Hepatic ICG removal in the pig depends on plasma protein and hematocrit: evidence of sinusoidal binding disequilibrium and unstirred water layer effects.

The influence of binding protein concentration and hematocrit on hepatic uptake of indocyanine green (ICG) was studied in anesthetized pigs during constant infusion of ICG. By exchange transfusions, we either substituted plasma protein with dextran 70 (n = 8) or changed hematocrit (n = 8). Intrinsic hepatic clearance of ICG, K, was calculated from plasma flow rate and concentrations in peripheral artery and liver vein after correction for extrahepatic distribution. By analyzing the relative change of K versus either the protein dilution factor or the change in plasma volume fraction (1-hct), we evaluated four current models for hepatic uptake of protein-bound substances even though a number of model parameters were unknown (parameter-free testing). Protein dilution factors (unitless) of 0.506 +/- 0.027, 0.673 +/- 0.011, and 0.749 +/- 0.028 were associated with inverse K ratios of 0.621 +/- 0.025, 0.758 +/- 0.021, and 0.817 +/- 0.013. These data rejected the traditional hypothesis that ICG uptake is proportional to the unbound concentration. They were compatible with development of binding disequilibrium along the sinusoidal lumen, an unstirred water layer close to the hepatocyte surface, or facilitated uptake from the bound pool. A plasma volume ratio [(1-hct2)/(1-hct1)] of 1.14 +/- 0.02 was associated with a K ratio of 1.07 +/- 0.02 (P = .01). Only sinusoidal binding disequilibrium predicted this finding, whereas an additional unstirred water layer effect could not be excluded. The observations could be simulated by a model that included both of these effects. Thus, neither the relative changes of K nor the absolute K values required the assumption of facilitated uptake from the bound pool. The parameter-free design presented may be useful with other ligands in intact animals.