www.elsevier.com/locate/jelekin

Journal of Electromyography and Kinesiology 15 (2005) 596–603

Analysis on the activation of the VMO and VLL muscles duringsemisquat exercises with and without hip adduction in

individuals with patellofemoral pain syndrome

Kelly Rafael Ribeiro Coqueiro a, Debora Bevilaqua-Grossi b, Fausto Berzin c,Alcimar Barbosa Soares d, Cecılia Candolo a, Vanessa Monteiro-Pedro a,*

a Federal University of Sao Carlos, Rodovia Washington Luıs, Km 235, Sao Carlos-SP 13565-905, Brazilb Ribeirao Preto School of Medicine, Sao Paulo University, SP, Brazilc Dentistry College of Piracicaba, University of Campinas, SP, Brazil

d Federal University of Uberlandia, MG, Brazil

Abstract

Objective: The purpose of this study was to investigate the effect of hip adduction on the activity of the Vastus Medialis Obliquus

(VMO) and Vastus Lateralis Longus (VLL) muscles during semisquat exercises.

Methods: Twenty female subjects, divided into two groups comprising healthy and patellofemoral pain syndrome (PFPS) subjects

(ten volunteers for each group), performed three double-leg semisquat exercise trials with maximum hip adduction isometric con-

traction (DLSS-HA) and three double-leg semisquat exercise trials without hip adduction (DLSS). The normalized electromyo-

graphic muscle data were analysed using Repeated Measure ANOVA (p 6 0.05).

Results: The electrical activity of both VMO and VLL muscles was significantly greater during DLSS-HA exercise than during

DLSS (p = 0.0002) for both groups. Additionally, an independent Repeated Measure ANOVA revealed that the electric activity

of the VLL muscle was significantly greater (p = 0.0149) than that of the VMO muscle during DLSS exercises only for the PFPS

group. However, no differences were found during DLSS-HA exercises.

Conclusions: Although there was no preferential VMO muscle activation, the association of hip adduction with squat exercise pro-

moted a greater balance between the medial and lateral portions of the quadriceps femoris muscle and could be indicated for the

conservatory treatment of PFPS patients. The association of isometric hip adduction with isometric semisquat exercises produced a

more overall quadriceps activity and could be indicated for clinical rehabilitation or muscle strengthening programs.

� 2005 Elsevier Ltd. All rights reserved.

Keywords: Surface electromyography; Knee rehabilitation; Squat exercise; Quadriceps femoris muscle

1. Introduction

Patellofemoral pain syndrome (PFPS), a disorder re-

lated to peripatellar or retropatellar pain, most fre-

quently occurs in athletes and women [6]. It is

aggravated by a number of factors such as sports activ-

1050-6411/$ - see front matter � 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jelekin.2005.03.001

* Corresponding Author. Tel.: +55 16 3351 8754; fax: +55 16 3361

2081.

E-mail address: vanessa@power.ufscar.br (V. Monteiro-Pedro).

ities, stairs climbing, kneeling or squatting and pro-

longed sitting with the knee flexed [7,21]. Although itsaetiology has not yet been clearly defined [26], it is pos-

sible to relate it to various components that can result in

patellar malalignment, such as an increase in the Q an-

gle, patella alta, excessive subtalar pronation and exter-

nal tibial rotation [24,31,36]. Also, the imbalance caused

by the atrophy and dysplasia [12] of the oblique portion

of the Vastus Medialis Obliquus muscle (VMO) plays an

important role in patellar malalignment. Such

K.R.R. Coqueiro et al. / Journal of Electromyography and Kinesiology 15 (2005) 596–603 597

imbalance occurs because the VMO muscle cannot op-

pose to the counteracting force produced by both the

vastus lateralis (VL) muscle and the iliotibial band [9].

The result is that the combination of all those elements

contributing to patellar malalignment can seriously im-

pair the recruitment pattern of the quadriceps musclein PFPS subjects.

Therefore, treatment protocols for PFPS must aim

primarily at the functional recovery of the VMO muscle

[5]. However, it is important to find out what the correct

protocol is and, even more important, to find out if the

chosen exercises are really contributing to prompt and

quick muscle recovery. In order to do so, one of the main

techniques used by researchers worldwide is to measurethe electric activity of muscles (electromyography) dur-

ing specific exercises [17,27]. But what would the best

exercises for PFPS treatment be? Unfortunately, no

agreement has been reached on that matter. Knee exten-

sion exercises in both open kinetic chain (OKC) and

closed kinetic chain (CKC), for instance, have been sug-

gested to be used in association with tibial rotation [1,21],

hip rotation [20] and hip adduction [14,15,27]. However,it is noteworthy that hip adduction exercises can be used

to selectively strengthen the VMO muscle [14,23,27,30]

since part of it originates in the adductor magnus muscle

and part of it originates in the adductor longus muscle

[4,22]. But what kind of chain (OKC or CKC), for hip

adduction exercises or for hip adduction exercises associ-

ated with knee extension, should be performed to pro-

duce the best selective strengthening of the VMOmuscle?A number of studies can be found in the literature

concerning the activation of the VMO and VL muscles

during different exercises. Nevertheless, most of them

cannot be applied to answer that question since they

are not directly related to it, are not conclusive and do

not agree as to many aspects. Some authors have re-

ported an increase in the VMO muscle activity as com-

pared to VL activity when OKC isometric hip adductionexercises are performed by healthy subjects [14]. Others,

however [37], have not found such effect. Also, when iso-

metric adduction is associated with OKC knee extension

exercises, no increase in the VMO muscle activity as

compared to VL activity has been reported for healthy

[6,19,21] or PFPS subjects [6,21].

CKC exercises such as squat have been widely indi-

cated for knee rehabilitation [11,28] since the functionalnature of those exercises, including co-contraction of the

quadriceps femoris and hamstrings, decreases tibiofem-

oral translation and compression forces on the tibiofem-

oral joint [36], thus promoting strengthening and normal

proprioceptive input [26,33]. Gresalmer and Klein [13]

and Escamilla [10] reported that such exercises should

be performed in PFPS subjects with the knee flexed at

0–45� or 0–50�.The electric activity of the VMO and VL muscles dur-

ing hip adduction exercises with knee extension have

been studied for healthy [6,9] and PFPS [6] subjects.

The results show that those exercises increase the electric

activity of both muscles. Furthermore, it has been

shown that exercises also lead to a higher electric activ-

ity of the VMO muscle as compared to the electric activ-

ity of the lateral portions of the quadriceps muscle forhealthy [15] and PFPS subjects [29].

It is noteworthy that, although a number of works

have described the activity of VMO and, considering

anatomic descriptions and the long portion of the VL

muscle, i.e., the Vastus Lateralis Longus muscle –

VLL [17,28,30] during different exercises, very little has

been reported about the association of double-leg semi-

squat exercises with hip adduction. To this date, theauthors of this study have found only one report of

hip adduction associated with mini-squat exercises in

healthy subjects [9]. Nothing has been found concerning

the electric activity of the VMO and the VLL muscles

during double-leg semisquat exercises associated with

hip adduction in PFPS subjects.

In order to fill that gap and to shed new light on fea-

sible treatment protocols for PFPS, the authors in thisstudy decided to investigate the effects of hip adduction

on the electric activity of the VMO and VLL muscles

during double-leg semisquat exercises at 45� knee flexionin healthy and PFPS subjects.

2. Methods

2.1. Subjects

The population for this study consisted of twenty fe-

male volunteers divided into two groups:

� Ten healthy subjects (age: 21.8 ± 2.52 years old,

height: 165 ± 0.043 cm, weight: 58.38 ± 5.88 kg): sub-

jects who did not report pain in the knee area and hadnever had any significant knee or lower limb pathol-

ogy or surgery.

� Ten sedentary PFPS subjects (age: 23.2 ± 2.65 years

old, height: 158 ± 0.056 cm, weight: 50.53 ±

5.83 kg): subjects who did not report any previous

muscle-skeletal damage to the hip, knee or ankle

joints and were submitted to physical evaluation con-

sisting of specific tests for those joints.

The following aspects defined the inclusion criteria

for PFPS volunteers in this work:

� Asymptomatic for at least three months prior to the

physical evaluation [7].

� Reports of episodes of anterior or retropatellar knee

pain when performing at least two of the followingactivities: prolonged sitting, squatting, ascending or

descending stairs, running, kneeling and jumping [7].

598 K.R.R. Coqueiro et al. / Journal of Electromyography and Kinesiology 15 (2005) 596–603

� Three or more of the following signs and symptoms

had to be detected: excessive pronation of the subtalar

joint, patella alta, hyper or hypomobile patella, tight-

ening of the iliotibial band (Ober�s Test), increase of

the Q angle, external tibial rotation, squeezing patel-

lae and pain on patella palpation [7,24].

The exclusion criteria were based on:

� History of knee or lower limb surgery [7].

� History of patellar dislocation or subluxation [7].

� Clinical evidence of meniscal or ligamentous lesions

and patellar tendon pathology [7].

Table 1 shows the percentage of incidence of the signs

and symptoms used in the inclusion criteria of PFPS

volunteers.

This study was approved of by the Ethics Committee

for Human Experimentation Research of the Federal

University of Sao Carlos and all the subjects were asked

to sign a consent form.

2.2. Instrumentation

The electrical activity of the VMO and VLL muscles

was detected by active differential surface EMG elec-

trodes, supplied by Lynx Electronics Technologies (Bra-

zil). The main features of those electrodes are:

� Detection surfaces: two Ag parallel bars – 10 · 2 mmof contact area.

� Distance between electrodes: 10 mm.

� Gain: 100·.� Input impedance: higher than 100 MX.� Common mode rejection ratio (CMRR): higher than

80 dB @ 60 Hz.

Signal conditioning included a further gain of 10·and a band pass filter tuned at 20 and 500 Hz. A 12-

bit A/D card (CAD 12/26–60 K by Lynx Electronics

Technologies) was used to sample the data at 2000 Hz.

The electromyographic (EMG) signals were synchro-

nously sampled (simultaneous sample and hold) and

stored for later processing.

Table 1

Percentage of the incidence of the chosen signs and symptoms in PFPS

volunteers (n = 10)

Signs and symptoms %

External tibial rotation 30

Excessive pronation of the subtalar joint 30

Increase of the Q angle 100

Hyper or hypomobile patella 10

Squeezing patellae 60

Patella alta 20

Pain on patella palpation 40

Tight iliotibial band (Ober�s Test) 20

A knee extension table was used for the maximum

isometric voluntary contraction (MIVC) exercises and

the angular position of the knee was measured by a uni-

versal goniometer. Hip adduction exercises were per-

formed against a mechanical resistance device, as

shown in Fig. 1 [29].

2.3. Procedures

A reference line, joining the anterosuperior iliac spine

to the centre of the patella, was drawn with a pen to help

define the correct position of the surface EMG elec-

trodes. The ‘‘VMO electrode’’ was fixed approximately

4 cm superior and medial to the superomedial borderof the patella [14] at a 50–55� angle to the reference line

[22]. The ‘‘VLL electrode’’ was fixed 15 cm superior to

the superolateral border of the patella at approximately

13.6� to the reference line [3]. The electrodes were posi-

tioned, with the subjects standing, on the midline of the

belly muscle with the detection surface (two parallel

bars) perpendicular to the direction of the muscle fibres,

as suggested by De Luca [8]. The reference electrode waspositioned on the volunteer�s left wrist and held in place

by Velcro tape. Adhesive tape was used to hold the ac-

tive electrodes in place. To minimise the contact imped-

ance, the skin was shaved, swabbed with alcohol 70%

and abraded with sandpaper. The electrodes were placed

on only one of the volunteers� legs. For the PFPS group,

a leg was randomly chosen; for the healthy group, the

leg with one or no PFPS signals was selected (in the casewhere both legs presented the same number of signals, a

random choice was also made).

Prior to the experiments, the volunteers were submit-

ted to a short training in order to become used to testing

procedures. This included a warm-up session and the

stretching of the adductor, quadriceps, hamstrings and

calf muscles for one minute.

The volunteer was then stabilised on the knee exten-sion table with the ankles, knees and hips flexed at 90�.

Fig. 1. Illustration of the mechanical resistance device designed to

provide resistance for hip adduction placed over the knee joint

interline.

ig. 2. Lateral view of DLSS and DLSS-HA exercises at 45� knee

exion.

K.R.R. Coqueiro et al. / Journal of Electromyography and Kinesiology 15 (2005) 596–603 599

A thoracic belt encircling the back of the table was fas-

tened under the axillae and a pelvic belt, also fixed to the

table, was fastened around the subject�s pelvis, just un-der the anterior superior iliac spines. A resistance pad

was positioned 2.5 cm above the medial malleolus, with

the knee flexed at 90�, in order to provide maximumresistance to the extension of knee.

The average of the RMS (root mean square) values of

three MIVC knee extension exercises performed on the

knee extension table was used as benchmark to normal-

ise the EMG recordings obtained during the double-leg

semisquat exercises. To do so, the volunteers were asked

to extend their knee as hard as possible and encouraged

by continuous verbal commands (Attention! Prepare!Go! Push! Push! Relax!). The three MIVCs, lasting 6 s

each, were performed following two-minute intervals

(resting period).

In this study, double-leg semisquat (DLSS) exercise

refers to semisquat exercise on two legs performed with

the knees flexed at 45�. Double-leg semisquat exercise

associated with hip adduction (DLSS-HA) refers to

semisquat exercise on two legs performed with the kneesflexed at 45� associated with hip adduction at MIVC

against a mechanical resistance (Fig. 1).

As described earlier, the angular position of the knee

was measured by a universal goniometer. The device�srotation centre was positioned directly over the joint�srotation centre. The proximal arm was fixed over the

thigh�s lateral line and aligned with the lateral midline

of the femur, using the greater throcanter as reference.The distal arm was fixed on the lateral aspect of the

lower leg and aligned with the lateral midline of the fib-

ula, using the lateral malleolus as reference [18].

DLSS-HA exercises were performed with the

mechanical device, which was created to provide resis-

tance to hip adduction, on the knee joint interline

throughout the whole exercise. This provided a 30� an-gle of abduction during the exercises. On the other hand,DLSS exercises used that device only to establish the ini-

tial position of the knee and then removed, just before

the beginning of the exercise. This ensured that both

exercises began with the knees in the same position.

The volunteers were also instructed to maintain the

trunk in the upright position, the feet in the neutral po-

sition and the shoulders flexed at 90� (Fig. 2) for both

exercises. This arrangement aimed to ensure that themovement was as stable as possible.

DLSS exercises were executed according to the fol-

lowing sequence:

1. descend to 45� knee flexion (Fig. 2),

2. hold for 6 s and

3. ascend to the initial position.

Three sequences were executed following two-minute

intervals. During the exercise the volunteer was verbally

F

fl

encouraged with commands such as Hold it! Hold it!Relax!

DLSS-HA exercises were executed as follows:

1. descend to 45� knee flexion (Fig. 2),

2. hold and perform a hip adduction MIVC against the

mechanical resistance for 6 s and

3. ascend to the initial position.

Three sequences were executed following two-minute

intervals. During the exercise the volunteer was verbally

encouraged with commands such as Squeeze it! Squeeze

it! Relax!

The volunteers executed the exercise sequences

(DLSS and DLSS-HA) at 4-minute resting intervals.

The order in which the sequences were executed was

randomly defined.EMG data acquisition began 2 s after the volunteer

got into the ‘‘hold’’ phase for DLSS exercises, and 2 s

600 K.R.R. Coqueiro et al. / Journal of Electromyography and Kinesiology 15 (2005) 596–603

after he began hip adduction at MIVC for DLSS-HA

exercises. These data were analysed in terms of their

RMS values, normalised by the mean RMS value of

the three knee extension MIVC executed on the resis-

tance knee extension table [14]

DLSS or DLSS�HA

knee Extension MIVC� 100.

2.4. Statistical analysis

Tests of Repeated Measure Analysis of Variance were

used to evaluate the differences between groups, muscles

and exercises. The analyses were performed between sub-

jects to evaluate groups and among subjects to evaluate

muscles and exercises. Significance levels were set at

p 6 0.05.

Independent repeated measure analyses of variance

were also performed for each group. Those were definedby tests of hypotheses among subjects for the effects of

muscles and exercises.

Prior to statistical analysis, statistical normality tests

were performed and descriptive analysis was carried out.

Calculations were performed using the R statistical free

software [16].

3. Results

The global analyses involving comparisons between

and among subjects showed that there were no signifi-

cant differences for groups (0.0935) and a small signifi-

cance level for muscles (p = 0.6122). On the other

hand, the analyses of exercise effects showed a high level

of significance (p = 0.0002). In other words, for bothgroups, the electric activity of the VMO and VLL mus-

cles were significantly greater during DLSS-HA exer-

cises than during DLSS exercises.

A separate analysis was also performed for the

healthy and PFPS groups. It showed a significant dif-

ference in muscle effect with greater electric activity

of the VLL muscle (p = 0.0049) than that of the

VMO muscle during DLSS exercise only for the PFPSgroup.

Table 2

Mean and standard deviation of normalised electromyographic readings (suc

VLL muscles during double-leg semisquat exercise (DLSS) and double-leg se

Healthy subjects

VMO VLL

DLSS 16.14 (±5.96) 22.64 (±6.79)

DLSS-HA 35.6 (±13.9)a 39.57 (±8.82)

a Electric activity of the VMO and VLL muscles was significantly greaterb Electric activity of the VLL muscle was significantly greater (p = 0.0049)

group.

The normalised RMS mean values for the electric

activity of the VMO and VLL muscles during the

exercises performed by both groups are shown in

Table 2.

4. Discussion

The experiments presented here show that there was a

significant increase in the activity of the VMO and the

VLL muscles when double-leg semisquat exercise was

associated with hip adduction (DLSS-HA) for both

healthy and PFPS groups. Those findings are in agree-

ment with previous studies that report an increase inthe electrical activity of the VMO and VL muscles when

isometric hip adduction was combined with dynamic

mini-squat at 30� knee flexion [9], wall-sit up to 60� kneeflexion [15] or wall-slide up to 45� knee flexion [6] in

healthy [6,9,15] and PFPS subjects [6]. Therefore, it

seems that, whatever the type of squat exercise, i.e.,

mini-squat, semi-squat, wall-sit or wall-slide exercises,

there is a tendency to an overall increase in the activityof the quadriceps femoris muscle when such exercise is

associated with hip adduction for both healthy and

PFPS subjects.

However, the authors of this study did not find any

evidence of preferential activation of the VMO muscle

as reported by Hodges and Richardson [15] and Miller

et al. [25], who studied the effect of hip adduction asso-

ciated with dynamic wall-sit at 60� knee flexion andwall-slide at 75� knee flexion. It is possible that dynamic

exercises and higher knee flexion angles could produce

such differences. However, some of those studies used

EMG data collected during dynamic contractions,

which is very controversial, since the EMG signal can

be altered by the relative movement of the electrode with

respect to the contracting muscle fibres. The fact is that

the authors of this study as well as many others, such asDe Luca, 1997 [8], normally recommend avoiding col-

lecting EMG data during anisometric contractions (as

it was the case for the experiments shown in this work).

But, even if there was no error in the technique used for

EMG data acquisition, other questions could still arise

for our experiments, such as ‘‘why did we choose 45�

h as percentage of knee extension at 90� knee flexion) of the VMO and

misquat exercise associated with MIVC of hip adduction (DLSS-HA)

PFPS subjects

VMO VLL

26.96 (±10.21) 35.53 (±10.55)b

a 34.96 (±13.05)a 41.45 (8.25)a

during DLSS-HA exercise than during DLSS exercise.

than that of the VMO muscle during DLSS exercise only for the PFPS

K.R.R. Coqueiro et al. / Journal of Electromyography and Kinesiology 15 (2005) 596–603 601

knee flexion? Why did we choose to normalise the data

using MIVC? . . .’’.Steinkamp et al. [34] suggested that CKC knee exten-

sion exercises in the functional range of motion should

be emphasised to strengthen the knee joint muscles be-

cause of its lower stress on the patellofemoral joint.Therefore, it seems that those exercises are good for

the PFPS treatment. However, what would optimum

knee flexion during squat exercises be? The answer could

be pointed by Escamilla [10], who stated �. . .because thepeak compressive forces generally occur near maximum

knee flexion, individuals with patellofemoral disorders

should avoid performing the squat at high knee flexion an-

gles�. But what would those ‘‘high knee flexion angles’’be? It certainly depends on each patient. In view of such

situation, the authors in this study elected 45� as a good

compromise for all the participant volunteers.

The normalisation of the raw data is an important

requirement if one intends to study and compare data

collected from different muscles and different individuals

in different situations. MIVC was elected as the normal-

isation unit based on the experience of many researchersworldwide and its use in similar studies concerning

PFPS and the muscles of the thigh [1,6,9,14,18,25,29].

However, we used data collected during MIVC exten-

sion of the knee at OKC to normalise CKC contrac-

tions. According to Schaub and Worrel [32], this is not

a problem either, since they have shown that muscle

recruitment patterns are very similar for OKC and

CKC exercises.Nevertheless, considering that there are differences

between the evaluated groups, especially regarding mus-

cle imbalance in PFPS subjects, widely reported in the

literature [1,7,9,20,21,23,24,29,33,35,36], we felt that a

separated statistical analysis comparing, for each group,

electric activity between the VMO and the VLL muscles

during DLSS and DLSS-HA exercises would be clini-

cally important for knee rehabilitation programs. Thisindependent analysis revealed that, the electrical activity

of the VLL muscle was significantly greater than that of

the VMO muscle during DLSS exercise for the PFPS

group and no differences were found during DLSS-HA

exercise. The greater electrical activity of the VLL mus-

cle when compared with that of the VMO muscle during

DLSQ exercise confirms the reports of imbalance

between the medial and lateral compartments of thequadriceps femoris muscle in PFPS subjects

[1,7,9,12,20,21,23,24,29,33,35,36]. However, this theory

should be carefully dealt with, since this work investi-

gated only EMG data and other tools, such as magnetic

resonance imaging (MRI) and computed tomography

(CT), should be added to explore the investigation of

muscle imbalance.

However, although there was no selective recruit-ment of the VMO muscle during DLSS-HA exercise,

considering that during DLSS exercise there was a sig-

nificant difference between the VLL and the VMO

muscles and that such difference had not been found

during DLSS-HA, we believe that DLSS-HA exercises

could promote a greater muscle balance between the

electric activity of the VMO and VLL muscles and

could benefit the treatment of PFPS patients. Thisgreater balance caused by the effect of hip adduction

on the activity of the VMO muscle could be explained

by the strong anatomical relation between the adductor

muscles and the VMO muscle. The action of adductor

muscles is essential to selectively strengthen the VMO

muscle. According to Hanten and Schulties [14],

‘‘. . .a strong VMO originating from weak adductors

would serve only to draw the adductor tendons toward

the patella . . .’’ and, on the other hand, ‘‘. . . strong hip

adductors give the VMO a stable origin from which to

contract . . .’’. Besides, another favourable factor for

the selective contraction of the VMO muscle during

DLSS-HA exercise refers to the length tension proper-

ties, as proposed by Beck and Wildermuth [2] who re-

ported that with the addition of hip adduction there is

a stretch to the VMO muscle, which would alter lengthtension properties, thus contributing to an enhanced

contraction force.

This study was designed with the primary goal of pro-

viding information regarding the effects of hip adduction

on the electric activity of the VMO and VLL muscles in

healthy and PFPS subjects. It has demonstrated that the

association of isometric hip adduction with isometric

semi-squat exercise produces a balanced activity of thequadriceps muscle group. Therefore, this exercise can

be indicated when overall quadriceps activity is desired

for clinical rehabilitation or muscle strengthening pro-

grams. Furthermore, because of the greater balance be-

tween the medial and lateral portions of the quadriceps

femoris muscle promoted by the association of semi-

squat exercise at 45� knee flexion and hip, it could also

be indicated for the conservatory treatment of PFPS.However, to confirm this statement, further EMG inves-

tigation would be necessary in order to compare the ef-

fects of muscle training programs based on DLSS-HA

exercises for both groups.

Acknowledgements

The authors thank Mrs. Lucia Batista for the illustra-

tions in this paper, CNPq-Brazil (Process No. 52419096-

8) and CAPES-Brazil for supporting this study.

References

[1] P.H. Andrade, D. Bevilaqua-Grosso, F. Berzin, I.A. Gil, V.

Monteiro-Pedro, Comparacao da atividade eletrica do musculo

vasto medial oblıquo e vasto lateral oblıquo em indivıduos com

602 K.R.R. Coqueiro et al. / Journal of Electromyography and Kinesiology 15 (2005) 596–603

disfuncao femoro-patelar, Revista de Fisioterapia da Universid-

ade de Sao Paulo 8 (2) (2001) 65–71.

[2] J.L. Beck, B.P. Wildermuth, The female athlete�s knee, Clinical

Sports Medicine 4 (1985) 345.

[3] D. Bevilaqua-Grossi, V. Monteiro-Pedro, G.C. Souza, Z. Silva,

F. Berzin, Contribution to the Anatomical Study of the Oblique

Portion of the Vastus Lateralis Muscle, Brazilian Journal of

Morphological Science 21 (1) (2004) 42–47.

[4] K. Bose, R. Kanagasuntherum, M. Osman, Vastus Medialis

Oblique: An Anatomical and Physiologic Study, Orthopedics 3

(1980) 880–883.

[5] M.J. Callagham, J.A. Oldham, The role of quadriceps exercise in

the treatment of patellofemoral pain syndrome, Sports Medicine

21 (1996) 384–391.

[6] K. Cerny, Vastus medialis oblique/ vastus lateralis muscle

activity ratios for selected exercises in persons with and without

patellofemoral pain syndrome, Physical Therapy 75 (1995) 672–

683.

[7] S.M. Cowan, K.B.L. Bennel, P. Hodges, K.M. Crossely, J.

McConnell, Delayed Onset of Electromyographic Activity of

Vastus Medialis Obliquos relative to Vastus Lateralis in subjects

with Patellofemoral Pain Syndrome, The Archives of Physical

Medicine Rehabilitation 82 (2001) 183–189.

[8] C.J. De Luca, The use of Surface Electromyography in Biome-

chanics, Journal of Applied Biomechanics 13 (1997) 135–163.

[9] J.E. Earl, B.L. Schimitz, B.L. Arnold, Activation of the VMO and

VL during dynamic mini-squat exercises with and without

isometric hip adduction, Journal of Electromyography and

Kinesiology 11 (1992) 381–386.

[10] R.F. Escamilla, Knee biomechanics of the dynamic squat

exercise, Medicine and Science in Sports and Exercise 33

(2001) 127–141.

[11] G.K. Fitzgerald, Open versus closed kinect chain exercise: issues

in rehabilitation after anterior cruciate ligament reconstructive

surgery, Physical Therapy 77 (1997) 1747–1754.

[12] T.A. Fox, Dysplasia of the quadriceps mechanism: hypoplasia of

the vastus medialis muscle as related to the hypermobile patella

syndrome, Surgical Clinics of North America 55 (1975) 199–206.

[13] R. Gresalmer, J. Klein, The biomechanics of the patellofemoral

joint, Journal of Orthopaedic and Sports Physical Therapy 28

(1998) 286–297.

[14] W.P. Hanten, S. Schulthies, Exercise effect on electromyographic

activity of the vastus medialis oblique and vastus lateralis muscles,

Physical Therapy 7 (1990) 561–565.

[15] P.W. Hodges, C.A. Richardson, The influence of isometric hip

adduction on quadriceps femoris activity, Scandinavian Journal of

Rehabilitation Medicine 25 (1993) 57–62.

[16] R. Ihaka, R.A. Gentleman, Language for Data Analysis and

Graphics, Journal of Computational and Graphical Statistics 5

(1996) 299–314.

[17] J. Insall, Current Concepts Review Patellar Pain, Journal of Bone

and Joint Surgery 64-A (1982) 147–152.

[18] J.A. Isear, J.C. Erickson, T.W. Worrel, EMG analysis of lower

extremity muscle recruitment patterns during an unloaded squat,

Medicine and Science in Sports and Exercise 29 (4) (1997) 532–

539.

[19] G.M. Karst, P.D. Jewett, Eletromyography Analysis of Exercises

Proposed for Differential Activation of Medial and Lateral

Quadriceps Femoris Muscle Components, Physical Therapy 73

(1993) 289–299.

[20] P.L. Lam, G.Y.F. Ng, Activation of the Quadriceps Muscle

during Semisquating with Different Hip and Knee Positions in

Patients with Anterior Knee Pain, Amerian Journal of Physical

Medicine and Rehabilitation 80 (2001) 804–808.

[21] J. Laprade, E. Culhan, B. Brouwer, Comparison of five

isometric exercises in the recrutment of the vastus medialis

oblique in persons with and without patellofemoral pain

syndrome, Journal of Orthopaedic and Sport Physical Therapy

27 (1998) 197–204.

[22] F.J. Lieb, J. Perry, Quadriceps functions: an anatomical and

mechanical study using amputed limbs, Journal of Bone and Joint

Surgery 53A (1968) 1535–1548.

[23] J. Mc Connell, The management of chondromalacia patellae: a

long-term solution, Australian Journal of Physiotherapy 32 (1986)

215–223.

[24] J. McConnell, J. Cook, Anterior Knee Pain. In: P. Burker, K.

Kham, Clinical Sports Medicine. 2000 (Chapter 24). Available

from:Æwww.clinicalsportsmedicine.comæ.[25] J.P. Miller, D. Sedory, R. Croce, Vastus Medialis Obliquus and

Vastus Lateralis Activity in Patients with and Without Patellofe-

moral Pain Syndrome, Journal of Sports Rehabilitation 6 (1997)

1–10.

[26] V. Monteiro-Pedro, M. Vitti, F. Berzin, D. Bevilaqua-Grosso,

Electromyographic activity of vastus medialis oblique muscle in

step-up and step-down exercises, Brazilian Journal of Morpho-

logical Science 14 (1997) 19–23.

[27] V. Monteiro-Pedro, M. Viiti, F. Berzin, D. Bevilaqua-Grosso,

The effect of free isotonic contraction exercises of the hip

adduction on vastus medialis oblique muscle: an electromyo-

graphic study, Electromyography and. Clinical Neurophysiology

39 (1999) 435–440.

[28] J.C. Ninos, J.J. Irrgang, R. Burdett, J.R. Weiss, Electromyo-

graphic analysis of the squat performed in self-selected lower

extremity neutral rotation and 30� of lower extremity turn-out

from self-selected neutral position, Journal of Orthopaedic and

Sport Physical Therapy 25 (1997) 307–315.

[29] C.V. Nunes, V. Monteiro-Pedro, Efeito do Exercıcio Isometrico

de Extensao do Joelho associado a Aducao Isometrica do

Quadril na Atividade Eletrica dos Musculos Vasto Medial

Oblıquo e Vasto Lateral Oblıquo em Indivıduos com Disfuncao

Femoro-patelar, Revista Brasileira de Fisioterapia 7 (2) (2003)

145–150.

[30] L. Reynolds, T.A. Levin, J.M. Medeiros, N.S. Adler, A.

Hallun, EMG activity of the Vastus Medialis oblique and the

vastus lateralis in their role in patellar alignment, American

Journal of Physical Medicine and Rehabilitation 62 (1983) 61–

70.

[31] M.T. Ruffin, R.B. Kinninghan, Anterior knee pain: the challenge

of patellofemoral syndrome, American Family Physician 43 (1993)

185–194.

[32] P.A. Schaub, T.W. Worrel, EMG activity of six muscles and

VMO:VL ratio determination during a maximal squat exercise,

Journal of Sports Rehabilitation 4 (1995) 203–209.

[33] H.A. Stiene, T. Brosky, M.F. Reinking, J. Nyland, M.B. Mason,

A comparison of closed kinetic chain and isokinetic joint

isolation exercise in patients with patellofemoral dysfunction,

Journal of Orthopaedic and Sports Physical Therapy 24 (1991)

136–141.

[34] L.A. Steinkamp, M.F. Dillingham, M.D. Markel, J.A. Hill, K.R.

Kaufman, Biomechanical considerations in patellofemoral joint

rehabilitation, American Journal of Sports Medicine 21 (1993)

438–444.

[35] R. Thomee, J. Augustsson, J. Karlsson, Patellofemoral Pain

Syndrome: A review of current issues, Sports Medicine 28 (4)

(1999) 245–262.

[36] E. Witvrouw, C. Sneyers, R. Lysens, J. Victor, J. Bellemans,

Reflex response times of vastus medialis oblique and vastus

lateralis in normal subjects and in subjects with patellofemoral

pain syndrome, Journal of Orthopaedic and Sports Physical

Therapy 24 (1996) 160–165.

[37] D. Zakaria, K.L. Harburn, J.F. Kramer, Preferential activation

of the Vastus Medialis oblique, Vastus lateralis and hip adductor

muscles during isometric exercises in females, Journal of Ortho-

paedic Sports Physical Therapy 26 (1) (1997) 23–28.

K.R.R. Coqueiro et al. / Journal of Electromyography and Kinesiology 15 (2005) 596–603 603

Kelly Rafael Ribeiro Coqueiro is physio-

therapist and concluded her Masters degree

in 2003. She is a candidate of the Post

Graduation Program in Physical Therapy

of the Federal University of Sao Carlos

(doctoral degree).

Dr. Debora Bevilaqua-Grossi is a currently

Professor of Physical Therapy curse at the

Department of Biomechanics, Medicine

and Rehabilitation of the Locomotor

Apparatus, Ribeirao Preto School of

Medicine, Sao Paulo University, Brazil. She

receive her bachelor�s and Master�s degree

from the University of Campinas, Brazil.

Dr. Fausto Berzin (DDS, PhD) is Dean

Professor of Electromyography Graduate

Course from Campinas State University

(FOP/UNICAMP) and Invited Professor of

Sao Carlos Federal University. He is

responsible for the Electromiography Lab-

oratory of FOP/UNICAMP, Chairman of

Brazilian Society Electromyography and

Kinesiology and Member of ISEK. He is

member of the Editorial Board of 5 Scien-

tific Journals. Under his supervision 6

Electromyography Laboratories in different

Universities were installed in Brazil. He has contributed for the project

and construction of a Brazilian Electromyograph as well as of its

software, which has been the most utilized equipment in Brazilian

research.He has also worked with Clinical Neurophysiology. Presently

his work focuses on myogenics disfunctions that cause pain.

Dr. Alcimar Soares graduated at the Fed-

eral University of Uberlandia (Brazil) in

1987 (Electrical Engineering) where he also

finished his MSc in Artificial Intelligence

(1990). In 1997 he received his PhD title

(Biomedical Engineering) by the University

of Edinbugh (UK). Since 1990 he has been

a lecturer at the Department of Electrical

Engineering of the Federal University of

Uberlandia (Brazil). Nowadays, he heads

the Biomedical Engineering and Automat-

ics Lab at that institution and supervises a

number of graduate and postgraduate students in the areas of bio-

mechanics, biomedical instrumentation and biomedical signal

processing.

Dr. Cecılia Candolo is currently Professor

of Statistics at the Federal University of

Sao Carlos, Brazil. She received her Ph.D.

in Statistics from the University of Sao

Paulo and her Bachelor�s and Master�sdegrees in Statistics from the University of

Campinas, Brazil.

Dr. Vanessa Monteiro-Pedro is a physio-

therapist. She is a senior lecturer in the

Departament of Physiotherapy in the Fed-

eral University of Sao Carlos (UFSCar).

She got her masters degree in Pathology

(1990) and her Phd. in Anatomy (1995)

from the State University of Campinas

(Brazil). She is member of the International

Association for the Study of Pain (LASP),

Brazilian Chapter and of the Brazilian

Society of Electromyography and Kinesi-

ology and member of International Society

of Electromyography and Kinesiology (ISEK).