close

Вход

Забыли?

вход по аккаунту

?

Limbic hypometabolism in Alzheimer's disease and mild cognitive impairment.

код для вставкиСкачать
Limbic Hypometabolism in Alzheimer’s
Disease and Mild Cognitive Impairment
Peter J. Nestor, PhD, FRACP,1 Tim D. Fryer, PhD,2 Peter Smielewski, PhD,2
and John R. Hodges, MD, FRCP1,3
The neural basis of the amnesia characterizing early Alzheimer’s disease (AD) remains uncertain. Postmortem pathological studies have suggested early involvement of the mesial temporal lobe, whereas in vivo metabolic studies have shown
hypometabolism of the posterior cingulate cortex. Using a technique that combined the anatomic precision of magnetic
resonance imaging with positron emission tomography, we found severe reductions of metabolism throughout a network
of limbic structures (the hippocampal complex, medial thalamus, mamillary bodies, and posterior cingulate) in patients
with mild AD. We then studied a cohort with mild cognitive impairment in whom amnesia was the only cognitive
abnormality and found comparable hypometabolism through the same network. The AD and mild cognitive impairment
groups were differentiated, however, by changes outside this network, the former showing significant hypometabolism in
amygdala and temporoparietal and frontal association cortex, whereas the latter did not. The amnesia of very early AD
reflects severe but localized limbic dysfunction.
Ann Neurol 2003;54:343–351
Alzheimer’s disease (AD) is the commonest degenerative dementia with an estimated annual incidence of
2.7% in the population aged 75 years and older.1 Clinical criteria for the diagnosis of probable AD mandate
a progressive decline in multiple cognitive domains
that must include memory impairment.2 Mild cognitive impairment (MCI) is defined as subjective memory complaints with objective evidence of memory impairment in the setting of normal general cognitive
function, intact activities of daily living, and no dementia.3 Epidemiological evidence suggests MCI is the
prodrome to AD with conversion rates from MCI to
AD, in studies, ranging from 6 to 25% per annum.3
Thus, isolated memory impairment is almost certainly
the earliest clinical manifestation of AD, and therefore
the study of MCI may offer insights into the earliest
neural correlates of the disease.
Lesion studies in humans have clearly documented
amnesia in subjects with damage to the hippocampi,4
the fornix,5 and the diencephalic projections of the
hippocampus: the mamillary bodies and/or thalamus.6,7 In keeping with these data, it is notable that
certain histopathological changes in AD, neurofibrillary
tangles (NFTs) and neuropil threads, are first seen in
the mesial temporal region.8 However, clinicopatholog-
ical studies suggest that patients may remain cognitively intact despite NFT pathology having spread beyond the mesial temporal lobe,9,10 although NFTs in
association with neuronal loss may be of greater relevance.11,12 It would seem probable, though, that pathology needs to reach a certain threshold to manifest
clinically; by the time such a threshold is reached in
the mesial temporal lobe, subclinical pathology is likely
to have appeared beyond this region. Studies of in vivo
regional brain metabolism offer a method of assessing
the functional significance of such pathology.
The calculation of cerebral metabolic rate of glucose
(CMRglc) with (18F)-2-fluoro-deoxy-D-glucose positron
emission tomography (FDG-PET) has been used extensively in AD. Hypometabolism of temporoparietal
and, later, frontal association cortices with relative sparing of primary sensory and motor cortices is a wellestablished finding,13–15 but the earliest metabolic reductions thus far identified with PET have been in the
posterior cingulate region.16 FDG-PET abnormalities
in mesial temporal lobe structures have, in contrast,
been more difficult to identify (eg, Ishii and colleagues17). Although this may relate to a variety of factors (eg, spatial resolution), one possible explanation
for this apparent paradox could lie in what FDG-PET
From the 1University of Cambridge, Neurology Unit; 2Wolfson
Brain Imaging Centre, Addenbrooke’s Hospital; and 3MRC Cognition and Brain Sciences Unit, Cambridge, United Kingdom.
Address correspondence to Dr Hodges, MRC Cognition and Brain
Sciences Unit, 15 Chaucer Road, Cambridge, CB2 2EF, UK.
E-mail: john.hodges@mrc-cbu.cam.ac.uk
Received Aug 5, 2002, and in revised form May 2, 2003. Accepted
for publication May 7, 2003.
© 2003 American Neurological Association
Published by Wiley-Liss, Inc., through Wiley Subscription Services
343
actually measures. Because CMRglc is primarily a
function of synaptic activity,18 examination of its efferent targets offers an alternative way of assessing
hippocampal complex (HC ⫽ hippocampus proper ⫹
entorhinal cortex) functional integrity. The principle
efferent projection of the hippocampus proper is via
the fornix to the mamillary bodies (MBs) and the laterodorsal nucleus of the thalamus. The MBs, in turn,
project to the anterior thalamic nucleus, while the anterior and laterodorsal nuclei have large projections to
the posterior cingulate cortex (PC).19 The projection
from the PC to the presubicular area20 completes the
so-called “circuit of Papez.”21 There is also a significant direct projection from entorhinal cortex to PC.19
Thus, in addition to studying metabolism in the HC
itself, damage potentially could be evaluated by studying CMRglc in the MBs (with knock-on effects to
thalamus) and PC (both from reduced thalamic and
entorhinal inputs). If changes throughout this network were found, they could help reconcile the observations that pathology begins in the mesial temporal lobe, whereas hypometabolism is first seen in the
posterior cingulate.
We proposed that the amnesia seen in MCI and AD
is a consequence of disruption of critical, focal nodes in
a network, whereas the global dementia of AD is
caused by a more general loss of synaptic density across
the isocortex. On the basis of focal lesion data, we reasoned that these nodes would most likely involve a network of limbic neurons, HC, MBs, “limbic” thalamic
nuclei, and PC, thus the a priori hypothesis of this
study was that hypometabolism at these points could
be common to both MCI and AD. We reasoned that
the failure to identify hypometabolism throughout this
network with voxel-based analysis (eg, Minoshima and
colleagues’ finding of selective posterior cingulate hypometabolism16) was likely to be methodological; for
instance, when CMRglc was calculated with magnetic
resonance imaging (MRI)–guided regions of interest,
hypometabolism in the hippocampus was identified in
MCI,22 suggesting that this method offers greater sensitivity for small brain regions.
The aims of this study therefore were threefold:
first, to establish whether this limbic network could
be shown, with PET, to be hypometabolic in AD; if
so, the second hypothesis was that an MCI group
would show comparable metabolic reductions to AD.
Reasoning that changes in this network would be beneath the resolution of voxel-based techniques such as
Statistical Parametric Mapping (SPM), we used an
MRI-guided region of interest (ROI) method. Each
subject’s CMRglc map was normalized to the CMRglc of their cerebellar vermis to minimize normal interindividual differences in brain metabolism (nCMRglc), and this was coregistered to their threedimensional MRI. The MRI was used to define ROIs
344
Annals of Neurology
Vol 54
No 3
September 2003
and provide a means of correcting the quantification
error in small structures (partial volume error). To
assess the specificity of such findings, our third aim
was to evaluate whether MCI and AD could be dissociated by findings outside this network; thus, we
evaluated amygdala metabolism with an ROI study
and also assessed the extent of isocortical damage using SPM. The former was of particular interest because, in the evolution of NFT changes in AD, the
amygdala is affected after involvement of the transentorhinal and cornu ammonis 1 region of the hippocampus but before the involvement of heteromodal
association cortex.8
Subjects and Methods
Subjects
Patients were recruited from our memory clinic. Subjects
(and caregivers) gave informed consent after detailed explanation of procedures; the study was approved by the local
ethics committee and the Administration of Radioactive Substances Advisory Committee, United Kingdom. Diagnosis
was classified after neurological and detailed neuropsychological assessments and an MRI brain scan. Those with alternate
neurological diagnoses or contraindication to PET were excluded. AD patients met National Institute of Neurological
and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Diseases Association criteria for
probable AD.2 All had amnesia plus impairment in nonmemory domains (scoring in the impaired range on tests in
at least one of the following areas: visuospatial function, attention, language and semantics, and executive function).
MCI was defined by insidious-onset memory symptoms, objective memory impairment (ie, scores ⬍1.5 standard deviations below the age-matched control mean on at least two of
four tests of delayed recall), but no other significant deficits
on detailed neuropsychological examination. The delayed recall tests were story recall (logical memory),23 recall of the
Rey figure,24 and the visual and verbal versions of the Doors
and People Test.25 The nonmnemonic tests used were as follows: attention (span), visuospatial (Visual Object and Space
Perception Battery26; copy of Rey figure24), semantic knowledge/language (Pyramid and Palm Tree Test27; 64-item picture naming28), and executive function (Wisconsin Card
Sort Test29). Controls were recruited from local community
groups: medical and neuropsychological screening showed no
evidence of neurological or major psychiatric illness (including memory impairment).
Imaging
Each subject’s PET and MRI scans were performed at the
same visit or, at most, within a couple of weeks of each
other. MRI was performed on a 3T Bruker scanner. Images
were acquired using a T1-weighted three-dimensional
SPGR sequence (echo time, 5 milliseconds; recovery time,
19.1 milliseconds; field of view, 25.6 ⫻ 22.0 ⫻ 18.0cm;
matrix size, 256 ⫻ 256 ⫻ 256). PET was performed on a
General Electric Advance scanner in three-dimensional
mode (voxel size, 2.35 ⫻ 2.35 ⫻ 4.5mm; field of view,
30.0 ⫻ 30.0 ⫻ 15.3cm). Subjects fasted for 8 hours and
were scanned in a dimly lit, quiet room; blindfolds and
earplugs were not used. A 10-minute preinjection transmission scan was performed using rotating 68Germanium rods
to calculate attenuation correction factors. Subjects then received 74MBq 18flourodeoyglucose over 30 to 60 seconds.
PET images used in this analysis were obtained from 35 to
55 minutes after injection, whereas arterial samples were
taken throughout scanning to define the FDG input function and blood glucose concentration. Images were reconstructed using the PROMIS algorithm,30 with corrections
applied for attenuation, dead time, scatter, and random coincidences.
Image Analysis
ROIs were drawn on
each subject’s MRI scan using AnalyzeAVW 3 (Biomedical
Imaging Resource, Mayo Foundation, Rochester, MN). To
maximize consistency, we oriented each subject’s scan to stereotaxic space (aligned to the ac-pc line), allowing individual
MRI scans to be compared directly with the format of the
Duvernoy brain atlas.31 ROIs were drawn for HC, MBs,
thalami, and PC for the first experiment and amygdala for
the second. Full details of the drawing method are at http://
www.wbic.cam.ac.uk. ROI tracing was performed blinded to
diagnostic group.
The CMRglc map was calculated using the Huang autoradiographic method.32 Each CMRglc map was coregistered to their MRI and resliced to MRI voxel size using
SPM99 (Wellcome Department of Cognitive Neurology,
London, UK) and Matlab5.2 software (Mathworks, Natick,
MA). SPM99 was also used to segment MRI into gray matter, white matter, and cerebrospinal fluid and smooth these
segments to the resolution of PET (6mm full-width at halfmaximum [FWHM] Gaussian). Within each ROI, the
smoothed segments were used to calculate the mean CMRglc in gray and white matter and hence correct the artefactual decrease (partial volume error) in mean ROI CMRglc seen because of the presence of cerebrospinal fluid,
using techniques similar to those previously published.33
To minimize intersubject variability, we then normalized
these mean CMRglc to the cerebellar ROI CMRglc to produce nCMRglc.34 Test-retest reliability of the ROI method
with Pearson’s correlation coefficient yielded r greater than
0.98 in all regions (derived from recalculating regional
CMRglc for ROIs redrawn from scratch 6 months after the
initial study).
REGION OF INTEREST METHOD.
Each subject’s CMRglc map was normalized to their cerebellar vermis
STATISTICAL PARAMETRIC MAPPING ANALYSIS.
CMRGlc (nCMRGlc, see above) and then coregistered to
their MRI. The MRI and hence the coregistered PET were
spatially normalized to the T1-MR template in SPM99 and
then smoothed (16mm FWHM Gaussian). Voxels with values greater than 40% of the whole brain mean nCMRglc
were included in the analysis.35 Statistical analysis involved a
one-way analysis of variance (ANOVA) of the three groups
and then post hoc unpaired t tests (control vs AD and control vs MCI). Statistical threshold was set at the p value corrected equals 0.05 but also analyzed at p value equals 0.1 to
ensure type 2 errors were minimized. To compare individual
cases with the control group, we selected a significance level
of p value uncorrected less than 0.001. Anatomical localization of clusters were ascertained using the coplanar stereotaxic atlas.36
Results
Cognitive Assessments
There was a significant difference in age between AD
and controls ( p ⫽ 0.02) but not in the other two
group comparisons. On global measures, the MCI
group scored at near normal levels on Mini-Mental Examination (MMSE)37 (28.4 ⫾ 1.5), although the
slight reduction was statistically significant ( p ⬍
0.005), controls being at the ceiling. The three groups
were separated from each other on the Addenbrooke’s
Cognitive Examination, a test designed to be more sensitive to MCI.38 The MCI group had significantly
more years education ( p ⫽ 0.03) than both AD and
controls (Table 1).
The AD and MCI groups detailed neuropsychological testing showed that both performed well on tests of
attention and both were impaired on tests of delayed
memory recall. In addition, as a group, the AD cases
performed significantly worse on language/semantics,
visuospatial, and executive tests (Table 2).
Region of Interest Study of the Limbic Network
One-way ANOVAs (degrees of freedom ⫽ 2, 32) of
the network ROIs showed significant group effects in
all regions: HC (right F ⫽ 10.4, p ⬍ 0.0005; left F ⫽
15.0, p ⫽ 0.0001), MBs (F ⫽ 16.1, p ⫽ 0.0001),
thalamus (right F ⫽ 6.5, p ⬍ 0.005; left F ⫽ 7.0, p ⬍
0.005), and PC (right F ⫽ 25.5, p ⫽ 0.0001; left F ⫽
30.7, p ⫽ 0.0001). When contrasted with controls,
post hoc two-tailed unpaired t tests showed significant
Table 1. Demographics of Atzheimer’s Disease, Mild Cognitive Impairment and Control Groups
Group
Sex (F:M)
Age (yr)
Education (yr)
MMSE37 (maximum ⫽ 30)
ACE38 (maximum ⫽ 100)
AD (n ⫽ 10)
MCI (n ⫽ 10)
Controls (n ⫽ 15)
4 F, 6 M
68.1 ⫾ 7.4
10.9 ⫾ 1.6
22.8 ⫾ 1.8
73.7 ⫾ 9.2
3 F, 7 M
63.3 ⫾ 5.9
14.3 ⫾ 4.0
28.4 ⫾ 1.5
84.8 ⫾ 4.3
7 F, 8 M
60.8 ⫾ 7.1
11.6 ⫾ 1.5
29.7 ⫾ 0.5
96.1 ⫾ 2.9
MMSE ⫽ Mini-Mental State Examination; ACE ⫽ Addenbrooke’s Cognitive Examination.
Nestor et al: Limbic System in AD and MCI
345
Table 2. Summary of Neuropsychological Results of AD and MCI Groups
AD
MCI
Aged Controls
(mean minus 1SD)
6.7 ⫾ 1
4.3 ⫾ 1.7
6.7 ⫾ 1.3
4.5 ⫾ 1.3
6.1
4.0
0.1 ⫾ 0.3
2.8 ⫾ 2.5
0.6 ⫾ 0.6
7.5 ⫾ 5.1
4.0
13.0
2.5 ⫾ 2.6
1.4 ⫾ 2.1
4.5 ⫾ 2.8
1.0 ⫾ 1.6
7.4
5.2
59.8 ⫾ 4.1
49 ⫾ 2.4
62.6 ⫾ 1.1a
51 ⫾ 1.3a
61.4
50.6
31.7 ⫾ 3.8
35 ⫾ 1.6a
32.6
17.3 ⫾ 2.3
8.1 ⫾ 1.6
18.6 ⫾ 1.6
8.8 ⫾ 1.8
14.6
7.8
3.4 ⫾ 1.9
5.3 ⫾ 0.9a
5.4
Group
Attention
Forward digit span
Backward digit span
Memory (delayed recall)
Logical memory23
Rey complex figure24
Doors and People test25
Visual
Verbal
Semantics/language
64-Item picture naming28
Pyramid and Palm Tree test (picture version)27
Visuospatial
Copy of Rey figure24
Visual Object and Space Perception Battery26
Object decision
Cube analysis
Executive
Wisconsin Card Sort Test (Max. ⫽ 6)29
p ⬍ 0.05 AD vs MCI.
AD ⫽ Alzheimer’s disease; MCI ⫽ mild cognitive impairment.
a
reductions of nCMRglc in AD for all regions (right
HC, p ⫽ 0.0001; left HC, p ⬍ 0.0001; MBs, p ⬍
0.0001; right thalamus, p ⬍ 0.005; left thalamus, p ⫽
0.001; right PC, p ⬍ 0.0000001; left PC, p ⬍
0.0000001), whereas in MCI all regions were significantly reduced with the exception of the right thalamus
that showed a strong trend to significance (right HC,
p ⫽ 0.03; left HC, p ⬍ 0.001; MBs, p ⬍ 0.005; right
thalamus, p ⫽ 0.055; left thalamus, p ⫽ 0.02; right
PC, p ⬍ 0.00005; left PC, p ⬍ 0.00001). There were
no statistically significant differences in any region between the AD and MCI groups (Fig 1 and Table 3).
Region of Interest Study of the Amygdalae
One-way ANOVAs (degrees of freedom ⫽ 2, 32) of
right and left amygdalae showed significant group effects (right F ⫽ 4.0, p ⫽ 0.03; left F ⫽ 9.1, p ⬍
0.001). Post hoc unpaired t tests showed significant bilateral reductions of nCMRglc in AD (right p ⫽ 0.01,
left p ⬍ 0.001), whereas the MCI group did not differ
significantly from controls (right p ⫽ 0.28, left p ⫽
0.17). Furthermore, there was a statistically significant
reduction in nCMRglc in AD when compared with
MCI on the left ( p ⫽ 0.02) though not for the right
( p ⫽ 0.17; see Fig 1 and Table 3).
Statistical Parametric Mapping Analysis
A one-way ANOVA of the three groups showed a significant group effect (data not shown). SPM analysis of
AD minus controls showed two abnormal clusters: posterior cingulate/retrosplenial cortex (ie, BA 23/29/30/
31) extending bilaterally to temporoparietal association
346
Annals of Neurology
Vol 54
No 3
September 2003
cortex and left frontal cortex. There was a single abnormal cluster in the MCI minus controls comparison
involving posterior cingulate/retrosplenial cortex and
cinguloparietal region (Fig 2 and Table 4). At a reduced statistical threshold (P corrected ⫽ 0.1), the distribution of clusters was similar in both groups, although each cluster was slightly larger (at p value
corrected ⫽ 0.1: AD clusters ⫽ 26,281 and 4,295 voxels; MCI ⫽ 7,541 voxels).
To ensure that the MCI result was not merely an
artefact of pooling cases of AD pathology with unaffected subjects, we also analyzed the MCI subjects as
individuals against the control group. Two subjects
showed clear evidence of association cortex abnormality
typical of AD and an additional three showed minor
abnormalities in the same areas. When these five cases
(MCIplus) were contrasted with the five with no such
abnormalities (MCIminus), the mean scores for the MB
ROI were identical (mean nCMRglc: MCIplus ⫽
0.683 ⫾ 0.10 vs MCIminus ⫽ 0.696 ⫾ 0.06, p ⫽
0.81; controls ⫽ 0.829 ⫾ 0.11), suggesting an equal
loss of hippocampal synaptic input.
Discussion
We have confirmed that AD is associated with dysfunction of the network of neurons implicated in human amnesia. This finding, although not unexpected,
is of interest in two respects. First, by using a partial
volume corrected ROI technique in which regions
were defined on a coregistered MRI, highly significant hypometabolism was seen in the hippocampal
complex, suggesting that earlier negative results were
Fig 1. Results for the region of interest studies. AD ⫽ Alzheimer’s disease; MCI ⫽ mild cognitive impairment; C ⫽control.
a consequence of lack of spatial resolution and/or
poor localization. More importantly, studies in which
mesial temporal hypometabolism has been observed
were conducted in established AD39,40 rather than its
amnesic prodrome, and therefore our finding of comparable results in a group with MCI is highly relevant. The result is consistent with a recent study of
the temporal lobe, using a similar MRI and PET approach to ours, that showed hypometabolism restricted to hippocampus and parahippocampal gyrus
in MCI, whereas AD showed additional temporal
neocortical hypometabolism.22 Second, the mamillary
bodies have not been imaged previously with PET.
They compose a dense population of synaptic termi-
Nestor et al: Limbic System in AD and MCI
347
Table 3. Mean ⫾ Standard Deviation by Group for Region of Interest Experiments
Region
Limbic network
Hippocampal complex
Right
Left
Mamillary bodies
Thalamus
Right
Left
Posterior cingulate
Right
Left
Amygdala
Right
Left
AD
MCI
Controls
0.66 ⫾ 0.13a
0.69 ⫾ 0.10a
0.61 ⫾ 0.09a
0.77 ⫾ 0.11b
0.75 ⫾ 0.08a
0.69 ⫾ 0.08a
0.85 ⫾ 0.08
0.89 ⫾ 0.09
0.83 ⫾ 0.11
1.04 ⫾ 0.17a
1.14 ⫾ 0.14a
1.14 ⫾ 0.11c
1.20 ⫾ 0.15b
1.24 ⫾ 0.13
1.34 ⫾ 0.12
1.02 ⫾ 0.14a
0.96 ⫾ 0.13a
1.10 ⫾ 0.17a
1.03 ⫾ 0.14a
1.38 ⫾ 0.10
1.37 ⫾ 0.15
0.61 ⫾ 0.16b
0.58 ⫾ 0.16a
0.71 ⫾ 0.12
0.73 ⫾ 0.10
0.76 ⫾ 0.10
0.79 ⫾ 0.10
Versus controls: ap ⬍ 0.01,
b
p ⬍ 0.05, cp ⫽ 0.055; no statistically significant differences between AD and MCI groups with the exception of the left amygdala, p ⬍ 0.05.
AD ⫽ Alzheimer’s disease; MCI ⫽ mild cognitive impairment.
nals whose cell bodies are located in the hippocampus
and can, therefore, serve as an in vivo marker of hippocampal functional integrity.
Observing comparable reductions in nCMRglc
throughout the network in MCI supports the hypothesis that dysfunction of these structures heralds the beginning of AD. Furthermore, hypometabolism of the
posterior cingulate previously observed in cases with
very early-stage AD (MMSE 25 ⫾ 1)16 was extended
in this study to patients with MCI (MMSE 28.4 ⫾
1.5). Because the MCI group did not show significant
reductions in nCMRglc in the amygdala or lateral cortical regions, we argue that our findings offer strong
evidence that damage to the limbic network is the first
clinically significant event in the evolution of AD.
From a methodological perspective, these data are in
agreement with those of De Santi and colleagues22 in
showing that greater sensitivity at identifying metabolic
changes in small structures can be achieved using MRIbased ROI analysis rather than SPM. Unlike voxelbased methods, this technique does not involve any
nonlinear spatial transformation, a significant source of
spatial localization error, or heavy smoothing. Visual
inspection of the coregistered PET and MRI images
showed good agreement (typically approximately 1mm
linear error). Quantifying metabolism in structures
whose size is comparable with the resolution of PET
(6mm FWHM on this scanner) is subject to significant
error because of the blurring on the radioactivity distribution. MRI-based partial volume correction was
used to minimize these artefactual errors; nevertheless,
partial volume error cannot completely be eliminated
because of the coregistration error and errors in segmenting the brain.
The extent to which the changes found in the limbic
network reflect pathology, primarily, at each node ver-
348
Annals of Neurology
Vol 54
No 3
September 2003
sus physiological deafferentation from pathology upstream, cannot be definitively answered in this study. It
is possible that the entire network abnormality is principally a consequence of HC damage. Reduction in cerebral perfusion (H215O-PET) has been demonstrated
in the thalamus and PC of subjects after mesial temporal lobectomy for epilepsy.41 Based on the Braak
staging system,8 the finding of reduced CMRglc in
thalamus and PC relative to amygdala would support
the deafferentation hypothesis given that the amygdala
is involved at an earlier stage than these other structures.
Finally, our results provide an interesting contrast
to the PET studies of presymptomatic subjects at risk
of genetically determined AD,42– 46 who show
changes in lateral temporoparietal association cortex
in a similar pattern to that seen in symptomatic cases.
In particular, the study of Reiman and colleagues46
explicitly excluded subjects with MCI by our current
definition and yet still found such changes. Given
that genetic factors (mutations of amyloid precursor
protein and presenilin genes, and, apolipoprotein E
status) are thought to confer their effect by modulating amyloid metabolism, it may be that the landscape
of early-stage familial and sporadic AD differ. Our
study may have been underpowered to detect subtle
changes in lateral temporoparietal association cortex
in MCI, but considering the profound changes that
we have shown in the limbic system it would seem
very unlikely that lateral cortical changes, if present,
are contributing significantly to the symptomatology.
We would expect, however, amnesic symptoms to
manifest in familial AD only when the limbic system
becomes hypometabolic. A direct comparison of preand postsymptomatic familial AD and sporadic MCI
Fig 2. Statistical Parametric Mapping–rendered brain images of hypometabolic regions
in the Alzheimer’s disease (AD) group and
mild cognitive impairment (MCI) group versus controls at p value corrected ⫽ 0.05.
Nestor et al: Limbic System in AD and MCI
349
Table 4. Talairach Coordinates of Abnormal Clusters from Statistical Parametric Mapping Analysis in AD and MCI Groups
Group
Cluster
Size (KE)
p
(corrected)
T
Coordinates
x, y, z (mm)
Location
20,143
0.000
0.000
0.000
0.000
0.001
0.002
0.002
0.002
0.026
0.040
0.006
0.010
0.025
0.026
0.026
0.027
10.19
10.06
9.44
8.65
7.07
6.77
6.75
6.70
5.44
5.20
6.23
5.99
5.46
5.44
5.43
5.42
54, ⫺62, 34
48, ⫺72, 26
6, ⫺54, 26
62, ⫺56, 2
⫺44, ⫺56, 28
20, ⫺58, 58
16, ⫺62, 56
⫺42, ⫺48, 38
⫺28, ⫺70, 36
⫺14, ⫺44, 2
⫺24, 60, 4
⫺24, 50, 26
2, 52, 14
⫺42, 34, 26
⫺40, 34, 30
⫺48, 34, 16
R. supramarginal gyrus
R. angular gyrus/inferior parietal lobule
Posterior cingulate
R. middle temporal gyrus
L. supramarginal gyrus/inferior parietal lobule
R. superior parietal lobule
R. superior parietal lobule
L. inferior parietal lobule
L. superior parietal lobule
L. retrosplenial cortex/posterior parahippocampal gyrus
L. frontal pole
L. superior frontal gyrus
Medial frontal gyrus
L. middle frontal gyrus
L. middle frontal gyrus
L. middle frontal gyrus
0.000
0.002
0.002
0.004
0.004
0.024
7.68
6.89
6.65
6.99
6.99
5.44
⫺12, ⫺38, 4
8, ⫺34, 0
2, ⫺24, 30
⫺10, ⫺62, 32
4, ⫺58, 30
⫺18, ⫺48, 22
L. retrosplenial cortex/posterior parahippocampal gyrus
R. retrosplenial cortex/posterior parahippocampal gyrus
R. posterior cingulate
L. posterior cingulate/precuneus
R. posterior cingulate
L. posterior cingulate
AD
1,999
MCI
5,330
AD ⫽ Alzheimer’s disease; MCI ⫽ mild cognitive impairment.
using the same techniques clearly would be of considerable interest.
This work was supported by a United Kingdom Medical Research
Council Programme Grant (G9724461, J.R.H.).
References
1. Paykel ES, Huppert FA, Brayne C. Incidence of dementia and
cognitive decline in over-75s in Cambridge: overview of cohort
study. Soc Psychiatry Psychiatr Epidemiol 1998;33:387–392.
2. McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of Alzheimer’s disease. Neurology 1984;34:939 –944.
3. Petersen RC, Stevens JC, Ganguli M, et al. Practice parameter:
early detection of dementia: mild cognitive impairment (an
evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology
2001;56:1133–1142.
4. Zola-Morgan S, Squire LR, Amaral DG. Human amnesia and
the medial temporal region: enduring memory impairment following a bilateral lesion limited to field CA1 of the hippocampus. J Neurosci 1986;6:2950 –2967.
5. D’Esposito M, Verfaellie M, Alexander MP, Katz DI. Amnesia
following traumatic bilateral fornix transection. Neurology
1995;45:1546 –1550.
6. Squire LR, Amaral DG, Zola-Morgan S, et al. Description of
brain injury in the amnesic patient N.A. based on magnetic
resonance imaging. Exp Neurol 1989;105:23–35.
7. Stuss DT, Guberman A, Nelson R, Larochelle S. The neuropsychology of paramedian thalamic infarction. Brain Cogn
1988;8:348 –378.
8. Braak H, Braak E. Neuropathological staging of Alzheimerrelated changes. Acta Neuropathol 1991;82:239 –259.
350
Annals of Neurology
Vol 54
No 3
September 2003
9. Price JL, Morris JC. Tangles and plaques in nondemented aging and “preclinical” Alzheimer’s disease. Ann Neurol 1999;45:
358 –368.
10. Delacourte A, David JP, Sergeant N, et al. The biochemical
pathway of neurofibrillary degeneration in aging and Alzheimer’s disease. Neurology 1999;52:1158 –1165.
11. Gomez-Isla T, Price JL, McKeel DW Jr, et al. Profound loss of
layer II entorhinal cortex neurons occurs in very mild Alzheimer’s disease. J Neurosci 1996;16:4491– 4500.
12. Gomez-Isla T, Hollister R, West H, et al. Neuronal loss correlates with but exceeds neurofibrillary tangles in Alzheimer’s disease. Ann Neurol 1997;41:17–24.
13. Cutler NR, Haxby JV, Duara R, et al. Clinical history, brain
metabolism, and neuropsychological function in Alzheimer’s
disease. Ann Neurol 1985;18:298 –309.
14. Foster NL, Chase TN, Mansi L, et al. Cortical abnormalities in
Alzheimer’s disease. Ann Neurol 1984;16:649 – 654.
15. Chase TN, Foster NL, Fedio P, et al. Regional cortical dysfunction in Alzheimer’s disease as determined by positron emission
tomography. Ann Neurol 1984;15:S170 –S174.
16. Minoshima S, Giordani B, Berent S, et al. Metabolic reduction
in the posterior cingulate cortex in very early Alzheimer’s disease. Ann Neurol 1997;42:85–94.
17. Ishii K, Sasaki M, Yamaji S, et al. Relatively preserved hippocampal glucose metabolism in mild Alzheimer’s disease. Dement Geriatr Cogn Disord 1998;9:317–322.
18. Kadekaro M, Crane AM, Sokoloff L. Differential effects of electrical stimulation of sciatic nerve on metabolic activity in spinal
cord and dorsal root ganglion in the rat. Proc Natl Acad Sci
USA 1985;82:6010 – 6013.
19. Baleydier C, Mauguiere F. The duality of the cingulate gyrus in
monkey. Neuroanatomical study and functional hypothesis.
Brain 1980;103:525–554.
20. Pandya DN, Van Hoesen GW, Mesulam MM. Efferent connections of the cingulate gyrus in the rhesus monkey. Exp Brain
Res 1981;42:319 –330.
21. Papez JW. A proposed mechanism of emotion. Arch Neurol
Psychiatry 1937;38:725–743.
22. De Santi S, de Leon MJ, Rusinek H, et al. Hippocampal formation glucose metabolism and volume losses in MCI and AD.
Neurobiol Aging 2001;22:529 –539.
23. Wechsler DA. Wechsler Memory Scale–Revised. San Antonio,
TX: Psychological Corporation, 1987.
24. Osterreith PA. Le test de copie d’une figure complexe. Arch
Psychol 1944;30:206 –256.
25. Baddeley AD, Emslie H, Nimmo-Smith I. The Doors and People Test: a test of visual and verbal recall and recognition. Bury
St Edmunds, UK: Thames Valley Test Company, 1994.
26. Warrington EK, James M. The Visual Object and Space Perception Battery. Bury St Edmunds, UK: Thames Valley Test
Company, 1991.
27. Howard D, Patterson K. Pyramids and Palm Trees: a test of
semantic access from pictures and words. Bury St Edmunds,
UK: Thames Valley Publishing Company, 1992.
28. Bozeat S, Lambon Ralph MA, Patterson K, et al. Non-verbal
semantic impairment in semantic dementia. Neuropsychologia
2000;38:1207–1215.
29. Heaton RK. Wisconsin card sorting test manual. Odessa, FL:
Psychology Assessment Resources, 1981.
30. Kinahan PE, Rogers JG. Analytic 3D image reconstruction using all detected events. IEEE Trans Nucl Sci 1989;36:964 –968.
31. Duvernoy HM. The human brain: surface, three-dimensional
sectional anatomy with MRI, and blood supply. New York:
Springer-Verlag, 1999.
32. Phelps ME, Huang SC, Hoffman EJ, et al. Tomographic measurement of local cerebral glucose metabolic rate in humans
with (F-18)2-fluoro-2-deoxy-D-glucose: validation of method.
Ann Neurol 1979;6:371–388.
33. Meltzer CC, Kinahan PE, Greer PJ, et al. Comparative evaluation of MR-based partial-volume correction schemes for PET.
J Nucl Med 1999;40:2053–2065.
34. Ichimiya A, Herholz K, Mielke R, et al. Difference of regional
cerebral metabolic pattern between presenile and senile dementia of the Alzheimer type: a factor analytic study. J Neurol Sci
1994;123:11–17.
35. Desgranges B, Baron JC, de la Sayette V, et al. The neural
substrates of memory systems impairment in Alzheimer’s disease. A PET study of resting brain glucose utilization. Brain
1998;121:611– 631.
36. Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain. 3-Dimensional proportional system: an approach to
cerebral imaging. Stuttgart, Germany: George Thieme Verlag,
1988.
37. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state.” A
practical method for grading the mental state of patients for the
clinician. J Psychiat Res 1975;12:189 –198.
38. Mathuranath PS, Nestor PJ, Berrios GE, et al. A brief cognitive
test battery to differentiate Alzheimer’s disease and frontotemporal dementia. Neurology 2000;55:1613–1620.
39. Ishii K, Kitagaki H, Kono M, Mori E. Decreased medial temporal oxygen metabolism in Alzheimer’s disease shown by PET.
J Nucl Med 1996;37:1159 –1165.
40. Ouchi Y, Nobezawa S, Okada H, et al. Altered glucose metabolism in the hippocampal head in memory impairment. Neurology 1998;51:136 –142.
41. Minoshima S, Cross DJ, Foster NL, et al. Discordance between
traditional pathologic and energy metabolic changes in very
early Alzheimer’s disease. Pathophysiological implications. Ann
N Y Acad Sci 1999;893:350 –352.
42. Kennedy AM, Frackowiak RS, Newman SK, et al. Deficits in
cerebral glucose metabolism demonstrated by positron emission
tomography in individuals at risk of familial Alzheimer’s disease. Neurosci Lett 1995;186:17–20.
43. Small GW, Mazziotta JC, Collins MT, et al. Apolipoprotein
E type 4 allele and cerebral glucose metabolism in relatives at
risk for familial Alzheimer disease. JAMA 1995;273:942–947.
44. Small GW, Ercoli LM, Silverman DH, et al. Cerebral metabolic and cognitive decline in persons at genetic risk for Alzheimer’s disease. Proc Natl Acad Sci USA 2000;97:6037–
6042.
45. Reiman EM, Caselli RJ, Chen K, et al. Declining brain activity
in cognitively normal apolipoprotein E epsilon 4 heterozygotes:
a foundation for using positron emission tomography to efficiently test treatments to prevent Alzheimer’s disease. Proc Natl
Acad Sci USA 2001;98:3334 –3339.
46. Reiman EM, Caselli RJ, Yun LS, et al. Preclinical evidence of
Alzheimer’s disease in persons homozygous for the epsilon 4
allele for apolipoprotein E. N Engl J Med 1996;334:752–758.
Nestor et al: Limbic System in AD and MCI
351
Документ
Категория
Без категории
Просмотров
6
Размер файла
237 Кб
Теги
mild, limbic, hypometabolism, impairments, disease, cognitive, alzheimers
1/--страниц
Пожаловаться на содержимое документа