作者
Davangere P. Devanand,Arthur Mikhno,Gregory H. Pelton,Katrina Cuasay,J.S. Dileep Kumar,Matthias H. Tabert,Gnanavalli Pradhaban,Neil Upton,Robert Lai,Vincenzo Libri,Roger N. Gunn,J. John Mann,Ramin V. Parsey
摘要
PET amyloid uptake of [11C]-6-OH-BTA-1 (Pittsburgh compound B or PIB) has been examined as an indicator of Alzheimer's disease (AD) pathology. While PIB uptake distinguishes AD from controls, some uncertainty remains about the utility of PIB uptake in separating MCI patients from AD and controls. Outpatients in a Memory Disorders Center with probable AD (n=16) and MCI (n=20) were recruited; age and sex-matched healthy controls (n=16) were also studied. A brief neuropsychological test battery was administered and PIB PET was done. MRI was done for coregistration and region of interest determination. Binding potential (BPnd) from the Logan graphical method using cerebellar gray matter as reference region was obtained from PET images coregistered to their corresponding MRI scans. Individual BPnd estimates were measured for four specific anatomical regions (cingulate, prefrontal cortex, hippocampus, parahippocampal gyrus). Across the entire sample, there were no age, sex, or education effects on BPnd in any ROI. Prefrontal cortex BPnd showed inverse correlations with MMSE (r=-0.53, p < .001), Selective Reminding Test (SRT) immediate recall (r=-0.50, p = .004) and SRT delayed recall scores (r=-0.60, p < .001), and a positive correlation with ADAS-cog scores (r=0.54, p < .001). Cingulate BPnd showed similar correlations with cognitive scores. Parahippocampal gyrus BPnd showed less significant correlations and hippocampus BPnd showed no significant correlations with cognitive scores. In ANOVAs on each ROI with subject group (control, MCI, AD) as the between subject factor, prefrontal cortex BPnd (F=18.0, p < .0001), cingulate BPnd (F=10.2, p < .0001) and parahippocampal gyrus were significant (F=5.7, p =006), but not hippocampus (F=.26, p=.77). In ANCOVA using a similar model with age and MMSE or the SRT measures included as covariates, prefrontal cortex and cingulate remained significant (p's < 0.02 to 0.05). Amnestic MCI patients (n=15) showed significantly greater prefrontal and cingulate BPnd compared to non-amnestic MCI patients (n=5). [11C]-6-OH-BTA-1 (PIB) PET BPnd showed strong associations with cognitive performance. Its ability to distinguish AD, MCI and controls after controlling for age and cognitive performance underlines its potential use in differential diagnosis and prediction of conversion from MCI to AD.