Manual Therapy 22 (2016) 16e21

Contents lists available at ScienceDirect

Manual Therapy
journal homepage: www.elsevier.com/math

Original article

The immediate effect of lumbopelvic manipulation on EMG of vasti

and gluteus medius in athletes with patellofemoral pain syndrome: A
randomized controlled trial
Alireza Motealleh, Elham Gheysari, Esmaeil Shokri, Sobhan Sobhani*
Department of Physical Therapy, School of Rehabilitation Sciences, Shiraz University of Medical Sciences, Shiraz, Iran

a r t i c l e i n f o a b s t r a c t

Article history: Objective: To evaluate the immediate effect of lubmopelvic manipulation on EMG activity of vastus
Received 13 October 2015 medialis, vastus lateralis and gluteus medius as well as pain and functional performance of athletes with
Received in revised form patellofemoral pain syndrome.
5 January 2016
Design: Randomized placebo-controlled trial.
Accepted 8 February 2016
Methods: Twenty eight athletes with patellofemoral pain syndrome were randomly assigned to two
groups. One group received a lubmopelvic manipulation at the side of the involved knee while the other
Keywords:
group received a sham manipulation. EMG activity of the vasti and gluteus medius were recorded before
Anterior knee pain
Chondromalacia
and after manipulation while performing a rocking on heel task. The functional abilities were evaluated
Electromyography using two tests: step-down and single-leg hop. Additionally, the pain intensity during the functional tests
Manual therapy was assessed using a visual analog scale.
Results: The onset and amplitude of EMG activity from vastus medialis and gluteus medius were,
respectively, earlier and higher in the manipulation group compared to the sham group. There were no
significant differences, however, between two groups in EMG onset of vastus lateralis. While the scores of
one-leg hop test were similar for both groups, significant improvement was observed in step-down test
and pain intensity in the manipulation group compared to the sham group.
Conclusions: Lubmopelvic manipulation might improve patellofemoral pain and functional level in
athletes with patellofemoral pain syndrome. These effects could be due to the changes observed in EMG
activity of gluteus medius and vasti muscles. Therefore, the lubmopelvic manipulation might be
considered in the rehabilitation protocol of the athletes with patellofemoral pain syndrome.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction point prevalence of 16.3% of PFPS has been reported in basketball

Patellofemoral pain syndrome (PFPS) causes pain and discom-
fort in the anterior part of the knee. Although the severity of pain
and physical impairment varies in PFPS patients, functional pos-
tures and physical activities (e.g. sustained sitting, squatting) usu-
ally increase the symptoms (Thomee
et al., 1999; Brechter and
Powers, 2002). PFPS is common among young active individuals
especially those who are involved in sports activities requiring high
levels of quadriceps activity with jumping, cutting and pivoting
activities (Nejati et al., 2011; Roush and Bay, 2012). For instance, a
* Corresponding author. 1st Abivardi Avenue, Chamran Blvd, School of Rehabili-
tation Sciences, Department of Physical Therapy, Shiraz, Iran.
E-mail addresses: sobhan132@gamil.com, sobhanis@sums.ac.ir (S. Sobhani).
players (Myer et al., 2010).
Although the etiology of PFPS has not been fully understood,
some possible risk factors include malalignment of the lower ex-
tremity, joint laxity, and neuro-motor dysfunction of the quadri-
ceps muscles (Thomee
et al., 1999; Waryasz and McDermott, 2008).
Among these factors, the latest has received great attention. It is
hypothesized that poor coordination between vastus medialis
obliquus (VMO) and vastus lateralis (VL) muscle activation may
cause lateral tracking of the patella. This neuromuscular imbalance
could overload the lateral aspect of the patellofemoral joint and
eventually leading to PFPS (Voight and Wieder, 1991; Witvrouw
et al., 1996; Coqueiro et al., 2005; Van Tiggelen et al., 2009; Briani
et al., 2015). The findings of a systematic review and meta-
analysis revealed a trend towards a delayed onset of VMO relative
to VL in patients with anterior knee pain (Chester et al., 2008).

http://dx.doi.org/10.1016/j.math.2016.02.002
1356-689X/© 2016 Elsevier Ltd. All rights reserved.
 A. Motealleh et al. / Manual Therapy 22 (2016) 16e21
However, because of heterogeneity of the included studies, no
definite conclusion can be drawn yet with respect to this associa-
tion (Chester et al., 2008).
To understand further the link between PFPS and neuro-motor
dysfunction, some researchers have paid attention to the activa-
tion patterns of muscles located more proximal to the knee joint.
Motealleh et al. showed a delayed onset of activity in some core
muscles during stair negotiation in patients with PFPS compared to
healthy controls (Motealleh et al., 2011, 2015). In another study,
Cowan et al. compared EMG activity of the anterior and posterior
parts of the gluteus medius (GM) between PFPS patients and
healthy controls during a stair-stepping task; and observed a delay
in activation of both parts of GM in patients with PFPS (Cowan et al.,
2009). In a recent comparative study of PFPS patients and healthy
controls, the PFPS group was found to have an earlier and longer
activation of their abdominal and erector spinae muscles following
an external lateral perturbation. For the GM, however, a delayed
response was observed in EMG activity (Shirazi et al., 2014).
Moreover, Brindle et al. identified a delay in EMG onset of GM in
people with anterior knee pain during both stair ascending and
descending (Brindle et al., 2003). It is claimed that such deficit in
neuromuscular control of the hip and trunk might aggravate the
PFPS symptoms. For instance, it has been shown that the timing of
gluteal muscles is correlated with greater internal rotation and hip
adduction (Willson et al., 2011). These abnormal hip mechanics
were found to be significant predictors of patellofemoral pain
(Nakagawa et al., 2013).
Previous studies have shown that lumbopelvic manipulation
could effectively increase quadriceps activation and strength as
well as knee extensor moment in patients with PFPS (Suter et al.,
1999; Hillermann et al., 2006). Similar results were also observed
in healthy individuals (Grindstaff et al., 2009). As mentioned earlier,
the activation timing of different components of quadriceps muscle
is considered as an important risk factor in developing PFPS. The
previous studies have mainly looked at the EMG amplitude of
quadriceps muscles in response to lumbopelvic manipulation with
no assessment of their EMG onsets or the EMG activity of muscles
proximal to the knee joint. The effects of lumbopelvic manipulation
on clinical outcomes such as pain or functional level have not been
much studied either. To our best of knowledge, there has been only
one study reporting some improvement in knee pain after lum-
bopelvic manipulation (Crowell and Wofford, 2012). This study,
however, lacked a placebo or control group, and additional studies
were recommended to verify these findings (Crowell and Wofford,
2012).
The aim of current study was, therefore, to assess immediate
effect of lumbopelvic manipulation on EMG activity of GM and vasti
muscles, lower limb function as well as pain intensity in athletes
with PFPS. Our first hypothesis was that lumbopelvic manipulation
could improve the EMG parameters of GM and vasti muscles. Our
second hypothesis was that an immediate improvement would be
observed in the clinical outcomes following the lumbopelvic
manipulation.
2. Materials and methods
2.1. Participants
This trial obtained approval from the Medical Ethics Committee
of Shiraz University of Medical Sciences (Registration No. CT-92-
6722), and all participants signed a written informed consent form
before study commencement. Athletes diagnosed with PFPS were
referred by an orthopedic surgeon to the outpatient physical ther-
apy clinic of Shiraz University of Medical Sciences. Inclusion and
exclusion criteria were similar to previous studies (Cowan et al.,
17
2001a; Van Tiggelen et al., 2009; Cowan et al., 2009). We
included patients if they were aged 40 years or less, to reduce the
likelihood of osteoporosis (a contraindication for manipulation)
and osteoarthritis in the knee and patellofemoral joint. Patients
were to have a non-traumatic unilateral anterior knee pain lasting
for no more than 6 months. They also needed to be athletes
involved in regular sports activity (at least 3 sessions per week).
The clinical criteria for diagnosis of PFPS were as follows: (a)
anterior knee pain or a pain on palpation of patellar facets provoked
by at least two of these activities: jumping, squatting, ascending/
descending stairs, kneeling and prolonged sitting, (b) positive sign
in eccentric step down test and/or patellar apprehension test and/
or vastus medialis coordination test, (c) a pain level of at least 3
points (0: no pain; 10: worst possible pain) on the numerical pain
rating scale during resisted knee extension, and (d) knee disability
level between 45 and 70 based on the Kujala patellofemoral
questionnaire (KPQ) (Kujala et al., 1993; Negahban et al., 2012). The
KPQ is a 13-item questionnaire with different categories. The total
score ranges from 0 to 100, with higher scores indicating lower
levels of disability (Kujala et al., 1993; Negahban et al., 2012). The
KPQ is a valid and reliable measure of disability in patients with
patellofemoral pain (Kujala et al., 1993; Negahban et al., 2012). The
patients were excluded if they had bilateral PFPS, a history of knee
surgery, meniscal lesion, patellar subluxation/dislocation, evidence
of tendinopathy or ligamentous pathologies, dislocation or fracture
in pelvic or spinal surgery, and neurologic disorders. Further, the
patients who had previously received physiotherapy treatment or
used analgesic drugs within 72 h prior to the experiment were
excluded.
2.2. Study design and sample size calculation
This study was a randomized, single-blind, placebo-controlled
trial. A convenience sample of patients with PFPS was randomly
assigned to two groups using a block randomization procedure.
One group received a general manipulation of the lumbopelvic
region whereas the other group received a sham manipulation.
Sample size calculation was based on EMG onset of the VMO during
a “rock” task reported in a previous study (Cowan et al., 2001b). The
sample size was estimated to be at least 26 patients (13 per group)
for a power of 80% and a ¼ 0.05 to detect a difference of 50 ms of
EMG onset between the two compared groups.
2.3. Outcome variables
The primary outcomes of interest included EMG activity of VMO,
VL and GM muscles. The secondary outcomes included functional
performance and knee pain.
2.4. EMG recordings
The ME6000 16-channel EMG telemetry system (Mega Elec-
tronics Ltd, Kuopio, Finland) was used to obtain and transmit the
data to the computer wirelessly. The system had a common mode
rejection ratio of 110 dB. EMG data sampled at the rate of 1000 Hz
and band-pass filtered at 8e500 Hz through a 14-bit analogue-to-
digital converter. The EMG activity of the vasti and GM was recor-
ded using round, pre-gelled, self-adhesive, Silvere-Silver Chloride
surface electrodes, with a 20 mm center-to-center distance. The
Megawin software (Mega Electronics Ltd, Kuopio, Finland, version
3.0), and a custom-made Matlab™ program (R2010a) were used to
receive and analyze EMG data respectively.
Before electrode attachment we prepared the skin by shaving,
cleaning with an alcohol wipe and rubbing with a fine sandpaper.
VMO electrodes were placed over the muscle belly approximately
18 A. Motealleh et al. / Manual Therapy 22 (2016) 16e21
4 cm superior and 3 cm medial to the upper border of the patella, in
an oblique direction of 550 deviated from vertical line (Cowan et al.,
2001b). The ground electrode was placed on the medial condyle of
femur. VL electrodes were placed 10 cm superior and 6e8 cm
lateral to the mid-upper border of the patella with a 150 deviation
from vertical line (Cowan et al., 2001b). The ground electrode was
placed on lateral femoral condyle. The electrodes for the GM were
positioned one-third the distance between the iliac crest and
greater trochanter in line of the anterior muscle fibers (Leis and
Trapani, 2000). The ground electrode was positioned over the
anterior superior iliac spine. Finally, the electrodes for the tibialis
anterior was placed 3 cm inferior to the tibial tubercle in line of the
muscle fibers and the ground electrode was placed on tibial tu-
berosity (Cowan et al., 2002). Manual muscle testing and observing
EMG signals during muscle contraction were used to confirm the
correct placement of electrodes. An anti-allergic adhesive tape was
used to maintain the cables on the skin in order to reduce motion-
induced signal artifacts.
2.5. Functional performance and pain
We used two functional tests in this study (Brechter and Powers,
2002): step-down test, and (Thomee
et al., 1999) one-leg hop test
(Ageberg et al., 1998; Loudon et al., 2002). To perform the step
down test, the patients had to stand on a 20-cm step with both legs.
We asked them to lower the sound leg to the ground, touch it, and
then return their leg to the initial position (full knee extension). The
number of repetitions of this task for 30 s was recorded for analysis
(Loudon et al., 2002). To perform the one-leg hop test, the patients
were asked to hop as far as possible with their affected leg, and land
on the same foot. The test was performed three times, and the
longest distance, measured from toe to toe, in 3 consecutive trials
was recorded and used for analysis (Ageberg et al., 1998). The test
was performed barefoot and the patients were allowed to practice
the test before measurements were taken. Immediately after per-
forming the functional tests, the patients' knee pain was verbally
assessed using an 11point numerical pain rating scale (0: no pain,
10: the maximal imaginable pain). This scale is a valid and reliable
measure of chronic pain intensity (Farrar et al., 2001).
2.6. Procedures
After initial evaluation, the EMG electrodes were attached to the
body. Patients stood barefoot with weight equally distributed on
two feet and with their arms by their side. Then, we instructed
them to rock back on their heels (rock task) as quickly as possible by
lifting their toes and contracting their tibialis anterior muscle
(Cowan et al., 2001b). The assessor raised her arm as a command for
patients to start the rock task. EMG recording procedure was
repeated 3 times and the average of the 3 trials were used in the
final analysis. After EMG recordings, the patients performed the
step-down and one-leg hop tests. The order of performing the two
tasks was randomized among participants in each group. The in-
tensity of knee pain was assessed after the functional tests.
Following the baseline testing, the lumbopelvic manipulation
was performed by a trained physical therapist. The patient was
positioned supine with the therapist standing on the opposite side
of the painful knee. After interlocking the fingers behind the head,
the patient was passively side-bent towards and rotated away from
the painful knee. Then, a low-amplitude high velocity thrust was
delivered to the anterior superior iliac spine in a posterior and
inferior direction by the therapist (Fig. 1) (Flynn et al., 2003). For the
sham group, the patient was held exactly in the same position for a
few seconds, without the manipulative thrust at the end. After the
Fig. 1. Lumbopelvic manipulation technique used in this study.
manipulation, the same measurements (EMG, function, and pain)
were repeated as described above for each patient.
2.7. EMG data analysis
The following EMG parameters were assessed: (a) the EMG
onset (ms), (b) the normalized EMG amplitude (% maximum
voluntary isometric contraction), and (c) the relative timing of VMO
compared to VL (D onset of vasti, ms). The EMG data were recorded
from 5 s before to 10 s after the start command, and 300 ms before
the beginning of activity was considered to determine the baseline
activity. Raw data were full-wave rectified and band-pass filtered
between 20 and 450 Hz. Each muscle onset time was defined as the
first point in time at which the amplitude of EMG activity exceeded
3 standard deviations of the mean baseline activity for at least
25 ms (Di Fabio, 1987). The EMG onsets of VMO, VL and GM were
expressed relative to the EMG onset of tibialis anterior as the prime
mover of the rock task, and then averaged over the 3 trials (Cowan
et al., 2001b). The D onset of vasti muscles (onset timing difference)
was quantified by subtracting the EMG onset of VMO from that of
VL (D onset of vasti ¼ VMO onset e VL onset). A positive difference
indicated VL pre-activation and a negative difference indicated
delayed VMO activation. The EMG onsets identified by the com-
puter algorithm were also confirmed visually (Hodges and Bui,
1996). For each muscle, the peak amplitudes during the rock task
were extracted from the rectified, low-pass filtered (6th order
Butterworth with a 50 Hz cut-off frequency) EMG signals. Then, the
average of three peak amplitudes were normalized to the 100%
maximum voluntary isometric contraction obtained in sitting (for
vasti) and side-lying (for GM) positions as previously described by
Konrad (Konrad, 2005). To assess within-subject reliability of EMG
measurements, 14 patients (7 randomly selected from each group)
were evaluated again after 24 h with identical procedures and
measurements as the initial testing day.
2.8. Statistical analysis
Descriptive statistics were used to summarize the demographic
characteristics of the population. To compare the baseline charac-
teristics of participants between the two groups, we used an in-
dependent t-test (for continuous variables) and chi-square test (for
categorical variables). The measurement reliability was evaluated
using intraclass correlation coefficient and standard error of mea-
surement (SEM) (Shrout and Fleiss, 1979; Weir, 2005). The Sha-
piroeWilk test showed a normal distribution (p > 0.05) for all
outcome variables, and therefore parametric statistics were used.
The within-group differences were analyzed using paired t-test. For
all outcome measures an analysis of co-variance (ANCOVA) was
used to compare the two groups in terms of pre-to post-interven-
tion change. Baseline values were included as a covariate in the
regression model. Adjusted mean differences and associated 95%
 A. Motealleh et al. / Manual Therapy 22 (2016) 16e21 19

confidence intervals were computed. The level of significance was

set at p < 0.05. Statistical analyses were performed using SPSS,
version 20 (SPSS, Chicago, IL, USA).

3. Results

In total, 28 people participated in the study with similar baseline

characteristics between the two intervention groups as shown in
Table 1. Participant flow diagram is presented in Fig. 2.
The ICC ± SEM for GM, VMO and VL onsets (ms) were
0.83 ± 11.43, 0.80 ± 13.43 and 0.78 ± 15.77, respectively, for the
control group, and 0.80 ± 17.86, 0.70 ± 14.44 and 0.67 ± 19.61,
respectively, for the intervention group. The ICC ± SEM for GM,
VMO and VL amplitudes (%MVC) were 0.93 ± 0.07, 0.84 ± 0.03 and
0.81 ± 0.04, respectively, for the control group, and 0.84 ± 0.00,
0.72 ± 0.05 and 0.94 ± 0.00, respectively, for the intervention group.
These results indicate a good repeatability of EMG measurements
of the VMO and GM in both groups and a moderate reliability for
the VL in the treatment group. Within the intervention group, there
was significant improvement in the EMG onset of VMO (p ¼ 0.005) Fig. 2. The flow diagram of participants through the stages of the study.
and GM (p ¼ 0.006). There was also increase in the EMG amplitudes
of VMO (p ¼ 0.001), VL (0.027) and GM (p < 0.001) in the inter-
vention group. These changes significantly differed from the con-
trol group. The D onset of vasti remained unchanged in the two be explained neurophysiologically and biomechanically. Because of
groups (p ¼ 0.128). Similarly, there was no significant change either the common innervation of the lumbopelvic regions and GM by
within or between the two groups for the VL muscle. The paired t- anterior primary division of L2-S2 (Cramer and Darby, 2013; Moore
test revealed significant improvement in the scores of one-leg hop et al., 2013), any modulation in the lumbopelvic afferent signals,
(p ¼ 0.125) and step down (p ¼ 0.004) tests. However, when due to manipulation, might have consequently affected the GM
compared to the control group, only the scores of the one-leg hop activation pattern (Grindstaff et al., 2009). In addition, there is
were statistically different. For the pain intensity, the within group evidence of a link between presence of myofascial trigger points
analysis showed significant change (p < 0.001) in the intervention and dysfunction in the adjacent joints (Ferna
ndez-de-las-Pen ~ as,
group, which significantly differed from the placebo group 2009). Since patients with PFPS have a high prevalence of myo-
(p < 0.001). An overview of pre- and post-intervention scores, the fascial trigger points in GM and quadratus lumborum muscles
mean group differences, and the p-values of between group ana- (Roach et al., 2013), deactivation of existing trigger points after the
lyses is presented in Table 2. lumbopelvic manipulation, therefore, might be a possible mecha-
nism related to improved onset of GM and its increased level of
activity. An unexpected result was the significant decrease in EMG
4. Discussion
amplitude of GM in the control group after the sham manipulation.
Unfortunately we could not find any reasonable explanation for this
Recently in research on PFPS, there has been more attention
change.
towards the activity of the muscles that are located more proximal
Neuromuscular imbalance between vasti muscles can lead to
to the knee joint. GM and gluteus maximus muscles control the
maltracking of the patella, and thus increase the patellofemoral
adduction and internal rotation moments of hip joint as a part of
contact pressure (Van Tiggelen et al., 2009). The manipulation
Lumbo-PelviceHip complex. Any dysfunction of these muscles, for
improved the EMG parameters of VMO muscle without significant
instance, can cause femoral adduction and internal rotation and
change in the EMG of VL muscle. The improvement in VMO onset
result in compensatory abduction of the tibia and affects the
was expected to improve the D onset of vasti (calculated using EMG
normal alignment of the knee (Powers et al., 2003). Some authors
onset of VL and VMO muscles). Following the manipulation, there
had previously observed delayed onset and/or earlier activation of
was an improvement in the D onset of vasti which was not signif-
GM in different tasks of lower extremity (Brindle et al., 2003;
icant (p ¼ 0.128) compared to the control group. Although not
Cowan et al., 2009; Willson et al., 2011).
statistically significant, the amount of change was considerably
Interestingly in the current work, the lumbopelvic manipulation
large (30 ms). Neptune et al. suggested that a 5 ms timing delay or
increased the EMG amplitude of the GM muscle which was
advance between the VLO and VM could lead to a biomechanical
significantly different from the sham group. In addition, the EMG
imbalance at the patellofemoral joint (Neptune et al., 2000).
onset of GM muscle became considerably shorter following the
Therefore, the observed change of 30 ms can be considered clini-
manipulation. Improved onset and increased activity of GM might
cally significant. A larger sample size could have provided a better
estimate of between group differences for this outcome. Improve-
Table 1 ment in the EMG amplitudes of vasti muscles compares favorably
Summary of participants' baseline characteristics. with previous studies (Suter et al., 1999, 2000; Hillermann et al.,
Intervention group (n ¼ 14) Control group (n ¼ 14) P value 2006; Grindstaff et al., 2009). These studies reported an increase
in EMG amplitude and force output of the quadriceps muscle as
Female (n) 8 8 0.99
Age (year) 26.9 ± 5.5 26.1 ± 3.9 0.63 well as increase in the knee extensor moment after spinal manip-
Height (cm) 166 ± 6.7 169 ± 9.5 0.27 ulation. The current study confirms the neuromuscular changes in
Weight (kg) 61.6 ± 12.1 70.4 ± 12.2 0.06 quadriceps following the lumbopelvic manipulation in people with
Pain (0e10) 5.4 ± 1.3 4.6 ± 1.3 0.14 PFPS. There is evidence for delayed onset of VMO relative to VL in
Kujala 65.8 ± 5.1 65.3 ± 4.3 0.78
people with PFPS (Chester et al., 2008). The D onset of vasti was
20 A. Motealleh et al. / Manual Therapy 22 (2016) 16e21

Table 2
Results of within and between-group comparisons for all outcome measures.

Variables Intervention group (mean ± SD) Control group (mean ± SD) Adjusted mean difference (95% CI)a P-value*

Before After Before After

Onset (mSec)
VMO 50.1 ± 54.6 7.9 ± 43.5** 31.4 ± 34.8 36.9 ± 56.5 54.5 (90.6; 18.3) 0.005
VL 13.7 ± 63.2 16.0 ± 56.2 12.7 ± 20.4 2.5 ± 39.0 19.9 (55.8; 16.0) 0.264
GM 104.1 ± 98.5 27.2 ± 83.4** 106.7 ± 80.1 108.7 ± 96.0 79.7 (134.1; 25.3) 0.006
D Onset of vasti (mSec) 36.4 ± 46.9 8.0 ± 41.5 18.6 ± 27.7 33.3 ± 57.7 30.5 (70.5; 9.4) 0.128
Amplitude (%MVC)
VMO 74.3 ± 19.7 92.3 ± 14.3** 71.0 ± 25.1 67.4 ± 26.7 22.8 (11.0; 0.35) 0.001
VL 78.6 ± 23.2 82.6 ± 22.7 56.6 ± 28.6 51.4 ± 25.8 15.5 (2.0; 29.1) 0.027
GM 30.0 ± 17.1 39.6 ± 13.0** 33.9 ± 27.8 28.5 ± 25.2** 14.1 (7.5; 20.9) <0.001
Pain (0e10) 5.4 ± 1.4 3.2 ± 1.5** 4.6 ± 1.3 5.2 ± 1.5 2.6 (3.5; 1.8) <0.001
Step-down (number) 14.0 ± 3.1 16.9 ± 3.5** 14.9 ± 3.0 15.2 ± 2.7 2.4 (0.8; 3.9) 0.004
One-leg hop (cm) 126.1 ± 43.6 130.1 ± 43** 118.1444.9 115.6 ± 44.2 6.80(2.0; 15.6) 0.125

VMO ¼ vastus medialis obliquus; VL ¼ vastus lateralis; GM ¼ gluteus medius; SD ¼ standard deviation; CI ¼ confidence interval.
*P values of between group comparisons analyzed by independent T-test. Significant differences are marked in bold (p < 0.05).
**Significantly different (p < 0.05) from baseline analyzed by paired t-test for within group comparison.
a
Mean change scores (adjusted for baseline values) were computed through ANCOVA.

positive at baseline in both groups indicating pre-activation of VL
relative to VMO. Therefore, it is worth mentioning that the same
phenomenon was observed in the rock task used in the current
study.
In line with our second hypothesis, the knee pain immediately
decreased after the lumbopelvic manipulation, while no significant
change was seen in the sham group. In the manipulation group, the
mean change in pain score was about 2 points which was statisti-
cally different from the change in the sham group. This observation
confirmed the finding of the previous quasi-experimental trial
(Crowell and Wofford, 2012). The results of functional performance
tests also revealed significant improvement after manipulation in
the scores of both one-leg hop and step down tests in the inter-
vention group. But only the score of one-leg hop test was statisti-
cally different between the two groups. The improved function and
decreased pain in the intervention group can be explained by the
regional interdependence model (Wainner et al., 2007). Based on
this model, an association between the main region of complaint
and impairment in remote anatomical regions might be considered
especially for the knee and lumbar problems (Wainner et al., 2007).
Immediate improvement in onset time and EMG amplitudes of GM
might have decreased the amount of femoral adduction and in-
ternal rotation, and thus restored the normal tracking of the patella
leading to less pain and better performance in functional tasks.
However, it should be noted that the response to lumbopelvic
manipulation might not be the same for all people with PFPS.
Previous research found that patients with an asymmetry greater
than 14 in the hip internal rotation, could experience more pain
relief after lumbopelvic manipulation (Iverson et al., 2008). Con-
trary to one-leg hop test, no improvement was seen in step down
test. The reason could be due to the eccentric activation of quad-
riceps muscle needed in step down task which could cause high
loads on the patellofemoral joint. In fact, this task is one of the most
painful activities for patients with PFPS (Guney et al., 2015).
A limitation of this study was that we only assessed one of the
core muscles (i.e. GM). Without considering other muscles, we have
a limited insight to contributing factors and efficacy of lumbopelvic
manipulation in PFPS patients. Studies with longer follow up are
also necessary to evaluate the long term effects of lumbopelvic
manipulation as a part of treatment protocol of patients with PFPS.
5. Conclusion
The lumbopelvic manipulation might improve patellofemoral
pain and functional level in athletes with patellofemoral pain
syndrome. These effects could be due to the changes observed in
EMG activity of gluteus medius and vasti muscles. Therefore, the
manipulation of the lumbopelvic region might be considered in the
rehabilitation protocol of the athletes with PFPS.
Acknowledgments
The authors wish to express their thanks to all the volunteers for
their kind participation in our study. We thank Karen Shashok
(Author AID in the Eastern Mediterranean) for improving the use of
English in the manuscript. We also thank Dr. Mehrdad Vosoughi for
statistical consultation. This study was supported by a grant
fromShiraz University of Medical Sciences, Shiraz, Iran.
References
Ageberg E, Z€ €m R, Moritz U. Stabilometry and one-leg hop test have high
atterstro
test-retest reliability. Scand J Med Sci Sports 1998;8(4):198e202.
Brechter JH, Powers CM. Patellofemoral joint stress during stair ascent and descent
in persons with and without patellofemoral pain. Gait Posture 2002;16(2):
115e23.
~o FA,
Briani RV, de Oliveira Silva D, Pazzinatto MF, de Albuquerque CE, Ferrari D, Araga
et al. Comparison of frequency and time domain electromyography parameters
in women with patellofemoral pain. Clin Biomech 2015;30(3):302e7.
Brindle TJ, Mattacola C, McCrory J. Electromyographic changes in the gluteus
medius during stair ascent and descent in subjects with anterior knee pain.
Knee Surg Sports Traumatol Arthrosc 2003;11(4):244e51.
Chester R, Smith TO, Sweeting D, Dixon J, Wood S, Song F. The relative timing of
VMO and VL in the aetiology of anterior knee pain: a systematic review and
meta-analysis. BMC Musculoskelet Disord 2008;9(1):64.
Coqueiro KRR, Bevilaqua-Grossi D, Be
rzin F, Soares AB, Candolo C, Monteiro-
Pedro V. Analysis on the activation of the VMO and VLL muscles during semi-
squat exercises with and without hip adduction in individuals with patellofe-
moral pain syndrome. J Electromyogr Kinesiol 2005;15(6):596e603.
Cowan SM, Bennell KL, Hodges PW, Crossley KM, McConnell J. Delayed onset of
electromyographic activity of vastus medialis obliquus relative to vastus later-
alis in subjects with patellofemoral pain syndrome. Arch Phys Med Rehabil
2001;82(2):183e9.
Cowan SM, Hodges PW, Bennell KL. Anticipatory activity of vastus lateralis and
vastus medialis obliquus occurs simultaneously in voluntary heel and toe raises.
Phys Ther Sport 2001;2(2):71e9.
Cowan SM, Bennell KL, Crossley KM, Hodges PW, McConnell J. Physical therapy
alters recruitment of the vasti in patellofemoral pain syndrome. Med Sci Sports
Exerc 2002;34(12):1879e85.
Cowan SM, Crossley KM, Bennell KL. Altered hip and trunk muscle function in in-
dividuals with patellofemoral pain. Br J Sports Med 2009;43(8):584e8.
Cramer GD, Darby SA. Clinical anatomy of the spine, spinal cord, and ANS. Elsevier
Health Sciences; 2013.
Crowell MS, Wofford NH. Lumbopelvic manipulation in patients with patellofe-
moral pain syndrome. J Man Manip Ther 2012;20(3):113e20.
Di Fabio RP. Reliability of computerized surface electromyography for determining
the onset of muscle activity. Phys Ther 1987;67(1):43e8.
 A. Motealleh et al. / Manual Therapy 22 (2016) 16e21
Farrar JT, Young JP, LaMoreaux L, Werth JL, Poole RM. Clinical importance of changes
in chronic pain intensity measured on an 11-point numerical pain rating scale.
Pain 2001;94(2):149e58.
Ferna
ndez-de-las-Pen ~ as C. Interaction between trigger points and joint hypo-
mobility: a clinical perspective. J Man Manip Ther 2009;17(2):74e7.
Flynn TW, Fritz JM, Wainner RS, Whitman JM. The audible pop is not necessary for
successful spinal high-velocity thrust manipulation in individuals with low
back pain. Arch Phys Med Rehabil 2003;84(7):1057e60.
Grindstaff TL, Hertel J, Beazell JR, Magrum EM, Ingersoll CD. Effects of lumbopelvic
joint manipulation on quadriceps activation and strength in healthy in-
dividuals. Man Ther 2009;14(4):415e20.
Guney H, Yuksel I, Kaya D, Doral MN. The relationship between quadriceps strength
and joint position sense, functional outcome and painful activities in patello-
femoral pain syndrome. Knee Surg Sports Traumatol Arthrosc 2015 [Epub ahead
of print].
Hillermann B, Gomes AN, Korporaal C, Jackson D. A pilot study comparing the ef-
fects of spinal manipulative therapy with those of extra-spinal manipulative
therapy on quadriceps muscle strength. J Manip Physiol Ther 2006;29(2):
145e9.
Hodges PW, Bui BH. A comparison of computer-based methods for the determi-
nation of onset of muscle contraction using electromyography. Electro-
encephalogr Clin Electromyogr Mot Control 1996;101(6):511e9.
Iverson CA, Sutlive TG, Crowell MS, Morrell RL, Perkins MW, Garber MB, et al.
Lumbopelvic manipulation for the treatment of patients with patellofemoral
pain syndrome: development of a clinical prediction rule. J Orthop Sports Phys
Ther 2008;38(6):297e312.
Konrad P. The ABC of EMG. A practical introduction to kinesiological electromy-
ography. 2005. p. 1.
Kujala UM, Jaakkola LH, Koskinen SK, Taimela S, Hurme M, Nelimarkka O. Scoring of
patellofemoral disorders. Arthrosc J Arthrosc Relat Surg 1993;9(2):159e63.
Leis AA, Trapani VC. Atlas of electromyography. Oxford: Oxford University Press;
2000.
Loudon JK, Wiesner D, Goist-Foley HL, Asjes C, Loudon KL. Intrarater reliability of
functional performance tests for subjects with patellofemoral pain syndrome.
J Athl Train 2002;37(3):256.
Moore KL, Dalley AF, Agur AM. Clinically oriented anatomy. Lippincott Williams &
Wilkins; 2013.
Motealleh A, Maroufi N, Sarrafzadeh J, Sanjari M-A. The test-retest reliability of the
onset of core and vasti electromyographic activity while ascending and
descending stairs in healthy controls and patellofemoral pain patients. Med J
Islam Repub Iran 2011;24(4):221e31.
Motealleh A, Maroufi N, Sarrafzadeh J, Sanjari MA, Salehi N. Comparative evaluation
of core and knee extensor mechanism muscle activation patterns in a stair
stepping task in healthy controls and patellofemoral pain patients. J Rehabil Sci
Res 2015;1(4):84e91.
Myer GD, Ford KR, Foss KDB, Goodman A, Ceasar A, Rauh MJ, et al. The incidence
and potential pathomechanics of patellofemoral pain in female athletes. Clin
Biomech 2010;25(7):700e7.
Nakagawa T, Serra ~o F, Maciel C, Powers C. Hip and knee kinematics are associated
with pain and self-reported functional status in males and females with
patellofemoral pain. Int J Sports Med 2013;34(11):997e1002.
Negahban H, Pouretezad M, Yazdi MJS, Sohani SM, Mazaheri M, Salavati M, et al.
Persian translation and validation of the Kujala Patellofemoral Scale in
21
patients with patellofemoral pain syndrome. Disabil Rehabil 2012;34(26):
2259e63.
Nejati P, Forogh B, Moeineddin R, Baradaran HR, Nejati M. Patellofemoral pain
syndrome in Iranian female athletes. Acta Medica Iran 2011;49(3):169e72.
Neptune R, Wright I, van den Bogert AJ. The influence of orthotic devices and vastus
medialis strength and timing on patellofemoral loads during running. Clin
Biomech 2000;15(8):611e8.
Powers CM, Ward SR, Fredericson M, Guillet M, Shellock FG. Patellofemoral kine-
matics during weight-bearing and non-weight-bearing knee extension in per-
sons with lateral subluxation of the patella: a preliminary study. J Orthop Sports
Phys Ther 2003;33(11):677e85.
Roach S, Sorenson E, Headley B, San Juan JG. Prevalence of myofascial trigger points
in the hip in patellofemoral pain. Arch Phys Med Rehabil 2013;94(3):522e6.
Roush JR, Bay RC. Prevalence of anterior knee pain in 18e35 year-old females. Int J
Sports Phys Ther 2012;7(4):396.
Shirazi ZR, Moghaddam MB, Motealleh A. Comparative evaluation of core muscle
recruitment pattern in response to sudden external perturbations in patients
with patellofemoral pain syndrome and healthy subjects. Arch Phys Med
Rehabil 2014;95(7):1383e9.
Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol
Bull 1979;86(2):420.
Suter E, McMorland G, Herzog W, Bray R. Decrease in quadriceps inhibition after
sacroiliac joint manipulation in patients with anterior knee pain. J Manip
Physiol Ther 1999;22(3):149e53.
Suter E, McMorland G, Herzog W, Bray R. Conservative lower back treatment re-
duces inhibition in knee-extensor muscles: a randomized controlled trial.
J Manip Physiol Ther 2000;23(2):76e80.
Thomee
R, Augustsson J, Karlsson J. Patellofemoral pain syndrome. Sports Med
1999;28(4):245e62.
Van Tiggelen D, Cowan S, Coorevits P, Duvigneaud N, Witvrouw E. Delayed vastus
medialis obliquus to vastus lateralis onset timing contributes to the develop-
ment of patellofemoral pain in previously healthy men a prospective study. Am
J Sports Med 2009;37(6):1099e105.
Voight ML, Wieder DL. Comparative reflex response times of vastus medialis obli-
quus and vastus lateralis in normal subjects and subjects with extensor
mechanism dysfunction an electromyographic study. Am J Sports Med
1991;19(2):131e7.
Wainner RS, Whitman JM, Cleland JA, Flynn TW. Regional interdependence: a
musculoskeletal examination model whose time has come. J Orthop Sports
Phys Ther 2007;37(11):658e60.
Waryasz GR, McDermott AY. Patellofemoral pain syndrome (PFPS): a systematic
review of anatomy and potential risk factors. Dyn Med 2008;7(1):9.
Weir JP. Quantifying test-retest reliability using the intraclass correlation coefficient
and the SEM. J Strength Cond Res 2005;19(1):231e40.
Willson JD, Kernozek TW, Arndt RL, Reznichek DA, Straker JS. Gluteal muscle acti-
vation during running in females with and without patellofemoral pain syn-
drome. Clin Biomech 2011;26(7):735e40.
Witvrouw E, Sneyers C, Lysens R, Victor J, Bellemans J. Reflex response times of
vastus medialis oblique and vastus lateralis in normal subjects and in subjects
with patellofemoral pain syndrome. J Orthop Sports Phys Ther 1996;24(3):
160e5.