Compartmental models: Limitation for measurement of CBF with PET

All tracer kinetic models currently employed with positron emission tomography (PET), are based on compartmental assumptions; that is, tracer transport can be interpreted in terms of compartments in which tracer concentrations are assumed to be spatially uniform. The assumption of uniformity of tracer concentration leads one to accept in the case of a one compartment model, for instance, that an inert radioactive tracer designed to measure CBF is instantaneously distributed throughout the vasculature and available extravascular tissue space upon entry into the brain. The author's studies were based upon in-vivo experimental data obtained from the bolus, intracarotid injection of a small aliquot (0.5 ml) of blood labeled with H/sub 2//sup 15/O in rhesus monkeys. Using the independent measurements compared the authors compared the prediction based on unit-impulse responses of four different models of increasing complexity for transport of an inert, diffusible tracer when given the measured and E and estimates of blood volume, tissue diffusion resistance, water space for transport of an inert, diffusible tracer. The author's results clearly indicate superiority of the two-barrier distributed model over the other models and emphasize the shortcomings of a one-compartment model