Amyloid load but not regional glucose metabolism predicts conversion to Alzheimer's dementia in a memory clinic population.

Abstract

The value of imaging regional glucose metabolism with [18F]FDG PET for the prediction of progression from mild cognitive impairment (MCI) to Alzheimer's dementia (AD) is controversial. The predictive value of imaging with [18F]FDG PET was therefore tested and compared with that of imaging beta-amyloid load with [11C]PIB PET in the same memory clinic population of MCI patients. Thirty-nine patients with MCI who had undergone [18F]FDG as well as [11C]PIB PET were identified from a single-centre clinical registry. [18F]FDG and [11C]PIB PET images were rated as positive or negative for the presence of regional hypometabolism typical of AD and beta-amyloid deposition, respectively. Raters were blinded to the clinical information. Patients were followed clinically for 2.7 ± 1.2years after PET. Cox proportional hazards models, adjusted for age and sex, were used to test the predictive value of [18F]FDG PET, [11C]PIB PET, and both in combination. [18F]FDG PET did not significantly predict conversion to AD (p > 0.1). By contrast, models including [11C]PIB PET only (p < 0.05) or both [18F]FDG and [11C]PIB PET (p < 0.05) significantly predicted conversion to AD. The hazard ratio for AD in patients with a positive [11C]PIB scan was 10.2 (95% confidence interval 1.3-78.1). The results were confirmed by analysis of semiquantitative measures using normalized [18F]FDG uptake and [11C]PIB standardized uptake value ratios in AD-typical regions as continuous predictors. In contrast to [11C]PIB PET, [18F]FDG PET did not predict conversion from MCI to AD in this clinical patient sample. Therefore, amyloid PET should be preferred for individual prediction and patient counselling in clinical practice.