Resting SPECT-neuropsychology correlation in very mild

Alzheimer’s disease

Flavio Nobilia,*, Andrea Brugnoloa, Piero Calvinic, Francesco Copellod, Caterina De Leoa,Nicola Girtlera, Silvia Morbellib, Arnoldo Piccardob, Paolo Vitalia, Guido Rodrigueza

aSection of Clinical Neurophysiology (DISEM), Department of Endocrinological and Metabolic Sciences,

University of Genoa, Viale Benedetto XV, 6, I-16132 Genoa, ItalybSection of Nuclear Medicine, Department of Internal Medicine, University of Genoa, Genoa, Italy

cINFM, Department of Physics, University of Genoa, Genoa, ItalydSection of Preventive Medicine, S. Martino Hospital, Genoa, Italy

Accepted 2 September 2004

Available online 7 October 2004

Abstract

Objective: To investigate the relationships between brain function and some of the most frequently impaired cognitive domains in the first

stages of Alzheimer’s disease (AD), we searched for correlation between the scores on 3 neuropsychological tests and brain perfusion,

assessed by single photon emission computed tomography (SPECT) in patients with very mild AD.

Methods: Twenty-nine consecutive outpatients (mean age 78.2G5.5) affected by probable AD in the very mild phase (i.e. with a score

R20 on the mini-mental state examination, MMSE) underwent brain SPECT with 99mTc-ethylcisteinate dimer. For correlative purposes,

word list learning (by the selective reminding test, SRT), constructional praxis test (CPT) and visual search test (VST) were chosen a priori

out of an extended battery employed to diagnose AD at first patient evaluation. Voxel-based correlation analysis was achieved by statistical

parametric mapping (SPM99) with a height threshold of PZ0.005. Age, years of education and the MMSE score were inserted in the

correlative analysis as confounding variables.

Results: The SRT score showed correlation with brain perfusion in 3 clusters of the left hemisphere, including the post-central gyrus, the

parietal precuneus, the inferior parietal lobule and the middle temporal gyrus, and in one cluster in the right hemisphere including the middle

temporal gyrus and the middle occipital gyrus. The CPT score was significantly correlated with brain perfusion in the parietal precuneus and

the posterior cingulate gyrus in the left hemisphere, whereas the VST score gave a significant correlation with brain perfusion in a left cluster

including the parietal precuneus and the superior temporal gyrus.

Conclusions: Cognitive impairment in very mild AD is reflected by brain dysfunction in posterior associative areas, with peculiar

topographical differences proper of each domain. The parietal precuneus was a common site of correlation of all 3 neuropsychological tests.

This region, together with the posterior cingulate and the superficial posterior temporal–parietal cortex, is thought to be affected by

disconnection from the mesial temporal lobe, besides being directly affected by increased oxidative stress and by atrophy as well. The

impairment of these areas is thought to contribute to cognitive decline in verbal memory, constructional praxis and visual sustained attention

which are indeed among the earliest signs of cognitive impairment in AD.

Significance: Assessing the relationships between neuropsychology and brain functional imaging is a key approach to clarify the

pathophysiology of cognitive failure in AD; the specificity of these findings in AD remains to be proven through comparison with correlation

achieved in matched controls.

q 2004 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

Keywords: Alzheimer’s disease; Brain SPECT; Neuropsychology; SPM; Verbal memory; Constructional apraxia; Sustained attention

1388-2457/$30.00 q 2004 International Federation of Clinical Neurophysiology.

doi:10.1016/j.clinph.2004.09.001

* Corresponding author. Tel.: C39 010 3537568; fax: C39 010 5556893.

E-mail address: flaviomariano.nobili@hsanmartino.liguria.it

(F. Nobili).

1. Introduction

Both brain perfusion single photon emission computed

tomography (SPECT) and neuropsychological assessment

Clinical Neurophysiology 116 (2005) 364–375

www.elsevier.com/locate/clinph

Published by Elsevier Ireland Ltd. All rights reserved.

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375 365

have been proven to be sensitive tools to assess the

functional deficit in the early stages of Alzheimer’s disease

(AD) (Arnaiz and Almkvist, 2003; Devous, 2002). Positron

Emission Tomography (PET) is also increasingly used,

yielding measurement of glucose utilization with a better

spatial resolution but at a consistently higher cost than

SPECT. SPECT and PET provide a topographic picture of

brain function and allow speculations on functional

connectivity among distant areas. On the other hand,

neuropsychology detects impairment in cognitive domains

by specific tasks and gives the experimental material to

design models of cognitive derangement in AD.

Two main approaches are used to investigate the brain

perfusion/metabolic substrates of cognitive performances.

One is mainly aimed to find out those brain areas that are

involved during the performance of a cognitive task in AD

patients in comparison with controls. From these ‘acti-

activation’ studies, that are especially carried out by PET

and by functional magnetic resonance imaging (fMRI),

information is drawn on putative circuitry that is activated in

AD patients while performing the task (Prvulovic et al.,

2002). The great advantage of activation studies is in

imaging the working brain and thus gaining information on

possibly alternative activity/inactivity of brain regions in

AD. The most serious drawbacks are the sophisticated

methodology and statistics required (making these studies

prone to artefacts), the high compliance required in subjects

(often hard to reach in AD patients) and the need of an a

priori cognitive model to interpret images (Shallice, 2003).

Very recently, the current way of interpretation of

‘activated’ areas in this provocative approach has been

seriously questioned (Sidtis et al., 2003).

The other investigation method is to search for

relationships between the scores on neuropsychological

tests applied to the patients either ad-hoc or while

performing the diagnosis and perfusion/metabolic scans

obtained at rest. This ‘resting’ method leads to understand

how much and where two sound tools to study brain

functional failure are correlated, allowing speculations on

the early involved sites, on how their impairment is reflected

by cognitive failure and possibly on the task that better

represents the brain impairment. The most obvious

advantages are that data is already available in a modern

memory clinic (thus not requiring extra costs to design ad-

hoc investigations) and that patient compliance is generally

good, especially when cognition is still mildly impaired.

The main limitation is that the two examinations are not

simultaneous and thus explore different moments of what

can well be a fluctuating cognitive state.

The two approaches yield very different information that

cannot be joined or directly compared, at least until a

complete theory of cognition is applied (Shallice, 2003). On

the other hand, both approaches are very useful to improve

our understanding of the disease’s pathophysiology from its

beginning.

A number of studies have investigated the relationships

between neuropsychological test scores and brain perfu-

sion/metabolism at rest (Bartenstein et al., 1997; Burns

et al., 1989; Desgranges et al., 1998; Elgh et al., 2002;

Goldenberg et al., 1989; Hirono et al., 2001; Montaldi et al.,

1990; O’Brien et al., 1992). However, most of them suffer

from several limitations. First, AD patients are examined

together, especially in early studies, irrespectively of the

severity of cognitive impairment and in some cases the

severity of the disease is not reported at all (Burns et al.,

1989; Goldenberg et al., 1989). In fact, it is now known that

both the cognitive profile and the perfusion/metabolic

deficits change substantially with the increasing severity

of the disease (Matsuda et al., 2002). Second, several studies

(Bartenstein et al., 1997; Burns et al., 1989; O’Brien et al.,

1992) have employed a correlative analysis between the

subscores of a general battery, such as those of the CAM-

COG, the ADAS-COG, or the MMSE scales, which are

often represented by too few points within each subscore to

allow meaningful linear correlations. Third, methods of

analysis are almost always based on Regions (ROIs) or

Volumes of Interest (VOIs) in SPECT/PET images, thus

inevitably missing a number of brain regions, or considering

in the same ROI/VOI a peculiar anatomical rather than a

functional region. It is especially noteworthy that mesial

brain regions, such as those of the temporal and parietal

lobes which are now known to be so relevantly impaired at

the beginning of the disease, were often not included in the

choice of ROIs or VOIs (O’Brien et al., 1992).

The aim of the present investigation is to assess the

correlation between the scores on specific neuropsycholo-

gical tests and brain perfusion SPECT on a voxel-based

basis in a consecutive series of patients with very mild AD,

taking into account the severity of general cognitive

impairment. The scores on word-list learning, construc-

tional praxis and visual sustained attention were chosen

among a wide standardized battery administered to patients

at the time of their first evaluation. In fact, impairment in

verbal learning is a well known typical hallmark of early

AD, constructional apraxia has been increasingly found in

the mild stages (Fisher et al., 1999), while sustained

attention is a ‘background’ function underlying all the

other tasks (Tales et al., 2002, 2004).

2. Methods

2.1. Patients

During a 1 year period, all the consecutive outpatients with

probable AD (according to the definition of the NINCDS-

ADRDA Work Group) (McKhann et al., 1984) in the early or

very mild stage of the disease (i.e. scoring 20 or higher on the

mini-mental state examination (MMSE)) (Folstein et al.,

1975) who came to our Centre for a first diagnostic evaluation

were considered eligible for this study.

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375366

All the patients underwent a complete diagnostic work-

up according to current standards, which include general

and neurological examinations, a standardized neuropsy-

chological assessment, basal computed tomography or

magnetic resonance imaging and other routine blood and

urine screening investigations to rule out secondary

dementias.

The presence of previous or present major psychiatric

disorders, serious neurological diseases, severe and uncon-

trolled arterial hypertension, diabetes mellitus, renal,

hepatic or respiratory failure, anaemia (Hb level!10 mg/dl) or malignancies were exclusion criteria.

In conformity with the diagnosis of AD, all patients

scored lower than 4 on the Hachinski ischaemic scale.

Lewy-body dementia, frontotemporal dementia and vascu-

lar dementia were excluded in all patients on the basis of

current clinical criteria (McKeith et al., 1999; Roman et al.,

1993; The Lund and Manchester Criteria, 1994).

All patients underwent brain perfusion SPECT examin-

ation by 99mTc-N,N 00-1,2-ethylene diylbis-L-cysteine diethyl

ester dihydrochloride (ECD) as part of the routine

diagnostic procedure in our Centre. The original group

was composed of 31 patients, but two left-handed patients

(Annett’s Questionnaire) were excluded. The study group

thus included 29 patients (23 women and 6 men), aged

66–86 years (mean 78.2G5.5), with years of full education

ranging from 3 to 17 (mean 7.2G4.2).

All the patients were at their first diagnostic evaluation

for cognitive complaints, thus none of them were under

treatment with Acetylcholinesterase inhibitors, nor with

other neuropsychoactive drugs. The patients (or their

relatives) were informed about the finalities of the study

and gave their informed consent.

2.2. Brain SPECT

Brain SPECT was performed employing a single-head

gamma-camera (Apex SPX-4, Elscint, Israel) equipped by a

low-energy, high-resolution collimator. All SPECT acqui-

sitions started between 30 and 120 min after the i.v.

injection of about 1000 MBq of 99mTc-ECD (Neurolite,

Dupont Pharma, USA), according to the guidelines of the

European Society of Nuclear Medicine (Tatsch et al., 2002).

A step-and-shoot acquisition protocol acquired images by

120 projections, with a radius of rotation !15 cm. Total

counts were between 7 and 10 million. SPECT images were

3-dimensionally (3D) reconstructed by the conjugate

gradient (CG) algorithm, a technique initially proposed in

a 2-D version (Formiconi et al., 1989). A 3D version has

been developed by means of the 2DC1 algorithm (Boccacci

et al., 1999), according to the procedure described else-

where (Nobili et al., 2001). The number of iterations was

between 13 and 15, depending on the number of counts

collected in the data. The 2DC1 version of the CG method

used in this study yielded a reconstruction consisting of

stacks of 64 transaxial slices, where the compensation for

attenuation had already been performed at the reconstruc-

tion stage by the iterative algorithm itself. In this case, the

attenuation contour was directly given by the skull contour

and the related anatomical shapes were carefully taken into

account. The CG reconstruction method has already been

employed in clinical studies on dementia, allowing better

spatial resolution and correlative results than the conven-

tional filtered back-projection method (Nobili et al., 2001;

Rodriguez et al., 2000).

2.3. Neuropsychological variables

Patients were carefully screened with the synoptic

batteries of the MMSE and of the CAM-COG battery

(short Italian version) (Neri et al., 1994). The specific

domains of sustained visual attention, verbal learning and

recall, language, abstract reasoning and constructional

praxis were assessed by a standardized battery in use at

our Center. The presence of behavioural and psychological

disturbances was assessed by a structured interview to the

principal caregiver by means of the neuropsychiatric

inventory.

The following neuropsychological tests were chosen a

priori out of the whole battery because they explore

cognitive domains that are typically and sometimes

alternatively impaired in the first stages of AD (verbal

memory and constructional praxis), or concern aspects of

cognition that are a common background for the execution

of several other tasks, such as visual sustained attention.

(i)

The Buschke–Fuld selective reminding test (SRT), in

its short version (Masur et al., 1989), is a word-list

learning of 12 words and 6 repetition trials. For the

purpose of this study, only the total number of words

recalled was considered, since delayed recall showed a

clear ground effect, leading to a consistent number of 0

scores. Abnormally low scores were identified accord-

ing to the normative values reported by Masur et al.

(1989);

(ii)

the constructional praxis test (Carlesimo et al., 1996;

Gainotti et al., 1977); only the subtest with planning

elements was considered for further analysis, since the

subtest of freehand copying yields a narrow score range

not suitable for linear correlative purposes. This subtest

requires the completion of 2 bi-dimensional and 8 tri-

dimensional figures, of which some parts have been

omitted. The items range from 2 to 10 for each figure

(maximum score 70) (Fig. 1), and abnormal values

were computed according to the reference literature

(Carlesimo et al., 1996);

(iii)

the visual search test (VST), which assesses sustained

attention, in the Italian validated version (Spinnler and

Tognoni, 1987). Three matrices of 110 (10!11)

numbers each, are presented in sequence to the patient

who must identify and mark the number 5 in the first

matrix (10 items), the numbers 2 and 6 in the second

Fig. 1. Two (a cube and a house, both with a tri-dimensional view) of the 10

figures that must be completed during the constructional praxis test.

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375 367

one (20 items) and the numbers 1, 4 and 9 in third one

(30 items). The score is formed by the sum of correct

items identified in 45 s of time for each of the 3

matrices (maximum score 60). The Italian normative

database (Spinnler and Tognoni, 1987) was used to

recognize pathological values.

Verbal learning is recognized as one of the first specific

memory complaints in AD, already detectable in prede-

mentia AD, such as mild cognitive impairment (Petersen et

al., 1999), and the SRT is widely employed in this context.

Constructional apraxia seems to be a somewhat later event

in a number of patients, but it is often the first cognitive

domain to be impaired in AD, besides memory (Fisher et al.,

1999). Finally, sustained visual attention is increasingly

recognized as one of the early deficits in AD (Baddeley et

al., 2001; Foster, 2001; Perry et al., 2000; Tales et al., 2002).

Category fluency verbal (CFVT) was tested for colours,

animals, fruits and towns in four 2 min periods. Letter

fluency verbal (LFVT) required three 1 min items, including

the letters F, L and P, according to the Italian normative

(Capitani et al., 1998). Abstract reasoning was evaluated by

Weigl’s sorting test (WST) which assesses the ability to

categorize 12 coloured items of different shape, colour,

thickness, dimension each engraved with card suits.

As for the 3 tests chosen for correlative analysis, scores

for CFVT, LFVT and WST were evaluated on the basis of

the Italian normative values, which require correction for

age, education and comparison with normal subjects.

Reference normal values are reported by Spinnler and

Tognoni (1987) for both the WST and the CFVT, and by

Capitani et al. (1998) for LFVT. Range values, individual

values, means and SDs for all the neuropsychological

tests performed in the present investigation are reported in

Table 1.

The correlation study between functional imaging and

other neuropsychological tests or scales would be of great

relevance to understand the pathophysiology of cognitive

failure in AD. However, performing more than 3 correlation

analyses seemed unjustified because of the dimension of the

present patient sample. Neuropsychological tests and

SPECT were performed within an interval of 3 months

(mean interval time 38 daysG16).

2.4. Statistical analysis

Correlation between the SPECT data and the neurop-

sychological scores were computed by statistical para-

metric mapping (SPM) (Friston et al., 1991), 1999 version

(SPM99), following the rules of the Consensus of the

European Alzheimer’s Disease Consortium (EADC)

(Frisoni et al., 2003). In a preprocessing step, data

sets were spatially normalized using a 12 point linear

affined transformation into Talairach space (Talairach and

Tournoux, 1988) and smoothed with an isotropic Gaussian

kernel of FWHM 12 mm. SPM99 co-registers the

individual SPECT with the 152 brains average of the

Montreal Neurological Institute. Since this template does

not completely match the Talairach brain, a correction of

the SPM{t} coordinates is needed. This was achieved

using the subroutine implemented by Matthew Brett

(http://www.mrc-cbu.cam.ac.uk/Imaging) which gives the

correspondence between SPM and the Talairach atlas

coordinates. The grey matter threshold was set at 0.8.

Normalization of global CBF to 50 was performed with

proportional scaling. The resulting set of values for

comparison constituted a statistical parametric map of

the statistic SPM{t}. Then, the SPM{t} maps were

transformed to the unit of normal distribution (SPM{z})

and reached a threshold at PZ0.005, which is an accepted

procedure in a clinical setting (Desgranges et al., 1998).

Because of the lack of any topographic a priori

hypothesis, the significance of identified regions was

assessed using P values corrected for multiple compari-

sons (P!0.05 was the first statistical significance to be

accepted at the cluster level).

Table 1

Neuropsychological test data in 29 outpatients with very mild Alzheimer’s disease

Patient no. Sex Age

yrs

Education

yrs

Short

CAM-COG

MMSE WST VST CFVT LFVT SRT CPs CPpe

Range 66–86 3–17 0–62 0–30 0–15 0–60 0–72 0–12 0–70

1 F 80 8 39 24a 10 50 10.67 14 29a 6a 68

2 F 83 5 40 29 3a 57 7.67 14.5 39 5a 60

3 F 86 5 27a 20a 8 46 4.67a 8.5 27a 5a 52a

4 F 81 4 21a 22a 2a 38 5.67a 9 22a 8 70

5 F 79 6 24a 23a 4 39 7.33 10.5 25a 5a 52a

6 F 81 5 26a 21a 6 44 3.0a 9.25 20a 5a 36a

7 F 72 5 26a 24a 4 32 3.0a 7 21a 8 70

8 F 79 5 14a 21a 1a 39 3.33a 7 27a 4a 56a

9 F 74 5 41 27 9 30a 15.33 13.25 29a 10 70

10 F 76 5 22a 20a 3a 55 6.67 11.25 22a 5a 55a

11 M 81 5 30a 27 5 45 8.0 6.5 25a 6a 60

12 M 78 17 33 24a 12 51 7.33 9a 26a 9 68

13 F 66 5 33 26 6 40 4.67a 13.25 21a 3a 24a

14 F 66 5 21a 20a 1a 26a 2.67a 8.25 19a 6a 52a

15 M 85 17 25a 20a 5a 36 8.67 12 18a 9 68

16 F 86 4 34 26 10 47 4.33a 13.5 29a 9 57a

17 F 79 6 37 26 4a 31 4.33a 10.5 24a 6a 58a

18 M 78 8 30 23a 6 34 2.33a 7.75 19a 2a 28a

19 F 73 5 39 29 10 45 11.33 14 25a 3a 60a

20 F 82 10 30 22a 5a 51 13.67 19.75 29a 7 68

21 F 81 6 21a 21a 6 48 7.0 9.5 25a 3a 55a

22 F 73 13 22a 20a 3a 36 7.67 12.5 26a 5a 49a

23 F 82 5 24a 22a 7 26a 5.0a 7.25 25a 5a 65

24 M 78 17 39 28 9 38 6.67 8.5a 10a 4a 67

25 F 82 5 36 25 7 51 4.67a 11 26a 6a 58a

26 M 72 13 47 28 13 30a 16.33 18.5 37 7a 69

27 F 86 3 22a 20a 3 51 4.0a 9.5 26a 5a 46a

28 F 70 3 32 26 10 48 9.0 16 22a 7 52a

29 F 80 8 45 27 10 47 11.0 14 29a 6 60a

Mean 23 F 78.24 7.17 30.34 23.83 6.28 41.76 7.1 11.22 24.9 5.83 57.0

SD 6 M 5.52 4.17 8.29 3.06 3.29 8.77 3.73 3.42 5.55 2.0 11.96

WST, Weigl’s sorting test; VST, visual search test; CFVT, category fluency verbal test; LFVT, letter fluency verbal test; SRT, selective reminding test; CPs,

constructional praxis test, simple (freehand copying of drawings); CPpe, constructional praxis test, with planning elements (copying drawings with landmarks).a Abnormally low value in comparison with normative database of the literature (see text in the Methods section for further details).

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375368

The correlative analysis was applied to SPECT examin-

ations by means of the ‘multi-subjects: co-variate only’

option. The co-variates were the scores on the selective

reminding test, constructional praxis with planning

elements and Visual search test. Age and years of education

were inserted as ‘nuisance’ variables, to account for their

confounding effect on both perfusion data (Salmon et al.,

2000) and neuropsychological scores. The MMSE score was

inserted as the third ‘nuisance’ variable to take into account

the degree of global cognitive impairment even in this

selected series of patients with very mild AD. The SPM

contrastC1 was defined in each correlation to search for

direct (positive) correlation.

3. Results

In this group of 29 patients with very mild AD, the

clinical dementia rating was 0.5 in 11 patients and 1 in 18

patients, whereas the MMSE score ranged from 20 to 29

(mean 23.8G3.1).

Table 1 reports the details of results of neuropsycholo-

gical tests. Briefly, all the patients showed an impairment in

at least two tests. The SRT was impaired in 37 patients

(94.9%), the CPT (freehand copy) in 20 (69%), CPT with

planning elements in 17 (58.6%), and VST in 4 (13.8%).

Among the other tests, the CFVT was altered in 13 patients

(44.8%), the WST in 8 patients (27.6%), and LFVT in 2

(6.9%) patients.

Table 2 reports in detail the sites of significant SPM

correlation between neuropsychological scores and brain

perfusion SPECT. The SRT score gave 3 clusters of

significant correlation in the left hemisphere and one cluster

in the right hemisphere. The 3 left-side clusters included the

post-central gyrus (Brodmann area, BA 3), the parietal

precuneus (BA 39), the inferior parietal lobule (BA 40) and

the middle temporal gyrus (BAs 39 and 21). The right

cluster gave less significant correlation at voxel level and

involved the middle temporal (BA 39) and occipital (BA 19)

gyri (Table 2; Fig. 2).

The constructional praxis score results of correlation

were significant in one cluster involving the left parietal

Table 2

Numerical results of SPM direct correlations (height threshold PZ0.005)

Correlation Cluster level Voxel level

Cluster

extent

Corrected

P value

Cortical region Z score of

maximum

Talairach

coordinates

Cortical region BA

Selective reminding test 861 0.05 L parietal 4.26 (K42, K22, 60) Post-central gy 3

1301 0.013 L parietal and

temporal

4.04 (K38, K62, 33) Precuneus 39

(K44, K39, 39) Inf. parietal lobule 40

(K51, K61, 29) Middle temporal gy 39

887 0.05 L temporal 3.74 (K63, K29, K5) Middle temporal gy 21

924 0.048 R temporal and

occipital

3.24 (36, K70, 29) Middle temporal gy 39

(32, K85, 19) Middle occipital gy 19

(36, K77, 17) Middle occipital gy 19

Constructional praxis test 442 0.05 L parietal 4.46 (K6, K53, 38) Precuneus 7

L posperior

cingulated

(K4, K35, 40) Posterior cingulate 31

Visual search test 1528 0.008 L temporal and

parietal

4.36 (K38, K51, 25) Superior temporal

gy

39

(K40, K48, 17) Superior temporal

gy

13

(K24, K45, 41) Precuneus 5

A value of P!0.05, corrected for multiple comparison at cluster level, was accepted as statistically significant. In the ‘cluster level’ section on left, the number

of voxels, the corrected P value of significance and the cortical region where the voxel is found, are all reported for each significant cluster of correlation. In the

‘voxel level’ section, all the coordinates of the correlation sites (with the Z score of the maximum correlation point), the corresponding cortical region and BA

are reported for each significant cluster. L, left; R, right; BA, Brodmann’s area. In the case that the maximum correlation is achieved in the white matter, the

nearest grey matter is indicated with the corresponding BA.

Fig. 2. z map of SPM correlation (height threshold PZ0.005) between brain perfusion and score on the Buschke–Fuld selective reminding test in a group of 29

patients with very mild AD. Significant areas (shown as glass brain) are found in the post-central gyrus, parietal precuneus, inferior parietal lobule and middle

temporal gyrus in the left hemisphere, and in the middle temporal and middle occipital gyri in the right hemisphere. See Table 2 for details of Talairach

coordinates.

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375 369

Fig. 3. z map of SPM correlation (height threshold PZ0.005) between brain perfusion and constructional apraxia score in a group of 29 patients with very mild

AD. Significant areas (shown as glass brain) are found in the left posterior cingulate and precuneus. See Table 2 for details of Talairach coordinates.

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375370

precuneus (BA 7) and the left posterior cingulate

gyrus (BA 31). This small region was very close to the

median line, though significance was reached in the left

hemisphere (Table 2; Fig. 3). The VST score was

Fig. 4. z map of SPM correlation (height threshold PZ0.005) between brain perfu

AD. Significant areas (shown as glass brain) are found in the left parietal prec

coordinates.

significantly correlated with brain perfusion in a left

cluster involving the parietal precuneus (BA 5) and the

superior temporal gyrus (BAs 13 and 39) (Table 2;

Fig. 4).

sion and score on visual search test in a group of 29 patients with very mild

uneus and superior temporal gyrus. See Table 2 for details of Talairach

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375 371

In summary, all the correlations were found in the

posterior regions of the brain, mainly in the posterior

parietal and temporal associative cortex, and in the posterior

cingulate. The left parietal precuneus was involved in all 3

correlations, whereas BAs 39 and 40 were widely involved

in the correlation sites of the SRT score and the VST score.

4. Discussion

By correcting for the confounding variable of the

severity of the disease, the study tries to assess the

relationships between specific cognitive functions and

brain perfusion, that are not merely the consequence of

the different degrees of AD severity, which is the single

strongest variable influencing both brain perfusion and

neuropsychological test scores. Therefore, this data is not

immediately comparable with those from previous studies,

which focussed—at best—on one stage of the disease

(i.e. mild or moderate). Even the studies that employed a

voxel-based analysis did not consider the severity of the

disease as a confounding variable (Bartenstein et al., 1997;

Desgranges et al., 1998, 2002a; Hirsch et al., 1997).

It should be remarked here that a comparison of data

obtained by SPM (or by other voxel-based analysis

methods) with or without correction for the severity of

cognitive impairment is a further issue deserving an ad-hoc

investigation that may lead to marked differences between

the two correlative approaches.

In early researches, the investigation of patients with

moderate or severe AD led to widespread correlation

between neuropsychological test scores and SPECT

(Montaldi et al., 1990). Data from patients with mild AD

was merged together with more severe cases thus producing

heterogeneous groups of patients, for instance, with a MMSE

score ranging from 7 to 29 (Bartenstein et al., 1997; Hirsch et

al., 1997). The issue of not taking into account the severity of

disease cannot be overemphasized. In fact, it has recently

been shown that when a group of patients with mild AD was

divided into a subgroup with a less severe disease and a

subgroup with a more severe disease, correlation sites

between brain metabolism and a story recall task were

found in either hemisphere alternatively (Desgranges et al.,

2002a).

A summary index of cognitive impairment had to be

chosen for taking into account the severity of general

cognitive impairment. The choice of the MMSE score was

made after a series of considerations. In fact, although its

sensitivity in very mild AD is limited, it has been used

worldwide for several years, thus easily allowing compari-

son among studies. The degree of correlation with the short-

CAM-COG score employed here was so high (rZ0.85) that

it rendered the two scores practically equivalent from a

statistical point of view and, to the best of our knowledge, it

is the only score expressing the global cognitive impairment

to be already used for this purpose (Benoit et al., 2004;

Desgranges et al., 2002b). On the other hand, the

computation of a novel, synoptic index starting from the

scores on individual tests would have required validation

from another sample of patients before entering SPM

correlation as a nuisance variable.

An intrinsic limitation of the present study (as well as of

other studies of this kind: Benoit et al., 2004; Desgranges

et al., 2002a,b; Goldenberg et al., 1989; Hirono et al., 2001)

derives from the lack of an age-matched control group,

which would have qualified the results as specific of patients

with very mild AD. Actually, the sites of correlation of the 3

test scores in control subjects are unknown, and the

possibility that the same or similar relationships could be

observed also in healthy individuals cannot be excluded.

The availability of an adequate control group will further

qualify the present results, provided the same methodology

is used. Given this limitation, the discussion of findings with

reference to the specific underlying pathology of AD can

only be matter of speculation, with reference to literature

data obtained in normals, when available. Notwithstanding,

these results may represent a useful point for the further

understanding of brain–neuropsychology relationships in

health and AD.

Verbal memory yielded wide correlation sites, mainly in

the left hemisphere, involving the parietal precuneus, the

inferior parietal lobule and the middle temporal gyrus,

besides the post-central gyrus. Word list learning and recall

is a complex task, with both a linguistic and memory aspect.

The wide clusters of correlation in the left hemisphere,

including the middle temporal gyrus and the inferior parietal

lobule, which are recognized key structures in verbal

function (Damasio, 1992), reflect mainly the linguistic

component in this sample. Similar results were obtained

with verbal tasks of the CAM-COG scale in earlier papers

(Bartenstein et al., 1997; Burns et al., 1989; Hirsch et al.,

1997; O’Brien et al., 1992). Consistent with the present

results, a correlation cluster in both temporal lobes, mainly

in the left one, were achieved by the Buschke–Fuld SRT in a

recent European cooperative study on patients with mild to

moderate AD (height threshold for SPM correlation not

specified) (Sutherland et al., 2003).

The extension of correlation sites to the left precuneus and

to the right middle temporal and occipital gyri, points to the

involvement of the wider memory network component of the

word list learning test. In a correlative analysis in mild AD

patients, the score on a similar test was found to correlate

with glucose metabolism in the left posterior cingulate and in

a wide right superficial region, involving the parietal

precuneus, the posterior cingulate, the middle occipital and

the inferior parietal gyri (Desgranges et al., 1998). The right-

sided correlation in posterior associative areas (BAs 21, 22

and 39) between brain perfusion and a word learning task has

been recently underlined in a small group of patients with

very mild AD by a VOI approach (Elgh et al., 2002).

Our finding of correlation in the precuneus seems

consistent with the observations of activation studies in AD

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375372

(Backman et al., 1999; Becker et al., 1996) as well as in

normals during the recall of episodic information (Cabeza

et al., 2003; Petrides et al., 1995; Shallice et al., 1994) or

during mental imagery processes (Mellet et al., 1995; Platel

et al., 1997), thus suggesting a role of the precuneus in the

memorization of imageable words (Fletcher et al., 1995).

Some activation studies in AD patients are consistent with

this interpretation. In fact, in a PET study employing a verbal

(episodic) memory task in mild AD (Desgranges et al.,

2002a), several areas of brain metabolic activation were

found in both hemispheres, including the bilateral precuneus

and posterior cingulate gyrus, besides mesial and superior

temporal cortex and the cerebellum. Even more interestingly,

a recent fMRI study has demonstrated that the mesial

parietal-posterior cingulated area is quickly deactivated in

young normal subjects during a semantic classification task

whereas AD patients maintained activation throughout the

task block (Lustig et al., 2003), showing an overlap between

the regions with the most pronounced metabolic deficit at rest

and those which are mostly activated during a semantic task.

Finally, functional activation in the mesial posterior regions

has been found to be correlated with the degree of

hippocampal atrophy during a verbal memory task (Garrido

et al., 2002).

The constructional praxis test yielded significant corre-

lation in the parietal precuneus and posterior cingulate gyrus

in the left hemisphere (very close to the median line, so that

the lack of correlation in the right hemisphere could merely

be the result of chance). Since we only evaluated

significance values corrected for multiple comparisons at

cluster level to account for type II error, the possibility that a

somewhat weaker correlation in homologous structures of

the right hemisphere did not survive correction should be

considered. The test employed in the present study is based

on the cued reproduction of a number of both geometric and

structured figures, the majority with a tri-dimensional view,

and was specifically addressed to investigate constructional

apraxia. Therefore, it is more extended and specific than the

constructional apraxia subtest of the CAM-COG scale

which has been the most widely used in correlation studies,

mainly reporting correlation in the right (Bartenstein et al.,

1997; Hirsch et al., 1997; O’Brien et al., 1992) or in both

(Burns et al., 1989) parietal lobes. The involvement of both

parietal lobes in constructional praxis function has been

recently demonstrated by fMRI in young volunteers as well

(Makuuchi et al., 2003). Giannakopoulos et al. (1998), by

using a constructional praxis test very similar to that used in

the present study, found a significant relationship between

the test score and neurofibrillary tangle density in the

superior parietal, posterior cingulate and occipital cortex of

both hemispheres, suggesting that constructional apraxia is

related to the disruption of cortical pathways mediating

visuospatial cognition in AD. It is therefore thought that AD

patients develop constructional apraxia early in the course

of disease as a likely consequence of impairment of both

parietal lobes (Spinnler, 1996).

The correlation achieved with sustained visual attention

mainly involved the left parietal precuneus and the posterior

part of the left temporal lobe. The majority of available data

in the literature has been obtained by activation PET and

fMRI studies. In healthy subjects employing an attention-

demanding search for a target shape, loci of activation were

found in the left parietal operculum, superior temporal

gyrus, parietal–occipital fissure and precuneus by PET

(Patel and Sathian, 2000), thus very similar to the site of

correlation we found in very mild AD patients. Both the

precuneus and the region of superior and inferior parietal

lobule, which is very close to the superior temporal gyrus

have repeatedly been shown to be involved in visual

sustained attention tasks in normal individuals (Astafiev

et al., 2003; Patel and Sathian, 2000) as well as in AD

patients (Astafiev et al., 2003; Buck et al., 1997; Fujimori

et al., 2000; Johannsen et al., 1999). By a picture description

task, it has been recently demonstrated that the posterior

part of the parietal lobes has a major role in sustained visuo-

spatial attention in AD. In fact, perfusion in a parietal lobe

has been shown to be correlated with visual attention toward

figures in the contralateral space (Meguro et al., 2001).

Finally, a recent fMRI study in healthy humans has

established that during visual search the regions in the

bilateral parietal sulcus are the most involved (Muller et al.,

2003).

The interpretation of these results as specific of the

disease process in AD is limited, however, by the lack of

findings of correlation of the same tests in normal

individuals, as already stated. Notwithstanding, some

speculation can be advanced on the basis of the present

results and of the literature data in AD patients. In fact, the

current model for the interpretation of brain hypometabo-

lism/hypoperfusion findings in early AD is based on the

theory of functional disconnection of posterior temporal–

parietal–occipital associative areas from the mesial tem-

poral cortex, where pathological changes are found early

(Braak and Braak, 1995). The precuneus and the posterior

cingulate have been shown to be among the first regions to

suffer from such a disconnection, both in AD (Matsuda

et al., 2002; Minoshima et al., 1997) and in mild cognitive

impairment (MCI) (Chetelat et al., 2003), and hypometa-

bolism has been confirmed even after correction for atrophy

(Ibanez et al., 1998). Neurochemical evidence suggests the

presence of an active oxidative stress in the posterior

temporal–parietal associative cortex, especially in the

inferior parietal lobule (Aksenov et al., 1999; Hensley

et al., 1995; Lovell et al., 1998). To further stress the

relevance of these regions, the right precuneus has been

reported to be the site of maximum perfusion decrease in a

group of patients with mild to moderate AD who did

not respond to continuous donepezil administration after

15 months (Nobili et al., 2002), and the left precuneus was

the site of maximum correlation between changes in

quantitative electroencephalography and brain perfusion in

a group of AD patients treated with donepezil after 1 year

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375 373

(Rodriguez et al., 2004). The converging mechanisms of

oxidative stress, disconnection and atrophy in these

posterior regions leading to hypoperfusion/hypometabolism

appear as a main substrate of cognitive decline in early AD,

as shown by correlation with the performances on

constructional praxis, visual attention and verbal learning.

In conclusion, impairment in 3 of the cognitive domains

which can be involved early in AD has been shown to

correlate with brain perfusion level in several areas in the

posterior parietal–temporal associative cortex, mainly

including the parietal precuneus, the posterior cingulate

gyrus, the inferior parietal lobule, besides other regions in

the temporal and occipital cortex, more linked to the

specificity of each task. These regions are the core of what is

today thought to suffer from the effects of disconnection

from the mesial temporal lobe, besides being directly

affected by increased oxidative stress and atrophy. The

dysfunction in these areas is thought to give a main

contribution to cognitive decline in verbal memory,

constructional praxis and visual sustained attention which

are early cognitive signs in AD. The future availability of

correlation results in healthy subjects could further qualify

these results obtained in early AD patients.

Acknowledgements

This study has been supported by the grant number

2002013422_004 of the Italian Ministry of University and

Research (MIUR). We are indebted to Mr Bruno Errani for

English editing.

References

Aksenov MY, Tucker HM, Nair P, Aksenova MV, Butterfield DA, Estus S,

Markesbery WR. The expression of several mitochondrial and nuclear

genes encoding the subunits of electron transport chain enzyme

complexes, cytochrome c oxidase and NADH dehydrogenase, in

different brain regions in Alzheimer’s disease. Neurochem Res 1999;

24:767–74.

Arnaiz E, Almkvist O. Neuropsychological features of mild cognitive

impairment and preclinical Alzheimer’s disease. Acta Neurol Scand

Suppl 2003;179:34–41.

Astafiev SV, Shulman GL, Stanley CM, Snyder AZ, Van Essen DC,

Corbetta M. Functional organization of human intraparietal and frontal

cortex for attending, looking, and pointing. J Neurosci 2003;23:

4689–99.

Backman L, Andersson JL, Nyberg L, Winblad B, Nordberg A,

Almkvist O. Brain regions associated with episodic retrieval in normal

aging and Alzheimer’s disease. Neurology 1999;52:1861–70.

Baddeley AD, Baddeley HA, Bucks RS, Wilcock GK. Attentional control

in Alzheimer’s disease. Brain 2001;124:1492–508.

Bartenstein P, Minoshima S, Hirsch C, Buch K, Willoch F, Mosch D,

Schad D, Schwaiger M, Kurz A. Quantitative assessment of cerebral

blood flow in patients with Alzheimer’s disease by SPECT. J Nucl Med

1997;38:1095–101.

Becker JT, Mintun MA, Aleva K, Wiseman MB, Nichols T, DeKosky ST.

Compensatory reallocation of brain resources supporting verbal

episodic memory in Alzheimer’s disease. Neurology 1996;46:692–700.

Benoit M, Clairet S, Koulibaly PM, Darcourt J, Robert PH. Brain perfusion

correlates of the apathy inventory dimensions in Alzheimer disease. Int

J Geriatr Psychiatry 2004;19:864–9.

Boccacci P, Bonetto P, Calvini P, Formiconi AR. A simple model for the

efficient correction of collimator blur in 3D SPECT imaging. Inverse

Probl 1999;15:907–30.

Braak H, Braak E. Staging of Alzheimer’s disease-related neurofibrillary

changes. Neurobiol Aging 1995;16:271–8.

Buck BH, Black SE, Behrmann M, Caldwell C, Bronskill MJ. Spatial- and

object-based attentional deficits in Alzheimer’s disease relationship to

HMPAO-SPECT measures of parietal perfusion. Brain 1997;120:

1229–44.

Burns A, Philpot MP, Costa DC, Ell PJ, Levy R. The investigation of

Alzheimer’s disease with single photon emission tomography. J Neurol

Neurosurg Psychiatry 1989;52:248–53.

Cabeza R, Dolcos F, Prince SE, Rice HJ, Weissman DH, Nyberg L.

Attention-related activity during episodic memory retrieval: a cross-

function fMRI study. Neuropsychologia 2003;41:390–9.

Capitani E, Laiacona M, Basso A. Phonetically cued word-fluency, gender

differences and aging: a reappraisal. Cortex 1998;34:779–83.

Carlesimo GA, Caltagirone C, Gainotti G. The Mental Deterioration

Battery: normative data, diagnostic reliability and qualitative analysis

of cognitive impairment. The group for the standardization of the

Mental Deterioration Battery. Eur Neurol 1996;36:378–84.

Chetelat G, Desgranges B, de la Sayette V, Viader F, Eustache F, Baron J-

C. Mild cognitive impairment. Can FDG-PET predict who is to rapidly

convert to Alzheimer’s disease? Neurology 2003;60:1374–7.

Damasio AR. Aphasia. New Engl J Med 1992;326:531–9.

Desgranges B, Baron J-C, de la Sayette V, Petit-Taboue M-C, Benali K,

Landeau B, Lechevalier B, Eustache F. The neural substrates of

memory systems impairment in Alzheimer’s disease. A PET study of

resting brain glucose utilisation. Brain 1998;121:611–31.

Desgranges B, Baron J-C, Lalavee C, Giffard B, Viader F, de la Sayette V,

Eustache F. The neural substrates of episodic memory impairment in

Alzheimer’s disease as revealed by FDG-PET: relationship to degree of

deterioration. Brain 2002a;125:1116–24.

Desgranges B, Baron J-C, Giffard B, Chetelat G, Lalevee C, Viader F, de la

Sayette V, Eustache F. The neural basis of intrusions in free recall and

cued recall: a PET study in Alzheimer’s disease. NeuroImage 2002b;17:

1658–64.

Devous Sr MD. Functional brain imaging in the dementias: role in early

detection, differential diagnosis, and longitudinal studies. Eur J Nucl

Med 2002;29:1685–7.

Elgh E, Sundstrom T, Nasman B, Ahlstrom KR, Nyberg L. Memory

functions and rCBF 99mTc-HMPAO SPET: developing diagnostics in

Alzheimer’s disease. Eur J Nucl Med 2002;29:1140–8.

Fisher NJ, Rourke BP, Bieliauskas LA. Neuropsychological subgroups of

patients with Alzheimer’s disease: an examination of the first 10 years

of CERAD data. J Clin Exp Neuropsychol 1999;21:488–518.

Fletcher PC, Frith CD, Baker SC, Shallice T, Frackowiak RS, Dolan RJ.

The mind’s eye-precuneus activation in memory-related imagery.

NeuroImage 1995;2:195–200.

Folstein MF, Folstein SE, McHugh PR. ‘Mini Mental State’: a practical

method for grading the cognitive state of patients for clinician.

J Psychiatr Res 1975;12:189–98.

Formiconi A, Passeri A, Pupi A. Compensation of spatial system response

in SPECT with conjugate gradient reconstruction technique. Phys Med

Biol 1989;34:69–84.

Foster JK. Selective attention in Alzheimer’s disease. Front Biosci 2001;6:

D135–D153.

Frisoni GB, Scheltens Ph, Galluzzi S, Nobili FM, Fox NC, Robert PH,

Soininen H, Wahlund LO, Waldemar G, Salmon E. Neuroimaging tools

to rate regional atrophy, subcortical cerebrovascular disease, and

regional cerebral blood flow and metabolism: consensus paper of the

EADC. J Neurol Neurosurg Psychiatry 2003;74:1371–81.

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375374

Friston KJ, Frith CD, Liddle PF. Comparing functional (PET) images: the

assessment of significant changes. J Cereb Blood Flow Metab 1991;11:

690–9.

Fujimori M, Imamura T, Hirono N, Ishii K, Sasaki M, Mori E. Disturbances

of spatial vision and object vision correlate differently with regional

cerebral glucose metabolism in Alzheimer’s disease. Neuropsychologia

2000;38:1356–61.

Gainotti G, Miceli G, Caltagirone C. Constructional apraxia in left brain

damaged patients: a planning disorder? Cortex 1977;13:109–18.

Garrido GE, Furuie SS, Buchpiguel CA, Bottino CM, Almeida OP, Cid CG,

Camargo CH, Castro CC, Glabus MF, Busatto GF. Relation between

medial temporal atrophy and functional brain activity during memory

processing in Alzheimer’s disease: a combined MRI and SPECT study.

J Neurol Neurosurg Psychiatry 2002;73:508–16.

Giannakopoulos P, Duc M, Gold G, Hof PR, Michel JP, Bouras C.

Pathologic correlates of apraxia in Alzheimer disease. Arch Neurol

1998;55:689–95.

Goldenberg G, Podreka I, Suess E, Deecke L. The cerebral localization of

neuropsychological impairment in Alzheimer’s disease: a SPECT

study. J Neurol 1989;236:131–8.

Hensley K, Hall N, Subramaniam R, Cole P, Harris M, Aksenov M,

Aksenova M, Gabbita SP, Wu JF, Carney JM, Lovell M,

Markesbery WR, Butterfield DA. Brain regional correspondence

between Alzheimer’s disease histopathology and biomarkers of protein

oxidation. J Neurochem 1995;65:2146–56.

Hirono N, Mori E, Ishii K, Imamura T, Tanimukai S, Kazui H,

Hashimoto M, Takatsuki Y, Kitagaki H, Sasaki M. Neuronal substrates

for semantic memory: a positron emission tomography study in

Alzheimer’s disease. Dement Geriatr Cogn Disord 2001;12:15–21.

Hirsch C, Bartenstein P, Minoshima S, Mosch D, Willoch F, Buch K,

Schad D, Schwaiger M, Kurz A. Reduction of regional cerebral blood

flow and cognitive impairment with Alzheimer’s disease: evaluation of

an observer-independent analystic approach. Dement Geriatr Cogn

Disord 1997;8:98–104.

Ibanez V, Pietrini P, Alexander GE, Furey ML, Teichberg D, Rajapakse JC,

Rapoport SI, Schapiro MB, Horwitz B. Regional glucose metabolic

abnormalities are not the result of atrophy in Alzheimer’s disease.

Neurology 1998;50:1585–93.

Johannsen P, Jakobsen J, Bruhn P, Gjedde A. Cortical responses to

sustained and divided attention in Alzheimer’s disease. NeuroImage

1999;10:269–81.

Lovell MA, Xie C, Markesbery WR. Decreased glutathione transferase

activity in brain and ventricular fluid in Alzheimer’s disease. Neurology

1998;51:1562–6.

Lustig C, Snyder AZ, Bhakta M, O’Brien KC, McAvoy M, Raichle ME,

Morris JC, Buckner RL. Functional deactivations: change with age and

dementia of the Alzheimer type. Proc Natl Acad Sci USA 2003;100:

14504–9.

Makuuchi M, Kaminaga T, Sugishita M. Both parietal lobes are involved in

drawing: a functional MRI study and implications for constructional

apraxia. Cogn Brain Res 2003;16:338–47.

Masur DM, Fuld PA, Blau AD, Thal LJ, Levin HS, Aronson MK.

Distinguishing normal and demented elderly with selective reminding

test. J Clin Exp Neuropsychol 1989;11:615–30.

Matsuda H, Kitayama N, Ohnishi T, Asada T, Nakano S, Sakamoto S,

Imabayashi E, Katoh A. Longitudinal evaluation of both morphologic

and functional changes in the same individuals with Alzheimer’s

disease. J Nucl Med 2002;43:304–11.

McKeith IG, Perry EK, Perry RH. Report of the second dementia with

Lewy body international workshop: diagnosis and treatment. Con-

sortium on Dementia with Lewy Bodies. Neurology 1999;53:902–5.

McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan E.

Clinical diagnosis of Alzheimer’s disease: report of the NINCDS

ADRDA work group under the auspices of the Department of Health

and Human Services Task Force on Alzheimer’s Disease. Neurology

1984;34:939–44.

Meguro K, Shimada M, Someya K, Horikawa A, Yamadori A. Hemispatial

visual-searching impairment correlated with decreased contralateral

parietal blood flow in Alzheimer disease. Neuropsychiatry Neuropsy-

chol Behav Neurol 2001;14:213–8.

Mellet E, Tzourio N, Denis M, Mazoyer B. A positron emission

tomography study of visual and mental spatial exploration. J Cogn

Neurosci 1995;7:433–45.

Minoshima S, Giordani B, Berent S, Frey K, Foster NL, Kuhl DE.

Metabolic reduction in the posterior cingulate cortex in very early

Alzheimer’s disease. Ann Neurol 1997;42:85–94.

Montaldi D, Brooks DN, McColl JH, Whyper D, Patterson J, Barron E,

McCulloch J. Measurements of regional cerebral blood flow and

cognitive performance in Alzheimer’s disease. J Neurol Neurosurg

Psychiatry 1990;53:33–8.

Muller NG, Donner TH, Bartelt OA, Brandt SA, Villringer A,

Kleinschmidta A. The functional neuroanatomy of visual conjunction

search: a parametric fMRI study. NeuroImage 2003;20:1578–90.

Neri M, Roth M, Mountjoy CQ, Andermarcher E, Rubichi S, Spano A,

Salvioli G, Cipolli C. Validation of the full and short forms of the

CAMDEX interview for diagnosing dementia. Cambridge Examination

for Mental Disorders of the Elderly. Dementia 1994;5:257–65.

Nobili F, Vitali P, Calvini P, Bollati F, Girtler N, Delmonte M, Mariani G,

Rodriguez G. Clinical correlative evaluation of an iterative method for

reconstruction of brain SPECT images. Nucl Med Biol 2001;28:

627–32.

Nobili F, Koulibaly M, Vitali P, Migneco O, Mariani G, Ebmeier K, Pupi A,

Robert PH, Rodriguez G, Darcourt J. Brain perfusion follow-up in

Alzheimer’s patients during treatment with acetylcholinesterase

inhibitors. J Nucl Med 2002;43:983–90.

O’Brien JT, Eagger S, Syed GMS, Sahakian BJ, Levy R. A study of

regional cerebral blood flow and cognitive performances in Alzheimer’s

disease. J Neurol Neurosurg Psychiatry 1992;55:1182–7.

Patel GA, Sathian K. Visual search: bottom-up or top-down? Front Biosci

2000;5:D169–D193.

Perry RJ, Watson P, Hodges JR. The nature and staging of attention

dysfunction in early (minimal and mild) Alzheimer’s disease:

relationship to episodic and semantic memory impairment. Neuropsy-

chologia 2000;38:252–71.

Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E.

Mild cognitive impairment: clinical characterization and outcome. Arch

Neurol 1999;56:303–8.

Petrides M, Alivisatos B, Evans AC. Functional activation of the human

ventrolateral frontal cortex during mnemonic retrieval of verbal

information. Proc Natl Acad Sci USA 1995;92:5803–7.

Platel H, Proce C, Baron JC, Wise R, Lambert J, Frackowiak RS,

Lechevalier B, Eustache F. The structural components of music

perception: a functional anatomical study. Brain 1997;120:229–43.

Prvulovic D, Hubl D, Sack AT, Melillo L, Maurer K, Frolich L,

Lanfermann H, Zanella FE, Goebel R, Linden DE, Dierks T. Functional

imaging of visuospatial processing in Alzheimer’s disease. NeuroImage

2002;17:1403–14.

Rodriguez G, Vitali P, Calvini P, Bordoni C, Girtler N, Taddei G, Mariani

G, Nobili F. Hippocampal perfusion in mild Alzheimer’s disease.

Psychiatry Res: Neuroimaging 2000;100:65–74.

Rodriguez G, Vitali P, Canfora M, Calvini P, Girtler N, De Leo C,

Piccardo A, Nobili F. Quantative EEG and perfusional single photon

emission computed tomography correlation during long-term donepezil

therapy in Alzheimer’s disease. Clin Neurophysiol 2004;115:39–49.

Roman GC, Tatemichi TK, Erkinjuntti T, Cummings JL, Masdeu JC,

Garcia JH, Amaducci L, Orgogozo JM, Brun A, Hofman A. Vascular

dementia: diagnostic criteria for research studies. Report of the NINDS-

AIREN International Workshop. Neurology 1993;43:250–60.

Salmon E, Collette F, Degueldre C, Lemaire C, Franck G. Voxel-based

analysis of confounding effects of age and dementia severity on cerebral

metabolism in Alzheimer’s disease. Hum Brain Mapp 2000;10:39–48.

Shallice T. Functional imaging and neuropsychology findings: how can

they be linked? NeuroImage 2003;20:S146–S154.

F. Nobili et al. / Clinical Neurophysiology 116 (2005) 364–375 375

Shallice T, Fletcher P, Frith CD, Grasby P, Frackowiak RS, Dolan RJ. Brain

regions associated with acquisition and retrieval of verbal episodic

memory. Nature 1994;368:633–5.

Sidtis JJ, Strother SC, Rottenberg DA. Predicting performance from

functional imaging data: methods matter. NeuroImage 2003;20:615–24.

Spinnler H. La malattia di Alzheimer. In: Denes G, Pizzamiglio L, editors.

Manuale di Neuropsicologia. 2nd ed. Bologna: Zanichelli; 1996. p.

912–74.

Spinnler H, Tognoni G. Standardizzazione e taratura italiana di test

neuropsicologici. Ital J Neurol Sci 1987;6(Suppl 8):1–120.

Sutherland JK, Dougall N, Ebmeier K. Brain behaviour relationships in

dementia (neuropsychology). In: Ebmeier KP, editor. SPECT in

dementia. Advances in biological psychiatry, vol. 22. Basel: Karger;

2003. p. 38–50.

Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain.

New York:: Thieme Medical; 1988.

Tales A, Butler SR, Fossey J, Gilchrist ID, Jones RW, Troscianko T. Visual

search in Alzheimer’s disease: a deficiency in processing conjunctions

of features. Neuropsychologia 2002;40:1849–57.

Tales A, Muir J, Jones R, Bayer A, Snowden RJ. The effects of saliency and

task difficulty on visual search performance in ageing and Alzheimer’s

disease. Neuropsychologia 2004;42:335–45.

Tatsch K, Asenbaum S, Bartenstein P, Catafau A, Halldin C, Pilowsky LS,

Pupi A. European Association of Nuclear Medicine procedure

guidelines for brain perfusion SPET using 99mTc-labelled radio-

pharmaceuticals. Eur J Nucl Med 2002;29:BP36–BP42.

The Lund and Manchester Groups. Clinical and neuropathological criteria for

frontotemporal dementia. J Neurol Neurosurg Psychiatry 1994;57:416–8.