Neuromuscular Excercise in Children with Down Syndrome. A Systematic Review

Objective to evaluate the effects of neuromuscular exercise, specifying the parameters and characteristics of effective interventions to improve balance, muscle strength and exibility in children with DS between the ages of 4 and 18 years. the the The methodology and results of the studies were critically appraised in compliance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyzes guidelines. Ten studies were included. The interventions included mechanotherapy, vibration, and use of different unstable surfaces. The exercise frequency ranged from three days to ve days a week, and the duration of each session was between six and 15 minutes. The frequency was between two and three times a week for 6 and 12 weeks and the intensity was between 60% and 80% of maximum resistance (MR).


Introduction
Children with Down syndrome (DS) exhibit delayed motor development compared with typically developing children 1 . DS involves alterations in balance, strength, and muscular endurance, which affect postural control and generate atypical motor development in addition to cognitive disability. Approximately 10% of children with DS are able to sit in an upright position and exhibit an independent walking pattern before the age of 3 years, and approximately 95% exhibit these abilities between 3 and 6 years of age 2 .
In general, children with DS retain the typical sequence of motor development, but basic motor skills are attained late in childhood 3 . The hypotonia and muscle weakness that are characteristic of DS interfere with intermuscular coordination, which affects balance and the processing of proprioceptive information and results in alterations in postural control, functionality and quality of life of the child. These characteristics restrict independence in mobility, which increases the demand for monitoring and support by caregivers, a situation that becomes more evident in the adolescent stage because of the new relationships established with friends and the school environment in which the child interacts 4,5 .
To promote mobility and independence in daily activities, children and adolescents with DS can participate in therapeutic physiotherapy interventions that aid the development of motor skills and improve functional performance through the use of therapeutic exercise, among other intervention alternatives 6 .
Therapeutic exercise involves the application of physical exercise parameters such as intensity, frequency, and duration for therapeutic purposes. Therapeutic exercise categories include aerobic, anaerobic, resistance, and neuromuscular exercise 7 . Neuromuscular exercise activates neurophysiological processes of intermuscular coordination that, together with increased muscle strength and postural responses, contribute to stability in the execution of motor patterns within the functional repertoire of children and adolescents with DS. There are different ways to perform neuromuscular exercise, including the use of unstable surfaces such as balls or mats, mechanotherapy equipment, isokinetic techniques, and bodyweight resistance exercises [8][9][10][11] .
To the best of our knowledge, no systematic reviews have addressed the effects of neuromuscular exercise in the pediatric population with DS. However, several primary studies have examined the effects of neuromuscular interventions in children and adolescents with DS. Although some published reviews on the effects of physical therapy interventions have included neuromuscular exercise in their analysis, the inclusion of recent primary studies may improve the certainty of the evidence 6,8 .
Previous systematic reviews of evidence regarding muscle exercise have included randomized controlled trials (RCTs) and quasi-experimental studies, making comparisons between studies despite the differences in their methodological quality. In addition, previous systematic reviews have examined different research questions, different populations, and different therapeutic interventions [12][13][14] .
The aim of the current study was to synthesize the existing research evidence on the effects of neuromuscular exercise, specifying the parameters and characteristics of effective interventions to improve balance, muscle strength and exibility in children with DS between the ages of 4 and 18 years. This synthesis of the best available evidence may help to facilitate the inclusion of these interventions in therapeutic approaches for this population and facilitate the development of motor skills and functional performance in children with DS.

Methods
Methods We adhered to the Preferred Reporting Items for Systematic Reviews and MetaAnalyses (PRISMA) guidelines 15 Eligibility criteria Design: A systematic review of the literature was conducted to identify RCTs.
Type of participants: Children with DS between the ages of 4 and 18 years.
Type of interventions: All neuromuscular interventions of therapeutic exercise with speci c prescription parameters in terms of intensity, frequency, duration, among others were included.
Outcomes: For inclusion, studies identi ed in the literature search had to include results for at least one of the outcomes included in the review: muscle strength, balance, and exibility.

Search and identi cation of studies
Search terms were generated from the Population, Intervention, Comparison and Outcome (PICO) components of the following question: What is the effect of neuromuscular therapeutic exercise on strength, balance, and exibility in children with DS? These terms were adapted according to the different databases explored. A systematic search was conducted in the Pubmed, EMBASE, SCIELO, Lilacs, Cochrane Library and Epistemonikos databases [16][17][18][19][20] . In addition, other sources of evidence were consulted to allow the identi cation and analysis of published or unpublished literature (gray literature) that had not been detected by the systematic search. This process was carried out through manual searches in reference lists of documents found in the review of databases. This process was developed over 7 months.

Selection of the studies
The nal selection of the studies was made independently by two reviewers who were both physiotherapists, one with training as an epidemiologist and the other with a Master's degree in Exercise and Sports Rehabilitation (MRTN and OCVP). The reviewers reviewed all titles and abstracts and excluded those that were considered irrelevant to the review because they did not meet the eligibility criteria of RCT design and inclusion of at least one of the prioritized outcomes.
Subsequently, the reviewers reviewed the full text of the studies verifying the eligibility criteria. Each reviewer generated Bib Tex les of the studies they considered eligible, and identi ed duplicates were eliminated using a bibliographic manager. The choice of each study was determined by consensus after independent review by the two reviewers. In cases where there was no consensus, a third reviewer decided on eligibility.

Extraction and management of variables
All variables considered relevant for the comparison of the studies and measurement of outcomes were extracted. Data on the type, mode, frequency, intensity, duration of the interventions, location in which the interventions were performed (e.g., outpatient clinic, home), person in charge of applying the intervention (e.g., physical therapist, other professional, family member or caregiver) were extracted from predesigned forms.
For the population, data were obtained regarding age, sex, sample size of each group and cognitive engagement.
Data were extracted for muscle strength, balance, and exibility outcomes. For balance, data were obtained on center of mass displacement or time to maintain a balanced posture. For muscle strength, data were reported in pounds, kilograms of force, or Newtons. For exibility, data on muscle elongation were obtained.
Evaluation of study quality Two independent assessors (MRTN and OCVP) assessed the risk of bias of each included study using the Cochrane Collaboration's risk of bias assessment tool 21 . The assessors rated each study as having a "low risk of bias," "high risk of bias," or "unclear risk of bias," taking into account six domains: random sequence generation (selection bias), allocation masking (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data, and selective reporting (reporting bias). The risk of bias rating was analyzed using Revman 5.1 software.
Disagreements in bias assessment were resolved by a third-party evaluator (EIRG).
Evaluation of the certainty of the evidence Evaluation of the certainty of the evidence found was carried out using the GRADE approach 22 . In this approach, the evaluation of evidence goes beyond the evaluation of the methodological quality, because the evidence found for each outcome was graded considering the risk of bias, inconsistency, directness or indirectness of the evidence and imprecision, risk of selective publication of outcomes, and dose-response gradient. The rst four criteria were rated using a three-level ordinal scale: very serious, serious and not serious. The classi cation options for risk of selective publication of outcomes were: not detected, strong suspicion. The classi cation options for effect size were: no, large, very large. The classi cation options for presence of confounding factors were: no, will reduce the demonstrated effect, suggests spurious effect. The classi cation options for dose-response gradient were: no, yes, no 23 .

Synthesis of data
The selected body of evidence was organized by prioritized outcomes. Within each outcome, we described the characteristics of the population, the parameters of the interventions including the mode of exercise applied, the frequency, intensity and duration of the interventions applied in these studies, and the quantitative results achieved with their level of signi cance. This information is presented in Table 1.
Meta-analysis was performed when similarities were found in the components of the PICO question in at least two studies, as well as in the instruments used for the measurement of the outcome and the quali cation of the risk of bias. When the measurement instruments were different, the data were converted to common units.
For meta-analyses, data on population characteristics, randomization methods, outcome measures, duration of follow-up and methods of analysis were extracted from each study in a database previously designed in Excel. Direct comparisons made between the intervention and a control group de ned as educational activities, recreational activities, or continuity with activities of daily living or another intervention of interest for this review were considered.
For the prioritized outcomes, averages and standard deviations were extracted from the available data. Standardized differences from the mean (SMD) and 95% con dence intervals (95% CI) were calculated to allow comparability of data. Heterogeneity between trials was assessed using the chi-square test, a statistical signi cance level of p < 0.05 and the value of statistic I 2 . When the data exhibited heterogeneity with an I 2 greater than 70%, the results were combined using the random effects model and the 95% CI was calculated 24,25 . All the analysis above was performed using Revman 5 software 26 .

Study selection
A total of 1,384 studies were identi ed in the systematic literature search. From other sources, including the bibliographic references of the studies found in the systematic search, 239 additional studies were identi ed, resulting in a total of 1,623 identi ed studies. Of these studies, 88 were excluded because of duplication and 1,159 were excluded on the basis of reviewing the titles and abstracts. In total, 376 studies were reviewed in full text by two reviewers, of which 366 were excluded for not meeting the eligibility criteria. Finally, 10 primary studies were included. Figure 1 shows a ow chart of studies identi ed and included in the body of evidence.

Assessment of the risk of bias of included studies Allocation (selection bias)
Six studies 10,11,27-30 presented a low risk of bias because they reported masking of randomization, whereas four studies 31-34 presented a high risk of bias because they did not report the methods by which participants were assigned, and the personnel in charge of maintaining random assignment were not masked.

Blinding
Because of the nature of the interventions used, the risk of bias assessment of each study took into account the blinding of the outcome assessors in each study. Three studies 28,31,32 presented a high risk of bias, while seven 10,11,27,29,30,32,34 presented a low risk of bias.

Outcomes with incomplete data
Three studies presented a high risk of bias because they reported incomplete data by not including data from participants who did not achieve the expected results 33 or did not indicate the number of participants included in the reported results 29,32 . The remaining studies reported all data from the study sample.

Selective report
Eight studies presented a low risk of selective reporting because they included all of the outcomes that were measured; one study had high risk 32 and one more had unclear risk of bias 29 .
This information is summarized in Figures 2 and 3.
Modes of application of neuromuscular exercise in physiotherapy interventions in children aged 4 to 18 years.
Two studies involved the use of mechanotherapy equipment available in gyms for muscle strength training 10,27 . Three studies applied therapeutic vibration 11,28,31 , which was combined with conventional physiotherapy interventions in two studies 11,28 . Two studies used different unstable surfaces for balance training 29,32 , and one study used Nintendo Wii and virtual reality 33 . One study used isokinetic exercise 30 and another used bicycle training 33 .

Frequency, intensity and duration of neuromuscular exercise
In the two studies that examined muscle strengthening exercises using mechanotherapy equipment, the frequency was twice a week for 10 weeks 10,27 . In the studies that applied vibration 11,28,31 , the frequency was between three times per week for 24 weeks 11 and three sessions every month for 20 weeks 31 to 12 weeks 28 .
Exercise using unstable surfaces was applied with a frequency between two and three times a week for 8 and 12 weeks and a volume of three sets of 12 repetitions for each muscle group that was exercised 29,32 .
Lin et al. 34 applied exercise three times a week for 6 weeks, whereas Eid et al. 29 applied a frequency of three times a week for 6 months. The duration of the sessions varied between 35 and 75 minutes and the intensity, a parameter mentioned in few of the selected studies, was between 60% and 80% of maximum resistance (MR) in the studies that included exercise using mechanotherapy equipment 10,27 . Lin et al. 34 used the speed and inclination of the treadmill as intensity criteria, which started at 2.0 kph (0% inclination) and ended with an average speed of 3.0 kph (58° elevation).
Outcomes assessed in the studies included in the review Of the proposed outcomes to be evaluated, no evidence was found for the outcome of exibility in this population. Table 1 Table 2 and Appendix 1). The certainty of the evidence for the outcome of muscle strength was between moderate and high. In contrast, the certainty of evidence for balance was highly variable, ranging from very low to high, mainly because of the imprecision of the results obtained in the primary studies. The information is presented in Table 2.

Discussion
The modalities of therapeutic exercise include neuromuscular exercise and neuromotor exercise. Although these two types of exercise can be similar, they are used in different training scenarios, some of which are therapeutic and some of which are focused on increasing performance in athletes 35 . Neuromuscular exercise aims to improve sensorimotor control (i.e., the ability to produce controlled movement through coordinated muscle activity) and to facilitate neuromuscular control. Thus, neuromuscular exercise seeks to improve the unconscious response of muscles to signals related to dynamic joint stability, which is the ability of the joint to remain stability during movement execution. To achieve this goal, neuromuscular exercise improves variables such as muscle strength, exibility, and balance 36,37 .
The present review included articles reporting neuromuscular exercise with outcomes including muscle strength, balance, and exibility. According to the American College of Sport Medicine (ACSM), some neuromotor exercise interventions incorporate other motor skills, such as coordination and agility 36 . That is, neuromotor exercise combines resistance and exibility training with activities such as Tai chi and Yoga. Thus, the characteristics of these two types of exercise are treated as interchangeable in the literature reviewed 37 .
The evidence identi ed was scarce regarding the interventions and outcomes selected, and the quality of the evidence was found to be low-moderate. Although the studies included in the present review were RCTs, they presented high risk and unclear risk in aspects that affect the internal validity of the study and certainty in the measurement of the effect. These risks were related to the masking of randomized allocation 31,34 , the masking of outcome assessment 28,31,33 , and selective data reporting 32 . In addition, the sample sizes of the studies were small, which may explain the wide con dence intervals 27,28,30,32,33 . The loss to follow-up and the impact on statistical power could be related to the non-signi cant differences found in the equilibrium outcome 32 , which may have been the result of systematic type 2 error 38 .
No evidence was found for the outcome of exibility. However, muscle stretching exercises were included as part of the training plan in two of the studies that examined muscle strength as the main outcome 30,34 . A previous study reported that the simultaneous training of exibility and muscular strength improved muscular performance and the maintenance of improvements in muscular elongation 39 . Thus, exibility may not have been included as a primary outcome in studies of the DS population because it was considered as a means to promote another outcome (such as muscle strength) and, in turn, as an indispensable element for effective and safe training 39,40 .
Another potential reason for the outcome of exibility to not have been included as a therapeutic objective in previous studies is that children with DS exhibit characteristic muscle hypotonia, which is a decrease in the residual tension of the muscles at rest 41 . Hypotonia is in turn associated with decreased muscle strength, increased joint laxity, and increased exibility in these children 42 . Therefore, improvement in exibility may not be a primary objective of therapeutic exercise, particularly when there is controversy regarding whether increased exibility directly promotes strength gains or, on the contrary, limits muscle strength gains 43 .
Regarding the parameters examined in the application of neuromuscular therapeutic exercise to improve muscle strength, the most common were frequency, intensity, and duration. The proposed therapeutic windows for the frequency of neuromuscular exercise in previous studies ranged from between two to three times per week, to 8 week 10,27 , to 24 weeks 11 . The approximate volume in this therapeutic window was three sets of 12 repetitions for each muscle group, the duration of each session varied between 35 and 75 minutes with an intensity between 60% and 80% of maximum resistance (MR) 10,27 .
Evidence regarding the use of neuromuscular exercise to improve balance reported the use of progressive muscle training 2-3 times per week in 45-60 minute sessions for 10-24 weeks. These studies included isometric and isokinetic exercises and muscle stretching, as well as therapeutic vibration, exercise bikes and virtual reality with devices such as Nintendo Wii 27,28,30,31,33 .
The therapeutic window proposed in this review is related to the ACSM's proposed guidelines for the evaluation and prescription of physical exercise 40 . The ACSM guidelines recommend a frequency greater than or equal to 2 or 3 days a week, a duration greater than or equal to 20 minutes, and an accumulated duration per week equal to or greater than 60 minutes, for the training of motor skills such as balance, coordination, gait, agility and proprioception in older adults and young people 40 .
The ACSM recommendations regarding prescription parameters are focused on physical exercise for potentially healthy adults 40 . There are no recommendations regarding parameters for the pediatric population because of the absence of supporting evidence in that age group. Future studies should evaluate the dose of neuromuscular exercise in potentially healthy children to de ne clear recommendations in this population. These can be extracted from the synthesis and analysis of scienti c evidence or from the adaptation of existing scienti c evidence regarding physical exercise in populations with speci c conditions. Thus, future studies on therapeutic exercise should include the prescription parameters to get closer to a therapeutic dose of exercise in children with DS.
Despite the scarcity of evidence regarding the effects of neuromuscular exercise in the pediatric population with DS, an effect has been reported in a population of adults with DS. Sugimoto et al. 12 , reported signi cant changes in muscle strength, and functional task performance in adults and youth with DS when they performed neuromuscular exercise. To the best of our knowledge, the current study is the rst systematic review to evaluate the effect of neuromuscular exercise in a pediatric population with DS from RCTs, contributing to the robustness of current evidence.

Conclusion
Despite the limited number of RCTs available in previous literature, neuromuscular exercise appears to be effective for the improvement of both lower limb and chest muscle strength in children over 8 years of age when isometric and isokinetic training is used as a mode of exercise in addition to other strategies, such treadmill training. There is also evidence for a positive effect on balance improvement when unstable surfaces and balance platforms, as well as lower limb muscle strength training, are used as modes of application. Future research is needed to investigate the effects of neuromuscular exercise in early childhood in more detail, as well as its effects on outcomes such as exibility.

Figure 3
Risk of bias graph: review authors' judgments about each risk of bias item presented as percentages across all included studies.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Appendix11.docx