AGE), four phenols (including p-cresol), six indoles (including indoxylsulfate), two hippurates, three polyamines, and two peptides, homocysteine and 3-carboxy-4-methyl-5-propyl-2-furanpropionic acid (CMPF). As then, three new compounds have been added to the list: phenylacetic acid, dinucleoside polyphosphates, and IL-18. During the last years, protein-bound compounds have been identified as some of the main toxins involved in vascular lesions of chronic kidney disease. The removal of these solutes by conventional hemodialysis (HD) is low because only the free fraction of the solute is available for diffusion. The increase in the convective part with hemodiafiltration improves the performance of depuration but convection only applies to the free fraction and its benefit is limited. One possibility to improve the removal of a protein-bound solute would be to stimulate its dissociation from the binding protein. This could be obtained in experiments by setting the dialysate flow rate and the dialyzer mass transfer area coefficient (KoA) at much higher levels than the plasma flow rate, or by adding to the dialysate a sorbent such as activated charcoal or albumin. In the future, specific adsorbents may be developed. Today, the only possibility is to use approaches such as daily HD and long HD which could allow better equilibration between extravascular and vascular compartments and consequently result in greater removal of protein-bound compounds.