Kt/V(urea) ratio is commonly used to assess the delivered dose of dialysis in maintenance hemodialysis (MHD) patients. This parameter only reflects the efficacy of dialytic treatments in removing small toxins, but not middle and protein-bound toxins. Erythrocyte glutathione transferase (e-GST), an enzyme devoted to cell depuration against a lot of large and small toxins, is overexpressed in uremic patients. Aim of the present study is to verify whether e-GST may represent a novel biomarker to assess the adequacy of different dialytic techniques complementary to Kt/V(urea) parameter. Furthermore, it will be investigated whether e-GST could reflect the 'average' adequacy of multiple dialytic sessions and not of a single one treatment as it occurs for Kt/V(urea). One hundred and three MHD patients and 82 healthy subjects were tested. Fourty four patients were treated with standard bicarbonate hemodialysis (HD) and 59 patients were on online hemodiafiltration (HDF). In all MHD patients e-GST activity was 60% higher than in healthy controls. In HDF, e-GST activity was lower than in HD subgroup (8.2±0.4 versus 10.0±0.4 U/g(Hb), respectively). Single-pool Kt/V(urea) and total weekly Kt/V(urea) were higher in HDF than in HD, but no correlation was found between e-GST activity and Kt/V(urea) data. e-GST, whose level is stable during the erythrocyte life-span, provides information on the long-term depurative efficacy of dialysis treatments.