Accepted Manuscript

Competitive swimmers with hypermobility have strength and fatigue deficits in

shoulder medial rotation

Behnam Liaghat, Birgit Juul-Kristensen, Thomas Frydendal, Camilla Marie

Larsen, Karen Søgaard, Aki Ilkka Tapio Salo

PII: S1050-6411(17)30364-4
DOI: https://doi.org/10.1016/j.jelekin.2018.01.003
Reference: JJEK 2153

To appear in: Journal of Electromyography and Kinesiology

Received Date: 10 September 2017

Revised Date: 8 December 2017
Accepted Date: 9 January 2018

Please cite this article as: B. Liaghat, B. Juul-Kristensen, T. Frydendal, C. Marie Larsen, K. Søgaard, A. Ilkka Tapio
Salo, Competitive swimmers with hypermobility have strength and fatigue deficits in shoulder medial rotation,
Journal of Electromyography and Kinesiology (2018), doi: https://doi.org/10.1016/j.jelekin.2018.01.003

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Competitive swimmers with hypermobility have strength and fatigue deficits in shoulder

medial rotation

Manuscript word count: 4710; abstract word count: 200

Behnam Liaghata,b, Birgit Juul-Kristensena, Thomas Frydendala, Camilla Marie Larsena,c,, Karen

Søgaarda, & Aki Ilkka Tapio Salob

a
Department of Sports Science and Clinical Biomechanics, University of Southern Denmark,

Odense, Denmark, bDepartment for Health, University of Bath, Bath, United Kingdom, cHealth

Sciences Research Centre, University College Lillebaelt, Odense, Denmark.

Corresponding author

Behnam Liaghat, Hjallesevej 117, 1.TH, 5230 Odense M, Denmark.

Email: [email protected], Phone: +45-26826801

Co-authors contact details

Birgit Juul-Kristensen, Department of Sports Science and Clinical Biomechanics, University of

Southern Denmark, Campusvej 55, 5230, Odense M, Denmark.

Thomas Frydendal, Physio- and Occupational Therapy, Lillebaelt Hospital – Vejle, Beriderbakken

4, 7100 Vejle, Denmark.

Camilla Marie Larsen, Department of Sports Science and Clinical Biomechanics, University of

Southern Denmark, Campusvej 55, 5230, Odense M, Denmark.

Karen Søgaard, Department of Sports Science and Clinical Biomechanics, University of Southern

Denmark, Campusvej 55, 5230, Odense M, Denmark.

Aki Salo, Department for Health, University of Bath, Applied Biomechanics Suite 1.309, Claverton

Down, Bath, BA2 7AY, United Kingdom.

1
1. Introduction

Generalised Joint Hypermobility (GJH) is described as lack of structural stability of the

passive system due to ligamentous or capsular looseness. The condition implies increased risk of

traumatic and non-traumatic shoulder dislocations (Cameron et al., 2010; Chahal et al., 2010), and

is a risk factor for overuse shoulder injuries (Zemek & Magee, 1996). The reason for this is not yet

known, but is probably due to decreased strength, muscle activity and coordination, and eventually

increased fatigue. Decreased isokinetic strength has been found in children and adults with

symptomatic and non-symptomatic GJH and knee hypermobility (Juul-Kristensen et al., 2012). In

contrast, recent studies of children with non-symptomatic GJH showed no reduced maximum

isometric knee strength and hop length (Jensen et al., 2013; Junge et al., 2015). In fact, these

studies reported that individuals with GJH use altered or compensatory muscle activation strategies

in both agonist and stabilizing leg muscles. To our knowledge no studies have reported shoulder

strength and muscle activity in individuals with GJH including hypermobile shoulders (GJHS).

Individuals with GHJS may also be characterised by multidirectional instability (MDI)

(Borsa et al., 2000), glenohumeral instability and/or increased risk of recurrent joint luxation

(Cameron et al., 2010; Chahal et al., 2010). In this aspect, previous findings indicate that these

individuals present more precisely with altered muscle activity of scapular stabilisers and rotator

cuff muscles (Barden et al., 2005), decreased shoulder muscle strength (Edouard et al., 2011),

scapular function (Struyf et al., 2011) and shoulder proprioception (Laudner et al., 2012). In line

with studies on MDI, it can be hypothesised that individuals with GJHS have functional deficits

and altered muscle activity in the shoulder; however, this remains scientifically unanswered.

Young competitive swimmers represent a group with high prevalence of GJHS (Junge et

al., 2016). This may be due to the demand of a greater range of shoulder motion to achieve a body

position that reduces drag and increases stroke length, which ultimately results in better swimming

performance (Wanivenhaus et al., 2012). Having GJHS in addition to being exposed to repetitive

shoulder rotation movements during competitive swimming, these individuals may be predisposed

2
to muscle-tendon overload, muscle fatigue and pain. Thus, the aim of this study was to investigate

whether young competitive swimmers with GJHS have reduced shoulder strength, increased

isokinetic fatigue development and altered muscle activity during swimming-related movements of

shoulder rotation.

2. Methods

2.1 Study Design and Recruitment

This study was cross-sectional comparing 13-17 years old competitive swimmers (highest national

level) with GJHS and control swimmers without GJHS, individually matched on age, sex and

swimming club. Participants were recruited from local sports clubs by initial email and phone

contacts with coaches and parents. Study procedures were approved by the Research Ethics

Approval Committee for Health at the University of Bath (EP 14/15 175). Participants and their

parent/guardian provided written informed consent according to the Helsinki Declaration.

Data were collected during two sessions: (i) the screening session consisting of clinical tests for

GJHS (approximately 10 min), performed by a second trained physiotherapist to ensure blinding of

the principal investigator to the participants’ health status (controls or GJHS); and (ii) the test

session comprising isokinetic and EMG measurements (approximately 90 min) conducted on

another day.

An element of shoulder laxity and GJHS is seen in many competitive swimmers, and both a

combination of acquired and inherent factors contribute to shoulder laxity in swimmers. Due to the

study design the included swimmers had to be inherently joint hypermobile and not only present

with acquired hypermobility due to the swimming exposure, hence the Beighton tests for GJH were

used during the screening session. Beighton tests have been found reproducible for testing GJH

(Juul-Kristensen et al., 2007) and include nine tests: apposition of the thumbs, hyperextension of

elbows and knees, dorsalflexion of the fifth fingers, all bilaterally, and forward bending in

standing. Each positive test scores one point with nine points being the highest possible score.

3
Since the Beighton tests do not specifically include the shoulder joint, a shoulder test included in a

Spanish test battery (Rotes-Querol, 1957) for GJH was added to ensure shoulder hypermobility.

The test was shoulder lateral rotation (positive score >90°) with the upper arm in neutral along the

side of the body, previously found to have satisfactory reliability (Juul-Kristensen et al., 2007).

The group with GJHS was defined with Beighton score ≥5 (Juul-Kristensen et al., 2007),

and positive shoulder hypermobility in at least one shoulder. Inclusion criteria for controls were a

Beighton score ≤3 and no shoulder hypermobility. Further inclusion criteria for both groups were:

being swimmer, having normal training and competition activity within the latest seven days, and

being matched for age and sex. Exclusion criteria were previous serious trauma to the upper

extremity, shoulder surgery, diagnoses of Ehlers Danlos syndrome or Marfan syndrome.

During the actual test session, the participants firstly answered questionnaires about

training activity, competitive swimming, other sports activities, previous injuries, perceived

shoulder instability by the Western Ontario Shoulder Instability Index (WOSI) questionnaire and

pain intensity measured by pain rating on Visual Analogue Scale (VAS) for current pain, pain

during the latest 24 hours, and pain during the latest seven days (Table 1). Subsequently,

information on anthropometric data was collected followed by the EMG electrode placement

procedure. Then the participants completed a standardised warm up program for 10 minutes,

comprising unilateral and bilateral shoulder movements (10 repetitions of flexion, extension,

horizontal abduction and adduction), scapular protractions against a wall, and standing push-ups

against a 90-cm high table, followed by simultaneous EMG recordings during maximum voluntary

isometric contraction (MVIC) tests, and isokinetic measurements. After the isokinetic testing

procedure, clinical tests for glenohumeral instability were performed.

2.2 Isokinetic Measurement

The isokinetic concentric shoulder medial and lateral rotation measurements were

performed on a calibrated Cybex NORM dynamometer (Cybex Inc., Ronkonkoma, New York,

4
USA) with the dominant arm defined as the side used for handwriting. One swimmer had GJHS on

his non-dominant shoulder, in which case the non-dominant shoulder was tested on him and his

matched control. The test position was selected to be close to the freestyle swimming stroke. The

participants were prone lying with 90° shoulder abduction and 90° elbow flexion (Figure 1), and

were fixated with belts around the mid lumbar spine and 10 centimetres above knee level. Centre of

rotation for the shoulders were placed to be in line with the rotation arm of the dynamometer. To

minimise risk of injuries in end range, total range of motion was set to 95°, ranging from 35° of

medial rotation to 60° of lateral rotation.

Insert Figure 1 about here…

Five repetitions of maximum shoulder rotation strength were performed at 60°/s, and 10

repetitions at 180°/s, with 60 seconds of rest periods between each test, and the current velocities

were selected as those most similar to the estimated velocities performed during swimming (Bak &

Magnusson, 1997). Before each test, participants had five trials to familiarise with movements and

velocities. No visual feedback was allowed, however, participants were encouraged to perform

maximally with verbal instructions. Outcome measures were peak torque (the maximal value of the

moment angle position curve) and maximum work (the repetition with highest value in torque x

angular displacement) for both directions at both velocities (60°/s and 180°/s) normalised to body

mass. Fatigue development in isokinetic strength was calculated as the decrease in work

(J/repetition) during repetitions from 2-10 at 180°/s.

2.3 Electromyography

During isokinetic testing, surface EMG (Telemyo DTS, Noraxon Inc. Scottsdale, USA)

was measured in the scapular stabilising muscles (mm. upper trapezius, lower trapezius and

serratus anterior), the primary medial rotation agonist (m. pectoralis major), and the primary lateral

rotation agonist muscle (m. infraspinatus). Once the exact point for electrode placement attachment

was identified, the skin was lightly shaved, rubbed and cleaned with alcohol to keep skin

impedance lower than 10 kOhm, measured with a digital multimeter. Bipolar electrodes (Ag/AgCl,

5
Ambu Blue Sensor, N-00-S/25, Ballerup, Denmark) were attached to the skin with a two-

centimetre inter-electrode distance, and in line with muscle fibres. The locations of electrodes were

as follows: upper trapezius, 20% medial to the mid distance between seventh cervical vertebra and

acromion’s lateral border (Holtermann et al., 2009); lower trapezius, 33% medial to the midpoint

between the medial border of the scapula and eighth thoracic vertebra (Holtermann et al., 2009);

serratus anterior, at the seventh rib, below the axilla, posterior to pectoralis major and anterior to

latissimus dorsi (Holtermann et al., 2010) in line with the xiphoid process and the axillary border;

infraspinatus, two and a half cm distal to the centre of the scapular spine (Barden et al., 2005);

pectoralis major, one third of the distance from tuberculum major to the xiphoid process (Pontillo

et al., 2007).

2.4 Maximum Voluntary Isometric Contraction

For each muscle, maximum voluntary electrical activity (MVE) signal was recorded during

MVIC in standardised anatomical positions. Upper trapezius was tested in standing with the arm in

90° elevation in scapula’s plane with the thumb pointing upwards, while performing elevation. For

serratus anterior, a similar testing position was used with the arm elevated to 135°. Lower

trapezius was tested in prone with the arm in 125° of abduction along muscle fibres, performing

horizontal abduction. Infraspinatus was tested in prone lying with the arm in 90° of abduction and

the elbow in 90° of flexion during performance of shoulder lateral rotation, and pectoralis major

was tested in supine lying with 90° of shoulder flexion during horizontal adduction performance.

An external load of 40 kg was used for resistance, and to stabilise the body a manual isometric

resistance was applied on the non-measured arm while the participants were fixated with belts in

the prone and supine lying positions. A total of three sets, with each contraction lasting five

seconds, were performed with 60 seconds rest period between each set.

2.5 Clinical Tests

Three clinical tests for glenohumeral instability were used, including Gagey hyperabduction, sulcus

and load-and-shift tests. The Gagey hyperabduction test for the inferior glenohumeral ligament was

6
considered positive with shoulder abduction beyond 105° (Gagey & Gagey, 2001). For the sulcus

test, multidirectional instability was defined as more than two centimeters widening between

acromion and the humeral head (Neer & Foster, 1980). In the load-and-shift test, severe ventral or

dorsal glenohumeral instability was defined as grade two or three (i.e. humeral head moved beyond

the glenoid labrum, and was relocated spontaneously or remained dislocated), on a scale from 0-3

(Bak & Faunø, 1997).

2.6 Analyses of EMG Signals

Raw EMG signals were amplified with gain 500 and bandpass filtered with 10-500 Hz

(Noraxon Inc. Scottsdale, USA). The analogue signal was recorded on a computer via laboratory

interface (CED Power 1401 16 bit, Spike2 software, Cambridge Electronic Design Limited, UK)

with analogue to digital converting at 1000 Hz. For medial and lateral rotation, muscle analyses

were performed with custom-made software (Hedera 3.0, The University of Southern Denmark). In

EMG signals, changed shoulder rotation directions were marked by triggers defining start/stop of

each medial and lateral rotation movement. EMG signals of each muscle during peak torque and

maximum work repetitions were normalised to the respective muscle’s MVE, defined as the

highest Root Mean Square (RMS) amplitude in a moving window of 100 ms across the whole

MVIC expressing relative MVE (%MVE).

2.7 Statistical Analysis

Baseline variables were tested for normality (Shapiro-Wilk, histograms and QQ-plots), and found

to be normally distributed except for pain intensity and WOSI. For demographic data, un-paired t-

tests were used to test for group differences on continuous data (Beighton score, age, height, body

mass, swimming competitive experience, swimming during practice and other sports activities),

while Mann-Whitney was used for non-normal data (pain intensity and WOSI), and Fisher’s exact

test for dichotomous data (Rotes-Querol and glenohumeral instability tests). For between-group

differences in the outcomes of peak torque, maximum work and %MVE, a linear regression model

was estimated adjusting for gender, age and body mass, without violating the assumption of

7
normality. For isokinetic fatigue (J/repetition) calculations during repetitions from 2-10 at 180°/s, a

linear regression mixed effect model was applied with age, gender and body mass as covariates,

using ID number as random factor. For the isokinetic fatigue development, a negative coefficient

denotes a decrease in maximum work.

Sample size was calculated to be minimum 16 per group based on previous isokinetic

values (Bak & Magnusson, 1997), with estimated standard deviation and minimum mean

difference of 0.10 Nm/kg and 0.07 Nm/kg , respectively, with  = 0.05 and =0.20. To

accommodate for drop-outs, 19 participants per group were recruited, however, there were no drop-

outs. P-values <0.05 were reported as statistically significant, while tendencies to significance were

defined as p-values >0.05 <0.10. All statistical analyses were performed using STATA (StataCorp,

2015, Stata Statistical Software: Release 14. College Station, TX: StataCorp LP.)

3. Results

In total, 97 competitive swimmers were screened for eligibility, of which 38 swimmers, 11 girls

and eight boys in each group, completed the study. The groups were comparable on demographics

(age, height, body mass, sports participation, previous and current pain levels, WOSI and clinical

tests for shoulder instability) except for the Beighton and Rotes-Querol tests, as expected due to the

study design (Table 1).

Insert Table 1 about here…

3.1 Isokinetic strength

In medial rotation at 60°/s, GJHS swimmers had significantly lower (12%) peak torque (0.53 vs

0.60 Nm/kg; p=0.047), and significantly lower (14%) maximum work (0.62 vs 0.71 J/kg; p=0.031)

compared with controls (Table 2).

Insert Table 2 about here…

8
3.2 Isokinetic fatigue

Based on linear regression models, swimmers with GJHS showed significantly larger isokinetic

fatigue development in medial rotation at 180°/s (-0.257 vs 0.064 J, SE=0.088; p=0.010) with a

between groups difference in time effects of 0.321 J/Repetition (SE= 0.124). There were no

significant group differences in lateral rotation (-0.428 vs -0.553 J, SE=0.097; p=0.362).

Insert Figure 2 about here…

3.3 Muscle activity

There were tendencies to lower %MVE (20%) for GJHS swimmers in infraspinatus in peak torque

(60°/s medial rotation) (29.1 ± 10.6 vs. 35.5 ± 11.4 %MVE; p=0.066), and lower %MVE (34%) in

maximum work in pectorals major (30.0 ± 22.1 vs. 42.3 ± 25.8 %MVE; p=0.092), but generally

there were no significant group differences in EMG activity (Figure 2).

4. Discussion

Swimmers with GJHS, despite having no formal diagnosis, displayed both lower isokinetic peak

torque and maximum work in medial rotation at 60°/s, and larger fatigue development in isokinetic

strength during medial rotation strength measurements at 180°/s. There were no significant group

differences in muscle activity during the corresponding isokinetic measurements at 60°/s.

4.1 Isokinetic strength

Velocities under 120°/s have been defined as those corresponding to strength (the amount

of force muscles can exert against an external load), while higher velocities such as 180 °/s have

been defined as corresponding to power (the ability to generate as much force and as fast as

possible) (St. Pierre et al., 1984). The significantly lower isokinetic performance at 60°/s (12-14%)

indicates that participants with GJHS lacked muscle strength rather than power. The lower medial

rotation strength in comparison (and no group differences in lateral rotation strength) was an

interesting finding, because the medial rotation movement is one of the main propulsive

9
movements throughout the acceleration phase in swimming strokes. In fact, up to 4000 strokes per

day in medial rotation with each arm are performed by high-level competitive swimmers (Bak &

Faunø, 1997). These findings suggest that swimmers with GJHS could exert less force during the

isokinetic test due to the musculoskeletal impairment itself, with the inclusion criteria of an

external shoulder rotation of more than 90°. Similar sized strength deficits in GJH (ranging from

18-19%), however, in knee strength, were found in 10-year old girls and women with GJH and

knee hypermobility (Juul-Kristensen et al., 2012). Since these two studies differ with respect to

body regions and athletic status (competitive swimmers vs non-athletes), comparison should be

made with caution. The reduced strength in the current study (12-14%) is in line with previous

studies of elite swimmers with shoulder pain (17.5% reduced medial rotation strength) compared

with those without pain (Bak & Faunø, 1997), and athletes with subacromial impingement (11%

reduced scapular retraction-protraction strength) compared with their non-injured shoulder (Cools

et al., 2004). The ‘safe’ limit of maximum 10% side difference before an increased risk of injuries

has been advocated for knees (Tol et al., 2014). Since the shoulder is anatomically and

biomechanically different from the weight bearing knee, it is unknown whether reduced shoulder

strength within the present level (12-14%) represents a risk factor for future injury development.

However, reduced strength in competitive sport may threaten joint integrity with potentially

detrimental consequences. Further, the present result of lower maximum work suggests that

swimmers with GJHS are weaker during the total range of swimming strokes, which may decrease

swimming propulsion and increase joint stress exposure close to end range.

4.2 Isokinetic Fatigue

Swimmers with GJHS further showed significantly larger fatigue development during

isokinetic strength measurements in medial rotation (180°/s). This corresponds with the current

reduced maximum work in medial rotation (60°/s) indicating lower endurance compared with

controls, which may alter swimming strike coordination (Suito et al., 2008) and hence increase the

risk of developing fatigue related shoulder pain (Crotty & Smith, 2000; Matthews et al., 2017). The

current result on shoulder fatigue is in line with previous studies of patients with GJH generally

10
showing increased fatigue (Voermans et al., 2010; Scheper et al., 2014), however, the long-term

consequences of this is unknown. If the included population develop clinical symptoms

(impingement, rotator cuff tendinopathy and labral tears), preventive measures to avoid this

development is important. Based on the current results, swimmers with GJHS may benefit from

medial rotation endurance and strength training in addition to lateral rotation strength (Cools et al.,

2015) to prevent such injuries. This may be an important topic for future studies.

4.3 Muscle activity

It was hypothesised that the group with GJHS would show lower muscle activity in

serratus anterior and lower trapezius indicating muscular imbalance in the scapular stabilisers,

which was not confirmed. There was only a tendency to lower muscle activity in medial rotation in

infraspinatus during peak torque and pectorals major during maximum work. The current results

indicate decreased dynamic stabilisation of the humeral head in the hypermobile shoulder due to

altered length-tension conditions for the medial rotators (which was found) and capsular looseness

(not measured). Although a previous study has shown swimmers with shoulder problems to have

lower activity of serratus anterior than swimmers without pain (Pink et al., 1991), such difference

could not be seen in those with GJHS, probably due to the non-pain population. Generally, a lower

activity of this muscle may influence scapular kinematics negatively by positioning the acromion to

impinge on the rotator cuff, with increased risk for pain development (Cools et al., 2004). Lower

muscle activity during medial rotation has previously been found in individuals with MDI as well

as in individuals with subacromial impingement (Barden et al., 2005; Struyf et al., 2011), and may

suggest altered muscle activation in the shoulders. The deeper lying medial rotator of subscapularis

muscle, which requires needle-EMG for measuring its muscle activity, has also shown significantly

lower muscle activity in swimmers with shoulder pain (Pink et al., 1991). Again, the present

population was pain-free, but may have shown same characteristics as swimmers with shoulder

pain with limited optimal working conditions (pain or altered length-tension conditions) for the

working muscles. However, since the subscapularis muscle activity was not measured in the

current study, its contribution to medial rotation remains unknown. The lack of statistically

11
significant group differences in EMG activity may be due to large inter-individual variability (SD

range between 19.5 and 34.6 %MVE), thereby increasing the risk of type 2 error. It is not known

whether a more functional task may have shown a clearer pattern of changed EMG-activity in

GJHS, as seen previously in GJH with knee hypermobility during functional tasks such as balance,

gait and jumping (Jensen et al., 2013; Junge et al., 2015).

4.4 Strength and limitations

One of the limitations in the current study may be the risk of selection bias since the

inclusion criteria were healthy competitive swimmers without pain, thereby eliminating a potential

effect of pain in GJHS. Although the inclusion criteria did not include instability, large positive

prevalence of at least one of the shoulder instability tests was seen in GJHS swimmers indicating

instability to be a potential inherent characteristic of GJHS. Further, clearer signs of fatigue

development, as shown in previous studies, may have been present with higher number of

repetitions and/or faster velocities than the present. This was, however, not chosen to avoid

exposing the current participants to potential injuries during the testing procedure. Another

limitation was increased risk of type 2 error in EMG measurements with large standard deviations

since the sample size was based on group differences in isokinetic variables.

The strengths of the study are the strict inclusion criteria, group similarity in demographics

and the standardised procedures in isokinetic and EMG measurements to ensure reliable data

(Edouard et al., 2011; Seitz & Uhl, 2012). Also, the standardized procedures used in the selected

clinical tests for shoulder instability provide satisfactory reliability (Eshoj et al., 2017). The strict

inclusion criteria of healthy swimmers may also be a strength, since it was possible to study the

clear effect of GJHS in swimmers without pain interference. The study population included 57.9%

girls and 42.1% boys making it possible to generalise data to both genders in contrast to previous

studies on GJH with few or no boys included (Juul-Kristensen et al., 2012; Jensen et al., 2013;

Junge et al., 2015).

12
5. Conclusion

In conclusion, young competitive swimmers with generalised and shoulder joint hypermobility

(GJHS), despite having no formal diagnosis, displayed both strength and fatigue deficits in medial

rotation, which may be some of the contributing mechanisms for development of shoulder injury.

Whether swimmers with GJHS will benefit from medial rotation strength training and not as per

now only lateral rotation strength as injury prevention is an important topic for future studies.

Acknowledgement

The authors would like to thank Henrik Baare Olsen for technical advice, Tore Bjarnason and Sune

Christensen for assistance in testing the swimmers, University College Lillebaelt in Odense,

Denmark, for their support with recruiting participants, and University of Southern Denmark for

providing facilities and equipment to complete this research project. The project was supported by

the Danish Society of Sports Physiotherapy, who was not involved in the conduction of research or

preparation of the article.

13
References

Bak K, Faunø P. Clinical findings in competitive swimmers with shoulder pain. Am J

Sports Med. 1997;25(2):254-260.
Bak K, Magnusson SP. Shoulder strength and range of motion in symptomatic and pain-
free elite swimmers. Am J Sports Med. 1997;25(4):454-459.
Barden JM, Balyk R, Raso VJ, Moreau M, Bagnall K. Atypical shoulder muscle activation
in multidirectional instability. Clin Neurophysiol. 2005;116(8):1846-1857.
Borsa PA, Sauers EL, Herling DE. Patterns of glenohumeral joint laxity and stiffness in
healthy men and women. Med Sci Sports Exerc. 2000;32(10):1685-1690.
Cameron KL, Duffey ML, DeBerardino TM, Stoneman PD, Jones CJ, Owens BD.
Association of generalized joint hypermobility with a history of glenohumeral joint
instability. J Athl Train. 2010;45(3):253-258.
Chahal J, Leiter J, McKee MD, Whelan DB. Generalized ligamentous laxity as a
predisposing factor for primary traumatic anterior shoulder dislocation. J Shoulder Elbow
Surg. 2010;19(8):1238-1242.
Cools AM, Johansson FR, Borms D, Maenhout A. Prevention of shoulder injuries in
overhead athletes: a science-based approach. Braz J Phys Ther. 2015;19:331-339.
Cools AM, Witvrouw EE, Declercq GA, Vanderstraeten GG, Cambier DC. Evaluation of
isokinetic force production and associated muscle activity in the scapular rotators during a
protraction-retraction movement in overhead athletes with impingement symptoms. Br J
Sports Med. 2004;38(1):64-68.
Crotty NM, Smith J. Alterations in scapular position with fatigue: a study in swimmers.
Clin J Sport Med. 2000;10(4):251-258.
Edouard P, Degache F, Beguin L, Samozino P, Gresta G, Fayolle-Minon I, Farizon F,
Calmels P. Rotator cuff strength in recurrent anterior shoulder instability. J Bone Joint
Surg Am. 2011;93(8):759-765.
Edouard P, Samozino P, Julia M, Gleizes Cervera S, Vanbiervliet W, Calmels P,
Gremeaux V. Reliability of isokinetic assessment of shoulder-rotator strength: a systematic
review of the effect of position. J Sport Rehabil. 2011;20(3):367-383.
Eshoj H, Ingwersen KG, Larsen CM, Hougs Kjær B, Juul-Kristensen B.Inter-tester
reliability of clinical shoulder instability and laxity tests in subjects with and without self-
reported shoulder problems. 2017;(accepted, BMJ Open).
Gagey OJ, Gagey N. The hyperabduction test. J Bone Joint Surg Am. 2001;83:69-74.
Holtermann A, Mork PJ, Andersen LL, Olsen HB, Søgaard K. The use of EMG
biofeedback for learning of selective activation of intra-muscular parts within the serratus
anterior muscle: a novel approach for rehabilitation of scapular muscle imbalance. J
Electromyogr Kinesiol. 2010;20(2):359-365.
Holtermann A, Roeleveld K, Mork PJ, Grönlund C, Karlsson JS, Andersen LL, Olsen HB,
Zebis MK, Sjøgaard G, Søgaard K. Selective activation of neuromuscular compartments
within the human trapezius muscle. J Electromyogr Kinesiol. 2009;19(5):896-902.
Jensen BR, Olesen AT, Pedersen MT, Kristensen JH, Remvig L, Simonsen EB, Juul-
Kristensen B. Effect of generalized joint hypermobility on knee function and muscle
activation in children and adults. Muscle Nerve. 2013;48(5):762-769.
Junge T, Henriksen P, Andersen HL, Byskov LD, Knudsen HK, Juul-Kristensen B. The
association between generalized joint hypermobility and active horizontal shoulder
abduction in 10-15 year old competitive swimmers. BMC Sports Sci Med Rehabil.
2016;8:19.
Junge T, Wedderkopp N, Thorlund JB, Søgaard K, Juul-Kristensen B. Altered knee joint
neuromuscular control during landing from a jump in 10-15 year old children with

14
Generalised Joint Hypermobility. A substudy of the CHAMPS-study Denmark. J
Electromyogr Kinesiol. 2015;25(3):501-507.
Juul-Kristensen B, Hansen H, Simonsen EB, Alkjær T, Kristensen JH, Jensen BR, Remvig
L. Knee function in 10-year-old children and adults with Generalised Joint Hypermobility.
Knee. 2012;19(6):773-778.
Juul-Kristensen B, Røgind H, Jensen DV, Remvig L. Inter-examiner reproducibility of
tests and criteria for generalized joint hypermobility and benign joint hypermobility
syndrome. Rheumatology (Oxford). 2007;46(12):1835-1841.
Laudner KG, Meister K, Kajiyama S, Noel B. The relationship between anterior
glenohumeral laxity and proprioception in collegiate baseball players. Clin J Sport Med.
2012;22(6):478-482.
Matthews MJ, Green D, Matthews H, Swanwick E. The effects of swimming fatigue on
shoulder strength, range of motion, joint control, and performance in swimmers. Phys Ther
Sport. 2017;23:118-122.
Neer CS.II, Foster CR. Inferior capsular shift for involuntary inferior and multidirectional
instability of the shoulder. A preliminary report. J Bone Joint Surg Am.1980;62:897-908.
Pink M, Perry J, Browne A, Scovazzo ML, Kerrigan J. The painful shoulder during
freestyle swimming: an electromyographic and cinematographic analysis of twelve
muscles, Am J Sports Med. 1991;19:569-576.
Pontillo M, Orishimo KF, Kremenic IJ, McHugh MP, Mullaney MJ, Tyler TF. Shoulder
musculature activity and stabilization during upper extremity weight-bearing activities. N
Am J Sports Phys Ther. 2007;2(2):90-96.
Rotes-Querol J. [Articular laxity considered as factor of changes of the locomotor
apparatus.] Rev Rhum Mal Osteoartic. 1957;24:535-539.
Scheper MC, de Vries JE, Juul-Kristensen B, Nollet F, Engelbert RH. The functional
consequences of generalized joint hypermobility: a cross-sectional study. BMC
Musculoskelet Disord. 2014;15:243.
Seitz AL, Uhl TL. Reliability and minimal detectable change in scapulothoracic
neuromuscular activity. J Electromyogr Kinesiol. 2012;22(6):968-974.
St Pierre RK, Andrews L, Allman, F, Fleming LL. The Cybex II evaluation of lateral ankle
ligamentous reconstructions. Am J Sports Med. 1984;12:52-56.
Struyf F, Nijs J, Baeyens JP, Mottram S, Meeusen R. Scapular positioning and movement
in unimpaired shoulders, shoulder impingement syndrome, and glenohumeral instability.
Scand J Med Sci Sports. 2011;21(3):352-358.
Suito H, Ikegami Y, Nunome H, Sano S, Shinkai H, Tsujimoto N. The effect of fatigue on
the underwater arm stroke motion in the 100-m front crawl. J Appl Biomech.
2008;24:316–324.
Tol JL, Hamilton B, Eirale C, Muxart P, Jacobsen P, Whiteley R. At return to play
following hamstring injury the majority of professional football players have residual
isokinetic deficits. Br J Sports Med. 2014;48(18):1364-1369.
Voermans NC, Knoop H, van de Kamp N, Hamel BC, Bleijenberg G, van Engelen BG.
Fatigue is a frequent and clinically relevant problem in Ehlers-Danlos Syndrome. Semin
Arthritis Rheum. 2010;40(3):267-274.
Wanivenhaus F, Fox AJ, Chaudhury S, Rodeo SA. Epidemiology of injuries and
prevention strategies in competitive swimmers. Sports Health. 2012;4(3):246-251.
Zemek MJ, Magee DJ. Comparison of glenohumeral joint laxity in elite and recreational
swimmers. Clin J Sport Med. 1996;6(1):40-47.

15
Table 1. Demographic variables, (mean ± standard deviation, median [range], or number

(percentage)) for self-reported ratings on pain (VAS), and shoulder-related instability and function

(WOSI), in addition to clinical tests for glenohumeral instability, for swimmers with Generalised

Joint Hypermobility and shoulder hypermobility (GJHS) and controls.

GJHS Controls p-value

(n = 19) (n = 19)

Gender: female, n (%) 11 (57.9) 11 (57.9) 1.000

Beighton score, 0-9 7.1 ± 1.1 1.1 ± 1.2 <0.001*

Rotes-Querol, positive, n (%) 19 (100) 0 (0) <0.001*

Age, years 14.8 ± 1.3 14.7 ± 1.1 0.678

Height, cm 172.5 ± 8.4 170.6 ± 9.8 0.515

Body mass, kg 65.8 ± 12.8 62.7 ± 10.9 0.423

Swimming competitive experience, years 4.3 ± 1.9 4.9 ± 1.7 0.339

Swimming practice duration, h/week 8.3 ± 3.9 8.7 ± 5.2 0.783

Other sports activities, h/week 6.1 ± 2.1 5.5 ± 3.0 0.516

Pain intensity

VAS 0-100 (mm) current pain 0 [0-20] 0.5 [0-10] 0.716

VAS 0-100 (mm) pain during latest 24 h 2 [0-50] 4 [0-39,5] 0.633

VAS 0-100 (mm) pain during latest seven days 2.5 [0-49] 9 [0-49] 0.175

WOSI overall score, 0-2100 132 [17-886] 294 [30-649] 0.609

Physical symptoms, 0-1000 50 [6-358] 84 [12-349] 0.280

Sports/recreation/work, 0-400 17 [2-235] 18 [0-127] 0.540

Lifestyle, 0-400 12 [0-151] 14 [1-71] 0.619

Emotions, 0-300 33 [1-178] 74 [5-183] 0.540

Glenohumeral instability tests

Gagey hyperabduction test, positive, n (%) 12 (63.2) 6 (31.6) 0.103

Sulcus test, positive, n (%) 3 (15.8) 0 (0) 0.230

Load-and-shift test, positive, n (%) 6 (31.6) 1 (5.3) 0.090

WOSI= Western Ontario Shoulder Instability Index.

Significant difference, p<0.05, is marked with *.

16
 Table 2. Peak torque (Nm/kg) and max work (J/kg), including mean ± standard deviation, during

isokinetic medial and lateral rotation movements at 60°/s (five repetitions) and 180°/s (10

repetitions), for swimmers with Generalised Joint Hypermobility and shoulder hypermobility

(GJHS) and controls.

Medial rotation Lateral rotation

Velocity GJHS Controls p-value GJHS Controls p-value

(degrees/s) (n = 19) (n = 19) (group (n = 19) (n = 19) (group

difference) difference)

Peak Torque

°
Nm/kg 60 /s 0.53 ± 0.10 0.60 ± 0.13 0.047* 0.38 ± 0,07 0.40 ± 0.07 0.274

°
Nm/kg 180 /s 0.48 ± 0.09 0.51 ± 0.11 0.323 0.32 ± 0.05 0.33 ± 0.05 0.927

Maximum Work

J/kg 60°/s 0.62 ± 0.12 0.71 ± 0.17 0.031* 0.41 ± 0.10 0.43 ± 0.08 0.373

J/kg 180°/s 0.58 ± 0.11 0.62 ± 0.15 0.236 0.34 ± 0.08 0.35 ± 0.08 0.615

Nm=Newton meter, J=Joules

Significant difference, p<0.05, is marked with *.

17
Figure 1. Experimental set-up for the electromyographic and isokinetic testing procedure.

Figure 2. Electromyographic activity of upper trapezius (UT), lower trapezius (LT), serratus

anterior (SA), pectoralis major (PM) and infraspinatus (INF) at peak torque (A-D) and maximum

work (E-H), expressed as percentage of maximal voluntary contraction (%MVE), during isokinetic

glenohumeral rotation movements at low (60 °/s) and high (180°/s) velocity, for swimmers with

Generalised Joint Hypermobility and shoulder hypermobility (GJHS) and controls. Bars are means,

and standard deviations are shown as error bars. (*) Tendency to significance, p<0.10.

18
Abstract

Generalised Joint Hypermobility including shoulder hypermobility (GJHS) in swimmers is

considered an intrinsic risk factor for shoulder injuries. The aim was to investigate the association

of GJHS with shoulder strength, fatigue development and muscle activity during swimming-related

shoulder rotations. Totally, 38 competitive swimmers (aged 13-17 years) participated, 19 were

competitive swimmers with GJHS and 19 were age, sex and club matched swimmers without

GJHS. Concentric isokinetic force in medial and lateral rotations were measured at 60°/s (5

repetitions) and 180°/s (10 repetitions). Electromyographic activity was measured from upper

trapezius, lower trapezius, serratus anterior, infraspinatus and pectoralis major muscles. Swimmers

with GJHS produced significantly lower peak torque (0.53 vs. 0.60 Nm/BW; p=0.047) and

maximum work (0.62 vs. 0.71 J/BW; p=0.031) than controls during medial rotation (60°/s).

Swimmers with GJHS showed significantly larger isokinetic fatigue at 180°/s (0.321 J/repetition;

p=0.010), and tendencies to lower levels of muscle activity in infraspinatus (20%, p=0.066) and

pectoralis major (34%, p=0.092) at 60°/s during medial rotation. Young competitive swimmers

with GJHS, despite no formal diagnosis, displayed strength and fatigue deficits in medial rotation,

potentially inherent with greater risk of shoulder injury. Whether GJHS swimmers benefit from

medial rotation strengthening is an important topic for future studies.

Keywords Joint instability; swimming; shoulder; muscle strength; electromyography;

adolescent

Disclosure of interest

The authors report no conflicts of interest.

19
20
21