Interpretation of simultaneous measurements of hepatic extraction fractions of indocyanine green and sorbitol: evidence of hepatic shunts and capillarization?

Sorbitol and indocyanine green (ICG) have high hepatic extraction fractions (E(sorb) and E(ICG)) in normal subjects. A curved relationship has been observed between E(sorb) and E(ICG) in liver disease. According to one interpretation, the decrease of E(sorb) is a result of intrahepatic shunting and 1 - E(sorb) is the fraction of shunted flow (the shunt hypothesis). Under the further assumption that capillarization of functioning sinusoids prevents hepatic uptake of plasma protein-bound ICG and allows uptake of water-soluble sorbitol, the difference E(sorb) - E(ICG) has been suggested as a measure of capillarization. We propose an alternative hypothesis: that the sinusoidal permeability-surface area products for sorbitol and ICG are reduced in proportion by liver disease (proportional reduction hypothesis). Based on the sinusoidal perfusion model, predictions were produced from both hypotheses for the relation between E(sorb) and E(ICG) and the additional effects of capillarization were described. By use of liver vein catheterization, E(sorb) and E(ICG) were simultaneously measured during continuous infusions in 53 human subjects with varying degrees of liver disease. The data were in better agreement with the predictions of the proportional reduction hypothesis than with the shunt hypothesis. Even though both intrahepatic portosystemic shunts and sinusoidal capillarization are known to occur in cirrhosis and also may have influenced our data, they appeared to be of minor importance from a kinetic point of view. These findings favor the proportional reduction hypothesis and do not support the use of systemic nonrenal clearance of sorbitol as a measure of "functional liver blood flow."