Printer Friendly

The within-day reliability of scapular and shoulder EMG measurements in asymptomatic individuals during shoulder abduction.


Electromyography (EMG) is used as a tool to assess muscle activity levels and recruitment patterns of muscles during functional activities (Konrad, 2005). Surface EMG (sEMG) is a non-invasive technique where electrodes are placed directly onto the skin and over the belly of the muscle that is investigated (Konrad, 2005). During joint movement or isometric muscle contraction, the electrodes detect electrical discharges from active motor units that can be used to assess neuromuscular disorders or as a research tool to identify abnormalities with motor control (Chowdhury et al., 2013).

Current literature regarding the reliability of EMG to the upper limb, specific to the shoulder complex is very limited. One study suggests that isometric maximal contraction, as a normalisation procedure, is linked to high EMG reliability for the upper limb muscles (Rota, Rogowski, Champely, & Hautier, 2013). Most published studies have assessed the reliability of EMG for measuring muscle activity of the trunk in patients with lower back pain (Dankaerts, O'Sullivan, Burnett, Straker, & Danneels, 2004) or lower limb muscles, such as vastus medialis obliquus (VMO), in patients with knee disorders (Worrell, Crisp, & LaRosa, 1998). Within-day EMG testing has excellent reliability (intraclass correlation coefficient (ICC) 0.91) in contrast to between-day reliability, which showed to have lower reliability (ICC 0.70) for vastus lateralis and VMO (Worrell et al., 1998). This variance could be due to several factors, such as, changes in the strength of contraction between tests and different placement of electrodes on different days (Oskouei, Paulin, & Carman, 2013; Worrell et al., 1998).

Findings from previous studies suggest that the method used for normalising EMG measurements can impact on reliability of measurements. Both submaximal and maximal methods were found to be linked to excellent reliability of EMG recordings (Oskouei et al., 2013). Normalising EMG recordings is not only essential for obtaining reliable measurements, but also for allowing appropriate interpretation of the EMG parameters (Burden, 2010).

The purpose of this study was to assess the within-day reliability of scapular and shoulder muscle EMG measurements in asymptomatic individuals as the evidence surrounding this topic is very limited. Findings from this study will help to interpret data from future studies assessing EMG recording when measuring muscle activity levels and recruitment patterns of scapular and shoulder muscles within the same day.


Study Design

This was a repeated measure cross-sectional study. Scapular and shoulder muscle activity was recorded using surface electromyography (sEMG). Muscles measured included the upper and lower trapezius; anterior, middle and posterior deltoids; supraspinatus; infraspinatus; and serratus anterior muscles. Each participant performed 10 repetitions of shoulder abduction with their dominant arm. The same task was then re-tested 10 minutes later.


Twenty asymptomatic individuals from the local community were included in this study. Participants signed an informed consent form prior to taking part in the study. A screening examination involved the cervical spine and shoulder. This involved full cervical active range of motion with overpressure, active shoulder elevation with overpressure, and maximum voluntary isometric shoulder internal and external rotation strength testing. Participants were excluded from the study if they presented with any pain or discomfort during these tests. The study was approved by the University of Otago Ethics Committee (reference number: H15/020).


We used a 16-channel wireless Noraxon TeleMyo 2400T G2 (Noraxon USA Inc., Arizona, USA) with a 3000 Hz sampling frequency, and gain of 500 to record muscle activity. Raw sEMG signals were sent wirelessly from the transmitter to a Noraxon TeleMyo EMG receiver. To record shoulder and scapular muscle activity, we used disposable, self-adhesive surface Ag/AgCl electrodes (Product SP-00-S/50, Ambu, DK-2750 Ballerup, Denmark). Electrodes were placed on the upper and lower trapezius; supraspinatus; infraspinatus; anterior, middle and posterior deltoid; and serratus anterior muscles.

We followed the Surface Electromyography for the Non-Invasive Assessment of Muscles (SENIAM) guidelines for electrode placement on upper and lower trapezius, anterior, middle and posterior deltoid muscles, and serratus anterior muscles (Hermens et al., 1999). We followed recommendations and descriptions by Criswell (Criswell, 2010) and Waite et al. (Waite, Brookham, & Dickerson, 2010), for electrode placement on the supraspinatus and infraspinatus muscles. Surface electrodes were placed parallel with muscle fibres, with a 2 cm interelectrode distance (Criswell, 2010; Hermens et al., 1999). We used the spinous process of the seventh cervical vertebrae for the reference electrode (Hermens et al., 1999).

We monitored the upper limb movements using a 3D motion analysis system (Motion Analysis Corporation, Santa Rosa, CA), with a frequency sample of 120 Hz. Reflective markers were placed over the acromion, lateral and medial epicondyles. During data processing, we monitored upper limb movements to identify concentric and eccentric phases of movement. We synchronised kinematic and EMG measurements using a trigger system.


To reduce EMG baseline noise, the participant's skin was shaved, wiped with a coarse towel and rubbed with an alcohol wipe. This was repeated several times until skin impedance was verified as less than 5kQ using a standard voltmeter, before placement of the electrodes. After this procedure, maximal isometric voluntary contraction (MIVC) was performed and measured for each individual muscle being monitored. The MIVC was then used to normalise EMG data gained during the study.

Following a rest period of 5 minutes, participants were asked to elevate their arms in the coronal plane 10 times. Participants were asked to elevate their arm until the end of the available active range of motion, without any compensatory movements. To standardise movement velocity, a metronome was used and set at 30 beats/min. This procedure was repeated 10 minutes later.

Data Processing

Processing of sEMG data was performed with MyoResearch XP Master Edition Software. Firstly, sEMG electrocardiogram artefact spikes were removed and the signal was processed using root mean square (RMS), with an average window of 50 ms. During shoulder abduction trials, sEMG data was normalised using the MIVC of muscle activity (during a 50 ms window of the 5s contraction). sEMG and kinematic data was processed using a tailored code written in MATLAB 7.12 (Mathworks, Inc., USA). EMG data were used to resample the kinematic data. The concentric and eccentric phases of shoulder abduction were identified by calculating the midpoint from the medial and lateral epicondyle reflective markers. The start and end of the concentric and eccentric phases were determined using the maximum and minimum values on the vertical axis of the midpoint. For each phase of abduction, the normalised mean muscle activity was calculated for each trial, for all participants. Following this, the mean for each of the 10 trials was calculated for every individual.

Primary outcome Measures

The mean level of muscle activity and standard deviation across the 10 trials for each monitored muscle was calculated and expressed as a percentage of the muscle-specific MIVC magnitudes.

Statistical Analyses

All statistical analyses were performed using SPSS software.

Within-Session Reliability of Muscle Activity Levels

EMG data were used to assess the within-day reliability of muscle activity between two measurements. The intraclass correlation coefficient (two-way mixed model, consistency definition ICC) was used to assess the within-session reliability (Hsu, Ho, Ho, & Hedman, 2000). The mean value of the 10 repetitions performed, for each muscle, for baseline and follow up measurements was used as input data for this model. It was assumed that the clinical researcher was the single rater. The standard error of measurement (SEM) was calculated from the data obtained from the 10 repetitions of shoulder abduction performed by participants, and was calculated using the square root of the error mean square term, obtained from the ANOVA test. This is beneficial as the SEM can be estimated independently from the ICC value (Weir, 2005).

Interpretation of Results

For the purposes of this study, ICC values were analysed using the categories: ICC values <0.40 suggested poor reliability;

ICC values ranging from 0.40 to 0.75 suggested fair to good reliability; >0.75 suggested excellent reliability (Fleiss, 1999). Currently, there are no clear guidelines explaining how to interpret SEM. Due to this, SEM values were qualitatively discussed.


Twenty participants took part in this study (6 males/12 females). The participants' anthropometric characteristics are reported on Table 1.

The concentric phase presented ICC values ranging from 0.87 to 0.98 (Table 2). As shown in Table 3, the SEM values ranged from 8.6% MVIC to 18.9% MVIC. With reference to the eccentric phase, ICC values ranged from 0.65 to 0.97 (Table 2). The SEM values ranged from 9.8% MVIC to 24.4% MVIC (Table 3).

Values for mean activity as a percentage of the MVIC during the concentric phase range from 10.2% to 33.01% and for the eccentric phase, range from 4.9% to 14.4% (Table 3).

The SEM values expressed as a percentage of the mean for the concentric phase range from 1.3% to 5.0% and for the eccentric phase, range from 1.0% to 12.0% (Table 3).


This study assessed the within-day reliability of scapular and shoulder EMG measurements in asymptomatic individuals. Our results suggest that EMG measurements have excellent reliability for concentric contractions and good to excellent reliability during the eccentric phase of shoulder abduction. The upper trapezius, lower trapezius, infraspinatus, anterior deltoids, middle deltoids, posterior deltoids, pectoralis major and latissimus dorsi muscles had excellent reliability for both the eccentric and concentric phases of muscle contractions, with reference to their ICC values. Moreover, the SEM percentages of the MVIC for all of the muscle groups during the eccentric phase were larger compared to the concentric phase (Table 2). Therefore, the eccentric phase seems to be less reliable than the concentric phase during unloaded shoulder abduction in asymptomatic individuals.

Numerous studies support our findings on excellent reliability for the concentric phase of contraction. One study investigated test-retest reliability of EMG during maximal concentric knee extensions (Larsson, Karlsson, Eriksson, & Gerdle, 2003). The results showed excellent reliability for the vastus lateralis (ICC>0.82), vastus medialis (ICC>0.88) and rectus femoris (ICC>0.83) muscles. Larsson et al. (2003) used a very similar method to our study, using maximal muscle contractions to normalise EMG data. Another study investigated test-retest reliability of onset concentric and eccentric EMG activity in stair stepping tasks and showed excellent reliability (ICC 0.91) during concentric contractions, similar to our findings (Cowan, Bennell, & Hodges, 2000). Despite both these studies assessing the activity of lower limb muscles, they both had a focus on dynamic contractions similar to our study.

Previous studies reported EMG recordings to be reliable, but with different reliability scores being reported for concentric and eccentric phases of movement. One study reported excellent reliability scores for both eccentric and concentric data when monitoring scapular and shoulder muscles (Ribeiro, Day, & Dickerson, 2017). Another study assessed the reliability of quadriceps femoris muscle (Finucane, Rafeei, Kues, Lamb, & Mayhew, 1998), and reported good to excellent reliability, with eccentric contractions associated with higher ICC scores (similar to our findings). Our results, on the other hand, showed noticeably lower ICC values and higher SEM values for EMG activity levels during the eccentric phase of movement, when compared to the concentric data. Variability on EMG recordings may arise due to electrode placement, within- and between subject variability when recruiting muscles to perform a task. These factors might explain the different findings reported by previous studies and our study.

In contrast, some studies have identified the eccentric phase of contraction to be more reliable compared to the concentric phase. Finucane et al. (1997) observed superior intra-rater and inter-rater ICC values in participants when performing eccentric isokinetic knee extensor contractions. All eccentric results were rated as excellent (intra-rater results ranging from 0.84 to 0.97, and inter-rater results ranging from 0.78 to 0.90). In comparison, concentric values ranged from good to excellent (intra-rater results ranging from 0.62 to 0.91, and inter-rater results ranging from 0.64 to 0.96). Moreover, another study (Finucane et al., 1998) found similar results, i.e. higher ICC values (intra-rater results ranging from 0.84 to 0.97, and interrater results ranging from 0.78 to 0.90), and lower SEM values (0.17 for intra-rater data and 0.27 for inter-rater data) for the eccentric phase of movement. Lower ICC values (intra-rater results ranging from 0.62 to 0.91, and inter-rater results ranging from 0.66 to 0.96), and higher SEM values (0.33 for intrarater and 0.33 for inter-rater), were found for concentric data (Finucane et al., 1998).


One of the main limitations for the use of sEMG is the potential crosstalk between adjacent muscles. Previous studies have proposed that sEMG may detect signals generated by a number of concurrently active motor units near the electrode (De Luca & Merletti, 1988; Solomonow et al., 1994). Regarding scapular and shoulder muscles, there is concern about potential cross-talk between the supraspinatus and upper trapezius muscles. A previous study suggests minimal crosstalk between the upper trapezius and supraspinatus, and the trapezius (upper and middle fibres) and infraspinatus (Waite et al., 2010) muscles. It is suggested that sEMG overestimates fine-wire recordings for the supraspinatus muscle. However, this was not due to cross-talk between these muscles, but due to muscle fibre orientation being recorded by sEMG and fine-wire EMG, muscle size and location. It was found that the supraspinatus muscle activity recorded by sEMG is lower when measuring submaximal muscle exertion or when the muscle is the primary mover (e.g. shoulder abduction, as analysed in this study) (Allen, Brookham, Cudlip, & Dickerson, 2013; Waite et al., 2010). Another limitation of sEMG is muscle activity measurement variability during dynamic contractions. During dynamic contractions, there is potential for displacement of the electrode in regards to the muscle being recorded. Consequently, this could reduce the intensity of the signal being recorded, affecting the reliability of the data (Masso et al., 2010). Moreover, as only surface muscles can be measured and analysed using sEMG, this study does not provide findings on the shoulder complex as a whole (Masso et al., 2010).


In asymptomatic individuals, EMG measurements of the shoulder and scapular during the concentric phase of shoulder abduction had excellent reliability and for the eccentric phase had good to excellent reliability of results. This study adds to current knowledge regarding the reliability of EMG measurements to muscles of the upper limb and helps to better understand the use of EMG as a reliable tool in assessing muscle activity levels and recruitment patterns to the shoulder complex. To date, literature concerning the reliability of EMG measurements to the upper limb is very limited which allows for the opportunity of further research on this topic.


1. EMG is a reliable tool for assessing scapular and shoulder muscle activity pattern.

2. Reliability of EMG recordings for scapular and shoulder muscles was higher during the concentric phase of shoulder abduction.


No funding was obtained to undertake this study. Authors report no conflicts of interest. Daniel Cury Ribeiro is supported by The Sir Charles Hercus Health Research Fellowship--Health Research Council of New Zealand.


Daniel Cury Ribeiro, School of Physiotherapy, University of Otago, PO Box 56, Dunedin. 9016. Telephone: 00 64 3 479 7455. Email:


Allen, T R., Brookham, R. L., Cudlip, A. C., & Dickerson, C. R. (2013). Comparing surface and indwelling electromyographic signals of the supraspinatus and infraspinatus muscles during submaximal axial humeral rotation. J Electromyogr Kinesiol, 23(6), 1343-1349. doi:10.1016/j. jelekin.2013.08.002.

Burden, A. (2010). How should we normalize electromyograms obtained from healthy participants? What we have learned from over 25 years of research. J Electromyogr Kinesiol, 20(6), 1023-1035. doi:10.1016/j. jelekin.2010.07.004.

Chowdhury, R. H., Reaz, M. B., Ali, M. A., Bakar, A. A., Chellappan, K., & Chang, T. G. (2013). Surface electromyography signal processing and classification techniques. Sensors (Basel), 73(9), 12431-12466. doi:10.3390/s130912431.

Cowan, S. M., Bennell, K. L., & Hodges, P W. (2000). The test-retest reliability of the onset of concentric and eccentric vastus medialis obliquus and vastus lateralis electromyographic activity in a stair stepping task. Physical Therapy in Sport, 7(4), 129-136. doi: ptsp.2000.0036.

Criswell, E. (2010). Cram's Introduction to Surface Electromyography: Jones & Bartlett Publishers.

Dankaerts, W., O'Sullivan, P. B., Burnett, A. F., Straker, L. M., & Danneels, L. A. (2004). Reliability of EMG measurements for trunk muscles during maximal and sub-maximal voluntary isometric contractions in healthy controls and CLBP patients. Journal of Electromyography and Kinesiology, 74(3), 333-342.

De Luca, C. J., & Merletti, R. (1988). Surface myoelectric signal crosstalk among muscles of the leg. Electroencephalography and clinical neurophysiology, 69(6), 568-575.

Finucane, S., Rafeei, T., Kues, J., Lamb, R., & Mayhew, T. (1998). Reproducibility of electromyographic recordings of submaximal concentric and eccentric muscle contractions in humans. Electroencephalography and Clinical Neurophysiology/Electromyography and Motor Control, 709(4), 290-296.

Fleiss, J. L. (1999). The design and analysis of clinical experiments (Wiley classics library ed.). New York: Wiley.

Hermens, H. J., Freriks, B., Merletti, R., Stegeman, D., Blok, J., Rau, G., ... Hagg, G. (1999). European recommendations for surface electromyography. Roessingh research and development, 8(2), 13-54.

Hsu, A.-T, Ho, L., Ho, S., & Hedman, T. (2000). Immediate response of glenohumeral abduction range of motion to a caudally directed translational mobilization: a fresh cadaver simulation. Archives of physical medicine and rehabilitation, 87(11), 1511-1516.

Konrad, P (2005). The abc of emg. A practical introduction to kinesiological electromyography, 7, 30-35.

Larsson, B., Karlsson, S., Eriksson, M., & Gerdle, B. (2003). Test-retest reliability of EMG and peak torque during repetitive maximum concentric knee extensions. Journal of Electromyography and Kinesiology, 73(3), 281-287.

Masso, N., Rey, F., Romero, D., Gual, G., Costa, L., & German, A. (2010). Surface electromyography applications. Apunts Medicina de l" Esport (English Edition), 45(166), 127-136.

Oskouei, A. H., Paulin, M. G., & Carman, A. B. (2013). Intra-session and inter-day reliability of forearm surface EMG during varying hand grip forces. Journal of Electromyography and Kinesiology 23(1), 216-222.

Ribeiro, D. C., Day, A., & Dickerson, C. R. (2017). Grade-IV inferior glenohumeral mobilization does not immediately alter shoulder and scapular muscle activity: a repeated-measures study in asymptomatic individuals. Journal of Manual & Manipulative Therapy, 25(5), 260-269. do i:10.1080/10669817.2017.1290310.

Rota, S., Rogowski, I., Champely, S., & Hautier, C. (2013). Reliability of EMG normalisation methods for upper-limb muscles. Journal of sports sciences, 31(15), 1696-1704.

Solomonow, M., Baratta, R., Bernardi, M., Zhou, B., Lu, Y., Zhu, M., & Acierno, S. (1994). Surface and wire EMG crosstalk in neighbouring muscles. Journal of Electromyography and Kinesiology 4(3), 131-142.

Waite, D. L., Brookham, R. L., & Dickerson, C. R. (2010). On the suitability of using surface electrode placements to estimate muscle activity of the rotator cuff as recorded by intramuscular electrodes. Journal of Electromyography and kinesiology, 20(5), 903-911.

Weir, J. P (2005). Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM. Journal of strength and conditioning research, 19(1), 231.

Worrell, T. W., Crisp, E., & LaRosa, C. (1998). Electromyographic reliability and analysis of selected lower extremity muscles during lateral step-up conditions. Journal of athletic training, 33(2), 156.


This study won the ML Roberts prize awarded for the best 4th year undergraduate research project at the School of Physiotherapy University of Otago in 2017.

Aishwarya Grime * BPhty

School of Physiotherapy, University of Otago, Dunedin, New Zealand.

Sophie Daines * BPhty

School of Physiotherapy, University of Otago, Dunedin, New Zealand.

Lucy Pringle * BPhty

School of Physiotherapy, University of Otago, Dunedin, New Zealand.

Lee Heang * BPhty

School of Physiotherapy, University of Otago, Dunedin, New Zealand.

Daniel Cury Ribeiro BPthy MSc, PhD

Senior Lecturer, Centre for Physical Activity, Heath, and Rehabilitation Research, School of Physiotherapy, University of Otago,

Dunedin, New Zealand.

* Undergraduate students at the time of the study.
Table 1: Demographic data for the 20 participants.

Participants                      Mean     SD       Range

Age (years)                       28.1    7.4       22-55
Height (m)                        168.6   9.5    152.5-189.1
Weight (kg)                       67.1    14.5    54-118.7
Body Mass Index (kg/[m.sup.2])    23.6    3.4     19.4-33.3
Gender (M/F)                      6/14

Notes: SD, standard deviation; M, male; F, female.

Table 2: Within-day reliability (intraclass correlation coefficient)
of muscle activity measurements between baseline and follow-up.

Muscles                Concentric ICC         Eccentric ICC

                          (95% CI)              (95% CI)

Upper Trapezius      0.97 (0.92 to 0.99)   0.83 (0.57 to 0.93)
Lower Trapezius      0.94 (0.84 to 0.98)   0.78 (0.44 to 0.91)
Supraspinatus        0.91 (0.77 to 0.96)   0.65 (0.11 to 0.86)
Infraspinatus        0.98 (0.95 to 0.99)   0.93 (0.81 to 0.97)
Anterior Deltoids    0.87 (0.67 to 0.95)   0.79 (0.47 to 0.92)
Middle Deltoids      0.91 (0.77 to 0.96)   0.80 (0.49 to 0.92)
Posterior Deltoids   0.96 (0.90 to 0.98)   0.89 (0.71 to 0.95)
Pectoralis Major     0.97 (0.92 to 0.99)   0.94 (0.84 to 0.98)
Latissimus Dorsi     0.96 (0.91 to 0.99)   0.97 (0.93 to 0.99)

Notes: ICC, intraclass correlation coefficient; CI, Confidence
Interval; SEM, standard error of measurement; MVIC, maximal
voluntary isometric contraction.

Table 3: Within-day reliability (standard error of measurement) of
muscle activity measurements between baseline and follow-up.

Muscles                Concentric phase

                        Mean activity     SEM        SEM
                        (% MVIC)          (%MVIC)    (%mean)

Upper Trapezius          33.01             8.6        2.8
Lower Trapezius           28.4             12.6       3.6
Supraspinatus             26.6             12.6       3.4
Infraspinatus             19.8             8.8        1.7
Anterior Deltoid          43.3             11.5       5.0
Middle Deltoid            27.3             11.2       3.1
Posterior Deltoid         14.1             9.4        1.3
Pectoralis Major          12.6             18.9       2.4
Latissimus Dorsi          10.2             10.2       1.4

Muscles                Eccentric phase

                       Mean activity       SEM        SEM
                        (% MVIC)           (%MVIC)    (%mean)

Upper Trapezius           12.0             19.8       12.0
Lower Trapezius           11.4             22.7       2.6
Supraspinatus             10.9             18.9       2.1
Infraspinatus             9.4              14.7       1.4
Anterior Deltoid          14.4             20.1       2.9
Middle Deltoid            9.9              16.4       1.6
Posterior Deltoid         5.8              15.0       1.0
Pectoralis Major          4.9              24.4       1.2
Latissimus Dorsi          5.1              9.8        1.0

Notes: ICC, intraclass correlation coefficient; CI, Confidence
Interval; SEM, standard error of measurement; MVIC, maximal voluntary
isometric contraction.
COPYRIGHT 2018 New Zealand Society of Physiotherapists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:electromyography
Author:Grime, Aishwarya; Daines, Sophie; Pringle, Lucy; Heang, Lee; Ribeiro, Daniel Cury
Publication:New Zealand Journal of Physiotherapy
Article Type:Report
Date:Jul 1, 2018
Previous Article:Organisational views on health care access for haua (disabled) Maori in Murihiku (southland), Aotearoa New Zealand: a mixed methods approach.
Next Article:What helps or hinders clinicians in their decision-making processes when using or prescribing mHealth apps in practice? an exploratory study.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters