Clinical Rehabilitation

http://cre.sagepub.com Constraint-induced movement therapy in the treatment of the upper limb in children with hemiplegic cerebral palsy: a Cochrane systematic review
Brian Hoare, Christine Imms, Leeanne Carey and Jason Wasiak Clin Rehabil 2007; 21; 675 DOI: 10.1177/0269215507080783 The online version of this article can be found at: http://cre.sagepub.com/cgi/content/abstract/21/8/675

Published by:
http://www.sagepublications.com

Additional services and information for Clinical Rehabilitation can be found at: Email Alerts: http://cre.sagepub.com/cgi/alerts Subscriptions: http://cre.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.co.uk/journalsPermissions.nav Citations http://cre.sagepub.com/cgi/content/refs/21/8/675

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

Clinical Rehabilitation 2007; 21: 675685

Constraint-induced movement therapy in the treatment of the upper limb in children with hemiplegic cerebral palsy: a Cochrane systematic review
Brian Hoare Victorian Paediatric Rehabilitation Service, Monash Medical Centre, Christine Imms Royal Childrens Hospital, Melbourne, Murdoch Childrens Research Institute and School of Occupational Therapy, La Trobe University, Melbourne, Leeanne Carey Division of Neurological Rehabilitation and Recovery, National Stroke Research Institute and School of Occupational Therapy, La Trobe University, Melbourne and Jason Wasiak Burns Unit, The Alfred Hospital, Victoria, Australia Received 17th April 2007; manuscript accepted 18th April 2007.

Background: Constraint-induced movement therapy (CIMT) is emerging as a treatment approach for children with hemiplegic cerebral palsy. It aims to increase spontaneous use of the affected upper limb and limit the effects of learned non-use. This review evaluates the effectiveness of CIMT, modified CIMT or Forced use in the treatment of children with hemiplegic cerebral palsy. Design and methods: Systematic Cochrane Review. The Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2006, Issue 3), MEDLINE (1966 to August Week 4 2006), CINAHL (1982 to July Week 3 2006), EMBASE (1980 to August 2006), PsychInfo (1985 to August Week 4 2006) and reference lists of all relevant articles were searched. Relevant randomized and controlled clinical trials were systematically reviewed. Results: Three studies met the inclusion criteria. One randomized controlled trial (RCT) showed a trend for positive treatment effect favouring CIMT using the Dissociated Movement subscale of the Quality of Upper Extremity Skills Test. A clinically controlled trial demonstrated a significant treatment effect favouring modified CIMT using the Assisting Hand Assessment at two and six months. Another inconsistently reported trial showed a significant treatment effect at six weeks on the self-care component of the WeeFIM using Forced use. Reviewers conclusions: Given the limited evidence, the use of CIMT, modified CIMT and Forced use should be considered experimental in children with hemiplegic cerebral palsy. Further research using adequately powered RCTs, rigorous methodology and valid, reliable outcome measures is essential to provide higher level support of the effectiveness of CIMT for children with hemiplegic cerebral palsy.

Background
Cerebral palsy and developmental disregard Hemiplegic cerebral palsy, characterized by a clinical pattern of unilateral motor impairment, is a common type of cerebral palsy, accounting for 35.1% of all cerebral palsy in Victoria, Australia,1 15.3% in Ontario, Canada, 40% in
10.1177/0269215507080783

Address for correspondence: Brian Hoare, Victorian Paediatric Rehabilitation Service, Monash Medical Centre, 246 Clayton Road, Clayton, Victoria 3168, Australia. e-mail: brian.hoare@southernhealth.org.au SAGE Publications 2007 Los Angeles, London, New Delhi and Singapore

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

676

B Hoare et al. reward for small improvements in task performance; (b) selecting tasks that are tailored to address the motor deficits of the individual patient; (c) helping the patient to carry out parts of a movement sequence, if they are incapable of completing the movement on their own at first; and (d) systematically increasing the difficulty of the task performed.10 Two different but linked mechanisms are considered to be responsible for increased use of the more affected extremity as a result of CIMT: overcoming learned non-use and inducing use-dependent cortical re-organization.10 CIMT attempts to change the contingencies of behavioural reinforcement so that the learned non-use of the more affected upper limb is counter-conditioned or lifted.10 The consequent increase in the use of the more affected limb is argued to induce expansion of the contralateral cortical area controlling movement of the more affected limb and to recruit new ipsilateral areas.10 It is proposed that this may serve as the neural basis for the permanent increase in use of the affected limb following treatment.10 Despite the conceptual difference in behavioural adaptation (learned non-use versus developmental disregard) as a result of an acquired condition (e.g. stroke) compared with a congenital condition (e.g. cerebral palsy), the core components of a CIMT programme including restraint and massed practice, aim to reverse the behavioural suppression of movement in the affected upper limb. As a result, CIMT is emerging as a treatment approach for children with cerebral palsy and it raises significant clinical practice issues for occupational therapists. Traditionally, therapists working with children with hemiplegia have attempted to encourage normal movement of the affected hand by utilizing bimanual tasks or voluntary repetitive practice of unilateral activities. In order to engage in these activities, often verbal and physical prompting is required to encourage repeated practice with the affected hand. Such persistent prompting can become frustrating for a child when movements are clumsy and effortful and often do not result in a successfully completed task, as perceived by the child. Negative behaviours can occur in response to this persistent prompting. Forced use of the affected limb through a CIMT programme however, removes

Sweden2 and 31.2% in northern England, UK.3 During growth and development, children with hemiplegic cerebral palsy learn strategies and techniques to manage daily tasks (e.g. play) with one hand. Performance of tasks is frequently discovered to be more efficient and effective using the non-affected hand, even if there is only mild impairment in the affected limb.4 Recently, DeLuca5 introduced the term developmental disregard to describe a child with hemiplegia who may disregard, or learn not to use, the affected limb during the development of motor function. This developmental disregard is akin to the learned non-use reported in the literature on adults who have sustained a stroke.6 Despite the similar behavioural mechanisms of reinforcement of the unaffected hand and suppression of use of the affected hand, as identified in adults, Eliasson et al. suggested that the learned non-use may be a different phenomenon in children who sustain an early brain lesion.7 Unlike an adult who has had a neurological insult later in life, a child with hemiplegia has not had the experience of normal motor function of the limb. There is not the potential to unmask motor function, as frequently described in adults. Therapy must, therefore, create the opportunity, experience and environment in which a child can learn how to use their affected limb. This experience must reverse the behavioural aspect of suppression of use of the affected limb and reward use of that limb in even the simplest tasks, such as stabilization of an object. Constraint-induced movement therapy is proposed as a method of achieving this.5

Constraint-induced movement therapy (CIMT) CIMT has its foundation in behavioural research with non-human primates, conducted by Taub and co-workers.8 Taub suggests that CIMT for the upper limb involves inducing use of the more-affected limb by using a sling or half glove to prevent use of the unaffected limb for a target of 90% of waking hours over a period of 23 weeks.6 During this time, concentrated, repetitive training of the more affected limb is provided daily for 6 hours using a technique to shape motor behaviour.9 The shaping procedure involves (a) providing explicit verbal feedback and verbal

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

Constraint-induced movement therapy for the upper limb the need for persistent prompting and may be a more effective way of improving outcomes. It may, on the other hand, lead to the opposite effect if a child demonstrates adverse behavioural responses to the restraint. Objectives To evaluate the effectiveness of CIMT, modified CIMT (mCIMT) or Forced use in the treatment of the affected upper limb in children with hemiplegic cerebral palsy as measured by objective outcomes of upper limb function.
Table 1 Sample search on MEDLINE

677

MEDLINE (Ovid) 1. Constraint adj3 therapy.tw 2. CIMT.tw 3. CI therapy.tw 4. Forced.tw 5. Massed practice.tw 6. or/15 7. Cerebral palsy 8. Cerebral pals$.tw 9. Hemiplegia/ 10. Hemiplegi$.tw 11. or/710 12. 6 and 11

Materials and methods

The systematic method used for this review was derived from the Cochrane Handbook for Systematic Reviews of Interventions.11 This review is a summarized version of a more extensively reported Cochrane Systematic Review.12 Literature search The literature was searched to identify both published and unpublished research using the following databases: Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2006, Issue 3), MEDLINE (1966 to August Week 3 2006), EMBASE (1980 to August 2006), CINAHL (1982 to August Week 3 2006) and PsychInfo (1985 to August Week 4 2006). In addition, reference lists of articles and conference abstracts were examined. No language restrictions were applied. Individual search strategies were developed and these were adapted for each database (see Table 1 for sample search). Selection criteria All randomized controlled trials (RCTs) and controlled clinical trials (CCTs) that compared the effect of any form of CIMT with traditional treatments (that is occupational therapy or physiotherapy) or no treatment were considered for inclusion in this review. Trials were included if participants were diagnosed with hemiplegic cerebral palsy, aged between 0 and 19 years and they received treatment of the affected upper limb. Outcomes included any objective measure of the upper limb. This could include domains of body function/structure, activity or participation, as defined by the International Classification of Functioning, Disability and Health (ICF).13 Studies that used any treatment approach that provided massed practice of movement to the affected upper limb (that is behavioural shaping or motor learning principles) when a restraint device was worn were included and categorized according to the following definitions as described by Taub (personal communication, 14 November 2002).  Constraint-induced movement therapy: Restraint of the unaffected upper limb is applied and more than 3 hours of therapy per day (massed practice) is provided for at least two consecutive weeks.  Modified constraint-induced movement therapy (mCIMT): Restraint of the unaffected upper limb is applied and less than 3 hours per day of therapy is provided to the affected limb.  Forced use: Restraint of the unaffected upper limb is applied but no additional treatment of the affected upper limb is provided. Due to the vastly different intensity of treatment provided within each type of therapy these terms are not interchangeable and studies investigating each treatment type were therefore reviewed separately.

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

678

B Hoare et al. identified 26 potentially relevant articles. Of the 26 assessed in full text form, 23 were excluded because they were abstracts, case reports, case series, narrative reviews or included children with diagnoses other than cerebral palsy. The remaining three trials met the inclusion criteria and formed the basis of this review.5,17,18

Evaluation of methodological quality Two authors independently reviewed titles and abstracts of articles retrieved following the search. Methodological quality was assessed using an adaptation of the method outlined in Schultz et al.14 This involved independent judgements of adequacy of randomization and allocation concealment processes, equivalence of participant characteristics at baseline, blinding of assessors and description of withdrawals and drop-outs. In addition, validity and reliability of outcome measures were evaluated for each study.

Data analysis Data from eligible studies were extracted using a data extraction tool and summarized independently by two reviewers. The authors of primary studies were contacted to provide information when missing or incomplete data were encountered. A combined analysis using pooled data was not appropriate due to the small number of studies located, inconsistency in outcome measures used and the variations in intensity of treatment. The reviewers followed the Cochrane preferred method for handling continuous variables.15 This approach assessed mean change scores (and standard deviation of the change) from baseline to time points following intervention, thus allowing for differences in baseline performance. In order to obtain the standard deviation of the mean difference authors were contacted where necessary. This method of analysis was used only for primary outcomes within studies. For all other outcomes, calculations and comparisons were based on mean differences between groups at postintervention and follow-up. All data were analysed by the review authors using RevMan16 and the standard mean difference (SMD) and 95% confidence intervals (95% CIs) of each outcome were calculated.

Results
Literature search A total of 214 references were identified. Independent scrutiny of the titles and abstracts

Study characteristics Characteristics of included studies are described in Table 2. The RCT by DeLuca5 (n 18) compared CIMT with traditional services. The restraint was a bivalved fibreglass cast on the less involved extremity from the upper arm to the fingertips. The protocol included the provision of occupational therapy and physiotherapy for 6 hours per day for 21 consecutive days using operant training (shaping). The trial was reported in an unpublished dissertation5 from the University of Alabama and later published in 2004.19 Although the trial was reported as an RCT with a crossover design,5 data from only one group (CIMT) were reported in the crossover period and the publication19 omitted both the crossover component of the trial and the Quality of Upper Extremity Skills Test (QUEST) data. As a result, the trial was reviewed as an RCT without the crossover component and results from both the unpublished dissertation and the publication are discussed in this review. In the largest trial (n 45), Eliasson et al.17 used a CCT design to compare a mCIMT protocol with traditional services. The restraint was a fabric glove with a built-in stiff volar plastic splint worn on the unaffected hand for 2 hours a day, 7 days a week for two months. The mCIMT was provided in the childs usual environment (home/ preschool). Parents and/or preschool teachers were responsible for accomplishing the intervention on a daily basis with occupational therapist supervision once weekly. Although recruitment was undertaken by open invitation to parents, and children were not randomly allocated to groups, matching of children for age and degree of hand impairment ensured baseline characteristics of children were similar on these variables. In 2005, Sung et al.18 (n 31) compared Forced use with conventional therapy services.

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

Table 2 DeLuca (2002)5 CIMT N 18 9 treatment, 9 control 13 male, 5 female Aged 7 months to 8 years Diagnosed with asymmetric involvement of the upper extremities All levels of impaired hand function All levels of impaired hand function Baseline, 2 and 6 months AHA Modified CIMT N 45 21 treatment, 20 control 20 male, 21 female Aged 18 to 51 months Hemiplegic cerebral palsy Eliasson et al. (2005)17

Sample and outcome characteristics of included studies Sung et al. (2005)18 Forced use N 31 18 treatment, 13 control 15 male, 16 female Aged less than 8 years Hemiplegic cerebral palsy

Variable

Intervention Sample

Assessment frequency

Children with severely impaired hand function were excluded Baseline and 6 weeks Box and Blocks Test, EDPA, WeeFIM

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

Outcome measures

Baseline, 3 and 6 weeks, 3 and 6 months (CIMT group only) QUEST Dissociated Movement subscale, CAUT, PMAL, EBS

Constraint-induced movement therapy for the upper limb 679

CIMT, constraint-induced movement therapy; QUEST, Quality of Upper Extremity Skills Test; CAUT, Child Arm Use Test; PMAL, Pediatric Motor Activity Log; EBS, Emerging Behaviours Scale; AHA, Assisting Hand Assessment; EDPA, Erhardt Developmental Prehension Assessment; WeeFIM, Paediatric Functional Independence Measure.

680

B Hoare et al. remained significant at six weeks (SMD 1.30, 95% CI 0.26 to 2.34), however the Quality of Use subscale did not show a significant treatment effect at post treatment (SMD 0.72, 95% CI 0.24 to 1.69) or at the six-week follow-up (SMD 0.67, 95% CI 0.29 to 1.63). The blindly rated measure, the Toddler Arm Use Test, did not show significant difference between groups as demonstrated in the ANCOVA analysis (F 1.42, P 0.25) reported in the dissertation. Further analyses reported in the publication19 found significant betweengroup differences after treatment in percentage of items for which children use their more affected arm and overall quality of use. Analysis of Eliasson et al.17 data demonstrated a significant treatment effect at two months (SMD 1.12, 95% CI 0.46 to 1.78) and at six months (SMD 0.74, 95% CI 0.10 to 1.37) based on change scores of the Assisting Hand Assessment (Figure 2). These outcomes supported the conclusion that over a six-month period mCIMT induced a rate of improvement in bimanual performance using the hemiplegic hand that was greater than with traditional services. The Sung et al.18 trial demonstrated a significant treatment effect at six weeks on one outcome measure, the self-care component of the Functional Independence Measure for Children (WeeFIM) (SMD 1.25, 95% CI 0.46 to 2.03) (Figure 3). At six weeks, all other measures including the Erhardt Developmental Prehension Assessment (SMD 0.57 95% CI 0.16 to 1.30), Box and Blocks Test (SMD 0.67 95% CI 0.07 to 1.40), motor component WeeFIM (SMD 0.40 95% CI 0.32 to 1.12), WeeFIM cognitive component (SMD 0.54 95% CI 0.19 to 1.27) and WeeFIM total score (SMD 0.54 95% CI 0.19 to 1.27) demonstrated a trend favouring Forced use but no significant treatment effects.

The restraint involved the application of a shortarm cast from below the elbow to the fingertips on the unaffected upper limb. Occupational therapy was provided twice weekly in 30-minute sessions for six weeks for both groups. Although the trial was reported as a prospective case series design in the abstract, the text reported random assignment of children to treatment or control group. This provided ambiguity regarding the true methodology of the trial. Due to the random assignment of children and a two-group design, the study was viewed as an RCT for the purposes of this review.

Methodological quality Comparative data on design characteristics impacting on methodological quality of each study is provided in Table 3. Of note was the lack of evidence related to psychometric properties for a number of outcomes within the DeLuca5 trial including the Pediatric Motor Activity Log, Emerging Behaviours Scale and the Toddler Arm Use Test. Each of these tools was developed specifically for this trial and their psychometric properties have not been reported. Sung et al.18 did not report on randomization or allocation concealment processes or on the level of masking, suggesting poor methodological quality.14

Findings from review analysis Analysis of DeLucas5 change scores from baseline level of performance to immediately post treatment at three weeks did not demonstrate a significant treatment effect on the QUEST Dissociated Movement subscale (SMD 0.91, 95% CI 0.08 to 1.89) (Figure 1). However, a trend favouring the CIMT group was evident. QUEST data were not reported at the six-week, threemonth or six-month follow-up. Results from the Pediatric Motor Activity Log Frequency of Use subscale (SMD 1.53, 95% CI 0.45 to 2.62) and Emerging Behaviors Scale (SMD 1.22, 95% CI 0.19 to 2.24), both parent-rated questionnairebased measures, indicated a significant treatment effect immediately after treatment (at three weeks). The treatment effect for the Pediatric Motor Activity Log Frequency of Use subscale

Discussion
Constraint-induced movement therapy (CIMT), mCIMT and Forced use are emerging as treatment approaches for children with hemiplegic cerebral palsy. The purpose of this review was to evaluate the effectiveness of these interventions using the highest-level evidence available. Three trials, with

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

Table 3 Eliasson et al. (2005)17 CCT NA Not used Four children from treatment group No statistically significant difference between groups for age and degree of hand function (using AHA) Validity AHA: yes20,25 Reliability AHA: yes25,26 Unclear RCT Not described Sung et al. (2005)18

Design characteristics impacting the quality of included studies

DeLuca (2002)5

Design Randomization

Allocation concealment Drop-outs Baseline characteristics

Single blind RCT Folding a piece of paper, taped closed and drawn from a jar set up before the beginning of the participant enrolment Unclear

Outcomes

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

Blinding of assessors

None No statistically significant differences between groups for age, gender, severity of hemiparesis, prior treatment history or baseline scores on outcome measures Validity Reliability QUEST: yes23,24 QUEST: yes23 CAUT: no CAUT: no PMAL: no PMAL: no EBS: no EBS: no CAUT only Yes

Not reported No statistically significant differences between groups for age, gender and baseline scores on outcome measures Validity Reliability Box & Blocks: Box & Blocks: yes30 yes27 EDPA: no EDPA: no WeeFIM: yes31,32 WeeFIM: yes28,19 Not reported

Constraint-induced movement therapy for the upper limb 681

RCT, randomized controlled trial; CCT, controlled clinical trial; QUEST, Quality of Upper Extremity Skills Test; CAUT, Child Arm Use Test; PMAL, Pediatric Motor Activity Log; EBS, Emerging Behaviours Scale; AHA, Assisting Hand Assessment; EDPA, Erhardt Developmental Prehension Assessment; WeeFIM, Functional Independence Measure for Children.

682

B Hoare et al.
Constraint induced movement therapy in the treatment of the upper limb in children with hemiplegic cerebral palsy 01 CIMT versus traditional services 01 QUEST "change" score from baseline to immediately post treatment (3 weeks)
N Treatment mean (SD) N Control mean (SD) 1.74(16.02) SMD (random) 95%CI Weight % SMD (random) 95%CI 0.91 [0.08, 1.89] 0.91 [0.08, 1.89]

Review: Comparison: Outcome:

Study or sub-category

01 Post-treatment (0 to 3 weeks "change" score) DeLuca 2002 9 14.23(17.48) 9 Subtotal (95%CI) 9 9 Test for heterogeneity: not applicable Test for overall effect: Z =1.81 (P=0.07)

100.00 100.00

10

10

Favours control

Favours CIMT

Figure 1

QUEST (dissociated movement) change scores from baseline to three weeks.

Constraint induced movement therapy in the treatment of the upper limb in children with hemiplegic cerebral palsy 02 mCIMT versus traditional services 01 Assisting hand assessment "change" score from baseline to 2 & 6 months
N CIMT mean (SD) N Traditional mean (SD) 0.24(0.65) SMD (random) 95%CI Weight % 100.00 100.00 SMD (random) 95%CI 1.12 [0.46, 1.78] 1.12 [0.10, 1.37]

Review: Comparison: Outcome:

Study or sub-category

03 Post-Treatment (0 to 2 months "change" score) Eliasson 2005 21 1.23(1.03) 20 Subtotal (95% CI) 21 20 Test for heterogeneity: not applicable Test for overall effect: Z =3.31(P=0.0009) 046 month follow up (0 to 6 months "change" score) Eliasson 2005 21 1.79(1.03) 20 Subtotal (95% CI) 21 20 Test for heterogeneity: not applicable Test for over all effect: Z = 2.27 (P = 0.02)

0.93(1.19)

100.00 100.00

0.74 [0.10, 1.37] 0.74 [0.10, 1.37]

10

10

Favours control Favours CIMT

Figure 2

Assisting Hand Assessment change score from baseline to two and six months.

Review: Constraint induced movement therapy in the treatment of the upper limb in children with hemiplegic cerebral palsy Comparison: 03 Forced Use versus traditional services Outcome: 06 WeeFIM - Total "change" scores
Study or sub-category CIMT mean (SD) Traditional mean (SD) SMD (random) 95%CI Weight % SMD (random) 95%CI

01 Post-treatment (0 to 6 weeks "change" score) Sung 2005 18 1.94(1.70) 13 Subtotal (95% CI) 18 13 Test for heterogeneity: not applicable Test for overall effect: Z=1.09 (P=0.28)

1.15(2.20)

100.00 0.40 [0.32, 1.12] 100.00 0.40 [0.32, 1.12]

10 5 0 5 10 Favours control Favours CIMT

Figure 3

WeeFIM total change score from baseline to six weeks.

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

Constraint-induced movement therapy for the upper limb Clinical messages  This systematic review found a significant treatment effect using modified constraintinduced movement therapy in a single nonrandomized trial.  A positive trend favouring CIMT and Forced use was demonstrated.  The use of CIMT, mCIMT and Forced use should remain within clinical trials until the evidence is clearer. varying methodological quality and sample sizes were identified.5,17,18 The three intervention protocols varied significantly, particularly in relation to the intensity of treatment, from no increase in conventional treatment18 to 6 hours a day for 21 consecutive days.5 As each trial examined different CIMT intervention protocols, CIMT, mCIMT and Forced use respectively, it was not possible to undertake a meta-analysis. Results from DeLuca5 on the Dissociated Movement subscale of the QUEST, which was scored by blinded raters and the only outcome measure with known validity and reliability, did not show statistically significant treatment effect for CIMT. This was despite a 14.23 (SD 17.48) point change in the CIMT group compared with 1.74 (16.02) in the control group, indicating that the result may have been clinically important for at least some individuals. The small sample size of the study (n 18) and high variability across individuals is likely to have resulted in inadequate power to detect a statistically significant difference between groups. Significant treatment effects favouring CIMT, however, were found by DeLuca5 on parent response questionnaires that were without reported psychometric properties (Pediatric Motor Activity Log Frequency of Use subscale, Emerging Behaviors Scale). Results from the Toddler Arm Use Test, administered by raters blind to group allocation, were not significant. The clinical importance of these results is difficult to determine in the absence of evidence related to validity and reliability of the measures. The significant treatment effect demonstrated by Eliasson et al.17 at two and six months post intervention using the Assisting Hand

683

Assessment,2022 has important implications for the use of CIMT protocols in clinical practice. A modified protocol using a removable mitt, as opposed to a long arm cast, and significantly reduced intensity of practice (2 hours per day) when compared with a CIMT protocol (6 hours per day) may be more clinically feasible and developmentally appropriate for young children. Additionally, the identification by Eliasson et al.17 that older children and those with very poor hand function improved more than younger and less impaired children highlights the need for further investigation into the individual characteristics of children who will benefit from CIMT and its modified versions. The Forced use trial by Sung et al.18 reported statistically significant results on all outcomes using what was described as non-parametric paired and unpaired t tests.18 The findings from these unclear analytical processes were not supported by further analysis using RevMan,16 which found a significant treatment effect for only one outcome: the self-care component of the WeeFIM. The clinical significance of this treatment effect is questionable with a mean difference between groups of one point on a 56-point scale. The authors of the review strongly recommend the findings from this trial be viewed with caution given the ambiguity of its specific methodology, lack of methodological rigour and inadequate reporting.

Reviewers conclusions
This systematic review found a significant treatment effect using mCIMT in a single non-randomized trial. A positive trend favouring CIMT and Forced use was also demonstrated. While these early results are encouraging, they are inconclusive. It is recommended that the use of CIMT, mCIMT and Forced use should remain within clinical trials until the evidence is clearer. Suggestions for future research include conducting appropriately powered studies and utilizing uniform and objective outcome measures which will allow pooling of data in future meta-analyses. Valid and reliable outcomes which measure the usefulness of the affected upper limb in bimanual tasks (e.g. the Assisting Hand Assessment) and in relation to individual client and family goals

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

684

B Hoare et al.
Institute, Royal Childrens Hospital, Melbourne, 2005. Howard J, Soo B, Kerr Graham H et al. Cerebral palsy in Victoria: Motor types, topography and gross motor function. J Paediatr Child Health 2005; 41: 47983. Jessen C, Mackie P, Jarvis S. Epidemiology of cerebral palsy. Arch Dis Child Fetal Neonatal Ed 1999; 80: 158. Kuhtz-Buschbeck JP, Krumlinde-Sundholm L, Eliasson A-C, Forssberg H. Quantitative assessment of mirror movements in children and adolescents with hemiplegic cerebral palsy. Dev Med Child Neurol 2000; 42: 72836. DeLuca S. Intensive movement therapy with casting for children with hemiparetic cerebral palsy: A randomised controlled trial. Dissertation, The University of Alabama at Birmingham, 2002. Taub E, Wolf SL. Constraint induction techniques to facilitate upper extremity use in stroke patients. Top Stroke Rehabil 1997; 3: 124. Eliasson A-C, Bonnier B, Krumlinde-Sundholm L. Clinical experience of constraint induced movement therapy in small children with hemiplegic cerebral palsy a day camp model. Dev Med Child Neurol 2003; 45: 35760. Taub E. Somatosensory deafferentation research with monkeys: implications for rehabilitation medicine. In Ince LP. ed. Behavioural psychology in rehabilitation medicine: clinical applications. Williams and Wilkins, 1980: 371401. Taub E, Uswatte G, Pidikiti R. Constraintinduced movement therapy: A new family of techniques with broad application to physical rehabilitation a clinical review. J Rehabil Res Dev 1999; 36: 23751. Morris DM, Taub E. Constraint-induced therapy approach to restoring function after neurological injury. Top Stroke Rehabil 2001; 8: 1630. Deeks JJ, Higgins JP, Altman DG. Analysing and presenting results. In Higgins JP ed. Cochrane Handbook for Systematic Reviews of Interventions 425: [updated May 2005]: Section 8 In The Cochrane Library 2005, Issue 3. John Wiley & Sons Ltd, 2005. Hoare BJ, Wasiak J, Imms C, Carey L. Constraint induced movement therapy in the treatment of the upper limb in children with hemiplegic cerebral palsy. Cochrane Database Syst Rev 2007, Issue 2. World Health Organization. International Classification of Impairments, Disabilities, and Handicaps (ICF). 2001. Available from: www3.who.int/icf/icftemplate.cfm?myurl homepage.html&mytitleHome%20Page.

(for example the Canadian Occupational Performance Measure and Goal Attainment Scaling) should be included. Investigations should pursue longer term outcomes and examine the individual characteristics of children who experience positive CIMT outcomes. The impact of varying intensities of treatment and the environment or social context in which the treatment is provided should also be examined with a view to optimizing treatment protocols and enhancing treatment outcomes. Competing interests None declared.

3 4

Contributors BH initiated the review, designed the review, wrote the paper, rated methodological quality, extracted and analysed data and was guarantor. CI monitored progress revising the review critically for intellectual content, extracted and analysed data and contributed to the discussion of the results. LC analysed data, monitored progress revising the review critically for intellectual content and contributed to the discussion of the results. JW contributed to the design of the review, conducted the search strategy and monitored progress.

6 7

Acknowledgements The authors appreciate the help and support of the Movement Disorders Cochrane Review Group. This paper is based on a Cochrane Review published in The Cochrane Library 2007, Issue 2 (see www.thecochranelibrary.com for information). Cochrane Reviews are regularly updated as new evidence emerges and in response to feedback, and The Cochrane Library should be consulted for the most recent version of the review.

10 11

12

References
13 1 Reid S, Lanigan A, Walstab J, Reddihough D. Third report of the Victorian cerebral palsy register. Department of Child Development and Rehabilitation, Murdoch Childrens Research

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008

Constraint-induced movement therapy for the upper limb

14 Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995; 273: 40812. Follmann D, Elliot P, Suh I, Cutler J. Variance imputation for overviews of clinical trials with continuous response. J Clin Epidemiol 1992; 45: 76973. Review Manager (RevMan) [Computer program]. The Cochrane Collaboration. 4.2 ed. The Nordic Cochrane Centre, 2003. Eliasson A-C, Krumlinde-Sundholm L, Shaw K, Wang C. Effects of constraint-induced movement therapy in young children with hemiplegic cerebral palsy: an adapted model. Dev Med Child Neurol 2005; 47: 26675. Sung IY, Ryu JS, Pyun SB, Yoo SD, Song WH, Park MJ. Efficacy of forced-use therapy in hemiplegic cerebral palsy. Arch Phys Med Rehabil 2005; 86: 219598. Taub E, Ramey SL, DeLuca S, Echols K. Efficacy of constraint-induced movement therapy for children with cerebral palsy with asymmetric motor impairment. Pediatrics 2004; 113: 30512. Holmefur M, Krumlinde-Sundholm L, Eliasson A-C. Interrater and intrarater reliability of the Assisting Hand Assessment. Am J Occup Ther 2007; 61: 7984. Krumlinde-Sundholm L, Holmefur M, Kottorp A, Eliasson A-C. The Assisting Hand Assessment: current evidence of validity, reliability and responsiveness to change. Dev Med Child Neurol 2007; 49: 25964. Krumlinde-Sundholm L, Eliasson A-C. Development of the Assisting Hand Assessment: a Rasch-built measure intended for children with

685

23

15

24

16 17

25 26

18

27

19

28 29

20

21

30

22

unilateral upper limb impairment. Scand J Occup Ther 2003; 10: 1626. DeMatteo C, Law M, Russel D, Pollock N, Rosenbaum P, Walter S. The reliability and validity of the Quality of Upper Extremity Skills Test. Phys Occup Ther Pediatr 1993; 13: 118. DeMatteo C, Law M, Russel D, Pollock N, Rosenbaum P, Walter S. QUEST: Quality of Upper Extremity Skills Test. McMaster University, Neurodevelopmental Clinical Research Unit, 1992. Mathiowetz V, Federman S, Wiemer D. Box and blocks test of manual dexterity: Norms for 619 year olds. Can J Occup Ther 1985; 52: 24145. Zivianni J, Ottenbacher KJ, Shephard K, Foreman S, Astbury W, Ireland P. Concurrent validity of the Functional Independence Measure for children (WeeFIM) and the Pediatric Evaluation of Disabilities Inventory in children with developmental disabilities and acquired brain injuries. Phys Occup Ther Pediatr 2001; 21: 91101. Ottenbacher KJ, Msall ME, Lyon NR. The WeeFIM instrument: its utility in detecting change in children with developmental disabilties. Arch Phys Med Rehabil 2000; 81: 131726. Mathiowetz V, Volland G, Kashman N, Weber K. Adult norms for the box and block test of manual dexterity. Am J Occup Ther 1985; 39: 38693. Ottenbacher KJ, Msall ME, Lyon NR, Duffy LC, Granger CV, Braun S. Interrater agreement and stability of the Functional Independence Measure for Children (WeeFIM): use in children with developmental disabilities. Arch Phys Med Rehabil 1997; 78: 130915. Ottenbacher KJ, Taylor ET, Msall ME et al. The stability and equivalence reliability of the functional independence measure for children (WeeFIM). Dev Med Child Neurol 1996; 38: 90716.

Downloaded from http://cre.sagepub.com by Suresh Mani on October 21, 2008