Content uploaded by Edward Draper
Author content
All content in this area was uploaded by Edward Draper on Apr 11, 2015
Content may be subject to copyright.
VOL. 82-B, N
O
. 7, SEPTEMBER 2000 1001
E. R. C. Draper, PhD, MIMechE, CEng, Principal Clinical Bioengineer
J. Sanchez-Ballester, FRCS Ed, Specialist Registrar
N. Hunt, FRCS, Specialist Registrar
J. R. Robinson, MRCS, Specialist Registrar
Department of Trauma and Orthopaedic Surgery, Imperial College School
of Medicine, 7th Floor East, Charing Cross Hospital, Fulham Palace Road,
London W6 8RF, UK.
R. K. Strachan, FRCS Ed, Consultant Orthopaedic Surgeon
J. M. Cable, RGN, P Tech Cert, Nurse Practitioner
Department of Orthopaedics, Ealing Hospital, Uxbridge Road, Southall,
Middlesex UB1 3HW, UK.
Correspondence should be sent to Dr E. R. C. Draper.
©2000 British Editorial Society of Bone and Joint Surgery
0301-620X/00/710638 $2.00
Improvement in function after valgus bracing
of the knee
AN ANALYSIS OF GAIT SYMMETRY
E. R. C. Draper, J. M. Cable, J. Sanchez-Ballester, N. Hunt,
J. R. Robinson, R. K. Strachan
From Imperial College School of Medicine and Ealing Hospital, London, England
T
he use of a valgus brace can effectively relieve the
symptoms of unicompartmental osteoarthritis of
the knee. This study provides an objective
measurement of function by analysis of gait symmetry.
This was measured in 30 patients on four separate
occasions: immediately before and after initial fitting
and then again at three months with the brace on and
off. All patients reported immediate symptomatic
improvement with less pain on walking. After fitting
the brace, symmetry indices of stance and the swing
phase of gait showed a consistent and immediate
improvement at 0 and 3 months, respectively, of
3.92% (p = 0.030) and 3.40% (p = 0.025) in the stance
phase and 11.78% (p = 0.020) and 9.58% (p = 0.005) in
the swing phase. This was confirmed by a significant
improvement at three months in the mean Hospital
for Special Surgery (HSS) knee score from 69.9 to
82.0 (p < 0.001). Thus, wearing a valgus brace gives a
significant and immediate improvement in the function
of patients with unicompartmental osteoarthritis of
the knee, as measured by analysis of gait symmetry.
J Bone Joint Surg [Br] 2000;82-B:1001-5.
Received 7 October 1999; Accepted after revision 7 April 2000
Osteoarthritis of the knee occurs more often in the medial
compartment,
1
presumably due to its increased loading
during gait. The load on the knee during the stance phase is
a combination of compression due to body mass and, with
the exception of a fleeting abduction at initial stance, an
adduction moment.
1
The consequence of this external
moment is that the centre of pressure within the knee
moves medially, increasing the load on the medial
compartment.
Traditional conservative methods of treatment, such as
weight loss, walking aids, wearing a heel wedge
2,3
and
analgesia, may temporarily ameliorate symptoms improv-
ing function and walking distance, but they fail to address
the underlying pathology. Many of these patients will
ultimately require total knee arthroplasty. Although osteo-
arthritis can be treated surgically, there are many patients
for whom this is inappropriate, because of medical com-
orbidity, old age or other circumstances. In young patients
it is desirable to delay primary arthroplasty. The object of
other surgical treatments, such as high tibial osteotomy or
unicompartmental arthroplasty, is to transfer the load to the
less affected lateral compartment.
4-7
Such procedures carry
the usual surgical risks, increase the possibility of compli-
cations for the subsequent arthroplasty
8
and are not suitable
for patients of all ages.
5
The concept of a valgus brace is to apply, during stance,
an external abduction moment to the knee, which directly
opposes the usual adductor moment and should unload the
degenerative medial compartment. The term ‘valgus brace’
derives from this applied abduction, or valgus, moment.
Horlick and Loomer
9
and Pollo et al
1
have shown that
symptomatic relief can be provided by these braces and
Lindenfeld et al
3
found that valgus braces could reduce the
adductor moment around the knee. It is clear that wearing a
valgus brace may delay the requirement for surgery in
relatively young patients.
1,3,9
The study of Kirkley et al
10
compared the results of a prospective, randomised clinical
trial using three types of treatment: medical treatment only
or with a neoprene sleeve or a valgus brace. Their results
indicated that patients who have osteoarthritis of the medial
compartment may benefit significantly from the use of a
knee brace in addition to standard medical treatment. The
group with valgus braces had, on average, significantly
better outcome scores than that with neoprene sleeves.
In these studies no objective measurements of function
when using a valgus brace were made. This was the aim of
the present study. The relatively new technique of gait
symmetry analysis (GSA)
11
uses a treadmill with left and
right forceplates under a belt to record consecutive steps
(usually between 20 and 60). Several parameters can be
calculated from each step, including the duration of stance
and swing phases. The measured parameters are then used
to compare the braced with the contralateral limb. A direct
ratio, the symmetry index, has been found to be the most
useful and easily understood method of presenting the
results. It defines perfect symmetry and shifts either
towards or away from it. An important advantage over
functional scoring such as the Hospital for Special Surgery
(HSS) knee score, is that GSA can measure changes
rapidly.
Patients and Methods
The patients were measured and assessed immediately
before fitting the brace. All had received conservative
treatment only. These data were taken as the control for
each patient and consequently, paired comparisons between
visits were possible, allowing the use of more sensitive
statistics (paired Student t-test). This trial was therefore a
prospective controlled comparison of function before and
after use of the brace.
We excluded patients who were over 70 or under 35
years of age and those who had a varus deformity greater
than 12° or a fixed flexion deformity greater than 10°,
significant hip, back or contralateral leg symptoms, an
arthroscopy of the knee within the preceding six months
and physical or mental inability to comply with the require-
ments of wearing a brace.
There were 18 men and 12 women with a mean age of
56.2 years (35 to 70) who attended a specialist knee clinic
between October 1997 and May 1999. They had radio-
logically demonstrable osteoarthritis of the medial compart-
ment of the knee (Larsen grades II to IV) on anteroposterior
and Schusse radiographs with relative sparing of the lateral
compartment (Larsen grade 0 to I) which was consistent
with their symptoms.
Each patient was fitted with a custom-made valgus brace,
the GII ADJ Unloader (GII Orthotics Europe, Belgium).
This comprises two semirigid plastic shells, for the thigh
and calf, linked by a polyaxial medial hinge set at 4° of
valgus. An adjustable tension strap crosses the lateral
aspect of the knee from below, posterolaterally to above,
anteromedially, from the calf to the thigh shells (Fig. 1).
The same technician obtained the negative casts, fitted
the braces and instructed the patients in the proper use and
care of the brace. The patients were measured and assessed
before and immediately after fitting the brace and again
after continuous wear of the brace for three months.
Instrumented treadmill. A standard running machine
(PowerJog 200; Sport Engineering, Birmingham, UK) was
modified to house two independent forceplates, one placed
on the left and the other on the right of the bed. Each plate
was supported at each corner by an electrical resistance
strain-gauge transducer designed to be sensitive only to the
vertical component of force. The signals from each trans-
ducer were amplified and digitised by a standard laboratory
system (National Instruments Corp, Austin, Texas). A pro-
gram written within LabView (National Instruments Corp)
on a personal computer recorded the signals for later
analysis.
Walking tests were performed only if the subject felt
confident enough to start and complete the exercise. Sub-
jects were asked to walk on the instrumented treadmill,
wearing their normal shoes (Fig. 2). Both the subject and
the operator were in a position to operate the controls at all
times. During the first test, patients selected their own
preferred walking speed; this was recorded and used for all
subsequent testing of that patient. Four tests were per-
formed, one immediately before and one immediately after
the initial fitting of the brace; the other two were performed
consecutively, and within minutes of each other, with and
1002 E. R. C. DRAPER, J. M. CABLE, J. SANCHEZ-BALLESTER, N. HUNT, J. R. ROBINSON, R. K. STRACHAN
THE JOURNAL OF BONE AND JOINT SURGERY
Fig. 1a Fig. 1b
Photograph (a) and diagram (b) showing the valgus
brace.
without the brace, after three months. Each test lasted for
one minute, during which time between 20 and 60 steps
were recorded.
Analysis began with a visual inspection of the data to
remove any invalid steps if the patient had faltered. A
macro was incorporated into the system which marked for
scrutiny, and possible exclusion, any step that exceeded two
SD
s of the mean for any measured parameter.
A series of Visual Basic macros in Microsoft Excel was
used in the analysis. They automatically detected the heel
strike and toe-off, thereby isolating each step. After this the
stance phase and swing times, among other parameters,
were calculated for each step, both left and right. Symmetry
indices (SI) were then calculated for the stance and swing
phases. An SI is defined as the ratio of the means of the
parameter measured from the affected limb divided by that
of the unaffected limb. These can be calculated for any
measurement; those for stance and swing phase alone were
used in this study.
Any change in an SI can easily be seen and understood.
An SI of 1 indicates a perfectly symmetrical gait. A shift
towards 1 with time indicates an improvement, whereas a
shift away indicates deterioration.
Clinical assessment. Patients were assessed clinically
using visual analogue pain scores for resting, standing,
walking and climbing stairs. A modified HSS knee score,
activity questionnaires and requirement for analgesia were
also recorded. The radiological grade of osteoarthritis in
each of the three compartments was noted and the mechan-
ical femorotibial axis was determined from full-length
weight-bearing radiographs.
GSA compared with HSS score. In order to validate the
use of GSA, HSS scores were calculated initially and after
three months. The initial score before bracing was com-
pared with the SI taken at the same time. The scores at the
second visit were assessed from activity during the pre-
vious three months in which the brace was worn. Conse-
quently, comparison at this point was with the SI while
wearing the brace and with that at the first visit. It can be
seen in Figure 3 that the three outcome measurements show
an improvement; in all cases this is significant (p < 0.05).
Although the greatest change in the mean improvement
1003IMPROVEMENT IN FUNCTION AFTER VALGUS BRACING OF THE KNEE
VOL. 82-B, N
O
. 7, SEPTEMBER 2000
Fig. 2
GSA walking test (using a modified, standard running
machine).
Fig. 3
Validation of symmetry indices by comparison with HSS score (mean ±
SD
).
appears in the HSS score, the differences at three months
between the measurements are not significant (p > 0.05). It
was felt therefore that the SI had a similar capacity to
detect any underlying change in function as the HSS score.
The advantage of GSA was that it could be performed
immediately before and after fitting the brace, whereas the
modified HSS scoring system requires the passage of time
between tests.
Results
At the initial examination the mean modified HSS score
was 69.93 (47 to 87). All patients reported moderate to
severe pain on walking, with a mean of 7.57 on the ten-
point visual analogue score for this activity. The mean
score for resting and standing was less at 2.30 (0 to 7) and
4.00 (0 to 7), respectively. All patients had Larsen grade II
or III (24 and 6, respectively) in the medial compartment
and grade 0 or I (16 and 14, respectively) in the lateral
compartment. All patients except one had an abnormal
varus mechanical axis at the knee (one patient had a valgus
mechanical axis of 2°) with a mean of 4.82° of varus
ranging from -2° (valgus axis) to +12°. All patients report-
ed subjective improvements in their symptoms, with less
pain on walking on the treadmill with the brace fitted.
With an antalgic gait, it can be expected that the subject
will tend to spend less time on the affected limb. This will
lead to a stance-phase SI of below 1 and a swing-phase SI
above 1, both of which have been observed in this study. A
more symmetrical gait, which is assumed to be an improve-
ment, will result in each SI shifting towards 1. This is
summarised in Table I and clearly demonstrated in Figure
4; the change in both indices is significant (p < 0.05). The
mean HSS score also improved significantly from 69.93
(±9.90
SD
) at the initial visit, to 82.04 (±10.65
SD
) at the
three-month visit.
Discussion
Patients with unicompartmental osteoarthritis of the knee
can be helped by the use of a valgus brace. Clinical studies
of this
2,9,10
have relied on subjective scoring systems. The
objective outcome measurements made in our study sup-
port these previous findings and show that the improvement
in function begins as soon as the brace is worn. Figure 4
clearly shows this improvement, with a significant shift
towards a value of 1 (perfect symmetry) for both the stance
and swing phases of gait. This effect is detectable at the
first fitting and is maintained three months later.
Although this shows that this design of brace is clinically
effective, it sheds little light on the mechanism. It is clear
that without the brace our patients tend to bear weight for
longer on the unaffected limb, while sparing the affected
limb for longer. This is typical of an antalgic gait. The
return to a more symmetrical pattern indicates less
discomfort.
There are, however, many parameters other than pain
which may also disturb the pattern of gait. These include
changes in proprioception and a lack of confidence in the
affected limb. Despite the fact that our study does not
1004 E. R. C. DRAPER, J. M. CABLE, J. SANCHEZ-BALLESTER, N. HUNT, J. R. ROBINSON, R. K. STRACHAN
THE JOURNAL OF BONE AND JOINT SURGERY
Table I. Mean (±
SD
) stance- and swing-phase symmetry indices
Initial visit Three-month visit
No brace Braced No brace Braced
Stance phase 0.97 ± 0.11 1.02 ± 0.10 0.97 ± 0.11 1.01 ± 0.12
p=0.030 p = 0.025
Swing phase 1.13 ± 0.32 1.01 ± 0.17 1.12 ± 0.29 1.02 ± 0.18
p=0.020 p = 0.005
Fig. 4
Symmetry indices with and without
valgus brace (mean ±
SD
).
explain the mechanisms which are responsible for improv-
ing symptoms, it does add objective evidence that valgus
bracing for unicompartmental osteoarthritis of the knee is
clinically effective.
Further work is required to determine if the effectiveness
of this type of brace is reduced with time, to investigate the
alteration in internal load of the knee when the brace is worn
and to determine whether there are biological consequences
of unloading a degenerative area of articular cartilage.
We thank Mr Shalaby and Mr Thomas for their support in this trial,
Professor Hughes for his tireless enthusiasm and Action Research for
providing the funding for the treadmill modifications (Project Number A/
P/0451).
One of more of the authors have received or will receive benefits for
personal or professional use from a commercial party related directly or
indirectly to the subject of this article.
References
1. Pollo FE, Otis JC, Wickiewicz TL, Warren RF. Biomechanical
analysis of valgus bracing for the osteoarthritic knee. Gait Posture
1994;2:63-5.
2. Pollo FE. Bracing and heel wedging for unicompartmental osteo-
arthritis of the knee. Am J Knee Surg 1998;11:47-50.
3. Lindenfeld TN, Hewett TE, Andriacchi TP. Joint loading with
valgus bracing in patients with varus gonarthrosis. Clin Orthop
1997;344:290-7.
4. Coventry MB. Upper tibia osteotomy for gonarthrosis: the evolution
of the operation in the last 18 years and long term results. Orthop Clin
North Am 1979;10:191-210.
5. Insall JN, Joseph DM, Msika C. High tibial osteotomy for varus
gonarthrosis: a long-term follow-up study. J Bone Joint Surg [Am]
1984;66-A:1040-8.
6. Noyes FR, Barber SD, Simon R. High tibia osteotomy and ligament
reconstruction in varus angulated, anterior cruciate ligaments-deficient
knees: a two- to seven-year follow-up study. Am J Sports Med
1993;21:2-12.
7. Noyes FR, Simon R. The role of high tibial osteotomy in the anterior
cruciate ligament-deficient knee with varus alignment. In: Orthopedic
sports medicine. principles and practice. Philadelphia: WB Saunders,
1993:1401-43.
8. Brooks PJ, Walker PS, Scott RD. Tibia component fixation in
deficient tibia bone stock. Clin Orthop 1984;184:302-8.
9. Horlick SG, Loomer RL. Valgus knee bracing for medial gon-
arthrosis. Clin J Sports Med 1993;3:251-5.
10. Kirkley A, Webster-Bogaert S, Litchfield R, et al. The effect of
bracing on varus gonarthrosis. J Bone Joint Surg [Am]
1999;81-A:539-48.
11. Draper ERC. A treadmill-based system for measuring symmetry of
gait. Med Eng Phys 2000; in press.
1005IMPROVEMENT IN FUNCTION AFTER VALGUS BRACING OF THE KNEE
VOL. 82-B, N
O
. 7, SEPTEMBER 2000
