Session 42
 - Osteoarthritis II
- VALENCIA B & C, Wed 8:00 AM - 10:00 AM
0262 46th Annual Meeting, Orthopaedic Research Society, March 12-15, 2000, Orlando, Florida

VALGUS BRACING FOR KNEE OSTEOARTHRITIS:
A BIOMECHANICAL AND CLINICAL OUTCOME STUDY
+*Otis, J C (A-Bledsoe Brace Systems and the Clark, Dana and Friese Foundations); *Backus, S I; *Campbell, D A; *Furman, G L (A-Bledsoe Brace Systems);
*Garrison, G (A-Bledsoe Brace Systems); *Warren, R F; *Wickiewicz, T L
+*Hospital for Special Surgery, New York, NY. 535 East 70th St./New York, NY 10021, 212 606-1088, Fax: 212 606-1490, otisj@hss.edu
Introduction:Valgus bracing has been used as a treatment for patients with
medial compartment OA for several years with variable amounts of pain
relief, functional improvement, and changes in medial joint loads. The basic
biomechanical philosophy is to apply a valgus load or correction about the
knee to partially offset the varus moment that the knee is subjected to during
the stance phase of gait. In doing so, joint compressive load is shifted from
the medial compartment to the lateral compartment. The efficacy of valgus
bracing using the Generation II Unloader brace has been previously
documented in an instrumented study of the brace in subjects with isolated
medial compartment osteoarthritis[1]. Patients reported a decrease in pain and
improvement in the performance of functional activities. The degree of pain
relief and functional improvement varied across subjects, however, as did
changes in the calculated medial joint loads.
The common denominator in the design of these braces is that, in theory, they
apply a valgus load about the knee. It is unclear, however, whether or not the
applied valgus load is actually a factor responsible for the clinical success of
valgus bracing. Other factors, e.g., the additional stability offered by the
brace or increased proprioception, may be responsible for the success of the
treatment. This randomized prospective controlled study examines the
hypothesis that valgus loading is responsible for the success observed
clinically. This is accomplished by comparing clinical outcomes of patients
treated with valgus bracing versus outcomes of those treated with sham
bracing (no valgus correction).
Methods:To date, forty-four subjects, 27 males and 17 females, (age 52 ± 11
years, ht 1.8 ± 0.1 m , wt 86 ± 16 kg) have been tested. Inclusion criteria
were: males and females 21 years of age or older; medial compartment knee
OA; and referral for unilateral bracing. Exclusion criteria were: fixed knee
flexion deformity greater than 5°; leg length discrepancy greater than 2 cm;
other neuromuscular deficits; skin or peripheral vascular disease preventing
brace application; and current lower extremity orthopaedic deficits, except for
the contralateral knee.
Subjects received an off-the-shelf, adjustable hinge correction, Bledsoe
Thruster brace at no charge. The braces were fit from the same orthotic
department. Informed consent was obtained, and treatment was randomized
into two groups. The treatment group received the prescribed amount of
valgus correction applied. A control group was fit with the same model brace
with no valgus correction applied (sham bracing). Upon completion of gait
analysis, the braces for these subjects were adjusted to the prescribed amount
of correction. The treatment group assignment was known only to the
orthotist. When a subject received the brace, an initial questionnaire was
completed that included visual analog scales for pain and function during
walking. Subjects used the brace according to physician referral when needed
during activities of daily living and sports. After at least two weeks of brace
use, gait analysis was conducted. Time-distance parameters, knee joint
kinematics and kinetics, and brace loads from instrumented braces were
determined. The questionnaire was also re-administered at this time.
Randomization was done with a block design. Each of the blocks consists of
two treatment cells that contain five subjects. Subjects are randomly assigned
to a cell when recruited for the study. At that time the leg is measured for the
brace, informed consent is obtained and each participant completes the initial
questionnaire. Included in the questionnaire were visual analog scales (VAS)
for pain at rest, during walking, stair negotiation and usual sports activity.
Subjects used the brace according to physician referral when needed during
activities of daily living, walking, exercising and sports. After two weeks, but
less than eight weeks, of using the brace, gait analysis was conducted with the
subject unbraced and braced. Time-distance parameters, knee joint
kinematics and kinetics, brace loads from instrumented braces and calculated
knee joint compartment loads were determined. A follow-up questionnaire
was also completed at the time of testing.
Each of the braces is instrumented to record bending moments. Three-dimensional video based gait analysis using a six-camera system (Motion
Analysis Corporation, Santa Rosa, CA) and two force platforms (Bertec, OH)
was completed at self-selected free speed walking. A minimum of three trials
of force platform contacts was collected for each of two conditions: braced
and unbraced free speed walking. A 2m by 1m by 2m volume was calibrated
and a standard unilateral lower extremity 11 marker set (Cleveland Clinic
configuration) was used. Hip, knee and ankle joint centers were calculated
according to standardized methodology. Pelvis and unilateral lower extremity
(hip, knee and ankle) joint kinematics, and unilateral lower extremity (hip,
knee and ankle) joint kinetics were determined. Only the affected lower
extremity was measured.
Average kinematics, kinetics and brace loads for each subject, for three trials
at each condition were calculated. In order to measure similarity between
treatment groups, two-sample t-test and chi square analysis were used to
compare demographics, time-distance parameters (velocity, cadence, and
stride length), and the pre-braced questionnaire results. Results were analyzed
to determine whether or not clinical success as measured by pain relief was
associated with treatment (t-test). In addition, the degree of load sharing by
the brace was analyzed for the two groups (bootstrapped 90% confidence
interval). A significance level of 0.05 was used.
Results:There were no differences between groups with respect to
demographics or time-distance parameters during braced or unbraced walking.
The unbraced and braced velocities for the control group were 113 and 112
cm/s, respectively. For the treatment group, these values were 113 and 114
cm/s, respectively. Changes in pain score during walking were –2.8 ± 3.0 and
–2.1 ± 2.1 in the control and treatment group respectively. A two-sample t-test demonstrated no difference in the pain relief or function for the two
groups. The average varus moment taken up by the brace during the stance
phase in the control group was –3.3 ± 2.9 Nm versus –2.9 ± 3.0 Nm for the
treatment group. A test of the difference between these means yielded p =
0.66.
Discussion: The results indicated that while load sharing occurred when
wearing the brace for all subjects (-3.3 ± 3.0 Nm, p < 0.001) there was no
difference between groups in the amount of valgus correction. The lack of
differences in pain and function between the two groups was consistent with
the comparable brace loads in both groups. As in previous work, there was
large variability in pain relief although both groups demonstrated significant
improvement between their pre and post visits (p < 0.001). The population is
representative of the spectrum of patients treated with valgus bracing and seen
by physical therapists during the course of conservative management. Valgus
bracing for medial compartment knee OA continues to be another option in
the treatment of these patients. A third group of 25 patients with valgus
correction greater than that used in the current treatment group is currently
being added to the study.
Reference:1. Otis, J.C., et al.: Gait and Posture, Vol. 4, No. 2, p. 189, 1996.