A quantitative model for the measurement of cerebral vascular extraction fraction in vivo following intravenous injection: simulation studies.

A mathematical method has been developed by which the cerebral vascular extraction fraction of flow- and diffusion-limited tracers injected intravenously can be measured quantitatively. Successive injections of three tracers are required: a test tracer such as 15O-labeled water; a completely diffusible tracer as a reference tracer and also as a flow tracer; and a tracer for cerebral blood volume (CBV). The arterial tracer concentration curves and total integrated head counts of the test and reference tracers, as well as CBF and CBV values, are required to calculate the extraction fraction. No calibration is required between the head counts and arterial curves. No decay correction of the head and arterial activity is required, because isotopic decay is explicitly included in the equation. The effect of nonextracted test tracer can be corrected. Simulation studies have shown that the calculated extraction fraction values are not sensitive to measurement error in CBF, CBV, partition coefficient, change in measurement time, or time shift effect between the arterial and head data. If there is mixing of two different tissues in the brain, the calculated extraction fraction values are close to the weighted mean values of extraction fraction by relative weight of the tissues. It is concluded that it is possible to apply this method to human studies with a positron emission tomograph scanner and to animal studies with external coincidence detectors.