Multicenter Standardized <sup>18</sup>F-FDG PET Diagnosis of Mild Cognitive Impairment, Alzheimer's Disease, and Other Dementias

Lisa Mosconi(New York University), Wai Hon Tsui(New York University), Karl Herholz(University of Manchester), Alberto Pupi(University of Florence), Alexander Drzezga(Ludwig-Maximilians-Universität München), Giovanni Lucignani(University of Milan), Eric M. Reiman(Banner - University Medical Center Phoenix), Vjera Holthoff‐Detto(TU Dresden), Elke Kalbe(University of Cologne), Sandro Sorbi(University of Florence), Janine Diehl‐Schmid(Ludwig-Maximilians-Universität München), Robert Perneczky(Ludwig-Maximilians-Universität München), Francesca Clerici(University of Milan), Richard J. Caselli(Arizona Research Center), Bettina Beuthien‐Baumann(Helmholtz-Zentrum Dresden-Rossendorf), Alexander Kurz(Ludwig-Maximilians-Universität München), Satoshi Minoshima(University of Washington), Mony J. de Leon(Nathan Kline Institute for Psychiatric Research)
Journal of Nuclear Medicine
February 20, 2008
Cited by 721Open Access
Full Text

Abstract

UNLABELLED: This multicenter study examined (18)F-FDG PET measures in the differential diagnosis of Alzheimer's disease (AD), frontotemporal dementia (FTD), and dementia with Lewy bodies (DLB) from normal aging and from each other and the relation of disease-specific patterns to mild cognitive impairment (MCI). METHODS: We examined the (18)F-FDG PET scans of 548 subjects, including 110 healthy elderly individuals ("normals" or NLs), 114 MCI, 199 AD, 98 FTD, and 27 DLB patients, collected at 7 participating centers. Individual PET scans were Z scored using automated voxel-based comparison with generation of disease-specific patterns of cortical and hippocampal (18)F-FDG uptake that were then applied to characterize MCI. RESULTS: Standardized disease-specific PET patterns were developed that correctly classified 95% AD, 92% DLB, 94% FTD, and 94% NL. MCI patients showed primarily posterior cingulate cortex and hippocampal hypometabolism (81%), whereas neocortical abnormalities varied according to neuropsychological profiles. An AD PET pattern was observed in 79% MCI with deficits in multiple cognitive domains and 31% amnesic MCI. (18)F-FDG PET heterogeneity in MCI with nonmemory deficits ranged from absent hypometabolism to FTD and DLB PET patterns. CONCLUSION: Standardized automated analysis of (18)F-FDG PET scans may provide an objective and sensitive support to the clinical diagnosis in early dementia.


Related Papers