Assessment of mild dementia with amyloid and dopamine terminal positron emission tomography.

We assessed the relationship between consensus clinical diagnostic classification and neurochemical positron emission tomography imaging of striatal vesicular monoamine transporters and cerebrocortical deposition of aβ-amyloid in mild dementia. Seventy-five subjects with mild dementia (Mini-Mental State Examination score≥18) underwent a conventional clinical evaluation followed by 11C-dihydrotetrabenazine positron emission tomography imaging of striatal vesicular monoamine transporters and 11C-Pittsburgh compound-B positron emission tomography imaging of cerebrocortical aβ-amyloid deposition. Clinical classifications were assigned by consensus of an experienced clinician panel. Neuroimaging classifications were assigned as Alzheimer's disease, frontotemporal dementia or dementia with Lewy bodies on the basis of the combined 11C-dihydrotetrabenazine and 11C-Pittsburgh compound-B results. Thirty-six subjects were classified clinically as having Alzheimer's disease, 25 as having frontotemporal dementia and 14 as having dementia with Lewy bodies. Forty-seven subjects were classified by positron emission tomography neuroimaging as having Alzheimer's disease, 15 as having dementia with Lewy bodies and 13 as having frontotemporal dementia. There was only moderate agreement between clinical consensus and neuroimaging classifications across all dementia subtypes, with discordant classifications in ∼35% of subjects (Cohen's κ=0.39). Discordant classifications were least frequent in clinical consensus Alzheimer's disease (17%), followed by dementia with Lewy bodies (29%) and were most common in frontotemporal dementia (64%). Accurate clinical classification of mild neurodegenerative dementia is challenging. Though additional post-mortem correlations are required, positron emission tomography imaging likely distinguishes subgroups corresponding to neurochemically defined pathologies. Use of these positron emission tomography imaging methods may augment clinical classifications and allow selection of more uniform subject groups in disease-modifying therapeutic trials and other prospective research involving subjects in the early stages of dementia.