Predilution haemofiltration -- the Second Sardinian Multicentre Study: comparisons between haemofiltration and haemodialysis during identical Kt/V and session times in a long-term cross-over study

Background. The potential superiority of various renal replacement treatment modalities consisting largely of convective mass transfer as opposed to primarily diffusive mass transfer, is still a matter of debate. The objective of the present study was to evaluate acute and long-term clinical effects of varying degrees of convection and diffusion in a group of 24 clinically stable patients with end-stage renal disease. Methods. The patients were prospectively assigned to three consecutive treatment schedules of 6 months each: phase I (HF1) (on-line predilution haemofiltration) → phase II (HD) (high-flux haemodialysis) → phase III (HF2; as phase I). We used the AK100/200 ULTRA monitor (Gambro), which prepares ultrapure dialysis fluid for HD and sterile, pyrogen-free substitution solution for HF. The membrane (polyamide), fluid composition, and treatment time were the same on HF and HD. The targeted equilibrated Kt/V was 1.2 for both treatment modes, creating a similar urea clearance. Results. Fifteen patients, mean age 62.8±8.4 years, completed the study according to the above conditions. Urea kinetics, nutritional parameters, and dry weight were similar in the three periods. The frequency of intra-treatment episodes of hypotension/patient/month was significantly lower on HF1 (1.24) and HF2 (1.27) than on HD (1.80) (P<0.04). It decreased progressively on HF1, then increased on HD, and decreased again during HF2. Patients had fewer muscular cramps on HF than on HD (P<0.03) and required significantly less saline and plasma expander during HF than HD sessions. The prevalence of inter-treatment symptoms, including fatigue and hypotension, was lower on HF than on HD (score difference P= 0.04). Quality of life, determined by the Laupacis method in all three periods, showed a tendency towards improvement during the study, reaching the best values during HF2. Conclusions. HF has a progressive stabilizing haemodynamic effect, producing a more physiological cardiovascular profile than HD. This long-term effect, observed in stable patients treated under strictly identical conditions, is probably due to the mechanism of convection, and is different from the acute effect observed mainly in unstable patients.