Modified constraintinduced. therapy for young children with hemiplegic cerebral palsy: a pilot study

Transcription

1 Modified constraintinduced movement therapy for young children with hemiplegic cerebral palsy: a pilot study CE Naylor* MSc MCSP, Senior Paediatric Physiotherapist, Royal Berkshire and Battle Hospitals NHS Trust, Reading; E Bower PhD FCSP, Honorary Senior Lecturer, University of Southampton, Southampton, UK. *Correspondence to first author at Dingley Child Development Centre, Battle Hospital, Oxford Road, Reading RG30 1AG, UK. cate.naylor@ntlworld.com The objective of the study was to evaluate the effectiveness of modified constraint-induced movement therapy in young children with hemiplegia. It was a single-case experimental design using children as their own controls. Assessment was at entry to the study and subsequently at 4-weekly intervals. A 4-week baseline period with no hand treatment, controlling for maturation, was followed by a 4-week treatment period and a second 4-week period with no hand treatment to measure carry-over. Treatment consisted of twice-weekly 1-hour sessions of structured activities with a therapist and a home programme for non-treatment days. Only verbal instruction and gentle restraint of the unaffected arm were used to encourage use of the affected arm. Nine children (six males, three females; median age 31mo, age range 21 to 61mo) presenting with congenital spastic hemiplegia (five right side, four left side) were involved in the study. Changes in hand function were evaluated with the Quality of Upper Extremity Skills Test. Improvement was seen throughout the study with statistical significance, using the Wilcoxon signed rank test, of 0.01 immediately after treatment. Results of this pilot study suggest that this modification of constraintinduced movement therapy may be an effective way of treating young children with hemiplegia. Future work is planned to consolidate and develop these results. See end of paper for list of abbreviations. Constraint-induced movement (CIM) therapy has been developed as a result of neuroscientific research with monkeys. Researchers found that by constraining the intact limb of a monkey whose other limb had been deafferented, the monkey was able to learn to use the affected limb again (Knapp et al. 1963). Pons et al. (1991) suggested that massive cortical reorganization occurred after somatosensory deafferentation of a limb, which might account for the therapeutic effect of CIM therapy. CIM therapy in humans is based on the hypothesis that in hemiplegia, disuse of the affected arm can occur as a result of learned non-use because it becomes more convenient to use the unaffected arm. CIM therapy involves constraining the unaffected arm while intensively training the affected arm by inducing concentrated repetitive practice. Neuroimaging has shown a significantly increased cortical representation of the affected hand after CIM therapy (Taub et al. 1999, 2002). Unlike adults with hemiplegia who have had function before the insult to the central nervous system, children with hemiplegia have usually never used their affected upper limb normally. On the basis of Taub s early work with deafferented monkeys, it has been suggested that plasticity of the brain could be a basis for rehabilitation with CIM therapy (Charles et al. 2001), which is becoming widely used in adults with hemiplegia (Taub et al. 1999) and is now being developed for use with children. Crocker et al. (1997) restrained the unaffected arm of two children aged 2 to 3 years in a splint during waking hours. The children were observed during normal therapy and free play. One child improved but the other did not tolerate the splintwearing regime. Charles et al. (2001) in a study of three children with hemiplegia aged 8 to 13 years constrained the unaffected arm in a sling for 6 hours a day for 2 weeks. The children were observed during functional and play activities while wearing the sling. Two of the children improved their hand function. In a crossover trial (Willis et al. 2002), 12 children received a plaster cast on the unaffected arm for 1 month, 13 control children did not; all the children continued their normal therapy. Greater improvement was seen in the treatment group, which was reconfirmed when the plaster cast was applied to the control group. Using a randomized controlled A B A design, Taub et al. (2004) randomly assigned 18 children aged between 7 months and 8 years to treatment and control groups. In the treatment group a bivalved plaster cast was applied to the unaffected arm for 21 days and functional training activities were given to the affected arm for 6 hours a day. Results demonstrated improvements in both amount of use and functional abilities in the treatment group. The aim of the present study was to investigate whether a modification of the method of restraint of the unaffected arm in children with congenital hemiplegic cerebral palsy (CP) and a treatment programme of activities was effective in improving function in the upper limb. The modification of the CIM therapy protocol provided a less invasive method of restraint than in other studies. A structured programme of activities was chosen, including play which is vital for the development of all children and is central to their lives (Pollock et al. 1997). Method PARTICIPANTS Children included in the study had a diagnosis of congenital spastic hemiplegic CP made by a consultant paediatrician, Developmental Medicine & Child Neurology 2005, 47:

2 were aged between 18 months and 5 years, and attended the Dingley Child Development Centre in Reading for treatment. They were excluded if they had unilateral signs without a diagnosis of hemiplegic CP, had bilateral signs, or were aged less than 18 months or more than 5 years. Sixteen children were nominated for the study, and 14 met the inclusion criteria. Of those 14 children, nine took part. Reasons for not taking part were parental work commitments (n=3), unexpected holiday (n=1) and parent didn t think the child needed the treatment (n=1). Median age of the participants (six males, three females) was 31 months, range 21 to 61 months. Distribution of affected side was five right and four left. Demographic data are shown in Table I and medical histories in Table II. Table I: Characteristics of children Child Sex Age (mo) Affected side Orthotics 1 M 21 Right Thumb abduction 2 F 23 Right Thumb abduction 3 M 25 Left Thumb abduction 4 M 29 Left None 5 M 31 Left None 6 M 32 Right None 7 F 50 Right None 8 M 56 Left None 9 F 61 Right Wrist extension splint Table II: Medical history of children Child History 1 Term, caesarean; failure to progress language delay and cognitive difficulties MRI shows left middle cerebral artery infarct 2 Term, caesarean; failure to progress neonatal seizures; medication only in neonatal period MRI shows left middle cerebral artery infarct 3 Preterm, 28wks; ventilated; hydrocephalus; shunt visual impairment, language delay MRI shows grade 4 right IVH 4 Term, normal delivery MRI normal 5 Preterm, 24wks; ventilated social and communication difficulties MRI shows right porencephalic cyst 6 Term, normal delivery MRI shows left middle cerebral infarct 7 Term, normal delivery visual impairment: nystagmus and strabismus MRI shows periventricular leucomalacia 8 Twin 2, born at 38wks (twin 1 no problems) MRI shows slight enlargement of right trigone 9 Elective caesarean for breech presentation at 38wks neonatal encephalopathy and seizures; medication only in neonatal period MRI shows left parietal infarction extending to basal ganglia MRI, magnetic resonance imaging; IVH, intraventricular leucomalacia. STUDY DESIGN As the study used a small number of participants, a singlecase A B A experimental design was chosen with a convenience sample of children who acted as their own controls. During A periods, the children had a 4-week rest from their regular hand therapy treatment (a combination of functional activities and play using both hands fortnightly or monthly for 1 hour, with the emphasis being on bilateral work and advice to parents). During B periods, children had 1 hour a day of modified CIM therapy. Constraint of the unaffected arm was achieved by gentle restraint, with an adult holding the child s unaffected hand during the activities. They were not restrained between activities but were allowed to have both hands free. Children were also encouraged verbally to use their other hand. The programme was administered twice weekly by the child s regular therapist and on other days as a home programme by parents. Children were assessed at the beginning of the study and at 4-weekly intervals in their usual treatment location by the researcher (Fig. 1). PROCEDURES Ethical approval for the study was obtained from the West Berkshire Local Research Ethics Committee and consent was given by each child s consultant paediatrician and parents. To reduce the risk of variables affecting the results, treatment sessions were at the same time of day and in the same location. All other interventions were continued as normal and a record was kept. The assessor was not present for the treatment sessions and, therefore, was not influenced by the child s performance during treatment. Intervention followed a detailed programme of fine motor and play activities aimed at improving fine motor skills. Instructions on how the child should perform each activity were given. Treatment sessions included action songs, playing with dough, sorting, threading, posting, jigsaws, and playing computer games with a touch screen (Appendix I). ASSESSMENT OF FINE MOTOR FUNCTION Fine motor function was assessed with the Quality of Upper Extremity Skills Test (QUEST; DeMatteo et al. 1993), which supplies information relating to movement and postural responses and an evaluation of the quality of upper limb function. The test has four domains: dissociated movement, grasps, protective extension, and weight bearing. It provides a total score for the four domains and a mean functional score is calculated for the two limbs combined. The test is validated for use with children aged 18 months to 8 years. It Entry Assessment 1 Assessment 2 Assessment 3 Assessment 4 Baseline observation period (A) Time (wks) Exit Figure 1: Trial design. Treatment period (B) Follow-up period (A) 366 Developmental Medicine & Child Neurology 2005, 47:

3 is assumed that a child with no disability would score perfectly in all areas except grasp at 18 months. DATA ANALYSIS This study used a single-case experimental design that used participants as their own controls, observed under two different conditions. Statistical significance was tested with the non-parametric Wilcoxon signed rank test because the sample size was small, and the QUEST generated ordinal data. Data were analyzed with SPSS for windows (version 10). Results AMOUNT OF TREATMENT RECEIVED The median number of sessions attended was eight (range six to eight) and the median length of treatment session was 51.3 minutes (range of 40.7 to 60.0). Table III shows the amount of treatment that each child received and their health and compliance during treatment at the centre. Parents were requested to undertake the programme with their children for 1 hour a day on non-treatment days. Each child could, therefore, have received 28 hours of modified CIM therapy over the 4 weeks. ASSESSMENT RESULTS Total QUEST scores for each child for the four assessments are presented in Table IV and the data are plotted in a box-andwhisker plot (Fig 2). The change in scores can be adjusted for the improvement occurring in the non-intervention periods by using the formula Score(3 2) [Score(2 1)+Score(4 3)]/2. This gives an indication of the true treatment score, with the effects of maturation and spontaneous recovery factored out. The adjusted treatment scores are shown in Table V. Table VI shows the mean and standard deviation, 95% confidence intervals, and significance values from the Wilcoxon signed rank test. Comparison of the means of scores 2 and 3 give p=0.008 and a significance of Comparison of the means of scores (1 and 2) and (3 and 4) both give p=0.028, with a significance of Discussion Results from this study are consistent with other studies in showing a significant improvement in upper limb function after CIM therapy in children (Willis et al. 2002, Taub et al. 2004). A key difference of this study and the new finding is that the improvements in upper limb function were achieved by using only gentle restraint and verbal instruction, unlike other studies, which have employed prolonged periods of physical restraint. In addition, these improvements were achieved with a structured programme of activities that aided the children s tolerance of the sessions. Crocker et al. (1997) found poor compliance a reason for withdrawal from the study. No side effects of prolonged restraint, such as muscle wasting or contractures due to immobilization, have been recorded in previous studies, although Charles et al. (2001) informed parents of the risks of splintage. This more gentle method of restraint is thought to reduce the possible risk of side effects. OUTCOME MEASURE AND DATA ANALYSIS The QUEST is a validated measure for bilateral upper-limb Table III: Treatment records Child Treatment sessions attended Mean length of treatment (min) per session Compliance with treatment Health status Variable Well Good Well Variable Well Good Well Variable Well Good Unwell 1 session Good Well Good Well Variable Well Table IV: Quality of Upper Extremity Skills Test scores Children Before intervention After intervention Score 1 Score 2 Score 3 Score Table V: Adjusted treatment scores Children Difference Difference Difference Amended treatment score Constraint-induced Movement Therapy in Children with Hemiplegia CE Naylor and E Bower 367

4 function, so a child with a normally functioning arm and a totally dysfunctioning hemiplegic arm would be expected to score 50. In this group, six of the children (1 to 6) were not of the developmental age to have a mature pencil grip and would not be expected to achieve a full score. The three children (1, 3, and 5) who scored below 50 in their first assessment also had learning difficulties which affected their ability to perform the test. Improvements after intervention were seen predominantly in the affected arm, with a very slight increase in pencil grip in the unaffected arm in two children (2 and 4). If it is assumed that improved function through maturation occurs bilaterally and that treatment is limited to the affected arm, it is reasonable to attribute an increase in QUEST score after treatment to the effect of the treatment itself. Further investigation of the use of QUEST in children with hemiplegia would establish the quantitative relationship between QUEST scores and unilateral involvement. Improvements seen in the baseline period might have been due to maturation, and in the final month to maintenance of skills after intervention and increased awareness by parents of the benefits of continuing the activities. OBSERVATIONS Improvements in the affected arm were seen in the domain of grasps. Child 3 progressed from never having grasped an Table VI: Comparison of means of QUEST scores Scores Mean Standard 95% confidence Wilcoxon deviation interval QUEST, Quality of Upper Extremity Skills Test (DeMatteo et al. 1993). Median and interquartile QUEST scores Intervention period Assessments Figure 2: Box and whisker plot of Quality of Upper Extremity Skills Test (QUEST) scores (n=9). Box covers interquartile range, line indicates median, and whiskers extend to minimum and maximum scores. object to a primitive grasp for the first time, whereas others (2, 4, 5, 6, and 7) refined their grasp. All children showed improvements in weight bearing and protective extension. The youngest child (child 1), presenting with learning difficulties, improved his proximal stability, and although he did not progress greatly in his fine motor skills, began to creep reciprocally, improving his gross motor skills. Child 8 improved the least, but had the mildest problems. Child 9, the oldest, improved the most; this was probably a combination of increased mobility without the splint and the effects of regular treatment (previously compliance with treatment had been a problem). BENEFITS OF THE STUDY The method of restraint in this study was well tolerated, it was easy to administer and parents were happy to continue its use at home. The activities were appropriate and easily adapted to the children s abilities. This study incorporated two assessments in the baseline period, providing a control for maturation; this is especially important with very young participants. LIMITATIONS OF THE STUDY A minimum age of 18 months was selected because the QUEST is validated for children aged between 18 months and 8 years. To reduce the risk of neglect of the affected arm and to take advantage of plasticity within the brain, it could be argued that this method of intervention should be implemented much earlier. The method of gentle restraint of the unaffected arm was accurately controlled during the time that the child was undergoing therapy but was difficult to control when the parents continued the regime at home. A further study would need to consider how to control and record home intervention accurately. FUTURE WORK A further project using a randomized controlled trial is planned to determine the generalizability of these results. An appropriate design would be an A B A/A A A design with children randomly selected to treatment or control groups. To facilitate participation in the study, a further B period of treatment for children in the control group would be included. An independent person masked to whether a participant was in the treatment or control group would undertake the assessments, which would include QUEST and a measure of functional ability. Associated disabilities would be taken into account before randomization, and children would be formally assessed for their classification and degree of loss of sensation. Sample size would be calculated from the results of this pilot study. Conclusions CIM therapy involves restraint of the unaffected arm and intensive treatment of the affected arm. A quantitative study of nine participants has been performed to investigate the use of CIM therapy in children with congenital spastic hemiplegic CP. Two modifications that differentiate the study from previous CIM therapy studies were introduced with the aim of improving tolerance of the treatment: first, the use of gentle restraint and verbal instruction instead of prolonged physical restraint, and second, the use of structured activities and play 368 Developmental Medicine & Child Neurology 2005, 47:

5 as therapy. Results suggest that a significant improvement occurred in upper limb function after the intervention in these children with hemiplegic CP. The use of gentle restraint and verbal instruction was found to be effective, and the activities were well tolerated. Both were easy to administer and were acceptable to the children. The statistical significance observed in this study is consistent with that found in other CIM therapy studies involving children and is supported by positive quantitative results obtained from the QUEST scores. Results of this small pilot study are sufficiently promising to justify further work in the form of a randomized controlled trial to determine more precisely the effectiveness of this modified method of CIM therapy. DOI: /S Accepted for publication 9th August References Charles J, Lavinder G, Gordon AM. (2001) Effects of constraint-induced therapy on hand function in children with hemiplegic cerebral palsy. Pediatr Phys Ther 13: Crocker M, MacKay-Lyons M, McDonnell E. (1997) Forced use of the upper extremity in cerebral palsy: a single case study. Am J Occup Ther 51: DeMatteo C, Law M, Russell D, Pollock N, Rosenbaum P, Walter S. (1993) The reliability and validity of the Quality of Upper Extremity Skills Test. Phys Occup Ther Pediatrics 13: Knapp HD, Taub E, Berman AJ. (1963) Movements in monkeys with deafferented limbs. Exp Neurol 7: Pollock N, Stewart D, Law M, Sahagian-Whalen S, Harvey S, Toal C. (1997) The meaning of play for young people with physical disabilities. Can J Occup Ther 64: Pons TP, Garraghty AK, Ommaya AK, Kaas JH, Taub E, Mishkin M. (1991) Massive cortical reorganization after sensory deafferentation in adult macaques. Science 252: Taub E, Landesman Ramey S, De Luca S, Echols K. (2004) Efficacy of constraint-induced movement therapy for children with cerebral palsy with asymmetric impairment. Pediatrics 113: Taub E, Uswatte G, Elbert T. (2002) New treatments in neurorehabilitation founded on basic research. Nat Rev Neurosci 3: Taub E, Uswatte G, Pidikiti R. (1999) Constraint-induced movement therapy: a new family of techniques with broad application to physical rehabilitation-clinical review. J Rehabil Res Dev 36: Willis JK, Morello A, Davie A, Rice JC, Bennett JT. (2002) Forced use treatment of childhood hemiparesis. Pediatrics 110: List of abbreviations CIM QUEST Constraint-induced movement therapy Quality of Upper Extremity Skills Test Appendix I TREATMENT PROGRAMME The children were treated individually, working with a parent and therapist. Groups of between two and four children were treated individually but on the same occasion, coming together for a warmup and a goodbye session. STARTING POSITION The child sat at a table with feet flat on the floor and hips and knees at 90. The table was at chest height so that the child s arms were supported. For the child needing more support for sitting, adaptive seating was provided and activities were performed on a tray positioned at chest height. Where objects needed to be held or steadied for the child to complete the task, this was done by the parent or therapist. WARM-UP SESSION Children were asked to shake their hands high, low, to the side, behind, and in front. Nursery rhymes with hand actions such as Peter Pointer where are you?, Wind the bobbin up and Incey wincey spider were sung. INDIVIDUAL ACTIVITIES All activities were adapted to the child s age and ability. PLAY WITH DOUGH If the child was unable to open out their hand, the therapist facilitated hand opening and placed the play dough into the child s hand. Ball of dough: roll the ball on the table and then squash it into a flat pancake. Sausage of dough: roll the sausage and then squeeze or squash it with their thumb and alternate fingers. THREADING For the younger child, a pipe cleaner; for the child with better hand function, stiff cord or thread. Different sizes of beads or buttons were used, depending on the child s age or ability. The bead was held by an adult with the child threading with their affected hand. Five beads or buttons were threaded using the affected hand. POSTING BOXES Posting boxes were used and 10 items were posted using the affected hand. JIGSAWS Form boards were used and the child was asked to place all the pieces back into it. SORTING The child was asked to sort a number of small objects by picking them up and placing them into different bowls. For children unable to pick up the objects, the objects were placed into their affected hand and they were asked to release them into the bowls. Sorting could be by colour, size, or shape, depending on the cognitive level of the child. COMPUTER The child used the touch screen on the computer with their affected hand and a programme appropriate to the child s cognitive level was initiated. Children were encouraged to reach forwards to touch the screen, and cause-and-effect programmes were used. Each child spent 5 minutes using the computer. FINGER GAMES Finger puppets were placed on fingers of the affected hand and the child was encouraged to tell a story with their therapist by using the finger puppets. ENDING THE SESSION A ball game was played, pushing a ball from one child to another using the affected hand. The session finished with a goodbye song. Constraint-induced Movement Therapy in Children with Hemiplegia CE Naylor and E Bower 369