Analysis of the influence of the infusion site on dialyser clearances measured in an in vitro system mimicking haemodialysis and haemodiafiltration.

ABSTRACT

BACKGROUND: Blood flow (QB), dialysate flow (QD), and dialyser characteristics are the three major factors driving dialysis efficacy. Haemodiafiltration has added an increased convective volume to increase efficacy. We aimed to assess the influence of the infusion site of the replacement fluid in an in vitro system emulating haemodiafiltration. METHODS: An in vitro system allowing us to control the dialysate temperature, concentration gradient, the flow of both dialyser sides over a range wider than that compatible with clinic, was set to evaluate the influence of the different parameters on dialysis efficacy. The total ion clearance was used as an accepted method for small molecule clearance assessment. Cellulose triacetate (CT190C, Baxter; FB170U, Nipro) and polysulfone (HF80, Fresenius) dialysers were included in the study. Dialysis as well as on-line diafiltration both with pre- and postdilutional infusion were assessed. The experimental conditions presented in this study included QD 620 and 970 ml/min. The convective flows ranged from 50 to 200 ml/min. RESULTS: For a QD = 620 ml/min and a QB = 350 ml/min the total ion clearance ranged from 269 to 274 for HF80, from 291 to 294 for FB170 and from 294 to 302 for CT190. The variability of the measurements was very low (SD < 1%). Total ion clearance increased by 17-21% when QB was raised from 300 to 400 ml/min. Increasing QD from 420 to 970 ml/min (for QB = 350 ml/min), resulted in an increase in total ion clearance which was more marked at lower QD (from 420 to 620 ml/min) and plateaued thereafter (from 620 to 970 ml/min). Postdilutional on-line diafiltration with 100 ml/min of infusate resulted in an additional increase in total ion clearance of 5.4-8.6%. This increase was proportional to the infused volume. On the contrary, predilutional on-line diafiltration resulted in a decrease in total ion clearance which was also proportional to the infused volume (between -5.1 and -6.9% at 100 ml/min infusion volume and -9.7 to -12.9% at 200 ml/min). CONCLUSIONS: The present in vitro system provided accurate and reproducible results on dialyser clearances. Our experiments confirmed previous studies on the influence of QB and QD on dialyser efficacy. Further, they show that the proportional increase in postdilutional on-line diafiltration is lesser than that previously reported. More importantly, they also show that pre-dilution infusion in high efficiency systems results in a drop in dialyser clearance compared to dialysis alone, again proportional to the infusion rate. Thus, increasing the convective flow may increase dialysis efficacy even more than increasing QD alone. However, the choice of infusion site is crucial to obtaining this benefit in small molecule clearances.