Anatomical Substrate and Scalp EEG Markers are Correlated in Subjects with Cognitive Impairment and Alzheimer's Disease

Abstract

Dementia is a syndromic diagnosis, encompassing various stage of severity and different anatomo-physiological substrates. The hippocampus is one of the first and most affected brain regions affected by both Alzheimer's disease (AD) and mild cognitive impairment (MCI). Moreover, chronic cerebrovascular disease (CVD) is one of the major risk factor for developing dementia. Recent studies have demonstrated different relationship between the anatomical substrate and scalp electroencephalography (EEG) markers. Indeed, modifications of EEG rhythmicity is not proportional to the hippocampal atrophy, whereas changes in EEG activity are directly proportional to the load of subcortical CVD. The computation of the EEG spectral power and the analysis of the functional coupling of brain areas, through linear coherence, are two of the most known processing methods in EEG research. Two specific EEG markers, theta/gamma and alpha3/alpha2 frequency ratio have been reliable associated to the atrophy of amygdalo-hippocampal complex. Moreover, theta/gamma ratio has been related to MCI conversion in dementia and alpha3/alpha2 ratio has been specifically related to MCI conversion in AD. The functional coupling of brain areas is also modulated by hippocampal atrophy. In the MCI subjects, hippocampal atrophy is linked to an increase of interhemispheric coherence seen on frontal and temporal regions whereas subcortical CVD is linked to a decrease of coherence in fronto-parietal regions. In the present study the most significant results of recent studies on correlation between scalp EEG, cognitive decline, and anatomical substrate have been reviewed, with particular attention to the relationships between EEG changes and hippocampal atrophy. The following review is not intended to provide a comprehensive summary of the literature. Rather it identifies and discusses selected studies that are designed to find the specific correlation between scalp EEG markers and anatomo-pathological substrate. The principal aim is to propose a plausible neurophysiological theoretical model of the cognitive decline as mirrored by both structural and functional tools of research.

Keywords: Alzheimer's disease; EEG markers; alpha rhythm; mild cognitive impairment; theta rhythm.

Figure 1

Statistical ANOVA interaction among Group factors, and relative band powers (delta, theta, alpha1, alpha2, alpha3), on the full scalp region. The groups are based on mean and standard deviations in a normal elderly sample. Group 1, no hippocampal atrophy; Group 2, mild hippocampal atrophy; Group 3, moderate hippocampal atrophy; Group 4 severe hippocampal atrophy. Post hoc results are indicated in the diagram (see Moretti et al., 2007).

Figure 2

Schematic diagrams of the possible structures involved in the thalamo-cortical arrhythmia in groups without hippocampal atrophy and with mild hippocampal atrophy. The EEG changes in the mild hippocampal-atrophy group are probably due to a prevalent depolarizing effect of the brainstem cholinergic system on the thalamus, because of initial neuronal loss in the cholinergic basal forebrain. On the other hand, the initial hippocampal atrophy gives rise to an increase in the theta activity, but it is not able to trigger thalamo-cortical synchronization activity (see Moretti et al., 2007). mPFC, medial prefrontal cortex; thalamus RE, nucleus reuniens of the thalamus. Black thick arrow, excitatory activity; blue arrow, synchronization effect; red arrow, desynchronization effect.

Figure 3

Schematic diagrams of the possible structures involved in the thalamo-cortical arrhythmia in groups with moderate and severe hippocampal atrophy. The progressive hippocampal atrophy, as in the moderate hippocampal-atrophy group, triggers thalamo-cortical synchronization, with increase in the alpha power band. This is likely to happen because the return pathway from the cortex to the hippocampus is not direct, but it mainly relays to the midline and mediodorsal thalamic nuclei. The decrease in the values for the alpha frequency markers in the moderate hippocampal-atrophy group also suggests a greater hyperpolarization state of thalamo-cortical pathways. As the hippocampal atrophy progresses, like in the group with severe hippocampal atrophy, the thalamo-cortical activity is sustained not by cortical activation, but by the prevailing cholinergic desynchronizing activity of the brainstem, with a decrease in the relative alpha power but the highest values for the alpha indices (see Moretti et al., 2007). mPFC, medial prefrontal cortex; thalamus RE, nucleus reuniens of the thalamus. Black thick arrow, excitatory activity; blue arrow, synchronization effect; red arrow, desynchronization effect.

Figure 4

ANOVA results of theta/gamma and alpha3/alpha2 relative power ratio. In the graph post hoc results are shown.

Figure 5

ANOVA statistical results. On the left part of the figure statistical ANOVA results of the first session analysis (normal-MCI whole group); on the right part of the figure statistical ANOVA results of the second session analysis (MCI subgroups-matched normal old group; (MCI-CHOL, MCI with greater cholinergic damage, MCI-CVD, MCI with greater cerebrovascular damage; MCI-HIPP, MCI with greater hippocampal atrophy; see Moretti et al., 2008a).

Figure 6

Theoretical, neurophysiological and clinical effects of damage to the cholinergic pathways (see Moretti et al., 2008b).

Figure 7

Correlation between alpha3/alpha2 power rhythm ratio and volumes of hippocampal subregions in AD patients.