Mapping distributed sources of cortical rhythms in mild Alzheimer's disease. A multicentric EEG study

Abstract

The study aimed at mapping (i) the distributed electroencephalographic (EEG) sources specific for mild Alzheimer's disease (AD) compared to vascular dementia (VaD) or normal elderly people (Nold) and (ii) the distributed EEG sources sensitive to the mild AD at different stages of severity. Resting EEG (10–20 electrode montage) was recorded from 48 mild AD, 20 VaD, and 38 Nold subjects. Both AD and VaD patients had 24–17 of mini mental state examination (MMSE). EEG rhythms were delta (2–4 Hz), theta (4–8 Hz), alpha 1 (8–10.5 Hz), alpha 2 (10.5–13 Hz), beta 1 (13–20 Hz), and beta 2 (20–30 Hz). Cortical EEG sources were modeled by low resolution brain electromagnetic tomography (LORETA). Regarding issue i, there was a decline of central, parietal, temporal, and limbic alpha 1 (low alpha) sources specific for mild AD group with respect to Nold and VaD groups. Furthermore, occipital alpha 1 sources showed a strong decline in mild AD compared to VaD group. Finally, distributed theta sources were largely abnormal in VaD but not in mild AD group. Regarding issue ii, there was a lower power of occipital alpha 1 sources in mild AD subgroup having more severe disease. Compared to previous field studies, this was the first investigation that illustrated the power spectrum profiles at the level of cortical (macroregions) EEG sources in mild AD patients having different severity of the disease with respect to VaD and normal subjects. Future studies should evaluate the clinical usefulness of this approach in early differential diagnosis, disease staging, and therapy monitoring.

Introduction

Quantitative analysis of electroencephalographic (EEG) rhythms in resting subjects is a low-cost and useful neurophysiological approach to the study of normal aging and dementia Brenner et al., 1986, Coben et al., 1983, Coben et al., 1990, Giaquinto and Nolfe, 1986, Gueguen et al., 1991, Leuchter et al., 1993, Maurer and Dierks, 1992, Schreiter-Gasser et al., 1993, Stigsby et al., 1981, Szelies et al., 1992, Visser et al., 1985. Indeed, scalp EEG rhythms are affected by Alzheimer's disease (AD; Besthorn et al., 1997, Chiaramonti et al., 1997, Schreiter-Gasser et al., 1994). Compared to normal subjects, Alzheimer's disease (AD) patients present an increase of delta (about 0.5–4 Hz) and theta (about 4–8 Hz) mean power along with a decrease of alpha (about 8–13 Hz) and beta (about 13–30 Hz) mean power.

EEG rhythms are also sensitive to the severity of dementia. Delta and/or theta rhythms do increase even in the earlier stages of AD (Schreiter-Gasser et al., 1994) and seem to predict disease progression Ihl et al., 1996, Nobili et al., 1999. In normal subjects, magnitude of alpha rhythm is maximal in scalp occipital areas. While alpha rhythm still peaks in posterior scalp areas in mild AD patients, it is either equally distributed over the scalp or localizes more anteriorly with disease progression Claus et al., 1998, Ihl et al., 1993, Ihl et al., 1996. Similarly, maximum beta rhythm is located more anteriorly in AD patients, as a function of disease severity.

The abnormality of EEG rhythms in dementia is associated with altered regional cerebral blood flow (rCBF)/metabolism and cognitive function, as revealed by positron emission tomography (PET), single photon emission computerized tomography (SPECT), and neuropsychological testing Celsis et al., 1990, Joannesson et al., 1977, Julin et al., 1995, Ihl et al., 1989, Passero et al., 1995, Rodriguez et al., 1998, Rodriguez et al., 1999a, Sheridan et al., 1988, Sloan et al., 1995, Szelies et al., 1992. Indeed, an inverse correlation between delta/theta rhythms and rCBF is observed in parieto-temporal regions of AD patients Buchan et al., 1997, Kwa et al., 1993, Passero et al., 1995, Stigsby et al., 1981. With few exceptions (Mueller et al., 1997), rCBF directly correlates with alpha rhythms in these regions Buchan et al., 1997, Rodriguez et al., 1998. Furthermore, EEG rhythms and rCBF seem to correlate with severity of AD as expressed by mini mental state evaluation (MMSE; Rodriguez et al., 1998). In vascular dementia (VaD), decrease of occipital alpha power positively correlates with the occipital and temporo-parietal glucose metabolism at rest (Szelies et al., 1999). In contrast, delta/theta power in VaD inversely correlates with the glucose metabolism, possibly reflecting subcortical lesions and cortical deafferentation (Szelies et al., 1999).

Cortical sources of scalp EEG rhythms have been successfully evaluated in AD patients by single dipole sources deeply located into a spherical brain model (Dierks et al., 1993). Single dipole sources of alpha or beta rhythms are located more anteriorly as a function of AD severity. Such “anteriorization” of the dipole source is observed in AD patients not only with respect to normal subjects but also with respect to subjects with mild cognitive impairment Dierks et al., 1993, Huang et al., 2000. Notably, the location of the dipole sources correlates with the reduction of rCBF in antero-posterior and latero-lateral brain axes (Dierks et al., 2000).

From a physiological point of view, EEG rhythms reflect the opening–closure (“gating function”) of bidirectional connections among several cortical and subcortical (i.e. brainstem, thalamus) structures Hari et al., 1997, Nunez, 1995, Pfurtscheller and Neuper, 1994, Pfurtscheller and Lopes da Silva, 1999. Therefore, a single dipole source indicates the “center of gravity” of the distributed cortical sources generating the EEG rhythms. An alternative approach for the modeling of these sources is called low resolution brain electromagnetic tomography (LORETA-KEY; Pascual-Marqui and Michel, 1994, Pascual-Marqui et al., 1999), which uses thousands of dipole sources within a 3-D brain model coregistered into Talairach space (Talairach and Tournoux, 1988). LORETA is a functional imaging technique belonging to a family of procedures (Valdés et al., 1998) in which the cortex can be modeled as a collection of volume elements (voxels) in the digitized Talairach atlas. LORETA accommodates neuroanatomical constraints and finds the linear inverse solutions that maximize only the synchronization of strength between neighboring neuronal populations (Pascual-Marqui et al., 2002). This roughly corresponds to the 3-D distribution of electric neuronal activity that has the maximum similarity (i.e. the maximum synchronization) in terms of orientation and strength among neighboring neuronal populations. In a previous review paper, it has been shown that LORETA was quite efficient when compared to other linear inverse algorithms like minimum norm solution, weighted minimum norm solution or weighted resolution optimization (Pascual-Marqui, 1999). Independent validation of LORETA solutions has been provided by recent studies Phillips et al., 2002, Yao and He, 2001.

LORETA solutions in resting AD patients have shown a significant coupling between rCBF pattern and distributed EEG sources of alpha and beta rhythms (Dierks et al., 2000). Moreover, LORETA provides a better spatial resolution with the advantage of 3-D representation of the cerebral activity compared to the classical scalp quantitative EEG and, with respect to the dipole modeling of cortical sources, no a priori decision of the dipole position is required in LORETA estimation.

The present multicentric study was aimed at defining (i) the distributed EEG sources specific for mild AD compared to VaD or normal aging (Nold) and (ii) the distributed EEG sources sensitive to the mild AD progression. For these aims, resting EEG was recorded from a large group of mild AD, VaD, and normal elderly (Nold) subjects. As normal elderly people, we mean people that could be age-matched with the demented patients and did not present any cognitive impairment or any potential condition altering the EEG profile. Frequency bands of EEG rhythms ranged from delta to beta bands. Cortical sources of these rhythms were modeled by LORETA solutions in macro-cortical regions. On the whole, the present study extended the previous field evidence in terms of fine topographical localization of the EEG sources (LORETA solutions into Talairach space) at specific frequency bands able to characterize the mild AD group.