DOI: 10.1007/s00259-017-3634-3Pages: 1-9

Value of FDG-PET scans of non-demented patients in predicting rates of future cognitive and functional decline

1. David Geffen School of Medicine at the University of California Los Angeles, Ahmanson Translational Imaging Division, Department of Molecular and Medical Pharmacology

Correspondence to:
Daniel H. S. Silverman
Tel: 310-983-1446




The aim of this study was to examine the value of fluorodeoxyglucose (FDG) positron emission tomography (PET) in predicting subsequent rates of functional and cognitive decline among subjects considered cognitively normal (CN) or clinically diagnosed with mild cognitive impairment (MCI).


Analyses of 276 subjects, 92 CN subjects and 184 with MCI, who were enrolled in the Alzheimer’s Disease Neuroimaging Initiative, were conducted. Functional decline was assessed using scores on the Functional Activities Questionnaire (FAQ) obtained over a period of 36 months, while cognitive decline was determined using the Alzheimer’s disease Assessment Scale-Cognitive subscale (ADAS-Cog) and Mini-Mental State Examination (MMSE) scores. PET images were analyzed using clinically routine brain quantification software. A dementia prognosis index (DPI), derived from a ratio of uptake values in regions of interest known to be hypometabolic in Alzheimer’s disease to regions known to be stable, was generated for each baseline FDG-PET scan. The DPI was correlated with change in scores on the neuropsychological examinations to examine the predictive value of baseline FDG-PET.


DPI powerfully predicted rate of functional decline among MCI patients (t = 5.75, p < 1.0E-8) and pooled N + MCI patient groups (t = 7.02, p < 1.0E-11). Rate of cognitive decline on MMSE was also predicted by the DPI among MCI (t = 6.96, p < 1.0E-10) and pooled N + MCI (t = 8.78, p < 5.0E-16). Rate of cognitive decline on ADAS-cog was powerfully predicted by the DPI alone among N (p < 0.001), MCI (t = 6.46, p < 1.0E-9) and for pooled N + MCI (t = 8.85, p = 1.1E-16).


These findings suggest that an index, derivable from automated regional analysis of brain PET scans, can be used to help predict rates of functional and cognitive deterioration in the years following baseline PET.

To access the full text, please Sign in

If you have institutional access, please click here

  • Accepted: Jan 16, 2017
  • Online: Mar 22, 2017

Article Tools