Effect of intravascular ligand binding on parameter estimates derived from tracer kinetic modelling.

The purpose of this study was to assess the effect of intravascular ligand binding on parameter estimates derived from tracer kinetic modelling. To this end intravascular ligand kinetics between the free and a bound compartment in plasma and the exchange of tracer between the capillary space and tissue were analysed using a simple compartment model. The effect of non-equilibrated intravascular compartments on parameter estimates was evaluated in a computer simulation. It was found that three kinetic situations must be distinguished. If the intravascular compartments are fully equilibrated when the ligand reaches the target organ, intravascular binding simply acts as a scale factor for the transport-related parameter K1. If the intravascular kinetics is very slow, only minimal binding will occur. In between there is a range where ongoing equilibration leads to time variability of K1. Since tracer kinetic modelling usually does not account for such time variability, the parameter estimates become biased, the degree of the bias depending on the intravascular binding kinetics. Furthermore the bias may be dependent on receptor density, meaning that model-derived receptor estimates are not linearly related to the true receptor density. It is concluded that intravascular ligand binding can severely affect parameter estimates derived from tracer kinetic modelling. Especially disturbing are effects due to ongoing intravascular equilibration following the arrival of the ligand in the target organ. These can be avoided by letting the ligand equilibrate with blood in a syringe prior to injection.