abstract
- UNLABELLED: PET has been used to quantify striatal 6-[18F]fluoro-L-dopa (FDOPA) uptake as a measure of presynaptic dopaminergic function. It has been suggested that the estimation of dopa-decarboxylation (DDC) rate, kD3, using a compartmental approach to dynamic FDOPA/PET data, can provide a better objective marker of parkinsonism. This modeling process, however, requires many assumptions to estimate DDC activity with acceptable errors. METHODS: We combined FDOPA 3-O-methyl-fluorodopa PET studies on three normal subjects and five Parkinson's disease patients. RESULTS: The contradicted modeling assumptions are: (a) the rate constants across the blood-brain barrier, KD1 and kD2, for 3OMFD and FDOPA were in similar range (ratio approximately equal to 1) and thus not equal to assumed values of KM1/KD1 of 2.3 derived from rat studies and applied to human FDOPA studies and (b) the KD1/kD2 ratio for frontal cortex was not equal to that for the striatum (0.70 +/- 0.15 versus 1.07 +/- 0.3; p < 0.002). Discriminant analyses indicate that simple estimates like the striatum-to-occipital ratio, or the graphically derived unidirectional transport rate constant (KiFD) separate normals from Parkinson's disease patients at least as accurately as estimates of striatal DDC activity (kD3). CONCLUSION: Measurements of striatal DDC activity with dynamic FDOPA/PET and compartmental modeling may be based on incorrect assumptions. Even though such complex models yield microparameters that may be applicable to certain clinical research demands, they may produce misleading results in other experimental settings.