Anterior cerebral artery infarction

Stroke mechanism and clinical-imaging study in 100 patients

Suk Y. Kang, MD ABSTRACT

Jong S. Kim, MD, PhD Background: Stroke mechanisms and clinical features of anterior cerebral artery (ACA) territory
infarction have rarely been investigated using MRI.
Objectives: To verify stroke mechanisms and to make clinical imaging correlation.
Address correspondence and
Methods: Clinical, MRI, and angiographic findings of 100 consecutive patients with ACA infarc-
reprint requests to Dr. Jong S. tion were studied.
Kim, Stroke Center and
Department of Neurology, Results: Motor dysfunction (n ⫽ 91) was the most common symptom, and severe motor dysfunc-
Asan Medical Center, Song-Pa tion was related to supplementary motor area/paracentral lobule involvement (p ⫽ 0.016). Hypob-
PO Box 145, Seoul 138-600,
Korea ulia/apathy (n ⫽ 43) was related to involvement of fontal pole (p ⫽ 0.002), corpus callosum/
[email protected] cingulate gyrus (p ⫽ 0.003), and superior frontal gyrus (p ⬍ 0.001), and occurred more frequently
in patients with bilateral lesions followed by left lesions. Urinary incontinence (n ⫽ 30) was not
related to any specific lesion locations. Grasp reflex (n ⫽ 25) was related to corpus callosum
involvement (p ⫽ 0.035). Angiographic (mostly MR angiography) results showed that 68 patients
had local ACA atherosclerosis, most often at A2 segment. The stroke mechanisms included car-
diogenic embolism in 10, internal carotid artery–ACA embolism in 6, and ACA atherosclerosis in
61 patients. In the latter group, detailed stroke mechanisms included local branch occlusion (n ⫽
20), in situ thrombotic occlusion (n ⫽ 20), artery-to-artery embolism (n ⫽ 12), and a combination
(n ⫽ 9). Patients with intrinsic ACA disease more often had hypobulia (p ⫽ 0.077) and corpus
callosal involvement (p ⫽ 0.016) than those with embolism either from the internal carotid artery
or the heart.
Conclusion: Anterior cerebral artery (ACA) atherosclerosis is the most important stroke etiology
in our population, causing infarction with various mechanisms. Topographic lesion patterns and
consequent clinical features of ACA infarction are determined by diverse pathogenic mechanisms
and the status of collateral circulation. Neurology® 2008;70:2386–2393

GLOSSARY
A-com ⫽ anterior communicating artery; AAE ⫽ artery-to-artery embolism; ACA ⫽ anterior cerebral artery; AIF ⫽ anterior
internal frontal artery; CM ⫽ callosomarginal artery; DWI ⫽ diffusion-weighted imaging; FP ⫽ frontal pole; FrP ⫽ frontopolar
artery; GR ⫽ gyrus rectus; ICA ⫽ internal carotid artery; ITO ⫽ in situ thrombotic occlusion; LBO ⫽ local branch occlusion;
MIF ⫽ middle internal frontal artery; MRA ⫽ MR angiogram; Orf ⫽ orbitofrontal artery; PC ⫽ precuneus; PCA ⫽ paracentral
artery; PIF ⫽ posterior internal frontal artery; PL ⫽ paracentral lobule; SFG ⫽ superior frontal gyrus; SMA ⫽ supplementary
motor area; SP ⫽ superior parietal artery; TE ⫽ echo time; TOAST ⫽ Trial of Org 10172 in Acute Stroke Treatment; TR ⫽
repetition time.

Anterior cerebral artery (ACA) territory infarction was first described in the early 20th
century based on autopsy results.1,2 Since then, several CT3-6 or MRI7,8 based studies
investigated etiologies and clinical features. However, the number of patients was gener-
ally too small to make a clinical–MRI correlation, and stroke mechanisms have not been
elucidated in detail, especially in those with intrinsic ACA atherosclerosis.6 The purpose
of the present study was to investigate the etiologic, clinical, and imaging findings in
ACA infarction using MRI and angiographic results. In particular, we wished to eluci-
date detailed stroke mechanisms in patients with intracranial ACA atherosclerosis.

From the Stroke Center and Department of Neurology, University of Ulsan, Asan Medical Center, Seoul, Korea.
Supported by a grant from Brain Research Center of the 21st Century Frontier Research Program funded by the Ministry of Science and
Technology of Korea (M103KV010010 06K2201 01010).
Disclosure: The authors report no disclosures.

2386 Copyright © 2008 by AAN Enterprises, Inc.

METHODS Between January 1996 and May 2007, we had
examined 110 consecutive patients with ACA territory in- Figure 1 Schematic illustration of anterior
cerebral artery territory and artery
farction (1.1% of total patients with ischemic stroke) who
branches
were admitted to the Asan Medical Center within 1 week
after stroke onset. Excluded were patients with 1) concomi-
tant infarcts in other vascular territories, 2) vasospasm sec-
ondary to subarachnoid hemorrhage, and 3) strictly
subcortical infarction, partly because the number of patients
was small (n ⫽ 2) and partly because of the uncertainty of
the vascular territory due to individual variability. In this
clinical-imaging correlation study, 10 patients were further
excluded because 1) three underwent CT scan only, 2) four
had significant residual neurologic deficits due to previous
strokes, and 3) three did not undergo vascular assessment.
Thus, 100 patients who underwent both MRI and angio-
graphic studies became the subject of our study.
All patients underwent blood tests, urine analysis, chest
Orf ⫽ orbitofrontal artery; FrP ⫽ frontopolar artery; AIF ⫽
X-ray, and 12-lead electrocardiogram. Transthoracic and
anterior internal frontal artery; MIF ⫽ middle internal frontal
transesophageal echocardiography and 24-hour electrocar-
artery; PIF ⫽ posterior internal frontal artery; CM ⫽ calloso-
diogram monitoring were performed in selected patients. marginal artery; PCA ⫽ paracentral artery; SP ⫽ superior pa-
MRI was performed with a 1.5 Tesla MR imaging unit (Sig- rietal artery; GR ⫽ gyrus rectus; FP ⫽ frontal pole; SFG ⫽
na; GE Medical Systems, Milwaukee, WI) with echoplanar superior frontal gyrus; SMA ⫽ supplementary motor area;
capabilities. The common MRI parameters for diffusion- PL ⫽ paracentral lobule; PC ⫽ precuneus.
weighted imaging (DWI) (repetition time [TR] of 7,500
msec, echo time [TE] of 84 msec, matrix number of 128 ⫻
128, and two b values of 0 and 1,000 s/mm2) and T2- were obtained from chart review. Muscle strength was
weighted imaging (TR of 4,500 msec, TE of 118 msec) were graded as I to V using the Medical Research Council scale,
slice thickness of 5 mm, interslice gap of 2 mm, 20 axial and motor dysfunction was considered severe when the
slices, and field of view of 250 mm. Common MR angiogram grade of the weakest muscle was ⱕ III. Because the corpus
(MRA) parameters included flip angle of 20°, matrix number callosum and adjacent cingulate gyrus are supplied by the
of 512 ⫻ 512, and field of view of 250 mm. Three- pericallosal artery branches and were usually involved si-
dimensional TOF MRA of circle of Willis was performed multaneously, we lumped these together in clinical-imaging
with TR of 25 msec and TE of 2 msec. Three-dimensional correlation study. The SMA, paracentral lobule, and precu-
contrast-enhanced MRA from the aortic arch to the level of neus were also lumped together for the same reason. The
the central skull base was obtained using IV bolus injection imaging analysis was done by one of the authors (S.Y.K.),
of 20 mL (3– 4 mL/second) of gadopentetate dimeglumine blinded to the clinical and angiographic findings. The angio-
(Magnevist; Schering, Berlin, Germany) with TR of 6 msec graphic result was assessed by a neuroradiologist who was
and TE of 1 msec. blinded to clinical findings and a stroke neurologist (J.S.K.).
Presumed etiologies were defined as follows: 1) large ar- We defined the result as abnormal only when both agreed.
tery atherosclerosis, further divided into a) internal carotid For statistical analyses, Student t test was used for nu-
artery (ICA) disease: the patients have significant (⬎50%) merical variables and ␹2 tests for categorical variables.
narrowing or occlusion of ipsilateral ICA but without signif- Fisher exact test was used when the number of cells was
icant ipsilateral ACA atherosclerosis or emboligenic heart small. All statistical tests were performed with the use of the
disease, and b) ACA disease: ipsilateral ACA stenosis/occlu- SPSS program (version 10.0) and p values ⬍0.05 were re-
sion without significant ipsilateral ICA disease or emboli- garded as indicating significance.
genic heart disease; 2) cardiogenic embolism: there should be
an emboligenic heart disease based on Trial of Org 10172 in RESULTS Demography and clinical-imaging corre-
Acute Stroke Treatment (TOAST) stroke classification crite- lation. There were 58 men and 42 women, ages
ria9 without atherosclerotic diseases in the ipsilateral ICA or ranging from 29 to 88 (mean 65.1) years. Risk fac-
ACA; 3) two or more causes: patients with two or more po- tors as defined previously13 included hypertension
tential causes of stroke such as those with ICA disease and in 70, diabetes mellitus in 38, hypercholesterol-
emboligenic heart disease; 4) undetermined etiology: The
emia in 37, cigarette smoking in 38, a previous
cause of a stroke cannot be determined after thorough evalu-
ation.
history of stroke in 24, and a history of coronary
Vascular territories and topographic localization were heart disease in 11 patients.
determined according to the previous anatomic studies (fig- MRI showed 55 left infarcts, 36 right infarcts,
ure 1).10-12 ACA territory includes the rostrum, genu, trunk, and 9 bilateral infarcts. The corpus callosum/
and splenium of the corpus callosum, cingulate gyrus, fron- cingulate gyrus was involved in 68 patients (cor-
tal pole/gyrus rectus (frontopolar and orbitofrontal artery
pus callosum in 63, cingulate gyrus in 28). In the
territory), medial aspect of superior frontal gyrus (anterior
corpus callosum, the genu was the most fre-
and middle internal frontal artery territory), supplementary
motor area (SMA) (posterior internal frontal artery terri- quently involved (n ⫽ 54), followed by trunk (n ⫽
tory), paracentral lobule (paracentral artery territory), and 32), rostrum (n ⫽ 10), and the splenium (n ⫽ 8).
precuneus (superior parietal artery territory). Clinical results The SMA/paracentral lobule was involved in 55,

Neurology 70 June 10, 2008 (Part 2 of 2) 2387

 Table 1 Clinical symptoms and correlated anatomic location

Correlated lesion location

Symptoms Prevalence (no. of patients) Comments

Motor dysfunction 91

Severe motor 43 SMA/paracentral lobule (30 of 56,

dysfunction 53.6%, vs 13 of 44, 29.5%,
p ⫽ 0.016)*

Hypobulia/apathy/ 43 Frontal pole/gyrus rectus (13 of Bilateral lesion most

indifference 16 patients, 81.3%, vs 30 of 84, common followed by left
35.7%, p ⫽ 0.002) and then right lesions

Corpus callosum/cingulate gyrus

(36 of 68, 52.9%, vs
7 of 32, 21.9%, p ⫽ 0.003)

Superior frontal lobe (25 of 41,

61%, vs 13 of 53,
24.5%, p ⫽ 0.000)

Urinary incontinence 30 None

Grasp reflex 25 Corpus callosum/cingulate gyrus More common in women

(21 of 68, 30.9%, vs than in men
4 of 32, 12.5%, p ⫽ 0.052)

Aphasia 18 Left lesions in 16, bilateral Global in 8, transcortical in 8,

lesions in 2 patients motor in 2

Sensory deficit† 20 Always with motor dysfunction

*Expressed as prevalence of symptom in patients with involvement of the region vs prevalence of symptom in patients with-
out.
†Tested in 81 patients.

the superior frontal gyrus in 41, frontal pole/gyrus was closely related to involvement of frontal pole/
rectus in 16, and the subcortical area (caudate gyrus rectus, corpus callosum/cingulate gyrus,
head/anterior putamen/internal capsule) was in- and the superior frontal lobe, and occurred most
volved in 6 patients. Whole ACA territory in- frequently in patients with bilateral lesions (6 of
volvement was rare, occurring in only in four 9, 66.7%), followed by left sided lesions (28 of 55,
patients. 50.9%) and right sided lesions (9 of 36, 25%) (p ⫽
The prevalence of symptoms and correlated 0.016). Hypobulia was closely related to grasp re-
anatomic location are summarized in table 1. Mo- flex (p ⫽ 0.001).
tor dysfunction (n ⫽ 91) was the most common Urinary incontinence, observed in 30 patients,
symptom: hemiparesis in 70, leg monoparesis in was not related to particular lesion location.
18, and paraparesis in 3 patients. In patients with Aphasia was present in 18 patients (transcortical
hemiparesis, the weakness was more severe in the mixed aphasia 8, global aphasia 8, motor aphasia
leg than in the arm in 37 patients. Thus, among 2). Global aphasia was closely related to superior
the 91 patients with motor dysfunction, leg weak- frontal gyrus involvement (p ⫽ 0.039) while
ness was dominant in 58 patients (64%). In 29 transcortical aphasia was not related to any par-
patients, the degree of motor dysfunction was ticular lesion location. Grasp reflex (n ⫽ 25) was
similar between arm and leg, while in four, arm more often observed in women (15 of 42, 35.7%)
was weaker than the leg. In these 33 patients, the than in men (10 of 58, 17.2%) (p ⫽ 0.035), and
shoulder was discrepantly weaker than the hand closely related to corpus callosum/cingulate gyrus
in the majority (n ⫽ 25). The motor dysfunction involvement. Alien hand sign was observed in 10
was severe in 43, and the severe motor dysfunc- patients. Although patients with involvement of
tion was closely related to the involvement of the corpus callosum/cingulate gyrus more often
SMA/paracentral lobule. had this symptom (9 of 68, 13.2%) as compared
Sensory examination was reliably performed to those without (1 of 32 patients, 3.1%), the dif-
in 81 patients while it was not in others due to ference was not significant (p ⫽ 0.162). Being
significant abulia or aphasia. Among them, 20 pa- women was marginally related to this symptom
tients (25%) had dysfunction: hemihypesthesia in (p ⫽ 0.09).
12 patients, sensory symptoms in the leg only in 7, Other miscellaneous symptoms included emo-
in the arm only in 1 patient. The sensory dysfunc- tional lability/incontinence14 (n ⫽ 10), drowsiness/
tion was always detected in the paretic limbs. Hy- somnolence (n ⫽ 8),5 acute confusion/agitation (n ⫽
pobulia/apathy/indifference shown in 43 patients 8), motor perseveration (n ⫽ 8), ideomotor apraxia

2388 Neurology 70 June 10, 2008 (Part 2 of 2)

 bolic infarction; 1 had a siphon stenosis while
Table 2 Etiologies of anterior cerebral infarction
others had proximal ICA disease.
No. of patients Emboligenic heart diseases were found in 12
Etiologies (n ⫽ 100)
patients: atrial fibrillation in 8, valvular heart dis-
Large artery atherosclerosis 73
ease in 1, atrial fibrillation and valvular disease in
ACA disease 61 1, cardiomyopathy in 1, and patent foramen ovale
Branch occlusion 20 with right to left shunt in 1 patient. Because two
Artery-to-artery embolism 12 of them had concomitant, significant atheroscle-
In situ thromboocclusion 20 rosis, 10 were considered to have embolic infarc-
Combined 9 tion due to a heart disease. Fifteen patients were
ICA disease (ICA to ACA embolism) 6 considered to have unknown mechanism.
Either ACA or ICA disease 6
Stroke mechanisms in ACA atherosclerosis. After an-

Cardiogenic embolism 10
alyzing MRI and angiographic data, we eluci-
dated the following stroke mechanisms in 61
Two or more causes
patients with intrinsic ACA disease.
ICA atherosclerosis ⫹ cardiac 1
embolism Local branch occlusion (LBO) (n ⫽ 20). The infarct
ACA atherosclerosis ⫹ 1 was localized to the area adjacent to the athero-
cardiac embolism sclerotic vessel, presumably due to an occlusion
Unknown 15 of the orifice of one or more of the perforators
supplying adjacent structures by the local throm-
ACA ⫽ anterior cerebral artery; ICA ⫽ internal carotid ar-
tery. bus formation.17-19 The infarct was usually limited
to the anterior part of the corpus callosum or the
cingulate gyrus (figure 2A). Occlusion of the fron-
(n ⫽ 7),5 amnesia (n ⫽ 6), altered consciousness (n topolar or anterior inferior frontal branch, which
⫽ 4), gaze deviation (n ⫽ 3), parkinsonian features usually stems from A2,10 may produce extension
(bradykinesia, cogwheel rigidity with or without of the lesion to the anterior part of the superior
tremor)15,16 (n ⫽ 3), and anosognosia (unawareness frontal gyrus (figure 2, B and C).
of the paralytic limbs) (n ⫽ 2 ). In situ thrombotic occlusion (ITO) (n ⫽ 20). The in-

Angiographic findings and presumed stroke mecha- farct extensively involved the whole or the most
nisms. Angiographies performed were as follows: part of the ACA territory (figure 2, D and E). It
MRA in 77, conventional angiography in 20, and can be explained by occlusion of the main ACA
CT angiography in 12 patients. Seven underwent trunk. However, a part of cortical areas may be
both MRA and conventional angiography and spared due to collateralization from the MCA or
two underwent MRA and CT angiography. Local contralateral ACA.
ACA atherosclerosis (40 stenoses and 28 occlu- Artery-to-artery embolism (AAE) (n ⫽ 12). The in-
sion) ipsilateral to the infarct was observed in 68 farcts were usually small, sometimes scattered,
patients, most frequently at A2 (n ⫽ 38) followed and situated at the areas supplied by distal corti-
by A1 (n ⫽ 22), A2–A3 junction (n ⫽ 5), and cal branches. The infarcts, which can be ex-
A1–A2 junction (n ⫽ 3). For statistics, A1–A2 plained by artery-to-artery embolism, are often
junction was included in A1 disease (A1 group, scattered in the ACA and MCA borderzone area
n ⫽ 25) and A2–A3 junction disease in A2 disease (figure 3A).
(A2 group, n ⫽ 43). A1 group patients more often Combination of LBO and AAE (n ⫽ 9). There were
had occlusion (16 of 25, 64%) than A2 group (12 infarcts located adjacent to the atherosclerotic
of 43, 27.9%) (p ⫽ 0.001). vessel as in LBO. In addition, scattered infarcts in
General stroke mechanisms. Presumed stroke the remote area were observed as in AAE (figure
mechanisms are summarized in table 2. Among 3, B and C). Unlike infarcts caused by ITO, the
the 68 patients with ACA disease, 6 had addi- lesions were not in continuity with each other.
tional significant ipsilateral ICA disease, and 1 We then compared patients with intrinsic ACA
had emboligenic heart disease. Therefore, 61 pa- atherosclerosis (n ⫽ 61) and those with embolism
tients had infarction caused by intrinsic ACA dis- from either the heart or ICA disease (n ⫽ 16). The
ease. Significant ipsilateral ICA disease was former group more often had corpus callosal in-
shown in 13 patients (5 occlusion, 8 stenosis), volvement (p ⫽ 0.016) and tended to have hypobu-
among whom 6 had concomitant ACA disease lia more often (p ⫽ 0.077) than the latter group.
and 1 had emboligenic heart disease. Thus, 6 pa- Stroke mechanisms in patients with bilateral ACA in-
tients were considered to have ICA to ACA em- farction. We identified nine patients with bilateral

Neurology 70 June 10, 2008 (Part 2 of 2) 2389

 Figure 2 Examples of stroke caused by local branch occlusion (A–C) and in situ thrombotic occlusion (D, E)

(A) T2-weighted MRI shows an infarct localized to the right anterior portion of the corpus callosum (left figure, small arrows),
which was caused by left anterior cerebral artery stenosis (right figure, large arrow). (B) Upper row: gadolinium-enhanced
T1-weighted MRI shows that the lesions involve the anterior part of the corpus callosum and cingulate gyrus. In addition, a
sagittal image shows involvement of anterior part of superior frontal cortex (black arrows). Lower row: MR angiogram shows
stenosis in the A2 and A3 portion of anterior cerebral artery (white arrows). (C) The diagram schematically illustrates the
occlusion of local branches supplying the corpus callosum due to atherothrombosis in the anterior cerebral artery. Dark area
⫽ thrombosis; gray area ⫽ infracted area. (D) Upper row: T2-weighted MRI shows a large infarct involving whole anterior
cerebral artery territory including the corpus callosum, frontal pole, cingulate gyrus, superior frontal lobe, supplementary
motor area, paracentral lobule, and precuneus. Lower row: MR angiogram shows occlusion of left anterior cerebral artery
(arrow). There was an absent right A1, and right anterior cerebral artery is supplied by a branch from the left A1. (E) The
diagram schematically illustrates extensive thrombotic occlusion of anterior cerebral artery producing whole territory infarc-
tion. Dark area ⫽ thrombosis; gray area ⫽ infracted area.

2390 Neurology 70 June 10, 2008 (Part 2 of 2)

 left A1 occlusion. Another had unilateral ICA ste-
Figure 3 Examples of stroke caused by artery-to-artery embolism (A) or
combined mechanism of local branch occlusion and artery-to-artery
nosis and bilateral A2 stenosis. The third had left
embolism (B) ICA occlusion and left A1 occlusion with the ab-
sence of right A1. One patient had emboligenic
heart disease, in whom the emboli probably trav-
eled to bilateral ACAs. Finally, the mechanism
was unknown in one patient.

DISCUSSION Although the prevalence of ACA

infarction in our series was similar to previous
results,4-7 the stroke mechanisms were different.
Studies from western countries showed that ACA
infarction is usually caused by embolism,4,5 while
a Japanese study reported that ACA atherosclero-
sis was the most important etiology.6 Our study
with a much larger number of patients confirmed
that intrinsic ACA disease is the most common
etiology in the East Asian region.
The difference reflects ethnic differences in the
location of cerebral atherosclerosis,20-22 but also
implies that the incidence of ACA atherosclerosis
could have been underestimated in previous stud-
ies where vascular imaging was incompletely per-
formed. The relative rarity of embolism in ACA
infarction as compared to MCA territory infarc-
tion19 may be related to the fact that the diameter
of A1 is only about half of the MCA and that
communications between both ACAs through the
A-com10 might play a role in the clearance of em-
boli.
We proposed novel stroke mechanisms in pa-
tients with intrinsic ACA diseases: LBO, AAE,
Left image: diffusion-weighted MRI shows scattered infarcts in anterior cerebral artery– and ITO. LBO represents occlusion of an orifice
middle cerebral artery borderzone area. Right image: MR angiogram shows significant steno-
sis in the A2–A3 junction of left anterior cerebral artery. (B) Upper row: diffusion-weighted
of smaller branches arising from the proximal
MRI shows infarcts at the left anterior corpus callosum and scattered infarcts at distal areas. ACA due to local thrombosis.17-19,23 Because ath-
Lower row: CT angiogram shows severe stenosis in the A2 portion of left anterior cerebral erosclerotic changes occurred most frequently at
artery (arrow). (C) The diagram schematically illustrates the stroke mechanism of combined
A2 segment followed by A1, and A2–A3 junction,
local branch occlusion and artery-to-artery embolism. Dark area ⫽ thrombosis; gray area ⫽
infracted area; dots ⫽ emboli. perforators as well as orbitofrontal, frontopolar,
or anterior internal frontal arteries arising from
these segments10 can be occluded, resulting in in-
infarcts; in four, the infarcts were considered to farcts usually restricted to the anterior part of the
be caused by proximal ACA disease. In three of corpus callosum, cingulate gyrus, and adjacent
them, the mechanism was AAE; one had occlu- frontal areas (figure 2, A and B). We also ob-
sion of both ACAs (right A2 and left A1–2 junc- served patients with AAE and those with a com-
tion), while another had both A1 occlusion. In bined mechanism of LBO and AAE. Thus, the
these two, AAE may have occurred from both stroke mechanisms seem to be similar to that of
ACA diseases. The third had left A1 stenosis, in intrinsic MCA atherosclerosis,18,19 except that
whom the emboli may have traveled to the oppo- ITO producing a large infarct is more prevalent in
site hemisphere through the intact anterior com- ACA infarction. However, even in patients with
municating artery (A-com). The fourth had left ITO, the cortical area was often spared, probably
azygos ACA supplying both hemispheres with the due to collateralization from the MCA or con-
absent right A1. The stroke mechanism was con- tralateral ACA. The embolic infarcts related to
sidered combined (LBO⫹AAE). In three patients, AAE were often scattered at the ACA and MCA
the mechanism was an embolism either from ICA borderzone area (figure 3), which may be ex-
or ACA. One patient had left ICA occlusion and plained by the insufficient washing out of emboli

Neurology 70 June 10, 2008 (Part 2 of 2) 2391

 in the hypoperfused area18,24 in the presence of se- more frequently in women than in men, which
vere ACA stenosis. Therefore, aside from stroke may be consistent with the previous notion that
mechanisms so far described, the status of collat- connecting fibers in the corpus callosum are more
eral circulations and hemodynamic factors ap- abundant in women than in men.37
pear to play additional roles in determining the There are limitations in our studies. First, this
pattern of ACA infarction and consequent clinical is a retrospective study, and inaccurate or insuffi-
syndromes. We also found that mechanisms for cient assessment might have been present. Sec-
bilateral infarcts included embolism from the ICA ond, because the patients were recruited from a
or the heart traveling to both hemispheres, simul- tertiary hospital, they may not represent the gen-
taneous embolization from both ACA stenoses, eral stroke population. To minimize referral bias,
and emboli from the A1 stenosis traveling to both we selected patients who were admitted within 1
hemispheres through the A-com. week after onset. Third, MRA was usually used
As expected, severe motor dysfunction was in this study, which is not sensitive enough in the
closely associated with SMA/paracentral lobule evaluation of distal arteries. Thus, at least some
involvement. Although predominant leg weak- of the patients classified as having unknown
ness was usual, identical weakness between arm mechanism may have arterial diseases at distal
and leg was not uncommon. Because the shoulder branches. Finally, all our patients were Korean,
was discrepantly weaker than the hand in the ma- and our data may not be generalizable to other
jority of these patients, a partial involvement of ethnic groups.
arm-presenting motor fibers2,5,25 seems to explain
Received September 28, 2007. Accepted in final form Febru-
the arm weakness, although the possible presence
ary 20, 2008.
of motor neglect26,27 cannot be ruled out.
Our data showed that callosal or anteromedial
REFERENCES
frontal damages are responsible for hypobulia/ 1. Foix C, Hillemand P. Les syndromes de l’artère antér-
apathy/indifference, which was consistent with ieure. Encéphale 1925;20:209–232.
previous observations.28,29 These symptoms are 2. Critchley M. The anterior cerebral artery and its syn-
more prevalent in patients with bilateral lesions dromes. Brain 1930;53:120–165.
3. Castaigne P, Lhermitte F, Escourelle R, et al. Etude
followed by those with left lesions. Although uri-
anatomopathologique de 74 infarcts de l’artère céré-
nary incontinence was considered to be related to brale anterieure (55 observations). Rev Med Toulouse
lesions affecting the superior frontal gyrus, cingu- 1975;suppl:339.
late, and the white matter in between,25,30 we 4. Gacs G, Fox AJM, Barnett HJM, Vinuela F. Occur-
failed to find any specific anatomic location re- rence and mechanisms of occlusion of the anterior cere-
lated to this symptom, as in previous studies.5,31,32 bral artery. Stroke 1983;14:952–959.
5. Bogousslavsky J, Relgi F. Anterior cerebral infarction
Perhaps urinary incontinence may be related to
in the Lausanne Stroke Registry: clinical and etiologi-
extensive lesions involving multiple neuronal cal patterns. Arch Neurol 1990;47:144–150.
pathways regulating bladder function. 6. Kazui S, Sawada T, Naritomi H, Kuriyama Y. Angio-
We found that transcortical aphasia character- graphic evaluation of brain infarction limited to the an-
istically occurred in patients with ACA infarc- terior cerebral artery territory. Stroke 1993;24:549–
tion.5,25,33,34, However, we also observed as many 553.
7. Kumral E, Bayulkem B, Evyapan D, Yunten N. Spec-
patients with global aphasia. The aphasia might
trum of anterior cerebral artery territory infarction:
be a manifestation of abulia. However, we con- clinical and MRI findings. Eur J Neurol 2002;9:615–
sidered that the speech dysfunction of these pa- 624.
tients was out of proportion of the abulia. 8. Kubis N, Guichard J-P. Isolated anterior cerebral ar-
Moreover, out of 16 patients with aphasia, 6 did tery infarcts: a series of 16 patients. Cerebrovasc Dis
not have abulia. Unlike the patients with MCA 1999;9:185–187.
9. Adams HP, Bendixen BH, Kappelle J, et al. The
infarction, the aphasias tended to improve rap-
TOAST Classification of subtype of acute ischemic
idly. Thus, we assumed that the initiation or reg- stroke: definitions for use in a multicenter clinical trial.
ulation of speech may be dysfunctional in patients Stroke 1993;24:35–41.
with medial frontal lesions,12,33,35,36 which could 10. Perlmutter D, Rhoton AL Jr. Microsurgical anatomy
result in transient aphasia. Our finding that of the distal anterior cerebral artery. J Neurosurg 1978;
global aphasia was closely related to medial supe- 49:204–228.
11. Tatu L, Moulin T, Bogousslavsky J, Duvernoy H. Ar-
rior frontal lobe involvement seems to be consis-
terial vascular territories of human brain: cerebral
tent with this theory. Grasp reflex was closely hemispheres. Neurology 1998;50:1699–1708.
related to corpus callosum/cingulate involvement. 12. Brust JCM, Sawada T, Kazui S. Anterior cerebral ar-
This and alien hand phenomenon tended to occur tery. In: Bogousslavsky J, Caplan LR, eds. Stroke Syn-

2392 Neurology 70 June 10, 2008 (Part 2 of 2)

 dromes. New York: Cambridge University Press, 2001: Pathophysiology, Diagnosis, and Treatment. Philadel-
439–450. phia: Churchill Livingstone, 1998:401–425.
13. Cho AH, Kang DW, Kwon SU, Kim JS. Is 1.5 cm crite- 26. Brust JCM. Lesions of the supplementary motor area.
rion for lacunar infarction still valid? A study on In: Luders HO, ed. Supplementary Sensorimotor Area.
strictly subcortical middle cerebral artery territory in- Vol 70. New York: Raven Press, 1996:237–248.
farction using diffusion weighted MRI. Cerebrovasc 27. Chamorro A, Marshall RS, Valls-Sole J, Tolosa E,
Dis 2007;23:14–19. Mohr JP. Motor behavior in stroke patients with iso-
14. Kim JS, Choi-Kwon S. Poststroke depression and emo- lated medial frontal ischemic infarction. Stroke 1997;
tional incontinence: correlation with lesion location. 28:1755–1760.
Neurology 2000;54:1805–1810. 28. Sussman NM, Gur RC, Gur RE, O’Connor MJ. Mut-
15. Klatka LA, Depper MH, Marini AM. Infarction in the ism as a consequence of callosotomy. J Neurosurg
territory of the anterior cerebral artery. Neurology 1983;59:514–519.
1998;17:69–75. 29. Ross MK, Reevers AG, Roberts DW. Post-
16. Kim JS. Involuntary movements after anterior cerebral commissurotomy mutism. Ann Neurol 1984;16:114.
artery territory infarction. Stroke 2001;32:258–261. 30. Andrew J, Natan PW. Lesions of the anterior frontal
17. Fisher CM. Capsular infarcts: the underlying vascular lobes and disturbances of micturition and defecation.
lesions. Arch Neurol 1979;36:65–73.
Brain 1964;87:233–262.
18. Wong KS, Gao S, Chan YL, et al. Mechanisms of acute
31. Gelber DA, Good DC, Laven LJ, Verhulst SJ. Causes
cerebral infarctions in patients with middle cerebral ar-
of urinary incontinence after acute hemiparetic stroke.
tery stenosis: a diffusion-weighted imaging and micro-
Stroke 1993;24:378–382.
emboli monitoring study. Ann Neurol 2002;52:74–81.
32. Nakayama H, Jorgensen HS, Pedersen PM, Raaschou
19. Lee DK, Kim JS, Kwon SU, Kang DW. Lesion patterns
and stroke mechanism in atherosclerotic middle cere- HO, Olsen TS. Prevalence and risk factors of inconti-
bral artery disease: early diffusion-weighted MRI nence after stroke: the Copenhagen stroke study.
study. Stroke 2005;36:2583–2588. Stroke 1997;28:58–62.
20. Gorelick PB, Caplan LR, Hier DB, Parker SL, Patel D. 33. Alexander MP, Schmitt MA. The aphasia syndrome of
Racial differences in the distribution of anterior circu- stroke in the left anterior cerebral artery territory. Arch
lation occlusive disease. Neurology 1984;34:54–59. Neurol 1980;37:97–100.
21. Caplan LR, Babiian V, Helgason C, et al. Occlusive 34. Ross ED. Left medial parietal lobe and receptive language
disease of the middle cerebral artery. Neurology 1985; functions: mixed transcortical aphasia after left anterior
35:975–982. cerebral artery infarction. Neurology 1980;30:144–151.
22. Kim JS, Kang DW, Kwon SU. Intracranial atheroscle- 35. Masdeu JC, Schoene WC, Funkenstein H. Aphasia fol-
rosis: incidence, diagnosis and treatment. J Clin Neurol
lowing infarction of the left supplementary motor area.
2005;1:1–7.
Neurology 1978;28:1220–1223.
23. Caplan LR. Intracranial branch atheromatous disease:
36. Bruton DB, Chelune GJ, Naugle RI, Bleasel A. Neurocog-
a neglected, understudied, and underused concept.
nitive studies in patients with supplementary sensorimo-
Neurology 1989;39:1246–1250.
24. Caplan LR, Hennerici M. Impaired clearance of emboli tor area lesions. In: Luders HO, ed. Supplementary
(washout) is an important link between hypoperfusion, Sensorimotor Area. Vol 70. New York: Raven Press,
embolism, and ischemic stroke. Arch Neurol 1998;55: 1996:249–261.
1475–1482. 37. De Lacoste-Utamsing C, Holloway RL. Sexual dimor-
25. Brust JCM. Anterior cerebral artery disease. In: Bar- phism in the human corpus callosum. Science 1982;216:
nett HJM, Mohr JP, Yatsu F, Stein B, eds. Stroke: 1431–1432.

Neurology 70 June 10, 2008 (Part 2 of 2) 2393