J Shoulder Elbow Surg (2024) 33, 1918–1927

www.elsevier.com/locate/ymse

Do therapeutic exercises impact supraspinatus

tendon thickness? Secondary analyses of the
combined dataset from two randomized
controlled trials in patients with rotator
cuff-related shoulder pain
Marc-Olivier Dub
e, PT, PhDa,b,c,*,1, Kim Gordon Ingwersen, PT, PhDd,1,
Jean-S
ebastien Roy, PT, PhDa,b, François Desmeules, PT, PhDe,f,
Jeremy Lewis, PT, PhDg,h,i,j, Birgit Juul-Kristensen, PT, PhDk, Jette Vobbe, MDl,
Steen Lund Jensen, MD, PhDm, Karen McCreesh, PT, PhDn,o

a
Center for Interdisciplinary Research in Rehabilitation and Social Integration, Quebec City, Quebec, Canada
b
School of Rehabilitation Sciences, Faculty of Medicine, Universit
e Laval, Quebec City, Quebec, Canada
c
La Trobe Sport and Exercise Medicine Research Centre, La Trobe University, Melbourne, Victoria, Australia
d
Research Unit in Physio – and Occupational Therapy, Department of Physio- and Occupational Therapy, Lillebaelt
Hospital - Vejle, Vejle, Denmark
e
Orthopaedic Clinical Research Unit, Maisonneuve-Rosemont Hospital Research Center, University of Montreal Affiliated
Research Center, Montreal, Canada
f
School of Rehabilitation, Faculty of Medicine, University of Montreal, Montreal, Canada
g
Therapy Department, Central London Community Healthcare National Health Service Trust, Finchley Memorial
Hospital, London, UK
h
School of Health Sciences, University of Nottingham, Nottingham, UK
i
School of Life and Health Sciences, University of Nicosia, Nicosia, Cyprus
j
Clinical Therapies, University of Limerick, Limerick, Ireland
k
Research Unit of Musculoskeletal Function and Physiotherapy, Department of Sports Science and Clinical Biomechanics,
University of Southern Denmark, Odense, Denmark
l
Orthopaedic Department, Shoulder Unit, Lillebaelt Hospital, Vejle Hospital, Vejle, Denmark

The RoCTEX trial was approved by the Regional Scientific Ethics Committee *Reprint requests: Marc-Olivier Dub
e, PT, PhD, Centre for Inter-
of Southern Denmark (S-20130071) and the Dub
e et al trial was approved by disciplinary Research in Rehabilitation and Social Integration, 525,
the Research Ethics Committee of Centre int
egr
e universitaire de sant
e et de Boulevard Wilfrid-Hamel, office H-0616, Quebec City, Quebec G1M
services sociaux de la Capitale-Nationale (CIUSSS-CN) (#2019-1762). 2S8, Canada.
1
Marc-Olivier Dub
e and Kim Gordon Ingwersen contributed equally as E-mail address: marc-olivier.dube.1@ulaval.ca (M.-O. Dub
e).
first co-authors.

1058-2746/Ó 2024 The Author(s). This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
https://doi.org/10.1016/j.jse.2024.03.055
Therapeutic exercises and supraspinatus tendon thickness 1919
m
Orthopaedic Department, Shoulder Unit, Aalborg University Hospital and Department of Clinical Medicine, Aalborg
University, Aalborg, Denmark
n
School of Allied Health, Faculty of Education and Health Sciences, University of Limerick, Limerick, Ireland
o
Ageing Research Centre, Health Research Institute, University of Limerick, Limerick, Ireland

Background: The mechanistic response of rotator cuff tendons to exercises within the context of rotator cuff-related shoulder pain
(RCRSP) remains a significant gap in current research. A greater understanding of this response can shed light on why individuals
exhibit varying responses to exercise interventions. It can also provide information on the influence of certain types of exercise on ten-
dons. The primary aim of this article is to explore if changes in supraspinatus tendon thickness (SSTT) ratio differ between exercise
interventions (high load vs. low load). The secondary aims are to explore if changes in SSTT ratio differ between ultrasonographic ten-
dinopathy subgroups (reactive vs. degenerative) and if there are associations between tendinopathy subgroups, changes in tendon thick-
ness ratio, and clinical outcomes (disability).
Methods: This study comprises secondary analyses of the combined dataset from two randomized controlled trials that compared high
and low-load exercises in patients with RCRSP. In those trials, different exercise interventions were compared: 1) progressive high-load
strengthening exercises and 2) low-load strengthening with or without motor control exercises. In 1 trial, there was also a third group that
was not allocated to exercises (education only). Ultrasound-assessed SSTT ratio, derived from comparing symptomatic and asymptom-
atic sides, served as the primary measure in categorizing participants into tendinopathy subgroups (reactive, normal and degenerative) at
baseline.
Results: Data from 159 participants were analyzed. Two-way repeated measures ANOVAs revealed significant Group (P < .001) and
Group
Time interaction (P < .001) effects for the SSTT ratio in different tendinopathy subgroups, but no Time effect (P ¼ .63).
Following the interventions, SSTT ratio increased in the "Degenerative" subgroup (0.14 [95% confidence interval {CI}: 0.09-0.19]),
decreased in the "Reactive" subgroup (0.11 [95% CI: 0.16 to 0.06]), and remained unchanged in the "Normal" subgroup
(0.01 [95% CI: 0.04 to 0.02]). There was no Time (P ¼ .21), Group (P ¼ .61), or Group
Time interaction (P ¼ .66) effect for
the SSTT ratio based on intervention allocation. Results of the linear regression did not highlight any significant association between
the tendinopathy subgroup (P ¼ .25) or change in SSTT ratio (P ¼ .40) and change in disability score.
Conclusion: Findings from this study suggest that, over time, SSTT in individuals with RCRSP tends to normalize, compared to the
contralateral side, regardless of the exercise intervention. Different subgroups of symptomatic tendons behave differently, emphasizing
the need to potentially consider tendinopathy subtypes in RCRSP research. Future adequately powered studies should investigate how
those different tendinopathy subgroups may predict long-term clinical outcomes.
Level of evidence: Level II; Randomized Controlled Trial; Treatment Study
Ó 2024 The Author(s). This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Keywords: Supraspinatus; ultrasound; shoulder; tendinopathy; thickness; rotator cuff-related shoulder pain

Shoulder pain affects up to 56% of the population Doppler ultrasound (US) imaging.8,34,42,52 One model of
yearly.17,35 A vast proportion of shoulder pain complaints tendon pain and pathology, known as the continuum model
may be labeled as rotator cuff-related shoulder pain of tendinopathy, classifies tendinopathy as reactive (char-
(RCRSP).57 Based on the definition of the International acterized by tendon thickening, and hypoechoic areas), or
Scientific Tendinopathy Symposium Consensus, RCRSP degenerative (characterized by tendon thinning and/or
may be defined as persistent pain and loss of function tearing).5,6,31,38
related to mechanical loading on the rotator cuff tendons.50 Many nonsurgical interventions have been suggested as
RCRSP leads to decreased function and quality of life29 as surgical management of RCRSP has not proven to be more
well as increased healthcare usage55 especially since 40%- effective than nonsurgical ones.22 Therapeutic exercise
50% of individuals with shoulder pain experience persistent being one of the advocated nonsurgical option, with sup-
or recurrent pain within the 12 months after initial porting evidence that it is associated with moderate effect
onset.7,13,26 sizes for pain and disability reduction when compared to a
Our understanding of RCRSP has evolved from a bio- placebo or no intervention.51 One aim of therapeutic ex-
logical model,28,43 to one that also includes psychosocial ercise is to induce beneficial mechanical stress to induce
factors that are likely to contribute to the persistence of changes in the rotator cuff tendons, notably decreased
pain.3,4 Although these psychological factors can play a tendon thickness in those whose tendinopathy can be
crucial role, it is important not to discard the biological classified as reactive and increased tendon thickness in
factors that may contribute to the onset and persistence of those whose tendinopathy can be classified as degenera-
pain. Structural pathology is commonly observed in tive.2,23 In healthy tendons, higher loads have been asso-
RCRSP, with changes such as tendon thickening or thin- ciated with an increased cross-sectional area,1,27 while the
ning, neovascularization, decreased tendon stiffness, and opposite seems to happen in pathological reactive tendons
hypoechoic appearance visible on greyscale or power where higher loads such as heavy-slow resistance exercise
1920 M.-O. Dub
e et al.

program have been associated with a decrease in cross- external rotation). Patients with other shoulder-related diagnoses
sectional area.25 There is no evidence regarding the not relating to RCRSP (eg, capsulitis, instability), bilateral
behavior of degenerative rotator cuff tendons following shoulder complaints, competing diagnoses, full-thickness rotator
loading in vivo in a symptomatic population. Trials of ex- cuff tears, large calcifications, or corticosteroid injection within
the past six weeks were excluded. Eligibility criteria were
ercise for reactive patellar and Achilles tendinopathy have
verified by a physiotherapist and participants were recruited
shown a decrease in tendon thickness in response to 12-
from shoulder units within the outpatient clinics of orthopedic
week exercise programs, with positive associations departments of 3 hospitals in the western part of Denmark for
between structural changes and patient-reported out- Ingwersen et al, and through advertisements in local physio-
comes.24,25,44 This could also apply for RCRSP, as the therapy clinics (Quebec, Canada), electronic mailing lists of
majority of cross-sectional studies evaluating tendon Universit
e Laval (Quebec, Canada) as well as social media for
thickness in symptomatic populations have mostly high- Dub
e et al.
lighted an increased tendon thickness.18,41,52 However, no
published trials have evaluated the changes in tendon Interventions
thickness in response to different exercise programs for
RCRSP. Both trials compared an exercise regimen with higher loads to an
Thus, this article presents secondary analyses of pooled exercise regimen with lower loads. More details about the in-
data from two randomized controlled trials10,21 looking at terventions are presented below and can be found in the respective
the effect of different exercise regimens on pain, disability, articles.11,10,19,21 In both trials, the low-load interventions
and supraspinatus tendon thickness (SSTT) in adults with included a larger number of repetitions (>15) and no significant
RCRSP. The primary aim of the current article is to explore external load. On the other hand, the high-load interventions
included in both trials aimed to increase strength by inducing a
if changes in SSTT differ between interventions (high load
sufficient load on the musculotendinous unit. In those in-
vs low load exercises). Based on previous findings in the
terventions, the number of repetitions was smaller (4-8 RM), and
lower limb literature, our hypothesis was that high load the external load applied was larger. Although not identical, the
exercises would lead to a greater reduction in tendon shared attributes of the interventions in both trials prompted us to
thickness compared to low load exercises. The secondary consider their integration to achieve the study’s objectives.
aims are to explore if changes in SSTT differ between ul-
trasonographic tendinopathy subgroups (reactive vs. RoCTEx trial
degenerative) and if there are associations between ten- In the RoCTEx trial, patients were randomly assigned to follow
dinopathy subgroups, changes in tendon thickness, and either the intervention program consisting of 12 weeks of pro-
clinical outcomes (disability). gressive high load exercises (PHLE) (gradually increasing the
load from 15 repetitions maximum (RM) in week 1, to 6 RM in
weeks 9 to 12) (n ¼ 28), or the control group consisting of
Materials and methods 12 weeks of low load exercises (LLE) (20-25 RM from weeks
1-12) (n ¼ 40).
In general, both groups received the same 6 exercises, spe-
Study design
cifically addressing the scapular and rotator cuff muscles. The
program consisted of 2 exercises for strengthening the rotator cuff
Results presented in this article are derived from secondary ana- muscles (push-up plus exercise and low row exercise), 2 exercises
lyses of pooled results from two randomized controlled trials for the scapulothoracic muscles (scaption and side-lying external
comparing high and low-load exercise for RCRSP.10,21 Both trials rotation), and 2 stretching and mobilizing exercises (posterior
obtained informed consent from participants. The RoCTEX trial capsule stretch and scapula retraction). The program was designed
was approved by the Regional Scientific Ethics Committee of to generate the same total load and time under tension in both
Southern Denmark (S- 20130071) and the Dub
e et al trial was groups. Further, attention and basic information about ergonomic
approved by the Research Ethics Committee of Centre Int
egr
e corrections related to daily use of the upper extremity were pro-
Universitaire de Sant
e et de Services Sociaux de la Capitale- vided. Both groups were recommended to perform home-based
Nationale (CIUSSS-CN) (#2019-1762). exercises 3 times per week and were offered 6 follow-up one-to-
one sessions at the physiotherapy department. Exercise
Participants instructions follow-up sessions were performed by trained phys-
iotherapists, following the standardized exercise regime described
Participants included in this study were adults with persistent in a previous article.19 Patients were blinded toward the specific
RCRSP who were included in either the Dub
e et al or Ingwersen difference regarding load adjustments in the 2 groups and were
et al (RoCTEx) trials.10,21 Detailed information about eligibility merely told that 2 different active treatment approaches were
criteria, which were similar in both trials, are presented in the compared.
respective articles.11,10,19,21 In summary, eligible participants One thing to note is that 34 participants (17 in both groups)
were adults aged 18-65 years with shoulder complaints lasting received a corticosteroid injection at baseline before starting their
for at least three months, presenting with a combination of 12-week exercise program. There was a significant interaction
positive clinical tests (painful arc, Neer and/or Hawkins Ken- between corticosteroid injection and type of exercise. This inter-
nedy and/or Jobe, and painful resisted isometric abduction or action highlighted a nonsignificant disability decrease in favor of
Therapeutic exercises and supraspinatus tendon thickness 1921

PHLE for patients receiving a corticosteroid injection. This is plane of the supraspinatus tendon just anterior to the acromion
relevant as both corticosteroid and anesthetic (eg, lidocaine) have with tuberculum major visible in the lateral part of the image.
the potential to reduce tenocyte numbers that may reduce the From the most lateral part of the insertion of the supraspinatus on
amount of proteoglycan which will reduce water content and lead the tuberculum major, a 20 mm line was drawn medial to a point
to a reduction in tendon thickness.32,30 where the line crosses the humeral head. From this point, a
perpendicular line from the caput was drawn to the superior
Dub

e et al trial border of the supraspinatus tendon, corresponding to the thick-
In the other trial, participants were allocated to an intervention ness of the supraspinatus tendon. Inter-rater reliability is excel-
comprising progressive high-load strengthening exercises com- lent (ICC: 0.85-0.90) for this standardized procedure.20 In the
bined with an education component or low-load motor control Dub
e et al trial, US measurements of supraspinatus tendon
exercises combined with an education component. In addition, thickness were obtained with the medial aspect of the wrist
there was a third group that only received the education compo- against the ipsilateral anterior superior iliac spine. Measures
nent without being instructed on any specific exercises. were obtained with the transducer perpendicularly, 1 cm behind
The strengthening exercises program was a 12-week progres- to the anterolateral aspect of the surface of the acromion. The
sive high-load exercises program that targeted humeral internal/ thickness of the SS tendon borders was defined inferiorly as the
external rotators, abductors, and the scapular muscles through 7 first hyperechoic region above the anechoic articular cartilage of
different exercises. Number of repetitions was 1 set of the the humeral head, and the hyperechoic superior border of the
maximum number of repetitions until muscular fatigue. The load tendon before the anechoic subdeltoid bursa. These US tendon
was 90% of 1 RM, which is the weight with which the participant measures are reliable (intraclass correlation coefficient [ICC] >
could only complete 1 repetition of the exercise. At each of the 6 0.92).37 Using ratio instead of the absolute value of supraspinatus
follow-up intervention sessions, shoulder strength was reassessed, tendon thickness measurement allowed us to pool US measure-
and the program was progressed accordingly. ment data obtained with 2 different methods. This prevented
As for the 12-week low-load motor control program, each of systematic differences that could have been observed between
the 6 follow-up sessions started with different shoulder symptom the 2 methods to influence our results.
modification procedures to alleviate symptoms during upper Participants were categorized into 3 subgroups of tendinop-
limb movements.33 If a technique reduced pain, that technique athy according to their ratio of baseline US measurements of
was then used during elevation exercises in 3 planes (flexion, SSTT in relation to the continuum model5,31: ‘‘degenerative’’
abduction, scaption) and incorporated into the participant’s tendinopathy for those with a ratio below 0.9, ‘‘reactive’’ ten-
everyday functional movements.49 Once participants had reached dinopathy for those with a ratio above 1.1 and ‘‘normal’’ tendons
pain-free execution, the program was progressed into re- for those that fell in-between 0.9 and 1.1. We considered a 10%
education exercises involving the whole body and mirroring cut-off value to divide the groups to account for measurement
the different functions performed by the shoulder. Participants error (minimal detectable change [MDC] ¼ 0.33 mm)20 and
performed around 6 different low-load motor control exercises within-subject variability between both sides.
with the number of repetitions varying from 1 to 3 sets of 15
repetitions.
Statistical analyses
Finally, the education component of the interventions included
2 30-minute face-to-face education sessions with a physiothera-
pist. During the first one, participants received oral and written Descriptive values were presented as mean standard deviation
advice and information related to the shoulder (anatomy and (SD) or median (range) per tendinopathy subgroup. Comparison
function), basic pain science, advice on pain management (night in tendon thickness ratio from baseline to 12-week follow-up
and day), activity modification (when to increase and decrease), between allocated interventions and tendinopathy subgroups
and on the importance of physical activity, sleep, healthy eating were performed with two-way repeated measures analysis of
habits, and managing stress. They were also instructed to watch a variance (ANOVAs). Those ANOVAs could reveal potential
series of 6 videos that addressed topics discussed during the initial Time (significant difference in the dependent variable between
session. both measurement timepoints [baseline and 12-week follow-up])
and/or Group (significant difference in the dependent variable
between groups [either the allocated intervention or the tendin-
Outcomes opathy subgroup]) effects as well as Time
Group interaction
(significant difference between groups in the change in the
In this article, the primary outcome was the change in the SSTT dependent variable over time). In addition, a linear regression
ratio from baseline to 12 weeks follow-up, calculated by dividing was used to explore the association between the baseline sub-
the measurement on the symptomatic side by the measurement group of tendinopathy and mean disability (DASH or Quick-
on the asymptomatic side. We did so using the supraspinatus DASH) score change, with disability change score acting as the
tendon thickness measurements obtained in millimeters dependent variable. Tendinopathy subgroup, exercise modality,
following standardized procedures on US-captured images.20,37 baseline disability score, and concomitant administration of
In both trials, only 1 investigator, blinded to exercise modality, corticosteroid injection (since it was a possible cointervention
captured all US images and performed all measurements of included in the RoCTEx trial) were imputed as variables in the
tendon thickness. In the RoCTEx trial, US examinations were model. Baseline disability, exercise modality, and corticosteroid
performed with the patient sitting and the shoulder positioned in were selected specifically to test for potential confounding of the
internal rotation by placing the dorsal side of the hand on the results from these variables. Baseline disability was considered
opposite gluteal region. Images were obtained in the longitudinal relevant, as it influences the amount of possible change the
1922 M.-O. Dub
e et al.

demographics and clinical characteristics are presented in

Table I based on tendinopathy subgroup.

Mean change in tendon thickness ratio between

baseline and 12-week follow-up

There was no significant difference in mean SSTT ratio

between or within exercise modalities as highlighted by the
absence of a Time (P ¼ .21), Group (P ¼ .61), or
Time
Group interaction (P ¼ .66) effect (Table II).
However, there were significant Group (P < .001) and
Group
Time interaction (P < .001) effects in SSTT ratio
for the different tendinopathy subgroups, but no Time effect
(P ¼ .63) (Table III). The mean tendon thickness ratio
increased significantly for the ‘‘Degenerative’’ subgroup
(0.14 [95% CI: 0.09 to 0.19]), remained unchanged for the
‘‘Normal’’ subgroup (0.01 [95% CI: 0.04 to 0.02]), and
decreased significantly for the ‘‘Reactive’’ subgroup (0.11
[95% CI: 0.16 to 0.06]) (Table III & Fig. 2). Both ex-
ercise modalities led to the same effects on the supra-
spinatus tendon, ie decreased tendon thickness ratio in the
‘‘Reactive’’ subgroup, no change in the ‘‘Normal’’ sub-
group, and increased tendon thickness ratio in the
‘‘Degenerative’’ subgroup. There was no significant change
for the intervention without exercises for the ‘‘Reactive’’
and ‘‘Degenerative’’ subgroups, but the trend observed in
the other 2 groups was the same in that group (Table IV).
Figure 1 Flowchart of included participants for this study. SE, However, there were no significant between-group differ-
strengthening exercises; PHLE, progressive high load exercises; ences for the mean tendon thickness ratio between baseline
Ed, education; MCE, motor control exercises; LLE, low load
and 12-week follow-up between the different allocated in-
exercises.
terventions within each tendinopathy subgroup (Table IV).

patient can achieve, allocated intervention was considered rele-

vant, as this is directly correlated with the amount of load the
Relation between tendinopathy subgroup and
patients were exposed to through the intervention, and the steroid change in disability
has potential negative effects on tendon tissue regeneration.47
Results of the linear regression did not highlight any as-
sociation between the tendinopathy subgroup and change in
Results disability score. The only significant independent variable
in the model was baseline disability (P < .001) which
Participant characteristics explained 34% of the variance (95%CI: 22%-44%). There
was no association between the change in disability and the
A total of 150 participants were assessed for eligibility in other independent variables included in the model: ten-
the RoCTEx trial. One hundred participants who fulfilled dinopathy subgroup (P ¼ .25), corticosteroid injection
the eligibility criteria and signed informed consent to (P ¼ .17), intervention allocated (P ¼ .09), and absolute
participate were included. Thirty-two participants did not difference in SSTT ratio (P ¼ .40).
take part in the 12-week follow-up US evaluation leaving
68 participants (PHLE: 28 and LLE: 40) for the current
analyses (Fig. 1). In the Dub
e et al trial, 123 adults were Discussion
eligible to be included out of 245 potential participants.
Thirty-two participants did not take part in the 12-week Main findings
follow-up US evaluation leaving 91 participants (strength-
ening/high load: 28; motor control/low load: 30; education When analyzed by baseline categories of tendinopathy
only: 33), for the current analyses (Fig. 1). Thus, data from subgroup, there were significant within and between-group
159 participants (high load: 56, low load: 70, no exercises: differences in SSTT ratio in adults with RCRSP. The
33) were combined for the current study. Baseline thickness of ‘‘Degenerative’’ tendons significantly
Therapeutic exercises and supraspinatus tendon thickness 1923

Table I Participants’ baseline characteristics presented as mean (SD)

Degenerative Normal tendon group Reactive tendinopathy All participants
tendinopathy group (n ¼ 84) group (n ¼ 49) (n ¼ 159)
(n ¼ 26)
Age (yr) 46 (13) 47 (13) 48 (13) 47 (13)
Sex 14 F; 12 M 44 F; 40 M 28 F; 21 M 86 F; 73 M
Height (cm) 172 (10) 173 (10) 173 (7) 173 (9)
Weight (kg) 76 (15) 78 (16) 78 (15) 78 (15)
BMI (kg/m2) 25.7 (4.1) 26.0 (4.2) 26.2 (4.4) 26.0 (4.2)
Disability score 26.5 (13.0) 30.4 (13.9) 32.3 (14.7) 30.4 (14.0)
(0 [no disability]-100
[maximum disability])
BMI, Body mass index; SD, standard deviation.
No significant between-group differences.

Table II Mean tendon thickness ratio at baseline and 12-week follow-up based on allocated intervention
Baseline 12 Weeks Within group change
Mean (SD) Mean (SD) Mean (95%CI)
No exercises (n ¼ 33) 1.04 (0.14) 1.03 (0.13) 0.01 (0.06 to 0.04)
Low-load exercises (n ¼ 70) 1.02 (0.17) 1.01 (0.16) 0.01 (0.05 to 0.03)
High-load exercises (n ¼ 56) 1.04 (0.13) 1.01 (0.13) 0.03 (0.08 to 0.01)
SD, standard deviation; CI, confidence interval.
No significant between-group differences.

Table III Mean tendon thickness ratio at baseline and 12-week follow-up based on tendinopathy subgroup
Baseline 12 Weeks Within group change
Mean (SD) Mean (SD) Mean (95% CI)
Degenerative tendinopathy (n ¼ 26) 0.80 (0.08) 0.94 (0.15) 0.14 (0.09 to 0.19)
Normal (n ¼ 84) 1.01 (0.06) 0.99 (0.12) 0.01 (0.04 to 0.02)
Reactive tendinopathy (n ¼ 49) 1.19 (0.08) 1.08 (0.15) 0.11 (0.16 to 0.06)
SD, standard deviation; CI, confidence interval.
At baseline, between-group differences were all lower than P < .001.
At 12 weeks, significant difference between reactive and degenerative (P < .001) and between reactive and normal (P ¼ .002).
For within-group change, between-group differences are all lower than P < .001.

increased, while the thickness of ‘‘Reactive’’ tendons This may serve as a caution for researchers investigating
significantly decreased, and no significant change changes in tendon morphology related to interventions. Our
occurred for tendons subgrouped as ‘‘Normal.’’ Results findings suggest comparing subgroups of RCRSP (ie,
were similar no matter the intervention the participants reactive, degenerative) may lead to very different findings
were allocated to, regardless of whether they performed than grouping all presentations under 1 blanket term. This
exercises or not. is not current common practice for RCRSP research, but it
is gaining popularity in studies looking at subgroups of
Structural changes in tendons patients presenting with Achilles tendinopathy with early
beneficial findings.16,15 Indeed, Achilles tendinopathy
Our findings indicate that over time SSTT normalize subgroups were associated with different recovery trajec-
compared to the contralateral unaffected tendon, in patients tories, which suggests the value of identifying patient
within a ‘‘Degenerative’’ or ‘‘Reactive’’ subgroup of subgroups in the management of tendinopathy.15
RCRSP, irrespective of the exercise modality they were Due to the small number of studies on the subject and the
allocated to. These results may illustrate how different small sample size of these studies, it is rather difficult to
subgroups of ‘‘symptomatic’’ tendons behave over time. compare our results with previous findings for RCRSP. A
1924 M.-O. Dub
e et al.

Tendon thickness in relation to patient-reported

outcome measures

Analyses did not highlight significant differences between

tendinopathy subgroups for the change in disability. The
only significant covariate was initial disability, which could
be expected since those with a higher baseline score have
more room for improvement and thus may exhibit a greater
change in outcome over the study period.
The absence of a significant difference between different
subgroups of imaging features is not unexpected as previ-
ous studies have shown that rotator cuff imaging findings
do not correlate well with symptoms.12,40,46,54 A 2018
systematic review including 56 articles found no significant
association, among high-quality cross-sectional studies,
between imaging features and symptoms for rotator cuff
structures.54 These results demonstrate that imaging fea-
tures are not useful on their own to inform the clinical state
of patients and their prognosis. However, the generalization
Figure 2 Mean tendon thickness ratio change between baseline
of the results to a large population, as is the case with cross-
and 12-week follow-up based on tendinopathy subgroup. Error
bars represent 95% confidence interval. sectional studies, does not make it possible to highlight the
specific characteristics of certain subgroups of patients for
whom morphological changes could play a more crucial
previous study found a US-assessed 1.4 mm reduction in role in the development and persistence of their shoulder
supraspinatus tendon thickness, after 5.5 months of exercise pain.
(single-case study),53 whereas a case series study of 6 pa- Furthermore, it is interesting to note that, within a sample
tients found no change in supraspinatus tendon thickness (by of individuals with Achilles tendinopathy, researchers have
magnetic resonance imaging), after 3 months of exercises.45 shown that patients classified within a Structure-Dominant
However, the different criteria used for the diagnosis of subgroup, characterized as patients with substantial tendon
RCRSP, in addition to the very small sample sizes, implying damage (increased tendon thickness and cross-sectional
a high risk of bias, may explain the differences in findings, area), experienced delayed improvement in symptoms.15 In
and making comparison with the current results limited. Our addition, it was the only subgroup among the 4 included in
current results are however in line with previous studies on their study (Activity-Dominant, Function-Dominant, Psy-
Achilles and patellar tendinopathy where a reduction in chosocial-Dominant, and Structure-Dominant) to not ach-
tendon thickness in ‘‘reactive/thickened’’ tendons was also ieve structural recovery after 24 weeks. In contrast to our
found after progressive high load or eccentric exer- own, those results demonstrate that there is potential for a
cises.24,25,44 In contrast, patients with ‘‘degenerative/thin- structure-focused subgroup to inform prognosis. However,
ned’’ tendons were also included in the current trial, since those results apply to the Achilles tendon and would not
this feature is commonly observed in the shoulder, whereas necessarily apply to the rotator cuff tendons.
only ‘‘reactive/thickened’’ tendons were included in the
Achilles and patellar tendinopathy studies as they are the Strengths and limitations
most common finding.6,36 Therefore, since there is no other
study evaluating the effect of long-term exercise programs Limitations are acknowledged. First, secondary analyses of
on structural changes in the ‘‘degenerative/thinned’’ sub- data are exploratory. No a-priori sample size calculation
group of tendinopathy, comparison of these results is limited. was performed for the analyses based on tendinopathy
To improve our understanding of long-term changes in subgroups, and thus a lack of power could mean that some
rotator cuff tendon thickness, it is worth looking at what significant results regarding SSTT ratio changes based on
happens in the short term, particularly in a population tendinopathy subgroups were possibly not detected in the
presenting with RCRSP. One study found an acute increase education group. However, with these exploratory analyses,
in SSTT after a single bout of fatiguing exercise to the RC, we hoped to provide hypotheses using the rationale for
in patients with RCRSP, which was sustained for subgrouping used in other tendinopathy research. Second,
>6 hours.39 Tendons may acutely become thickened due to we pooled results from 2 trials with some differences in
increased tendon fluid, whereas the observed long-term methods and interventions. This is why we chose to use
changes may be due to additional collagen synthesis or ratio instead of the absolute value of supraspinatus tendon
restoration of fibril density and small-size fibrils.9,14,48,56 thickness measurement. Doing so allowed the pooling of
Therapeutic exercises and supraspinatus tendon thickness 1925

Table IV Mean change in tendon thickness ratio in each exercise group, from baseline to 12-week follow-up based on tendinopathy
subgroup
No exercises High-load exercises Low-load exercises Between-group difference
Mean (95% CI) Mean (95% CI) Mean (95% CI) between both exercise
(n ¼ 33) (n ¼ 56) (n ¼ 70) groups
Mean (95% CI)
Degenerative 0.12 (0.02 to 0.26) 0.18 (0.04 to 0.31) 0.13 (0.07 to 0.19) 0.04 (0.10 to 0.18)
tendinopathy (n ¼ 3) (n ¼ 7) (n ¼ 16)
(n ¼ 26)
Normal (n ¼ 84) 0.01 (0.04 to 0.06) 0.03 (0.07 to 0.02) 0.01 (0.06 to 0.04) 0.01 (0.09 to 0.07)
(n ¼ 23) (n ¼ 28) (n ¼ 33)
Reactive tendinopathy 0.12 (0.32 to 0.07) 0.11 (0.18 to 0.04) 0.11 (0.19 to 0.02) 0.01 (0.14 to 0.13)
(n ¼ 49) (n ¼ 7) (n ¼ 21) (n ¼ 21)
CI, confidence interval.
No significant between-group differences.

data and reduced some of the systematic differences that

may have been observed if absolute values of supraspinatus future studies with adequate sample size, calculated
tendon thickness between the 2 US measurement methods a-priori, should be conducted to confirm and expand
were used. We also believed that the results of both high- upon our current findings.
load and low-load exercise regimens could be grouped
because their content was similar, although not identical,
especially regarding their prescription parameters. The
strengths of the current study are that only 1 investigator, Acknowledgment
blinded to exercise modality, captured all US images and
performed all measurements of tendon thickness for each of The authors thank all individuals who participated in the
the trials using validated reliable methods. A control group, original trials. We acknowledge the support and contri-
including participants who did not perform exercises, was bution of the physiotherapy departments at Hospital
part of the analyses. Finally, subgrouping participants into Lillebaelt – Vejle Hospital, Aalborg University Hospital
baseline tendinopathy stages allowed us to highlight some – Farsø Hospital, and Odense municipality – Holluf-
differences between different clinical presentations of gaard Rehabilitation Department. Steen Rasmussen
RCRSP. helped in the initiation of the project at the Shoulder
Unit at Aalborg University Hospital. We also want to
acknowledge Jean Tittley and Fr
ed
erique Dupuis for
their help with data collection.
Conclusion
After the 12-week intervention, the supraspinatus tendon
thickness ratio significantly increased back to a normal Disclaimers:
range for those in the ‘‘Degenerative’’ subgroup, while it
significantly decreased back to a normal range for those Funding: The RoCTEx was funded by the Region of
in the ‘‘Reactive’’ subgroup and did not significantly Southern Denmark’s Research fund, The Danish Rheu-
change for those in the ‘‘Normal’’ subgroup. Those matism Association, The Ryholts Foundation, and the
findings were true for those allocated to exercise in- Danish Physiotherapy Association Foundation for
terventions, as those significant changes were not found Advancement of Research, Education and Information,
in those who received education only, but this lack of while the Dub
e et al trial was supported by the Quebec
difference can be explained by a lack of power as sta- Rehabilitation Research Network (REPAR). KGI was
tistical trends observed were the same as in the other funded by 1-year fellowships from 1) Hospital Lillebaelt
groups. However, the type of exercise did not influence – Vejle Hospital, 2) University of Southern Denmark,
the results. Analyses did not reveal an association be- and 3) Region of Southern Denmark. M-OD received a
tween tendinopathy subgroups or change in tendon Doctoral Training Scholarship from the Fonds de
thickness and change in disability. Since this was an Recherche Qu
ebec-Sant
e (FRQ-S). J-SR and FD are
exploratory study based on secondary analyses of data, supported by salary awards from the FRQ-S. The study
1926 M.-O. Dub
e et al.

13. Failla MJ, Mintken PE, McDevitt AW, Michener LA. Trajectory of
sponsors had no role in the design, collection of data, patient-rated outcomes and association with patient acceptable
symptom state in patients with musculoskeletal shoulder pain. J Man
analysis, interpretation of data, the writing of the article, Manip Ther 2023;31:279-86. https://doi.org/10.1080/10669817.2022.
or the decision to submit the publication. 2137350
Conflicts of interest: The authors, their immediate fam- 14. Han S, Gemmell SJ, Helmer KG, Grigg P, Wellen JW, Hoffman AH,
ilies, and any research foundation with which they are et al. Changes in ADC caused by tensile loading of rabbit achilles
affiliated have not received any financial payments or tendon: evidence for water transport. J Magn Reson 2000;144:217-27.
15. Hanlon SL, Pohlig RT, Silbernagel KG. Differences in recovery of
other benefits from any commercial entity related to the tendon health explained by midportion achilles tendinopathy sub-
subject of this article. groups: a 6-month follow-up. J Orthop Sports Phys Ther 2023;53:1-
Data sharing: Data are available upon reasonable 18. https://doi.org/10.2519/jospt.2023.11330
request. The data that support the findings of this study 16. Hanlon SL, Pohlig RT, Silbernagel KG. Beyond the diagnosis: using
are available from the corresponding author upon patient characteristics and domains of tendon health to identify latent
subgroups of achilles tendinopathy. J Orthop Sports Phys Ther 2021;
reasonable request. 51:440-8. https://doi.org/10.2519/jospt.2021.10271
17. Huisstede BMA, Wijnhoven HA, Bierma-Zeinstra SM, Koes BW,
Verhaar JA, Picavet S. Prevalence and characteristics of complaints of
the arm, neck, and/or shoulder (CANS) in the open population. Clin J
Pain 2008;24:253-9. https://doi.org/10.1097/AJP.0b013e318160a8b4
References 18. Hunter DJ, Rivett DA, McKiernan S, Snodgrass SJ. Acromiohumeral
1. Bohm S, Mersmann F, Arampatzis A. Human tendon adaptation in distance and supraspinatus tendon thickness in people with shoulder
response to mechanical loading: a systematic review and meta- impingement syndrome compared to asymptomatic age and gender-
analysis of exercise intervention studies on healthy adults. Sports matched participants: a case control study. BMC Muscoskel Disord
Med Open 2015;1:7. https://doi.org/10.1186/s40798-015-0009-9 2021;22:1004. https://doi.org/10.1186/s12891-021-04885-3
2. Brumitt J, Cuddeford T. Current concepts of muscle and tendon adaptation 19. Ingwersen KG, Christensen R, Sørensen L, Jørgensen HR, Jensen SL,
to strength and conditioning. Int J Sports Phys Ther 2015;10:748-59. Rasmussen S, et al. Progressive high-load strength training compared
3. Chester R, Jerosch-Herold C, Lewis J, Shepstone L. Psychological with general low-load exercises in patients with rotator cuff tendin-
factors are associated with the outcome of physiotherapy for people opathy: study protocol for a randomised controlled trial. Trials 2015;
with shoulder pain: a multicentre longitudinal cohort study. Br J Sports 16:27. https://doi.org/10.1186/s13063-014-0544-6
Med 2018;52:269-75. https://doi.org/10.1136/bjsports-2016-096084 20. Ingwersen KG, Hjarbaek J, Eshoej H, Larsen CM, Vobbe J, Juul-
4. Chester R, Khondoker M, Shepstone L, Lewis JS, Jerosch-Herold C. Kristensen B. Ultrasound assessment for grading structural tendon
Self-efficacy and risk of persistent shoulder pain: results of a Classi- changes in supraspinatus tendinopathy: an inter-rater reliability study.
fication and Regression Tree (CART) analysis. Br J Sports Med 2019; BMJ Open 2016;6:e011746. https://doi.org/10.1136/bmjopen-2016-
53:825-34. https://doi.org/10.1136/bjsports-2018-099450 011746
5. Cook JL, Purdam CR. Is tendon pathology a continuum? A pathology 21. Ingwersen KG, Jensen SL, Sørensen L, Jørgensen HR, Christensen R,
model to explain the clinical presentation of load-induced tendinop- Søgaard K, et al. Three months of progressive high-load versus
athy. Br J Sports Med 2009;43:409-16. https://doi.org/10.1136/bjsm. traditional low-load strength training among patients with rotator cuff
2008.051193 tendinopathy: primary results from the double-blind randomized
6. Cook JL, Rio E, Purdam CR, Docking SI. Revisiting the continuum controlled RoCTEx trial. Orthop J Sports Med 2017;5:
model of tendon pathology: what is its merit in clinical practice and 2325967117723292. https://doi.org/10.1177/2325967117723292
research? Br J Sports Med 2016;50:1187-91. https://doi.org/10.1136/ 22. Karjalainen TV, Jain NB, Page CM, L€ahdeoja TA, Johnston RV,
bjsports-2015-095422 Salamh P, et al. Subacromial decompression surgery for rotator cuff
7. Croft P, Pope D, Silman A. The clinical course of shoulder pain: disease. Cochrane Database Syst Rev 2019;1:CD005619. https://doi.
prospective cohort study in primary care. Primary Care Rheumatology org/10.1002/14651858.CD005619.pub3
Society Shoulder Study Group. BMJ 1996;313:601-2. 23. Khan KM, Scott A. Mechanotherapy: how physical therapists’ pre-
8. Dean BJF, Dakin SG, Millar NL, Carr AJ. Review: emerging concepts scription of exercise promotes tissue repair. Br J Sports Med 2009;43:
in the pathogenesis of tendinopathy. Surgeon 2017;15:349-54. https:// 247-52. https://doi.org/10.1136/bjsm.2008.054239
doi.org/10.1016/j.surge.2017.05.005 24. Kongsgaard M, Kovanen V, Aagaard P, Doessing S, Hansen P,
9. Dideriksen K, Sindby AKR, Krogsgaard M, Schjerling P, Holm L, Laursen AH, et al. Corticosteroid injections, eccentric decline squat
Langberg H. Effect of acute exercise on patella tendon protein syn- training and heavy slow resistance training in patellar tendinopathy.
thesis and gene expression. SpringerPlus 2013;2:109. https://doi.org/ Scand J Med Sci Sports 2009;19:790-802. https://doi.org/10.1111/j.
10.1186/2193-1801-2-109 1600-0838.2009.00949.x
10. Dub
e M-O, Desmeules F, Lewis JS, Roy J-S. Does the addition of 25. Kongsgaard M, Qvortrup K, Larsen J, Aagaard P, Doessing S, Hansen P,
motor control or strengthening exercises to education result in better et al. Fibril morphology and tendon mechanical properties in patellar
outcomes for rotator cuff-related shoulder pain? A multiarm rando- tendinopathy: effects of heavy slow resistance training. Am J Sports
mised controlled trial. Br J Sports Med 2023;57:457-63. https://doi. Med 2010;38:749-56. https://doi.org/10.1177/0363546509350915
org/10.1136/bjsports-2021-105027 26. Laslett M, Steele M, Hing W, McNair P, Cadogan A. Shoulder pain in
11. Dub
e M-O, Desmeules F, Lewis J, Roy J-S. Rotator cuff-related primary care–part 2: predictors of clinical outcome to 12 months. J
shoulder pain: does the type of exercise influence the outcomes? Rehabil Med 2015;47:66-71. https://doi.org/10.2340/16501977-1885
Protocol of a randomised controlled trial. BMJ Open 2020;10: 27. Lazarczuk SL, Maniar N, Opar DA, Duhig SJ, Shield A, Barrett RS,
e039976. https://doi.org/10.1136/bmjopen-2020-039976 et al. Mechanical, material and morphological adaptations of healthy
12. Dunn WR, Kuhn JE, Sanders R, An Q, Baumgarten KM, Bishop JY, lower limb tendons to mechanical loading: a systematic review and
et al. Symptoms of pain do not correlate with rotator cuff tear severity: meta-analysis. Sports Med 2022;52:2405-29. https://doi.org/10.1007/
a cross-sectional study of 393 patients with a symptomatic atraumatic s40279-022-01695-y
full-thickness rotator cuff tear. J Bone Joint Surg Am 2014;96:793- 28. Lewis J. The end of an era? J Orthop Sports Phys Ther 2018;48:127-9.
800. https://doi.org/10.2106/JBJS.L.01304 https://doi.org/10.2519/jospt.2018.0102
Therapeutic exercises and supraspinatus tendon thickness 1927

29. Lewis J. Rotator cuff related shoulder pain: assessment, management decreased thickness at follow up. Br J Sports Med 2004;38:8-11;
and uncertainties. Man Ther 2016;23:57-68. https://doi.org/10.1016/j. discussion 11. https://doi.org/10.1136/bjsm.2001.000284
math.2016.03.009 45. Østeras H, Torstensen TA, Øster as B. High-dosage medical exercise
30. Lewis JS. Subacromial impingement syndrome: a musculoskeletal therapy in patients with long-term subacromial shoulder pain: a ran-
condition or a clinical illusion? Phys Ther Rev 2011;16:388-98. domized controlled trial. Physiother Res Int 2010;15:232-42. https://
https://doi.org/10.1179/1743288X11Y.0000000027 doi.org/10.1002/pri.468
31. Lewis JS. Rotator cuff tendinopathy: a model for the continuum of 46. Pietroski AD, Burdick GB, Warren JR, Franovic S, Muh SJ. Patient-
pathology and related management. Br J Sports Med 2010;44:918-23. Reported Outcomes Measurements Information System (PROMIS)
https://doi.org/10.1136/bjsm.2008.054817 upper extremity and pain interference do not significantly predict ro-
32. Lewis JS. Rotator cuff tendinopathy. Br J Sports Med 2009;43:236-41. tator cuff tear dimensions. JSES Int 2021;6:56-61. https://doi.org/10.
https://doi.org/10.1136/bjsm.2008.052175 1016/j.jseint.2021.10.003
33. Lewis JS, McCreesh K, Barratt E, Hegedus EJ, Sim J. Inter-rater 47. Puzzitiello RN, Patel BH, Forlenza EM, Nwachukwu BU, Allen AA,
reliability of the shoulder symptom modification procedure in people Forsythe B, et al. Adverse impact of corticosteroids on rotator cuff
with shoulder pain. BMJ Open Sport Exerc Med 2016;2:e000181. tendon health and repair: a systematic review of basic science studies.
https://doi.org/10.1136/bmjsem-2016-000181 Arthrosc Sports Med Rehabil 2020;2:e161-9. https://doi.org/10.1016/j.
34. Lewis JS, Raza SA, Pilcher J, Heron C, Poloniecki JD. The prevalence asmr.2020.01.002
of neovascularity in patients clinically diagnosed with rotator cuff 48. Riley GP, Harrall RL, Constant CR, Chard MD, Cawston TE,
tendinopathy. BMC Musculoskelet Disord 2009;10:163. https://doi. Hazleman BL. Glycosaminoglycans of human rotator cuff tendons:
org/10.1186/1471-2474-10-163 changes with age and in chronic rotator cuff tendinitis. Ann Rheum
35. Luime JJ, Koes BW, Hendriksen IJM, Burdorf A, Verhagen AP, Dis 1994;53:367-76.
Miedema HS, et al. Prevalence and incidence of shoulder pain in the 49. Roy J-S, Moffet H, H
ebert LJ, Lirette R. Effect of motor control and
general population; a systematic review. Scand J Rheumatol 2004;33: strengthening exercises on shoulder function in persons with
73-81. https://doi.org/10.1080/03009740310004667 impingement syndrome: a single-subject study design. Man Ther
36. Matthews W, Ellis R, Furness JW, Rathbone E, Hing W. Staging 2009;14:180-8. https://doi.org/10.1016/j.math.2008.01.010
achilles tendinopathy using ultrasound imaging: the development and 50. Scott A, Squier K, Alfredson H, Bahr R, Cook JL, Coombes B, et al.
investigation of a new ultrasound imaging criteria based on the con- ICON 2019: international scientific tendinopathy symposium
tinuum model of tendon pathology. BMJ Open Sport Exerc Med 2020; consensus: clinical terminology. Br J Sports Med 2020;54:260-2.
6:e000699. https://doi.org/10.1136/bmjsem-2019-000699 https://doi.org/10.1136/bjsports-2019-100885
37. McCreesh KM, Anjum S, Crotty JM, Lewis JS. Ultrasound measures 51. Steuri R, Sattelmayer M, Elsig S, Kolly C, Tal A, Taeymans J, et al.
of supraspinatus tendon thickness and acromiohumeral distance in Effectiveness of conservative interventions including exercise, manual
rotator cuff tendinopathy are reliable. J Clin Ultrasound 2016;44:159- therapy and medical management in adults with shoulder impingement:
66. https://doi.org/10.1002/jcu.22318 a systematic review and meta-analysis of RCTs. Br J Sports Med 2017;
38. McCreesh K, Lewis J. Continuum model of tendon pathology - where 51:1340-7. https://doi.org/10.1136/bjsports-2016-096515
are we now? Int J Exp Pathol 2013;94:242-7. https://doi.org/10.1111/ 52. Suzuki Y, Maeda N, Sasadai J, Kaneda K, Shirakawa T, Urabe Y.
iep.12029 Ultrasonographic evaluation of the shoulders and its associations with
39. McCreesh KM, Purtill H, Donnelly AE, Lewis JS. Increased supra- shoulder pain, age, and swim training in masters swimmers. Medicina
spinatus tendon thickness following fatigue loading in rotator cuff (Kaunas) 2020;57:29. https://doi.org/10.3390/medicina57010029
tendinopathy: potential implications for exercise therapy. BMJ Open 53. Torstensen TA, Meen HD, Stiris M. The effect of medical exercise
Sport Exerc Med 2017;3:e000279. https://doi.org/10.1136/bmjsem- therapy on a patient with chronic supraspinatus tendinitis. Diagnostic
2017-000279 ultrasound–tissue regeneration: a case study. J Orthop Sports Phys
40. Meroni R, Piscitelli D, Valerio S, Boria P, Perin C, De Vito G, et al. Ther 1994;20:319-27.
Ultrasonography of the shoulder: asymptomatic findings from 54. Tran G, Cowling P, Smith T, Bury J, Lucas A, Barr A, et al. What
working-age women in the general population. J Phys Ther Sci 2017; imaging-detected pathologies are associated with shoulder symp-
29:1219-23. https://doi.org/10.1589/jpts.29.1219 toms and their persistence? A systematic literature review. Arthritis
41. Michener LA, Subasi Yesilyaprak SS, Seitz AL, Timmons MK, Care Res (Hoboken) 2018;70:1169-84. https://doi.org/10.1002/acr.
Walsworth MK. Supraspinatus tendon and subacromial space param- 23554
eters measured on ultrasonographic imaging in subacromial 55. Virta L, Joranger P, Brox JI, Eriksson R. Costs of shoulder pain and
impingement syndrome. Knee Surg Sports Traumatol Arthrosc 2015; resource use in primary health care: a cost-of-illness study in Sweden.
23:363-9. https://doi.org/10.1007/s00167-013-2542-8 BMC Musculoskelet Disord 2012;13:17. https://doi.org/10.1186/1471-
42. Millar NL, Silbernagel KG, Thorborg K, Kirwan PD, Galatz LM, 2474-13-17
Abrams GD, et al. Tendinopathy. Nat Rev Dis Primers 2021;7:1. 56. Wearing SC, Smeathers JE, Hooper SL, Locke S, Purdam C, Cook JL.
https://doi.org/10.1038/s41572-020-00234-1 The time course of in vivo recovery of transverse strain in high-stress
43. Neer CS. Anterior acromioplasty for the chronic impingement syn- tendons following exercise. Br J Sports Med 2014;48:383-7. https://
drome in the shoulder: a preliminary report. J Bone Joint Surg Am doi.org/10.1136/bjsports-2012-091707
1972;54:41-50. 57. van der Windt DA, Koes BW, de Jong BA, Bouter LM. Shoulder
44. Ohberg L, Lorentzon R, Alfredson H. Eccentric training in patients disorders in general practice: incidence, patient characteristics, and
with chronic Achilles tendinosis: normalised tendon structure and management. Ann Rheum Dis 1995;54:959-64.