Effect of acetate-free biofiltration on the anaemia of haemodialysis patients: a prospective cross-over study

Background. The discussion about the pathogenesis of renal anaemia, whether it is primarily due to relative erythropoietin (Epo) deficiency or to urae mic inhibition of erythropoiesis, is still open, Although it has so far not been possible to identify or isolate a substance retained in uraemia with a suppressive action directed specifically against red-cell production, dia lysis therapy can improve the effect of both residual endogenous Epo and ex ogenous rHuEpo. To what extent the mode and/or the dose of dialysis influen ce Epo efficacy is as yet poorly understood. Methods. This study was performed as a single-centre trial. The protocol in cluded a run-in period of 4 months followed by a prospective cross-over stu dy including 6 months each of acetate-free biofiltration (AFB) with a high- flux. biocompatible membrane and standard bicarbonate dialysis (BD) with a low-flux cellulosic membrane in a random sequence. AFB is a haemodiafiltrat ion technique based on a continuous post-dilution infusion of a sterile iso tonic, bicarbonate solution. At the start of the run-in period (and for the entire length of the study), rHuEpo administration was withdrawn; patients whose haemoglobin (Hb) levels dropped at a level < 8.0 g/dl at one single monthly check, had to be withdrawn from the study. A blood sample was colle cted every month for the blood gas analysis, and for the determination of b lood urea nitrogen, serum creatinine, sodium, potassium, calcium, phosphoru s, Hb, erythrocyte, reticulocyte, leukocyte and thrombocyte cell counts, me an globular volume and haematocrit. An equilibrated single pool Kt/V-urea>1 .2 was mandatory in both treatment modalities. Serum iron, total iron-bindi ng capacity, and ferritin were checked every 3 months. Results. Twenty-three of 137 haemodialysis patients were considered eligibl e for the trial on the basis of the entry criteria. Of these, 15 volunteere d and only 10 completed the study. No significant differences in the haemat ological indices, in the biochemical parameters assessing body iron stores, or in i.v. iron dosage was observed when comparing AFB with BD treatments. The equilibrated single pool Kt/V-urea was always > 1.2 and in no case was a significant difference observed when comparing AFB with BD treatments. T reatment time was significantly different between the two treatments (262 /-2 min in BD and 249 +/-1 in AFB, P <0.0001). Neither pre- nor post-dialys is systolic and diastolic blood pressures, pre-dialysis serum bicarbonate a nd pH, pre-dialysis serum sodium, potassium, calcium, or phosphorus were si gnificantly different when comparing the two treatment modalities. All 10 p atients completed the 1-year follow-up without any major side-effects. Conclusions. Our study did not show any improvement of anaemia when treatin g a highly selected patient group, in the absence of any Epo therapy, with AFB compared with standard BD. Even though these conclusions cannot be exte nded in toto to the entire dialysis population, in which there is a large p roportion of Epo-treated patients with Hb levels around 11 g/dl, we may nev ertheless conclude that when patients are well selected, adequately dialyse d, and not iron- and/or vitamin-depleted, the effect of a haemodiafiltratio n technique with a high-flux biocompatible membrane is less than might be e xpected from the results of uncontrolled studies.