Assessment of the association between increasing membrane pore size and endotoxin permeability using a novel experimental dialysis simulation set-up

BACKGROUND Membranes with increasing pore size are introduced to enhance removal of large uremic toxins with regular hemodialysis. These membranes might theoretically have higher permeability for bacterial degradation products. In this paper, permeability for bacterial degradation products of membranes of comparable composition with different pore size was investigated with a new in vitro set-up that represents clinical flow and pressure conditions. METHODS Dialysis was simulated with an AK200 machine using a low-flux, high-flux, medium cut-off (MCO) or high cut-off (HCO) device (n = 6/type). A polyvinylpyrrolidone-solution (PVP) was recirculated at blood side. At dialysate side, a challenge solution containing a filtrated lysate of two water-borne bacteria (Pseudomonas aeruginosa and Pelomononas saccharophila) was infused in the dialysate flow (endotoxin ≥ 4EU/ml). Blood and dialysate flow were set at 400 and 500 ml/min for 60 min. PVP was sampled before (PVPpre) and after (PVPpost) the experiment and dialysate after 5 and 55 min. Limulus Amebocyte Lysate (LAL) test was performed. Additionally, samples were incubated with a THP-1 cell line (24 h) and IL-1β levels were measured evaluating biological activity. RESULTS The LAL-assay confirmed presence of 9.5 ± 7.4 EU/ml at dialysate side. For none of the devices the LAL activity in PVPpre vs. PVPpost was significantly different. Although more blood side PVP solutions had a detectable amount of endotoxin using a highly sensitive LAL assay in the more open vs traditional membranes, the permeability for endotoxins of the 4 tested dialysis membranes was not significantly different but the number of repeats is small. None of the PVP solutions induced IL-1β in the THP-1 assay. CONCLUSIONS A realisitic in vitro dialysis was developed to assess membrane translocation of bacterial products. LAL activity on the blood side after endotoxin exposure did not change for all membranes. Also, none of the PVPpost solutions induced IL-1β in the THP-1 bio-assay.