Kijkunasathian et al.

BMC Musculoskeletal Disorders (2023) 24:255 BMC Musculoskeletal

https://doi.org/10.1186/s12891-023-06372-3
Disorders

RESEARCH Open Access

The preferable shoulder position can isolate

supraspinatus activity superior to the classic
empty can test: an electromyographic study
Chusak Kijkunasathian1, Supajed Niyomkha1, Patarawan Woratanarat1 and Chaiyanun Vijittrakarnrung1*

Abstract
Background  Supraspinatus (SSP) strength tests are an important shoulder examination tool for clinical evaluations
of patients with a suspected SSP tear. While the empty can (EC) test is widely used to diagnose SSP dysfunction, the
test cannot selectively activate SSP activity. The aim of this study was to access the electromyographic (EMG) activity
within SSP, deltoid, and surrounding periscapular muscles after resisted abduction force to determine which shoulder
position helps best isolate SSP from deltoid activity.
Methods  A controlled laboratory EMG study was conducted. Specifically, we conducted an EMG analysis of
the seven periscapular muscles (i.e., the middle deltoid, anterior deltoid, SSP, upper trapezius, posterior deltoid,
infraspinatus, and pectoralis major) in 21 healthy participants, without any history of shoulder disorder, aged 29 ± 0.9
years old with a dominant-right arm. EMG activities were measured during resisted abduction force according to
comprehensive shoulder positions in abduction, horizontal flexion, and humeral rotation. The supraspinatus to middle
deltoid (S:D) ratio was calculated using the standardized weighted EMG and the maximum voluntary isometric
contraction of the SSP and middle deltoid muscles, for each shoulder position to determine the best isolated SSP
muscle strength test position. Results were analyzed with the Kruskal–Wallis test for non-normally distributed data.
Results  Shoulder abduction, horizontal flexion, and humeral rotation significantly affected the activity of the middle
deltoid, SSP, and S:D ratio (P < 0.05). The S:D ratio increased significantly in lower degrees of shoulder abduction, lower
degrees of horizontal flexion, and external humeral rotation over internal rotation. The greatest S:D ratio (3.4 (0.5–9.1))
occurred at the shoulder position of 30° shoulder abduction combined with 30° horizontal flexion and external
humeral rotation. Conversely, the classic EC position manifested nearly the smallest S:D ratio (0.8 (0.2–1.2)).
Conclusion  Application of the SSP strength test in the shoulder position of 30 degrees abduction, 30 degrees
horizontal flexion, and external humeral rotation offers the best position to isolate the abducting activity of the
SSP from that of the deltoid, which could help with diagnosis among patients with chronic shoulder pain with a
suspected SSP tear condition.
Keywords  Rotator cuff tear, Supraspinatus tear, Electromyographic Study, Shoulder physical examination, S:D ratio

*Correspondence:
Chaiyanun Vijittrakarnrung
[email protected]
1
Department of Orthopedics, Faculty of Medicine Ramathibodi Hospital,
Mahidol University, 270, Rama VI Road, Ratchathewi District,
Bangkok 10400, Thailand

© The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use,
sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and
the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included
in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The
Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available
in this article, unless otherwise stated in a credit line to the data.
Kijkunasathian et al. BMC Musculoskeletal Disorders (2023) 24:255 Page 2 of 12

Background conclusion, the diagnostic accuracy of these classic EC

Rotator cuff tears are a major cause of chronic shoulder and FC tests are still unsatisfactory, the more proper
disability [1]. Among such tears, the supraspinatus (SSP) shoulder position which could isolate and specific with
is the most affected tendon [2, 3]. An increased recogni- SSP activity should be further determined.
tion and understanding of SSP tear pathology have led to To increase SSP muscle strength test accuracy, the
further assessment of the related clinical diagnosis. The proper shoulder position must be accountable for maxi-
SSP muscle strength tests remain an essential tool for the mizing the abducting contribution of SSP and mini-
clinical evaluation of patients with a suspected SSP tear. mizing the deltoid abducting activity. Chalmers et al.
Acting as the main humeral depressor, SSP also functions advocated the SSP and middle deltoid ratio (S:D ratio)
as a prime initiator of glenohumeral joint elevation [4]. as a parameter to represent and quantify how well each
Nevertheless, due to the complex anatomy of the shoul- shoulder position isolated SSP activity from deltoid activ-
der joint, overlapping muscle function could affect the ity [17]. Several previous studies showed that a lower
interpretation of this specific physical examination [5, degree of shoulder abduction might be more specific to
6]. Accordingly, physicians must isolate the SSP function SSP function than to that of the deltoid function [17–
from the deltoid abduction force to identify and diagnose 19]. However, no previous study examined the potential
SSP tears. relationship between periscapular muscles, especially
The most common special test used to examine the the SSP and deltoid, and comprehensive shoulder posi-
integrity of SSP is the “empty can” (EC) or “Jobe’s” test. tion with respect to abduction, horizontal flexion, and
Initially proposed by Jobe and Moynes in 1982 [7], the humeral rotation. To fill that research gap, the primary
EC test involves resistance being applied to abduction objectives of our study were to [1] conduct an electro-
in 90 degrees shoulder abduction, 30 degrees shoulder myographic (EMG) study to determine which shoulder
horizontal flexion, and full internal humeral rotation. The position best isolates SSP from deltoid activity and [2]
researchers explained that, with this shoulder position, evaluate the EMG activity within the SSP, deltoid, and
SSP activity could be isolated based on only one subject surrounding shoulder muscle after resisted abduction
[7]. Subsequently, the “full can” (FC) modification test force in various shoulder positions. We hypothesized that
was introduced by Kelly et al. in 1996 [8]. They claimed lower degrees of abduction and horizontal flexion would
that FC would be less painful than EC due to avoiding an better isolate SSP abduction activity from the deltoid
impingement position, which could result in more reli- activity.
able results for SSP tear diagnoses.
Traditionally, the shoulder physical examination has Materials and methods
been a cornerstone of the diagnostic process. Largely Study design and participants
based on the result of these two original studies [7, 8], The controlled laboratory EMG study was conducted
both EC and FC tests have turned into classic used clini- at the Department of Orthopedics, Faculty of Medicine
cal examination for diagnosing SSP pathology. Never- Ramathibodi Hospital. Participants who were normal
theless, the result of their previous EMG studies [7, 8] healthy individuals aged 18–40 years old without any
provided insufficient information to support the conclu- history of shoulder instability, major shoulder trauma,
sion that the EC and FC tests can specifically isolate SSP shoulder surgery, shoulder or periscapular pain were
activity. Certainly, previous EMG studies suggest that included. A complete shoulder physical examination was
the EC and FC tests extremely activate deltoid muscle [9, performed in every participant by the orthopedic sur-
10], infraspinatus [9, 11] as well as SSP. In clinical prac- geon. Participants with medical comorbidity-affected
tice, the EC and FC test could be painful and difficult to shoulder motion, or abnormal shoulder examination
achieve for patients, resulting in apparent weakness from were excluded. Experimental testing was conducted on a
the pain-mediated reflex inhibitor of the muscle. Many healthy participant as it was determined that interpreta-
previous studies have manifested the unsatisfactory diag- tion of SSP testing in patients with SSP pathology has to
nostic accuracy of these tests in terms of clinical appli- be established base on a precise understanding of normal
cation [5, 12–14). Longo et al. [14] conducted a review EMG muscle activation [13]. Besides, our protocol was
article on clinical testing for SSP pathology, they found set up in the same fashion as many previous EMG studies
that the EC test mostly had a sensitivity lesser than 80% [4, 11, 13, 17].
(4 out of 6 studies), and a specificity of less than 80% (5 The sample size was calculated using STATA 15.0 and a
out of 6 studies). Correspondingly, they also found that reference from a previous study [8]. The following values
the FC test mostly had a sensitivity and specificity lesser were used to calculate the sample size: an alpha error of
than 80% (3 out of 4 studies). Some studies also demon- 0.05, power of study of 0.8, mean SSP MVIC in 90° shoul-
strated sensitivity, specificity, and accuracy of the EC test der flexion and external rotation of 7.65 volts, SD of 1.58
as low as 30%, 35% and 50%, respectively [8, 15, 16]. In volts, mean SSP MVIC in 0° shoulder flexion and external
Kijkunasathian et al. BMC Musculoskeletal Disorders (2023) 24:255 Page 3 of 12

rotation of 6.69 volts, SD of 2.96 volts, 23 measurements, electrodes (Fig.  1A–B) were applied to skin, parallel to
1 baseline measurement, and a between-measurement the muscle direction with 2  cm between center-center,
correlation of 0.8. The total sample size needed was 21 as described in a related study [21]. The electrodes were
participants. After informed consent was obtained from applied over seven muscles (i.e., middle deltoid, anterior
patients, baseline characteristics—including age, gender, deltoid, supraspinatus, upper trapezius, posterior deltoid,
and body mass index (BMI)—were recorded. A total of infraspinatus, and pectoralis major) by a single experi-
21 participants were included in the final sample. All par- enced physiotherapist (Fig. 1C–D; Table 1). These seven
ticipants were males aged 29 ± 0.9 years old with a domi- periscapular muscles are selected based on previous rel-
nant-right arm. The mean BMI was 24.6 ± 2.9 kg/m2. evant EMG studies that demonstrated an activation of
All participants were unaware of the study hypoth- these muscles during shoulder abduction [8, 11, 13]. Sub-
esis. This study was ethically approved by our hospital’s scapularis and Latissimus dorsi were excluded due to its
institutional research board committee (IRB number prime function as an internal rotation and low activity
MURA2017/582). All methods were performed in accor- during shoulder abduction [8, 13, 22]. The EMG signals
dance with the Helsinki guidelines and relevant CIOMS were sampled by computer at 1000 Hz. Eight integrated
guidelines. channels were used for signal filtering (10 and 400  Hz,
Butterworth) and rectification. The isometric contraction
Experiment protocol was measured for a total of 5 s interval [8].
All eligible participants dressed in the proper attire and A single well-trained examiner conducted all testing to
exposed their upper trunks and extremities. Only the ensure the replication of the same resistance and posi-
dominant arm was considered for measurement. The tions. The maximum voluntary isometric contraction
scapular plane position was measured using a standard (MVIC) of the seven muscles was measured by perform-
goniometer (Supplementary Fig.  1) according to previ- ing manual maximum isometric resistance in standard
ous proposed method [20]. The skin around the shoul- posture references, as described in Table 1. The test order
der was prepared using an alcohol rub. After the surface for individual participants was a randomized sequence.
anatomical landmarks were outlined, electromyography Each contraction was set in the same 5-second pattern
(Wireless Myon 320 Surface Electromyography System®; used in a previous study [11]. Three trials of each muscle
Schwarzenberg, Switzerland) was used; surface-adhesive testing were conducted, with a minimum rest interval of

Fig. 1  Wireless Myon 320 Surface Electromyography System®; Schwarzenberg, Switzerland. The system consists of transmitters (A) and receivers (B).
Surface electrode placement on seven muscles identified by different-colored electrodes for the anterior side (C) and posterior side (D)
Kijkunasathian et al. BMC Musculoskeletal Disorders (2023) 24:255 Page 4 of 12

Table 1  Descriptions of the surface electrodes placement and Data collection & outcome measurement
MVIC normalization positions for each of the seven shoulder We used the EMG activity of the middle deltoid to rep-
muscles examined resent the deltoid abduction activity. The best shoulder
Muscle Surface electrode place- MVIC normalization
position to isolate SSP from deltoid activity was quanti-
ment [11, 23] positions [24, 25]
fied using S:D ratio [17], which can be calculated using a
Middle deltoid Intersection of the midpoint Resisted abduction
between the anterior and at 0° elevation in percentage of the standardized weighted EMG (%sEMG)
posterior deltoid muscles scapular plane and of the SSP divided by the %sEMG of the middle del-
and the midpoint between elbow flexion to 90° toid. The %sEMG of each muscle was calculated using
the acromion and deltoid the relevant standardized weighted EMG divided by the
tuberosity
relevant MVIC. Thus, the higher S:D ratio represents a
Anterior deltoid 3.5 cm below the anterior Resisted forward flex-
angle of the acromion ion at 0° elevation in
higher contribution of SSP activity compared to that of
scapular plane and the middle deltoid.
elbow flexion to 90°
Supraspinatus Midpoint and 2 Resisted elevation Statistical analysis
finger-breadths at 90° elevation in Statistical analyses were calculated using Stata 15 soft-
anterior to scapular spine scapular plane, 45° ware (StataCorp, College Station, TX, USA). The baseline
internal rotation and
elbow extension
characteristics—such as age (years), body weight (kg),
Upper trapezius Supero-medial and infero- Resisted shoulder
height (cm), BMI (kg/m2), scapular plane (degree), MVIC,
lateral to a point 2 cm lateral shrug with subject and %sEMG—of seven muscles were reported using the
to one-half the distance seated and arm at mean and standard deviation. Data were analyzed for
between the C7 spinous side completeness and normality using the Shapiro–Wilk test
process and the lateral tip of combined with a normal distribution plot. The Mann–
the acromion
Whitney test was used for continuous variable that did
Posterior deltoid 2 cm below the posterior Resisted extension at
angle of the acromion 0° elevation in scapu- not satisfy normality. The between-group difference in
lar plane and elbow the MVIC and S:D ratio was determined by analyzing the
flexion to 90° variable (normally distributed: one-way analysis of vari-
Infraspinatus Parallel to spine of scapulae, Resisted external ance; non-normally distributed: Kruskal–Wallis test).
approximately 4 cm below, rotation at 90° eleva- The multilevel mixed-effects linear regression analysis
over the infrascapular fossa tion in scapular plane
was used to identify the coefficient comparing groups of
and neutral rotation
higher degrees of motion. Multiple regression with a par-
Pectoralis major 3.5 cm medial to the ante- Resisted horizontal
rior axillary line adduction at 90° simonious model and 95% confidence interval was used
elevation in scapular to identify and control confounding factors. A P value
plane and elbow less than 0.05 was considered statistically significant.
flexion to 90°
Results
30 s between trials. Every trial was closely monitored to Participant demographic data
avoid compensatory movement from the trunk and scap- The total number of recruited samples was 21 partici-
ular. After recording the MVIC value for each muscle ref- pants. The mean scapular angle was 26.2° ± 4.2°. The
erence, examiner instructed participants to isometrically mean MVIC was 2.2 ± 0.3 volts for SSP and 2.2 ± 0.2 volts
hold a standard 1-kg dumbbell for 5 s in 24 total shoulder for the middle deltoid, respectively. The participant’s
positions (i.e., factorial of 30°/60°/90° shoulder abduction; demographic data and MVIC values for other muscles
0°/30°/60°/scapular plane; and internal/external humeral are reported in Table 2.
rotation) (Fig.  2A–C). The EMG activity of each muscle
after participants held the 1-kg weight was collected as Standardized weighted EMG testing
the standardized weighted EMG according to the indi- We normalized all standardized weighted EMG activi-
vidual position. The participants’ shoulder angle mea- ties to %sEMG, which can be calculated by dividing the
surements were done using 2-plane goniometer (Fig. 2D), sEMG of the activities by their MVIC. The highest per-
which has been previously demonstrated excellent reli- centage %sEMG for SSP was detected in the 90° shoul-
ability and validity when compared with a digital incli- der abduction combined with 0° horizontal flexion and
nometer for measuring shoulder range of motion [26]. To internal humeral rotation. The %sEMG values for the
prevent the fatigue effect, participants were given a mini- seven muscles in all shoulder positions are represented in
mum 30-second rest interval after each measurement Table 3.
[27]. According to the relationship between shoulder
positions and %sEMG (Supplementary Fig.  2), the
Kijkunasathian et al. BMC Musculoskeletal Disorders (2023) 24:255 Page 5 of 12

Fig. 2  Comprehensive shoulder motion tasks for 30°/60°/90° shoulder abduction (A), 0°/30°/60°/scapular plane horizontal flexion (B), and full internal/
external humeral rotation (C). Example of two-plane goniometer measurement of 90° shoulder abduction, 30° horizontal flexion, and full internal humeral
rotation, i.e., EC position (D)

infraspinatus and pectoralis major demonstrated unrelate abduction, 30 and 60-degree horizontal flexion compared
activity in any shoulder position, illustrated as a nearly with 0-degree horizontal flexion) (Table 4).
flat graph over all positions. Additionally, the upper tra- The highest S:D ratio represented the best shoulder
pezius and posterior deltoid manifested similar trend to position to isolate the SSP from deltoid activity, and the
SSP and middle deltoid respectively, but with a relative highest ratio occurred when the shoulder was in the
lower magnitude of EMG changes. position of 30° shoulder abduction combined with 30°
To simplify the analysis, further analysis and the S:D horizontal flexion and external humeral rotation, and the
ratio calculation were only completed for the middle next highest ratio occurred at the position of 30° shoul-
deltoid and SSP. The middle deltoid activity significantly der abduction combined with 60° horizontal flexion and
increased in higher degrees of shoulder abduction, lower external humeral rotation, which was significant with
degrees of horizontal flexion, and internal humeral rota- a P-value of 0.0001 from the Kruskal–Wallis test (com-
tion over external rotation (P < 0.05; Fig.  3). SSP activ- pare among 24 shoulder positions) (Table 5). Using mul-
ity significantly increased in higher degrees of shoulder tiple regression with the parsimonious model, we found
abduction, lower degrees of horizontal flexion, and exter- that factors inversely associated with the outcomes were
nal humeral rotation over internal rotation (P < 0.05; shoulder position, body weight, and scapular plane, with
Fig.  4). The S:D ratio significantly increased in lower P-value from F-test < 0.0001, which indicated overall sig-
degrees of shoulder abduction, lower degrees of horizon- nificant in the regression analysis (Table  6), this infor-
tal flexion, and external humeral rotation over internal mation highlighted the significance of shoulder position
rotation (P < 0.05; Fig.  5). The multilevel mixed-effects related to S:D ratio values as proposed in our primary
linear regression analysis revealed how the coefficients objective
differed between groups with higher degrees of motion
(60 and 90-degree abduction compared with 30-degree
Kijkunasathian et al. BMC Musculoskeletal Disorders (2023) 24:255 Page 6 of 12

Table 2  Participants’ demographic characteristics and the Our results showed that the shoulder position of 30°
maximum voluntary isometric contraction for each of the seven shoulder abduction combined with 30° horizontal flexion
muscles and external humeral rotation best isolates the abducting
Demographic characteristics Study
activity of the SSP from the abducting activity of the del-
population
toid. Meanwhile, the 90° shoulder abduction combined
N = 21
Age (years)∞
with 30° horizontal flexion and internal humeral rotation
  Mean ± SD 29.0 ± 0.9
(i.e., the EC test position) manifests the least abducting
Genderµ
activation of the SSP relative to the deltoid in proposition
 Male 21(100%) to the nearly lowest S:D ratio. By using S:D ratio com-
Height (cm)∞ parison, our proposed position contributes up to 4.25-
  Mean ± SD 175.6 ± 5.3 fold better in SSP isolation than the classic EC position.
Weight (kg)∞ The clinical application of using this shoulder position in
  Mean ± SD 75.9 ± 10.3 SSP strength tests can potentially improve the accuracy
BMI (kg/m2)∞ of physical examination using SSP testing. Nevertheless,
  Mean ± SD 24.6 ± 2.9 in our analysis, we only considered the EMG activity of
Scapular Plane (degrees)∞ the middle deltoid to represent a deltoid abduction activ-
  Mean ± SD 6.2 ± 4.2 ity; we did not include anterior deltoid activity. Compara-
Maximum voluntary contraction (unit)∞ tively, a previous study suggested that the anterior deltoid
  Middle deltoid 2.2 ± 0.2 contributed force only 2% for significant abduction
  Anterior deltoid 2.1 ± 0.2 touque [31]. Despite this difference, the shoulder position
 Supraspinatus 2.2 ± 0.3 of 30° shoulder abduction combined with 30° horizontal
  Upper trapezius 1.4 ± 0.5 flexion and external humeral rotation still offers the best
  Posterior deltoid 2.1 ± 0.4 position when considering the S:D ratio calculation with
 Infraspinatus 1.6 ± 0.4 the mean deltoid activity from the anterior and middle
  Pectoralis major 1.4 ± 0.5 deltoid (Supplementary Fig. 3).
∞: value presented as mean ± standard deviation The results of the present study are consistent with sev-
µ: value presented as the number of volunteers with that condition (percentage) eral published EMG studies. Wickham et al. measured
EMG activity in a wide variety of periscapular muscles,
Discussion demonstrating a comparable trend of muscle activa-
Rotator cuff tears, especially SSP tears, are a highly com- tion to that in our study during the shoulder abduction
mon cause of chronic shoulder disability, leading to moment. Corresponding to the Wickham et al. study, the
decreased quality of life, decreased functionality, and abduction motion also significantly increased both SSP
increased utilization of healthcare resources [28]. Due to and deltoid activity, but the magnitude of change was
clinical readiness accessibility, SSP muscle strength tests higher in the deltoid than in the SSP [32]. In addition,
are typically the first mandatory screening tool used for Chalmer et al. performed a specific shoulder examina-
patients suspected of having an SSP tear [29]; the tests tion test called the champagne toast test, which involves
provide additional information and identify the need a 30° shoulder abduction combined with relative external
for further investigation [30]. This study aimed to deter- humeral rotation. The researchers suggested that resisted
mine which shoulder position can best isolate SSP from abduction strength testing in lower degrees of shoulder
deltoid activity based on the EMG activity analysis of abduction could deactivate deltoid function and isolated
the periscapular musculature during resisted abduction SSP activity [17]. Moreover, Kelly et al. conducted an
strength testing. EMG study of the periscapular muscles at various shoul-
To determine the best SSP muscle strength test posi- der positions. The result of the study demonstrated better
tion, the proper shoulder position must be accountable isolation of SSP activity in external humeral rotation than
for maximizing the abducting contribution from SSP. in internal humeral rotation [8]. Additionally, Malanga et
Our result showed that both the SSP and the deltoid have al. examined EMG testing of the SSP and deltoid at both
increased activity related to higher degrees of abduction, the empty can and Blackburn positions. The result of that
but in a different magnitude, the deltoid seems to become examination showed that neither testing position can iso-
increasingly dominant when compared to the SSP. How- late SSP from deltoid activity [9]. Overall, a comparison
ever, resisted abduction strength still depends on both of the findings suggests that our proposed position (i.e.,
the SSP and deltoid activity. With the aim of our study shoulder position of 30° shoulder abduction combined
to isolate SSP activity, the lower degrees of abduction with 30° horizontal flexion and external humeral rota-
should be considered to minimize the deltoid activity tion) better isolates the SSP than the classic EC position.
related to SSP activity.
Table 3  The percentage of standardized weighted EMG (%sEMG) within all seven muscles (i.e., middle deltoid, anterior deltoid, supraspinatus, upper trapezius, posterior deltoid,
infraspinatus, and pectoralis major) regarding comprehensive shoulder positions
Position no. Abduction Horizontal flex- Rotation Middle deltoid Anterior deltoid Supraspinatus Upper Trapezius Posterior Infra Spinatus Pec-
(degrees) ion (degrees) deltoid toralis
major
1 30 0 IR 15.0 (7.5) 10.7 (8.9) 11.3 (4.5) 6.7 (5.3) 9.7 (4.0) 4.8 (7.2) 2.7 (1.7)
2 30 0 ER 7.8 (4.3) 6.7 (5.5) 11.7 (5.2) 8.4 (6.4) 4.2 (2.5) 4.7 (2.7) 2.5 (1.5)
3 30 30 IR 9.2 (4.6) 7.9 (7.9) 9.5 (4.1) 8.9 (7.4) 5.5 (2.6) 2.7 (1.3) 2.7 (1.8)
4 30 30 ER 3.6 (2.6) 6.8 (4.0) 9.3 (4.3) 9.2 (8.0) 2.2 (3.0) 7.3 (11.9) 2.7 (1.6)
Kijkunasathian et al. BMC Musculoskeletal Disorders

5 30 60 IR 8.3 (7.7) 6.9 (5.6) 10.2 (9.3) 14.7 (29.6) 3.5 (3.5) 4.4 (8.3) 3.0 (1.8)
6 30 60 ER 2.0 (1.1) 6.4 (3.6) 7.2 (6.4) 7.5 (8.2) 1.0 (0.5) 4.4 (2.8) 3.2 (2.0)
7 30 SP IR 11.2 (5.1) 9.1 (7.4) 10.4 (5.0) 10.8 (11.6) 6.4 (2.5) 2.8 (1.1) 2.7 (1.7)
8 30 SP ER 3.9 (2.1) 7.6 (4.5) 9.7 (4.8) 9.4 (8.4) 1.8 (1.2) 5.1 (2.8) 2.7 (1.5)
9 60 0 IR 29.7 (14.0) 22.5 (16.3) 20.1 (8.3) 14.0 (9.3) 19.2 (7.1) 5.9 (2.4) 2.8 (1.8)
10 60 0 ER 12.7 (5.7) 12.7 (8.1) 17.0 (7.1) 16.0(11.2) 5.9 (3.8) 7.2 (5.9) 2.7 (1.6)
(2023) 24:255

11 60 30 IR 18.4 (6.2) 16.3 (12.8) 16.9 (6.7) 15.4 (10.6) 10.9 (4.0) 5.4 (2.7) 3.8 (4.6)
12 60 30 ER 9.2 (2.7) 12.7 (5.2) 14.5 (4.7) 17.3(13.6) 4.4 (6.5) 8.4 (6.0) 4.3 (7.1)
13 60 60 IR 13.2 (7.6) 14.2 (11.3) 14.8 (8.2) 13.8 (11.3) 7.4 (3.6) 9.2 (7.8) 6.7 (8.5)
14 60 60 ER 8.8 (3.3) 14.9 (5.4) 12.3 (6.5) 15.6(11.0) 2.9 (1.7) 7.3 (2.8) 4.3 (2.4)
15 60 SP IR 21.5 (9.8) 18.2 (12.6) 17.6 (7.1) 14.7 (9.6) 13.1 (6.2) 8.2 (11.0) 2.9 (1.7)
16 60 SP ER 9.8 (3.3) 13.1(5.7) 15.6 (6.2) 16.6(10.8) 3.6 (1.7) 7.8 (5.1) 2.9 (1.6)
17 90 0 IR 39.9 (13.4) 30.9 (17.6) 27.1 (11.9) 20.9 (13.8) 24.8 (10.1) 8.3 (4.0) 2.9 (1.6)
18 90 0 ER 17.2 (8.1) 18.3(12.5) 22.1 (9.9) 20.4(13.7) 8.2 (5.6) 7.6 (4.5) 3.3 (3.2)
19 90 30 IR 31.7 (11.3) 25.6 (17.4) 22.7 (9.5) 21.4 (13.5) 18.8 (8.2) 7.9 (3.2) 3.0 (1.7)
20 90 30 ER 15.1 (5.9) 17.5 (9.5) 18.6 (7.5) 22.1(14.7) 5.3 (1.7) 8.5 (3.9) 3.3 (2.1)
21 90 60 IR 23.2 (8.8) 20.9 (11.1) 20.3 (10.3) 22.3 (14.7) 11.8 (4.4) 9.6 (6.0) 3.2 (1.8)
22 90 60 ER 14.8 (4.8) 19.9 (7.8) 15.6 (6.5) 23.7(20.0) 4.7 (1.6) 10.7 (8.7) 4.3 (3.2)
23 90 SP IR 32.1 (13.0) 26.1 (17.6) 24.3 (11.8) 21.6 (14.7) 18.9 (9.2) 8.4 (4.1) 3.0 (1.7)
24 90 SP ER 15.7 (7.1) 17.5 (8.0) 19.7(10.1) 21.2(13.3) 5.9 (2.4) 8.5 (4.4) 3.0 (1.6)
Value presented as mean(standard deviation)
*IR = internal rotation, ER = external rotation, SP = scapular plane
Page 7 of 12
Kijkunasathian et al. BMC Musculoskeletal Disorders (2023) 24:255 Page 8 of 12

Fig. 3  Comparison of the percentage of standardized weighted EMG (%sEMG) of middle deltoid between the groups graded by degree of shoulder
abduction, degree of shoulder horizontal flexion, and humeral rotation. Error bars indicate the interquartile range (IQR) of the median. Black dots indicate
values above the upper fence (1.5*IQR). * above the lines spanning between groups indicates significant P-values < 0.05

Fig. 4  Comparison of the percentage of standardized weighted EMG (%sEMG) of supraspinatus between the groups graded by degree of shoulder
abduction, degree of shoulder horizontal flexion, and humeral rotation. Error bars indicate the interquartile range (IQR) of the median. Black dots indicate
values above the upper fence (1.5*IQR). * above the lines spanning between groups indicates significant P-values < 0.05
Kijkunasathian et al. BMC Musculoskeletal Disorders (2023) 24:255 Page 9 of 12

Fig. 5  Comparison of the supraspinatus: middle deltoid (S:D) ratio between the groups graded by degree of shoulder abduction, degree of shoulder
horizontal flexion, and humeral rotation. Error bars indicate the interquartile range (IQR) of the median. Black dots indicate values above the upper fence
(1.5*IQR). * above the lines spanning between groups indicates significant P-values < 0.05

Table 4  The multilevel mixed-effects linear regression revealed the differing coefficients between groups with higher degrees of
motion
Parameters Coefficient 95% Conf. interval P-value
Middle deltoid
Abduction
  60° (compare with 30°) 7.7 6.32 – 9.08 < 0.001**
  90° (compare with 30°) 16 14.62 – 17.38 < 0.001**
Horizontal flexion (degrees)
  30° (compare with 0°) -5.82 -7.2 – (-4.4) < 0.001**
  60° (compare with 0°) -8.68 -10.05 – (-7.3) < 0.001**
Supraspinatus
Abduction
  60° (compare with 30°) 6.08 4.87 – 7.28 < 0.001**
  90° (compare with 30°) 11.2 10 – 12.4 < 0.001**
Horizontal flexion
  30° (compare with 0°) -2.99 -4.1 – (-1.88) < 0.001**
  60° (compare with 0°) -4.84 -5.95 – (-3.73) < 0.001**
S:D ratio
Abduction
  60° (compare with 30°) -0.95 -1.37 – (-0.53) < 0.001**
  90° (compare with 30°) -1.23 -1.65 – (-0.81) < 0.001**
Horizontal flexion
  30° (compare with 0°) 0.45 0.06 – 0.85 0.023*
  60° (compare with 0°) 0.86 0.47 – 1.26 < 0.001**
*Significant at level 0.05
**Significant at level 0.01
Kijkunasathian et al. BMC Musculoskeletal Disorders (2023) 24:255 Page 10 of 12

Table 5  The supraspinatus: middle deltoid (S:D) ratio for 24 as plain shoulder radiograph, and MRI, to diagnostic
shoulder positions urgency conditions, such as SSP tear.
Position no. Abduction Horizontal Rotation Median The present study has several strengths. The first is
(degrees) flexion (range)
(degrees)
due to the limitation of previous studies when restricted
1 30 0 IR 0.7 (0.3–1.5) planes of shoulder motion were evaluated. Our study
2 30 0 ER 1.5 (0.7–2.8) design allows for a simultaneous comprehensive three-
3 30 30 IR 1.0 (0.3–3.3) plane motion assessment, especially with the horizontal
4 30 30 ER 3.4 (0.5–9.1) flexion and scapular plane, which potentially helps iden-
5 30 60 IR 1.5 (0.1–11.6) tify the best position of the shoulder to maximize the
6 30 60 ER 3.0 (0.6–21.9) abducting contribution of SSP and minimize the deltoid
7 30 SP IR 0.9 (0.4–1.7) abducting activity. Our study also advocates for each
8 30 SP ER 2.5 (0.9–8.0) participant to serve as an internal control, and measure-
9 60 0 IR 0.7 (0.4–1.3) ments were normalized for each participant. Further, the
10 60 0 ER 1.4 (0.5–2.6) electromyography measurements were completed by a
11 60 30 IR 1.0 (0.5–1.7) single well-trained physiotherapist, and an adequate sam-
12 60 30 ER 1.6 (0.7–4.4) ple size was obtained to determine the clinically signifi-
13 60 60 IR 1.2 (0.3–7.3) cant result.
14 60 60 ER 1.5 (0.3–4.1) Our study also has several limitations. First, to reduce
15 60 SP IR 0.9 (0.4–1.8) participants’ discomfort, surface-adhesive electrodes
16 60 SP ER 1.6 (0.5–4.0) were used for EMG in our study due to their non-inva-
17 90 0 IR 0.7 (1.6–1.2) siveness. Previous studies demonstrated some relation-
18 90 0 ER 1.4 (0.3–2.8) ships for estimating the EMG muscle activity between
19 90 30 IR 0.8 (0.2–1.2) the surface and fine-needle electrode [23, 33]. Despite
20 90 30 ER 1.4 (0.2–2.9) this variation, our significant result remains consistent
21 90 60 IR 1.0 (0.1–1.5) with multiple previous studies that used fine-needle
22 90 60 ER 1.2 (0.1–2.1) electrodes. Second, we did not include subscapularis
23 90 SP IR 0.9 (0.2–1.3) in our study, which might interfere with the abduction
24 90 SP ER 1.4 (0.2–3.2) torque in some positions [31]. Third, due to the limita-
P-value from Kruskal–Wallis (compared among 24 shoulder positions) = 0.0001*
tions of the study design, the sample only included nor-
*IR = internal rotation, ER = external rotation, SP = scapular plane
mal healthy individuals. However, we did not perform
any further investigation to quantify that these partici-
Table 6  The multiple regression with parsimonious model pants are exclusively normal, though all participants were
demonstrated variable factors associated with supraspinatus:
otherwise of a young age, free of shoulder pain, and in a
middle deltoid (S:D) ratio
total good health. Finally, since our experimental study
Variables Coefficient 95% con- Stan- P-value
fidence dard included normal healthy participants with a specified age
interval error and gender, whether this finding will translate to clini-
Shoulder position -0.05 -0.7, -0.03 0.11 < 0.001* cal applications among patients with chronic shoulder
Body weight (kg) -0.02 -0.3, -0.00 0.01 0.020* pain and a suspected SSP tear remains unclear. In order
Scapular plane -0.04 -0.8, -0.00 0.02 0.042* to optimize the clinical translation of our study, we con-
(degrees) ducted both isometric testing and maximum voluntary
Constant 4.6 3.15, 6.04 0.74 < 0.001* contraction against a standardized load with accordance
Adjusted R-square 0.0545 results. Nevertheless, further clinical testing in terms of
P-value from the < 0.0001* diagnostic accuracy is still required.
F-test
*Significant at level < 0.01
Conclusion
Application of the SSP strength test in the shoulder posi-
Our study revealed that the most specific shoulder tion of 30° shoulder abduction combined with 30° hori-
position testing for isolated SSP activity is 30° shoulder zontal flexion and external humeral rotation is the best
abduction combined with 30° horizontal flexion and position to isolate the abducting activity of the SSP from
external humeral rotation. To apply these significant find- the abducting activity of the deltoid. This finding might
ings to clinical management, the SSP strength test in this potentially help physicians seeking to diagnose patients
preferred shoulder position would provide more spe- with chronic shoulder pain and a suspected SSP tear
cific results and the needs of further investigation, such condition.
Kijkunasathian et al. BMC Musculoskeletal Disorders (2023) 24:255 Page 11 of 12

List of Abbreviations elevation task based on activities of daily living: an electromyographic study.
SSP supraspinatus J Orthop Res. 2012;30(1):53–60.
EC empty can test 5. Hegedus EJ, Goode A, Campbell S, Morin A, Tamaddoni M, Moorman CT 3, et
EMG electromyographic al. Physical examination tests of the shoulder: a systematic review with meta-
S:D ratio supraspinatus to middle deltoid ratio analysis of individual tests. Br J Sports Med. 2008;42(2):80–92. discussion.
BMI body mass index 6. Kelly SM, Brittle N, Allen GM. The value of physical tests for subacromial
MVIC maximum voluntary isometric contraction impingement syndrome: a study of diagnostic accuracy. Clin Rehabil.
%sEMG standardized weighted EMG 2010;24(2):149–58.
7. Jobe FW, Moynes DR. Delineation of diagnostic criteria and a rehabilitation
program for rotator cuff injuries. Am J Sports Med. 1982;10(6):336–9.
Supplementary Information 8. Kelly BT, Kadrmas WR, Speer KP. The manual muscle examination for rota-
The online version contains supplementary material available at https://doi. tor cuff strength. An electromyographic investigation. Am J Sports Med.
org/10.1186/s12891-023-06372-3. 1996;24(5):581–8.
9. Malanga GA, Jenp YN, Growney ES, An KN. EMG analysis of shoulder position-
Supplementary Material 1 ing in testing and strengthening the supraspinatus. Med Sci Sports Exerc.
1996;28(6):661–4.
10. Reinold MM, Macrina LC, Wilk KE, Fleisig GS, Dun S, Barrentine SW, et al.
Acknowledgements Electromyographic analysis of the supraspinatus and deltoid muscles during
The authors would like to thank Department of Orthopedics, Faculty of 3 common rehabilitation exercises. J Athl Train. 2007;42(4):464–9.
Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand for all 11. Boettcher CE, Ginn KA, Cathers I. Standard maximum isometric voluntary
of the kindly help and permission to carry out the study. contraction tests for normalizing shoulder muscle EMG. J Orthop Res.
2008;26(12):1591–7.
Author Contribution 12. Park HB, Yokota A, Gill HS, El Rassi G, McFarland EG. Diagnostic accuracy of
CK conceptualized and designed the study. CK and SN performed an clinical tests for the different degrees of subacromial impingement syn-
experiment and helped in the acquisition of data. CK, PW and CV were drome. J Bone Joint Surg Am. 2005;87(7):1446–55.
responsible for analysis and interpretation of data. CK and CV helped to draft 13. Boettcher CE, Ginn KA, Cathers I. The ‘empty can’ and ‘full can’ tests do not
the manuscript. CV helped to revise the manuscript critically for important selectively activate supraspinatus. J Sci Med Sport. 2009;12(4):435–9.
intellectual content. All authors read and approved the final manuscript. 14. Longo UG, Berton A, Ahrens PM, Maffulli N, Denaro V. Clinical tests for the
diagnosis of rotator cuff disease. Sports Med Arthrosc Rev. 2011;19(3):266–78.
Funding 15. Bak K, Sørensen AK, Jørgensen U, Nygaard M, Krarup AL, Thune C, et al.
No external funding was obtained for this study. The value of clinical tests in acute full-thickness tears of the supraspinatus
tendon: does a subacromial lidocaine injection help in the clinical diagnosis?
Data Availability A prospective study. Arthroscopy. 2010;26(6):734–42.
The datasets generated and/or analysed during the current study are not 16. Ostör AJ, Richards CA, Tytherleigh-Strong G, Bearcroft PW, Prevost AT, Speed
publicly available due to limitations of ethical approval involving the patient CA, et al. Validation of clinical examination versus magnetic resonance imag-
data and anonymity but are available from the corresponding author on ing and arthroscopy for the detection of rotator cuff lesions. Clin Rheumatol.
reasonable request. 2013;32(9):1283–91.
17. Chalmers PN, Cvetanovich GL, Kupfer N, Wimmer MA, Verma NN, Cole BJ,
et al. The champagne toast position isolates the supraspinatus better than
Declarations the Jobe test: an electromyographic study of shoulder physical examination
tests. J Shoulder Elbow Surg. 2016;25(2):322–9.
Ethics approval and consent to participate 18. Colachis SC Jr, Strohm BR. Effect of suprascauular and axillary nerve blocks on
This experimental study was performed in accordance with the Declaration muscle force in upper extremity. Arch Phys Med Rehabil. 1971;52(1):22–9.
of Helsinki and was received the written approval by the Institutional Review 19. Ruckstuhl H, Krzycki J, Petrou N, Favre P, Horn T, Schmid S, et al. Shoulder
Boards in Mahidol University (certificate of approval no. MURA2017/582). All abduction moment arms in three clinically important positions. J Shoulder
participants signed the informed consent from before participate. Elbow Surg. 2009;18(4):632–8.
20. Short N, Mays M, Ford R, Fahrney E. Proposed method for goniomet-
Consent for publication ric measurement of scapular protraction and retraction. J Hand Ther.
Not applicable. 2021;34(1):148–50.
21. Sakaki Y, Kaneko F, Watanabe K, Kobayashi T, Katayose M, Aoki N, et al. Effects
Competing interests of different movement directions on electromyography recorded from the
The authors declare that they have no competing interests. shoulder muscles while passing the target positions. J Electromyogr Kinesiol.
2013;23(6):1362–9.
Received: 17 December 2022 / Accepted: 25 March 2023 22. Pennock AT, Pennington WW, Torry MR, Decker MJ, Vaishnav SB, Provencher
MT, et al. The influence of arm and shoulder position on the bear-hug,
belly-press, and lift-off tests: an electromyographic study. Am J Sports Med.
2011;39(11):2338–46.
23. Waite DL, Brookham RL, Dickerson CR. On the suitability of using surface elec-
trode placements to estimate muscle activity of the rotator cuff as recorded
by intramuscular electrodes. J Electromyogr Kinesiol. 2010;20(5):903–11.
References 24. Kelly BT, Williams RJ, Cordasco FA, Backus SI, Otis JC, Weiland DE, et al. Dif-
1. Yamamoto A, Takagishi K, Osawa T, Yanagawa T, Nakajima D, Shitara H, et al. ferential patterns of muscle activation in patients with symptomatic and
Prevalence and risk factors of a rotator cuff tear in the general population. J asymptomatic rotator cuff tears. J Shoulder Elbow Surg. 2005;14(2):165–71.
Shoulder Elbow Surg. 2010;19(1):116–20. 25. Hislop HJ, Avers D, Brown M, Daniels L. Daniels and Worthingham’s muscle
2. Ellman H. Diagnosis and treatment of incomplete rotator cuff tears.Clin testing: techniques of manual examination and performance testing. 2014.
Orthop Relat Res. 1990(254):64–74. 26. Kolber MJ, Fuller C, Marshall J, Wright A, Hanney WJ. The reliability and
3. Gschwend N, Ivosević-Radovanović D, Patte D. Rotator cuff tear–relationship concurrent validity of scapular plane shoulder elevation measurements
between clinical and anatomopathological findings. Arch Orthop Trauma using a digital inclinometer and goniometer. Physiother Theory Pract.
Surg. 1988;107(1):7–15. 2012;28(2):161–8.
4. Hawkes DH, Alizadehkhaiyat O, Fisher AC, Kemp GJ, Roebuck MM, Frostick SP. 27. Willardson JM. A brief review: factors affecting the length of the rest interval
Normal shoulder muscular activation and co-ordination during a shoulder between resistance exercise sets. J Strength Cond Res. 2006;20(4):978–84.
Kijkunasathian et al. BMC Musculoskeletal Disorders (2023) 24:255 Page 12 of 12

28. Minns Lowe CJ, Moser J, Barker K. Living with a symptomatic rotator cuff 32. Wickham J, Pizzari T, Stansfeld K, Burnside A, Watson L. Quantifying ‘normal’
tear ‘bad days, bad nights’: a qualitative study. BMC Musculoskelet Disord. shoulder muscle activity during abduction. J Electromyogr Kinesiol.
2014;15(1):228. 2010;20(2):212–22.
29. Jain NB, Wilcox RB 3rd, Katz JN, Higgins LD. Clinical examination of the rotator 33. Semciw AI, Neate R, Pizzari T. A comparison of surface and fine wire EMG
cuff. Pm r. 2013;5(1):45–56. recordings of gluteus medius during selected maximum isometric voluntary
30. Nazarian LN, Jacobson JA, Benson CB, Bancroft LW, Bedi A, McShane JM et contractions of the hip. J Electromyogr Kinesiol. 2014;24(6):835–40.
al. Imaging algorithms for evaluating suspected rotator cuff disease: Society
of Radiologists in Ultrasound consensus conference statement. Radiology.
2013;267(2):589 – 95.
31. Escamilla RF, Yamashiro K, Paulos L, Andrews JR. Shoulder muscle activity Publisher’s Note
and function in common shoulder rehabilitation exercises. Sports Med. Springer Nature remains neutral with regard to jurisdictional claims in
2009;39(8):663–85. published maps and institutional affiliations.