Permeability of cat skeletal muscle capillaries to small solutes.

Permeability-surface area products (PS) for 51Cr-EDTA, [3H]mannitol, [14C]urea, and 22Na were measured in isolated, perfused, lower hindlimb muscles of anesthetized cats. The tracers were added stepwise to the arterial inflow, and Evans blue-labeled albumin was the reference indicator. At flow rates > 70 ml.min-1.100 g-1, the PS values (+/- SE) were 5.0 +/- 0.5, 7.6 +/- 1.2, 17.8 +/- 1.4, and 21.0 +/- 1.4 (n = 7, 4, 5, and 3 animals, respectively). The ratio of simultaneous PS measurements of mannitol and urea was 0.42 +/- 0.02 (n = 9), significantly less than the free diffusion coefficient ratio (0.49), indicating the presence of restricted diffusion. PS measurements were also made during osmotic flow (4.2 +/- 0.6 ml.min-1.100 g-1) induced by 20% NaCl. The data clearly showed that osmotic transients did not alter small solute permeability. Pore models were used to show that the PS data and previously reported reflection coefficient data were consistent with a single description of the capillary wall. This model contained a water-only pathway containing 60% of the hydraulic capacity and an extracellular route modeled by pores of 4 nm radius having 21,000 cm of area per unit membrane thickness (A/delta x).