Content uploaded by Michael Decker
Author content
All content in this area was uploaded by Michael Decker on Apr 24, 2015
Content may be subject to copyright.
Serratus Anterior Muscle Activity During
Selected Rehabilitation Exercises*
Michael J. Decker,† MS, Robert A. Hintermeister, PhD, Kenneth J. Faber, MD, and
Richard J. Hawkins, MD
From the Steadman Hawkins, Sports Medicine Foundation, Vail, Colorado
ABSTRACT
The purpose of this study was to document the elec-
tromyographic activity and applied resistance associ-
ated with eight scapulohumeral exercises performed
below shoulder height. We used this information to
design a continuum of serratus anterior muscle exer-
cises for progressive rehabilitation or training. Five
muscles in 20 healthy subjects were studied with sur-
face electrodes for the following exercises: shoulder
extension, forward punch, serratus anterior punch, dy-
namic hug, scaption (with external rotation), press-up,
push-up plus, and knee push-up plus. Electromyo-
graphic data were collected from the middle serratus
anterior, upper and middle trapezius, and anterior and
posterior deltoid muscles. Each exercise was parti-
tioned into phases of increasing and decreasing force
and analyzed for average and peak electromyographic
amplitude. Resistance was provided by body weight,
an elastic cord, or dumbbells. The serratus anterior
punch, scaption, dynamic hug, knee push-up plus, and
push-up plus exercises consistently elicited serratus
anterior muscle activity greater than 20% maximal vol-
untary contraction. The exercises that maintained an
upwardly rotated scapula while accentuating scapular
protraction, such as the push-up plus and the newly
designed dynamic hug, elicited the greatest electro-
myographic activity from the serratus anterior muscle.
Normal shoulder motion results from a complex interplay
of the scapulohumeral, acromioclavicular, sternoclavicu-
lar, and scapulothoracic articulations. The coordination of
these articulations provides the shoulder with an ample
range of motion necessary for overhead sporting activities.
Proper positioning of the humerus in the glenoid cavity,
known as scapulohumeral rhythm,
6
is critical to the
proper function of the glenohumeral joint during overhead
motion. A disturbance in normal scapulohumeral rhythm
may cause inappropriate positioning of the glenoid rela-
tive to the humeral head, resulting in injury.
16,18,22
One of the primary muscles responsible for maintaining
normal rhythm and shoulder motion is the serratus ante-
rior.
8,32
Lack of strength or endurance in this muscle
allows the scapula to rest in a downwardly rotated posi-
tion, causing the inferior border to become more promi-
nent (scapular winging).
32
Scapular winging may precip-
itate or contribute to persistent symptoms in patients with
orthopaedic shoulder abnormalities.
19
Thus, the injured
shoulder with subsequent immobilization or disuse may
benefit from a rehabilitation program that reconditions
the serratus anterior muscle.
Shoulder exercises intended to strengthen the serratus
anterior muscle and other scapular stabilizers have been in-
cluded in shoulder rehabilitation protocols.
4,16,23,25,26,29,34
Ex-
ercise selection has been based on EMG data
4,23,25,33
or clinical
experience.
26,29,34
Several researchers have recommended
that exercises be performed with the arm below 90° of humeral
elevation during the initial stages of rehabilitation to prevent
excessive strain on the rotator cuff and shoulder liga-
ments.
15,23,27,32
Despite these recommendations, a progres-
sion of shoulder rehabilitation exercises that specifically target
the serratus anterior muscle while constraining humeral ele-
vation below shoulder height has not, to our knowledge, been
investigated. The purpose of this study was to document the
EMG activity and applied resistance associated with eight
scapulohumeral exercises performed below shoulder height
that target the serratus anterior muscle and to design a con-
tinuum of serratus anterior muscle exercises for progressive
rehabilitation or training.
MATERIALS AND METHODS
Subject Preparation
Twenty male subjects (age, 30.4 6 5.1 years [average 6
SD]; height, 1.8 6 0.1 meters; weight, 80.1 6 9.6 kg) with
* Presented at the the American College of Sports Medicine annual con-
ference, Denver, Colorado, May 1997.
† Address correspondence and reprint requests to Michael J. Decker, MS,
Steadman Hawkins Sports Medicine Foundation, 181 West Meadow Drive,
Suite 1000, Vail, CO 81657.
No author or related institution has received any financial benefit from
research in this study. See “Acknowledgements” for funding information.
0363-5465/99/2727-0784$02.00/0
THE AMERICAN JOURNAL OF SPORTS MEDICINE, Vol. 27, No. 6
© 1999 American Orthopaedic Society for Sports Medicine
784
no history of shoulder injury gave their written informed
consent to participate in this study. Before testing, all
subjects participated in an orientation session to practice
the exercise techniques and maximum voluntary contrac-
tion (MVC) protocols.
Pregelled, silver-silver chloride bipolar surface elec-
trodes (NDM, Dayton, Ohio) were placed over five mus-
cles: the upper trapezius, middle trapezius, anterior del-
toid, posterior deltoid, and middle serratus anterior.
Surface electrodes were used because they are noninva-
sive and provide a reliable means to access the greatest
portrayal of motor unit activity of a superficial muscle.
10
The electrodes were placed according to the method de-
scribed by Basmajian and DeLuca,
3
with an interelectrode
distance of approximately 25 mm. Electrode placements
are described in Table 1 and in a previous study.
12
Elec-
trode placements were confirmed from a manual muscle
test of the primary muscle.
Experimental Protocol
On the day of testing, the subject warmed up by walking
on a treadmill at a self-selected pace for 5 minutes. The
testing session began with a series of five isometric MVCs
for each muscle. The 3-second maximum contractions
were interspersed with 3 to 5 seconds of rest. Maximum
voluntary contraction protocols for all muscles are de-
scribed in Table 1. All MVCs were performed in a joint
configuration that maximized EMG activity. Resistance
was provided by chains mounted to a wall and aligned
parallel to the line of pull. A consistent shoulder joint
position was ensured by placing rigid wedges between the
trunk and arm.
Applied force and EMG data were collected (1000 Hz)
for eight exercises that were performed below shoulder
height. Applied force was measured with a force plate
(Bertec Corp., Columbus, Ohio) or a force transducer (En-
tran Devices Inc., Fairfield, New Jersey) (maximal range,
61115 N) placed in series with an elastic resistance device
(Body Lines, Innovation Sports, Irvine, California). The
EMG data (bandwidth 10 to 2000 Hz) were collected using
Myosoft software (Noraxon USA Inc., Scottsdale, Arizo-
na). The input impedance of the EMG amplifier was less
than 10 megaohms, with a common-mode rejection ratio of
85 dB and a gain of 1000.
To assist in the delineation of exercise phases, a lateral
view of each subject was recorded using a 30-Hz video
camera. A manual timing signal was recorded with the
Myosoft software and video camera, providing for synchro-
nization of video and EMG data.
All exercises were completed in a slow, controlled man-
ner with the aid of a metronome. Each phase of the exer-
cises was performed at 54 beats per minute (bpm), except
the press-up (60 bpm), shoulder extension (100 bpm), and
serratus anterior punch (100 bpm) exercises. The exer-
cises using the elastic resistance device for resistance
were performed at a distance where the subject could
perform no more than 10 repetitions while maintaining
consistent metronome speed. The exercise order was ran-
domly selected for the first subject and subsequently bal-
anced to eliminate any order effects. The exercises were
performed as follows.
The push-up plus began with the subject in a prone
position with the hands shoulder-width apart and the
chest near the ground.
25
The subject then extended his
elbows to a standard push-up position and continued to
rise up by protracting the scapula. The subject returned to
the starting position by retracting the scapula and flexing
the elbows.
The knee push-up plus, a modification of the push-up
plus, was performed exactly like the push-up plus except
that body weight was supported by the hands and knees,
rather than the hands and feet.
The press-up was performed from a seated position,
with feet off the floor, hands at the level of the buttocks,
and the trunk and elbows slightly flexed.
31
The subject
then slowly raised his body up off the chair by straighten-
ing his arms. The subject held this position for 3 seconds
and then slowly returned to the starting position.
Shoulder extension was performed while the subject
stood with his chest facing the wall, knees slightly bent,
and feet shoulder-width apart in a split stance.
13,23
The
lead leg was the leg opposite the hand gripping the elastic
TABLE 1
Muscles Tested, Electrode Position, and MVC Protocol
Muscle Electrode position MVC joint position MVC Action
Upper trapezius At the angle of the neck and shoulder, over
the belly of the muscle in line with the
muscle fibers.
Arms fully extended, subject
standing on two chains grasping
the handles at mid-thigh.
Shoulder shrug
(Scapular
elevation and
retraction)
Middle trapezius Centered vertically between the medial
border of the scapula and the spines of
the thoracic vertebrae (T-1 to T-6).
Elbow flexed 15°, shoulder flexed
90° and internally rotated 45°.
Scapular
retraction
Anterior deltoid Two to three finger-breadths below the
acromion process, over the muscle belly,
in line with the fibers.
Elbow flexed to 90°, no shoulder
flexion.
Shoulder
flexion
Posterior deltoid Two finger-widths behind the angle of the
acromion, over the muscle belly, in line
with the fibers.
Elbow flexed 90°, shoulder abducted
45°, and internally rotated 45°.
Shoulder
extension
Serratus anterior Below the axilla, anterior to the latissimus,
placed vertically over the ribs 4–6.
Elbow flexed 45°, shoulder abducted
75° and internally rotated 45°.
Scapular
protraction
Vol. 27, No. 6, 1999 Serratus Anterior Muscle Activity and Shoulder Rehabilitation 785
resistance device, and the other leg was behind the mid-
line of the sagittal plane. The handle of the elastic resis-
tance device was grasped at waist height with the elbow
flexed 90° and the humerus in the neutral position. With
the arm kept at the side of the body, the subject fully
extended the humerus while flexing the elbow and then
slowly returned to the starting position.
The serratus anterior punch was performed while the
subject stood with his back to the wall, knees slightly bent,
and feet shoulder-width apart in a split stance.
29
The
handle of the elastic resistance device was grasped at
shoulder height with the elbow fully extended, the hu-
merus was internally rotated 45°, and the scapula was in
a retracted position. The subject then protracted and re-
tracted the scapula.
The forward punch was performed as the subject stood
with his back to the wall, with knees slightly bent and feet
shoulder-width apart in a split stance.
5,34
The subject
grasped the elastic resistance device with his arm at the
side of his body and his elbow flexed 90°. The subject
flexed the shoulder and extended the elbow until his hand
was at shoulder height with the elbow slightly flexed.
Scaption was performed while standing with the knees
slightly bent and the feet shoulder-width apart.
31
A dumb-
bell was firmly grasped in one hand at the side of the body
with the elbow fully extended and the shoulder externally
rotated 45° (thumb up). The subject performed humeral
elevation in the scapular plane up to shoulder height, then
slowly returned to the starting position.
The dynamic hug was performed while standing with
the back toward the wall, knees slightly bent, and the feet
shoulder-width apart. The subject began with the elbow
flexed 45°, the arm abducted 60°, and the shoulder inter-
nally rotated 45°. The subject then horizontally flexed the
humerus by following an arc described by his hands (hug-
ging action) (Fig. 1). Once the subject’s hands touched
together (maximum scapular protraction), he slowly re-
turned to the starting position.
Analysis
Each exercise was divided into phases of increasing and
decreasing force. The press-up exercise had an additional
isometric phase called the press-up hold.
25
The push-up
and plus phases for the knee push-up and push-up plus
exercises were further divided into phases of increasing
and decreasing force. The increasing force phase of the
push-up began with the chest near the ground (bottom)
and ended with the arms fully extended (top). The increas-
ing force phase of the push-up plus began with the arms
fully extended and continued until the scapulae were max-
imally protracted (plus). The decreasing force phase of the
push-up plus and push-up were simply the reverse, with
one repetition ending as the chest neared the ground. The
same phase demarcations were used for the knee push-up
plus.
All EMG data were processed with a 50-ms, root-mean-
square, moving window algorithm. The window duration
was chosen to smooth random myoelectric signals such
that the resulting envelope was as repeatable as possible
Figure 1. The dynamic hug exercise horizontally flexes the
humerus at a constant 60° of humeral elevation while the
hands follow an imaginary arc until maximum protraction is
attained.
Figure 2. Mean peak force (in newtons) and standard devi-
ations for the increasing (Inc Force) and decreasing force
(Dec Force) phases for the eight shoulder rehabilitation ex-
ercises: forward punch (FP), serratus anterior punch (SAP),
scaption (SCP), shoulder extension (SE), dynamic hug
(HUG), press-up (PRU), knee push-up plus (KPU1), and
push-up plus (PU1).
786 Decker et al. American Journal of Sports Medicine
without compromising true muscle activity.
11
Maximum
EMG reference values were calculated for each muscle
using the average of the five peak EMG signals and rep-
resented 100% MVC. Muscle activity during the exercise
was categorized as minimal (0% to 20% MVC), moderate
(21% to 50% MVC), or marked (.50% MVC).
23
Ten trials of EMG and force data for each subject were
analyzed to determine average and peak amplitudes for
all exercises during each phase. In an effort to be clinically
relevant, we expressed the EMG data as a percentage of
MVC (%MVC) and provided a relative measure of muscle
activity. Exercises that stimulated muscles with moderate
or marked average amplitudes (.20% MVC) in both the
increasing and decreasing force phases were reported.
Statistics
Descriptive statistics were calculated for peak and aver-
age amplitudes of muscle activity and force within each
phase for all muscles. Statistical trends of peak and aver-
age serratus anterior muscle activity, for both the increas-
ing and decreasing force phases, were determined through
simple regression analyses. The mean EMG activity of
each exercise was rank ordered, and the least-squares
statistical method was used to determine significant lin-
ear relationships. An adjusted r
2
value near 1.0 indicated
a strong relationship between the ranked exercise order
and progressive serratus anterior muscle activity. The
level of statistical significance was set at P # 0.05.
RESULTS
Group means and standard deviations for the applied
peak force are graphically presented in Figure 2. Mean
peak force (for one arm) ranged from 42 N in scaption to
386 N in the press-up. All other exercises resulted in a
peak force ranging from 225 to 325 N.
Group means and standard deviations for average and
peak EMG activity are presented in Table 2. Only those
muscles that elicited average EMG activity greater than
20% MVC during both the increasing and decreasing force
phases are presented.
As is typical with EMG data, there was substantial
intersubject variation in muscle activity for the serratus
anterior muscle, with coefficients of variation ranging
from 28% to 97%. However, reliability for average and
peak amplitude was excellent, as indicated by intraclass
correlation coefficients that ranged from r 5 0.977 to
0.990. Thus, despite the large variability between sub-
jects, EMG activity was highly consistent and reproduc-
ible within subjects for all exercises.
Serratus anterior muscle activity ranged from 3% to
109% MVC. The dynamic hug, scaption, serratus anterior
punch, and the knee and push-up plus exercises resulted
in the greatest muscle activity for the serratus anterior
muscle. The push-up plus elicited the greatest average
serratus anterior muscle activity (63% MVC), and the
dynamic hug the greatest peak serratus anterior muscle
activity (109% MVC), both in the increasing force phase.
Upper trapezius muscle activity ranged from 5% to 97%
MVC. Scaption and the dynamic hug exercises resulted in
the highest activity for the upper trapezius muscle. Scap-
tion elicited the greatest average (44% MVC) and peak
(97% MVC) upper trapezius muscle activity, both during
the increasing force phase.
Middle trapezius muscle activity ranged from 9% to 91%
MVC. Scaption was the only exercise that consistently
elicited activity greater than 20% MVC. Scaption had the
greatest average (41% MVC) and peak (91% MVC) middle
trapezius muscle activity, both during the increasing force
phase.
Anterior deltoid muscle activity ranged from 7% to
185% MVC. The forward punch, serratus anterior punch,
scaption, dynamic hug, press-up, and the knee and
push-up plus exercises demonstrated muscle activity
above 20% MVC. The push-up plus evoked the largest
average (103% MVC) and peak (185% MVC) anterior del-
toid muscle activity, both in the increasing force phase.
Posterior deltoid muscle activity ranged from 6% to
124% MVC. The shoulder extension and scaption exer-
cises consistently elicited EMG activity greater than 20%
MVC. Shoulder extension evoked the greatest average
(43% MVC) and peak (124% MVC) posterior deltoid mus-
cle activity, both in the increasing force phase.
Continuum Design
The four regression analyses of the rank-ordered exercises
on serratus anterior muscle EMG activity are displayed in
Figures 3 through 6. All regression analyses demon-
strated a significant linear trend (P 5 0.05) and accounted
for 64% to 86% of the variance in EMG activity.
The four regression analyses were incorporated into an
overall continuum for serratus anterior muscle activity.
Muscle activity was rank-ordered by exercise, with equal
weighting given to average and peak EMG amplitudes,
and to increasing and decreasing force phases (Table 3).
The exercise that consistently elicited the greatest EMG
activity for the serratus anterior muscle represented the
top-ranking exercise. The forward punch and press-up
had the same mean ranking. Because the forward punch
had a higher peak EMG magnitude, it was given the
higher rank in the overall continuum.
DISCUSSION
We designed a continuum of scapulohumeral exercises
based on serratus anterior muscle activity. The impor-
tance of a conditioned serratus anterior muscle has been
highlighted in EMG studies of sports such as swimming,
28
throwing,
9
and tennis.
30
A fatigued serratus anterior mus-
cle will reduce scapular rotation and protraction and allow
the humeral head to translate anteriorly and superiorly,
possibly leading to secondary impingement and rotator
cuff tears.
1
This study provides the clinician with several
serratus anterior muscle exercises that can be imple-
mented within a range of motion that is mechanically
optimal for early shoulder rehabilitation.
The push-up plus, dynamic hug, serratus anterior
punch, scaption, and the knee push-up plus exercises em-
Vol. 27, No. 6, 1999 Serratus Anterior Muscle Activity and Shoulder Rehabilitation 787
phasized scapular rotation and protraction and elicited
the greatest serratus anterior muscle activity. The results
of the push-up plus and scaption exercises are in agree-
ment with previous studies that have recommended both
of these exercises for scapular strengthening.
4,20,25
An
unreported knee push-up plus was determined to be more
user friendly than the push-up plus because it used less
applied force and elicited similar EMG amplitudes. This
study also provides support for the previously unsubstan-
tiated serratus anterior punch
29
and dynamic hug exer-
cises, since they both resulted in marked serratus anterior
muscle activity.
In contrast, the forward punch, shoulder extension, and
press-up exercises did not activate the serratus anterior
muscle above minimal EMG levels. The EMG activity we
saw during the forward punch exercise was similar to that
reported in a paper by Hintermeister et al.,
12
despite the
fact that they reported lower applied forces. In addition to
the small scapular range of motion provided by the for-
ward punch, the pectoralis major muscle may have been
used as the prime mover and may account for the low
serratus anterior muscle EMG amplitudes. The shoulder
extension and press-up exercises used the serratus ante-
rior muscle eccentrically as an antagonist to control scapular
retraction. Low EMG amplitudes were expected because, at
a given load, eccentric muscle actions demonstrate lower
amplitudes than concentric muscle actions.
24
Thus, for these
two exercises, EMG amplitude may not be an optimal tool to
measure exercise proficiency.
A primary emphasis in shoulder rehabilitation is to
restore the strength and coordination of the upper trape-
zius and serratus anterior muscles because they work
together to upwardly rotate the scapula for overhead
movements. Scaption and the dynamic hug exercises both
activated the upper trapezius and serratus anterior mus-
cles above minimal levels to perform upward scapular
rotation. Furthermore, scaption was the only exercise to
activate the middle trapezius muscle above 20% MVC.
These results were similar to those of other studies where
progressive humeral elevation in the scapular plane in-
duced middle and upper trapezius and serratus anterior
muscle activity.
2,14,22
Thus, while both exercises may be
used to effectively train the natural force couple between
the serratus anterior and the upper trapezius muscles,
scaption may be more effective.
Interestingly, the knee and push-up plus did not sub-
stantially recruit the upper trapezius muscle, even though
Figure 3. Average EMG amplitude (%MVC) and standard
deviations recorded from the serratus anterior muscle during
the increasing force phase for eight exercises. A significant
linear trend (
P
5 0.05) was demonstrated for the rank-
ordered data. See legend at Figure 2 for abbreviations. SEE,
standard error of estimate.
TABLE 2
Means and Standard Deviations (in Parentheses) Expressed as a Percentage of MVC for Peak and Average Amplitude in Muscles
with Average Amplitudes Greater than 20% MVC
Exercise Muscle
Peak amplitude
Increasing Decreasing
Forward punch Anterior deltoid 177.8 (79.0) 136.4 (69.1)
Serratus anterior punch Anterior deltoid 160.7 (79.3) 129.3 (62.9)
Serratus anterior 94.4 (30.8) 76.5 (27.5)
Scaption Upper trapezius 97.0 (41.1) 83.7 (40.9)
Middle trapezius 91.1 (34.6) 75.4 (28.2)
Anterior deltoid 183.3 (91.8) 157.2 (84.2)
Posterior deltoid 76.6 (41.8) 62.9 (34.6)
Serratus anterior 92.2 (28.5) 85.8 (28.6)
Shoulder extension Posterior deltoid 123.7 (75.4) 81.2 (54.8)
Dynamic hug Upper trapezius 51.8 (27.9) 46.0 (23.3)
Anterior deltoid 173.8 (82.3) 129.1 (70.8)
Serratus anterior 109.0 (30.7) 74.1 (24.7)
Bottom to Hold Hold Hold to Bottom
Press-up Anterior deltoid 104.1 (78.2) 123.0 (84.6) 53.8 (41.5)
Top to Bottom Bottom to Top Top to Plus Plus to Top
Knee push-up plus Anterior deltoid 89.5 (55.6) 127.0 (72.7) 82.8 (57.0) 70.6 (44.7)
Serratus anterior 47.3 (16.7) 68.5 (22.4) 72.0 (27.0) 65.3 (28.4)
Push-up plus Anterior deltoid 147.6 (81.7) 185.2 (85.1) 170.5 (80.5) 142.0 (68.8)
Serratus anterior 73.0 (29.2) 100.0 (37.5) 104.0 (38.0) 91.6 (33.1)
788 Decker et al. American Journal of Sports Medicine
the scapula maintained an upwardly rotated position
throughout both exercises. Load-bearing exercises for the
shoulder, regardless of whether the hand is fixed or mov-
ing,
7
are intended to promote proximal humeral stability
with scapular stabilization.
17
Increased proximal joint
stability, derived from body weight compressing the joint,
may retain upward scapular rotation, resulting in reduced
levels of upper trapezius muscle activity.
All exercises induced either anterior or posterior deltoid
muscle activity greater than 20% MVC. This finding high-
lights the important relationship between the deltoid and
scapular rotator muscles during humeral elevation. The
deltoid muscles attain maximum range of motion and
power production when the scapula is upwardly rotated
throughout humeral elevation.
14,22
Because of the angle
of pull of the deltoid, marked activity of this muscle may
cause superior humeral migration and, thus, shoulder
pain.
17,27
However, these muscles provide scapulo-
humeral stability in the range of 30° to 60° of abduction by
compressing the humerus into the glenoid.
21
Therefore,
exercises that begin with, and maintain, humeral eleva-
tion (such as the dynamic hug) may assist the rotator cuff
muscles with joint stability and avoid excessive superior
shear forces that may be detrimental to rehabilitation.
Figure 4. Peak EMG amplitude (%MVC) and standard de-
viations recorded from the serratus anterior muscle during
the increasing force phase for eight exercises. A significant
linear trend (
P
5 0.05) was demonstrated for the rank-
ordered data. See legend at Figure 2 for abbreviations. SEE,
standard error of estimate.
Figure 5. Average EMG amplitude (%MVC) and standard
deviations recorded from the serratus anterior muscle during
the decreasing force phase for eight exercises. A significant
linear trend (
P
5 0.05) was demonstrated for the rank-
ordered data. See legend at Figure 2 for abbreviations. SEE,
standard error of estimate.
Figure 6. Peak EMG amplitude (%MVC) and standard de-
viations recorded from the serratus anterior muscle during
the decreasing force phase for eight exercises. A significant
linear trend (
P
5 0.05) was demonstrated for the rank-
ordered data. See legend at Figure 2 for abbreviations. SEE,
standard error of estimate.
TABLE 2
Continued
Average amplitude
Increasing Decreasing
77.5 (32.9) 43.1 (20.1)
81.0 (41.6) 58.3 (26.6)
47.0 (13.5) 31.4 (9.8)
43.7 (19.0) 31.3 (13.6)
41.3 (15.0) 28.4 (9.9)
83.3 (40.5) 52.4 (26.3)
30.3 (17.1) 20.4 (10.4)
37.3 (12.6) 24.2 (7.9)
42.8 (31.2) 22.6 (15.5)
23.1 (12.6) 20.3 (8.9)
89.7 (43.8) 50.2 (26.0)
54.1 (14.6) 28.1 (9.5)
Bottom to Hold Hold Hold to Bottom
51.1 (40.1) 52.2 (34.8) 53.8 (41.5)
Top to Bottom Bottom to Top Top to Plus Plus to Top
42.5 (27.7) 61.7 (34.7) 45.1 (29.2) 36.0 (21.2)
22.9 (7.9) 33.9 (10.1) 42.1 (15.4) 35.2 (12.7)
73.9 (43.8) 100.2 (47.8) 102.9 (50.5) 75.5 (37.4)
35.3 (14.0) 52.1 (16.9) 63.1 (19.9) 53.0 (17.7)
Vol. 27, No. 6, 1999 Serratus Anterior Muscle Activity and Shoulder Rehabilitation 789
Peak and average EMG amplitudes, from both force
phases, were integrated into an overall exercise contin-
uum. In a dynamic situation, the relationship between
EMG activity and muscle force is multifactorial.
3
How-
ever, investigators have generally inferred muscular ten-
sion to increase, whether linearly or nonlinearly, with the
EMG signal.
3,12
Thus, exercises with greater peak EMG
amplitudes may be important for strength training or
used to determine a safe upper limit for postsurgical mus-
cle activity. The average EMG amplitude represents the
arithmetic mean of muscular activity within a force phase
and may be important for endurance training. Exercises
with larger average amplitudes may offer greater muscu-
lar challenges and require greater physiologic efforts. If
either the peak or average EMG amplitude is of primary
concern in a rehabilitation protocol, one of the four param-
eter and phase-specific continuums may be applied rather
than the overall continuum.
Applied force was not included in the overall design of
the continuum because each exercise demonstrated scap-
ular movements that were different in magnitude and
direction (for example, protraction, depression). Exercises
that did not use scapular protraction, rotation, or a com-
bination, elicited minimal levels of serratus anterior mus-
cle activity, regardless of the peak force. Therefore, it may
be more important to select an exercise based on the
direction of the applied force and the associated scapular
actions, while adjusting the resistance to accommodate
individual needs.
SUMMARY
Average and peak serratus anterior muscle activity was
documented for eight shoulder rehabilitation exercises
performed below 90° of humeral elevation, and an exercise
continuum was designed. The data illustrate that sub-
stantial serratus anterior muscle activity can be achieved
in a range of motion that is recommended and safe for
most patients with shoulder lesions. The selection of an
appropriate exercise progression focused on serratus an-
terior muscle activity but may depend on other factors as
well (for example, shear or compressive forces, user friend-
liness, or number of muscles recruited). However, the
exercise continuum is used most efficiently when the cli-
nician chooses an exercise that matches the movement
limitations and goals of the patient.
ACKNOWLEDGMENTS
This project was supported by a grant from the NFL
Charities, New York, New York. The authors thank Den-
nis O’Connor for his technical support and recognize Mi-
chael R. Torry, Sherry L. Werner, and Tricia A. Murray for
their assistance.
REFERENCES
1. Allegrucci M, Whitney SL, Irrgang JJ: Clinical implications of secondary
impingement of the shoulder in freestyle swimmers.
J Orthop Sports Phys
Ther 20:
307–318, 1994
2. Bagg SD, Forrest WJ: Electromyographic study of the scapular rotators
during arm abduction in the scapular plane.
Am J Phys Med 65:
111–124,
1986
3. Basmajian JV, DeLuca CJ:
Muscles Alive: Their Functions Revealed by
Electromyography.
Fifth edition. Baltimore, Williams & Wilkins, 1985
4. Belle RM, Hawkins RJ: Dynamic electromyography analysis of the shoul-
der muscles during rotational and scapular strengthening exercises, in
Post M, Morrey BF, Hawkins RJ (eds):
Surgery of the Shoulder.
St. Louis,
Mosby Year Book, 1990, pp 32–35
5. Burkhead WZ Jr, Rockwood CA Jr:Treatment ofinstability ofthe shoulder
with an exercise program.
J Bone Joint Surg 74A:
890–896, 1992
6. Codman EA:
The Shoulder.
Boston, Thomas Todd, 1934
7. Dillman CJ, Murray TA, Hintermeister RA: Biomechanical differences of
open and closed chain exercises with respect to the shoulder.
J Sport
Rehabil 3:
228–238, 1994
8. Dvir Z, Berme N: The shoulder complex in elevation of the arm: A mech-
anism approach.
J Biomech 11:
219–225, 1978
9. Glousman R, Jobe F,Tibone J, etal: Dynamic electromyographic analysis
of the throwing shoulder with glenohumeral instability.
J Bone Joint Surg
70A:
220–226, 1988
10. Greis PE, Kuhn JE, Schultheis J, et al: Validation of the lift-off test and
analysis of subscapularis activity during maximal internal rotation.
Am J Sports Med 24:
589–593, 1996
11. Hershler C, Milner M: An optimality criterion for processing electromyo-
graphic (EMG) signals relating to human locomotion.
IEEE Trans Biomed
Eng 25(5):
413–420, 1978
12. Hintermeister RA, Lange GW, Schultheis JM, et al: Electromyographic
activity and applied load during shoulder rehabilitation exercises using
elastic resistance.
Am J Sports Med 26:
210–220, 1998
13. Hughes M, Neer CSII: Glenohumeral joint replacement and postoperative
rehabilitation.
Phys Ther 55:
850–858, 1975
14. Inman VT,SaundersJBDM, Abbot LC: Observationsonthe function of the
shoulder joint.
J Bone Joint Surg 26:
1–30, 1944
15. Jobe FW, Moynes DR, Brewster CE: Rehabilitation of shoulder joint
instabilities.
Orthop Clin North Am 18:
473–482, 1987
16. Jobe FW, Pink M: Classification and treatment of shoulder dysfunction in
the overhead athlete.
J Orthop Sports Phys Ther 18:
427–432, 1993
17. Keirns MA: Conservative management of shoulder impingement, in An-
drews JR, Wilk KE (eds):
The Athlete’s Shoulder.
New York, Churchill
Livingstone, 1994, pp 605–622
18. Kibler WB: Normal shoulder mechanics and function.
Instr Course Lect
46:
39–42, 1997
19. Kibler WB: Role of the scapula in the overhead throwing motion.
Contemp
Orthop 22:
525–532, 1991
20. Lear LJ, Gross MT: An electromyographical analysis of the scapular
stabilizing synergists during a push-up progression.
J Orthop Sports Phys
Ther 28:
146–157, 1998
21. Liu J, Hughes RE, Smutz WP, et al: Roles of deltoid and rotator cuff
muscles in shoulder elevation.
Clin Biomech 12:
32–38, 1997
22. Ludewig PM, Cook TM, Nawoczenski DA: Three-dimensional scapular
orientation and muscle activity at selected positions of humeral elevation.
J Orthop Sports Phys Ther 24:
57–65, 1996
23. McCann PD, Wootten ME, Kadaba MP, et al: A kinematic and electro-
myographic study of shoulder rehabilitation exercises.
Clin Orthop 288:
179–188, 1993
24. Moritani T, Muramatsu S, Bacharach D: Comparison of intramuscular and
surface electromyograms during concentric and eccentric contractions, in
Jonsson B (ed):
Biomechanics X-A.
Champaign, IL, Human Kinetics,
1987, pp 221–225
TABLE 3
Final Rank Order of Exercises for Activating the Serratus
Anterior Muscle
Final
order
Exercise
Increasing
force
phase
a
Decreasing
force phase
AA PA AA PA
1 Push-up plus 1211
2 Dynamic hug 2154
3 Serratus anterior punch 3343
4 Scaption 5462
5 Knee push-up plus 4627
6 Forward punch 6575
7 Press-up 7736
8 Shoulder extension 8888
a
AA, average amplitude; PA, peak amplitude.
790 Decker et al. American Journal of Sports Medicine
25. Moseley JB Jr, Jobe FW, Pink M, et al: EMG analysis of the scapular
muscles during a shoulder rehabilitation program.
Am J Sports Med 20:
128–134, 1992
26. Paine RM, Voight M: The role of the scapula.
J Orthop Sports Phys Ther
18:
386–391, 1993
27. Pappas AM, Zawacki RM, McCarthy CF: Rehabilitation of the pitching
shoulder.
Am J Sports Med 13:
223–235, 1985
28. Pink M, Perry J, Browne A, et al: The normal shoulder during freestyle
swimming. An electromyographic and cinematographic analysis of twelve
muscles.
Am J Sports Med 19:
569–576, 1991
29. Regan D: The role of scapular stabilization in overhead motion.
Strength
Conditioning 18(1):
33–38, 1996
30. Ryu RKN, McCormick J, Jobe FW, et al: An electromyographic analysis
of shoulder function in tennis players.
Am J Sports Med 16:
481–485,
1988
31. Townsend H, Jobe FW, Pink M, et al: Electromyographic analysis of the
glenohumeral muscles during a baseball rehabilitation program.
Am J Sports Med 19:
264–272, 1991
32. Watson CJ, Schenkman M: Physical therapy management of isolated
serratus anterior muscle paralysis.
Phys Ther 75:
194–202, 1994
33. Wilk KE, Arrigo C: Current concepts in the rehabilitation of the athletic
shoulder.
J Orthop Sports Phys Ther 18:
365–378, 1993
34. Young DC, Rockwood CA Jr: Complications of a failed Bristow procedure
and their management.
J Bone Joint Surg 73A:
969–981, 1991
Vol. 27, No. 6, 1999 Serratus Anterior Muscle Activity and Shoulder Rehabilitation 791
