Factors influencing measurement of protein reflection coefficient by filtered volume technique.

In isolated perfused organs, the protein reflection coefficient (sigma) can be estimated by comparing increases in hematocrit (Hct) and protein concentration (CP) during transvascular fluid filtration. In this study, we developed an equation for sigma to examine the potential influences of perfusate leak, evaporation, and hemolysis-induced changes in red blood cell volume and perfusate water. We also performed experiments in isolated ferret lungs to quantitate the magnitude of these potential sources of error and the effects of free hemoglobin on measurements of CP. These studies demonstrated that 1) perfusate leak does not cause an error because its effects on changes in Hct and CP counteract each other; 2) evaporation causes an overestimation of sigma, but in our experiments this effect was small; 3) hemolysis-induced changes in red blood cell and perfusate water volumes may cause an over- or underestimation of sigma, but these effects are small; 4) overestimations of CP due to increasing free perfusate hemoglobin concentration can cause substantial overestimations of sigma; and 5) values of sigma calculated from previous equations and from our equation were virtually identical, suggesting that the assumptions necessary for the previous equations were not significant sources of error. In agreement with previous workers, we conclude that the most important potential source of error is hemolysis-induced increases in free perfusate hemoglobin.