P2-209 Regional specificity of brain in metabolite changes in ad

P2-209 Regional specificity of brain in metabolite changes in ad

Poster Session P2: Diagnosis and Disease Progression - Neuroimaging prominent parietal/medial temporal signals in AD. FDG-PET for FTD patients showed ...

343KB Sizes 1 Downloads 14 Views

Poster Session P2: Diagnosis and Disease Progression - Neuroimaging prominent parietal/medial temporal signals in AD. FDG-PET for FTD patients showed decreased frontal/temporal glucose metabolism, correlating well with increased [F-18]FDDNP binding (see figure). Conclusion: [F18]FDDNP-PET has permitted, for the first time, the visualization of tauopathies in living patients. The FFD patients studied presented prominent frontal and temporal signals compared with controls, suggesting potential utility of [F-18]FDDNP-PET in differentiating FTD from the parietal and temporal signals in AD, a clinical distinction important for identifying patients that might benefit from cholinesterase inhibitor therapies. Further, the ability of [F-18]FDDNP to label tanopathies in vivo provides a tool for monitoring the effect of therapies designed to prevent or eliminate NFT accumulation.



CLINICAL PROGRESSION OF DEMENTIA AND PREDICTIVE VALUE OF HMPAO-SPECT: FOLLOW-UP STUDY OF 97 PATIENTS FROM A MULTIDISCIPLINARY MEMORY CLINIC BASED IN NEUROLOGY

Peter HCgh*, Ann-Sofie Teller, Steen G. Hasselbalch, Gunhild Waldemar.

Memory Disorders Research Unit, Dept of Neurology, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark. Contact e-mail: [email protected] Background: Previous studies of HMPAO-SPECT in dementia have shown that SPECT may be an important diagnostic adjunct in the identification of - and discrimination between various etiologies to dementia. The rate of clinical progression in dementia may vary significantly, and prognostic markers have generally not been identified. Only few previous studies have investigated the predictive value of SPECT in discriminating between progressive and non-progressive patients with dementia. Objective(s): The objective of the study was to examine the role of HMPAO-SPECT as a predictor for disease progression in dementia. Methods: Ninety-seven patients with dementia (DSM IV) were identified in the memory clinic database. The inclusion criteria were: all consecutively referred patients with a SPECT-HMPAO study, a MMSE score higher than 20 and age over 60 years at time of admission. The mean age of the patients was 75.7 years, mean MMSE score 23.4. The mean follow-up period was 2.1 years. SPECT images were visually assessed by two experienced raters in consensus, blinded to the clinical status of the patients. Region of interest (ROI) based parametric analysis of regional cerebral blood flow (rCBF) was also performed. Disease progression was registered as a drop of 2 points or more in the MMSE score. Results: Seventy-seven % of patients with globally abnormal SPECT progressed significantly during follow-up, compared to 50% of patients with globally normal SPECT. SPECT images visually assessed as globally abnormal or with reduced rCBF in the left temporal region on admission clearly separated progressive from non-progressive patients (i0 < 0.05). Conclusions: Decreased global or left temporal rCBF on HMPAO-SPECT may be a useful prognostic finding in patients meeting clinical crireria for dementia.



CORPUS CALLOSUM ATROPHY IN A MIXED ELDERLY POPULATION

$289

(WMH) and degree of cognitive impairment, in a mixed population of non-demented elderly subjects with WMH. Methods: Sixty-two subjects (34/32 F,qVl, mean age = 74.7 years, range 65-84 years) participated in the Danish cohort of a European longitudinal study on "Leukoaraiosis And DISability" (LADIS). Subjects were recruited from an outpatient memory clinic and from a population of healthy elderly and had no or minor disability (ADL). For all subjects high-resolution 3D sagittal MPRAGE of the whole head was acquired on a 1,5 T Vision Siemens scanner. CC cross-sectional area was manually delineated on the midsagittal slice. Subsequently, the CC cross-section was automatically subdivided into 5 regions (rostrum and genu, rostral body, midbody, isthmus and splenium) in an matero-posterior direction. The severity of W M H were categorized according to Fazekas scale in mild, moderate or severe WMH. Cognitive impairment was assessed with the Mini-Mental State Examination (MMSE). Results: Mean MMSE was 28.2 (range 22-30). The mean total CC area was 518.6 m m 2 (range 293.8779.5; ICV-normalized values). Smaller CC area correlated significantly (Spearman) with lower MMSE (CCtotal p < 0.039; Rostral body p < 0.001; Midbody p < 0.006; Isthmus p < 0.041) and increased W M H load (Rostral body p < 0.006; Midbody p < 0.032). We also found a significantly higher WMH load (p < 0.002) and lower MMSE score (p < 0.045) with increasing age. However when we corrected for age the only significant correlations were those between MMSE and Rostral body (F 5.447; p = 0.023), and between MMSE and Midbody (F 2.034; p = 0.047). Conclusion: Loss of CC tissue correlates significantly with impaired cognitive function and with increased W M H load as well as age.



REGIONAL SPECIFICITY OF BRAIN METABOLITE CHANGES IN AD

Xiaoping Zhu .1 , Norbert Schuff 1, Karl Young ~, Brian Soher 2, Bruce L. Miller 3 , William J. Jagust 4, Michael M. Weiner I . 1VA Medical

Center/UCSE San Francisco, CA, USA; 2University of Miami, School of Medicine, Miami, FL, USA; 3 UCSE San Francisco, CA, USA; 4 UC Davis, Sacramento, CA, USA. Contact e-mail: [email protected] Background: Previously, this laboratory demonstrated with MR spectroscopic imaging (MRSI) that AD was associated with reduced N-Acetyl Aspartate (NAA, a neuronal marker) in parietal cortex gray matter, but not in white matter or in frontal lobe. Others have reported increased myoInositol (mI, a glial marker) in AD using short TE single voxel MRS, but the anatomical pattern of mI changes in AD has not been described. Objective: To determine the regional distribution of NAA and mI in AD and normal aging using MRSI. Methods: Sixteen patients with AD and 21 cognitive normal elderly (CN) were studied using short echo time (TE = 25ms) multislice MRSI, positioned to cover parietal and frontal lobes. Results: Figure 1 shows metabolite images of mI (left) and NAA (right) from a 77 years old normal elderly subject. AD patients had less (p > 0.05) NAA and more (p < 0.05) mI in parietal cortical gray matter (GM), and a 16% increase of m//NAA in parietal lobe GM for AD compared to CN (p < 0.005). In contrast there were no significant changes of NAA or mI in white matter or frontal lobe GM. ml/NAA in AD was proportional to the amount of GM.

MI

NAA

Charlotte Ryberg* ~, Egill Ros~up 1, Ellen Garde 1, Anne Metre Hejl 2, Anne Mette Leffers 1, Gunhild Waldemar 2. On behalf of the LADIS group,

University of Florence, Florence, Italy; 1Danish Research Center Jor Magnetic Resonance, Copenhagen University Hospital, Hvidovre, Denmark; 2Memory Disorders Research Unit, Dept. of Neurology, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark. Contact e-mail: ryberg @magnet, drcmr,dk Background: Corpus Callosum (CC) pathways communicates cognitive and learned information between the two hemispheres. Effects on regional callosal structure has been reported in Alzheimer's Disease and multiple sclerosis and suggest that CC tissue loss may be a marker for early phase degenerative disease. To evaluate this hypothesis we investigated the relationship between CC size and the severity of white matter hyperintensities

Fig. I. Short TE metabolite images.

Conclusions: In AD, mI (a glial marker) is increased in parietal lobe GM, which is the same region in which NAA (a neuronal marker) is reduced, suggesting that neuronal injury and reactive gliosis are linked in AD.

$290 ~

Poster Session P2: Diagnosis and Disease Progression - Neuroimaging R E G I O N A L MR BRAIN V O L U M E S IN THE PREDICTION OF COGNITIVE DECLINE IN AD

Wes S. Houston* 1, Christine Fennema-Notestine1'2, Nikki R. Home 3, Mark W. Bondi L2, Sarah L. Archibald 1'2, Terry L. Jernigan 1'2, David P. Salmon 4. 1VA San Diego Healthcare System, San Diego, CA, USA;

e University of California, San Diego, Department of Psychiatry, San Diego, CA, USA; SSan Diego State University~University of California, San Diego Joint Doctoral Program in Clinical Psychology. San Diego, CA, USA. 4 University of California, San Diego, Department of Neurosciences, San Diego, CA, USA. Contact e-mail: [email protected] Background: The characteristic pattern of cognitive deterioration in Alzheimer's disease (AD) is a decline in episodic memory followed by a progressive worsening of other cognitive abilities including executive, language and visuomotor skills. Consistent and predictable neuropathological changes have also been observed in AD. In its earliest stages, AD neuropathology originates in the mesial temporal areas including the hippocampus and entorhinal cortices followed by a spreading to cortical association areas (Braak & Braak, 1998). Objective: In an effort to predict annual rate of cognitive decline, we examined regional structural magnetic resonance (MR) brain volumes in mild-to-moderate AD. Given the pattern of progressive cortical atrophy, we expected cortical brain changes in mild-to-moderate AD patients to be more highly correlated with language and visuomotor abilities than with memory ahihties, given its primacy and prominence in the earliest stages of the disease. Methods: Thirty-one patients with mild-to-moderate AD were administered a measure of general cognitive abilities, Dementia Rating Scale (DRS, Mattis, 1973) both at the time of receiving a structural MR scan and approximately one year later. One-year change in general cognitive abilities was calculated by subtracting the baseline DRS total score from the DRS total the following year. Multiple regression and simple correlations were calculated between brain volumes of 8 cortical regions (i.e., left and right frontal, temporal, parietal and occipital areas). Results: The mean DRS decline was 8.29 points (SD = 12.37). A correlation matrix indicated that left and right occipital gray (r = 0.39; p = 0.02, r = 0.35; p = 0.03, respectively) and left parietal gray volumes (1" = 0.37; p = 0.02) had the strongest linear relation with a one-year decline in cognitive functioning. Multiple regression analysis revealed that only the left occipital cortical area was retained in the final model (R2 = 0.15; p = 0.03). Conclusion: Cognitive decline in this sample of mild-to-moderate AD patients was best predicted by atrophy in regions other than the temporal lobes, consistent with the known pattern of neuropathologic spreading in this stage of AD. Support: NIH/NIA P50AG05131, AG12674, & AG04085, DVA Medical Research Service



NEURAL CORRELATES OF IMPAIRED SEMANTIC MEMORY IN ALZHEIMER'S DISEASE

Murray Grossman*, James Gee, Peachie Moore, Melissa Work, Phyllis Koenig. University of Pennsylvania, Philadelphia, PA, USA.

Contact e-mail: [email protected] Background: Our model of semantic memory includes feature knowledge for categories, and processes such as categorization that use this knowledge to make semantic decisions. AD patients have impaired semantic memory that may involve degraded semantic feature knowledge and impaired categorization processes. Objective: Determine the basis for impaired semantic memory in AD. Methods: 123 mild-to-moderate AD patients performed a semantic category membership judgment task. Feature representations were assessed by examining judgments for natural vs manufactured stimuli. Semantic processes were examined by assessing judgments of foils: Errors discriminating between category members and related non-members (e.g. for the category VEGETABLE, a related foil is "apple") suggest a deficit using rule-based processes to distinguish between partially-overlapping stimuli; Errors discriminating between category members and unrelated non-members (e.g. for the category VEGETABLE, an unrelated foil is "chair") suggest a deficit with similarity-based processes used to distinguish between non-overlapping stimuli. Regression analyses related semantic performance to cortical volume in a subgroup of 12 AD

patients using voxel-based morphometry analyses of spatially normalized and segmented structural MtlI images obtained at 1.5T (voxels 0.9 x 0.9 x 1.3 mm). Results: AD patients had significantly impaired overall category membership judgments. Individual patient analyses revealed a significant impairment in 24% of AD patients, according to z-scores relative to 25 matched controls. Judgments of the natural category were significantly less accurate than the manufactured category. Error analyses revealed a deficit in the rule-based categorization process needed to discriminate between category exemplars and partially-overlapping foils that are not category members. This categorization impairment was more evident in the natural than the manufactured category. MRI correlations associated overall semantic judgment performance with dorsolateral prefrontal and inferior temporal cortex of the left hemisphere in AD. This overlapped with the MR/correlations for the rule-based categorization processing impairment, but not with correlations for the degraded natural category of knowledge. Conclusions: Category-specific semantic deficits in AD may be due in part to impaired categorization processes. Consistent with fMRI activation studies in healthy adults and AD patients, MRI correlations relate impaired semantic categorization to disease in a left hemisphere network involving multimodal frontal and temporal association cortex.



AGE-RELATED CORTICAL GREY MATTER REDUCTIONS IN NONDEMENTED DOWN'S SYNDROME ADULTS DETERMINED BY MAGNETIC RESONANCE IMAGING W I T H VOXEL-BASED MORPHOMETRY

Stefan J. Teipel* 1, Gene E. Alexander 2, Marc B. Schapiro 3, Hans-Jtkgen M~511er1, Stanley I, Rapoport 4, Harald Hampel I . 1Alzheimer

Memorial Center, Department of Psychiatry, Ludwig-Maximilian University, Munich, Germany; SNeuroimage Analysis Laboratory, Department of Psychology, Arizona State University, Tempe, AZ, USA; 3Division of Neurology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA; 4Brain Physiology and Metabolism Section, National Institute on Aging, National Institutes of Health, Bethesda, MD, USA. Contact e-mail: Stefan. Teipel@reed. uni-muenchen.de Background: Aging in Down's syndrome (DS) is accompanied by amyloid and neurofibriUary pathology whose distribution replicates pathological features of Alzheimer's disease (AD). Based on these findings, DS has been proposed as a model to study the predementia stages of AD. Objective: To investigate age effects on regional cortical grey matter in predementia stages of DS. Methods: We applied the automated and objective technique of voxel-based morphometry, implemented in SPM 99 (Wellcome Department of Imaging Neuroscience, London, UK), to the analysis of structural MRI from 27 nondemented DS adults (mean age 41.1 years, 15 women). Results: Regional grey matter volume was decreased with advancing age in bilateral parietal cortex (mainly the precuneus and inferior parietal lobule), bilateral frontal cortex with left side predominance (mainly middle frontal gyms), left occipital cortex (mainly lingual cortex), fight precentral and left postcentral gyms, left transverse temporal gyms, and right parahippocampal gyms. The reductions were unrelated to gender, intracranial volume or general cognitive function. Grey matter volume was relatively preserved in subcortical nuclei, periventricular regions, the basal surface of the brain (bilateral orbitofrontal and anterior temporal), and the anterior cingnlate gyms, Conclusions: Our findings suggest grey matter reductions in allocortex and association neocortex in the predementia stage of DS. The most likely substrate of these changes is alterations or loss of allocortical and neocortical neurons due to AD type pathology.