Available online at www.sciencedirect.

com

Journal of Electromyography and Kinesiology 19 (2009) 105–112

www.elsevier.com/locate/jelekin

Muscular activation patterns during active prone

hip extension exercises
Ana Cristina Lamounier Sakamoto, Luci Fuscaldi Teixeira-Salmela *,
Fátima Rodrigues de Paula-Goulart, Christina Danielli Coelho de Morais Faria,
Cristiano Queiroz Guimarães
Department of Physical Therapy, Universidade Federal de Minas Gerais, Avenida Antônio Carlos, 6627,
Campus Pampulha, 31270-901 Belo Horizonte, Minas Gerais, Brazil

Received 14 March 2007; received in revised form 5 July 2007; accepted 5 July 2007

Abstract

Background: Changes in activation patterns of hip extensors and pelvic stabilizing muscles are recognized as factors that cause low back
disorders and these disturbances could have an impact on the physiological loading and alter the direction and magnitude of joint reac-
tion forces.
Objective: To investigate activation patterns of the gluteus maximus, semitendinosus and erector spinae muscles with healthy young indi-
viduals during four different modalities of therapeutic exercise.
Methods: Thirty-one volunteers were selected: (16 men and 15 women), age (24.5 ± 3.47 years), body mass of 66.89 ± 11.89 kg and a
height of 1.70 ± 0.09 m). They performed four modalities of therapeutic exercise while the electromyographic activity of the investigated
muscles was recorded to determine muscle pattern activation for each exercise.
Results: Repeated measure ANOVA revealed that muscle activation patterns were similar for the four analyzed exercises, starting with the
semitendinosus, followed by the erector spinae, and then, the gluteus maximus. The gluteus maximus was the last activated muscle during
hip extension associated with knee flexion (p < 0.0001), knee extension (p < 0.0001), and with lateral rotation and knee flexion (p < 0.05).
Conclusion: Findings of the present study suggested that despite individual variability, the muscle firing order was similar for the four
therapeutic exercises.
Ó 2007 Elsevier Ltd. All rights reserved.

Keywords: Gluteus maximus; Therapeutic exercises; Electromyography; Muscle strengthening

1. Introduction achieving coordinated activity between all muscles within

Deficiencies in movement patterns play a major role in
the development of musculoskeletal dysfunctions, espe-
cially in the peripheral part of the locomotive system (Jull
and Janda, 1987). Studies suggest that low back disorders
are associated with muscle imbalance (Cholewicki et al.,
2003; Hungerford et al., 2003; Leinonen et al., 2000; Moo-
ney, 1999, 2001; Norris, 1995; Vogt et al., 2003) and cur-
rently, emphasis has been placed on the importance of
*
Corresponding author. Tel./fax: +55 31 3499 4783.
E-mail address: lfts@ufmg.br (L.F. Teixeira-Salmela).
a balanced muscular system for the prevention and treat-
ment of low back pain (Cholewicki et al., 2003; Jull and
Janda, 1987; Norris, 1995).
Normal functioning of the spine depends not only on
passive joint mobility, but also on normal muscle activity
and central nervous system regulation. Muscles produce
and control movement and are also dynamic stabilizers
of the spine, protecting it from overloads that can be inher-
ent in normal function (Jull and Janda, 1987; Norris, 1995).
According to Panjabi’s stability concept, disturbances in
the musculoskeletal system can either be the causes or con-
sequences of pathological syndromes in the lumbar spine
(Panjabi, 1992, 2003).

1050-6411/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jelekin.2007.07.004
106 A.C.L. Sakamoto et al. / Journal of Electromyography and Kinesiology 19 (2009) 105–112
Gluteus maximus has a major functional importance in
the early stance in walking and generates most of the sup-
port and prevents collapse of the hip, knee and ankle
(Anderson and Pandy, 2003). Loading response is charac-
terized by the free fall of body gravity and 60% of the body
weight is transferred in 0.02 s, resulting in abrupt loading
of the forward limb. It is at in this moment that the gluteus
maximus compresses the sacroiliac joint to provide stability
(Hossain and Nokes, 2005). The inappropriate activation
of the gluteus maximus in gait is thought to be a cause of
low back pain, resulting in a deficiency in the shock absorp-
tion mechanism at the sacroiliac joint (Hossain and Nokes,
2005). Biceps femoris has been shown to have an early
onset in sacroiliac joint pain subjects (Hungerford et al.,
2003), and is thought to be a compensation for the weak-
ness of the gluteus maximus (Hossain and Nokes, 2005).
Coordinated muscular activity seems to be important in
preventing and treating low back pain. There is evidence
that delayed gluteus maximus activity and early biceps
femoris activity in the supporting leg, when the contralat-
eral hip is flexed, occurs in patients with sacroiliac joint pain
(Hungerford et al., 2003). Activation patterns of the gluteus
maximus in subjects with low back pain during exercise or
functional activities has been reported to be either inhibited
(Hungerford et al., 2003; Kankaanpaa et al., 1998; Leinon-
en et al., 2000; Mooney, 1999) or hyperactive (Clark et al.,
2002, 2003; Mooney et al., 2001; Vogt et al., 2003). In sub-
jects where the gluteus maximus was inhibited, earlier acti-
vation of the hamstrings and erector spinae muscles
occurred to stabilize the lumbar spine (Sahrmann, 2002;
Vogt and Banzer, 1997). If these disturbances occur, they
can be detrimental, i.e., they may cause low back disorders
by having detrimental effects on physiological loading.
Studies that have analyzed the activation patterns of the
hip extensors in the prone positions have shown contradic-
tory results (Bullock-Saxton et al., 1994; Lehman et al.,
2004; Pierce and Lee, 1990; Vogt and Banzer, 1997). At this
point, we do not know what the normal activation patterns
are in activities that use the gluteus maximus muscles, even
in normal subjects. The understanding of normal muscular
activation patterns presumably would provide a basis for
comparisons with individuals with low back pain or sacro-
iliac disorders, and perhaps do not reveal whether any
abnormalities that are subsequently found are the cause
or the effect.
Therefore, the aim of this study was to investigate acti-
vation patterns of the gluteus maximus, semitendinosus
and erector spinae muscles in healthy young individuals
during four different modalities of therapeutic exercise in
the prone position.
2. Methods and measures
2.1. Subjects
Thirty-one healthy young subjects of both genders were
recruited from the community and participated in the study.
Inclusion criteria were the absence of shortening of hip flexors,
determined by a negative Thomas test (Vogt and Banzer, 1997),
and complaints of pain, history of surgery in the lumbar spine or
hip, congenital alterations in the lower limbs or any other limi-
tations that could prevent data collection, such as difficulty in
performing the exercises. Individuals with recent episodes of ankle
sprain (grade II or III) were also not included (Bullock-Saxton
et al., 1994).
2.2. Instrumentation
The activation patterns of the gluteus maximus, semitendino-
sus and erector spinae muscles were assessed by an electromyo-
graph (MP150WSW, Biopac Systems, Inc.Ó, Santa Barbara, CA,
USA). This device had two amplifiers connected to a microcom-
puter, which have an input impedance of 2 MX and CMRR of
1000 MX and allows data acquisition at frequencies from 10 to
1000 Hz. Data were collected at a frequency of 1000 Hz. Surface
electrodes, active, bipolar, TSD 150, with diameters of 13.5 mm
and impedances of 100 MX were used for data collection.
The beginning of the movement was detected by the motion
capture system ProReflex MCU Qualisys (QUALISYS MEDI-
CAL AB, Gothenburg, Sweden), with capture rates of 120 Hz and
digital cameras MCU 120, equipped with a set of infra red light
senders that were reflected by spherical passive markers of 12 mm
of diameter, adhered to specific anatomic bony marks. Procedures
of linearization and calibration were performed according to
instructions in the manufacturer’s manual. Three cameras were
employed to capture the images for each exercise and were posi-
tioned in such a way that all markers were captured during all
investigated movements.
2.3. Procedures
Before the initiation of data collection, subjects were informed
about the objectives of the study and invited to sign a consent form,
which was previously approved by the University Ethical Review
Board. Demographic data were collected on all subjects to docu-
ment their age, as well as other clinically relevant information.
To obtain EMG data, subjects were instructed to lie in a prone
position and passive markers were placed over their iliac crest,
anterior superior iliac spine, posterior superior iliac spine, greater
trochanter, middle point of the thigh and the lateral epicondyle of
the femur of the evaluated lower limb. All markers were 12 mm in
diameter with the exception of the one placed over the anterior
superior iliac spine that was 5 mm to avoid discomfort during
exercises.
Surface electrodes (Ag/AgCl) were placed in pairs and parallel
to the muscle fibers (Cram et al., 1998). For the gluteus maximus,
the electrodes were placed at the midpoint of a line running from
the last sacral vertebrae to the greater trochanter; for semitendi-
nosus, medially on the mid-distance between gluteal fold and knee
joint; and for the erector spinae muscles, at the L3 level, bilater-
ally 2 cm lateral to the spinal processes and parallel to the lumbar
spine. The interelectrode spacing was 2 cm from center to center.
The reference electrode was placed over the lateral malleolus. Skin
preparation included shaving, rubbing and cleaning with alcohol.
A lamp was placed in front of the subjects and they were
instructed to start the required movement at their natural speed
when the light went off. Activation patterns of the gluteus maxi-
mus, semitendinosus and erector spinae muscles were analyzed
 A.C.L. Sakamoto et al. / Journal of Electromyography and Kinesiology 19 (2009) 105–112
during four modalities of therapeutic exercise in the prone posi-
tion often used in clinical practice. As illustrated in Fig. 1, the hip
extension exercises were performed in four positions: with knee
extension (KE), knee flexion (KF), lateral hip rotation and knee
extension (LHR–KE), lateral hip rotation and knee flexion
(LHR–KF). As demonstrated in Fig. 1b, a wooden device was
employed to position the subject’s leg and guarantee relaxation
during the performance of the KF exercise.
All exercises were randomly assigned and three trials were
obtained for each modality with a 2-min rest period between each
trial. The mean of the three trials for each exercise was used for
analysis. The beginning of the movement was determined by
changes in angular displacement of the rigid segment, obtained
from the motion capture system. A trigger mechanism was used to
synchronize the EMG and the motion capture system data, after
assuring EMG silence.
2.4. Data reduction
The motion capture system data processing was performed
using the software Qualysis Track Manager 1.6.0.X and later the
data were exported to the MATLABÒ for analysis. The signal
processing acquired for the motion capture system used the raw
values because the displacement was consistent and the noise did
not interfere with the measurements. Joint angles were calculated
only in the sagittal plane, using X and Z coordinates. Two straight
lines from the pelvic and lower limb segments were traced and the
prolongation of these lines provided information about the joint
angle.
EMG data processing was performed using the Acknowledge
software. The EMG signals were full wave rectified and low-pass
Fig. 1. Hip extension exercises: (a) knee extension (KE); (b) knee flexion (KF); (c) lateral hip rotation and knee extension (LHR–KE) and (d) lateral hip
rotation and knee flexion (LHR–KF).
107
and high-pass filtered with cut-off frequencies of 500 and 10 Hz,
respectively, and were recorded at the sampling rate of 1000 Hz.
Muscular activation patterns were described after determining
the EMG onset for each muscle. The onset of muscular activity
was considered to occur when the value exceeded two standard
deviations from the mean value observed at baseline for a 50 ms
period (Brindle et al., 1999; Hodges and Bui, 1996). The onset of
movement was calculated by a specific routine developed in
MATLABÒ and was recorded when angular velocity was positive
during performance of all exercises and when displacement
exceeded one degree and remained constant.
2.5. Data analysis
Descriptive statistics and tests for normality and homogeneity
of variance were calculated for all outcome variables, using the
software SPSS 13.0 for Windows (SPSS Inc.Ó, Chicago, IL).
Repeated measure ANOVAs followed by planned contrasts were
used to investigate differences in activation patterns for the four
modalities of exercises, with a significance level of a < 0.05.
3. Results
3.1. Subject characteristics
Thirty-one volunteers (16 men and 15 women) partici-
pated in the study, with the ages ranging from 20 to
36 years (24.5 ± 3.47), body masses from 46 to 90 kg
(66.89 ± 11.89 kg), heights from 1.50 to 1.84 m (1.70 ±
108 A.C.L. Sakamoto et al. / Journal of Electromyography and Kinesiology 19 (2009) 105–112
0.09), and body mass indices of 22.09 ± 2.22 and
23.75 ± 3.49 kg/m2, respectively, for the women and men.
Because of technical problems with MATLAB in read-
ing some motion analysis system files, data from three sub-
jects were excluded for analysis for the exercises performed
with KE and KF and from five others when exercises were
performed with LR–KE and LR–KF.
3.2. Activation patterns
For the exercises performed with KE, LR–KE and LR–
KF, the muscle activation patterns were initiated by semi-
tendinosus, followed by the contralateral erector spinae,
ipsilateral erector spinae muscles and finished by the glu-
teus maximus (Fig. 2). Fig. 3 illustrates a typical activation
pattern during the exercise with KE.
3.2.1. Knee extension
ANOVA demonstrated significant differences in laten-
cies for the semitendinosus ( 473.5 ± 257.6 ms) in relation
to the contralateral erector spinae ( 393.4 ± 291.0 ms;
p = 0.006), ipsilateral erector spinae ( 362.8 ± 278.8 ms;
p = 0.001), and gluteus maximus (22.0 ± 374.7; p =
0.0001), indicating that during this exercise, the movement
was initiated by the semitendinosus muscle in 68% of the
cases.
Significant differences were also found for the gluteus
maximus latency (p = 0.0001) in relation to the other mus-
cles, indicating that in 82% of the cases, the gluteus maxi-
mus was the last activated muscle and, in 50%, it was
activated after the initiation of the movement.
*
* LHR–KF (p = 0.001), indicating that the gluteus maximus
**
**
* KE
**
-900 -800 -700 -600 -500 -400 -300 -200 -100
Latency (ms)
Ipsilateral Spinae Contralateral Spinae
Semitendinosus Gluteus Maximus
Fig. 2. Muscular activation patterns for the four modalities of therapeutic
exercises.
3.2.2. Knee flexion
This exercise showed large variability in muscular acti-
vation, since the movement was initiated in 32% of the
cases by the semitendinosus ( 456.1 ± 156.3) or ipsilateral
erector spinae ( 422.0 ± 166.5 ms), and in 22% by the con-
tralateral erector spinae muscles ( 412.3 ± 232.6 ms). In
79% of the cases, the gluteus maximus was also the last
muscle to be activated and in only 21% of the time, its
onset was after the beginning of the movement. Significant
differences were found for the gluteus maximus latency
(p = 0.0001) in relation to the other muscles, indicating
that on average, the gluteus maximus was the last muscle
to be activated.
3.2.3. Lateral hip rotation and knee extension
The movement was initiated by the semitendinosus mus-
cle ( 541.6 ± 236.7) in 31% of the cases. This exercise also
showed great variability regarding muscular activation,
since in 39% of the cases, the movement was initiated by
the contralateral erector spinae ( 512.9 ± 224.9), in 15%
by the ipsilateral erector spinae ( 512.4 ± 241.1) and the
gluteus maximus ( 390.0 ± 259.7). In 54% of the cases,
the gluteus maximus was the last activated muscle and in
only 8%, its activation occurred after the beginning of the
movement. Significant differences were found for the glu-
teus maximus latency in relation to ipsilateral erector spi-
nae (p = 0.009), contralateral erector spinae (p = 0.002),
and semitendinosus muscles (p = 0.002), indicating that
during this modality of exercise, the gluteus maximus was
also the last activated muscle.
3.2.4. Lateral hip rotation and knee flexion
The movement was initiated by the semitendinosus mus-
cle in 46% of cases, by the contralateral erector spinae in
19%, by the ipsilateral erector spinae in 12%, and by the
**p<0,001 gluteus maximus in 23%. No significant differences were
*p<0,01
observed when the onset of the four investigated muscles
were compared, demonstrating that during this exercise,
HLR-KF the four muscles showed similar latencies.
When the gluteus maximus latency was analyzed sepa-
rately for each exercise, significant differences were found
HLR-KE
between the exercises performed with KE and those per-
formed with KF (p < 0.0001), LHR–KE (p < 0.0001), and
was first activated with LHR–KE, followed by LHR–KF,
KF KF and finally by KE (Fig. 4).
4. Discussion
The present study showed that the muscle firing order
was similar for the following modalities of exercises: KE,
0 100 LHR–KE and LHR–KF, beginning with the semitendino-
sus and followed by the contralateral erector spinae, ipsi-
lateral erector spinae, and finally by the gluteus maximus
muscles. For the KF modality, the onset sequence was
almost the same, however, the ipsilateral erector spinae
fired before the contralaterals. In fact, activation patterns
 A.C.L. Sakamoto et al. / Journal of Electromyography and Kinesiology 19 (2009) 105–112 109

Fig. 3. Activation patterns determined by the onset of EMG activity during the exercise of HE-KE. The dotted line represents the beginning of the
movement and the straight line the beginning of muscular activity.

determined by the EMG onset of each investigated mus-

cle, were consistent despite the individual variability
HLR-KF
observed during performance of the various modalities
of exercise.
HLR-KE

**
4.1. Knee extension
KF
For the exercise with KE, the firing order was initiated
KE
by semitendinosus, followed by contralateral erector spi-
nae, ipsilateral erector spinae muscles, and finally, by the
-700 -600 -500 -400 -300 -200 -100 0 100
gluteus maximus. The semitendinosus was the first muscle
Latency (ms)
to be activated and its onset time was almost simultaneous
**p<0.001
with the ipsilateral and contralateral erector spinae, with a
Fig. 4. Gluteus maximus latency (ms) for the four modalities of time span difference of only 34 ms, which was not signifi-
therapeutic exercises. cant. Lehman et al. (2004) also did not find significant
110 A.C.L. Sakamoto et al. / Journal of Electromyography and Kinesiology 19 (2009) 105–112
differences in hamstring latency compared to the lumbar
erector spinae muscle latency, indicating that they were
simultaneously activated (Lehman et al., 2004). Bullock-
Saxton et al. (1994) also found such consistent data, with
the beginning of hip extension usually initiated by the ham-
string. However, in contrast, they did not find significant
differences in the onset times between the four investigated
muscles. Based on these findings, Bullock-Saxton et al.
(1994) suggested that under normal circumstances, the
motor program is fixed and the activation occurs practi-
cally in a simultaneous way. Pierce and Lee (1990) also
observed activation of the biceps femoris before the begin-
ning of the hip extension movement in the prone position,
with this muscle being the first to be activated. However,
they found inconsistent activation patterns, although they
analyzed the movement from 30° of hip flexion to the neu-
tral position (Pierce and Lee, 1990).
On the other hand, Vogt and Banzer (1997) reported
consistent muscular activation patterns, but the firing
sequence was different from the one observed in the present
study, first being initiated by the ipsilateral erector spinae,
followed by semitendinosus, contralateral erector spinae
and gluteus maximus muscles. No significant differences
were found between the latencies of the erector spinae
and semitendinosus muscles, showing that they had almost
simultaneous activation. They justified the activation of the
lumbar erector spinae during the pre-movement phase as
preparation to stabilize the trunk to control the pelvis dur-
ing leg lifting (Vogt and Banzer, 1997). The initial semiten-
dinosus activation observed in the present study suggests
that with an appropriate length-tension relationship, this
muscle can be better recruited than the gluteus maximus.
In the present study, the mean onset time for the gluteus
maximus showed statistically significant differences when
compared to the onset of all other muscles. The gluteus
maximus was the last muscle to be activated during exer-
cises with KE in the majority of cases, and in 50% of cases,
its activation occurred after the beginning of the move-
ment. These findings also corroborated other studies that
analyzed the same movement and also demonstrated that
the gluteus maximus was the last activated muscle (Bull-
ock-Saxton et al., 1994; Lehman et al., 2004; Vogt and
Banzer, 1997).
Delays in onset of the gluteus maximus were observed in
subjects with ankle sprains (Bullock-Saxton et al., 1994):
reported value for onset with healthy subjects was at
249 ms, while with subjects with ankle sprains, was at
97 ms. The authors also found differences in the time
between the firing of the first (semitendinosus) and last
muscle (gluteus maximus) (Bullock-Saxton et al., 1994).
Lehman et al. (2004) demonstrated that the gluteus maxi-
mus was the last activated muscle during hip extension in
the prone position in healthy subjects who were carefully
assessed for detection of muscle skeletal disorders. How-
ever, five subjects exceeded the average time span that
was considered abnormal in the study of Bullock-Saxton
et al. (1994). In addition, the present results showed a mean
onset value of 13.63 ms for the gluteus maximus, with a
delay value that was also higher than the abnormal values
reported by Bullock-Saxton et al. (1994). In the present
study, the subjects were healthy, as was the case of Lehman
et al. (2004), which makes it difficult to determine the nor-
mal onset value limits.
Only one study was found that evaluated muscular firing
order in populations with any kind of injury (Bullock-Sax-
ton et al., 1994). They assessed subjects with ankle sprains
(grade II or III) and studies applying the same methodol-
ogy with other pathologies, such as low back pain, are cer-
tainly necessary, since studies have already shown altered
muscle patterns in other functional activities (Kankaanpaa
et al., 1998; Leinonen et al., 2000; Vogt et al., 2003).
It is difficult to compare the onset time values reported
between these studies because of methodological differ-
ences. In the present study, onset was considered to occur
when the value exceeded two standard deviations of the
mean value observed at baseline (Hodges and Bui,
1996). Studies that investigated the order of the muscle
firing for exercises performed with KE used different
methodologies for the determination of muscular onset,
such as 5% of the maximum amplitude (Bullock-Saxton
et al., 1994), 10% of the maximum rectified peak ampli-
tude (Lehman et al., 2004; Vogt and Banzer, 1997), and
even visually with the first increases in EMG activity
(Pierce and Lee, 1990). The method selected to determine
onset in the present study has been shown to be more reli-
able when compared to other methods (Hodges and Bui,
1996). They reported that the use of two standard devia-
tions avoids both type I and type II errors. Type I errors
can occur when one standard deviation is used and the
muscle may be identified as active when actually it is
not. Type II errors can occur when three standard devia-
tions are used and may indicate a failure to identify EMG
onset. According to these authors, the determination of
the onset through percentage of peak values is potentially
inaccurate due to the sensitivity of this technique, the
magnitude of the peak, and the rate of increase in ampli-
tude which are likely to vary between muscles (Hodges
and Bui, 1996).
Delays in the gluteus maximus activation are believed to
lead to lumbar dysfunctions (Hungerford et al., 2003;
Mooney et al., 2001; Vogt et al., 2003). The movement per-
formed with KE is also often used as a test of lumbar pelvic
function. This test is performed by palpation of the head of
the femur or gluteus maximus and hamstring muscles dur-
ing hip extension (Vogt and Banzer, 1997). Other authors
considered altered muscle activation patterns when ham-
string and erector spinae muscles are activated before the
gluteus maximus, whose activity is delayed (Jull and Janda,
1987; Vogt and Banzer, 1997). The present study and oth-
ers that investigated this movement were unanimous in
relation to the onset of the gluteus maximus as the last
muscle to be recruited and this information should be con-
sidered for the interpretation of the test proposed by Sahr-
mann (2002).
 A.C.L. Sakamoto et al. / Journal of Electromyography and Kinesiology 19 (2009) 105–112
4.2. Knee flexion
This exercise presented large variability in the order of
muscle firing. It was only during this exercise that the ipsi-
lateral erector spinae muscles were activated before the
contralaterals. However, this difference in onset was only
14 ms, showing that the activation of these muscles were
almost simultaneous. Only the gluteus maximus onset time
showed statistically significant differences when compared
to other investigated muscles and were the last to be acti-
vated. For this exercise modality, gluteus maximus activa-
tion occurred before the beginning of movement in a
majority of cases.
4.3. Lateral hip rotation and knee extension
This exercise also showed large variability in activation
patterns and no significant differences were found
between the onset time of the semitendinosus and bilat-
eral erector spinae. The gluteus maximus was again the
last activated muscle in the majority of cases and, as
was observed with the previous exercises, its onset was
significantly different when compared to other analyzed
muscles. However, when compared to the other exercises,
its activation had either the earliest or the lowest pre-
movement onset.
4.4. Lateral hip rotation and knee flexion
This exercise showed the highest variability in the order
of muscle firing and no significant differences were found
between onset times of all investigated muscles. This higher
variability may be due to the fact that this exercise starts
from an uncomfortable position. The exercises associated
with lateral rotation showed greater anticipated activation
of the gluteus maximus, probably because the subjects were
more careful in performing, since they had to extend the
hip and keep it laterally rotated at the same time.
4.5. Limitations
One could argue that the fact the speed of the move-
ments was not controlled in the present study. It is well
known that the magnitude of the EMG signal can be
directly influenced by several factors, such as speed, accel-
eration, range of movement, load and repetitions. How-
ever, although the speed was not controlled, the subjects
were instructed to perform the movements at their natural
speed in order to reproduce a situation similar to that
employed in clinical practice.
4.6. Clinical implications
The present findings demonstrated that independently of
the way the exercises were performed, the activation pat-
terns were similar. Studies involving subjects with low back
pain or sacroiliac dysfunctions are necessary and these
111
findings could be employed as reference values for
comparisons.
5. Conclusions
The present findings indicated that despite of the indi-
vidual variability, the order of muscle firing was similar
for all four exercise modalities and initiated by the semiten-
dinosus, then followed by lumbar erector spinae and glu-
teus maximus. The gluteus maximus was the last muscle
activated for the exercises performed with KE, KF, and
LHR–KE.
Acknowledgements
Brazilian Government Agencies (CNP and FAPEMIG).
References
Anderson FC, Pandy MG. Individual muscle contributions to support in
normal walking. Gait Posture 2003;17(2):159–69.
Brindle TJ, Nyland J, Shapiro R, Caborn DN, Stine R. Shoulder
proprioception: latent muscle reaction times. Med Sci Sports Exercise
1999;31(10):1394–8.
Bullock-Saxton JE, Janda V, Bullock MI. The influence of ankle sprain
injury on muscle activation during hip extension. Int J Sports Med
1994;15(6):330–4.
Cholewicki J, Van Dieen JH, Arsenault AB. Muscle function and
dysfunction in the spine. J Electromyogr Kinesiol 2003;13(4):303–4.
Clark BC, Manini TM, Mayer JM, Ploutz-Snyder LL, Graves JE.
Electromyographic activity of the lumbar and hip extensors during
dynamic trunk extension exercise. Arch Phys Med Rehabil
2002;83(11):1547–52.
Clark BC, Manini TM, Ploutz-Snyder LL. Derecruitment of the lumbar
musculature with fatiguing trunk extension exercise. Spine
2003;28(3):282–7.
Cram JR, Kasman GS, Holtz J. Introduction to surface electromyogra-
phy. 1st ed. Maryland: Aspen Publishers; 1998.
Hodges P, Bui BH. A comparison of computer-based methods for the
determination of onset of muscle contraction using electromyography.
Electromyogr Clin Neurophysiol 1996;101(6):511–9.
Hossain M, Nokes LD. A model of dynamic sacro-iliac joint instability
from malrecruitment of gluteus maximus and biceps femoris muscles
resulting in low back pain. Med Hypotheses 2005;65(2):278–81.
Hungerford B, Gilleard W, Hodges P. Evidence of altered lumbopelvic
muscle recruitment in the presence of sacroiliac joint pain. Spine
2003;28(14):1593–600.
Jull GA, Janda V. Muscles and motor control in low back pain:
assessment and management. In: Twomey LT, Taylor JR, editors.
Physical therapy of the low back. New York: Churchill Livingstone;
1987. p. 253–78.
Kankaanpaa M, Taimela S, Laaksonen D, Hanninen O, Airaksinen O.
Back and hip extensor fatigability in chronic low back pain patients
and controls. Arch Phys Med Rehabil 1998;79(4):412–7.
Lehman GJ, Lennon D, Tresidder B, Rayfield B, Poschar M. Muscle
recruitment patterns during the prone leg extension. BMC Musculo-
skeletal Disord 2004;5(3).
Leinonen V, Kankaanpaa M, Airaksinen O, Hanninen O. Back and hip
extensor activities during trunk flexion/extension: effects of low back
pain and rehabilitation. Arch Phys Med Rehabil 2000;81(1):32–7.
Mooney V. Coupled motion of contralateral latissimus dorsi and gluteus
maximus: its role in sacroiliac stabilization. In: Vleeming A, editor.
Movement, stability & low back pain: the essential role of the
pelvis. London: Harcourt Publishers; 1999. p. 115–22.
112 A.C.L. Sakamoto et al. / Journal of Electromyography and Kinesiology 19 (2009) 105–112
Mooney V, Pozos R, Vleeming A, Gulick J, Swenski D. Exercise
treatment for sacroiliac pain. Orthopedics 2001;24(1):29–32.
Norris CM. Spinal stabilization: muscle imbalance and the low back.
Physiotherapy 1995;81(3):127–37.
Panjabi MM. The stabilizing system of the spine. Part II. Neutral zone
and instability hypothesis. J Spinal Disord 1992;5(4):390–6.
Panjabi MM. Clinical spinal instability and low back pain. J Electromyogr
Kinesiol 2003;13(4):371–9.
Pierce MN, Lee WA. Muscle firing order during active prone hip
extension. J Orthop Sports Phys Ther 1990;12(1):2–9.
Sahrmann SA. Diagnosis and treatment of movement impairment
syndromes. 1st ed. St Louis: Mosby; 2002.
Vogt L, Banzer W. Dynamic testing of the motor stereotype in prone hip
extension from neutral position. Clin Biomech 1997;12(2):122–7.
Vogt L, Pfeifer K, Banzer W. Neuromuscular control of walking with
chronic low-back pain. Manual Ther 2003;8(1):21–8.
Ana Cristina Lamounier Sakamoto received her
B.Sc. in physical therapy from the Universid-
ade Federal de Minas Gerais in Belo Hori-
zonte, Brazil in 2000 and her M.Sc. in
rehabilitation sciences at the same university.
Currently, she is an assistant professor at the
Centro Universitário de Belo Horizonte, Bra-
zil. She has been involved in research activities
focusing on low back pain, disability and
functional movements.
Luci Fuscaldi Teixeira-Salmela is an associate
professor in the department of physical therapy
of the Universidade Federal de Minas Gerais in
Belo Horizonte, Brazil. She received her B.Sc.
in physical therapy in Brazil, an M.Sc. in
Rehabilitation Sciences and a Ph.D. in Anat-
omy at Queen’s University, Canada. Her areas
of interest include kinesiological biomechanics,
gait and task analysis and the understanding of
factors which limit functional performance.
Currently, she is the director of the graduate
program in Rehabilitation Sciences at the Universidade Federal de Minas
Gerais.
Fátima Rodrigues de Paula-Goulart is an asso-
ciate professor in the department of physical
therapy of the Universidade Federal de Minas
Gerais in Belo Horizonte, Brazil. She received
her B.Sc. in physical therapy in Brazil and a
doctorate in neurosciences from the Universi-
dad Autonoma de Barcelona. She has been
involved in research activities focusing on
motor control, motor-neuronal excitability, sit
to stand, postural adjustments, electromyog-
raphy and transcranial magnetic stimulation.
Christina Danielli Coelho de Morais Faria
received her B.Sc. in physical therapy from the
Universidade Federal de Minas Gerais in Belo
Horizonte, Brazil in 2005. She then registered
for the M.Sc. program in rehabilitation sci-
ences at the same university in 2006. Presently,
she is a Ph.D. student in rehabilitation sciences
at the Universidade Federal de Minas Gerais in
collaboration with the University of Montreal.
Her research interests include biomechanics
and functional activities.
Cristiano Queiroz Guimarães received his B.Sc.
in physical therapy with a specialization in
orthopedics and physical therapy in sport from
the Universidade Federal de Minas Gerais in
Belo Horizonte, Brazil. Currently, he is pres-
ently a masters student in rehabilitation sci-
ences at the same university. He has been
involved in research activities focusing on low
back pain, disability and functional
movements.