• Rik Ossenkoppele, Wiesje M van der Flier, Sander C J Verfaillie, Hugo Vrenken, Adriaan Versteeg, Ronald A van Schijndel, Sietske A Sikkes, Jos Twisk, Sofie M Adriaanse, Marissa D Zwan, Ronald Boellaard, Albert D Windhorst, Frederik Barkhof, Philip Scheltens, Adriaan A Lammertsma, and Bart N M van Berckel.
• From the Department of Radiology & Nuclear Medicine (R.O., S.C.J.V., H.V., A.V., R.A.v.S., S.M.A., M.D.Z., R.B., A.D.W., F.B., A.A.L., B.N.M.v.B.), Alzheimer Center & Department of Neurology, Neuroscience Campus Amsterdam (R.O., W.M.v.d.F., S.A.S., S.M.A., M.D.Z., P.S.), and Department of Epidemiology & Biostatistics (W.M.v.d.F., S.A.S., J.T.), VU University Medical Center, Amsterdam, the Netherlands. r.ossenkoppele@vumc.nl.
• Neurology. 2014 May 20;82(20):1768-75.

ObjectiveTo assess how amyloid deposition, glucose hypometabolism, and cerebral atrophy affect neuropsychological performance in patients with Alzheimer disease (AD) dementia, patients with mild cognitive impairment (MCI), and controls over time.MethodsA total of 41 patients with AD dementia, 28 patients with MCI, and 19 controls underwent [(11)C]-Pittsburgh compound B ((11)C-PiB) and [(18)F]-2-fluoro-2-deoxy-d-glucose ((18)F-FDG)-PET and MRI scans at baseline. We extracted global binding potential for (11)C-PiB, the number of abnormal voxels for (18)F-FDG, and gray matter volumes using SIENAX for MRI as measures of amyloid, hypometabolism, and atrophy. In addition, repeat neuropsychological testing was performed, including memory, attention, language, and executive tasks (mean follow-up 2.2 ± 0.7 years). Cross-sectional and longitudinal relationships between imaging markers and cognition were assessed using linear mixed models, including terms for the imaging markers, time, sex, age, diagnosis, and interactions for imaging marker × time and imaging marker × time × diagnosis.ResultsLinear mixed models showed that baseline hypometabolism and atrophy were associated with poorer baseline performance on attention and executive functions (p < 0.05), whereas amyloid was not related to baseline cognition. Hypometabolism and amyloid were strongly associated with longitudinal decline in essentially all cognitive domains (pinteraction < 0.05), whereas atrophy was related specifically to future decline in Mini-Mental State Examination and memory (pinteraction < 0.05).ConclusionGlucose hypometabolism and brain atrophy were associated with concurrent cognitive function, whereas brain amyloid was not. Amyloid deposition and glucose hypometabolism were predictors for decline of a wide variety of cognitive functions, while brain atrophy specifically predicted memory deterioration.© 2014 American Academy of Neurology.

36 keywords...

hide…