Journal of the Neurological Sciences 203 – 204 (2002) 85 – 89

www.elsevier.com/locate/jns

Single stroke dementia: Insights from 12 cases in Singapore

Alexander P. Auchus *, Christopher P.L.H. Chen, Swati N. Sodagar,
Melissa Thong, Eugene C.S. Sng
Department of Neurology, Singapore General Hospital, Outram Road, Singapore 169608, Singapore

Abstract

Background: Vascular dementia (VaD) is occasionally caused by a single, strategically located stroke. In this report, we describe the
clinical and anatomical features of 12 cases of strategic single infarct dementia (SSID) from Singapore. Methods: Each patient completed a
standardized diagnostic evaluation including history, neurological and neuropsychological examination, laboratory testing, and brain
imaging. Dementia was diagnosed using the Diagnostic and Statistical Manual, 3rd edition, revised (DSM-III-R) criteria, and VaD was
diagnosed using the National Institute of Neurologic Disorders and Stroke and the Association Internationale pour la Recherche et
l’Enseignement en Neurosciences (NINDS – AIREN) criteria. VaD patients whose brain imaging study revealed a single cerebrovascular
event were diagnosed with SSID. Results: We identified 12 cases of SSID among 125 VaD patients (9.6%). Stroke mechanism was lacunar
infarction in five cases, embolism in four cases, large vessel thrombosis in two cases, and parenchymal hemorrhage in one case. The most
commonly impaired cognitive domains on neuropsychological testing were visual memory, visuoconstruction, and language. In 11 of the 12
SSID cases, the stroke was located in the left hemisphere. The thalamus, either alone or as the proximal portion of a posterior cerebral artery
infarction, was involved in 8 of the 12 cases. Stroke locations in the nonthalamic SSID cases included left angular gyrus, subcortical left
frontal lobe including minor forceps, left basal forebrain and medial frontal lobe plus anterior corpus callosum (proximal anterior cerebral
artery infarction), and anterior corpus callosum alone. Conclusions: Various stroke mechanisms may produce SSID. In our SSID cases,
vascular damage almost always involved the left hemisphere and frequently involved the thalamus and major interhemispheric or
intrahemispheric white matter pathways.
D 2002 Elsevier Science B.V. All rights reserved.

Keywords: Dementia; Single stroke; Left angular gyrus

1. Introduction 2. Methods

Vascular dementia (VaD) is an important clinical syn-
drome throughout the world, particularly in Asia [1 – 3].
Diverse pathological subtypes of VaD including multiple
large vessel infarctions, multiple lacunar strokes, the diffuse
leukoencephalopathy of Binswanger’s disease, and others
have been described [4]. Occasionally, a single stroke may
produce VaD. This condition is called strategic single infarct
dementia (SSID) and has not been widely studied. In this
paper, we report the demographic, clinical, neuropsycho-
logical, and neuroanatomical features of SSID in 12 patients
from Singapore. The salient clinicoanatomic features shared
amongst these cases support a cerebral disconnection
hypothesis explaining the resulting dementia.
*
Corresponding author. Tel.: +65-6326-5003; fax: +65-6220-3321.
E-mail address: gnrale@sgh.gov.sg (A.P. Auchus).
The Neurodegenerative Diseases Program at Singapore
General Hospital maintains a research database of all
patients who have completed a standardized diagnostic
evaluation for dementia. This evaluation includes history
of dementia presentation, neurological examination, neuro-
psychological testing, screening laboratory tests, and brain
imaging. The dementia syndrome is diagnosed using the
Diagnostic and Statistical Manual, 3rd edition, revised
(DSM-III-R) criteria [5]. Probable and possible VaD are
diagnosed using the National Institute of Neurologic Dis-
orders and Stroke and the Association Internationale pour la
Recherche et l’Enseignement en Neurosciences (NINDS –
AIREN) criteria [4]. We diagnosed SSID in patients with
VaD whose brain imaging studies revealed a single cere-
brovascular event.
Each patient underwent neuropsychological assessment
using a battery of cognitive tests. The assessment was

0022-510X/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 2 - 5 1 0 X ( 0 2 ) 0 0 2 7 2 - 1
86 A.P. Auchus et al. / Journal of the Neurological Sciences 203 – 204 (2002) 85–89

Table 1
Demographic and clinical features of 12 patients with single stroke dementia
Age (years) Gender Hemisphere Stroke location Stroke mechanism Neuropsychology
(impaired domains)
61 F Left thalamus lacune VsM, VC
80 M Left thalamus lacune VbM, VsM, VC
72 M Left thalamus hemorrhage A, L, VsM, VC
69 M Left thalamus + hypothalamus lacune L, VsM
85 F Left thalamus + subthalamus + IC(pl) lacune VsM, VC
59 M Left subcortical frontal lobe including minor forceps giant lacune L, VbM, VsM, VC
82 F Left angular gyrus MCA embolism A, L, VbM, VsM
54 M Left anterior corpus callosum ACA thrombosis VbM, VsM, VC
86 M Left anterior CC + basal forebrain + medial frontal lobe ACA thrombosis A, L, VsM
58 F Left thalamus + occipital lobe PCA embolism L, VsM, VC
54 F Left thalamus + hippo + splenium + IC(pl) PCA embolism L, VbM, VsM, VC
68 M Right thalamus + hippo + splenium + occipital lobe PCA embolism VbM, VsM, VC
IC(pl) = internal capsule, posterior limb; CC = corpus callosum; hippo = hippocampus; MCA = middle cerebral artery; ACA = anterior cerebral artery;
PCA = posterior cerebral artery; A = attention; L = language; VbM = verbal memory; VsM = visual memory; VC = visuoconstruction.

conducted at least 1 month (median interval = 2 months)
following the stroke, when the patient was in a steady state
condition. Our battery includes tests for attention, language
(naming and verbal fluency), verbal memory (recall and
recognition), visual memory (recall and recognition) and
visuoconstruction, together with the mini-mental state
examination (MMSE) [6] and screening for depression.
Normative data, including cut-points defining impairment,
have been determined locally for use in elderly Singapor-
eans. Patients with vision, hearing, or language impairments
severe enough to prevent neuropsychological assessment
are not included in the database.
Demographic information was abstracted from chart
review. The mechanism of stroke in each case was deter-
mined by evaluation of clinical onset and course, together
with neuroimaging and ultrasonographic data. Stroke loca-
tion was determined directly from brain imaging studies.
3. Results
One hundred twenty-five patients meeting the NINDS –
AIREN criteria for VaD were identified in the database. Of
these, 12 patients had SSID (9.6%). There were seven men
and five women with SSID. Their average age was 69 years
(range 54 –86 years). Seven patients had hypertension, five
had diabetes mellitus, two had prior myocardial infarction,
but none had atrial fibrillation.
In neuropsychological evaluation, the average MMSE
score was 17.5 with a range from 9 to 24. All SSID patients
demonstrated impairment in multiple cognitive domains.
The most commonly impaired cognitive domain was visual
memory (all 12 patients), followed by visuoconstruction (8
of 12) and language (7 of 12).
In all but one patient, the stroke was located in the left
cerebral hemisphere. The thalamus was involved in 8 of the
12 patients, and major white matter pathways (anterior
corpus callosum, splenium, and minor forceps) were
involved in five patients. The most common stroke mech-
anism, present in five cases, was lacunar infarction. Embo-
lism produced the stroke in four cases, large vessel
thrombosis occurred in two cases, and parenchymal hemor-
rhage was seen in one case. Demographic and clinical
features of the 12 cases are summarized in Table 1.
4. Discussion
Dementia following a single stroke is not common. In
our hospital, we identified only three to four cases a year
from approximately 1500 annual stroke admissions. For
dementia to complicate a single stroke, the stroke must
impair multiple cognitive functions and multiple brain
processes. More often, this requires multiple strokes.
The adjective ‘‘strategic’’ in SSID might suggest that
single stroke dementia occurs primarily from small lesions.
Indeed, some prior descriptions of SSID have focused on
individual lacunes affecting critical brain regions [7,8]. In
this report, we also recognize SSID following small lacunar
infarcts. However, we further describe cases where a single
cerebrovascular event damaged several regions of the brain
and produced dementia. Our sole case involving the right
cerebral hemisphere is a good example of this phenomenon.
In this case, a posterior cerebral artery embolism fragmented
to produce scattered infarction involving the splenium,
thalamus, occipital lobe, and hippocampus (Fig. 1). Thus,
our findings indicate that SSID follows a single stroke, but
not always a small stroke or a single lesion locus.
Several studies have addressed the question of risk
factors in predicting dementia after stroke [9,10]. Factors
predicting post-stroke dementia include stroke location
within the left hemisphere, evidence of underlying cerebral
atrophy, advanced patient age, and social factors such as low
education [11 –15]. We clearly observed a predominance of
left hemisphere strokes in this sample of SSID cases.
However, our 12 patients with SSID were not significantly
different from our non-SSID VaD patients (n = 113) in age
or educational level.
 A.P. Auchus et al. / Journal of the Neurological Sciences 203 – 204 (2002) 85–89 87

Fig. 1. Diffusion-weighted MR images showing areas of acute infarction (bright signal) in the splenium, thalamus, occipital lobe, and hippocampus.

Dementia can complicate a strategically located small
stroke, as what may occur with infarction of the left angular
gyrus [16] or the genu of the internal capsule [8]. In this
case series, we identified one case of dementia with infarc-
tion of the left angular gyrus, but no cases of dementia with
infarction of the capsular genu. However, we had multiple
cases of dementia following thalamic infarction.
Four arteries are known to supply blood to the thalamus.
We were able to identify the likely artery producing the
thalamic stroke in six of eight patients. The damaged tissue
in the thalamic hemorrhage case did not respect anatomical
arterial boundaries. However, of the seven other cases of
thalamic stroke, two cases involved infarction in the terri-
tory of the paramedian thalamic artery, two involved the
territory of the thalamogeniculate artery, one involved the
polar artery, one involved the posterior choroidal artery, and
one was impossible to determine. Neuropsychiatric signs
and symptoms are known to complicate infarctions in the
territories of each of these arteries, except for the thalamo-
geniculate artery [17,18]. In our two cases with thalamoge-
niculate infarction, the thalamus was involved as the
proximal portion of a larger posterior cerebral artery embo-
lism. In these cases, the resulting cognitive impairment
likely resulted from the combined injury to the thalamus
and to other critical brain regions, namely, the ipsilateral
hippocampus and splenium.
Thalamic structures supplied by the polar, paramedian, or
posterior choroidal arteries and known to be involved in
cognitive processes include the dorsomedial nucleus, the
intralaminar nuclei, the posterior nuclei, and the mamillo-
thalamic tract. These structures are reciprocally connected
with cortical brain regions important for many higher
88 A.P. Auchus et al. / Journal of the Neurological Sciences 203 – 204 (2002) 85–89
cognitive functions. The more anterior thalamic nuclei have
interconnections with prefrontal, orbitofrontal, and cingulate
cortices, the more medial and posterior thalamic nuclei
interconnect with insular, temporal, and parietal regions
[19]. Thus, damage to these thalamic structures may pro-
duce dementia via disconnection (or diaschisis) of multiple
brain regions important for normal cognitive function.
Similarly, in prior studies of patients with dementia follow-
ing unilateral capsular genu infarction, functional deactiva-
tion of ipsilateral cortical brain regions has been inferred
from reduced hemispheric perfusion on SPECT and xenon
rCBF studies [8].
The neuropsychological findings in our cases suggest
significant bilateral brain dysfunction even though the
strokes were confined to a single cerebral hemisphere.
Impaired visual memory was seen in all 12 cases. As our
assessment of visual memory includes drawing-based tasks,
we found that most of the patients’ poor performance on
visual memory tasks was partly related to concomitant
ideational apraxia, presumably due to functional disconnec-
tion of multiple brain regions involved in such complex
tasks. Similarly, impaired visuoconstruction function was
commonly observed and was likely due to either ideational
or ideomotor apraxia [20]. We do not ascribe our patients’
visuoconstruction failure to motor impairment since, at the
time of neuropsychological assessment, only two patients
had sufficient hemiparesis to impair drawing. Finally, lan-
guage impairment was seen in 7 of our 12 cases—all with
left-sided strokes. In four of the seven language-impaired
cases, the stroke involved the left thalamus and produced a
mild thalamic aphasia [21]. In the three nonthalamic cases
with language impairment, two demonstrated impaired
verbal fluency associated with left frontal lobe infarction,
Fig. 2. T2-weighted MR images showing acute infarction (bright signal) of the anterior corpus callosum.
and one manifested anomia associated with infarction of the
left angular gyrus.
An interaction between cerebrovascular events and
degenerative dementias has been recognized [22], and
sometimes a stroke may precipitate the presentation of
dementia in a patient who already has neurodegeneration
and subclinical Alzheimer’s disease. In such situations, the
resulting dementia could be mistakenly attributed solely to
the stroke. In our patients, it is impossible to exclude with
complete certainty the presence of underlying Alzheimer’s
disease. However, none of our patients had historical or
clinical evidence of cognitive decline prior to their stroke.
Similarly, longitudinal neuropsychological follow-up has
not suggested concomitant Alzheimer’s disease. Significant
hippocampal atrophy was not present on brain imaging,
even in those patients over 80 years of age.
Although we did not find evidence for concomitant
Alzheimer’s disease in our patients, we did encounter one
case of single stroke dementia in a man who historically
abused alcohol. This was the case of anterior corpus
callosum infarction following thrombosis of the left anterior
cerebral artery (Fig. 2). This patient was acutely confused
with repetitive speech and inappropriate behavior. Although
only 54 years old, diffuse cerebral atrophy was present on
MRI, and subsequent neuropsychological testing revealed
storage-type verbal and visual memory impairment. We
postulate the presence of a subclinical, alcohol-related
amnestic disorder in this man. His acute dementia likely
resulted from the combined effects of this memory impair-
ment and the abrupt disconnection of the two cerebral
hemispheres produced by the corpus callosum infarct.
Lacunar infarction was the most common stroke mech-
anism producing dementia in our case series. However, our
 A.P. Auchus et al. / Journal of the Neurological Sciences 203 – 204 (2002) 85–89
experience also calls attention to several other stroke mech-
anisms which can produce single stroke dementia. These
include embolism, large vessel thrombosis, and parenchy-
mal hemorrhage. In our institution, intracranial hemorrhages
are generally triaged to the neurosurgery department. We
thus recognize a selection bias against, and a probable
underreporting of hemorrhage as a cause of single stroke
dementia in this report.
Our study has additional limitations. The study is retro-
spective, and consequently, our patients do not have neuro-
psychological testing to document normal cognitive function
prior to their stroke. However, we do have informant-based
information in the form of the informant questionnaire on
cognitive decline in the elderly (IQCODE) [23,24] scores
obtained at the time of stroke in three cases. The Quality
Standards Subcommittee of the American Academy of
Neurology has recently endorsed the IQCODE as an accept-
able informant-based screening tool for dementia [25]. In
each of these cases, the IQCODE score suggested no
cognitive decline antedating the stroke. In the cases without
IQCODE scores, family members confirmed patients’ inde-
pendent cognitive function prior to stroke. Our study also
lacks neuropathology. Without results from autopsy, we are
unable to exclude concomitant Alzheimer’s disease or other
pathologies in our cases. Unfortunately, the strong cultural
taboo against autopsy held by many Asians makes it difficult
to overcome this limitation. One additional criticism of this
study concerns the long-term follow-up of these cases. Quite
often, cognitive impairment following stroke is transient. For
instance, patients suffering thalamic infarction may gradu-
ally progress from stupor or coma, through a protracted
period of apathy and abulia, and eventually make substantial
cognitive recovery [26]. We conducted neuropsychological
assessments in a minimum of 1 month post-stroke. In some
cases, this may not be sufficiently long enough after the
stroke to conclude that the dementia is permanent. However,
in four of our cases, we have longitudinal neuropsycholog-
ical follow-up data for up to 28 months post-stroke. Each of
these four patients continues to show impairment in multiple
cognitive domains with little change in MMSE score. We
plan to conduct annual neuropsychological reassessments on
the remaining SSID patients.
These 12 cases of vascular dementia following a single
stroke illustrate the diversity of stroke mechanisms capable
of producing SSID. Our cases confirm the importance of left
hemisphere injury in dementia following stroke, and support
thalamocortical and corticocortical disconnection as critical
features in the pathogenesis of SSID.
References
[1] Yoshitake T, Kiyohara Y, Kato I, et al. Incidence and risk factors of
vascular dementia and Alzheimer’s disease in a defined elderly Japa-
nese population: the Hisayama study. Neurology 1995;45:1161 – 8.
[2] Chiu HF, Lam LC, Chi I, et al. Prevalence of dementia in Chinese
elderly in Hong Kong. Neurology 1998;50:1002 – 9.
89
[3] Ikeda M, Hokoishi K, Maki N, et al. Increased prevalence of vascular
dementia in Japan: a community-based epidemiological study. Neu-
rology 2001;57:839 – 44.
[4] Roman GC, Tatemichi TK, Erkinjuntti T, et al. Vascular dementia:
diagnostic criteria for research studies. Report of the NINDS – AIREN
International Workshop. Neurology 1993;43:25 – 260.
[5] American Psychiatric Association. Diagnostic and statistical manual
of mental disorders. 3rd ed. Washington, DC: American Psychiatric
Association; 1994.
[6] Folstein M, Folstein S, McHugh PR. Mini-mental state: a practical
method for grading the cognitive state of patients for the clinician. J
Psychiatr Res 1975;12:189 – 98.
[7] Tatemichi TK. How acute brain failure becomes chronic: a view of
the mechanisms of dementia related to stroke. Neurology 1990;40:
1652 – 9.
[8] Tatemichi TK, Desmond DW, Prohovnik I, et al. Confusion and mem-
ory loss from capsular genu infarction: a thalamocortical disconnec-
tion syndrome? Neurology 1992;42:1966 – 79.
[9] Loeb C, Gandolfo C, Croce R, Conti M. Dementia associated with
lacunar infarction. Stroke 1992;9:1225 – 9.
[10] Tatemichi TK, Paik M, Bagiella E, et al. Risk of dementia after a
stroke in a hospitalized cohort: results of a longitudinal study. Neu-
rology 1994;44:1885 – 91.
[11] Liu CK, Miller BL, Cummings JL, et al. A quantitative MRI study of
vascular dementia. Neurology 1992;42:138 – 43.
[12] Loeb C. Dementia due to lacunar infarctions: a misnomer or a clinical
entity? Eur Neurol 1995;35:187 – 92.
[13] Fein G, Di Sclafani V, Tanabe J, et al. Hippocampal and cortical
atrophy predict dementia in subcortical ischemic vascular disease.
Neurology 2000;55:1626 – 35.
[14] Pohjasvaara T, Mantyla R, Salonen O, et al. How complex interactions
of ischemic brain infarcts, white matter lesions, and atrophy relate to
poststroke dementia. Arch Neurol 2000;57:1295 – 300.
[15] Pohjasvaara T, Mantyla R, Salonen O, et al. MRI correlates of demen-
tia after first clinical ischemic stroke. J Neurol Sci 2000;181:111 – 7.
[16] Benson DF, Cummings JL. Angular gyrus syndrome simulating Alz-
heimer’s disease. Arch Neurol 1982;39:616 – 20.
[17] Graff-Radford NR, Damasio H, Yamada T, Eslinger PJ, Damasio AR.
Nonhaemorrhagic thalamic infarction. Clinical, neuropsychological
and electrophysiological findings in four anatomical groups defined
by computed tomography. Brain 1985;108:485 – 516.
[18] Bogousslavsky J, Regli F, Uske A. Thalamic infarcts: clinical syn-
dromes, etiology, and prognosis. Neurology 1988;38:837 – 48.
[19] Jones EG. The thalamus. New York: Plenum; 1985.
[20] Heilman KM, Rothi LJ, Valenstein E. Two forms of ideomotor apra-
xia. Neurology 1982;32:342 – 6.
[21] Tuszynski MH, Petito CK. Ischemic thalamic aphasia with pathologic
confirmation. Neurology 1988;38:800 – 2.
[22] Snowdon DA, Greiner LH, Mortimer JA, Riley KP, Greiner PA, Mar-
kesbery WR. Brain infarction and the clinical expression of Alzhei-
mer’s disease. The nun study. JAMA 1997;277:813 – 7.
[23] Jorm AF, Scott R, Cullen JS, MacKinnon AJ. Performance of
informant questionnaire on cognitive decline in the elderly (IQ-
CODE) as a screening test for dementia. Psychol Med 1991;21:
785 – 90.
[24] Fuh JL, Teng EL, Lin KN, et al. The informant questionnaire on
cognitive decline in the elderly (IQCODE) as a screening tool for
dementia for a predominantly illiterate Chinese population. Neurology
1995;45:92 – 6.
[25] Petersen RC, Stevens JC, Ganguli M, Tangalos EG, Cummings JL, De
Kosky ST. Practice parameter: early detection of dementia: mild cog-
nitive impairment (an evidence-based review). Report of the Quality
Standards Subcommittee of the American Academy of Neurology.
Neurology 2001;56:1133 – 42.
[26] Krolak-Salmon P, Croisile B, Houzard C, Setiey A, Girard-Madoux P,
Vighetto A. Total recovery after bilateral paramedian thalamic infarct.
Eur Neurol 2000;44:216 – 8.