Introduction
Recently the 11C labelled compound PIB has been demonstrated to image brain β-amyloid in Alzheimer's disease (AD) 1 . Here we report the use of net accumulation and unidirectional influx of PIB for inter-subject differentiation of patients with clinical AD diagnosis.
Methods
Uptake data were acquired on an ECAT EXACT HR+ camera during 60 min following bolus administration of approximately 300 MBq of PIB. The time course of radioactivity in arterial plasma corrected for labelled metabolites was determined and used as input function. Four healthy controls (HC) and four AD patients were examined. Parametric maps were constructed using a linear algorithm 2 .
Results
The data distributions in Gjedde-Patlak plots followed asymptotically straight lines after 30 min, characteristic for irreversible kinetics. The tracer is thus far from equilibrium during the first hour after administration, and consequently models with irreversible kinetics are appropriate. The simple model with one reversible and one irreversible tissue compartment and three rate constants was applied. The unidirectional influx rate constant K1 (right part of figure 1) was found to be of the order of 0.2–0.3 min−1 (whole brain), corresponding to relatively high extraction (40–60%) and indicating that K1 might be a quite good index of CBF. The accumulation of PIB in frontal and parietal cortex measured with the macro parameter (net accumulation rate constant) Kacc = K1k3/(k2+k3) was found to give a good discrimination between the AD-patients and the HC subjects (left part of figure 1), except for two AD patients where PIB net accumulation was close to the low values found in the HC subjects (cf. upper left in the figure 1). These two patients had lower K1 than the HC subjects in many cortical areas, close to the values found for the other two patients with high net accumulation of PIB (cf. lower right in the figure 1). The MMSE scores for the former patients were close to normal, in the range 28–29.
Conclusion
A group of patients with clinical diagnosis AD but with very low PIB accumulation could not be discriminated from the HC subjects with only PIB net accumulation as measure. Also in this AD group we found comparatively low cortical K1 values, suggesting that their symptoms can be related to a vascular deficit rather than β-amyloid deposition.