Original Research Article
Dement Geriatr Cogn Disord 2000;11:114–118
Medial Temporal Lobe Width on CT Scanning in
Alzheimer’s Disease: Comparison with Vascular
Dementia, Depression and Dementia with
Lewy Bodies
J.T. O’Brien a S. Metcalfe a A. Swann a J. Hobson a K. Jobst b C. Ballard a
I. McKeith a A. Gholkar a
a Institute for the Health of the Elderly, University of Newcastle-upon-Tyne, and b Glasgow Homeopathic Hospital
and the University Department of Medicine, Glasgow, UK
Key Words
Dementia W Diagnosis W CT
Abstract
A simple linear measurement of the minimum width of
the medial temporal lobe (MTL) on angled CT scans has
been suggested as an accurate ante-mortem marker for
Alzheimer’s disease (AD). To determine the clinical utility
and specificity of this finding, we performed angled CT
scans with 5-mm slices in 116 subjects referred to a geo-
graphically based Old Age Psychiatry service in Newcas-
tle. Diagnoses were of NINCDS/ADRDA AD (n = 69, 36
probable and 33 possible). NINDS/AIREN vascular de-
mentia (VaD, n = 25), consensus criteria for dementia
with Lewy bodies (DLB, n = 9) and DSM-IV criteria for
major depression (n = 13). Subjects were well matched
for age. Minimum MTL width was significantly greater in
depressed subjects (13.7 mm) compared to those with
dementia, though no differences were seen within the
dementia groups (AD 10.8, VaD 10.4, and DLB 10.9 mm).
An MTL width below 11.5 mm had a sensitivity of 54%
(56/103) and a specificity of 77% (10/13) for distinguish-
ing dementia from depression. We conclude that a single
© 2000 S. Karger AG, Basel
ABC
1420–8008/00/0112–0114$17.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/journals/dem
Accepted: May 15, 1999
cross-sectional measurement of MTL width on CT does
not help differentiate between different types of demen-
tia, though it may provide some supportive evidence
when distinguishing depression from dementia.
Copyright © 2000 S. Karger AG, Basel
Introduction
Accurate ante-mortem diagnosis of Alzheimer’s dis-
ease (AD), in particular its differentiation from other
causes of dementia, remains problematic. The recent
introduction of symptomatic treatments for AD [1] and
new therapeutic options for vascular dementia (VaD),
together with the need to recognise and avoid antipsy-
chotic medication in those with dementia with Lewy bod-
ies (DLB) [2], has emphasised the need for early and accu-
rate diagnosis of these conditions. It has previously been
reported that the minimal width of the medial temporal
lobe (MTL) opposite the brain stem on CT scans angled
along the temporal lobe axis may be an accurate diagnos-
tic marker for AD and other MTL dementias [3]. In a
series of 118 subjects with autopsy confirmation of diag-
nosis and 119 controls, a sensitivity of 85% and a specific-
J.T. O’Brien, MD
137.73.144.138 - 1/30/2018 11:00:45 PM
Wolfson Research Centre, Institute for the Health of the Elderly
Newcastle General Hospital, Westgate Road
Newcastle-upon-Tyne, NE4 6BE (UK)
Tel. +44 191 256 3323, Fax +44 191 219 5051, E-Mail j.t.o’brien@ncl.ac.uk
King's College London
Downloaded by:
Table 1. Subject characteristics and
differences in medial temporal lobe width AD VaD DLB DEP Significance
(n = 69)a (n = 25)b (n = 9)c (n = 13) (ANOVA)

Age, mean B SD 80.2B7.4 77.4B7.9 80.9B3.4 76.1B5.0 NS

Sex, M/F 26/43 15/10 3/6 5/8 NS
MMSE score, mean B SD 16.6B5.8 20.4B17.4 13.0B6.9 25.8B2.2 p = 0.007d
MTL width, mean B SD 10.8B3.0 10.4B3.0 10.9B4.0 13.7B3.4 p = 0.02e

DEP = Depression.
a 36 probable and 33 possible AD.
b 11 probable and 14 possible VaD.
c 8 probable and 1 possible DLB.
d F = 4.27, d.f. (3, 112); post-hoc tests show that the MMSE score in the depressed group is
significantly higher than in the other groups (p ! 0.05).
e F = 3.4, d.f. (3, 112); post-hoc tests reveal that MTL width is significantly higher in the
depressed subjects than in the other groups (p ! 0.05).

ity of 78%%% was achieved [3]. There were some non-AD
cases with confirmed diagnoses (24, including 8 with VaD
and 4 with DLB). However, to fully examine the specifici-
ty of this finding, further comparison with larger numbers
of non-AD cases is necessary, as well as comparison with
subjects with psychiatric disorders like depression, which
has itself been associated with atrophy of MTL structures
[4] and is a risk factor for AD [5]. We sought to determine
whether MTL width on CT scanning would distinguish
between AD and other disorders which may cause diag-
nostic confusion such as depression, VaD and DLB in a
clinically representative sample of patients presenting to
Old Age Psychiatry services.
Materials and Methods
Subjects
Consecutive referrals over an 18-month period to a geographical-
ly based Old Age Psychiatry service in Newcastle who underwent
temporal-lobe-angled CT scanning as part of routine diagnostic
investigations for dementia or depression were included. Other
assessments included full history and mental state examination, stan-
dardised physical and neurological examination, cognitive screening
using the Mini-Mental State Examination (MMSE) [6], chest X-ray
and blood screen (including B12, folate and syphilis serology). Clini-
cal diagnosis was made by two experienced independent psychia-
trists using all clinical information available but, to maintain blind-
ness, without actually viewing the CT or having access to any infor-
mation regarding the presence and extent of any atrophy of MTL
structures. Information regarding any vascular or other changes on
scan was made available to allow the standardised diagnostic criteria
to be rigorously applied. The following criteria were used: NINCDS/
ADRDA criteria for AD [7], NINDS/AIREN criteria for VaD [8],
consensus criteria for DLB [2] and DSM-IV criteria for major depres-
sion (APA) [9]. From 130 subjects initially considered for inclusion,
14 were excluded either because they did not meet entry diagnostic
criteria, underwent standard (not angled) CT or because there was
diagnostic doubt. This left 116 subjects who were included in the
study. Subject characteristics are shown in table 1.
CT Scanning
CT scans were obtained on an IGE CT 9800 head scanner. After
an initial sagittal scout, angled scans 5 mm through the temporal
lobes were acquired approximately 20–25 ° C caudal to the orbito-
meatal line as previously described [10]. This was achieved by
obtaining a lateral tomogram to visualise the bony landmarks of the
skull, and a plane passing from the top of the mastoids to the front of
the hard palate was established. Slices 5 mm parallel to this plane
were obtained which usually resulted in three slices passing through
the temporal lobes. The minimum width of the MTL was measured
from hard copies using callipers, through the section that correspond-
ed most closely to that passing through the mid-point of the temporal
lobes. The minimum width of the MTL on either side between the
anterior and posterior margins of the brain stem was chosen for anal-
ysis [10, 13]. All measurements were performed by a single rater
blind to diagnosis. The coefficient of variation for inter-rater reliabil-
ity for repeat MTL measurements was 5%, and inter-rater reliability
(30 scans measured by 3 different raters) was 11.6%. Statistical anal-
ysis was undertaken using SPSS for Windows (version 8.0) with ¯2
tests or ANOVA (followed by post-hoc Scheffé test to determine
group differences) as appropriate. Correlations were examined using
Pearson’s r.
Results
Results are shown in table 1. As can be seen, groups
were well matched for age and sex. As expected, MMSE
scores were significantly higher in the depressed group.
There were no significant differences in MTL width

MTL Width on CT in AD Dement Geriatr Cogn Disord 2000;11:114–118 115

137.73.144.138 - 1/30/2018 11:00:45 PM
King's College London
Downloaded by:
 a b

Fig. 1. Temporal-lobe-angled axial CT scans demonstrating the measurement of minimum MTL width (arrows) in
(a) a depressed subject (MTL width = 14.5 mm) and (b) a patient with Alzheimer’s disease (MTL width = 5 mm).

between the three dementia groups, though all were signif-
icantly smaller than in the depressed group (p ! 0.05 for
all groups). There were no significant differences between
those with probable and possible AD. Examples of CT
scans for AD and depressed subjects are shown in fig-
ure 1. Not surprisingly, given the similar mean MTL mea-
surements, no cut-off of MTL width provided good dis-
crimination between dementia groups. A previously sug-
gested cut-off of 11.5 mm for detecting AD [10, 11] had a
sensitivity of 51% (35/69) for detecting AD with specifici-
ties of 77% (10/13) for depressed subjects, but only 56%
(5/9) for DLB and 32% (8/25) for VaD. For distinguishing
between dementia (irrespective of aetiology) and depres-
sion, sensitivity was 54% (56/103) with a specificity of
77% (10/13). For all subjects, there was a significant
inverse correlation between age and MTL width (r =
–0.27, p ! 0.01). Within AD subjects only, MTL width
correlated with MMSE score (r = 0.35, p ! 0.01).
Discussion
The measurement of MTL width on temporal-lobe-
angled CT from hard copies is simple, quick and demon-
strates good intra- and inter-rater reliability. We found a
significant reduction in the minimum width of the MTL
of approximately 3 mm in subjects with dementia com-
pared with those with depression. Mean MTL width was
10.8 mm in AD subjects and showed a modest but highly
significant correlation with MMSE score and an inverse
correlation with age. This may explain why our MTL
width was slightly less than the 12.0 mm described by Fri-
soni et al. [12] in a younger (mean age 70) and slightly less
cognitively impaired group, but higher than the 7.3 mm
reported by Jobst et al. [3] in a more cognitively impaired
sample (mean MMSE 12). Our groups were well matched
for age, though these discrepant results clearly indicate
that the clinical interpretation of the significance of
reduced MTL width can only be made after considering
both the age of the patients and their dementia severity.
Although depressed subjects had a significantly greater

116 Dement Geriatr Cogn Disord 2000;11:114–118 O’Brien/Metcalfe/Swann/Hobson/Jobst/

Ballard/McKeith/Gholkar
137.73.144.138 - 1/30/2018 11:00:45 PM
King's College London
Downloaded by:
MTL width (mean 13.7 mm), some overlap between Although this study is based on clinical, rather than
groups still existed, and the diagnostic precision of this neuropathological, diagnosis, we carefully applied rigor-
measurement was modest (specificity 77%). However, as ous standardised sets of diagnostic criteria, all of which
has been shown for MRI [13], we suggest that this may have been shown to have a positive predictive value
still sometimes be clinically useful given the often difficult above 80% when judged by post-mortem diagnosis [19,
differentiation between depression and dementia in the 20]. As such, we do not feel that clinical misdiagnosis
elderly. Since depression is a risk factor for AD, only care- could explain our findings of no differences in MTL width
ful long-term follow-up of cases reported here who have between groups. Moreover, neuropathological confirma-
atrophy will determine whether they are at increased risk tion of diagnosis has been obtained on 7 subjects (all accu-
of developing AD. Overall, although mild volume reduc- rately diagnosed during life): 2 cases with AD (MTL width
tions in the hippocampus of 10–15% have been reported 9.5 and 12 mm), 3 cases of VaD (MTL 15, 7.5 and
in chronic depression [4], our results are consistent with 9.5 mm), and 2 cases of DLB (MTL width 9 and 12 mm).
the findings of Ebmeier et al. [14] who reported that mean Even from this small sample with neuropathological con-
MTL width on MRI was significantly greater in depressed firmation, it is clear that MTL width cannot reliably dis-
subjects than in those with AD. Although structural brain tinguish between different dementia groups, at least by
changes and enduring cognitive impairments are known the time they present to Old Age Psychiatry services. Evi-
to occur in elderly depressed subjects [14, 15]. MTL width dence from both CT and MRI studies suggests that longi-
on CT is still preserved in most cases. tudinal assessment of change may be more helpful than a
Importantly, we found no differences in MTL width single cross-sectional measure [21, 22]. Further study
between the three subtypes of dementia studied, with should assess clinically representative populations using
both VaD and DLB associated with a reduction in MTL thinner (2 mm) CT slices, investigate MTL width as a
width of a similar magnitude to AD. Our results, while in diagnostic marker earlier in the disease process, and com-
agreement with other reports of medial temporal lobe pare the accuracy and clinical utility of both cross-section-
atrophy in AD [3, 16], question the specificity of this find- al and longitudinal change on CT an MRI in terms of
ing regarding the other important subtypes of dementia. improving the differential diagnosis of dementia.
Using high resolution coronal MRI, we have recently
shown that DLB is associated with less hippocampal atro-
phy than AD and VaD, which themselves showed a con- Acknowledgments
siderable degree of overlap [17, 18]. In the current study,
We thank Bob Barber and staff from the Regional Neurosciences
however, we did not find differences between DLB and
Centre for assistance with collating necessary data and the Medical
other groups. Our CT measurements were based on 5-mm Research Council for financial support.
cuts, and it is possible that thinner slices may provide bet-
ter discrimination, though, consistent with our results,
Jobst et al. [3] found MTL atrophy in 2/4 DLB and 7/8
VaD subjects. Similarly, Pasquier et al. [16] reported atro-
phy in 8/16 DLB subjects and 8/8 VaD subjects. Taken
overall, the results of our study and others indicate that a
single measurement of MTL width on 5-mm CT sections,
while being helpful in adding support to a clinical diagno-
sis of dementia, will not assist in differentiating between
the different causes of dementia, at least in subjects at the
time of their presentation to Old Age Psychiatry services.
The use of this technique at an early stage remains to be
determined, since it has been suggested that MTL atrophy
in non-AD dementias may be a relatively late feature [3],
in contrast to AD where it occurs early. In clinical prac-
tice, other features on CT, such as the presence of definite
infarcts or extensive white matter change, will help distin-
guish AD from VaD.

MTL Width on CT in AD Dement Geriatr Cogn Disord 2000;11:114–118 117

137.73.144.138 - 1/30/2018 11:00:45 PM
King's College London
Downloaded by:
 References
1 Rogers SL, Farlow MR, Doody RS, Mohs R,
Friedhoff LT, and the Donepezil Study Group:
A 24-week, double-blind, placebo-controlled
trial of donepezil in patients with Alzheimer’s
disease. Neurology 1998;50:136–145.
2 McKeith IG, Galasko D, Koska K, et al: Con-
sensus guidelines for the clinical and pathologic
diagnosis of dementia with Lewy bodies
(DLB): Report of the consortium on DLB inter-
national workshop. Neurology 1996;47:1113–
1124.
3 Jobst KA, Lin PD, Shepstone BJ: Accurate pre-
diction of histologically confirmed Alzheimer’s
disease and the differential diagnosis of demen-
tia: The use of NINCDS-ADRDA and DSM-
III-R criteria, SPECT, X-ray CT, and Apo E4
in medial temporal lobe dementias. Int Psycho-
geriatr 1998;10:271–302.
4 Sheline YI, Wang PW, Gado MH, Csernansky
JG, Vannier MW: Hippocampal atrophy in
recurrent major depression. Proc Natl Acad Sci
USA 1996;93(9):3908–3913.
5 Jorm AF, van Dujin CM, Chandra V, Frati-
glioni L, Graves AB, Heyman A, Kokmen E,
Kondo K, Mortimer JA, Rocca WA, et al: Psy-
chiatric history and related exposures as risk
factors for Alzheimer’s disease: A collaborative
re-analysis of case-controlled studies. EURO-
DEM Risk Factors Research Group. Int J Epi-
demiol 1991;20(suppl 2):S43–47.
6 Folstein MF, Folstein SE, McHugh PR: ‘Mini-
Mental State’: A practical method for grading
the cognitive state of patients for the clinician.
J Psychiatr Res 1975;12:189–198.
7 McKhann G, Drachman D, Folstein M, Katz-
man R, Price D, Stadlan E: Clinical dignosis of
Alzheimer’s disease: Report of the NINCDS/
ADRDA work group under the auspices of the
Dept. of Health and Human Task Force in Alz-
heimer’s Disease. Neurology 1984;34:939–
944.
118 Dement Geriatr Cogn Disord 2000;11:114–118
8 Roman GC, Tatemichi T, Erkinjuntti T, et al:
Vascular dementia: Diagnostic criteria for re-
search studies. Report of the NINDS AIREN
International Workshop. Neurology 1993;43:
250–260.
9 American Psychiatric Association: Diagnostic
and Statistical Manual of Mental Disorders,
ed 4. Washington, American Psychiatric Asso-
ciation, 1994.
10 Jobst KA, Smith AD, Szatmari M, et al: Detec-
tion in life of confirmed Alzheimer’s disease
using a simple measurement of medial tempo-
ral lobe atrophy by computed tomography.
Lancet 1992;340:1179–1183.
11 Henon H, Pasquirer F, Durieu I, Pruvo JP,
Leys D: Medial temporal lobe atrophy in stroke
patients: Relation to pre-existing dementia. J
Neurol Neurosurg Psychiatry 1998;65:641–
647.
12 Frisoni GB, Bianchetti A, Geroldi C, Trabuc-
chi M, Beltramello A, Weiss C: Measures of
medial temporal lobe atrophy in Alzheimer’s
disease. J Neurol Neurosurg Psychiatry 1994;
57(11):1438–1439.
13 O’Brien JT, Desmond P, Ames D, Schweitzer
I, Chiu E, Tress B: Temporal lobe magnetic res-
onance imaging can differentiate Alzheimer’s
disease from normal ageing, depression, vascu-
lar dementia and other causes of cognitive im-
pairment. Psychol Med 1997;27:1267–1275.
14 Ebmeier K, Prentice N, Ryman A, Halloran E,
Rimmington JE, Best JK, Goodwin G: Tempo-
ral lobe abnormalities in dementia and depres-
sion: A study using high resolution single pho-
ton emission tomography and magnetic reso-
nance imaging. J Neurol Neurosurg Psychiatry
1997;63:597–604.
15 O’Brien J, Desmond P, Ames D, Schweitzer I,
Harrigan S, Tress B: A magnetic resonance
imaging study of white matter lesions in de-
pression and Alzheimer’s disease. Br J Psychia-
try 1996;168:477–485.
16 Pasquier F, Hamon M, Lebert F, Jacob B, Pru-
vo JP, Petit H: Medial temporal lobe atrophy in
memory disorders. J Neurol 1997;244:175–
181.
17 Harvey GT, O’Brien JT, Hughes J, et al: Mag-
netic resonance imaging differences between
dementia with Lewy bodies and Alzheimer’s
disease. Psychol Med 1999;29:181–187.
18 Barber R, Gholkar A, Ballard C, et al: Tempo-
ral lobe atrophy on MRI in dementia with
Lewy bodies: A comparison with Alzheimer’s
disease, vascular dementia and normal ageing.
Neurology 1999;52:1153–1158.
19 Burns A, Luthert P, Levy R, Jacoby R, Lantos
P: Accuracy of clinical diagnosis of Alzheimer’s
disease. Br Med J 1990;301:1026.
20 McKeith IG, Ballard C, Perry R, Ince P, Jarosh
E, Neill D, O’Brien J: Predictive accuracy of
clinical diagnostic criteria for dementia with
Lewy bodies: A prospective neuropathological
validation study. Neurology 1998;50(4):181.
21 Jobst KA, Smith AD, Szatmari M, Esiri M, Jas-
kowski A, Hindley N, McDonald B, Molyneux
AJ: Rapidly progressive atrophy of medial tem-
poral lobe Alzheimer’s. Lancet 1994;343:829–
830.
22 Fox MC, Freeborough PA, Rossor MN: Visual-
isation and quantification of rates of atrophy in
Alzheimer’s disease. Lancet 1996;348:94–97.
O’Brien/Metcalfe/Swann/Hobson/Jobst/
Ballard/McKeith/Gholkar
137.73.144.138 - 1/30/2018 11:00:45 PM
King's College London
Downloaded by: