Convective treatments with on-line production of replacement fluid: a clinical experience lasting 6 years

BACKGROUND: The introduction of techniques with on-line (OL) production of replacement fluid by filtration of dialysis fluid raises concerns about exposure of dialysis patients to pyrogenic substances. This work was undertaken to evaluate safety and feasibility of OL preparation of replacement fluid for haemodiafiltration (HDF). METHODS: OL HDF was carried out with commercially available monitors without any adjustment in the operational organization of our Centre. Bicarbonate dialysis fluid was filtered twice before being reinjected into the patients. The effects of acute load of OL fluid were assessed by very sensitive in vitro and in vivo tests; the chronic effects were assessed by monitoring the patients for the appearance of any untoward clinical manifestations and by measuring their cytokine response. RESULTS: In a pilot study the membrane filter culture technique of replacement fluid yielded no bacteria or mycetes growth, while LAL test was < 0.01 EU/ml. The normal human monocyte production of TNF alpha, IL-1 beta and IL-1Ra was not significantly different when cells were incubated with OL or commercial replacement fluid. The patients' body temperature profile (continuous recording during treatments and the following 24 h) overlapped with that of the control procedure. Over 6 years we performed 4284 OL treatments (total amount reinjected fluid 102,900 litres) on 13 patients treated for 26 +/- 9 months. In none of these treatments did we observe pyrogenic reactions. In comparison with the previous period on standard bicarbonate haemodialysis, OL HDF afforded significantly better cardiovascular tolerance to fluid removal and higher Kt/V values. The nutritional status did not deteriorate, while the acute-phase reactants and serum beta 2M levels did not increase. Moreover, no translucent cysts or destructive arthropathy were observed on bone X-rays. The patients' plasma cytokine levels and monocytes cytokines production, measured either before or after a single OL HDF, were comparable with the values obtained in controls treated with standard HDF. CONCLUSIONS: We conclude that OL-prepared replacement fluid is as safe as that of the commercial bags with regard to sterility and non-pyrogenicity. OL HDF can be readily implemented in any dialysis centre without bringing any further burden on the staff.