Effect of blood flow (QB) and infusion flow rate (Qi) on efficacy of on-line hemodiafiltration

On-line HDF is a technique which combines diffusion with convection and use s pyrogen-free dialysate as replacement fluid. During postdilution on-line HDF additional convective removal is possible. In the present study we have evaluated the influence of variations in QB and Qi on removal of small and high molecules. Twelve patients were included in this study, 8 males and 4 females. Every p atient received 7 on-line HDF sessions with a Fresenius 4008 B machine, QD 800 ml/min, 1.9 m(2) high-flux polysulfone, UF 0.5 L/h and Td 60 min. Only QB or Qi was changed, 2 sessions with QB 300 ml/min (Qi 25 & 100 ml/min), 2 with QB 400 (Qi 25 & 100) and 3 with QB 500 (Qi 25 100 & 150). Arterial pr essure, venous pressure, transmembrane pressure and effective QB (eQB) were monitored. Plasma urea, creatinine, phosphate, uric acid and beta(2)m conc entrations were measured at the beginning (Ini) and at the end of dialysis from arterial (Art) and venous (Ven) blood lines, and arterial blood line w ith slow flow method (per). Recirculation, dialyzer solutes clearance, K = (eQB-0.0085 x R x eQB) x (Art-Ven)/per, Kt/V = Ln (Ini/Per) and solute redu ction rates, SRR = 100 x (Ini-Per)/Ini, were calculated. Mean recirculation was 10% lower for all study situations. Mean dialyser ur ea K was: at QB 300: 236 +/- 13 ml/min (Qi 25) and 250 +/- 10 (Qi 100); at QB 400: 275 +/- 78 (Qi 25) and 292 +/- 14 (Qi 100); at QB 500: 293 +/- 17, 314 +/- 13, 323 +/- 10 mi/min (Qi 25 100 and 150 ml/min, respectively). Sim ilar behaviour was observed for creatinine, phosphate and uric acid althoug h with lower clearances. No significant differences in dialyzer beta(2)m cl earance was found with QB changes during on-line HDF, but a significant inc rease when Qi was raised: 70 +/- 11 (Qi 25), 117 +/- 16 (Qi 100) (p < 0.001 ) and 152 +/- 16 mi/min (Qi 150) (p < 0.001). The beta(2)m reduction ratio was 40.6 +/- 8.2% with Qi 25 and it was increased to 54.1 +/- 7.1% (p < 0.0 1) and 59.8 +/- 8.4% (p < 0.01) with Qi 100 and 150 ml/min respectively. In conclusion, in on-line HDF small molecule removal is better when both QB and Qi are raised; beta(2)m removal depends only on Qi and is not influenc ed by QB. High QB is necessary to achieve high Qi.