Clinical Neurology and Neurosurgery 112 (2010) 406–412

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery

journal homepage: www.elsevier.com/locate/clineuro

Impaired ability to shift weight onto the non-paretic leg in right-cortical

brain-damaged patients
Fumiyasu Ishii a,b , Noriyuki Matsukawa a,∗ , Mitsuya Horiba b , Takehiko Yamanaka c , Manabu Hattori a ,
Ikuo Wada b , Kosei Ojika a
a
Department of Neurology, Nagoya City University Medical School, Mizuho-ku, Nagoya 467-8601, Japan
b
Division of Rehabilitation Medicine, Nagoya City University Medical School, Mizuho-ku, Nagoya 467-8601, Japan
c
Department of Information Technology for Human Welfare, Nihon Fukushi University, Toumi-cho, Handa 475-0012, Japan

a r t i c l e i n f o a b s t r a c t

Article history: Background: : Stroke patients experience postural instability that can impede functional improvements in
Received 10 November 2008 their gait. However, the precise functions of the dominant and non-dominant hemispheres in controlling
Received in revised form 2 February 2010 static standing posture and weight-bearing remain unclear.
Accepted 7 February 2010
Objective: : To investigate differences in balancing ability between right-handed patients with right and
Available online 12 March 2010
left hemispheric lesions.
Methods: : Weight shifting was quantitatively evaluated to determine the ability of patients to control
Keywords:
their balance in a static posture and during conscious weight shifting onto the paretic or non-paretic leg.
Instability
Weight shift
Participants were enrolled from a consecutive series of stroke patients attending a rehabilitation program
Stroke (n = 49; 31 male, 18 female; mean age 69.3 ± 9.4 years). Age-matched normal controls were recruited as
Right hemisphere volunteers (n = 12; 4 male, 8 female; mean age 67.9 ± 4.9 years).
Anosognosia Results: : Patients with cortical lesions in the right hemisphere were able to shift less weight onto the non-
paretic leg than patients with cortical lesions in the left hemisphere (p < 0.05). There were no correlations
between the existence of unilateral spatial neglect and the percentage of weight shifted onto the non-
paretic leg, static standing posture (r = 0.27, p = 0.40) or dynamic standing posture (r = −0.37, p = 0.24). In
contrast, there was a significant correlation between the percentage of weight consciously shifted onto
the non-paretic leg and the existence of anosognosia (r = 0.74, p = 0.006), but not between static standing
posture and anosognosia (r = −0.15, p = 0.63).
Conclusion: : Patients with right cortical hemispheric lesions were able to shift less body weight onto
their non-paretic leg. These patients should be encouraged to practice shifting their weight towards their
non-paretic leg to improve their balance.
© 2010 Elsevier B.V. All rights reserved.

1. Introduction synergistic organization of movements, muscle weakness, sensory

Stroke is one of the commonest causes of death in developed
countries. Although therapeutic advances have reduced the mortal-
ity during the acute stages of stroke, patients still experience severe
neurological disabilities [1]. Deficits in motor control, abnormal
Abbreviations: LI, left hemispheric infarction in territory of middle cerebral
artery; RI, right hemispheric infarction in territory of middle cerebral artery; LCI, left
hemispheric infarction in territory of middle cerebral cortical artery; RCI, right hemi-
spheric infarction in territory of middle cerebral cortical artery; LPI, left hemispheric
infarction in territory of middle cerebral perforating artery; RPI, right hemispheric
infarction in territory of middle cerebral perforating artery; NR, normal control sub-
jects through right leg; NL, normal control subjects through left leg; 10s, 20s and
60s, 10, 20 and 60 s after starting to keep static position.
∗ Corresponding author. Tel.: +81 52 853 8094; fax: +81 52 852 3590.
E-mail address: norim@med.nagoya-cu.ac.jp (N. Matsukawa).
deficits, and loss of range of motion can all reduce the quality of life
in stroke patients [2,3]. The reluctance to bear weight on the paretic
leg may persist and cause consequent gait disturbances, despite
conventional physical therapy to correct asymmetric standing pos-
ture [4]. Moreover, an asymmetric posture could foster further
disuse and hinder the recovery of motor function in the lower limbs
during motor convalescence [5].
Some stroke patients have an unstable gait and a tendency to
repeatedly fall to the paretic side from an early stage [6]. These
patients include those who fall towards their paretic side and have
problems with balance because of their paresis [7,8].
In 1985, Davies reported patients who exhibited paradoxical
behavior as having “Pusher Syndrome”; when pushed from the
paretic side, these stroke patients used their non-paretic leg to
push toward the paretic side, leading to postural imbalance [9].
Karnath et al. suggested that the mismatch between the visual

0303-8467/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.clineuro.2010.02.006
 F. Ishii et al. / Clinical Neurology and Neurosurgery 112 (2010) 406–412
vertical and the tilted orientation of body verticality formed part of
the underlying mechanism of Pusher syndrome, and also suggested
that the posterolateral thalamus could be anatomically damaged in
patients with this syndrome [10,11]. Lafosse et al. also reported that
contraversive pushing after right brain damage may relate with a
spatial higher-order processing deficit underlying the higher fre-
quency and severity of the axial postural abnormalities [12]. Visual
cues that allow alignment of the body axis to the earth vertical may
be helpful aspects of physiotherapy for patients with this instability
[11,13]. Moreover, the imbalance in this syndrome is not necessar-
ily related to spatial neglect, anosognosia or apraxia [10,14].
Some patients with stroke, however, may show postural imbal-
ance while standing or sitting, despite the absence of Pusher
syndrome or severe weakness in the paretic leg. Visual cues do not
necessarily help to correct the postural imbalance in these patients,
and lesions are not limited to the posterolateral thalamus. However,
the mechanisms underlying this symptom remain unknown.
Patients with lesions in the right hemisphere have been found
to have a larger sway area and more lateral displacement than
patients with left hemispheric lesions, indicating the importance
of the right hemisphere in controlling stability during standing
and walking [15]. The right and left hemispheres may play dif-
ferent roles in controlling standing posture [15], but their precise
functions in controlling static standing posture and weight-bearing
remain unclear.
In this study, we investigated the involvement of the non-paretic
leg in postural imbalance in stroke patients who showed deficits in
balance and a tendency to fall towards their paretic side, despite the
absence of Pusher syndrome or severe weakness in the paretic leg.
We examined the percentage of body weight borne by the non-
paretic leg when standing still, and the maximum percentage of
body weight voluntarily shifted onto the paretic or non-paretic leg
in patients with right or left hemispheric lesions, including cortical
lesions and/or lesions in the territory of the perforating artery. We
also examined the association between poor balance on standing
and unilateral spatial neglect and/or anosognosia in patients with
right cortical lesions.
2. Methods
2.1. Participants
Japanese patients with clinically and neuroradiologically diag-
nosed stroke were recruited (n = 49; 31 male, 18 female; mean
age 69.3 ± 9.4 years). Participants were medical admission patients
from Nagoya City University Medical Center and were registered
as a consecutive series of right-handed stroke patients who were
attending a rehabilitation program. The inclusion criteria were: (i)
right-handed; (ii) being able to stand alone; (iii) neuroradiologically
verified single vascular lesion in the territory of the middle cere-
bral artery of one hemisphere, and (iv) within 10–30 days of onset
at the start of the study (Fig. 1). To eliminate potential effects of
varying degrees of neurological dysfunction, including motor, sen-
sory and cerebellar function, all the registered participants were
in Brunnstrom stage 5–6, with no neurological sensory or cerebel-
lar dysfunction [16]. Patients who were able to bend their knee
joint while maintaining an extended position of the hip joint were
included. Any patients who were unable to understand the study
were excluded. The vibration sense, joint position sense, pain sense,
touch sense and cerebellar ataxia were examined neurologically,
and patients with neurological sensory or cerebellar dysfunctions
were excluded. Patients with contraversive pushing were also
excluded on the basis of clinical assessment scale for contraversive
pushing [11]. To determine the involvement of the cortical region,
we separated each hemispheric group by magnetic resonance (MR)
407
Fig. 1. Enrolled participants.
Forty-nine participants were enrolled in this rehabilitation program, out of 128
consecutive stroke patients.
images and MR angiography into patients with infarction in the ter-
ritory of the cortical artery and those with infarction in the territory
of the perforating artery. The characteristics of the patients and age-
matched normal controls are summarized in Table 1. Age-matched
right-handed normal volunteers were recruited from the families
of patients (n = 12: 4 male, 8 female; mean age 67.9 ± 4.9 years).
We considered patients to be right-handed if they met all the cri-
teria of the Western Aphasia Battery [17]. Briefly, 27 patients had
left hemispheric lesions (17 male, 10 female; mean age 68.9 ± 8.8
years), and 22 patients had right hemispheric lesions (14 male, 8
female; mean age 69.1 ± 11.3 years). There was no significant dif-
ference in duration from stroke to time of examination between
patients with left hemispheric infarctions (LI) and those with right
hemispheric infarctions (RI) (LI, 15.8 ± 7.1 days; RI, 16.2 ± 7.6 days)
(unpaired samples Tukey’s post-hoc test; p = 0.73) (Table 1). There
was no significant difference in age or Brunnstrom stage between LI
and RI patients (unpaired samples Tukey’s post-hoc test; p = 0.363
and p = 0.994, respectively). Informed consent was received from all
participants prior to carrying out physical and neuropsychological
tests. This study was approved by the medical ethical committee of
Nagoya City University Graduate School of Medical Sciences.
2.2. Clinical assessment
We screened the patients with right hemispheric cortical lesions
(RCI) for the presence of unilateral spatial neglect using a standard-
ized test battery, including line crossing and line bisection, which
form part of the Behavioral Inattention Test (BIT) [18]. Anosognosia
was assessed using Bisiach’s scale [19].
2.3. Apparatus
Weight-bearing was assessed using F-scan Tactile Sensor
System computerized force platforms, as described previously
(Tekscan Inc., Boston, MA, United States). Briefly, the sensor was
composed of thin films (0.1 mm) that displayed electrical resistance
proportional to the physical forces applied. The system consisted of
a single platform connected to a personal computer allowing inde-
pendent real-time or recorded measurements of the vertical force
between the feet and the surface of the platform [20].
2.4. Procedures
Weight-bearing measurements were performed in three differ-
ent standing positions. In the first series, participants were asked to
stand with a level, shoulder-width stance, with eyes open (Fig. 2a),
and their weight-bearing distribution was recorded and calcu-
lated during a 0.1-ms period at 10, 30 and 60 s after starting the
static position. In the second series, participants were encour-
408 F. Ishii et al. / Clinical Neurology and Neurosurgery 112 (2010) 406–412

Table 1
Characteristics of the subjects.

RI (n = 22) LI (n = 27) Normal p value

Characteristic RCI RPI LCI LPI

Subjects (male/female) 12 (8/4) 10 (6/4) 5 (3/2) 22 (14/8) 12 (4/8)

Age (years) 68.6 ± 9.9 70.8 ± 12.1 73.8 ± 12.2 67.6 ± 7.4 67.9 ± 4.9 p = 0.363
Duration of illness (days) 15.8 ± 9.4 16.8 ± 6.5 17.2 ± 4.7 15.1 ± 7.9 p = 0.73

Brunnstrom stage
V 5 5 3 12 p = 0.994
VI 7 5 2 10

One-way analysis of variance (ANOVA) and the Turkey were used to compare means of both age and duration of illness among five groups.
After analyzing the equalities of variance using an F-test, the sampling distributions of Brunnstrom stage in the two groups, RI and LI, were analyzed using Pearson’s chi-
squared tests.
RI = right hemispheric infarction; LI = light hemispheric infarction; RCI = right hemispheric infarction in territory of middle cerebral cortical artery; RPI = right hemispheric
infarction in territory of middle cerebral perforating artery; LCI = light hemispheric infarction in territory of middle cerebral cortical artery; LPI = light hemispheric infarction
in territory of middle cerebral perforating artery.

aged to shift as much weight as possible onto the non-paretic leg,
while the paretic leg remained on the platform. The percentage of
forced weight was measured as for the static standing measure-
ment (Fig. 2b). In the third series, participants were encouraged to
bear maximal weight on the paretic leg and maintain this forced
position, and the percentage of weight-bearing was calculated as
described above (Fig. 2c). Rest periods of 2–3 min were allowed
between each position. Testing in the three positions was repeated
three times on the same day, and the mean results of the three
measurements in each position were analyzed statistically.
2.5. Statistical analysis
After analyzing the equalities of variance using an F-test, the
sampling distributions of Brunnstrom stage between the two
groups, RI and LI, were analyzed using Pearson’s chi-squared tests.
Data were analyzed by one-way ANOVA to identify any significant
differences in characteristics of patients, age and duration of illness,
and percentage of weight shifting through the paretic or non-
paretic leg among the groups. Parametric comparisons between
the groups were carried out using Tukey’s post-hoc tests. To evalu-
ate the significance of neuropsychological impairment, Spearman’s
rank correlation was calculated to analyze the relationship between
unilateral spatial neglect or anosognosia, and the average percent-
age of weight shifted onto each leg in nine measurement times.
Statistical significance was set at p < 0.05.
3. Results
All participants with stroke demonstrated asymmetrical
weight-bearing during static posture. Participants with LI bore
>50% of their weight on the non-paretic leg in static standing pos-
ture (Table 2a), while participants with RI bore most of their weight
on the paretic leg (Table 2a). The percentage of weight borne on
the non-paretic leg during static standing was significantly differ-
ent between patients with LI and those with RI (LI; 10 s 53.7 ± 3.5,
30 s 52.8 ± 4.5, 60 s 53.1 ± 4.9, RI; 10 s 48.8 ± 5.4, 30 s 49.6 ± 3.1, 60 s
48.4 ± 5.0) (unpaired samples Tukey’s post-hoc test: p < 0.05). Age-
matched control participants divided their weight equally between
each leg in the static position (NR; 10 s 50.1 ± 1.9, 30 s 50.8 ± 3.2,
60 s 50.4 ± 3.8, NL; 10 s 49.9 ± 1.9, 30 s 49.2 ± 3.2, 60 s 49.6 ± 3.8)
(Table 2a).
We also examined the maximum percentage of weight volun-
tarily shifted between the paretic and non-paretic legs. Patients
with LI were able to transfer significantly more of their body weight
to the non-paretic leg than those with RI (LI; 10 s 81.5 ± 8.6, 30 s
79.7 ± 10.2, 60 s 76.7 ± 13.3, RI; 10 s 70.4 ± 10.6, 30 s 68.7 ± 11.0,
60 s 64.9 ± 13.8) (unpaired samples Tukey’s post-hoc test: p < 0.05)
(Table 2b). However, there was no significant difference in the abil-
ity to shift weight to the paretic leg between the two groups (LI;
10 s 81.8 ± 7.8, 30 s 81.8 ± 7.7, 60 s 80.4 ± 9.4, RI; 10 s 78.8 ± 10.9,
30 s 77.4 ± 11.2, 60 s 77.7 ± 10.5) (Table 2b). Control participants
were able to shift >80% of their weight onto either leg (NR; 10 s

Fig. 2. Schematic presentation of static posture and maximum voluntary weight shift.
Subjects maintained the basic position during rest. They then shifted their weight maximally onto the non-paretic leg, and then changed the weight shift to the paretic leg.
The pressure measured through each leg was shown as a percentage of the total weight, measured by the foot scan computerized force platform system.
(a) Static position, (b) maximum voluntary weight shifting onto left leg, (c) maximum voluntary weight shifting onto right leg.
Table 2a
Percentage of weight sustained in static position in RI, LI, NR and NL groups (Tukey’s post-hoc test; p values).

Percentage sustained through non-paretic leg

10 s RI LI NR NL 30 s RI LI NR NL 60 s RI LI NR NL

RI (n = 22) 48.8 ± 5.4 – 0.001 0.858 0.800 RI 49.6 ± 3.1 – 0.026 0.986 0.810 RI 48.4 ± 5.0 – 0.005 0.629 0.894
LI (n = 28) 53.7 ± 3.5 – 0.024 0.033 LI 52.8 ± 4.5 – 0.037 0.461 LI 53.1 ± 4.9 – 0.360 0.146
NR (n = 12) 50.1 ± 1.9 – 1.000 NR 50.8 ± 3.2 – 0.698 NR 50.4 ± 3.8 – 0.973

F. Ishii et al. / Clinical Neurology and Neurosurgery 112 (2010) 406–412

NL (n = 12) 49.9 ± 1.9 – NL 49.2 ± 3.2 – NL 49.6 ± 3.8 –

Table 2b
Maximum voluntary percentage of weight sustained in RI, LI, NR and NL groups (Tukey’s post-hoc test; p values).

10 s RI LI NR NL 30 s RI LI NR NL 60 s RI LI NR NL

Percentage sustained through non-paretic leg

RI (n = 22) 70.4 ± 10.6 – 0.001 0.001 0.001 RI 68.7 ± 11.0 – 0.001 0.001 0.001 RI 64.9 ± 13.8 – 0.001 0.001 0.001
LI (n = 28) 81.5 ± 8.6 – 0.220 0.209 LI 79.7 ± 10.2 – 0.289 0.707 LI 76.7 ± 13.3 – 0.094 0.406
NR (n = 12) 88.1 ± 4.0 – 0.582 NR 86.8 ± 5.2 – 0.930 NR 87.5 ± 5.7 – 0.904
NL (n = 12) 88.1 ± 5.9 – NL 84.6 ± 7.8 – NL 84.4 ± 7.6 –

Percentage sustained through paretic leg

RI (n = 22) 78.8 ± 10.9 – 0.833 0.021 0.019 RI 77.4 ± 11.2 – 0.455 0.026 0.127 RI 77.7 ± 10.5 – 0.938 0.033 0.232
LI (n = 28) 81.8 ± 7.8 – 0.091 0.086 LI 81.8 ± 7.7 – 0.300 0.707 LI 80.4 ± 9.4 – 0.085 0.444
NR (n = 12) 88.1 ± 4.0 – 1.000 NR 86.8 ± 5.2 – 0.936 NR 87.5 ± 5.7 – 0.866
NL (n = 12) 88.1 ± 5.9 – NL 84.6 ± 7.8 – NL 84.4 ± 7.6 –

409
410 F. Ishii et al. / Clinical Neurology and Neurosurgery 112 (2010) 406–412

88.1 ± 4.0, 30 s 86.8 ± 5.2, 60 s 87.5 ± 5.7, NL; 10 s 88.1 ± 5.9, 30 s

0.675
0.985

0.956
0.997
0.007
84.6 ± 7.8, 60 s 84.4 ± 7.6) (Table 2b).

NL

–
To determine the involvement of the cortical region, we com-
pared weight-bearing in relation to the infarct-related artery. We

0.909

0.739
1.000
0.002
NR
separated each group into patients with infarction in the territory

–
of the cortical artery and those with infarction in the territory of

0.126
0.636
0.019
the perforating artery. Patients with a left hemispheric infarction

LPI

–
in the territory of middle cerebral cortical artery (LCI) put >50% of
their body weight on the non-paretic leg when standing still, while
0.001
0.002
LCI
patients with a right hemispheric infarction in the territory on mid-

–
dle cerebral cortical artery (RCI) put <50% of their weight on the
0.974

non-paretic leg (LCI; 10 s 55.7 ± 3.9, 30 s 56.2 ± 3.4, 60 s 58.3 ± 3.7,

RPI

RCI; 10 s 47.4 ± 5.8, 30 s 48.6 ± 3.5, 60 s 47.7 ± 4.5) (unpaired sam-

ples Tukey’s post-hoc test: p < 0.05) (Table 3a). Conversely, patients
RCI

with perforating artery infarctions were able to distribute their

–

weight evenly between both legs. There was no significant differ-

ence in percentage of weight on the non-paretic leg in the static
RCI 47.7 ± 4.5
RPI 49.2 ± 5.6
LCI 58.3 ± 3.7
LPI 51.5 ± 3.7
NR 50.4 ± 3.8
NL 49.6 ± 3.8

posture between patients with left hemispheric infarction in the

territory of the middle cerebral perforating artery (LPI) and those
with right hemispheric infarction in the territory of the middle
60 s

cerebral perforating artery (RPI) (LPI; 10 s 52.9 ± 3.1, 30 s 51.6 ± 3.6,

60 s 51.5 ± 3.7, RPI; 10 s 50.5 ± 4.9, 30 s 50.9 ± 2.1, 60 s 49.2 ± 5.6)
0.647

0.965
0.861
1.000
0.038

(Table 3a).
NL

We also examined the maximum percentage of weight volun-

–

tarily shifted between the paretic and non-paretic legs. Patients

0.999
0.874

0.294
0.003

with RCI were able to shift less weight onto their non-paretic leg,
NR

compared with those with LCI (LCI; 10 s 81.8 ± 7.7, 30 s 81.8 ± 9.0,
0.125
0.976
0.094

60 s 80.7 ± 11.1, RCI; 10 s 66.9 ± 9.5, 30 s 64.8 ± 10.8, 60 s 58.4 ± 7.6)

LPI

(unpaired samples Tukey’s post-hoc test: p < 0.05) (Table 3b). There
was no significant difference in maximum voluntary weight shift-
0.051
0.001
LCI

ing onto the non-paretic leg between the LPI and RPI groups
–
Percentage of weight sustained in static position in RCI, RPI, LCI, LPI, NR and NL groups (Tukey’s post-hoc test; p values).

(LPI; 10 s 81.6 ± 9.0, 30 s 79.2 ± 10.6, 60 s 75.6 ± 13.8, RPI; 10 s

0.678

74.7 ± 10.8, 30 s 73.4 ± 9.9, 60 s 72.6 ± 10.0) (Table 3b).

RPI

The four groups demonstrated similar maximum volun-

–

tary weight shifting onto the paretic leg (LCI; 10 s 81.0 ± 7.6,
30 s 77.8 ± 10.1, 60 s 72.0 ± 16.4, RCI; 10 s 78.9 ± 10.4, 30 s
RCI

76.6 ± 10.5, 60 s 76.1 ± 10.7, LPI; 10 s 82.1 ± 8.1, 30 s 82.8 ± 6.8, 60 s

82.3 ± 6.9, RPI; 10 s 78.9 ± 12.1, 30 s 78.3 ± 12.6, 60 s 79.7 ± 10.6)
RCI 48.6 ± 3.5
RPI 50.9 ± 2.1
LCI 56.2 ± 3.4
LPI 51.6 ± 3.6
NR 50.8 ± 3.2
NL 49.2 ± 3.2

(Table 3b).
We performed neuropsychological tests in 12 patients with
RCI to clarify the role of unilateral spatial neglect or anosognosia.
30 s

These included line bisection and line crossing, which form part
of the BIT battery, and Bisiach’s anosognosia scale. The results in
seven of the 12 were indicative of unilateral spatial neglect, with
0.241
0.500
1.000

1.000
0.004
NL

cut-off points below seven for the line bisection score and 34 for
–

the line crossing score in the BIT Japanese Version. There was
0.999
0.570

0.190
0.003

no correlation between the existence of unilateral spatial neglect

NR

and the percentage of weight shifted onto the non-paretic leg,

static standing posture (paired samples Spearman’s rank correla-
0.461
0.155
0.001
LPI

tion: r = 0.27, p = 0.40) or dynamic standing posture (paired samples

–

Spearman’s rank correlation: r = −0.37, p = 0.24). Moreover, eight

0.001
0.011

of the 12 patients also showed neurological symptoms of anosog-

LCI
Percentage sustained through non-paretic leg

nosia. There was a significant correlation between the percentage

of weight shifted onto the non-paretic leg and the existence of
0.400
RPI

anosognosia (paired samples Spearman’s rank correlation: r = 0.74,

–

p = 0.006), but not between static posture and anosognosia (paired

samples Spearman’s rank correlation: r = −0.15, p = 0.63). To con-
RCI

firm the involvement of anosognosia in instability of RCI patients,

we divided patients into two groups, eight RCI patients with
RCI (n = 12) 47.4 ± 5.8
RPI (n = 10) 50.5 ± 4.9

LPI (n = 22) 52.9 ± 3.1

NR (n = 12) 50.1 ± 1.9
NL (n = 12) 49.9 ± 1.9

anosognosia and four RCI patients without anosognosia. And we

LCI (n = 5) 55.7 ± 3.9

re-calculated the ability of their weight shifting onto non-paretic

leg at each period of 10 s, 30 s and 60 s in both static and dynamic
postures. There was no significant difference among RCI with-
out anosognosia, LCI, RPI and LPI, whereas RCI with anosognosia
Table 3a

10 s

showed even significant difference compared with LCI, RPI and LPI
(data not shown).
 F. Ishii et al. / Clinical Neurology and Neurosurgery 112 (2010) 406–412 411

4. Discussion

0.384
0.978

0.271

0.946
0.967
0.100

0.850
0.073
1.000
0.001
NL

–
In this study, we found that all stroke patients adopted

0.985
0.075

0.394

0.471
0.001
0.017

0.047

0.014
asymmetrical weight-bearing when standing still with their feet

NR
shoulder-width and eyes open. However, the percentage of body

–
weight shifted to the non-paretic leg differed significantly, depend-

0.862
0.790

0.628
0.997
0.193
0.001
ing on the side of the lesion. Patients with LI put >50% of their weight
LPI

–
on their non-paretic leg, while those with RI put <50% of their
0.376

0.869
0.479
0.001
weight on their non-paretic leg. There was also a significant dif-
LCI

–
ference in the maximum percentage of weight voluntarily shifted
onto the non-paretic leg between patients with LI and those with

0.948
0.025
RPI

RI. Patients with RCI were less able to consciously shift weight onto
–

–
their non-paretic leg, compared with those with LCI. This indicates
the importance of right cortical function in controlling stability
RCI

–
during standing, and in consciously shifting weight to the non-
paretic leg. This instability could be associated with the presence
RCI 76.1 ± 10.7
RPI 72.6 ± 10.0

RPI 79.7 ± 10.6

LCI 80.7 ± 11.1

LCI 72.0 ± 16.4

LPI 75.6 ± 13.8

of anosognosia, rather than unilateral spatial neglect.

RCI 58.4 ± 7.6

LPI 82.3 ± 6.9

NR 87.5 ± 5.7

NR 87.5 ± 5.7
NL 84.4 ± 7.6

NL 84.4 ± 7.6
The ability to maintain balance while standing still is controlled
by interplay between the motor, sensory and cognitive systems
60 s

[21]. Impairment of the mechanism responsible for postural con-

trol can lead to unsteadiness and instability during standing [22].
In this study, we only recruited patients who were able to stand
0.767
0.989

0.275
0.594

0.992
0.703
0.970
1.000
0.001
0.034

independently, without the use of an assistive device, to exclude

NL

any possible effects of differences in motor or sensory dysfunction;

0.998
0.329

0.267
0.082

0.420
0.701

there were no significant differences in motor or sensory function

0.001
0.006
NR

scores among the evaluated groups.

–

Our results also showed that patients with RCI bore more weight
0.249
0.911

0.559

0.922
0.890
0.001

on their paretic leg when standing still, compared with those with
LPI

LCI. In contrast, patients with RPI and LPI showed similar dis-
0.153

1.000
1.000
0.001

tributions of body weight on the paretic leg. This suggests that

LCI
Maximum voluntary percentage of weight sustained in RCI, RPI, LCI, LPI, NR and NL groups (Tukey’s post-hoc test; p values).

the cortical area of the right hemisphere could be important for

unconsciously shifting body weight onto the paretic leg, leading to
0.180

0.998
RPI

instability during standing.

–

Recent reports have emphasized the important role of spa-

tial perception disorders, such as hemi-neglect, in imbalance after
RCI

stroke [23]. Instability is more frequent, and takes longer to recover

from following right hemispheric lesions compared with left hemi-
RCI 64.8 ± 10.8

RCI 76.6 ± 10.5

RPI 78.3 ± 12.6
LCI 77.8 ± 10.1
LPI 79.2 ± 10.6

spheric lesions [24]. In the current study, there was no correlation

RPI 73.4 ± 9.9
LCI 81.8 ± 9.0

LPI 82.8 ± 6.8

NR 86.8 ± 5.2

NR 86.8 ± 5.2
NL 84.6 ± 7.8

NL 84.6 ± 7.8

between the degree of instability and the existence of unilateral

spatial neglect, indicating that the asymmetrical weight-bearing
30 s

in our patients with RCI was independent of unilateral spatial

neglect.
Anosognosia is one of the neurological symptoms found in
0.955
0.294

0.148
0.110

0.695
0.200
1.000

1.000
0.001
0.002

patients with RCI; patients may deny the existence of a deficit

NL

and claim that their paralyzed limbs can still move, despite obvi-
ous motor impairment. It has been suggested that this symptom
0.959

0.116
0.309

0.156

0.211
0.706
0.001
0.002
NR

is caused by deficits in parts of the sensory-motor or cognitive

–

systems, and that the causal lesion involves the frontal cortex,
0.123

0.987
0.990

0.992
1.000
0.001

including areas 6, 44 and the insula [25]. Previous reports have

LPI

indicated that stroke patients with anosognosia showed instability

0.197

0.994
0.995
0.205 0.001

when standing still, and swayed during sitting, indicative of Pusher

LCI

syndrome. This lack of static and dynamic balance is thought to be

–

–
Percentage sustained through non-paretic leg

due to transfer of weight-bearing onto the paretic leg [14,26]. Ten

1.000
RPI

out of 12 of our participants with RCI had lesions in the frontal lobe,
Percentage sustained through paretic leg
–

and eight of 12 also had symptoms of anosognosia (Table 4).

In this study, we demonstrated that patients with RCI had
RCI

reduced abilities to shift their weight onto their non-paretic leg.

A previous report by Goldie et al. reported that stroke patients
RCI (n = 12) 78.9 ± 10.4
RPI (n = 10) 74.7 ± 10.8

RPI (n = 10) 78.9 ± 12.1

were able to transfer less weight to their non-paretic leg than aged-
RCI (n = 12) 66.9 ± 9.5

LPI (n = 22) 82.1 ± 8.1

LPI (n = 22) 81.6 ± 9.0
NR (n = 12) 88.1 ± 4.0

NR (n = 12) 88.1 ± 4.0

NL (n = 12) 88.1 ± 5.9

NL (n = 12) 88.1 ± 5.9

LCI (n = 5) 81.8 ± 7.7

LCI (n = 5) 81.0 ± 7.6

matched controls (control subjects, 95%; stroke patients, 80–85%)

[27]. Our results suggest that the instability of patients with RCI
could be due to this reduction in the conscious ability to shift
weight, as well as an unconscious inclination toward the paretic
leg. The results of this study suggest that physical therapists should
Table 3b

10 s

encourage patients to consciously shift their weight onto their

non-paretic leg, and to avoid unconscious shifting of weight onto
412 F. Ishii et al. / Clinical Neurology and Neurosurgery 112 (2010) 406–412

Table 4
Characteristics of the subjects with right hemispheric infarction in the territory of the cortical artery.

Cognition Weight Balance

Case Age, years USN Anosognosia Static (%) non-paretic leg Shift non-paretic leg (%) Infarction region

A 60 (+) (+) 39 64 Frontal lobe

B 77 (−) (+) 47 62 Frontal and temporal lobe
C 73 (−) (−) 51 85 Frontal and parietal lobe
D 76 (+) (−) 46 70 Parietal and occipital lobe
E 62 (+) (+) 52 63 Frontal lobe
F 57 (−) (+) 50 45 Frontal and parietal lobe
G 55 (−) (+) 53 55 Frontal and parietal lobe
H 70 (+) (+) 53 54 Frontal and parietal lobe
I 77 (−) (+) 48 61 Frontal and parietal lobe
J 58 (+) (−) 43 63 Frontal and parietal lobe
K 72 (+) (−) 50 77 Occipital lobe
L 78 (+) (+) 44 63 Frontal and parietal lobe

The numbers in either static or shift onto non-paretic leg show the average percentage in nine measurement times as representative value.
USN: unilateral spatial neglect (cut-off point: line bisection 7 and line crossing 34, respectively).

the paretic leg. This could improve balance during standing and
walking in patients with cortical damage to the non-dominant
hemisphere. Further studies, including functional image, will be
needed to clarify the mechanism of instability, why patients with
anosognosia could not well shift onto non-paretic leg, and to
improve rehabilitation method for patients with instability.
5. Conclusions
We determined that patients with RCI were able to shift less
weight onto their non-paretic leg, compared with those with LCI,
while there was no significant difference in terms of their abilities
to shift weight onto their non-paretic leg between patients with RPI
and those with LPI. We also revealed that the presence of anosog-
nosia, but not unilateral spatial neglect, was correlated with the
percentage of weight shifted onto the non-paretic leg in patients
with RCI.
References
[1] Goldstein LB. Acute ischemic stroke treatment in 2007. Circulation
2007;116(13):1504–14.
[2] Indredavik B, Bakke F, Slordahl SA, Rokseth R, Haheim LL. Stroke unit treat-
ment improves long-term quality of life: a randomized controlled trial. Stroke
1998;29(5):895–9.
[3] Ones K, Yilmaz E, Cetinkaya B, Caglar N. Quality of life for patients poststroke
and the factors affecting it. J Stroke Cerebrovasc Dis 2005;14(6):261–6.
[4] Aruin AS, Hanke T, Chaudhuri G, Harvey R, Rao N. Compelled weightbearing
in persons with hemiparesis following stroke: the effect of a lift insert and
goal-directed balance exercise. J Rehabil Res Dev 2000;37(1):65–72.
[5] Garland SJ, Willems DA, Ivanova TD, Miller KJ. Recovery of standing balance and
functional mobility after stroke. Arch Phys Med Rehabil 2003;84(12):1753–9.
[6] Nyberg L, Gustafson Y. Patient falls in stroke rehabilitation. A challenge to
rehabilitation strategies. Stroke 1995;26(5):838–42.
[7] Bohannon RW, Smith MB, Larkin PA. Relationship between independent sitting
balance and side of hemiparesis. Phys Ther 1986;66(6):944–5.
[8] Dettmann MA, Linder MT, Sepic SB. Relationships among walking performance,
postural stability, and functional assessments of the hemiplegic patient. Am J
Phys Med 1987;66(2):77–90.
[9] Davies PM. Steps to follow: a guide to the treatment of adult hemiplegia. New
York, NY: Sprinder; 1985.
[10] Karnath HO, Ferber S, Dichgans J. The neural representation of postural control
in humans. Proc Natl Acad Sci USA 2000;97(25):13931–6.
[11] Karnath HO, Ferber S, Dichgans J. The origin of contraversive pushing: evidence
for a second graviceptive system in humans. Neurology 2000;55(9):1298–304.
[12] Lafosse C, Kerckhofs E, Vereeck L, Troch M, Van Hoydonck G, Moeremans M,
et al. Postural abnormalities and contraversive pushing following right hemi-
sphere brain damage. Neuropsychol Rehabil 2007;17(3):374–96.
[13] Karnath HO, Johannsen L, Broetz D, Ferber S, Dichgans J. Prognosis of contraver-
sive pushing. J Neurol 2002;249(9):1250–3.
[14] Pedersen PM, Wandel A, Jorgensen HS, Nakayama H, Raaschou HO, Olsen TS.
Ipsilateral pushing in stroke: incidence, relation to neuropsychological symp-
toms, and impact on rehabilitation. The Copenhagen stroke study. Arch Phys
Med Rehabil 1996;77(1):25–8.
[15] Spinazzola L, Cubelli R, Della Sala S. Impairments of trunk movements follow-
ing left or right hemisphere lesions: dissociation between apraxic errors and
postural instability. Brain 2003;126(Pt 12):2656–66.
[16] Brunnstrom S. Motor testing procedures in hemiplegia: based on sequential
recovery stages. Phys Ther 1966;46(4):357–75.
[17] Shewan CM, Kertesz A. Reliability and validity characteristics of the Western
Aphasia Battery (WAB). J Speech Hear Disord 1980;45(3):308–24.
[18] Hartman-Maeir A, Katz N. Validity of the behavioral inattention test (BIT):
relationships with functional tasks. Am J Occup Ther 1995;49(6):507–16.
[19] Bisiach E, Vallar G, Perani D, Papagno C, Berti A. Unawareness of disease
following lesions of the right hemisphere: anosognosia for hemiplegia and
anosognosia for hemianopia. Neuropsychologia 1986;24(4):471–82.
[20] Randolph AL, Nelson M, Akkapeddi S, Levin A, Alexandrescu R. Reliability of
measurements of pressures applied on the foot during walking by a comput-
erized insole sensor system. Arch Phys Med Rehabil 2000;81(5):573–8.
[21] Jacobs JV, Horak FB. Cortical control of postural responses. J Neural Transm
2007;114(10):1339–48.
[22] Dickstein R, Abulaffio N. Postural sway of the affected and nonaffected
pelvis and leg in stance of hemiparetic patients. Arch Phys Med Rehabil
2000;81(3):364–7.
[23] Rode G, Tiliket C, Charlopain P, Boisson D. Postural asymmetry reduction by
vestibular caloric stimulation in left hemiparetic patients. Scand J Rehabil Med
1998;30(1):9–14.
[24] Laufer Y, Sivan D, Schwarzmann R, Sprecher E. Standing balance and func-
tional recovery of patients with right and left hemiparesis in the early stages
of rehabilitation. Neurorehabil Neural Repair 2003;17(4):207–13.
[25] Berti A, Bottini G, Gandola M, Pia L, Smania N, Stracciari A, et al.
Shared cortical anatomy for motor awareness and motor control. Science
2005;309(5733):488–91.
[26] Hartman-Maeir A, Soroker N, Katz N. Anosognosia for hemiplegia in stroke
rehabilitation. Neurorehabil Neural Repair 2001;15(3):213–22.
[27] Goldie PA, Matyas TA, Evans OM, Galea M, Bach TM. Maximum voluntary
weight-bearing by the affected and unaffected legs in standing following stroke.
Clin Biomech (Bristol, Avon) 1996;11(6):333–42.