Journal Pre-proof

Exercise therapy is effective for improvement in range of motion,

function and pain in patients with frozen shoulder: a systematic
review and meta-analysis

Michel GCAM Mertens , Lotte Meert , Filip Struyf ,

Ariane Schwank , Mira Meeus

PII: S0003-9993(21)01366-6
DOI: [Link]
Reference: YAPMR 58322

To appear in: Archives of Physical Medicine and Rehabilitation

Received date: 19 January 2021

Revised date: 7 July 2021
Accepted date: 9 July 2021

Please cite this article as: Michel GCAM Mertens , Lotte Meert , Filip Struyf , Ariane Schwank ,
Mira Meeus , Exercise therapy is effective for improvement in range of motion, function and pain in
patients with frozen shoulder: a systematic review and meta-analysis, Archives of Physical Medicine
and Rehabilitation (2021), doi: [Link]

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.

© 2021 Published by Elsevier Inc. on behalf of The American Congress of Rehabilitation Medicine.
 1

Title: Exercise therapy is effective for improvement in range of motion, function and pain in

patients with frozen shoulder: a systematic review and meta-analysis

Authors:

1. Michel GCAM Mertens, Research Group MOVANT, Department of Rehabilitation

Sciences and Physiotherapy (REVAKI), University of Antwerp, Wilrijk, Belgium; Pain in

Motion International Research Group, Belgium. [Link]@[Link]

2. Lotte Meert, Research Group MOVANT, Department of Rehabilitation Sciences and

Physiotherapy (REVAKI), University of Antwerp, Wilrijk, Belgium; Pain in Motion

International Research Group, Belgium. [Link]@[Link]

3. Filip Struyf, Research Group MOVANT, Department of Rehabilitation Sciences and

Physiotherapy (REVAKI), University of Antwerp, Wilrijk, Belgium.

[Link]@[Link]

4. Ariane Schwank, Research Group MOVANT, Department of Rehabilitation Sciences and

Physiotherapy (REVAKI), University of Antwerp, Wilrijk, Belgium; Pain in Motion

International Research Group, Belgium; Department of Physiotherapy, Kantonsspital

Winterthur, Winterthur, Switzerland. [Link]@[Link]

5. Mira Meeus, Research Group MOVANT, Department of Rehabilitation Sciences and

Physiotherapy (REVAKI), University of Antwerp, Wilrijk, Belgium; Pain in Motion

International Research Group, Belgium; Department of Rehabilitation Sciences and

Physiotherapy, Ghent University, Ghent, Belgium. [Link]@[Link]

Running head: Exercise therapy in patients with frozen shoulder

Word Count: 298 words (Abstract)

5000 words (Introduction, Method, Results, and Discussion,)

 2

References: 102

Tables: 7

Figures: 17

Competing interests: there is no conflict of interest to report

Source(s) of support: not applicable

Acknowledgements: Lotte Meert is a PhD research fellow funded by the Research

Foundation – Flanders (FWO) [11E5720N].

Funding: This research did not receive any specific grant from funding agencies

in the public, commercial, or not-for-profit sectors

Correspondence: Mira Meeus, Research Group MOVANT, Department of Rehabilitation

Sciences and Physiotherapy (REVAKI), University of Antwerp,

Universiteitsplein 1, 2610 Wilrijk, Belgium,

[Link]@[Link]
 3

Exercise therapy is effective for improvement in range of motion, function and pain in

patients with frozen shoulder: a systematic review and meta-analysis

ABSTRACT

Objective: To determine 1) the effect of exercise therapy alone or in combination with other

interventions compared to solely exercises and programs with or without exercises? And 2)

what kind of exercise therapy or combination with other interventions is most effective?

Design: Systematic review and meta-analysis.

Participants: Patients with frozen shoulder.

Intervention: Exercise therapy as sole intervention or combined with other physical therapy

interventions.

Outcome measures: Range of motion (ROM), function, disability pain, muscle strength and

patient satisfaction.

Results: Thirty-three studies were included in the qualitative and 19 in the meta-analysis.

Preliminary evidence was found for supervised exercises to be more beneficial than home

exercises for ROM and function. Multimodal programs comprising exercises may result in

little to no difference in ROM compared to solely exercises.

Programs comprising muscle energy techniques show little to no difference in ROM when

compared to programs with other exercises. Adding stretches to a multimodal program with

exercises may increase ROM. There is uncertain evidence that there is a difference between

those programs regarding function and pain.

Preliminary evidence was found for several treatment programs including exercises to be
 4

beneficial for improvement in both passive and active ROM, function, pain, and muscle

strength. No studies used patient satisfaction as an outcome measure.

Conclusion: ROM, function and pain improve with both solely exercises and programs with

exercises, but for ROM and pain there was little to no difference between programs and for

function the evidence was uncertain. Adding exercises improve active ROM compared to a

program without exercises, while adding physical modalities has no beneficial effect. Muscle

energy techniques are a beneficial type of exercise therapy for improving function compared

to other types of exercise. Unfortunately, no conclusion can be drawn about the results in the

long-term and most effective dose of exercise therapy.

Key words:

frozen shoulder; exercise therapy; physical therapy; rehabilitation; meta-analysis.

LIST OF ABBREVIATIONS

FS: frozen shoulder

DM: Diabetes Mellitus

PT: interventions performed by physical therapists

ROM: range of motion

WoS: Web of Science

CENTRAL: Cochrane Central Register of Controlled Trials

PICO: Patient, Intervention, Comparison, Outcome

MD: mean difference

SMD: standardized mean difference

SPADI: shoulder pain and disability index

 5

CMS: Constant Murley score

RC: rotator cuff

CPM: continuous passive motion

PROM: passive range of motion

AROM: active range of motion

DASH: disabilities of arm shoulder and hand

VAS: visual analogue scale

ER: external rotation

 6

INTRODUCTION

Frozen shoulder (FS) is a common condition characterized by a spontaneous, progressive

inflammation and fibrosis of the shoulder joint capsule and the rotator interval, resulting in

functional restrictions of both active and passive shoulder range of motion (ROM).1-5 The

non-dominant shoulder is most affected6-8 and about 6%-34% of those affected will develop a

FS in the opposite shoulder.6, 8-11 FS usually develops between the ages of 40-60 years6, 8, 10

with the incidence increasing with age.12 The prevalence of primary FS in the general

population is 2-5%1, 3, 13-16 and usually more women than men are affected.6, 8-12, 17-19

Over the last two decades, there has been an increase in the incidence and prevalence of FS,

possibly due to an increase in sedentary jobs with physically low activity.12 It seems that the

occurrence of FS is higher in patients with these jobs.12 Furthermore, up to 39% of patients

with Diabetes Mellitus (DM) will develop a FS 16, 20 and they have a 5 to 7 times higher risk

of developing a FS.14

Interventions performed by physical therapists (PT) are commonly used and often

recommended for FS. Treatment of patients with a FS by a physical therapist usually starts

when the patient experiences a progressive loss of ROM and persistence of pain. PT are most

consistently prescribed to maintain and improve motion and function, but there is a lack of

consensus about which PT are most effective.17, 21-23 Traditional treatment with PT consists of

patient education, physical applications (heating or electrotherapy), joint mobilization and

exercises.14, 24 Exercises aim to improve ROM and muscle function by restoring shoulder

mobility and stability through range.25 In general, exercises include any purposeful movement

of a joint, muscle contraction or prescribed activity.26

 7

In chronic diseases and a range of musculoskeletal conditions, including FS, it has been

suggested that exercise therapy is the most effective component of PT, and as effective as

medical treatment.25, 27-32 Exercise therapy (as part of programs including multiple

interventions (multimodal protocols)) was found to be effective in reducing pain and

disability in several shoulder disorders.31, 32

It is proposed that exercise therapy might help to reduce pain and restore the range,

coordination and/or control of movements in patients with FS,33 however, this information

might be outdated, not specific for FS, or not systematically reviewed. Hence, it is uncertain

what the effects of exercises are, to what extent they work besides or in combinations with

other modalities, and which format of exercise therapy is the most effective. Indeed, exercise

therapy is usually part of a multimodal program and is often not provided as a single

intervention. However, it is interesting to know whether a program with solely exercises is as

effective as a multimodal program and what combination of interventions are most effective,

in relation to various outcome measures in the short and/or long term.

The research questions regarding patients with FS and the outcome measures ROM,

function/disability, pain, muscle strength and patient satisfaction were:

1. What is the effect of exercise therapy alone or in combination with other interventions

compared to solely exercises and programs with or without exercises?

2. What kind of exercise therapy or combination with other interventions is the most

effective?

METHOD
 8

Design

Systematic review and meta-analysis of randomized controlled trials. The review was not

prospectively registered.

Identification and selection of studies

PubMed, Web of Science (WoS) and Cochrane Central Register of Controlled Trials

(CENTRAL) were searched to identify relevant studies concerning exercise therapy in

patients with FS. The search strategy and search terms are based on a Patient, Intervention,

Comparison, and Outcomes (PICO) design. The full search strategy for PubMed is presented

in Table 1. The search terms for the three different components were combined into one

search strategy. The full strategies for WoS and CENTRAL are presented in Supplemental

Appendix S1.

The reference lists of included studies and interesting systematic reviews and meta-analyses

concerning exercise therapy in patients with FS 11, 13, 14, 17, 21, 25, 26, 33-50 were hand searched

additionally. The last search took place on 18 April 2019 and was updated 25 May 2021.

After searching the three different databases duplicates were removed by the use of Endnote

X9. The remaining studies were screened for fulfilling the inclusion criteria (Box 1) on title

and abstract by two independent reviewers (MM and LM) with the help of Rayyan.51 If title

and abstract were unclear concerning fulfilling the eligibility criteria, the full text was

retrieved and screened together with the remaining studies once more, again by the two

independent reviewers. Differences were discussed in a consensus meeting, if consensus

could not be reached the last author made the final decision.

Quality of evidence
 9

Two reviewers (MM and LM) determined the risk of bias independently by the use of the

Risk of Bias 2.0 tool.52 The ratings of both reviewers were compared and potential differences

were discussed in a consensus meeting. If disagreements occurred after the consensus

meeting, they were resolved by consulting the last author. A distinction between clinician

reported outcome measures, like ROM and patient reported outcome measures, like pain and

questionnaires was used to determine the quality of evidence for the different outcome

measures. Afterwards, the overall quality of the evidence for each outcome was rated with the

GRADE approach by the first author.53

Data analysis

All included full texts were read and information was extracted about origin, characteristics of

study participants, eligibility criteria, characteristics of exercise therapy (exercises, duration,

frequency), outcome measures, and main results. Two independent reviewers (MM and LM)

performed data extraction in a pre-defined template.

The synthesis of results was performed through meta-analysis, with the software Rev Man

5.3. Clinical homogenous studies were grouped based on intervention applied and outcome

measures used, next the I2 test determined statistical heterogeneity. With low statistical

heterogeneity (I2≤50%) the fixed effects method was used for data-analysis, else the random

effects method was used. Depending on the results in the included studies the mean difference

(MD) [95% confidence interval (CI)] was used for outcomes with the same measurement tool,

the standardized mean difference (SMD) [95% CI] was used for outcomes with a different

measurement tool. Effect measures were determined for ROM, function/disability, pain,

muscle strength, and patient satisfaction (if appropriate). If median and range or quartiles

were reported, the mean and standard deviation were estimated based on the formulas of Wan

et al.54 When included studies compared three intervention groups, groups were combined, as
 10

recommended by the Cochrane Handbook,55 depending on the comparison. The magnitude of

the effect sizes was determined based on the minimal detectable change and minimal

clinically important difference if available, otherwise arbitrary borders were determined based

on previous literature. Finally, results are presented with their effect in the short (<3 months

follow up), mid- (3-9 months follow up) and/or long term (>9 months follow up).

RESULTS

Flow of studies through the review

The study selection process is shown in Figure 1. Finally, 33 studies were included in the

qualitative analysis, of which 19 were used in the meta-analysis as well. For the first

screening, there was a 96% agreement rate between the two reviewers and for the second

screening; there was an 84.1% agreement rate. Full agreement was reached after discussion

between the two reviewers.

Quality of evidence

The risk of bias within and between studies is presented in Figure 2. Regarding clinician

reported outcome measures, being ROM, muscle strength and scapular position, overall three

studies 56-59 had high quality, four studies 59-62 had moderate quality and 20 studies had low

quality.63-83 Regarding the patient reported outcome measures, like pain and self-reported

questionnaires (e.g. Shoulder Pain and Disability Index (SPADI) and Constant Murley Score

(CMS)), overall two studies 57-59 had high quality, two studies 59, 60 had moderate quality and

28 studies 61, 62, 64-88 had low quality. Low quality was mainly due to a lack of reporting about
 11

adherence to the intervention (domain ‘Deviations from intended interventions’) in most

studies, and lack of blinding participants in studies with patient reported outcome measures.

The initial agreement rate between the two reviewers for quality assessment was 77.8%,

reaching full agreement after discussing the differences. Most differences occurred in the

deviations from the intended interventions, where one reviewer made some assumptions. The

reviewing team decided to use only information that was published.

Table 2 shows the quality of evidence determined by the GRADE approach for the different

research questions with their outcome measures. For several interventions only preliminary

evidence is available, these results are shown in Table 3.

Study characteristics

The characteristics of the included studies are presented in Table 4-7 and summarized below.

Study population

FS patients included in the studies were termed as adhesive capsulitis,56, 58, 59, 61, 62, 64-67, 72, 73, 76,
78-80, 83, 84
FS,57, 68-70, 74, 75, 82, 86, 87 FS syndrome,81, 85 periarthritis of the shoulder,60, 77 diabetic

FS,63 diabetic adhesive capsulitis,71 and scapulohumeral periarthritis.88 Twenty-one studies

did not specify the FS phase,56-59, 61-65, 67-69, 74, 79, 81-86, 88 while 7870seven studies included phase

260, 66, 71, 72, 75-77 and one study included acute phase FS,78 phase 1 and/or 2,70 phase 3 or 4,73

chronic phase87 and subacute and chronic phase.80

Treatments

One study compared supervised group exercises with home exercises,82 whereas eight studies

compared a multimodal program including exercises with solely exercises.57, 59, 65, 68, 70, 75, 82, 85

Four studies compared a multimodal program including exercises, with a multimodal program
 12

without exercises,66, 78, 79, 86 and 24 studies compared two multimodal programs (identical PT

modalities) including different exercises with each other.56, 58-64, 66, 67, 69, 71-74, 76-78, 80, 81, 83, 84, 87,
88

Treatment period varied from 1 session66 to 2 years85, with 4 weeks61, 63, 72, 73, 75, 84, 86, 88, 89 as

most common period. Other treatment durations were 2 weeks,57, 74, 77, 78, 80 3 weeks,59, 64, 79, 87

5 weeks,56, 65 6 weeks,68, 69, 82 8 weeks58, 62 and 12 weeks.60, 70, 71, 81 Most studies used the same

follow up period as their treatment period,56, 60, 61, 63-66, 70-72, 74, 75, 77, 78, 80, 83, 84, 86-88 however,

some studies used a longer follow up period up to 3 months,57, 59, 69 24 weeks,79 6 months,58, 62

8 months,68 1 year,82 and 2 years.73, 85 In addition, one study used a shorter follow up period of

8 weeks.81 The treatment frequency in the included studies varied from 2 to 6 times a week

for supervised treatment, home exercises were usually daily recommended.

Exercises

As part of the multimodal program or solely, most common types of exercises were isometric

or strengthening exercises of rotator cuff (RC), trapezius, scapular, and glenohumeral

muscles,56, 61, 63, 66, 67, 74, 76, 80 muscle energy technics (e.g. Proprioceptive Neuromuscular

Facilitation (PNF)),57, 64, 66, 69, 72, 75, 77, 84, 87 wand/wall exercises,62, 64, 66, 76, 83, 84 (Codman)

pendulum exercises,56, 59, 62, 65-67, 70-72, 75-78, 80, 83-85, 87 and stretching exercises.60, 65, 67, 69, 71, 74, 80,
84

ROM exercises,59, 67, 69, 71 functional exercises (e.g. daily activities),57, 88 scapulothoracic

exercises,58, 69 cycle ergometer exercise,56 yoga,86 position induced movement re-education,78

exercise circuit (combination of various exercises),82 and not further defined active exercises81

were less common.

Several studies incorporated a home exercise program,61, 62, 67-73, 79, 82, 85 that included various

of the above-mentioned exercises, like pendulum, ROM, wall, and scapular exercises.
 13

Treatment programs

The multimodal programs with and without exercises consisted of combinations of the

following interventions: thermotherapy,58, 60, 62, 64, 66, 69, 71, 73, 74, 77-80, 82-84 ultrasound,56, 63, 64, 66,
67, 71, 72, 87
electrotherapy,61, 62, 66, 67, 69, 71, 76, 77 manual therapy,56-58, 61-63, 65, 73-79, 81-83, 85, 87 oral

medication,68, 69, 79, 86 continuous passive motion (CPM),70, 71 laser therapy,59, 72, 78 infrared

therapy,67 a progressive stretch device,73 spray and stretch technique,76 electromagnetic

therapy,60 tuina (kind of manual therapy),88 sleepers stretch,80 and not further defined physical

modalities.81

Outcome measures

Both passive and active ROM (PROM and AROM) were measured with a goniometer in all

included studies. Apley’s scratch test was used as an alternative measurement for AROM in

one study.83 Scapular dyskinesis was assessed in only two studies and they used different

outcome measures: scapular tipping 63 and the lateral slide test.66

The included studies used various outcome measures for function/disability and pain. For

function/disability the following outcome measures were used SPADI,58, 59, 61, 64, 65, 70, 71, 75-80,
83, 84, 86
CMS,56, 69-71, 82, 85, 87, 88 Simple Shoulder Test,66 Modified Upper Extremity Motor

Activity Log,57 University of California Los Angeles scale,67 Oxford Shoulder Score,82

Disabilities of Arm, Shoulder and Hand (DASH),74 PENN score,74 patient-specific functional

scale,61 and FLEX-SF.81 Most common used measures for pain were Visual Analogue Scale

(VAS) 56, 59, 61, 64-71, 73-77, 84, 88, Numeric Rating Scale,72, 83 and McGill Pain Questionnaire.57 In

some studies the outcome measure for pain and functional ability was an ordinal scale60 or a

self-constructed measure.62
 14

Muscle strength was used in only two studies61, 76 as an outcome measure and they used a

sphygmomanometer76 and a handheld dynamometer.61

Effect of intervention

Supervised exercises compared to unsupervised exercises

Only one study82 compared supervised and unsupervised exercise interventions, class versus

home exercises, for ROM and function in the long term. There is preliminary evidence that an

exercise class increases ROM (MD: 10.96° [7.54°, 14.37°]) and function/disability (MD:

CMS, 16.10 [10.25, 21.95] points) compared to a home exercise program.

Exercises in a multimodal program compared to solely exercises

Eight included studies 57, 59, 65, 68, 70, 75, 82, 85 evaluated exercises in a multimodal program

compared to solely exercises in the short and long term. Unfortunately, one study 75 could not

be used in the meta-analysis because of a lack of information in the study and upon

information request (not answered). Figure 3-5 show the results of the meta-analysis for these

interventions on PROM, function/disability, and pain, respectively.

Four studies found that solely exercises may result in little to no difference inPROM into

flexion (MD: -3.32 [-7.23, 0.58])57, 59, 68, 70 and slightly increase internal rotation (MD: -7.64 [-

11.548, -3.75])57, 59, 65, 70 compared to exercises in a multimodal program in the short term.

Five studies 57, 59, 65, 68, 70 found that exercises in a multimodal program may result in no

difference in PROM into abduction (MD: -1.06 [-4.92, 2.80]) and external rotation (ER, MD:

-6.86 [-10.13, -3.58]) compared to solely exercises. The excluded study75 for meta-analysis

preliminary showed that exercises in a multimodal program improve active and passive ER

ROM compared to solely exercises.

 15

The efficacy of exercises in a multimodal program versus solely exercises on

function/disability and pain in the short and long term was investigated by four59, 70, 82, 85 and

three59, 65, 70 studies, respectively. The evidence is uncertain about the effect of exercises in a

multimodal program on function/disability (SMD: -0.04 [-0.56, 0.64]) compared to solely

exercises. In addition, exercises in a multimodal program may not reduce pain (MD: -1.13 [-

2.61, 0.35]) compared to solely exercises. The excluded study75 for meta-analysis preliminary

showed that exercises in a multimodal program improve function/disability compared to

solely exercises.

Exercises in a multimodal program compared to a multimodal program without exercises

Four included studies66, 78, 79, 86 evaluated exercises in a multimodal program and compared it

to a multimodal program without exercises in the short and midterm. Figure 6-9 show the

results of the meta-analysis for these interventions on PROM, AROM, function, and pain,

respectively.

Two studies78, 79 found that exercises in a multimodal program results in little to no difference

in PROM into abduction (MD: 6.12 [2.96, 9.28]) and ER (MD: 4.53 [2.22, 6.83]) compared to

a program without exercises. In addition, preliminary evidence78 was found that in the short

term programs comprising exercises slightly increase flexion (MD: 10.35 [6.20, 14.50])

ROM, but not internal rotation (MD: 2.85 [0.83, 4.87]) and extension (MD: 0.10 [-2.45,

2.56]) ROM compared to a program without exercises.

For AROM only preliminary evidence66 was found that a program with exercises increases

flexion (MD: 16.00 [14.07, 17.93]) and slightly increases abduction (MD: 9.00 [4.38, 13.62])

ROM, compared to a program without exercises.

The efficacy of these treatment programs on function/disability and pain was investigated by

three66, 78, 86 and two66, 86 studies, correspondingly. The evidence is uncertain about the effect
 16

of a program with exercises compared to one without exercises on function/disability (SMD: -

0.78 [-2.06, 0.49]), while programs comprising exercises probably do not reduce pain (SMD:

-0.06 [-0.42, 0.30]) compared to one without exercises.

Exercises in a multimodal program compared to different exercises in a multimodal program

Twenty-four studies 56, 58-64, 66, 67, 69, 71-74, 76-78, 80, 81, 83, 84, 87, 88 compared various exercises in

different programs with each other. Due to heterogeneity in outcome measures and exercise

programs, only a meta-analysis could be performed for studies comparing muscle energy

techniques (e.g. PNF) with other types of exercises, for studies comparing programs with and

without static stretching and programs comparing physical modalities with sham treatment.

Muscle energy techniques compared to different exercises in a treatment program

Seven studies 64, 66, 72, 77, 78, 84, 87 compared a type of muscle energy techniques with another

type of exercise for PROM, AROM, function and pain in the short term. The results of the

meta-analysis for these outcome measures are shown in Figure 10-13, respectively.

Based on three studies64, 77, 78 it is likely that muscle energy techniques have similar effects for

PROM (MD: 4.88° [3.24-6.51°]) and AROM (MD: 6.35 [-8.83, 21.63]),66, 72compared to

other types of exercises. Muscle energy techniques 66, 72 may improve function/disability

(SMD: -0.62 [-1.28, 0.04]),64, 66, 77, 78, 84, 87 compared to other exercises. Furthermore, the

evidence64, 66, 72, 77, 84 is uncertain about the effect of muscle energy techniques on pain (SMD:

-0.36 [-1.24, 0.52]) compared to other exercises.

Static stretching combined with exercise vs exercises in a multimodal program

Two studies 73, 80 compared adding static stretching to a multimodal program (including

thermotherapy and home exercises) to the same program without static stretching in the short
 17

and long term. The results of the meta-analysis for PROM and function are shown in Figure

14 and 15, respectively.

The evidence is uncertain about the effect of adding stretches to a multimodal program on

PROM (MD: 16.40 [7.41, 25.38]) and function/disability (SMD: -0.60 [-2.92, 1.72])

compared to the same program without stretching.

Physical modalities combined with exercises compared to sham with exercises

Two studies59, 60 compared physical modalities combined with exercises with sham treatment

in the short term. The results of the meta-analysis for PROM and pain are shown in Figure 16

and 17, respectively.

Physical modalities do not improve PROM (Overall MD: 1.51 [-4.14, 7.16]) and pain (MD:

0.10 [-0.26, 0.46]).

Various exercises in a treatment program

Thirteen studies 56, 58, 61-63, 67, 69, 71, 74, 76, 81, 83, 88 compared different types of exercises with each

other on various outcome measures. An overview of these preliminary results is shown in

Table 3 and summarized below. The results show the effect of the treatment programs in the

short term, unless indicated otherwise.

For local exercises with US a large increase in PROM in flexion was found.56 Adding

instrument-assisted soft-tissue massage,83 RC strengthening exercises,61 spray and stretch

technique,76 dynamic scapular recognition exercise58 and end-range mobilizations,81 mirror

therapy67 and local exercises with US56 61765881improve PROM (in at least one direction), and
83
CPM71 and adding instrument-assisted soft-tissue massage83 and scapulothoracic exercises69

slightly improve PROM (in at least one direction) compared to a control intervention with

exercises. 60
 18

Mirror therapy67 increases AROM (in at least one direction), while adding instrument-assisted

soft-tissue massage83 and CPM71 slightly increases AROM (in at least one direction)

compared to a control intervention with exercises.

Compared to a control intervention with exercises, an increase in function/disability was

found with mirror therapy,67 local exercises with US56 and adding spray and stretch

technique76 and RC strengthening exercises.61 In addition, a slight increase in function was

found with adding scapulothoracic exercises69 and CPM71, while no effect was found with yi

jin jing88 and additional instrument-assisted soft-tissue massage,83 scapular recognition

exercise58 and end-range mobilization.81

For pain as an outcome, only adding spray and stretch techniques showed a decrease

compared to an intervention without spray and stretch.76 Furthermore, a slight decrease in

pain was found with mirror therapy,67 adding scapulothoracic exercises69and RC

strengthening exercises,61 CPM,71 local exercises with US56 61and yi jin jing,88 while no effect

was found for additional instrument-assisted soft-tissue massage83 and additional

mobilizations74 60

Adding spray and stretch technique76 was found to increase muscle strength, and additional

RC strengthening exercises61 slightly increased muscle strength compared to a control

intervention without these interventions. Furthermore, adding lower trapezius exercises to a

program already containing exercises63 slightly decreased scapular tipping. Finally,

functional ROM was changed after additional instrument-assisted soft-tissue massage,83

however, the magnitude was unclear and scapular upward rotation did not change with an

additional scapular recognition exercise.58

DISCUSSION
 19

The first aim of the current study was to determine the effect of solely exercise or combined

with other interventions in patients with FS. Preliminary evidence shows an improvement in

ROM and function/disability of an exercise class compared to a home exercise program.

Furthermore, solely exercises may result in little to no difference in PROM and pain

compared to a multimodal program including exercises and the evidence for

function/disability is uncertain. Adding exercises to a multimodal program results in little to

no difference in PROM, probably do not reduce pain, and the evidence is uncertain about the

effects of these programs on function/disability.

The second aim was to determine what kind of exercise therapy or combined with other

interventions is most effective on ROM, function/disability, pain, muscle strength, and patient

satisfaction in these patients. It is likely that the type of exercises (muscle energy techniques

versus other type) do not result in a difference in PROM and AROM, while

function/disability may improve with muscle energy techniques. Finally, the evidence for the

effect on pain of different types of exercises is uncertain. Adding static stretches to

multimodal programs including exercises may increase ROM, but the evidence is uncertain

about the effect on function/disability.

Clinical and research implications

The results from this review implicate that exercises improve ROM, function/disability and

pain and that the type of exercise has little or no influence on this. Although the latter can

only be concluded for muscle energy techniques compared to other exercise types. For

strength training or ROM exercises not sufficient data was available to draw any conclusions.
 20

Adding (physical) modalities to exercises has no benefit for treatment outcome. Due to

heterogeneity of modalities added to the exercises no specific modalities can be excluded.

However, exercises can be performed in a home program or combined with an exercise class,

this seems to be effective as well and is more efficient and cost-effective. Although the effect

of exercise class with home program should be confirmed in future research.

Programs with exercises result in larger AROM gains than programs without exercises, no

difference was found for other outcomes. In these programs the exercises comprised mostly of

supervised exercises. The effect of a home program compared to a program without exercises

should be confirmed in future research.

The evidence for additional static stretches is uncertain, the effect on PROM is promising, but

should be confirmed with higher quality studies. Passive stretching was not included in this

review and a more extensive comparison of the effect of stretching compared to exercises was

not possible and should be investigated in future research.

Our results are in line with several other reviews, that indicate that exercises are an effective

intervention.14, 38, 48, 50 However, in most reviews, exercises were part of a multimodal

program and a more extensive comparison is not possible.

Limitations included evidence

These results were influenced by several factors, including methodological issues and

substantive differences between studies. In the next section the influence of the quality of

evidence, differences in patient characteristics, applied treatment programs, and selected

outcome measures will be discussed.

Quality of evidence
 21

In studies with modalities as intervention and subjective outcome measures, several

challenges need to be countered to blind participants.90 In addition, an intervention as exercise

therapy is difficult to compare to a placebo exercise, because the placebo exercise needs to

have the same characteristics as the ‘real’ exercise.90 Therefore in many cases this will result

in a high risk of bias. As a consequence of these results, the quality of evidence according to

the GRADE will be downgraded with one or two levels.

Another difficulty within our review was the consideration of publication bias. After the

creation of homogenous groups, the number of studies was not sufficient (5-10 is

recommended) to create a funnel plot for detection of publication bias. Which also might have

influenced the quality of evidence.

Finally, we believe that it is not fair to determine the GRADE for comparisons that include

only one study, because few domains (inconsistency, imprecision) cannot be scored correctly.

Therefore we did not rate these studies with a certainty level, but we proposed them as

preliminary evidence.

Patient characteristics

Comparison of the results between studies based on patient characteristics is difficult for

various reasons. First, there is moderate evidence of early recovery that slows with time.91 So

studies that included patients in an earlier phase could have found larger benefits of the

intervention compared to studies that included patients in a later phase. Although diagnosing

disease stage is difficult, comparability of patients could be done with tissue irritability levels.

Second, there is conflicting evidence whether patients with FS and DM have a worse

prognosis for recovery34, 92-95 and therefore it is uncertain whether these studies can be

compared to each other.

 22

Treatment programs

There is a large heterogeneity in type of exercise (e.g. supervised, home, strength training,

ROM exercises) and dose between studies that provided exercise programs solely or as part of

a treatment program. In addition, not in all studies the dose of exercises is clearly described.

These limitations make comparison between studies difficult and insufficient to prove the

most effective dose for exercise therapy. Furthermore, the heterogeneity in content and dose

of the multimodal programs prevent to provide evidence for the most effective multimodal

program as well.

Most studies use short treatment and follow up periods. Since FS is a chronic disorder with an

average disease duration of 1 to 3 years8 and time to greatest improvement from 12 to 48

months91 these short time frames may not be sufficient for realizing effective treatments. As a

consequence of these short treatment periods and time to greatest improvement from at least

12 months,91 a large improvement in most studies cannot be expected.

Outcome measures

Not all outcome measures may be valid for the FS population. In the included studies, the

CMS is one of the most commonly used outcome measures regarding function/disability.

However, its use is (up to now) only advised for patients with subacromial shoulder

disorders96 and it is only validated in English.97 If patients are unable to achieve 90° abduction

(which is the case in many patients with FS) they should receive the score zero and this might

not reflect the actual strength of this patient, but more the restriction of ROM.97, 98 In addition,

pain is measured in two elements, during self-report and as factor within pain-free ROM.97 As

a consequence of these constructs the CMS is not valid in patients with FS, because the

majority of these patients are not able to abduct their shoulders sufficient to lift the weight

reliably 99 and might move their shoulder beyond pain free range as well. This might be a
 23

reason for not finding a difference between treatment programs regarding function/disability.

For patient reported outcome measures in patients with FS, it is recommended to use the

DASH, the American Shoulder and Elbow Surgeons shoulder scale, or the SPADI.14

Clinical relevant changes for ROM, function/disability (SPADI), and pain (VAS 0-100) were

suggested to be at least 15°,100 8-13,101 and 12 mm,102 respectively. However, not for all

outcome measures minimal detectable change and minimal clinically important difference

values are present. Therefore, for some outcome measures (e.g. CMS, muscle strength) it was

difficult to determine effect sizes.

Finally, another shortcoming is the limited studies about the effect of exercises regarding the

outcomes muscle strength and patient satisfaction. Both outcomes should be more emphasized

in future studies.

Strengths

This study had several strengths, first a comprehensive set of search terms was used to search

three databases for relevant studies. Second, a hand search was performed to prevent

overlooking of relevant studies. Third, two independent reviewers performed the screening,

risk of bias assessment, and data extraction. Fourth, there was sufficient homogeneity between

studies to perform a meta-analysis.

Study limitations

Due to the lack of multiple studies investigating solely exercise programs and the

heterogeneity of the other studies comparing exercises in a multimodal program no meta-

analysis could be performed for these studies. In addition, we might have overlooked some

relevant studies, despite our comprehensive set of search terms and searching three databases.

We only selected studies written in English or Dutch, we did not search for gray literature,
 24

and we could have searched additional databases. Finally, the GRADE assessment was only

performed by one reviewer, which could have resulted in bias.

Conclusion

In conclusion, exercises (in a program or on their own) improve ROM, function/disability and

pain. However, only little to no difference was found in PROM and pain between the

programs and the effects in function/disability are uncertain. Adding physical modalities to

exercises has no benefit for treatment outcome. Compared to a program without exercises,

adding exercises improve the AROM. Regarding type of exercise can be concluded that

muscle energy techniques only improve function/disability more than other exercise types,

while no difference was found for other outcomes.

Future research should focus on the effect of exercises on muscle strength and patient

satisfaction as outcomes and results in the long term should be investigated. Moreover, the

effect of solely exercises (as class, home program or combined) should be confirmed. Finally,

the dose of exercises should be standardized to draw a conclusion.

 25

REFERENCES

1. Bunker TD. Frozen shoulder: unravelling the enigma. Ann R Coll Surg Engl

1997;79(3):210-3.

2. Hand GC, Athanasou NA, Matthews T, Carr AJ. The pathology of frozen shoulder. J

Bone Joint Surg Br 2007;89(7):928-32.

3. Nagy MT, Macfarlane RJ, Khan Y, Waseem M. The frozen shoulder: myths and

realities. Open Orthop J 2013;7:352-5.

4. Tamai K, Akutsu M, Yano Y. Primary frozen shoulder: brief review of pathology and

imaging abnormalities. J Orthop Sci 2014;19(1):1-5.

5. Zuckerman JD, Rokito A. Frozen shoulder: a consensus definition. J Shoulder Elbow

Surg 2011;20(2):322-5.

6. Shaffer B, Tibone JE, Kerlan RK. Frozen shoulder. A long-term follow-up. J Bone Joint

Surg Am 1992;74(5):738-46.

7. Griggs SM, Ahn A, Green A. Idiopathic adhesive capsulitis. A prospective functional

outcome study of nonoperative treatment. J Bone Joint Surg Am 2000;82(10):1398-407.

8. Hand C, Clipsham K, Rees JL, Carr AJ. Long-term outcome of frozen shoulder. J

Shoulder Elbow Surg 2008;17(2):231-6.

9. Bulgen DY, Binder A, Hazleman BL, Park JR. Immunological studies in frozen

shoulder. J Rheumatol 1982;9(6):893-8.

10. Rizk TE, Pinals RS. Frozen shoulder. Semin Arthritis Rheum 1982;11(4):440-52.

11. Hannafin JA, Chiaia TA. Adhesive capsulitis. A treatment approach. Clin Orthop Relat

Res 2000(372):95-109.

12. White D, Choi H, Peloquin C, Zhu Y, Zhang Y. Secular trend of adhesive capsulitis.

Arthritis Care Res (Hoboken) 2011;63(11):1571-5.

 26

13. Brue S, Valentin A, Forssblad M, Werner S, Mikkelsen C, Cerulli G. Idiopathic

adhesive capsulitis of the shoulder: a review. Knee Surg Sports Traumatol Arthrosc

2007;15(8):1048-54.

14. Kelley MJ, Shaffer MA, Kuhn JE, Michener LA, Seitz AL, Uhl TL et al. Shoulder pain

and mobility deficits: adhesive capsulitis. J Orthop Sports Phys Ther 2013;43(5):A1-31.

15. Hettrich CM, DiCarlo EF, Faryniarz D, Vadasdi KB, Williams R, Hannafin JA. The

effect of myofibroblasts and corticosteroid injections in adhesive capsulitis. J Shoulder

Elbow Surg 2016;25(8):1274-9.

16. Prodromidis AD, Charalambous CP. Is There a Genetic Predisposition to Frozen

Shoulder?: A Systematic Review and Meta-Analysis. JBJS Rev 2016;4(2).

17. Sheridan MA, Hannafin JA. Upper extremity: emphasis on frozen shoulder. Orthop Clin

North Am 2006;37(4):531-9.

18. Milgrom C, Novack V, Weil Y, Jaber S, Radeva-Petrova DR, Finestone A. Risk factors

for idiopathic frozen shoulder. Isr Med Assoc J 2008;10(5):361-4.

19. Aydeniz A, Gursoy S, Guney E. Which musculoskeletal complications are most

frequently seen in type 2 diabetes mellitus? J Int Med Res 2008;36(3):505-11.

20. Zreik NH, Malik RA, Charalambous CP. Adhesive capsulitis of the shoulder and

diabetes: a meta-analysis of prevalence. Muscles Ligaments Tendons J 2016;6(1):26-34.

21. Neviaser AS, Hannafin JA. Adhesive capsulitis: a review of current treatment. Am J

Sports Med 2010;38(11):2346-56.

22. Hanchard NC, Goodchild L, Thompson J, O'Brien T, Davison D, Richardson C. A

questionnaire survey of UK physiotherapists on the diagnosis and management of

contracted (frozen) shoulder. Physiotherapy 2011;97(2):115-25.

 27

23. Maund E, Craig D, Suekarran S, Neilson A, Wright K, Brealey S et al. Management of

frozen shoulder: a systematic review and cost-effectiveness analysis. Health Technol

Assess 2012;16(11):1-264.

24. Georgiannos D, Markopoulos G, Devetzi E, Bisbinas I. Adhesive Capsulitis of the

Shoulder. Is there Consensus Regarding the Treatment? A Comprehensive Review.

Open Orthop J 2017;11:65-76.

25. Green S, Buchbinder R, Hetrick S. Physiotherapy interventions for shoulder pain.

Cochrane Database Syst Rev 2003(2):CD004258.

26. Page MJ, Green S, Kramer S, Johnston RV, McBain B, Chau M et al. Manual therapy

and exercise for adhesive capsulitis (frozen shoulder). Cochrane Database Syst Rev

2014(8):CD011275.

27. Hagen KB, Dagfinrud H, Moe RH, Osteras N, Kjeken I, Grotle M et al. Exercise

therapy for bone and muscle health: an overview of systematic reviews. BMC Med

2012;10:167.

28. Gebremariam L, Hay EM, van der Sande R, Rinkel WD, Koes BW, Huisstede BM.

Subacromial impingement syndrome--effectiveness of physiotherapy and manual

therapy. Br J Sports Med 2014;48(16):1202-8.

29. Pedersen BK, Saltin B. Exercise as medicine - evidence for prescribing exercise as

therapy in 26 different chronic diseases. Scand J Med Sci Sports 2015;25 Suppl 3:1-72.

30. O'Keeffe M, Hayes A, McCreesh K, Purtill H, O'Sullivan K. Are group-based and

individual physiotherapy exercise programmes equally effective for musculoskeletal

conditions? A systematic review and meta-analysis. Br J Sports Med 2017;51(2):126-

32.
 28

31. Hawk C, Minkalis AL, Khorsan R, Daniels CJ, Homack D, Gliedt JA et al. Systematic

Review of Nondrug, Nonsurgical Treatment of Shoulder Conditions. J Manipulative

Physiol Ther 2017;40(5):293-319.

32. Marik TL, Roll SC. Effectiveness of Occupational Therapy Interventions for

Musculoskeletal Shoulder Conditions: A Systematic Review. Am J Occup Ther

2017;71(1):7101180020p1-p11.

33. Hanchard NCA, Goodchild LM, Thompson J, O'Brien T, Richardson C, Davison D et

al. Evidence-based clinical guidelines for the diagnosis, assessment and physiotherapy

management of contracted (frozen) shoulder. 2011 2011.

34. Alsubheen SA, Nazari G, Bobos P, MacDermid JC, Overend TJ, Faber K. Effectiveness

of Nonsurgical Interventions for Managing Adhesive Capsulitis in Patients With

Diabetes: A Systematic Review. Arch Phys Med Rehabil 2019;100(2):350-65.

35. Ewald A. Adhesive capsulitis: a review. Am Fam Physician 2011;83(4):417-22.

36. Grubbs N. Frozen shoulder syndrome: a review of literature. J Orthop Sports Phys Ther

1993;18(3):479-87.

37. Hsu JE, Anakwenze OA, Warrender WJ, Abboud JA. Current review of adhesive

capsulitis. J Shoulder Elbow Surg 2011;20(3):502-14.

38. Jain TK, Sharma NK. The effectiveness of physiotherapeutic interventions in treatment

of frozen shoulder/adhesive capsulitis: a systematic review. J Back Musculoskelet

Rehabil 2014;27(3):247-73.

39. Page P, Labbe A. Adhesive capsulitis: use the evidence to integrate your interventions.

N Am J Sports Phys Ther 2010;5(4):266-73.

40. Zavala-Gonzalez J, Pavez-Baeza F, Gutierrez-Espinoza H, Olguin-Huerta C. The

effectiveness of joint mobilization techniques for range of motion in adult patients with
 29

primary adhesive capsulitis of the shoulder: a systematic review and meta-analysis.

Medwave 2018;18(5):e7265.

41. Yamshon LJ. Frozen shoulder: methods for bringing about early mobilization. Calif

Med 1958;89(5):333-4.

42. Trojian T, Stevenson JH, Agrawal N. What can we expect from nonoperative treatment

options for shoulder pain? J Fam Pract 2005;54(3):216-23.

43. Ramirez J. Adhesive Capsulitis: Diagnosis and Management. Am Fam Physician

2019;99(5):297-300.

44. Chan HBY, Pua PY, How CH. Physical therapy in the management of frozen shoulder.

Singapore Med J 2017;58(12):685-9.

45. Anton HA. Frozen shoulder. Can Fam Physician 1993;39:1773-8.

46. Brun SP. Idiopathic frozen shoulder. Aust J Gen Pract 2019;48(11):757-61.

47. Redler LH, Dennis ER. Treatment of Adhesive Capsulitis of the Shoulder. J Am Acad

Orthop Surg 2019;27(12):e544-e54.

48. Challoumas D, Biddle M, McLean M, Millar NL. Comparison of Treatments for Frozen

Shoulder: A Systematic Review and Meta-analysis. JAMA Netw Open

2020;3(12):e2029581.

49. Zhang J, Zhong S, Tan T, Li J, Liu S, Cheng R et al. Comparative Efficacy and Patient-

Specific Moderating Factors of Nonsurgical Treatment Strategies for Frozen Shoulder:

An Updated Systematic Review and Network Meta-analysis. Am J Sports Med

2020:363546520956293.

50. Nakandala P, Nanayakkara I, Wadugodapitiya S, Gawarammana I. The efficacy of

physiotherapy interventions in the treatment of adhesive capsulitis: A systematic

review. Journal of Back and Musculoskeletal Rehabilitation 2021;34(2):195-205.

 30

51. Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A. Rayyan-a web and mobile app

for systematic reviews. Syst Rev 2016;5(1):210.

52. Higgins JS, JAC; Savović, J; Page, MJ; Hróbjartsson, A; Boutron, I; Reeves, B;

Eldridge, S;. A revised tool for assessing risk of bias in randomized trial. Cochrane

Database of Systematic Reviews 2016(10).

53. Schunemann H, Brozek J, Guyatt G, Oxman A. Handbook for grading the quality of

evidence and the strength of recommendations using the GRADE approach. GRADE

Handbook. 2013.

54. Wan X, Wang W, Liu J, Tong T. Estimating the sample mean and standard deviation

from the sample size, median, range and/or interquartile range. BMC Med Res

Methodol 2014;14:135.

55. Higgins J, Green S, editors. Cochrane Handbook for Systematic Reviews of

Interventions Version 5.1.0. The Cochrane Collaboration; 2011.

56. Gutierrez Espinoza HJ, Pavez F, Guajardo C, Acosta M. Glenohumeral posterior

mobilization versus conventional physiotherapy for primary adhesive capsulitis: a

randomized clinical trial. Medwave 2015;15(8):e6267.

57. Horst R, Maicki T, Trabka R, Albrecht S, Schmidt K, Metel S et al. Activity- vs.

structural-oriented treatment approach for frozen shoulder: a randomized controlled

trial. Clin Rehabil 2017;31(5):686-95.

58. Mohamed AA, Jan YK, El Sayed WH, Wanis MEA, Yamany AA. Dynamic scapular

recognition exercise improves scapular upward rotation and shoulder pain and disability

in patients with adhesive capsulitis: a randomized controlled trial. J Man Manip Ther

2020;28(3):146-58.
 31

59. Atan T, Bahar-Ozdemir Y. Efficacy of high-intensity laser therapy in patients with

adhesive capsulitis: a sham-controlled randomized controlled trial. Lasers in Medical

Science 2021;36(1):207-17.

60. Leclaire R, Bourgouin J. Electromagnetic treatment of shoulder periarthritis: a

randomized controlled trial of the efficiency and tolerance of magnetotherapy. Arch

Phys Med Rehabil 1991;72(5):284-7.

61. Rawat P, Eapen C, Seema KP. Effect of rotator cuff strengthening as an adjunct to

standard care in subjects with adhesive capsulitis: A randomized controlled trial. J Hand

Ther 2017;30(3):235-41 e8.

62. Rizk TE, Christopher RP, Pinals RS, Higgins AC, Frix R. Adhesive capsulitis (frozen

shoulder): a new approach to its management. Arch Phys Med Rehabil 1983;64(1):29-

33.

63. Abd Elhamed HB, Koura GM, Hamada HA, Mohamed YE, Abbas R. Effect of

strengthening lower trapezius muscle on scapular tipping in patients with diabetic

frozen shoulder: a randomized controlled study. Biomedical research (india)

2018;29(3):442‐ 7.

64. Akbas E, Guneri S, Tas S, Erdem EU, Yuksel I. The Effects of Additional

Proprioceptive Neuromuscular Facilitation over Conventional Therapy in Patients with

Adhesive Capsulitis. Turkish Journal of Physiotherapy Rehabilitation-Fizyoterapi

Rehabilitasyon 2015;26(2):78-85.

65. Ali SA, Khan M. Comparison for efficacy of general exercises with and without

mobilization therapy for the management of adhesive capsulitis of shoulder - An

interventional study. Pak J Med Sci 2015;31(6):1372-6.

 32

66. Balci NC, Yuruk ZO, Zeybek A, Gulsen M, Tekindal MA. Acute effect of scapular

proprioceptive neuromuscular facilitation (PNF) techniques and classic exercises in

adhesive capsulitis: a randomized controlled trial. J Phys Ther Sci 2016;28(4):1219-27.

67. Baskaya MC, Ercalik C, Karatas Kir O, Ercalik T, Tuncer T. The efficacy of mirror

therapy in patients with adhesive capsulitis: A randomized, prospective, controlled

study. J Back Musculoskelet Rehabil 2018;31(6):1177-82.

68. Binder A, Hazleman BL, Parr G, Roberts S. A controlled study of oral prednisolone in

frozen shoulder. Br J Rheumatol 1986;25(3):288-92.

69. Celik D. Comparison of the outcomes of two different exercise programs on frozen

shoulder. Acta Orthop Traumatol Turc 2010;44(4):285-92.

70. Dundar U, Toktas H, Cakir T, Evcik D, Kavuncu V. Continuous passive motion

provides good pain control in patients with adhesive capsulitis. Int J Rehabil Res

2009;32(3):193-8.

71. Ekim AA, Inal EE, Gonullu E, Hamarat H, Yorulmaz G, Mumcu G et al. Continuous

passive motion in adhesive capsulitis patients with diabetes mellitus: A randomized

controlled trial. J Back Musculoskelet Rehabil 2016;29(4):779-86.

72. Elhafez HM, Elhafez SM. Axillary Ultrasound and Laser Combined With Postisometric

Facilitation in Treatment of Shoulder Adhesive Capsulitis: A Randomized Clinical

Trial. J Manipulative Physiol Ther 2016;39(5):330-8.

73. Hussein AZ, Ibrahim MI, Hellman MA, Donatelli R. Static progressive stretch is

effective in treating shoulder adhesive capsulitis: Prospective, randomized, controlled

study with a two-year follow-up. European Journal of Physiotherapy 2015;17(3):138-

47.

74. Junaid M, Burq SIA, Rafique S, Malik s, Rasool A, Mubeen I et al. A comparative

study to determine the efficacy of routine physical therapy treatment with and without
 33

Kaltenborn mobilization on pain and shoulder mobility in frozen shoulder patients.

International Journal of Physiotherapy 2016;3(3).

75. Kalita A, Milton A. The Combined Effectiveness of Glenohumeral End-Range

Mobilization and Contract-Relax Technique for Glenohumeral Internal Rotators in

Subjects with Adhesive Capsulitis. International Journal of Physiotherapy

2015;2(5):691-7.

76. Kumar G, Sudhakar S, Sudhan S, Jyothi N. Subscapularis muscle spray and stretch

technique with conventional physical therapy for the management of adhesive

capsulitis. Biomedicine (india) 2017;37(4):511‐ 7.

77. Lokesh M, Raja R, Prashantha S, Rajeev A. Comparison of effectiveness of the

combination of muscle energy techniques and conventional physiotherapy alone in

periarthritis of shoulder: a randomized study. Journal of Evolution of Medical and

Dental Sciences-Jemds 2015;4(4):545-54.

78. Muhammed AA, Shanmugam S, Kumar D. Is position induced movement re-education

helpful on early functional recovery in acute adhesive capsulitis? A randomised

controlled trial. Journal of clinical and diagnostic research 2018;12(1):YC08‐ YC13.

79. Pajareya K, Chadchavalpanichaya N, Painmanakit S, Kaidwan C, Puttaruksa P,

Wongsaranuchit Y. Effectiveness of physical therapy for patients with adhesive

capsulitis: a randomized controlled trial. J Med Assoc Thai 2004;87(5):473-80.

80. Sule K, Rathi M, Palekar T, Anwer S. Comparison of conventional therapy versus

sleeper stretch with conventional therapy in Adhesive Capsulitis. International Journal

of Health Sciences and Research 2015;5(11):7.

81. Yang JL, Jan MH, Chang CW, Lin JJ. Effectiveness of the end-range mobilization and

scapular mobilization approach in a subgroup of subjects with frozen shoulder

syndrome: a randomized control trial. Man Ther 2012;17(1):47-52.

 34

82. Russell S, Jariwala A, Conlon R, Selfe J, Richards J, Walton M. A blinded, randomized,

controlled trial assessing conservative management strategies for frozen shoulder. J

Shoulder Elbow Surg 2014;23(4):500-7.

83. Aggarwal A, Saxena K, Palekar TJ, Rathi M. Instrument assisted soft tissue

mobilization in adhesive capsulitis: A randomized clinical trial. Journal of Bodywork

and Movement Therapies 2021;26:435-42.

84. Contractor E, Agnihotri D, Patel R. Effect of Spencer Muscle Energy Technique on

pain and functional disability in cases of adhesive capsulitis of shoulder joint.

International Archives of Integrated Medicine 2016;3(8):6.

85. Diercks RL, Stevens M. Gentle thawing of the frozen shoulder: a prospective study of

supervised neglect versus intensive physical therapy in seventy-seven patients with

frozen shoulder syndrome followed up for two years. J Shoulder Elbow Surg

2004;13(5):499-502.

86. Jain M, Tripathy PR, Manik R, Tripathy S, Behera B, Barman A. Short term effect of

yoga asana - An adjunct therapy to conventional treatment in frozen shoulder. J

Ayurveda Integr Med 2020;11(2):101-5.

87. Nellutla M, Giri P. Comparative Study between Efficacy of PNF Movement Patterns

Versus Conventional Free Exercises on Functional Activities Among Patiens With

Chronic Peri-Arthritis of Shoulder. Indian Journal of Physiotherapy and Occupational

Therapy 2011;5(3):62-7.

88. Shen ZF, Zhu GF, Shen QH, Wu YJ, Xu J. Effect of Yi Jin Jing (Sinew-transforming

Qigong Exercises) plus tuina on scapulohumeral periarthritis. Journal of Acupuncture

and Tuina Science 2017;15(4):285-9.

 35

89. Anjum R, Aggarwal J, Gautam R, Pathak S, Sharma A. Evaluating the Outcome of Two

Different Regimes in Adhesive Capsulitis: A Prospective Clinical Study. Med Princ

Pract 2020;29(3):225-30.

90. Fregni F, Imamura M, Chien HF, Lew HL, Boggio P, Kaptchuk TJ et al. Challenges and

recommendations for placebo controls in randomized trials in physical and

rehabilitation medicine: a report of the international placebo symposium working group.

Am J Phys Med Rehabil 2010;89(2):160-72.

91. Wong CK, Levine WN, Deo K, Kesting RS, Mercer EA, Schram GA et al. Natural

history of frozen shoulder: fact or fiction? A systematic review. Physiotherapy

2017;103(1):40-7.

92. Vastamaki H, Ristolainen L, Vastamaki M. Range of motion of diabetic frozen shoulder

recovers to the contralateral level. J Int Med Res 2016;44(6):1191-9.

93. Vermeulen HM, Rozing PM, Obermann WR, le Cessie S, Vliet Vlieland TP.

Comparison of high-grade and low-grade mobilization techniques in the management of

adhesive capsulitis of the shoulder: randomized controlled trial. Phys Ther

2006;86(3):355-68.

94. Ando A, Sugaya H, Hagiwara Y, Takahashi N, Watanabe T, Kanazawa K et al.

Identification of prognostic factors for the nonoperative treatment of stiff shoulder. Int

Orthop 2013;37(5):859-64.

95. Rill BK, Fleckenstein CM, Levy MS, Nagesh V, Hasan SS. Predictors of outcome after

nonoperative and operative treatment of adhesive capsulitis. Am J Sports Med

2011;39(3):567-74.

96. Vrotsou K, Avila M, Machon M, Mateo-Abad M, Pardo Y, Garin O et al. Constant-

Murley Score: systematic review and standardized evaluation in different shoulder

pathologies. Qual Life Res 2018;27(9):2217-26.

 36

97. Roy JS, MacDermid JC, Woodhouse LJ. A systematic review of the psychometric

properties of the Constant-Murley score. J Shoulder Elbow Surg 2010;19(1):157-64.

98. Hirschmann MT, Wind B, Amsler F, Gross T. Reliability of shoulder abduction strength

measure for the Constant-Murley score. Clin Orthop Relat Res 2010;468(6):1565-71.

99. Othman A, Taylor G. Is the constant score reliable in assessing patients with frozen

shoulder? 60 shoulders scored 3 years after manipulation under anaesthesia. Acta

Orthop Scand 2004;75(1):114-6.

100. Sharma SP, Baerheim A, Kvale A. Passive range of motion in patients with adhesive

shoulder capsulitis, an intertester reliability study over eight weeks. BMC

Musculoskelet Disord 2015;16:37.

101. Roy JS, MacDermid JC, Woodhouse LJ. Measuring shoulder function: a systematic

review of four questionnaires. Arthritis Rheum 2009;61(5):623-32.

102. Kelly AM. The minimum clinically significant difference in visual analogue scale pain

score does not differ with severity of pain. Emerg Med J 2001;18(3):205-7.
 37

LEGENDS

Table 1: Full search strategy for the different elements of the PICO for PubMed. Different

elements were combined with AND.

Table 2: Pooled quality of evidence, based on the GRADE, for the different comparisons and

each outcome measure.

Table 3: Overview of results for various treatment programs incorporating exercise therapy,

with the mean difference [95% confidence interval] and the effect size.

Table 4: Characteristics of studies comparing solely exercises in different formats

Table 5: Characteristics of studies comparing physical therapy programs including exercises

with solely exercises

Table 6: Characteristics of studies comparing a physical therapy program including exercises

with a program without exercises

Table 7: Characteristics of studies comparing 2 physical therapy programs both including

exercises

Box 1: Inclusion criteria

 38

Figure 1. Flowchart of study selection

 39

Figure 2. Overview of within and between studies risk of bias for both clinician and patient

reported outcome measures.

 40

Figure 3. Pooled results of PT program incl. exercise compared to solely exercises for

PROM.

Figure 4. Pooled results of PT program incl. exercise compared to solely exercises for

function.
 41

Figure 5. Pooled results of PT program incl. exercise compared to solely exercises for pain

(VAS).

Figure 6. Pooled results of PT program incl. exercise compared to a PT program without

exercise for PROM.

 42

Figure 7. Pooled results of PT program incl. exercise compared to a PT program without

exercise for AROM.

Figure 8. Pooled results of PT program incl. exercise compared to a PT program without

exercise for function.

 43

Figure 9. Pooled results of PT program incl. exercise compared to a PT program without

exercise for pain.

 44

Figure 10. Pooled results of MET in a PT program compared to a PT program with different

exercises for PROM.

 45

Figure 11. Pooled results of MET in a PT program compared to a PT program with different

exercises for AROM.

 46

Figure 12. Pooled results of MET in a PT program compared to a PT program with different

exercises for function.

Figure 13. Pooled results of MET in a PT program compared to a PT program with different

exercises for pain.

 47

Figure 14. Pooled results of stretching added to a PT program including exercises compared

to the same PT program for PROM.

 48

Figure 15. Pooled results of stretching added to a PT program including exercises compared

to the same PT program for function.

 49

Figure 16. Pooled results of physical modalities compared to sham treatment added to a PT

program including exercises for PROM.

 50

Figure 17. Pooled results of physical modalities compared to sham treatment added to a PT

program including exercises for pain.

 51

Table 1: Full search strategy for the different elements of the PICO for PubMed. Different elements were combined with AND.

Patient Intervention Comparison Outcome

Frozen shoulder OR “Rehabilitation” [MeSH] OR “Range of motion, articular” [MeSH] OR
Adhesive capsulitis OR “Exercise Therapy” [MeSH] OR “Pain” [MeSH] OR
Stiff shoulder OR "Exercise Movement Techniques"[Mesh] OR “Musculoskeletal Pain” [MeSH] OR
“Periarthritis” [MeSH] OR "Resistance Training"[Mesh] OR “chronic pain” [MeSH] OR
Periarthritis OR "Plyometric Exercise"[Mesh] OR “Shoulder Pain” [MeSH] OR
Pericapsulitis “High-Intensity Interval Training” [MeSH] OR "Muscle Strength"[Mesh] OR
“Physical Therapy Modalities” [MeSH] OR “activities of daily living” [MeSH] OR
“Physical Therapy Specialty” [MeSH] OR “Sports” [MeSH] OR
Exercise therapy OR “Quality of life” [MeSH] OR
Exercise training OR "Patient Satisfaction"[Mesh] OR
Exercise movement techniques OR Pain OR
Muscle strengthening exercises OR Shoulder pain OR
Resistance training OR Mobility OR
Resistance exercise OR Range of motion OR
Plyometric training OR Muscle strength OR
Plyometric exercise OR Functionality OR
Proprioceptive training OR Functional ability OR
strength training OR Activities of daily living OR
rehabilitation OR Sports OR
aerobic exercise OR Quality of life OR
anaerobic exercise OR Patient satisfaction
high-intensity interval training OR
anaerobic training OR
aerobic training OR
physical therapy
 52

Table 2: Pooled quality of evidence, based on the GRADE, for the different comparisons and each outcome measure.

Outcome Result Weighted (S)MD [95%CI] Evidence

Multimodal program including exercises compared to solely exercises
PROM (°) No difference -4.91 [-6.76, -3.06] Low
Function (various) No difference 0.04 [-0.56, 0.64] Very low
Pain (VAS) No difference -1.13 [-2.61, 0.35] Low
MM program including exercises compared to MM program without exercises
PROM (°) No difference 4.51 [2.10, 6.91] High
AROM (°) MM including exercises more effective 12.83 [6.00, 19.66] Preliminary
Function (various) No difference -0.78 [-2.06, 0.49] Very low
Pain (various) No difference -0.06 [-0.42, 0.30] Moderate
MM program including MET compared to MM program including other exercises
PROM (°) No difference 4.88 [3.24, 6.51] Moderate
AROM (°) No difference 6.35 [-8.93, 21.63] Low
Function (various) MET more effective -0.62 [-1.28, 0.04] Low
Pain (various) No difference -0.36 [-1.24, 0.52] Very low
MM program including static stretching compared to MM program without stretching
PROM (°) Static stretching more effective 16.40 [7.41, 25.38] Very low
Function (various) No difference -0.60 [-2.92, 1.72] Very low
MM program including physical modalities compared to MM program including sham treatment
PROM (°) No difference 1.51 [-4.14, 7.16] Moderate
Pain (VAS) No difference 0.10 [-0.26, 0.46] High
MM: multimodal; PROM: passive range of motion; AROM: active range of motion; MET: muscle energy techniques;
 53

Table 3: Overview of results for various treatment programs incorporating exercise therapy, with the mean difference [95% confidence interval]

and the effect size.

Study Intervention MD between groups Effect size

[95% CI]
PROM abduction (°)
Aggarwal et al., 202183 Addition of IASTM -6.60 [-13.42, 0.22] No effect
Baskaya et al., 2018 67 Mirror therapy compared to 19.10 [5.47, 32.37] Moderate
no mirror
Ekim et al., 2016 71 CPM compared to additional 11.00 [0.77, 21.23] Small
stretching
56
Gutierrez-Espinoza et al., 2015 Local exercises with US 21.90 [17.65, 26.15] Moderate
compared to aerobic with
mobilization
Mohamed et al., 20205860 Scapular recognition 2.29 [-1.63, 6.21] No effect
exercise compared to
placebo exercise
Rawat et al., 2017 61 Addition of RC 17.72 [8.36, 27.08] Moderate
strengthening exercises
PROM external rotation (°)
Aggarwal et al., 202183 Addition of IASTM 1.40 [-6.18, 8.98] No effect
Baskaya et al., 2018 67 Mirror therapy compared to 3.10 [-5.82, 12.02] No effect
no mirror
Celik, 2010 69 Addition of scapulothoracic 2.50 [-4.47, 9.47] No effect
exercises
71
Ekim et al., 2016 CPM compared to additional 3.60 [-6.42, 13.62] No effect
stretching
56
Gutierrez-Espinoza et al., 2015 Local exercises with US 26.80 [22.75, 30.85] Moderate
compared to aerobic with
mobilization
Kumar et al., 2017 76 Addition of spray & stretch 19.00 [15.76, 22.24] Moderate
Mohamed et al., 20205860 Scapular recognition 9.16 [4.58, 13.74] Small
exercise compared to
placebo exercise
Rawat et al., 2017 61 Addition of RC 26.05 [18.34, 33.76] Moderate
 54

strengthening exercises
Yang et al., 2012 81 (no values per Addition of end range 23.4 [8.2, 37.3] Moderate
intervention specified) mobilization
PROM internal rotation (°)
Aggarwal et al., 202183 Addition of IASTM -1.40 [-8.04, 5.24] No effect
Baskaya et al., 2018 67 Mirror therapy compared to 3.40 [-6.00, 12.80] No effect
no mirror
Celik, 2010 69 Addition of scapulothoracic 0.00 [-4.72, 4.72] No effect
exercises
71
Ekim et al., 2016 CPM compared to additional 8.90 [-0.05, 17.85] Small
stretching
Rawat et al., 2017 61 Addition of RC 18.43 [13.33, 23.53] Moderate
strengthening exercises
Yang et al., 2012 81 Addition of end range -0.03 [-0.11, 0.05] No effect
mobilization
PROM flexion (°)
Aggarwal et al., 202183 Addition of IASTM 6.20 [-4.59, 16.99] No effect
Baskaya et al., 2018 67 Mirror therapy compared to 22.00 [9.63, 34.37] Moderate
no mirror
Celik, 2010 69 Addition of scapulothoracic 12.21 [4.39, 20.03] Small
exercises
71
Ekim et al., 2016 CPM compared to additional 11.50 [4.33, 18.67] Small
stretching
Gutierrez-Espinoza et al., 2015 56 Local exercises with US 37.30 [28.73, 45.87] Large
compared to aerobic with
mobilization
Mohamed et al., 20205860 Scapular recognition 10.60 [5.46, 15.74] Small
exercise compared to
placebo exercise
Rawat et al., 2017 61 Addition of RC 7.05 [-5.32, 19.42] Small
strengthening exercises
AROM abduction (°)
Aggarwal et al., 202183 Addition of IASTM -4.90 [-19.42, 9.62] No effect
Baskaya et al., 2018 67 Mirror therapy compared to 21.70 [6.75, 36.65] Moderate
no mirror
Ekim et al., 2016 71 CPM compared to additional 11.90 [2.47, 21.33] Small
 55

stretching
AROM external rotation (°)
Aggarwal et al., 202183 Addition of IASTM 2.00 [-5.48, 9.48] No effect
Baskaya et al., 2018 67 Mirror therapy compared to 4.30 [-4.33, 12.93] No effect
no mirror
Ekim et al., 2016 71 CPM compared to additional 2.50 [-7.49, 12.49] No effect
stretching
AROM internal rotation (°)
Aggarwal et al., 202183 Addition of IASTM -0.13 [-7.20, 6.94] No effect
Baskaya et al., 2018 67 Mirror therapy compared to 7.10 [-2.67, 16.87] Small
no mirror
Ekim et al., 2016 71 CPM compared to additional 7.70 [-1.90, 17.30] Small
stretching
AROM flexion (°)
Aggarwal et al., 202183 Addition of IASTM 5.20 [-5.64, 16.04] No effect
Baskaya et al., 2018 67 Mirror therapy compared to 24.10 [11.60, 36.60] Moderate
no mirror
Ekim et al., 2016 71 CPM compared to additional 11.60 [4.25, 18.95] Small
stretching
Functional ROM (apley’s scratch test, overall)
Aggarwal et al., 202183 Addition of IASTM -0.02 [-1.61, 1.58] No effect
Function (diverse)
Celik, 2010 69 (modified CMS) Addition of scapulothoracic 9.00 [2.77, 15.23] Small
exercises
Baskaya et al., 2018 67 (UCLA) Mirror therapy compared to 6.00 [2.48, 9.52] Moderate
no mirror
71
Ekim et al., 2016 (CMS) CPM compared to additional 7.40 [3.08, 11.72] Small
stretching
56
Gutierrez-Espinoza et al., 2015 Local exercises with US 20.60 [16.82, 24.38] Moderate
(CMS) compared to aerobic with
mobilization
Kumar et al., 2017 76 (SPADI) Addition of spray & stretch -21.00 [-26.21, -15.79] Moderate
Mohamed et al., 202058 Scapular recognition -8.84 [-3.27,-14.41] No effect
exercise compared to
placebo exercise
Rawat et al., 2017 61 (SPADI) Addition of RC -19.62 [-25.56, -13.68] Moderate
 56

strengthening exercises
Shen et al., 2017 88 (CMS) Yi jin jing compared to 3.20 [0.96, 5.44] No effect
functional
Yang et al., 2012 81 (FLEX-SF) Addition of end range 0.74 [-0.17, 1.66] No effect
mobilization
Pain (VAS, unless indicated otherwise )
Baskaya et al., 2018 67 Mirror therapy compared to -1.48 [-2.34, -0.62] Small
no mirror
69
Celik, 2010 Addition of scapulothoracic -1 (-1.59, -0.41] Small
exercises
71
Ekim et al., 2016 CPM compared to additional -1.10 [-1.90, -0.30] Small
stretching
Gutierrez-Espinoza et al., 2015 56 Local exercises with US -1.00 [-1.50, -0.50] Small
compared to aerobic with
mobilization
Junaid et al., 2016 74 Addition of mobilization -0.75 [-1.24, -0.26] No effect
Kumar et al., 2017 76 Addition of spray & stretch -2.00 [-2.72, -1.28] Moderate
60
Leclaire & Bourgouin, 1991 Addition of electromagnetic 0.10 [-0.26, 0.46] No effect
therapy (ordinal scale)
61
Rawat et al., 2017 Addition of RC -1.29 [-2.01, -0.57] Small
strengthening exercises
88
Shen et al., 2017 Yi jin jing compared to -1.80 [-2.46, -1.14] Small
functional
Muscle strength
Kumar et al., 2017 76 Addition of spray & stretch 32.00 [26.23, 37.77] Moderate
mmHg
Rawat et al., 2017 61 (multiple Addition of RC 2.10 [1.67, 2.52] lb. Small
directions) strengthening exercises
Scapular tipping (cm)
Abd Elhamed et al., 2018 63 Addition of lower trapezius -3.09 [-4.33, -1.85] Small
strengthening
Scapular upward rotation (°)
Mohamed et al., 202058 Scapular recognition 2.43 [-1.50, 6.36] No effect
exercise compared to
placebo exercise
MD: mean difference; CI: confidence interval; PROM: passive range of motion; CPM: continuous passive motion; US:
 57

ultrasound; RC: rotator cuff; AROM: active range of motion; VAS: visual analogue scale; cm: centimeter.

Table 4: Characteristics of studies comparing solely exercises in different formats

Source & Participants Experimental Control Dose Follow-up Outcome measures Results
origin Group Inclusion Exclusion intervention intervention
composition
and patient
characteristics
Russell et Frozen shoulder  Insidious onset of  Other shoulder HEP HEP F: 2x/w 6 weeks, 6 Function (CMS) All FU: E↑, C↑, E>C
l., 2014 82 pain & stiffness disorders, surgery or + D: 6 w months,
75 T (51.1 (40-  Clinically significant trauma exercise class I: 50 min and 1 year Oxford shoulder score All FU: E↑, C↑, E>C,
United 65) reduction in  Local CSI or any PT (class)
Kingdom ROM, >50% ER intervention within last HEP ROM (°)
25 E  No radiologic 3 months continued - flexion All FU: E↑, C↑, E>C,
? abnormalities  Bilateral frozen after 6 - ER All FU: E↑, C↑, E>C,
26 C  At least 3 months shoulder weeks
? complaints  Presence of
comorbidities
 Active medicolegal
involvement
T: total group study; E: experimental group; C: control group; ROM: range of motion; ER: external rotation; CSI: corticosteroid injection; PT: physical therapy; HEP: home exercise program; F:
requency; w: week; D: duration; I: intensity; CMS: Constant Murley Score; FU: follow up;

↑: improved
=: not improved
>: improved more than
<: improved less than
 58

Table 5: Characteristics of studies comparing physical therapy programs including exercises with solely exercises

Source & Participants Experimental Control Dose Follow-up Outcome measures Results
origin Group Inclusion Exclusion intervention intervention
composition and
patient
characteristics
Ali & Adhesive  One sided  Additional shoulder General General F: 3x/w After 5 Pain (VAS) E↑, C↑, E=C
Khan, 2015 capsulitis shoulder or cervical pathology exercises exercises D: 5 w weeks
5
involvement  Presence of + I: 45 min treatment ROM (°)
43 T  Complaints of comorbidities Manual therapy (pre-post) - abduction E↑, C↑, E=C
Pakistan pain & shoulder  Severe trauma of (Maitland - ER E↑, C↑, E=C
22 E (51.31) ROM restriction fracture mobilization) - IR E↑, C↑, E=C
11 ♀ (50%) for more than 3  Pregnancy
11 ♂ (50%) months Function (SPADI) E↑, C↑, E=C

21 C (51.71)
unknown
Atan et al., Adhesive  18 to 65 years  History of bilateral E1: Therapeutic Therapeutic F: 5x/w After 3 Pain (VAS) All FU: E1↑, E2=, C=,
2021 59 capsulitis  Passive ER concurrent adhesive exercises exercises D: 3 w weeks E1>E2, E1>C, E2=C
restriction <50% capsulitis, shoulder + I: 25 min. treatment
Turkey 31 T of contralateral trauma, fracture, High intensity exercises, 15 and at 12 Pain and disability All FU: E1↑, E2↑, C↑
22 ♀ (71.0%) shoulder shoulder surgery, laser therapy min. week (SPADI) 3 w: E1=E2=C
9 ♂ (29.0%)  Normal calcific tendinopathy, laser/sham follow-up 12 w: E1>E2, E1>C
radiograph GH OA, E2: Therapeutic
11 E1 finding of the inflammatory exercises Quality of life (SF-
(56.00±11.63) affected shoulder rheumatic diseases, + 36)
7 ♀ (63.6%)  Complaints of tumor and infection sham laser - PF All FU: E1↑, E2=, C=,
4 ♂ (36.4%) shoulder  History of CSI in the E1=E2=C
restriction with shoulder during last 3 - RLPH All FU: E1↑, E2=, C=,
10 E2 severe pain for at months E1=E2=C
(60.80±8.32) least 1 month  History of recent - RLE All FU: E1↑, E2=, C=,
8 ♀ (80%)  Literate and lung, breast, or E1=E2=C
 59

2 ♂ (20%) ability to bypass - EF All FU: E1↑, E2=, C=,

comprehend surgery/radiotherapy E1>E2, E1>C, E2=C
10 C verbal  History of cervical - EWB All FU: E1↑, E2=, C=,
(58.50±7.29) instructions in radiculopathy/brachia E1=E2=C
7 ♀ (70%) our language l plexus lesion - SF All FU: E1↑, E2=, C=,
3 ♂ (30%)  History of E1=E2=C
neuromuscular -P All FU: E1↑, E2=, C=,
disease E1>E2, E1>C, E2=C
 History of physical - GH All FU: E1↑, E2=, C=,
therapy program for E2=C
the same shoulder 3 w: E1=E2=C
last 6 months 12 w: E1>E2, E1>C
- HC All FU: E1↑, E2=, C=,
E1=E2=C

AROM (°) All FU: E1↑, E2↑, C↑,

- flexion E1=E2=C
- abduction
- ER
- IR

PROM (°) All FU: E1↑, E2↑, C↑,

- flexion E1=E2=C
- abduction
- ER
- IR

Binder et 40 patients with  Painful stiff  Generalized arthritis, HEP HEP HEP Fortnightly Pain (VAS)
l., 1986 68 frozen shoulder shoulder at least sensory symptoms or + F: every for 6 - night 2, 4, 6 w, 3, 4 m: E>C
1 month signs in the arm or Oral steroid hour weeks, 5, 6, 7, 8 m: E=C
United 40 T (54.8 (45-  Pain with sleep radiation of pain to (prednisolone) D: 6 w monthly - movement All FU: E=C
Kingdom 76)) disturbance the neck I: 2-3 min for a - rest All FU: E=C
24 ♀ (60%)  Restricted  Peptic ulceration, further 6
16 ♂ (40%) AROM and serious infection or Steroid months ROM (°)
PROM with ER contra-indications to F: daily - total flexion All FU: E=C
20 E at least 50% systemic steroid D: 6 w - GH flexion All FU: E=C
? therapy I: 10 mg (4 - total abduction All FU: E=C
 60

w), 5 mg (2 - GH abduction All FU: E=C

20 C w) - ER All FU: E=C
?
Diercks & Idiopathic frozen  >50% restriction  Significant injury to Standardized PT Exercises only F: - 24 months CMS All FU: E<C
Stevens, shoulder GHJ in all ipsilateral shoulder or (including D: 2 y with 3
2004 85 syndrome directions for 3 arm exercises) I: - month
months or more  Surgical procedures intervals
The 77 T on the shoulder, arm,
Netherlands cervical spine, thorax
32 E (51±7) or breast within
21 ♀ (65.6%) previous 2 years
11 ♂ (34.4%)  Intra-articular
deformities,
45 C (50±6) degenerative or
26 ♀ (57.7%) inflammatory arthritis
19 ♂ (42.3%)  DM
Dundar et Primary frozen  Gradually  Other shoulder HEP HEP HEP After 4- Pain (VAS)
l., 2009 70 shoulder phase 1 increasing disorders or + + F: daily and 12- - rest Both FU: E↑, C↑, E>C
and/or 2 shoulder pain significant trauma CPM CPT (exercises) D: 12 w weeks - movement Both FU: E↑, C↑, E>C
Turkey and stiffness  Secondary frozen I: - treatment - night Both FU: E↑, C↑, E>C
57 T shoulder
CPM/CPT PROM (°)
29 E (56.3±7.8) F: 5x/w - flexion Both FU: E↑, C↑, E=C
20 ♀ (69.0%) D: 4 w - abduction Both FU: E↑, C↑, E=C
9 ♂ (31.0%) I: 1 h/day - IR Both FU: E↑, C↑, E=C
- ER Both FU: E↑, C↑, E=C
28 C (57.1±8.3)
19 ♀ (67.9%) Function
9 ♂ (32.1%) - CMS Both FU: E↑, C↑, E=C
- SPADI
● pain Both FU: E↑, C↑, E>C
● disability Both FU: E↑, C↑, E=C
Horst et al., Frozen shoulder  Limited range of  Additional symptoms Structural Activity F: 5x/w After 2 McGill pain 2 w: E=C
2017 57 motion of dizziness and a oriented (MT & oriented D: 2 w weeks questionnaire 3 m: E<C
66 T  Pain in the case history of PNF exercises) (exercise only) I: 30 min treatment
Poland 25 ♀ (37.9%) shoulder region headaches and 3 Function 2 w: E=C (3/5); E<C
41 ♂ (62.1%)  Prescription for  Pain and/or limited months (MUEMAL) (2/5)
 61

PT by orthopedic ROM in the cervical 3 m: E=C (2/5); E<C

33 E (44±16) surgeon spine and/or (3/5)
13 ♀ (31%) temporomandibular ROM (°)
20 ♂ (61%) joint - flexion
- extension Both FU: E=C
33 C (47±17) - IR Both FU: E=C
12 ♀ (36%) - ER Both FU: E<C
21 ♂ (64%) - abduction Both FU: E<C
- adduction Both FU: E=C
Both FU: E<C
Strength
- flexion
- extension Both FU: E<C
- IR Both FU: E=C
- ER Both FU: E<C
2 w: E=C
- abduction 3 m: E<C
- adduction Both FU: E<C
Both FU: E<C
Kalita & Frozen shoulder  Unilateral  DM Pendulum Pendulum F: 2x/w After 4 Pain (VAS) -
Milton, involvement  History of surgery on exercises exercises D: 4 w weeks
2015 75 60 T  Painful stiff particular shoulder + I: treatment Function (SPADI)
shoulder for at  Other shoulder GH end-range (pre-post) - pain E>C
ndia 30 E least 3 months disorders or surgery mobilization - disability E>C
 Restriction>50% and contract - total E>C
30 C passive ER & relax technique
restricted for GH IR AROM (°)
overhead reach - ER E>C

PROM (°)
- ER E>C
Russell et Frozen shoulder  Insidious onset  Other shoulder HEP C1: HEP F: 2x/w 6 weeks, 6 Function (CMS) All FU: E↑, C1↑, C2↑,
l., 2014 82 of pain & disorders, surgery or + + D: 6 w months, E<C1, E>C2, C1>C2
75 T (51.1 (40- stiffness significant trauma Individual Exercise class I: 50 min and 1 year
United 65)  Clinically  Local CSI or any PT multimodal PT (class) Oxford shoulder All FU: E↑, C1↑, C2↑,
Kingdom reduction in intervention within C2: HEP score E<C1, E>C2, C1>C2
25 E1 ROM, >50% ER last 3 months HEP
 62

?  No radiologic  Bilateral frozen continued

24 E2 abnormalities shoulder after 6 ROM (°)
?  At least 3 months  Presence of weeks - flexion All FU: E↑, C1↑, C2↑,
26 C complaint comorbidities E=C1, E>C2, C1>C2,
?  Active medicolegal - ER All FU: E↑, C1↑, C2↑,
involvement E=C1, E>C2, C1>C2,
T: total group study; E: experimental group; C: control group; ♀: female; ♂: male; ROM: range of motion; F: frequency; w: week; D: duration; I: intensity; VAS: Visual Analogue Scale; ER: external
otation; IR: internal rotation; SPADI: Shoulder Pain and Disability Index; SF-36: 36-item short form health survey; PF: physical functioning; RLPH: role limitations due to physical health; RLE: role
imitations due to emotional problems; EF: energy/fatigue; EWB: emotional well-being; SF: social functioning; P: pain; GH: general health; HC: health change; AROM: active range of motion; PROM:
passive range of motion; HEP: home exercise program; min: minutes; GH: glenohumeral; GHJ: glenohumeral joint; DM: Diabetes Mellitus; PT: physical therapy; CMS: Constant Murley Score; CPM:
ontinuous passive motion; CPT: conventional physical therapy; MT: manual therapy; PNF: Proprioceptive Neuromuscular Facilitation; MUEMAL: Modified Upper Extremity Motor Activity Log; CSI:
orticosteroid injection;

↑: improved
=: not improved
>: improved more than
<: improved less than
 63

Table 6: Characteristics of studies comparing a physical therapy program including exercises with a program without exercises

Source & Participants Experimental Control Dose Follow-up Outcome measures Results
origin Group Inclusion Exclusion intervention intervention
composition and
patient
characteristics
Balci et al., Unilateral  Pain in the  History of surgery or E1: PT PT modalities F:once After 1 Pain (VAS) E1↑, E2=, C↑, E1=E2=C
2016 66 adhesive shoulder for at MUA modalities D: once session
capsulitis stage II least 3 months  Pain or disorders of + I: 1 h Scapular dyskinesis E1=, E2=, C=, E1=E2=C
Turkey the cervical spine, PNF exercises (LSST)
53 T elbow, wrist or hand
40 ♀ (75.5%)  Other pathological E2:PT AROM (°)
13 ♂ (24.5%) conditions (including modalities - flexion E1↑, E2↑, C↑, E1=E2=C
neurologic) involving + - abduction E1↑, E2↑, C↑, E1=E2=C
18 E1 (56.7±7.7) the shoulder Classic
14 ♀ (77.8%) exercises Function (SST) E1↑, E2↑, C↑, E1=E2=C
4 ♂ (22.2%)

18 E2 (58.1±8.4)
15 ♀ (83.3%)
3 ♂ (16.7%)

17 C (58.6±11.3)
11 ♀ (64.7%)
6 ♂ (35.3%)
ain et al., Frozen shoulder  Pain &  Prior history of trauma Standard care Standard care F: daily After 1, 2, Pain & Disability
2020 86 limitation in or arthritis + D: 4 w and 4 (SPADI)
72 T both active and  Bilateral involvement Supervised SGA I: 30 min weeks - pain FFU: E↑, C↑, E=C
41 ♀ (56.9%) passive  Major psychiatric (yoga) (yoga) treatment - disability FFU: E↑, C↑, E=C
31 ♂ (43.1%) movements of problems - total FFU: E↑, C↑, E=C
GHJ
36 E  Moderate to
(49.61±11.27) severe pain and
20 ♀ (55.6%) stiffness for 6
 64

16 ♂ (44.4%) months

36 C
(49.08±11.78)
21 ♀ (58.3%)
15 ♂ (41.7%)
Muhammed Acute stage  Complaints <3  History of trauma, E1: PIMR, Maitland F: 5x/w After 2 Pain & disability E1↑, E2↑, C↑, E1>E2>C
t al., 2018 adhesive months shoulder dislocation, LLLT and home mobilization D: 2 w weeks (SPADI)
8
capsulitis  Radiographic cervical radiculopathy care program and PT modality I: ±20 min treatment
evidence for  Fibromyalgia PROM (°)
ndia 30 T adhesive  Hemiplegic shoulder E2: Codman - flexion E1↑, E2↑, C↑, E1>E2,
13 ♀ (43.3%) capsulitis  RA pendulum E1>C, E2=C
17 ♂ (56.7%)  Reduction  Shoulder pain>3 exercises and - extension E1↑, E2↑, C↑, E1=E2=C
shoulder months LLLT - abduction E1↑, E2↑, C↑, E1>C,
10 E1 (53±6.61) movements E1=E2, E2=C
6 ♀ (60%) - ER E1↑, E2↑, C↑, E1>C,
4 ♂ (40%) E2>C, E1=E2
- IR E1↑, E2↑, C↑, E1>C,
10 E2 E2>C, E1=E2
(50.7±6.34)
3 ♀ (30%)
7 ♂ (70%)

10 C (54.9±5.38)
6 ♀ (60%)
4 ♂ (40%)
Pajareya et Primary adhesive  Shoulder pain  Secondary adhesive Medication and Medication and Medication 3, 6, 12, SPADI 3 w: E>C
l., 2004 79 capsulitis  Limitation of capsulitis advice advice : and 24
PROM in all  Intrinsic and extrinsic + F: daily weeks (6, PROM (°)
Thailand 119 T directions causes of shoulder hospital based D: 3 w 12 and 24 - abduction 3 w: E>C
6 ♀ (60%) problems PT program I: 3x/day only - ER 3 w: E=C
4 ♂ (40%)  Generalized arthritis (including successful - IR 3 w: E>C
 Bilateral involvement exercises) and PT treatment)
60 E (56.3±10.6)  Contra-indication for HEP program Treatment 3 w: E>C
36 ♀ (60%) NSAIDs F: 3x/w satisfaction
24 ♂ (40%)  Bleeding tendencies D: 3 w
I: ±60 min Successful 3, 6 w: E>C
 65

59 C (57.7±10) treatment (self- 12, 24 w: E=C

45 ♀ (76.3%) rated disappearance
14 ♂ (23.7%) of complaints)
T: total group study; E: experimental group; C: control group; ♀: female; ♂: male; MUA: manipulation under anesthesia; PT: physical therapy; PNF: proprioceptive neuromuscular facilitation; F:
requency; w: week; D: duration; I: intensity; VAS: Visual Analogue Scale; LSST: lateral scapular slide test; AROM: active range of motion; SST: Simple Shoulder Test; GHJ: glenohumeral joint; SGA:
tanding group asana; min: minutes; SPADI: Shoulder Pain and Disability Index; FFU: final follow up; RA: rheumatoid arthritis; PIMR: position induced movement re-education; LLLT: low level laser
herapy; PROM: passive range of motion; ER: external rotation; IR: internal rotation; HEP: home exercise program;

↑: improved
=: not improved
>: improved more than
<: improved less than
 66

Table 7: Characteristics of studies comparing 2 physical therapy programs both including exercises

Source & Participants Experimental Control Dose Follow-up Outcome measures Results
origin Group Inclusion Exclusion intervention intervention
composition and
patient
characteristics
Abd Diabetic frozen  Shoulder pain &  Bilateral shoulder Traditional Traditional F: 3x/w After 4 weeks Scapular tipping (A-T
Elhamed, et shoulder restriction in involvement treatment treatment D: 4 w treatment (pre- distance)
l., 2018 63 ROM (50% loss  Other GHJ or AC (including (including I: ±15 min post) - supine E↑, C=, E>C
30 T (40-60) of PROM of the disorders or home home (w/o - supine with E↑, C=, E>C
Egypt ? shoulder relative surgery program) program) exercises) retraction
to unaffected  Presence of co- + - standing E↑, C=, E>C
15 E side in at least 3 morbidities Strengthening - standing with E↑, C=, E>C
(25.06±3.36) directions) for a  Pregnancy exercises retraction
duration of >3  Unwillingness to lower fibers
15 C months participate trapezius
(26.06±3.39)  No treatment
other than
analgesics was
prescribed within
last 3 months
 No abnormal
radiographic
findings
Aggarwal Adhesive  Between 35 and  Past UE injuries in Hydrocollator Hydrocollator F: 3x/w After 2 weeks Pain (NPRS) All FU: E↑, C↑, E=C
t al., 2021 capsulitis 60 years last 6 months pack, pack, D: 4 w treatment and
3
 Showing presence  History of surgeries exercises, exercises, I: - at 4 weeks FU Function (SPADI) All FU: E↑, C↑, E=C
30 T of capsular of arm Maitland Maitland
ndia 23 ♀ (76.7%) pattern  Open wounds, mobilizations mobilizations PROM (°)
7 ♂ (23.3%) unhealed sutures, (grade III, IV), (grade III, IV), - flexion 2 w: E↑, C↑, E=C
hypersensitivity, stretches stretches 4 w: E↑, C↑, E>C
15 E generalized + - extension 2 w: E↑, C↑, E=C
(52.67±6.25) infections and IASTM 4 w: E↑, C↑, E>C
10 ♀ (66.7%) uncontrolled - abduction All FU: E↑, C↑, E>C
5 ♂ (33.3%) hypertension - ER 2 w: E↑, C↑, E=C
 67

4 w: E↑, C↑, E>C

- IR All FU: E↑, C↑, E>C
15 C
(46.13±8.66) AROM (°)
13 ♀ (86.7%) - flexion 2 w: E↑, C↑, E=C
2 ♂ (13.3%) 4 w: E↑, C↑, E>C
- extension 2 w: E↑, C↑, E=C
4 w: E↑, C↑, E>C
- abduction All FU: E↑, C↑, E>C
- ER All FU: E↑, C↑, E>C
- IR All FU: E↑, C↑, E=C

Functional ROM
(Apley’s scratch test)
- overhead All FU: E↑, C↑, E=C
- behind back All FU: E↑, C↑, E>C
- across body All FU: E↑, C↑, E>C
Akbas et Adhesive  Grade 2 or 3  Other GHJ HEP HEP F: 5x/w After 3 weeks Pain (VAS)
l., 2015 64 capsulitis adhesive disorders or surgery + + D: 3 w treatment (pre- - rest E=, C=, E=C
capsulitis  Being unable to Before Before I: 25 min (w/o post) - night E↑, C=, E=C
Turkey 36 T cooperate with exercises PT exercises PT exercises) - activity E↑, C↑, E=C
(54.35±10.52) exercises modalities modalities
16 ♀ (44.4%) + PROM (°)
20 ♂ (55.6%) PNF exercises - flexion E↑, C↑, E>C
- abduction E↑, C↑, E>C
18 E - ER E↑, C↑, E=C
(53.94±9.38) - IR E↑, C↑, E=C
7 ♀ (38.9%)
11 ♂ (61.1%) Functional (SPADI)
- pain E↑, C↑, E>C
18 C - disability E↑, C↑, E=C
(54.81±11.96) - total E↑, C↑, E=C
9 ♀ (50%)
9 ♂ (50%)
Atan et al., Adhesive  18 to 65 years  History of bilateral Therapeutic Therapeutic F: 5x/w After 3 weeks Pain (VAS) All FU: E↑, C=, E>C
2021 59 capsulitis  Passive ER concurrent adhesive exercises exercises D: 3 w treatment and
restriction <50% capsulitis, shoulder + + I: 25 min. at 12 week FU Pain and disability All FU: E↑, C↑, E>C
 68

Turkey 21 T of contralateral trauma, fracture, High intensity sham laser exercises, 15 (SPADI)
15 ♀ (71.4%) shoulder shoulder surgery, laser therapy min.
6 ♂ (28.6%)  Normal calcific laser/sham Quality of life (SF-
radiograph tendinopathy, GH 36)
11 E finding of the OA, inflammatory - PF All FU: E↑, C=, E=C
(56.00±11.63) affected shoulder rheumatic diseases, - RLPH All FU: E↑, C=, E= C
7 ♀ (63.6%)  Complaints of tumor and infection - RLE All FU: E↑, C=, E=C
4 ♂ (36.4%) shoulder  History of CSI in - EF All FU: E↑, C=, E>C
restriction with the shoulder during - EWB All FU: E↑, C=, E=C
10 C severe pain for at last 3 months - SF All FU: E↑, C=, E=C
(60.80±8.32) least 1 month  History of recent -P All FU: E↑, C=, E>C
8 ♀ (80%)  Literate and lung, breast, or - GH All FU: E↑, C=
2 ♂ (20%) ability to bypass 3 w: E=C
comprehend surgery/radiotherap 12 w: E>C
verbal y - HC All FU: E↑, C=, E=C
instructions in our  History of cervical
language radiculopathy/brach AROM (°) All FU: E↑, C↑, E=C
ial plexus lesion - flexion
 History of - abduction
neuromuscular - ER
disease - IR
 History of physical
therapy program for PROM (°) All FU: E↑, C↑, E=C
the same shoulder - flexion
last 6 months - abduction
- ER
- IR
Balci et al., Unilateral  Pain in the  History of surgery PT modalities PT modalities F:once After 1 session Pain (VAS) E↑, C=, E=C
2016 66 adhesive shoulder for at or MUA + + D: once
capsulitis stage II least 3 months  Pain or disorders PNF exercises Classic group I: 1 h Scapular dyskinesis E=, C=, E=C
Turkey of the cervical exercises (LSST)
53 T * spine, elbow,
40 ♀ (75.5%) wrist or hand AROM (°)
13 ♂ (24.5%)  Other pathological - flexion E↑, C↑, E=C
conditions - abduction E↑, C↑, E=C
18 E (56.7±7.7) (including
14 ♀ (77.8%) neurologic) Function (SST) E↑, C↑, E=C
 69

4 ♂ (22.2%) involving the

shoulder
18 C (58.1±8.4)
15 ♀ (83.3%)
3 ♂ (16.7%)
Baskaya et Adhesive  Pain in a single  Hemiplegia Exercises with Exercises with F: 10 sessions Pre and post Pain (VAS) E↑, C↑, E>C
l., 2018 67 capsulitis shoulder  DM reflecting side non-reflecting D: ? treatment
 <135° shoulder  Excessive of a mirror side of a I: 1 h AROM (°)
Turkey 30 T elevation limitation & pain + mirror - flexion E↑, C↑, E=C
(56.63±9.49)  Limitation related to head Standard PT + - abduction E↑, C↑, E=C
21 ♀ (70%) shoulder and neck program Standard PT - IR E=, C=, E=C
9 ♂ (30%) movement only movements (including program - ER E=, C=, E=C
at GHJ  Strength sensory exercises and (including
15 E (54.4±7.6) or reflex deficit in a HEP). exercises and PROM (°)
9 ♀ (60%) UE a HEP). - flexion E↑, C↑, E=C
6 ♂ (40%)  Other GHJ - abduction E↑, C↑, E=C
disorders - IR E↑, C↑, E=C
15 C (59.8±10.6)  Major trauma - ER E↑, C↑, E=C
12 ♀ (80%) history
3 ♂ (20%) Functional (UCLA) E↑, C↑, E>C
 History of intra-
articular injections
in preceding 3
months or PT in
preceding 6
months
Celik, 2010 Frozen shoulder  ROM ER,  Radiculopathy PT modalities, PT modalities, F: 5x/w 6 & 12 weeks Function (Modified Both FU: E↑, C↑, E=C
9
abduction &  TOS NSAID, NSAID, D: 6 w CMS)
29 T (52.1 (38- flexion<50%  Rheumatologic exercises exercises I: ±45 min
Turkey 65)) compared to disorders (including (including (w/o Pain (VAS) Both FU: E↑, C↑
22 ♀ (75.9%) contralateral side  Fractures & PNF & HEP) PNF & HEP) exercises) 6 w: E>C
7 ♂ (24.1%)  Normal tumors of the UE + 12 w: E=C
radiography  Neurological ST exercises
15 E (49.6 (38-  Secondary frozen disorders causing PROM (°)
62)) shoulder with muscle weakness - flexion Both FU: E↑, C↑
13 ♀ (86.7%) MRI showing in the shoulder 6 w: E=C
2 ♂ (13.3%) small RC tear 12 w: E>C
- ER Both FU: E↑, C↑, E=C
 70

14 C (54.78 (42-  Secondary frozen - IR Both FU: E↑, C↑, E=C

65)) shoulder with
9 ♀ (64.3%) type II SAI
5 ♂ (35.7%)
Contractor Adhesive  Having painful  RC tears CPT CPT F: 3x/wk After 4 weeks Pain (VAS) E↑, C↑, E=C
t al., 2016 capsulitis stiff shoulder for  History of RA (including (including D: 4 w treatment (pre-
4
at least 3 months  Adhesive exercises) exercises) I: 20 min (w/o post) Function (SPADI) E↑, C↑, E>C
30 T  Idiopathic capsulitis + exercises)
ndia adhesive secondary to Muscle Energy
15 E capsulitis fracture, Techniques
 Subjects with dislocation, reflex
15 C DM sympathetic
 Limited ROM dystrophy,
abduction & ER neurological
 Bi/unilateral disorder, TOS &
adhesive peripheral nerve
capsulitis injury
Ekim et al., Adhesive  marked loss of  presence of co- PT modalities, PT modalities, Supervised After 4- and Pain (VAS)
2016 71 Capsulitis (phase AROM & morbidities HEP HEP F: 5x/w 12-weeks - night Both FU: E↑, C↑, E>C
2) and DM (w/o PROM (>50%  stiff shoulder due + + D: 4 w treatment - rest Both FU:E↑, C↑, E>C
Turkey complications) ER loss) to fracture, CPM CPT treatment I: 45 min - movement Both FU:E↑, C↑
 shoulder pain dislocation, treatment (exercises) 4 w: E>C
41 T and stiffness calcific tendonitis, HEP (after 4 12 w: E=C
(phase 2) reflex sympathetic weeks):
20 E (60.5±8.1)  pain at extremes dystrophy F: - AROM (°)
13 ♀ (65%) of all shoulder  intra-articular D: 8 w - flexion Both FU: E↑, C↑, E>C
7 ♂ (35%) motions injections to the I: - - abduction Both FU: E↑, C↑, E>C
 normal findings shoulder last 3 - ER Both FU: E↑, C↑, E=C
21 C (60.4±6.7) on radiographs months CPM, CPT - IR Both FU: E↑, C↑, E=C
13 ♀ (61.9%)  type 2 DM for at F: daily
8 ♂ (38.1%) least 2 years D: 4 w PROM (°)
I: 1 h - flexion Both FU: E↑, C↑, E>C
- abduction Both FU: E↑, C↑, E>C
- ER Both FU: E↑, C↑, E=C
- IR Both FU: E↑, C↑
4 w: E=C
12 w: E>C
 71

Function
- CMS Both FU: E↑, C↑, E>C

- SPADI
* pain Both FU: E↑, C↑, E>C
* disability Both FU: E↑, C↑, E>C
Elhafez et Unilateral  painful, restricted  local CSI to the E1: Traditional PT F: 3x/w After 4 weeks Pain (NRS) E1↑, E2↑, C↑, E2>C,
l., 2016 72 Adhesive AROM & shoulder within Traditional PT (including D: 4 w treatment E2>E1, E1=C
Capsulitis stage II PROM last 3 months or (including laser (different I: 30 min (w/o
Egypt  capsular pattern current CS laser, region), exercises) AROM
45 T (40-60) of motion therapy supervised supervised - flexion E1↑, E2↑, C↑, E2>C
restriction  shoulder exercises & exercises & HEP: - abduction E1↑, E2↑, C↑, E2>C,
15 E1  absence of symptoms due to HEP) HEP) F: daily E2>E1, E1=C
(50.06±5.3) radiologic other causes or D: 4 w - ER E1↑, E2↑, C↑,
8 ♀ (53.3%) evidence of GHJ history of E2: I: 1-2/d E2>E1>C
7 ♂ (56.7%) arthritis shoulder surgery Traditional PT
 pregnancy (including Postisometric
15 E2 (49.5±4.6)  presence of laser, facilitation
10 ♀ (67.7%) comorbidities supervised F: 3x/w
5 ♂ (33.3%) exercises & D: 4 w
HEP) & I: 9-13 min
15 C (50.4±5.3) postisometric
9 ♀ (60%) facilitation
6 ♂ (40%) technique
Gutierrez Primary adhesive  Unilateral  Secondary to UE cycle CPT F: 2 or 3x/w Pre and post PROM (°)
Espinoza et capsulitis adhesive other shoulder ergometer, GH (including D: 10 sessions treatment - ER E↑, C↑, E>C
l., 2015 56 capsulitis disorders or posterior exercises) I: at least 15 - flexion E↑, C↑, E>C
57 T surgery mobilization min - abduction E↑, C↑, E>C
Chile 46 ♀ (80.7%)  High level of and distraction
11 ♂ (19.3%) irritability (Kaltenborn Pain (VAS) E↑, C↑, E>C
 Non-steroid anti- III)
29 E inflammatory Function (CMS) E↑, C↑, E>C
23 ♀ (79.3%) drug infiltration or
6 ♂ (20.7%) CSI in the last 6
months
28 C  Stroke
 72

23 ♀ (82.1%)  Previously treated

5 ♂ (17.9%) with release
technique and/or
MUA
Hussein et Adhesive  Globally limited  Bilateral shoulder Traditional PT Traditional PT Traditional After 4 weeks PROM (°)
l., 2015 73 capsulitis stage 3 GH translation involvement (including (including PT: treatment and - abduction All FU: E>C
or 4  Loss of PROM  Previous shoulder HEP) HEP) F: 3x/w after 12, 24, - ER All FU: E>C
USA >50% compared surgery + D: 4 w 52, 104 weeks
60 T to non-affected  Any Static I: 20 min AROM (°)
31 ♀ (51.7%) side neuromuscular progressive - abduction All FU: E>C
29 ♂ (48.3%)  No radiographic disorders stretching HEP:
findings on AP,  DM F: daily Function (DASH) 4 w E=C
30 E (51.9) axillary or  CSI previous 6 D: 4 weeks All other FU: E>C
scapular y-view months I: 3x10 rep
30 C (51.2) shoulder  Prior trauma
Stretching: Pain (VAS)
 Any intrinsic GH
F: daily - rest 4, 12 and 104 w: E=C
pathology
D: 4 w 24 and 52 w: E>C
 CRPS
I: w 1: 1x30
 Pulmonary
min.
disease
w 2, 3: 2x30
 Contra-indications min.
to treatment w 4: 3x30
min.
unaid et Frozen shoulder  No recent injury,  Diabetic patients Routine PT Routine PT F: 4x/w After 2 weeks Pain (VAS) E=C
l., 2016 74 fracture, cancer  Major (including (including D: 2 w treatment
52 T (48.90 (30- and no metabolic musculoskeletal exercises) exercises) I: E: 40 min; ROM (°)
Pakistan 60)) diseases problems + C: 25 min. - abduction E>C
26 ♀ (50%)  Red flags Kaltenborn - flexion E>C
26 ♂ (50%)  History of mobilization - extension E>C
shoulder trauma - IR E>C
26 E or prolonged - ER E>C
immobilization
26 C due to neurologic Function (PENN E>C
disorder shoulder scale)
 Suffering with
Neuralgia/hemiple
 73

gia
 Bilateral frozen
shoulder
Kumar et Primary Adhesive  Primary  Previous shoulder conservative Conservative F: 4x/w Pre- Pain (VAS) E↑, C↑, E>C
l., 2017 76 Capsulitis idiopathic surgeries to management management D: - posttreamtent
adhesive affected shoulder, (including (including I: - ROM (°)
ndia 30 T capsulitis with neck, elbow exercises) exercises) - ER E↑, C↑, E>C
trigger points in  Secondary +
15 E subscapularis adhesive Spray & Function (SPADI) E↑, C↑, E>C
 Painful stiff capsulitis stretch
15 C shoulder >3  Other technique Muscle strength
months comorbidities) - ER E↑, C↑, E>C
 Male/female  CSI in affected
 Unilateral shoulder in
condition with preceding 4 weeks
50% ROM  Other
compared to inflammatory
unaffected side conditions
 Allergic to spray
Leclaire & Periarthritis of  Shoulder pain >2  Presence of co- PT modalities PT modalities F: 3x/w After 4, 8, and ROM (°)
Bourgouin, the shoulder months morbidities and exercises and exercises (exercises 12 weeks - flexion All FU: E↑, C↑, E=C
1991 60  Limited AROM  RC rupture + + daily) treatment - extension All FU: E=, C=, E=C
47 T (58±6.9) and PROM  X-ray Electromagnet Sham therapy D: 12 w - abduction All FU: E↑, C↑, E=C
France 29 ♀ (61.7%)  Pain on resisted calcification ic therapy I: 35 min - adduction All FU: E=, C=, E=C
18 ♂ (38.3%) abduction, IR or  Severe adhesive (supervised), - ER All FU: E↑, C↑, E=C
ER capsulitis 20 min - IR All FU: E↑, C↑, E=C
22 E  Impaired GHJ (flexion<100°, (exercises)
motion abduction<90° or Pain (ordinal scale)
25 C global rotations - rest 4 w: E↓, C↑, E<C
<20°) Other FU: ↑, C↑, E=C
 Receiving - motion All FU: E↑, C↑, E=C
anticoagulants or - lying down All FU: E↑, C↑, E=C
anti-inflammatory
drugs or received Self-rating disability
CSI scale
- functional All FU: E↑, C↑, E=C
- pain All FU: E↑, C↑, E=C
 74

Lokesh et Periarthritis  Capsular pattern  Shoulder trauma HEP and CPT HEP and CPT F: 6x/w After 2 weeks ROM (°)
l., 2015 77 shoulder of restriction or disorders + D: 2 w treatment - flexion FFU: E↑, C↑, E>C
 History of pain  Neurological muscle energy I: - - abduction FFU: E↑, C↑, E>C
ndia 30 T (40-60) for 3-18 months disorders techniques. Before 3rd, 6th, - IR FFU: E↑, C↑, E>C
 Radiating pain 9th and 12th - ER FFU: E↑, C↑, E>C
?E  Neoplastic treatment
conditions session Pain (VAS) FFU: E↑, C↑, E>C
?C
Function (SPADI)
- pain FFU: E↑, C↑, E>C
- disability FFU: E↑, C↑, E>C
- total FFU: E↑, C↑, E>C
Mohamed Unilateral  Inability to  Presence of any Hot pack and Hot pack and F: 3x/w After 2 weeks, Scapular upward 2 w: E↑, C=, E>C
t al., 2020 adhesive elevate the arm shoulder scapular scapular D: 2 months 2 and 6 rotation (°) 2, 6 m: E↑, C↑, E>C
8
capsulitis above 100 condition that is a mobilization mobilization I: 40 min months
degrees in the contraindication + + ROM (°)
Egypt 60 T plane of the for exercising the Dynamic Placebo active - flexion 2 w: E↑, C=, E>C
26 ♀ (43%) scapula shoulder joint scapular shoulder 2, 6 m: E↑, C↑, E>C
34 ♂ (57%)  Limitation in both  No signs of recognition exercise with - abduction 2 w: E↑, C=, E>C
active and passive scapular exercise uninvolved 2, 6 m: E↑, C↑, E>C
30 E shoulder ROM dyskinesis shoulder - ER 2 w: E=, C=, E=C
(51.93±6.16)  Presence of pain 2 m: E↑, C↑, E>C
12 ♀ (40%) interfering with 6 m: E↑, C=, E>C
18 ♂ (60%) activities of daily
living Pain and disability 2 w: E↓, C↓, E<C
30 C (SPADI) 2 m: E↓, C↓, E<C
(50.06±5.87) 6 m: E↓, C=, E<C
14 ♀ (47%)
16 ♂ (53%)
Muhammed Acute stage  Complaints <3  History of trauma, PIMR, LLLT Codman F: 5x/w After 2 weeks Pain & disability E↑, C↑, E1>C
t al., 2018 adhesive months shoulder and home care pendulum D: 2 w treatment (SPADI)
8
capsulitis  Radiographic dislocation, program exercises and I: ±20 min
evidence for cervical LLLT PROM (°)
ndia 30 T * adhesive radiculopathy - flexion E↑, C↑, E>C
13 ♀ (43.3%) capsulitis  Fibromyalgia - extension E↑, C↑, E=C
17 ♂ (56.7%)  Restricted  Hemiplegic - abduction E↑, C↑, E=C
shoulder shoulder - IR E↑, C↑, E=C
 75

10 E (53±6.61) movements  RA - ER E↑, C↑, E=C

6 ♀ (60%)  Shoulder pain>3
4 ♂ (40%) months

10 C (50.7±6.34)
3 ♀ (30%)
7 ♂ (70%)
Nellutla & Chronic frozen  Restricted ROM  none PT modality, PT modality, Conventional After 3 weeks CMS E↑, C↑, E=C
Giri, 2011 shoulder  Limitations in mobilizations mobilizations F: 6x/w treatment (pre-
7
ADL (GH, AC, SC, (GH, AC, SC, D: 3 w post)
40 T 56.15±8.71  Pain score 10 or ST) ST) I: ±20 min
ndia 16 ♀ (60%) 15 on CMS + + (w/o
24 ♂ (40%) PNF Conventional exercises)
free exercises
20 E (including PNF
6 ♀ (60%) HEP) F: daily
4 ♂ (40%) D: 3 w
I: 3x/day
20 C HEP
6 ♀ (60%) F: daily
4 ♂ (40%) D: 3 w
I: 2x/d
Rawat et Adhesive  1-3 months onset  OA or signs of HEP HEP F: 3x/w After 4 weeks Pain (VAS) E>C
l., 2017 61 capsulitis of pain & bony damage + + D: 4 w treatment (pre-
stiffness  Hypermobility PT modality, PT modality, I: - post) ROM (°)
ndia 42 T  ROM restriction and instability mobilization mobilization - flexion E=C
45 ♀ (76.3%) in ER, abduction  Neurological + - abduction E>C
14 ♂ (23.7%) & flexion <50% disorder causing RC muscle - IR E>C
compared to muscle weakness strengthening - ER E>C
21 E contralateral side  Any local or
(56.00±10.42)  Pain during sleep systemic disease Function
11 ♀ (52.4%)  Difficulty with  Upper limb nerve - PSFS E>C
10 ♂ (47.6%) grooming, tension testing - SPADI E>C
dressing and reproduces the
21 C reaching to symptoms Muscle strength
(54.19±8.33) shoulder level, - flexors E=C
7 ♀ (33.3%) behind the back - extensors E>C
 76

14 ♂ (66.7%) and overhead - abductors E>C

- adductors E>C
- IR E>C
- ER E>C
Rizk et al., Adhesive  Pain on resisted  Bony or HEP HEP HEP: Monthly up to ROM (°)
1983 62 capsulitis abduction and/or neurological + + F: daily 8m - IR All FU: E↑, C↑, E>C
IR or ER disorders PT modality CPT D: ? - ER All FU: E↑, C↑, E>C
USA 50 T (56 (40-70))  Localization of  Polyarthritis combined with (including I: 5 3x/d Weekly for 8 - flexion All FU: E↑, C↑, E>C
32 ♀ (64%) impaired traction exercises) weeks, - extension All FU: E↑, C↑, E>C
18 ♂ (36%) movement to Supervised PT monthly for 6 - adduction All FU: E↑, C↑, E>C
GHJ exclusively F: 4x/w - 3x/w m. - abduction All FU: E↑, C↑, E>C
24 E  Maximum D: 4 w - 4 w
? PROM < 110° I: E: 2 h; C: - Function 1 m: E=, C=, E=C
abduction, 50° 2, 3 m: E↑, C=, E>C
26 C ER, 70° IR and 4, 5, 6, 7, 8 m: E↑, C↑,
? 140° flexion E>C

Night pain 1, 2, 3, 4 m: E↑, C=,

E>C
All other FU: E↑, C↑,
E>C
Shen et al., Scapulohumeral  Chronic onset  Experienced acute Tuina Tuina Tuina: After 1-month Pain intensity (VAS) E↑, C↑, E>C
2017 88 periarthritis  History of injury inflammation of treatment treatment F: 3-4x/w treatment (pre-
 Deficiency of qi the shoulder (mobilization, (mobilization, D: 1 month post) Function (CMS) E↑, C↑, E>C
China 30 T and blood  Shoulder injury or manipulation) manipulation) I: 20 min.
coupled with bone fracture + +
15 E (55.3±6.7) external  Shoulder tumor Yi jin jing Shoulder joint Exercises:
10 ♀ (67.7%) contraction of  Severe heart, (exercises) functional F: daily
5 ♂ (33.3%) wind, cold and brain or kidney exercise D: 1 month
dampness diseases I: -
15 C (57.6±8.7)  Shoulder pain,  History of mental
8 ♀ (53.3%) aggravate at disorder
7 ♂ (46.7%) night
 Induced by
weather change
or fatigue
 Limited shoulder
 77

joint movement
 Incidence of
shoulder muscle
atrophy
 Pressing pain on
shoulder
 Negative X-ray
 Did not receive
therapy last 2
months
Sule et al., Adhesive  Prediagnosed  History of CPT CPT F: 5x/w After 2 weeks ROM
2015 80 capsulitis adhesive uncontrolled DM (including (including D: 2 w (10 d) treatment (at - flexion E↑, C↑, E>C
(subacute & capsulitis  Recent fracture exercises) exercises) I: - 10th day) - extension E↑, C↑, E>C
ndia chronic stage)  Subacute & upper limb + - abduction E↑, C↑, E=C
chronic stage  Elbow pathology Sleepers - IR E↑, C↑, E>C
30 T  Both male & restricting ROM stretch - ER E↑, C↑, E>C
(56.27±5.20) female  Cervical - horizontal adduction E↑, C↑, E>C
 Having at least radiculopathy - horizontal abduction E↑, C↑, E=C
15 E 90° shoulder
abduction and SPADI
15 C elbow flexion - pain E↑, C↑, E=C
- function E↑, C↑, E=C
Yang et al., Frozen shoulder  >50% loss of  History of stroke E1: Standardized F: 2x/w After 4 and 8 FLEX-SF 4 w: E1=E2, E1=C,
2012 81 syndrome PROM in 2 or with residual Standardized treatment D: 3 months weeks of E2=C
more directions upper extremity treatment (including I: - treatment 8 w: E1>E2, E1=C,
Taiwan 34 T  Duration of involvement (including exercises) E2<C
complaints >3  Presence of co- exercises)
10 E1 (56.8±7.2) months morbidities + ROM (°)
7 ♀ (70%)  Other shoulder End range - IR All FU: E1>E2, E1=C,
3 ♂ (30%) disorders or mobilization E2<C
surgery & scapular - ER 4 w: E1=E2, E1=C,
12 E2  Pain or disorders mobilization E2<C
(54.9±10.3) of the cervical 8 w: E1>E2, E1=C,
10 ♀ (83.3%) spine, elbow, E2: E2<C
2 ♂ (16.7%) wrist or hand Standardized
treatment
10 C (54.3±7.6) (including
 78

5 ♀ (50%) exercises)
5 ♂ (50%) +
Passive mid-
range
mobilization

**

* study with 3 experimental groups, only the relevant groups for this comparison are shown.
** patients with less kinematics as 8° scapular posterior tipping, 97° humeral elevation & 39° humeral ER during elevation received E1 or E2, patients with larger kinematic received the control
ntervention.

T: total group study; E: experimental group; C: control group; ROM: range of motion; PROM: passive range of motion; GHJ: glenohumeral joint; AC: acromioclavicular; F: frequency; w: week; D:
duration; I: intensity; min.: minutes; w/o: without: A-T: acromion-table; ♀: female; ♂: male; UE: upper extremity; IASTM: instrument assisted soft tissue mobilization; NPRS: numeric pain rating scale;
SPADI: Shoulder Pain and Disability Index; ER: external rotation; IR: internal rotation; AROM: active range of motion; FU: follow up; HEP: home exercise program; PT: physical therapy; PNF:
Proprioceptive Neuromuscular Facilitation; VAS: Visual Analogue Scale; SF-36: 36-item short form health survey; PF: physical functioning; RLPH: role limitations due to physical health; RLE: role
imitations due to emotional problems; EF: energy/fatigue; EWB: emotional well-being; SF: social functioning; P: pain; GH: general health; HC: health change; MUA: manipulation under anesthesia; h:
hour; LSST: lateral scapular slide test; SST: Simple Shoulder Test; DM: Diabetes Mellitus; UCLA: University of California Los Angeles scale; RC: rotator cuff; SAI: subacromial impingement; TOS:
horacic outlet syndrome; ST: scapula-thoracic; CMS: Constant Murley Score; RA: rheumatoid arthritis; CPT: conventional physical therapy; CPM: continuous passive motion; CSI: corticosteroid
njection; GH: glenohumeral; AP: anterior-posterior; CRPS: complex regional pain syndrome; DASH: Disabilities of Arm, Shoulder and Hand; FFU: final follow up; PIMR: position induced movement
e-education; LLLT: low level laser therapy; ADL: activities of daily living; SC: sternoclavicular; OA: osteoarthritis; PSFS: patient specific functional score; m: months

↑: improved
=: no change
↓: deteriorated
>: scored better than
<: scored worse than
 79

Box 1: Inclusion criteria

Design
 Randomized controlled trials
Participants
 Patients with frozen shoulder
 Primary or secondary (systemic and intrinsic) frozen shoulder
 Humans >18 years
Intervention
 Exercise therapy
Outcome measures
 Pain
 Range of motion
 Muscle strength
 Functional ability
 Patient satisfaction
Language
 English or Dutch