INTRODUCTION AND PURPOSE
Contractures, hip pathologies, and spinal malalignments1–3 are common complications for children with neuromuscular disabilities, including cerebral palsy (CP), muscular dystrophies, and neural tube defects. To address complications and promote independence in these children, considerable therapeutic resources are used such as orthoses, therapy equipment, and therapy time.4–7 Therapists frequently prescribe and encourage compliance to a variety of stretch interventions including (1) active stretching, (2) passive stretching, (3) prolonged positioning through supported standing, or (4) prolonged stretching through casting and orthoses.8,9 The clinical rationale for using these interventions is to avoid or defer surgery, decrease complications such as contractures, and promote function.8,9 Proposed causes of contractures that have been hypothesized include agonist-antagonist muscle imbalance, muscle fiber atrophy, spasticity, static positioning, and structural changes to muscle tendon tissue (eg, the reduction of in-series or in-parallel sarcomeres).8,10,11 Regardless of the cause, research shows that contractures interfere with activities of daily living, cause pain, sleep disturbance, and increase the burden of care.10 Stretching, positioning, and active movement are proposed to prevent contractures and malalignment by avoiding the reduction of the number of in-series sarcomeres that decreased movement causes.8 Despite the common practice of prescribing stretch interventions, these clinical rationales have not been validated as there is limited and varied evidence about the actual causes of contractures, the proposed physiological theory of stretching, and the clinical effectiveness of stretching in the human model.8,10
For all therapeutic interventions, clinicians need to consider potential benefits and harms to the child and family.12 This clinical decision is even more important when there is limited evidence to guide practice. Although physical therapists have clinical rationales for the possible benefits of stretch interventions, the possible harmful effects also need to be examined. For example, continuous postural management can have a negative effect on sleep hygiene,13,14 and assisted stretching is frequently reported as the most common daily activity that causes pain for children with CP.15 Complying with an intervention that compromises a child's sleep or induces pain can place a significant emotional burden on the child, caregivers, and parents.13,16
Considering the routine prescription of stretch interventions and the burden and cost of implementation, a systematic analysis of the efficacy of these interventions is needed. The objective of this systematic review (SR) is to determine whether casting, orthoses, stretching programs, or supported standing programs are effective in improving or maintaining body functions and structures, activity, or participation in children and youth with neuromuscular disabilities.
English language titles were searched from the earliest date available until December 31, 2014, in the following electronic databases: CINAHL, EMBASE/Ovid, EBMR/Ovid, MEDLINE/PubMed, MEDLINE/EBSCO, and Physiotherapy Evidence Database. See the Appendix for the detailed electronic database search strategy. We did not use population-specific search terms (eg, CP and muscular dystrophy) to get comprehensive search results to later limit by inclusion criteria. Preliminary searches did not yield any articles with the same objective of this review. Details of the protocol for this SR were registered on September 19, 2014, on the International Prospective Register of Systematic Reviews (PROSPERO) and can be accessed at: http://www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42014013807.17
Inclusion. The inclusion criteria of this review were: (a) studies published in peer-reviewed journals appraised as Level 1 or 2 Oxford Centre of Evidence-Based Medicine (OCEBM) levels of evidence;18 (b) study participants were younger than 19 years and had a confirmed neuromuscular disability; (c) studies contained a stretch intervention; and (d) studies evaluated the effect of stretch interventions on any body structure, body function, activity, or participation provided that there was a primary outcome measure of flexibility. Studies were included if cointerventions involved education and/or other exercise prescription (eg, aquatic therapy, aerobic training, and strength training) as long as one of the interventions was a stretch intervention. These types of cointerventions were included to allow for comprehensive programs that can be fully delegated to members of the child's team (as stretch interventions often are) under the supervision, but not direct treatment, of a physical therapist.
To address studies that included both pediatric and adult subjects, the following was determined a priori to determine study eligibility: (a) individual clinical studies must have 50% or more pediatric subjects, or a mean participant age of less than 19 years; and (b) SRs must have 50% or more studies that met the pediatric criteria, or they must provide subanalyses of the pediatric population.
Exclusion. Exclusion criteria of this review were (a) observational studies and surveys; (b) studies included able-bodied youth or youth with disabilities not considered neuromuscular in nature; (c) studies involved concurrent treatment of other physiotherapeutic interventions directly provided by a physical therapist or other health care provider (eg, acupuncture, Botox, electric modalities, manual therapy, massage, or neurodevelopmental treatment); (d) where casting was used as part of constraint-induced movement therapy; or (e) where recent surgery was done.
For the purpose of this review, the definition of a neuromuscular disability is any chronic disease or syndrome that impairs the function of skeletal muscles. This impairment can affect the muscle structure itself and/or the signal sent to the muscle. Examples of neuromuscular disabilities that were considered for review include CP, muscular dystrophies, neural tube defects, spinal cord injuries, spinal muscular atrophies, traumatic brain injury, and other rare neuromuscular diseases. The definition of a stretch intervention is an intervention aimed at maintaining or increasing joint mobility by influencing the extensibility of soft tissues spanning joints.10 The following were preidentified as possible stretch interventions: bracing, casting, orthoses, positioning programs, self-administered stretches, splinting, stretches by caregivers, and yoga programs. Bracing, splinting, and orthoses were considered to be one treatment category, herein “orthoses,” to improve clarity and knowledge translation. Both active and passive range of motion (ROM) and stretch programs were included. For the purpose of this review, flexibility was defined as the ability to move a joint through its complete ROM19 and could have been measured with a goniometer, through gait analysis or with another valid instrument.
Selection of Studies
One reviewer (JC) screened the titles and abstracts of found articles using the inclusion and exclusion criteria stated earlier. Full-text copies of any study that appeared to meet the inclusion criteria were obtained for further inspection. Two reviewers (JC and GW or RM) independently read each article and recommended inclusion or exclusion. In cases of disagreement, a third reviewer (CH) was consulted and discussion occurred until agreement could be met.
Two review authors (JC and GW or RM) independently performed data extraction of included studies using internally made data extraction forms. For each study, we collected information on the authors' main conclusions, level of evidence,18 outcome measures coded by the International Classification of Functioning, Disability and Health (ICF),20 participant baseline characteristics, sample size, study design, study methods, and type of intervention(s).
Assessment of Risk of Bias and Study Quality in Included Studies
For randomized controlled trials (RCTs), we assessed the risk of bias of individual studies by using a domain-based evaluation recommended by the Cochrane Collaboration21 because the use of scales for assessing quality or risk of bias is explicitly discouraged.22 The quality of SRs was assessed using an OCEBM appraisal sheet23 and recorded on the data extraction form. OCEBM levels of evidence were also assigned during the data extraction process and could be downgraded due to study bias or upgraded because of large effect sizes.18
We used the Grading of Recommendation Assessment, Development and Evaluation (GRADE) approach to assess risk of bias across studies.24 For purposes of SRs, GRADE defines the quality of a body of evidence as high, moderate, low, or very low for a particular outcome measure.24 Three review authors (JC, GW, and CH) first independently performed the quality of evidence grading for each identified outcome measure. A decision about the final grade assigned was reached through discussion and consensus.
The strength of clinical recommendations was made using the GRADE approach12 and Evidence Alert Traffic Light System (EATLS)25,26 to facilitate knowledge translation. A GRADE strength of recommendation is the extent to which one can be confident that the desirable consequences of an intervention outweigh its undesirable consequences.12 The recommendations are graded as either strong or weak, and in support of or against an intervention for a particular outcome measure.12 The EATLS rates interventions based on the quality of evidence according to the following criteria: green, go (ie, high-quality evidence supporting the effectiveness of this intervention, therefore use this approach); yellow, measure (ie, low-quality or conflicting evidence supporting the effectiveness of this intervention, therefore measure the outcomes of the intervention when using this approach to ensure the patient's goal is met); red, stop (ie, high-quality evidence demonstrating this intervention is unsafe or ineffective, therefore do not use this approach).25,26 Three review authors (JC, GW, and CH) independently performed a GRADE strength of recommendation and EATLS rating for each identified outcome measure. A decision about the final grading and strength of recommendations assigned was reached through discussion and consensus. Reporting of SRs followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement.27
Results of the Search
Electronic database and hand searching yielded 24 930 references. After removing duplicates and screening titles and abstracts, 81 studies were eligible for full-text review. There were 10 studies that required a third reviewer to determine eligibility. After inspecting the full reports, 16 articles were included (See Figure 1).
Figure 1 provides a summary of reasons for exclusion of studies (See Supplemental Digital Content 1, available at http://links.lww.com/PPT/A106, which lists all articles excluded at the full-text level). The most common reasons for exclusion were lower level of evidence (n = 26), adult population (n = 9), no stretch intervention (n = 8), cointerventions (n = 7), and downgraded due to risk of bias (n = 6). All articles downgraded due to risk of bias were determined to have unacceptable risk of attrition, detection, performance, selection, and other biases. Other biases included confounding cointerventions, poor compliance to intervention or poor reporting of compliance, lack of statistical analyses to determine whether groups were similar at baseline, and sample populations not representative of exposed cohort.
Levels of Evidence and Risk of Bias in Included Studies. Among the 16 included articles, 12 were SRs and 4 were randomized controlled or crossover trials (Table 1).28–43 Two of the SRs were appraised as Level 1 evidence as they had both high methodological quality and only included RCTs or SRs.28,29 The 10 other SRs were appraised as Level 2 evidence due to inclusion of lower levels of evidence and/or poor methodology. All of the included RCTs were graded as Level 2 evidence. In general, the methodological quality of included articles was poor. Common methodological weaknesses in the studies included lack of reporting, or inadequate randomization methods, allocation concealment, reporting of dropouts, and controlling for confounding cointerventions (see Supplemental Digital Content 2, available at http://links.lww.com/PPT/107, which summarizes methodological quality of included studies).
Sample sizes of the included studies ranged from 14 to 1110 (Table 1). The effectiveness of stretch interventions was investigated in the following populations: CP (n = 9), mixed disabilities (n = 4), Charcot-Marie-Tooth (n = 2), and Duchenne muscular dystrophy (n = 1). The age of participants ranged from 20 months to 30 years, with all studies having a median age less than 19 years.
The included studies evaluated the effectiveness of casting (n = 5), orthoses (n = 10), passive stretching or positioning (n = 5), and supported standing programs (n = 6). A wide range of casting protocols, orthoses configurations, prescription of stretching programs, and supported standing equipment was evident from the literature. Not all research reports described this in adequate detail to be replicated. Comparison interventions were explicitly mentioned in all included RCTs30–33; however, SRs often did not record comparison interventions. Where this information was well reported, comparison groups were often lacking because of study design (eg, case report) (Table 1).
Intervention dosing parameters such as treatment duration, frequency, and intensity were well recorded in all of the RCTs; however, these parameters were inconsistently reported in the SRs. When recorded, casting intervention lasted for 3 to 5 weeks with casting protocols not being well documented.28,29,34–36 There was a lack of information about the specific orthotic intervention dosing protocols recorded by the SRs.28,29,37–39 RCTs investigating night splinting required splints to be worn all night30,32,33 or every other night31 for a duration of 4 weeks,32 6 weeks,33 12 months,31 or 30 months.30 Passive stretching or positioning dosing reported in the SRs noted that 30 minutes' total stretch program was the most commonly chosen session time, with each stretch typically being held for 30 to 60 seconds and repeated for several repetitions.35,40,41 One SR noted an average duration of passive stretching or positioning study length to be 8.2 weeks with a mean frequency of intervention to be to 4.5 times per week.40 Supported standing program dosing was well recorded in 1 SR with the following evidence-based dosage recommended: 5 days/wk positively affects bone mineral density (BMD) (60-90 min/d), hip stability (60 min/d in 30°-60° degrees hip abduction), ROM of hip knee and ankle (45-60 min/d), and spasticity (30-65 min/d).42 The majority of included studies in the 2 other reviews noted 30 minutes as the common duration of supported standing; however, there was a large variation in study duration from just 1 session to 9 months of intervention.35,43 No studies commented on the intensity of the intervention, besides 1 RCT that mentioned that if night splinting interfered with sleeping, participants were to use them during day rest periods instead.31 It is important to note that although several of the SRs did not record specific dosing parameters, they noted a lack of long-term follow-up in their included studies.28,29,35,37,40
ROM, prevention of contractures, BMD, gait analyses, and spasticity were the most studied body function and structure measures (Table 1). Functional mobility assessment (eg, sit to stand) and the Gross Motor Function Measure were the most common activity measures. The Canadian Occupational Performance Measure was the only participation measure identified in any study.
Analysis of the Evidence
The effectiveness of all of the interventions as coded by ICF levels, GRADE quality of evidence, and by the previously mentioned knowledge translation tools is summarized in Table 2. Because of the large heterogeneity and lack of reporting of interventions and outcome measures used in individual studies and SRs, effect-size estimation and meta-analysis of the data were not performed.
Two RCTs,31,32 and 2 SRs4,36 on casting and orthoses reported on adverse events. Adverse events such as bruising and blistering were seen in 13% of subjects who had serial casting,32 whereas the majority of participants in a study on knee-ankle-foot-orthoses (KAFO) reported frequent pain because of muscle strain and pressure spots, as well as sleep disturbance.31 Additional complaints of night-time use of KAFOs included hot or sweating legs, itching, cramping, and bed-wetting.31 The most common adverse events of casting cited from the SRs included skin irritation, skin breakdown, and pain.34,36
Evidence. Consistent but very low evidence supports the use of 3 to 5 weeks of ankle casting for the positive short-term effects that it has on passive ankle dorsiflexion.28,29,34,35 Short-term improvements in gait parameters such as self-selected pace and stride length following ankle casting have also been noted29,34; however, this review did not identify any Level 1 or 2 evidence supporting or refuting long-term benefits on gait and ROM. There is insufficient research on the effectiveness of casting for other lower extremity joints. One SR that assessed the effectiveness of upper extremity casting for children with neurological conditions concluded that there is insufficient high-quality evidence regarding the effect or long-term effects to either support or abandon upper extremity casting.36 No studies identified in this review included the direct measurement of the effect of casting on activity or participation of children with neuromuscular disabilities.
Clinical Recommendation From the Evidence. Using the EATLS, serial casting for short-term improvement of ankle ROM is rated as a green intervention supported by very low evidence. All other outcome measures have insufficient evidence, thus casting is rated as a yellow intervention for these measures.
Evidence. The most consistent finding among studies identified in this review is that there is very low evidence that ankle-foot orthotic (AFO) devices that restrict plantarflexion improve gait kinematics and kinetics while the device is worn (Table 2).28,29,35,37,38 One SR that compared articulated and rigid AFOs for children with CP found significant differences in peak dorsiflexion, reduction in double-support time, increase in gait speed, and reduction in energy expenditure with the use of an articulated orthosis.39 There is both conflicting and insufficient evidence on the effectiveness of orthoses for the prevention of contractures, either by the use of AFOs or by wearing night splints.28–33,35,37,39 Two randomized trials showed that night ankle splints or KAFOs do not improve ROM in children with CP or Charcot-Marie-Tooth, whereas 1 study found that the expected annual change in tendoachilles contracture for boys with Duchenne muscular dystrophy was 23% less in the night splint and passive stretch group compared with the passive stretch-only group.30,31,33 One RCT found that at 4 weeks of postserial night casting, the experimental group had significant but small increase in ankle dorsiflexion; however, these effects were not maintained with stretching at 8 weeks.32 There is also conflicting and insufficient evidence in the studies identified to support the use of orthotics for promoting activity or participation while the device is worn. Two studies reported that wearing a lower extremity device might make functional activities, such as rising up from the floor, more difficult,28,35 whereas another study showed that orthoses have a positive effect on functional activities related to mobility.37 Most SRs reported that there is insufficient evidence to support or refute the use of orthoses in improving function,28,29,35 whereas the majority of randomized trials30–32 showed no functional difference between experimental and control groups or did not include a functional measure.33 Several authors mention that the wide variety of lower limb orthoses investigated as well as different terms used for the same orthoses made a systematic evaluation difficult.28,37,38 Poor compliance and tolerance of night orthoses has also been cited as a limitation in determining the effectiveness of this intervention.31,32 No high-quality studies that assessed the effectiveness of upper extremity orthoses were found.
Clinical Recommendation From the Evidence. According to the EATLS, orthoses that restrict plantarflexion to improve a child's gait while the device is worn is a green intervention. Orthotic use for the prevention of contractures and promotion of activity and participation are yellow interventions.
Positioning and Stretching
Evidence. The most commonly reported outcome measures for positioning and stretching programs were prevention of contractures and ROM (Table 1). Two SRs reported that there is insufficient evidence to support or refute the use of passive ROM35 or positioning29 to prevent contractures. One review29 concluded that manual stretching is ineffective for contracture prevention in the short to medium term (<7 mo) based on a comprehensive and robust meta-analysis; however, this conclusion was based mainly on one review10 that included mostly adults, only looked at static stretches, and was not able to define or standardize the control condition of usual care. Two SRs identified in this study noted that there is limited or weak evidence to support the effectiveness of passive stretching for improving ROM and spasticity.40,41 Although there was limited/weak evidence, generally studies showed an increase in ROM poststretching or a loss of ROM after stretching stopped.40,41 One author concluded that it appeared that sustained stretching of longer duration was preferable to improve range of movements and to reduce spasticity of muscles around the targeted joints.41 No studies identified by this review reported the effect of positioning, ROM, or stretching programs on the activity or participation of children with neuromuscular disabilities.
Clinical Recommendation From the Evidence. There is insufficient evidence to make any recommendations in regard to the use of positioning or stretching programs; therefore, all are yellow interventions in accordance with the EATLS.
Supported Standing Programs
Evidence. Six SRs evaluated the effectiveness of supported standing programs.29,35,38,40,42,43 In this review, 1 SR was identified that evaluated the effectiveness of static weight-bearing in children with CP. It found that other than the findings of increased BMD and temporary reduction in spasticity, there is limited evidence supporting the intervention because of weak research methodology.43 Another study, we identified assessed the effectiveness of standing programs for children with atypical development, concluded that there is evidence that the intervention positively affects BMD, hip stability, ROM of the hip knee and ankle and spasticity.42 The authors concluded that to see these positive results, standing program dosage should be between 30 and 90 min/d.42 The most consistent finding reported in all studies we identified is that there is very low evidence that supported standing programs using an external device increase lower limb and vertebral BMD.29,35,38,40,42,43 There is no evidence that improved BMD prevents pathological fractures.35,43 Although there is insufficient evidence to refute or support the effectiveness of this intervention for promoting ROM and/or preventing contractures, 2 studies showed a positive effect on hip ROM or hip migration percentage.38,42 Two SRs we identified also found that there is evidence that supported standing programs temporarily decrease spasticity.42,43 This review identified no quantitative research on the effect of supported standing programs on the activity or participation of children with neuromuscular disabilities.
Clinical Recommendation From the Evidence. Using the EATLS, supported standing programs using an external device are rated as green interventions for increasing lower extremity BMD. Supported standing programs are yellow interventions for all other outcome measures.
Summary of Main Findings
The primary outcome measure for this SR was flexibility. For this outcome measure, the strongest evidence found in this review was for the use of casting to increase passive ankle dorsiflexion in the short-term. Conflicting evidence was found for the use of orthotics to prevent contractures in populations of children with CP, Charcot-Marie-Tooth, and Duchenne muscular dystrophy. Insufficient evidence was found to support or refute stretching/positioning programs or supported standing programs for the increase of ROM or prevention of contractures.
In addition to the flexibility measure, the strongest evidence identified in this review supports the use of orthoses for improving gait kinetics and kinematics while worn, and for supported standing programs for improving lower extremity BMD. The majority of research has focused on assessing short-term outcomes of the lower extremity at the body structure and function level. Of the limited reporting of activity and participation measures, no treatment effect was found when compared with controls.30–32 For example, although AFOs that restrict plantarflexion have a favorable effect on improving gait kinetics and kinematics, we identified no consistent research to indicate this increases the child's quality of life (Table 2). Treatment interventions varied immensely with a wide range of the following interventions being recorded: casting protocols, orthoses configurations, prescription of stretching programs, and supported standing program equipment. Adverse events were rarely mentioned and when recorded they included skin irritation and pain from casting and orthoses.31,32,34,36 Only one review35 has looked at the same breadth of interventions for multiple pediatric disabilities as this current research has presented. The comparable review35 included lower levels of evidence and no quality assessment was conducted. All other studies have focused on 1 population28–34,37,39,40,41,43 and/or on 1 or 2 stretch interventions.28,30–34,36,37,39,41–43 To date, this is the most comprehensive review on all stretch interventions for a large population treated by pediatric physical therapists. In addition, this is only the second review article29 on pediatric stretch interventions that has used the GRADE approach and the EATLS to categorize treatments based on the quality and strength of evidence. These knowledge translation tools allow clinicians to quickly implement research into practice. This is valuable, as previous research has shown that although pediatric physical therapists have a positive attitude toward evidence-based practice, they routinely self-report that they are unable to implement this information into practice.44 Although this review confirms the effectiveness of some stretch interventions, it is clear that there is still a large gap between clinical practice, treatment rationales, and the available evidence. More specifically, there is no high-quality evidence to support or refute that stretch interventions can avoid or defer surgery, decrease complications such as contractures, or promote function as reported by pediatric physical therapists.8,9 The yellow rating of many interventions identified in this review highlights the importance of clinicians using not only the best-available research regarding function and basic science muscle/tendon physiology but also the 2 other tenets of evidence-based practice: clinician experience and patient values.45,46 For example, clinicians should consider their historical knowledge about how disuse, muscle imbalances, and immobility affect flexibility and function.45 When the evidence in the literature is unclear, clinicians are encouraged to use outcome measures specific and meaningful to their clients to track the effectiveness of and to modify treatment as necessary.46
Although this review only included Levels 1 and 2 evidence, 10 of the 12 identified SRs did incorporate lower levels of evidence. There is a possibility that some relevant studies of lower levels of evidence may have been missed were they not included in these reviews, and could have provided additional insight into this research question. However, although case studies or series may have demonstrated more polarized results for the individuals (larger positive or negative effects) compared with groups (because of regression to the mean), it is unlikely that their results could have been generalized to the population investigated in this review. Inclusion criteria requiring 1 primary outcome measure of flexibility may have also excluded articles that only looked at other measures of body functions and structures or primarily on activity or participation measures. This may mean that other benefits of these interventions, apart from those on flexibility measures, were missed. The large umbrella search strategy, however, should have still found these articles, and only 3 studies were excluded only on the premise of not having a flexibility measure (see Table, Supplemental Digital Content 1, which lists all articles excluded at the full-text level).
Implications for Research and Clinical Practice
Assessment. Future research is needed to verify whether stretch interventions have a clinically significant effect on the activity, participation, and quality of life of children with neuromuscular disabilities. Several of the included studies note the need to use outcome measures of activity and participation that are meaningful for the child.28,29,32,35,37,38,43 This is significant in that both clinicians and parents identify quality of life as the most important domain to assess in this population.47 The majority of stretch intervention studies have focused at the body function and structure level, and it is rare to see a study that assessed all ICF levels at once. When functional measurements were assessed, no difference between experimental and control groups were detected.30–32 The intervention itself might be ineffective for functional changes; however, an alternative explanation is that the activity and participation measures currently designed may not be reliable, sensitive, or valid enough to detect change during the course of intervention(s). For example, in SRs of activity and participation measures in children with CP, many of the tools did not have sound psychometric properties.48,49 Thus, it is evident from the literature that even if clinicians and researchers start to implement more activity and participation assessment, further development of outcome measures is needed to address the reliability, sensitivity, and validity of these tools.
Treatment. Another observation when functional measures were assessed was that improvements at the body function level did not correlate with improvements in activity or participation.29,40 A probable explanation for the lack of carryover between ICF levels uses the motor learning principles of bottom-up and top-down interventions. Bottom-up treatment interventions focus on remediating an underlying impairment or motor deficit, whereas top-down treatment interventions typically use a problem-solving approach to motor skill development or task-specific interventions focused on the direct teaching of a skill.50 Employing an intervention at the bottom (eg, body structure and function level) and assuming that there is overflow40 or translation upstream to the activities level29 is not as logical as implementing an intervention aimed at improving activity and participation and then assessing for change at this level. Distinguishing bottom-up and top-down interventions is important as a recent SR found that the majority of green interventions for children with CP were top-down therapy approaches, aimed at improving activities performance.29 Thus, not only should clinicians and researchers assess functional and quality of life measures with sensitive tools, but functional interventions should be considered in conjunction with interventions targeted at the body structure and function level. For the clinician, assessment and treatment should be organized according to findings and hypotheses about impairments and limitations of the specific individual client in the context of best evidence, clinician experience, and client values.
Research Considerations. Future studies also need to address study design and rigor to improve the quality of the research and thus the applicability of the knowledge into clinical practice. More specifically, studies need to account for dropouts, blind assessors and report on concurrent activities and therapies. Researchers also need to look at long-term follow-up (eg, >1 y) and report on adverse events and safety considerations, as the majority of the included studies did not address these events. This is paramount because the majority of stretch interventions are considered yellow interventions, meaning that there is low-quality or conflicting evidence supporting the effectiveness of the interventions.25,26 For all interventions, especially where there is insufficient evidence to support or refute its effectiveness, the possible desirable effects of an intervention have to be weighed against its undesirable effects, such as adverse events and burden of care.12 Possible negative effects of a stretch intervention may include increased emotional burden on caregivers,13,16 pain,15,31,34,36 sleep disturbances,13,14,31 or even a negative physiological effect.51 Research on other pediatric populations has demonstrated better coreporting of primary outcome physiological measures along with secondary activity and participation outcome measures including adverse events. For example, extensive research has been conducted on the clinical effectiveness of orthoses for adolescent idiopathic scoliosis along with the coreporting of pain, quality of life, psychological issues, and self-image.52,53 Similar research should be conducted for children with neuromuscular disabilities. In order to perform high-quality clinical trials, researchers should conduct sample size calculations a priori to provide adequate power. This was only mentioned in one of the included studies of this review.31 The reporting of sample size calculation in physical medicine and rehabilitation research has been identified as being inadequate given current publication guidelines.54 Where RCTs are unethical or impossible to conduct,10 long-term prospective cohorts that address the previously identified methodological weaknesses may be useful. Multicenter trials organized by using databases, such as CP registries, can acquire the large sample sizes needed for this type of research. Following a large group of children over a long period, while using sensitive outcome measures and employing top-down interventions, will likely yield the most meaningful evidence.
Three green interventions were found: ankle casting for improving passive ankle dorsiflexion, orthoses for improving gait kinetics and kinematics, and supported standing programs for improvement of lower extremity BMD. All other stretch interventions are yellow interventions for a particular outcome measure, meaning that clinicians can continue to implement these interventions but need to use sensitive outcome measures to see whether these interventions have helped the child reach their goal. No red stretch interventions were found, indicating that, at this time, no interventions need to be discontinued on the basis of ineffectiveness or detrimental effects. Further investigation of stretch interventions is warranted because the gap between clinical practice and the lack of clear scientific evidence can have implications that could influence the future allocation and use of pediatric physical therapy services.
The authors thank the BC Centre of Ability, Queen Alexandra Centre for Children's Health, and the University of British Columbia for supporting this research, and Sarah Craig, MEd, for her assistance with editing.
1. Nordmark E, Hägglund G, Lauge-Pedersen H, Wagner P, Westbom L. Development of lower limb range of motion from early childhood to adolescence in cerebral palsy: a population-based study. BMC Med. 2009;7:65. doi:10.1186/1741-7015-7-65.
2. Gordon GS, Simkiss DE. A systematic review of the evidence for hip surveillance in children with cerebral palsy. J Bone Joint Surg Br. 2006;88(11):1492–1496. doi:10.1302/0301-620X.88B11.18114.
3. Ferrari A, Ferrara C, Balugani M, Sassi S. Severe scoliosis in neurodevelopmental disabilities: clinical signs and therapeutic proposals. Eur J Phys Rehabil Med. 2010;46(4):563–580.
4. Bourke-Taylor H, Cotter C, Stephan R. Young children with cerebral palsy: families self-reported equipment needs and out-of-pocket expenditure. Child Care Heal Dev. 2014;40(5):654–662. doi:10.1111/cch.12098.
5. Bowles D, Wasiak R, Kissner M, et al. Economic burden of neural tube defects in Germany. Public Health. 2014;128(3):274–281. doi:10.1016/j.puhe.2013.12.001.
6. Landfeldt E, Lindgren P, Bell CF, et al. The burden of Duchenne muscular dystrophy: an international, cross-sectional study. Neurology. 2014;83(6):529–536. doi:10.1212/WNL.0000000000000669.
7. Wingstrand M, Hägglund G, Rodby-Bousquet E. Ankle-foot orthoses in children with cerebral palsy: a cross sectional population based study of 2200 children. BMC Musculoskelet Disord. 15(327):1–7. doi:10.1186/1471-2474-15-327.
8. Wiart L, Darrah J, Kembhavi G. Stretching with children with cerebral palsy: what do we know and where are we going? Pediatr Phys Ther. 2008;20(2):173–178. doi:10.1097/PEP.0b013e3181728a8c.
9. Taylor K. Factors affecting prescription and implementation of standing-frame programs by school-based physical therapists for children with impaired mobility. Pediatr Phys Ther. 2009;21(3):282–288. doi:10.1097/PEP.0b013e3181b175cd.
10. Katalinic OM, Harvey LA, Herbert RD, Moseley AM, Lannin NA, Schurr K. Stretch for the treatment and prevention of contracture (Review). Cochrane Database Syst Rev. 2010;9:CD007455. doi:10.1002/14651858.CD007455.pub2.
11. Skalsky AJ, McDonald CM. Prevention and management of limb contractures in neuromuscular diseases. Phys Med Rehabil Clin N Am. 2012;23(3):675–687.
12. Andrews J, Guyatt G, Oxman AD, et al. GRADE guidelines: 14. Going from evidence to recommendations: the significance and presentation of recommendations. J Clin Epidemiol. 2013;66(7):719–725. doi:10.1016/j.jclinepi.2012.03.013.
13. Gough M. Continuous postural management and the prevention of deformity in children with cerebral palsy: an appraisal. Dev Med Child Neurol. 2009;51(2):105–110. doi:10.1111/j.1469-8749.2008.03160.x.
14. Lloyd C, Logan S, McHugh C, et al. Sleep positioning for children with cerebral palsy [Protocol]. Cochrane Database Syst Rev. 2012;10. doi:10.1002/14651858.CD009257.
15. Swiggum M, Hamilton ML, Gleeson P, Roddey T. Pain in children with cerebral palsy: implications for pediatric physical therapy. Pediatr Phys Ther. 2010;22(1):86–92. doi:10.1097/PEP.0b013e3181cd18a7.
16. Hutton E, Coxon K. Posture for learning: meeting the postural care needs in mainstream primary schools in England a research into practice exploratory study. Disabil Rehabil. 2011;33(19–20):1912–1924. doi:10.3109/09638288.2010.544837.
17. Craig J, Hilderman C, Misovic R, Wilson G. The effectiveness of stretching programs, bracing, casting and orthoses for improving or maintaining body functions and structure, activities, or participation in children or youth with neuromuscular disabilities: a systematic review. PROSPERO: International Prospective Register of Systematic Reviews. http://www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42014013807
. Updated March 13, 2015. Accessed May 20, 2015.
18. OCEBM Levels of Evidence Working Group. The oxford 2011 levels of evidence. Oxford Centre for Evidence-Based Medicine. http://www.cebm.net/index.aspx?o=5653
. Updated 2011. Accessed March 4, 2015.
19. American College of Sports Medicine. ACSM's Guidelines for Exercise Testing and Prescription. 8th ed. Philadelphia, PA: Lippincott Williams and Wilkins; 2010:8.
21. Higgins JPT, Green S, eds. Table 8.5.a: The Cochrane Collaboration's Tool for Assessing Risk of Bias: Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. The Cochrane Collaboration. http://www.cochrane-handbook.org
. Updated March 2011. Accessed March 5, 2015.
22. Higgins JPT, Green S, eds. 8.3.3: Quality Scales and Cochrane Reviews: Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. The Cochrane Collaboration. http://www.cochrane-handbook.org
. Updated March 2011. Accessed March 5, 2015.
24. Balshem H, Helfand M, Schünemann, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011;64(4):401–406. doi:10.1016/j.jclinepi.2010.07.015.
25. Novak I, McIntyre S. The effect of education with workplace supports on practitioners' evidence. Aust Occup Ther J. 2010;57(6):386–393. doi:10.1111/j.1440-1630.2010.00861.x.
27. Moher D, Liberati A, Tetzlaff J, Altman DG. The PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Med. 2009;6(7):e1000097. doi:10.1371/journal.pmed1000097.
28. Autti-Rämö I, Suoranta J, Anttila H, Malmivaara A, Makela M. Effectiveness of upper and lower limb casting and orthoses in children with cerebral palsy. Am J Phys Med Rehabil. 2006;85(1):89–103. doi:10.1097/01.phm.0000179442.59847.27.
29. Novak I, McIntyre S, Morgan C, et al. A systematic review of interventions for children with cerebral palsy: state of the evidence. Dev Med Child Neurol. 2013;55(10):885–910. doi:10.1111/dmcn.12246.
30. Hyde S, FlŁytrup I, Glent S, et al. A randomized comparative study of two methods for controlling tendo achilles contracture in Duchenne muscular dystrophy. Neuromuscul Disord. 2000;10(4–5):257–263.
31. Maas J, Dallmeijer A, Huijing P, et al. A randomized controlled trial studying efficacy and tolerance of a knee ankle orthosis used to prevent equinus in children with spastic cerebral palsy. Clin Rehabil. 2014;28(10):1025–1038. doi: 10.1177/0269215514542355.
32. Rose KJ, Raymond J, Refshauge K, North KN, Burns J. Serial night casting increases ankle dorsiflexion range in children and young adults with Charcot-Marie-Tooth disease: a randomised trial. J Physiother. 2010;56(2):113–119.
33. Refshauge KM, Raymond J, Nicholson G, van den Dolder PA. Night splinting does not increase ankle range of motion in people with Charcot-Marie-Tooth disease: a randomised, cross-over trial. Aust J Physiother. 2006;52(3):193–199.
34. Blackmore A, Boettcher-Hunt E, Jordan M, Chan MD. A systematic review of the effects of casting on equinus in children with cerebral palsy: an evidence report of the AACPDM. Dev Med Child Neurol. 2007;49:781–790.
35. Effgen SK, McEwen IR. Review of selected physical therapy interventions for school age children with disabilities. Phys Ther Rev. 2008;13(5):297–312. doi:10.1179/174328808×309287.
36. Lannin N, Novak I, Cusick A. A systematic review of upper extremity casting for children and adults with central nervous system motor disorders. Clin Rehabil. 2007;21(11):963–976. doi:10.1177/0269215507079141.
37. Figueiredo EM, Ferreira GB, Maia Moreira RC, Kirkwood RN, Fetters L. Efficacy of ankle-foot orthoses on gait of children with cerebral palsy: systematic review of literature. Pediatr Phys Ther. 2008;20(3):207–223. doi:10.1097/PEP.0b013e318181fb34.
38. Montero SM, Gómez-Conesa A. Technical devices in children with motor disabilities: a review. Disabil Rehabil Assist Technol. 2014;9(1):3–11. doi:10.3109/17483107.2013.785034.
39. Neto HP, Collange Grecco LA, Galli M, Santos Oliveira C. Comparison of articulated and rigid ankle-foot orthoses in children with cerebral palsy: a systematic review. Pediatr Phys Ther. 2012;24(4):308–312. doi:10.1097/PEP.0b013e318268a4fc.
40. Franki I, Desloovere K, De Cat J, et al. The evidence-base for basic physical therapy techniques targeting lower limb function in children with cerebral palsy: a systematic review using the International Classification of Functioning, Disability and Health as a conceptual framework. J Rehabil Med. 2012;44(5):385–395. doi:10.2340/16501977-0983.
41. Pin T. The effectiveness of passive stretching in children with cerebral palsy. Dev Med Child Neurol. 2006;48(2):855–862.
42. Paleg GS, Smith BA, Glickman LB. Systematic review and evidence-based clinical recommendations for dosing of pediatric supported standing programs. Pediatr Phys Ther. 2013;25(3):232–247. doi:10.1097/PEP.0b013e318299d5e7.
43. Pin TW. Effectiveness of static weight-bearing exercises in children with cerebral palsy. Pediatr Phys Ther. 2007;19(1):62–73. doi:10.1097/PEP.0b013e3180302111.
44. Schreiber J, Stern P, Marchetti G, Provident I, Turocy P. School-based pediatric physical therapists' perspectives on evidence-based practice. Pediatr Phys Ther. 2008;20(4):292–302. doi:10.1097/PEP.0b013e31818bc475.
45. Gorter JW, Becher J, Oosterom I. To stretch or not stretch in children with cerebral palsy [Letter to the Editor]. Dev Med Child Neurol. 2007;49:797–800.
46. Sullivan K. Evidence for physical therapist practice: how can we reconcile clinical guidelines and patient-centered care? J Neurol Phys Ther. 2010;34:52–53. doi:10.1097/NPT.0b013e3181d055c2.
47. Vargus-Adams JN, Martin LK. Measuring what matters in cerebral palsy: a breadth of important domains and outcome measures. Arch Phys Med Rehabil. 2009;90(12):2089–2095. doi:10.1016/j.apmr.2009.06.018.
48. Harvey A, Robin J, Morris ME, Graham HK, Baker R. A systematic review of measures of activity limitation for children with cerebral palsy. Dev Med Child Neurol. 2008;50(3):190–198. doi:10.1111/j.1469-8749.2008.02027.x.
49. Sakzewski L, Boyd R, Ziviani J. Clinimetric properties of participation measures for 5- to 13-year-old children with cerebral palsy: a systematic review. Dev Med Child Neurol. 2007;49(3):232–240. doi:10.1111/j.1469-8749.2007.00232.x.
50. Barnhart RC, Davenport M, Epps S, Nordquist VM. Developmental coordination disorder. Phys Ther. 2003;83(8):722–731.
51. Chang CH, Chen YY, Wang CJ, Lee ZL, Kao H-K, Kuo KN. Dynamic displacement of the femoral head by hamstring stretching in children with cerebral palsy. J Pediatr Orthop. 2010;30(5):475–478. doi:10.1097/BPO.0b013e3181df8948.
52. Negrini S, Minozzi S, Bettany-Saltikov J, et al. Braces for idiopathic scoliosis in adolescents. Cochrane Database Syst Rev. 2010;1:CD006850. doi:10.1002/14651858.CD006850.pub2.
53. Weinstein S, Dolan L, Wright J, Dobbs M. Effects of bracing in adolescents with idiopathic scoliosis. N Engl J Med. 2013;369(16):1512–1521. doi:10.1056/NEJMoa1307337.
54. Latif L, Amadera J, Pimentel D, Pimentel T, Fregni F. Sample size calculation in physical medicine and rehabilitation: a systematic review of reporting, characteristics, and results in randomized controlled trials. Arch Phys Med Rehabil. 2011;92(2):306–315. doi:10.1016/j.apmr.2010.10.003.
APPENDIX Search Strategy
The following search strategy was employed along with relevant MESH terms for each database: “stretches” or “stretch” or “stretching” or “range of motion” or “passive range of motion” or “active range of motion” or “positioning” or “casting” or “casts” or “cast” or “splints” or “splinting” or “splint” or “bracing” or “braces” or “brace” or “yoga” or “orthotics” or “orthoses” or “orthotic” (and) “passive range of motion” or “active range of motion” or “joint mobility” or “flexibility” or “flexible” or “pain” or “quality of life” or “spastic” or “spasticity” or “activities of daily living” or “activity of daily living” or “participation” or “contracture” or “contractures.”