Clinical implications of a three-pore model of peritoneal transport.

The peritoneal barrier exchange characteristics are in this article described in terms of a three-pore model of membrane permselectivity. The peritoneal membrane during continuous ambulatory peritoneal dialysis (CAPD) is thus simulated to have a large number of small pores of radius 40-55 A, a small number of large pores of radius 200-300 A, and an abundance of transcellular pores of radius 4-5 A. Due to the heteroporous nature of the peritoneal membrane, peritoneal small solute sieving coefficients are of the order of 0.5-0.6, and not near unity, as predicted for a homoporous membrane having 50 A (radius) equivalent pores, but lacking transcellular pores. As a consequence, the dialysate during CAPD is diluted during the first 50-100 minutes of the dwell. Furthermore, there is a marked coupling between the increased net transperitoneal volume flow, occurring early in the cycle, and the transfer of "small" macromolecules, such as beta 2-microglobulin and albumin, across the peritoneal membrane. This coupling is, however, small for "large" macromolecules, such as IgG and IgM, or for small solutes. Increasing the peritoneal surface area, in computer simulations of peritoneal transport according to the three-pore model, causes the simulated intraperitoneal (i.p.) volume vs. time (V(t)) curves to peak earlier than during control, while the maximum volume ultrafiltered is not markedly affected. However, selectively increasing the glucose PS (mass transfer area coefficient) causes a reduction both in the peak time and the peak "height" of the V(t) curves. The latter pattern is also seen when the dialysate volume is reduced. It is concluded that a three-pore model of membrane permselectivity selectivity can adequately describe the kinetics of peritoneal transport of small and large solutes and of fluid.