Three-Dimensional Scapular Orientation and
Muscle Activity at selected Positions of
Paula M. ludewig, MA, PT'
Thomas M. Cook, PhD, PT *
Deborah A. Nawoczenski, PhD, PT3
levation of the arm for
Downloaded from www.jospt.org at on March 29, 2021. For personal use only. No other uses without permission.
overhead activities is ac-
complished by combined
motion at multiple articu-
lations of the shoulder,
including the sternoclavicular, acro-
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
mioclavicular, and glenohumeral
joints (24,48). Some authors include
the scapulothoracic articulation when
describing shoulder anatomy and ki-
nesiology (24,29). Due to the liga-
mentous and capsular attachments of
the scapula to the clavicle and the
clavicle to the sternum, scapulotho-
racic movement requires motion of
Journal of Orthopaedic & Sports Physical Therapy®
the clavicle on the thorax at the ster-
noclavicular joint, motion of the
scapula relative to the clavicle at the
acromioclavicular joint, o r some com-
bination of both (24). Scapulotho-
racic motion, therefore, is a summa-
tion of sternoclavicular and acromio-
clavicular motion, and, subsequently,
elevation of the arm is frequently de-
scribed in terms of scapulothoracic
and glenohumeral components.
Cathcart, while observing arm
movements in living subjects, first
suggested that glenohumeral and
scapulothoracic motion occur syn-
chronously when lifting the arm over-
head (7). Codman later termed this
synchronous motion, scapulohumeral
rhythm (8). Since that time, a great
deal of research in shoulder kinemat-
ics has been directed toward the
study of scapulohumeral rhythm (2,
14,19,24,45,48).The majority of this
JOSIT Volume 24 Number 2 August 1996
Abnormal scapular kinematics and associated muscle function presumably contribute to
shoulder pain and pathology. An understanding of scapular kinematic and electromyographic
profiles in asymptomatic individuals can provide a basis for evaluation of pathology. The purpose
of this study was to describe normal three-dimensional scapular orientation and associated muscle
activity during humeral elevation. Twenty-five asymptomatic subjects, 19-37 years old, were
evaluated. Digitized coordinate data and surface electromyographic signals from the trapezius
(upper and lower), levator scapulae, and serratus anterior were colleaed at static positions of 0,
90, and 140" of humeral elevation in the scapular plane. The scapula demonstrated a pattern of
progressive upward rotation, decreased internal rotation, and movement from an anteriorly to a
posteriorly tipped position as humeral elevation angle increased. Electromyographic activity of all
muscles studied increased with increased humeral elevation angles. Differences between mean
values at all elevation angles for all variables were significant (p < .05), except for the lower
trapezius between the 90 and 140" humeral angles. The results of this study suggest assessment of
scapular tipping and internal rotation as well as upward rotation may be necessary to understand
pathologies of the shoulder that are related to abnormal scapular kinematics.
Key Words: scapula, electromyography, shoulder joint
' Doctoral student, Physical Therapy Graduate Program, 2600 Steindler Building, The University of lowa,
lowa City, IA 52242
Associate Professor, Department of Preventive Medicine and Physical Therapy Graduate Program, The
University of Iowa, lowa City, IA
' Associate Professor, Department of Physical Therapy, lthaca College-Rochester campus, Rochester, NY
previous research is limited to two- allel to the scapular spine has been
dimensional (2-D) studies of scapular termed flexion/extension (13), ante-
upward rotation about an axis per- rior/posterior or forward/backward
pendicular to the plane of the scap tilting (36,39), or anterior/posterior
ula during humeral elevation. tipping (40). Around a vertical axis,
Scapular motion is known to oc- scapular motion has been defined as
cur about other axes as well (13,34, anterior/posterior rotation (39). in-
36,55). Poppen and Walker described ternal/external rotation (36), or
a twisting movement of the scapula winging (40). In this paper, rotation
occurring in combination with u p about an axis perpendicular to the
ward rotation as the arm was elevated plane of the scapula is defined as u p
(45). A variety of terms has been ward/downward rotation, rotation
used to define this motion. Scapular about an axis parallel to the scapular
rotation around an axis roughly par- spine is described as anterior/poste-
57
 RESEARCH STUDY

descriptions of scapular motion (2,

l5,24).
Although there are minimal re-
search data describing the contribu-
tions of accessory scapular motions of
internal/external rotation and t i p
ping to normal scapular kinematics,
these accessory motions are fre-
quently discussed and evaluated clini-
cally (5,38,42). Abnormal scapular
kinematics in one or all planes and
associated abnormal muscle function
FIGURE 1. Local coordinate systems and axis orientation for the scapula and twnk. X,, Y,, and Z, defined the are believed to contribute to shoul-
anatomical coordinate system embedded in the scapula, with the psifive X, axis directed medial to lateral along der pain and pathology (20,21,SO).
the spine of the scapula, the positive Y, axis directed anteriorly perpendicular to the plane of the scapula, and Culham and Peat (10) reported alter-
the positive Z, axis directed superiorly. IRBR = Internal/externalrotation. UWDR = Upward/downward rotation. ations in resting scapular orientation
ATPT = Anterior/posterior tipping. X,, Y,, andZ, defined the coordinate system embedded in the twnk with the
positive X, axis directed medial to lateral, the positive Y, axis directed anteriorly, and the positive Z, axis directed
related to age and spinal posture in
superiorly. Orientation was described for the scapula relative to the trunk. females. Significant changes in scapu-
Downloaded from www.jospt.org at on March 29, 2021. For personal use only. No other uses without permission.

lar internal rotation and increases in

anterior tilt of the scapula occurred
nor tipping, and rotation around a the scapula (27,48). In addition, with increased age and increasing
vertical axis is referred to as internal/ nearly all of these previous investiga- slope of the upper thoracic spine,
external rotation (Figure 1). tions (26,31,34,55) report only mean while the scapular upward rotation
angle was not affected by spinal pos-
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Despite the knowledge of poten- values for their particular sample,

tial three-dimensional (3-D) scapular
motion, inherent difficulties in scapu- the variability known to occur be-
lar measurement associated with over- tween subjects.
lying skin movement have hindered
3-D kinematic analyses of the shoul-
der. Investigations of shoulder mo-
tion completed in 3-D frequently
have been limited to humeral mo-
Journal of Orthopaedic & Sports Physical Therapy®
without providing an indication of ture or age (10). In addition, a study
comparing scapular position (upward
rotation and protraction) in relaxed
To produce the complex kine- standing between patients with shoul-
matics at the shoulder during hu- der overuse injuries and healthy sub-
meral elevation, complementary ac- jects found n o significant diKerences
tion of scapulothoracic and gleno- between groups for these scapular
humeral muscles is required (24,49). variables (22). The authors suggest
that a lack of assessment of scapular

tions relative to the trunk (16,25) or Electromyographic (EMG) activity of

they have been based on cadaver the scapulothoracic muscles has been tipping and internal rotation may
data (1). Other previous 3-D studies studied by numerous authors (3,4,6, have contributed to their nonsignifi-
are often difficult to interpret clini- 11.1 7,24,49,57). Although these stud- cant findings (22). The results of pre-
cally (23,26,31) because the chosen ies verify the activity of all portions of vious investigations indicate that as-
angular descriptions differ from com- the trapezius, the levator scapulae, sessment of 3 D scapular orientation,
mon clinical understanding. rhomboids, and lower serratus ante- including internal rotation and t i p
Descriptions of scapular rest posi- rior during arm elevation, disagree- ping, is warranted in future kine-
tion have been reported in 3-D (10, ment exists as to their relative contri- matic studies of patient populations.
34,36). However, the shoulder litera- butions to this activity. The use of A more complete understanding of
ture, providing clinically meaningful differing electrodes (needle, fine scapular motion and muscle activity
3-D descriptions of scapular motion wire, or surface), a lack of normaliza- in functional planes of elevation in
o r orientation beyond rest positions, tion procedures in most studies, a asymptomatic individuals is needed
is sparse (34,36,55). Further, the ma- lack of standard electrode place- in order to provide a basis for further
jority of these previous 3-D studies ments, and differences in instrumen- understanding of shoulder dysfunc-
have assessed scapular position dur- tation and methods all contribute to tion in symptomatic individuals.
ing humeral elevation in the cardinal this variability between studies. Over- The purposes of this paper are:
planes (26,34,55). Elevation of the all, the general understanding of I ) to describe and compare 3-D scap
arm for functional activities, however, muscle activity of the scapular rota- ular orientation at static humeral an-
rarely occurs in the cardinal planes, tors during arm elevation is predomi- gles of elevation in the scapular
instead, occurring approximately nately qualitative. Furthermore, asso- plane and 2) to describe and com-
midway between these planes in what ciated discussions of muscle function pare muscle activity of the upper tra-
has been described as the plane of are commonly related to only 2-D pezius, lower trapezius, levator scapu-

Volume 24 a Number 2 August 1996 JOSPT


- lengths were completed with custom
Variable X SD Minimum Maximum
Age (years) 25.9 5.2 19.0
Height (rn) 1.7 0.1 1.6
Weight (kg) 66.9 11.4 47.7
TABLE 1. Descriptive statistics on demographic data.
lae, and serratus anterior at these
same humeral positions.
METHODS
Subjects
Twenty-five subjects (14 women
and 11 men), without a history of
shoulder pain, pathology, or range of
motion restriction, voluntarily partici-
pated in this study. The eligible sub-
Downloaded from www.jospt.org at on March 29, 2021. For personal use only. No other uses without permission.
ject age range was limited to 18-40
years old. Demographic information
for the subjects is presented in Table
1. All but two of the subjects (92%)
were right-hand dominant. Each sub-
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
ject was given a verbal and written
summary of information about the
study and signed a consent form
prior to participation.
Design
A one-way repeated measures de-
sign was used to determine the effect
Journal of Orthopaedic & Sports Physical Therapy®
of humeral angle on scapular orienta-
tion and muscle activity. The inde-
pendent variable humeral angle had
three levels: 0, 90, and 140" of hu-
meral elevation relative to the trunk.
These angles were selected to provide
data in the functional range of eleva-
tion. The dependent variables were
scapular upward rotation angle, inter-
nal rotation angle, and tipping angle
all relative to the trunk (Figure 1)
and EMG values from the levator
scapulae, upper trapezius, lower tra-
pezius, and serratus anterior normal-
ized as a percentage of maximum
voluntary isometric contraction. Data
collection was completed on the right
shoulder only for all subjects.
Instrumentation
A position control system was
fabricated for this study to allow the
JOSPT Volume 24 Number 2 August 1%6
RESEARCH STUDY
software programs. Similar linkage
37.0
1.9
digitizers have been used in previous
86.4 E shoulder kinematic research (37,47).
The digitizer and related software
were calibrated for the measurement
of 3-D linear distances over the work-
ing range. Resultant worst case linear-
ity, hysteresis, repeatability, and accu-
racy were 1 mm, 2 mm, 2 mm, and 4
mm, respectively. Worst case angular
errors were determined to be no
FIGURE 2. Diagram of a subject positioned for data
collection. A) Upper thoracic stabilization; B) lumbar greater than 2" (35).
support and stabilization; C) anterior trunk stabiliza- Silver-silver chloride surface elec-
tion; D) arm cuff with attached pendulum potentiom- trode assemblies with an interelec-
eter; and E) meter system for visual feedback of arm trode distance of 20 mm, 8 m m diam-
position.
eter active electrodes, and on-site
preamplification (Therapeutics Un-
subjects to consistently maintain the limited, Iowa City, IA) were used to
desired humeral angle within the record EMG activity of the selected
scapular plane during the data collec- muscles. The signals were passed to a
tion process and replicate the posi- GCS 67 amplifier with adjustable gain
tions for subsequent measurements. settings, high input impedance
A gravity-referenced pendulum po- (greater than 15 megohms at 100
tentiometer was attached to a cuff on Hz), a common mode rejection ratio
the humerus (Figure 2). A neutral of 87 dB at 60 Hz, and a bandwidth
arm position (anatomical position of 40-4000 Hz. The raw signals were
with the arm at the subject's side) root mean square processed with a
was obtained, and the measurement time constant of 55 msec. All signals
system was zeroed to this position. were collected on-line to a microcom-
Subsequent elevation of the humerus puter simultaneously as voltages were
from this position resulted in a volt- collected from the digitizer. Electro-
age output that was displayed on an myographic signals from each muscle
analog meter placed in front of the were monitored on an oscilloscope
subject (Figure 2). Voltage outputs (Tektronix 7313, Beaverton, OR)
consistent with the desired elevation throughout data collection in order
angles were marked on the meter, to verify signal quality.
thus allowing the subjects visual feed-
back to maintain positions within and Procedures
between trials.
Digitized 3-D coordinate loca- Electrode placement locations
tions were obtained via an electrome- were chosen based on previous stud-
chanical linkage digitizer constructed ies collecting surface EMG from the
for this project. Positioning of the muscles of interest (33,50,53,58). All
pointer of the digitizer to a location electrodes were aligned in the direc-
within its working range resulted in tion of the muscle fibers. The levator
voltage output from each of three scapulae electrode was placed be-
precision rotary potentiometers. Cus- tween the posterior margin of the
tom data acquisition software and a sternocleidomastoid muscle and the
12 bit A/D board (Dash 16F, Metra- anterior malgin of the upper trape-
byte Corporation, Stoughton, MA) zius. In this region, the levator scapu-
were used for on-line data collection lae is superficially located (50,53).
to a microcomputer. Calculations of The upper trapezius electrode was
x, y, z Cartesian coordinates from raw placed one-third of the distance be-
voltage values and linkage arm tween C7 and the acromion process.
 RESEARCH STUDY

The lower trapezius electrode was collected while subjects maintained rotation of the scapula about an axis
placed one-half of the distance be- each position. Three trials were com- perpendicular to the plane of the
tween the inferior angle of the scap pleted in each position. During each scapula, and rotations about X, de-
ula and the thoracic spine. The serra- trial, the medial inferior edge of the fined anterior/posterior tipping of
tus anterior electrode was placed spine of the scapula, the posterolat- the scapula about a medial-lateral
over the muscle fibers just lateral to era1 tip of the acromion, and the in- axis (Figure 1).
the inferior angle of the scapula ferior angle of the scapula were pal- Root mean square EMG voltage
(12). A ground electrode was placed pated and digitized. The three points values were averaged for each posi-
on the distal ulna of the left wrist. used to describe the arm and trunk tion and across the %second period
After electrode placement, verifica- orientations were digitized from the for each maximum voluntary isomet-
tion of signal quality was completed humeral cuff and trunk reference ric contraction. Resting EMG voltage
for each muscle at each arm position. frame, respectively. The reference values were then subtracted from
Three maximum voluntary iso- frame to which the trunk was aligned both the maximal contraction and
metric contractions for each muscle was digitized at the beginning of the arm position data. The highest aver-
were collected, sampling for 3 sec- data collection session. During digi- age maximum contraction of the
onds after the first second of effort. tizing, EMG signals were simulta- three trials for each muscle and posi-
Maximum contractions were col- neously collected from the muscles of tion was used as the normalization
lected for the lower trapezius and interest. Data collection began with reference. Average values for each
Downloaded from www.jospt.org at on March 29, 2021. For personal use only. No other uses without permission.

serratus anterior in traditional man-
ual muscle test positions (28). The
maximum contractions for the re-
maining muscles were collected with
the subject seated. Upper trapezius
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
the examiner triggering a foot switch,
and data were collected at 300 Hz for
0.5 seconds at each point. The sub-
ject was allowed 30 seconds of rest
between each position.
arm position were normalized as a
percentage of the respective maxi-
mum voluntary isometric contraction.
The degree of similarity among
the three trials at each arm position
was assessed using type 3, 1 intraclass

voluntary isometric contractions were Digitized angular values for hu-

collected with maximum resistance to meral positions were displayed on the correlation coefficients (ICCs) for
arm abduction at a position of 90" of computer terminal immediately upon each dependent variable (32.46). The
humeral abduction in the scapular completing data collection at each standard error of measurement was
plane (51). Levator scapulae volun- position. If digitized values were out- also calculated for the kinematic
tary isometric contractions were col- side a range of 25" of the targeted data. The mean of the three trials at
lected with maximum resistance to position, then that position was re- each arm position was used in all
right lateral flexion of the head, peated before collecting data in the subsequent analysis. Statistical Analy-
while the arm was actively held by the next position. sis System software (SAS Institute,
Cary, NC) was utilized to compute a
Journal of Orthopaedic & Sports Physical Therapy®

subject at 90" of abduction in the

scapular plane (51) .
Subjects were seated and stabi-
lized to a lumbar support by a strap
placed around their waist. Their
trunk was aligned vertically to the
trunk reference frame on the chair.
Limitation of trunk movement from
this position was controlled by place- scapula are illustrated in Figure 1.
ment of stabilizing bars, anteriorly
contacting the sternum and posteri-
orly contacting the upper thoracic
spine. This stabilization did not re-
strict scapular motion or muscle ac-
tivity (Figure 2). Shoulder elevation
in the plane of the scapula was con-
trolled by having the subject elevate
the right arm along a flat planar sur- about Z, defined scapular internal/
face angled 30" anterior to the coro-
nal plane.
The subject actively obtained
each humeral position in a prese-
lected random order, and data were
Data Reduction and Analysis
A local reference frame was es-
tablished for each body segment us-
ing the three noncollinear points col-
lected for that segment (56). The
axis orientations for the trunk and
The trunk reference frame was estab-
lished coincident with the cardinal
planes. The threedimensional angu-
lar orientation of the scapula and
humerus at each arm position was
described relative to the trunk using
a Z, Y, X" ordered Cardan angle ro-
tation sequence (9.56). Rotations
external rotation relative to the coro-
nal plane. This first rotation can be
considered as defining the orienta-
tion of the scapular plane. Rotations
about Y, defined upward/downward
one-way repeated measures analysis
of variance (ANOVA) for each de-
pendent variable of interest. A signifi-
cance level of .05 was used to test for
statistical differences in the overall
model. When a significant main ef-
fect for humeral angle was found,
Tukey follow-up tests were performed
to adjust the significance level and
control the experiment-wise error
rate at a .05 level.
RESULTS
Intraclass correlation coefficients
assess the degree of similarity among
the three trials at each arm position
for each subject. The values for each
of the kinematic and EMG variables
across all humeral angles are pre-
sented in Table 2. The ICCs ranged
from .78 to .9Qand are indicative of

Volume 24 Number 2 August 1996 JOSPT

 RESEARCH STUDY
------..--. -.

w humeralaigb
140' hurneral angle
t

TABLE 2. lntraclass correlation coetiicients (ICCI tor

kinematic and electromyographic variables based on
three trials within the same session.

good within-session reliability for all hator Upper Lorn Senahm

Scapulm Traperha Traperha Antsnw
dependent variables of interest (46).
Additionally, the standard error of Kinematic Vatlable
Muscle
measurement calculated for the scap FIGURE 3. Mean scapular orientation across subjects.
ular angle data estimates the average FIGURE 4. Mean electromyographic activity of the
four scapular rotators. 'MVIC = Maximum voluntary
error of' the measurement for any
magnitude, 100% of the individual isometric contraction.
Downloaded from www.jospt.org at on March 29, 2021. For personal use only. No other uses without permission.

given trial. These calculations reflect

subjects followed this general pattern
the repeatability of the measurement
from anterior to posterior tipping sented with either a further increase
method as a whole, including repeat-
with increased arm elevation. Results in activity as compared with the 90"
ability of the instrumentation, repeat-
of the repeated measures ANOVA angle, a maintenance of the 90" mus-
ability of palpation of bony land-
revealed a significant main effect of cle activity level (defined as a value
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

marks, and the ability of the subjects

humeral angle ( p < .001) for each of falling within the 95% confidence
to replicate the specific arm position.
the three scapular orientation vari- interval for that subject's mean val-
The standard error of measurement
ables. Follow-up testing of means ue), or a decrease in activity from the
was consistently 2" for scapular t i p
across humeral angles resulted in sig- 90" level.
ping angle at all three arm positions.
nificant differences for all pair-wise For the upper trapezius, 64% of
For upward rotation and internal ro-
comparisons. the subjects followed the first pattern
tation angles, the standard error of
Means and standard deviations of progressively increased activity,
measurement ranged from 2 to 3"
across subjects for the muscle activity 24% plateaued at the 90" activity lev-
across arm positions.
variables are presented graphically in el, and 12% demonstrated slightly
Journal of Orthopaedic & Sports Physical Therapy®

Means and standard deviations

Figure 4 and numerically in Table 4. decreased activity at 140". For the
across subject5 for the three scapular
The mean EMG activity increased lower trapezius, 48% followed a pat-
angular orientation variables are pre-
progressively as humeral angle in- tern of progressively increased activ-
sented graphically in Figure 3 and
creased for all of the muscles studied. ity, 20% plateaued in activity at the
numerically in Table 3. The mean
All subjects demonstrated increased 90" level, and 32% demonstrated de-
scapular upward rotation angle in-
activity for all muscles at a 90" hu- creased activity at 140".Serratus ante-
creased progressively as the humeral
meral angle as compared with the 0" rior activity was more consistent
angle increased. Although differing
humeral angle, except for levator across subjects, with 92% following a
in magnitude, 100% of the individual
scapulae values from one subject pattern of progressively increased
subjects followed this pattern of pro-
which remained unchanged at 1%. activity and 8% (two subjects) dem-
gressive upward rotation. Mean scap
At the 140" humeral angle, three pat- onstrating a decrease in activity from
ular internal rotation decreased pro-
terns were possible. Subject5 pre- 90-140". For the levator scapulae,
gressively as humeral angle increased.
Eighty-four percent of the subjects . -
followed this general pattern of pro- Hurneral Bwation Angle
gressive external rotation, while 16% Kinematic Variable 0" 90" 140"
maintained approximately the same - - -
scapular plane orientation at all lev- X SD X SD X SD
els of humeral elevation. Mean t i p Upward rotation 2 6 21 5 36 4
ping angle increased progressively Internal rotation 33 9 28 6 20 6
Tipping* -8 4 -2 3 7 5
from an anterior to a posterior
tipped position as humeral angle in- * tY(ydt~\
e \ ,I/w\ fridft dte ,intcwor tfpprry; porrti\ e \ ' h e ; ir~drc,~te
pocterfor tfpprn,q
creased. Again, although differing in TABLE 3. Summary data (x, SD) of scapular angular orientation in degrees.

JOSPT Volume 24 Number 2 Augusr 1 9 6

 RESEARCH STUDY

Humeral Elevation Angle increase in the group mean value

between 90 and 140".
Muscle 0° 90' 140' Between-subject variability in
- - -
X SD X SD X SD scapular upward rotation values has
Levator scapulae 1 1 7 7 16 20 frequently been noted in the 2-D lit-
Upper trapezius 1 1 25 10 34 13 erature (14,19). The magnitude of
Lower trapezius 0 1 20 7 23 11 variability in the present study, with
Serratus anterior 1 1 21 10 40 16 standard deviations ranging from 4 to
TABLE 4. Summary data (f,SDI of electromyographic activity for scapulothoracic muscles in percent 6", is quite consistent with the re-
maximum voluntary isometric contraction. ported 2-D values. In the previous
3-D literature (26,31,34,55), between-
subject variability in scapular angles
at a given humeral angle is often not
-- ----- -
addressed. In the present study, the
Kinematic Variable culharn peat Laumann McQuade Ludewig et a1 magnitude of between-subject vari-
(10) (34) (36) (presentstudy) ability at all humeral angles tested
Upward rotation 1.3 3.0 (23) -1.7 (25.7) 2 (26) was greatest for the scapular internal
Internal rotation 30.2 30.8 37.4 (25.9) 33 (29) rotation angle, with standard devia-
Downloaded from www.jospt.org at on March 29, 2021. For personal use only. No other uses without permission.

Tipping* -9.0 -18.9 (210) -20.6 (25.2) -8 (24) tions ranging between 6 and 9". This
* Negative values indicate anterior tipping. suggests that scapular plane orienta-
TABLE 5. Mean angular values (293, when available) for scapular orientation at rest or at a 0"humeral
tion is quite variable between s u b
elevation angle. ject.., although at rest, it averages very
close to the 50" position previously
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

defined in the literature (27). It

72% of the subjects followed a pat-
tern of increased activity across hu-
meral elevation angles and 28% pla-
teaued at the 90" activity level.
Results of the repeated measures
ANOVA revealed a significant main
effect of humeral angle ( p < .001)
for each of the four EMG variables.
Journal of Orthopaedic & Sports Physical Therapy®
measures. When attempting to com-
pare abnormal kinematics or muscle
activity in a patient population, the
expected variability from the mean
among asymptomatic subjects is im-
portant to consider. Additionally, par-
ticularly with EMG variables, all
asymptomatic subjects may not follow
should be noted that despite be-
tween-subject differences in the mag-
nitude of scapular orientation values,
the patterns of orientation between
arm positions were quite consistent
among subjects, with 100% ckmon-
strating progressive upward rotation
and posterior tipping and 84% dem-

Follow-up testing of means across hu- the pattern represented by the mean onstrating progressive external rota-
meral angles resulted in significant of the group. Between-subject vari- tion. A lack of progressive upward
differences for all pair-wise compari- ability has been reported to be pre- rotation or posterior tipping in an
sons except for the lower trapezius individual subject could thus be con-
sent but has not been quantified in
values. For this muscle, the mean at a previous shoulder EMG literature sidered abnormal relative to this sam-
0" humeral angle was significantly ple.
analyzing shoulder elevation (5,57).
different from the means at the 90 Despite differences in experimen-
The variability in patterns of
and 140" humeral angles, but the tal design and selection of subject
EMG activity reported in this study
values at the 90 and 140" angles were groups, the results of this study at a
may explain the apparent contradic- 0" humeral elevation angle can be
not significantly different from one
tion in reports of lower trapezius ac- compared with previous studies de-
another.
tivity among preceding investigations. scribing an average rest position of
Previous authors report increasing the scapula. These comparisons are
DISCUSSION activity of the lower trapezius at in- presented in Table 5. Measurements
The results of this study provide creased humeral elevation angles (3, of mean upward rotation and inter-
a clinically relevant description of 3-D 17), decreasing activity after 90" (49). nal rotation angles from the present
scapular orientation and associated or activity changes being dependent study are consistent with the results
muscle activity with humeral eleva- on the plane of elevation (24). In the of these previous investigations. Fur-
tion in the plane of the scapula. Per- present study, patterns of increasing ther, all studies report the scapula to
haps equally important to the under- and decreasing activity were nearly be in an anteriorly tipped position at
standing of normative profiles is the equally present in individuals (48% rest. The magnitude of the mean t i p
variability that is evident between s u b increased, 52% decreased), conse- ping angle measured in the present
jects for both kinematic and EMG quently resulting in no significant study is consistent with the findings

Volume 24 Number 2 August IN6 JOSPT

 of Culham and Peat (10). Both Lau-
mann and McQuade define the t i p
ping angle relative to different planes
or axes than the present study, and,
therefore, absolute comparisons of
this angle between these studies are
not appropriate (34.36).
Comparisons of this study's re-
sults to previous studies of 3-D scapu-
lar orientation during humeral eleva-
tion are constrained by further
complicating factors. Differences in
instrumentation (31,34,36), planes of
analysis (26,31,34,55), loading condi-
tions (36), and definitions of axis
orientations and determination of
angular values prevent the direct
comparison of magnitudes of angular
Downloaded from www.jospt.org at on March 29, 2021. For personal use only. No other uses without permission.
orientations. Nonetheless, the gen-
eral patterns of scapular rotations as
the humerus is elevated can be com-
pared. All previous studies describe
progressive upward rotation and pos-
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
terior tipping with increasing hu-
meral elevation. The results of the
present study demonstrate that these
same patterns are present during hu-
meral elevation in the scapular plane.
The description of scapular rotations
in the plane of the scapula as re-
ported in the present study allows a
closer approximation to movements
Journal of Orthopaedic & Sports Physical Therapy®
present during unconstrained func-
tional elevation. Additionally, the
progressive external rotation of the
scapula observed in the present study
supports Van Der Helm and Pronk's
contention that the scapular plane is
steadily moving throughout humeral
elevation (55). Kondo et al and
McQuade, however, describe a more
constant scapular plane orientation
during humeral elevation (31,36).
Altered patterns of upward rota-
tion of the scapula have been hypoth-
esized to contribute to shoulder
problems. Synchronous upward rota-
tion of the scapula as the arm is ele-
vated is believed necessary to main-
tain an appropriate length tension
relationship for the deltoid. Alter-
ations in this relationship may impact
power production and total range of
elevation in all planes (24). Addition-
JOSPT Volume 24 Number 2 August 1996
ally, Poppen and Walker assessed
scapulohumeral ratios of both symp
tomatic and asymptomatic subjects
during 2-D scapular plane abduction
(45). These authors concluded that
abnormal ratios were associated with
disease, but normal ratios did not
rule out disease (45). Decreases from
the normal upward rotation have also
been theorized to decrease the s u b
acromial space and contribute to im-
pingement (20.42).
In addition to changes in upward
rotation, abnormalities in scapular
tipping patterns warrant increased
attention in patients with impinge-
ment symptoms. Flatow et al, based
on simulated humeral elevation in
-- -- --------
The scapula
demonstrafed a
pa ffern of progressive
upward rotation,
decreased internal
rotation, and
movement from
an anteriorly
to a posteriorly
tipped position as
humeral elevation
angle increased.
- - -
cadaver specimens, described a sub-
acromial contact pattern located be-
neath the anterior acromial surface
(18). Decreases in the normal pat-
tern of movement from anterior to
posterior tipping would increase the
proximity of the anterior acromion
to the rotator cuff tendons as the ten-
dons are attempting to pass beneath
this surface when the arm is elevated.
Subsequently, a lack of normal poste-
rior tipping may predispose an indi-
vidual to impingement or exacerbate
preexisting impingement symptoms.
Instability of the shoulder has
also been related to abnormal scapu-
lar kinematics of both upward rota-
tion and internal rotation angles.
Ozaki demonstrated a significant de-
crease in scapular upward rotation
values during humeral elevation in
patients with involuntary inferior and
multidirectional instability as com-
pared with asymptomatic volunteers
(41). In addition, Kibler hypothe-
sized that excess "antetilt" of the gle-
noid during a throwing motion may
increase stress on the anterior gleno-
humeral structures and predispose
the subject to glenoid labral tears
and subsequent anterior instability
(30). This excessive "antetilted" ori-
entation would be defined as exces-
sive internal rotation, using the ter-
minology of the present study.
If an abnormal scapular motion
pattern is considered contributory to
a clinical pathology, the rehabilita-
tion approach should consider mus-
cle forces that may restrict or en-
hance the desired scapular kine-
matics. As the pectoralis minor in-
serts into the coracoid process, excess
active or passive tension in this mus-
cle may impede normal posterior t i p
ping. Likewise, normal upward rota-
tion may be restricted by excess
tension in the rhomboids or levator
scapulae.
Electromyographic descriptions
of trapezius and serratus anterior ac-
tivities during humeral elevation have
commonly been related to the 2-D
kinematic patterns of scapular u p
ward rotation (2,15,24,44). Little dis-
cussion has related muscle activity to
the patterns of tipping and external
rotation that are also occurring with
humeral elevation. Due to its inser-
tion into the medial border and infe-
rior angle of the scapula, the progres-
sive activity seen in the serratus
anterior with humeral elevation
should contribute to posterior t i p
ping and external rotation of the
scapula as well as upward rotation.
The role of the trapezius in contrib
 RESEARCH STUDY
uting to scapular tipping and inter-
nal/external rotation is more difficult
to visualize. However, the downward
and medially directed fibers of the
lower trapezius may also assist in pro-
ducing posterior tipping and external
rotation moments as the humerus is
elevated. Alterations in patterns of
EMG activity from the scapular mus-
cles, which have been found in some
symptomatic subject groups (21,43,
52). may impact the normal rotations
of the scapula about all three axes as
the humerus is elevated. Consequent-
ly, rehabilitation approaches to scap
ular muscle strengthening should
consider the effect.. on tipping and
internal/external rotation motions as
Downloaded from www.jospt.org at on March 29, 2021. For personal use only. No other uses without permission.
well as upward rotation.
Limitations should be noted
when interpreting the surface EMG
signal describing levator scapulae ac-
tivity. The levator scapulae is superfi-
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
cially located at the level of the fifth
cervical vertebrae between the poste-
rior margin of the sternocleidomas-
toid and the anterior margin of the
upper trapezius. For subject.. with
small necks or poorly defined muscu-
lature, this area was somewhat diffi-
cult to locate. The levator scapulae
EMG signal may have been influ-
Journal of Orthopaedic & Sports Physical Therapy®
enced by cross talk from the upper
trapezius in some subjects, although
the overall low level of activity ob-
served from the levator suggests no
substantial impact of upper trapezius
cross talk on the mean values. Al-
though collected in a similar fashion
in previous studies (3,43), the acqui-
sition of serratus anterior EMG data
with surface electrodes also has limi-
tations. However, the normalization
of the data minimizes the influence
of electrode location and allows an
adequate representation of this mus-
cle's activity.
A complete description of muscle
function of the scapular rotators is
contingent upon a thorough under-
standing of 3 D scapular kinematics.
Moments created by these muscles
are dependent upon their line of ac-
tion and the axis of rotation of the
scapula, which has yet to be de-
scribed using a 3-D model. In addi-
tion, the functional roles of these
muscles will change substantially
throughout the range of shoulder
elevation due to constant changes in
the axis of rotation and lines of ac-
tion (54). Objective assessment of
scapular upward rotation alone may
not provide adequate information
toward understanding pathologies of
the shoulder that are believed related
to mechanisms of abnormal kinemat-
ics and muscle activity. A comprehen-
sive kinematic assessment of the
shoulder should include scapular t i p
ping and internal/external rotation
measurements as well.
SUMMARY AND CONCLUSIONS
This study describes the 3-D ori-
entation of the scapula and corre-
sponding EMG activity of the scapu-
lar rotators during humeral elevation
in the scapular plane. The scapula
demonstrated a pattern of progres-
sive upward rotation, decreased inter-
nal rotation, and movement from an
anteriorly to a posteriorly tipped posi-
tion as humeral elevation angle in-
creased. Electromyographic activity of
the levator scapulae, upper trapezius,
lower trapezius, and serratus anterior
also increased progressively with in-
creased humeral elevation angle. The
secondary rotations of the scapula
(tipping and internal/external rota-
tion) are believed to be clinically sig-
nificant contributors to abnormal
kinematics in shoulder dysfunction.
Investigation of these accessory rota-
tions in clinical populations is war-
ranted. In addition, further 3-D kine-
matic analysis of the shoulder com-
plex is needed in combination with
EMG data to enhance our under-
standing of shoulder muscle func-
tion. IOSPT
ACKNOWLEDGMENTS
The authors would like to thank
Dr. Gary Smidt and Dr. Rich Shields
for their input at various stages of
this project.
REFERENCES
1. An KN, Browne AO, Korinek S, Tanaka
S, Morrey BF: Three-dimensional kine-
matics of glenohumeral elevation. J Or-
thop Res 9:143-149, 1991
2. Bagg SD, Forrest WJ: A biomechanical
analysis of scapular rotation during arm
abduction in the scapular plane. Am 1
Phys Med Rehabil67:238-245, 1988
3. Bagg SO, Forrest WJ: Electromyo-
graphic study of the scapular rotators
during arm abduction in the scapular
plane. Am J Phys Med Rehabil65:lll-
124, 1986
4. Behrsin JF, Macguire K: Levator scapu-
lae action during shoulder movement:
A possible mechanism for shoulder
pain of cervical origin. Aust J Phys-
iother 32: 7 0 1- 106, 1 986
5. Boublik M, Hawkins RJ: Clinical exam-
ination of the shoulder complex. J Or-
thop Sports Phys Ther 18(1):379-385,
1993
6. Bull ML, De Freitas V, Vitti M : Electro-
myographic study of the trapezius (pars
superior) and serratus anterior (pars in-
ferior) in free movements of the arm.
Anat Anz 17 1:125- 133, 1990
7. Cathcart CW: Movements of the shoul-
der girdle involved in those of the arm
on the trunk. ] Anat Physiol 18:211-
218, 1884
8. Codman EA: The Shoulder, Boston,
MA: Thomas Todd, 1934
9. Craig JJ: Introduction to Robotics: Me-
chanics and Control (2nd Ed), pp 19-
67. New York: Addison-Wesley Com-
pany, 1989
10. Culham E, Peat M : Functional anatomy
of the shoulder complex. J Orthop
Sports Phys Ther 18(1):342-350, 1993
11. De Freitas V, Vitti M, Furlani J: Electro-
myographic study of levator scapulae
and rhornboideus major muscles in
movements of the shoulder and arm.
Electromyogr Clin Neurophysiol 20:
205-2 16, 1 980
12. Delagi EF, lazelti J, Perotta A, Morrison
D: Anatomic Guide for the Electromyo-
grapher (2nd Ed), pp 106-107. Spring-
field, IL: Charles C. Thomas, 1980
13. Dempster WT: Mechanisms of shoul-
der movement. Arch Phys Med Rehabil
46A:49-70, 1965
14. Doody SG, Freedman L, Waterland JC:
Shoulder movements during abduction
in the scapular plane. Arch Phys Med
Rehabil5 1.595-604, 1970
15. Dvir Z, Berme N: The shoulder complex
in elevation of the arm: A mechanism
approach. J Biomech 11:2 19-225, 1978
16. Engin AE, Chen SM: Statistical data
Volume 24 Number 2 0 August 1996 JOSPT

base for the biomechanical properties
of the human shoulder complex-I. Ki-
nematics of the shoulder complex.
J Biomech Eng 108:215-22 1, 1986
17. Filho JG, Furlani J, De Freitas V: Elec-
tromyographic study of the trapezius
muscle in free movements of the arm.
Electromyogr Clin Neurophysiol 3 1 :
93-98, 1991
18. Flatow EL, Soslowsky LJ, Ticker JB,
Pawluk RJ, Hepler M, Ark I, Mow VC,
Bigliani LU: Excursion of the rotator
cuff under the acromion. Patterns of
subacromial contact. Am J Sports Med
22(6):779-788, 1 994
19. Freedman L, Munro RR: Abduction of
the arm in the scapular plane: Scapular
and glenohumeral movements. 1 Bone
Joint Surg 48A: 1503- 15 10, 1966
20. Fu FH, Harner CD, Klein AH: Shoulder
impingement syndrome: A critical re-
view. Clin Orthop 269: 162- 173, 1991
Downloaded from www.jospt.org at on March 29, 2021. For personal use only. No other uses without permission.
2 1. Glousman R, Jobe F, TiboneJ, Moynes
D, Antonelli D, Perry J: Dynamic elec-
tromyographic analysis of the throwing
shoulder with glenohumeral instability.
] Bone Joint Surg 70A:220-226, 1988
22. Greenfield B, Catlin PA, Coats PW,
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
Green E, McDonald 11, North C: Pos-
ture in patients with shoulder overuse
injuries and healthy individuals. Or-
thop Sports Phys Ther 2 1(5):287-295,
1995
23. Hogfors C, Peterson B, Sigholm G, Her-
berts P: Biomechanical model of the
human shoulder joint-11. The shoulder
rhythm. 1 Biomech 24:699-709, 1991
24. lnman VT; Saunders JB, Abbott LC: Ob-
servations on the function of the shoul-
Journal of Orthopaedic & Sports Physical Therapy®
der joint. J Bone ] Surg 26A: 1-30, 1944
25. Johnson GR, Anderson JM: Measure-
ment of three-dimensional shoulder
movement by an electromagnetic sen-
sor. Clin Biomech 5: 13 1- 136, 1 990
26. Johnson GR, Stuart PR, Mitchell S: A
method for the measurement of three-
dimensional scapular movement. Clin
Biomech 8:269-273, 1993
27. Johnston TB: The movements of the
shoulder joint. A plea for the use of the
plane of the scapula as the plane of
reference for movements occurring at
the humero-scapularjoint. Br J Surg 25:
252-260, 1937
28. Kendall HO, Kendall FP: Muscles Test-
ing and Function, Baltimore: Williams
& Wilkins, 1949
29. Kent BE: Functional anatomy of the
shoulder complex: A review. Phys Ther
5 1 :867-887, 1971
30. Kibler WB: Role of the scapula in the
JOSFT Volume 24 Number 2 August 1996
overhead throwing motion. Contemp
Orthop 22525-532, 1991
31. Kondo M, Tazoe S, Yamada M:
Changes of the tilting angle of the scap-
ula following elevation of the arm. In:
Bateman JE, Wehh RP (eds), Surgery of
the Shoulder, pp 12-1 6. Philadelphia:
B.C. Decker lnc., 1984
32. Lahey MA, Downey RG, Saal FE: Intra-
class correlations: There's more than
meets the eye. Psycho1 Bull 93586-
595, 1983
33. Lannersten L, Harms-Ringdahl K: Neck
and shoulder muscle activity during
work with different cash register sys-
tems. Ergonomics 33:49- 65, 1990
34. Laumann U: Kinesiology of the shoul-
der joint. In: Kolbel R, Helbig B, Blauth
W, (eds), Shoulder Replacement, pp
23-3 1. Berlin: Springer-Verlag, 1 987
35. Ludewig PM: The effect of head posi-
tion on scapular rotation and muscle
activity during humeral elevation. Un-
published master's thesis, University of
lowa, lowa City, IA, 1994
36. McQuade KJ: The scapulohumeral
rhythm: A three-dimensional kinematic
analysis of the effects of load and fa-
tigue during elevation of the arm in the
scapular plane. Unpublished doctoral
thesis, University of lowa, lowa City,
IA, 1994
37. McQuade KJ, Wei SH, Smidt GL: Ef-
fects of local muscle fatigue on three-
dimensional scapulohumeral rhythm.
Clin Biomech 10:144-148, 1995
38. Miller NE, Merritt JL: Don't forget the
serratus. J Am Paraplegia Soc 13:46-
47, 1990
39. Money BF, An KN: Biomechanics of the
shoulder. In: Rockwood CA, Matsen FA
(eds), The Shoulder, pp 208-245. Phila-
delphia: W.B. Saunders Company, 1990
40. Norkin CC, Levangie PK: Joint Structure
and Function: A Comprehensive Anal-
ysis, pp 2 15-2 16. Philadelphia: F.A.
Davis Company, 1992
4 1. Ozaki J: Glenohumeral movements of
the involuntary inferior and multidirec-
tional instability. Clin Orthop 238: 107-
1 1 1, 1989
42. Paine RM, Voight M: The role of the
scapula. / Orthop Sports Phys Ther
18(1):386-391, 1993
43. Peat M, Grahame RE: Electromyo-
graphic analysis of soft tissue lesions
affecting shoulder function. Am J Phys
Med 56:223-240, 1977
44. Perry J: Normal upper extremity kinesi-
ology. Phys Ther 58:265-278, 1978
45. Poppen NK, Walker PS: Normal and
RESEARCH STUDY
abnormal motion of the shoulder.
J Bone Joint Surg 58A: 195-20 1, 1976
46. Portney LC, Watkins MP: Foundations
of Clinical Research: Applications to
Practice, pp 505-528. Norwalk, CT:
Appleton & Lange, 1993
47. Pronk GM, Van Der Helm FCT: The
palpater: An instrument for measuring
the positions of bones in three dimen-
sions. J Med Eng Techno1 l5:15-20,
1991
48. Saha AK: Theory of Shoulder Mecha-
nism: Descriptive and Applied, pp 29-
53. Springfield, IL: Charles C. Thomas,
1961
49. Saha AK, Das NN, Chakravarty BG:
Studies on electromyographic changes
of muscles acting on the shoulder joint
complex. Calcutta Med 1 53:409-4 13,
1956
50. Schuldt K: On neck muscle activity and
load reduction in sitting postures. An
electromyographic and biomechanical
study with applications in ergonomics
and rehabilitation. Scand J Rehabil
Med 19(S~ppl): 1-49, 1 988
51. Schuldt K, Harms-Ringdahl K: Activity
levels during isometric test contractions
of neck and shoulder muscles. Scand J
Rehabil Med 20: 1 17- 127, 1988
52. Scovazzo ML, Browne A, Pink M, lobe
FW, Kerrigan 1: The painful shoulder
during freestyle swimming: An electro-
myographic cinematographic analysis
of twelve muscles. Am J Sports Med
19577-582, 1991
53. Sundelin G, Hagberg M: The effects of
different pause types on neck and
shoulder EMG activity during VDU
work. Ergonomics 32527-537, 1989
54. Van Der Helm FCT: Analysis of the
kinematic and dynamic behavior of the
shoulder mechanism. J Biomech 27:
527-550, 1994
55. Van Der Helm FCT, Pronk GM: Three-
dimensional recording and description
of motions of the shoulder mechanism.
J Biomech Eng 1 l7:27-4O, 1995
56. Wei SH, McQuade KJ, Smidt GL:
Three-dimensional joint range of mo-
tion measurements from skeletal coor-
dinate data. J Orthop Sports Phys Ther
18(1):687-691, 1993
57. Yamshon LJ, Bierman W: Kinesiologic
electromyography. 11. The trapezius.
Arch Phys Med Rehabil 29:647-651,
1 983
58. Zipp P: Recommendations for the stan-
dardization of lead positions in surface
electromyography. Eur J Appl Physiol
50:4 1-54, 1 982