Effectiveness of Partial Body Weight-Supported Treadmill Training on Various Outcomes in Different Contexts among Children and Adolescents with Cerebral Palsy: A Systematic Review and Meta-Analysis

The efficiency of partial body weight-supported treadmill training (PBWSTT) for treating various conditions in children and adolescents with cerebral palsy (CP) in diverse contexts of rehabilitation, households, or schools is unknown. The major objective of this systematic review and meta-analysis was to analyze the effectiveness of PBWSTT on various outcomes in different contexts among children and adolescents with CP. We incorporated full-text, randomized controlled trial studies that specifically assessed the effects of PBWSTT walking, motor function, stride, balance, and endurance in children and adolescents aged 3 to 18 years with CP. The literature search was carried out using Google Scholar, PubMed, Web of Science, CINAHL Plus, Scopus, PEDro, and ResearchGate databases. The methodological quality was evaluated using a Cochrane risk of bias instrument. A meta-analysis of pooled data from 10 studies with 255 participants demonstrated that PBWSTT for 4–12 weeks in rehabilitation (mean difference [MD] = 1.94, 95% confidence interval [CI] = 1.40—2.48, p < 0.0001), at home or in a school context (MD = 13.5, 95% CI = 13.9—16.0, p < 0.0001), was significantly effective for treating various conditions in children and adolescents suffering with CP. The period of 4–12 weeks of PBWSTT in rehabilitation and at-home/school settings is effective on various outcomes in children or adolescents with CP.


Introduction
Cerebral palsy (CP) is a significant global health issue that affects between one and four of every thousand live births worldwide [1,2].In the United States of America, approximately 2.4 of every 1000 children are affected by CP [3].Additionally, the prevalence of CP can vary among males and females.For instance, males generally experience a higher prevalence of CP compared to females, with males in Europe having a rate that is 1.3 times higher [4].The occurrence of CP among Arabic-speaking countries has been documented as 1.8 cases per 1000 live births [5,6].For example, Saudi Arabia has one of the lowest rates of CP at 1.6 per 1000 live births [5], although this is still higher than the global average [7].
While there is currently no known cure for CP, certain supportive therapies, such as physical [8] and occupational therapy [9], can help partially manage it, specifically treadmill training [10].Treadmill training has recently drawn attention as a helpful gait training strategy, in addition to traditional methods, and has many benefits [11].Partial body weight-supported treadmill training, abbreviated as PBWSTT, is a specific type of training where a person is supported by a harness or other supportive device that reduces the amount of weight they bear while walking on the treadmill [12].It has been demonstrated

Materials and Methods
We followed the recommendations and checklists of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses, abbreviated as PRISMA [17], for the development of the review [18].The Cochrane Collaboration's recommendations were followed for conducting a literature review, applying selection criteria, extracting data, and performing statistical analysis [19].This study, as a literature review, is exempted from Institutional Review Board approval.The registration of this systematic review has been documented in PROSPERO, with the number code CRD42023405443.

Search Strategy
Systematic searches of randomized controlled trial studies were conducted on the Web of Science, PubMed, CINAHL Plus, Google Scholar, the Physiotherapy Evidence Database (PEDro), Scopus, and ResearchGate up to 28 September 2022.These databases were specifically selected for this review based on their reputations as professional databases relating to rehabilitation and physiotherapy.Search terms included specific terms, such as "children", "young", "adolescent" "cerebral palsy", "cerebral palsies", "infantile palsies", "treadmill training", "partial body weight treadmill", "body weight-supported training", "treadmill therapy", "partial body weight treadmill", and "PBWSTT".Before reading the complete text, articles were initially chosen based on their keywords, and then on their titles and abstracts.

Eligibility Criteria
The selection criteria for inclusion in this study encompassed articles composed in the English language, with all of them utilizing randomized controlled trials.Studies comprising children and teenagers with CP aged 3 to 18 years were included.Trials were eligible for inclusion that were available in full text, used PBWSTT as the exposure, and its effectiveness on various outcomes (such as walking, motor function, stride, balance, endurance, etc.), in different settings as an outcome.We excluded studies that used combined interventions (treadmill and other therapy) and where the outcome data were not clear.

Data Extraction
Based on the data given in the abstracts and the titles of the studies, we meticulously reviewed and removed any information that appeared to be irrelevant.Afterwards, we carefully examined the complete content of the remaining articles and compared them against the predetermined criteria for eligibility to individually identify and choose the pertinent ones.Subsequently, the final set of eligible studies was chosen based on specific inclusion/exclusion criteria.In this meta-analysis, we focused on incorporating previous studies that reported the mean and ninety-five percent confidence interval (95% CI) to assess the efficiency of PBWSTT on various outcomes in different settings.

Assessment of Methodological Quality
We assessed the methodology of each included study independently, employing the PEDro scale.This particular scale was selected because of its utilization in clinical trials and its important role in assessing the methodological components of such trials.It has also been used in rehabilitative therapies and for restorative purposes and results.Study quality is rated using the PEDro scale, which has a scale from 0 to 10.The experts decided on the PEDro score elsewhere [4], classifying it as 0-3 for poor, 4-5 for fair, 6-8 for good, and 9-10 for excellent.The PEDro score has shown "fair" to "excellent" inter-rater reliability (ICC, Intraclass Correlation Coefficient = 0.53-0.91)for clinical trials.Similarly, this reliability for the specific items of the PEDro scale in physiotherapy trials varies from "fair" to "almost perfect" (k = 0.36-1.00).The PEDro scale's validity and reliability have been documented in the past [20].

Assessment of the Risk of Bias
We employed the Cochrane Collaboration tool to evaluate bias risk in each of the randomized controlled trials [21].The tool is structured with a predefined collection of bias categories that focus on different aspects of trial planning, implementation, and reporting.Each evaluation conducted using the tool focuses on a particular randomized controlled trial result.A set of inquiries (designated as "signaling questions") within each domain tries to extract details regarding characteristic features of the trial that are relevant to the risk of bias.Using the feedback to the signaling queries, an algorithm makes recommendations about the potential bias risk associated with each domain.The algorithm classifies the bias risk as either "High", "Low", or "Unclear" [21].The Cochrane Collaboration tool's validity and reliability were fair [22].

Statistical Analysis
We looked at multivariate-adjusted data on various outcomes, presented as a mean and 95% CI, to assess the effectiveness of PBWSTT in any setting (primary outcome) and its differences in various contexts (secondary outcome) among patients with CP.The risk of bias summary and risk of bias graph are displayed in traffic light form.The fixed-effect model combined the findings if there was additional relevant study heterogeneity.In other situations, a random-effects model was utilized.The values were shown as mean difference (MD) and 95% CI.The I 2 statistic was utilized for assessing heterogeneity.The degree of study heterogeneity was classified as moderate (30-59%), significant (60-89%), or great (90-100%) [23].Funnel plots were employed to assess the likelihood of bias of publication affecting both the primary and secondary outcomes [24].All data analyses were carried out using RevMan version 5.4 software (the Cochrane Collaboration, Denmark) for Windows.Statistical significance was determined as p < 0.05, a p-value below 0.05.

Systematic Review
A pool of 232 records was found after database searches (Figure 1).The abstracts and titles of these articles were reviewed.After a manual recheck and deduplication, 227 main titles were reviewed, and 207 of them were eliminated.Figure 1 presents the reason for the elimination.Of twenty studies, ten studies (four in rehabs, four in schools, and two with at-home settings) [3,16,[25][26][27][28][29][30][31][32] with 255 children were found to be potentially appropriate for the review and included in a full-text review for qualitative synthesis and quantitative analysis.The remaining 10 studies were disqualified for a variety of reasons, including combination therapies, age over 18 years, lack of randomized control trial research, and unclear outcome data.

Systematic Review
A pool of 232 records was found after database searches (Figure 1).The abstracts and titles of these articles were reviewed.After a manual recheck and deduplication, 227 main titles were reviewed, and 207 of them were eliminated.Figure 1 presents the reason for the elimination.Of twenty studies, ten studies (four in rehabs, four in schools, and two with at-home settings) [3,16,[25][26][27][28][29][30][31][32] with 255 children were found to be potentially appropriate for the review and included in a full-text review for qualitative synthesis and quantitative analysis.The remaining 10 studies were disqualified for a variety of reasons, including combination therapies, age over 18 years, lack of randomized control trial research, and unclear outcome data.

Quality and Risk of Bias
The articles selected for analysis [3,16,[25][26][27][28][29][30][31][32] received PEDro values ranging from 3 to 8, along with 6 median scores, indicating articles of good quality.The risk of bias in the selected studies [3,16,[25][26][27][28][29][30][31][32] is shown in Figure 2. Every single one of these studies received a 100% rating for low-risk reporting and selection biases.All of these studies, however, had at least one element that was highly susceptible to bias.Because the participant, assessor, and therapist were not blinded to treatment allocation, there is a 100% chance of performance and detection biases in all these included studies.A substantial risk of attrition bias (40%) exists in four studies [3,29,30,32] because of insufficient follow-up or data.Due to insufficient outcome data, three studies [3,16,26] were assigned a high risk of

Characteristic Features of the Included Studies
Characteristics of the included studies are shown in Table 1.In total, 255 participants (38.4% boys), with an average age of 9 years, were split into intervention (n = 115) or

Characteristic Features of the Included Studies
Characteristics of the included studies are shown in Table 1.In total, 255 participants (38.4% boys), with an average age of 9 years, were split into intervention (n = 115) or control (n = 100) groups at random, in a rehab center, a home, or a school, for the 10 articles that were included.The sample size of all included studies (n = 10) was smaller than the required size [33].Most of the patients had a mean age of 9 years and had spastic-type CP.The majority of study settings were rehabilitation (n = 4) and school (n = 4), followed by home (n = 2).Most studies were carried out in Australia (n = 3), followed by the United States (n = 2).The details of the intervention, treatment time, assessment, outcome variables, and summary of findings have been presented in Table 2. Most of the individuals in the experimental group underwent training sessions, on average lasting 20 to 30 min in the beginning from 4 to 12 weeks and 12 to 24 weeks.The findings from most studies [3,[25][26][27][28][30][31][32] show that 4-12 weeks of PBWSTT is significantly effective at improving walking speed, gross motor functions, stability, and balance in various contexts for children with CP.However, two studies [16,29] show there was no significant difference between PBWSTT and exercise in improving gait speed, cadence, or endurance for children with CP in a home or school environment.Specifically, three studies [28,30,31] that used PBWSTT therapies showed an improvement in walking speed in CP patients.In a Turkish rehabilitation center, one intervention [28] consisted of twenty 45 min sessions held twice a week for four weeks.In an Australian school context, the other intervention [30] involved a walking test lasting 10 weeks.An investigation [31] revealed that after 36 sessions of PBWSTT, i.e., three times a week for thirty minutes for 12 weeks, in the Saudi Arabian rehabilitation environment, children with CP could safely and effectively walk.Following PBWSTT, children with CP had improved gross motor capabilities in three investigations [3,25,27].One of these three trials used PBWSTT for 20 min per week for 12 weeks in a Taiwanese rehabilitation environment, involving two or three sessions [25].In a Chinese school, another investigation [3] ran for 262 min throughout 14 PBWSTT sessions over 12 weeks.In an Australian school context, the study conducted two 30 min PBWSTT sessions every week for eight weeks [27].

Discussion
The effectiveness of PBWSTT, particularly in rehabilitation, at-home, or school settings, for children aged 3 to 18 years with CP was assessed for the first time in this study.The evidence from this review demonstrated that 4-12 weeks of PBWSTT was effective, especially in rehabilitation, at-home, or school settings.There were insufficient data to

Discussion
The effectiveness of PBWSTT, particularly in rehabilitation, at-home, or school settings, for children aged 3 to 18 years with CP was assessed for the first time in this study.The evidence from this review demonstrated that 4-12 weeks of PBWSTT was effective, especially in rehabilitation, at-home, or school settings.There were insufficient data to determine the effects of PBWSTT over a period of 12-24 weeks.
In the qualitative synthesis, studies from Turkey [23], Australia [30], and Saudi Arabia [31] indicated that PBWSTT over 4-12 weeks might increase walking speed in children with CP in rehabilitation or school settings.The three other trials, carried out in Taiwan [25], China [3], and Australia [27], demonstrated that PBWSTT exhibited enhanced gross motor skills in identical settings.In an Indian rehabilitation context [32], PBWSTT resulted in a considerable improvement in walking speed, cadence, stride, and gross motor function.Furthermore, three investigations [16,26,29] found no evidence of a significant effect of PBWSTT on these results.
The evidence from this review is consistent with the best evidence that was available in 2019 [34] and earlier [35,36].The study demonstrated that PBWSTT helps CP children function better and accomplish tasks more effectively.In children suffering from CP, PBWSTT has been documented to improve walking speed in a previous evaluation of 41 trials, including 11 randomized trials [37].Another review of 24 randomized trials found that PBWSTT was more effective than resistance training at enhancing gait speed in children with CP [38].Additionally, the outcomes of previous studies revealed that PBWSTT enhances children's walking ability [28] and improves gross motor skills [3].A prior systematic review, however, did not conclude that PBWSTT helps children with CP [39].Possible reasons could be the lack of participants, the individuals' varying degrees of skill, and the poor quality of the trials.
The results of this review, however, consist of a prior clinical trial that investigated the outcomes of a school-based, twice-weekly PBWSTT program for six weeks on the endurance and gait speed of children with CP [30].The trial's findings suggested that PBWSTT would be a good gait training alternative and it might be possible to implement a similar program in a classroom setting.The results from a two-period randomized crossover design with repeated measures study show that PBWSTT improved gross motor skills in low-functioning children and adolescents with nonspastic CP who were enrolled in special schools [3].However, the results from a different randomized trial carried out at a special-needs school in Singapore revealed that PBWSTT is not more effective than overground training in enhancing the qualities of walking and other functions in children with mild to moderate CP [27].A possible reason for these contradictory results in a learning environment could be attributed to interventional parameters, such as speed, body weight support, time per session, and the frequency and duration of use of various types of treadmill training.It is currently impossible to speculate on the criteria that might lead to successful results.
The notable strength of this study lies in the fact that all of the studies incorporated were randomized trials, a design which, under similar conditions, is associated with a high level of evidence [40].However, there are certain limitations to this study.First, there were methodological issues with the studies included in this evaluation, including a high risk of bias brought on by the lack of blinding.Second, PBWSTT is not well established as the most commonly utilized gait speed outcome [31,41], especially for modest clinically relevant differences.Future studies should aim to adopt a consistent methodology and to pinpoint clinically significant changes.Third, in most rehabilitation trials, it is challenging to entirely blind therapists or participants, making a PEDro score of 8 the best possible result for this review.Finally, the reduced sample size and uneven quality of the research make this study limited.Therefore, it is important to interpret these results with caution.
The findings of this research could facilitate multidisciplinary cooperation amongst rehabilitation, health, and education experts in order to tackle the intricacy of variables involved in the application of PBWSTT in various contexts.A pediatrician, physiotherapist, occupational therapist, and speech therapist are examples of an interdisciplinary team that may be included in the early stages of CP treatment [42].Every member of the team plays a crucial role in providing the affected child with care.The physiotherapist, for instance, assesses and treats muscular tone, strength, and gait (walking) [43].An occupational therapist evaluates the child's capacity for self-care and self-help skills, such as feeding and manual dexterity [9].The child's ability to communicate, understand speech, and talk is assessed by the speech therapist [44].

Conclusions
Our research demonstrates that children or adolescents with CP can benefit from 4 to 12 weeks of PBWSTT in a rehabilitation, at-home, or school setting.There is a lack of research to support the long-term effects of this intervention in many circumstances.Therefore, more research and high-quality randomized trials are undoubtedly necessary to establish the advantages and efficacy that back up the continued use of this intervention in these circumstances.Future investigations, particularly randomized trials with large sample numbers, must also include follow-up measures to ascertain whether children with CP will experience any long-lasting and meaningful advantages.

Figure 1 .
Figure 1.The flow of articles identified in the electronic database search.

Figure 1 .
Figure 1.The flow of articles identified in the electronic database search.

Figure 2 .
Figure 2. Risk of bias of all included randomized controlled trials.Positive symbols denote low bias risk, negative symbols indicate high bias risk, and question marks suggest unclear for bias risk [3,16,25-32].

Figure 2 .
Figure 2. Risk of bias of all included randomized controlled trials.Positive symbols denote low bias risk, negative symbols indicate high bias risk, and question marks suggest unclear for bias risk [3,16,25-32].

14 Figure 5 .
Figure 5. Funnel plot.The funnel is depicted by dashed lines, and the individual studies' effects are represented by open circles.

Figure 5 .
Figure 5. Funnel plot.The funnel is depicted by dashed lines, and the individual studies' effects are represented by open circles.

Table 1 .
Characteristic features of included studies.
Abbreviations: n, a number of participants, E, experimental group; C, control group; F, female; M, male; yr, year; CP, cerebral palsy.

Table 2 .
Intervention and main findings of included studies.