Hypercapnia affects the functional coupling of resting state electroencephalographic rhythms and cerebral haemodynamics in healthy elderly subjects and in patients with amnestic mild

Clinical Neurophysiology xxx (2013) xxx–xxx

Contents lists available at ScienceDirect

Clinical Neurophysiology

journal homepage: www.elsevier .com/locate /c l inph

Hypercapnia affects the functional coupling of resting stateelectroencephalographic rhythms and cerebral haemodynamicsin healthy elderly subjects and in patients with amnestic mild cognitiveimpairment

1388-2457/$36.00 � 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.http://dx.doi.org/10.1016/j.clinph.2013.10.002

⇑ Corresponding author at: Department of Physiology and Pharmacology,University of Rome ‘‘La Sapienza’’, Rome, Italy. Tel.: +39 06 49910989; fax: +3906 49910917.

E-mail address: [email protected] (C. Babiloni).

Please cite this article in press as: Babiloni C et al. Hypercapnia affects the functional coupling of resting state electroencephalographic rhythms anbral haemodynamics in healthy elderly subjects and in patients with amnestic mild cognitive impairment. Clin Neurophysiol (2013), http://dx.d10.1016/j.clinph.2013.10.002

Claudio Babiloni a,b,⇑, Fabrizio Vecchio b,c, Riccardo Altavilla d, Francesco Tibuzzi c, Roberta Lizio b,Claudia Altamura d, Paola Palazzo d, Paola Maggio d, Francesca Ursini d, Matilde Ercolani c,Andrea Soricelli e,g, Giuseppe Noce e, Paolo Maria Rossini b,f, Fabrizio Vernieri d

a Department of Physiology and Pharmacology, University of Rome ‘‘La Sapienza’’, Rome, Italyb IRCCS San Raffaele Pisana, Rome, Italyc A.Fa.R. Dip. Neurosci, Ospedale ‘San Giovanni Calibita’ Fatebenefratelli, Isola Tiberina, Rome, Italyd Unità di Neurologia, Università Campus Bio-Medico di Roma, Rome, Italye IRCCS SDN, Napoli, Italyf Department of Geriatrics, Neuroscience & Orthopedics, Institute of Neurology Catholic University ‘‘Sacro Cuore’’, Rome, Italyg Department of Studies of Institutions and Territorial Systems, University of Naples Parthenope, Naples, Italy

a r t i c l e i n f o

Article history:Accepted 3 October 2013Available online xxxx

Keywords:Hypercapnia (CO2)Amnesic mild cognitive impairment (MCI)Resting stateElectroencephalography (EEG)Near-infrared spectroscopy (NIRS)

a b s t r a c t

Objective: Cerebral vasomotor reactivity (VMR) and coherence of resting state electroencephalographic(EEG) rhythms are impaired in Alzheimer’s disease (AD) patients. Here we tested the hypothesis thatthese two variables could be related.Methods: We investigated VMR and coherence of resting state EEG rhythms in nine normal elderly (Nold)and in 10 amnesic mild cognitive impairment (MCI) subjects. Resting state eyes-closed EEG data wererecorded at baseline pre-CO2 (ambient air, 2 min), during 7% CO2/air mixture inhalation (hypercapnia,90 s) and post-CO2 (ambient air, 2 min) conditions. Simultaneous frontal bilateral near-infrared spectros-copy (NIRS) was performed to assess VMR by cortical oxy- and deoxy-haemoglobin concentrationchanges. EEG coherence across all electrodes was computed at 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), beta 2 (20–30 Hz) and gamma (30–40 Hz) bands.Results: In Nold subjects, ‘total coherence’ of EEG across all frequency bands and electrode pairsdecreased during hypercapnia, with full recovery during post-CO2. Total coherence resulted lower inpre-CO2 and post-CO2 and presented poor reactivity during CO2 inhalation in MCI patients comparedwith Nold subjects. Hypercapnia increased oxy-haemoglobin and decreased deoxy-haemoglobin concen-trations in both groups. Furthermore, the extent of changes in these variables during CO2 challenge wascorrelated with the EEG coherence, as a reflection of neurovascular coupling.Conclusions: Hypercapnia induced normal frontal VMR that was detected by NIRS in both Nold andamnesic MCI groups, while it produced a reactivity of global functional coupling of resting state EEGrhythms only in the Nold group.Significance: In amnesic MCI patients, global EEG functional coupling is basically low in amplitude anddoes not react to hypercapnia.� 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights

reserved.

1. Introduction (single-domain) or with (multi-domain) decline of other cognitive

Elderly subjects with amnesic mild cognitive impairment (MCI)are characterised by an objective impairment of memory, without

functions (Petersen et al., 2001; Petersen, 2004; Petersen andNegash, 2008). In these subjects, the impairment of cognitive andsocial skills does not yet fulfil the clinical picture of dementia(Petersen et al., 2001; Petersen, 2004; Petersen and Negash,2008). However, the amnesic MCI condition is a relevant risk factorfor Alzheimer’s disease (AD) and could be considered its prodromalform (Bachman et al., 1993; Gao et al., 1998; Galluzzi et al., 2001;

d cere-oi.org/

http://dx.doi.org/10.1016/j.clinph.2013.10.002

mailto:[email protected]


http://www.sciencedirect.com/science/journal/13882457

http://www.elsevier.com/locate/clinph



2 C. Babiloni et al. / Clinical Neurophysiology xxx (2013) xxx–xxx

Petersen et al., 2001; Scheltens et al., 2002; Arnaiz and Almkvist,2003; Fisk et al., 2003; Frisoni et al., 2004), especially when MCIpatients display cerebrospinal fluid, structural and molecular neu-roimaging biomarkers of this pathological condition (Albert et al.,1991).

Nowadays, no single biomarker allows a clear-cut early diagno-sis of AD. Furthermore, some licensed biomarkers are too invasive(e.g., cerebrospinal fluid) and/or expensive (e.g., molecular bio-markers based on positron emission tomography (PET)) to beexamined in serial recordings in elderly subjects at risk of AD.For these reasons, other candidate biomarkers are under evalua-tion to simplify early diagnosis and disease progression of AD. Inthis line, an electroencephalogram (EEG) probes general neuro-physiological mechanisms of cortical neural synchronisation firingallowing temporal summation of post-synaptic potentials at thebasis of scalp EEG rhythms and seems to be an ideal candidatefor understanding the progressive loss of connectivity across theneuronal circuitry related to AD neurodegeneration. When com-pared to normal elderly (Nold) subjects, AD patients were charac-terised by high power density of delta (0–4 Hz) and theta (4–7 Hz)rhythms and low power density of posterior alpha (8–12 Hz) and/or beta (13–30 Hz) rhythms (Dierks et al., 1993, 2000; Huang et al.,2000; Ponomareva et al., 2003; Jeong, 2004; Babiloni et al., 2004).Furthermore, posterior alpha rhythms showed a decrement of thepower density in MCI compared to Nold subjects (Zappoli et al.,1995; Elmstahl and Rosen, 1997; Huang et al., 2000; Jelic et al.,2000; Koenig et al., 2005a,b; Babiloni et al., 2006b; for a reviewsee Rossini et al. (2007)).

The above results on the power density of resting state eyes-closed EEG rhythms have led to the widely supported hypothesisthat neuronal networks of time-coordinated brain activity acrossdifferent regional brain structures underpin cognitive functionand denote AD neurodegeneration (see D’Amelio and Rossini(2012) for a review). Failure of integration within a network maylead to cognitive dysfunction in prodromal and clinically definedAD; thus, AD can be viewed, at least, in part, as a disconnectionsyndrome (Bokde et al., 2009). In this theoretical framework, thespectrum of EEG power density per se may not fully capture theimpairment of functional neural connectivity. More specific mark-ers of functional neural connectivity may be derived from the mea-surement of the functional coupling of resting state eyes-closedEEG rhythms between pairs of electrodes. The linear componentsof such coupling, functional coordination and mutual informationexchange can be evaluated by the analysis of EEG spectral coher-ence (Gerloff et al., 1998; Gevins et al., 1998; Thatcher et al.,1986; Rappelsberger and Petsche, 1988). Spectral coherence is anormalised value that quantifies the temporal synchronisation oftwo EEG time series between pairs of electrodes in the frequencydomain of the oscillations. It can be derived by fast Fourier trans-form (FFT) (Rappelsberger and Petsche, 1988; Pfurtscheller and An-drew, 1999), and its theoretical assumption is based on theobservation that when the oscillatory activity of two cortical areasis functionally coordinated, their EEG rhythms show linear correla-tion and high spectral coherence. In general, decreased coherencereflects reduced linear functional coupling and information trans-fer (i.e., functional uncoupling or unbinding) among cortical areasor the reduced modulation of common areas by a third region.On the contrary, an increase in EEG coherence values can be inter-preted as an enhancement of the linear functional connections andinformation transfer (i.e., functional coupling or binding), thusreflecting the interaction of different cortical structures for a giventask. Increased coherence in alpha or higher frequencies reflects agreater ‘facilitation’ or connectivity, while in delta frequencies itsuggests an increase of ‘inhibition’ or a functional disconnection.It has been demonstrated that EEG spectral coherence is enhancedfollowing perceptive, cognitive and motor processes in the cortical

Please cite this article in press as: Babiloni C et al. Hypercapnia affects the funcbral haemodynamics in healthy elderly subjects and in patients with amnesti10.1016/j.clinph.2013.10.002

regions involved in task-related processing (Sauseng et al., 2005;Babiloni et al., 2006c; Vecchio et al., 2007, 2010, 2012) as a func-tion of the extension and type of the engaged neural networks(Pfurtscheller and Lopes da Silva, 1999; von Stein and Sarnthein,2000). In addition, spectral coherence may reflect the integrity ofcortical neural pathways (Locatelli et al., 1998).

Functional coupling of resting state eyes-closed cortical EEGrhythms differs across Nold, MCI and AD subjects. Previous EEGstudies mainly reported a prominent decrease in the coherenceof alpha rhythms in AD subjects compared with Nold subjects(Cook and Leuchter, 1996; Jelic et al., 1997, 2000; Almkvist et al.,2001; Locatelli et al., 1998; Wada et al., 1998a,b; Knott et al.,2000; Adler et al., 2003; Leuchter et al., 1987, 1992). This effectwas also associated with ApoE-related genetic risk and washypothesised to be mediated by a cholinergic deficit (Jelic et al.,1997). However, previous studies showed contradictory results,with either a decrease or an increase in low-band EEG coherenceof delta and theta rhythms (Locatelli et al., 1998; Adler et al.,2003; Leuchter et al., 1987; Brunovsky et al., 2003). A recent studyhas reconciled these conflicting results by computing ‘total coher-ence’, obtained by averaging the EEG spectral coherence across allcombinations of electrode pairs (Babiloni et al., 2010b). It was re-ported that delta total coherence gradually decreases from AD toMCI to Nold subjects (Babiloni et al., 2010b). Furthermore, alphatotal coherence was lower in AD than in MCI and Nold subjects.Of note, these EEG coherence values were negatively correlatedto (moderate to high) cholinergic lesions across MCI subjects (Babi-loni et al., 2010b).

What are the physiological mechanisms subtending the abnor-mal generation of the resting state EEG rhythms in manifest ADand in amnesic MCI patients? One possible mechanism is an abnor-mal neurovascular coupling and, in particular, an impairment ofcerebral haemodynamics in terms of vasomotor reactivity (VMR)to hypoxia and hypercapnia. Hypoxia, even in vitro, can increasethe production of amyloid beta (Ab) in different cell types (for a re-view see Peers et al., 2009). Hypoxia was also described to alterexpression of the Ab precursor, amyloid precursor protein (APP),as well as of secretase enzymes, which cleave Ab from APP (Peerset al., 2009). Other studies suggest that hypoxia increases Ab levelsby means of a reduced Ab degradation. Such variability may beattributable to cell-specific responses to hypoxia (Peers et al.,2009). At a macroscopic level, the effect of hypoxia on resting stateEEG was studied in healthy subjects by the inhalation of low oxy-gen–gas mixtures (Gastaut et al., 1961; Burykh, 2005; Schellart andReits, 2001; Thiebauld et al., 1983; Ozaki et al., 1995; Forster et al.,1975; Selvamurthy et al., 1978; Shi et al., 1987). A slowing of rest-ing state EEG rhythms in terms of increased delta/theta and de-creased alpha oscillations has been reported repeatedly (Gastautet al., 1961; Burykh, 2005; Schellart and Reits, 2001; Thiebauldet al., 1983; Ozaki et al., 1995; Forster et al., 1975; Selvamurthyet al., 1978; Shi et al., 1987), which nicely resembles the EEG ‘slow-ing’ observed in AD patients.

The ability of the neurovascular unit to locally augment bloodflow during neuronal activation is termed ‘neurovascular coupling’and is supposed to be compromised in AD patients even at prodro-mal stages (Claassen et al., 2009). It is debated whether an abnor-mal neurovascular coupling is the result of a decreased demand bythe impaired neural tissue or of a compromised cerebral supply ofcerebral blood flow (CBF) in activated areas (Nicolakakis andHamel, 2011). An abnormal neurovascular coupling may activatehypoxia-sensitive pathways culminating in the up-regulation ofproteins, such as Ab and transforming growth factor-b1 (TGF-b1)(Sun et al., 2006), which have been shown to be detrimental forcerebrovascular structure and function (Tong et al., 2005; Hanet al., 2008). Cerebral VMR, the peculiar capability of brain arteri-oles to dilate further after a stimulus, represents the amount of

tional coupling of resting state electroencephalographic rhythms and cere-c mild cognitive impairment. Clin Neurophysiol (2013), http://dx.doi.org/



C. Babiloni et al. / Clinical Neurophysiology xxx (2013) xxx–xxx 3

blood supply the brain tissue has at its disposal in critical situa-tions (Vernieri et al., 1999). Cerebral VMR to hypercapnia (7%CO2/air mixture) has been successfully investigated by near-infra-red spectroscopy (NIRS) (Smielewski et al., 1995; Vernieri et al.,2004). NIRS evaluates CBF and VMR, by revealing cortical arterioleand capillary modifications, in a completely non-invasive way.Functional MRI recordings during a CO2 challenge showed im-paired cerebral VMR in MCI and AD patients at the individual level(Cantin et al., 2011). Hypercapnia induced an increased flow pulsa-tility and a lower VMR in patients with cerebrovascular dementiaand AD compared with controls, as indicators of an impairmentof cerebral microvasculature circulation (Silvestrini et al., 2006;Vicenzini et al., 2007).

Summarising, previous studies have shown that EEG coherenceis impaired in amnesic MCI and AD patients (Babiloni et al., 2010a,b) and that cerebral VMR is altered in AD patients (Silvestrini et al.,2006). In this exploratory study, the effect of hypercapnia on EEGcoherence and CBF was, for the first time, investigated in Noldand in amnesic MCI subjects. Resting state eyes-closed EEG datawere recorded before, during and after hypercapnia with 7% CO2/air mixture inhalation. Simultaneous frontal NIRS was performedto assess cerebral VMR by cortical oxy- and deoxy-haemoglobinconcentration changes. EEG coherence was computed for the stan-dard frequency bands from delta to gamma and the correlation be-tween EEG coherence and NIRS variables was computed to test thehypothesis that the link between these variables reflects neurovas-cular coupling in both Nold and amnesic MCI subjects.

2. Methods

2.1. Subjects and diagnostic criteria

A total of 10 Nold subjects (five males; age = 70.8 ± 2.5 standarderror, SE; MMSE score = 29.1 ± 0.7 SE) and 10 amnesic MCI patients(six males; age = 70.7 ± 2.43 SE; MMSE score = 25.4 ± 1.02 SE) wererecruited. Data obtained from one Nold case could not be used dueto a technical failure; hence, the EEG data analysis was developedin nine Nold subjects. All subjects underwent a simultaneousrecording of EEG and NIRS data before, during and after inhalationof air enriched with 7% of CO2.

The local institutional ethics committee approved the study. Allexperiments were performed with the informed and overt consentof each participant or caregiver, in line with the Code of Ethics ofthe World Medical Association (Declaration of Helsinki) and thestandards established by the Author’s Institutional Review Board.

The inclusion and exclusion criteria for MCI were based on pre-vious seminal reports (Albert et al., 1991; Devanand et al., 1997;Flicker et al., 1991; Petersen et al., 1995, 1997, 2001; Rubin et al.,1989; Zaudig, 1992; Portet et al., 2006). Summarising, the inclusioncriteria were as follows: (i) objective memory impairment onneuropsychological evaluation, as defined by performances P1.5standard deviation below the mean value of age- and education-matched controls for a neuropsychological test battery of neuro-psychological tests, to assess cognitive performance in the domainsof memory (i.e., Buschke-Fuld and Memory Rey tests), language,executive function/attention and visuo-construction; (ii) noinvolvement of other cognitive domains; (iii) normal activities ofdaily living as documented by medical history and evidence ofindependent living; and (iv) clinical dementia rating score of 0.5.The exclusion criteria included: (i) mild AD, as diagnosed by stan-dard protocols including NINCDS-ADRDA (McKhann et al., 1994);(ii) evidence (including diagnostic MRI procedures) of concomitantdementia such as frontotemporal, vascular dementia, reversibledementias (including depressive pseudodementia), marked fluctu-ations in cognitive performance compatible with Lewy body


dementia and/or features of mixed dementias; (iii) evidence ofconcomitant extra-pyramidal symptoms; (iv) clinical and indirectevidence of depression as revealed by Geriatric Depression Scalescores higher than 13; (v) other psychiatric diseases, epilepsy, drugaddiction, alcohol abuse (as revealed by a psychiatric interview)and use of psychoactive drugs including acetylcholinesteraseinhibitors or other drugs enhancing brain cognitive functions;and (vi) current or previous uncontrolled or systemic diseases(including diabetes mellitus and atrial fibrillation) or traumaticbrain injuries. Furthermore, subjects affected by carotid steno-occlusive disease were excluded, as previous studies reported acorrelation among carotid stenosis/occlusion, alteration in VMRand cognitive status in humans (Marshall et al., 2012; Silvestriniet al., 2009; Balucani et al., 2012). Of note, benzodiazepines, anti-depressant and/or anti-hypertensive drugs were withdrawn about24 h before the EEG recordings.

Neuropsychological tests were performed to assess cognitiveperformances in the domains of memory, language, executivefunction/attention and visuo-construction abilities. We followedthe general international field standards as the clinical data collec-tion was not specifically focussed on the topic of the present study(i.e., occipital alpha rhythms). Immediate and delayed recall mea-sure of the Rey Auditory Verbal Learning Test (Rey, 1958; Carlesi-mo et al., 1996), the delayed recall of Rey figures (Rey, 1968), thedelayed recall of a three-word list (Chandler et al., 2004) and thedelayed recall of a story (Spinnler and Tognoni, 1987) were usedto assess memory. Language was tested with the 1 min verbal flu-ency for letters (Novelli, 1986); the 1 min verbal fluency for fruits,animals or car trades (Novelli, 1986); and the Token test (De Renziand Vignolo, 1962; Spinnler and Tognoni, 1987). Trail Making Testparts A and B (Reitan, 1958) and the Digit Forward and the DigitBackward (Orsini et al., 1987) tests were used to evaluate execu-tive function and attention. Finally, visuo-construction perfor-mances were tested with Rey figures’ copying (Rey, 1968),Raven’s progressive matrices (Raven, 1965) and the Clock Drawingtest (Shulman et al., 1993).

Nold subjects underwent physical and neurological examina-tions as well as cognitive screening.

2.2. Experimental recordings

EEG recordings were performed by a specialised clinical unit ina dimly lit room at late morning. During the EEG recording, sub-jects were seated on a comfortable reclined chair. EEG data werecollected (0.3–70 Hz bandpass; cephalic reference) from 19 elec-trodes positioned according to the International 10–20 System(Fig. 1; i.e., Fp1, Fp2, F7, F3, Fz, F4, F8, T3, C3, Cz, C4, T4, T5, P3,Pz, P4, T6, O1 and O2). These recordings were performed duringstandard resting state eyes-closed (2 min) and eyes-open (2 min)conditions as well as during the hypercapnia session. In the hyper-capnia session, the EEG recordings were performed under the fol-lowing conditions: baseline (respiration of ambient air for 2 min;pre-CO2), during the respiration of air enriched with 7% of CO2

(90 s; CO2 condition) and during the recovery period (respirationof ambient air for 2 min; Post-CO2). To monitor eye movements,the horizontal and vertical electro-oculography (0.3–70 Hz band-pass) data were also collected. All data were digitised in a contin-uous recording mode. In order to keep vigilance constant, theexperimenters controlled the subject and the EEG traces continu-ously. The recording unit personnel were familiar with the issueof vigilance in resting state elderly subjects. The experimentersverbally alerted the subject any time when behavioural and/orEEG drowsiness appeared, especially during the eyes-closedcondition.

To probe the neurovascular coupling during the EEG experi-ment, we simultaneously performed NIRS measurements from left




Fig. 1. Grand average of the total coherence spectra averaged across all electrode pairs for the two groups of subjects (Nold, MCI) in the three experimental conditions (pre-CO2, CO2, post-CO2). The Nold group shows maximum total coherence at alpha 1. Both groups show a slight reduction of total coherence at several frequency bands in the CO2

condition.


and right frontal locations overlying the prefrontal cortex. Despitethe complexity of the combined experimental procedures (i.e., EEGand NIRS), clean records were successfully carried out and the NIRSsignals could be obtained in 10 MCI patients and in six Nold sub-jects. The mentioned NIRS measurements were carried out withan ISS oximeter (Model 96208 Two-channel Non-Invasive TissueOximeter, ISS Inc., Champaign, IL (Fantini et al., 1995)). The ISSOximeter allows the measurement of oxygenated (HbO2) anddeoxygenated (Hb) haemoglobin concentrations in tissues. Thedevice works by beaming NIR light into tissues at four knowndistances (2, 2.5, 3 and 3.5 cm, respectively) and collecting thebackscattered light through optical fibre cables. Light of two differ-ent wavelengths (750 and 825 nm), produced by solid-state lasers,is used for this purpose. Light is modulated at a frequency of110 MHz to allow the measurements of phase and modulation ofthe collected light. From these raw data, the absorption and scat-tering coefficients of the medium are determined. Assuming thatin the brain tissue haemoglobin is the only significant absorbingmolecule which varies over time, once the absorption and scatter-ing coefficients are determined, the oxygenated and deoxygenatedhemoglobin concentrations are calculated.

NIRS recordings allowed monitoring the concentration changesof HbO2, Hb and, therefore, total haemoglobin content (THC), in theperiods corresponding to the baseline (pre-CO2), the CO2 conditionand post-CO2. CO2 values for NIRS parameters were obtained after90 s of 7% CO2 in air mixture inhalation. We used only the HbO2

and Hb values for subsequent analyses. Cerebral VMR was consid-ered as the percent increase in values of HbO2 and Hb from base-line to the end of the CO2 inhalation period (CO2 value)according to the following formula: ðValueCO2 � ValueBaseline=

ValueBaselineÞ � 100 (Vernieri et al., 2004).

2.3. Preprocessing of the EEG data

The recorded EEG data were fragmented offline in consecutiveepochs of 2 s. The EEG epochs with ocular, muscular and othertypes of artefacts were preliminarily identified by a computerisedautomatic procedure. EEG epochs with sporadic blinking artefacts(<10% of the total) were corrected by an autoregressive method.Two independent experimenters, blind to the diagnosis, manuallyconfirmed the EEG segments accepted for further analysis.

2.4. Spectral analysis of the EEG data

The digital FFT-based power spectrum analysis (Welch tech-nique, Hanning windowing function and no phase shift) was eval-uated in order to calculate the individual alpha frequency (IAF)


peak, defined as the frequency associated with the strongest EEGpower at the extended alpha range (Klimesch, 1999). The meanIAF peak was 9.3 Hz (±0.2 SE) in the MCI subjects, 9.1 Hz (±0.2SE) in the Nold subjects. No statistically significant ANOVA differ-ence was found between these two groups (p > 0.05). Nevertheless,the IAF peak was used as a covariate (together with age, education,gender and MMSE) in the statistics. Indeed, the IAF frequency isvery important, as it is associated with the maximum power ofresting eyes-closed EEG rhythms (Klimesch, 1999). The above pro-cedure minimised the possibility that small differences in the IAFpeak could confound the comparisons between the MCI and theNold groups.

2.5. Total coherence analysis

EEG spectral coherence is a normalised measure of the couplingbetween two EEG signals at any given frequency (Rappelsbergerand Petsche, 1988; Pfurtscheller and Andrew, 1999). The coherencevalues were calculated at each frequency bin as:

CohxyðkÞ ¼ jRxyðkÞj2 ¼jfxyðkÞj2

fxxðkÞfyyðkÞ:

The numerator contains the cross-spectrum of two EEG signalsx and y (fxy) for a given frequency bin (k), while the denominatorcontains the respective autospectra for x(fxx) and y(fyy). The coher-ence value (Cohxy) is obtained by squaring the magnitude of thecomplex correlation coefficient R. This procedure produces a realnumber between 0 (no coherence) and 1 (max coherence).

As mentioned above, the global functional coupling of the EEGrhythms was indexed by the mean of spectral coherence for allcombinations of electrode pairs, namely the total coherence (Babi-loni et al., 2010a,b). The main steps for the computation of the totalcoherence were as follows:

(1) For each subject, the spectral coherence was calculated foreach pair of channels from all artefact-free EEG trials. Thefrequency bands of interest were delta (2–4 Hz), theta(4–8 Hz), alpha 1 (8–10 Hz), alpha 2 (10–13 Hz), beta 1(13–20 Hz), beta 2 (20–30 Hz) and gamma (31–40 Hz) incontinuity with a bulk of previous studies of this Consortiumon resting eyes-closed EEG rhythms in ageing (Babiloni et al.,2004, 2006a–e).

(2) For a given frequency band of interest, the coherence valuefor any EEG electrode was estimated as the average of thecoherence values between that electrode and each of theother 18 electrodes (‘electrode coherence’).





(3) For a given frequency band of interest, the ‘electrode coher-ence’ of the 19 electrodes was averaged to form the totalcoherence.

The choice of the fixed EEG bands did not account for the IAFpeak. However, this should not affect the results, as more than90% of the subjects had the IAF peak within the alpha 1 band (8–10.5 Hz).

2.6. Statistical analysis of the total coherence values

Statistical analysis aimed at evaluating the working hypothesisthat the total coherence of the resting state cortical EEG rhythms isreduced in magnitude in the amnesic MCI compared with the Noldgroup, possibly due to the inhalation of CO2. Total coherence valueswere considered as a dependent variable for an ANOVA design,with subjects’ age, education, gender, MMSE and IAF peak as covar-iates. The ANOVA factors (levels) were Group (Nold and MCI), Con-dition (pre-CO2, CO2 and post-CO2) and Band (delta, theta, alpha 1,alpha 2, beta 1, beta 2 and gamma). Mauchly’s test evaluated thesphericity assumption. Correction of the degrees of freedom wasmade with the Greenhouse–Geisser procedure. The Duncan testwas used for post hoc comparisons (p < 0.05). Specifically, theworking hypothesis would be confirmed by the following statisti-cal results: (i) a statistical ANOVA effect including the factor Group(p < 0.05) and (ii) a post hoc test indicating a statistically significantreduction of the total coherence values in the MCI compared withthe Nold subjects as a function of the mixture inspiration (Duncantest, p < 0.05).

Concerning the cerebral haemodynamics evaluation, weassessed VMR to hypercapnia in frontal areas by an ANOVA designusing HbO2 and Hb values as dependent variables. Any ANOVA de-sign included Group (Nold and MCI; independent variables), Time(pre-CO2, CO2 and post-CO2) and Hemisphere (left and right) asfactors. A significant interaction including the Group and the Timefactors was expected, indicating VMR in the CO2 condition(p < 0.05).

To probe the neurovascular coupling during the experiment, weperformed an exploratory correlation analysis between the totalEEG coherence during the CO2 period and the NIRS parameters(i.e., HbO2 and Hb) during the CO2 period in the subgroup of sub-jects in whom the NIRS measurements were successfully recorded(i.e., six Nold and 10 MCI subjects; Pearson test, p < 0.05). Wehypothesised a correlation between these NIRS and EEG variablesas a reflection of the underlying neurovascular interactions.

Fig. 2. ANOVA interaction between group and condition factors (Fig. 2, F(2,34) = 4.46; p < 0.019). The planned post hoc testing indicated that the totalcoherence across all electrode pairs and band frequencies is higher in magnitudein the Nold than in the MCI group during both pre- and post-CO2 conditions(p < 0.01). Compared to these conditions, the CO2 shows a greater reduction of thetotal coherence values in the Nold than in the MCI group (p < 0.0005).

3. Results

For illustrative purposes, Fig. 1 plots the grand average of thetotal coherence spectra averaged across all electrode pairs for thetwo groups of subjects (Nold and MCI) in the three experimentalconditions (pre-CO2, CO2 and post-CO2). In the pre-CO2 condition,the Nold group presented the highest total coherence in the alpha1 frequency, an intermediate total coherence in delta, theta and al-pha 2 frequencies, while the lowest magnitude values of the totalcoherence were observed in beta 1, beta 2 and gamma frequencies.In the CO2 condition, Nold subjects showed a reduction of the totalspectral coherence from delta to gamma, with a full recoveryduring the post-CO2 condition. Compared to Nold subjects, MCI pa-tients were characterised by lower total spectral coherence acrossfrequency bands in the baseline, pre-CO2 and in the post-CO2 con-dition, especially at the frequency bands including alpha, beta andgamma. Furthermore, MCI subjects showed a poor reactivity of thetotal coherence in the CO2 condition compared to the pre-CO2 andpost-CO2 conditions, especially at these high frequency bands.


The results showed statistically significant ANOVA interactionbetween the Group and the Condition factor (Fig. 2, F(2,34) = 4.46; p < 0.019). The planned post hoc analysis indicatedthat the total coherence across all electrode pairs and band fre-quencies is higher in magnitude in the Nold than in the MCIgroup during both pre- and post-CO2 conditions (p < 0.01). Com-pared to these conditions, the CO2 showed a greater reduction ofthe total coherence values in the Nold than in the MCI group(p < 0.0005).

Concerning the cerebral haemodynamics study, Fig. 3 illus-trates the grand average of the NIRS values of HbO2 and Hb con-centrations in the MCI and in the Nold group before the CO2

inspiration (pre-CO2 considered as a reference baseline period),during CO2 inhalation (CO2 condition) and after CO2 inspiration(post-CO2). It is worth nothing that that in both groups, theHbO2 and Hb mean values were stable in the period precedingthe CO2 condition. Afterwards, these values showed clear changespeaking at the end of CO2 inhalation. Finally, the HbO2 and Hbmean values recovered at the end of the CO2 period. The HbO2

and Hb mean values showed a similar time course and amplitudein the two groups. Two ANOVAs showed a main effect Time usingthe HbO2 (F(1,14) = 14.82) and Hb (F(1,14) = 23.69) values asdependent variables, regardless of the factors Group (Nold andMCI) and the Hemisphere (left and right). A post hoc test indi-cated that with respect to baseline, the CO2 condition was relatedto a significant increase of the HbO2 (p < 0.005) and to a decreaseof Hb (p < 0.0005).

Another analysis explored the correlation between the EEG totalcoherence and the NIRS indices (i.e., HbO2 and Hb,) during CO2

inspiration across all Nold and MCI subjects as a whole group.The main findings are as follows: both Right HbO2 and Hb percen-tual variations (VMR) were positively correlated with CO2 totalcoherence values (r = 0.54, p < 0.03 and r = 0.51, p < 0.04, respec-tively) as reported in Fig. 4, thus unveiling the underlying neuro-vascular interactions.




Fig. 3. Grand average of HbO2 and Hb values before the CO2 inhalation (pre-CO2,considered as the reference baseline period), during the inspiration (CO2), and afterCO2 inspiration (post-CO2) in the MCI and in the Nold group. These values show asimilar time course and amplitude in the two groups.


4. Discussion

In the present exploratory study, we tested the hypothesis thathypercapnia affects both total EEG coherence (i.e., coherenceacross the average of all combinations of the electrode pairs; Babi-loni et al., 2010a,b) and CBF, in Nold and in amnesic MCI subjects.Before discussing the results, it should be remarked that this studyis characterised by some methodological limitations. It was per-formed in relatively small groups of amnesic MCI patients and Noldsubjects. To minimise this disadvantage, we made the two groupsas homogeneous as possible for age, gender and education vari-ables. Furthermore, we had no access to subjects’ MR for quantita-tive analyses to correlate MR findings (i.e., vascular load in thewhite matter) and EEG variables. Moreover, the experimentalprocedures were engaging; hence, we decided not to involve ADpatients as a further experimental group at this stage. Finally, thesimultaneous recording of EEG and frontal NIRS was not techni-cally easy. In four Nold subjects, the quality of NIRS recordingswas insufficient to extract reliable values of cortical oxy-anddeoxy-haemoglobin (i.e., an effective placement of the NIRSsensors was prevented by EEG electrodes). This problem occurredin the first four examined subjects. Due to this observation, wechanged the NIRS probes for the further experiments. The newapproach improved the quality of recorded NIRS signals in allremaining subjects, namely six Nold and 10 MCI individuals. De-spite these limitations, this study allowed exploring the possibleeffects of a simple and robust experimental model of hypercapniaon the cortical neural synchronisation on the basis of resting stateEEG rhythms. The main results have an important heuristic valueto motivate future investigations in larger cohorts of Nold, amnesic


MCI and AD subjects, using measurements of structural MRI andscanning of regional CBF as revealed by SPECT/PET-FGD or MRIarterial spin label.

The main results of the present study provide a first interestingconfirmation of the working hypothesis. In the Nold group, ‘totalcoherence’ of EEG across all frequency bands and electrode pairsdecreased during the CO2 condition, with full recovery during thepost-CO2 condition. Compared with the Nold group, the MCI groupshowed lower ‘total coherence’ in the pre-CO2 and post-CO2 condi-tions and poor reactivity of this coherence in the CO2 condition.Cerebral VMR was normal and did not differ between the twogroups. Furthermore, cerebral VMR and EEG coherence indiceswere correlated in all Nold and MCI subjects unveiling a possiblemechanism of neurovascular coupling. Previous evidence demon-strated an impairment of cerebral VMR to hypercapnia in MCI sub-jects and in demented patients (Silvestrini et al., 2006; Vicenziniet al., 2007; Cantin et al., 2011). Our results extend the finding ofan altered response to CO2 also to the EEG functional coupling, asa reflection of cortical connectivity. Furthermore, the present find-ings enlarge previous evidence showing an amplitude reduction ofcerebral response in AD patients, which probably reflects animpairment of synaptic function produced by cerebral hypoxia(Adler, 1991).

At the present early stage of the research, we can just speculatethat these anomalies of cerebral VMR might either be a determi-nant or a consequence of the neurodegeneration process; whateverthe explanation, this abnormal mechanism chronically exposesMCI and therefore AD patients to the effects of hypoxia. Duringsleep, repeated hypoxic events might affect the respiratory cholin-ergic mechanism, respiratory regulation, upper airway patency andcerebral oxygenation (Peers et al., 2009; Scragg et al., 2005; Dau-latzai, 2010). During daytime, hypo-oxygenation of the brain mightbe also due to chronic cerebrovascular and vasomotor deficits, pro-voked by a poor cholinergic tone. These alterations may cause Abdeposition at the extracellular level, as well as neurofibrillarycytopathology in cholinergic and other neurons at the intracellularlevel (Peers et al., 2009; Scragg et al., 2005; Daulatzai, 2010).Future investigations should test experimentally the impact ofhypoxia on the brain of AD patients.

The results of the present study are in line with those showing aprominent decrease in alpha and beta coherence (Cook and Leuch-ter, 1996; Almkvist et al., 2001; Locatelli et al., 1998; Wada et al.,1998a,b; Knott et al., 2000; Adler et al., 2003; Leuchter et al.,1987, 1992; Jelic et al., 1997, 2000) and in total alpha coherencein AD (Babiloni et al., 2010a,b). The present results also extend tothe MCI condition the findings of previous EEG studies inducingmild acute hypoxia in healthy subjects by gas mixtures with lowoxygen concentration (Gastaut et al., 1961; Burykh, 2005; Schellartand Reits, 2001; Thiebauld et al., 1983; Ozaki et al., 1995; Forsteret al., 1975; Selvamurthy et al., 1978; Shi et al., 1987). In thosestudies, a general ‘slowing’ of the resting state EEG rhythms wasreported in terms of increased delta/theta power density and ofdecreased alpha power density (Gastaut et al., 1961; Burykh,2005; Schellart and Reits, 2001; Thiebauld et al., 1983; Ozakiet al., 1995; Forster et al., 1975; Selvamurthy et al., 1978; Shiet al., 1987).

Is neurovascular coupling one of the physiological mechanismsat the basis of the modulation of the resting state EEG rhythms inNold and in amnesic MCI subjects? A control analysis of the pres-ent study supports this hypothesis. During hypercapnia, a strictcorrelation between frontal haemodynamic processes (as revealedby NIRS) and the global functional coupling of the EEG rhythms inthe resting state condition (as revealed by total coherence) was ob-served in Nold and MCI subjects considered as a unique group. Thecontrol analysis showed that both Nold and MCI groups were char-acterised by an effective frontal VMR to hypercapnia, as revealed




Fig. 4. Scatterplot showing the correlation between total coherence values and HbO2 or Hb in the right hemisphere in all Nold and MCI subjects as a whole group. The r- andp-values relative to the Pearson correlation are reported within the diagram.


by the expected opposite variations of the oxy- and deoxy-haemo-globin to the CO2 challenge. Due to the small number of Nold andMCI subjects with a successful recording of frontal NIRS variables,the present results do not allow conclusive considerations aboutthe different effects of hypercapnia on these variables in compari-son between Nold and MCI subjects. However, the results of thisstudy suggest that in the stage of amnestic MCI, total EEG coher-ence in the resting state might be more sensitive to reveal an im-paired neurovascular coupling than haemodynamic changesdetected by NIRS.

In conclusion, the results of our study suggest that the com-bined recording of NIRS and EEG parameters before, during andafter hypercapnia may be considered a promising approach toquantitatively measure neurovascular coupling in elderly subjectsduring the resting state condition. In this line, hypercapnia may bean effective ‘challenge’ for brain function to develop and validateprocedures for diagnostic and prognostic purposes. Namely, it


can be hypothesised that amnesic MCI patients showing a weakerresponse of EEG total coherence to hypercapnia are more prone tosubsequently develop AD. These are fascinating working hypothe-ses for future studies.

Acknowledgements

This research was developed thanks to the financial support ofTosinvest Sanita’ (Cassino, Pisana). The study was partially sup-ported by PRIN2010–2011 (prot. 2010SH7H3F) entitled ‘CON-NAGE’. There are no conflicts of interest or financial disclosures.

References

Adler G. Hyperventilation as a model for acute ischaemic hypoxia of the brain:effects on cortical auditory evoked potentials. Eur Arch Psychiatry Clin Neurosci1991;240:367–9.


http://refhub.elsevier.com/S1388-2457(13)01103-6/h0005






Adler G, Brassen S, Jajcevic A. EEG coherence in Alzheimer’s dementia. J NeuralTransm 2003;110:1051–8.

Albert M, Smith LA, Scherr PA, Taylor JO, Evans DA, Funkenstein HH. Use of briefcognitive tests to identify individuals in the community with clinicallydiagnosed Alzheimer’s disease. Int J Neurosci 1991;57:167–78.

Almkvist O, Jelic V, Amberla K, Hellström-Lindahl E, Meurling L, Nordberg A.Responder characteristics to a single oral dose of cholinesterase inhibitor: adouble-blind placebo-controlled study with tacrine in Alzheimer patients.Dement Geriatr Cogn Disord 2001;12:22–32.

Arnaiz E, Almkvist O. Neuropsychological features of mild cognitive impairmentand preclinical Alzheimer’s disease. Acta Neurol Scand Suppl 2003;179:34–41.

Babiloni C, Binetti G, Cassetta E, Cerboneschi D, Dal Forno G, Del Percio C, et al.Mapping distributed sources of cortical rhythms in mild Alzheimers disease. amulti-centric EEG study. Neuroimage 2004;22:57–67.

Babiloni C, Binetti G, Cassarino A, Dal Forno G, Del Percio C, Ferreri F, et al. Sources ofcortical rhythms in adults during physiological aging: a multi-centric EEGstudy. Hum Brain Mapp 2006a;27:162–72.

Babiloni C, Binetti G, Cassetta E, Dal Forno G, Del Percio C, Ferreri F, et al. Sources ofcortical rhythms change as a function of cognitive impairment in pathologicalaging: a multi-centric study. Clin Neurophysiol 2006b;117:252–68.

Babiloni C, Benussi L, Binetti G, Bosco P, Busonero G, Cesaretti S, et al. Genotype(cystatin C) and EEG phenotype in Alzheimer disease and mild cognitiveimpairment: a multicentric study. Neuroimage 2006c;29:948–64.

Babiloni C, Benussi L, Binetti G, Cassetta E, Dal Forno G, Del Percio C, et al.Apolipoprotein E and alpha brain rhythms in mild cognitive impairment: amulticentric EEG study. Ann Neurol 2006d;59:323–34.

Babiloni C, Frisoni G, Steriade M, Bresciani L, Binetti G, Del Percio C, et al. Frontalwhite matter volume and delta EEG sources negatively correlate in awakesubjects with mild cognitive impairment and Alzheimer’s disease. ClinNeurophysiol 2006e;117:1113–29.

Babiloni C, Frisoni GB, Vecchio F, Pievani M, Geroldi C, De Carli C, et al. Globalfunctional coupling of resting EEG rhythms is related to white-matter lesionsalong the cholinergic tracts in subjects with amnesic mild cognitiveimpairment. J Alzheimers Dis 2010a;19:859–71.

Babiloni C, Visser PJ, Frisoni G, De Deyn PP, Bresciani L, Jelic V, et al. Cortical sourcesof resting EEG rhythms in mild cognitive impairment and subjective memorycomplaint. Neurobiol Aging 2010b;31:1787–98.

Bachman DL, Wolf PA, Linn RT, Knoefel JE, Cobb JL, Belanger AJ, et al. Incidence ofdementia and probable Alzheimer’s disease in a general population: theFramingham study. Neurology 1993;43:515–9.

Balucani C, Viticchi G, Falsetti L, Silvestrini M. Cerebral hemodynamics andcognitive performance in bilateral asymptomatic carotid stenosis. Neurology2012;79:1788–95.

Bokde AL, Ewers M, Hampel H. Assessing neuronal networks: understandingAlzheimer’s disease. Prog Neurobiol 2009;89:125–33.

Brunovsky M, Matousek M, Edman A, Cervena K, Krajca V. Objective assessment ofthe degree of dementia by means of EEG. Neuropsychobiology 2003;48:19–26.

Burykh EA. Relations of the EEG local and spatialtemporal spectral characteristicschanges under hypoxia in humans. Ross Fiziol Zh Im I M Sechenova2005;91:1260–80.

Cantin S, Villien M, Moreaud O, Tropres I, Keignart S, Chipon E, et al. Impairedcerebral vasoreactivity to CO2 in Alzheimer’s disease using BOLD fMRI.Neuroimage 2011;58:579–87.

Carlesimo GA, Caltagirone C, Gainotti G. The group for the standardization of themental deterioration battery. The mental deterioration battery: normative data,diagnostic reliability and qualitative analyses of cognitive impairment. EurNeurol 1996;36:378–84.

Chandler MJ, Lacritz LH, Cicerello AR, Chapman SB, Honig LS, Weiner MF, et al.Three-word recall in normal aging. J Clin Exp Neuropsychol 2004;26:1128–33.

Claassen JA, Diaz-Arrastia R, Martin-Cook K, Levine BD, Zhang R. Altered cerebralhemodynamics in early Alzheimer disease: a pilot study using transcranialDoppler. J Alzheimers Dis 2009;17:621–9.

Cook IA, Leuchter AF. Synaptic dysfunction in Alzheimer’s disease: clinicalassessment using quantitative EEG. Behav Brain Res 1996;78:15–23.

D’Amelio M, Rossini PM. Brain excitability and connectivity of neuronal assembliesin Alzheimer’s disease: from animal models to human findings. Prog Neurobiol2012;99:42–60.

Daulatzai MA. Conversion of elderly to Alzheimer’s dementia: role of confluence ofhypothermia and senescent stigmata – the plausible pathway. J Alzheimers Dis2010;21:1039–63.

De Renzi E, Vignolo LA. The token test: a sensitive test to detect receptivedisturbances in aphasics. Brain 1962;85:665–78.

Devanand DP, Folz M, Gorlyn M, Moeller JR, Stem J. Questionable dementia: clinicalcourse and predictors of outcome. J Am Geriatr Soc 1997;45:321–8.

Dierks T, Ihl R, Frolich L, Maurer K. Dementia of the Alzheimer type: effects on thespontaneous EEG described by dipole sources. Psychiatry Res 1993;50:51–162.

Dierks T, Jelic V, Pascual-Marqui RD, Wahlund LO, Julin P, Linden DEJ, et al. Spatialpattern of cerebral glucose metabolism (PET) correlates with localization ofintracerebral EEG-generators in Alzheimer’s disease. Clin Neurophysiol2000;111:1817–24.

Elmstahl S, Rosen I. Postural hypotension and EEG variables predict cognitivedecline: results from a 5-year follow-up of healthy elderly women. DementGeriatr Cogn Disord 1997;8:180–7.

Fantini S, Franceschini MA, Maier JS, Walker SA, Barbieri B, Gratton E. Frequency-domain multichannel optical detector for non-invasive tissue spectroscopy andoximetry. Opt Eng 1995;34:32–42.


Fisk JD, Merry HR, Rockwood K. Variations in case definition affect prevalence butnot outcomes of mild cognitive impairment. Neurology 2003;61:1179–84.

Flicker CS, Ferris H, Reisberg B. Mild cognitive impairment in the elderly: predictorsof dementia. Neurology 1991;41:1006–9.

Forster HV, Soto RJ, Dempsey JA, Hosko MJ. Effect of sojourn at 4300 m altitude onelectroencephalogram and visual evoked response. J Appl Physiol1975;39:109–13.

Frisoni GB, Padovani A, Wahlund LO. The predementia diagnosis of Alzheimerdisease. Alzheimer Dis Assoc Disord 2004;18:51–3.

Galluzzi S, Cimaschi L, Ferrucci L, Frisoni GB. Mild cognitive impairment: clinicalfeatures and review of screening instruments. Aging (Milano)2001;13:183–202.

Gao S, Hendrie HC, Hall KS, Hui S. The relationships between age, sex, and theincidence of dementia and Alzheimer disease: a meta-analysis. Arch GenPsychiatry 1998;55:809–15.

Gastaut H, Fischgold J, Meyer JS. EEG in acute cerebral hypoxia in man. In: Meyer JS,Gastaut H, editors. Cerebral hypoxia and the electroencephalogram. Springfield,IL: Thomas; 1961. p. 599–617.

Gerloff C, Richard J, Hadley J, Schulman AE, Honda M, Hallett M. Functional couplingand regional activation of human cortical motor areas during simple, internallypaced and externally paced finger movements. Brain 1998;121:1513–31.

Gevins A, Smith ME, Leong H, McEvoy L, Whitfield S, Du R, et al. Monitoring workingmemory load during computer-based tasks with EEG pattern recognitionmethods. Hum Factors 1998;40:79–91.

Han BH, Zhou ML, Abousaleh F, Brendza RP, Dietrich HH, Koenigsknecht-Talboo J,et al. Cerebrovascular dysfunction in amyloid precursor protein transgenicmice: contribution of soluble and insoluble amyloid-beta peptide, partialrestoration via gamma secretase inhibition. J Neurosci 2008;28:13542–50.

Huang C, Wahlund LO, Dierks T, Julin P, Winblad B, Jelic V. Discrimination ofAlzheimer’s disease and mild cognitive impairment by equivalent EEG sources:a cross-sectional and longitudinal study. Clin Neurophysiol 2000;11:1961–7.

Jelic V, Johansson SE, Almkvist O, Shigeta M, Julin P, Nordberg A, et al. Quantitativeelectroencephalography in mild cognitive impairment: longitudinal changesand possible prediction of Alzheimer’s disease. Neurobiol Aging2000;21:533–40.

Jelic V, Julin P, Shigeta M, Nordberg A, Lannfelt L, Winblad B, et al. Apolipoprotein Eepsilon4 allele decreases functional connectivity in Alzheimer’s disease asmeasured by EEG coherence. J Neurol Neurosurg Psychiatry 1997;63:59–65.

Jeong J. EEG dynamics in patients with Alzheimer’s disease. Clin Neurophysiol2004;115:1490–505.

Klimesch W. EEG alpha and theta oscillations reflect cognitive and memoryperformance: a review and analysis. Brain Res Brain Res Rev 1999;29:169–95.

Knott V, Mohr E, Mahoney C, Ilivitsky V. Electroencephalographic coherence inAlzheimer’s disease: comparisons with a control group and population norms. JGeriatr Psychiatry Neurol 2000;13:1–8.

Koenig T, Prichep L, Dierks T, Hubl D, Wahlund LO, John ER, et al. Decreased EEGsynchronization in Alzheimer’s disease and mild cognitive impairment.Neurobiol Aging 2005a;26:165–71.

Koenig T, Prichep L, Dierks T, Hubl D, Wahlund LO, John ER, et al. EEGsynchronization in Alzheimer’s disease and mild cognitive impairment.Neurobiol Aging 2005b;26:165–71.

Leuchter AF, Newton TF, Cook IA, Walter DO, Rosenberg-Thompson S, LachenbruchPA. Changes in brain functional connectivity in Alzheimer-type and multi-infarct dementia. Brain 1992;115:1543–61.

Leuchter AF, Spar JE, Walter DO, Weiner H. Electroencephalographic spectra andcoherence in the diagnosis of Alzheimer’s-type and multi-infarct dementia. Apilot study. Arch Gen Psychiatry 1987;44:993–8.

Locatelli T, Cursi M, Liberati D, Franceschi M, Comi G. EEG coherence in Alzheimer’sdisease. Electroencephalogr Clin Neurophysiol 1998;106:229–37.

Marshall RS, Festa JR, Cheung YK, Chen R, Pavol MA, Derdeyn CP, et al. Cerebralhemodynamics and cognitive impairment: baseline data from the RECON trial.Neurology 2012;78:250–5.

McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinicaldiagnosis of Alzheimer’s disease: report of the NINCDS – ADRDA work groupunder the auspices of Department of Health and Human Services Task Force onAlzheimer’s disease. Neurology 1994;34:939–44.

Nicolakakis N, Hamel E. Neurovascular function in Alzheimer’s disease patients andexperimental models. J Cereb Blood Flow Metab 2011;31:1354–70.

Novelli G, Papagno C, Capitani E, Laiacona M, Vallar G, Cappa SF. Tre test clinici diricerca e produzione lessicale: taratura su soggetti normali. Arch Psicol NeurolPsichiatr 1986;47:477–506.

Orsini A, Grossi D, Capitani E, Laiacona M, Papagno C, Vallar G. Verbal and spatialimmediate memory span: normative data from 1355 adults and 1112 children.Ital J Neurol Sci 1987;8:539–48.

Ozaki H, Watanabe S, Suzuki H. Topographic EEG changes due to hypobaric hypoxiaat simulated high altitude. Electroencephalogr Clin Neurophysiol1995;94:349–56.

Peers C, Dallas ML, Boycott HE, Scragg JL, Pearson HA, Boyle JP. Hypoxia andneurodegeneration. Ann NY Acad Sci 2009;1177:169–77.

Petersen RC, Doody R, Kurz A, Mohs RC, Morris JC, Rabins PV, et al. Current conceptsin mild cognitive impairment. Arch Neurol 2001;58:1985–92.

Petersen RC, Negash S. Mild cognitive impairment: an overview. CNS Spectr2008;13:45–53.

Petersen RC, Smith GE, Ivnik RJ, Tangalos EG, Schaid SN, Thibodeau SN, et al.Apolipoprotein E status as a predictor of the development of Alzheimer’sdisease in memory-impaired individuals. JAMA 1995;273:1274–8.


















































































































































































Petersen RC, Smith GE, Waring SC, Ivnik RJ, Kokmen E, Tangelos EG. Aging, memory,and mild cognitive impairment. Int Psychogeriatr 1997;1:65–9.

Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med2004;256:183–94.

Pfurtscheller G, Andrew C. Event-related changes of band power and coherence:methodology and interpretation. J Clin Neurophysiol 1999;16:512–9.

Pfurtscheller G, Lopes da Silva FH. Event-related EEG/MEG synchronization anddesynchronization: basic principles. Clin Neurophysiol 1999;110:1842–57.

Ponomareva NV, Selesneva ND, Jarikov GA. EEG alterations in subjects at highfamilial risk for Alzheimer’s disease. Neuropsychobiology 2003;48:152–9.

Portet F, Ousset PJ, Visser PJ, Frisoni GB, Nobili F, Scheltens P, et al. MCI workinggroup of the European consortium on Alzheimer’s disease mild cognitiveimpairment (MCI) in medical practice: a critical review of the concept and newdiagnostic procedure. Report of the MCI working group of the Europeanconsortium on Alzheimer’s disease. J Neurol Neurosurg Psychiatry2006;77:714–8.

Rappelsberger P, Petsche H. Probability mapping: power and coherence analyses ofcognitive processes. Brain Topogr 1988;1:46–54.

Raven JC. Guide to using coloured progressive matrices. London UK: K Lewis; 1965.Reitan RM. Validity of the trail making test as an indication of organic brain damage.

Percept Mot Skill 1958;23:271–6.Rey A. Memorisation d‘une serie de 15 mots en 5 repetitions. In: Rey A, editor.

L‘examen clinique en psychologie. Paris: Presses Universitaires de France; 1958.Rey A. Reattivo Della Figura Complessa Manuale. Firenze: Organizzazioni Speciali 1968.Rossini PM, Rossi S, Babiloni C, Polich J. Clinical neurophysiology of aging

brain: from normal aging to neurodegeneration. Prog Neurobiol2007;83:375–400.

Rubin EH, Morris JC, Grant EA, Vendegna T. Very mild senile dementia of theAlzheimer type. I. Clinical assessment. Arch Neurol 1989;46:379–82.

Sauseng P, Klimesch W, Schabus M, Doppelmayr M. Fronto-parietal EEG coherencein theta and upper alpha reflect central executive functions of working memory.Int J Psychophysiol 2005;57:97–103.

Schellart NA, Reits D. Transient and maintained changes of the spontaneousoccipital EEG during acute systemic hypoxia. Aviat Space Environ Med2001;72:462–70.

Scheltens P, Fox N, Barkhof F, De Carli C. Structural magnetic resonance imaging inthe practical assessment of dementia: beyond exclusion. Lancet Neurol2002;1:13–21.

Scragg JL, Fearon IM, Boyle JP, Ball SG, Varadi G, Peers C. Alzheimer’s amyloidpeptides mediate hypoxic up-regulation of L-type Ca2+ channels. FASEB J2005;19:150–2.

Selvamurthy W, Saxena RK, Krishnamurthy N, Suri ML, Malhotra MS. Changes inEEG pattern during acclimatization to high altitude (3500 m) in man. AviatSpace Environ Med 1978;49:968–71.

Shi Z, Zhao D, Gu Z. The influence of acute and chronic hypoxia on theelectroencephalogram of human body. Ann Physiol Anthropol 1987;6:111–21.

Shulman KI, Gold DP, Cohen CA, Zucchero CA. Clock drawing and dementia in thecommunity: a longitudinal study. Int J Geriatr Psychiat 1993;8:487–96.

Silvestrini M, Pasqualetti P, Baruffaldi R, Bartolini M, Handouk Y, Matteis M, et al.Cerebrovascular reactivity and cognitive decline in patients with Alzheimerdisease. Stroke 2006;37:1010–5.

Silvestrini M, Paolino I, Vernieri F, Pedone C, Baruffaldi R, Gobbi B, et al. Cerebralhemodynamics and cognitive performance in patients with asymptomaticcarotid stenosis. Neurology 2009;72:1062–8.


Smielewski P, Kirkpatrick P, Minhas P, Pickard JD, Czosnyka M. Can cerebrovascularreactivity be measured with near-infrared spectroscopy? Stroke1995;26:2285–892.

Spinnler H, Tognoni G. Standardizzazione e taratura italiana di testneuropsicologici. Ital J Neurol Sci 1987;6(Suppl. 8):1–120.

Sun X, He G, Qing H, Zhou W, Dobie F, Cai F, et al. Hypoxia facilitates Alzheimer’sdisease pathogenesis by up-regulating BACE1 gene expression. Proc Natl AcadSci USA 2006;103:18727–32.

Thatcher RW, Krause PJ, Hrybyk M. Cortico-cortical associations and EEGcoherence: a two-compartmental model. Electroencephalogr ClinNeurophysiol 1986;64:123–43.

Thiebauld C, Van Mullen J, Lintermans J, Sprumont P. Testing in a hypobaricchamber drugs claimed to improve impaired brain function. Lancet1983;2:225–6.

Tong XK, Nicolakakis N, Kocharyan A, Hamel E. Vascular remodeling versus amyloidbeta-induced oxidative stress in the cerebrovascular dysfunctions associatedwith Alzheimer’s disease. J Neurosci 2005;25:11165–71114.

Vecchio F, Babiloni C, Ferreri F, Buffo P, Cibelli G, Curcio G, et al. Mobile phone emissionmodulates inter-hemispheric functional coupling of EEG alpha rhythms in elderlycompared to young subjects. Clin Neurophysiol 2010;121:163–71.

Vecchio F, Babiloni C, Ferreri F, Curcio G, Fini R, Del Percio C, et al. Mobile phoneemission modulates interhemispheric functional coupling of EEG alpharhythms. Eur J Neurosci 2007;25:1908–13.

Vecchio F, Tombini M, Buffo P, Assenza G, Pellegrino G, Benvenga A, et al. Mobilephone emission increases inter-hemispheric functional coupling ofelectroencephalographic alpha rhythms in epileptic patients. Int JPsychophysiol 2012;84:164–71.

Vernieri F, Pasqualetti P, Passarelli F, Rossini PM, Silvestrini M. Outcome of carotidartery occlusion is predicted by cerebrovascular reactivity. Stroke1999;30:593–8.

Vernieri F, Tibuzzi F, Pasqualetti P, Rosato N, Passarelli F, Rossini PM, et al.Transcranial Doppler and near-infrared spectroscopy can evaluate thehemodynamic effect of carotid artery occlusion. Stroke 2004;35:64–70.

Vicenzini E, Ricciardi MC, Altieri M, Puccinelli F, Bonaffini N, Di Piero V, et al.Cerebrovascular reactivity in degenerative and vascular dementia: atranscranial Doppler study. Eur Neurol 2007;58:84–9.

Von Stein A, Sarnthein J. Different frequencies for different scales of corticalintegration: from local gamma to long range alpha/theta synchronization. Int JPsychophysiol 2000;38:301–13.

Wada Y, Nanbu Y, Kikuchi M, Koshino Y, Hashimot T, Yamaguchi N. Abnormalfunctional connectivity in Alzheimer’s disease: intrahemispheric EEG coherenceduring rest and photic stimulation. Eur Arch Psychiatry Clin Neurosci1998a;248:203–8.

Wada Y, Nanbu Y, Koshino Y, Yamaguchi N, Hashimoto T. 1Reducedinterhemispheric EEG coherence in Alzheimer disease: analysis during restand photic stimulation. Alzheimer Dis Assoc Disord 1998b;12:175–81.

Zappoli R, Versari A, Paganini M, Arnetoli G, Muscas GC, Gangemi PF, et al. Brainelectrical activity (quantitative EEG and bit-mapping neurocognitive CNVcomponents), psychometrics and clinical findings in presenile subjects withinitial mild cognitive decline or probable Alzheimer-type dementia. Ital J NeurolSci 1995;16:341–76.

Zaudig M. A new systematic method of measurement and diagnosis of ‘‘mildcognitive impairment’’ and dementia according to ICD-10 and DSM-III-Rcriteria. Int Psychogeriatr 1992;2:203–19.

















































































































Documents

Hypercapnia affects the functional coupling of resting state electroencephalographic rhythms and cerebral haemodynamics in healthy elderly subjects and in patients with amnestic mild