A Dose-Response Relationship Between Amount of Weight-

Bearing Exercise and Walking Outcome Following

Cerebrovascular Accident
Julie A. Nugent, BAppSc (Phty), Karl A. Schurr, BAppSc (Phty), Roger D. Adams, PhD

ABSTRACT. Nugent JA, Schurr KA, Adams RD. A dose-response relationship between amount of weight-
bearing exercise and walking outcome following cerebrovascular accident. Arch Phys Med Rehabil 1994;75:
399-402.
l To supplement a motor relearning approach to rehabilitation for patients following cerebrovascular accident
(CVA), a weight-bearing exercise was used that was designed specifically to strengthen the leg extensor muscles
to simulate the output required in the single support phase of walking. For patients who were initially able to
stand on their affected leg and step forward with their other leg, a dose-response relationship was found between
an increasing number of repetitions of this weight-bearing exercise and improved walking outcome as measured
on the Motor Assessment Scale (MAS) for stroke. For patients who could not stand and step forward at initial
assessment (MAS-0), there was no relationship between the amount of practice done and walking outcome,
however, all of the patients who practiced the exercise achieved independent walking for at least a 3m distance,
which gave a final MAS score of three or greater.
0 1994 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and
Rehabilitation

Independent walking after stroke rehabilitation has been
reported as varying from 26.7%’ to 75%.* In a study compar-
ing the outcomes of three different physiotherapy ap-
proaches, Dickstein, and colleagues3 found that at discharge
a total of 50.3% of their patients were able to walk indepen-
dently, with 44.2% requiring the use of a walking aid. No
difference was found between the outcomes of the different
therapies. Partridge, and colleagues4 attempted to establish
normative data for outcomes after stroke. They found that
53.3% of a group of 368 stroke patients were able to walk
independently indoors after 8 weeks. A recent study of stroke
outcome using a Motor Relearning Programme for Stroke5
reported that 72.3% of the patients reported on improved
their walking performance on discharge from physiotherapy
and 61.5% of these patients were able to walk independently
for a minimum of 3m.6
Although it would now seem that more than 50% of stroke
patients might achieve independent walking using current
rehabilitation techniques, there is still room for improvement
in physiotherapy intervention. This may increase the percent-
age of patients who learn how to walk poststroke and in-
crease the skill of patients already able to walk. One way
to improve the current level of walking outcomes for patients
may be to include practice of a specific skill component of
walking.
This study investigates the influence of a weight-bearing
From the Stroke Unit (Ms. Nugent, Mr. Schurr), Lidcombe Hospital; and the
Behavioral Sciences Department (Dr. Adams). University of Sydney, Lidcombe, Syd-
ney, NSW.
Submitted for publication June 7, 1993. Accepted in revised form November 16.
1993.
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit upon the authors or upon any
organizations with which the authors are associated.
Reprint requests to Julie A Nugent, BAppSc (Phty), Physiotherapy Department,
Lidcombe Hospital, Joseph Street, Lidcombe, Sydney, NSW 2141.
0 1994 by the American Congress of Rehabilitation Medicine and the American
Academy of Physical Medicine and Rehabilitation
0003.9993/94/7504-0143$3.00/0
leg extensor exercise (WBE) on patients’ walking outcomes
when this was added to The Motor Relearning Programme
for Stroke.’ The rationale for selecting this leg extensor
exercise was based on the functional requirement of estab-
lishing an overall extensor moment on the stance leg when
walking.7.”
It has been observed clinically that stroke patients com-
monly have a decreased stance time on their affected leg
and consequently a shortened step length with their nonaf-
fected leg.’ This implies an impaired ability to generate an
appropriate support moment on their affected leg during sin-
gle support, thus decreasing the opportunity to step forward
with the nonaffected leg. It might be anticipated that the
skill level of walking would improve with the amount of
practice,‘O as would leg extensor strength.
The Motor Assessment Scale (MAS)” was used to mea-
sure change in walking ability. This is a seven point ordinal
scale that rates motor function for eight everyday tasks in-
cluding walking (see Appendix 1). The MAS scale was cho-
sen because it has face validity, has been shown to have
concurrent validity,” and the interrater and intrarater reli-
ability coefficients of the MAS have been assessed at r
= 0.98 and r = 0.95 respectively.” Using this scale, patients
who walk with a walking aid can only achieve a maximum
score of three, irrespective of their walking speed.
A score of four requires a minimum unaided walking
speed of 0.3m/sec for a distance of at least 5m, whereas a
score of five requires a minimum walking speed of O.Xm/
set for 20m as well as the ability to bend down, pick up a
sandbag from the floor, and turn around. The MAS is thus
able to distinguish between patients who require physical
assistance and/or walking aids, as well as those who walk
unaided with increasing levels of skill.
METHOD
The MAS scores for walking at the time of the initial
and final physiotherapy assessments were recorded for 109

Arch Phys Med Rehabil Vol 75, April 1994

400 A DOSE-RESPONSE RELATIONSHIP, Nugent

consecutive patients admitted to our stroke unit between

January and September 199 1.
Patients were accepted in the WBE program if they: were
admitted for inpatient rehabilitation during the time of the
study; had suffered a cerebrovascular accident (CVA) (pa-
tients who had previous CVAs were also included); had an
MAS score of greater than zero and less than six at the
initial physiotherapy assessment or at any time during their
inpatient rehabilitation; were able to perform the WBE inde-
pendently or with the assistance of one person and/or a table
or rail for support. (If more assistance than this was required,
the patients did not undertake the exercise until they could
attempt as noted previously); had given their verbal consent
to participate.
Twenty-five patients who had an MAS score of zero for
walking at both the initial and final physiotherapy assess-
ments were not able to practise the WBE during their hospital
stay, and could not be included. Ten patients who had an
MAS score of six for walking at the initial physiotherapy
assessment were also not asked to participate.
An additional 19 patients who met the inclusion criteria
did not participate because of pain while attempting the exer-
cise, inability to understand the instructions, or inability to
provide informed consent. Others simply declined to partici-
pate.
The group who practiced the WBE included 25 men and
29 women, with a mean age of 69.6 years (SD = 10.6). In
3 patients, both cerebral hemispheres were affected; in 24
patients the left hemisphere only was affected and in 26
patients the right hemisphere was affected. One patient did
not show a lesion on the computed tomography (CT) scan.
The mean time between stroke onset and admission to the
unit was 31 days, and the mean length of stay in the unit
was 41 days.

Description of the Weight-Bearing Exercise

The patient stood with the affected leg on a wooden block
with a minimum height of 5cm and a maximum height of
12cm (fig 1). Assistance was provided to achieve this posi-
tion if required. The patient then lifted their nonaffected leg
from the floor by extending their affected hip and knee,
without using their nonaffected leg to achieve foot clearance,
or permitting the affected knee to hyperextend. If necessary,
manual guidance was used to prevent hyperextension of the
affected knee.
A maximum of 60 repetitions of this exercise were done
per working day but no minimum number of repetitions was
specified. Some patients also practiced during weekends and
this was recorded. The number of repetitions per day and
the total number of both repetitions and practice days was
recorded by a physiotherapist. Practice continued until the
patient scored a 6 on the MAS scale for walking, or was
discharged from physiotherapy or from the unit.

RESULTS
Overall, there was a statistically significant and moder-
ately strong Pearson correlation between the number of repe-
titions of the WBE and the change in MAS score for walking, Fig l-The WBE being performed with physical assistance.
r = 0.45, p < .Ol. For patients with an initial non-zero

Arch Phys Med Rehabil Vol 75, April 1994

 A DOSE-RESPONSE RELATIONSHIP, Nugent 401

3000 I I I I I I I patients able to stand but not walk unaided (MAS initial of
one, two, or three) this positive relationship stood in contrast
to an absence of relationship between improvement and days
2500 available for practice, thereby discounting spontaneous re-
covery as a joint explanation for both MAS score and total
number of WBE repetitions. This is consistent with skill
i 2000 acquisition theory which notes that skill in performance is
.- . directly related to the amount of task-relevant practice John-
Y
Y . son3 and supports the function-relatedness of the WBE task
a,
k 1500 to the rehabilitation target of independent walking. Apart
L
.
from the obvious physical benefits of the WBE, the patients
0 . may also benefit psychologically from their participation,
$ 1000 . through improved self-esteem and body-perception on the
I-
. one hand and through improved ability to maintain task-
. . focused attention. Such effects as a result of exercise in the
500 neurologically disabled population have been reviewed by
; !
a . Davis and Glaser,” and Adams”’ respectively.
. : For the patients who could walk unaided on admission
0 1 T I I I I I
(MAS initial of 4 or 5) there was less scope to show improve-
0 1 2 3 4 5 6 ment using the MAS scale. Measurement of walking speed
may have been useful to show discrete changes not detected
MAS difference
by the measurement tool chosen. The moderate relationship
Fig 2-Scatterplot of the total number of repetitions of WBE between amount of improvement and days available for prac-
performed against the difference between initial and final MAS
tice may be a reflection of the shorter length of stay for this
scores for the 28 cases with an initial MAS score of greater
than zero. subgroup (14.9 days) compared with the whole group.
Improvement in walking ability in proportion to the
amount of practice was not seen in the patients who began
MAS (n = 28). the correlation between the amount of WBE
with an MAS score of zero. This may be accounted for in
practice and MAS outcome was 0.71 (fig 2).
part by the heterogeneity of this group, in that a single repeti-
When this group was further divided a strong relationship
tion was not of the same value for all patients. Included in
was also obtained for the subgroup of patients who had an
initial MAS score of either one, two, or three (n = 12), for this group were patients who were not able to initiate any
whom the Pearson correlation was r = 0.62, p < .03. A muscle activity in their affected leg, as well as those who
moderate relationship was obtained for patients (n = 16) could initiate muscle activity, but were unable to organize
who had an initial MAS score of four or five (r = 0.42, p
< .l). Thus for patients with an initial MAS score of one 3000 I I I
I I
.1 I
or greater, the “reward for work rule” held, ie, the more
WBE practice they did, the better outcome they tended to
have at discharge. 2500 .
There was no relationship between the number of repeti-
tions of the WBE done and the change in MAS score for t
the subgroup of patients with an initial MAS score of zero
(1. = -0.06). However, all of the patients who performed
the WBE scored three or more on the MAS for walking at
the final physiotherapy assessment. The scatterplot is pre-
sented in fig 3.
Finally, correlations were calculated between the number
of days on which practice could have taken place and the
change in MAS score for walking. This relationship was
significantly different from zero for patients with an initial
non-zero MAS (Y = .38, p < .05),but when this group was
separated into those with initial scores of one, two, or three
and those with initial scores of four or five the values were
r = -0.3 1 and r = 0.44 (p < .I) respectively.
0 1 2 3 4 5 6
DISCUSSION
For patients initially able to stand on their affected leg MAS difference
and step forward with their other leg (MAS 1-5) the results Fig 3-Scatterplot of the total number of repetitions of WBE
are quite straightforward; the more WBE practice done, the performed against the difference between initial and final MAS
better the associated walking outcome. For the group of scores for the 26 cases with an initial MAS score of zero.

Arch Phys Med Rehabil Vol75, April 1994

402 A DOSE-RESPONSE
it effectively to stand and step. The following two examples
illustrate the differences in patient ability in this group.
Subject one, who practiced 615 repetitions of the weight-
bearing exercise, showed the maximum possible change in
MAS score from 0 on admission to 6 on discharge. This
was achieved after 12 days of WBE practice. In contrast,
subject two practiced the WBE the most, performing a total
of 2,930 repetitions over 50 days, only achieved an MAS
walking score of 4 on discharge. This patient did not begin
to practice until 1 month after admission because she was
unable to generate a support moment in the affected leg until
that time. When practice began, maximal assistance was
required by the patient to perform the WBE and the quality
of her performance was poor. Thus, before she was able to
benefit from it, she first had to increase her ability to perform
it. By putting the subject in a situation which places a de-
mand on the leg extensor muscles this improvement could
occur. The necessity for some stroke patients to attempt
multiple repetitions of a specific task to improve perfor-
mance has been noted by Canning’6 in a study of standing-
up training.
However, the use of physical assistance for patients only
barely able to practice the WBE meant that for this group
all repetitions were not equivalent, and this may have masked
the relationship between the amount of practice done and
MAS outcome observed with patients who required no assis-
tance. Although it seems that the WBE may be a useful
adjunct to the current daily training poststroke, there remain
many patients who are unable to perform it. Researchers in
our facility are currently investigating simpler versions of
this exercise to increase the number of patients who may
benefit from it. Weight-bearing exercise in a semireclined
position on a tilt table is one possibility, as is taping the
affected knee to provide additional support during the early
phase of training.
Acknowledgments: The authors thank the physiotherapy staff who as-
sisted in the data collection for the study. Our thanks also go to Janet Carr
and Roberta Shepherd for consultation.
References
1. Matsamura S. Effect of physical therapy on elderly hemiplegic patients.
Proceedings of the Tenth International Congress World Confederation
for Physical Therapy, Sydney: 1987, 985-9.
2. Smith ME, Garraway WM. Smith DL, Akhtar AJ. Therapy impact on
functional outcome in a controlled trial of stroke rehabilitation. Arch
Phys Med Rehabil 1982;63:21-4.
3. Dickstein R, Hocherman S, Pillar T, Shaham R. Stroke rehabilitation:
three exercise therapy approaches. Phys Ther 1986;66:1233-8.
Arch Pbya Med Rehabil Vol75, April 1994
RELATIONSHIP, Nugent
4. Partridge CJ, Johnston M, Edwards S. Recovery from physical disabil-
ity after stroke: normal patterns as a basis for evaluation. Lancet
1987; 14:373-5.
5. Carr JH, Shepherd RA. A motor relearning programme for stroke. 2nd
ed. London: Heinemann, 1987.
6. Dean C. Mackev F. Motor assessment scale scores as a measure of
rehabilitation o&ome following stroke. Aust J Physiother 1992;38:31-
5.
Winter DA. Overall principle of lower limb support during stance phase
of gait. J Biomech 1980; 13:923-7.
Winter DA. The Biomechanics and motor control of human gait. On-
tario: University of Waterloo, 1987.
Lehmann JF, Condon SM. Price R, deLateur BJ. Gait abnormalities in
hemiplegia: their correction by ankle-foot orthoses. Arch Phys Med
Rehabil 1987;68:763-71.
10. Gentile AM. Skill acquisition: Action, movement and neuromotor pro-
cesses. In: Carr JH. Shepherd RB, editors. Movement science: founda-
tions for physical therapy in rehabilitation. Rockville, MD: Aspen,
1987, 93-1.54.
11. Carr JH, Shepherd RB, Lynne D, Nordholm L. Investigation of a new
motor assessment scale for stroke patients. Phys Ther 1985;65:175-80.
12. Poole LJ, Whitney SL. Motor assessment scale for stroke patients
concurrent validity and interrater reliability. Arch Phys Med Rehabil
1988:69:195-7.
13. Johnston P. The acquisition of skill. In: Smyth M, Wing A editors.
The psychology of human movement. London: Academic, 1984, 215-
39.
14. Davis G, Glaser R. Cardiorespiratory fitness following spinal cord in-
jury. In: Ada L, Canning C editors. Key issues in neurological physio-
therapy. Oxford: Heinemann, 1990, 155-96.
15. Adams R. Attention control training and behaviour management. In:
Ada L, Canning C, editors. Key issues in Neurological Physiotherapy,
Oxford: Heinemann, 1990, 81-98.
16. Canning C. Training standing up following stroke: a clinical trial. Pro-
ceedings of WCPT Congress, Sydney, Australia. World Confederation
for Physical Therapy, London: 1987, 915-19.
APPENDIX
Guidelines for Scoring Walking on the Motor
Assessment Scale”
1. Stands on affected leg and steps forward with other
leg. (Weighbearing hip must be extended. Therapist
may give standby help).
2. Walks with standby help from one person.
3. Walks 3m alone or using any aid but with no standby
help.
4. Walks 5m with no aid in 15 seconds.
5. Walks 1Om with no aid, turns around, picks up a small
sandbag from the floor and walks back in 25 seconds.
(May use either hand).
6. Walks up and down four steps with or without an aid,
but without holding on to the rail three times in 35
seconds.