-
journal of orthopaedic & sports physical therapy | volume 40
| number 10 | october 2010 | 633
[ research report ]
Shoulder pain is the third most common musculoskeletal problem,
accounting for at least 21% of all musculoskeletal complaints.36
The most common cause of shoulder pain is rotator cuff disease
(RCD), which includes impingement
syndrome, bursitis, tendinitis, tendinosis, and partial- and
full-thickness tears.37 Upper quadrant postural deviations have
been linked to upper
NitiN Kalra, PT, MS1 Amee L. Seitz, PT, PhD, OCS2 N. DougLAS
BoArDmAN iii, MD3 Lori A. micheNer, PT, PhD, ATC, SCS4
1 Orthopaedic Physical Therapist, Select Physical Therapy,
Fairfax, VA; Masters student [at time of study], Department of
Anatomy and Physical Therapy, Virginia Commonwealth University,
Richmond, VA. 2 Research Associate, Department of Physical Therapy,
Virginia Commonwealth University, Richmond, VA. 3 Associate
Professor, Department of Orthopedic Surgery, Virginia Commonwealth
University Health Systems, Richmond, VA. 4 Associate Professor,
Department of Physical Therapy, Virginia Commonwealth University,
Richmond, VA. This study was completed as partial fulfillment for a
Masters Degree in Anatomical Sciences from Virginia Commonwealth
University. The study was approved by The Institutional Review
Board at Virginia Commonwealth University. Funding for this study
was provided partially by the A. D. Williams Fund for Research at
Virginia Commonwealth University, the Department of Physical
Therapy at Virginia Commonwealth University, and by the Foundation
for Physical Therapy. Address correspondence to Dr Lori A.
Michener, Department of Physical Therapy, 1200 E Broad St,
Richmond, VA, 23298. E-mail: [email protected]
Effect of Posture on Acromiohumeral Distance With Arm Elevation
in
Subjects With and Without Rotator Cuff Disease Using
Ultrasonography
t StuDY DeSigN: Controlled laboratory study.t oBJectiVeS: To
examine the effects of altering posture on the subacromial space
(SAS) in sub-jects with rotator cuff disease and subjects without
shoulder pain.
t BAcKgrouND: Poor upper quadrant posture has been linked to
altered scapular mechanics, which has been theorized to excessively
reduce SAS. However, no study has examined the direct effects of
altering upper quadrant posture on SAS. We hypothesized that
upright posture would increase and slouched posture would decrease
the SAS, as compared to a normal posture, when measured both with
the shoulder at rest along the side of the trunk and when
maintained in 45 of active shoulder abduction.
t methoDS: Participants included 2 groups: the subjects with
shoulder pain and rotator cuff disease, as diagnosed via magnetic
resonance imaging (n = 31), and control subjects without shoulder
pain (n = 29). The SAS was imaged with ultrasound using a 7.5-MHz
linear transducer placed in the coronal plane over the posterior to
midportion of the acromion. The SAS was mea-sured on ultrasound
images using the acromio-humeral distance (AHD), defined as the
shortest
distance between the acromion and the humerus. The AHD was
measured in 2 trials at 2 arm angles (at rest along the trunk and
at 45 of active abduc-tion) and across 3 postures (normal,
slouched, and upright), and averaged for data analysis.
t reSuLtS: Two mixed-model analyses of variance, 1 for each arm
angle, were used to compare AHD across postures and between groups.
There was no interaction between group and posture, and no
sig-nificant main effect of group for either arm position. There
was no significant main effect of posture for the arm at rest (P =
.26); however, there was a significant main effect of posture on
AHD at the 45 abduction arm angle (P = .0002), with a significantly
greater AHD in upright posture (mean AHD, 9.8 mm), as compared to
normal posture (mean AHD, 8.6 mm).
t coNcLuSioN: The effect of posture on SAS, as measured by the
2-dimensional AHD using ultra-sound of the posterior to middle
aspect of the SAS, is small. The AHD increased with upright posture
by 1.2 mm compared to normal posture, when the arm was in 45 active
abduction. J Orthop Sports Phys Ther 2010;40(10):633-640.
doi:10.2519/jospt.2010.3155
t KeY WorDS: impingement, posture, rotator cuff, shoulder,
subacromial space
extremity impairments8,20,22 and RCD.16,18,21,25,27 A forward
head pos-ture has been linked to pain related to shoulder
overuse.16 Adopting a slouched posture has been shown to decrease
gle-nohumeral abduction strength,20 while adopting an upright
posture resulted in increased glenohumeral elevation.8,22 Slouched
posture may limit shoulder motion due to impingement beneath the
acromion, creating a mechanical block to shoulder elevation coupled
with tissue impingement.11,27,28 Taping applied to the posterior
trunk parallel to the thoracic spine and over the scapula of
patients with shoulder pain increased thoracic extension, reduced
pain with shoulder elevation, and improved resting scapular
position.22 Slouched posture is linked to shoulder pain, changes in
scapular posi-tion, shoulder strength, and range of mo-tion, which
may contribute to disability.
Mechanistically, upper quadrant pos-ture can affect scapular
motion or ki-nematics,9,12,20,21 which may reduce the subacromial
space (SAS) and contribute to shoulder pain and the development of
RCD.6,16,21,23,24,27 Increased thoracic spine kyphosis or slouched
thoracic posture has been shown to decrease scapular upward
rotation,20 posterior tilting,9,12,20 and ex-ternal
rotation.9,20,21 Patients with RCD,
40-10 Kalra.indd 633 9/21/10 2:38 PM
-
634 | october 2010 | volume 40 | number 10 | journal of
orthopaedic & sports physical therapy
[ research report ]
est distance between the acromion and humerus in the SAS.
Studies using ultra-sound imaged the SAS over the middle SAS
space,2,3 or did not state the exact lo-cation of the ultrasound
transducer dur-ing imaging.14 Ultrasound as an imaging modality is
less costly and more practi-cal than MRI, and has established
con-current validity with radiographic AHD measures (r =
0.77-0.85).2,3
The link between upper quadrant posture, shoulder pain, and RCD
is not well understood. Investigating the effects of posture on the
SAS of patients with RCD will expand our understanding of possible
mechanisms of posture associ-ated with RCD. Moreover, it may
provide mechanistic evidence for postural correc-tion often used to
treat these patients in rehabilitation. The purpose of this study
was to determine the effect of posture on SAS, using ultrasound
images of the SAS to measure the AHD in 2 arm positions across 3
postures, in subjects with and without RCD. We hypothesized that in
both groups slouched posture would lead to a reduction in SAS and
upright posture would increase the SAS when compared to normal
posture, at both arm positions of rest and 45 of active
abduction.
methoDS
Subjects
Subjects with shoulder pain and RCD (n = 31) and a control group
without shoulder pain (n = 29) were
recruited through the physician offices at the University Health
System and with flyers posted at the Virginia Common-
wealth University Health System and Vir-ginia Commonwealth
University campus (tABLe 1). The Institutional Review Board at the
Virginia Commonwealth Univer-sity approved the study protocol and
the informed consent form, which was signed prior to the data
collection procedures.
Inclusion criteria for the RCD group were (a) an age of at least
18 years, (b) shoulder pain, (c) MRI confirmation of RCD
(tendinosis, bursitis, impinge-ment, or partial- or full-thickness
rota-tor cuff tear), and (d) ability to lift the arm up to 90 of
elevation, which would ensure their ability to easily obtain and
easily maintain the 45 arm abduction test position. If a subject
had 2 or more diagnoses, the more severe RCD pathol-ogy diagnosis
was used to classify group membership. Inclusion criteria for the
control group subjects were an age of at least 18 years and having
no pain and no known previous or current shoulder pa-thology.
Exclusion criteria for all subjects were (a) a cervical range of
motion that reproduced shoulder pain, (b) pain below the elbow
indicative of cervical or nerve pathologies, (c) past shoulder
surgery, and (d) glenohumeral joint arthritis, as indicated in the
MRI report.
Sample size was determined from a pilot study,26 using AHD
measures ob-tained from ultrasound images taken on 10 subjects with
RCD and 10 subjects without shoulder pain. Two ultrasound images
were collected with the arm at rest and 45 active abduction, with
the subject in a neutral posture. This proce-dure was subsequently
repeated 15 to 30 minutes later. The same methods were
specifically, impingement syndrome, have demonstrated these same
altered scapular kinematic patterns.23 These al-tered scapular
kinematics may decrease the SAS and clearance of the humeral head
beneath the acromion during arm elevation.6,7,13,34,35 Studies
provide some support for this relationship. Artificially induced
scapular protraction, compared to a position of scapular
retraction, de-creased the 2-dimensional (2-D) linear distance
between the acromion and hu-merusthe acromiohumeral distance
(AHD)as measured on magnetic reso-nance images (MRIs).35 Another
study found a reduced AHD measured on ultra-sound images in tennis
players without shoulder pain but with scapular dyskine-sis as
compared to those without scapular dyskinesis.34
Direct effects of posture on SAS for individuals with RCD are
unknown. Subjects with acquired or idiopathic thoracic
hyperkyphosis have a smaller AHD as measured on radiographic images
compared to nonkyphotic sub-jects.17 However, this study used
subjects without shoulder pain and measured AHD with the arm at
rest only. More-over, slouched posture is not analogous to
hyperkyphosis, thereby limiting the generalizability of these
results. Studies examining the direct effects of posture on SAS in
patients with RCD are need-ed to elucidate the relationship between
posture and RCD.
Generally, the AHD is smaller in pa-tients with RCD as compared
to healthy shoulders.1,3,14,15,19 Studies using MRI re-ported a
smaller AHD in patients with impingement syndrome as compared to
healthy shoulders.1,15,19 Studies using ultrasound imaging
corroborate these findings with a smaller AHD in patients with
RCD3,14 and a smaller AHD with in-creased severity of the RCD.3
Although these studies consistently indicated a smaller AHD with
RCD, they varied on the location of the measurement of the SAS. MRI
studies1,15,19 depicted the loca-tion as the anterior to middle
aspect of the SAS, or at the location of the small-
tABLe 1 Descriptive Statistics*
Abbreviation: RCD, rotator cuff disease.* Data presented as mean
SD (range).
Subjects With rcD (n = 31) control Subjects (n = 29)
Age (y) 53.5 13.7 (20.0-80.0) 31.9 10.7 (23.0-62.0)
Height (cm) 170.2 9.6 (152.4-187.9) 169.8 8.8 (151.1-1.9)
Body mass (kg) 80.5 13.4 (60.2-115.1) 72.1 13.7 (56.6-113.3)
Pain (mo) 18.2 17.3 (2.0-84.0) 0.0 0.0 (0.0-0.0)
40-10 Kalra.indd 634 9/21/10 2:38 PM
-
journal of orthopaedic & sports physical therapy | volume 40
| number 10 | october 2010 | 635
used as described in this study, using the same investigator to
collect data and make AHD measurements. For the rest and 45 arm
positions, respectively, the standard error of the measure (SEM)
was 2.3 and 2.5 mm. Based on these data, we calculated our desired
sample size using a desired minimal difference of 2.5 mm between
posture conditions. The average standard deviation was
conservatively es-timated at 4.0 mm. For a power of 90% and an
alpha of .05, sample size calcu-lations indicated 20 subjects per
group. As data collection ensued, the majority of subjects in the
RCD group had rota-tor cuff tears; therefore, recruitment was
extended to purposefully recruit subjects with impingement to
represent the entity of RCD.
ProceduresAfter signing the informed consent form, subjects
completed an intake form and the American Shoulder and Elbow
Sur-geons patient self-report shoulder score to assess shoulder
functional loss and dis-ability (tABLe 2). Next, ultrasound images
of the SAS were collected for AHD mea-surement. The examiner was
not blinded to group assignment to conduct the ultra-sound imaging,
but was blinded for AHD measurements.Subacromial Space Measurement
The outlet of the SAS was measured on ul-trasound-generated 2-D
images via the AHD. The AHD is a 2-D linear measure defined as the
shortest distance between the acromion and the humerus (Figure 1).
An ultrasound unit (The Pyramid 764; Pyramid Management LLC, Los
Alami-tos, CA) with a 7.5-MHz linear ultrasound transducer was
utilized. Placement of the ultrasound transducer was standardized,
with its location on the posterior to mid-dle portion of the
acromion in the coronal plane, with the transducer placed parallel
to the flat superior aspect of the acromion so that both the
acromion and humerus were visualized (Figure 2). All ultrasound
images were saved on a computer for AHD measurements performed
later. Two ultrasound images were taken for
each of 3 postures and 2 arm positions in each posture.Posture
and Arm Angle Subjects were tested in 3 sitting postures: (1)
normal resting posture, (2) slouched posture, and (3) upright
posture with scapular re-traction. For normal posture, the subject
was asked to sit in a chair with the back supported, feet flat on
the floor, hips and
knees at 90 of flexion, head and shoul-der in their habitual
posture, looking straight ahead (Figure 3). Slouched pos-ture was
achieved by having each subject move forward in the chair so that
the subjects back was a minimum of 15 cm away from the back
support, then slump forward and down to attain a flexed tho-racic
and lumbar spine, forward head, and rounded shoulder posture
(Figure 3). For the slouched posture position, subjects were
instructed to slouch down but look straight ahead, to get flexion
in the thoracic and lower cervical spine, ex-tension in the upper
cervical spine, and
Figure 1. The line represents the acromiohumeral distance
(AHD).
tABLe 2 Subject Demographics
Abbreviations: ASES, American Shoulder and Elbow Surgeons
Self-Report Form; RCD, rotator cuff disease.* Mean SD (range), 0-50
points, with 50 as no pain. Mean SD (range), 0-50 points, with 50
as no functional loss. Mean SD (range), 0-100 points, with 100 as
no pain and functional loss.
Subjects With rcD (n = 31) control Subjects (n = 29)
Gender (n)
Male 11 14
Female 20 15
Dominant shoulder (n)
Right 22 27
Left 5 2
Ambidextrous 4 0
RCD diagnosis (n)
Impingement 15
Partial-thickness tear 9
Full-thickness tear 7
ASES pain score* 31.3 11.8 (10.0-50.0) 50.0 0.0 (50.0-50.0)
Function score 25.2 10.3 (5.0-45.0) 50.0 0.0 (50.0-50.0)
Total score 56.9 17.8 (28.0-95.0) 100.0 0.0 (100.0-100.0)
Figure 2. Ultrasound probe positioning on the acromion.
40-10 Kalra.indd 635 9/21/10 2:38 PM
-
636 | october 2010 | volume 40 | number 10 | journal of
orthopaedic & sports physical therapy
[ research report ]
elevation. Moreover, measurement of the SAS at angles greater
than 45 may not provide clinically relevant information because,
above 35 to 40 of glenohumer-al elevation, the supraspinatus tendon
has likely already passed underneath the acromion and may no longer
be at risk of impingement in the SAS.4
Each ultrasound image was an indi-vidual measure. Prior to each
ultrasound image, the subjects rolled their shoulders and moved
about to change their pos-ture, then were asked to assume the
des-ignated posture and reposition the arm in the appropriate arm
angle. Between measures at the 45 arm angle, subjects took their
arm out of the sling and then repositioned the arm for the second
ul-trasound image. The sequence of pos-tural alterations was
counterbalanced by changing the posture sequence, but the resting
arm position was always tested prior to the 45 angle. Images were
saved to the computer, and retrieved later for AHD measurement.AHD
Measurement The examiner who captured the ultrasound images also
measured the AHD. Ultrasound images were randomly retrieved, with
the ex-aminer blinded to group, arm angle, and posture. The AHD was
measured using the software Universal Desktop Ruler (AVP Inc,
Voronezh, Russia) to measure distances on screen. A mark was first
placed at the most inferior aspect of the acromion, then a second
mark was placed at the humeral head to measure the
shortest distance between the acromion and humerus. The 2 AHD
measures at each arm angle for each posture were av-eraged for data
analysis.
Data AnalysisTo examine the effect of posture on AHD, 2 separate
mixed-model analy-ses of variance (ANOVAs), 1 for each arm position
(rest and 45 abduction), were performed. These included effects for
group (RCD and control), posture (normal, upright, slouched), and
group-posture interaction, with posture as the repeated factor.
Post hoc testing was per-formed using contrasts. All analyses used
a significance set at = .05. Test-retest intrarater reliability was
calculated using an intraclass correlation coefficient (ICC) 2-way
random analysis for each arm po-sition. Error was calculated using
SEM (SD 1ICC) and minimal detectable change (MDC) (SEM 2) values.
The 90% confidence bounds were calculated by multiplying error
values by the z score of 1.64.
reSuLtS
the AHD means and standard deviations are reported in tABLe 3.
Test-retest reliability analysis on
60 subjects revealed that for the rest and 45 abduction
positions, respectively, ICCs (2-way random) were 0.92 (95% CI:
0.87, 0.95) and 0.76 (95% CI: 0.63, 0.86), SEMs were 0.9 and 1.6
mm, and MDCs were 1.3 and 2.2 mm.
There were no significant group-
rounded shoulders. For upright posture, subjects were asked to
sit back against the back rest of the chair with a pillow between
their back and the back support, then were instructed to sit up
straight and pull their shoulders back and look straight ahead, to
achieve retracted shoulders and extension in the thoracic and
cervical spine (Figure 3). For both upright and slouched postures,
subjects altered both the spine and shoulder pos-ture from their
normal resting posture. Therefore, changes in AHD between postures
can result from changes in spine posture (thoracic or cervical
spine) or shoulder posture (humerus or scapula), or a combination
of these.
In all 3 postures, ultrasound images were obtained at 2 arm
angles: at rest with arm at the side and in 45 actively maintained
coronal plane shoulder ab-duction. For the 45 position, the arm was
initially suspended in an adjustable sling, but the arm was held
actively above the sling for imaging (Figure 4). This sling
al-lowed the patient to rest the arm between measurements and
during setup. The subject was instructed to hold the arm up and off
the sling for a few seconds, while the arm angle was verified with
a bubble inclinometer. Then, ultrasound images were captured with
active arm elevation. An elevated arm angle was used because
patients with RCD complain of pain dur-ing shoulder elevation. In
pilot testing, it was technically difficult to obtain ad-equate
ultrasound images above 45 of
Figure 4. Subject with arm supported at 45 abduction.
Figure 3. (A) Normal posture. (B) Slouched posture. (C) Upright
posture.
40-10 Kalra.indd 636 9/21/10 2:38 PM
-
journal of orthopaedic & sports physical therapy | volume 40
| number 10 | october 2010 | 637
0.7, 1.7) with the arm at rest and 0.1 mm (95% CI: 0.8, 0.9) for
the 45 abduc-tion position. There was no significant main effect of
posture with the arm at rest (F2,116 = 1.4, P = .26); however,
there was a statistically significant main effect of posture at 45
of abduction (F2,115.4 = 9.1, P = .0002). Post hoc testing, using
contrasts for the 45 arm abducted po-sition, revealed a
statistically significant difference (t = 3.1, P = .002) of a
greater AHD in the upright posture (mean SD AHD, 9.8 2.0 mm)
compared to nor-mal posture (mean SD AHD, 8.6 1.9 mm), with a mean
difference of 1.2 mm (95% CI: 0.3, 2.0). However, there was no
significant difference (t = 1.6, P = .106) between slouched posture
(mean SD AHD, 9.2 1.9 mm) and normal posture at the 45 abduction
position.
DiScuSSioN
the posture impairment theory links postural deviations with
ana-tomical changes, impairments, and
pain of the shoulder.6 Slouched posture can alter scapular
kinematics9,12,20,21; but this posture is only theoretically linked
to a change in the outlet of the SAS.7,13,35 This study has
demonstrated a link between upper quadrant posture and SAS, with a
small change in the AHD
posture interactions with the arm at rest (F2,116 = 0.4, P =
.658) and 45 abduction (F2,115.4 = 1.0, P = .364). Therefore, the
ef-fect of posture was determined to be in-dependent of group
classification. Means and standard deviations of all subjects are
reported in tABLe 3, and 95% CIs are
represented in Figure 5. There were no statistically significant
main effects of group classification for either the rest (F1,58 =
0.6, P = .431) or 45 abduction (F1,58.1 = 0.04, P = .839)
positions. There-fore, the difference between the control and RCD
groups was 0.5 mm (95% CI:
Slouch Normal Upright
7
8
9
10
11
12
13
14
Ac
rom
iohu
mer
al D
ista
nce
(mm
)
At rest (0) 45 abduction
Figure 5. Acromiohumeral distance in millimeters for 3 postures
and for 2 arm positions (at rest and 45 abduction). Data represent
combined results for the control and patients with rotator cuff
disease. Vertical bars represent 95% confidence intervals.
tABLe 3 Acromiohumeral Distance*
Abbreviations: Abd, shoulder abduction; RCD, rotator cuff
disease.* Data presented as mean SD mm. Subgroup definition based
on magnetic resonance imaging. Significantly different from normal
posture (P = .002).
rest 45 Abd rest 45 Abd rest 45 Abd
All subjects 12.1 2.6 8.6 1.9 12.5 3.1 9.2 1.9 12.6 2.5 9.8
2.0
Groups
Control 11.8 2.5 8.7 1.9 12.2 2.5 9.4 2.0 12.5 2.3 9.6 1.9
RCD 12.5 2.6 8.5 2.0 12.8 3.6 9.0 2.0 12.7 2.6 9.9 2.1
Subgroups of RCD
Impingement 12.8 2.5 8.6 2.4 12.9 4.0 8.3 1.7 12.8 2.4 9.8
1.7
Partial-thickness tear 12.0 2.1 8.6 1.8 13.6 2.6 9.4 1.8 12.8
2.2 11.1 2.5
Full-thickness tear 12.3 3.8 8.4 1.8 11.6 4.2 9.8 2.4 11.8 3.6
8.9 1.7
Normal Posture Slouched Posture upright Posture
Posture and Arm elevation Angle
40-10 Kalra.indd 637 9/21/10 2:38 PM
-
638 | october 2010 | volume 40 | number 10 | journal of
orthopaedic & sports physical therapy
[ research report ]2-D measure in the coronal plane of the
posterior to middle portion of the SAS. Upright posture increased
the AHD, as compared to normal posture, when the arm was actively
held at 45 abduc-tion but not when the arm was at rest. Slouched
posture did not change AHD as compared to normal posture.
We examined the direct effect of posture on the AHD linear
measure in patients with RCD and control subjects at 2 arm angles.
At 45 active abduc-tion, AHD increased by a mean of 1.2 mm with a
change from normal to the upright posture. This supports the
hy-pothesis that upright posture increases AHD. This increase in
AHD may have the effect of relieving the symptoms of compression of
the SAS structures. The meaningful change in AHD, the amount needed
to change patient symptoms and shoulder function is unknown.
Subjects without shoulder pain but with thoracic hyperkyphosis had
1.4 to 1.7 mm smaller mean AHD than those without hyper-kyphosis.17
A study of 4 healthy subjects using MRI images of the midcoronal
plane, which approximated the middle SAS, noted a mean AHD increase
of 0.5 mm (range, 0.3-1.5 mm) with scapular retraction as compared
to protraction.35 In our study, the mean change of 1.2 mm in AHD
with upright posture was great-er than that of the scapular
protraction-retraction study, but smaller than the study of
thoracic hyperkyphosis. The MDC, the distribution-based error for
our AHD measure, was 2.2 mm. A change in AHD of 2.2 mm from normal
to upright posture was experienced by 17 of 60 subjects (28%) in
this study. Although the AHD change with up-right posture is
statistically significant, it was less than the MDC in 72% of the
subjects. The relationship of a 1.2-mm change in AHD to patient
symptoms is unknown. Research is needed to deter-mine the
meaningful amount of change in AHD.
Slouched posture was expected to de-crease AHD when compared to
normal posture in both subjects with RCD and
control. However, our results did not confirm this hypothesis.
What may par-tially explain this lack of difference is the subject
report of difficulty and pain while maintaining their arm at 45 of
abduc-tion when in the slouched posture. Sub-jects might have
elevated their scapula to relieve pain, and this substitution
move-ment might have prevented a reduction in the AHD. Also,
scapular muscle ac-tivity might have been altered with the
postures, and this might have an effect on AHD. We did not monitor
scapular motion or muscle activity.
Upper quadrant posture is a com-bination of thoracic and
cervical spine posture, and shoulder posture of the humerus and
scapula. The com-ponents of upper quadrant posture were not
measured in the 3 postures. Changes in shoulder and spine pos-ture
were inferred with the postures. However, changes in components of
the upper quadrant posture could have been inconsistent across
subjects and, therefore, would explain the lack of dif-ferences
across postures.
The posterior to middle aspect of the SAS was the best position
to obtain the landmarks for AHD measures with ultrasound images.
Prior studies using ultrasound to measure the AHD in indi-viduals
with RCD imaged the SAS over the middle2,3 or anterior10 aspect of
the SAS space, or did not describe location14 of the ultrasound
probe. Findings from these ultrasound studies were generally
consistent with those using MRI to im-age the SAS,1,15,19 which
depicted the AHD measure of the anterior to middle aspect of the
SAS or described it as the smallest distance between the acromion
and hu-merus regardless of location in the SAS. There is evidence
of greater humeral contact on the anterior aspect of the
ac-romion13 and a decrease of the anterior aspect of the SAS with
clinical impinge-ment maneuvers30; however, the purpose of this
study was not to look at contact or absolute values but to examine
the ef-fects of change in AHD with change in posture. Additionally,
no evidence indi-
cates that the anterior aspect differs from the posterior aspect
with respect to AHD change during arm movement or altered postures.
A recent study33 comparing changes in AHD with arm elevation
be-tween the anterior and posterior aspect of SAS indicated that
changes in AHD with arm elevation were not significantly dif-ferent
when AHD was measured at the anterior versus the posterior aspect
of the SAS. AHD was significantly smaller in the anterior aspect of
SAS compared to posterior aspect, but the change in AHD with arm
elevation was not significantly different. However, this does not
exclude the possibility that different effects of posture on AHD
might have occurred if the anterior SAS was imaged. Measure-ment of
the anterior SAS is indicated in future studies.
The AHD measure does not represent the entire SAS, rather, only
a 2-D lin-ear distance of a portion of the outlet of the SAS. AHD
measured on ultrasound images are reliable2,3,10,38 and have
dem-onstrated concurrent validity with ra-diographs (r =
0.77-0.85),2,3 and a high correlation has been demonstrated
be-tween AHD measures taken with radio-graphs and those with MRI (r
= 0.81).32 With 2-D imaging, significant projection variations31
can be avoided with stan-dardization of subject and transducer
position. The reliability of our AHD measures were excellent, with
all ICC values greater than 0.75. Our results are comparable to a
recent reliability study of AHD measured on ultrasound images of
patients with RCD, with reported reli-ability of 0.92 and 0.90 at
rest and 60 abduction, respectively.29
There was no effect of the presence or absence of RCD on AHD
across pos-tures and arm angles. Subjects in the RCD group
represented the broad spectrum of the disease as identified by
their MRI, ranging from impingement syndrome to partial-thickness
and full-thickness rota-tor cuff tear. This supports external
validi-ty. However, this heterogeneity might have limited the
internal validity and, therefore, the ability to detect differences
between
40-10 Kalra.indd 638 9/21/10 2:38 PM
-
journal of orthopaedic & sports physical therapy | volume 40
| number 10 | october 2010 | 639
coNcLuSioN
We examined the direct effects of upper quadrant posture on the
SAS, using ultrasound to image
the SAS, performing a linear AHD mea-sure in patients with RCD
and control subjects without shoulder pain. Upright posture
increased AHD in both groups of subjects when tested in 45 actively
held abduction position. The AHD did not differ between slouched
posture as com-pared to normal posture, and between those with or
without the presence of RCD in the 2 arm positions. The AHD measure
in this study represents only the posterior to midportion of the
SAS in the coronal plane; therefore, results may dif-fer for the
anterior SAS. t
KeY PoiNtSFiNDiNgS: Upright posture resulted in a 1.2-mm
increase in SAS, as measured by AHD, with the arm at 45 abduc-tion,
when compared to normal posture. Change in AHD with upright posture
was not dependent on the presence or absence of RCD. Slouched
posture did not induce a change in AHD, as com-pared to normal
posture.imPLicAtioN: Upright posture can in-crease the SAS, as
measured by the AHD. However, the magnitude of the change is small
and within the range of measurement error for the majority of
subjects.cAutioN: Posture was artificially in-duced. Furthermore,
the SAS was as-sessed using the AHD, which is a 2-D measure and
taken from ultrasound images of the posterior to middle aspect of
the SAS.
groups. In a prior study that measured the AHD from ultrasound
images taken over the middle acromion,3 RCD severity was linked to
AHD, with a smaller AHD in those with more severe RCD. Our RCD
subgroup data do not descriptively sup-port this premise, which may
be due to the lower number of subjects or differences in methods,
as we generated our ultrasound images from the middle to posterior
as-pect of the acromion.
Patients with RCD have pain with shoulder elevation, suggesting
that the SAS should be measured at various po-sitions of shoulder
elevation. However, evidence suggests that clinical relevance of
measuring SAS at elevation angles greater than 35 to 40 of
glenohumeral elevation may not be important, as the supraspinatus
tendon has likely already passed underneath the acromion and thus
may no longer be at risk of impinge-ment in the SAS.4 Measurements
of the SAS and orientation of supraspinatus tendon found that,
anatomically, the su-praspinatus tendon was at greatest risk of
impingement between the acromion and greater tuberosity of humerus
between 27.7 to 36.1.4,5
The healthy group was not age and gender matched. Age of control
subjects (mean SD, 31.9 10.7 years) was sig-nificantly less (P.001)
than the age of the RCD group (mean SD, 53.5 13.7 years), with a
mean difference of 21.8 years (95% CI: 15.2, 27.9). With aging,
there is a potential of osteophyte forma-tion on the inferior
acromion that could affect the AHD measure. Gender was not matched;
however, there were no signifi-cant differences in distribution
between groups (P.05).
Postures were artificially induced in a laboratory setting,
therefore they may not represent faulty postures seen in cli-ents
without shoulder pain or in patients with shoulder pain. No
intervention for posture correction was given, so results cannot be
applied to the effects of a pos-tural treatment. Lastly, the tester
was not blinded to group membership during ul-trasound imaging of
the SAS.
reFereNceS
1. Allmann KH, Uhl M, Gufler H, et al. Cine-MR imaging of the
shoulder. Acta Radiol. 1997;38:1043-1046.
2. Azzoni R, Cabitza P. Sonographic versus radio-graphic
measurement of the subacromial space width. Chir Organi Mov.
2004;89:143-150.
3. Azzoni R, Cabitza P, Parrini M. Sonographic
evaluation of subacromial space. Ultrasonics. 2004;42:683-687.
http://dx.doi.org/10.1016/j.ultras.2003.11.015
4. Bey MJ, Brock SK, Beierwaltes WN, Zauel R, Kolowich PA, Lock
TR. In vivo measure-ment of subacromial space width during shoulder
elevation: technique and preliminary results in patients following
unilateral rota-tor cuff repair. Clin Biomech (Bristol, Avon).
2007;22:767-773.
http://dx.doi.org/10.1016/j.clinbiomech.2007.04.006
5. Bey MJ, Zauel R, Brock SK, Tashman S. Validation of a new
model-based tracking technique for measuring three-dimensional, in
vivo glenohumeral joint kinematics. J Bio-mech Eng.
2006;128:604-609. http://dx.doi.org/10.1115/1.2206199
6. Borstad JD. Resting position variables at the shoulder:
evidence to support a posture-impair-ment association. Phys Ther.
2006;86:549-557.
7. Brossmann J, Preidler KW, Pedowitz RA, White LM, Trudell D,
Resnick D. Shoulder impingement syndrome: influence of shoulder
position on ro-tator cuff impingement--an anatomic study. AJR Am J
Roentgenol. 1996;167:1511-1515.
8. Bullock MP, Foster NE, Wright CC. Shoulder impingement: the
effect of sitting posture on shoulder pain and range of motion. Man
Ther. 2005;10:28-37.
http://dx.doi.org/10.1016/j.math.2004.07.002
9. Culham E, Peat M. Functional anatomy of the shoulder complex.
J Orthop Sports Phys Ther. 1993;18:342-350.
10. Desmeules F, Minville L, Riederer B, Cote CH, Fremont P.
Acromio-humeral distance variation measured by ultrasonography and
its associa-tion with the outcome of rehabilitation for shoul-der
impingement syndrome. Clin J Sport Med. 2004;14:197-205.
11. Donatelli R. Physical Therapy of the Shoulder (Clinics in
Physical Therapy). 4th ed. New York, NY: Churchill-Livingstone;
2004.
12. Finley MA, Lee RY. Effect of sitting posture on
3-dimensional scapular kinematics measured by skin-mounted
electromagnetic tracking sen-sors. Arch Phys Med Rehabil.
2003;84:563-568. http://dx.doi.org/10.1053/apmr.2003.50087
13. Flatow EL, Soslowsky LJ, Ticker JB, et al. Excur-sion of the
rotator cuff under the acromion. Pat-terns of subacromial contact.
Am J Sports Med. 1994;22:779-788.
14. Girometti R, De Candia A, Sbuelz M, Toso F, Zuiani C,
Bazzocchi M. Supraspinatus tendon US morphology in basketball
players: correla-tion with main pathologic models of secondary
impingement syndrome in young overhead athletes. Preliminary
report. Radiol Med. 2006;111:42-52.
15. Graichen H, Bonel H, Stammberger T, et al. Three-dimensional
analysis of the width of the subacromial space in healthy subjects
and patients with impingement syndrome. AJR Am J Roentgenol.
1999;172:1081-1086.
16. Greenfield B, Catlin PA, Coats PW, Green E,
40-10 Kalra.indd 639 9/21/10 2:38 PM
-
640 | october 2010 | volume 40 | number 10 | journal of
orthopaedic & sports physical therapy
[ research report ]
@ more iNFormAtioNwww.jospt.org
McDonald JJ, North C. Posture in patients with shoulder overuse
injuries and healthy individu-als. J Orthop Sports Phys Ther.
1995;21:287-295.
17. Gumina S, Di Giorgio G, Postacchini F, Postac-chini R.
Subacromial space in adult patients with thoracic hyperkyphosis and
in healthy vol-unteers. Chir Organi Mov. 2008;91:93-96.
http://dx.doi.org/10.1007/s12306-007-0016-1
18. Hawkins RJ, Kennedy JC. Impingement syndrome in athletes. Am
J Sports Med. 1980;8:151-158.
19. Hebert LJ, Moffet H, Dufour M, Moisan C. Ac-romiohumeral
distance in a seated position in persons with impingement syndrome.
J Magn Reson Imaging. 2003;18:72-79.
http://dx.doi.org/10.1002/jmri.10327
20. Kebaetse M, McClure P, Pratt NA. Tho-racic position effect
on shoulder range of motion, strength, and three-dimensional
scapular kinematics. Arch Phys Med Rehabil. 1999;80:945-950.
21. Lewis JS, Green A, Wright C. Subacromial impingement
syndrome: the role of posture and muscle imbalance. J Shoulder
Elbow Surg. 2005;14:385-392.
http://dx.doi.org/10.1016/j.jse.2004.08.007
22. Lewis JS, Wright C, Green A. Subacromial impingement
syndrome: the effect of chang-ing posture on shoulder range of
movement. J Orthop Sports Phys Ther. 2005;35:72-87.
http://dx.doi.org/10.2519/jospt.2005.1578
23. Ludewig PM, Reynolds JF. The association of scapular
kinematics and glenohumeral joint pathologies. J Orthop Sports Phys
Ther. 2009;39:90-104. http://dx.doi.org/10.2519/jospt.2009.2808
24. Lukasiewicz AC, McClure P, Michener L, Pratt N, Sennett B.
Comparison of 3-dimensional
scapular position and orientation between sub-jects with and
without shoulder impingement. J Orthop Sports Phys Ther.
1999;29:574-583; discussion 584-576.
25. McCann PD, Bigliani LU. Shoulder pain in tennis players.
Sports Med. 1994;17:53-64.
26. Michener LA, Kalra N, Pinkstaff S, Ericksen J, Boardman N.
Measurement of the subacromial space using ultrasonography. J Athl
Train. 2007;42:126.
27. Michener LA, McClure PW, Karduna AR. Ana-tomical and
biomechanical mechanisms of sub-acromial impingement syndrome. Clin
Biomech (Bristol, Avon). 2003;18:369-379.
28. Neer CS, 2nd. Impingement lesions. Clin Orthop Relat Res.
1983;70-77.
29. Pijls BG, Kok FP, Penning LI, Guldemond NA, Arens HJ.
Reliability study of the sonographic measurement of the
acromiohumeral distance in symptomatic patients. J Clin Ultrasound.
2010;38:128-134. http://dx.doi.org/10.1002/jcu.20674
30. Roberts CS, Davila JN, Hushek SG, Tillett ED, Corrigan TM.
Magnetic resonance imaging anal-ysis of the subacromial space in
the impinge-ment sign positions. J Shoulder Elbow Surg.
2002;11:595-599. http://dx.doi.org/10.1067/mse.2002.127095
31. Rozing PM, Obermann WR. Osteometry of the glenohumeral
joint. J Shoulder Elbow Surg. 1999;8:438-442.
32. Saupe N, Pfirrmann CW, Schmid MR, Jost B, Werner CM, Zanetti
M. Association between rotator cuff abnormalities and reduced
acro-miohumeral distance. AJR Am J Roentgenol. 2006;187:376-382.
http://dx.doi.org/10.2214/AJR.05.0435
33. Seitz AL, Michener L, Lynch SS, Zirker C, Stith
TR, Boardman N. Anterior versus posterior
subacromial space changes with elevation mea-
sured in vivo using ultrasonography [abstract]. J
Orthop Sports Phys Ther. 2010;40:A30-31.
34. Silva RT, Hartmann LG, Laurino CF, Bil JPR. Clinical and
ultrasonographic correlation
between scapular dyskinesia and subacromial
space measurement among junior elite tennis
players. Br J Sports Med. 2008;http://dx.doi.
org/10.1136/bjsm.2008.046284
35. Solem-Bertoft E, Thuomas KA, Westerberg CE. The influence of
scapular retraction and protrac-
tion on the width of the subacromial space. An
MRI study. Clin Orthop Relat Res. 1993;99-103.
36. Urwin M, Symmons D, Allison T, et al. Estimat-ing the burden
of musculoskeletal disorders in
the community: the comparative prevalence of
symptoms at different anatomical sites, and the
relation to social deprivation. Ann Rheum Dis.
1998;57:649-655.
37. van der Windt DA, Koes BW, de Jong BA, Bouter LM. Shoulder
disorders in general practice:
incidence, patient characteristics, and manage-
ment. Ann Rheum Dis. 1995;54:959-964.
38. Wang HK, Lin JJ, Pan SL, Wang TG. Sonographic evaluations in
elite college baseball athletes.
Scand J Med Sci Sports. 2005;15:29-35. http://
dx.doi.org/10.1111/j.1600-0838.2004.00408.x
FIND Author Instructions & Tools on the Journals
WebsiteJOSPTs instructions to authors are available at
www.jospt.org by clicking AUTHOR TOOLS & INSTRUCTIONS in the
upper right-hand column of the home page, or by visiting
INFORMATION FOR AUTHORS, located in the sites navigation bar in the
left-hand column. The Journals editors have assembled a list of
useful tools and links for authors as well as reviewers.
40-10 Kalra.indd 640 9/21/10 2:38 PM
633JOSPToct10.p1634JOSPToct10.p1635JOSPToct10.p1636JOSPToct10.p1637JOSPToct10.p1638JOSPToct10.p1639JOSPToct10.p1640JOSPToct10.p1