Hepatic removal of two fractions of indocyanine green after bolus injection in anesthetized pigs.

In the anesthetized pig, we studied the kinetics after intravenous bolus injection of two fractions of indocyanine green (ICG): the genuine ICGg (95-99% of total) and a degradation product, ICGdp (1-5%). Plasma concentrations were followed in the carotid artery and a hepatic vein. ICGg disappearance curves (n = 7) were biexponential with rate constants alpha = 0.189 +/- 0.021 min-1 and beta = 0.0356 +/- 0.0061 min-1. The hepatic extraction fraction was constant with time. A detailed mathematical analysis showed this to be in disagreement with the conventional assumption that the biexponential plasma disappearance curve is a result of backflux from the liver storage to plasma. In contrast, our observations were predicted by an alternative model assuming temporary extrahepatic, extravasal redistribution during first-order, one-way hepatic uptake. Nevertheless, when a large bolus of sulfobromophthalein (BSP) was injected 20 min after ICG, a net backflux of ICG could be demonstrated, presumably due to countertransport. Thus a sufficient description of ICGg kinetics must include the complex kinetic behavior of the hepatic membrane carrier involved. Mass spectrometry suggested that ICGdp is formed by two ICGg molecules. Plasma elimination of ICGdp was slower (alpha = 0.0094 +/- 0.0007 min-1). Analysis of the bile after bolus injection (n = 2) of ICGdp revealed two possible metabolites of ICGdp that were not found in urine. Since BSP injection did not alter the ICGdp disappearance curve, ICGdp is probably not taken up by the same hepatic membrane carrier as ICGg.