Journal of Sport Rehabilitation, 2013, 22, 27-32
© 2013 Human Kinetics, Inc.
www.JSR-Journal.com
ORIGINAL RESEARCH REPORT
Effect of a SERF Strap on Pain and Knee-Valgus Angle
During Unilateral Squat and Step Landing
in Patellofemoral Patients
Lee Herrington
Context: A valgus position of the knee on functional loading tasks has been reported to be associated with
patellofemoral-joint pain. Training programs to reduce knee valgus have been shown to be effective but take
time. It would appear logical to use a brace or strap to help control this knee motion to reduce symptoms. Objective: To assess the impact of the SERF strap on knee valgus and patellofemoral-joint pain. Design: Repeated
measures. Setting: University human performance laboratory. Participants: 12 women with patellofemoral pain
(mean age 24 ± 3.2 y). Intervention: Application of SERF strap. Main Outcome Measures: Knee-valgus angle
on single-leg squat and step landing and visual analog scale pain score. Results: The application of the SERF
brace significantly reduced the pain (P < .04) and knee valgus (P < .034) during both tasks. Conclusion: The
SERF brace brings about a significant reduction in pain during functional tasks. Although the brace brought
about a significant reduction in knee valgus, this failed to exceed the smallest-detectable-difference value,
so the difference is likely to be related to measurement error. The mechanism as to why this the reduction in
pain occurs therefore remains unclear, as this study in line with many others failed to demonstrate meaningful
changes in kinematics that could provide an obvious explanation.
Keywords: patellofemoral joint pain, bracing, landing mechanics, motion control
A valgus position of the knee on functional loading
tasks has been reported to be associated with a number
of different knee injuries.1–3 Females have been found to
present with greater knee valgus during loading tasks than
males,4 and this is believed at least in part to explain the
higher incidence of patellofemoral pain in females.3,4 The
presence of a valgus knee position has been associated
with muscle weakness in the hip abductors and external
rotators.5 Considerable attention has been given to how
best to train these muscles with some success.5 However,
these neuromuscular changes take time, with training
programs lasting 4 to 6 weeks.5 As the changes brought
about by training take time to occur, clinicians have
sought means of bringing about immediate improvements
in limb alignment and so reducing patellofemoral stress
by using means such as external supports.
External supports have been used to aid performance
of patients with a number of knee pathologies. The
research into the role of braces in treating patellofemoral
joint pain is limited.6 Some studies have used knee braces
to provide local control to the patellofemoral joint,6,7
but the findings are mixed. One of the main reasons for
the mixed results might be the nature of the braces. The
The author is with the School of Health Sciences, University
of Salford, Salford, UK.
sleeve-type braces, though creating increased compression and so improved contact area, also increase joint
pressure, with the combination of these effects possibly
creating no net reduction in stress.8 While the wrap-style
patellar-stabilization brace allows knee pressure to be
decreased in magnitude and creates a proximal shift in the
location of the peak patellofemoral pressure,8 improving
stress loading on the patellofemoral joint, these effects
can be inconsistent and depend on optimal application
of the brace,
The role of knee valgus in patellofemoral pain has
been investigated by a number of authors.2,3 Previous
interventions have aimed to change local patella position through taping or bracing. It would appear to be
a logical step to develop a strap to augment control at
the hip with the aim of minimizing hip internal-rotation
and -adduction elements of knee-valgus collapse, so
improving lower extremity kinematics to change local
patellofemoral-joint alignment. Biomechanical research
would appear to support this, with Elias et al9 finding
that changing Q angle has a more significant effect on
patellofemoral-joint load than changing local muscle
activity in their biomechanical model.
The Stability Through External Rotation of the
Femur (SERF) strap (Don Joy Orthopedics Inc, Vista,
CA) was developed with the aim of assisting lower limb
kinematics, decreasing knee valgus through supporting
27
28 Herrington
femoral abduction and external rotation. To date there
has been no published literature on the effectiveness of
this brace in either improving lower limb kinematics or
relieving pain in patients with patellofemoral pain. The
purpose of this study was to assess the influence of the
SERF strap on lower limb kinematics and pain during
single-leg squat and step-down task in patients with
patellofemoral pain.
Method
Participants
Twelve women with patellofemoral pain (mean age 24
± 3.2 years) participated in the study. The inclusion and
exclusion criteria for the study subjects are as follows.10
Inclusion Criteria
Symptoms of anterior knee pain for at least 1 month
Average pain level of 3 or more on a 10-cm visual
analog scale (VAS) during stepping up and down
off a 30-cm-high bench
Anterior or retropatellar knee pain on at least 2 of the
following activities: prolonged sitting, climbing
stairs, squatting, running, kneeling, and hopping/
jumping
Presence of 2 of the following clinical criteria on
assessment: pain during apprehension test, pain
during the patellar-compression test, and crepitation during the compression test
Exclusion Criteria
Previous knee surgery or arthritis
History of patellar dislocation, patellar subluxation,
or ligament laxity
Patellar-tendon pathology or chondral damage
Referred spinal pain
History of other abnormalities such as leg-length
inequality (>2 cm)
Medication as a part of the treatment
Previous physical therapy or acupuncture treatment
for the knee within the previous 30 days
All subjects gave written informed consent, and the study
was approved by the university research ethics committee.
Procedures
Subjects performed a single-leg squat and unilateral steplanding task on their symptomatic leg. They performed
3 test trials both with and without the SERF strap. For
each task, the sequence of either braced or unbraced
was block ordered, with alternate subjects carrying out
testing unbraced first. The brace was applied by the
same researcher each time (brace is shown in Figure 1).
Each subject had 3 practice trials of both tasks and then
a 3-minute recovery; the same 3-minute recovery was
used between all the test trials.
Figure 1 — Application of the SERF brace.
The procedure for single-leg squat involved subjects standing on the test limb, facing the video camera.
Participants were asked to squat down as far as possible,
to at least 45° knee flexion, over a period of 5 seconds.
Knee-flexion angle was checked during practice trials
(maximum of 3) using a standard goniometer (GaiamPro, Physiomed, UK) by the same examiner throughout
the trials. There was also a counter for each participant
over this 5-second period; the first count initiates the
movement, the third indicates the lowest point of the
squat, and the fifth indicates the end of the movement
with the subject returning to the start position. This standardized the test for each participant, therefore reducing
the effect of velocity on knee angles. Trials were only
accepted if the subjects squatted to the minimum degree
of knee flexion (45°) and maintained balance throughout
while keeping their hands on their iliac crests.11 While
the task was carried out, perceived pain was recorded
using a 10-cm VAS.
Step landing involved stepping off a 30-cm-high
bench, landing on a mark 30 cm from the bench (this
standardized depth and distance of the step and landing),
and holding the landing for a minimum of 3 seconds.
Subjects were asked to take a unilateral stance on the
contralateral limb and to step forward to drop onto the
Bracing and Knee Valgus 29
mark corresponding, ensuring that the contralateral leg
made no contact with any other surface.11 While carrying
out each task, the subject recorded perceived pain on a
10-cm VAS.10
Two-dimensional (2D) frontal-plane projection angle
of knee-valgus alignment was measured.2 A digital video
camera (Sony Handycam DCR-HC37, Sony Corp, Japan,
filming at 25 frames/s) was placed at each subject’s knee
height, 2 m anterior to the landing target, and aligned
perpendicular to the frontal plane. The digital images
were imported into a digitizing software program (Quintic 4, Quintic Consultancy Ltd, UK). The knee-valgus
angle was acquired from the angle formed between the
line formed between markers at the anterosuperior iliac
spine and middle of the tibiofemoral joint, and the line
drawn between markers on the middle of the tibiofemoral joint to the middle of the ankle mortise.11 Markers
were placed by the same researcher for all subjects. The
angle was captured from the frame that corresponded to
the lowest point of the landing or squatting phase—the
frame before the initiation of upward movement. The
same individual digitized all the data from all subjects.
The average knee-valgus-angle value from 3 trials was
used for further analysis.
Analysis
All statistical analysis was conducted using SPSS for
Windows version 16.0 (SPSS Inc, Chicago, IL). The
effect of the brace on knee-valgus angle for each of the
tasks was analyzed using paired t tests with α = .05, and
the effect of brace on VAS score was similarly analyzed.
Pilot work was carried out to ascertain intratester
reliability of angle capture from the video. Ten kneevalgus-angle images (5 from single-leg squat, 5 from
step landing) were selected randomly for reassessment
of angle; comparison of first and second measurement
revealed a strong correlation (ICC3,1) of r = .90 (P <
.001) and r = .92 (P < .001) between the 2 measurements (single-leg squat and step landing, respectively).
The test–retest reliability of this method for assessing
knee valgus has been established from our laboratory
and has been reported elsewhere.11 The findings of
this study showed good reliability ICC3,1 of .72 with
a smallest detectable difference (SDD) of 8.93° for
single-leg squat and ICC of .82 and SDD of 7.9° for
step landing.
Results
The results of the study are shown in Table 1 and Figures
2 and 3. Mean knee-valgus angle during single-leg squat
was 16.8° ± 5.4°. The application of the SERF brace
significantly reduced the knee valgus (P = .023); mean
decrease in knee valgus was 8.9° ± 2.3°. Reported pain
during single-leg squat was reduced significantly (P =
.001) with the application of the brace, from a mean score
of 5 ± 1.3 unbraced to 1.1 ± 1.1 with the brace in situ.
Mean knee-valgus angle during single-leg step landing was 13.9° ± 6.8°. The application of the SERF brace
significantly reduced the knee valgus (P = .034); mean
decrease in knee valgus was 6.9° ± 4.1°. Reported pain
during single-leg step landing was reduced significantly
(P = .04) with the application of the brace, from a mean
score of 4 ± 1.9 unbraced to 1.0 ± 1.3 with the brace in
situ.
Discussion
The application of the SERF brace in female patients
with patellofemoral pain would appear to improve kneevalgus angle, bringing about a significant reduction in
angle during both tasks. The presence of the brace also
significantly reduced pain in female patients with patellofemoral pain during unilateral functional loading tasks.
The decrease in pain associated with the application
of the SERF brace is consistent with the findings of other
studies where either local knee braces or patella tape
were applied to the patellofemoral joint.6,12 The change
in femoral alignment found in this study may create sufficient change in internal tissue homeostasis13 to relieve
symptoms, as has been hypothesized previously in relation to patella taping.14
There must be caution applied when interpreting our
results in relation to the change in knee-valgus angle. The
mean change in knee-valgus angle brought about using
the brace during both tests did not exceed the SDD value
previously reported from our laboratory11 for these tests.
The SDD was 8.9° for single-leg squat and 7.9° for step
landing, with mean change found in the study being 8.9°
and 6.9°, respectively. The SDD statistic is useful for
distinguishing real changes from meaningless fluctuation. It represents reliability in context of measurement
error, with SDD being the minimal change required to
be 90% confident that difference between individual
Table 1 Effect of Brace on Knee-Valgus Angle and Pain Score
Single-Leg Squat
Condition
Step Landing
Knee-valgus angle (°)
Pain score
Knee-valgus angle (°)
Pain score
Braced
8.2 ± 2.6
1.1 ± 1.1
7.3 ± 4.3
1 ± 1.3
Unbraced
16.8 ± 5.4
5 ± 1.3
13.9 ± 6.8
4 ± 1.9
Figure 2 — Mean (± SD) knee-valgus angle during both tasks with and without the brace.
Figure 3 — Mean (± SD) visual-analog-scale (VAS) pain scores for both tasks with leg braced and unbraced.
30
Bracing and Knee Valgus 31
pre–post measures is due to real change.15 This would
indicate that the differences between the prebracing and
postbracing values in knee valgus are more likely to be
due to measurement error (or random chance) than an
actual effect of the brace itself, despite the statistically
significant difference.
McLean et al16 reported that 2-D video analysis was
suitable to screen individuals with excessive knee valgus
and the effects of interventions to improve knee valgus
in those individuals. Both McLean et al16 and Willson
and Davis2 stated that this conclusion should be viewed
with caution, as this method lacks sensitivity to small
changes in angle. McLean et al16 found average peak
2-D knee-valgus angle accounted for 58% to 64% of
the variance in average peak 3-dimensional knee-valgus
angle between subjects during side-step and side-jump
activities. Willson and Davis2 found that 2-D knee valgus
reflected 23% to 30% of the variance of 3-dimensional
measurements during single-leg squat but found knee
valgus to be significantly correlated with knee external
rotation (r = .54, P = .001) and hip adduction (r = .32, P
= .04), which are major components of the “medial collapse” 2-D knee-valgus angle aims to represent. As Souza
and Powers5 found femoral internal rotation to be the
significant discriminator between controls and patients
with patellofemoral pain, as opposed to hip adduction;
the role of abnormal transverse-plane motion must not be
discounted. The SERF strap, by the nature of its orientation and direction of pull, is likely to have an influence,
predominantly on hip rotation. It may be that 2-D video
analysis is insensitive to changes in this plane, even
though the knee-valgus-angle measure is a composite
movement including hip rotation. It may be that analysis
of any changes brought about in the transverse plane
would require 3-dimensional motion-capture systems.
The changes reported by this study in pain with brace
application cannot be clearly linked to the change in knee
valgus, as explained herein, because of the possibility of
the differences occurring due to random chance. However, small changes in position are likely to have occurred.
It may be that the change in perceived pain could be
explained by Dye et al17 and Dye,13 who believed that
bracing allowed an expansion of the “envelope of function,” helping the patient overcome the initial metabolic
homeostatic crisis, taking normal activities of daily living
out of the zone of traumatic supraphysiological loading.
The level of tissue offloading required to bring about
relief of stress is not known but may prove to be only a
small change in the local joint environment.13
In the clinical context, the findings of this study have
value, as they indicate that use of this brace will bring
about an immediate reduction in pain during a number
of limb-loading tasks. If the patient is able to participate
in limb-loading tasks without pain, this will allow the
clinician to further progress the patient down a loading
continuum, increasing their envelope of function.13,17
Conclusion
In common with a number of other interventions such
as local knee braces and patella taping, the SERF brace
brings about a significant reduction in pain during
functional tasks. The mechanism by which this occurs,
though, remains unclear, as this study in line with many
others failed to demonstrate valid changes in kinematics
that could provide an obvious explanation.
References
1.Hewett TE, Myer G, Ford K, Heidt R, Colosimo A,
McLean S. Biomechanical measures of neuromuscular
control and valgus loading of the knee predict anterior
cruciate ligament injury risk in female athletes: a prospective study. Am J Sports Med. 2005;33:492–501. PubMed
doi:10.1177/0363546504269591
2.Willson JD, Davis I. Utility of frontal plane projection
angle in females with patellofemoral pain. J Orthop Sports
Phys Ther. 2008;38:606–615. PubMed
3. Myer GD, Ford K, Barber Foss K, et al. The incidence and
potential pathomechanics of patellofemoral pain in female
athletes. Clin Biomech (Bristol, Avon). 2010;25:700–707.
PubMed doi:10.1016/j.clinbiomech.2010.04.001
4. Russell KA, Palmieri R, Zinder S, Ingersoll C. Sex differences in valgus knee angle during a single-leg drop jump.
J Athl Train. 2006;41:166–171. PubMed
5.Souza RB, Powers C. Differences in hip kinematics,
muscle strength and muscle activation between patients
with and without patellofemoral pain. J Orthop Sports
Phys Ther. 2009;39:12–19. PubMed
6.Powers CM, Doubleday K, Escudero C. Influence of
patellofemoral bracing on pain, knee extensor torque
and gait function in females with patellofemoral pain.
Physiother Theory Pract. 2008;24:143–150. PubMed
doi:10.1080/09593980701665793
7. Denton J, Willson J, Ballantyne B, Davis I. The addition
of the Protonics brace system to a rehabilitation protocol
to address patellofemoral joint syndrome. J Orthop Sports
Phys Ther. 2005;35:210–219. PubMed
8.Wilson NA, Mazahery B, Koh J, Zhang L. Effect of
bracing on dynamic patellofemoral contact mechanics. J
Rehabil Res Dev. 2010;47:531–541. PubMed doi:10.1682/
JRRD.2009.12.0204
9.Elias JJ, Bratton DR, Weinstein DM, Cosgarea AJ.
Comparing two estimations of the quadriceps force
distribution for use during patellofemoral simulation.
J Biomech. 2006;39(5):865–872. doi:10.1016/j.jbiomech.2005.01.030
10.Herrington L, Al-Shehri A. A controlled trial of open
versus closed kinetic chain exercises for patellofemoral
pain. J Orthop Sports Phys Ther. 2007;37:155–160.
PubMed
11. Munro A, Herrington L, Carolan M. Reliability of 2-dimensional video assessment of frontal-plane dynamic knee
32 Herrington
valgus during common athletic screening tasks. J Sport
Rehabil. 2012;21:7–11. PubMed
12. Herrington L. The effect of patellar taping on quadriceps
peak torque and perceived pain: a preliminary investigation. Phys Ther Sport. 2001;2:23–28. doi:10.1054/
ptsp.2000.0034
13.Dye SF. The pathophysiology of patellofemoral pain:
a tissue homeostasis perspective. Clin Orthop Relat
Res. 2005;40:100–110. PubMed doi:10.1097/01.
blo.0000172303.74414.7d
14.Herrington L. The difference in a clinical measure of
patella lateral displacement between patients with patellofemoral pain and matched controls. J Orthop Sports Phys
Ther. 2008;38:59–62. PubMed
15.Eliasziw M, Young S, Woodbury M, Fryday-Field K.
Statistical methodology for concurrent assessment of
interrater and intrarater reliability: using goniometric measurements as an example. Phys Ther. 1994;74:777–788.
PubMed
16. McLean SG, Walker K, Ford K, Myer G, Hewett T, Van
den Bogert A. Evaluation of a two dimensional analysis
method as a screening and evaluation tool for anterior cruciate ligament injury. Br J Sports Med. 2005;39:355–362.
PubMed doi:10.1136/bjsm.2005.018598
17.Dye S, Staubli H, Biedert R, Vaupel G. The mosaic of
pathophysiology causing patellofemoral pain: therapeutic implications. Oper Tech Sports Med. 1999;7:46–54.
doi:10.1016/S1060-1872(99)80014-8