does evidence support physiotherapy management of adult complex regional pain syndrome type one a...

8/10/2019 Does Evidence Support Physiotherapy Management of Adult Complex Regional Pain Syndrome Type One a Syste

1/15

Review

Does evidence support physiotherapy management of adult

Complex Regional Pain Syndrome Type One? A systematic review

Anne E. Daly a,b,*, Andrea E. Bialocerkowski b

a Department of Physiotherapy, Austin Hospital, P.O. Box 5555, Heidelberg 3084, Victoria, Australiab School of Physiotherapy, The University of Melbourne, Parkville 3010, Victoria, Australia

a r t i c l e i n f o

Article history:Received 26 November 2007

Received in revised form 19 March 2008

Accepted 1 May 2008

Available online 10 July 2008

Keywords:

Complex Regional Pain Syndrome Type One

Physiotherapy

Systematic review

Critical review form

a b s t r a c t

Objective: To source and critically evaluate the evidence on the effectiveness of Physiotherapy to manageadult CRPS-1.

Design: Systematic literature review.

Methods: Electronic databases, conference proceedings, clinical guidelines and text books were searched

for quantitative studies on CRPS-1 in adults where Physiotherapy was a sole or significant component of

the intervention. Data were extracted according to predefined criteria by two independent reviewers.

Methodological quality was assessed using the Critical Review Form.

Results: The search strategy identified 1320 potential articles. Of these, 14 articles, representing 11 stud-

ies, met inclusion criteria. There were five randomised controlled trials, one comparative study and five

case series. Methodological quality was dependent on study type, with randomised controlled trials being

higher in quality. Physiotherapy treatments varied between studies and were often provided in combina-

tion with medical management. This did not allow for the stand-alone value of Physiotherapy to be

determined. Heterogeneity across the studies, with respect to participants, interventions evaluated and

outcome measures used, prevented meta-analysis. Narrative synthesis of the results, based on effect size,

found there was good to very good quality level II evidence that graded motor imagery is effective in

reducing pain in adults with CRPS-1, irrespective of the outcome measure used. No evidence was foundto support treatments frequently recommended in clinical guidelines, such as stress loading.

Conclusions: Graded motor imagery should be used to reduce pain in adult CRPS-1 patients. Further, the

results of this review should be used to update CRPS-1 clinical guidelines.

2008 European Federation of Chapters of the International Association for the Study of Pain. Published

by Elsevier Ltd. All rights reserved.

1. Introduction

Complex Regional Pain Syndrome Type One (CRPS-1), which is

characterized by pain, swelling, autonomic and motor disturbances

in the absence of a peripheral nerve injury, is a disabling and dis-

tressing condition that is difficult to treat effectively. Many treat-

ments have been developed to manage this condition due to the

spectrum of symptoms and an incomplete understanding of the

pathophysiology behind CRPS-1. These treatments include allied

health therapies, such as physiotherapy, complementary and alter-

native medicine, medical, pharmacological and psychological ap-

proaches (Wilson et al., 2005).

There is some evidence to support the use of pharmacological

approaches in the management of CRPS-1. For example, anticon-

vulsants, sodium channel blockers, oral corticosteroids, calcium

regulating drugs and free radical scavengers and the management

of neuropathic pain strategies (such as the use of antidepressants,

anticonvulsants, NMDA-receptor antagonists) appear to reduce

CRPS-1 symptoms (Baron, 2005; Cepeda et al., 2002). However,

other treatment approaches, such as spinal cord stimulation and

modulation of the sympathetic nervous system via surgical sympa-

thectomy, a-adrenergic blockade, intravenous regional block and

local anaesthetic block have equivocal evidence (Baron, 2005;

Cepeda et al., 2002; Nelson and Stacey, 2006; Stanton-Hicks,

2006), based on the results of systematic reviews (Cepeda et al.,

2002; Forouzanfar et al., 2002; Jadad et al., 1995) and meta-analy-

ses (Kingery, 1997; Perez et al., 2001).

Physiotherapy is advocated in CRPS clinical guidelines (Harden

et al., 2006c; Stanton-Hicks et al., 2002), textbooks on pain

(Giamberardino, 2002; Harden et al., 2001; Janig and Stanton-

Hicks, 1996; Justins, 2005; McMahon and Koltzenburg, 2005;

Wilson et al., 2005) and narrative CRPS reviews (Baron and Janig,

2004; Berthelot, 2006; Birklein, 2005; Dommerholt, 2004; Harden

et al., 2006a; Quisel et al., 2005; Sharma et al., 2006; Wasner et al.,

2003). Despite this advocacy, the role of physiotherapy in the

1090-3801/$36.00 2008 European Federation of Chapters of the International Association for the Study of Pain. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.ejpain.2008.05.003

* Corresponding author. Address: Department of Physiotherapy, Austin Hospital,

P.O. Box5555, Heidelberg 3084, Victoria, Australia. Tel.: +61 3 94965461; fax: +613

9457 6326.

E-mail address:[email protected](A.E. Daly).

European Journal of Pain 13 (2009) 339353

Contents lists available at ScienceDirect

European Journal of Pain

j o u r n a l h o m e p a g e : w w w . E u r o p e a n J o u r n a l P a i n . c o m
mailto:[email protected]://www.sciencedirect.com/science/journal/10903801http://www.europeanjournalpain.com/http://www.europeanjournalpain.com/http://www.sciencedirect.com/science/journal/10903801mailto:[email protected]


2/15

treatment of CRPS has not been rigorously evaluated using system-

atic methodologies. Therefore, physiotherapy for CRPS is not pres-

ently underpinned by research evidence (Cepeda et al., 2002).

Although the scope and complexity of physiotherapy practice

varies world wide, physiotherapy can be thought of as

The treatment of disorders with physical agents and methods

(Anderson, 2002)

Physical agents and methods include manual therapy, electro-

therapy, transcutaneous electrical nerve stimulation, massage

and therapeutic exercise. Often these treatments are contextua-

lised by theories, such as operant behavioural theory (Fordyce,

1976; Main et al., 2002), cognitive behavioural therapy (Main

et al., 2002; Turk et al., 1983), fear avoidance and graded exposure

in vivo theory (Vlaeyen et al., 2001, 2002). Additionally, empower-

ment and education of the patient underpin the delivery of physio-

therapy treatment (Australian Physiotherapy Association, 2004).

Althoughsome physiotherapy evidence has been includedin the

work by Forouzanfar et al. (2002), the entire body of physiotherapy

evidence has not been synthesized. Therefore, the objective of this

systematic review was to source, critically evaluate and synthesize

current physiotherapy research evidence for the treatment of adult

CRPS-1. The results of this study could then be used to make evi-

dence-based recommendations regarding the physiotherapy treat-

ments for CRPS-1 and to update CRPS treatment guidelines.

2. Method

2.1. Search strategy

An electronic search was performed in September, 2007 cover-

ing the previous two decades of the following databases: CINAHL

(19872007), Medline (19872007), Embase (19872007), ISI

Web of Science (19872007), and the following evidence-based

practice resources; Cochrane Library (19872007) (The Cochrane

Collaboration, 2007), Turning Research into Practice (TRIP) Data-base (complete site) (TRIP Database, 2007), Physiotherapy Evi-

dence Database (PEDro) (complete site) (Centre for Evidence

Based Physiotherapy, 2007), Joanna Briggs Institute (complete site)

(Joanna Briggs Institute, 2007). The websites for these databases

are included in the reference list.

Key words used to describe CRPS-1 and physiotherapy were

listed. Multiple keywords describing CRPS-1 were used as different

terms were used prior to 1994 when the International Association

for the Study of Pain (IASP) introduced the term Complex Regional

Pain Syndrome Type One (Merskey and Bogduk, 1994). Key words

describing physiotherapy were also numerous reflecting both the

range of physiotherapy interventions and the terminology used to

describe such interventions (Table 1). Where possible, key words

were used to identify relevant Medical Subject Headings (MeSH)

thatwere thenexploded. To gaina listof potentially relevantpapers,

the CRPS-1 keywords were combined using OR. This strategy was

repeated for the physiotherapy keywords. To identify papers on

physiotherapy interventionsfor CRPS-1the twogroupsof keywords

were combined using AND. Truncation and wild card symbols

were usedto retrieve all relevantkeywords. In non-MeSH databases

the following CRPS-1 keywords were used: Complex Regional Pain

Syndrome, CRPS, Reflex SympatheticDystrophy, RSDand were com-

bined using OR. The term Complex Regional Pain Syndrome was

used in databases that would only accept one term.

In addition, secondary search strategies were performed by con-

ducting hand searches of

Reference lists of textbooks on pain (Giamberardino, 2002; Jus-

tins, 2005; McMahon and Koltzenburg, 2005) and CRPS (Harden

et al., 2001; Janig and Stanton-Hicks, 1996; Wilson et al., 2005).

Proceedings of the 11th World Congress on Pain (Flor et al.,

2006).

Reference lists of all papers meeting the inclusion criteria.

Internet searches of 30 contributors to three CRPS clinical

guidelines were undertaken (Harden et al., 2006b; Harden et al.,

2006c; Stanton-Hicks et al., 2002) to identify other relevant

publications.

2.2. Inclusion criteria

All titles and abstracts were read to identify relevant papers. Pa-

pers were included in this systematic review if they met the inclu-

sion criteria listed in Table 2.The type of outcome measure used to

evaluate the effect of physiotherapy was not an inclusion criterion

for this review. Currently, consensus is lacking with regards to a

gold standard of outcome measure for both physiotherapy andCRPS-1. Furthermore, as physiotherapy generally aims to affect

multiple outcomes such as pain intensity, strength, weight bearing,

swelling, activity limitation and participation restriction this al-

lowed for a broad consideration of the effectiveness of

physiotherapy.

When there was uncertainty regarding the eligibility of the pa-

per from the abstract, the full text version of the paper was re-

trieved and evaluated against the inclusion criteria. The full text

version of all papers that met the inclusion criteria were retrieved

for quality assessment and data extraction.

Table 1

Keywords used to develop the search strategy

Population Intervention

Complex regional pain syndrome Physiotherapy Therapeutic exercise

CRPS Physical therapy Exercise

Complex regional pain syndrome type one Manual therapy Mirror visual feedback

CRPS-1 Manipulative Physiotherapy Mirror therapy

Reflex Sympathetic Dystrophy Manipulative therapy Graded motor imagery

RSD Hand therapy Graded imagery

Reflex Sympathetic Dystrophy syndrome Rehabilitation Stress loading program

RSDS Electrotherapy Stress loading

Sympathetic Dystrophy Heat therapy Scrub and carry

Neuropathic pain Cryotherapy Massage

Shoulder hand syndrome Ultrasound Sensory motor integration

Sudecks atrophy Transcutaneous electrical nerve stimulation Water exercise

Algodystrophy TENS Aquatic Physiotherapy

Low frequency electrical cutaneous therapy Aquatic therapy

Magnetic Aquatic exercise

Feldenkrais Hydrotherapy

340 A.E. Daly, A.E. Bialocerkowski/ European Journal of Pain 13 (2009) 339353


3/15

2.3. Quality assessment

Critical appraisal of each included study was conducted by

determining:

The level of evidence on the Australian National Health and

Medical Research Council (NHMRC) Hierarchy of Evidence (Aus-tralian National Health and Medical Research Council, 1999)

(Table 3). This provides a broad indication of bias based on study

design. Studies higher on the hierarchy potentially contain less

bias than those that are lower on the hierarchy.

The methodological quality of the study using the Critical

Review Form Quantitative studies (Law et al., 1998a). This tool

can be used to appraise all types of quantitative studies ranging

from RCTs to case series. Thus all quantitative studies on phys-

iotherapy management of CRPS-1 were included in this review

and evaluated for quality using the same tool. This made the

quality results comparable between the different study designs

(Petticrew, 2003). Standardised guidelines on the interpretation

and scoring of each item were used (Law et al., 1998b). Items

were scored as 1 (completely fulfils the criterion) or 0 (doesnot completely fulfil the criterion). The scores of the 16 closed

ended questions were tallied to provide an overall score of qual-

ity where the maximum score of 16 indicated excellent quality

(Bialocerkowski et al., 2005). Two researchers independently

scored the studies and where disagreement occurred, consensus

was achieved by discussion. Quality scores were arbitrarily

divided into five categories: poor (score6 8), fair (score = 9

10), good (score = 1112), very good (score = 1314) and excel-

lent (score = 1516).

The Critical Review Form Quantitative studies (Law et al.,

1998a) includes 17 of the 22 items that are contained in the

CONSORT statement (Altman et al., 2001; Moher et al., 2001).

It does not include items one (study design stated in title or

abstract), eight, nine and 10 (randomisation: sequence genera-

tion, allocation concealment and implementation respectively)

or 19 (adverse events). The CONSORT statement was not

designed to evaluate methodological quality (Altman et al.,

2001). However, in this review, it was noted whether these five

CONSORT criteria were fulfilled by the RCTs. This would providethe reader with further methodological quality information.

2.4. Data extraction

Data were extracted by the authors using the PICO approach

(Stone, 2002)

Participants: area of involvement of CRPS, age, sex, duration of

symptoms, type of referral source and diagnostic criteria.

Interventions: type, intensity, duration, in combination or stand-

alone physiotherapy.

Comparison: to another treatment, no treatment or usual

treatment. Outcomes: domains and tools used to measure the effects of the

intervention. These were grouped according to the World Health

Organisation (WHO) International Classification of Functioning,

Disability and Health: abnormalities of body function and struc-

ture; activity limitation; and participation restriction (World

Health Organisation, 2001). Additional domains included quality

of life and cost effectiveness.

Data on the effectiveness of the physiotherapy treatment were

also extracted for each study. To determine the effect of the

physiotherapy treatment on each outcome measure, the mean

and 95% confidence interval for the between-group differences

Table 3

Study design, based on the Australian National Health and Medical Research Council Hierarchy of Evidence (Australian National Health and Medical Research Council, 1999)

Level Definition Studies

I Evidence obtained from a systematic review of all relevant randomised controlled trials

II Evidence obtained from at least one properly-designed randomised controlled trial Durmus et al. (2004), Moseley (2004, 2005, 2006); Oerlemans

studya

III-1 Evidence obtained from well-designed pseudo-randomised controlled trials (alternate

allocation or some other method)

III-2 Evidence obtained from comparative studies (including systematic reviews of such studies)

with concurrent controls and allocation not randomised, cohort studies, casecontrol studies,

or interrupted time series with a control group

McCabe et al. (2003)

III-3 Evidence obtained from comparative studies with historical control, two or more single arm

studies, or interrupted time series without a parallel control group

IV Evidence obtained from case series, either post-test or pre-test/post-test de Jong et al. (2005), Pleger et al. (2005), Robaina et al. (1989),

Singh et al. (2004), Watson and Carlson (1987)

a Oerlemans et al. (1999a,b, 2000b), Severens et al. (1999).

Table 2

Inclusion criteria used in the systematic review

Criterion Justification

1 English language Major journals in this area are published in this language

2 19872007 Twenty years captures the most frequently used physical modalities in clinical practice today

3 Studied the outcome of Physiotherapy alone or in combination

with other medical or psychological therapies

As interdisciplinary management of CRPS-1 is considered optimal (Stanton-Hicks et al., 2002),

studies that involved Physiotherapy in combination with other interventions were included

4 Humans over 18 years of age This increased the homogeneity of participants between the studies

5 Stated diagnostic criteria based on clinical presentation Diagnostic criteria must be stated so that the reader can determine how likely it was that astudy population actually had CRPS 1 and how similar the study population was compared to

their own CRPS 1 population

6 Quantitative study design including randomised controlled trials

(RCTs), non-randomized clinical trials or case series

Study designs other than RCTs were included in this review as they provide complementary

and relevant clinical detail to the current state of our knowledge and its limitations (Barton,

2000; Norris and Atkins, 2005). Case studies were not included because of the low level of

evidence they provide

A.E. Daly, A.E. Bialocerkowski/ European Journal of Pain 13 (2009) 339353 341


4/15

was calculated for RCTs and comparative studies, based on the

results provided in each article (Herbert, 2000). Moreover the

mean change between pre and post treatment (and 95% confi-

dence intervals) were calculated for the case series. Pain reduc-

tion of more than 20%, irrespective of the measurement tool

used, was considered clinically worth-while (Farrar et al., 2001;

Ferreira et al., 2002). It was expected that there would be heter-

ogeneity in participants, interventions, comparisons and out-

comes. Therefore the results of the studies were synthesized in

a narrative format.

3. Results

3.1. Search strategy yield

Initially 1320 hits were gained from database and secondary

search strategies. After review of the titles and abstracts, duplicates

and those articles that did not meet the inclusion criteria were re-

moved. After reviewing the full text of 180 articles, 166 articles

were excluded as they did not meet the inclusion criteria. There-

fore 14 articles were included in this systematic review (Fig. 1)

(de Jong et al., 2005; Durmus et al., 2004; McCabe et al., 2003;

Moseley, 2004, 2005, 2006; Oerlemans et al., 1999a,b, 2000b; Pleg-

er et al., 2005; Robaina et al., 1989; Severens et al., 1999; Singh

et al., 2004; Watson and Carlson, 1987). However, four of these

articles (Oerlemans et al., 1999a,b, 2000b; Severens et al., 1999)

were derived from one large RCT but were published separately

due to the diversity of outcomes measured. These articles will be

referred to as the Oerlemans study. Consequently, 11 studies were

included in this systematic review.

3.2. Critical appraisal

3.2.1. Hierarchy of evidence

There were five RCTs (Durmus et al., 2004; Moseley, 2004, 2005,

2006, Oerlemans study), one comparative study (McCabe et al.,

2003), and five case series (de Jong et al., 2005; Pleger et al.,

2005; Robaina et al., 1989; Singh et al., 2004; Watson and Carlson,

1987)(Table 3).

3.2.2. Methodological quality

There was 98% agreement in scoring between the researchers

conducting the systematic review. Disagreements were resolved

by discussion. Variation in methodological quality was noted (Ta-

ble 4), with scores ranging from two (Robaina et al., 1989) to 13

(Moseley, 2006; Oerlemans et al., 2000b) (mean = 10/16). The

majority of papers were good in quality (de Jong et al., 2005; Dur-

mus et al., 2004; Moseley, 2004, 2005; Oerlemans et al., 1999a,b;

Severens et al., 1999; Singh et al., 2004), two were poor (Robaina

et al., 1989; Watson and Carlson, 1987), two were fair (McCabe

et al., 2003; Pleger et al., 2005) and two were very good (Moseley,

2006; Oerlemans et al., 2000b). No papers were rated as excellent.

In general, the level II studies had higher quality scores than the le-

vel III study, which was higher in quality than the level IV studies

(mean = 12, 10, 8 points respectively).

Table 4 provides details regarding the criteria that were fulfilled

on the Critical Appraisal Form Quantitative studies (Law et al.,

1998a). It demonstrates that only half the studies provided ade-

quate detail to allow for reproduction of their interventions (crite-

rion 10). Only two studies reported the reliability of all their

measurement tools (criterion 9), one justified sample size (crite-

rion 6), and no studies were free from major biases (criterion 4)

or reported the validity of all their measurement tools (criterion 8).

3.2.3. CONSORT criteria 1, 8, 9, 10 and 19

Table 4 also provides details regarding the fulfillment of the

CONSORT criteria. None of the RCTs complied with item nine by

reporting the method used to implement the random allocation se-quence. OnlyMoseley (2004, 2005, 2006) complied with item 10

by reporting who generated the allocation sequence, enrolled par-

ticipants and assigned participants to their groups. No studies

complied with item 19 by reporting whether there were any ad-

verse events in any intervention group.

3.2.4. Types of intervention

In all studies physiotherapy was a major (Durmus et al., 2004;

Moseley, 2004, 2005, 2006, Oerlemans study;Pleger et al., 2005;

Singh et al., 2004) or sole (de Jong et al., 2005; McCabe et al.,

2003; Robaina et al., 1989; Watson and Carlson, 1987) source of

treatment. Details of the specific physiotherapy treatments evalu-

ated are found inTable 5. In summary, physiotherapy consisted of

Education (de Jong et al., 2005).

Graded exposure in vivo to activities, movements and light

touch (de Jong et al., 2005).

Sensorimotor treatment: pain adapted exercises and desensitisa-

tion activities (Pleger et al., 2005).

Exercise: stretching, active range of motion (Durmus et al., 2004),

water therapy (Singh et al., 2004), stress loading (Watson and

Carlson, 1987).

Mirror visual feedback: a mirror is placed between the affected

and unaffected limb, obscuring the affected limb and exercises

are performed while the participant looks at the image of the

unaffected limb (McCabe et al., 2003).

Graded motor imagery: a motor imagery program (MIP) where

the subject recognises the laterality of a photograph of a handor foot, followed by imagining moving their painful limb into

Review of title & abstract

Removal of duplicates

Ineligible if:

Not published in English language

Did not study Physiotherapy

Review article, letter or comment

Conference abstract or dissertation

Articles retrieved

n = 1320

Not eligible n = 1140Potentially eligible

n = 180

Review of full text

Ineligible if:

Did not describe diagnostic procedures

Physiotherapy was not a significant intervention

Subjects aged < 18 years

Eligible articles

n = 14

Individual studies

n = 11

Not eligible n = 166

Fig. 1. Retrieval and review process.



5/15

Table 4

Study quality using the Critical Review Form Quantitative studies (Law et al., 1998a,b) and the CONSORT statement ( Moher et al., 2001)

Criterion Critical

Review Form

de Jong et al.

(2005)

Durmus et al.

(2004)

McCabe et al.

(2003)

Moseley

(2004)

Moseley

(2005)

Moseley

(2006)

Oerlemans

studyaPleger et al.

(2005)

Robaina et a

(1989)

1 Purpose clearly stated 1 1 1 1 1 1 1 1 0

2 Literature review

relevant

1 1 1 1 1 1 1 1 0

3 Study design

appropriate to study

aims

1 1 1 1 1 1 1 1 0

4 No biases present 0 0 0 0 0 0 0 0 0

5 Sample described in

detail

1 1 1 1 1 1 1 0 0

6 Sample size justified 0 0 0 0 0 1 0 0 0

7 Informed consent

gained

1 1 0 1 1 1 1 1 0

8 Validity of outcome

measures used

0 0 0 0 0 0 0 0 0

9 Reliability of outcome

measures used

0 0 1 0 0 0 0 0 0

10 Intervention described

in detail

1 1 1 1 1 1 0 0 0

11 Statistical reporting of

results

1 1 0 1 1 1 1 1 0

12 Appropriate statistical

analysis

1 1 1 1 1 1 1 1 0

13 Clinical importance

reported

1 1 1 1 1 1 1 1 1

14 Appropriate

conclusions

1 1 1 1 1 1 1 1 0

15 Clinical implications

reported

1 1 1 1 1 1 1 1 1

16 Study limitations

acknowledged

1 1 0 1 1 1 1 0 0

Total 12 12 10 12 12 13 11 9 2

Quality categoryb Good Good Fair Good Good Very good Good Fair Poor

Criterion CONSORT statement

1 Study design stated in

title or abstract

U U U U U

8 Randomisation:

sequence generation

U U U U U

9 Randomisation:

allocation concealment

10 Randomisation:

implementation

U U U

19 Adverse events

a Oerlemans et al. (1999a,b), Severens et al. (1999); quality score = 11,Oerlemans et al. (2000b),criteria 6 and 9 also fulfilled, quality score = 13, U = criterion fulfilled, = cb Quality category: poor (score 6 8), fair (score = 910), good (score = 1112), very good (score = 1314) and excellent (score = 1516).


6/15


7/15


8/15

Table 5 (continued)

Author Participants Intervention Outcomes

n Sample characteristics Diagnostic criteria Treatment Control Outcome instr

Oerlemans

study1,2,3,4135 RSD UL< 1 year duration recruitment:

outpatient clinics, University Hospitals

in Nijmegan or Amsterdam, The

Netherlands, female = 95, mean age 53,

right UL 45%, mean duration 3.6

Months

Veldman et al.

(1993)

Physiotherapy or Occupational

Therapy with specific treatment

objectivese and fixed protocol medical

treatment

Social workf in

addition to fixed

protocol medical

treatment

Impairmen

ation of Per

Active rang

of all UL joi

Sensory los

Grip streng

Pain inten

pain at pre

effort, MPQ

Impairmen

score (ISS)3

Disability: M

RASQ3, Radb

Handicap:

(SIP)3,4

Cost effect

tion, differe

tal costs,

visits, trave

and work

tiveness4

1No signific

OT and cont

to permane

follow-up.

and to a less

value over Sp< 0.05, O

p< 0.10) a

increased j

(p< 0.05) b

months. Qu

the PT grou3Adjuvant P

extent OT (

relevant th

ment which

low-up. 4P

effective t

CRPS-1 com

(SW). Costs

treatment a

other medic

Pleger et al.

(2005)

6 CRPS-1of the whole hand, including all

digits, recruitment: BG-Kliniken,

Bergmannsheil, Bochum, Germany left

UL = 4, duration range 152 months

Bruehlet al. (1999) Pain adapted sensorimotor treatment

protocol (graded desensitisation and

motor tasks) and usual drug therapy

No Pain intens

now and ov

Immobility

points

Cortical m

(fMRI)

Peripheral

crimination

Graded sen

and 6 mont

intensity (p

tactile discr

regaining o

(p = 0.02) a


9/15


10/15

the position in the photograph. This is progressed to moving

both limbs into the position while observing the unaffected limb

in a mirror which obscures the affected limb (Moseley, 2004,

2005, 2006). Physical therapy: no description provided (Singh et al., 2004).

Pain management techniques including improving pain control,

optimizing coping, relaxation, connective tissue massage,

improving skills, extinguishing pain source, developing compen-

satory skills (Oerlemans study).

Transcutaneous electrical nerve stimulation (Robaina et al.,

1989).

Electromagnetic field treatment (Durmus et al., 2004).

3.2.5. Use of control groups

Control interventions varied between the Level II studies, and

included the

Maintenance of existing medical and physiotherapy manage-

ment (Moseley, 2004).

Implementation of standard medical and physiotherapy man-

agement including a home exercise program at the same inten-

sity as the experimental group (Moseley, 2006).

Use of the same drug and exercise program as the experimental

group and a sham form of electromagnetic field treatment (Dur-

mus et al., 2004).

Use of the same medical management as the experimental

group and added passive attention via social work sessions (Oer-

lemans study).

3.2.6. Outcome measures

There was great variability in outcome measurement across the

studies, in terms of the number and type of outcome measures

used and the number of occasions on which they were used

(Table 6). The majority of studies used outcome instruments that

addressed disorders of body structure or function (World Health

Organisation, 2001). Only five studies measured other constructssuch as activity limitation (de Jong et al., 2005; Moseley, 2005,

2006; Oerlemans et al., 2000b; Severens et al., 1999; Singh et al.,

2004), participation restriction (Singh et al., 2004), cost effective-

ness (Oerlemans et al., 2000b; Severens et al., 1999), quality of life

(Oerlemans et al., 2000b; Severens et al., 1999) or treatment

credibility and expectancy (de Jong et al., 2005). Pain was the most

frequently measured variable; however it was not evaluated in a

consistent manner across the studies. For example, the verbal rat-

ing scale, Neuropathic Pain Scale (NPS), the intensity score of the

NPS, the McGill Pain Questionnaire, the 11-point numeric rating

scale and various visual analogue scales were used to quantify a

change in pain following physiotherapy treatment. Moreover,

measurement periods were variable, ranging from two weeks

(Moseley, 2004) to three years (Robaina et al., 1989), but often rep-resented the immediate and longer term effects of physiotherapy.

3.3. Participants

Participant characteristics varied across the studies. Three stud-

ies included participants with both upper and lower limb CRPS-1

(de Jong et al., 2005; McCabe et al., 2003; Moseley, 2006), whereas

the remaining eight studies included participants with only upper

limb CRPS-1. Mean age ranges varied from 33 years (McCabe et al.,

2003) to 53 years (Oerlemans study), although three studies did

not provide this detail (de Jong et al., 2005; Moseley, 2006; Pleger

et al., 2005). Males and females were recruited in varying

proportions ranging from 100% female (de Jong et al., 2005) to

37% female (Watson and Carlson, 1987) (mean = 67%). Average

duration of symptoms varied from 3.6 months (Oerlemans study)

Table 6

Classification of outcomes according to the World Health Organisations 2001 International Classification of Functioning, Disability and Health (WHO, 2001)

Concept ICF 2001 terminology Domain identified in

CRPS-1 studies

Tools used in CRPS-1 studies

Disorder of body structure

or function

Abnormality of body

structure and function

Pain intensity and

quality

Visual analogue scale, 11 point rating scale, Verbal pain scale, Neuropathic pain scale,

McGill pain questionnaire, Diagnostic criteria

Weight bearing

tolerance

Weight scale and timer

Pressure tolerance Force gauge and timerSensation 2 point discriminator

Strength Jamar dynamometer

Range of motion Goniometer

Swelling Water displacement volumeter, Digital circumference tape measure

Dexterity Radboud dexterity test, JebsenTaylor test

Limb laterality

recognition

Response time

Autonomic changes Diagnostic criteria, Resting sweat output conductivitycell, Muscle blood flow blood

pressure cuff, Point temperature infrared thermography

Psychological and

behavioural attributes

CES-D scale, Pain locus of control, Pain cognitions questionnaire, Pain situations

questionnaire, Tampa scale for Kinesiophobia, Photograph series of daily activities

Other Three phase bone scan, Biochemical markers, Functional MRI, Impairment level

sumscore, AMA Evaluation of Permanent Impairment

Abnormality in function at

the level of the person

Activity limitation Radboud skills questionnaire, Walking stairs questionnaire, Rising and sitting down

questionnaire, Self rating on 11 point scale of present ability to complete tasks,

Modified Greentest, Baltimore therapeutic work simulator test

Inability to fulfill normalroles Participation restriction Goal achievement Diary, Visual analogue scale, Baltimore therapeutic work simulator

Cost effectiveness Differential cost

analysis

Hospital costs, patient diary of medical visits, travel, medications, home care and work

absence

Treatment

effectiveness

Impairment level sumscore, Modified Greentest, Sickness impact profile

Quality of life Sickness impact profile

Treatment credibility and

expectancy

Visual analogue scale

MRI, magnetic resonance imaging, AMA, American Medical Association, CES-D, Centre for Epidemiological Studies Depression.



11/15

to 36 months (de Jong et al., 2005) (mean = 11.4 months, n= 10

studies).

Sample sizes in the studies ranged from six (Pleger et al., 2005)

to 135 (Oerlemans study) (mean = 56). Subjects were recruited

from University hospitals and specialist clinics in all studies with

Moseley also recruiting participants from Physiotherapy depart-

ments and local General Practice clinics (Moseley, 2004, 2005,

2006). All studies listed or referenced their diagnostic criteria forCRPS-1 (Table 7). Most frequently the IASP criteria (Merskey and

Bogduk, 1994) or the proposed modified research diagnostic crite-

ria for CRPS-1 (Bruehl et al., 1999), were used(Durmus et al., 2004;

McCabe et al., 2003; Moseley, 2004, 2005, 2006; Pleger et al., 2005;

Singh et al., 2004).

3.4. Effectiveness of physiotherapy

Data gained from the RCTs could not be pooled because of the

heterogeneity of the outcome measures and comparison groups.

Tables 8a and 8billustrate the effect sizes of the studies included

in this systematic review. Effect sizes could not be calculated for

two case studies (Robaina et al., 1989; Singh et al., 2004) and

95% confidence intervals could not be calculated for a furthertwo studies (Watson and Carlson, 1987; Moseley, 2005).

Table 8a illustrates that graded motor imagery is more effec-

tive than usual physiotherapy plus medical management in

reducing pain, irrespective of the pain outcome measure used

(Moseley, 2004, 2006). A statistically significant reduction in pain

intensity occurred immediately after a six week program of

graded motor imagery (Moseley, 2004, 2006) and this effect was

maintained up to six months following treatment (Moseley,

2004, 2006). These results are clinically relevant, as pain was re-

duced on average by more than 20 points on a 100 point scale

(Farrar et al., 2001; Ferreira et al., 2002). Graded motor imagery

also reduced swelling in the fingers (Moseley, 2004), however it

is difficult to interpret the magnitude of effect size as a clinically

important change for this variable has not been determined. Theorder in which the components of graded motor imagery are pre-

sented affect the magnitude of pain reduction, with laterality rec-

ognition followed by imagined movements and mirror

movements producing the greatest pain reduction (Table 8b)

(Moseley, 2005). The Oerlemans study found statistically signifi-

cant changes in the impairment sum score, temperature differ-

ence between the hands, and pain (measured by the McGill Pain

Questionnaire), such that physiotherapy and medical manage-

ment was more effective than both occupational therapy andmedical management and social work and medical management.

However, it is likely that these changes were not clinically signif-

icant due to their small effect sizes.

The data contained in Table 8b illustrate that significant

changes between pre and post measurements were gained for

two types of physiotherapy. Sensorimotor treatment reduced pain

and increased tactile discrimination (Pleger et al., 2005) whereas a

graded exposure in vivo program significantly reduced pain related

fear and pain disability (de Jong et al., 2005). Mirror visual feed-

back appears to reduce the perception of pain in early CRPS-1

and stiffness in intermediate CRPS-1 (Table 8b) (McCabe et al.,

2003) and an interdisciplinary pain management program in-

creases function in CRPS-1 (Table 5) (Singh et al., 2004). There

was no evidence to support the effectiveness of transcutaneouselectrical nerve stimulation (Robaina et al., 1989) and stress load-

ing exercise (Watson and Carlson, 1987). The study results, as de-

scribed by the authors and the statistical precision of the effect (p

value) are listed inTable 5.

4. Discussion

4.1. Interpretation of the results

This is the first systematic review that has considered the qual-

ity of individual studies when synthesizing the results. The major

findings were

1. There is good to very good quality level II evidence that physio-therapy consisting of graded motor imagery, combined with

Table 7

Diagnostic criteria

Diagnostic set Criterion 1 Criterion 2 Criterion 3 Sensitivity Specificity

Watson and Carlson

(1987)

Pain out of proportion to

trauma

Swelling and vasomotor changes Stiffness

Robaina et al. (1989) Pain Sudomotor and vasomotor alterations

Veldman et al. (1993) 4 out of 5: Pain, Oedema

Altered skin colour Altered

skin temperature Reducedrange of motion

Signs and symptoms must be present in

area much larger than and distal to

primary injury

Signs and symptoms must be

aggravated by activity of the limb

0.65 0.85

Perez et al. (2005)

International

Association for the

Study of Pain criteria

isMerskey and

Bogduk (1994)

Continuing pain, allodynia

or hyperalgesia with which

the pain is disproportionate

to the inciting event

Evidence at sometime of oedema,

changes in skin blood flow or abnormal

sudomotor activity in the region of the

pain

This diagnosis is excluded by the

existence of conditions that would

otherwise account for the degree of

pain and dysfunction

0.98 0.36

Bruehl and Harden

(2001)

Proposed modified

research diagnostic

criteria for CRPS-1,

Bruehl et al. (1999)

Continuing pain that is

disproportionate to any

inciting event

Must report at least 1 symptom in each

of the 4 following categories:

Must display at least 1 sign in 2 or more

of the following categories:

0.70 0.96

Baron and Janig (2004)

Sensory: reports of hyperaesthesia Sensory: evidence of hyperalgesia (to

pinprick) and/or allodynia (to light

touch)

Vasomotor: reports of temperature

asymmetry and/or skin colour changes

and/or skin colour asymmetry

Vasomotor: evidence of temperature

asymmetry and/or skin colour changes

and/or skin colour asymmetry

Sudomotor/oedema: reports of oedema

and/or sweating changes and/or

sweating asymmetry

Sudomotor/oedema : evidence of oedema

and/or sweating changes and/or

sweating asymmetryMotor/trophic: Reports of decreased

range of motion and/or motor

dysfunction (weakness, tremor,

dystonia) and/or trophic changes (hair,

nails, skin)

Motor/trophic: Evidence of decreased

range of motion and/or motor

dysfunction (weakness, tremor,

dystonia) and/or trophic changes (hair,

nails, skin)



12/15

medical management is more effective than physiotherapy and

medical management for upper or lower limb CRPS-1 (Moseley,

2004, 2005, 2006). Graded motor imagery reduces pain inten-

sity by a clinically relevant amount, and this reduction is main-

tained for up to six months.

2. There is good quality level II evidence that pain management

physiotherapy combined with medical management is more

effective than occupational therapy or social work combined

with medical management in patients with upper limb CRPS-

1 (Oerlemans study). However, the clinical relevance of this

finding is questionable.

3. There is good quality level II evidence that electromagnetic field

treatment offers no added benefit to calcitonin and exercise in

upper limb CRPS-1 patients (Durmus et al., 2004).

In addition, there is good quality emerging evidence which sug-

gests that a graded exposure in vivo program (de Jong et al., 2005)

and a four week interdisciplinary outpatient treatment program

(Singh et al., 2004) may be effective for CRPS-1. There is fair quality

emerging evidence that a sensorimotor treatment program (Pleger

et al., 2005) and mirror visual feedback (McCabe et al., 2003) may

also be effective.

Table 8a

Effect of Physiotherapy treatments on outcomes measured on a continuous scale, produced from randomized controlled trials with intention to treat analyses

Physiotherapy vs comparison (outcome measure and possible range)

(reference time of follow-up)

Sample size Quality score Mean change difference (95% CI)

EMFT + C + Ex vs C+ Ex (pain intensity: 0100 mm VAS)

Durmus et al. (2004) end 6 weeks Rx (pain at rest) 40 12 12 [0 to 23]

Durmus et al. (2004) end 6 weeks Rx (pain with activity) 40 12 13 [1 to 26]

EMFT+ C + Ex vs C + Ex (swelling: mls)

Durmus et al. (2004) end 6 weeks Rx 40 12 6.5 [43.3 to 30.4]

GMI vs PT + MM (pain intensity: 0100 VAS)

Moseley (2006) end 6 weeks of Rx (av. pain over last 2 days) 51 13 12.9 [1.7 to 21.4]

Moseley (2006) 6 month follow-up (av. pain over last 2 days) 51 13 20.5 [8.2 to 32.8]*

GMI vs PT + MM (pain: NPS: 0100)

Moseley (2004) end 6 weeks of Rx 13 12 22 [10.1 to 29.9]*

Moseley (2004) 12 week follow-up 13 12 22 [13.4 to 30.6]*

GMI vs PT + MM (pain intensity: NPS item on intensity: 010)

Moseley (2004) end 6 weeks of Rx 13 12 3.0 [2.6 to 5.4]*

Moseley (2004) 12 week follow-up 13 12 3.0 [2.8 to 5.6]*

GMI vs PT + MM (finger circumference: mm)


Moseley (2004) 12 week follow-up 13 12 10.0 [2.6 to 17.3]*

GMI vs PT + MM (function: NRS: 010)


Moseley (2006) 6 month follow-up 51 13 2.2 [1.1 to 3.3]*

Rec + Im+ Mir vs Im + Rec + Im (pain: NPS: 0100)

Moseley (2005) end of 6 weeks of Rx 20 12 5 [7.0 to 11]a

Moseley (2005) 12 week follow-up 20 12 7 [3.0 to 17]a,*

Rec+ Im + Mir vs Rec + Mir+ Rec (pain intensity: NPS: 0100)

Moseley (2005) end of 6 weeks of Rx 20 12 10 [3 to 20]a,*

Moseley (2005) 12 week follow-up 20 12 18 [5 to 25]a,*

PT + MM vs OT + MM (impairment sum score: 050)

Oerlemans studyb 12 month follow-up 135 13 1.6 [1 to 2.2]*

PT + MM vs SW + MM (impairment sum score: 050)

Oerlemans studyb 12 month follow-up 135 13 5.1 [4.6 to 5.6]*

PT + MM vs OT + MM (temperature difference:C)


PT + MM vs SW + MM (temperature difference: C)


PT + MM vs OT + MM (pain intensity: 0100 VAS)

Oerlemans studyb 12 month follow-up (pain with effort) 135 13 5.2 [3.3 to 7.1]

PT + MM vs SW + MM (pain intensity: 0100 VAS)

Oerlemans studyb 12 month follow-up (pain with effort) 135 13 4.5 [10.1 to 19.1]

PT + MM vs OT + MM (pain: 078: McGill Pain Questionnaire)Oerlemans studyb 12 month follow-up 135 13 2.8 [1.4 to 4.2]

PT + MM vs SW+ MM (pain: 078: McGill Pain Questionnaire)


PT + MM vs OT + MM (volume difference between hands: mls)

Oerlemans studyb 12 month follow-up 135 13 4.1 [6.1 to 2.1]

PT + MM vs SW+ MM (volume difference between hands: mls)


PT + MM vs OT + MM (UL AROM: 05)

Oerlemans studyb 12 month follow-up 135 13 0.3 [0.6 to 0]

PT + MM vs SW + MM (UL AROM: 05)


Positive effect sizes favour a physiotherapy treatment effect.

EMFT, electromagnetic field treatment; C, calcitonin; Ex, exercise; VAS, visual analogue scale; Rx, treatment; av, average; mls, milliliters; GMI, graded motor imagery; PT,

physiotherapy; MM, medical management; NPS, neuropathic pain scale; mm, millimeters; NRS, numerical rating scale; OT, occupational therapy; SW, social work; UL, upper

limb; AROM, active range of motion; Rec, laterality recognition; Im, imagined movements; Mir, mirror movements.a Estimated from a graph.b Effect sizes could only be calculated from data presented in Oerlemans (2000).* Statistically significant.



13/15

There is no evidence of sufficient quality to draw conclusions on

the effectiveness of transcutaneous electrical nerve stimulation

(Robaina et al., 1989) or stress loading exercise (Watson and Carl-

son, 1987).

There was insufficient primary evidence on the same interven-

tion to allow grouping of results, which is a limitation of this re-

view. Although every study on this topic may not have been

sourced, this is unlikely, as the primary search strategies used eight

databases as well as extensive secondary searching. The reviewers

of the studies were not blind to author or source of the study and

this may have affected the methodological quality rating. However,

there was close agreement between the reviewers suggesting this

is an unlikely source of bias.The majority of studies included in this systematic review

(including all RCTs), investigated the effect of physiotherapy in

combination with medical management. This means that while

the added benefit of physiotherapy can be determined in a study

that compares usual treatment with usual treatment plus physio-

therapy (Durmus et al., 2004; McCabe et al., 2003; Moseley,

2004, 2005, 2006), the stand-alone value of physiotherapy cannot

be determined. In case series that involved usual medical manage-

ment plus physiotherapy, (Pleger et al., 2005; Singh et al., 2004),

the findings on physiotherapy effectiveness must be interpreted

with caution as there was no comparison group and these studies

were of lower methodological quality. Clinical guidelines support

interdisciplinary management (Harden et al., 2006c; Stanton-Hicks

et al., 2002) and the very nature of CRPS-1 requires that multiple

disciplines contribute to the management of this painful, disabling

and distressing condition. Therefore perhaps determining the ef-

fect of physiotherapy in isolation is not a helpful exercise.

4.2. Generalisability

Consideration must be given to the generalisability of the stud-

ies to current clinical practice, particularly with respect to partici-

pant characteristics. Participants were recruited from University

hospitals and specialist pain centres. These participants may have

differing characteristics to other populations, such as severity and

duration of symptoms, levels of depression and substance abuse

and recalcitrance to primary management (Covington, 1996).

Moseley excluded participants who had previously had a positiveresponse to sympathetic blockade (2004, 2005). Thus participants

may have had sympathetically independent pain, which is associ-

ated with a poorer outcome for CRPS-1.

With the exception ofMoseley (2006), all RCTs recruited partic-

ipants with only upper limb CRPS-1. Therefore their findings may

not be generalisable to patients with lower limb CRPS-1. On the

whole, participants were mostly female (66.8%), aged between 33

and 53 years and had CRPS-1 symptoms in their upper limb for

an average of 11.4 months following a noxious event, usually a

fracture. These characteristics reflect epidemiological data on

CRPS-1 (Janig and Stanton-Hicks, 1996; Low et al., 1996) but

should be considered when generalising the results of this review.

For example, the chronicity of participants in this review may inad-

equately represent the effect some treatments may have on acute

CRPS-1. Other reviews of mostly medical treatments for CRPS-1

Table 8b

Effect of Physiotherapy treatment on outcomes measured on a continuous scale, produced from other study designs

Outcome measure (reference time of follow-up) Sample size Quality score Mean change between pre and post treatment (95% CI)

Pain adapted sensorimotor treatment (pain: NRS 010)

Pleger et al. (2005) end of Rxa (pain over the last 4 weeks) 6 9 3.8 [1.2 to 6.4]*

Pain adapted sensorimotor treatment (% impaired tactile discrimination)

Pleger et al. (2005) end of Rxa (2 point discrimination threshold) 6 9 23.0 [6.8 to 39.2]*

Stress loading program (pain intensity: Verbal scale: 110 cm)

Watson and Carlson (1987) 16 month follow-up 52 4 5.0 [N/A]

Stress loading program (distance from finger pulp to palm crease: mm)

Watson and Carlson (1987) 16 month follow-up 52 4 2.0 [N/A]

Stress loading program (grip strength: kg)

Watson and Carlson (1987) 16 month follow-up 52 4 16 [N/A]

MVF+ usual Rx (pain intensity: VAS: 010)

McCabe et al. (2003) end 6 weeks of Rx 8 10 4.1 [1.5 to 6.7]

MVF+ usual Rx (temperature difference (painful vs non-painful)C)

McCabe et al. (2003) end 6 weeks of Rx 8 10 0.9 [0.2 to 1.6]

Graded exposure in vivo (TSK: 068)

de Jong et al. (2005) end of Rx 8 12 34.5 [19.9 to 49.1]*

de Jong et al. (2005) 6 month follow-up 8 12 35.0 [20.2 to 39.8]*

Graded exposure in vivo (PHODA: 0100)

de Jong et al. (2005) end of Rx 8 12 71.9 [68.3 to 75.5]*

de Jong et al. (2005) 6 month follow-up 8 12 73.1 [69.9 to 76.3]*

Rec (pain intensity: NPS: 0100)

Moseley (2005) 20 12 8 [N/A]*

Rec (function: NRS: 050)

Moseley (2005) 20 12 7 [N/A]*

Rec + Im (pain intensity: NPS: 0100)

Moseley (2005) 20 12 16 [N/A]*

Rec + Im (function: NRS: 050)

Moseley (2005) 20 12 15 [N/A]*

Im + Mir (pain intensity: NPS: 0100)

Moseley (2005) 20 12 14 [N/A]*

Im + Mir (function: NRS: 050)

Moseley (2005) 20 12 19 [N/A]*

Positive effect sizes favour a physiotherapy treatment effect.

NRS, numeric rating scale; RX, treatment; %, percentage; N/A, not available; MM, mirror movements; TSK, Tampa Scale for Kinesiophobia; PHODA, Photograph series of Daily

Activities; Rec, laterality recognition; Im, imagined movements; Mir, mirror movements; MVF, mirror visual feedback.a Lasting between 1 and 6 months.* Statistically significant.



14/15

have not reported participant characteristics (Forouzanfar et al.,

2002; Kingery, 1997; Perez et al., 2001).

4.3. Implications of variable diagnostic criteria

It is essential to consider diagnostic criteria to ensure that the

same condition is being evaluated in the studies contained in a sys-

tematic review (Perez et al., 2007). Comparison of the diagnostic

criteria reveals that important differences are present (Table 7).

The studies by Watson and Carlson (1987) and Robaina et al.

(1989)use basic signs and symptoms that are inadequate to diag-

nose CRPS-1. On this basis alone, little weight can be placed on the

results of their studies. Veldman criterion one (Veldman et al.,

1993), has moderate sensitivity (0.65) and good specificity (0.85)

(Perez et al., 2005). The IASP diagnostic criteria for CRPS-1 (Mers-

key and Bogduk, 1994) have high sensitivity (0.98), but low speci-

ficity (0.36) (Bruehl and Harden, 2001). These criteria have been

revised to provide diagnostic criteria for research purposes (Bruehl

et al., 1999). The proposed modified research diagnostic criteria for

CRPS-1 have lower sensitivity (0.70) but higher specificity (0.96)

(Baron and Janig, 2004).Moseley (2004, 2005, 2006) and de Jong

et al. (2005)used the research diagnostic criteria in their studies.

This suggests that they may have more accurately identified and

excluded participants without CRPS-1 than those studies using

the Veldman or IASP criteria.

4.4. Implications of variable outcome measures

The use of outcome measurement also requires consideration

when interpreting the results of this systematic review. The com-

parison of results between studies is limited due to the wide range

of measurement domains and tools that were used (Table 6). Addi-

tionally, the reliability and validity of outcome measures were

rarely reported in the studies. This information is required so that

the reader can determine the confidence they have in the measure-

ment of the study variables and therefore the results of the study

(Law et al., 1998b).Evidence of acceptable validity and reliability for neuropathic

pain conditions such as CRPS-1exists forthe Neuropathic Pain Scale

(Galer and Jensen, 1997; Jensen et al., 2005). The Impairment Level

Sum Score is reliable and valid for upper limb CRPS-1 (Oerlemans

et al., 1998) and lower limb CRPS-1 (Perez et al., 2003). The Rad-

boud Skills Questionnaire has acceptable validity and reliability

for upper limb CRPS-1 (Oerlemans et al., 2000a). The Walking Stairs

Questionnaire and the Questionnaire Rising and Sitting Down are

reliable for measuring activity limitation in lower limb CRPS-1 pa-

tients (Perez et al., 2002). Greater credence can be attributed to the

results of studies in which valid and reliable outcome measures for

CRPS-1 were used (de Jong et al., 2005; Moseley, 2004, 2005, 2006;

Oerlemans et al., 1999a,b, 2000b; Severens et al., 1999).

4.5. Implications for further research

Further studies would be strengthened by the use of the pro-

posed modified research diagnostic criteria for CRPS-1 (Bruehl

et al., 1999). The reporting of all CRPS-1 research would greatly

benefit from international and interdisciplinary consensus on out-

come measurement. This discussion should be informed by the

WHOs International Classification of Functioning, Disability and

Health (World Health Organisation, 2001). Critical appraisal and

interpretation of future RCTS would be enhanced by adoption of

the CONSORT statement checklist and flow diagram (Altman

et al., 2001). However, while RCTs provide the highest levels of evi-

dence, other forms of clinical investigation such as comparative

studies and case series can be designed to provide clinically rele-vant evidence that can later be tested by RCT. The findings of this

systematic review suggest that robust RCTs should be designed to

test the treatments for which there was good quality evidence

from case series, i.e. graded exposure in vivo (de Jong et al.,

2005) and interdisciplinary outpatient treatment programs (Singh

et al., 2004).

4.6. Implications for clinical guidelines

Existing CRPS clinical guidelines (Harden et al., 2006c;Stanton-

Hicks et al., 2002) should be interpreted with caution, as the rec-

ommendations for specific physiotherapy treatments are not based

on current evidence. The findings of this review would support the

inclusion of graded motor imagery (Moseley, 2004, 2005, 2006) for

which there is good to very good quality level II evidence from

three RCTs produced by one research group. These results need

to be replicated by others. Stress loading exercise (Watson and

Carlson, 1987), which is included in clinical guidelines on CRPS

(Harden et al., 2006c; Stanton-Hicks et al., 2002) has very little evi-

dence supporting its effectiveness. We recommend that clinical

guidelines should be evidence based.

The findings of this systematic review can be used to update

recommendations for the physiotherapy management of CRPS-1

to ensure these recommendations are clinically effective, relevant

and based on the best available evidence; to stimulate interdisci-

plinary discussion about physiotherapy and to inform future

research.

References

Altman D, Schulz K, Moher D. The revised CONSORT statement for reporting

randomized trials: explanation and elaboration. Ann Int Med 2001;134:66394.

Anderson K. Mosbys Medical, Nursing and Allied Health Dictionary. St. Louis: C.V.

Mosby; 2002.

Baron R. Complex regional pain syndromes. In: McMahonS, Koltzenburg M, editors.

Wall and Melzacks textbook of pain. 5th ed. Philadelphia: Elsevier/Churchill

Livingstone; 2005.

Baron R, Janig W. CRPS how do we escape the diagnostic trap? Lancet

2004;364:173941.

Barton S. Which clinical studies provide the best evidence? BMJ 2000;321:2556.

Berthelot J. Current management of Reflex Sympathetic Dystrophy syndrome

(complex regional pain syndrome type 1). Joint, Bone, Spine 2006;73:4959.

Bialocerkowski A, Kurlowicz K, Vladusic S, Grimmer K. Effectiveness of primary

conservative management for infants with obstetric brachial plexus palsy. Int J

Evidence Based Healthcare 2005;3:2744.

Birklein F. Complex regional pain syndrome. J Neurol 2005;252:1318.

Bruehl S, Harden R. An empirical approach to modifying IASP diagnostic criteria for

CRPS. In: Harden R, Baron R, Janig W, editors. Complex regional pain

syndrome. Seattle: IASP Press; 2001. p. 30313.

Bruehl S, Harden R, Galer B, Saltz S, Bertram M, Backonja M, et al. External

validation of IASP diagnostic criteria for Complex Regional Pain Syndrome and

proposed research diagnostic criteria. Pain 1999;81:14754.

Cepeda M, Lau J, Carr D. Defining the therapeutic role of local anesthetic

sympathetic blockade in complex regional pain syndrome: a narrative and

systematic review. Clin J Pain 2002;18:21633.

Covington E. Psychological issues in Reflex Sympathetic Dystrophy. In: Janig W,

Stanton-Hicks M, editors. Reflex Sympathetic Dystrophy: a

reappraisal. Seattle: IASP Press; 1996.de JongJ, Vlaeyen J, Onghena P, Cuypers C, denHollander M, Ruijgrok J. Reduction of

pain related fear in Complex Regional Pain Syndrome Type One: the application

of graded exposure in vivo. Pain 2005;116:26475.

DommerholtJ. Complex regional pain syndrome 2: Physical therapy management.

J Bodywork Movement Ther 2004;8:2418.

Durmus A, Cakmak A, Disci R, Muslumanoglu L. The efficiency of electromagnetic

field treatment in Complex Regional Pain Syndrome Type One. Disabil Rehab

2004;26:53745.

Farrar J, Young J, La Moreaux L, Werth J, Poole R. Clinical importance of changes in

chronic pain intensity measured on an 11 point numerical pain rating scale.

Pain 2001;94:14958.

Ferreira M, Ferreira P, Latimer J, Herbert R, Maher C. Does spinal manipulative

therapy help people with chronic low back pain? Aust J Physiother

2002;48:27784.

Flor H, Kalso E, Dostrovsky J, editors. Proceedings of the 11th world congress on

pain. The 11th world congress on pain. Sydney, Australia: IASP Press; 2006.

Fordyce W. Behavioural methods for chronic pain and illness. St

Louis: Mosby; 1976.

Forouzanfar T, Koke A, van Kleef M, Weber W. Treatment of Complex Regional PainSyndrome Type One. Eur J Pain 2002;6:10522.



15/15

Galer B, Jensen M. Development and preliminary validation of a pain measure

specific to neuropathic pain: the Neuropathic Pain Scale. Neurology

1997;48:3328.

Giamberardino MA, editor. Pain 2002 An updated review, refresher course

syllabus. Seattle: IASP Press; 2002.

Harden R, Baron R, Janig W, editors. Complex regional pain syndrome. Progress in

pain research and management. Seattle: IASP Press; 2001.

Harden N, Swan M, King A, Costa B, Barthel J. Treatment of CRPS: functional

restoration. Clin J Pain 2006a;22:4204.

Herbert R. How to estimate treatment effects from reports of clinical trials I:

Continuous outcomes. Aust J Physiother 2000;46:22935.Jadad A, Carroll D, Glynn C, McQuay H. Intravenous regional sympathetic blockade

for pain relief in Reflex Sympathetic Dystrophy: a systematic review and a

randomised double blind crossover study. J Pain Symptom Manage

1995;10:1320.

Janig W, Stanton-Hicks M, editors. Reflex Sympathetic Dystrophy: a reappraisal.

Progress in pain research and management. Seattle: IASP Press; 1996.

Jensen M, Dworkin R, Gammaitoni A, Olaleye D, Oleka N, Galer B. Assessment of

pain quality in chronic neuropathic and nociceptive pain clinical trials with the

Neuropathic Pain Scale. J Pain 2005;6:98106.

Justins DM, editor. Pain 2005 An updated review, refresher course

syllabus. Seattle: IASP Press; 2005.

Kingery W. A critical review of controlled clinical trials for peripheral neuropathic

pain and complex regional pain syndromes. Pain 1997;73:12339.

Low P, Wilson P, Sandroni P, Willner C, Chelimsky T. Clinical characteristics of

patients with Reflex Sympathetic Dystrophy (sympathetically maintained pain)

in the USA. In: Janig W, Stanton-Hicks M, editors. Reflex Sympathetic

Dystrophy: a reappraisal. Seattle: IASP Press; 1996. p. 966.

Main C, Keefe F, Rollman G. Psychological assessment and treatment of the pain

patient. In: Giamberardino MA, editor. Pain 2002 An Updated Review

Refresher Course Syllabus. Seattle: IASP Press; 2002. p. 281301.

McCabeC, Haigh R, RingE, HalliganP, Wall P,Blake D. A controlledpilot study of the

utility of mirror visual feedback in the treatment of Complex Regional Pain

Syndrome Type One. Rheumatology 2003;42:97101.

McMahon S, Koltzenburg M, editors. Wall and Melzacks Textbook of

Pain. Philadelphia: Elsevier/Churchill Livingstone; 2005.

Merskey H, BogdukN, editors. Classification of Chronic Pain, Descriptions of Chronic

Pain Syndromes and Definitions of Pain Terms. Seattle: IASP Press; 1994.

Moher D, Schulz K, Altman D. The CONSORT statement: revised recommendations

for improving the quality of reports of parallel-group randomized trials. Ann

Intern Med 2001;134:65762.

Moseley G. Graded motor imagery is effective for long standing Complex Regional

Pain Syndrome: A randomised controlled trial. Pain 2004;108:1928.

Moseley G. Is successful rehabilitation of Complex Regional Pain Syndrome due to

sustained attention to the affected limb? A randomised clinical trial. Pain

2005;114:5461.

Moseley G. Graded motor imagery for pathologic pain: A randomised controlled

trial. Neurology 2006;67:212934.NelsonD, StaceyB. Interventional therapiesin themanagement of complex regional

pain syndrome. Clin J Pain 2006;22:43842.

Norris S, Atkins D. Challenges in using nonrandomized studies in systematic

reviews of treatment interventions. Ann Int Med 2005;142:11129.

Oerlemans H, Goris R, Oostendorp R. Impairment level sumscore in Reflex

Sympathetic Dystrophy of one upper extremity. Arch Phys Med Rehabil

1998;79:97990.

OerlemansH, Goris R, de BooT, Oostendorp R. Do physical therapy andoccupational

therapy reduce the impairment percentage in Reflex Sympathetic Dystrophy?

Am J Phys Med Rehabil 1999a;78:5339.

Oerlemans H, Oostendorp R, de Boo T, Goris R. Pain and reduced mobility in

Complex Regional Pain Syndrome type one: Outcome of a prospective

randomised controlled trial of adjuvant physical therapy versus occupational

therapy. Pain 1999b;83:7783.

OerlemansH, CupE, BooT, Goris R, Oostendorp R. TheRadboud skillsquestionnaire:

construction and reliability in patients with Reflex Sympathetic Dystrophy of

one upper extremity. Disabil Rehabil 2000a;20:23345.

Oerlemans H, Oostendorp R, de Boo T, Laan L, Severens J, Goris R. Adjuvant physical

therapy versus occupational therapy in patients with Reflex SympatheticDystrophy/Complex Regional Pain Syndrome Type One. Arch Phys Med Rehabil

2000b;81:4956.

Perez R, Kwakkel G, Zuurmond W, de Lange J. Treatment of Reflex Sympathetic

Dystrophy (CRPS type 1): a research synthesis of 21 randomised clinical trials. J

Pain Symptom Manage 2001;21:51126.

Perez R, Roorda L, Zuurmond W, Bannin I, Vranken J, de Lange J. Measuring

perceived activity limitations in lower extremity Complex Regional Pain

Syndrome type 1 (CRPS 1): testretest reliability of two questionnaires. Clin

Rehabil 2002;16:45460.

Perez R, Oerlemans HM, Zuurmond W, de Lange J. Impairment level SumScore for

lower extremity Complex Regional Pain Syndrome type I. Disabil Rehabil

2003;25:98491.

Perez R, Keijzer C, Bezemer P, Zuurmond W, de Lange J. Predictive value of symptom

level measurements for complex regional pain syndrome type 1. Eur J Pain

2005;9:4956.

Perez R, Collins S, Marinus J, Zuurmond W, de Lange J. Diagnostic criteria for CRPS-

1: Differences between patient profiles using three different diagnostic sets. Eur

J Pain 2007;11:895902.

Petticrew M. Why certain systematic reviews reach uncertain conclusions. BMJ

2003;326:7568.

Pleger B, Tegenthoff M, Ragert P, Forster A, Dinse H, Schweinkreis P, et al.

Sensorimotor returning in Complex Regional Pain Syndrome parallels pain

reduction. Ann Neurol 2005;57:4259.

Quisel A, Gill J, Witherell P. Complex regional pain syndrome: which treatments

show promise? J Fam Pract 2005;54:599603.

Robaina F, Rodriguez J, de Vera J, Martin M. Transcutaneous electrical nervestimulation and spinal cord stimulation for pain relief in Reflex Sympathetic

Dystrophy. Stereotact Funct Neurosurg 1989;52:5362.

Severens J, Oerlemans H, Wessels A, vant Hof M, Oostendorp R, Goris R.

Cost effective analysis of adjuvant physical or occupational therapy for

patients with Reflex Sympathetic Dystrophy. Arch Phys Med Rehabil 1999;80:

103843.

Sharma A, Williams K, Raja S. Advances in treatment of complex regional pain

syndrome: recent insights on a perplexing disease. Curr Opin Anaesthesiol

2006;19:56672.

Singh G, Willen S, Boswell M, Janta J, Chelimsky T. The value of interdisciplinary

pain management in Complex Regional Pain Syndrome Type One: A prospective

outcome study. Pain Physician 2004;7:2039.

Stanton-Hicks M. Complex regional pain syndrome: manifestations and the role of

neurostimulationin its management. J Pain Symptom Manage 2006;31:S204.

Stanton-Hicks M, Burton A, Bruehl S, Carr D, Harden N, Hassenbusch S, et al. An

updated interdisciplinary clinical pathway for CRPS: Report of an expert panel.

Pain Practice 2002;2:116.

Stone P. Popping the (PICO) question in research and evidence based practice. Appl

Nurs Res 2002;16:1978.

Turk D, Meichenbaum D, Genest M. Pain and behavioural medicine: theory,

research and clinical guide. New York: Guilford; 1983.

Veldman P, Reynen H, Goris R. Signs and symptoms of Reflex Sympathetic

Dystrophy: prospective study of 829 patients. Lancet 1993;342:10126.

Vlaeyen J, de Jong J, Geilen M, Heuts P, van Breukelen G. Graded exposure in vivo in

the treatment of pain-related fear: a replicated single case experimental

design in four patients with chronic low back pain. Behav Res Ther

2001;39:15166.

Vlaeyen J, de Jong J, Geilen M, Heuts P, van Breukelen G. The treatment of fear of

movement/reinjury in chronic low back pain: further evidence on the

effectiveness of exposure in vivo. Clin J Pain 2002;18:25161.

Wasner G, Schattschneider J, Binder A, Baron R. Complex regional pain syndrome

diagnostic, mechanisms, CNS involvement and therapy. Spinal Cord

2003;41:6175.

Watson H, Carlson L. Treatment of Reflex Sympathetic Dystrophy of the hand with

an active stress loading program. J Hand Surg (Am) 1987;12A5:77985.

Wilson P, Stanton-Hicks M, HardenR, editors. CRPS: Current Diagnosis andTherapy.

Progress in Pain Research and Management. Seattle: IASP Press; 2005.

Web references

Australian National Health and Medical Research Council (1999: accessed 2007).

How to review the evidence: systematic identification and review of the

scientific literature. From: www.nhmrc.gov.au/publications/synopses/_files/

cp65.pdf.

Australian Physiotherapy Association (2004: accessed 2007). APA Position

Statement. From: www.apa.advsol.com.au/staticcontent/staticpages/

position_statements/public/RoleDefinition.pdf.

Centre for Evidence Based Physiotherapy (accessed 2007). Physiotherapy Evidence

Database. From:www.pedro.fhs.usyd.edu.au.

Harden R, Bruehl S, Burton A, Swan M, Costa B, Barthel J and King A (2006b:

accessed 2007). Complex Regional Pain Syndrome: Treatment Guidelines. From:

www.guideline.gov/summary/summary.aspx.

Harden R, Swan M, Costa B, Barthel J and King A (2006c: accessed 2007). CRPS

Treatment Guidelines: Interdisciplinary Management. From: www.rsds.org/3/

clinical_guidelines/index.Joanna Briggs Institute (accessed 2007). Clinical On-Line Network of Evidence for

Care and Therapeutics. Fromwww.jbiconnect.org/.

LawM, Stewart D, Letts L, Pollock N, Bosch J and Westmoreland M (1998a: accessed

2007). Critical Review Form Quantitative Studies. From:

www.fhs.mcmaster.ca/rehab/ebp/pdf/qualreview.pdf.

LawM, Stewart D, Pollock N, Letts L, Bosch J andWestmoreland M (1998b: accessed

2007). Guidelines for critical review form quantitative studies. From:

www.fhs.mcmaster.ca/rehab/ebp/pdf/qualguidelines.pdf.

The Cochrane Collaboration (accessed 2007). The Cochrane Library. From:

www.cochrane.org/reviews/clibintro.htm.

Turning Research Into Practice Database (accessed 2007). TRIP Database, From:

www.tripdatabase.com.

World Health Organisation (2001: accessed 2007). International Classification of

Functioning, Disability and Health. From:www.who.int/classifications/icf/site/

icftemplate.cfm.

http://www.nhmrc.gov.au/publications/synopses/_files/cp65.pdfhttp://www.nhmrc.gov.au/publications/synopses/_files/cp65.pdfhttp://www.apa.advsol.com.au/staticcontent/staticpages/position_statements/public/RoleDefinition.pdfhttp://www.apa.advsol.com.au/staticcontent/staticpages/position_statements/public/RoleDefinition.pdfhttp://www.pedro.fhs.usyd.edu.au/http://www.pedro.fhs.usyd.edu.au/http://www.guideline.gov/summary/summary.aspxhttp://www.rsds.org/3/clinical_guidelines/indexhttp://www.rsds.org/3/clinical_guidelines/indexhttp://www.jbiconnect.org/http://www.fhs.mcmaster.ca/rehab/ebp/pdf/qualreview.pdfhttp://www.fhs.mcmaster.ca/rehab/ebp/pdf/qualguidelines.pdfhttp://www.cochrane.org/reviews/clibintro.htmhttp://www.tripdatabase.com/http://www.who.int/classifications/icf/site/icftemplate.cfmhttp://www.who.int/classifications/icf/site/icftemplate.cfmhttp://www.who.int/classifications/icf/site/icftemplate.cfmhttp://www.who.int/classifications/icf/site/icftemplate.cfmhttp://www.tripdatabase.com/http://www.cochrane.org/reviews/clibintro.htmhttp://www.fhs.mcmaster.ca/rehab/ebp/pdf/qualguidelines.pdfhttp://www.fhs.mcmaster.ca/rehab/ebp/pdf/qualreview.pdfhttp://www.jbiconnect.org/http://www.rsds.org/3/clinical_guidelines/indexhttp://www.rsds.org/3/clinical_guidelines/indexhttp://www.guideline.gov/summary/summary.aspxhttp://www.pedro.fhs.usyd.edu.au/http://www.apa.advsol.com.au/staticcontent/staticpages/position_statements/public/RoleDefinition.pdfhttp://www.apa.advsol.com.au/staticcontent/staticpages/position_statements/public/RoleDefinition.pdfhttp://www.nhmrc.gov.au/publications/synopses/_files/cp65.pdfhttp://www.nhmrc.gov.au/publications/synopses/_files/cp65.pdf

does evidence support physiotherapy management of adult complex regional pain syndrome type one a...

Documents