Accepted: 8 September 2016

DOI: 10.1111/ane.12690

ORIGINAL ARTICLE

Ipsilateral hemiparesis in ischemic stroke patients

Y. Inatomi1 | M. Nakajima2 | T. Yonehara1 | Y. Ando2

1
Department of Neurology, Saiseikai
Kumamoto Hospital, Kumamoto, Japan
2
Department of Neurology, Graduate School
of Medical Sciences, Kumamoto University,
Kumamoto, Japan
Correspondence
Y. Inatomi, Department of Neurology,
Saiseikai Kumamoto Hospital, Kumamoto,
Japan.
Email: y.inatomix@silk.ocn.ne.jp
Objectives: To investigate clinical characteristics of ipsilateral hemiparesis in ischemic
stroke patients.
Materials and methods: Patients with acute ischemic stroke were prospectively exam-
ined. Ipsilateral hemiparesis was defined as hemiparesis ipsilateral to recent stroke le-
sions. Patients with ipsilateral hemiparesis were examined with functional neuroimaging
studies including transcranial magnetic stimulation (TMS) and functional MRI.
Results: Of 8360 patients, ipsilateral hemiparesis was detected in 14 patients (0.17%,
mean age 71±6 years, eight men). Lesions responsible for the recent strokes were lo-
cated in the frontal cortex in three patients, corona radiata in seven, internal capsule
in one, and pons in three. These lesions were located along the typical route of the
corticospinal tract in all but one patient. Thirteen patients also had a past history of
stroke contralateral to the recent lesions; 12 of these had motor deficits contralateral
to past stroke lesions. During TMS, ipsilateral magnetic evoked potentials were evoked
in two of seven patients and contralateral potentials were evoked in all seven.
Functional MRI activated cerebral hemispheres ipsilaterally in eight of nine patients
and contralaterally in all nine.
Conclusions: Most patients with ipsilateral hemiparesis had a past history of stroke
contralateral to the recent one, resulting in motor deficits contralateral to the earlier
lesions. Moreover, functional neuroimaging findings indicated an active crossed corti-
cospinal tract in all of the examined patients. Both findings suggest the contribution of
the uncrossed corticospinal tract contralateral to stroke lesions as a post-­stroke com-
pensatory motor system.

KEYWORDS
cerebrovascular diseases, ipsilateral hemiparesis, strokes

1 |  INTRODUCTION Studying ipsilateral hemiparesis may contribute to our inves-

Cases of ipsilateral hemiparesis or deterioration of hemiparesis ip-
silateral to the cerebral injury are known; to the best of our knowl-
1–13
edge, there are reports of 17 such patients (Table 1). Ipsilateral
hemiparesis is hypothesized to be caused by injury to the uncrossed
corticospinal tract in patients without decussation of the corticospi-
4,5,9–11,13
nal tract, injury to the uncrossed corticospinal tract acquired
after prior brain injury,1,6,7 injury to the ipsilateral extrapyramidal
motor pathway or premotor cortex,8,14,15 or transhemispheric diaschi-
sis from the affected side.16,17
tigations of post-­stroke plastic reorganization,14,15 especially the
compensatory function of the corticospinal tract contralateral to
the injury. Post-­stroke plastic reorganization has been assessed by
functional neuroimaging studies including positron emission tomog-
raphy,18 transcranial magnetic stimulation (TMS),19–21 functional
MRI (fMRI),18,22 magnetoencephalography,23 and diffusion tensor
imaging.21,24 Case-­control studies or serial observation studies, in
which ordinary stroke patients were examined by these functional
neuroimaging methods, have provided only indirect evidence about
the reorganization system because it is uncertain that atypical regions

Acta Neurol Scand 2016; 1–10 wileyonlinelibrary.com/journal/ane © 2016 John Wiley & Sons A/S.  |  1
Published by John Wiley & Sons Ltd
T A B L E   1   Clinical characteristics of patients with ipsilateral hemiparesis reported in the literature
|

Past stroke Recent stroke

2      

DTR Babinski sign Transcranial motor stimulation Functional MRI

Ipsi Contra Contra

stimulation stimulation Ipsi tapping tapping
Interval Previous
Age, between motor Side and severity Ipsi Contra Ipsi Contra Ipsi Contra Ipsi Contra
sex Type and lesion strokes deficit Type and lesion of hemiparesis Ipsi Contra Ipsi Contra MEP MEP MEP MEP activ activ activ activ Ref.

75F BI, R basal ganglia 22 days BI, L internal capsule R complete→ L, ++ ++ + + n.t. n.t. n.t. n.t. n.t. n.t. n.t. n.t. 1
to basal gangliaa slight on Day 4
39F None -­ -­ AVM, L insular cortexb L, slight loss ++ ++ − − n.t. n.t. n.t. n.t. n.t. n.t. n.t. n.t. 2
62M None -­ -­ SAH and ICH, L L, moderate → no − − − − n.t. n.t. n.t. n.t. n.t. n.t. n.t. n.t. 3
insular cortex deficit on Day 3
60M None -­ -­ ICH, R internal capsule R, moderate ++ n.d. + − − − − + n.t. n.t. n.t. n.t. 4
to basal ganglia 10 months later
62M None -­ -­ ICH, R internal capsule R, moderate ++ n.d. + − − − − + + − − + 5
to basal ganglia 10 months later
59M ICH, R putamen 5 years Mild BI, L corona radiata L, deteriorated n.d. n.d. n.d. n.d. n.t. n.t. n.t. n.t. + − + − 6
markedly
62M BI, L corona radiata 9 years Good BI, L corona radiata L, MRC 3+/4 ++ n.d. + − n.t. n.t. n.t. n.t. + + + − 7
ICH, R thalamus 1 year recov-
and subcortex ery
41M BI, R corona radiata 1 month Full BI, L corona radiata L, MRC 4+/4+ ++ n.d. + − n.t. n.t. n.t. n.t. + + + − 7
recov-
ery
74M BI, L corona radiata, 11 years Residual BI, R thalamus R, MRC 2/3 → 3/4 ++ ++ + − n.t. n.t. n.t. n.t. n.t. n.t. n.t. n.t. 8
cortical lesions in
MCA territory
76M BI, L corona radiata, 11 years Residual BI, R corona radiata R, MRC 3/3 → 4/4 ++ n.d. + − n.t. n.t. n.t. n.t. n.t. n.t. n.t. n.t. 8
cortical lesions in
MCA territory
64M BI, L paraventricular 4 years Residual BI, R corona radiata R, MRC 3/3 → 4/4 n.d. n.d. n.d. n.d. n.t. n.t. n.t. n.t. n.t. n.t. n.t. n.t. 8
region, occipital
lobe
55M None (HGPPS) -­ -­ BI, L corona radiata L, MRC 4/4 → n.d. n.d. n.d. n.d. n.t. n.t. n.t. n.t. n.t. n.t. n.t. n.t. 9
improved
35M None (agenesis of -­ -­ BI, R cortical and R, MRC 4/4 n.d. n.d. n.d. n.d. + − − + n.t. n.t. n.t. n.t. 10
corpus callosum) subcortical regions
of MCA territory
INATOMI

(Continues)
et al
 INATOMI et al |
      3

activated neurophysiologically actually work to compensate for dam-

Ipsi, ipsilateral side to recent lesion; Contra, contralateral side to recent lesion; MEP, magnetic evoked potential; AVM, arteriovenous malformation; SAH, subarachnoid hemorrhage; BI, brain infarction; ICH,
intracerebral hemorrhage; R, right; L, left; MRC, manual motor power test using Medical Research Council grading (arm/leg); n.t., not tested; n.d., not described; HGPPS, horizontal gaze palsy with progressive
Ref.

11
12
aged ­function in the stroke patients. On the other hand, clinical and

12

13
neurophysiological findings of ipsilateral hemiparesis may identify ana-

Contra
activ
tomical and functional characteristics of the contralateral hemispheric

n.t.
n.t.

n.t.

n.t.
compensatory motor system that had actually been functional until
tapping
Contra

activ
onset of the recent stroke had damaged them.
Ipsi

n.t.
n.t.

n.t.

n.t.
Contra Here, we present a study in which ischemic stroke patients with
Functional MRI

activ ipsilateral hemiparesis were evaluated with further neurological exam-

n.t.
n.t.

n.t.

n.t.
Ipsi tapping

inations and functional neuroimaging.

activ
Ipsi

n.t.
n.t.

n.t.

n.t.
2 | METHODS
Contra
Transcranial motor stimulation

MEP
stimulation

2.1 | Study patients
n.t.
n.t.

n.t.

n.t.
Contra

MEP

This was a prospective study of consecutive acute ischemic stroke

Ipsi

n.t.
n.t.

n.t.

n.t.

patients who were admitted to our hospital within 7 days of the onset
Contra

of ischemic stroke including transient ischemic attack from April 1,

MEP
stimulation

n.t.
n.t.

n.t.

n.t.

2003, to March 31, 2015. All of the patients were initially examined
by neurologists on admission, and were evaluated with carotid echog-
MEP
Ipsi

Ipsi

n.t.
n.t.

n.t.

n.t.

raphy and brain MRI/MRA; if these were not available, then CT scans
were conducted. The study protocol followed the principles outlined
Babinski sign

Contra

She was anatomically diagnosed by only CT scan. In all of the others, MRI including diffusion-­weighted imaging was performed.
n.d.
n.d.

n.d.

n.d.

in the Declaration of Helsinki, and all patients provided informed writ-

ten consent to the use of their medical data.
n.d.
n.d.

n.d.

n.d.
Ipsi
Contra

2.2 | Criteria of ipsilateral hemiparesis

n.d.
n.d.

n.d.

n.d.

Ipsilateral hemiparesis was defined as hemiparesis in patients satis-

DTR

n.d.
n.d.

n.d.

n.d.
Ipsi

fying the following criteria: 1. Patients who experienced obvious oc-

persisted on Day 7

currence, recurrence, or worsening of hemiparesis, as well as those

L, n.d. → improved
R, MRC 4/4 → not
Side and severity

R, crescendo TIA
L, MRC 3-­4/5 →

with novel bilateral hemiparesis due to the recent stroke. Patients

of hemiparesis

with motor deficits, including ataxia, bradykinesia, motor neglect, or

resolved

apraxia were excluded. 2. Patients in which the recent stroke lesions

were located over the upper medulla and ipsilateral to their hemipare-
sis. Patients with old lesions due to past brain injury were included. 3.
BI, L parietal-­superior

Patients with other recent ischemic lesions in the brain contralateral

BI, R corona radiata
Type and lesion

to the hemiparesis detected by MRI, and those who demonstrated

ICH, L putamen
Recent stroke

She was anatomically diagnosed by CT scan and pathological study.

frontal lobe
BI, R, frontal
operculum

hypoperfused areas with decreased cerebral blood flow due to arte-

rial severe stenosis or occlusion on MRA or carotid echography in the
contralateral cerebral hemisphere were excluded. Patients with diffi-
culty in communicating to explain their past and recent motor defi-
Previous

Residual

cits were also excluded from the present study, as were patients who
deficit
motor

could not be examined by MRI or SPECT due to their critical status,

n.d.
-­

-­

metal implants, or claustrophobia.

between
Interval

strokes

n.d.

n.d.

2.3 | Clinical consultation and physical

-­

-­

examinations
T A B L E   1   (Continued)

BI, R basal ganglia

Type and lesion

None (HGPPS)

All patients and their family members were interviewed during their
BI, R, internal
Past stroke

hospital stay by neurologists concerning their past history of stroke,

capsule

side of their residual motor deficits, and activity of daily life. Their
None

premorbid modified Rankin Scale scores were estimated based on the

scoliosis.

questionnaire results. Neurologists also evaluated their neurologi-

55M
57M

67M

55M
Age,
sex

cal findings including manual muscle power using Medical Research

b
a
 |
4       INATOMI
Council (MRC) grading, deep tendon reflexes, and pathological re-
flexes on admission and at discharge.
2.4 | Neuroimaging studies
MRI studies were performed using a Siemens MAGNETOM Vision
1.5-­Tesla MR unit with echo-­planar capability, and were carried out
on admission and in the subacute phase if the responsible lesions
were not detected on admission. Type and location of past and cur-
rent strokes were established from interviews of the patients and
their family members, recent MRI findings, and medical records from
previous injuries if available. Leukoaraiosis, as one of the findings of
subcortical microvascular disease, was evaluated as supratentorial
white matter hyperintensity on fluid-­attenuated inversion recovery
images and graded according to the Fazekas scale25 on the basis of
visual assessment of periventricular areas (0=absent, 1=caps or pencil
lining, 2=smooth halo, and 3=irregular periventricular hyperintensities
extending into deep white matter).
SPECT was performed using a double-­headed GCA7200A/PI
(Toshiba Medical Systems Corp, Otawara, Japan) or a triple-­headed
Prism 3000 (Shimadzu Corp, Kyoto, Japan) scanner. The patients were
123
injected with 167 MBq of I N-­isopropyl-­p-­iodoamphetamine or 600
99m
MBq of Tc-­ethylcysteinate dimer. SPECT acquisition parameters
for the scanners were as follows: 90 frames and a matrix of 64×64
pixels (1 pixel=4.3 mm) for the GCA7200A/PI system and 120 frames
and a matrix of 128×128 pixels (1 pixel=1.93 mm) for the Prism 3000
system. The data were reconstructed using a standard Butterworth-­
filtered back projection.
If SPECT revealed occult low-­uptake areas accompanied by critical
stenotic or occlusive lesions in the proximal arteries on echography
or MRA in the hemisphere contralateral to the side of recent stroke,
the patients were excluded from this study due to criteria 3. If low-­
uptake areas without infarcts or proximal occlusive arterial diseases
were demonstrated, the area was regarded as dysfunctional due to
mechanisms other than ischemia such as diaschisis, and the patients
were included in this study.
As TMS was performed in another institute (Kumamoto Kinoh
Hospital), those patients, who could not be transferred because of
severe general condition and/or lack of familial support for trans-
portation, were not examined by TMS. Magnetic evoked potentials
(MEPs) were evoked at rest by a single-­pulse TMS with a High Power
Magstim 200 machine and a figure-­eight coil with mean loop diam-
eters of 70 mm. The magnetic stimulus had a nearly monophasic
pulse configuration with a rise time of approximately 100 μs, de-
caying back to zero over approximately 0.8 ms. The orientation of
coil stimulation was chosen because motor threshold is minimum
when the induced electrical current in the brain flows approxi-
mately perpendicular to the line of the central sulcus. MEPs were
recorded from electrodes over the right and left first dorsal interos-
seous muscles or abductor pollicis brevis muscles if the first dorsal
interosseous muscles were atrophic. These bilateral MEPs were re-
corded after a single-­pulse TMS of the contralateral primary motor
cortex. The signal was amplified, band-­pass filtered (20-­3000 Hz,
et al
Synax 1200, NEC), and acquired at a sampling rate of 5 kHz on a
personal ­computer for off-­line analysis (Signal Software, Cambridge
Electronic Design, Cambridge, UK). The optimum positions for ac-
tivation of the first dorsal interosseous or abductor pollicis brevis
muscles were determined by moving the coil in 1-­cm steps around
the presumed motor cortex. The site at which stimuli of slightly su-
prathreshold intensity consistently produced the largest MEPs in
the target muscle was marked with a grease pencil as the “motor
hot spot”. Each patient’s active maximal threshold over the motor
cortex was decided by where the first dorsal interosseous muscle
was evoked. TMS series at maximal stimulator output were repeated
under strong (<100% maximum) voluntary traction.
Functional MRI was performed using Toshiba medical systems
Vantage 1.5-­tesla XGV power plus package. Patients with too severe
muscle weakness or cognitive disorder to follow the technologist’s
commands of finger tapping were excluded. Initially, the patients
maintained a resting position, and baseline images (T1-­weighted im-
ages TR:1800 TE:10 TI:600 ms) were obtained. The patients were then
scanned while conducting a simple motor task consisting of three rep-
etitions of self-­paced finger tapping for 30 s followed by rest for 45 s.
Activated images were performed using T2* weighted imaging (FE-­EPI
sequence, TR:1000 ms TE:40 ms, ETS: 0.8, 3 min). Finally, these ac-
tivated images were mapped on the T1-­weighted images to identify
activated areas.
3 | RESULTS
A total of 8360 patients with the diagnosis of acute ischemic
stroke were admitted to our hospital during the 11-­year study
period. Fifteen of these patients were diagnosed with ipsilateral
hemiparesis. One of these with a medial medullary infarction in the
left side and left hemiparesis was not examined by our consulta-
tion and SPECT due to severe apnea, and was therefore excluded
from the present study. Consequently, 14 patients (0.17%, mean
age 71±6 years, eight men) were enrolled in the present study
(Tables 2 and 3). They were admitted to our hospital on Day 1
­(median; range 0-­7), and the median length of hospital stay was
14.5 (range 7-­29) days.
Twelve of these 14 patients had a past history of abrupt motor
deficits (hemiparesis in 11 and unilateral limb ataxia in one) ipsilat-
eral to the hemiparesis in the recent stroke. In all patients, the le-
sions responsible for previous motor deficits were identified to be
contralateral to their motor deficits in past strokes. Patient 2 had an
old and silent stroke lesion contralateral to the recent stroke lesion.
Patient 12 did not have any past history of motor deficits or old le-
sions. The median estimated premorbid modified Rankin Scale esti-
mated was 3 (range 0-­3). Vascular risk factors included hypertension
in 10 patients, diabetes mellitus in eight, dyslipidemia in seven, cur-
rent smoking in one, atrial fibrillation in two, and ischemic coronary
disease in one.
Ipsilateral hemiparesis occurred on the right side in eight patients
and on the left in six. None of the patients had bilateral hemiparesis.
 INATOMI
et al

T A B L E   2   Clinical characteristics of the present patients

Past stroke Recent stroke

Initial MRC Final MRC Initial DTR (arm/ Final DTR (arm/ Initial pathological Final pathological
Interval Side of Length of grading (arm/leg) grading (arm/leg) leg) leg) reflex (arm/leg) reflex (arm/leg)
Age, Side of between hemi-­ hospital
No. sex hemi-­paresis strokes Pre-­mRS paresis stay Ipsi Contra Ipsi Contra Ipsi Contra. Ipsi Contra Ipsi Contra Ipsi Contra

1 68F R 17 years 2 R 16 4/4 5/5 5/5 5/5 ++/++ +/+ ++/++ +/+ −/− −/− +/− −/−
2 73M None -­ 0 L 29 4/4 5/5 4/4 5/5 ±/± ±/± ±/± ±/± −/− −/− −/− −/−
3 59M L 12 years 3 L 16 3/4 5/5 4/5 5/5 +/+ +/+ ++/++ +/+ −/− −/− −/− −/−
4 60M L 2 years 1 R 7 3/4 5/5 4/4 5/5 ++/++ +/+ +/+ +/+ −/+ −/− −/+ −/−
5 72M R 1 year 2 R 19 3/3 5/5 4/4 5/5 ++/++ ++/++ ++/++ ++/++ −/+ −/+ −/+ −/+
R 5 months
6 65F L 11 years 3 L 11 3/3 5/5 5/5 5/5 ++/++ +/+ ++/++ +/+ +/− −/− +/− −/−
7 72F R ataxia 3 months 3 R 12 4/4 5/5 5/5 5/5 +/++ +/+ +/+ +/+ −/+ −/− −/+ −/−
8 71F L 2 years 1 L 11 4/3 5/5 4/4 5/5 ++/+ +/+ ++/++ +/++ −/− −/− −/− −/−
9 78F R 1 year 3 R 12 4/2 5/5 5/4 5/5 ±/+ −/+ +/+ ±/+ −/− −/− −/− −/−
10 68F L 2 years 3 L 14 4/3 5/5 4/4 5/5 −/− −/+ +/+ +/+ −/− −/− −/+ −/−
11 80M R 5 months 2 R 12 4/4 5/5 4/4 5/5 +/+ −/− +/+ −/− −/− −/− −/− −/−
12 73M None -­ 0 R 18 5/4 5/5 5/5 5/5 +/+ +/+ +/+ +/+ −/− −/− −/− −/−
13 70M R 12 year 1 L 15 4/4 5/5 5/5 5/5 +/+ +/+ −/− −/− −/− −/− −/− −/−
14 79M R 10 months 1 R 17 4/4 5/5 5/4 5/5 ±/± ±/± −/− −/− −/− −/− −/− −/−

mRS, modified Rankin Scale; MRC, manual motor power test using Medical Research Council grading (arm/leg); DTR, deep tendon reflex; Ipsi, ipsilateral side to recent lesion; Contra, contralateral side to recent
lesion; R, right; L, left.
|
      5
 |
6       INATOMI et al

T A B L E   3   Past and present stroke characteristics and TMS and fMRI findings

Present stroke

Transcranial motor stimulation Functional MRI

Ipsi stimulation Contra stimulation Ipsi tapping Contra tapping

Fazekas
Type and lesion of scale Ipsi Contra. Ipsi Contra Ipsi Contra Ipsi Contra
No. past stroke grade Lesion of infarct MEP MEP MEP MEP activated activated activated activated

1 BI, L corona radiata 2 R side of pons n.t. n.t. n.t. n.t. n.t. n.t. n.t. n.t.
2 BI, R globus 2 L corona radiata 20.6 19.2 19.2 n.e. sf pc pc -­
pallidum,
asymptomatic
3 BI, L thalamus, 1 L side of pons n.e. n.e. 20.6 n.e. sf pc, mf pc mf
asymptomatic
ICH, R thalamus
4 ICH, R putamen 2 L precentral gyrus, n.e. 18.8 19.2 n.e. n.t. n.t. n.t. n.t.
superior parietal
lobule
5 BI, L side of medulla 2 R corona radiata n.t. n.t. n.t. n.t. n.t. n.t. n.t. n.t.
BI, L side of pons
6 BI, R corona radiata 1 L corona radiata to n.t. n.t. n.t. n.t. n.t. n.t. n.t. n.t.
putamen
7 BI, R side of medulla 1 R corona radiata n.e. 19.9 20.5 n.e. n.t. n.t. n.t. n.t.
8 BI, R corona radiata 1 L cingulate gyrus, n.e. 17.0 19.8 n.e. pc pc pc -­
corpus callosum
9 BI, L internal 2 R side of pons n.t. n.t. n.t. n.t. pc, sf, mf pc, sf, mf pc, sf, mf pc, sf, mf
capsula to
occipital lobe
10 BI, R internal 2 L internal capsula n.t. n.t. n.t. n.t. -­ pc, mf pc, mf -­
capsula
11 BI, L corona 2 R corona radiata to n.e. 24.0 23.3 n.e. pc, mf pc, sf pc, sf -­
radiata to internal putamen
capsula
12 None 2 R corona radiata, n.t. n.t. n.t. n.t. pc pc pc -­
sup. frontal gyrus
13 BI, L putamen to 2 L cingulate gyrus, n.t. n.t. n.t. n.t. pc, mf pc, mf pc, mf pc, mf
internal capsula precentral gyrus
14 BI, L corona radiata 1 R corona radiata n.e. 20.8a 23.3a 26.9 pc, mf pc pc, sf -­

MEP, magnetic evoked potential (distal latency, ms); BI, brain infarction; ICH, intracerebral hemorrhage; n.t., not tested; n.e., not evoked on TMS or not
activated on functional MRI; pc, precentral gyrus; sf, superior frontal gyrus; mf, middle frontal gyrus.
a
Recorded at abductor pollicis brevis.

The median MRC grading was 4 (range 3-­5) in the arms and 4 (range
2-­4) in the legs on admission. Thirteen patients had a hemiparesis MRC
grading of 3 or more. Patient 9 had severe residual hemiparesis MRC
grading of 2 caused by a past stroke. At discharge, hemiparesis had
ameliorated in all patients, and the median MRC grading of the hemi-
paresis was 4.5 (range 4-­5) in the arms and 4 (range 4-­5) in the legs.
Hyperreflexia in the hemiparetic limbs was present in seven pa-
tients on admission and in five at discharge. Hyperreflexia in the non-­
hemiparetic limbs was present in one patient on admission and in two
at discharge. Pathological reflexes in the hemiparetic limbs were pres-
ent in four patients on admission and in one at discharge. Pathological
reflexes in the non-­hemiparetic limbs were present in one patient on
admission and in one at discharge.
Hemi-­sensory loss ipsilateral to the hemiparesis was observed in
six patients (patients 1, 4, 6, 9, 10, 14). They did not remember the
degree of the sensory loss before their recent stroke, and could not
discern an obvious change of sense. Patients 2 and 14 had noticed
dysesthesia since their past stroke; however, they denied any change
of degree of their dysesthesia. None of the patients had cortical defi-
cits, including aphasia or unilateral spatial neglect. The mirror phenom-
enon was observed in patient 14.
MRI in patients for which recent responsible lesions were de-
tected was performed on Day 1.5 (median, range Day 0-­8). Table 2 and
Figure 1 show the locations of the lesions responsible for the patients’
previous (13 patients) and recent strokes. The responsible lesions of
the recent ischemic stroke were along the corticospinal tract, whose
INATOMI et al
anatomical situation was established classically and neurophysiologi-
26,27
cally in 13 patients. By contrast, the responsible lesion in patient
8 was limited to the cingulate gyrus and corpus callosum. None of the
14 patients had agenesis of the corpus callosum, a lacunar state, or
severe leukoaraiosis (see Fazekas grade Table 3).
None of the 14 patients had any hypoperfusion area associated
with proximal arterial occlusive diseases in the cerebral hemisphere
contralateral to the recent stroke. Moreover, hypoperfusion lesions
without arterial occlusive diseases were also not observed in any of
the patients, including patient 8 who did not have a lesion located
along the corticospinal tract.
TMS was performed in seven patients on Day 9 (median, range
4-­26). There were no differences between those with or without TMS
examinations, including in the degree of motor weakness caused by
recent strokes, although the percentage of patients with premorbid
mRS>1 was higher in those who did undergo the study (50% in TMS
groups vs 63% in non-­TMS groups). Crossed MEPs were evoked in all
seven patients. Uncrossed MEPs ipsilateral to the recent lesion were
evoked in patient 2, and contralateral to the recent lesion in patient
14. None of the seven had only uncrossed and no crossed MEPs.
Functional MRI was performed in nine patients on Day 9 (median,
range 5-­12). Patients 5 and 6 had severe hemiparesis and/or cognitive
impairment and could not perform the necessary tapping for the study,
F I G U R E   1   MRI findings in the present patients. Left: T2-­weighted images showing responsible lesions of past stroke (arrow). Right:
Diffusion-­weighted images showing responsible lesions of recent stroke (arrow head)
|
      7
and the instruments were unfortunately under maintenance during the
stay of patients 1, 4, and 7. However, a comparison of those patients
with and without fMRI studies showed that there were no differences
in the degree of motor weakness caused by recent strokes between
the two groups, but the percentage of patients with right hemiparesis
was higher in those who did not undergo the examinations (44% in
fMRI group vs 80% in non-­fMRI group). Cortical activation contralat-
eral to the tapping side was observed in all nine patients and ipsilateral
to the tapping side in eight of them. None of the patients exhibited
only ipsilateral activation but not contralateral activation.
4 | DISCUSSIONS
The present study of ipsilateral hemiparesis in ischemic stroke pa-
tients has three major findings. First, most patients had a past his-
tory of stroke contralateral to the side of the recent stroke, and their
past stroke had caused unilateral motor deficits contralateral to the
lesions. Second, the recent stroke lesions responsible for the ipsilat-
eral hemiparesis were located along the corticospinal tract in all but
one of the patients. Finally, functional neuroimaging studies revealed
findings indicating an active crossed corticospinal tract in all of the
examined patients. As none of the 14 patients had findings indicating
 |
8       INATOMI
the existence of severe microvascular disease nor misery perfusion in
the contralateral brain on MRI and SPECT, we concluded that the po-
tential effects of ischemia in the brain contralateral to their ipsilateral
hemiparesis were minimum.
To focus on the dynamic activity of the crossed and uncrossed
corticospinal tracts, we divide the pathophysiological mechanisms of
ipsilateral hemiparesis into three possible injury types (Figure 2): Type
I is an injury of the corticospinal tract in patients who congenitally
have no decussation of the tract; Type II is an injury of the uncrossed
(lateral) corticospinal tract in patients who do have active crossed and
uncrossed tracts; Type IIa is an injury of the uncrossed corticospinal
tract, in patients with both crossed and uncrossed active tracts, that
arose secondarily to a previous injury of the crossed tract.
Previous strokes contralateral to the recent strokes were present
in 12 of our patients and in eight of 17 reported patients (Table 1)
with ipsilateral hemiparesis.1,6–8,12 Moreover, these past strokes man-
ifested unilateral motor deficits contralateral to the lesions in all of the
present and previously reported patients. Thus, we suggested that all
of the 12 patients had additional hemiparesis that could not be ex-
plained by residual motor deficits caused by the past strokes. The high
frequency of previous strokes contralateral to the recent strokes sug-
gests that prior damage of the corticospinal tract contralateral to the
recent stroke might be associated with the pathogenesis of ipsilateral
hemiparesis. In other words, their past strokes might have generated
the conditions for the ipsilateral hemiparesis. Although patient 2 did
not have past history of unilateral motor deficit, he did have old and
silent stroke lesions contralateral to the recent stroke. In contrast, pa-
tient 12 had no history of brain injury; therefore, the above-­mentioned
hypothesis may not pertain to this or other reported patients without
past brain injury.2–4,9–12
The responsible lesions of ipsilateral hemiparesis were located
along the corticospinal tract in all 17 of the previously reported pa-
tients.1–13 Moreover, there was no observed hemiparesis contralateral
to the recent ischemic lesions in any of the present patients. There
was only one reported patient with bilateral hemiparesis due to unilat-
eral stroke.1 One possible explanation for these two findings regard-
ing ipsilateral hemiparesis might be that the non-­crossed and crossed
Type I Type II
Corticospinal tract
Active
㽢 㽢 㽢1
Inactive
Brain injury 㽢
1st attack 㽢䠍
2nd attack 㽢䠎
Contralateral MEP (–) (+)
Requirement of
(–) (–)
past ipsilateral injury
et al
corticospinal tracts might be located closely but clearly separated, and
the lesions might have involved only the crossed tract and spared the
non-­crossed tract.
In patient 8, the responsible lesions were located in the cingulate
gyrus and corpus callosum, far away from the corticospinal tract. On
SPECT, hypoperfusion areas indicating cerebral diaschisis were not
detected along the corticospinal tract, thus suggesting that the ipsi-
lateral hemiparesis had another pathophysiological mechanism than
the others.
Crossed MEPs on TMS were not found in any of the three previ-
ously reported patients so tested,4,5,10 nor was contralateral activation
on fMRI found in three of the four tested patients.5–7 The TMS and
fMRI findings in the present study indicate that crossed cerebrospinal
tracts were preserved in many of our patients with ipsilateral hemi-
paresis. On the other hand, uncrossed MEPs ipsilateral to the recent
lesion were evoked in only 28% of the present patients, although they
were evoked in all three of the reported patients.4,5,10 The frequency
of ipsilateral MEPs in the present study was lower than in healthy
subjects or acute-­stroke patients, where it is over 60%.19 A possible
explanation for the low frequency of uncrossed MEPs in the present
study might be that the areas with lesions responsible for the recent
ipsilateral stroke might have acquired strong innervation from the
non-­crossed corticospinal tract.
To further examine the pathophysiological mechanisms associated
with ipsilateral hemiparesis, we also divided the patients reported in
the literature and those in the present study into the three aforemen-
tioned types of possible injuries. In Type I, the absence of crossed
corticospinal tracts should be neurophysiologically confirmed. In the
reported studies, three patients4,5,10 (17.7%) were compatible with
Type I injuries, while none of our patients satisfied these criteria. In
Type II, neurophysiological evidence of crossed corticospinal tracts
and no history of injury of the crossed tract are needed. None of the
patients reported in the literature fulfilled these criteria, and only pa-
tient 12 (7.1%) in the present study was compatible with Type II. In
Type IIa, crossed corticospinal tracts should be neurophysiologically
evident, and the crossed tracts should have been injured prior to the
recent damage of the uncrossed tract. In the reported studies, three
Type IIa
㽢2
(+)
F I G U R E   2   Three possible
pathophysiological mechanisms of
(+)
ipsilateral hemiparesis
INATOMI et al
patients6,7 (17.7%) were compatible with Type IIa, and in our study,
patient 3, 4,7-­11, 13, and 14 (64.3%) satisfied these criteria. In the
other five (35.7%) present and 11 reported (64.7%) patients, it is dif-
ficult to categorize the injuries because of incomplete examination
or the injuries did not conform to any of the three types (e.g., our
patient 8).
The frequency of ipsilateral hemiparesis among patients with isch-
emic stroke was only 0.17%. In contrast, ipsilateral MEPs have been
detected in over 60% of tested patients with ischemic stroke.19–21 To
resolve this discrepancy between frequency of ipsilateral hemipare-
sis and findings of functional neuroimaging studies, we propose pro-
spective studies in which patients are examined by serial functional
neuroimaging studies from the time of their first stroke until their
recurrence.
In the present study, TMS and fMRI were not performed in all of
the patients. Patients with high premorbid mRS were more frequently
in non-­TMS group and right hemiparesis more frequent in the non-­
fMRI group. Moreover, TMS was performed at various times after the
onset of symptoms (range 4–24 days). These differences between the
groups with or without functional neuroimaging might have generated
a bias: for example, functional neuroimaging was only performed in
patients with mild cognitive dysfunction. However, there were no dif-
ference in motor weakness caused by recent strokes between the two
groups; therefore, we suggest that the two groups did not have signifi-
cant differences in motor system dysfunctions. Finally, the fMRIs were
not analyzed with Z-­scores and mapping.7 A Z-­score analysis might
improve the reliability of fMRI findings in the present study.
In conclusion, we studied ipsilateral hemiparesis in ischemic stroke
using functional neuroimaging studies. The majority of these patients
were classified as having injuries consistent with Type IIa ipsilateral
hemiparesis, in which patients had a past history of stroke contralat-
eral to the recent stroke and ordinary contralateral motor deficits, and
both the crossed and uncrossed corticospinal tracts were neurophys-
iologically detected. These findings suggest that Type IIa ipsilateral
hemiparesis might indicate actual contribution of the uncrossed corti-
cospinal tract contralateral to stroke lesions as a post-­stroke compen-
satory motor system.
ACKNOWLE DGME N TS
We thank Dr. Kaoru Matsunaga (Kumamoto Kinoh Hospital, Onjaku
Hospital) for conducting the transcranial magnetic stimulations.
CO NFLI CT OF I NTERE STS
The authors have no conflict of interests to declare. No targeted fund-
ing reported.
AUT HOR CONTRI BUTI O N S
YI helped in acquisition of data, analysis and interpretation, and writ-
ing the manuscript; MN carried out critical revision of the manuscript;
TY contributed to critical revision of the manuscript; and YA helped
|
      9
in critical revision of the manuscript. All authors approval of the final
manuscript.
REFERENCES
1. Miller Fisher C. Concerning the mechanism of recovery in stroke
hemiplegia. Can J Neurol Sci. 1992;19:57–63.
2. Schneider RC, Crosby E. Ipsilateral symptoms caused by an arteriove-
nous malformation of the second or supplementary sensory area of
the island of Reil. Neurosurgery. 1983;12:557–560.
3. Kudo T, Uno T. Ipsilateral hemiparesis caused by subarachnoid hem-
orrhage in a patient with a ruptured middle cerebral artery aneurysm:
a case report. Neurosurgery. 1984;15:727–729.
4. Hosokawa S, Tsuji S, Uozumi T, Matsunaga K, Toda K, Ota S. Ipsilateral
hemiplegia caused by right internal capsule and thalamic hemorrhage.
Neurology. 1996;46:1146–1149.
5. Terakawa H, Abe K, Nakamura M, Okazaki T, Obashi J, Yanagihara T.
Ipsilateral hemiparesis after putaminal hemorrhage due to uncrossed
pyramidal tract. Neurology. 2000;54:1801–1805.
6. Ago T, Kitazono T, Ooboshi H, et  al. Deterioration of pre-­existing
hemiparesis brought about by subsequent ipsilateral lacunar infarc-
tion. J Neurol Neurosurg Psychiatry. 2003;74:1152–1153.
7. Song YM, Lee JY, Park JM, Yoon BW, Roh JK. Ipsilateral hemiparesis
caused by a corona radiata infarct after a previous stroke on the op-
posite side. Arch Neurol. 2005;62:809–811.
8. Yamamoto S, Takasawa M, Kajiyama K, Baron JC, Yamaguchi T.
Deterioration of hemiparesis after recurrent stroke in the unaf-
fected hemisphere: three further cases with possible interpretation.
Cerebrovasc Dis. 2007;23:35–39.
9. Ng AS, Sitoh YY, Zhao Y, Teng EW, Tan EK, Tan LC. Ipsilateral stroke in
a patient with horizontal gaze palsy with progressive scoliosis and a
subcortical infarct. Stroke. 2011;42:e1–e3.
10. Kang K, Choi NC. Ipsilateral hemiparesis and spontaneous hor-
izontal nystagmus caused by middle cerebral artery territory in-
farct in a patient with agenesis of the corpus callosum. Neurol Sci.
2012;33:1165–1168.
11. Alurkar A, Karanam LSP, Atre A, Nirhale S, Nayak S, Oak S. Ipsilateral
stroke with uncrossed pyramidal tracts and underlying right internal
carotid artery stenosis treated with percutaneous transluminal angio-
plasty and stenting. A rare case report and review of the literature.
Neuroradiol J. 2012;25:237–242.
12. Saada F, Antonios N. Existence of ipsilateral hemiparesis in ischemic
and hemorrhagic stroke: two case reports and review of the literature.
Eur Neurol. 2014;71:25–31.
13. Yamada S, Okita Y, Shofuda T, et al. Ipsilateral hemiparesis caused by
putaminal hemorrhage in a patient with horizontal gaze palsy with
progressive scoliosis: a case report. BMC Neurol. 2015;15:25.
14. Rossini PM, Calautti C, Pauri F, Baron JC. Post-­stroke plastic reorgan-
isation in the adult brain. Lancet Neurol. 2003;2:493–502.
15. Langhorne P, Coupar F, Pollock A. Motor recovery after stroke: a sys-
tematic review. Lancet Neurol. 2009;8:741–754.
16. Andrews RJ. Transhemispheric diaschisis. A review and comment.
Stroke. 1991;22:943–949.
17. Bakar M, Kirshner HS, Wertz RT. Crossed aphasia. Functional brain
imaging with PET or SPECT. Arch Neurol. 1996;53:1026–1032.
18. Calautti C, Baron JC. Functional neuroimaging studies of motor recov-
ery after stroke in adults: a review. Stroke. 2003;34:1553–1566.
19. Alagona G, Delvaux V, Gérard P, et al. Ipsilateral motor responses to
focal transcranial magnetic stimulation in healthy subjects and acute-­
stroke patients. Stroke. 2001;32:1304–1309.
20. Schwerin S, Dewald JP, Haztl M, Jovanovich S, Nickeas M, MacKinnon
C. Ipsilateral versus contralateral cortical motor projections to a
shoulder adductor in chronic hemiparetic stroke: implications for the
expression of arm synergies. Exp Brain Res. 2008;185:509–519.
 |
10       INATOMI et al

21. Stinear CM, Barber PA, Smale PR, Coxon JP, Fleming MK, Byblow
WD. Functional potential in chronic stroke patients depends on corti-
cospinal tract integrity. Brain. 2007;130:170–180.
22. Favre I, Zeffiro TA, Detante O, Krainik A, Hommel M, Jaillard A. Upper
limb recovery after stroke is associated with ipsilesional primary
motor cortical activity: a meta-­analysis. Stroke. 2014;45:1077–1083.
23. Oshino S, Kato A, Hirata M, et  al. Ipsilateral motor-­related hyper-
activity in patients with cerebral occlusive vascular disease. Stroke.
2008;39:2769–2775.
24. Liu J, Qin W, Zhang J, Zhang X, Yu C. Enhanced interhemispheric func-
tional connectivity compensates for anatomical connection damages
in subcortical stroke. Stroke. 2015;46:1045–1051.
25. Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR signal
abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. Am
J Roentgenol. 1987;149:351–356.
26. Davidoff RA. The pyramidal tract. Neurology. 1990;40:332–339.
27. Han BS, Hong JH, Hong C, et al. Location of the corticospinal tract at
the corona radiata in human brain. Brain Res. 2010;1326:75–80.
How to cite this article: Inatomi, Y., Nakajima, M., Yonehara, T. and
Ando, Y. (2016), Ipsilateral hemiparesis in ischemic stroke patients.
Acta Neurologica Scandinavica, 00: 1–10. doi: 10.1111/ane.12690