Premium
P3–116: Significant longitudinal change in neostriatal volume in premanifest and manifest Huntington's disease: The IMAGE‐HD study
Author(s) -
Wilkes Fiona,
Looi Jeffrey,
Walterfang Mark,
Velakoulis Dennis,
Stout Julie,
Churchyard Andrew,
Chua Phyllis,
Egan Gary,
GeorgiouKaristianis Nellie
Publication year - 2013
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2013.05.1187
Subject(s) - huntington's disease , atrophy , intraclass correlation , medicine , analysis of variance , biomarker , nuclear medicine , cardiology , psychology , disease , clinical psychology , biochemistry , chemistry , psychometrics
Background:Many studies have been trying to evaluate AD as a disconnection syndrome, however very few of them have combined multi-modal information to asses whether the damage in the withe matter (WM) tracts disrupt the organization of the functional networks. Here we integrate the analysis of magnetoencephalographic (MEG) data withWM integrity quantification from diffusion weighted imaging (DWI), to achieve this aim. Methods: The sample comprised 143 subjects divided into 52 healthy-elderly (HE), 37 amnestic MCI (aMCI) and 54 multi-domain MCI (mMCI). DWIs were acquired on a GE 1.5T MRI scanner. 25 diffusion directions with b1⁄41000mm/s 2. Fractional anisotropy (FA) images were obtained fromDTI and introduced in the TBSS [1] pipeline, yielding one FA skeleton image per subject in MNI space.MEG recordings were consisted of 3 minutes closed eyes resting state in a Elekta Neuromag system. Non-artifacted trials (4s) were filtered in 5 frequency bands. Phase Locking Value [2] was computed between sensors and averaged across trials. Two graph metrics (normalized clustering and normalized characteristic path length [3]) were obtained for the PLV networks.We studied dependency between MEG and FA through a General Linear Model (GLM). We built 20 GLM, (2 metrics x 2 groups x 5 frequencies). The results were thresholded at p-value1⁄40.01. Results: TBSS. HE subjects showed higher FA in comparison with the two groups of MCI subjects in several brain regions including the medial temporal lobe and prefrontal regions. GLM. For the aMCI we obtained significant correlations in the cingulum of the hippocampus (L/R) for all frequency bands with the normalized clustering except for gamma. For the mMCI we observed significant correlations in superior and inferior longitudinal fasciculus (L/R). Conclusions: Due to the lower WM integrity showed in MCI patients and the correlation found between MEG and FA, our results indicate that the lower the FA in certain brain regions the higher the random the network, as is the case for the MCI patients. These results point out that these white matter regions may be responsible of a crucial functioning of the whole functional networks, and that when impaired, e.g. due to demyelination, dementia process may be highly influenced.