Advertisement

Hypometabolism of the posterior cingulate cortex is not restricted to Alzheimer's disease

      Background

      Hypometabolism of the posterior cingulate cortex (PCC) is associated with Alzheimer’s disease (AD). PCC vulnerability, however, could also be present in other neurodegenerative diseases, such as behavioural variant of frontotemporal dementia (bvFTD), and normal aging. The aims of this study were to assess 1) the proportion of AD, bvFTD and cognitive normal subjects (CN) with PCC hypometabolism, and 2) the relationship between PCC metabolism and demographical, neuropsychological, and neurobiological characteristics in AD, bvFTD and CN.

      Methods

      We included 33 probable bvFTD patients (Neary criteria, low likelihood of AD pathophysiology based on CSF amyloid-beta1-42 >550ng/l), 82 probable AD patients (National Institute on Aging-Alzheimer’s Association workgroups core criteria, high likelihood of AD pathophysiology based on CSF tau/amyloid-beta1-42 >.52ng/l), and 26 CN (22 subjects with subjective memory complaints and 4 healthy controls, CSF amyloid-beta1-42 >550ng/l). Glucose metabolism was assessed using [18F]FDG-PET. Parametric images of standardized uptake value ratios (SUVr) using cerebellar grey matter as reference tissue were generated. First, we defined PCC hypometabolism based on the Receiver Operating Characteristic (ROC) separating AD from CN, to assess the relative prevalence of PCC hypometabolism in AD, bvFTD and CN. Second, we explored relationships between PCC metabolism and demographics (age, sex, and education), Mini-Mental State Examination (MMSE), neuropsychological tests, CSF biomarkers (amyloid-beta1-42, tau and phosphorylated tau), and APOE genotype within diagnostic groups using linear regression analyses or ANOVA where appropriate.

      Results

      Mean age was 63±7 (AD), 65±8 (bvFTD) and 61±8 (CN) years old. PCC [18F]FDG SUVr was CN > bvFTD > AD. Based on optimal discrimination of AD and CN, the PCC [18F]FDG SUVr cut-off was set at 1.052, resulting in PCC hypometabolism in 78% (AD), 33% (bvFTD), and 23% (CN). PCC [18F]FDG SUVr was associated with age in CN (beta±SE: -.007±.002, p=.002) and bvFTD (beta±SE: -.007±.003, p=.011), not in AD.

      Conclusions

      PCC hypometabolism was present in 33% of bvFTD and 23% of CN. PCC [18F]FDG SUVr was associated with age in bvFTD and CN. In the context of diagnostic work-up of dementia, it is important to realize that PCC vulnerability is not restricted to AD, but could be present in bvFTD and normal aging.