Impact of contamination with long-lived radionuclides on PET kinetics modelling in multitracer studies.

Abstract

An important issue in multitracer studies is the separation of signals from the different radiotracers. This is especially the case when an early tracer has a long physical half-life and kinetic modelling has to be performed, because the early tracer can confer a long-lived contaminating background not only to images but also to a measured input function derived from blood samples. In this study, we examined data from a sequential multitracer infection study involving In (t1/2=2.8 days), investigating the influence on gamma counting of blood samples and on the kinetic modelling of subsequent PET tracers. Blood sample counts were corrected by recounting the samples a few days later. A more optimal choice of energy window was also explored. The effect of correction versus noncorrection was investigated using a two-tissue kinetic model with irreversible uptake (K1, k2, k3). K1 was least affected and k3 was most affected by the contamination, corresponding to the effect being relatively larger on the late part of the blood input function. A narrower energy window reduced the problem, but this will not be possible for all types of contaminating background. Gamma counting of blood samples can lead to a contaminating background not observed in PET imaging and this background can affect kinetic modelling. If the contaminating tracer has a much longer half-life than the foreground tracer, then the problem can be solved by late recounting of the samples.