Hepatic transport kinetics and plasma disappearance curves: distributed modeling versus conventional approach.

The conventional approach to estimating hepatic transfer coefficients from the plasma disappearance curve ignores the effects of blood flow in the splanchnic and peripheral circulations. The effect of these simplifying assumptions on derived estimates of the rate constants has never been studied. To examine this problem we have constructed a distributed model that takes account of intrahepatic concentration profiles, nonuniform blood-flow distribution in the sinusoids, and delayed mixing in the peripheral circulation. Solutions to the new model constructed by numerical inversion of the Laplace transform afford a comparison between the new model and the conventional one. The sensitivity of the conventional equations to experimental error has also been evaluated. The results indicate that conventional estimates of the rate constants for hepatic uptake and cell-to-plasma efflux are subject to a systematic underestimate, the errors increasing rapidly with the initial extraction fraction. Estimates of the uptake constant obtained from the initial slope are especially susceptible to circulatory distortions and proved unacceptable even at low values of the initial extraction fraction. The liver content at 3 min did not in general provide a reliable index to these errors. In contrast to these problems, the conventional model returns generally accurate estimates of the steady-state plasma clearance and the rate constant for excretion.