Clinical Rehabilitation 2009; 23: 622–638
Early loading in physiotherapy treatment after
full-thickness rotator cuff repair: a prospective
randomized pilot-study with a two-year follow-up
Ingrid Hultenheim Klintberg Department of Physiotherapy, Sahlgrenska University Hospital; Department of Orthopaedics and
Institute of Neuroscience and Physiology/Physiotherapy, Sahlgrenska Academy at University of Gothenburg,
Ann-Christine Gunnarsson Department of Physiotherapy, Sahlgrenska University Hospital and Department of Orthopaedics,
Sahlgrenska Academy at University of Gothenburg, Ulla Svantesson Department of Physiotherapy, Sahlgrenska University
Hospital; Department of Orthopaedics and Institute of Neuroscience and Physiology/Physiotherapy, Sahlgrenska Academy at
University of Gothenburg, Jorma Styf and Jo ` n Karlsson Department of Orthopaedics, Sahlgrenska Academy at University of
Received 8th July 2008; returned for revisions 28th September 2008; revised manuscript accepted 17th December 2008.
Objective: To describe the clinical changes following two different physiotherapy
treatment protocols after rotator cuff repair.
Design: A prospective, randomized pilot study with a two-year follow-up.
Subjects: Five women and nine men, 55 (40–64) years old, were included.
Intervention: The progressive group (n¼7) started with dynamic, specific muscle
activation of the rotator cuff the day after surgery as well as passive range of motion.
After four weeks of immobilization the loading to the rotator cuff increased and in a
progressive manner throughout the rehabilitation. In the traditional group (n¼7) the
rotator cuff was protected from loading. Patients were immobilized for six weeks and
started with passive range of motion the day after surgery. No specific exercises to
the rotator cuff were introduced during this period.
Main measures: A clinical evaluation was made preoperatively, 3, 6, 12 and 24
months after surgery. Pain rating during activity and at rest, patient satisfaction,
active range of motion and muscle strength, Constant score, hand in neck, hand in
back and pour out of a pot, as well as Functional Index of the Shoulder were used.
Results: At two years follow-up, the progressive group and traditional group scored
pain during activity visual analogue scale (VAS) 2/0mm and pain at rest 0/0mm,
respectively. The groups attained 170/175?in active abduction in standing and 70/90?
in passive external rotation while lying in supine. Using Constant score, the groups
attained 82/77 points respectively.
Conclusion: The present study showed that the progressive protocol produced no
adverse effects compared with the traditional protocol.
Address for correspondence: Ingrid Hultenheim Klintberg,
SU/Mo ¨ lndal, 431 80 Mo ¨ lndal, Sweden.
? SAGE Publications 2009
Los Angeles, London, New Delhi and Singapore10.1177/0269215509102952
Shoulder function depends on a well-coordinated
interplay as well as correct temporal sequencing
the between muscles acting on humerus and sca-
pula.1,2The role of the rotator cuff is to provide
stability in the glenohumeral joint.3It is mainly
the upward pull by the deltoid muscle and the
anterior translation by the pectoralis major and
latissimus dorsi of the humeral head that must
The main complaint in patients with rotator cuff
tears is loss of function due to pain.6,7The com-
plaints also include stiffness of the shoulder and
loss of strength in abduction and external rota-
tion.8In a prospective study by Smith et al.,7
86% of the patients with rotator cuff tears were
unable to sleep on the affected side.
The primary goal of rotator cuff repair is pain
relief together with functional
including range of motion, strength and endur-
ance.9To achieve the goal of pain-free function of
the shoulder after rotator cuff repair, the rehabili-
tation protocol must place special emphasis on
several factors. Tendon healing, the endurance
training of muscles protecting the rotator cuff
from impinging and shearing, flexibility of the pos-
terior glenohumeral capsule, the appropriate thor-
acic posture and the correct positioning of the
scapula throughout full range of motion are impor-
tant parts of rotator cuff rehabilitation.2,8,10–15
Controlled and gradually increased loading of
movements and exercises are prerequisites for
optimal tendon healing.16Exercises should be
started in the proliferative phase of healing and
should continue through the remodelling phase.17
Collagen that is stressed regains its formation and
tensile strength better than collagen that is not
stressed.17,18Short periods of exercises should be
spread out over the day.19–21
The second goal after rotator cuff repair is
strength8,9,12,22and regain tensile strength23in
tendon tissue is estimated to be approximately
To our knowledge, the role of physiotherapy
and the optimal of load on the rotator cuff
during rehabilitation after rotator cuff repair
have not been evaluated.
The aim of this study was to evaluate the out-
come of two different physiotherapy treatment
protocols for patients with full-thickness rotator
cuff repair up to two years after surgery. The
hypothesis was that after rotator cuff repair
patients would attain greater shoulder function
at an earlier stage and maintain the function
with a well-designed and controlled physiotherapy
programme consisting of early activation and pro-
gressive loads on the rotator cuff.
Fourteen patients, five women and nine men,
with a median age of 55 (40–64) years, all with a
full-thickness rotator cuff tear, who underwent
surgery involving subacromial decompression24
and repair of the torn rotator cuff completed this
study. The same surgical method was used in all
patients.25All shoulders were immobilized in a
brace with the arm at the side.
Inclusion to the study was performed in a con-
secutive manner when patients were listed for sur-
gery. A flow diagram showing the inclusion
process is shown in Figure 1.
The exclusion criteria were: no previous rotator
cuff repair to the involved shoulder, other diag-
noses that could interfere with the treatment or
shoulder function (e.g. rheumatoid arthritis, dia-
betes, neurological or psychological diseases) as
well as difficulties in reading and writing in
This study was approved by the Human Ethics
Committee at the University of Gothenburg. All
the patients gave their written informed consent
before being included in the study.
After they had been given an appointment for
surgery, patients were contacted by the authors
responsible for the study to ask if they agreed to
participate in the study. Patients who agreed were
siotherapist, who was not involved in the treat-
ment and blinded to the patients’ treatment
group. The pre-op evaluation was performed
Early loading after rotator cuff repair623
between 1 and 14 days before surgery. During sur-
gery, patients were randomized using sealed num-
bered envelopes following a random list, into two
groups: progressive group and traditional group.
The patients in the progressive group were
treated under the responsibility of the authors.
The traditional group was treated by co-worker
physiotherapists, who were familiar with the
1. Lists with names of
patients appointed for
surgery sent to IHK and
ACG n = 36
3. Patients contacted by IHK
or ACG by telephone for
information and asked to
agree to participate in the
study n = 22
4. Did not agree to participate,
n = 4
criteria n = 22
2. Fulfilled exclusion criteria n = 14
4. Agreed to participate and
contacted for pre-op
examination n = 18
5. Randomization with sealed envelopes during surgery by IHK or ACG
6. Progressive group n = 9 6. Traditional group n = 9
7. Excluded n = 2
(Rheumatoid arthritis, severe
8. Progressive group n = 7 and Traditional group n = 7 fulfilled the study
7. Excluded n = 1
(Irreparable rotator cuff)
Lost to follow-up, n = 1
Flowchart showing the inclusion process into the study.
624IH Klintberg et al.
guidelines for rehabilitation after repair of the
rotator cuff at our department.
Patients were evaluated before surgery, 6 and 12
months after surgery with all test parameters.
At three months, only pain during activity and
pain at rest were assessed. At two years, pain
during activity and pain at rest, Constant score
and Functional Index of the Shoulder (Appendix
1) were assessed.
Intensity of pain
Patients rated the intensity of pain during activ-
ity and pain at rest on the actual occasion of
assessment using the visual analogue scale (VAS
0–100mm). The anchor points were labelled ‘No
pain’ and ‘Worst imaginable pain’. VAS is found
to be valid and reliable.26Activity was not speci-
fied, but was explained to patients to refer to
patients’ daily activities at home, at work if applic-
able and during leisure-time.
Patients’ independent, subjective opinion about
the results of surgery and physiotherapy was
assessed using a Likert scale with five options.
The patients were asked to answer the following
question: How satisfied are you with the result of
treatment of your shoulder? 1: very unsatisfactory,
2: quite unsatisfactory, 3: neither satisfactory nor
unsatisfactory, 4: quite satisfactory and 5: very
satisfactory. The scale steps were defined by the
authors and the patients graded the outcome
according to their own subjective opinion.
Active range of motion
Active range of motion was measured using a
handheld goniometer,27,28in flexion, extension
and abduction when standing. External rotation
External and internal rotation in 90?of abduction
was measured while lying supine.
Measures of active range of motion were found
to be reliable.29
Muscle strength was evaluated with an isoki-
netic dynamometer (KIN-COM 500H; Kinetic
strength in the scapular plane, the patient was
seated on a standard chair, secured with Velcro
straps over the opposite shoulder. The chair posi-
tion in relation to the dynamometer was adjusted
to place the shoulder joint in 80?of elevation in
the scapular plane.32The resistance pad was
placed just distal of the elbow, which was held
straight during testing. The rotational axis of the
the shoulder joint. Gravity correction was used.
The test started with the non-injured arm.
Measurements for rotation strength were per-
formed in a standard chair with a Velcro belt
strapped over the opposite shoulder of the patient.
The dynamometer head was tilted at 60?and the
chair was adjusted in a position that determined
the shoulder joint in approximately 30?of eleva-
tion in the scapular plane. The elbow was sup-
ported by a ‘V-pad’ and the shin pad with the
load cell was placed on the distal part of the fore-
arm. The axis of rotation was aligned through the
elbow and humerus. The test ranged from 50?
internal rotation to 50?external rotation.30,31
Prior to the strength measurements, the patient
performed a 5-minute warm-up. To familiarize
themselves with the testing procedure, the patients
practised the test at submaximal level three times.
The evaluation of internal and external rotation
strength consisted of three maximal concentric
muscle actions in each direction with the angular
velocity set at 120?/s. The highest total work (J)
was selected for analysis. The evaluation of eleva-
tion strength in the scapular plane consisted of
three maximal isometric repetitions held for 5 sec-
onds. There was an interval of at least 10 seconds
between each trial. The highest peak torque (Nm)
was selected for analysis.
Patients were interviewed and examined to fill in
the subjective and objective part of the Constant
score.33,34A higher score reflects better function.
The Constant score had to be modified due to lack
of strength measures from the Isobex dynam-
ometer at the preoperative evaluation (Constant
score 75 points).
Isometric strength was evaluated at 6, 12 and
24 monthsusing anIsobex dynamometer
Early loading after rotator cuff repair 625
(Cursor AG) to fulfil the Constant score (100
points). The Isobex has been found to be a valid
and reliable evaluation tool.35The patient was
seated on a standard chair with the arm in 90?
of elevation in the plane of the scapula. The
band from the dynamometer was placed just
proximally to the processus styloideus ulnae. The
other arm hung loosely at the patient’s side.
Patients kept their legs straight and crossed at
the ankles with the heel of one foot resting on
the floor. Patients were told to keep the trunk in
an erect, straight position during the test. The test
consisted of three maximum isometric muscle
actions held for 5 seconds. The trial with the high-
est muscle action (lb) was selected for analysis.
Three standardized composite active move-
ments described, validated and tested for reliabil-
ity by Solem-Bertoft10,36– hand in neck, hand in
back and pour out of a pot – were used. In the
hand in neck test, the patient was instructed to put
both hands behind the neck with the elbows posi-
tioned as far to the side as possible. The test was
performed bilaterally to avoid compensatory
movements of the trunk, instead of one hand as
described by Solem-Bertoft. During the hand in
back test, the patient was instructed to put both
hands against the back and lift them as far as
possible in the midline. During the pour out of a
pot test, the patient was instructed to lift a pot
containing a litre of water and pour the water
into a sink in front of the patient. The patient
was asked to freeze the final position for 3–5 sec-
onds in order to facilitate the observation. The
movement-pattern and compensatory movements
were registered in each movement according to
photographs and descriptions of the scale steps.36
The scale steps were adjusted by the authors to
facilitate statistical calculations in the hand in
neck test to 0–6, in the hand in back test and
in the pour out of a pot test to 0–5.
The Functional Index of the Shoulder was
developed by the authors to evaluate shoulder
function (Appendix 1). Patients rated their ability
to perform 10 specific tasks using visual analogue
scales (0–100mm, 0¼can perform without pro-
blems; 100mm¼cannot perform). The Functional
Index of the Shoulder form was patient-adminis-
tered. The reliability of the Functional Index
of the Shoulder has been evaluated and found to
Physiotherapy treatment protocol
The physiotherapy treatment started the day
after surgery in both groups. All the patients
were asked to fill in a compliance log for the
home training programme. Patients in both
groups were instructed to perform the exercises
in front of a mirror to enhance the quality of per-
formance. Guidelines throughout the rehabilita-
tion period were based on strengthening exercises
being modified to allow pain-free motion in both
groups.11The main outlines showing differences
(marked in bold) between the treatment protocols
in the progressive group and traditional group are
shown in Appendix 2. Specific exercises for the
rotator cuff, increased loading and aquatic train-
ing38were introduced at an earlier stage in the
progressive group. For details of the home train-
ing programmes, see Appendices 3 and 4.
after rotator cuff repair
Measurement values of pain (VAS) on all test occasions for the progressive group (PG) and the traditional group (TG)
Preop 3 months6 months12 months 24 months
Pain during activity (mm)
Pain at rest (mm)
All values are presented as median (ranges).
626 IH Klintberg et al.
The physiotherapy treatment protocol for the
traditional group followed the existing guidelines
used at the department. The physiotherapy treat-
ment protocol in the progressive group was devel-
oped by the authors (IHK, ACG) and based on
(1) Increase metabolic and aerobic capacity in
collagen tissue by activating the rotator
(2) Restore humeral head depression exerted by
supraspinatus, infraspinatus, teres minor
and the long head of the biceps12,17,20,39
(3) Regain normal flexibility of the posterior
capsule to allow the humeral head to
remain centred in the fossa throughout full
range of motion1,11–13,41
(4) Regain stability and normal movement pat-
terns of the scapula by serratus anterior, tra-
pezius superior and inferior2,11,12,19,42
(5) Restore correct thoracic posture and the
retracted position of the scapula.10,12
The training in the physiotherapy department
for both groups with regard to type of exercise,
loading, sets and reps was modified depending
on the clinical progress of the individual patient.
Exercises including eccentric loading were per-
formed at the physiotherapy department with
pulley systemsor dumbbells
patients to prolong the eccentric phase for the
rotator cuff during the exercises.
According to the protocol guidelines there were
no differences in the number of visits to the phy-
siotherapy department between the two groups.
Supervised physiotherapy was continued until
patients had returned to their earlier activity
level during work and leisure time or when no
more progress was judged to be possible by the
physiotherapist in charge.
Non-parametric tests were used due small
sample sizes. For changes over time within
groups, between follow-ups, Friedman’s two-way
analysis of variance was used. Only patients
attending all follow-ups were included in the
paired analysis of change over time.
An unpaired comparison of the differences
between preoperative values and each follow-up
assessment was made between the two groups
with the Mann–Whitney U-test. A level of
P?0.05 was regarded as statistically significant.
There were no significant differences in terms of
age, gender or any of the evaluated outcome para-
meters between the two groups at the preoperative
evaluation. The tear size is measured in centi-
According to the system of grading tear size pre-
sented by DeOrio and Cofield43the progressive
group (n¼7) consisted of six medium tears
(1–3cm) and one large tear (3–5cm), while the
traditional group (n¼7), consisted of three
medium tears, three large tears and one massive
In the progressive group as well as in the tradi-
tional group the median size of retraction was 2cm
(1–3cm). In the progressive group there were four
patients and in the traditional group five patients
who had ruptured the rotator cuff by a trauma.
No patients had concomitant rupture of the long
head of biceps.
From the traditional group, one man was
excluded because of irreparable tear of the rotator
cuff. One man with incomplete reconstruction who
re-ruptured the rotator cuff two weeks after sur-
gery attended the follow-up at 3 and 6 months, but
not at one and two years. From the progressive
group, one man was excluded the day after surgery
because of rheumatologic disorder, not known
to the authors at the time of inclusion to the
study. This patient developed postoperative infec-
tion and underwent repeated surgical debride-
ment. A second man was excluded one month
after surgery, because of postoperative infection
and surgical debridement. Seven patients in each
group completed the study over two years.
One subject in the progressive group did not
attend the three-month follow-up, one subject in
the progressive group did not attend the six-month
follow-up and for a third subject in the progressive
group the KIN-COM was out of order at the
Early loading after rotator cuff repair627
The progressive group showed significant reduc-
tions in pain at rest and during activity (Figure 2,
Table 1) and in Functional Index of the Shoulder
(Figure 3, Table 2) as well as significant increase in
the Constant score (Figure 4, Table 2), hand in
neck and pour out of a pot (Table 2) over time.
The traditional group showed significant reduc-
tion in pain during activity (Figure 2, Table 1) and
in Functional Index of the Shoulder (Figure 3,
Table 2) as well as significant increase in range
of motion in abduction (Figure 5), isometric
strength in abduction and pour out of a pot
(Table 2) over time.
Tables 1 and 2 show the results at all test occa-
sions on all available data.
The progressive group (n¼7) showed signifi-
cantly larger reduction in pain during activity
compared with the traditional group (n¼7)
between the preoperative test occasion and two-
year follow-up (P?0.05) (Figure 2). The progres-
sive group showed significantly larger reduction in
pain at rest compared with the traditional group
between the preoperative test occasion and the
one-year follow-up (P?0.05), and the two-year
follow-up (P?0.05) (Figure 2).
All the patients in the progressive group (n¼7)
reported that they were satisfied with the results
(4 or 5 on the Likert scale) at the one-year follow-
up and very satisfactied (5 on the Likert scale) at
the two-year follow-up.
In the traditional group (n¼7) all the patients
reported satisfaction with the results (4 or 5 on the
Likert scale) at one year. Six patients reported
satisfaction with the results and one reported dis-
satisfaction (2 on the Likert scale) at the two-year
The Constant score showed that all the patients
in the progressive group experienced undisturbed
sleep at one and two years. In the traditional
group all the patients experienced undisturbed
sleep at two years.
The present study showed that the progressive
group produced no adverse effects compared
with the traditional protocol. Therefore, we feel
that a more progressive approach may be a step
Pre-op 3 months 6 months12 months 24 months
Pain during activity, TG
Pain during activity, PG
Pain at rest, TG
Pain at rest, PG
the traditional group (TG). Pre-op values denote absolute values and each follow-up values denote median of individual
reduction in relation to pre-op values. Comparisons between groups showed that the progressive group reached
significantly larger reductions in pain during activity at the two-year follow-up and for pain at rest at the one- and two-year
Reduction over time of pain during activity and pain at rest (VAS, 0–100mm) in the progressive group (PG) and
628 IH Klintberg et al.
forward into new ideas for rehabilitation of
patients after rotator cuff repair without taking
any substantial risks. Previous studies have
shown that surgery and physiotherapy signifi-
cantly improve pain, the ability to sleep on the
tion.3,6,9,16,22,41,44–46A large number of scores
and methods for evaluating postoperative results
have been used.3,6,9,16,22,41,44–46At the time of set-
up of this study, no consensus in evaluating results
after rotator cuff repair existed in the literature.
Consequently specific comparisons involving the
present study are difficult.
In the present study, pain during activity and at
rest decreased by approximately by 50% within
three months postoperatively in both groups.
Patients are more affected by pain during activity
than at rest. Preoperatively, the progressive group
scored 73mm during activity compared with
27mm at rest using the VAS. Although the pro-
gressive group reported a slightly higher assess-
ment of pain preoperatively, the progressive
group reported more pain reduction at all
follow-ups than the traditional group.
With regard topatient
patients were very or quite satisfied with the
results after surgery and physiotherapy. In the
traditional group, the only patient who was
unable to return to full-time work as a smith was
dissatisfied with the results at two years. The
patients’ independent, subjective opinion of the
results provides valuable information and should
be regarded as a complement to other outcome
In our series of patients, both groups regained a
higher active range of motion in flexion than that
reported by Harryman et al.41and better abduc-
tion than that reported by Hawkins et al.6
In Harryman and colleagues’ series,41patients
with small tears improved to 129?and with
medium-sized tears 119?in active flexion. In the
study by Hawkins et al.,6patients improved to
123?in active abduction. In the present study,
patients in both groups improved to 150?in
active flexion and to 170?in active abduction.
This might be due to a specific rehabilitation pro-
gramme with more emphasis on restoring muscu-
lar capacity than was used at the time of the
investigations by Harryman et al.41and Hawkins
In a randomized study, Lastayo et al.16com-
pared manual passive range of motion versus con-
tinuous passive motion for the first four weeks
postoperatively. Thereafter patients followed a
Pre-op 6 months 12 months 24 months
and the traditional group (TG). Pre-op values denote absolute values and each follow-up value denotes median of individual
reduction in relation to preoperative values.
Improvement of function in the Functional Index of the Shoulder (100–0) over time in the progressive group (PG)
Early loading after rotator cuff repair 629
rehabilitation protocol similar to the traditional
group’s protocol in our study for 2–5 months.
The follow-up time ranged from 6 to 45 months.
Patient in the groups attained active flexion of
150?and 120?respectively. We agree with the
authors who acknowledge the difficulty in analys-
ing the effect of the interventions after such long
period of time.
Hayes et al.48conducted a randomized study
home training. Patients were followed up to two
years postoperatively. The majority of patients
had a full-thickness tear, although patients with
a partial thickness tear were also included.
Patients were evaluated with estimation of range
of motion (whether active or passive mode is not
stated), manual muscle testing (1–5) and the
Shoulder Service Questionnaire. We argue that
the methods used have a low sensitivity to detect
a clinical change. Together with short postopera-
tive follow-up it is difficult to conclude any
clinical message from this study or to compare
According to Neer24,49pain itself inhibits appro-
priate muscle actions and thus becomes one of the
factors influencing the progression of the degen-
erative process in the impingement syndrome.
We believe, however, that pain-free shoulder
the plane of the scapula) preoperative values and outcome at follow-ups for the progressive group (PG) and the traditional
group (TG) after rotator cuff repair
Active range of motion and muscular strength, functional assessment and isometric strength (in 90?of elevation in
Preop6 months 12 months 24 months
in adduction (degrees)
in abduction (degrees)
in abduction (degrees)
internal rotation (J)
external rotation (J)
Strength, Elevation (Nm)
Constant score (75p) 71
Constant score (100p)
All values are presented as median (ranges).
p, points; HIN, hand in neck test; HIB, hand in back test; POP, pour out of a pot test; FIS, Functional Index of the Shoulder.
630IH Klintberg et al.
function is achieved by better muscle function
in the rotator cuff itself. Rokito et al.9and
Kirschenbaum et al.46discussed the possibility of
using subacromial lidocaine injections before
isokineticmuscle testing toeliminate pain
inhibition. We believe that the assessment of
strength can be regarded as a way of evaluating
the capacity of the rotator cuff muscles to oppose
the cranial migration of caput humeri during
strenuous work.4Therefore a muscle test with
Pre-op 6 months12 months
group (TG). Pre-op values denote absolute values and each follow-up value denotes median of individual improvement in
relation to preoperative values.
Improvement in active range of motion over time in abduction in the progressive group (PG) and the traditional
Preop6 months 12 months24 months
group (TG). Pre-op values denote absolute values and each follow-up value denotes median of individual improvement in
relation to preoperative values.
Improvement of function in the Constant score (75p) over time in the progressive group (PG) and the traditional
Early loading after rotator cuff repair631
lidocaine would not show the true capacity of the
rotator cuff. It is our belief that muscle endurance
of the rotator cuff must be achieved in order to
prevent relapses when patients return to strenuous
In our series, we found a strength deficit at
the one-year follow-up (Table 2). It is estimated
that recovery of strength takes one year for
small and medium-sized tears (51–3cm) and
up to 18 months for large or massive tears
(3–45cm).4,8,12,22,23In our series, patients had
medium to massive tears (tear size of between
In our series, outcomes in strength and in inter-
nal and external rotation are presented as total
work. This was chosen to reflect the dynamic
capacity of the rotator cuff. In previous studies,
peak torque is most often presented. In internal
rotation both groups improved up to one year,
but not all individuals within the two groups had
regained their preoperative values. In external
rotation, 2/7 patients in the progressive group con-
tinued to lose strength in external rotation from
six months until one year. Both were men with
strenuous work. It is our belief that it is difficult
to compensate for 8 hours of labour with strength-
ening exercises, even if they are performed on a
daily basis. The level of poor outcome in strength
at one year may suggest that one year is too short
a time to evaluate final outcome, especially in
terms of strength development.
No postoperative evaluations of the rotator cuff
with magnetic resonance imaging or ultrasound
were included in this study. We therefore do not
know if the poor result in strength in internal rota-
tion could be due to attenuation or failure of the
Previous studies evaluating isokinetic strength
have presented the results in relation (%) to the
non-involved shoulder.9,22,44,46Rokito et al. and
Walker et al.9,22are critical of the appropriateness
of this method. A comparison between shoulders
might give the wrong impression as the non-
injured shoulder might not be regarded as
normal. We decided to present our results within
the same shoulder and have therefore not com-
pared our results with previous studies in this
Gazielly et al.3evaluated 100 shoulders pre- and
postoperatively using the Constant score with an
average follow-up of four years. Their patients
attained 82 out of 100 points. Gerber et al.50fol-
lowed 29 patients with rotator cuff repair between
2 and 5 years; in his series in patients with
points using the Constant score. In our series,
the progressive group attained 82 points and
the traditional group 77 points at the two-year
Normative values for the Constant score based
on age and gender have been established by
Katolik et al.51For men between 50 and 59
years the value is 93 points and for women in
the same age category 83 points. As we only
have seven patients in each group we decided not
to present our results for men and women sepa-
rately. Katolik et al.51showed that men produce
a significantly higher Constant score and most of
the difference emerges from differences in strength
performance. In future studies, with larger series,
it would be more appropriate to present the results
for men and woman separately.
Rahme et al.52found that the combination of
normal performance of hand in neck (HIN¼5)
and subnormal performance of pour out of a pot
(POP54) predicted a positive outcome of subacro-
mial decompression. During the hand in neck test,
normal performance is restricted by pain gener-
ated by tension in the rotator cuff. During the
pour out of a pot test, normal performance is
restricted by pain generated by the compression
of the rotator cuff and the subacromial bursa.52
All the patients in the present study gave a high
preoperative performance in the hand in neck test
and all the patients also gave a high preoperative
performance in the pour out of a pot test.
Consequently, we found that neither hand in
neck nor pour out of a pot was sufficiently sensi-
tive to reveal any loss of function in patients with
rotator cuff tears and therefore not as useful an
outcome measurement as we had expected.
According to the Functional Index of the
Shoulder questionnaire, the present study revealed
a large improvement in function at one and two
years. Seven patients regained full ability to per-
form the 10 specific tasks in the Functional Index
of the Shoulder.
One limitation of this study is the small number
of patients. However, to our knowledge, there are
632IH Klintberg et al.
shoulder physiotherapy focusing on restoring
muscular capacity and the biomechanics of the
shoulder treatment after rotator cuff repair.
In the future, it will be important to conduct
patient-oriented outcome research focusing on
the effectiveness and appropriateness of treat-
system for shoulder outcome measurement must
be established to facilitate and produce acceptable
comparisons of different treatment models.
We feel that it is important to find evidence of
how rehabilitation can be made more effective.
With a small patient population these results
should be seen as a pilot study, but they should
nevertheless encourage further research to confirm
these findings and the development of specific
physiotherapy treatment protocols for full-thick-
ness rotator cuff tears.
The authors would like to thank Karin Lind for
technical support, Gudrun Hagberg for help with
the statistical analysis and Anna-Karin Larsson
for graphical work.
Financial support was provided by University
of Gothenburg, the Va ¨ stra Go ¨ taland (research
and education) research fund and the Centre for
Sports Medicine Research.
Conflict of interest
1Kamkar A, Irrgang JJ, Whitney SL. Nonoperative
management of secondary shoulder impingement
syndrome. J Orthop Sports Phys Ther 1993; 17:
2Kibler WB. The role of the scapula in athletic
shoulder function. Am J Sports Med 1998; 26:
Gazielly DF, Gleyze P, Montagnon C. Functional
and anatomical results after rotator cuff repair.
Clin Orthop Relat Res 1994; 43–53.
Sharkey NA, Marder RA. The rotator cuff opposes
superior translation of the humeral head. Am J
Sports Med 1995; 23: 270–75.
Townsend H, Jobe FW, Pink M, Perry J.
Electromyographic analysis of the glenohumeral
muscles during a baseball rehabilitation program.
Am J Sports Med 1991; 19: 264–72.
Hawkins RJ, Misamore GW, Hobeika PE. Surgery
for full-thickness rotator-cuff tears. J Bone Joint
Surg Am 1985; 67: 1349–55.
Smith KL, Harryman II DT, Antoniou J et al.
A prospective, multipractice study of shoulder
function and health status in patients with
documented rotator cuff tears. J Shoulder Elbow
Surg 2000; 9: 395–402.
Meister K, Andrews JR. Classification and
treatment of rotator cuff injuries in the overhand
athlete. J Orthop Sports Phys Ther 1993; 18:
Rokito AS, Zuckerman JD, Gallagher MA,
Cuomo F. Strength after surgical repair of the
rotator cuff. J Shoulder Elbow Surg 1996; 5: 12–17.
Solem-Bertoft E. Physiotherapeutic analysis of pain
in the shoulder. Uppsala, Uppsala University, 1995.
Brewster C, Schwab DR. Rehabilitation of the
shoulder following rotator cuff injury or surgery.
J Orthop Sports Phys Ther 1993; 18: 422–26.
Bo ¨ hmer AS, Staff PH, Brox JI. Supervised
exercises in relation to rotator cuff disease
(impingement syndrome stages II and III):
A treatment regimen and its rationale. Physioter
Theory Pract 1998; 14: 93–105.
Millett PJ, Wilcox 3rd RB, O’Holleran JD,
Warner JJ. Rehabilitation of the rotator cuff: an
evaluation-based approach. J Am Acad Orthop
Surg 2006; 14: 599–609.
McClure P, Balaicuis J, Heiland D et al.
A randomized controlled comparison of stretching
procedures for posterior shoulder tightness.
J Orthop Sports Phys Ther 2007; 37: 108–14.
Lewis JS, Wright C, Green A. Subacromial
impingement syndrome: the effect of changing
posture on shoulder range of movement. J Orthop
Sports Phys Ther 2005; 35: 72–87.
Lastayo PC, Wright T, Jaffe R, Hartzel J.
Continuous passive motion after repair of the
rotator cuff. A prospective outcome study. J Bone
Joint Surg Am 1998; 80: 1002–11.
? A more proactive and progressive phy-
slightly faster recovery.
? No adverse affects were noted.
Early loading after rotator cuff repair 633
17Jozsa L, Kannus P. Healing and regeneration of
tendons. In Jozsa L, Kannus P. eds. Human
tendons. Champaign, IL, Human Kinetics, 1997,
Kannus P, Jozsa L, Renstro ¨ m P et al. The efffects
of training, immobilization and remobilization on
muskuloskeletal tissue. 2. Remobilization and
prevention of immobilization atrophy. Scand J
Med Sci Sports 1992; 2: 164–76.
Gazielly DF. A model of rehabilitation for repair
of the rotator cuff. In Vastamaki M, Jalovaara P.
eds. Surgery of the shoulder. New York, Elsevier
Science, 1995, 455–67.
Thompson LV. Skeletal muscle adaptations with
age, inactivity, and therapeutic exercise. J Orthop
Sports Phys Ther 2002; 32: 44–57.
Jackins S. Postoperative shoulder rehabilitation.
Phys Med Rehabil Clin N Am 2004; 15: 643–82.
Walker SW, Couch WH, Boester GA, Sprowl DW.
Isokinetic strength of the shoulder after repair of a
torn rotator cuff. J Bone Joint Surg Am 1987; 69:
Woo SL, Gomez MA, Woo YK, Akeson WH.
Mechanical properties of tendons and ligaments. II.
The relationships of immobilization and exercise
on tissue remodeling. Biorheology 1982; 19:
Neer II CS. Anterior acromioplasty for the
chronic impingement syndrome in the shoulder:
a preliminary report. J Bone Joint Surg Am 1972;
Rockwood A. Authors preferred method for
surgical treatment of full-thickness rotator cuff
tears. In Rockwood A, Matsen F, Wirth M,
Lippit S. eds. The shoulder. Vol 2. Philadephia, PA,
Saunders, 2004, 864–66.
McCormack HM, Horne DJ, Sheather S. Clinical
applications of visual analogue scales: a critical
review. Psychol Med 1988; 18: 1007–19.
Boone DC, Azen SP, Lin CM et al. Reliability of
goniometric measurements. Phys Ther 1978; 58:
Riddle DL, Rothstein JM, Lamb RL. Goniometric
reliability in a clinical setting. Shoulder measure-
ments. Phys Ther 1987; 67: 668–73.
Gajdosik RL, Bohannon RW. Clinical measure-
ment of range of motion. Review of goniometry
emphasizing reliability and validity. Phys Ther
1987; 67: 1867–72.
Dvir Z. Isokinetics of the shoulder muscles.
In Dvir Z. ed. Muscle testing, interpretation and
clinical applications. London, Churchill
Livingstone, 1995, 171–91.
Wilk KE, Arrigo CA, Andrews JR. Standardized
isokinetic testing protocol for the throwing
shoulder: the throwers’ series. Isokinet Exerc Sci
1991; 1: 63–71.
Greenfield BH, Donatelli R, Wooden MJ,
Wilkes J. Isokinetic evaluation of shoulder rota-
tional strength between the plane of scapula and
the frontal plane. Am J Sports Med 1990; 18:
Constant CR, Murley AH. A clinical method of
functional assessment of the shoulder. Clin Orthop
Relat Res 1987; 214: 160–64.
Kirkley A, Griffin S, Dainty K. Scoring systems for
the functional assessment of the shoulder.
Arthroscopy 2003; 19: 1109–20.
Leggin BG, Neuman RM, Iannotti JP,
Williams GR, Thompson EC. Intrarater and
interrater reliability of three isometric
dynamometers in assessing shoulder strength.
J Shoulder Elbow Surg 1996; 5: 18–24.
Westerberg CE, Solem-Bertoft E, Lundh I. The
reliability of three active motor tests used in painful
shoulder disorders. Presentation of a method of
general applicability for the analysis of reliability in
the presence of pain. Scand J Rehabil Med 1996; 28:
Hultenheim Klintberg I, Gunnarsson AC, Styf J,
Karlsson J. Early activation or a more
protective regime after arthroscopic subacromial
decompression – a description of clinical
changes with two different physiotherapy treatment
protocols – a prospective, randomized pilot study
with a two-year follow-up. Clin Rehabil 2008; 22:
Brady B, Redfern J, MacDougal G, Williams J.
The addition of aquatic therapy to rehabilitation
following surgical rotator cuff repair: a feasibility
study. Physiother Res Int 2008; 13: 153–61.
Fyfe I, Stanish WD. The use of eccentric training
and stretching in the treatment and prevention of
tendon injuries. Clin Sports Med 1992; 11: 601–24.
Jozsa L, Kannus P. Treatment principles in tendon
injuries and other tendon disorders. In Jozsa L,
Kannus P. eds. Human tendons. Champaign, IL,
Human Kinetics, 1997, 491–521.
Harryman 2nd DT, Mack LA, Wang KY et al.
Repairs of the rotator cuff. Correlation of
functional results with integrity of the cuff. J Bone
Joint Surg Am 1991; 73: 982–89.
Moseley Jr JB, Jobe FW, Pink M, Perry J,
Tibone J. EMG analysis of the scapular muscles
during a shoulder rehabilitation program. Am J
Sports Med 1992; 20: 128–34.
DeOrio JK, Cofield RH. Results of a second
attempt at surgical repair of a failed initial rotator-
cuff repair. J Bone Joint Surg Am 1984; 66: 563–67.
634IH Klintberg et al.
44Hartsell HD, Forwell L. Postoperative eccentric
and concentric isokinetic strength for the shoulder
rotators in the scapular and neutral planes.
J Orthop Sports Phys Ther 1997; 25: 19–25.
Iannotti JP, Bernot MP, Kuhlman JR, Kelley MJ,
Williams GR. Postoperative assessment of
shoulder function: a prospective study of
full-thickness rotator cuff tears. J Shoulder
Elbow Surg 1996; 5: 449–57.
Kirschenbaum D, Coyle Jr MP, Leddy JP et al.
Shoulder strength with rotator cuff tears. Pre- and
postoperative analysis. Clin Orthop Relat Res 1993;
Roddey TS, Cook KF, O’Malley KJ,
Gartsman GM. The relationship among strength
and mobility measures and self-report outcome
scores in persons after rotator cuff repair surgery:
impairment measures are not enough. J Shoulder
Elbow Surg 2005; 14: 95S–98S.
Hayes K, Ginn KA, Walton JR, Szomor ZL,
Murrell GA. A randomised clinical trial evaluating
the efficacy of physiotherapy after rotator cuff
repair. Aust J Physiother 2004; 50: 77–83.
Neer II CS. Impingement lesions. Clin Orthop Relat
Res 1983; 173: 70–77.
Gerber C, Fuchs B, Hodler J. The results of repair
of massive tears of the rotator cuff. J Bone Joint
Surg Am 2000; 82: 505–15.
Katolik LI, Romeo AA, Cole BJ et al.
Normalization of the Constant score. J Shoulder
Elbow Surg 2005; 14: 279–85.
Rahme H, Solem-Bertoft E, Westerberg CE et al.
The subacromial impingement syndrome.
A study of results of treatment with special
emphasis on predictive factors and
pain-generating mechanisms. Scand J Rehabil Med
1998; 30: 253–62.
Kannus P, Jozsa L, Renstro ¨ m P et al. The
efffects of training, immobilization and
remobilization on muskuloskeletal tissue.1.
Training and immobilization. Scand J Med Sci
Sports 1992; 2: 100–18.
Appendix 1 – Functional Index of the Shoulder (FIS)
How do you manage the following activities?
(1) Carry a bag or string bag
(2) Participate in sports
(3) Pour from a one litre can
(4) Do your normal work
(5) Throw a ball with your hand above your head
(6) Sleep on the affected side
(7) Put your hand in your back pocket
(8) Wash the opposite axilla
(9) Comb your hair
(10) Perform push-ups.
Early loading after rotator cuff repair635
Appendix 2 – Main outlines and differences in exercises
Main outlines and differences (bold) in exercises as well as in timing between the physiotherapy treat-
ment protocols in the progressive group (PG) and traditional group (TG). The PG were instructed using
a home training programme 1(A), 2(A) and 3(A), while the TG were instructed with 1(B), 2(B) and 3(B).
In aquatic training, both the PG and TG were instructed using the same programmes.
ROM, range of motion.
Aquatic training was performed in warm water (34?C) supervised by a physiotherapist.
Progressive group (PG) Traditional group (TG)
First day post-op Home training programme 1(A), including
activation of the rotator cuff, performed 3
times per day. Passive ROM (2–3 times/week) was
performed by the physiotherapist throughout the
physiotherapy period or until full ROM was
Removal of immobilizing sling
Home training programme 2(A), including exercises
with increased loading of the rotator cuff and
active, assisted ROM in flexion and elevation in
the plane of the scapula, performed 3 times per
Supervised physiotherapy on an outpatient basis
2–3 times/week was started
Aquatic training programme 1; active assisted
exercises performed with the arms in front of the
Increased loading of the rotator cuff by the home
training programme; isometrics and active ROM in
flexion and abduction with elbows flexed
(short lever arm)
Passive ROM including emphasis on restoring
Increase of load during supervised physiotherapy
Home training programme 3(A), increased load on
the rotator cuff by rubber-elastics, performed 2
times per day
Dynamic strengthening exercises for the rotator
cuff and scapular muscles through-out full ROM
during supervised physiotherapy
Active ROM with elbow straight (long lever arm)
Aquatic training programme 2; active,
water-resisted exercises performed throughout full
range of motion, 2 times/week
Eccentric load on the rotator cuff during
Home training programme 1(B), performed
3 times per day.
Passive ROM (2–3 times/week) was performed
by the physiotherapist throughout the phy-
siotherapy period or until full ROM was
Four weeks post-op
Six weeks post-op
Removal of immobilizing sling
Home training programme 2(B), including
activation of the internal and external rota-
tors and active assisted ROM in flexion and
elevation in the plane of the scapula
Supervised physiotherapy on an outpatient
basis 2–3 times/week was started
Eight weeks post-op
10 weeks post-op
Aquatic training programme 1, active assisted
exercises performed with the arms in front of
the body, once/week
12 weeks post-op
16 weeks post-opHome training programme 3(B)
Supervised physical therapy with dynamic
strengthening exercises for the rotator cuff and
scapular muscles through-out full ROM
Aquatic training, programme 2; active,
water-resisted exercises performed throughout
full range of motion, 2 times/week
Eccentric load on the rotator cuff during
24 weeks post. op
636IH Klintberg et al.