Application of Virtual Reality-Assisted Exergaming on the Rehabilitation of Children with Cerebral Palsy: A Systematic Review and Meta-Analysis

Background: Rehabilitation programs for children with cerebral palsy (CP) aim to improve their motor and cognitive skills through repeated and progressively challenging exercises. However, these exercises can be tedious and demotivating, which can affect the effectiveness and feasibility of the programs. To overcome this problem, virtual reality VR-assisted exergaming has emerged as a novel modality of physiotherapy that combines fun and motivation with physical activity. VR exergaming allows children with CP to perform complex movements in a secure and immersive environment, where they can interact with virtual objects and scenarios. This enhances their active engagement and learning, as well as their self-confidence and enjoyment. We aim to provide a comprehensive overview of the current state of research on VR exergaming for CP rehabilitation. The specific objectives are: To identify and describe the existing studies that have investigated the effects of VR exergaming on motor function and participation outcomes in children with CP. In addition, we aim to identify and discuss the main gaps, challenges, and limitations in the current research on VR exergaming for CP rehabilitation. Finally, we aim to provide recommendations and suggestions for future research and practice in this field. Methods: In June 2023, we conducted a systematic search on Scopus, Web of Science, PubMed, Cochrane, and Embase for randomized trials and cohort studies that applied VR-assisted exergaming to rehabilitating patients with CP. The inclusion criteria encompassed the following: (1) Randomized controlled trials (RCTs) and cohort studies involving the rehabilitation of children with CP; (2) the application of VR-based exergaming on the rehabilitation; (3) in comparison with conventional rehabilitation/usual care. The quality of the selected RCTs was evaluated using Cochrane’s tool for risk of bias assessment bias includes. Whereas the quality of cohort studies was assessed using the National Institutes of Health (NIH) tool. Results: The systematic search of databases retrieved a total of 2576 studies. After removing 863 duplicates, 1713 studies underwent title and abstract screening, and 68 studies were then selected as eligible for full-text screening. Finally, 45 studies were involved in this review (n = 1580), and 24 of those were included in the quantitative analysis. The majority of the included RCTs had a low risk of bias regarding study reporting, participants’ attrition, and generating a random sequence. Nearly half of the RCTs ensured good blinding of outcomes assessors. However, almost all the RCTs were unclear regarding the blinding of the participants and the study personnel. The 2020 retrospective cohort study conducted at Samsung Changwon Hospital, investigating the effects of virtual reality-based rehabilitation on upper extremity function in children with cerebral palsy, demonstrated fair quality in its methodology and findings. VR-assisted exergaming was more effective than conventional physiotherapy in improving the Gross Motor Function Measurement (GMFM)-88 score (MD = 0.81; 95% CI [0.15, 1.47], p-value = 0.02) and the GMFM walking and standing dimensions (MD = 1.45; 95% CI [0.48, 2.24], p-value = 0.003 and MD = 3.15; 95% CI [0.87, 5.42], p-value = 0.007), respectively. The mobility and cognitive domains of the Pediatric Evaluation of Disability Inventory score (MD = 1.32; 95% CI [1.11, 1.52], p-value < 0.001) and (MD = 0.81; 95% CI [0.50, 1.13], p-value < 0.0001) were also improved. The Canadian Occupational Performance Measure performance domain (MD = 1.30; 95% CI [1.04, 1.56], p-value < 0.001), the WeeFunctional Independence Measure total score (MD = 6.67; 95% CI [6.36, 6.99], p-value < 0.0001), and the Melbourne Assessment of Unilateral Upper Limb Function-2 score (p-value < 0.001) improved as well. This new intervention is similarly beneficial as conventional therapy in improving other efficacy measures. Conclusions: Our findings suggest that VR-assisted exergaming may have some advantages over conventional rehabilitation in improving CP children’s functioning and performance in daily life activities, upper and lower limb mobility, and cognition. VR-assisted exergaming seems to be as effective as conventional physiotherapy in the other studied function measures. With its potential efficacy, better feasibility, no reported side effects, and entertaining experience, VR-assisted exergaming may be a viable complementary approach to conventional physiotherapy in rehabilitating children with CP.


Introduction
Cerebral palsy (CP) is a permanent neurodevelopmental disorder [1].The term covers a group of conditions that result from a non-progressive lesion affecting the developing brain before, during, or after birth.This lesion causes motor impairment that affects mainly patients' movement and posture development [2][3][4].Motor impairment might also be accompanied by sensory, cognitive, or communication deficits or even behavioral abnormalities.Epilepsy and learning difficulties usually coincide with CP as well [5][6][7].Cerebral palsy prevalence has been increasing over the past years, and it is considered the most common cause of childhood neurological disability worldwide [2,8].
Body mass growth is normally accompanied by an increase in muscle mass, however, this increase in muscle mass is defective in children with CP [9].Children with CP have motor abnormalities, such as abnormal tone, weakness, and control that affect balance, posture, and coordination, and cause contractures and deformities.These symptoms impair their daily life activities [5,[10][11][12][13].
Children with CP need rehabilitation to increase muscle mass and motor function, which can boost their independence and daily life performance.Effective neurorehabilitation requires repeated and varied tasks with increasing difficulty, which can help the brain form new muscle synergies for specific goals [7,[14][15][16]].However, for several reasons, the difficulty comes in sustaining physiotherapy throughout the individual's whole life span.
Virtual reality technology provides an interactive computerized simulation of a real-world environment.The technology reacts with the users' real-time movement using threedimensional sensors, which are devices that can measure the strength or direction of a magnetic or electric field in three dimensions, i.e., along the x-, y-, and z-axes.They can be used for various applications, such as position detection, motion tracking, gesture recognition, and robotics.The users move and engage in the video game with all their senses (touch, hearing, and vision), and the sensors give direct encouraging feedback on their performance [22][23][24][25].VR exergaming has different forms of immersion.Non-immersive VR uses a small screen and a keyboard/mouse or a joystick.Semi-immersive VR uses a large screen and body parts.Immersive VR uses an HMD or a CAVE, headphones, and motion sensors.Immersive VR gives the feeling of immersion and presence.Commercial VR-based video games usually require fast movements and are not specific or task-oriented.Therefore, special video games are used for CP children's neurorehabilitation [26,27].Developing and validating VR games for CP is challenging.It needs collaboration among researchers, clinicians, game designers, and patients to ensure the safety, usability, efficacy, and ethics of VR.Some examples of VR games for CP are: RehabMaster™, a task-specific VR system that tracks upper limb movements with visual feedback; and RGS, a VR system that uses Kinect to monitor upper limb movements with multisensory feedback.
Home-based VR for CP can be convenient, accessible, affordable, and personalized, but also challenging due to technical, safety, supervision, and social issues.The effects of home-based vs. in-facility VR may vary by CP type and severity, VR quality and availability, therapist and caregiver support, and patient preferences and goals.More research is needed to compare different VR modalities for CP rehabilitation.
For rehabilitation programs to achieve fruitful results, sustainability and repetition are important [28].These are difficult to maintain over the individual's lifespan due to many barriers.Distance and cost could limit accessibility to physiotherapy centers.In addition, repetition over a long time limits children's engagement and motivation.The use of VR-assisted exergaming in the rehabilitation of CP children has come with many advantages.Participation in these games makes the performed tasks meaningful and enjoyable for those children and therefore enhances their active engagement with the training [19,29,30].In addition, VR technologies allow for the practice of tasks that require large physical space or cannot be performed in the real world.It can also safely simulate dangerous situations [31][32][33].Virtual reality-based training can be provided at a low cost at home [26,27,34,35].
Several studies [20,27,29,34] were conducted to investigate the effectiveness and safety of VR-based exergaming in the rehabilitation of children with CP.We aim to provide a comprehensive overview of the current state of research on VR exergaming for CP rehabilitation.The specific objectives are: To identify and describe the existing studies that have investigated the effects of VR exergaming on motor function and participation outcomes in children with CP.In addition, we aim to identify and discuss the main gaps, challenges, and limitations in the current research on VR exergaming for CP rehabilitation.Finally, we aim to provide recommendations and suggestions for future research and practice in this field.

Methods
We conducted this secondary research following the Cochrane Handbook for systematic reviews of intervention [36].The study was then reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) checklist [37].The specific details of the PRISMA checklist are in Supplementary File S1.

Literature Searching and Search Strategy
On 15 June 2023, we systematically searched five online databases and updated them on 5 September 2023.The following search strategy was used in all the databases: (("Cerebral palsy OR Cerebral palsies ") AND ("virtual reality" OR VR OR videogame OR videogames OR "video games" OR "virtual gam*" OR "augmented reality" OR "mixed reality" OR exergame* OR "Immersive virtual reality" OR IVR OR "Wii games" OR "Wii game" OR "Nintendo Wii" OR Wii OR "exergaming" OR "serious games" OR "interactive games")).No filters or limitations were applied to the search.A search on the reference lists of the selected eligible studies was conducted manually for any additional relevant studies.

Eligibility Criteria
The inclusion criteria encompassed the following: Specific studies were excluded from our analysis: (1) studies not published in English; (2) studies comprising solely of abstracts; (3) single-arm studies; and (4) studies that involved adults with cerebral palsy.

Selection Process
The retrieved search results were introduced to EndNote software V.20 and the duplicated results were deleted.Thereafter, we performed a screening of the titles and abstracts of the studies for initial eligibility, which means full-text screening if the paper seems to be eligible from its title and abstract.Then, the selected studies' full texts were carefully screened for final eligibility.
In the first phase, two reviewers independently screened the titles and abstracts of the retrieved records and excluded those that did not meet the eligibility criteria.In the second phase, the same two reviewers independently assessed the full texts of the remaining records and applied the same eligibility criteria.Any disagreement between the reviewers was resolved by discussion or consultation with a third reviewer.

Risk of Bias Assessment
The quality of the selected RCTs was evaluated by two investigators using Cochrane's tool for risk of bias assessment.Bias includes (1) random sequence generation; (2) allocation concealment; (3) blinding of participants and personnel; (4) blinding of outcome assessment; (5) incomplete outcome data; (6) selective reporting; and (7) other bias [36].Whereas the quality of cohort studies was assessed using the National Institutes of Health (NIH) tool [38].The tool was composed of 12 questions about population and sample size justification, the research question, control definition, inclusion criteria and cases, event time, blindness, and the reporting of confounders.The authors' opinion is classified as "good", "fair", or "poor" according to scores obtained during the assessment.We categorized the results of the appraisal into these three categories based on the following thresholds.Good: The study met all or most of the quality criteria and had minimal risk of bias; Fair: The study met some of the quality criteria and had a moderate risk of bias; Poor: The study met few or none of the quality criteria and had a high risk of bias.To ensure accuracy and consistency, any discrepancies during the evaluation process were resolved through discussions between the investigators or involving a third assessor.

Data Extraction
Summary and baseline participants' characteristics data were extracted from each study.These data include the study design, site, and arms, the inclusion criteria, the duration of VR-based exergaming sessions, the follow-up duration, the study's primary outcomes, and the study conclusion.In addition, these data described the studied type of CP, as well as participants' age, gender, Gross Motor Function Classification System (GMFCS) level, Manual Ability Classification System (MACS) level, and Pediatric Balance Scale (PBS) score.Finally, we extracted the following outcomes in the data extraction step: GMFM score, PBS score, PEDI score, COPM score, WeeFIM score, MA-2 score, QUEST score, and ABILHAND-Kids test score.In addition, the specific definitions of each outcome are described below.Two reviewers independently extracted data from each study and cross-checked their results for accuracy and consistency.Any discrepancy between the reviewers was resolved by discussion or verification with the original source.

Gross Motor Function Measurement (GMFM) Score
This outcome was measured through studies using two scales, the GMFM-88 scale and the GMFM-66 scale.The GMFM-88 scale consists of five dimensions: running and jumping, standing and walking, crawling and kneeling, sitting, and lying and rolling.
The performance of each item was measured on a scale of five points, with higher scores indicating better capacity.The GMFM-66 scale is a subset of the GMFM-88 scale acquired via software [39][40][41].

PBS Score
The PBS score measures the child's ability to balance dynamically.The scale consists of 14 items; each item has a score of four points with higher scores indicating better abilities [42].

Pediatric Evaluation of Disability Inventory (PEDI) Score
The PEDI scale assesses the child's functioning and skills in three domains: mobility, self-care, and social functioning.Each item in the PEDI is ranked on a scale of 0-100, with a higher score indicating better performance [43].

Canadian Occupational Performance Measure (COPM) Score
The COPM score measures the individual's performance and satisfaction regarding daily life activities.It assesses these aspects in three domains: self-care, productivity, and leisure.Each item is self-rated on a scale of 10 points, with a higher score indicating better performance and satisfaction [44].
2.6.5.WeeFunctional Independence Measure (WeeFIM) Score The WeeFIM instrument evaluates the child's functioning in daily life activities.Functioning in 18 items was assessed on a scale of 7 points, with a higher score indicating better performance.Six subscales are included in WeeFIM: self-care, mobility, social cognition, communication, transfer, and sphincter control [45].
2.6.6.Melbourne Assessment of Unilateral Upper Limb Function-2 (MA-2) Score This measure evaluates the function of a unilateral upper limb.Fourteen tasks are performed and recorded in video, thereafter, the movements are scored.The MA-2 score is calculated as a percentage of the maximum possible score in four subscales: range of motion, accuracy, dexterity, and fluency [46].

Quality of Upper Extremity Skills Test (QUEST) Score
The QUEST score is a measure of the upper limb function.The QUEST scores 33 upper limb activities on a scale of two points.Finally, the score is calculated as a percentage of the maximum score.Four domains are evaluated by the QUEST: grasps, dissociated movements, protective extensions, and weight bearing [47].
2.6.8.ABILHAND-Kids Test Score ABILHAND-Kids test assesses hand disability.The test evaluates the difficulty of performing 21 bimanual activities on a scale of three points.Each activity is scored, and a higher score indicates easier performance [48].

Data Synthesis
We carried out a quantitative synthesis which is defined as collecting numerical data and analyzing them using statistical methods, and it aims to produce objective, empirical data that can be measured and expressed in numerical terms.Analysis was carried out using RevMan software version 5.3 in the inverse variance method.The effect estimate was calculated as a mean difference (MD) and a 95% confidence interval (CI).The heterogeneity of the results across the studies was initially evaluated by direct inspection of the forest plot.Thereafter, the results of the I-squared (I 2 ) and chi-squared tests were checked.A random effect model was applied to the analysis wherever heterogeneity was detected.Then, we tried to resolve the heterogeneity by applying the "leaving one out method".Otherwise, analysis was conducted using the fixed-effect model [49].
Moreover, we performed a qualitative synthesis, which is defined as collecting nonnumerical data such as words, images, and sounds, and it aims to produce rich and detailed descriptions of the phenomenon being studied and to uncover new insights and meanings.

Literature Search
The systematic search of databases retrieved a total of 2576 studies.After removing 863 duplicates, 1713 studies underwent title and abstract screening, and 68 studies were then selected as eligible for full-text screening.Finally, 45 studies were judged eligible in this systematic review.A total of 24 studies were included in the quantitative synthesis [50-74], whereas 21 studies were only eligible for qualitative synthesis  (Figure 1).The manual search did not retrieve any additional relevant studies.

Description of the Included Studies
We included 44 RCTs and one retrospective cohort study [54] in this systematic review.The total number of enrolled CP patients was 1580 (801 for VR-based exergaming and 779 controls).The included studies were carried out in different countries: Turkey, Egypt, South Korea, Taiwan, Australia, Iran, India, the UK, Italy, Russia, Finland, Brazil, Saudi Arabia, Canada, USA, Spain, Belgium, Denmark, and the Netherlands.The duration of physiotherapy sessions ranged from 25 min to 90 min, and the frequency fluctuated between once per week and seven times a week.The studies followed the patients for a period that ranged from four to 20 weeks.Further description of the included studies' design, conclusion, and baseline characteristics of the enrolled children is available in Table 1.

Description of the Included Studies
We included 44 RCTs and one retrospective cohort study [54] in this systematic review.The total number of enrolled CP patients was 1580 (801 for VR-based exergaming and 779 controls).The included studies were carried out in different countries: Turkey, Egypt, South Korea, Taiwan, Australia, Iran, India, the UK, Italy, Russia, Finland, Brazil, Saudi Arabia, Canada, USA, Spain, Belgium, Denmark, and the Netherlands.The duration of physiotherapy sessions ranged from 25 min to 90 min, and the frequency fluctuated between once per week and seven times a week.The studies followed the patients for a period that ranged from four to 20 weeks.Further description of the included studies' design, conclusion, and baseline characteristics of the enrolled children is available in Table 1.

Quality Assessment Findings
The majority of the included RCTs had a low risk of bias regarding study reporting, participants' attrition, and generating a random sequence.Allocation of the children in the study groups was concealed in less than half of the trials.Nearly half of the RCTs ensured good blinding of outcomes assessors.However, almost all the RCTs were unclear regarding the blinding of the participants and the study personnel (Figure 2).The retrospective cohort study (Chang et al., 2020) [54] had a fair quality (Supplementary File S2).The GMFM-88 score was reported in four RCTs that enrolled 84 CP patients (43 for VR-based exergaming and 41 for controls) [51][52][53]60].The meta-analysis revealed homogenously significant improvement in the GMFM-88 score with the use of VR-based exergaming (MD = 0.81; 95% CI [0.15, 1.47], p-value = 0.02), (p-value = 0.32, I 2 = 15%) (Figure 3).
Three trials reported the dissociated movements and grasp domains of the QUEST score separately, with a sample of 65 CP children (34 for VR-based exergaming and 31 controls) [54,61,64]

Qualitative Synthesis of Other Efficacy Outcomes
Park et al., 2021 reported a significantly greater improvement in the static and dynamic sitting balance (assessed via Wii Balance Board and modified functional reach test, respectively), postural swing speed and distance, as well as trunk stability with VR-based intervention when compared to the control [75].Abo-Zaid et al., 2021 compared VR-based intervention with usual care and task-oriented therapy.Their analysis results favored task-oriented therapy over VR-based intervention and usual care regarding step length, stride length, cadence, and weight support as evaluated via the 3D motion analysis system [76].Avcil et al., 2020 reported a similar improvement in the motor functions with the VR-based exergaming and the control.However, exergaming was more effective in improving manual dexterity as assessed via the Minnesota manual dexterity test [77].
Aran et al., 2019 reported that the VR-based exergaming was more effective in comparison with the control on improving cognitive functioning as evaluated by the dynamic occupational therapy cognitive assessment for children [78].Ökmen et al., 2019 concluded that VR-based exergaming is more effective than the control in regards to hand function (as assessed via the bimanual fine motor function scale), functional level (assessed by the GMFCS), and mobility (evaluated by the functional mobility scale) [81].Pourazar et al., 2019 reported better improvement in balance ability (in the anterior, posteromedial, and posterolateral directions) with VR-based exergaming when compared to the control [82].Tarakci et al., 2019 found out that VR-based exergaming is equally beneficial as conventional physiotherapy [83].

GRADE Assessment
According to GRADE, all our comparisons in the different outcomes were at different levels of certainty (from very low to moderate).The causes of their downgrading were: the heterogeneity of pooled studies in each assessed outcome and the publication bias as the observational studies are more attributable to it.Moreover, the specific details and s of the publication bias are in Supplementary File S4.

Discussion
This study aimed to systematically review the evidence on the application of VR-based exergaming in rehabilitating children with CP.Our review included 45 studies (44 RCTs and one retrospective cohort study), with a total of 1580 (801 for VR-based exergaming and 779 controls).We conducted meta-analyses that revealed that VR-assisted exergaming was more effective in improving the GMFM-88 total score (p-value = 0.02) and the GMFM walking and standing dimensions (p-value < 0.05).VR-assisted exergaming modality was also more effective in improving the mobility and cognitive domains of the PEDI score (p-value < 0.001), the COPM performance domain (p-value < 0.001), and the WeeFIM total score (p-value < 0.001).Regarding upper limb function, VR-based exergaming was more effective (p-value < 0.001) in improving the four subscales of the MA-2 (accuracy, dexterity, range of motion, and fluency).VR-assisted exergaming intervention was similarly beneficial as conventional physiotherapy in improving other function assessment measures.No safety issues were observed with this proposed intervention in the included studies.
Rehabilitation by VR-assisted exergaming adds elements of fun and excitement to children's experience with physiotherapy.These elements keep the children engaged and facilitate active participation in task-oriented training [97].The provided direct encouraging and rewarding feedback (visual, auditory, and tactile) from the game with the aforementioned elements motivates the child to keep repeating the specific task.Eventually, the experience becomes more entertaining with higher intensity and difficulty may be increased [98,99].Motivated active engagement, in addition to repetition of tasks and increasing difficulty, positively modulate neuroplasticity [100][101][102][103].These facts might explain why children undergoing rehabilitation via VR-assisted exergaming have enhanced responses to rehabilitation programs.The boosted response to physiotherapy, along with the improved feasibility in terms of accessibility and cost, makes VR-assisted exergaming a promising advanced substitute to conventional physiotherapy [104,105].
The role of the Internet of Things (IoT) in VR-assisted exergaming is to enable the connection and communication between different devices and applications that are involved in the exergaming process.IoT can help to collect and process data from sensors, motion trackers, controllers, and other devices that monitor the user's movements, performance, and feedback in the virtual environment.IoT can also help to provide remote access and support for the user, such as by allowing the therapist or caregiver to observe, guide, or join the exergaming session from a different location.IoT can enhance the value and effectiveness of VR-assisted exergaming for rehabilitation by providing more data, feedback, personalization, and social interaction [106].
Our study provides an update on the previous systematic review conducted by Fandim et al. in 2020 [107].A total of 12 new studies are added in this update [50-57,60,75-77], and the number of included CP patients in the analyses is greater than before.Moreover, we conducted a separately specified meta-analysis for each function measurement scale.The results of our analyses were consistent with those reported by Fandim et al. concerning a child's functioning and upper and lower limb mobility [106].Heterogeneity in the results was detected in some of our meta-analyses, as in Fandim et al., 2020 [106].This heterogeneity might emerge from the variation between the studies in the types of CP included in the study or dissimilarities in the baseline level of functioning.Moreover, the variation in the performed training tasks their difficulty and repetition, and the frequency and duration of rehabilitation sessions may contribute to the observed heterogeneity.
This systematic review and meta-analysis summarize updated evidence of the efficacy and safety of VR-assisted exergaming when applied in rehabilitating children with CP.Our study is strengthened by including a variety of studies with several exergaming techniques applied and different groups of muscles targeted.These include Bicycle exercise: This technique involves pedaling a stationary bike that is connected to a VR system that simulates different environments and scenarios.The child can choose the level of difficulty and the type of terrain they want to explore.Resistance training: This technique involves using elastic bands, weights, or machines that provide resistance to the child's movements.The child can perform various exercises that target specific muscle groups, such as biceps curls, shoulder presses, chest presses, leg extensions, and leg curls.The resistance can be adjusted according to the child's ability and progress.This technique can improve the strength and power of the upper and lower limb muscles.Aquatic training: This technique involves performing exercises in water that is heated to a comfortable temperature.The water provides buoyancy and resistance to the child's movements, which can reduce pain and spasticity.The child can perform various exercises that involve moving their arms, legs, trunk, and head in different directions.This technique can improve the flexibility and range of motion of all muscle groups.Balance training: This technique involves using a VR system that provides visual and auditory feedback to the child's balance performance.The child can stand on a platform that tilts or vibrates according to their movements.The VR system can display different scenes and challenges that require the child to maintain their balance and posture.This technique can improve the stability and coordination of the core and lower limb muscles.
This review is also strengthened by the different types of CP.However, this variation might have limited our study due to the resulting heterogeneity that could not be resolved in some of the analyses.In addition, the inconsistency in blinding the studies' participants and personnel induces a risk of performance bias.This is explained by the inability to blind them due to the nature of the applied intervention.However, outcome assessors were properly blinded in the majority of the trials.
Many of the studies we reviewed had some methodological limitations, such as small sample sizes and lack of blinding, and most of them were preliminary tests.Therefore, we should carry out more rigorous trials on this topic and account for the potential confounders that may skew the findings.Moreover, a limitation of this systematic review is that we could not compare the effects of different types of VR (non-immersive, semi-immersive, or immersive) or different types of controls (active or passive) on the outcomes of VR exergaming for CP rehabilitation, due to the lack of information and consistency in the included studies

Conclusions
Our findings suggest that VR-assisted exergaming may have some advantages over conventional rehabilitation in improving CP children's functioning and performance in daily life activities, upper and lower limb mobility, and cognition.VR-assisted exergaming seems to be as effective as conventional physiotherapy in the other studied function measures.With its potential efficacy, better feasibility, no reported side effects, and entertaining experience, VR-assisted exergaming may be a viable complementary approach to conventional physiotherapy in rehabilitating children with CP.

Figure 2 .
Figure 2. Risk of bias summary for randomized controlled trials.

Figure 2 .
Figure 2. Risk of bias summary for randomized controlled trials.

Figure 3 .
Figure 3. Forest plot of the analysis Gross Motor Function Measurement 88 score.

Figure 3 .
Figure 3. Forest plot of the analysis Gross Motor Function Measurement 88 score.

Figure 5 .
Figure 5. Forest plot of the analysis Canadian Occupational Performance Measure score (Performance domain).

Figure 5 .
Figure 5. Forest plot of the analysis Canadian Occupational Performance Measure score (Performance domain).

Figure 5 .
Figure 5. Forest plot of the analysis Canadian Occupational Performance Measure score (Performance domain).

Figure 7 .
Figure 7. Forest plot of the analysis Melbourne Assessment of Unilateral Upper Limb Functionversion 2 scale score.

Figure S11 :
Figure S11: Forest plot of the analysis of ABILHAND kids' test scores; Figure S12: Forest plot of the analysis of ABILHAND kid's test scores after leaving one out.Supplementary File S4 detailing VR compared to control for Cerebral Palsy.

Table 1 .
Summary of the included studies and characteristics of the enrolled patients.