Yong Soo Shim

Abnormal integrity of corticocortical tracts in mild cognitive impairment: a diffusion tensor imaging study.

Mild cognitive impairment (MCI) has been defined as a transitional state between normal aging and Alzheimer disease. Diffusion tensor imaging (DTI) can estimate the microstructural integrity of white matter tracts in MCI. We evaluated the microstructural changes in the white matter of MCI patients with DTI. We recruited 11 patients with MCI who met the working criteria of MCI and 11 elderly normal controls. The mean diffusivity (MD) and fractional anisotropy (FA) were measured in 26 regions of the brain with the regions of interest (ROIs) method. In the MCI patients, FA values were significantly decreased in the hippocampus, the posterior limb of the internal capsule, the splenium of corpus callosum, and in the superior and inferior longitudinal fasciculus compared to the control group. MD values were significantly increased in the hippocampus, the anterior and posterior limbs of the internal capsules, the splenium of the corpus callosum, the right frontal lobe, and in the superior and the inferior longitudinal fasciculus. Microstructural changes of several corticocortical tracts associated with cognition were identified in patients with MCI. FA and MD values of DTI may be used as novel biomarkers for the evaluation of neurodegenerative disorders.

Hippocampal and entorhinal structures in subjective memory impairment: a combined MRI volumetric and DTI study

BACKGROUND Subjective memory impairment (SMI) is common among older adults. Increasing evidence suggests that SMI is a risk factor for future cognitive decline, as well as for mild cognitive impairment and dementia. Medial temporal lobe structures, including the hippocampus and entorhinal cortex, are affected in the early stages of Alzheimers disease. The current study examined the gray matter (GM) volume and microstructural changes of hippocampal and entorhinal regions in individuals with SMI, compared with elderly control participants without memory complaints. METHODS A total of 45 participants (mean age: 70.31 ± 6.07 years) took part in the study, including 18 participants with SMI and 27 elderly controls without memory complaints. We compared the GM volume and diffusion tensor imaging (DTI) measures in the hippocampal and entorhinal regions between SMI and control groups. RESULTS Individuals with SMI had lower entorhinal cortical volumes than control participants, but no differences in hippocampal volume were found between groups. In addition, SMI patients exhibited DTI changes (lower fractional anisotropy (FA) and higher mean diffusivity in SMI) in the hippocampal body and entorhinal white matter compared with controls. Combining entorhinal cortical volume and FA in the hippocampal body improved the accuracy of classification between SMI and control groups. CONCLUSIONS These findings suggest that the entorhinal region exhibits macrostructural as well as microstructural changes in individuals with SMI, whereas the hippocampus exhibits only microstructural alterations.

P3-018: Difference of amnestic pattern in MCI patients with and without subcortical vascular disease: Preliminary study

Background: Mild cognitive impairment (MCI) has the clinical and etiological heterogeneity. Therefore it is difficult to predict into which type of dementia MCI progresses. Whereas Alzheimer’s disease (AD) shows prominently the defect of memory encoding and storage, the impaired retrieval is associated with the subcortical dementia such as vascular dementia (VaD). Objectives: We investigated the patterns of memory deficit in Seoul Verbal Learning Test (SVLT) between vascular MCI (vMCI) and non-vascular MCI (nvMCI) by the Erkinjuntti’s imaging criteria. Methods: Among 70 MCI patients by the Petersen’s criteria visiting the Memory Clinic of St. Mary’s Hospital, 49 MCI patients (including 24 nvMCIs and 25 vMCIs) were evaluated. They were performed by the Korean mini-mental state examination (K-MMSE), the geriatric depression scale (GDS), the Korean instrumental activity of daily living (K-IADL) and Hachinski’s ischemic scale (HIS), and by the SVLT for memory test. MRI was performed for excluding the other causes such as tumor and trauma and for subdivision into vMCI and nvMCI by Erkinjuntti’s imaging criteria. The differences of the SVLT parameters were compared between vMCI and nvMCI. Results: Between the two groups, the distribution of age, sex, education level, K-MMSE score, K-IADL score, and GDS score were not different. HIS score was higher in vMCI (5.00 2.62) than that of nvMCI (1.72 2.11). Pattern analysis of the SVLT performance revealed that the scores of the true positive recognition, the discrimination index and the combination index (total recall plus discrimination index) were higher in vMCI (p 0.002, 0.02 and 0.05, respectively). Although not statistically significant, nvMCI patients recalled fewer words than vMCI patients. vMCI patients showed better recognition than nvMCI patients. Conclusions: This finding suggests that vMCI, compared to nvMCI has the retrieval dysfunction with little disturbance of recall (encoding and storage), which is similar to the memory dysfunction of VaD. When the further clinical studies are added, the Erkinjuntti’s imaging criteria may be useful for the subclassification of MCI patients as the predictor of the progression into the AD or VaD.

DEVELOPMENT OF A PROGRAM FOR QUICK BRAIN VOLUMETRY

ambiguous (“Hard”) and unambiguous (“Easy”) words for animacy. Modulatory capacity was measured by contrasting hard vs. easy in functional activation. Age, amyloid and crystallized ability were entered in a multiple regression model for each of the 8 brain regions in the frontoparietal cognitive control network to assess whether amyloid and crystallized knowledge accounted for variance in modulation beyond age. Results:We first replicated our previous findings that older adults had decreased modulatory capacity in the frontoparietal network. After controlling for age, amyloid accumulation was related to declined modulatory capacity in parietal and right prefrontal but not left prefrontal cortex, and such depletive effects were found in middle-aged (40-60 years) but not older (60-69 years) adults. After controlling for age and continuous accumulation of amyloid, better crystallized knowledge predicted higher modulation in the frontoparietal network. Because we were particularly interested in whether the facilitating role of crystallized knowledge in modulatory capacity was related to amyloid positivity status, in a second analysis, we conducted multiple regressions with age and crystallized knowledge predicting brain modulation for amyloid negative and positive individuals, respectively.We found that better crystallized knowledge predicted higher modulatory capacity in prefrontal regions and left angular gyrus for amyloid negative individuals but only mildly (p1⁄4 0.057) in medial superior frontal gyrus for amyloid positive individuals. Conclusions: Our findings suggested that brain modulatory capability is impaired with age and amyloid. Crystallized knowledge protects modulatory capacity, which seems to have stronger effects for amyloid negative individuals, compared to amyloid positive individuals.

Behavioral effects of memantine add-on therapy to rivastigmine patches in AD

caregivers, mixed ANOVA with two sub-scores resulting from the factor analysis revealed the deterioration in combination treatment group was focused on the first factor named ‘Physically Non-Aggressive Behavior’. Conclusions: This study showed that there were no significant benefits on agitation in mild to moderate AD patients receiving combination therapy with rivastigmine patch and memantine. To date, clinical trial support is greater for memantine use in combination with an AChE inhibitor in some aspect, while more data are needed to evaluate its efficacy as combination therapy in AD.

O2-04-07: Treatment with donepezil in Korean Alzheimer patients with and without cerebrovascular factors: One-year follow-up study

randomized to receive either Vitamin E 800 IU vitamin C 200 mg alpha lipoic acid 600 mg/day; coenzyme Q 2400 mg per day (800mg 3 times per day), or placebo, for 16 weeks. CSF and plasma were obtained at baseline and after 16 weeks of treatment. Physical examination, blood tests and adverse event reports were used to monitor safety. The MMSE and ADCS-ADL were administered at baseline and week 16. CSF levels of A-beta42, tau and P-tau181 were measured by multiplex assays. F2-isoprostanes are measured by LC-MS methods. Clinical and CSF data were compared between groups using ANCOVA models. Results: 78 subjects with AD (50% women) were randomized. Study medications were well-tolerated, without differences in adverse events between groups. 74 subjects completed treatment and had successful serial lumbar punctures. At baseline, means (SD) were: age 72.8 (9.1), MMSE 22.9 (3.7); these did not differ between treatment arms. CSF levels of A-beta42, tau and P-tau-181 did not differ between treatment arms at baseline. Changes in CSF levels of A-beta42, tau and P-tau181 from baseline to week 16 did not differ between either treatment arm and placebo. Measurements of F2-isoprostanes are in progress. Conclusions: Treatment with a cytosolic antioxidant combination or with CoQ did not alter CSF biomarkers related to A-beta or tau in AD. The supplements, at the doses used, did not appear to influence AD pathologic processes in the brain sufficiently to alter CSF biomarkers. Supported by NIA grant AG10483.

IC-P1-055: Regional volumetric comparison of gray matter in patients with subcortical ischemic vascular dementia to Alzheimer's disease

Background: Subcortical ischemic vascular dementia(SIVD) is a homogenous and common subtype of vascular dementia(VaD), and clinically it shows the gradual progression similar to Alzheimer’s disease(AD), contrary to the usual VaD. Moreover, the findings like the atrophy of brain and the dilatation of ventricle can be also shown on MRI of the patients with SIVD, and some patients with SIVD have the pathology of AD. Therefore there have been opinions that SIVD is actually but the mixed type with AD not the true VaD. We compared the regional volumes of the gray matters between AD and SIVD with the hypothesis that the brain atrophy of the patients with SIVD shown on MRI is a relative result from the atrophy of the white matter, not from the atrophy of the gray matter. Methods: Twelve AD patients meeting the criteria of the NINCDS-ADRDA, and 13 SIVD patients meeting the probable VaD criteria of the NINDS-AIREN and Erkinjuntti’s imaging criteria of SIVD were included in this study. Eleven controls were also included. On coronal T1W MRI with the SPGR, the volumes of the bilateral frontal, temporal, parietal and occipital cortex, and hippocampus and entorhinal cortex were obtained. After the correction with the ratio to the intracranial volume, the regional volumes were compared among AD, SIVD and the controls by ANOVA with multiple comparisons. Results: Whole brain volumes were not different among three groups. The ventricular volumes of the patients with AD and SIVD were smaller than that of the controls(p 0.001). But, there was no difference between AD and SIVD. For the gray matters, the volumes of all regions were not different between the controls and SIVD patients. The volumes of the bilateral hippocampus(right; p 0.009, left; p 0.008) and entorhinal cortex(all p 0.002) were the lowest in the AD patients. Conclusions: There can be also observed that the brain atrophy and ventricular dilatation in SIVD as well as in AD. However, these findings of SIVD are not from the cortical atrophy such as the hippocampus and entorhinal cortex, which is responsible for AD, and might be from the lesions of the white matters. P2-079 CORTICAL AMYLOID DEPOSITION AND PARAHIPPOCAMPAL ATROPHY IN ALZHEIMER’S DISEASE AND MILD COGNITIVE IMPAIRMENT

IC-P-005: White matter changes associated with dementia development in patients with subcortical ischemic white matter lesions: A diffusion tensor image study

observations, the present study characterized default mode activity using analysis of low frequency functional correlations between brain regions (Biswal et al., 1995 MRM; Fox et al., 2005 PNAS). BOLD-contrast fMRI imaging was used and the correlation coefficient between regions was the dependent measure. Methods: Functional correlations were explored in a large sample of 38 young adults (age 22.4 y; 18 M), 55 nondemented older adults (76.5 y; 18 M) and 17 older adults (76.9 y; 12 M) with a clinical diagnosis of AD in accordance with validated criteria using standard assessment protocols for both clinical and neuropsychological measures. Data were from Lustig et al. (2003 PNAS). To minimize anatomical biases, seed regions used for correlation analyses were defined in a separate data set of an equal number of young, nondemented older, and AD participants (n 24 total). Many participants also underwent diffusion tensor imaging (DTI) at nearby times to assess white matter integrity. Conclusions: Results revealed that functional correlations were dramatically reduced between anterior and posterior regions comprising the default mode in nondemented older adults compared to the young. Considerably more modest reductions in correlations were noted between the nondemented and AD older adult groups. We also examined relationships between functional correlation measures, performance on neuropsychological tests, and DTI in the same group of individuals, as they are useful to understand the interaction between potentially related changes in aging. Our results are consistent with the hypothesis that aging is associated with disconnection of distributed brain networks that show coordinated activity in young adults. Future analyses will explore whether our observed functional disconnection in aging is related to findings of white-matter degradation (e.g., O’Sullivan et al., 2000 Neurology; Head et al., 2004 CC) and also whether functional disconnection is prominent in nondemented aging in the absence of AD.

IC-P-005

observations, the present study characterized default mode activity using analysis of low frequency functional correlations between brain regions (Biswal et al., 1995 MRM; Fox et al., 2005 PNAS). BOLD-contrast fMRI imaging was used and the correlation coefficient between regions was the dependent measure. Methods: Functional correlations were explored in a large sample of 38 young adults (age 22.4 y; 18 M), 55 nondemented older adults (76.5 y; 18 M) and 17 older adults (76.9 y; 12 M) with a clinical diagnosis of AD in accordance with validated criteria using standard assessment protocols for both clinical and neuropsychological measures. Data were from Lustig et al. (2003 PNAS). To minimize anatomical biases, seed regions used for correlation analyses were defined in a separate data set of an equal number of young, nondemented older, and AD participants (n 24 total). Many participants also underwent diffusion tensor imaging (DTI) at nearby times to assess white matter integrity. Conclusions: Results revealed that functional correlations were dramatically reduced between anterior and posterior regions comprising the default mode in nondemented older adults compared to the young. Considerably more modest reductions in correlations were noted between the nondemented and AD older adult groups. We also examined relationships between functional correlation measures, performance on neuropsychological tests, and DTI in the same group of individuals, as they are useful to understand the interaction between potentially related changes in aging. Our results are consistent with the hypothesis that aging is associated with disconnection of distributed brain networks that show coordinated activity in young adults. Future analyses will explore whether our observed functional disconnection in aging is related to findings of white-matter degradation (e.g., O’Sullivan et al., 2000 Neurology; Head et al., 2004 CC) and also whether functional disconnection is prominent in nondemented aging in the absence of AD.

IC-P-003

Background: Mild cognitive impairment (MCI) refers to a transitional state between normal aging and dementia, and it is clinically heterogeneous. Although most imaging studies of MCI have focused on gray matter alterations, many MCI patients have subcortical vascular disease on magnetic resonance imaging (MRI) scan. However it is not known whether these findings are associated with the cognitive dysfunction. Diffusion tensor imaging (DTI) detects the microstructural alterations in white matter by measuring the directionality of molecular diffusion. Objective: We investigated the alterations of white matter as well as hippocampus in MCI patients according to the presence or the absence of subcortical vascular disease by using DTI, and compared the differences between two groups. Methods: Forty consecutive patients with memory complaints, at least one neuropsychological memory test below 1.5 standard deviation of the normal for age and education, and maintained activities of daily living, were included. 21 patients with MCI had no ischemic lesions and 19 patients were found to have subcortical vascular changes, by the Erkinjuntti’s imaging criteria. For 21 non-vascular MCI (nvMVI), 19 vascular MCI (vMCI) and 17 controls, mean diffusivity (MD) and fractional anisotropy (FA) were measured in the bilateral temporal, frontal, parietal and occipital white matter regions as well as in the bilateral hippocampi, sentrum semiovale and the corpus callosum (genu and splenum), and the differences were compared by analysis of variance and multiple comparison. Results: All patients with MCI, both vascular and non-vascular, showed that decreased FA and increased MD values in the other regions except occipital white matter, compared to the controls. For the FA, both types of MCI patients had decreased FA values in the corpus callosum, frontal and temporal regions, compared to the controls. In the parietal areas and centrum semiovale, vMCI patients had more decreased FA values than nvMCI patients and the controls. In the hippocampus, FA values were lowest in the nvMCI patients. FA values of vMCI patients were also significantly lower than the controls. The findings of region-specific MD increases were similar to those of FA. Conclusions: The DTI can be a useful tool to quantify MCI pathology in vivo, and to evaluate the alterations of intracerebral microstructure.