Hemodialysis is the major technique of renal suppleance and membrane improvements allow an efficient depuration. During the past fifty years, improving the architecture of the membranes increased constantly the clearance of uremic toxins. Online-hemodiafiltration allowed a potential interest for decreasing morbi-mortality, but this technique is not available in all hemodialysis centers yet. Theranova membrane, by its innovating architecture, could be a valuable option for the enhancement of toxin removal in patients who are not eligible for online-hemodiafiltration techniques. Interestingly, many works pointed adsorptive capacities of dialysis membranes and these observations suggest that adsorption cannot be considered as an epiphenomenon. However, there is a lack of knowledge about the extent of its benefits in hemodialysis. The efficiency of a session of hemodialysis is routinely based on the clearance of only few generic toxins (urea, ß2-microglobulin, myoglobin). The eKT/V formula permitted to link a good depuration of urea with a better survival. EuTox described in a recent review, a large panel of middle molecules considered as uremic toxins. As there are many involved toxins, modern evaluation of hemodialysis efficiency can't be still based on the only description of ß2-microglobulin and urea removals. Inflammation status is routinely measured with blood levels of CRP. However, its accumulation is not leading to a toxicity. Many others inflammatory markers have been linked with morbidity and atherosclerosis, the main cause of mortality in ESRD patients. Obviously, removal of inflammatory markers by different dialyzers needs to be assess concomitantly but single conventional dosages are time-consuming.

A minimal albumin loss is considered as another important factor for a safe hemodialysis. Due to its multiple biological functions, albumin may be one of the most important protein for the maintenance of body homeostasis. Recent works were interested in the potential toxicity of some isoforms of the albumin and especially in chronic kidney disease. Modified forms of albumin have been linked with toxic properties and the removal of modified forms could be beneficial. Furthermore, albumin oxidation can lead to an under-estimation of albumin levels with the routinely used assay. A modern evaluation of the capacities of depuration of dialyzers clearly needs to integrate the evaluation of the clearance and generation of modified forms of albumin and proteins.

Proteomic techniques allow an interesting non-selective integrative approach. This exhaustive view would be particularly adequate for building a snapshot of blood uremic toxins, permeative and adsorptive capacities of each membrane. Due to a high sensitiveness, proteomic techniques can improve dramatically the characterization of the depuration capacities of different hemodialysis membranes considering a large range of uremic toxins especially with middle molecular weight, a large panel of inflammatory markers and modified forms of albumin. Furthermore, proteomic approach can help to build a quasi-exhaustive description of all the proteins removed by a membrane.

This project aims to compare the removal capacities of each membrane on a large range of proteins and forms of albumin (native and modified) with a highly-sensitive LC-MS/MS proteomic approach. We will also build the complete proteome of depuration of each tested membrane.

Inclusion Criteria:

• Age ≥ 18 years old
• Patient on hemodialysis for at least 1 year
• Patient with a functional vascular access permitting at least 300 mL/min of blood flow.

Exclusion Criteria:

• Running infection or other active disease (scheduled surgery, cancer, hospitalisation),
• Catheter as vascular access,
• Recirculation of vascular access over 10%.