Research ReportHippocampus–precuneus functional connectivity as an early sign of Alzheimer's disease: A preliminary study using structural and functional magnetic resonance imaging data
Highlights
► Structural atrophy (SA) with Alzheimer's disease (AD) was in the hippocampus (HP). ► Functional connectivity (FC) between the HP and precuneus was reduced from AD. ► FC between the post central gyrus and precuneus was significantly lower from AD. ► Effect size of FC levels was significantly greater than effect size of SA levels. ► Aberrant FC than SA seems to be a valuable biomarker for early detection of AD.
Introduction
Structural abnormalities of the human brain, as measured via non-invasive magnetic resonance imaging (MRI), have been implicated in various neurodegenerative illnesses, including Alzheimer's disease (AD) (Frisoni et al., 2010, Dickerson and Sperling, 2008). The characteristic traits estimated from the structural MRI (sMRI) data include atrophies of regional volumes and reduced gray matter (GM) intensities, especially in the medial temporal lobe (MTL) (Frisoni et al., 2010, Risacher et al., 2009). Frisoni et al. (2010) conducted a volume analysis (VA) using sMRI data and then used the results as a clinical assessment tool for suspected AD patients. In this study, the authors suggested that the atrophies within the MTL area, including the hippocampus (HP), were a valid biomarker to represent AD status for the assessment of individual patients. Furthermore, abnormally altered functional connectivity (FC) patterns, as measured by blood oxygenation level-dependent (BOLD) functional MRI (fMRI), have also been reported to be a potential biomarker for the diagnosis of AD (Greicius et al., 2004, Bai et al., 2009, Wang et al., 2007).
A growing body of evidence suggests that both structural atrophy and functional alterations are implicated in the early stages of AD (Gili et al., 2011, Sorg et al., 2007). For example, Gili et al. (2011) estimated GM intensities using sMRI data with a voxel-based morphometry (VBM) method and estimated FC patterns using resting-state fMRI data of mild cognitive impairment (MCI) and AD groups with a group independent component analysis (ICA) method. The greater degree of aberrant FC patterns observed in both the MCI and AD groups were in the posterior cingulate cortex (PCC) and medial prefrontal cortex relative to the rest of the brain regions (Gili et al., 2011). On the other hand, the observed structural atrophies in the MCI group were significantly limited, with only marginal changes observed in the PCC and precuneus relative to the AD group (Gili et al., 2011). Thus, it is possible that aberrant FC patterns may appear earlier than GM atrophies during the progression of AD pathology. However, there have been limited attempts to explicitly identify potentially crucial information about structural atrophies and levels of aberrant FC patterns in the early stage of AD using non-invasive MRI.
The aim of the present study was to provide preliminary evidence on the link between the degree of structural atrophy and the level of aberrant FC patterns that are associated with an early stage of AD. More specifically, we hypothesized that the functional aberrance estimated with FC levels would precede the structural abnormality estimated with structural atrophy. Furthermore, the spatial patterns (SPs) with aberrant FC levels can be identified from the seed brain regions with structural atrophy and possibly even from brain regions without structural atrophy as evaluated with an effect size. As a result, we will suggest a brain region that would represent the potentially early sign of AD status, so that the FC level in this region can be adopted as a potential biomarker to assist a diagnosis of the AD status. To this end, brain regions were automatically parcellated from sMRI data using the VA method, and cortical atrophies were identified at the voxel level using the VBM method. Then, the average BOLD time series (TS) were calculated from the seed regions-of-interest (ROIs) within brain regions that either displayed atrophies or did not display atrophies. These BOLD TS were then used as references for the subsequent FC analyses. Next, FC patterns were calculated with voxel-wise partial correlation coefficients (CCs) using the reference BOLD TS selected from the two seed ROIs in each hemisphere and the BOLD TS from the voxels in the remaining brain areas. For the group-level statistical analyses, potentially significant changes in the FC patterns between healthy control (HC) and AD groups were compared using two-sample t-tests. Finally, an effect size of severity associated with AD was compared between observed structural abnormalities and aberrant FC levels.
Section snippets
Results
Fig. 1 shows the marginal levels of structural atrophies in terms of volume sizes observed in the left HP (p<0.03) and right HP (p<0.04) in the AD group relative to the HC group. The normalized regional volumes of the bilateral PCC and precuneus as well as the PoCG were not significantly different between the two groups (p>0.05). The intra-cranial volume (ICV) values between the two groups were also not significantly different (AD: 1.27×103 cm3±0.18×103 cm3; HC: 1.27×103±0.19×103 cm3; p=0.96).
Summary
In this study, we presented preliminary evidence that there is a close link between abnormal structural atrophies and corresponding aberrant FC patterns that are both associated with AD status. The structural atrophies were observed in the bilateral HP, especially in the rostral portion of the left HP, as revealed with the VBM method. Compared to the marginal levels of structural atrophies that have been implicated in AD, the FC levels between the HP and rest of the brain were substantially
Overview
Fig. 6 illustrates an overall flow diagram of the procedures used in the present study. First, the seed regions with and without atrophy were identified by determining regional volume segmentations with VA and voxel-based identifications of GM intensities with VBM. Second, after the brain regions with and without atrophy were identified as seed regions, FC analyses were performed between the seed regions and the remaining whole brain areas. Third, the characteristic traits of GM intensities and
Funding sources
This work was supported by the WCU (World Class University) program through the National Research Foundation (NRF) of Korea and was funded by the Ministry of Education, Science and Technology (R31-10008) and the Basic Science Research Program, NRF grant of Korea (2012–0002342).
Acknowledgments
The authors would like to thank Dr. Seung-Schik Yoo and Mr. Dong-Youl Kim for their valuable discussions and logistic support in preparation of this manuscript.
References (37)
- et al.
Abnormal functional connectivity of hippocampus during episodic memory retrieval processing network in amnestic mild cognitive impairment
Biol. Psychiatry
(2009) - et al.
Cortical surface-based analysis I. Segmentation and surface reconstruction
Neuroimage
(1999) - et al.
Functional abnormalities of the medial temporal lobe memory system in mild cognitive impairment and Alzheimer's disease: insights from functional MRI studies
Neuropsychologia
(2008) - et al.
Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain
Neuron
(2002) - et al.
The precuneus/posterior cingulate cortex plays a pivotal role in the default mode network: evidence from a partial correlation network analysis
Neuroimage
(2008) - et al.
Multisubject fMRI studies and conjunction analyses
Neuroimage
(1999) - et al.
Dynamic causal modelling
Neuroimage
(2003) - et al.
Voxel-based assessment of gray and white matter volumes in Alzheimer's disease
Neurosci. Lett.
(2010) - et al.
Group inference of default-mode networks from functional magnetic resonance imaging (fMRI) data: comparison of random- and mixed-effect group statistics
Int. J. Imaging Syst. Technol.
(2012) - et al.
Investigation of spectrally-coherent resting-state networks using nonnegative matrix factorization for fMRI data
Int. J. Imaging Syst. Technol.
(2011)