Functional connectivity in patients with early Alzheimers disease, MCI and healthy controls assessed by fMRI and EEG

Alzheimer's disease (AD) is a neurodegenerative disorder selectively involving cerebral neuronal networks facilitating higher cognitive functions. Coherence measured by electroencephalography (EEG) is a sensitive marker for functional connectivity in the brain, whereas functional magnetic resonance imaging (fMRI) allows the detection of activation patterns. The aim of this study was to correlate fMRI activation patterns and EEG-coherence in patients with AD, mild cognitive impairment (MCI) and age-matched healthy controls (HC). The aim of this study was to correlate fMRI activation patterns and EEG-coherence in patients AD, MCI and HC, thus investigating differences of functional connectivity between the groups.

There were statistically significant differences between AD and MCI patients for theta band coherences (6,5–8Hz) between anterior cingulate gyrus (ACC) and left temporal gyrus (GT); between AD and HC for theta band coherence between ACC and right GT, between ACC and left hippocampus, and between ACC and right parietal gyrus (GP). Furthermore, MCI-subjects showed reduced coherence compared with healthy controls between ACC and left frontal superior gyrus within delta, theta and alpha1-bands (p<0,05). Theta coherence was statistically significantly lower in patients with MCI compared with controls between anterior and posterior cingulate gyrus, ACC and left/right temporal gyrus, posterior cingulate gyrus and superior frontal gyrus (MCI<controls, p<0,01), and between right and left temporal gyri (MCI<controls, p<0,05). Activation patterns (fMRI, eLORETA) were different between groups with less activation in the groups of patients, especially within the ACC.

Coherence as a measure of functional connectivity is an important research tool in psychiatry. By using exact low resolution brain electromagnetic tomography (eLORETA) distributed cortical networks can be localized and investigated. The cortical networks in AD and MCI as identified in this study are in agreement with the literature. The assessment of voxel-based connectivity between cortical regions may help identifying local and network abnormalities in neurodegenerative disorders. EEG coherence measurements based on eLORETA analyses seem to be a useful approach to investigate connectivity between regions of interest as defined by fMRI and resting EEG activation patterns. EEG coherence could serve a sensitive indicator for cognitive decline at a very early stage of neurodegenerative disorders.

Fig.1: Significant connectivity differences of ROIs between Alzheimer patients (AD) and healthy controls.θ Theta band (6.5–8.0Hz); * p<0.05; ** p<0.01

Fig.2: Significant connectivity differences of ROIs between Alzheimer patients (AD) and healthy controls.θ Theta band (6.5–8.0Hz); α1 Alpha-1-band (8.0–10.0Hz)(*) p<0.10; * p<0.05; ** p<0.01

Fig.3: Significant connectivity differences of ROIs between Alzheimer patients (AD) and MCI patients.θ Theta band (6.5–8.0Hz), * p<0.05

Fig.4: Comparison of healthy elderly subjects vs. MCI patients. fMRI activations (p<0.001) and EEG-sLORETA current source density activations. Correspondence in the frontal, but not in posterior and deeper regions.

Fig.5: Comparison of healthy elderly subjects vs. AD patients. fMRI activations (p<0.001) and EEG-sLORETA current source density activations. Correspondence in the frontal, but not in posterior and deeper regions.

Fig.6: Comparison of MCI vs. AD patients. fMRI activations (p<0.001) and EEG-sLORETA current source density activations. Correspondence in posterior regions.