Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.

com
REVIEWS

Neuroimaging in dementia:
a practical guide
Alex M Mortimer, Marcus Likeman, Timothy T Lewis

▸ Additional material is ABSTRACT regions and, eventually, of all cortical

published online only. To view
Over 800 000 people in the UK are demented. regions. Atrophy of the whole brain (and
please visit the journal online
(http://dx.doi.org/10.1136/ Alzheimer’s disease, dementia with Lewy bodies, hippocampus in particular) accelerates as
practneurol-2012-000337). vascular dementia and frontotemporal lobar the patient moves from normality to cog-
degeneration account for the majority. Although nitive impairment,4 and as dementia pro-
Department of Neuroradiology,
Frenchay Hospital, Bristol, UK detailed clinical assessment forms the basis of gresses.5 The cortical volume loss in AD
evaluating a patient with cognitive impairment, leads to ventricular enlargement, white
Correspondence to structural and functional imaging techniques are matter tract atrophy (including the
Dr Alex Mark Mortimer,
Department of Neuroradiology,
increasingly being used. Neuroimaging can corpus callosum), and smaller total brain
Frenchay Hospital, Frenchay Park identify changes to supplement the clinical volume (figure 1).6–9
Road, Bristol, BS16 1LE; diagnosis and help to distinguish dementia Atrophy within the medial temporal
[email protected] subtypes. This may be important for treatment, structures is the most studied structural
prognosis and care planning. Furthermore, early biomarker of AD (figure 2). These struc-
changes on structural and functional imaging tures atrophy at a greater rate than in
may have a role in preclinical detection, perhaps normal controls, and this correlates with
allowing people to start any treatments early. In progressive neurofibrillary tangle depos-
this review, we explain the tools available to the ition and also with cognitive tests scores.8
neuroradiologist and examine the implications of Although volumetric techniques are used
imaging findings in assessing patients with in the research setting to assess the
cognitive impairment or dementia. medial temporal lobe, these are not estab-
lished in clinical practice; the mainstay of
NEUROIMAGING IN DEMENTIA: assessment is visual inspection supple-
A PRACTICAL GUIDE mented by visual rating scales and linear
Over 800 000 people in the UK are measurements10 11 (figures 3 and 4).
demented.1 Alzheimer’s disease (AD) Scheltens’ medial temporal lobe atrophy
accounts for 60% of cases; dementia with (MTA) score is the most widely published
Lewy bodies, vascular dementia and fron- visual rating scale.12 This uses coronal
totemporal lobar degeneration account T1-weighted MRI acquired parallel to
for most of the remainder.1 Although the brainstem, but is also reliable when
detailed clinical assessment forms the using multidetector CT coronal recon-
basis of evaluating a patient with cognitive structions13 (figure 4 and table 1).
impairment, imaging techniques are MTA is characteristic of AD, but also
increasingly being used, for instance, occurs in other forms of dementia,
structural imaging with CT and MR scans, including frontotemporal lobar degener-
and functional imaging (for selected ation, vascular dementia and dementia
cases), using Tc99-HMPAO SPECT, with Lewy bodies.14–16 However, MTA
18
F-FDG PET or FP-CIT (decreased dopa- scores are lower in other dementias, and
mine transporter (DAT) scan). the atrophy is often asymmetrical in fron-
totemporal lobar degeneration. In some
ALZHEIMER’S DISEASE patients with AD ( particularly if early or
Structural imaging atypical), MTA is mild and difficult to
Longitudinal neuroimaging studies2 3 detect.17 18 In early AD at or before the
show that patients with early mild cogni- minimal cognitive impairment stage,
tive impairment have medial temporal there may be parietal atrophy, particularly
To cite: Mortimer AM, lobe involvement. As cognitive decline of the precuneus and posterior cingulate
Likeman M, Lewis TT. Pract progresses, there is increasing involve- regions; this also occurs in patients with
Neurol 2013;13:92–103. ment of the temporal neocortex, parietal posterior cortical atrophy (see figures 5

92 Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

and 6)19 20. Parietal atrophy may also help to distin-

guish between AD and frontotemporal lobar degener-
ation,21 summarised as an imaging scoring system20
(table 2).

Functional imaging
In patients with suspected AD whose structural
imaging is equivocal, one option is to repeat and
compare the imaging after 12–24 months, looking for
focal atrophy. Alternatively, functional imaging can
help. The classical abnormality on 99mTc-HMPAO
SPECT and 18F-FDG PET imaging is temporoparietal
hypoperfusion and hypometabolism; this is usually
symmetrical, and with relative preservation of basal
ganglia and cerebellum (figure 7).22–24 With disease
progression, there may be frontal hypoperfusion/
Figure 1 Coronal CT reformat showing structural imaging
metabolism.23 99mTc-HMPAO SPECT imaging differ-
appearances in advanced Alzheimer’s disease with severe entiates AD patients from healthy controls with a
medial temporal atrophy (arrows) and ventricular enlargement. sensitivity of 89% and specificity 80%.25 Also,
99m
Tc-HMPAO SPECT can distinguish AD from other

Figure 2 Coronal T1-weighted image showing important medial temporal structures in Alzheimer’s disease. (A) temporal horn; (B)
choroidal fissure; (C) hippocampus; (D) parahippocampal gyrus.

Figure 3 Axial CT image orientated parallel to the temporal lobes demonstrating linear measures of the medial temporal lobes. (A)
radial width of the temporal horn (using 8 mm as a cut off, an initial publication10 suggested high sensitivity and specificity of ~92%
but this has subsequently been questioned11) (B) minimum thickness of the temporal lobe (a threshold value for Alzheimer’s disease
of >5.3 mm gives a sensitivity of 93% and specificity of 97%11).

Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337 93

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

Figure 4 Scheltens’ scale of medial temporal atrophy. Scores 0–4 indicate progressive medial temporal volume loss. Medial
temporal lobe atrophy (MTA) is assessed using a 5-point scale from 0–4, with scores deduced through assessment of the
hippocampal height and width of the choroid fissure and temporal horn. For subjects younger than 75 years, a MTA score of 2 or
more is abnormal, while for subjects older than 75 years, a MTA score of 3 or more is abnormal (table 1).

Table 1 Scheltens’ medial temporal lobe atrophy (MTA) score dementias, such as frontotemporal lobar degeneration
(N=normal) and vascular dementia, with a sensitivity and specifi-
Height of city of 70–79%.26
Width of Width of hippocampal
Score choroid fissure temporal horn formation Vascular changes in AD
0 N N N AD patients have more white matter disease than con-
1 ↑ N N trols.27 28 However, it is unclear whether the white
2 ↑↑ ↑ ↓ matter T2-hyperintensity of AD reflect a secondary
3 ↑↑↑ ↑↑ ↓↓ pathological process, relate to cerebral amyloid angio-
4 ↑↑↑ ↑↑↑ ↓↓↓ pathy or represent coexisting vascular pathology from
Please also refer to figure 4.
hypertension, smoking and diabetes mellitus.

Figure 5 Sagittal, coronal and axial images demonstrating parietal atrophy. White arrowheads indicate widening of the posterior
cingulate sulcus. Black arrowheads indicate parieto-occipital sulcal widening. Solid arrows indicate lateral parietal sulcal widening.

Figure 6 Posterior cortical atrophy. T1-weighted axial and sagittal MRI (left and centre) plus FLAIR coronal image (right)
demonstrating severe parieto-occipital atrophy. Contrary to typical Alzheimer’s disease, patients with posterior cortical atrophy have
well preserved memory and language but show a rapid and relatively selective decline in vision and/or literacy skills.

94 Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

Table 2 Koedam’s parietal atrophy score.20 (In case of different

scores on different orientations (eg, score 1 on sagittal direction
and score 2 on axial direction), the highest score is considered)
Grade Description
0 ▪ Closed posterior cingulate- and parieto-occipital sulcus
▪ Closed sulci of the parietal lobes and precuneus
1 ▪ Mild widening of the posterior cingulate- and parieto-occipital
sulcus,
▪ Mild atrophy of the parietal lobes and precuneus
2 ▪ Substantial widening of the posterior cingulate- and
parieto-occipital sulcus
▪ Substantial atrophy of the parietal lobes and precuneus
3 ▪ End-stage atrophy with evident widening of both sulci and
knife-blade atrophy of the parietal lobes and precuneus

Microbleeds are another important feature. These

appear as focal hypointensities on T2*-weighted or
susceptibility-weighted images. They are found in
about 20% of AD patients,29 probably relating to cere-
bral amyloid angiopathy, found in >90% of autopsied
AD brains.30 Microbleeds occur mainly in the cortical–
subcortical area in patients with suspected cerebral
amyloid angiopathy or AD (figure 8).31 Microbleeds
on susceptibility-weighted images may predict cogni- Figure 8 T2* gradient echo image in a patient with
tive decline and possibly increased risk of conversion Alzheimer’s disease demonstrating multiple subcortical
from minimal cognitive impairment to AD.30 microhaemorrhages seen as punctuate foci of hypointensity.

FRONTOTEMPORAL LOBAR DEGENERATION below the age of 65 years,33 and can be categorised
Frontotemporal lobar degeneration is a heterogeneous into behavioural variant frontotemporal dementia and
group of clinicoanatomical subtypes with various primary progressive aphasia. Primary progressive
underlying pathologies. These include tau and ubiqui- aphasia is itself divided into three subtypes, non-fluent
tin inclusions, ‘fused in sarcoma’ pathology, ‘dementia progressive aphasia, semantic dementia and logopenic
lacking distinctive histology’ and even AD-type path- progressive aphasia.
ology.32 This group of conditions is as common as AD

Behavioural variant frontotemporal lobe dementia

This variant is characterised by atrophy of the frontal
and temporal lobes (including medial temporal struc-
tures such as the hippocampus) and largely sparing
the parietal lobe (figure 9),21 with an anterior to pos-
terior gradient, and often asymmetric, with the left
side more atrophic34–36 (table 3).
99m
Tc-HMPAO SPECT shows reduced perfusion of the
frontal and anterior temporal lobes22 26 37 (figure 10).
The frontal hypometabolism and perfusion are asym-
metrical and can be predominantly left or right
dominant.37

Non-fluent progressive aphasia

This is marked by atrophy in the left inferior frontal
regions of the pars opercularis and a small region of
the pars orbitalis38 (figure 11). There is usually
obvious grey matter loss in the left precentral gyrus,
Figure 7 Typical 99mTc-HMPAO SPECT changes with
Alzheimer’s disease. Source images (A) and statistical parametric
and more anteriorly in the middle frontal gyrus.
18
mapping (B) glass brain and (C) surface rendered F-FDG PET and 99mTc-HMPAO SPECT show asym-
representations of the SPECT data show bilateral symmetrical metric left frontal hypometabolism/perfusion, often
regions of temporoparietal hypoperfusion. involving the insular region39–41 (figure 12).

Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337 95

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

Figure 9 Structural imaging in behavioural variant frontotemporal lobar degeneration. Axial and coronal CT images demonstrating
marked frontal atrophy along with asymmetrical medial temporal atrophy, most marked on the left.

Semantic dementia
Table 3 Structural changes in Alzheimer’s disease (AD) versus This is marked by bilateral atrophy in the medial and
‘behavioural variant’ frontotemporal lobar degeneration, modified lateral portions of the anterior temporal lobes, includ-
from Frisconi36 ing amygdala/anterior hippocampus, anterior inferior,
and middle and superior temporal gyri and the anter-
Frontotemporal
AD lobar degeneration ior fusiform gyri. The atrophy is commonly asymmet-
rical and maximal on the left but with progressive
Frontal lobes +− +++
changes on the right also. Again, there is an anterior
Frontal horns of lateral +− ++++
ventricles
to posterior gradient34 38 (figure 13). 18F-FDG PET
and 99mTc-HMPAO SPECT in semantic dementia
Temporal horns of lateral + +++
ventricles shows prominent anterior temporal hypometabolism
Amygdala +++ ++++ or perfusion, left greater than right.40–42 (figure 14).
Entorhinal cortex ++ +
Hippocampus +++ + Logopenic progressive aphasia
This is atrophy involving the posterior temporal
cortex and inferior parietal lobule. There may also be

Figure 10 99mTc-HMPAO SPECT abnormalities in behavioural variant frontotemporal lobar degeneration. Source images (A), and
statistical parametric mapping (B) glass brain and (C) surface rendered representations showing a predominant pattern of bifrontal
hypoperfusion.

96 Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

Figure 11 Structural imaging changes in non-fluent progressive aphasia. Coronal and axial CT images demonstrating left-sided
peri-insular volume loss.

marked atrophy in the left anterior hippocampus, in VASCULAR DEMENTIA

the right angular gyrus and in the precuneus.38 This can result from large-vessel (cortical) or small-
18F-FDG PET and 99mTc-HMPAO SPECT show the vessel (subcortical) vascular disease. Cognitive decline
greatest metabolic and perfusion lesions in the left may develop as a direct result of the vascular injury
parietal lobe (inferior lobule) and posterolateral tem- (eg, multi-infarcts or a single stroke affecting a stra-
poral lobes (superior and middle temporal gyri)41–43 tegic location) and/or lowering the threshold for the
(figure 15). expression of concomitant pathology, such as AD. In
1993, the National Institute of Neurological
Disorders and Stroke and the Association
Internationale pour la Recherche et l’Enseignement en
Neurosciences reported diagnostic criteria for the
diagnosis of vascular dementia for research studies.44
This includes CT or MR criteria (table 4).

Lacunar infarcts and leukoaraiosis

Subcortical vascular dementia is the most common
form of vascular dementia, and refers to cognitive
impairment secondary to lacunar infarction and small-
vessel disease, or leukoaraiosis. Binswanger’s disease is
the eponym associated with extensive white matter
ischaemic change.
Lacunar infarctions result from arteriosclerosis, lipo-
Figure 12 99mTc-HMPAO SPECT abnormalities in non-fluent hyalinosis or embolism affecting the deep perforator
progressive aphasia. Note the left frontal and peri-insular arteries, such as the lenticulostriate, thalmoperforating
hypoperfusion. and long medullary arterioles. Lacunar infarctions,

Figure 13 Structural imaging in semantic dementia. Axial and coronal CT images show asymmetrical medial temporal atrophy,
maximal on the left, with associated anterolateral atrophy as well. A coronal FLAIR MRI in another (semantic dementia) patient (right)
shows similar structural changes.

Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337 97

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

Figure 14 99mTc-HMPAO SPECT abnormalities in semantic

dementia. Source images (A) and statistical parametric mapping
(B) glass brain and (C) surface rendered representations
demonstrating bilateral anterior temporal hypoperfusion,
maximal on the left.
Figure 15 99mTc-HMPAO SPECT abnormalities in logopenic
aphasia. Source images (A) and statistical parametric mapping
(B) glass brain and (C) surface rendered representations. Note,
hypoperfusion in the left inferior parietal lobule and
posterolateral temporal lobes.

Table 4 International Workshop of the National Institute of Neurological Disorders and Stroke (NINDS) and the Association
Internationale pour la Recherche et l’Enseignement en Neurosciences (AIREN) reported diagnostic criteria for the diagnosis of vascular
dementia for research studies
Topography
Large-vessel stroke Large-vessel stroke is an infarction defined as a parenchymal defect in an arterial territory involving the
cortical grey matter
Anterior cerebral artery Only bilateral anterior cerebral artery infarctions are sufficient to meet the NINDS-AIREN criteria
Posterior cerebral artery Infarcts in the posterior cerebral artery territory can be included only when they involve the following
regions.
Paramedian thalamic infarction: the infarct includes the cortical grey matter of the temporal/occipital lobe
and extends into the paramedian part (defined as extending to the third ventricle) of the thalamus; the
extension may be limited to the gliotic rim of the infarct that surrounds the parenchymal defect
Inferior medial temporal lobe lesions
Middle cerebral artery Middle cerebral artery infarction needs to involve the following regions:
Parietotemporal: the infarct involves both the parietal and temporal lobe (eg, angular gyrus)
Temporo-occipital: the infarct involves both the temporal and occipital lobe
Border zones A watershed infarction is defined as an infarction in the border zone between the middle and posterior
cerebral arteries or the middle and anterior cerebral arteries in the following regions:
Superior frontal region
Parietal region
Small-vessel disease Ischaemic pathology resulting from occlusion of small perforating arteries may manifest itself as lacunes
or white matter lesions. A lacune is a lesion with CSF-like intensity on all sequences on MRI (water
density on CT) surrounded by white matter or subcortical grey matter >2 mm. Care should be taken not
to include Virchow–Robin spaces, which typically occur at the vertex and around the anterior commissure
near the substantia perforata. Ischaemic white matter lesions circumscribed abnormalities with high signal
on T2-weighted images not following CSF signal (mildly hypodense compared with surrounding tissue on
CT) with a minimum diameter of 2 mm
Multiple basal ganglia and frontal white matter lacunes—criteria are met when there are at least 2
lacunes in the basal ganglia region (including thalamus and internal capsule) and at least 2 lacunes in
the frontal white matter
Extensive periventricular white matter lesions—lesions in the white matter abutting the ventricles and
extending irregularly into the deep white matter, or deep/subcortical white matter lesions. Smooth caps
and bands by themselves are not sufficient. Gliotic areas surrounding large-vessel strokes should not be
included here
Bilateral thalamic lesions—to meet the criteria, at least 1 lesion in each thalamus should be present
Continued

98 Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

Table 4 Continued
Topography
Severity
Large-vessel disease Large-vessel disease of the dominant hemisphere—if there is a large-vessel infarction as defined above,
to meet the criteria it has to be in the dominant hemisphere. In the absence of clinical information, the
left hemisphere is considered dominant
Bilateral large-vessel hemispheric strokes—1 of the infarctions should involve an area listed under
topography but is in the non-dominant hemisphere, while the infarction in the dominant hemisphere
does not meet the topography criteria
Small-vessel disease Leukoencephalopathy involving at least ¼ of the total white matter—extensive white matter lesions are
considered to involve ¼ of the total white matter when they are confluent (grade 3 in the Fazekas scale)
in at least 2 regions of the Fazekas scale and beginning confluent (grade 2 in the Fazekas scale) in 2
other regions. A lesion is considered confluent when >20 mm or consists of ≥2 smaller lesions fused by
more than a connecting bridge
Fulfilment of radiological criteria for Large-vessel disease—both the topography and severity criteria should be met (a lesion must be scored
probable vascular dementia in at least 1 subsection of both topography and severity)
Small-vessel disease—for white matter lesions, both the topography and severity criteria should be met
(a lesion must be scored in at least 1 subsection of both topography and severity); for multiple lacunes
and bilateral thalamic lesions, only the topography criterion is sufficient
Strategic infarction locations include the left angular gyrus, inferomesial temporal and mesial frontal locations, thalami, left capsular genu, and caudate
nuclei.

therefore, occur in the basal ganglia, putamen,

internal capsule, thalamus, corona radiata and
centrum semiovale and lateral brainstem.45 Silent
lacunes occur in 10–25% of patients aged
>65 years.46 Patients with lacunar infarctions develop
dementia 4–12 times more frequently than the
normal population.47
Lacunar infarctions are (radiologically) 3–15 mm in
size. They are cavitating, round, oval or slit-like, and
on CT are hypodense. On MRI, they often follow
cerebrospinal fluid signal intensity, being hypointense
on T1, and hyperintense on T2.46 48 FLAIR sequences
may show hypointensity with a high signal rim.45
Small-vessel disease, or leukoaraiosis, is another
player in subcortical vascular dementia. White matter
lesions on MRI correspond histologically to variable
combinations of myelin and axonal loss and scattered
microinfarctions, astrogliosis, and dilatation of periven-
tricular spaces.49 These lesions are exceptionally
common.46 White matter disease on CT shows as hypo-
dense areas. On MR T2 and FLAIR imaging, it appears
as hyperintense areas in the periventricular and deep
white matter45 (figure 16). MRI is more sensitive than
CT for detecting leukoaraiosis, although modern multi-
slice CT may be as good.13 Cognitive decline relates to
the severity of white matter hyperintensity50 51 and is
Figure 16 White matter hyperintensities on T2-weighted MRI. more pronounced in patients with these changes in a
Fazekas 0: none or a single punctate white matter periventricular location. There are rating scales to grade
hyperintensity lesions; Fazekas 1: multiple punctate lesions; the severity of white matter disease in population-based
Fazekas 2: beginning confluency of lesions (bridging); Fazekas studies, the most simple and commonly used being the
3: large confluent lesions. The Fazekas scale provides an overall
Fazekas scale52 (figure 16).
impression of the presence of white matter hyperintensities in
the entire brain. It can be scored on transverse FLAIR or
T2-weighted images. As well as dementia, severe white matter
hyperintensities (Fazekas 3) are associated with gait Microbleeds
abnormalities, falls and depression, and recently, they have These appear as focal hypointensities on
been shown to contribute to disability in the elderly population. T2*-weighted or susceptibility-weighted images.

Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337 99

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

Figure 17 Cortical atrophy scale devised by Pasquier and colleagues. Scores 0–3 represent absent, mild, moderate and severe
cortical atrophy respectively. Mild changes are said to be present when there is sulcal opening peripherally. Moderate changes are
seen when there is widening along the length of the sulcus. Severe atrophy is present when there is gyral thinning.

Although the number of microbleeds is associated DEMENTIA WITH LEWY BODIES

with age (occurring in 10% of the healthy elderly This accounts for 10% of dementias.1 Structural
population), 65% of patients with vascular dementia imaging shows mild generalised atrophy; there is
in one series had microbleeds.29 In sporadic small- often hippocampal atrophy but it is milder than in
vessel disease, microbleeds are located centrally.53 The AD.15 16 An absence of MTA strongly suggests demen-
number of cerebral microbleeds is more closely asso- tia with Lewy bodies rather than AD or vascular
ciated with lacunar infarction and leukoaraiosis than it dementia.15 Patients with dementia with Lewy bodies
is with age.53 54 Multiple microbleeds are associated develop severe nigrostriatal dopaminergic degener-
with poorer executive function and processing speed, ation: this does not occur in AD or most other
and suggest an increased risk of vascular dementia.55 31 dementia subtypes. Therefore, DAT uptake in the
Mortality is increased in patients with both micro- basal ganglia is a potential biomarker (figure 18).
bleeds and global cortical atrophy.56 McKeith et al,62 in a phase III trial, the largest study
of dementia with Lewy bodies to date, showed that
Volume loss abnormal scans had a mean sensitivity of 78% for
Patients with vascular dementia also have global detecting clinically probable dementia with Lewy
volume loss,57–60 perhaps due to subclinical diffuse bodies, and a specificity of 90% for excluding it, the
ischaemia. The global cortical atrophy scale of alternative diagnosis predominantly being AD. The
Pasquier et al (figure 17)61 relies on assessment of overall diagnostic accuracy was 86%, positive predict-
sulcal opening and gyral narrowing. It was designed ive value was 82% and negative predictive value
for MRI, but there is good agreement between MR was 88%.
and multidetector CT.13 It has good intrarater vari- Table 5 summarises the main imaging features
ability but poor inter-reader variability. encountered with the four most common dementias.

Figure 18 Dopamine transporter scan demonstrating normal configuration of basal ganglia uptake (left) and that seen in a patient
with dementia with Lewy bodies (right).

100 Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

Table 5 A summary of the principal findings encountered in the four most common dementias
Behavioural variant
frontotemporal lobar Primary progressive Dementia with Lewy
Condition Alzheimer’s disease degeneration aphasia Vascular dementia bodies
Structural Medial temporal lobe Frontal and anterior Semantic dementia: Left Cortical and/or Lacunar Global volume loss
imaging atrophy temporal atrophy anterior, inferior and lateral infarctions
Parietal atrophy Medial temporal lobe temporal atrophy Deep and periventricular
Ventriculomegaly atrophy Non-fluent progressive white matter CT
Global volume loss Often asymmetry aphasia: Perisylvian atrophy hypodensity/T2
Logopenic progressive hyperintensity
aphasia: Perisylvian atrophy Global volume loss
Mild medial temporal
lobe atrophy
Functional Temporoparietal Frontal and temporal Semantic dementia: Left Non-specific patterns of Occipital hypoperfusion
imaging hypoperfusion or hypoperfusion or anterior temporal hypoperfusion Reduction in basal
hypometabolism hypometabolism hypoperfusion ganglia uptake with
Non-fluent progressive dopamine transporter
aphasia: left forntal, scan
periinsular hypoperfusion or
hypometabolism
Logopenic progressive
aphasia: Left temporoparietal
hypoperfusion

Contributors All coauthors contributed to the medial temporal lobe atrophy by computed tomography.
compilation and editing of the paper. Lancet 1992;340:1179–83.
11 Frisoni GB, Beltramello A, Weiss C, et al. Linear measures of
Competing interests None.
atrophy in mild Alzheimer disease. Am J Neuroradiol
Provenance and peer review Commissioned; externally 1996;17:913–23.
peer reviewed. This paper was reviewed by Ian 12 Scheltens P, Leys D, Barkhof F, et al. Atrophy of medial
Turnbull, Bangor, UK. temporal lobes on MRI in ‘probable’ Alzheimer’s
disease and normal ageing: diagnostic value and
REFERENCES neuropsychological correlates. J Neurol Neurosurg Psychiatry
1 Dementia UK. Alzheimer’s Society: http://alzheimers.org.uk 1992;55:967–72.
(accessed March 2012). 13 Wattjes MP, Henneman WJ, van der Flier WM, et al.
2 McDonald CR, McEvoy LK, Gharapetian L, et al. Regional Diagnostic imaging of patients in a memory clinic: comparison
rates of neocortical atrophy from normal aging to early of MR imaging and 64-detector row CT. Radiology
Alzheimer disease. Neurology 2009;73:457–65. 2009;253:174–83.
3 Whitwell JL, Przybelski SA, Weigand SD, et al. 3D maps from 14 van de Pol LA, Hensel A, van der Flier WM, et al.
multiple MRI illustrate changing atrophy patterns as subjects Hippocampal atrophy on MRI in frontotemporal lobar
progress from mild cognitive impairment to Alzheimer’s degeneration and Alzheimer’s disease. J Neurol Neurosurg
disease. Brain 2007;130:1777–86. Psychiatry 2006;77:439–42.
4 Fotenos AF, Snyder AZ, Girton LE, et al. Normative estimates 15 Barber R, Gholkar A, Scheltens P, et al. Medial temporal lobe
of cross-sectional and longitudinal brain volume decline in atrophy on MRI in dementia with Lewy bodies. Neurology
aging and AD. Neurology 2005;64:1032–9. 1999;52:1153–8.
5 Jack CR, Weigand SD, Shiung MM, et al. Atrophy rates 16 Burton EJ, Barber R, Mukaetova-Ladinska EB, et al. Medial
accelerate in amnestic mild cognitive impairment. Neurology temporal lobe atrophy on MRI differentiates Alzheimer’s
2008;70:1740–52. disease from dementia with Lewy bodies and vascular cognitive
6 Busatto GF, Diniz BS, Zanetti MV. Voxel-based morphometry impairment: a prospective study with pathological verification
in Alzheimer’s disease. Expert Rev Neurother of diagnosis. Brain 2009;132:195–203.
2008;8:1691–702. 17 Frisoni GB, Testa C, Sabattoli F, et al. Structural correlates of early
7 Thompson PM, Hayashi KM, De Zubicaray GI, et al. Mapping and late onset Alzheimer’s disease: voxel based morphometric
hippocampal and ventricular change in Alzheimer disease. study. J Neurol Neurosurg Psychiatry 2005;76:112–14.
Neuroimage 2004;22:1754–66. 18 Galton CJ, Patterson K, Xuereb JH, et al. Atypical and typical
8 Chaim TM, Duran FL, Uchida RR, et al. Volumetric reduction presentations of Alzheimer’s disease: a clinical,
of the corpus callosum in Alzheimer’s disease in vivo as neuropsychological, neuroimaging and pathological study of 13
assessed with voxel-based morphometry. Psychiatry Res cases. Brain 2000;123:484–98.
2007;154:59–68. 19 Tondelli M, Wilcock GK, Nichelli P, et al. Structural MRI
9 Frisoni GB, Fox NC, Jack CR, et al. The clinical use of changes detectable up to ten years before clinical Alzheimer’s
structural MRI in Alzheimer disease. Nat Rev Neurol disease. Neurobiol Aging 2012;33:825. e25–36.
2010;6:67–77. 20 Koedam ELGE, Lehmann M, van der Flier WM, et al. Visual
10 Jobst KA, Smith AD, Szatmari M, et al. Detection in life of assessment of posterior atrophy development of a MRI rating
confirmed Alzheimer’s disease using a simple measurement of scale. Eur Radiol 2011;21:2618–25.

Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337 101

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

21 Lehmann M, Rohrer JD, Clarkson MJ, et al. Reduced cortical 39 Nestor PJ, Graham NL, Fryer TD, et al. Progressive non-fluent
thickness in the posterior cingulate gyrus is characteristic of aphasia is associated with hypometabolism centred on the left
both typical and atypical Alzheimer’s disease. J Alzheimers Dis insula. Brain 2003;126:2406–18.
2010;20:587–98. 40 Borroni B, Grassi M, Agosti C, et al. Latent profile analysis in
22 Lojkowska W, Ryglewicz D, Jedrzejczak T, et al. SPECT as a frontotemporal lobar degeneration and related disorders:
diagnostic test in the investigation of dementia. J Neurol Sci clinical presentation and SPECTfunctional correlates. BMC
2002;203:215–19. Neurology 2007;7:9–21.
23 Brown DR, Hunter R, Wyper DJ, et al. Longitudinal changes 41 Rabinovici GD, Jagust WJ, Furst AJ, et al. Aβ amyloid and
in cognitive function and regional cerebral function in glucose metabolism in three variants of primary progressive
Alzheimer’s disease: a SPECT blood flow study. J Psychiatr Res aphasia. Ann Neurol. 2008;64:388–401.
1996;30:109–26. 42 Diehl-Schmid J, Grimmer T, Drzezga A, et al. Longitudinal
24 Silverman DH. Brain 18F-FDG PET in the diagnosis of changes of cerebral glucose metabolism in semantic dementia.
neurodegenerative dementias: comparison with perfusion Dement Geriatr Cogn Disord 2006;22:346–51.
SPECT and with clinical evaluations lacking nuclear imaging. 43 Gorno-Tempini ML, Brambati SM, Ginex V, et al. The
J Nucl Med 2004;45:594–607. Logopenic/phonological variant of primary progressive aphasia.
25 Jobst KA, Barnetson LP, Shepstone BJ. Accurate prediction of Neurology 2008;71:1227–34.
histologically confirmed Alzheimer’s disease and the 44 Roman GC, Tatemichi TK, Erkinjuntti T, et al. Vascular dementia:
differential diagnosis of dementia: the use of diagnostic criteria for research studies. Report of the NINDS-
NINCDS-ADRDA and DSM-III-R criteria, SPECT, X-ray CT, AIREN International Workshop. Neurology 1993;43:250–60.
and Apo E4 in medial temporal lobe dementias. Oxford 45 Guermazi A, Miaux Y, Rovira-Cañellas A, et al.
Project to Investigate Memory and Aging. Int Psychogeriatr Neuroradiological findings in vascular dementia. Neuroradiol
1998;10:271–302. 2007;49:1–22.
26 Dougall NJ, Bruggink S, Ebmeier KP. Systematic review of the 46 Román GC, Erkinjuntti T, Wallin A, et al. Subcortical
diagnostic accuracy of 99 mTc-HMPAO-SPECT in dementia. ischaemic vascular dementia. Lancet Neurol 2002;1:426–36.
Am J Geriatr Psychiatry 2004;12:554–70. 47 Loeb C, Gandolfo C, Croce R, et al. Dementia associated with
27 Capizzano AA, Ación L, Bekinschtein T, et al. White matter lacunar infarction. Stroke 1992;23:1225–9.
hyperintensities are significantly associatedwith cortical atrophy 48 Brown JJ, Hesselink JR, Rothrock JF. MR and CT of lacunar
in Alzheimer’s disease. J Neurol Neurosurg Psychiatry infarcts. AJR Am J Roentgenol 1988;151:367–72.
2004;75:822–7. 49 Udaka F, Sawada H, Kameyama M. White matter lesions and
28 Holland CM, Smith EE, Csapo I, et al. Spatial distribution of dementia: MRI-pathological correlation. Ann N Y Acad Sci
white-matter hyperintensities in Alzheimer disease, cerebral 2002;977:411–15.
amyloid angiopathy, and healthy. Aging Stroke 50 Prins ND, Van Dijk EJ, den Heijer T, et al. Cerebral white
2008;39:1127–33. matter lesions and the risk of dementia. Arch Neurol
29 Cordonnier C, van der Flier WM, Sluimer JD, et al. 2004;61:1531–4.
Prevalence and severity of microbleeds in a memory clinic 51 Dufouil C, Godin O, Chalmers J, et al. PROGRESS MRI
setting. Neurology 2006;66:1356–60. Substudy Investigators. Severe cerebral white matter
30 Cordonnier C, van der Flier WM. Brain microbleeds and hyperintensities predict severe cognitive decline in patients
Alzheimer’s disease: innocent observation or key player? Brain with cerebrovascular disease history. Stroke 2009;40:2219–21.
2011;134:335–44. 52 Fazekas F, Chawluk JB, Alavi A, et al. MR signal abnormalities
31 Poels MM, Ikram MA, van der Lugt A, et al. Cerebral at 1.5 T in Alzheimer’s dementia and normal aging. Am J
microbleeds are associated with worse cognitive function: The Roentgenol 1987;149:351–6.
Rotterdam Scan Study. Neurology 2012;78:326–33. 53 Yamada S, Saiki M, Satow T, et al. Periventricular and deep
32 Rabinovici GR, Miller BL. Frontotemporal lobar degeneration white matter leukoaraiosis have a closer association with
—epidemiology, pathophysiology, diagnosis and management. cerebral microbleeds than age. Eur J Neurol 2012;19:98–104.
CNS Drugs 2010;24:375–98. 54 Koennecke HC. Cerebral microbleeds on MRI. Prevalence,
33 Ratnavalli E, Brayne C, Dawson K, et al. The prevalence of associations, and potential clinical implications. Neurology
frontotemporal dementia. Neurology 2002;58:1615–21. 2006;66:165–71.
34 Rosen HJ, Gorno-Tempini ML, Goldman WP, et al. Patterns of 55 Qiu C, Cotch MF, Sigurdsson S, et al. Cerebral microbleeds,
brain atrophy in frontotemporal dementia and semantic retinopathy, and dementia: the AGES—Reykjavik Study.
dementia. Neurology 2002;58:198–208. Neurology 2010;75:2221–8.
35 Boccardi M, Laakso MP, Bresciani L, et al. The MRI pattern of 56 Henneman WJ, Sluimer JD, Cordonnier C, et al. MRI
frontal and temporal brain atrophy in fronto-temporal biomarkers of vascular damage and atrophy predicting mortality
dementia. Neurobiol Aging 2003;24:95–103. in a memory clinic population. Stroke 2009;40:492–8.
36 Frisconi GB. Structural imaging in the clinical diagnosis of 57 van de Pol L, Gertz HJ, Scheltens P, et al. Hippocampal
Alzheimer’s disease: problems and tools. J Neurol Neurosurg atrophy in subcortical vascular dementia. Neurodegener Dis
Psychiatry 2001;70:711–18. 2011;8:465–9.
37 McMurtray AM, Chen AK, Shapira JS, et al. Variations in 58 Logue MW, Posner H, Green RC, et al., MIRAGE Study
regional SPECT hypoperfusion and clinical features in Group. Magnetic resonance imaging-measured atrophy and its
frontotemporal dementia. Neurology. 2006;66:517–22. relationship to cognitive functioning in vascular dementia and
38 Gorno-Tempini ML, Dronkers NF, Rankin KP, et al. Cognition Alzheimer’s disease patients. Alzheimers Dement
and anatomy in three variants of primary progressive aphasia. 2011;7:493–500.
Ann Neurol 2004;55:335–46.

102 Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com
REVIEWS

59 Seo SW, Lee JM, Im K, et al. Cortical thinning related to 61 Scheltens P, Pasquier F, Weerts JG, et al. Qualitative assessment of
periventricular and deep white matter hyperintensities. cerebral atrophy on MRI: inter- and intra-observer reproducibility
Neurobiol Aging 2012;33:1156–67. in dementia and normal aging. Eur Neurol 1997;37:95–9.
60 Wen W, Sachdev PS, Chen X, et al. Gray matter reduction is 62 McKeith I, O’Brien J, Walker Z, et al., DLB Study Group.
correlated with white matter hyperintensity volume: a Sensitivity and specificity of dopamine transporter imaging
voxel-based morphometric study in a large epidemiological with 123I-FP-CIT SPECT in dementia with Lewy bodies: a
sample. Neuroimage 2006;29:1031–9. phase III, multicentre study. Lancet Neurol 2007;6:305–13.

Mortimer AM, et al. Pract Neurol 2013;13:92–103. doi:10.1136/practneurol-2012-000337 103

 Downloaded from http://pn.bmj.com/ on October 2, 2017 - Published by group.bmj.com

Neuroimaging in dementia: a practical guide

Alex M Mortimer, Marcus Likeman and Timothy T Lewis

Pract Neurol 2013 13: 92-103

doi: 10.1136/practneurol-2012-000337

Updated information and services can be found at:

http://pn.bmj.com/content/13/2/92

These include:

Supplementary Supplementary material can be found at:

Material http://pn.bmj.com/content/suppl/2013/03/05/practneurol-2012-000337
.DC1
References This article cites 61 articles, 10 of which you can access for free at:
http://pn.bmj.com/content/13/2/92#BIBL

Email alerting Receive free email alerts when new articles cite this article. Sign up in the
service box at the top right corner of the online article.

Notes

To request permissions go to:

http://group.bmj.com/group/rights-licensing/permissions

To order reprints go to:

http://journals.bmj.com/cgi/reprintform

To subscribe to BMJ go to:

http://group.bmj.com/subscribe/