Physiotherapy

in the Management
of Frozen Shoulder
A Randomised Controlled Trial
of Physiotherapy Modalities
Julia Walton
Specialist Shoulder Physiotherapist

Acknowledgments
Sarah Russell
Physiotherapy department at
Wrightington

University of Central Lancashire

A blinded, randomized, controlled trial
assessing conservative management
strategies for frozen shoulder.
Russell S, Jariwala A, Conlon R, Selfe J, Richards J,
Walton M.
J Shoulder Elbow Surg. 2014 Apr;23(4):500-7.

Physiotherapy
Most initial presentations will be in
primary care to GP or
Physiotherapy

Treat effectively
Refer appropriately

Physiotherapy
CSP Guidelines
Hanchard N, Goodchild L, Thompson J, OBrien T,
Richardson C, Davison D, Watson H, Wragg M,
Mtopo S, Scott M. (2011) Evidence-based clinical
guidelines for the diagnosis, assessment and
physiotherapy management of contracted (frozen)
.shoulder v.1.6, standard physiotherapy
Hanchard NC, Goodchild L, Thompson J, O'Brien T,
Davison D, Richardson C Evidence-based clinical
guidelines for the diagnosis, assessment and
physiotherapy management of contracted (frozen)
shoulder: quick reference summary. Physiotherapy.
2012 Jun;98(2):117-20.

CSP Guidelines

Current Evidence
Lots of Probably
Limited quality
research

Current Evidence
Recent vogue for more

interventional procedures

Hydrodilitation
MUA / Capsular Release
Significant cost
implications

What we need...
Better evidence for physiotherapy
Randomnised controlled trials
Good quality methodology

Study Design
Randomised Controlled Trial of
Physiotherapy Modalities

Group Exercise Class

Individual Multimodal Physiotherapy
Home Exercise Programme

Study Design
All primary care referrals to

Wrightington physiotherapy
department with a diagnosis of
Primary Idiopathic Frozen Shoulder

Inclusion Criteria

Age 40 to 70 years old

Marked loss of active and passive global shoulder motion, with

at least 50% loss of external rotation

Normal x-rays on anteroposterior and axillary radiographs of

the glenohumeral joint

Spontaneous onset of a painful stiff shoulder

Symptoms present for at least three months
Patient reported local shoulder pain, frequently present over
either the anteromedial aspect of the shoulder extending
distally into the biceps region, or over the lateral aspect of the
shoulder extending into the lateral deltoid region.

Exclusion Criteria

Radiographic pathological findings or glenohumeral

osteoarthritis on x-ray

Local corticosteroid injection or any physiotherapy intervention

to the affected shoulder within the last three months

Prior surgery, dislocation or trauma to the affected shoulder

Inflammatory joint disease affecting the shoulder
Active medico legal involvement

Exclusion Criteria

Clinical evidence of significant cervical spine disease

Thyroid disease

Cerebral vascular accident affecting the shoulder

Bilateral frozen shoulder due to possible underlying systemic
cause

Any coronary event, post coronary artery by-pass or

catheterisation prior to the clinical appearance of frozen
shoulder

Study Design
Ethical approval
Clinical trial registration 05/Q1401/86

Conformed to CONSORT statement

(Altman et al 2001)

Computer-generated permutated

block randomisation (Statistician!)

Study Design
Baseline evaluation
Constant Shoulder Score
Oxford Shoulder Score
SF-36
Repeated 6 weeks, 6 months, and 1 year
All assessments made by independent
blind physiotherapist

Group Exercise
Class

Twice per week for 6 weeks

50min exercise circuit - 12x4 min
stations

physio (>10yrs experience shoulder

Senior
therapy)

Home Exercise Sheet and

Information Booklet

Individual Physiotherapy
2 Sessions of individual mulitimodal

physiotherapy per week for 6 weeks

Physio (>10yrs experience of shoulder

Senior
therapy)

Maitland mobilisations, soft tissue

massage, myofacial trigger point
release, heat, stretches

Home Exercise Sheet and Information

Booklet

Home Exercise
Home Exercise Sheet and
Information Booklet

pathology
pain management
posture

Statistical Analysis
Tested for Normal Distribution
Repeated measures one-way

analysis of variance (RM-ANOVA)

Pairwise comparison using Least

Squares Difference

Power Calculation
80% Power & 5% Significance
Minimal Clinically Important

Difference for Constant Score of 15

No accepted MCID
General local consensus
117 Patients, 39 per Group

Results
850 patients referred over 12
months

705 (83%) did not meet inclusion

criteria for Primary Idiopathic
Frozen Shoulder

70 declined to participate

Results
75 Patients
Group Exercise Class - 25
Individual Physiotherapy - 24
Home Exercises - 26

Demographics
Mean Age 51.1 (40-65)
Male:Female 1:1.14
Dominant Arm 53% (73% RightHanded)

Mean duration of Symptoms 5.79

months (4-10)

Results
1 pt from EC died
1 patient from IP referred for
injection at 6 months

2 patients from HE referred for

injection at 6 months

Intention to treat principal

Results
No difference between groups at
baseline

Significant improvement in

Constant Score at 6 weeks in all

groups (p<0.001)

Continued improvement in all

groups at 1 year

Results - Constant
Baseline 39.8 (18-64)
Group Exercise Class
6 Weeks 71.4 (60-89)
1 year 88.1 (71-96)

Results - Constant
Baseline 39.8 (18-64)
Group Exercise Class
6 Weeks 71.4 (60-89)
1 year 88.1 (71-96)

Results - Oxford

Results
Pairwise
Comparison
Exercise Class
vs
Physiotherapy

Mean
Difference

Standard
Error

p Value

10.7

2.871

<0.001

Exercise Class
vs Home
Exercises

20.304

2.936

<0.001

Physiotherapy
vs Home
Exercises

9.606

2.970

0.002

Results
Improvement in Constant Score was

significantly greater in Group Exercise

Class than individual physiotherapy
(p<0.001) or home exercises (p<0.001)

Individual Multimodal Physiotherapy

Significant better then HE (p=0.002)

Significance demonstrated for all domains

of Constant Score and also Oxford Score

Results
Improvement in Constant Score was

significantly greater in Group Exercise

Class than individual physiotherapy
(p<0.001) or home exercises (p<0.001)

Individual physiotherapy significantly

better then home exercises (p=0.002)

Significance demonstrated for all domains

of Constant Score and Oxford Score

Results
Significant improvement over time
Baseline and 6 weeks (p<0.001)
6 week and 6 months (p<0.001)
6 months and 1 year (p<0.001)
Both Constant and Oxford Scores

Results
Short Form - 36
Significant improvements in

bodily pain (p=0.011) mental

health (p=0.009) and social
function (p<0.001)

No other significant differences

between groups or over time

Results
SF-36 does not appear to

accurately reflect shoulder

symptoms and change

We would not recommend SF-36 as

a PROM for shoulder pathology

(Beaton 1996, Griggs 2000 Carette, 2003,

Buckbinder 2004)

Discussion
91% of patients in the group

exercise class had a clinically

important improvement in constant
score within 6 weeks

68% with individual physiotherapy

41% with home exercise
programme

Discussion
Group exercise class gives

significant improvement in
symptoms of frozen shoulder

Improvement is greater than with

individual physiotherapy

Both better than a home exercise

programme

Discussion
Hopefully removed some
Probablies....

X
X

Discussion
Group Therapy
Psychological Impact
Discuss condition with similar
patients

Reassurance
Competition

HADS analysis - improvement

significantly improved with

group and physio intervention
over home exercise group

Discussion
Primary care diagnosis of
Frozen Shoulder

ONLY 17% of referrals

had accurate diagnosis

Significant implications
for primary care
management / triage
services / care in the
community

Discussion
Prevalence: 2-3% with
female predisposition

? Based on inaccurate

primary care diagnosis

True prevalence much

lower

Equal gender

distribution (Bunker et
al)

Limitations
Principal limitation is not meeting
Power

Based upon inaccurate prevalence

and referral data

Unable to collect enough patients in

timescale

Still significant despite smaller

numbers

Limitations
Ethically unable to offer
no treatment arm

Home exercises provides a

control against
physiotherapy intervention

Home exercises probably

close to natural history

Conclusions
Group exercise class - Cost effective and
Time effective treatment option for
frozen shoulder

Individual physiotherapy more effective

than a home exercise programme

Highlights non-responders after 6 week

course

Appropriate referrals to Surgeons

Thank you