Solute and Water Transport Across the Peritoneal Barrier

Distributed Nature of the Barrier

Fig. 180.1 displays the elements of the peritoneal barrier, which is much more complex than the concept of a single “peritoneal membrane.” As illustrated, the barrier has the following three components: (1) the anatomic peritoneum, (2) the cell-interstitial matrix, and (3) blood capillary endothelium lining the vasculature, which is distributed within the tissue. The anatomic peritoneum consists of a single layer of mesothelial cells overlying several layers of connective tissue. The visceral peritoneum has been dissected and measured to be 90 mm thick in the normal state. Although many nephrologists consider the anatomic peritoneum the barrier to transport, experiments in both humans and rodents have demonstrated that the peritoneum is not a barrier to solute and water transport. Complete destruction of the peritoneum in rodents has had no effect on the transfer of small solutes or the osmotic filtration of fluid from the peritoneal cavity into a transport chamber. There have been parallel findings in patients who undergo extensive peritonectomy for treatment of peritoneal carcinomatosis; in one report, clearance of mitomycin C from the peritoneal cavity was not significantly affected by an extensive peritoneal resection. On the other hand, if the overall peritoneal thickness increases with uremic inflammation and fibrosis from chronic contact over months with dialysis fluid, the transfer of water but not solute would be affected. However, unless there were ongoing inflammation in the abdominal cavity, the mesothelium and the underlying tissue should be relatively normal in AKI without other changes.

The two other major components of the peritoneum therefore make up the barrier. The cell-interstitial matrix restricts movement of solutes and water between the blood capillary walls and the peritoneal cavity, slowing transport and making it less efficient than if the blood vessels were in direct contact with the dialysis solution. Because the muscle of the abdominal wall and the gut constitute the vast majority of the peritoneal surface in contact with the dialysis solution, the vessels of these tissues dominate transport. The endothelium of most smooth muscles, capillaries, and venules is known to be size selective. As illustrated in Fig. 180.1 , these vessels, through which the blood flows, are distributed within these tissues, which are surrounded by cells and the interstitial matrix.

Effects of the Interstitial Matrix on Transport

The interstitial matrix, once considered to be inert, “sticky” mucopolysaccharides and termed “ground substance,” is now known to be an orderly structure of the tissue. Collagen fibers, which provide the skeleton of the interstitial network, are linked to interstitial cells and possibly pericytes through adhesion molecules such as β ₁ -integrins. These collagen fibers can stretch and contract as the cells to which they are attached are stimulated in different ways. Wrapped around the collagen fibers and, in some cases, attached to them are large (1-40 megadaltons) molecules of hyaluronan, with proteoglycan molecules bound to the hyaluronan molecules. The hyaluronan molecules within the collagen matrix are highly negatively charged, imbibe large amounts of water, and restrict the passage of negatively charged proteins. Proteins are typically restricted to about 50% of the interstitial space, which translates into a protein space of 6% to 10% of the entire tissue space available to proteins for transport if the typical interstitial space is only 12% to 20%.

The rates of transport through the tissue depend on the interstitial matrix. Transport includes diffusion, which can be described by the effective diffusivity as follows:

where D _eff is effective tissue diffusivity; D _isf is diffusion coefficient in the interstitium; θ _isf is the interstitial fraction (fraction of the total tissue space available to the solute); and τ is tortuosity (factor to account for the convoluted path of the solute around cells and through the interstitial matrix). For water transport and substances that are transported chiefly through convection or solvent drag, the hydraulic conductivity of the tissue space (K _tiss ) has been shown to depend on the interstitial fraction and the concentrations of collagen, proteoglycan, and hyaluronan.

Dialysis solutions infused into the peritoneal cavity typically cause intraperitoneal hydrostatic pressures (IPP) above 3 or 4 mm Hg, which alter the surrounding tissue space. Intraperitoneal pressures depend on the size and position of the patient, and on the infusion volume used ( Fig. 180.2A ). IPPs of 4 mm Hg would seem to be a very small increase, but the tissue responds by absorbing significant amounts of fluid. This absorption occurs particularly in the abdominal wall, where there is a positive-pressure gradient from the serosa to the subcutaneous space ( Fig. 180.2B ). Studies in rats have demonstrated that the extracellular space doubles with a rise in IPP from zero to 3 mm Hg, and the hydraulic conductivity increases four to five times. In experiments in the rat, sampling of the interstitial fluid after 4 hours of dialysis showed a 50% decrease in colloid osmotic pressure. Expansion of the interstitial space and decreases in collagen and hyaluronan concentrations raise the rates of diffusion and convection within the tissue. Clinical complications can occur from this intraabdominal pressure that may result in abdominal wall hernias or inhibit diaphragmatic movement of ventilated ICU patients.

Nature of the Endothelial Barrier

The endothelial barrier is depicted in Fig. 180.3 as a transcellular pore, called an aquaporin , and different intercellular gaps lined with matrix material, called the glycocalyx ; this concept represents a necessary modification of the three-pore model of peritoneal transport to account for alterations in pathologic states. The aquaporin permits only water through its channel and is responsible for much of the osmotically induced filtration from the plasma. Intercellular gaps lined by the glycocalyx are the second portion of the barrier, which permit the transfer of solutes and water, depending on the density of the glycocalyx.

The discovery of aquaporins by Agre and colleagues has brought new understanding to the transfer of water across blood capillaries into the tissue and subsequently into the peritoneal cavity. Because the aquaporin does not permit any solute to transfer, it represents the perfect semipermeable membrane across which any solute concentration difference exerts osmotic pressures that result in filtration. The functional significance of the aquaporins has been demonstrated in numerous experiments. Carlsson et al. showed that in vivo inhibition of aquaporins with mercuric chloride resulted in a significant decrease in volume of osmotically filtered fluid from the tissue. Sixty-six percent inhibition of water flow through the aquaporins was verified subsequently by Yang et al. in aquaporin 1—knockout mice. When mice were dialyzed with a hypertonic solution, the filtration in the knockout mice was 40% of that in normal mice. Another study in rodents has demonstrated both the structural appearance and the functionality of the endothelial aquaporins.

Solute transport depends on the density of the glycocalyx in the intercellular gap. A denser glycocalyx restricts the passage of larger solutes (functionally equivalent to the “small pore” of the three-pore model), with a less dense glycocalyx allowing protein leakage (the equivalent of the “large pore” of the three-pore model). In the normal situation, the vast majority of the intercellular spaces are densely packed with glycocalyx and restrict the passage of macromolecules, making up 95% of the total capillary permeable area; these spaces are responsible for 40% to 50% of the osmotic filtration. Vlahu et al. examined the endothelial glycocalyx in patients with chronic renal failure; using Sidestream Darkfield imaging, they demonstrated significant damage to the glycocalyx barrier. Subsequent studies in rodents did not demonstrate major changes to the peritoneal glycocalyx with exposure to dialysis solutions. These authors concluded that this area needs further research. The remainder of the capillary is made up of intercellular junctions, which permit proteins to leak out.

The rate of transfer from the plasma to the interstitial space of the surrounding tissue can be represented in a simplified fashion as follows:

where J _endo is solute transfer rate across the endothelium (mass/time/tissue mass); p is overall endothelial permeability, including the effect of all intercellular passages; a is capillary surface area/mass of tissue; C _plasma is solute concentration in plasma; and C _isf is solute concentration in interstitial fluid. More complicated mathematical approaches can be employed to include the three elements of the endothelial barrier.