The functional ultrastructure of the blood-lymph barrier. Computer analysis of data from dog heart-lymph experiments using theoretical models.

Abstract

AbstractA generalized two‐pore model including differences in permeability, surface area and pressure conditions along the capillary, was formulated and certain simplifications were tested with respect to their influence on the estimated parameters. A simplified version of the model was tested for compatibility and parameter confidence against a series of macromolecular sieving data obtained from Starling dog heart‐lung preparations. In these experiments heart‐lymph was collected during steady state conditions and the venous pressure was changed.During “normal” conditions the blood lymph barrier of the dog heart‐lung preparations could be adequately described by an equivalent two‐pore membrane with small pores of radius 35–60 Å, and an additional set of larger pores of radius 120–160 Å, with the density of one large pore per 10,000 small pores. The mean transcapillary pressure difference was calculated to be less than 1 cm H2O.During elevated venous pressure the effective surface area for fluid exchange was found to decrease to 1/2–1/3 of the normal value, due to a pronounced decrease in small pore radius. The concomitant increase in large pore radius caused an increased flow of fluid through the large pores. These results, in addition to the corresponding pressure perturbation data, have given some support to the pressure compensatory mechanisms suggested by Wiederhielm (1968).The hypothesis that the transport of macromolecules takes place by way of cytopempsis was shown to be tenable if the ad hoc assumption was made that the vesicle size or density is pressure dependent.