HA

CCLESAa

Sb

c

d

Re

f

Aim[DtNlitT3ps(Ahd41sntaanfpm

*FMEhASSeMnr

Neuroscience 145 (2007) 942–954

0d

OMOCYSTEINE AND ELECTROENCEPHALOGRAPHIC RHYTHMS IN

LZHEIMER DISEASE: A MULTICENTRIC STUDY

tcpEr

Kme

Qrnd1Ww(aeBa(e(2isweE2

WrebfraImG2rsiFot

. BABILONI,a,b,c* P. BOSCO,d R. GHIDONI,b

. DEL PERCIO,a,b R. SQUITTI,c G. BINETTI,b

. BENUSSI,b R. FERRI,d G. FRISONI,b,c B. LANUZZA,d

. CASSETTA,c G. ANELLO,d M. GURZÌ,b

. BARTESAGHI,e R. LIZIO,a M. TOMBINIf

ND P. M. ROSSINIb,c,f

Dipartimento di Fisiologia Umana e Farmacologia, Università deglitudi di Roma “La Sapienza,” P.le Aldo Moro 5, Rome, Italy

IRCCS “Centro S. Giovanni di Dio-F.B.F.,” Brescia, Italy

A. Fa.R., Dip. Neurosci. Osp. FBF, Isola Tiberina, Rome, Italy

Department of Neurology, Oasi Institute for Research on Mentaletardation and Brain Aging, Troina, Italy

Ospedale “Sacra Famiglia” Fatebenefratelli AFaR, Erba (CO), Italy

Clinica Neurologica University “Campus Biomedico,” Rome, Italy

bstract—High plasma concentration of homocysteine is anndependent risk factor for Alzheimer’s disease (AD), due to

icrovascular impairment and consequent neural lossSeshadri S, Beiser A, Selhub J, Jacques PF, Rosenberg IH,’Agostino RB, Wilson PW, Wolf PA (2002) Plasma homocys-

eine as a risk factor for dementia and Alzheimer’s disease.Engl J Med 346(7):476–483]. Is high plasma homocysteine

evel related to slow electroencephalographic (EEG) rhythmsn awake resting AD subjects, as a reflection of known rela-ionships between cortical neural loss and these rhythms?o test this hypothesis, we enrolled 34 mild AD patients and4 subjects with mild cognitive impairment (MCI). Enrolledeople were then subdivided into four sub-groups of 17 per-ons: MCI and AD subjects with low homocysteine levelMCI� and AD�, homocysteine level <11 �mol/l); MCI andD subjects with high homocysteine level (MCI� and AD�,omocysteine level >11 �mol/l). Resting eyes-closed EEGata were recorded. EEG rhythms of interest were delta (2–Hz), theta (4–8 Hz), alpha 1 (8–10.5 Hz), alpha 2 (10.5–

3 Hz), beta 1 (13–20 Hz), and beta 2 (20–30 Hz). EEG corticalources were estimated by low-resolution brain electromag-etic tomography (LORETA). Results showed that delta (fron-al and temporal), theta (central, frontal, parietal, occipital,nd temporal), alpha 1 (parietal, occipital, and temporal), andlpha 2 (parietal and occipital) sources were stronger in mag-itude in AD� than AD� group. Instead, no difference wasound between MCI� and MCI� groups. In conclusion, highlasma homocysteine level is related to unselective incre-ent of cortical delta, theta, and alpha rhythms in mild AD,

Correspondence to: C. Babiloni, Dipartimento di Fisiologia Umana earmacologia, Università degli Studi di Roma “La Sapienza,” P.le Aldooro 5, Rome, Italy. Tel: �39-06-49910989; fax: �39-06-49910917.-mail address: claudio.babiloni@uniroma1.it (C. Babiloni); URL:ttp://hreeg. ifu.uniroma1.it/.bbreviations: AD, Alzheimer disease; CDR, Clinical Dementia Ratingcale; EEG, electroencephalographic; GDS, Geriatric Depressioncale; IAF, individual alpha frequency; LORETA, low-resolution brainlectromagnetic tomography; MCI, mild cognitive impairment; MMSE,ini Mental State Examination; Nold, normal elderly subjects showing

Ao evidence of cognitive impairment; PD, Parkinson disease; ROI,egion of interest.

306-4522/07$30.00�0.00 © 2007 IBRO. Published by Elsevier Ltd. All rights reseroi:10.1016/j.neuroscience.2006.12.065

942

hus unveiling possible relationships among that level, mi-rovascular concomitants of advanced neurodegenerativerocesses, and synchronization mechanisms generatingEG rhythms. © 2007 IBRO. Published by Elsevier Ltd. Allights reserved.

ey words: homocysteine, mild cognitive impairment, Alzhei-er’s disease, electroencephalography, low resolution brain

lectromagnetic tomography.

uantitative analysis of electroencephalographic (EEG)hythms in awake resting subjects is a low-cost and usefuleurophysiological approach to the study of Alzheimerisease (AD) (Maurer and Dierks, 1992; Szelies et al.,992; Leuchter et al., 1993; Schreiter-Gasser et al., 1993).hen compared with normal elderly subjects, AD patientsere characterized by excessive delta (0–4 Hz) and theta

4–7 Hz) rhythms, and a significant decrement of posteriorlpha rhythms (8–12 Hz; Dierks et al., 1993, 2000; Huangt al., 2000; Ponomareva et al., 2003; Jeong, 2004;abiloni et al., 2004a). These EEG abnormalities weressociated with altered regional cerebral blood flowrCBF)/metabolism and with global cognitive function asvaluated by Mini Mental State Examination (MMSE)Sloan et al., 1995; Rodriguez et al., 1998, 1999; Jeong,004). Similar EEG abnormalities have been also reported

n subjects with mild cognitive impairment (MCI), a clinicaltate between elderly normal cognition and dementia inhich subjects present memory complaints and objectivevidence of cognitive impairment (Zappoli et al., 1995;lmstahl and Rosen, 1997; Huang et al., 2000; Jelic et al.,000; Babiloni et al., 2006e; Koenig et al., 2005).

Keeping in mind these data, a logical question is,hich is the physiological meaning of abnormal EEG

hythms in AD? It has been reported that, in AD patients,arly pathological processes include loss of cholinergicasal forebrain neurons projecting to hippocampus andronto-parietal areas and that alpha and slower EEGhythms can be modulated by these neurons (Helkala etl., 1996; Holschneider et al., 1999; Mesulam et al., 2004).

n contrast, brainstem cholinergic innervation of the thala-us would be relatively spared in AD (Mash et al., 1985;eula and Mesulam, 1989, 1996, 1999; Mesulam et al.,004). It has been also reported that cholinergic and se-otoninergic neurons are the most important factors re-ponsible for the block of spindles and delta rhythms dur-

ng arousal (Dringenberg, 2000; Dringenberg et al., 2002).inally, delta rhythms increase in AD patients as a functionf atrophy in mesial-temporal, posterior, and/or frontal cor-ical areas (Fernandez et al., 2003; Babiloni et al., 2006a).

bnormal delta rhythms can be also observed for other

ved.

km1douta

ictlteacrr(

ih(eei2mlm(

hpdsaewiaespsbre

wcsAps2io

pb(Moe2auce2a

S

WcsgMsu(a

eots

(pai(HeaBaohcdAddbsoanvnwnbdwuop

B. Claudio et al. / Neuroscience 145 (2007) 942–954 943

inds of brain alterations such as cerebral vascular, trau-atic, tumor lesions (Harmony et al., 1993; Murri et al.,998; de Jongh et al., 2003; Hensel et al., 2004). Theistribution of abnormal delta rhythms depends on the typef lesions; namely, localized delta rhythms increase fornderlying lesions of connection white-matter (“disconnec-ion” delta), whereas diffuse delta increases by thalamicnd troncoencephalic lesions (Gloor et al., 1977).

The physiological meaning of EEG abnormalities in ADs enlightened by studies investigating the gene coding forystatin C, a protein with neuro-protective effects. Haplo-ype B of this gene is related to a decrease in cystatin Cevels of AD patients, which would reduce protection fromoxic insults and stem-cell-mediated regeneration (Vinterst al., 1990; Taupin et al., 2000; Deng et al., 2001; Levy etl., 2001; Sastre et al., 2004). AD and MCI subjects havingystatin C gene haplotype B show an increase of deltahythms and a decrease of alpha rhythms, as a possibleeflection of increasing neurodegenerative processesBabiloni et al., 2006b).

Another biological risk factor for the development of ADs the plasma level of amino acid homocysteine, which isigher in AD and MCI than in normal elderly subjectsMcCaddon et al., 1998, 2003; Kalmijn et al., 1999; Mizrahit al., 2004; Guidi et al., 2006; Quadri et al., 2005; Laut-nschlager et al., 2005). Plasma homocysteine level pos-

tively correlates with amyloid beta protein (Shea et al.,002; Irizarry et al., 2005) and produces its effects byeans of several possible actions. High homocysteine

evel might induce direct neurotoxicity, potentiate gluta-ate neurotoxicity, and might be implied in oxidative stress

Shea et al., 2002).An important study on 1092 subjects without dementia

as indicated a strong, graded association betweenlasma homocysteine level and the risk of AD, possiblyue to its relationship with microvascular impairment (Nil-son et al., 2006) and consequent neural loss (Seshadri etl., 2002). This hypothesis would be supported by previousvidence showing that higher plasma homocysteine levelsere related to more cortical atrophy and vascular lesions

n elderly non-demented and in AD subjects (Gottfries etl., 2001; den Heijer et al., 2003; Scott et al., 2004; de Silvat al., 2005; Folin et al., 2005). It can hence be hypothe-ized that high plasma homocysteine level is related to theower of slow EEG rhythms in awakening relaxed ADubjects, as a reflection of the well-known effects of cere-ral atrophy or lesion on the enhancement of thesehythms (Harmony et al., 1993; Murri et al., 1998; de Jonght al., 2003; Hensel et al., 2004).

In the present study, a preliminary control hypothesisas that EEG sources have amplitude sensitive to theognitive status of Nold (i.e. Nold�normal elderly subjectshowing no evidence of cognitive impairment), MCI, andD subjects; namely, whether delta and alpha sourcesresented different amplitude across Nold, MCI, and ADubjects in line with previous evidence (Babiloni et al.,006b,c,d). Instead, experimental hypothesis stated an

ncrement of abnormalities of EEG rhythms in sub-groups

f AD and MCI subjects having high compared with low b

lasma homocysteine. The EEG sources were estimatedy low-resolution brain electromagnetic tomographyLORETA; Pascual-Marqui and Michel, 1994; Pascual-arqui et al., 1999, 2002), which belongs to a large familyf tomographic procedures based on EEG linear inversestimation (Valdes et al., 1998; Pascual-Marqui et al.,002; Sekihara et al., 2005; Fuchs et al., 2001; Gross etl., 2001). In precedence, LORETA has been successfullysed for the study of aging in the same experimentalondition and setting of the present investigation (Anderert al., 1998a,b, 2003; Dierks et al., 2000; Huang et al.,002; Saletu et al., 2002; Goforth et al., 2004; Babiloni etl., 2004a, 2006a,b,c,d,e).

EXPERIMENTAL PROCEDURES

ubjects and diagnostic criteria

e enrolled 34 AD patients and 34 MCI subjects. They were notonsecutive cases received in the research unit’s ambulatories,ince we had to balance as much as possible age, education, andender across the groups. Furthermore, we could include onlyCI and AD subjects with a certain degree of general cognitive

tatus (see in the following). Part of the individual data sets wassed for previous EEG studies with absolutely other aimsBabiloni et al., 2004a, 2006a,b,c,d). In none of these studies weddressed the specific issue of the present study.

Local institutional ethics committees approved the study. Allxperiments were performed with the informed and overt consentf each participant or caregiver, in line with the Code of Ethics ofhe World Medical Association (Declaration of Helsinki) and thetandards established by the Author’s institutional review board.

Probable AD was diagnosed according to NINCDS-ADRDAMcKhann et al., 1984) and DSM IV criteria. The recruited ADatients underwent general medical, neurological, and psychiatricssessments. Patients were also rated with a number of standard-

zed diagnostic and severity instruments that included MMSEFolstein et al., 1975), Clinical Dementia Rating Scale (CDR;ughes et al., 1982), Geriatric Depression Scale (GDS, Yesavaget al., 1983), Hachinski Ischemic Score (HIS, Rosen et al., 1980),nd Instrumental Activities of Daily Living scale (IADL, Lawton androdie, 1969). Neuroimaging diagnostic procedures (CT or MRI)nd complete laboratory analyses were carried out to excludether causes of progressive or reversible dementias, in order toave a clinically homogenous mild AD patient group. Exclusionriteria included, in particular, any evidence of (i) frontotemporalementia, (ii) vascular dementia, diagnosed according to NINDS-IREN criteria (Roman et al., 1993), (iii) extra-pyramidal syn-romes, (iv) reversible dementias (including pseudodementia ofepression); and (v) Lewy body dementia, according to the criteriay McKeith et al. (1999). The participating AD patients had MMSEcores ranging from 14.4–26.7 (MMSE corrected). The detectionf an eventual vascular component in dementia and MCI wasccounted for based on previous theoretical guidelines from ouretwork, in order to exclude AD and MCI subjects with a majorascular component (Frisoni et al., 1995; Galluzzi et al., 2005). Ofote, benzodiazepines, antidepressants and/or antihypertensivesere suspended for about 24 h before EEG recordings. This didot insure a complete washout of the drug, but the drug conditionecome comparable across the patients. A complete washout ofrugs would have required a too long suspension of the treatmentith high risks for the patients. Of note, about 10% of AD patientsnderwent therapy with acetylcholinesterase inhibitors at the timef EEG recording (donepezil; 5–10 mg per day; rivastigmine 3 mger day).

The inclusion and exclusion criteria for MCI diagnosis were

ased on previous seminal studies (Zaudig, 1992; Devanand et

aitdfeba(aEptucstpdpdetdsr

rtiei((spe

E

PA

swt(eotttmv

Ca

Top2awlhcsaacM

dvt((us

E

E(jwtFOrdE(mwmrc

TN

NAGME

T

NAGME

h

B. Claudio et al. / Neuroscience 145 (2007) 942–954944

l., 1997; Petersen et al., 1995, 1997, 2001) and aimed at select-ng elderly persons with objective cognitive deficits, especially inhe memory domain, who did not meet criteria for a diagnosis ofementia or AD. The inclusion criteria for MCI subjects were theollowing: (i) objective memory impairment at neuropsychologicalvaluation, as defined by performance �1.5 standard deviationelow the mean value of age- and education-matched controls fortest battery including Busckhe-Fuld and Memory Rey tests;

ii) normal activities of daily living, as documented by the historynd evidence of independent living; and (iii) CDR score of 0.5.xclusion criteria for MCI were: (i) mild AD, as diagnosed by therocedures described above (ii) evidence of concomitant demen-ia such as frontotemporal, vascular dementia (special care wassed, given the putative effects of plasma homocysteine on mi-rocirculation), reversible dementias (including pseudo-depres-ive dementia), fluctuations in cognitive performance, and/or fea-ures of mixed dementias; (iii) evidence of concomitant extra-yramidal symptoms; (iv) clinical and indirect evidence ofepression as revealed by GDS scores lower than 14; (v) othersychiatric diseases, epilepsy, drug addiction, alcohol depen-ence, and use of psychoactive drugs including acetylcholinest-rase inhibitors or other drugs enhancing brain cognitive func-ions; and (vi) current or previous systemic diseases (includingiabetes mellitus) or traumatic brain injuries. The participating MCIubjects had MMSE scores ranging from 23.7–29 (MMSE cor-ected).

A control group of healthy elderly subjects (Nold) was alsoecruited (mostly among the patients’ spouses), in order to ascer-ain preliminarily that the selected AD and MCI subjects presentedndeed the typical EEG rhythmicity changes associated with dis-ase and cognitive impairment. All Nold subjects underwent phys-

cal and neurological examinations as well as cognitive screeningincluding MMSE). Subjects affected by chronic systemic illnessesi.e. diabetes mellitus or organ failure) were excluded, as wereubjects receiving psychoactive drugs. Subjects with a history ofresent or previous neurological or psychiatric disease were alsoxcluded. All Nold subjects had a GDS score lower than 14.

valuation of plasma homocysteine level

lasma total homocysteine levels were measured in all MCI andD subjects according to standard procedures. In brief, all plasma

able 1. Demographic and neuropsychological data of interest ofold, MCI, and mild AD subjects

Nold MCI AD

34 34 34ge (y) 69.2�1.6 S.E. 68.4�1.3 S.E. 74.9�1.1 S.E.ender (F/M) 22/12 22/12 22/12MSE 28.4�0.2 S.E. 26.6�0.3 S.E. 20.6�0.6 S.E.ducation (y) 6.1�0.3 S.E. 5.9�0.5 S.E. 4.8�0.4 S.E.

Values are means�standard error.

able 2. Demographic and neuropsychological data of MCI and AD s

MCI� MC

17 17ge (y) 67.4�2.2 S.E. 69.ender (F/M) 13/4 9/8MSE 26.5�0.4 S.E. 26.ducation (y) 6.1�0.9 S.E. 5.8

Values are means�standard error. MCI and AD subjects with lo

omocysteine level (MCI� and AD�).

pecimens were stored at or below �20 °C. Homocysteine levelsere determined with the use of high-performance liquid chroma-

ography with fluorometric detection as previously describedAraki and Sako, 1987; Vester and Rasmussen, 1991; Bartesaghit al., 2001). Of note, we were unable to recruit a sufficient numberf Nold subjects for the determination of plasma total homocys-eine levels, since many of them refused the blood sample extrac-ion. This made the analysis of the relationship between plasmaotal homocysteine levels and the characteristics of EEG rhyth-icity in Nold subjects compared with MCI and AD subjects not

iable, an important issue for future research.

omposition of the experimental groups of ADnd MCI subjects

he AD and MCI subjects were then subdivided in four sub-groupsf 17 persons based on a criterion roughly referring to normativelasma homocysteine values in elderly subjects (Anello et al.,004): MCI and AD subjects with low homocysteine level (MCI�nd AD�, homocysteine level �11 �mol/l); MCI and AD subjectsith high homocysteine level (MCI� and AD�, homocysteine

evel �11 �mol/l). The two sub-groups of AD and MCI subjectsad same size and were comparable in terms of demographic andlinical features. Of note, here we were unable to collect biologicalamples from a sufficient number of Nold subjects for the evalu-tion of plasma homocysteine level. This fact did not allow thenalysis of the relationship between homocysteine levels and theharacteristics of EEG rhythmicity in Nold subjects compared withCI and AD subjects. It is an important issue for future research.

Table 1 summarizes the relevant demographic and clinicalata of Nold, MCI and AD participants. Table 2 reports the rele-ant demographic and clinical data of participants subdivided inwo genetic sub-groups: MCI/AD with low homocysteine levelMCI� and AD�) and MCI/AD with high homocysteine levelMCI� and AD�), respectively. Age, education, and gender weresed as covariates in the statistical evaluation of the corticalources of EEG rhythms, to remove possible confounding effects.

EG recordings

EG data were recorded (EB-Neuro (Firenze, Italy) and MicromedMogliano Veneto, Treviso, Italy) EEG machines) in resting sub-ects (eyes-closed) by specialized clinical units. EEG recordingsere performed (0.3–70 Hz bandpass) from 19 electrodes posi-

ioned according to the International 10–20 System (i.e. Fp1, Fp2,7, F3, Fz, F4, F8, T3, C3, Cz, C4, T4, T5, P3, Pz, P4, T6, O1,2). A specific kind of reference electrode was not imposed to all

ecording units of this multi-centric study, given that preliminaryata analysis and LORETA source analysis were carried out afterEG data were re-referenced to a common average reference

i.e. all electrodes were included in the average reference). Toonitor eye movements, electrooculogram (0.3–70 Hz bandpass)as also collected. All data were digitized in continuous recordingode (5 min of EEG; 256 Hz sampling rate). In all subjects, EEG

ecordings were performed at late morning. During all EEG re-ordings, to keep state of vigilance at constant level, an operator

each subdivided in four sub-groups of 17 persons

AD� AD�

17 17. 73.9�1.2 S.E. 75.8�1.8 S.E.

11/6 11/6. 20.6�0.8 S.E. 20.6�0.8 S.E.. 4.6�0.6 S.E. 4.7�0.3 S.E.

cysteine level (MCI� and AD�); MCI and AD subjects with high

ubjects,

I�

6�1.5 S.E

7�0.4 S.E�0.5 S.E

w homo

ca

se1Rprvs

uta(disEim

S

AHdft(cd1hCqw1PF12l

awmopNp9ege

2arE

Cb

LEiu

apadIce

cpsLatnvfvouw2aw

atape(2alpc1tdpt

trwtcftimLsm

Ts

L

FCPTO

B. Claudio et al. / Neuroscience 145 (2007) 942–954 945

ontrolled on-line the EEG traces and verbally alerted the subjectny time there were sign of behavioral drowsiness.

Of note, EEG recordings lasting 5 min allowed the compari-on of the present results with several previous AD studies usingither recording EEG periods no longer than 5 min (Buchan et al.,997; Muller et al., 1997; Pucci et al., 1999; Szelies et al., 1999;odriguez et al., 2002; Babiloni et al., 2004a, 2006c,d,e) or EEGeriods shorter than 1 min (Dierks et al., 1993, 2000). Longeresting EEG recordings in AD patients would have reduced theariability of the data but would have increased the possibility oflowing of EEG oscillations due to reduced vigilance and arousal.

EEG data were analyzed and fragmented off-line in consec-tive epochs of 2 s. EEG epochs with ocular, muscular and otherypes of artifact were preliminary identified by a computerizedutomatic procedure. The EEG epochs including ocular artifactsless than 15% of the total ones) were then corrected by a stan-ard autoregressive method whose performances were validated

n a previous study (Moretti et al., 2003). In brief, this methodubtracted the projection of electrooculogram (EOG) artifact onEG data based on estimated weights (Moretti et al., 2003). Two

ndependent experimenters manually confirmed the EEG seg-ents accepted for further analysis.

pectral analysis of the EEG data

digital FFT-based power spectrum analysis (Welch technique,anning windowing function, no phase shift) computed powerensity of EEG rhythms with 0.5 Hz frequency resolution. Theollowing standard band frequencies were studied: delta (2–4 Hz),heta (4–8 Hz), alpha 1 (8–10.5 Hz), alpha 2 (10.5–13 Hz), beta 113–20 Hz), and beta 2 (20–30 Hz). These band frequencies werehosen averaging those used in previous relevant EEG studies onementia (Leuchter et al., 1993; Jelic et al., 1996; Besthorn et al.,997; Chiaramonti et al., 1997; Rodriguez et al., 1999a,b) andave been successfully used in recent studies on AD of thisonsortium (Babiloni et al., 2004a, 2006c,d,e). Sharing of a fre-uency bin by two contiguous bands was allowed to be consistentith the bulk of EEG literature on dementia (Cook and Leutcher,996; Jelic et al., 1996; Besthorn et al., 1997; Nobili et al., 1998;ucci et al., 1997; Kolev et al., 2002; Holschneider et al., 1999).urthermore, it followed Klimesch and others’ (Klimesch, 1996,999; Klimesch et al., 1997, 1998; Babiloni et al., 2004b,,d,e,f,005) theoretical consideration that near EEG rhythms may over-

ap at their frequency borders.The choice of fixed EEG bands did not account for individual

lpha frequency (IAF) peak, defined as the frequency associatedith the strongest EEG power at the extended alpha range (Kli-esch, 1999). However, this may not affect the results, since mostf the subjects had IAF peaks within alpha 1 band (8–10.5 Hz). Inarticular, mean (�standard error) IAF peak was 9.4�0.2 Hz inold subjects, 9.5�0.2 Hz in MCI subjects, and 8.6�0.2 Hz in ADatients. In the two MCI subgroups, the mean IAF peak was.4�0.3 Hz in MCI� and 9.6�0.3 Hz in MCI�. To control for theffect of IAF on the EEG comparisons between these two sub-roups, the IAF peak was used as a covariate (together with age,ducation and gender) for further statistics.

We could not use narrow frequency bands for beta 1 (13–0 Hz) and beta 2 (20–30 Hz) because of the variability of betand gamma peaks in the power spectra. Therefore, LORETAesults for the beta bands could suffer from sensitivity limitations ofEG spectral analyses for large bands (Szava et al., 1994).

ortical source analysis of the EEG rhythmsy LORETA

ORETA computed 3-D linear solutions (LORETA solutions) forEG inverse problem within a three-shell spherical head model

ncluding scalp, skull, and brain compartments (http://www.

nizh.ch/keyinst/NewLORETA/LORETA01.htm) (Pascual-Marqui i

nd Michel, 1994; Pascual-Marqui et al., 1999, 2002). Brain com-artment was restricted to the cortical gray matter/hippocampus ofhead model co-registered to Talairach probability brain atlas andigitized at the Brain Imaging Center of the Montreal Neurologicnstitute (Talairach and Tournoux, 1988). This compartment in-luded 2394 voxels (7 mm resolution), each voxel containing anquivalent current dipole.

LORETA solutions from EEG time series consisted of powerurrent density values at cortical voxels (cortical sources) able toredict EEG spectral power density at scalp electrodes. Theseolutions are reference free, in that one obtains the sameORETA cortical source distribution for EEG data referenced tony reference electrode including common average. To enhancehe topographical results, data normalization was obtained byormalizing the LORETA power current density at each corticaloxel with the LORETA power current density averaged across allrequencies (0.5–45 Hz) and across all 2.394 voxels of the corticalolume. After the normalization, the LORETA solutions lost theriginal physical dimension and were represented by an arbitrarynit scale. This procedure reduced inter-subjects variability andas used in previous our EEG study (Babiloni et al., 2004a,006c,d,e). The general procedure fitted the LORETA solutions inGaussian distribution and reduced inter-subject variability (Nu-er, 1988; Leuchter et al., 1993).

Solutions of the EEG inverse problem are underdeterminednd ill-conditioned when the number of spatial samples (elec-rodes) is lower than that of the unknown samples (current densityt each voxel). To account for that, the cortical LORETA solutionsredicting scalp EEG spectral power density were regularized tostimate distributed rather than punctual EEG source patternsPascual-Marqui and Michel, 1994; Pascual-Marqui et al., 1999,002). Based on this regularization procedure, LORETA providesunique deterministic source solution to the EEG inverse prob-

em. Of course, this does not mean that LORETA solutions areerfect models of EEG sources. Although LORETA is quite effi-ient (Pascual-Marqui and Michel, 1994; Pascual-Marqui et al.,999, 2002), it is not able to overcome the intrinsic limitations inhe spatial resolution of deep EEG sources common to all proce-ures of the field. Furthermore, we used the standard LORETAackage, which estimates EEG cortical sources within a spherehat just approximates head volume conductor (Ding et al., 2005).

LORETA package associates each LORETA solution relativeo a voxel with a Brodmann area. In line with the low spatialesolution of the LORETA technique, we used our MATLAB soft-are to collapse the voxels of LORETA solutions at frontal, cen-

ral, parietal, occipital, and temporal regions of the brain modeloded into Talairach space. The Brodmann areas listed in Table 3ormed each of these regions of interest (ROIs). The “low-resolu-ion” features of regional LORETA solutions could reliably takento account the slight change in the cortical volume (resolution in

illimeters) present in the mild stages of AD. Of note, theORETA procedure has been successfully used in our previoustudies on aging (Babiloni et al., 2004a, 2006c,d,e). Finally, theain advantage of the regional analysis of LORETA solutions was

able 3. Brodmann areas included in the cortical ROIs of the presenttudy

ORETA Brodmann areas into the ROIs

rontal 8, 9, 10, 11, 44, 45, 46, 47entral 1, 2, 3, 4, 6arietal 5, 7, 30, 39, 40, 43emporal 20, 21, 22, 37, 38, 41, 42ccipital 17, 18, 19

LORETA solutions were collapsed in frontal, central, parietal, occip-

tal, and temporal ROIs.

tadtLl

S

TaaaChcu

pf(fsa

ebMngdaTLssfc

te

lu1af

Ci

Ftas2patwbvdagao

Gsspfaa

Fdh averageT

B. Claudio et al. / Neuroscience 145 (2007) 942–954946

hat our modeling could disentangle rhythms of contiguous corticalreas. For example, the rhythms of the occipital source wereisentangled with respect to those of the contiguous parietal andemporal sources etc. This was made possible by the fact thatORETA uses an explicit source model coregistered into Ta-

airach space.

tatistical analysis of the LORETA solutions

he normalized regional LORETA solutions (cortical sources) rel-tive to AD, MCI and Nold subjects served as inputs to ANOVAnalyses. Subjects’ age, education, gender and IAF peak serveds covariates. Mauchly’s test evaluated the sphericity assumption.orrection of the degrees of freedom was made with the Green-ouse-Geisser procedure. The Duncan test was used for post hocomparisons (P�0.05). In particular, three ANOVA designs weresed in the present study.

For the evaluation of the control hypothesis (sensitivity of theresent LORETA approach), the first ANOVA used the followingactors: Group (Nold, MCI, AD; independent analysis), Banddelta, theta, alpha 1, alpha 2, beta 1, beta 2), and ROI (central,rontal, parietal, occipital, temporal). The existence of LORETAource differences among the groups would be confirmed by thestatistical ANOVA effect including the factor Group (P�0.05).

The second and third ANOVAs evaluated the working hypoth-sis; in particular, the second ANOVA evaluated the relationshipetween plasma homocysteine level and EEG cortical sources ofCI subjects, namely the existence of differences in the regionalormalized LORETA solutions between the MCI� and MCI�roups. This ANOVA design used Group (MCI�, MCI�; indepen-ent analysis), Band (delta, theta, alpha 1, alpha 2, beta 1, beta 2),nd ROI (central, frontal, parietal, occipital, temporal) as factors.he planned Duncan post hoc testing assessed the followingORETA pattern MCI��MCI�. This can be considered as a quiteensitive and safe procedure, given the relatively small number ofubjects and the number of bands and ROIs. Bonferroni correctionor systematic application of correlation analysis would have in-reased the risk of false negative results.

The third ANOVA evaluated the effect of plasma homocys-eine level on EEG cortical sources of AD patients, namely the

ig. 1. Grand average of LORETA solutions (i.e. normalized relative celta, theta, alpha 1, alpha 2, beta 1, and beta 2 bands in Nold, Memisphere. Color scale: all power estimates were scaled based on thehe maximal value of power is reported under each column.

xistence of differences in the regional normalized LORETA so- t

utions between the AD� and AD� groups. This ANOVA designsed Group (AD�, AD�) Band (delta, theta, alpha 1, alpha 2, beta, beta 2), and ROI (central, frontal, parietal, occipital, temporal)s factors. The planned Duncan post hoc testing assessed theollowing LORETA pattern AD��AD�.

RESULTS

ontrol analysis to validate LORETA EEG sourcesn the recruited subjects’ groups

or illustrative purposes, Fig. 1 maps the grand average ofhe LORETA solutions (i.e. relative power current densityt cortical voxels) modeling the distributed EEG corticalources for delta, theta, alpha 1, alpha 2, beta 1, and betabands in Nold, MCI, and AD groups. The Nold group

resented alpha 1 sources with the maximal values ofmplitude distributed in parieto-occipital regions. Delta,heta, and alpha 2 sources had moderate amplitude valueshen compared with alpha 1 sources. Finally, beta 1 andeta 2 sources were characterized by lowest amplitudealues. Compared with Nold group, MCI group showed aecrease in amplitude of parietal, occipital, and temporallpha 1 sources. With respect to Nold and MCI groups, ADroup showed an amplitude increase of widespread deltand theta sources, along with a strong amplitude reductionf parietal, occipital, and temporal alpha 1 sources.

Kolmogorov-Smirnov test was used to evaluate theaussian distribution of the regional normalized LORETA

olutions in Nold, MCI, and AD subjects. The resultshowed that all normalized regional LORETA solutionsresented a Gaussian distribution of the residuals in theour groups (P�0.1). Therefore, these data were used forn ANOVA control analysis to test the hypothesis that MCInd AD subjects had typical EEG features as described by

nsity at the cortical voxels) modeling the distributed EEG sources forroups. The left side of the maps (top view) corresponds to the leftd maximum value (i.e. alpha 1 power value of occipital region in Nold).

urrent deCI, AD g

he literature.

atapeaip

La

L16tci(g

afi(h

gtiaacCeas

Sa

TtafashE

tA(t

Fd(

B. Claudio et al. / Neuroscience 145 (2007) 942–954 947

LORETA solutions showed a statistical ANOVA inter-ction (F(40,1980)�1.68; MSe�0.57; P�0.004) among

he factors Group (Nold, MCI, AD), Band (delta, theta,lpha 1, alpha 2, beta 1, beta 2), and ROI (central, frontal,arietal, occipital, temporal). In line with the control hypoth-sis some EEG sources showed different amplitudecross Nold, MCI, and AD subjects: namely, delta sources

n occipital and temporal areas and alpha 1 sources inarietal, temporal, and occipital areas.

ORETA EEG sources in MCI/AD subjects with highnd low homocysteine levels

evels of plasma homocysteine (�mol/l) ranged from 5.6–0.8 in the MCI�, from 11.3–20.2 in the MCI�, from.2–10.7 in AD�, and from 11 to 38.9 in the AD�. Statis-ical ANOVA analysis showed that level of plasma homo-ysteine was lower (F(1,29)�70.6; MSe�3.4; P�0.0001)n MCI� than MCI� group. In the same vein, it was lowerF(1,29)�25.7; MSe�42.8; P�0.0001) in AD� than AD�roup.

For illustrative purposes, Fig. 2 maps the grand aver-ge of the LORETA modeling the distributed EEG sourcesor delta, theta, alpha 1, alpha 2, beta 1, and beta 2 bandsn MCI and AD subjects with low homocysteine levelMCI�, AD�) as well as in MCI and AD subjects with highomocysteine level (MCI�, AD�). Both MCI� and AD�

ig. 2. Grand average of LORETA solutions (i.e. normalized relative c

elta, theta, alpha 1, alpha 2, beta 1, and beta 2 bands in MCI and AD grohomocysteine level �11 mmol/l), and AD� and MCI� (homocysteine level �

roups presented alpha 1 sources with maximal values ofhe relative current density distributed in the parieto-occip-tal regions. Delta, theta, and alpha 2 sources had moder-te relative current density values when compared withlpha 1 sources. Finally, beta 1 and beta 2 sources wereharacterized by the lowest relative current density values.ompared with MCI�, MCI� showed very slight differ-nces. On the contrary, AD� group showed a markedmplitude increase of delta, theta, alpha 1, and alpha 2ources compared with AD� group.

tatistical comparisons of LORETA EEG sourcess a function of homocysteine level

he ANOVA analysis for the evaluation of differences inhe regional normalized LORETA solutions between MCI�nd MCI� groups showed neither a significant main effector factor Group (P�0.8) nor a significant statistical inter-ction including factor Group (P�0.6). These resultshowed that, in MCI subjects, the different level of plasmaomocysteine was not related to differences in LORETAEG sources.

ANOVA analysis for the evaluation of differences inhe regional normalized LORETA solutions betweenD� and AD� groups showed a statistical interaction

F(20,1698)�1.91; MSe�0.17; P�0.01; Fig. 3) amonghe factors Group (AD�, AD�), Band (delta, theta, alpha

nsity at the cortical voxels) modeling the distributed EEG sources for

urrent de ups subdivided in four sub-groups of 17 persons: AD� and MCI�11 �mol/l).

1phAtpPPog

og

Cs

Ijy(iesdad

iTsi2Hmperaac(sNaij

E(q2ltl

FAnrm requencyd

B. Claudio et al. / Neuroscience 145 (2007) 942–954948

, alpha 2, beta 1, beta 2), and ROI (central, frontal,arietal, occipital, temporal). The planned Duncan postoc testing, assessing the LORETA source patternD��AD�, showed that amplitude of delta (frontal and

emporal areas, P�0.01 to P�0.02), theta (central, frontal,arietal, occipital and temporal areas, P�0.00009 to�0.03), alpha 1 (parietal, occipital and temporal areas,�0.000005 to P�0.000002), and alpha 2 (parietal andccipital areas, P�0.0003 to P�0.02) sources was stron-er in AD� compared with AD� group.

These results indicated that, in AD subjects, the levelf plasma homocysteine was related to differences in re-ional LORETA EEG sources.

DISCUSSION

ontrol analysis validated the procedures forubjects’ selection and EEG data analysis

n the preliminary control analysis, AD� and AD� sub-ects were considered as a single AD group. This anal-sis allowed the modeling of resting EEG sourcesLORETA solutions) sensitive to global cognitive statusn the recruited Nold, MCI, and AD subjects. With refer-nce to Nold subjects, EEG rhythms in MCI and ADubjects were characterized by a marked magnitudeecrease of the alpha 1 sources in parietal, occipital,nd temporal areas. Furthermore, the magnitude of the

ig. 3. Regional normalized LORETA solutions (mean across subjecD�), Band (delta, theta, alpha 1, alpha 2, beta 1, beta 2), and ROormalized regional LORETA solutions as a dependent variable. Subjeegional LORETA solutions modeled the EEG relative power spectraacrocortical ROIs. The rectangles indicate the cortical regions and fifferent values between AD� and AD� groups (P�0.05).

elta sources in occipital, and temporal areas increased n

n AD subjects compared with Nold and MCI subjects.hese results are globally in line with previous evidencehowing: (i) a magnitude decrease of the alpha rhythms

n AD compared with Nold subjects (Dierks et al., 1993,000; Jelic et al., 1996, 2000; Rodriguez et al., 1999a,b;uang et al., 2000; Babiloni et al., 2004a, 2006d); (ii) aagnitude decrease of the alpha rhythms in MCI com-ared with Nold subjects (Babiloni et al., 2006d; Koenigt al., 2005); and (iii) an enhancement of the deltahythms in AD compared with Nold subjects (Prichep etl., 1994; Jelic et al., 2000; Wolf et al., 2003; Babiloni etl., 2004a). Furthermore, the present results are in ac-ordance with previous evidence on Parkinson diseasePD) patients showing an increase of magnitude of deltaources (Tanaka et al., 2000; Soikkeli et al., 1991;eufeld et al., 1988) and a decrease of magnitude oflpha sources (Soikkeli et al., 1991; Neufeld et al., 1988)

n demented PD subjects compared with normal sub-ects.

As an exception, our procedure could not discloseEG changes at beta frequencies in AD vs. Nold subjects

Claus et al., 2000, 1998). This may be due to the use ofuite large frequency bands for beta 1 (13–20 Hz) and beta(20–30 Hz), together with the average of LORETA so-

utions across large cortical regions. As mentioned above,he procedure served to reduce the mismatch between theimited number of EEG channels (i.e. 19) and the high

e to a statistical ANOVA interaction among the factors Group (AD�,central, parietal, occipital, temporal). This ANOVA design used the

education, gender and IAF peak were used as covariates. Normalizedled by a sort of “virtual” intracranial macro-electrodes located on thebands in which LORETA solutions presented statistically significant

ts) relativI (frontal,cts’ age,as revea

umber of LORETA voxels (i.e. 2394).

pctE

Ca

Aapoo(wrsetaltnrpcpsw(rpra(

lltorlthncasrwapm

qdAc1

Lorsgratwccact28a“

timstbslrFpcacobdr1eta1twpist(

wcbpsiAf

B. Claudio et al. / Neuroscience 145 (2007) 942–954 949

From a methodological viewpoint, the validation of therocedures for subjects’ selection and EEG data analysisorroborated the novel results of the present study, namelyhe relationships among plasma homocysteine level andEG cortical sources in AD patients.

ortical sources of EEG rhythms in AD subjectsre related to plasma homocysteine level

s original results, we demonstrated that cortical sourcest delta (frontal and temporal areas), theta (central, frontal,arietal, occipital and temporal areas), alpha 1 (parietal,ccipital and temporal areas), and alpha 2 (parietal andccipital areas) were stronger in magnitude in the AD�i.e. “�” means more plasma homocysteine concentration)ith respect to AD� subjects. It is unlikely that the present

esults are affected by inhibitors of acetylcholinesterase,ince they were administrated with drug dose practicallyqual in AD patients of both sub-groups. Rather, the men-ioned effects of homocysteine on cortical EEG rhythmsre reasonably in relationship with the marked brain neural

oss accompanying high plasma homocysteine concentra-ion in AD subjects (i.e. with effects on microcirculation,eurotoxicity, oxidative stress, and stem cell-mediatedegeneration). Several lines of evidence support this ex-lanation. It has been reported that higher plasma homo-ysteine levels were related to more atrophy of the hip-ocampus and cortical regions in elderly non-dementedubjects at risk of AD (den Heijer et al., 2003) and tohite-matter hyperintensity in psychogeriatric patients

Scott et al., 2004). Furthermore, it was observed that withespect to control elderly subjects, AD patients had higherlasma homocysteine concentration, lower medial tempo-al lobe thickness (de Silva et al., 2005), more global braintrophy (Gottfries et al., 2001), and more vascular lesionsFolin et al., 2005).

The above explanation does not mean that neuronaloss and some abnormalities in the vasculature are pecu-iar features of AD patients with a high plasma homocys-eine level, given that these features can be practicallybserved in all AD patients. Rather, it suggests that neu-onal loss and corresponding abnormalities in the vascu-ature might be more pronounced in AD patients with highhan low plasma homocysteine levels, and that this mightave possible negative implications for therapy and prog-osis. However, at the present stage of the research, weould not evaluate the exact nature of the relationshipsmong plasma homocysteine concentration, vascular le-ion, brain atrophy, and abnormal cortical sources of EEGhythms in AD patients. Causality of the relationshipsould have required serial measurements of these vari-bles across time, together with a regression analysis. Theresent results strongly motivate the remarkable invest-ents to do so.

For the understanding of the present results, a crucialuestion is, Why was the magnitude of alpha other thanelta and theta rhythms stronger in AD� compared withD� subjects? Indeed, alpha rhythms of normal amplitudeharacterize healthy wakeful brain (Steriade and Llinas,

988; Rossini et al., 1991; Brunia, 1999; Pfurtscheller and a

opes da Silva, 1999). To formulate a tentative explanationf the paradox, generation mechanisms producing EEGhythms should be briefly outlined. During slow-waveleep, corticofugal slow oscillations (�1 Hz) are effective inrouping thalamic-generated delta (1–4 Hz) and spindlinghythms (7–14 Hz; Steriade, 2003). In the case of brainrousal, spindles as well as high and low components ofhe delta rhythms are blocked by the inhibition of oscillatorsithin, respectively, reticulo-thalamic (7–14 Hz), thalamo-ortical (1–4 Hz), and intracortical (�1 Hz) neuronal cir-uits. These rhythms are replaced by fast oscillations (betand gamma), which are mainly induced by forebrain (nu-leus basalis) cholinergic inputs to hippocampus and cor-ex as well as by thalamocortical projections (Steriade,003). In the healthy condition of awake rest, alpha (about–12 Hz) dominates the EEG rhythms and delta is low inmplitude; this is reminiscent of a condition of possiblereciprocal inhibition.”

Within this theoretical framework, one can speculatehat AD� subjects are characterized by a more markedmpairment of the cholinergic basal forebrain altering the

echanism of “reciprocal inhibition” between alpha andlower EEG rhythms. This would produce a specific pat-ern of EEG synchronization across several frequencyands, where the selectivity for frequency of EEG power iseverely affected. Indeed, acetylcholine might provide se-

ectivity of cortical neural synchronization (generating EEGhythms) rather than simple excitation or facilitation effects.urthermore, acetylcholine systems might be more im-aired in AD subjects with higher plasma levels of homo-ysteine. This speculation is in accordance with previousnimal studies showing that proper modulation of acetyl-holine in the cortex (i.e. chemical release; administrationf cholinergic agonists; electrical stimulation of the fore-rain cholinergic neurons) reduces the amplitude of theelta EEG rhythms and increases the amplitude of fasterhythms (Semba and Fibiger, 1989; Metherate et al.,992). Conversely, the intra-cortical application of cholin-rgic (muscarinic) antagonists and experimental lesions ofhe basal forebrain both increase the amplitude of the deltand decrease that of faster EEG rhythms (Stewart et al.,984; Buzsaki et al., 1988; Ray and Jackson, 1991; Me-herate et al., 1992). Finally, the above explanation agreesith recent evidence of the present Consortium. In com-arison to AD non-responders to long-term acetylcholine

nhibitors therapy (i.e. donepezil), AD non-respondershowed paradoxical higher magnitude of alpha rhythmsogether with higher magnitude of slower EEG rhythmsBabiloni et al., 2006f).

Remarkably, a paradoxical result of the present studyas that in contrast to the cortical sources of EEG rhythms,ognitive status of the AD subjects did not statistically differetween the sub-groups with high (AD�) and low (AD�)lasma homocysteine concentration. Indeed the MMSEcore was similar in both AD sub-groups. Therefore, the

ncrease of amplitude in delta, theta and alpha sources inD� compared with AD� sub-group was not due to dif-

erence in global cognitive status. These results pointed to

dissociation between cortical EEG deterioration and cog-

ncrnspoA2bpiticbs

tApthbt(aTbmaa

WrAsstsailbrtducvttsnatfi

spnbF

AttbI

A

A

A

A

A

B

B

B

B

B

B

B

B. Claudio et al. / Neuroscience 145 (2007) 942–954950

itive status in the two sub-groups of AD subjects, andomplement those showing that cortical sources of EEGhythms (delta and alpha 1) correlated with subject’s cog-itive status across normal elderly, MCI, and mild ADubjects (Babiloni et al., 2006a,d). Furthermore, they com-lement the previous results indicating that the amplitudef alpha sources decreased with increasing dementia inD and PD patients (Babiloni et al., 2004a; Tanaka et al.,000). Of note, focusing on linear or simple relationshipsetween plasma homocysteine concentration, delta or al-ha sources, and MMSE in pathological aging may be

nsufficient to capture the complex interactions amonghese variables. The present results motivate future stud-es using artificial neural networks or other non-linear pro-edures to reconstruct the complex relationships amongiological risk factors, single EEG cortical sources, andubject’s cognitive status in pathological aging.

Another paradoxical result of the present study washat in contrast to the cortical sources of EEG rhythms inD patients, those in MCI subjects (including alpha 1)resented no statistically significant difference between

he sub-groups with high (MCI�) and low (MCI�) plasmaomocysteine concentration. In reality, these results cane explained by lower difference in the plasma homocys-eine concentration between MCI� and MCI� subjectsmax difference from 5.6–20.2 �mol/l) than between AD�nd AD� patients (max difference from 6.2–38.9 �mol/l).his explanation agrees with previous reports showing thatrain neural loss correlates with the severity of AD and isuch more evident in AD then MCI subjects (Rombouts etl., 2000; Baron et al., 2001; Ohnishi et al., 2001; Frisoni etl., 2002; Testa et al., 2004; Wolf et al., 2003).

CONCLUSIONS

e hypothesized that high plasma homocysteine levels areelated to cortical sources of slow EEG rhythms in awakeningD subjects at rest, as a reflection of the well-known relation-hips between cortical neural loss and these rhythms. Re-ults showed that not only delta (frontal and temporal) andheta (central, frontal, parietal, occipital, and temporal)ources but also alpha 1 (parietal, occipital, and temporal)nd alpha 2 (parietal and occipital) sources were stronger

n magnitude in AD� (�higher plasma homocysteineevel) than AD� group. Instead, no differences were foundetween the MCI� and MCI� groups. These results rep-esent the first demonstration that high plasma homocys-eine levels are related to unselective increment of corticalelta, theta, and alpha rhythms in mild AD patients, thusnveiling possible relationships among high plasma homo-ysteine concentration, microvascular concomitants of ad-anced neurodegenerative processes, and synchroniza-ion mechanisms regulating the relative amplitude of delta–heta compared with alpha EEG rhythms. However, ithould be noted that because of the purely descriptiveature of the present study and the just hypothetical mech-nism(s) through which plasma homocysteine concentra-ion confers a risk to AD, the significance of the present

ndings will have to be elucidated by future multi-modal

tudies in AD subjects combining the quantification oflasma homocysteine concentration, brain vascular lesion,eural loss, and EEG rhythms. This will be invaluable forasic research and clinical applications (e.g. prognosis).ig 3.

cknowledgments—We thank Prof. Fabrizio Eusebi for his con-inuous support. We thank Roberta Lizio for the valuable help inhe development of the present study. The research was fundedy the Association Fatebenefratelli for Research (AFaR) andRCCS “Oasi” of Troina (Italy).

REFERENCES

nderer P, Pascual-Marqui RD, Semlitsch HV, Saletu B (1998a) Dif-ferential effects of normal aging on sources of standard N1, targetN1 and target P300 auditory event-related brain potentials re-vealed by low resolution electromagnetic tomography (LORETA).Electroencephalogr Clin Neurophysiol 108(2):160–174.

nderer P, Saletu B, Semlitsch HV, Pascual-Marqui RD (1998b) Elec-trical sources of P300 event-related brain potentials revealed bylow resolution electromagnetic tomography. 2. Effects of nootropictherapy in age-associated memory impairment. Neuropsychobiol-ogy 37(1):28–35.

nderer P, Saletu B, Semlitsch HV, Pascual-Marqui RD (2003) Non-invasive localization of P300 sources in normal aging and age-associated memory impairment. Neurobiol Aging 24(3):463–479.

nello G, Gueant-Rodriguez RM, Bosco P, Gueant JL, Romano A,Namour B, Spada R, Caraci F, Pourie G, Daval JL, Ferri R (2004)Homocysteine and methylenetetrahydrofolate reductase polymor-phism in Alzheimer’s disease. Neuroreport 15:859–861.

raki A, Sako Y (1987) Determination of free and total homocysteinein human plasma by high-performance liquid chromatography withfluorescence detection. J Chromatogr 422:43–52.

abiloni C, Binetti G, Cassetta E, Cerboneschi D, Dal Forno G, Del PercioC, Ferreri F, Ferri R, Lanuzza B, Miniussi C, Moretti DV, Nobili E,Pascual-Marqui RD, Rodriguez G, Romani GL, Salinari S, Tecchio F,Vitali P, Zanetti O, Zappasodi F, Rossini PM (2004a) Mapping distrib-uted sources of cortical rhythms in mild Alzheimer’s disease. A multi-centric EEG study. Neuroimage 22(1):57–67.

abiloni C, Babiloni F, Carducci F, Cincotti F, Del Percio C, DellaPenna S, Franciotti R, Pignotti S, Pizzella V, Rossini PM, SabatiniE, Torquati K, Romani GL (2004b) Human alpha rhythms duringvisual delayed choice reaction time tasks. A MEG study. HumBrain Map 24(3):184–192.

abiloni C, Babiloni F, Carducci F, Cappa S, Cincotti F, Del Percio C,Miniussi C, Moretti DV, Rossi S, Sosta K, Rossini PM (2004c)Human cortical rhythms during visual delayed choice reaction timetasks. A high-resolution EEG study on normal aging. Behav BrainRes 153(1):261–271.

abiloni C, Miniussi C, Babiloni F, Carducci F, Cincotti F, Del Percio C,Sirello G, Sosta K, Nobre AC, Paolo M, Rossini PM (2004d)Sub-second “temporal attention” modulates alpha rhythms. A high-resolution EEG study. Cogn Brain Res 19(3):259–268.

abiloni C, Babiloni F, Carducci F, Cappa S, Cincotti F, Del Percio C,Miniussi C, Moretti DV, Rossi S, Sosta K, Rossini PM (2004e)Human cortical responses during one-bit short-term memory. Ahigh-resolution EEG study on delayed choice reaction time tasks.Clin Neurophysiol 115(1):161–170.

abiloni C, Babiloni F, Carducci F, Cappa S, Cincotti F, Del Percio C,Miniassi C, Moretti DV, Pasqualetti P, Rossi S, Sosta K, RossiniPM (2004f) Human cortical EEG rhythms during long-term episodicmemory task. A high resolution EEG study of the HERA model.Neuroimage 21(4):1576–1584.

abiloni C, Cassetta E, Chiovenda P, Del Percio C, Ercolani M, MorettiDV, Moffa F, Pasqualetti P, Pizzella V, Romani GL, Tecchio F,

Zappasodi F, Rossini PM (2005) Frontomedial alpha hyper-reac-

B

B

B

B

B

B

B

B

B

B

B

B

C

C

C

C

d

d

D

d

D

D

D

D

D

D

E

F

F

F

F

F

F

B. Claudio et al. / Neuroscience 145 (2007) 942–954 951

tivity in mild demented patients during visual delayed responsetasks. A MEG study. Brain Res Bull 65(6):457–470.

abiloni C, Frisoni G, Steriade M, Bresciani L, Binetti G, Del Percio C,Geroldi C, Miniussi C, Nobili F, Rodriguez G, Zappasodi F, Carf-agna T, Rossini P (2006a) Frontal white matter volume and deltaEEG sources negatively correlate in awake subjects with mildcognitive impairment and Alzheimer’s disease. Clin Neurophysiol117(5):1113–1129.

abiloni C, Benussi L, Binetti G, Bosco P, Busonero G, Cesaretti S,Dal Forno G, Del Percio C, Ferri R, Frisoni G, Ghidoni R, Rodri-guez G, Squitti R, Rossini PM (2006b) Genotype (cystatin C) andEEG phenotype in Alzheimer disease and mild cognitiveimpairment: a multicentric study. Neuroimage 29(3):948–64. Epub2005 Oct 6.

abiloni C, Benussi L, Binetti G, Cassetta E, Dal Forno G, Del PercioC, Ferreri F, Ferri R, Frisoni G, Ghidoni R, Miniussi C, RodriguezG, Romani GL, Squitti R, Ventriglia MC, Rossini PM (2006c) Apo-lipoprotein E and alpha brain rhythms in mild cognitive impairment:A multicentric EEG study. Ann Neurol 59(2):323–334.

abiloni C, Binetti G, Cassetta E, Dal Forno G, Del Percio C, Ferreri F,Ferri R, Frisoni G, Hirata K, Lanuzza B, Miniussi C, Moretti DV,Nobili F, Rodriguez G, Romani GL, Salinari S, Rossini PM (2006d)Sources of cortical rhythms change as a function of cognitiveimpairment in pathological aging: a multi-centric study. Clin Neu-rophysiol 117(2):252–268. Epub 2005 Dec 27.

abiloni C, Binetti G, Cassarino A, Dal Forno G, Del Percio C, FerreriF, Ferri R, Frisoni G, Galderisi S, Hirata K, Lanuzza B, Miniussi C,Mucci A, Nobili F, Rodriguez G, Romani GL, Rossini PM (2006e)Sources of cortical rhythms in adults during physiological aging: amulti-centric EEG study. Hum Brain Mapp 27(2):162–172.

abiloni C, Cassetta E, Dal Forno G, Del Percio C, Ferreri F, Ferri R,Lanuzza B, Miniussi C, Moretti DV, Nobili F, Pascual-Marqui RD,Rodriguez G, Luca Romani G, Salinari S, Zanetti O, Rossini PM(2006f) Donepezil effects on sources of cortical rhythms in mildAlzheimer’s disease: Responders vs. non-responders. Neuroim-age. 2006f Apr 4;[Epub ahead of print].

aron JC, Chételat G, Desgranges B, et al. (2001) In vivo mapping ofgrey matter loss with voxel-based morphometry in mild Alzheimer’sdisease. Neuroimage 14:298–309.

artesaghi S, Accinni R, De Leo G, Cursano CF, Campolo J, GalluzzoC, Vegezzi PG, Parodi O (2001) A new HPLC micromethod tomeasure total plasma homocysteine in newborn. J Pharm BiomedAnal 24(5–6):1137–1141.

esthorn C, Zerfass R, Geiger-Kabisch C, Sattel H, Daniel S, Schre-iter-Gasser U, Forstl H (1997) Discrimination of Alzheimer’s dis-ease and normal aging by EEG data. Electroencephalogr ClinNeurophysiol 103(2):241–248.

runia CH (1999) Neural aspects of anticipatory behavior. Acta Psy-chol (Amst) 101(2–3):213–242.

uchan RJ, Nagata K, Yokoyama E, Langman P, Yuya H, Hirata Y,Hatazawa J, Kanno I (1997) Regional correlations between theEEG and oxygen metabolism in dementia of Alzheimer’s type.Electroencephalogr Clin Neurophysiol 103(3):409–417.

uzsaki G, Bickford RG, Ponomareff G, Thal LJ, Mandel R, Gage FH(1988) Nucleus basalis and thalamic control of neocortical activityin the freely moving rat. J Neurosci 8(11):4007–4026.

hiaramonti R, Muscas GC, Paganini M, Muller TJ, Fallgatter AJ,Versari A, Strik WK (1997) Correlations of topographical EEGfeatures with clinical severity in mild and moderate dementia ofAlzheimer type. Neuropsychobiology 36(3):153–158.

laus JJ, Kwa VI, Teunisse S, Walstra GJ, van Gool WA, Koelman JH,Bour LJ, Ongerboer de Visser BW (1998) Slowing on quantitativespectral EEG is a marker for rate of subsequent cognitive andfunctional decline in early Alzheimer disease. Alzheimer Dis AssocDisord 12(3):167–174.

laus JJ, Ongerboer De Visser BW, Bour LJ, Walstra GJ, Hijdra A,Verbeeten B Jr, Van Royen EA, Kwa VI, van Gool WA (2000)

Determinants of quantitative spectral electroencephalography in

early Alzheimer’s disease: cognitive function, regional cerebralblood flow, and computed tomography. Dement Geriatr CognDisord 11(2):81–89.

ook IA, Leuchter AF (1996) Synaptic dysfunction in Alzheimer’sdisease: clinical assessment using quantitative EEG. Behav BrainRes 78(1):15–23.

e Jongh A, Baayen JC, de Munck JC, Heethaar RM, Vandertop WP,Stam CJ (2003) The influence of brain tumor treatment on patho-logical delta activity in MEG. Neuroimage 20(4):2291–2301.

en Heijer T, Vermeer SE, Clarke R, Oudkerk M, Koudstaal PJ,Hofman A, Breteler MM (2003) Homocysteine and brain atrophy onMRI of non-demented elderly. Department of Epidemiology andBiostatistics, Erasmus Medical Center, Rotterdam, The Nether-lands. Brain 126(Pt 1):170–175.

eng A, Irizarry MC, Nitsch RM, Growdon JH, Rebeck GW (2001)Elevation of cystatin C in susceptible neurons in Alzheimer’s dis-ease. Am J Pathol 159(3):1061–1068.

e Silva HA, Gunatilake SB, Johnston C, Warden D, Smith AD (2005)Medial temporal lobe atrophy, apolipoprotein genotype, andplasma homocysteine in Sri Lankan patients with Alzheimer’s dis-ease. Exp Aging Res 31(3):345–354.

evanand DP, Folz M, Gorlyn M, Moeller JR, Stern J (1997) Ques-tionable dementia: clinical course and predictors of outcome. J AmGeriatr Soc 45:321–328.

ierks T, Ihl R, Frolich L, Maurer K (1993) Dementia of the Alzheimertype: effects on the spontaneous EEG described by dipolesources. Psychiatry Res 50(3):151–162.

ierks T, Jelic V, Pascual-Marqui RD, Wahlund LO, Julin P, LindenDEJ, Maurer K, Winblad B, Nordberg A (2000) Spatial pattern ofcerebral glucose metabolism (PET) correlates with localization ofintracerebral EEG-generators in Alzheimer’s disease. Clin Neuro-physiol 111:1817–1824.

ing L, Lai Y, He B (2005) Low resolution brain electromagnetictomography in a realistic geometry head model: a simulation study.Phys Med Biol 50(1):45–56.

ringenberg HC (2000) Alzheimer’s disease: more than a “cholinergicdisorder”: evidence that cholinergic-monoaminergic interactionscontribute to EEG slowing and dementia. Behav Brain Res115(2):235–249.

ringenberg HC, Rubenstein ML, Solty H, Tomaszek S, Bruce A(2002) Electroencephalographic activation by tacrine, deprenyl,and quipazine: cholinergic vs. non-cholinergic contributions. EurJ Pharmacol 447(1):43–50.

lmstahl S, Rosen I (1997) Postural hypotension and EEG variablespredict cognitive decline: results from a 5-year follow-up of healthyelderly women. Dement Geriatr Cogn Disord 8(3):180–187.

ernandez A, Arrazola J, Maestu F, Amo C, Gil-Gregorio P, Wien-bruch C, Ortiz T (2003) Correlations of hippocampal atrophy andfocal low-frequency magnetic activity in Alzheimer disease: volu-metric MR imaging-magnetoencephalographic study. AJNR Am JNeuroradiol 24(3):481–487.

olin M, Baiguera S, Gallucci M, Conconi MT, Di Liddo R, Zanardo A,Parnigotto PP (2005) A cross-sectional study of homocysteine-,NO-levels, and CT-findings in Alzheimer dementia, vascular de-mentia and controls. Biogerontology 6(4):255–260.

olstein MF, Folstein SE, McHugh PR (1975) “Mini Mental State”: apractical method for grading the cognitive state of patients forclinician. J Psychiatr Res 12:189–198.

risoni GB, Beltramello A, Binetti G, Bianchetti A, Weiss C, Scuratti A,Trabucchi M (1995) Computed tomography in the detection of thevascular component in dementia. Gerontology 41(2):121–128.

risoni GB, Testa C, Zorzan A, et al. (2002) Detection of gray matterloss in mild Alzheimer’s disease with voxel-based morphometry.J Neurol Neurosurg Psychiatry 73:657–664.

uchs M, Wagner M, Kastner J (2001) Boundary element methodvolume conductor models for EEG source reconstruction. Clin

Neurophysiol 112(8):1400–1407.

G

G

G

G

G

G

G

G

G

H

H

H

H

H

H

H

I

J

J

J

K

K

K

K

K

K

K

L

L

L

L

M

M

M

M

M

M

M

B. Claudio et al. / Neuroscience 145 (2007) 942–954952

alluzzi S, Sheu CF, Zanetti O, Frisoni GB (2005) Distinctive clinicalfeatures of mild cognitive impairment with subcortical cerebrovas-cular disease. Dement Geriatr Cogn Disord 19(4):196–203.

eula C, Mesulam M (1989) Special properties of cholinesterases inthe cerebral cortex of Alzheimer’s disease. Brain Res 498(1):185–189.

eula C, Mesulam MM (1996) Systematic regional variations in theloss of cortical cholinergic fibers in Alzheimer’s disease. CerebCortex 6(2):165–177.

eula C, Mesulam MM (1999) Cholinergic system in Alzheimer’sdisease. In: Alzheimer disease, 2nd ed. (Terry RD et al., eds),pp 69–292. Philadelphia, PA: Lippincott Williams & Wilkins.

loor P, Ball G, Schaul N (1977) Brain lesions that produce deltawaves in the EEG. Neurology 27(4):326–333.

oforth HW, Konopka L, Primeau M, Ruth A, O’Donnell K, Patel R,Poprawski T, Shirazi P, Rao M (2004) Quantitative electroen-cephalography in frontotemporal dementia with methylphenidateresponse: a case study. Clin EEG Neurosci 35(2):108–111.

ottfries J, Blennow K, Lehmann MW, Regland B, Gottfries CG (2001)One-carbon metabolism and other biochemical correlates of cog-nitive impairment as visualized by principal component analysis.J Geriatr Psychiatry Neurol 14(3):109–114.

ross J, Kujala J, Hamalainen M, Timmermann L, Schnitzler A,Salmelin R (2001) Dynamic imaging of coherent sources: Studyingneural interactions in the human brain. Proc Natl Acad Sci U S A98(2):694–699.

uidi I, Galimberti D, Lonati S, Novembrino C, Bamonti F, Tiriticco M,Fenoglio C, Venturelli E, Baron P, Bresolin N, Scarpini E (2006)Oxidative imbalance in patients with mild cognitive impairment andAlzheimer’s disease. Neurobiol Aging 27(2):262–269. Epub 2005Feb 17.

armony T, Fernandez-Bouzas A, Marosi E, Fernandez T, Bernal J,Rodriguez M, Reyes A, Silva J, Alonso M, Casian G (1993) Cor-relation between computed tomography and voltage and currentsource density spectral EEG parameters in patients with brainlesions. Electroencephalogr Clin Neurophysiol 87(4):196–205.

elkala EL, Hanninen T, Hallikainen M, Kononen M, Laakso MP,Hartikainen P, Soininen H, Partanen J, Partanen K, Vainio P,Riekkinen P Sr (1996) Slow-wave activity in the spectral analysis ofthe electroencephalogram and volumes of hippocampus insubgroups of Alzheimer’s disease patients. Behav Neurosci110(6):1235–1243.

ensel S, Rockstroh B, Berg P, Elbert T, Schonle PW (2004) Left-hemispheric abnormal EEG activity in relation to impairment andrecovery in aphasic patients. Psychophysiology 41(3):394–400.

olschneider DP, Waite JJ, Leuchter AF, Walton NY, Scremin OU(1999) Changes in electrocortical power and coherence in re-sponse to the selective cholinergic immunotoxin 192 IgG-saporin.Exp Brain Res 126(2):270–280.

uang C, Wahlund LO, Dierks T, Julin P, Winblad B, Jelic V (2000)Discrimination of Alzheimer’s disease and mild cognitive impair-ment by equivalent EEG sources: a cross-sectional and longitudi-nal study. Clin Neurophysiol 11:1961–1967.

uang C, Wahlund LO, Svensson L, Winblad B, Julin P (2002) Cin-gulate cortex hypoperfusion predicts Alzheimer’s disease in mildcognitive impairment. BMC Neurol 2(1):9.

ughes CP, Berg L, Danziger WL, Cohen LA, Martin RL (1982) A newclinical rating scale for the staging of dementia. Br J Psychiatry140:1225–1230.

rizarry MC, Gurol ME, Raju S, Diaz-Arrastia R, Locascio JJ, Tennis M,Hyman BT, Growdon JH, Greenberg SM, Bottiglieri T (2005) Associ-ation of homocysteine with plasma amyloid beta protein in aging andneurodegenerative disease. Neurology 65(9):1402–1408.

elic V, Shigeta M, Julin P (1996) Quantitative electroencephalogra-phy power and coherence in Alzheimer’s disease and mild cogni-tive impairment. Dementia 7:314–323.

elic V, Johansson SE, Almkvist O, Shigeta M, Julin P, Nordberg A,

Winblad B, Wahlund LO (2000) Quantitative electroencephalogra-

phy in mild cognitive impairment: longitudinal changes andpossible prediction of Alzheimer’s disease. Neurobiol Aging 21(4):533–540.

eong J (2004) EEG dynamics in patients with Alzheimer’s disease.Clin Neurophysiol 115(7):1490–1505.

almijn S, Launer LJ, Lindemans J, Bots ML, Hofman A, Breteler MM(1999) Total homocysteine and cognitive decline in a community-based sample of elderly subjects: the Rotterdam Study. Am JEpidemiol 150(3):283–289.

limesch W (1996) Memory processes, brain oscillations and EEGsynchronization. Int J Psychophysiol 24(1–2):61–100.

limesch W (1999) EEG alpha and theta oscillations reflect cognitiveand memory performance: a review and analysis. Brain Res Rev29:169–195.

limesch W, Doppelmayr M, Pachinger T, Russegger H (1997) Event-related desynchronization in the alpha band and the processing ofsemantic information. Brain Res Cogn Brain Res 6(2):83–94.

limesch W, Doppelmayr M, Russegger H, Pachinger T, Schwaiger J(1998) Induced alpha band power changes in the human EEG andattention. Neurosci Lett 244(2):73–76.

oenig T, Prichep L, Dierks T, Hubl D, Wahlund LO, John ER, Jelic V(2005) Decreased EEG synchronization in Alzheimer’s diseaseand mild cognitive impairment. Neurobiol Aging 26(2):165–171.

olev V, Yordanova J, Basar-Eroglu C, Basar E (2002) Age effects onvisual EEG responses reveal distinct frontal alpha networks. ClinNeurophysiol 113(6):901–910.

autenschlager NT, Flicker L, Vasikaran S, Leedman P, Almeida OP(2005) Coil-to-helix transition and ligand release of Bombyx moripheromone-binding protein. Biochem Biophys Res Commun335(4):1044–1050.

awton MP, Brodie EM (1969) Assessment of older people: self main-taining and instrumental activity of daily living. J Gerontol 9:179–186.

euchter AF, Cook IA, Newton TF, Dunkin J, Walter DO, RosenbergTompson S, Lachenbruch PA, Weiner H (1993) Regional differ-ences in brain electrical activity in dementia: use of spectral powerand spectral ratio measures. Electroencephalogr Clin Neuro-physiol 87:385–393.

evy E, Sastre M, Kumar A, Gallo G, Piccardo P, Ghetti B, TagliaviniF (2001) Codeposition of cystatin C with amyloid-beta protein in thebrain of Alzheimer disease patients. J Neuropathol Exp Neurol60(1):94–104.

ash DC, Flynn DD, Potter LT (1985) Loss of M2 muscarine receptorsin the cerebral cortex in Alzheimer’s disease and experimentalcholinergic denervation Science 228(4703):1115–1117.

aurer K, Dierks T (1992) Functional imaging procedures indementias: mapping of EEG and evoked potentials. Acta NeurolScand Suppl 139:40–46.

cCaddon A, Davies G, Hudson P, Tandy S, Cattell H (1998) Totalserum homocysteine in senile dementia of Alzheimer type. Int JGeriatr Psychiatry 13(4):235–239.

cCaddon A, Hudson P, Hill D, Barber J, Lloyd A, Davies G, ReglandB (2003) Alzheimer’s disease and total plasma aminothiols. BiolPsychiatry 53(3):254–260.

cKeith IG, Perry EK, Perry RH (1999) Report of the second dementiawith Lewy body international workshop: diagnosis and treatment.Consortium on Dementia with Lewy Bodies. Neurology 53:902–905.

cKhann G, Drachman D, Folstein M, Katzman R, Price D, StadlanEM (1984) Clinical diagnosis of Alzheimer’s disease: report of theNINCDS- ADRDA Work Group under the auspices of Departmentof Health and Human Services Task Force on Alzheimer’s disease.Neurology 34:939–944.

esulam M, Shaw P, Mash D, Weintraub S (2004) Cholinergic nu-cleus basalis tauopathy emerges early in the aging-MCI-AD con-

tinuum. Ann Neurol 55(6):815–828.

M

M

M

M

M

N

N

N

N

O

P

P

P

P

P

P

P

P

P

P

P

Q

R

R

R

R

R

R

R

R

R

S

S

S

S

S

S

S

B. Claudio et al. / Neuroscience 145 (2007) 942–954 953

etherate R, Cox CL, Ashe JH (1992) Cellular bases of neocorticalactivation: modulation of neural oscillations by the nucleus basalisand endogenous acetylcholine. J Neurosci 12(12):4701–4711.

izrahi EH, Bowirrat A, Jacobsen DW, Korczyn AD, Traore F, PetotGJ, Lerner AJ, Debanne SM, Adunsky A, Dibello PM, Friedland RP(2004) Plasma homocysteine, vitamin B12 and folate in Alzhei-mer’s patients and healthy Arabs in Israel. J Neurol Sci227(1):109–113.

oretti DV, Babiloni F, Carducci F, Cincotti F, Remondini E, RossiniPM, Salinari S, Babiloni C (2003) Computerized processing ofEEG-EOG-EMG artifacts for multicenter studies in EEG oscilla-tions and event-related potentials. Int J Psychophysiol 47(3):199–216.

uller TJ, Thome J, Chiaramonti R, Dierks T, Maurer K, Fallgatter AJ,Frolich L, Scheubeck M, Strik WK (1997) A comparison of qEEGand HMPAO-SPECT in relation to the clinical severity of Alzhei-mer’s disease. Eur Arch Psychiatry Clin Neurosci 247(5):259–263.

urri L, Gori S, Massetani R, Bonanni E, Marcella F, Milani S (1998)Evaluation of acute ischemic stroke using quantitative EEG: acomparison with conventional EEG and CT scan. NeurophysiolClin 28(3):249–257.

eufeld MY, Inzelberg R, Korczyn AD (1998) EEG in demented andnon-demented parkinsonian patients. Acta Neurol Scand 78(1):1–5.

ilsson K, Gustafson L, Hultberg B (2006) Plasma homocysteine andvascular disease in psychogeriatric patients. Dement Geriatr CognDisord 21(3):148–154. Epub 2006 Jan 2.

obili F, Taddei G, Vitali P, Bazzano L, Catsafados E, Mariani G,Rodriguez G (1998) Relationships between 99m Tc-HMPAOceraspect and quantitative EEG observations in Alzheimer’s dis-ease. Arch Gerontol Geriatr 6:363–368.

uwer MR, Quantitative EEG (1988) I: techniques and problems offrequency analysis and topographic mapping. J Clin Neurophysiol5:1–43.

hnishi T, Matsuda H, Tabira T, Asada T, Uno M (2001) Changes inbrain morphology in Alzheimer disease and normal ageing: isAlzheimer disease an exaggerated aging process? AJNR Am JNeuroradiol 22:1680–1685.

ascual-Marqui RD, Michel CM (1994) LORETA (low resolution brainelectromagnetic tomography): new authentic 3D functional images ofthe brain. ISBET Newsletter ISSN 5:4–8.

ascual-Marqui RD, Lehmann D, Koenig T, Kochi K, Merlo MC, Hell D,Koukkou M (1999) Low resolution brain electromagnetic tomogra-phy (LORETA) functional imaging in acute, neuroleptic-naive, first-episode, productive schizophrenia. Psychiatry Res 90(3):169–179.

ascual-Marqui RD, Esslen M, Kochi K, Lehmann D (2002) Functionalimaging with low resolution brain electromagnetic tomography(LORETA): a review. Methods Find Exp Clin Pharmacol 24:91–95.

etersen RC, Smith GE, Ivnik RJ, Tangalos EG, Schaid SN, Thi-bodeau SN, Kokmen E, Waring SC, Kurland LT (1995) Apolipopro-tein E status as a predictor of the development of Alzheimer’sdisease in memory-impaired individuals. JAMA 273:1274–1278.

etersen RC, Smith GE, Waring SC, Ivnik RJ, Kokmen E, TangelosEG (1997) Aging, memory, and mild cognitive impairment. IntPsychogeriatr 9 (Suppl 1):65–69.

etersen RC, Doody R, Kurz A, Mohs RC, Morris JC, Rabins PV,Ritchie K, Rossor M, Thal L, Winblad B (2001) Current concepts inmild cognitive impairment. Arch Neurol 58(12):1985–1992.

furtscheller G, Lopes da Silva F (1999) Event-related EEG/MEGsynchronization and desynchronization: basic principles. Clin Neu-rophysiol 110:1842–1857.

onomareva NV, Selesneva ND, Jarikov GA (2003) EEG alterations insubjects at high familial risk for Alzheimer’s disease. Neuropsy-chobiology 48(3):152–159.

richep LS, John ER, Ferris SH, Reisberg B, Almas M, Alper K,

Cancro R (1994) Quantitative EEG correlates of cognitive deteri-

oration in the elderly. Neurobiol Aging 15(1):85–90. Erratum in:Neurobiol Aging 1994;15(3):391.

ucci E, Cacchiò G, Angeloni R, Belardinelli N, Nolfe G, Signorino M,Angeleri F (1997) EEG spectral analysis in Alzheimer’s diseaseand different degenerative dementias. Arch Gerontol Geriatr26:283–297.

ucci E, Belardinelli N, Cacchio G, Signorino M, Angeleri F (1999)EEG power spectrum differences in early and late onset forms ofAlzheimer’s disease. Clin Neurophysiol 110(4):621–631.

uadri P, Fragiacomo C, Pezzati R, Zanda E, Tettamanti M, Lucca U(2005) Homocysteine and B vitamins in mild cognitive impairmentand dementia. Clin Chem Lab Med 43(10):1096–1100.

ay PG, Jackson WJ (1991) Lesions of nucleus basalis alter ChATactivity and EEG in rat frontal neocortex. Electroencephalogr ClinNeurophysiol 79(1):62–68.

odriguez G, Nobili F, Rocca G, DeCarli F, Gianelli MV, Rosadini G(1998) Quantitative electroencephalography and regional cerebralblood flow: discriminant analysis between Alzheimer’s patients andhealthy controls. Dement Geriatr Cogn Disord 9:238–274.

odriguez G, Copello F, Nobili F, Vitali P, Perego G, Nobili F (1999a)EEG spectral profile to stage Alzheimer’s disease. Clin Neuro-physiol 110:1831–1837.

odriguez G, Nobili F, Copello F, Vitali P, Gianelli MV, Taddei G,Catsafados E, Mariani G (1999b) 99mTc-HMPAO regional cere-bral blood flow and quantitative electroencephalography in Alzhei-mer’s disease: a correlative study. J Nucl Med 40:522–529.

odriguez G, Vitali P, De Leo C, De Carli F, Girtler N, Nobili F (2002)Quantitative EEG changes in Alzheimer patients during long-termdonepezil therapy. Neuropsychobiology 46:49–56.

oman GC, Tatemichi TK, Erkinjuntti T, Cummings JL, Masdeu JC,Garcia JH, Amaducci L, Orgogozo JM, Brun A, Hofman A, et al.(1993) Vascular dementia: diagnostic criteria for research studies.Report of the NINDS-AIREN International Workshop. Neurology43(2):250–260.

ombouts SA, Barkhof F, Witter MP, Scheltens P (2000) Unbiasedwholebrain analysis of grey matter loss in Alzheimer’s disease.Neurosci Lett 285:231–233.

osen WG, Terry RD, Fuld PA, Katzman R, Peck A (1980) Patholog-ical verification of ischemic score in differentiation of dementias.Ann Neurol 7(5):486–488.

ossini PM, Desiato MT, Lavaroni F, Caramia MD (1991) Brain excit-ability and electroencephalographic activation: non-invasive eval-uation in healthy humans via transcranial magnetic stimulation.Brain Res 567(1):111–119.

aletu B, Anderer P, Saletu-Zyhlarz GM, Pascual-Marqui RD (2002)EEG topography and tomography in diagnosis and treatment ofmental disorders: evidence for a key-lock principle. Methods FindExp Clin Pharmacol 24 (Suppl D):97–106.

astre M, Calero M, Pawlik M, Mathews PM, Kumar A, Danilov V,Schmidt SD, Nixon RA, Frangione B, Levy E (2004) Binding ofcystatin C to Alzheimer’s amyloid beta inhibits in vitro amyloid fibrilformation. Neurobiol Aging 25(8):1033–1043.

chreiter-Gasser U, Gasser T, Ziegler P (1993) Quantitative EEGanalysis in early onset Alzheimer’s disease: a controlled study.Electroencephalogr Clin Neurophysiol 86(1):15–22.

cott TM, Tucker KL, Bhadelia A, Benjamin B, Patz S, Bhadelia R,Liebson E, Price LL, Griffith J, Rosenberg I, Folstein MF (2004)Homocysteine and B vitamins relate to brain volume and white-matter changes in geriatric patients with psychiatric disorders.Am J Geriatr Psychiatry 12(6):631–638.

ekihara K, Sahani M, Nagarajan SS (2005) Localization bias andspatial resolution of adaptive and non-adaptive spatial filters forMEG source reconstruction. Neuroimage 25(4):1056–1067.

emba K, Fibiger HC (1989) Organization of central cholinergic sys-tems. Prog Brain Res 79:37–63.

eshadri S, Beiser A, Selhub J, Jacques PF, Rosenberg IH,

D’Agostino RB, Wilson PW, Wolf PA (2002) Plasma homocysteine

S

S

S

S

S

S

S

S

S

T

T

T

T

V

V

V

Y

W

Z

Z

B. Claudio et al. / Neuroscience 145 (2007) 942–954954

as a risk factor for dementia and Alzheimer’s disease. N EnglJ Med 346(7):476–483.

hea TB, Lyons-Weiler J, Rogers E (2002) Homocysteine, folatedeprivation and Alzheimer neuropathology. J Alzheimers Dis4(4):261–267.

loan EP, Fenton GW, Kennedy NSJ, MacLennan JM (1995) Electro-encephalography and single photon emission computed tomogra-phy in dementia: a comparative study. Psychol Med 25:631–638.

oikkeli R, Partanen J, Soininen H, Paakkonen A, Riekkinen P Sr(1991) Slowing of EEG in Parkinson’s disease. ElectroencephalogrClin Neurophysiol 79(3):159–165.

teriade M (2003) Neuronal substrates of sleep and epilepsy, p 522.Cambridge, UK: Cambridge University Press.

teriade M, Llinas RR (1988) The functional states of the thalamusand the associated neuronal interplay. Physiol Rev 68(3):649–742.

tewart GR, Frederickson CJ, Howell GA, Gage FH (1984) Cholin-ergic denervation-induced increase of chelatable zinc in mossy-fiber region of the hippocampal formation. Brain Res 290(1):43–51.

zava S, Valdes P, Biscay R, Galan L, Bosch J, Clark I, Jimenez JC(1994) High resolution quantitative EEG analysis. Brain Topogr6(3):211–219.

zelies B, Grond M, Herholz K, Kessler J, Wullen T, Heiss WD (1992)Quantitative EEG mapping and PET in Alzheimer’s disease. J Neu-rol Sci 110:46–56.

zelies B, Mielke R, Kessler J, Heiss WD (1999) EEG power changesare related to regional cerebral glucose metabolism in vasculardementia. Clin Neurophysiol 110(4):615–620.

alairach J, Tournoux P (1988) Co-planar stereotaxic atlas of thehuman brain. Stuttgart: Thieme.

anaka H, Koenig T, Pascual-Marqui RD, Hirata K, Kochi K, LehmannD (2000) Event-related potential and EEG measures in Parkinson’sdisease without and with dementia. Dement Geriatr Cogn Disord

11(1):39–45.

aupin P, Ray J, Fischer WH, Suhr ST, Hakansson K, Grubb A, GageFH (2000) FGF-2-responsive neural stem cell proliferation requiresCCg, a novel autocrine/paracrine cofactor. Neuron 28(2):385–397.

esta C, Laakso MP, Sabattoli F, Rossi R, Beltramello A, Soininen H,Frisoni GB (2004) A comparison between the accuracy of voxel-based morphometry and hippocampal volumetry in Alzheimer’sdisease. J Magn Reson Imaging 19(3):274–282.

aldes P, Picton TW, Trujillo N, Bosch J, Aubert E, Riera J (1998)Constraining EEG-MEG source imaging with statistical neuroanat-omy. Neuroimage 4:635.

ester B, Rasmussen K (1991) High performance liquid chromatog-raphy method for rapid and accurate determination of homocys-teine in plasma and serum. Eur J Clin Biochem 29:549–554.

inters HV, Nishimura GS, Secor DL, Pardridge WM (1990) Immuno-reactive A4 and gamma-trace peptide colocalization in amyloidoticarteriolar lesions in brains of patients with Alzheimer’s disease.Am J Pathol 137(2):233–240.

esavage JA, Brink TL, Rose TL, Lum O, Huang V, Adey M, Leirer VO(1983) Development and validation of a geriatric depression screen-ing scale: a preliminary report. J Psychiatr Res 17(1):37–49.

olf H, Jelic V, Gertz H-J, et al. (2003) A critical discussion of the roleof neuroimaging in mild cognitive impairment. Acta Neurol Scand107(Suppl 179):52–76.

appoli R, Versari A, Paganini M, Arnetoli G, Muscas GC, GangemiPF, Arneodo MG, Poggiolini D, Zappoli F, Battaglia A (1995) Brainelectrical activity (quantitative EEG and bit-mapping neurocogni-tive CNV components), psychometrics and clinical findings in pre-senile subjects with initial mild cognitive decline or probable Alzhe-imer-type dementia. Ital J Neurol Sci 16(6):341–376.

audig M (1992) A new systematic method of measurement anddiagnosis of “mild cognitive impairment” and dementia according toICD-10 and DSM-III-R criteria. Int Psychogeriatr 4(Suppl 2):

203–219.

(Accepted 19 December 2006)(Available online 22 February 2007)