ARTICLE IN PRESS

Effects of a 4-Week Self-Ankle Mobilization with Movement

Intervention on Ankle Passive Range of Motion, Balance, Gait,
and Activities of Daily Living in Patients with Chronic Stroke:
A Randomized Controlled Study

D1X XDonghwan Park, D2XMS,

X PT, D3X XJi-Hyun Lee, D4XPhD,
X PT, D5X XTae-Woo Kang, D6XMS,
X PT,
and D7X XHeon-seock Cynn, D8XPhD,
X PT

Goal: To compare the effects of a 4-week self-ankle mobilization with movement

training program with those of self-ankle mobilization with movement with a 10°
inclined board in patients with chronic stroke. Materials and Methods: A randomized
controlled assessor-blind trial was conducted. The patients were randomized into 2
arms. Subjects were 28 chronic stroke patients with hemiplegia. Both arms attended
standard rehabilitation therapy for 30 minutes per session. In addition, self-ankle
mobilization with movement and self-ankle mobilization with movement with a
10° inclined board trainings were performed 3 times per week for 4 weeks. Ankle
dorsiflexion passive range of motion, static balance ability, Berg balance scale, gait
parameters (walking speed, cadence, and step length), and activities of daily living
were used to assess changes in motor function after training. Findings: After 4 weeks
of training, all dependent variables were significantly improved in both arms as
compared with their baseline values. Furthermore, relative to the self-ankle mobili-
zation with movement arm, the self-ankle mobilization with movement with a 10°
inclined board arm demonstrated significantly improved ankle dorsiflexion passive
range of motion, static balance ability, gait speed, cadence, and affected-side step
length. Conclusions: Our results support the hypothesis that self-ankle mobilization
with movement with a 10° inclined board combined with standard rehabilitation
was superior to self-ankle mobilization with movement combined with standard
rehabilitation with respect to the improvement in motor function in the patients
with chronic stroke.
Key Words: Stroke rehabilitation—ankle—balance—gait—self-mobilization with
movement
© 2018 National Stroke Association. Published by Elsevier Inc. All rights reserved.

From the Department of Physical Therapy, Graduate School, Yon-
sei University, Wonju, Gangwon-do, Republic of Korea.
Received June 8, 2018; revision received July 14, 2018; accepted
August 5, 2018.
Financial Disclosures: No funding or grants or equipment provided
for the project from any source. No financial benefits to the authors.
Address correspondence to Heon-seock Cynn, PT, PhD, Depart-
ment of Physical Therapy, Graduate School, Yonsei University,
Wonju, Gangwon-do 26493, Republic of Korea. E-mail:
cynn@yonsei.ac.kr
1052-3057/$ - see front matter
© 2018 National Stroke Association. Published by Elsevier Inc. All
rights reserved.
https://doi.org/10.1016/j.jstrokecerebrovasdis.2018.08.010
Introduction
The most commonly recognized stroke-induced defi-
ciency is motor impairment, characterized by limited
functional muscle control, decreased range of motion
(ROM), and reduced mobility.1 A normal gait requires an
ankle dorsiflexion passive ROM (DF-PROM) ranging
from 10° to 15°, allowing the tibia to properly move over
the talus.2 However, previous investigations indicated
that stroke patients demonstrate approximately 50%
reduction in DF-PROM relative to healthy individuals.3
As a result, patients with stroke often present with
impaired balance and decreased independence during
walking.4,5
Various interventions have been used to increase DF-
PROM and functional ability in stroke patients, including

Journal of Stroke and Cerebrovascular Diseases, Vol. &&, No. && (&&), 2018: pp 1
9 1
 ARTICLE IN PRESS
2 D. PARK ET AL.
ankle joint exercises, joint mobilization, and mobilization
with movement (MWM).6-8 Nakayama et al8 applied an
ankle inclined board exercise to poststroke hemiparetic
patients, which produced significant improvements in
walking velocity, decreased the number of steps in the 10
m walking test, and decreased the timed up-and-go test
performance time. Furthermore, Kluding and Santos6
reported an increase in ankle ROM and a decrease in the
time required to transition from a seated to a standing
position following joint mobilization combined with func-
tional task practice. Lastly, An and Jo7 performed a ran-
domized controlled trial in stroke patients and
demonstrated increased ankle DF-PROM, strength, and
gait function in patients receiving MWM as compared
with the control arm (ie, no therapy received).
MWM facilitates the anterior glide of the tibia over a
fixed talus during ankle dorsiflexion exercises in a
weight-bearing position, improving dorsiflexion ROM,
and enabling proper functional activity.9,10 MWM can
also be administered in combination with a manual force
applied by a therapist to stabilize the talus during active
motion, permitting active ankle dorsiflexion in a weight-
bearing position.7,9,11 However, as MWM for ankle dorsi-
flexion requires the presence of a trained therapist to sta-
bilize the anterior talus,9 MWM is practically challenging
for an individual to perform independently.12 To account
for this technical difficulty, Mulligan was the first to pro-
pose self-MWM (S-MWM) by using a strap to enable
unrestricted movement without producing pain at addi-
tional joints in the body.10,13 Furthermore, ankle mobiliza-
tion with a strap can provide for the posterior glide of the
talus during closed chain DF movements.11
Although MWM in combination with stool has been
shown to improve ankle ROM and gait functions in indi-
viduals with patients with strokes,7,14 the effectiveness of
S-MWM using inclined board versus S-MWM using stool
in patients with chronic strokes has not yet been investi-
gated. In addition, conducting S-MWM with a 10°
inclined board may help to effectively control the pulling
force angle of the strap in the posterior-inferior direction
in chronic stroke patients with limited ankle DF-PROM.
Therefore, the objective of this investigation was to com-
pare the relative efficacy of S-MWM with that of S-MWM
in combination with a 10° inclined board (S-MWM10°)
applied over 4 weeks with respect to improving ankle DF-
PROM, static balance ability (SBA), Berg balance scale
(BBS), gait parameters (walking speed, cadence, and step
length), and activities of daily living (ADLs; modified Bar-
thel index) in patients with chronic stroke.
Materials and Methods
Study Design and Participants
This randomized controlled trial was used. A pilot test
was performed in 8 volunteers (n = 4 patients each in the
S-MWM and S-MWM10° arms), the results of which were
used to conduct a power analysis to determine the sample
size required to achieve a significance level of a = .05,
power of .8, and an effect size of 1.65. The sample size
needed to adequately power the study was estimated
using the unaffected-side step-length outcome measure,
with the largest estimate retained among the pilot study
outcomes. The results of the power analysis demonstrated
that 6 patients were required per study arm. The G-power
version 3.1.2 software was used for the power analyses
(Franz Faul, University of Kiel, Kiel, Germany). Twenty-
eight chronic stroke inpatients from Gyeong-in Hospital,
Incheon, volunteered to participate in this study. The
inclusion criteria were as follows: (1) a diagnosis of hemi-
plegia due to hemorrhagic or ischemic stroke, after more
than 6 months had elapsed after stroke;15 (2) <8° of ankle
DF-PROM on the affected side;6 (3) able to perform a sin-
gle-leg lunge, independently placing the lower limb of the
affected side onto an inclined board from a standing posi-
tion; (4) able to independently walk a distance of 10 m
without the use of assistive devices;16 (5) receiving contin-
uous standard rehabilitation therapy according to the
daily inpatient treatment program; (6) no concurrent
training in any interventions related to balance and gait
from other institutions; and (7) a minimum score of 24 in
the Korean Mini-Mental State Examination.17 The exclu-
sion criteria were as follows: (1) history of surgery in the
lower extremities; (2) lower limb bone fractures; (3) a
diagnosis of neurological disease; (4) hip and knee flexion
contractures; and (5) presence of any contraindication of
ankle joint mobilization (eg, ankle joint hypermobility,
trauma, or inflammation). For each participant, baseline
data were collected, including sex, age, weight, height,
time from stroke onset, stroke type, hemiplegic side, and
K-MMSE score. The baseline characteristics of the 24
stroke patients recruited are summarized in Table 1. All
patients provided written informed consent for participa-
tion in this study, and the Yonsei University Wonju Insti-
tutional Review Board issued protocol approval no.
1041849-201705-BM-050-02. The trial was registered
online at The Clinical Research Information Service (ID
no. KCT0002547).
Interventions and Procedures
The principal investigator administered the study inter-
vention. The participants were randomized to either the
S-MWM (n = 14) or S-MWM10° (n = 14) arm by using an
online randomization program (http://www.randomiza
tion.com). The patients’ characteristics and all outcome
measures were assessed on study day 1 and 1 day after 4
weeks of treatment, measured by 2 physical therapists
with 8 years of practical clinical experience who were
blinded to the grouping. Specifically, ankle DF-PROM,
SBA, BBS, gait parameters (eg, gait speed, cadence, and
step lengths), and the Korean version of the modified Bar-
thel index (K-MBI) served as the outcome measures. All
 ARTICLE IN PRESS
SELF-ANKLE MOBILIZATION WITH MOVEMENT ON PATIENTS WITH CHRONIC STROKE
Table 1. Clinical information of the patients with stroke
Characteristics S-MWM (n = 14)
Age (year) 57.6 § 9.0
Height (cm) 164.6 § 7.1
Weight (kg) 63.8 § 10.1
Gender(male/female) 7/7
Hemiplegic side (left/right) 8/6
Type of stroke (ischemia/hemorrhage) 10/4
Disease duration (months) 9.1 § 1.9
K-MMSE (0-30) 26.7 § 2.2
modified Ashworth scale 1.75 § .62
K-MMSE, Korean Mini-Mental State Examination; S-MWM, self-mobilization with movement; S-MWM10˚, self-mobilization with
movement with 10˚ incline board. Values are expressed as mean § standard deviation or frequency.
the measurements were repeated 3 times and averaged.
At the start of the trial, the participants in both arms were
respectively familiarized with either S-MWM or S-
MWM10°, depending on the randomization. Familiariza-
tion was implemented for approximately 15 minutes/day
over the course of 1-3 days, depending on the participant.
The familiarization period concluded when the partici-
pant demonstrated the ability to conduct either the S-
MWM or S-MWM10° protocol. All the enrolled partici-
pants, regardless of study arm, demonstrated acclimation
to the protocol after the familiarization period. Through-
out the trial, all the subjects underwent the same rehabili-
tative training for gait and the lower limbs in accordance
with the daily inpatient treatment program schedule. All
the participants received the standard rehabilitation ther-
apy for 30 minutes per session, and the amount and type
of rehabilitation therapy were based on a standardized
protocol to maintain similarities between the participants
in both arms. The first 5 minutes of the active and passive
ROM exercises were allocated for the lower limb of the
affected side. The next 15 minutes were spent on weight-
bearing training during sitting and standing. The final 10
minutes were spent on walking. Both S-MWM and S-
MWM10° were performed 3 times per week for 4 weeks
in the same location and under the supervision of the
principal investigator to ensure both the proper perfor-
mance of the protocol and the safety of all the enrolled
participants. Both interventions were conducted in a one-
on-one basis. Any participant with pain during treatment
was excluded from the study Fig 1.
Self-Mobilization with Movement Using a Strap on
the Stool
While in a lunging stance, the participants performed
ankle S-MWM with a nonelastic strap approximately 40
cm long. The foot of the affected side was placed on the
stool. The length and width of the stool were approxi-
mately 20 and 15 cm, respectively. Black tape (2 cm width)
was attached to the middle of the stool, and the middle of
the heel and second toe were placed along the taped line.
3
S-MWM10˚ (n = 14) P value
64.1 § 9.7 .082
165.4 § 5.1 .739
63.0 § 7.6 .818
7/7 1.000
9/5 .699
9/5 .686
11.28 § 3.6 .065
26.4 § 2.3 .736
01.75 § .45 .932
The opposite foot was then placed on the ground behind
the participant in the lunge position, with a backward force
applied by pulling the strap. The front of the strap was
placed around the anterior aspect of the talus, positioned
just inferior to the medial and lateral malleoli of the
affected ankle on the stool, and the back of the strap was
placed around the medial region of the opposite foot to
provide the posterior-inferior pulling force during S-
MWM. The participant was then asked to perform S-
MWM with the strap pulled taut in the initial lunge posi-
tion. Specifically, while lunging, the knee of the affected leg
is moved forward along a straight line to affect a lunge dur-
ing S-MWM. The patient was required to move the knee
forward without discomfort and pain until the soleus mus-
cle of the front leg was sufficiently stretched. This end posi-
tion was maintained with the constant application of
resistance without lifting the heel for 20 seconds; the
patient then returned to the initial position. S-MWM was
performed 15 times, with 10 seconds of rest between exer-
cises. If at any point the patient required assistance, they
were instructed to lightly grasp a safety bar.
Self-Mobilization with Movement Using a Strap on
the 10° Inclined Board
All the procedures were identical to S-MWM except for
the inclusion of the 10° inclined board. The length and
width of the inclined board were approximately 20 and
15 cm, respectively. The inclined board was used to con-
trol the pulling-force angle of the strap in the posterior-
inferior direction on the affected ankle. S-MWM10° was
performed 15 times, with 10 seconds of rest between exer-
cises. If the patient required any assistance, they were
instructed to lightly grasp a safety bar Fig 2.
Outcome Measures
Ankle DF-PROM was measured in the prone position
and asked to flex the knee to 90°. The central axis of a 14
in. plastic goniometer was placed on the lateral malleolus.
The stationary arm of the goniometer was placed parallel
to the lateral side of the fifth metatarsal bone. The moving
 ARTICLE IN PRESS
4 D. PARK ET AL.
arm of the goniometer was placed parallel to the center of
the fibular head, and the 3 axes were marked with a dot.
The 3 marked dots were maintained in the same position
throughout the duration of the test. One physical therapist
maintained a neutral subtalar joint position while apply-
ing force to the plantar surface of the forefoot and midfoot
until further movement was firmly restricted. The second
physical therapist confirmed the neutral subtalar joint
position and independently measured the ankle DF-
PROM. Measurements of ankle DF-PROM was repeated 3
times, with results averaged for ankle DF-PROM data
analysis.
SBA was measured using the Biodex Balance System
version 1.08 software (Biodex, Inc., Shirley, NY). During
static measurements, the stability index was measured as
the angular excursion of the patient's center of gravity.
High SBA scores indicate several repositioning move-
ments and poor overall balance. This overall stability
score is reported to be an accurate indicator of the overall
balance capability of the patient.18 The test was performed
for 30 seconds and repeated 3 times, with results averaged
for final analysis.
BBS measures participant balance ability to quantify
both static and dynamic balance abilities. This scale is
comprised of 14 items related to functional tasks routinely
performed during ADLs. Points are assigned for 3
spheres, namely sitting, standing, and changing postures.
Possible scores for each task range from a minimum of 0
to a maximum of 4, with a maximum score of 56 points.
This measurement tool has reportedly high reliability and
validity, with intra- and interexaminer reliabilities of
r = .99 and r = .98, respectively.19
Spatial-temporal gait parameters were evaluated using
the GAITRite system (CIR Systems, Easton, PA). Specifi-
cally, as the patient walks on the gait mat, the machine
records the loads produced by the feet at a sampling rate
of 80 Hz, digitally compiled in the attached computer.
The reliability of the machine for testing gait speed,
cadence, single and double limb support periods, step
length, and stride length is reportedly excellent (intraclass
correlation coefficients between .82 and .92 and coeffi-
cients of variations between 1.4% and 3.5%).20 All the par-
ticipants were instructed to stand in front of the gait board
and then walk across at a comfortable, self-selected speed
until they arrive at the end of the board. Gait parameters
were measured 3 times and averaged for final analysis.
The K-MBI, an inventory for quantifying participant
performance of ADLs, is comprised of 10 items, namely
personal hygiene, bathing alone, eating, using the bath-
room, ascending stairs, dressing, defecating, controlling
urine, walking or using a wheelchair, and moving to a
chair or bed. Each item is scored from 0 to 5, that is, (0)
independent conduct, (1) minimum assistance, (2) inter-
mediate assistance, (3) maximum assistance, and (5)
impossible to conduct.21 Higher K-MBI scores indicate
greater impairment in performance of ADLs.
Statistical Analysis
The PASW Statistics 18 software (SPSS, Chicago, IL)
was used for all statistical analyses. Normality of data dis-
tribution was examined using the one-sample Kolmo-
gorov-Smirnov test. Data are presented as mean §
standard deviation for continuous variables as appropri-
ate and as proportions for categorical variables. Baseline
demographic variables were compared between the arms
by using an independent t test for continuous data and
the chi-square test of independence for categorical data.
The Mann-Whitney test was used to compare the baseline
differences in MAS score between the arms. After the
treatment, independent t tests were used to compare
between-group means, and paired t tests were used to
compare within-group means. Furthermore, effect sizes
were calculated to determine meaningful changes
between groups; an effect size of .20 indicates a small
change; .50, a moderate change; and .80, a large change.22
Statistical significance was set at P < .05.
Results
Outcome measures are presented in Table 2. Between
arms, no significant differences were observed in any of
the measured baseline values (Table 2). After treatment,
however, ankle DF-PROM was significantly increased in
the S-MWM10° arm relative to the S-MWM arm (P =
.033). Furthermore, both arms demonstrated significant
improvements in ankle DF-PROM after treatment,
increasing by 27.4% in the S-MWM arm and by 54.9% in
the S-MWM10° arm (P < .001).
Post-treatment SBA was significantly decreased in the
S-MWM10° arm as compared with the S-MWM arm (P <
.001). Furthermore, both arms demonstrated significant
D9X X
improvements in SBA after treatment, decreasing by
29.2% in the S-MWM arm and by 53.1% in the
S-MWM10° arm (P < .001). Moreover, post-treatment BBS
was not significantly different between the study arms
(P = .79). However, both arms demonstrated significant
improvements in BBS after treatment, increasing by 17.4%
in the S-MWM arm and by 28.8% in the S-MWM10° arm
(P < .001).
Post-treatment gait speed was significantly increased
in the S-MWM10° arm relative to the S-MWM arm
(P = .01). Furthermore, both arms demonstrated signifi-
cant improvements in gait speed after treatment,
increasing by 23.6% in the S-MWM arm and by 42.8%
in the S-MWM10° arm (P < .001). In addition, cadence
was significantly increased after treatment in the
S-MWM10° arm relative to the S-MWM arm (P = .005).
Furthermore, both arms demonstrated significant
improvements in cadence after treatment, increasing
by 11.4% in the S-MWM arm and by 24.0% in the
S-MWM10° arm (P < .001). Similarly, post-treatment
step length on the affected side was significantly
increased in the S-MWM10° arm relative to the
 ARTICLE IN PRESS
SELF-ANKLE MOBILIZATION WITH MOVEMENT ON PATIENTS WITH CHRONIC STROKE 5

BBS, Berg balance scale; DF-PROM, dorsiflexion passive range of motion; K-MBI, Korean version of the Modified Barthel Index; SBA, static balance ability. Values are expressed as mean §
S-MWM arm (P = .033). Furthermore, both arms dem-

Effect Size
onstrated significant improvements in step length on

2.44
3.52
3.44
1.45
1.36
1.12
.68
2.38
the affected after treatment, increasing by 13.2% in the
S-MWM arm and by 19.4% % in the S-MWM10° arm
Self-mobilization with movement with 10˚ incline board (n = 14)

(P < .001). Although the post-treatment step length of

16.21 (13.32, 19.11) y

17.11 (11.99, 22.23) y
the unaffected side was not significantly different

11.93 (9.83, 14.02)

Change score (CI)
y

6.53 (4.61, 8.45) y

10.5 (8.13, 12.87)
-.51 (-.59, -.44) y
3.37 (2.96, 3.79)

7.73 (5.62, 9.83)

between the arms (P = .98). However, both arms dem-
onstrated significant improvements in step length of
the unaffected side after treatment, increasing by
19.5% in the S-MWM arm and by 27.1% in the
S-MWM10° arm (P < .001).
The post-treatment K-MBI was not significantly differ-
54.02 § 12.19*
88.44 § 12.48*

36.25 § 11.28*

ent between the trial arms (P = .18). However, the K-MBI

09.51 § 1.21*

46.93 § 3.58*

40.18 § 6.25*

69.07 § 4.7*

scores were significantly improved after treatment rela-

0.45 § .12*

tive to the baseline scores in both arms, by 15.5% in the

Posttest

S-MWM arm and 20.9% in the S-MWM10° arm (P < .001).

Discussion
37.81 § 10.22
71.33 § 12.76

28.52 § 11.40
06.14 § 1.55

36.43 § 2.53

57.14 § 5.33
33.65 § 5.4
0.96 § .17

The objective of this study was to compare the efficacy

Table 2. Changes in the intervention in each group

Pretest

of 4 weeks of S-MWM and S-MWM10° therapies for

improving ankle DF-PROM, SBA, BBS, gait parameters
(walking speed, cadence, and step length), and ADLs in
Effect Size

chronic stroke patients. After 4 weeks of study interven-

1.47
1.81
1.99
.74
.58
.63
.58
.97

tion, the S-MWM10° arm demonstrated significant

increases in DF-PROM, SBA, gait speed, cadence, and
*P < .05 indicates a significant difference between pre- and postinterventions within the group.

affected-side step length as compared with the S-MWM

arm after completion of the trial intervention. Our find-
Change score (CI)

7.99 (5.43, 10.56)

P < .05 indicates a significant difference between the change scores between the groups.
1.82 (1.44, 2.19)

6.57 (5.65, 7.49)

7.5 (3.55, 11.45)

4.06 (2.19, 5.92)
5.91 (3.54, 8.28)
8.78 (7.46, 10.1)

ings support our hypothesis, as S-MWM10° was more

-.28 (-.34, -.22)

effective for improving ankle DF-PROM, balance, gait

Self-mobilization with movement (n = 14)

parameters, and ADLs relative to S-MWM alone. To the

best of our knowledge, this is the first investigation to
demonstrate both a long-term training benefit and to com-
pare the differential effect(s) of S-MWM and S-MWM10°
standard deviation. Change score = post-test
pretest. CI, confidence interval.
41.90 § 10.69*
73.56 § 13.27*

36.35 § 10.67*

in chronic stroke patients.

08.46 § 1.27*

44.43 § 3.65*

34.90 § 6.18*

65.43 § 8.61*
0.68 § .15*

Compared with the S-MWM arm, ankle DF-PROM sig-

Post-test

nificantly increased by 85.2% in the S-MWM10° arm after

4 weeks of study intervention. Our results are consistent
with those of several previous studies.6,12 Specifically, the
addition of ankle joint mobilization (8 sessions over 4
33.90 § 10.99
66.06 § 12.77
37.86 § 2.93

30.84 § 6.81
30.43 § 9.59
56.64 § 9.58

weeks) was shown to significantly increase ankle ROM

06.64 § 1.2
0.96 § .16

and decrease the time required to transition from a seated

Pretest

to a standing position as compared with functional task

practice alone in patients with stroke-induced hemipare-
sis.6 Furthermore, Jeon et al12 also demonstrated that
Unaffected

ankle self-stretching using a strap significantly increased

Affected

the ankle DF-ROM, passive ankle DF-ROM, and lunge

angle compared with static muscle stretching alone in
patients with limited ankle DF. This recovery of accessory
movement at the talocrural joint was largely facilitated by
Gait speed (cm/sec)
Cadence (step/min)

the application of a posteroanterior glide of the tibia on

Step length (cm)

K-MBI (score)
DF-PROM (˚)

the talus during ankle MWM and through a possible gas-

SBA (score)
BBS (score)
Parameters

trocsoleus complex stretch obtained through sustained

and repeated maximal DF-ROM during MWM.9,10 In the
present study, S-MWM exercise uses a strap to facilitate
†

posterior glide of the talus during closed-chain

 ARTICLE IN PRESS
6 D. PARK ET AL.

dorsiflexion activities.11,12 Also, during a lunge, the mid-
dle of the heel and the second toe were aligned directly
over a taped line to minimize subtalar pronation and
other compensatory movements.
In addition, the 10° incline board facilitates easy
application of a posterior-inferior gliding force to the
talus through the strap during lunge position because
it permits the use of a greater stretching force to the
ankle plantar flexors. It may even provide a greater
stretching force than a stool. These factors likely con-
tribute to the significant increases in ankle DF-PROM
in the S-MWM10° arm relative to that in the S-MWM
arm. Therefore, our results indicate that S-MWM10°
combined with the standard rehabilitation is likely far
more effective than the traditional S-MWM combined
with the standard rehabilitation.
After the completion of the study interventions, SBA
was significantly improved by 82.1% in the S-
MWM10° arm relative to the S-MWM arm. Again,
these results are consistent with those of several previ-
ous studies.7,23 Mecagni et al23 reported a decreased
performance in balance testing associated with
restricted ankle motion, likely attributable to the pres-
ence of noncontractile tissues (eg, capsule, ligaments,
and bone) in elderly, community-dwelling women.
Furthermore, An and Jo7 examined the effects of ankle
MWM on both ankle DF-PROM and weight-bearing
ability during standing or gait in stroke patients with
limited ankle mobility. Their results indicated that
MWM combined with the standard rehabilitation
improved both ankle kinetics and weight-bearing
capabilities in the chronic stroke patients. These find-
ings suggest significant improvements in the ability of
the paretic lower limb to shift and support body
weight during standing or walking. Therefore, in the
present investigation, the 4-week intervention of joint
mobilization may help to stretch the capsule and liga-
ments of the ankle, possibly increasing mechanorecep-
tor activity or improving their sensory output as
gamma motor neurons are activated with tissue trac-
tion. Indeed, the classical functions of these stretch
receptors, Golgi tendon organs, and other propriocep-
tors are to aid in the maintenance of organismal bal-
ance.24,25 Furthermore, the 10° inclined board in
combination with the strap may facilitate the applica-
tion of the required posterior-inferior directional force
at the talus during S-MWM10°.12 As a result, S-
MWM10° may enhance mechanoreceptor activity

Figure 1. Flowchart of the participant selection procedure. Abbreviations: DF-PROM, dorsiflexion passive range of motion; K-MBI, the Korean version of the
modified Barthel index; S-MWM, self-mobilization with movement; S-MWM10°, self-mobilization with movement with a 10° inclined board.
 ARTICLE IN PRESS
SELF-ANKLE MOBILIZATION WITH MOVEMENT ON PATIENTS WITH CHRONIC STROKE
relative to S-MWM alone, possibly explaining the
larger increases in SBA. Together, our results indicate
that treatment with S-MWM10° combined with the
standard rehabilitation is demonstrably more effica-
cious for improving SBA.
Following 4 weeks of the trial protocol, gait speed,
cadence, and affected side step length were significantly
improved in the S-MWM10° arm as compared with the S-
MWM arm by 102.6%, 128.1%, and 61.8%, respectively.
Similarly, our results mirror those of previous
Figure 2. Application of self-ankle mobilization with movement (top row: start to end position) and self-mobilization with movement with a 10° inclined board
(bottom row: start to end position).
7
investigations.26,27 Teixeira-Salmela et al26 reported that
chronic stroke patients demonstrated significant improve-
ments in gait velocity, cadence, and stride length after 10
weeks of therapy, further reporting improvements in
walking pattern. Specifically, these improvements were
determined by increases in both plantar flexor and hip
flexor/extensor muscle power generation, and by positive
work performed by the hip and ankle joint ROM. Christi-
ansen27 demonstrated that a hip and ankle flexibility pro-
gram significantly increased hip and knee ROM, ankle
 ARTICLE IN PRESS
8 D. PARK ET AL.
ROM, and gait speed relative to a control group of elderly,
community-dwelling patients. These results suggest that
compared with control patients, patients who were receiv-
ing hip and ankle flexibility therapy presented with
increased combined hip and knee ROM, ankle ROM, and
gait speed. The increased hip and ankle joint passive
motions were further associated with increases in gait
speed, indicating a possible link between joint motion and
gait speed. Mechanistically, movement limitations of the
ankle joint prolong the preswing phase of the gait, leading
to inadequate momentum, and poor progression of the
affected limb in patients with slow gait velocity. In turn,
this causes the development of compensatory mecha-
nisms to correct for these deficiencies.6,27 Generally, stroke
patients are treated for selective motor control, focused
primarily on muscle strengthening and mobility. Further-
more, strategies for patients with poor motor recovery
should focus on isolated, selected joint movement training
in an effort to improve patient gait pattern.6,28 The antero-
posterior glide component of the weight-bearing MWM
may reduce any positional fault, minimize anterior dis-
placement of the talus, and restore normal ankle-joint
kinematics and accessory movement.9,10 Both S-MWM
and S-MWM10° are performed in a single-leg lunge
stance, followed by a forward weight shift. This single-leg
lunge causes a muscular contraction of the gastrocnemius,
including an eccentric contractile component, resulting in
plantar flexor contraction and thus improved plantar
flexor strength.12,29 The 10° inclined board used by the
participant performing S-MWM10° may be a useful tool
to control the pulling-force angle of the strap during the
forward shift of the participant's body weight while in the
lunge stance. As a result, the 10° inclined board likely pro-
vides a greater stretching force than that achieved using a
stool alone. This greater stretching force may be more
effective for the stimulation of ankle dorsiflexor and plan-
tar flexor contractions, likely producing the observed
increases in gait speed, cadence, and affected-side step
length. Therefore, our results indicate that S-MWM10°
combined with the standard rehabilitation may be more
effective with respect to enhancing gait speed, cadence,
and affected-side step length.
The K-MBI scores were not significantly different
between the S-MWM and S-MWM10° arms. However,
the K-MBI scores were significantly improved after treat-
ment relative to the baseline scores in both arms. This
result is consistent with those in previously published
reports,30,31 both demonstrating enhanced balance and
gait parameters, and improvements in performance in
ADLs.
This study has several limitations. First, owing to the
small sample size and eligibility criteria, the findings
of this investigation may not be generalizable to all
stroke patients. Specifically, the patients enrolled in
this trial had chronic stroke accompanied by mild to
moderate physical impairment, likely not
representative of the overall stroke patient population.
Second, long-term follow-up was not performed, pre-
venting the investigation of any carryover effect of S-
MWM. Future investigations with extensive follow-up
periods are required to elucidate these effects, if any.
Finally, this study did not assess specific outcomes
related to either lower extremity proprioception or
active ankle ROM. Future studies will be required to
appropriately address these specific issues.
In conclusion, our findings indicate that S-MWM10°
combined with the standard rehabilitation arm demon-
strated more-significant improvements in ankle DF-
PROM, SBA, gait speed, cadence, and affected-side step
length relative to the S-MWM combined with the stan-
dard rehabilitation arm after 4 weeks of study interven-
tion. Taken together, these findings will likely serve as a
foundation for the development of novel therapeutic
interventions and strategies for patients with chronic
stroke.
References
1. Wade DT. Measurement in neurological rehabilitation.
Curr Opin Neurol 1992;5:682-686.
2. Saltzman CL, Nawoczenski DA. Complexities of foot
architecture as a base of support. J Orthop Sports Phys
Ther 1995;21:354-360.
3. Chung SG, van Rey E, Bai Z, et al. Biomechanic changes
in passive properties of hemiplegic ankles with spastic
hypertonia. Arch Phys Med Rehabil 2004;85:1638-1646.
4. Langhorne P, Coupar F, Pollock A. Motor recovery after
stroke: a systematic review. Lancet Neurol 2009;8:741-
754.
5. Michael KM, Allen JK, Macko RF. Reduced ambulatory
activity after stroke: the role of balance, gait, and cardio-
vascular fitness. Arch Phys Med Rehabil 2005;86:1552-
1556.
6. Kluding PM, Santos M. Effects of ankle joint mobiliza-
tions in adults poststroke: a pilot study. Arch Phys Med
Rehabil 2008;89:449-456.
7. An CM, Jo SO. Effects of talocrural mobilization with
movement on ankle strength, mobility, and weight-bear-
ing ability in hemiplegic patients with chronic stroke: a
randomized controlled trial. J Stroke Cerebrovasc Dis
2017;26:169-176.
8. Nakayama Y, Iijima S, Kakuda W, et al. Effect of home-
based training using a slant board with dorsiflexed
ankles on walking function in post-stroke hemiparetic
patients. J Phys Ther Sci 2016;28:2353-2357.
9. Vicenzino B, Branjerdporn M, Teys P, et al. Initial
changes in posterior talar glide and dorsiflexion of the
ankle after mobilization with movement in individuals
with recurrent ankle sprain. J Orthop Sports Phys Ther
2006;36:464-471.
10. Mulligan B. Manual therapy “NAGS”,“SNAGS”,“MWMS”.
5th ed Wellington, New Zealand: Plane View Service; 2004.
11. Sahrmann S. Movement system impairment syndromes
of the extremities, cervical and thoracic spines. Elsevier
Health Sci 2010;1:439-509.
12. Jeon IC, Kwon OY, Yi CH, et al. Ankle-dorsiflexion range
of motion after ankle self-stretching using a strap. J Athl
Train 2015;50:1226-1232.
 ARTICLE IN PRESS
SELF-ANKLE MOBILIZATION WITH MOVEMENT ON PATIENTS WITH CHRONIC STROKE
13. Hall T, Chan HT, Christensen L, et al. Efficacy of a C1-C2
self-sustained natural apophyseal glide (SNAG) in the
management of cervicogenic headache. J Orthopaedic
Sports Phys Ther 2007;37:100-107.
14. Kim S-L, Lee B-H. Effect of Mulligan's mobilization with
movement technique on gait function in stroke patients.
J Phys Ther Sci 2016;28:2326-2329.
15. Flansbjer UB, Downham D, Lexell J. Knee muscle
strength, gait performance, and perceived participa-
tion after stroke. Arch Phys Med Rehabil 2006;87:
974-980.
16. Ada L, Dean CM, Vargas J, et al. Mechanically assisted
walking with body weight support results in more inde-
pendent walking than assisted overground walking in
non-ambulatory patients early after stroke: a systematic
review. J Physiotherapy 2010;56:153-161.
17. Kang Y, Na DL, Hahn S. A validity study on the Korean
Mini-Mental State Examination (K-MMSE) in dementia
patients. J Korean Neurol Assoc 1997;15:300-308.
18. Testerman C, Vander Griend R. Evaluation of ankle insta-
bility using the Biodex Stability System. Foot Ankle Int
1999;20:317-321.
19. Berg K, Wood-Dauphinee S, Williams J. The Balance
Scale: reliability assessment with elderly residents and
patients with an acute stroke. Scand J Rehabil Med
1995;27:27-36.
20. Menz HB, Latt MD, Tiedemann A, et al. Reliability of the
GAITRiteÒ walkway system for the quantification of tem-
poro-spatial parameters of gait in young and older peo-
ple. Gait Posture 2004;20:20-25.
21. Jung HY, Park BK, Shin HS, et al. Development of the
Korean version of Modified Barthel Index (K-MBI):
multi-center study for subjects with stroke. J Korean
Acad Rehabil Med 2007;31:283-297.
9
22. Cohen J. Statistical power analysis. Curr Dir Psychol Sci
1992;1:98-101.
23. Mecagni C, Smith JP, Roberts KE, et al. Balance and ankle
range of motion in community-dwelling women aged 64
to 87 years: a correlational study. Phys Ther 2000;80:1004.
24. Hiller CE, Refshauge KM, Bundy AC, et al. The Cumber-
land ankle instability tool: a report of validity and reli-
ability testing. Arch Phys Med Rehabil 2006;87:1235-1241.
25. Nashner L. Adapting reflexes controlling the human pos-
ture. Exp Brain Res 1976;26:59-72.
26. Teixeira-Salmela LF, Nadeau S, Mcbride I, et al. Effects of
muscle strengthening and physical conditioning training
on temporal, kinematic and kinetic variables during gait
in chronic stroke survivors. J Rehabil Med 2001;33:53-60.
27. Christiansen CL. The effects of hip and ankle stretching
on gait function of older people. Arch Phys Med Rehabil
2008;89:1421-1428.
28. Chen CL, Chen HC, Tang SF, et al. Gait performance with
compensatory adaptations in stroke patients with differ-
ent degrees of motor recovery. Am J Phys Med Rehabil
2003;82:925-935.
29. J€
onhagen S, Halvorsen K, Benoit DL. Muscle activation
and length changes during two lunge exercises: implica-
tions for rehabilitation. Scand J Med Sci Sports 2009;19:
561-568.
30. Kang TW, Lee JH, Cynn HS. Six-week nordic treadmill
training compared with treadmill training on balance,
gait, and activities of daily living for stroke patients: a
randomized controlled trial. J Stroke Cerebrovasc Dis
2016;25:848-856.
31. Macko RF, Benvenuti F, DrEng VM, et al. Adaptive
physical activity improves mobility function and qual-
ity of life in chronic hemiparesis. J Rehabil Res Dev
2008;45:323.