P1-248: decreased resting-state functional connectivity strength in amnestic mci-to-ad converters and non-converters

Can Sheng,Mingrui Xia,Zhongjie Hu,Ying Han
DOI: https://doi.org/10.1016/j.jalz.2014.05.487
2014-01-01
Alzheimer s & Dementia
Abstract:AMCI patients have been proved to be a high risk group of developing to AD [1]. Combining the advances in resting-state fMRI (rs-fMRI) technique and developments in graph theoretical methods, it allows to study the abnormality of the topological properties of the functional brain connectome non-invasively in aMCI patients. Recent studies based on rs-fMRI data have demonstrated aMCI-related aberrant nodal centrality [i.e., functional connectivity strength (FCS)] in several brain functional systems, such as the default mode and the visual networks [2-4]. However, it still remains largely unknown whether such abnormalities differ from aMCI-to-AD converters (aMCI-P) to aMCI-stable (aMCI-S). Here, we employed rs-fMRI to construct the voxel-wise functional brain networks, and further assess the differences on FCS between aMCI patients and NCs, and particularly between the aMCI-P and aMCI-S. Resting-state fMRI data of 32 aMCI patients and 28 well matched NCs were collected. During 1∼2 years follow up, 13 of the 32 aMCI progressed to probable AD (i.e. aMCI-P) and others remained stable (i.e. aMCI-S). The baseline data were used here, and after preprocessing [slice timing, head motion correction, spatial normalization, spatial smoothing (6mm), temporal filtering (0.01∼0.1 Hz) and regression of six head-motion profiles, global signal, CSF signal and white-matter signal], we obtained voxel-wise FCS map for each subject. For a given voxel, FCS was computed by summing the Fisher r-to-z transformed Pearson correlation coefficients (r>0.2) of all the voxels to the voxel [5,6,7]. To examine between-group differences on FCS, voxel-wise linear models were performed with age, gender and education removed. Multiple comparisons were corrected with Monte Carlo simulations at P2106 mm3. Compared with NC, aMCI patients showed decreased FCS in the bilateral precuneus. Compared with aMCI-S, aMCI-P showed decreased FCS in the left superior temporal gyrus, the bilateral insula and bilateral sensorimotor and visual cortices. We demonstrated decreased voxel-wise FCS in the visual and sensorimotor cortices, and the salience network in aMCI-to-AD converters as compared to non-converters. According to these preliminary findings, aMCI-P groups might be recognized in their early stages, thus could be crucial for studying imaging-based biomarkers for AD prediction and diagnosis. Abnormal FCS in aMCI compared with NC Abnormal FCS in aMCI-P compared with aMCI-S
What problem does this paper attempt to address?