PURPOSE: The aim of the study was to estimate normal ranges and test-retest measures for various parameters characterising dopamine metabolism from a prolonged (18)F-dopa positron emission tomography (PET) measurement using a reference tissue model and compare their value for the detection of early Parkinson's disease (PD).

METHODS: Healthy volunteers (n = 9) and patients (n = 36) in an early stage of PD underwent an (18)F-dopa PET measurement lasting 4 h. The influx rate constant k(occ) and the effective distribution volume ratio (EDVR, its inverse is an indicator for dopamine turnover) were estimated by a graphical approach using dynamic data in the striatum and, as a reference region, the occipital cortex. Furthermore, ratios of activity concentrations between striatum and occipital brain taken for three time intervals completed the data analysis. All parameters were determined both in eight small volumes of interest placed in the striatum as well as averaged for caudate nucleus and putamen. For the control group, reproducibility was checked in a second study 3 months later and ranges for normal values were derived from mean ± 2 standard deviations. Receiver-operating characteristic (ROC) analyses were performed to assess the value of the parameters for diagnostic purposes.

RESULTS: Patients with early-stage PD and healthy volunteers could be separated by the values of the putamen, not the caudate nucleus. The normal ranges of the putamen were 0.0151-0.0216/min for the influx rate constant k(occ) and 2.02-3.00 for EDVR. For the various time intervals used the striato-occipital ratios yielded 2.24-3.06, 2.43-3.42 and 2.35-3.21, respectively. Patients were characterised by significantly lower values (p < 0.001) and significant differences between ipsi- and contralateral sides (p < 0.001) with regard to their clinical symptoms and a rostrocaudal gradient. EDVR as well as k(occ) for the putamen were able to effectively differentiate between groups (sensitivity >97%, specificity 100%). In contrast, striato-occipital ratios showed a sensitivity of about only 85%.

CONCLUSION: For clinical applications, our data do not demonstrate any superiority of the EDVR determination compared to influx rate constant, while requiring long and tedious acquisition protocols. The normal range estimates do not represent absolute quantitative measures for dopamine metabolism but are specific for the chosen acquisition and processing procedures.