G Model
JSAMS-1239; No. of Pages 6
ARTICLE IN PRESS
Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Science and Medicine in Sport
journal homepage: www.elsevier.com/locate/jsams
Original research
Does load influence shoulder muscle recruitment patterns during
scapular plane abduction?
Darren Reed a,∗ , Ian Cathers a , Mark Halaki b , Karen A. Ginn a
a
b
Discipline of Biomedical Science, School of Medical Sciences, Sydney Medical School, The University of Sydney, Australia
Discipline of Exercise and Sport Science, Faculty of Health Sciences, The University of Sydney, Australia
a r t i c l e
i n f o
Article history:
Received 27 July 2015
Received in revised form 12 October 2015
Accepted 29 October 2015
Available online xxx
Keywords:
Shoulder
Abduction
Muscle activation
Electromyography
a b s t r a c t
Objectives: Load is used to increasingly challenge muscle function and has been shown to increase muscle activity levels with no change in activation patterns during shoulder flexion, extension, adduction
and rotation. However, the effect of load during shoulder abduction, a movement commonly used in
assessment of shoulder dysfunction and to improve shoulder function, has not been comprehensively
examined. Therefore, the purpose of this study was to determine if load influences shoulder muscle
activation patterns and levels during scapular plane abduction in normal subjects.
Design: Experimental study.
Methods: Fourteen volunteers performed shoulder abduction in the scapular plane at 25%, 50% and 75%
of maximum load. Eight shoulder muscles were investigated using a combination of indwelling and surface electromyographic recordings: middle deltoid, infraspinatus, subscapularis, supraspinatus, serratus
anterior, upper and lower trapezius and rhomboid major.
Results: All muscles tested showed increasing average muscle activation levels with increasing load and
strong correlations in the activation patterns between loads.
Conclusions: Increasing shoulder abduction load not only increases activity in middle deltoid but also in
the rotator cuff (infraspinatus, subscapularis, supraspinatus) and axioscapular (serratus anterior, upper
and lower trapezius, rhomboid major) muscles. The functional stabilising role of both the rotator cuff and
axioscapular muscles is considered an important contribution to the increased activation levels in these
muscle groups as they function to counterbalance potential translation forces produced by other muscles
during shoulder abduction. The activation patterns of all shoulder muscle groups during abduction can
be trained at low load and progressively challenged with increasing load.
© 2015 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
1. Introduction
Increasing load is commonly used in shoulder rehabilitation or
gym based exercise programs to progressively challenge muscles
with the aim of improving strength and muscle function.1,2 Electromyography (EMG) has been used to confirm muscle activation
levels and patterns during rehabilitation exercises and it has been
suggested that exercises producing low muscle activation levels
be used early in rehabilitation programs3 while higher activation
levels are described as more demanding and better suited to later
in rehabilitation programs.4,5 Previous EMG studies have shown
that normal shoulder muscle activation levels increase not only
in the torque producing muscles but also in the rotator cuff and
axioscapular muscles with increased load during dynamic shoulder
∗ Corresponding author.
E-mail address: darren.reed@sydney.edu.au (D. Reed).
flexion,6 extension,6 rotation7,8 and isometric adduction.9 Furthermore, in these studies shoulder muscle activation patterns did
not change with increasing load with the same shoulder muscles
recruited at low load also recruited at medium and high load. This
implies that the normal recruitment pattern of muscles involved in
moving and stabilising the humerus and scapula during shoulder
movements can be established at low load and as load increases,
all increase their activity.
Scapular plane abduction has been described as the most
functional plane of shoulder abduction2,10 and is a commonly recommended exercise used in rehabilitation and exercise programs
to improve function at the shoulder.4,5 Although shoulder muscle
recruitment patterns and levels have been investigated in previous studies during scapular plane abduction at low load1,4,11 and
high load12 significant methodological considerations such as differences in the EMG normalisation process, makes comparison
between these studies at different loads invalid. Two previous EMG
studies have concurrently investigated the effects of increasing
http://dx.doi.org/10.1016/j.jsams.2015.10.007
1440-2440/© 2015 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Reed D, et al. Does load influence shoulder muscle recruitment patterns during scapular plane
abduction? J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.10.007
G Model
JSAMS-1239; No. of Pages 6
ARTICLE IN PRESS
D. Reed et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
2
load during scapular plane abduction; however, direct comparison
between muscle activation at different loads is still problematic.2,13
One of these studies only investigated inner range abduction (0◦ to
90◦ )13 and neither study included load as a factor in the statistical
analysis so that significant differences between muscle activation levels at different loads cannot be confirmed.2,13 Additionally
while middle deltoid and rotator cuff muscles were investigated in
both studies, the only axioscapular muscle investigated was upper
trapezius and only in the 0–90◦ abduction range.13
As no single EMG study has comprehensively investigated the
effect of load on normal shoulder muscle activation levels and
patterns during scapular plane abduction, it is still not known if
increasing load during abduction systematically increases activation levels with no change in activation patterns in all shoulder
muscle groups. Therefore, the aim of this experiment was to quantify and compare activation levels and recruitment patterns of
middle deltoid (abduction torque producing muscle), supraspinatus, infraspinatus and subscapularis (rotator cuff muscles) and
serratus anterior, upper and lower trapezius and rhomboid major
(axioscapular muscles) in normal subjects through full range scapular plane abduction at low, medium and high loads.
2. Material and methods
The shoulder of the arm used for writing of fourteen asymptomatic subjects (5 female and 9 male) aged between 18 and
49 years (mean age 22.5 years) was investigated in this study.
Participants had no history of shoulder pain, no pain on maximally resisted rotation tests and demonstrated full normal range
of shoulder abduction (180◦ ) with normal scapulohumeral rhythm
(assessed visually by an experienced physiotherapist). The study
was approved by the Human Research Ethics Committee of the University of Sydney and all subjects gave written informed consent
before participating.
A power analysis using G power software14 was performed
to calculate the required sample size. Mean activity levels
recorded from middle deltoid (38 ± 11%MVC) and supraspinatus
(39 ± 11%MVC) in a previous EMG study investigating shoulder
abduction11 were chosen for this analysis as the muscles most
responsible for producing shoulder abduction torque. With power
set at 0.80, significance level ˛ = 0.05, a mean detectable difference of 6%MVC and a correlation value of 0.80, a sample size of
13 (middle deltoid data) and 14 (supraspinatus data) subjects was
required. Consequently a sample size of 14 subjects was chosen for
this current study.
Following subject familiarisation with the exercise protocol,
pairs of silver/silver chloride surface electrodes (Red Dot, 2258,
3M) were placed over middle deltoid and upper trapezius15 at
an inter electrode distance of 20 mm and impedance <5 k. Bipolar intramuscular electrodes16 were used to record activity from
the remaining muscles i.e. deep muscles inaccessible to surface
electrodes (supraspinatus, subscapularis, rhomboid major), muscles where crosstalk (infraspinatus17 ) and geometric displacement
(serratus anterior18 ) have been shown to significantly affect surface electrode recordings and thin muscles (lower trapezius) where
underlying muscles could produce crosstalk. The intramuscular
electrode insertion sites were prepared with an anaesthetic gel
(Xylocaine 2% jelly, AstraZeneca Pty Ltd, NSW, Australia), an antiseptic solution (Betadine, Faulding Healthcare Pty Ltd, Virginia,
Australia) and cleaned with alcohol. A sterile insertion technique
using a 23 ga hypodermic needle as a cannula was used to place
indwelling electrodes. The placement of electrodes in all muscles,
except for subscapularis, were according to the recommendations
of Geiringer.19 A single bipolar electrode was placed in the midsection of subscapularis using the method described by Kadaba
et al.20 Due to the thin nature of lower trapezius and close proximity of the underlying rhomboid major, insertion of the electrodes
into these muscles was guided and confirmed by a digital ultrasonic
diagnostic imaging system (Mindray, DP-9900). Correct placement
of the other intramuscular electrodes was confirmed by visual
inspection of the raw EMG signals during submaximal contractions expected to produce moderate activity in the target muscle
with low activity in adjacent muscles.15 A surface ground electrode
(Universal Electrosurgical Pad, 9160F, 3M) was placed over the
spine of the scapula on the contralateral shoulder. The EMG signals
were amplified and filtered (Iso-DAM 8 amplifiers, World Precision
Instruments, gain = 100 or 1000 depending on the signal voltage
as to not saturate the amplifier and provide good digital resolution, bandpass between 10 Hz and 1 kHz, Common Mode Rejection
Ratio:100 dB at 50 Hz) before transferring to a personal computer
with a 16 bit analog to digital converter (1401, Cambridge Electronics Design) at a sampling rate of 2564 Hz using Spike2 software
(version 4.00, Cambridge Electronics Design).
Before application/insertion of the electrodes subjects were
trained to abduct the arm from the side through full range of
abduction in the scapular plane leading with the thumb. Feet were
comfortably spaced apart and the opposite hand rested on the
adjacent hip to limit compensatory trunk movements. Speed was
standardised to a count of 3 s in the concentric phase, a second at
full range abduction and 3 s in the eccentric phase of motion. Tape
markings on the floor were a guide to indicate the scapular plane
of movement and an examiner standing on this line provided an
additional visual target for the subject. Using dumbbell weights,
the maximum load (100% load) able to be lifted in one repetition with normal scapulohumeral rhythm and no compensatory
trunk movement was determined for each subject. A minimum
rest interval of 30 s was given between attempts and a maximum
of five attempts were required to determine the 100% load in this
cohort.
Following electrode placement and signal verification, resting
EMG activity was recorded for each muscle. Five standardised maximum isometric shoulder normalisation tests15,21 were performed
in sitting, three times each in random order with a 30 s break
between repetitions and 2 min between tests. These included:
resisted abduction at 90◦ abduction with humerus externally
rotated; resisted extension in 30◦ abduction; resisted horizontal
adduction at 90◦ flexion; resisted flexion in 125◦ flexion; resisted
internal rotation at 90◦ abduction. These tests have a high likelihood of generating maximum activity in the shoulder muscles
tested.15,21
Abduction was performed in the scapular plane while holding
a dumbbell weight corresponding to 25%, 50% and 75% of maximum abduction load. Two full repetitions of shoulder abduction
were completed at each load. Different load conditions were performed in random order with a 30 s break between each load
condition. Synchronisation of the EMG signal with movement was
achieved by a draw-wire sensor (Micro-Epsilon, WPS-1000-MK46P10, Germany), attached to the wrist of the arm being tested.6 If
compensatory trunk or scapular movements occurred, or timing
varied, the test was repeated.
EMG signals were high pass filtered (10 Hz, zero lag 8th order
Butterworth), rectified then low pass filtered (5 Hz, zero lag 8th
order Butterworth) using Matlab 7.1 (The Mathworks). The maximum value obtained during the normalisation tests for each muscle
was used to normalise EMG recordings during shoulder abduction.
The signals were time normalised over the period of the combined
concentric and eccentric phases of movement (0–100% representing the duration of movement). Group mean data (±SD) for each
muscle was generated from the averages of the two trials at each
load. Three out of 672 signals (14 subjects × 3 loads × 8 muscles × 2
trials) or less than 1% of data was eliminated due to electrode
Please cite this article in press as: Reed D, et al. Does load influence shoulder muscle recruitment patterns during scapular plane
abduction? J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.10.007
G Model
ARTICLE IN PRESS
JSAMS-1239; No. of Pages 6
D. Reed et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
80
25% load
50% load
75% load
‡
‡
‡
EMG (% MVC)
70
‡
‡
‡
60
50
3
*
*
*
*
*
‡
‡
40
30
20
10
0
middle
deltoid
supraspinatus infraspinatus subscapularis
upper
trapezius
serr atus
anterior
lower
trapezius
rhomboid
major
MUSCLES
Fig. 1. Average (±SD) normalised EMG activation levels (%MVC) of the eight muscles tested during scapular plane shoulder abduction at 25%, 50% and 75% loads. ‡ Indicates
significantly greater activation levels at 75% load than 25% load. * Indicates significantly greater activation levels at 50% load than 25% load.
failure in subscapularis in one subject, serratus anterior in another
and rhomboid major in another.
A two factor repeated measures ANOVA (Statistica, version 7.1,
Statsoft) was performed to compare the activation levels of the
eight muscles examined across the three loads (25%, 50% and 75%
of maximum load). Tukey post hoc analysis was used to identify
specific differences when significant ANOVA results were obtained.
Statistical significance was set at p < 0.05.
Pearson’s correlation analysis was conducted between the time
normalised group mean EMG signals from each muscle at different
loads to determine the consistency of muscle activation patterns
between 25/50%, 50/75% and 25/75% loads and between muscles. Correlation strengths were defined as “weak” for 0.1 < r < 0.3,
“medium” 0.3 < r < 0.5 and “strong” 0.5 < r < 1.0.22
3. Results
The average normalised muscle activation levels of the eight
shoulder muscles examined during abduction performed in the
scapular plane at 25%, 50% and 75% maximum load are shown
in Fig. 1. Time normalised and averaged EMG muscle activation
patterns during the concentric and eccentric phases of full range
shoulder abduction in the scapular plane at 25%, 50% and 75% maximum load are illustrated in Fig. 2. The lowest correlation coefficient
value between 25% and 50%, 50% and 75%, and 25% and 75% loads
for each muscle is reported, with a strong correlation (r ≥ 0.65,
p < 0.05) indicating a similar pattern of activation between loads
for all muscles tested.
There were significant ANOVA main effects for load (F2,26 = 87.4,
p < 0.01), muscles (F7,91 = 4.2, p < 0.01) and significant interaction
between load and muscles (F14,182 = 3.6, p < 0.01). Tukey post hoc
testing revealed that significantly greater muscle activity occurred
at 50% than at 25% load and at 75% than at 50% load (p < 0.01)
and that there was a significant increase in activity levels with
increasing load for all muscles (p < 0.01). To illustrate the interaction effects, comparison of average activation levels between
muscles at the low, medium and high loads are shown in Fig. 3.
Muscles are ordered by decreasing level of activity during shoulder
abduction and grouped by boxes indicating no significant differences in activity between muscles within a box and with associated
p-values from Tukey post hoc tests. There were significantly different activity levels between muscles in different boxes (p < 0.05).
Correlation analysis between muscle activation patterns indicated:
• the pattern of activation of middle deltoid was similar to
infraspinatus and subscapularis across all loads with a strong correlation (r ≥ 0.67, p < 0.05) and to supraspinatus with a medium
correlation (0.47 ≥ r ≥ 0.42, p < 0.05). Middle deltoid activity pattern had a strong correlation to the pattern of all axioscapular
muscles investigated (r ≥ 0.59, p < 0.05).
• the pattern of activation of all rotator cuff muscles was similar
with a strong correlation (r ≥ 0.75, p < 0.05). Infraspinatus and
subscapularis activity patterns had a strong correlation with the
pattern of all axioscapular muscles investigated (r ≥ 0.57, p < 0.05)
and supraspinatus had a medium or strong correlation with the
axioscapular muscles (r ≥ 0.45, p < 0.05).
• the pattern of activation of all axioscapular muscles investigated
was similar with a strong correlation (r ≥ 0.67, p < 0.05).
4. Discussion
The results of this study examining shoulder abduction in the
scapular plane indicate that as load increases, average activation levels also increase systematically in all muscles tested. This
includes the torque producing muscle (middle deltoid), rotator
cuff muscles (supraspinatus, infraspinatus and subscapularis) and
axioscapular muscles (upper trapezius, lower trapezius, serratus
anterior and rhomboid major). The systematic increase in all muscles through full range abduction found in this current study would
suggest that shoulder muscle recruitment during abduction, does
not follow the ‘law of minimal activation’ which states that ‘the
muscles with least synergistic activity will be recruited first and
then as load increases other muscles are recruited’.23 Rather these
results suggest a ‘law of proportional activation’, where the complex coordinated muscle activation pattern required to produce
shoulder abduction in the scapular plane is established at low load
and maintained as load increases. This would imply that normal
muscle activation patterns during abduction can be trained at low
load and then progressively challenged by adding resistance to the
exercise through part or full range abduction. This is consistent
with previous shoulder EMG research investigating normal muscle activation during shoulder adduction,9 flexion,6 extension,6 and
Please cite this article in press as: Reed D, et al. Does load influence shoulder muscle recruitment patterns during scapular plane
abduction? J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.10.007
G Model
JSAMS-1239; No. of Pages 6
4
ARTICLE IN PRESS
D. Reed et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
Fig. 2. Time normalised and averaged EMG muscle activation patterns of the eight muscles tested during shoulder abduction in the scapular plane at 25%, 50% and 75% loads.
Time of 0% cycle represents start of movement and 100% cycle represents return to the starting position. Draw wire sensor trace, time normalised to maximum voltage is
included as a final graph. r = lowest correlation value between 25/50%, 50/75% and 25/75% loads for each muscle.
rotation7,8 which has shown systematic increases in middle deltoid,
rotator cuff and axioscapular muscle activity with increasing load.
These results also support the conclusions of previous research
which have reported increasing muscle activity trends in the rotator cuff and middle deltoid with increasing load during scapular
plane abduction from 0 to 90◦ .2,13 However, in the range greater
than 90◦ abduction, the systematic increases in activation levels in
the current study do not concur with the only previous study investigating the effect of load during abduction in this range, where
inconsistent results were reported.2
The results of the current study demonstrate that as load
increases during shoulder abduction the main abduction torque
producing muscle (middle deltoid) increases its activation levels to produce movement of the arm. As middle deltoid activity
increases during abduction, the superior translatory forces on the
humerus24,25 would also increase, resulting in greater potential
for impingement of subacromial structures between the humeral
head and the coracoacromial arch. A subsequent increase in activity
in parts of the rotator cuff able to prevent this superior translation (e.g. mid to lower sections of infraspinatus and subscapularis)
would be expected.24,26 The results from this current study provide
strong evidence to support this interaction between middle deltoid and both infraspinatus and subscapularis. Activation levels
in infraspinatus and subscapularis increased systematically with
increasing middle deltoid activity. In addition, there was a strong
correlation between middle deltoid and both infraspinatus and
subscapularis at all loads, indicating a similar pattern of activation
between these muscles. Activation patterns in infraspinatus and
subscapularis were also strongly correlated with that recorded in
supraspinatus supporting the belief that, as part of the rotator cuff,
supraspinatus is contributing to dynamic stability at the shoulder
by providing a medial stabilising force on the humeral head during
abduction.24,26
Although activity in infraspinatus and subscapularis was
strongly correlated and highly likely to be activating with a common stabilising function, infraspinatus was activated at higher
levels than subscapularis during abduction at all loads tested
(Fig. 3). This may reflect the additional role of infraspinatus to
externally rotate the humerus to prevent the greater tubercle
from impacting on the acromion during the abduction movement.
Similarly the greater activation levels in supraspinatus, compared
with that recorded from subscapularis, in this study may reflect
its additional role of producing abduction torque (Fig. 3). Biomechanical studies have suggested that supraspinatus has a more
favourable abduction moment arm than middle deltoid in the early
stages of shoulder abduction24 and therefore, is a more effective abductor than middle deltoid in this range of movement.27
The results of the current study support the above hypothesis,
Please cite this article in press as: Reed D, et al. Does load influence shoulder muscle recruitment patterns during scapular plane
abduction? J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.10.007
G Model
JSAMS-1239; No. of Pages 6
ARTICLE IN PRESS
D. Reed et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
5
Fig. 3. Muscles ordered by level of activity during scapular plane abduction at 25%, 50% and 75% load and grouped by boxes indicating no significant differences between
muscles within a box and with associated p-values from Tukey post hoc tests. Significant differences are indicated for muscles in different boxes (p < 0.05).
although it has been clearly shown that supraspinatus does not
initiate abduction.28 Supraspinatus activation levels were twice
that of middle deltoid at the start of the abduction movement
(Fig. 2), peak activation levels in supraspinatus occurred earlier
in range than that of middle deltoid (Fig. 2) and there was only
a moderate correlation between the activity patterns in these two
muscles.
Full range abduction at the shoulder depends not only on
humeral elevation but also on the synchronised upward rotation
of the scapula,25 with the axioscapular muscles (serratus anterior,
upper trapezius and lower trapezius) acting as the torque producing muscles for this rotation.10,25 However, as increased load on
the arm does not directly increase load on the scapula, the torque
producing role of the axioscapular muscles does not fully explain
their increase in activation levels during abduction with load. The
axioscapular muscles must also stabilise the scapula to provide a
fixed attachment site for the scapulohumeral muscles (middle deltoid and the rotator cuff) to enable them to produce glenohumeral
joint torque and stability, rather than their origin and insertion
being reversed and potentially producing lateral movement of the
scapula.6,27,29 A strong correlation between the activation patterns of the axioscapular and scapulohumeral muscles suggests
that as middle deltoid, infraspinatus and subscapularis increase
their activation levels with increased load, they exert an increased
lateral pull on the scapula necessitating a corresponding increase
in activation of axioscapular muscles to counterbalance this pull.
This argument is further strengthened by the finding that, similar to other axioscapular muscles, rhomboid major was activated
at moderate, increasing average levels as abduction load increased
(20–44%MVC). As rhomboid major is a downward rotator of the
scapula30 it cannot be functioning to upwardly rotate the scapula
during the shoulder abduction task under investigation. The strong
correlation between the pattern of activation in rhomboid major
and all other axioscapular muscles and scapulohumeral muscles
tested in this current study indicates that it is providing a stabilising
force to prevent lateral movement of the scapula.
5. Conclusion
The results of this study show that in asymptomatic subjects,
increasing load during scapular plane abduction increases activation levels in the main abduction torque producing muscle
(middle deltoid) and consequently in the rotator cuff muscles (infraspinatus, subscapularis, supraspinatus) and axioscapular
muscles (serratus anterior, upper and lower trapezius and rhomboid major). This increase in activation levels is systematic and
indicates that the pattern of individual muscle activation does not
change as load increases. The shoulder muscle recruitment pattern
at low loads is similar to that at high loads.
Practical implications
• Training with increasing abduction load will result in strength
increases in rotator cuff and axioscapular muscles as well as deltoid with potential fatigue implications in any of these muscles
at higher loads.
• The strength increases in the rotator cuff and axioscapular
muscles which occur with increasing abduction load are predominantly as result of their role as dynamic stabilisers:
o increased deltoid activity with greater abduction load will
result in increased translational forces on the humeral head
requiring greater rotator cuff activity as they function to counterbalance these destabilising forces,
o increased deltoid and rotator cuff activity will result in
increased translation forces on the scapula requiring increased
Please cite this article in press as: Reed D, et al. Does load influence shoulder muscle recruitment patterns during scapular plane
abduction? J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.10.007
G Model
JSAMS-1239; No. of Pages 6
ARTICLE IN PRESS
D. Reed et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
6
axioscapular muscle activity to counterbalance these destabilising forces.
• The activation patterns of all shoulder muscle groups during
abduction can be trained at low load and progressively challenged
with increasing load.
Institutional review board
Ethics approval was granted by The University of Sydney Human
Ethics Committee with approval number 07-2007/10/10122.
Acknowledgement
No external funding was received for this study.
References
1. McCann PD, Wootten ME, Kadaba MP et al. A kinematic and electromyographic
study of shoulder rehabilitation exercises. Clin Orthop 1993;(288):179–188.
2. Alpert SW, Pink MM, Jobe FW et al. Electromyographic analysis of deltoid and
rotator cuff function under varying loads and speeds. J Shoulder Elbow Surg 2000;
9(1):47–58.
3. Kelly BT, Kadrmas WR, Kirkendall DT et al. Optimal normalization tests for
shoulder muscle activation: an electromyographic study. J Orthop Res 1996;
14(4):647–653.
4. Townsend H, Jobe FW, Pink M et al. Electromyographic analysis of the glenohumeral muscles during a baseball rehabilitation program. Am J Sports Med 1991;
19(3):264–272.
5. Moseley Jr JB, Jobe FW, Pink M et al. EMG analysis of the scapular muscles during
a shoulder rehabilitation program. Am J Sports Med 1992; 20(2):128–134.
6. Wattanaprakornkul D, Cathers I, Halaki M et al. The rotator cuff muscles have
a direction specific recruitment pattern during shoulder flexion and extension
exercises. J Sci Med Sport 2011; 14(5):376–382.
7. Dark A, Ginn KA, Halaki M. Shoulder muscle recruitment patterns during commonly used rotator cuff exercises: an electromyographic study. Phys Ther 2007;
87(8):1039–1046.
8. Boettcher CE, Cathers I, Ginn KA. The role of shoulder muscles is task specific. J
Sci Med Sport 2010; 13(6):651–656.
9. Reed D, Halaki M, Ginn K. The rotator cuff muscles are activated at low
levels during shoulder adduction: an experimental study. J Physiother 2010;
56(4):259–264.
10. Bagg SD, Forrest WJ. Electromyographic study of the scapular rotators during
arm abduction in the scapular plane. Am J Phys Med 1986; 65(3):111–124.
11. Wickham J, Pizzari T, Stansfeld K et al. Quantifying ‘normal’ shoulder muscle
activity during abduction. J Electromyogr Kinesiol 2010; 20(2):212–222.
12. Ekstrom RA, Donatelli RA, Soderberg GL. Surface electromyographic analysis of
exercises for the trapezius and serratus anterior muscles. J Orthop Sports Phys
Ther 2003; 33(5):247–258.
13. Yasojima T, Kizuka T, Noguchi H et al. Differences in EMG activity in scapular
plane abduction under variable arm positions and loading conditions. Med Sci
Sports Exerc 2008; 40(4):716–721.
14. Faul F, Erdfelder E, Lang A et al. G*Power 3: a flexible statistical power analysis
program for the social, behavioral, and biomedical sciences. Behav Res Methods
2007; 39:175–191.
15. Boettcher CE, Ginn KA, Cathers I. Standard maximum isometric voluntary
contraction tests for normalizing shoulder muscle EMG. J Orthop Res 2008;
26(12):1591–1597.
16. Basmajian JV, DeLuca CJ. Muscles alive: their functions revealed by electromyography, Sydney, Williams and Wilkins, 1985.
17. Johnson VL, Halaki M, Ginn KA. The use of surface electrodes to record infraspinatus activity is not valid at low infraspinatus activation levels. J Electromyogr
Kinesiol 2011; 21(1):112–118.
18. Hackett L, Reed D, Halaki M et al. Assessing the validity of surface electromyography for recording muscle activation patterns from serratus anterior.
J Electromyogr Kinesiol 2014; 24(2):221–227.
19. Geiringer S. Anatomic localisation for needle electromyography, Sydney, Hanley
and Belfus Inc., 1994.
20. Kadaba MP, Cole A, Wootten ME et al. Intramuscular wire electromyography of
the subscapularis. J Orthop Res 1992; 10(3):394–397.
21. Ginn KA, Halaki M, Cathers I. Revision of the Shoulder Normalization Tests
is required to include rhomboid major and teres major. J Orthop Res 2011;
29(12):1846–1849.
22. Cohen J. Statistical power analysis for the behavioral sciences, New York, NY, Academic Press, 1988.
23. MacConaill MA, Basmajian JV. Muscles and movements, a basis for human kinesiology, New York, NY, Robert E. Krieger Publishing Co., 1977.
24. Poppen NK, Walker PS. Forces at the glenohumeral joint in abduction. Clin Orthop
1978; 135:165–170.
25. Inman VT, Saunders JB, Abbott LC. Observations of the function of the shoulder
joint. J Bone Joint Surg 1944; 26:1–30.
26. Sharkey N, Marder R. The rotator cuff opposes superior translation of the
humeral head. Am J Sports Med 1995; 23(3):1–11.
27. Reinold MM, Escamilla RF, Wilk KE. Current concepts in the scientific and clinical
rationale behind exercises for glenohumeral and scapulothoracic musculature.
J Orthop Sports Phys Ther 2009; 39(2):105–117.
28. Reed D, Cathers I, Halaki M et al. Does supraspinatus initiate shoulder abduction?
J Electromyogr Kinesiol 2013; 23(2):425–429.
29. Kibler WB. The role of the scapula in athletic shoulder function. Am J Sports Med
1998; 26(2):325–337.
30. Palastanga N, Field D, Soames R. Anatomy and human movement—structure and
function, Edinburgh, Butterworth-Heinemann Elsevier, 2012.
Please cite this article in press as: Reed D, et al. Does load influence shoulder muscle recruitment patterns during scapular plane
abduction? J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.10.007