Gait & Posture 33 (2011) 102107

Contents lists available at ScienceDirect

Gait & Posture

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

Effects of an ankle-foot orthosis with oil damper on muscle activity in adults

after stroke
Koji Ohata a,*, Tadashi Yasui b, Tadao Tsuboyama a, Noriaki Ichihashi a
a
b

Human Health Sciences, Graduate School of Medicine, Kyoto University, Kyoto, Japan
R&D Division, Kawamuragishi Co., Ltd, Osaka, Japan

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 9 June 2010
Received in revised form 4 October 2010
Accepted 12 October 2010

Background and objective: An ankle-foot orthosis with an oil damper (AFO-OD) was developed to resist
plantarexion motion, thereby improving hemiplegic gait performance. The purpose of this study was to
determine the effect of AFO-OD on muscle activity during the gait cycle in individuals affected by stroke.
Methods: Electromyography (EMG) was used to assess gait at a self-selected speed while wearing an
AFO-OD or an AFO with a plantarexion stop (AFO-PS) worn on the affected side in 11 stroke survivors
and on the right side in 11 age-matched healthy adults. EMG signals were obtained from the tibialis
anterior (TA), gastrocnemius (GAS), and soleus (SOL) muscles. In addition, the ankle joint angle under
both braces and the plantarexion resistance torque (PFRT) under AFO-OD were monitored.
Results: Peak PFRT under AFO-OD was observed during the loading response phase (LRP) in both groups.
AFO-OD promoted adequate plantarexion during LRP in the stroke group, whereas AFO-PS did not.
Compared with the AFO-PS, the AFO-OD signicantly reduced GAS EMG amplitude during LRP in the
stroke group, which was signicantly correlated with peak PFRT during LRP.
Conclusion: AFO-OD assisted the heel rocker function and reduced GAS muscle EMG amplitude during
LRP.
2010 Elsevier B.V. All rights reserved.

Keywords:
Stroke
Ankle-foot orthosis
Electromyography
Oil damper
Hemiplegic gait
Muscle activity

1. Introduction
For adults with hemiplegia after stroke, regaining the ability to
walk is crucial for performing activities of daily life. Common
problems after hemiplegic stroke include decreased gait speed and
an asymmetrical gait pattern [1,2], which increase the energetic
cost [3,4]. After a stroke, individuals often have impaired ankle
function due to muscle weakness [5], increased passive stiffness
[6], and excessive muscle coactivation [7]. These dysfunctions
affect gait because ankle motion and related muscle activities play
important roles in walking.
Wong et al. [8] suggested that the hemiplegic gait after stroke
may lack the typical heel strike and push-off mechanisms, i.e., heel
and forefoot rocker functions as described by Perry [9], thus
altering ground reaction forces and changing the foot contact
pattern to a pathologic shape. The heel rocker uses the heel as a
fulcrum during the loading response phase (LRP). During this
phase, rapid loading of the heel generates plantarexion torque,
which drives the foot toward the oor. The pretibial muscles

* Corresponding author at: Department of Health Science, Graduate School of

Medicine, Kyoto University, 53 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto 606-8507,
Japan. Tel.: +81 75 751 3918; fax: +81 75 751 3918.
E-mail address: oohata@hs.med.kyoto-u.ac.jp (K. Ohata).
0966-6362/$ see front matter 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.gaitpost.2010.10.083

decelerate the foot drop and draw the tibia forward when the foot
rolls into plantarexion. Insufcient eccentric dorsiexion muscle
activity after a stroke reduces the heel rocker function, as shown by
a positive relationship between dorsiexor strength and gait
velocity [10]. The forefoot rocker uses the metatarsophalangeal
joint as a fulcrum during the pre-swing (PSw) phase [9]. When the
limb is rapidly unloaded by the transfer of body weight to the other
limb, residual plantarexion action progresses to the tibia. As a
result, limb progression with knee exion occurs during the PSw
phase. The hemiplegic gait is characterized by impaired swing
initiation in the affected limb [11] due to inadequate leg propulsion
by the plantarexor [5]. Thus, these two plantarexion actions
during the LRP and PSw phase are critical for recovering gait after
hemiplegic stroke.
An ankle-foot orthosis (AFO) can improve the gait of hemiplegic
individuals [12,13]; however, the limited ankle motion associated
with an AFO with plantarexion stop [14,15], AFO with bilateral
stop [13], or unarticulated AFO [16], seems to be disadvantageous
for ankle function, because both heel and forefoot rocker functions
require adequate plantarexion range. An AFO with an oil damper
(AFO-OD) was developed to assist the heel rocker function [17,18];
however, differences between an AFO-OD and an AFO with limited
motion are not clear. The aim of this study was to determine
electromyography (EMG) changes of the lower limb muscles in
stroke patients wearing an AFO-OD.

K. Ohata et al. / Gait & Posture 33 (2011) 102107

2. Methods
2.1. Participants
Adult patients with stroke were recruited from the AFO-OD users list of the
manufacturer (Kawamuragishi Co. Ltd.). After providing informed consent, 13 men
with hemiplegia who lived in Osaka or Kyoto and had walked with an AFO-OD for at
least 1 month were selected for this study. Inclusion criteria were (1) a single stroke
at least 6 months prior to the study, (2) living at home independently with family
support, (3) ability to walk independently using an ankle foot orthosis and/or Tcane, (4) no gait symptoms from parkinsonism or ataxia, (5) no pain during gait due
to orthopedic disease, (6) no limitation of activity due to heart disease, (7) resting
heart rate <120 beats per minute, systolic blood pressure between 120 and
200 mm Hg, (8) no difculty understanding experimental tasks due to cognitive
problems, and (9) no excessive contracture inhibiting dorsiexion beyond neutral
(08) in a gait cycle. The control group consisted of 13 age-matched healthy men, who
consented to participate in this study. They exhibited a normal range of joint motion
and muscle strength and had no apparent gait abnormality.
In the control group, data from two participants were excluded; one participant
wore a shoe that inuenced his gait pattern, and the other participant was excluded
because of motion artifacts. In the stroke group, EMG data from one participant
were lost due to an error in the measurement procedure, and one participant

[(Fig._1)TD$IG]

103

showed limited ankle motion during gait due to excessive contracture. The
functional status of the participants in the stroke group was evaluated by the
Brunnstrom stage of recovery and the modied Rankin scale, which is the functional
outcome measure most often used in stroke research [19]. In addition, passive range
of motion of the ankle joint was measured and muscle tone was rated according to
the modied Ashworth scale. Isometric muscle strength of the ankle dorsiexor and
plantarexor was determined with a hand-held dynamometer in the sitting and
prone positions, respectively. Intraclass correlation coefcients (ICCs; [1.1]) were
0.97 and 0.91 for paretic and non-paretic plantarexors, 0.99 and 0.91 for paretic
and non-paretic dorsiexors, respectively, in the stroke group, and 0.97 and 0.96 for
plantarexors and dorsiexors, respectively, in the control group.
All procedures were approved by the ethics committee of Kyoto University
Graduate School and Faculty of Medicine and were consistent with the Declaration
of Helsinki.
2.2. Bracing conditions
The bracing conditions were similar in the stroke and control groups. After
sufcient practice, all participants walked twice on a 10-m walkway at a selfselected comfortable speed with or without a cane in each of three bracing
conditions: with an AFO-OD, with a conventional AFO with a plantarexion stop

Fig. 1. Oil damper unit for the ankle-foot orthosis and position of load cell. (A) Gait solution design of ankle-foot orthosis with an oil damper (AFO-OD). (B) Schematic design of
the ankle joint with the oil damper. Ankle joint consists of a piston rod (1), spring (2), cylinder (3), adjusting screw (4), spacer (5), and load cell (6). (C) Typical change in the
ankle joint angle and plantarexion resistive torque (PFRT) under the AFO-OD in the control group. PFRT shows two peaks during the loading response and pre-swing phases.

104

K. Ohata et al. / Gait & Posture 33 (2011) 102107

Table 1
Participant characteristics in the stroke and control groups.
Stroke group (n = 11)

Control group (n = 11)

Age (years)
Height (cm)
Weight (kg)
Etiology (n): ischemia/hemorrhage
Time post-stroke (months): median (range)
Affected side (n): right/left
Modied Rankin scale (n): I/II/III
Brunnstrom stage (n): III/IV/V/VI

52.1  13.6
169.4  5.3
63.0  6.9
4/7
20 (6116)
7/4
5/5/1
1/6/2/2

52.1  9.3
169.7  5.6
70.5  10.7

n.s.
n.s.
n.s.

Ankle strength
DF on paretic (non-dominant) side (Nm/kg)
PF on paretic (non-dominant) side (Nm/kg)
DF on non-paretic (dominant) side (Nm/kg)
PF on non-paretic (dominant) side (Nm/kg)

0.21  0.13
0.28  0.11
0.47  0.10
0.66  0.16

0.64  0.13
0.82  0.27
0.70  0.13
0.90  0.30

ROM
Dorsiexion: 5108/158

2/9

MAS
DF (n): MAS 0/MAS 1
PF (n): MAS <2/MAS 2

11/0
8/3

<0.001
<0.001
<0.001
0.032

Data (excluding time post-stroke) are expressed as mean  SD. Group differences were determined by unpaired t-test. ROM: range of motion, MAS: modied Ashworth scale, DF:
dorsiexor, PF: plantarexor, and n.s.: not signicant.
(AFO-PS), and without a brace. The mean gait speed under each bracing condition
was recorded for each participant.
The AFO-OD (Gait Solution Design; Kawamura Gishi, Osaka, Japan; Fig. 1A)
carries an oil damper unit on the lateral side of the ankle joint. A small hydraulic
cylinder is inserted in the oil damper unit to provide resistance to plantarexion as
needed (Fig. 1B). As the ankle joint plantarexes at initial contact, a piston rod is
pushed upward into an oil-lled cylinder with resistance. A spring returns the
piston to its initial position after plantarexion motion. The resistive force of the oil
damper can be easily changed by adjusting a screw. During measurements, the
screw position was maintained at a constant value that allowed comfortable
walking in the stroke group. In the control group, the screw was set to the same
position for all subjects. The AFO-PS was set to limit ankle plantarexion at 08. Both
braces allowed free dorsiexion and were worn on the affected side in the stroke
group and on the right side in the control group.

2.3. Measurement procedure

EMG measurements were performed at 1500 Hz with the TeleMyo system
(Noraxon Inc., USA). Bipolar silversilver chloride disposable surface electrodes
were placed over the muscle bellies of the three lower limb muscles that serve as
the main agonist muscles of dorsiexion and plantarexion tibialis anterior (TA),
lateral gastrocnemius (GAS), and soleus (SOL) on the paretic in the stroke group
and on the right side in the control group. Electrode placement on the TA was at 1/3
on the line 12-cm lateral to the tibia. The GAS electrode was placed at 1/3 on the
line between the head of the bula and the heel. The SOL electrode was placed at 1/2
to 2/3 on the line between the head of the medial condyles of the femur and the tip
of the medial malleolus. Two foot switches were positioned at the rst metatarsal
head and the heel on the paretic side to record the gait cycle.
To measure plantarexion resistance torque (PFRT) during gait using the AFOOD, a load cell was inserted above the hydraulic cylinder in the oil damper unit. As
the hydraulic cylinder produced a resistive force, the hydraulic cylinder pushed the
load cell as a counterforce; therefore, measurement of the counterforce reected
the PFRT at the oil damper. The angle of the ankle joint with AFO-OD was
simultaneously monitored with a potentiometer attached to the joint. Fig. 1C shows
a typical change during the gait cycle in a healthy control. Resistance of the oil
damper occurred when the joint angle exceeded 08 plantarexion; therefore, PFRT
showed two peaks during the LRP and PSw phases. The angle of the ankle joint with
AFO-PS was also simultaneously monitored with a potentiometer.
Five to 10 gait cycles were used to determine the EMG parameters and obtain
data from the load cell and the potentiometer of each subject. Time during a gait
cycle was expressed as percentage of the gait cycle (%GC). EMG recordings were
band-pass ltered between 16 and 500 Hz. Full wave rectication was performed
using the root mean square smoothing algorithm at a window interval of 50 ms. To
determine the change in muscle activity during LRP, peak EMG amplitudes from
0%GC to 10%GC were normalized using the maximum value from 20%GC to 100%GC.
The load cell and the potentiometer were connected to the TeleMyo transmitter
system with EMG and foot switch. EMG signals were recorded at a sampling rate of
1500 Hz and smoothed with a low-pass lter at 20 Hz. Peak PFRT was obtained, and
changes in ankle joint angle were calculated from initial contact to plantarexion
peak during LRP. Dorsiexion angles were expressed as positive values.
In the stroke group, these measurements were repeated 2 weeks later to
determine testretest reliability. Ten of 11 individuals in the stroke group

participated in the second measurement; one participant was absent for health
reasons. ICCs (1, k) for changes in ankle joint angle from initial contact were 0.89 for
both AFO-OD and AFO-PS, and ICC (1, k) for peak PFRT under the AFO-OD during LRP
was 0.95.
2.4. Statistical analysis
One-way repeated measurement analysis of variance and multiple comparisons
(Bonferroni) were used to compare gait speed among brace conditions (AFO-OD,
AFO-PS and without brace) in each group. Peak PFRT and changes in ankle joint
angle during LRP were compared between groups with the MannWhitney U test.
EMG amplitudes during LRP were compared with the Wilcoxon signed-rank test in
each group. To determine the importance of peak PFRT with the AFO-OD, the
relationship between peak PFRT and gait speed with AFO-OD use and the percent
reduction of EMG amplitude with AFO-OD use compared with AFO-PS use was
determined by partial correlation coefcients adjusted by body weight in each
group. Statistical signicance was set at p < 0.05.

3. Results
3.1. Increased gait speed with AFO-PS and AFO-OD
Table 1 shows the patient demographic and clinical characteristics. Age, height, and weight were not signicantly different
between the stroke and the control groups (unpaired t-test);
however, ankle muscle strength on both sides was lower in
the stroke group than in the control group. Gait speed without a
brace was 84.1  13.2 m/min (control group) and 28.3  11.0 m/min
(stroke group). Use of AFO-PS increased gait speed to 88.0  13.6 m/
min (control group) and 32.0  11.3 m/min (stroke group), whereas
AFO-OD increased gait speed to 90.1  14.6 m/min (control group)
and 34.8  13.9 m/min (stroke group). A signicant difference in gait
speed among the three bracing conditions was observed in the stroke
group (p = 0.002). Multiple comparison analysis revealed that both
AFO-PS and AFO-OD improved gait speed (p = 0.012 and 0.007,
respectively).
3.2. Brace-dependent change in PFRT, EMG amplitude, and ankle
motion
Fig. 2 shows typical PFRT data and ankle joint angle with both
braces in the control and stroke groups, and typical EMG patterns
of each lower limb muscle produced by both braces in the stroke
group. Use of the AFO-OD produced a similar peak in PFRT during
LRP in both groups (Table 2); however, an additional peak in PFRT
was observed during the PSw phase in the control group (Fig. 2A).

[(Fig._2)TD$IG]

K. Ohata et al. / Gait & Posture 33 (2011) 102107

105

A. PFRT
a. Control

b. Stroke

(Nm)
5

(Nm)
5

0
5

(Nm)

0
100%GC

0
100%GC

100%GC

B. Ankle joint angle

c. AFO-OD
(degree)
40

Stroke
Control

30

20

Stroke
Control

15
10

20

10

0
-10

100%GC

20%GC

-5

d. AFO-PS
(degree)
40

20

30

15

20

10

10

-10

100%GC

20%GC

-5

C. Electromyography
(uV)

(uV)
500 Tibialis anterior

500 Tibialis anterior

-500
500 Gastrocnemius

-500
500 Gastrocnemius

-500
500 Soleus

-500
500 Soleus

-500

-500

LR

e. AFO-OD

LR

f. AFO-PS

Fig. 2. Typical plantarexion resistive torque (PFRT), ankle joint angle, and electromyography (EMG) amplitudes during gait cycle. (A) PFRT in the control group (a: left
column) shows two peaks during the loading response (LR) and pre-swing (PSw) phases. However, PFRT in the stroke group (b: middle and right columns) shows low or no
peak during PSw. (B) The ankle joint angle during the entire gait cycle (left) and from 0% to 20% of gait cycle (right) using AFO-OD (c) and AFO-PS (d). Positive values represent
dorsiexion. Solid and dash lines indicate the stroke and control group. (C) Raw EMG data using AFO-OD (e) and AFO-PS (f).

In most subjects in the stroke group, PFRT was low during the PSw
phase with no clear peak. Furthermore, peak PFRT during the PSw
phase using the AFO-OD was signicantly correlated with gait
speed in the control group (r = 0.78, p = 0.008).
The peak plantarexion angle after initial contact did not differ
between stroke and control groups when using the AFO-OD
(Table 2). In contrast, the ankle joint angle was signicantly

different between the groups when using the AFO-PS (p = 0.033).

In particular, the stroke group showed dorsiexion immediately
after initial contact (Fig. 2B).
In the control group, there was no signicant brace-dependent
difference in EMG amplitude of the three lower limb muscles
(Table 2). However, in the stroke group, GAS muscle activity was
signicantly lower with AFO-OD than with AFO-PS during LRP

K. Ohata et al. / Gait & Posture 33 (2011) 102107

106

Table 2
Brace-dependent changes in stroke patients and control subjects.
Stroke group (n = 11)
AFO-OD
Peak PFRT during LR (Nm)
Peak PFRT during PSw
Change of ankle angle from initial contact (8)
TA activity during LR (%)
GAS activity during LR (%)
SOL activity during LR (%)

Control group (n = 11)

AFO-PS

1.5  1.1
N.A.
1.8  1.4
62.5  37.3
73.7  35.3
84.6  26.0

0.9  1.2#
56.0  25.9
120.5  68.9
128.1  92.6

AFO-OD

AFO-PS

0.508
0.041*
0.594

1.3  0.9
2.8  1.4
1.7  1.1
55.2  15.9
8.4  4.0
13.6  4.7

0.1  0.5
53.7  19.7
9.9  5.0
13.5  5.6

0.328
0.182
0.534

Data are expressed as mean  SD. Group differences were determined by Wilcoxon signed-rank test. PFRT: plantar exor resistive torque, LR: loading response phase, PSw: preswing phase, and N.A.: not available. Plantar exion angle in Change of ankle angle from initial contact was expressed as a negative value.
#
p<0.05 compared with control group.
*
p<0.05 between two brace conditions.

Table 3
Relation between peak plantar exor resistive torque and reduction in electromyography (EMG) amplitude during loading response phase in stroke patients and
control subjects.
Stroke group
(n = 11)

TA (% reduction)
GAS (% reduction)
SOL (% reduction)

Control group
(n = 11)

0.05
0.78*
0.34

0.892
0.007
0.334

p
0.62
0.56
0.21

0.055
0.093
0.568

r: partial correlation coefcient adjusted for body weight. TA: tibialis anterior, GAS:
gastrocnemius, and SOL: soleus.
*
p < 0.05; signicant correlation between peak plantar exor resistive torque
and the percent reduction of EMG amplitude under AFO-OD from the value under
AFO-PS during loading response phase.

(p = 0.041, Fig. 2C and Table 2). This difference in EMG amplitude

between AFO-OD and AFO-PS (percent reduction) was signicantly
correlated with peak PFRT during LRP (Table 3).

4. Discussion
In the present study, a peak in PFRT was observed with AFO-OD
use in stroke patients and control subjects during LRP. Most of the
subjects in the stroke group dorsiexed the ankle immediately
after initial contact when using the AFO-PS. With the AFO-OD,
however, the ankle joint in the stroke group showed adequate
plantarexion from initial contact, similar to the control group. In
the stroke group, the AFO-OD decreased GAS muscle activity
compared with the AFO-PS during LRP. Further, the percent
reduction in GAS muscle EMG amplitude with AFO-OD, compared
with AFO-PS, was signicantly correlated with peak PFRT during
LRP. The control group produced an additional peak in PFRT during
the PSw phase when using AFO-OD, whereas the stroke group did
not. PFRT during the PSw phase was signicantly correlated to gait
speed.
Eccentric contraction of the TA muscle decreases the rate of
plantarexion during LRP. This contraction draws the tibia forward
as the foot drops. In the hemiplegic gait, dorsiexion inability leads
to insufcient toe clearance during the swing phase; the TA muscle
is required for toe clearance during the swing phase and to
progress the tibia during LRP. The AFO-PS can compensate for the
TA to improve the toe clearance during the swing phase through
limited plantarexion; however, this leads to excessive tibial
progression during LRP. In fact, this study showed that the ankle
joint began to dorsiex immediately after initial contact in stroke
patients using the AFO-PS, indicating that the AFO-PS blocks
adequate plantarexion. In contrast, adequate plantarexion was
observed during LRP in stroke patients using the AFO-OD,
consistent with results of a case series study reporting that

AFO-OD achieves sufcient plantarexion of the ankle by proper

PFRT during LRP [18].
In the present study, the triceps surae produced high EMG
amplitudes during LRP in the stroke group wearing the AFO-PS,
probably because of an excessive stretch reex due to dorsiexion
immediately after initial contact. This dorsiexion may also
produce higher plantarexor activity. AFO-OD reduced the EMG
activity of the GAS muscle during LRP in the stroke group.
Furthermore, this reduction in EMG amplitude was related to the
peak PFRT during LRP. The smooth plantarexion motion achieved
with the AFO-OD may reduce excessive activity caused by the
stretch reex.
PFRT is a convenient method to assess plantarexion torque
during the PSw phase. Peak PFRT during the PSw phase was
signicantly correlated with gait speed in the control group;
however, in the stroke group, PFRT did not show a denite peak
during the PSw phase. The loss of plantarexor force during the
PSw phase and its relationship with gait function in patients after
stroke have been previously reported [5,11,20].
The present study has several limitations. Muscle activity was
not measured at the peroneus longus or extensor hallucis longus.
Kinematic analysis of the knee or other joints is also lacking in this
study; thus it is not known whether the AFO-OD inuences these
parameters. Furthermore, the difference in gait speed between the
stroke and control groups probably inuenced the presented
results, such as the peak PFRT during PSw and ankle joint motion.
Further study with proper adjustments of gait speed is necessary to
evaluate the difference between both groups more strictly. In
addition, participants in the present study had been using the AFOOD for at least one month. It is not clear whether similar changes
would be observed immediately with initial use of the AFO-OD.
In conclusion, the AFO-OD assists the heel rocker function by
producing adequate plantarexion. The main effect of PFRT during
LRP is not to reduce TA activity but to decrease GAS activity to
avoid an excessive stretch reex; however, the decit in ankle
plantarexion torque during the PSw phase remains a major
problem.
Acknowledgements
This study was supported by a grant from Kawamuragishi Co.
Ltd., Japan. We wish to thank the participants who volunteered.
Conict of interest
The authors declared a potential conict of interest as follows:
Tadashi Yasui is employed by Kawamuragishi Co. Ltd.
References
[1] Brandstater ME, de Bruin H, Gowland C, Clark BM. Hemiplegic gait: analysis of
temporal variables. Arch Phys Med Rehabil 1983;64:5837.

K. Ohata et al. / Gait & Posture 33 (2011) 102107

[2] Wall JC, Turnbull GI. Gait asymmetries in residual hemiplegia. Arch Phys Med
Rehabil 1986;67:5503.
[3] BARD G. Energy expenditure of hemiplegic subjects during walking. Arch Phys
Med Rehabil 1963;44:36870.
[4] Detrembleur C, Dierick F, Stoquart G, Chantraine F, Lejeune T. Energy cost,
mechanical work, and efciency of hemiparetic walking. Gait Posture
2003;18:4755.
[5] Nadeau S, Gravel D, Arsenault AB, Bourbonnais D. Plantarexor weakness as a
limiting factor of gait speed in stroke subjects and the compensating role of hip
exors. Clin Biomech (Bristol Avon) 1999;14:12535.
[6] Lamontagne A, Malouin F, Richards CL. Contribution of passive stiffness to
ankle plantarexor moment during gait after stroke. Arch Phys Med Rehabil
2000;81:3518.
[7] Lamontagne A, Malouin F, Richards CL, Dumas F. Mechanisms of disturbed
motor control in ankle weakness during gait after stroke. Gait Posture
2002;15:24455.
[8] Wong AM, Pei YC, Hong WH, Chung CY, Lau YC, Chen CP. Foot contact pattern
analysis in hemiplegic stroke patients: an implication for neurologic status
determination. Arch Phys Med Rehabil 2004;85:162530.
[9] Perry J. Gait analysis: normal and pathological function. Thorofare, NJ: Slack;
1992.
[10] Lin PY, Yang YR, Cheng SJ, Wang RY. The relation between ankle impairments
and gait velocity and symmetry in people with stroke. Arch Phys Med Rehabil
2006;87:5628.
[11] Chen G, Patten C, Kothari DH, Zajac FE. Gait differences between individuals
with post-stroke hemiparesis and non-disabled controls at matched speeds.
Gait Posture 2005;22:516.

107

[12] Lehmann JF, Condon SM, Price R, deLateur BJ. Gait abnormalities in hemiplegia: their correction by ankle-foot orthoses. Arch Phys Med Rehabil 1987;68:
76371.
[13] Hesse S, Werner C, Matthias K, Stephen K, Berteanu M. Non-velocity-related
effects of a rigid double-stopped ankle-foot orthosis on gait and lower limb
muscle activity of hemiparetic subjects with an equinovarus deformity. Stroke
1999;30:185561.
[14] Fatone S, Gard SA, Malas BS. Effect of ankle-foot orthosis alignment and footplate length on the gait of adults with poststroke hemiplegia. Arch Phys Med
Rehabil 2009;90:8108.
[15] Fatone S, Hansen AH. Effect of ankle-foot orthosis on roll-over shape in adults
with hemiplegia. J Rehabil Res Dev 2007;44:1120.
[16] Park JH, Chun MH, Ahn JS, Yu JY, Kang SH. Comparison of gait analysis between
anterior and posterior ankle foot orthosis in hemiplegic patients. Am J Phys
Med Rehabil 2009;88:6304.
[17] Yamamoto S, Hagiwara A, Mizobe T, Yokoyama O, Yasui T. Development of an
ankle-foot orthosis with an oil damper. Prosthet Orthot Int 2005;29:20919.
[18] Yokoyama O, Sashika H, Hagiwara A, Yamamoto S, Yasui T. Kinematic effects
on gait of a newly designed ankle-foot orthosis with oil damper resistance: a
case series of 2 patients with hemiplegia. Arch Phys Med Rehabil 2005;86:
1626.
[19] Van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJA, Van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke
1988;19:6047.
[20] Neptune RR, Kautz SA, Zajac FE. Contributions of the individual ankle plantar
exors to support, forward progression and swing initiation during walking. J
Biomech 2001;34:138798.