The effect of the adequacy of dialysis on the response to recombinant human erythropoietin (rHuEpo) therapy is still incompletely understood because of many confounding factors such as iron deficiency, biocompatibility of dialysis membranes, and dialysis modality that can interfere. We investigated the relationship between Kt/V and the weekly dose of rHuEpo in 68 stable haemodialysis (HD) patients (age 65+/-15 years) treated with bicarbonate HD and unsubstituted cellulose membranes for 6-343 months (median 67 months). Inclusion criteria were HD for at least 6 months, subcutaneous rHuEpo for at least 4 months, transferrin saturation (TSAT) > or = 20%, serum ferritin > or = 100 ng/ml, and haematocrit (Hct) level targeted to 35% for at least 3 months. Exclusion criteria included HBsAg and HIV positivity, need for blood transfusions or evidence of blood loss in the 3 months before the study, and acute or chronic infections. Hct and haemoglobin (Hb) levels were evaluated weekly for 4 weeks; TSAT, serum ferritin, Kt/V, PCRn, serum albumin (sAlb), and weekly dose of rHuEpo were evaluated at the end of observation. No change in dialysis or therapy prescription was made during the study. The results for the whole group of patients were: Hct 35 +/- 1.2%, Hb 12.1 +/- 0.6 g/dl, TSAT 29 +/- 10%, serum ferritin 204 +/- 98 ng/ml, sAlb 4.1 +/- 0.3 g/dl, Kt/V 1.33 +/-0.19, PCRn 1.11+/- 0.28 g/kg/day, weekly dose of rHuEpo 123 +/- 76 U/kg. Hct did not correlate with Kt/V, whereas rHuEpo dose and Kt/V were inversely correlated (r = -0.49; P < 0.0001). Multiple regression analysis with rHuEpo as dependent variable confirmed Kt/V as the only significant variable (P < 0.002). Division of the patients into two groups according to Kt/V (group A, Kt/V < or = 1.2; group B, Kt/V > or = 1.4), showed no differences in Hct levels between the two groups, while weekly rHuEpo dose was significantly lower in group B than in group A (group B, 86 +/- 33 U/kg; group A, 183 +/- 95 U/kg, P < 0.0001). In iron-replete HD patients treated with rHuEpo in the maintenance phase, Kt/V exerts a significant sparing effect on rHuEpo requirement independent of the use of biocompatible synthetic membranes. By optimizing rHuEpo responsiveness, an adequate dialysis treatment can contribute to the reduction of the costs of rHuEpo therapy.