Hyaluronan and peritoneal ultrafiltration: a test of the "filter-cake" hypothesis

Adding hyaluronan (HA) to the dialysis fluid seems to improve the efficiency of peritoneal dialysis (PD). This effect may be explained by the gradual formation of a HA "filter-cake" that decreases the tissue hydraulic conductivity. A filter cake (concentration hyperpolarization layer) can be formed when a large, slowly diffusible molecule, such as HA, is partly sieved by the pores of a membrane during the process of transmembrane ultrafiltration. The filter cake then forms at the "uphill" membrane-fluid interface, thereby increasing the resistance to fluid flow across the membrane. To test the filter-cake hypothesis, we investigated the effects of intraperitoneal (IP) HA on peritoneal small solute and water transport by administering HA either during the dwells or as incubations before PD dwells in rats. In the first set of experiments, HA, 0.01% (n = 7), 0.05% (n = 6), and 0.1% (n = 7) was given in 20 mL dialysis fluid (3.86% Dianeal). Control group was instilled with 20 mL of dialysis fluid. Evans Blue (EB) albumin was given as an intra-arterial (IA) bolus and (51)Cr-EDTA as an intravenous (IV) infusion. Plasma and dialysate were sampled up to 240 minutes to determine total peritoneal clearance (Cl), clearance from dialysate to plasma (Cl-->P) of (125)I-albumin (RISA), clearance from plasma to dialysate (Cl-->D) of EB-albumin, and mass transfer area coefficients (MTAC or permeability-surface area products, PS) of (51)Cr-EDTA and glucose. Peritoneal ultrafiltration (UF) was determined from RISA dilution. In the second set of experiments, rats were first incubated with 4 mL of phosphate-buffered saline (PBS) or PBS containing 0.1% HA for 120 minutes. Rats were then dialyzed with HA-free PD fluid, and sampling of plasma and dialysate was performed for 60 minutes. For HA concentrations exceeding 0.01%, UF volumes increased with increasing doses of HA. Small solute MTACs and initial UF were unaffected when HA was either given during the dwell or as a preincubation. Compared with control, there was a significant decrease in RISA-Cl for 0.05% HA and 0.1% HA. Also, Cl-->P decreased significantly compared with control for 0.1% HA. In conclusion, the present data clearly demonstrate that small solute MTACs and the glucose-induced osmotic water transport occurring early in the dwell are not affected by HA. Only the back-filtration of fluid from peritoneum to plasma was affected.