Abstract
Objective:
This systematic review and meta-analysis investigates the effects of strength training program in children and adolescents with cerebral palsy to improve function, activity, and participation.
Data sources:
Five electronic databases (MEDLINE-Pubmed, Cochrane Library, PEDro, CINAHL, and SPORTDiscus) were systematically searched for full-text articles published from inception to 30 June 2021.
Review methods:
Randomized controlled trials were included, who compared: (i) child population with spastic cerebral palsy population between 0 and 22 years; (ii) studies in which a muscle strength training program was performed and included dosing information; (iii) studies comparing strength training with other physical therapy technique(s) or untreated control group. Studies with similar outcomes were pooled by calculating standardized mean differences. Risk of bias was assessed with Cochrane Collaboration’s tool for assessing the risk of bias and PROSPERO’s registration number ID: CRD42020193535.
Results:
Twenty-seven studies, comprising 847 participants with spastic cerebral palsy. The meta-analyses demonstrated significant standardized mean differences in favor of strength training program compared to other physical therapy technique(s) or untreated control group(s) for muscle strength at the knee flexors, at the knee extensor, at the plantarflexors, maximum resistance, balance, gait speed, GMFM (global, D and E dimension) and spasticity.
Conclusion:
A strength training program has positive functional and activity effects on muscle strength, balance, gait speed, or gross motor function without increasing spasticity for children and adolescents with cerebral palsy in Gross Motor Function Classification System levels I, II, and III when adequate dosage and specific principles are utilized.
Introduction
Based on the systematic review of the evidence for all interventions in cerebral palsy by Novak et al., 1 physical training has demonstrated positive effects on muscle strength, 2 aerobic capacity, 2 and energy expenditure (reducing sedentary lifestyle). 1 The types of effects seen may depend on the dose and training parameters performed in order to target certain objectives. In their review, muscle strengthening was shown to improve fitness, physical activity, ambulation, mobility, participation, and quality of life, but it is necessary to combine this with other intervention to achieve a positive effect on functional mobility or other gross motor skills. 1
Muscle strengthening has been studied by various authors with inconsistent conclusions across studies, such as some state that muscle strength improves but not function, or that the results should be taken with caution; 3 while other authors report that there is an improvement in motor activity4,5 and in specific functions such as gait.5,6 Finally, there is another group of studies reporting no benefit of any kind both for strengthening of the muscle groups of the lower 7 and upper limbs. 8 Most if not all authors conclude that there is the need for more research and a higher level of evidence to be able to determine the functional benefits of muscle strengthening in children and adolescents with cerebral palsy.3–8
Therefore our objective was to review the most current scientific evidence on strength training in children and adolescents with cerebral palsy, due to an increase in the number of clinical studies since the last systematic review in 2014. It is a recommended practice to update evidence reviews or guidelines every 3–5 years. Our focus was on the effects of strength training on function, activity, and participation and on providing insights on the dosing parameters and training strategies that are associated with more positive results, in terms of strength, gait, balance, function, or spasticity.
Methodology
Before starting, this review was registered in the PROSPERO database (registration number ID: CRD42020193535). The recommendations and the checklist of the Preferred Reporting Elements for Systematic Reviews and Meta-Analysis (PRISMA) 9 were followed for its development.
Search strategy
Systematic searches were conducted in MEDLINE (PubMed), PEDro, Cochrane Library, CINAHL, and SPORTDiscus. From inception to 30 June 2021 searches were conducted for randomized trials comparing the effects of strength training with other types of interventions or no intervention. Search terms included specific terms such as “cerebral palsy,” strength training,” “strength exercise,” “resistance training,” “children,” “young,” “adolescent,” “infant,” “pediatric,” and “spastic.” The search strategy for the different databases is shown in Table A1 of the Supplemental Appendix. To complete the systematic literature search, the references of eligible articles were examined; where articles were first selected by keywords, then by title and abstract; the last step being the full text reading.
Study selection
The included studies had to meet the following inclusion criteria: (i) child population with spastic cerebral palsy; (ii) cerebral palsy population between 0 and 22 years old; (iii) randomized clinical trials; (iv) studies in which a muscle strength training program was performed and included dosing information. These studies could involve strength training of a single body segment or one or more limbs or larger body regions; (v) studies comparing strength training with other physical therapy technique(s) or untreated control group; (vi) no language restrictions.
The exclusion criteria for the studies were the following: (i) studies where strength training is combined with aerobic training; (ii) studies that combined strength training with botulinum toxin type A treatment or any type of medical-surgical intervention (with a minimum of six months of application); (iii) not randomized clinical trials; and (iv) studies that do not specify the strength training dose.
Data extraction and risk of bias assessment
Data extraction was done by two investigators and they created a table summarizing the following information from the studies: (1) author; (2) country; (3) year of publication; (4) characteristics of the sample (sample size, distribution by age, and type of population); (5) dose of the intervention; (6) type of force intervention; (7) measured variables; and (8) comparison groups. The Cochrane Collaboration’s tool for assessing risk of bias (RoB2) 10 was used to assess the risk of bias of the included randomized clinical trials. Assessment with this tool includes six domains: randomization process, deviations from planned interventions, missing outcome data, outcome measurement, and reported outcome selection. Each domain can be scored as: low risk of bias, some concerns, or high risk of bias. The literature search, data extraction, and quality assessment were performed by two independent reviewers, and a third reviewer was included when inconsistencies remained.
Assessment of methodological quality
The reviewed studies were composed of randomized clinical trials. Therefore, GRADE (Grading of Recommendations Assessment, Development, and Evaluation) scores were calculated to estimate the methodological quality of these studies. Randomized clinical trials are controlled study designs known to minimize the risk of bias. The GRADEpro application, 11 whose utility has been approved, was used. The GRADE scale rates the level of evidence of the results, taking into account aspects related to the methodology and results of the study, such as (1) risk of bias, (2) inconsistency, (3) indirect evidence, (4) imprecision, and (5) other considerations (dose-response and reported biases).
Statistical analysis and data synthesis
To calculate the pooled estimate of the effect size and confidence intervals (CI) of 95% used the method DerSimonian and Kacker, 12 using random effects models. A standardized mean difference score was calculated for each subgroup, using the Hedges g-index as the effect size statistic, in which negative effect size values indicate a positive effect of specific strength training versus the control group or another intervention, for the improvement of each variable. Hedges g values represented the following: (i) weak effects when the values were around 0.2, (ii) moderate effects when the values were around 0.5, (iii) strong effects when the values were around 0.8, and (iv) very strong effects when the values were greater than 1.0. 13 Heterogeneity of results between studies was assessed using the I 2 statistic. The I 2 values were interpreted as: may not be important (0%–40%); may represent moderate heterogeneity (30%–60%); substantial heterogeneity (50%–90%); or considerable heterogeneity (75%–100%). Values with statistical significance set at P < 0.10 were also considered. These statistical calculations were performed as long as there were a minimum of four studies reporting data on the same outcome and the data were valid for meta-analysis. The software Comprehensive Meta-Analysis version 2.0 software (Biostat, Englewood, NJ, USA) for the statistical analyzes was used.
Results
Systematic review
Twenty-seven studies14–40 (Figure 1) analyzed the influence of muscle strength training in children and adolescents with spastic cerebral palsy. A total of 873 subjects participated in the analyzed studies, of which 847 were diagnosed with spastic cerebral palsy and 26 subjects with no diagnosis of neurological impairments.14,32 The age range was between 3 and 22 years. The subjects were divided into unilateral spastic (221 subjects) and bilateral (517 subjects) cerebral palsy, while the distribution in the Gross Motor Function Classification System levels of the studies that reported these data was 147 in level I, 170 in level II, 106 at level III, and 9 at level IV, (not all datas were reported). Of the total sample of children with diagnosed with spastic cerebral palsy, strength training was performed by 452 participants (Table A2 in the Supplemental Appendix).

Flowchart of included studies.
The results of assessment of the strength of the evidence in references are shown in Table A3 in the Supplemental Appendix.
Of the 27 studies, 15 performed strength training in a clinical physiotherapy session, 8 were done in the natural environment of the home and 4 in the school environment. The main outcome variables evaluated were strength, spasticity, gait, balance, energy expenditure, and motor function with some variation in the types of outcomes and how these were measured across studies (Table A2 in the Supplemental Appendix).
Of the 27 studies, 22 reported measures of muscle strength in various muscles and using a range of methods. At the knee joint, the strength of the extensors was measured in 15 studies and of the flexors in 10 studies. At the hip joint, the strength of the flexors was measured in five studies, of the extensors in four studies; and of the abductors in six studies. Finally, at the ankle, plantarflexors strength was measured in five studies and of the dorsiflexors in three studies. The global strength of the lower limbs was measured in four studies. Maximum resistance was measured by one-repetition maximum in six studies and by six-repetitions maximum were measured in two studies. Others strength measures were anaerobic muscle power and muscle power sprint.
With respect to gross motor function, 15 studies measured this outcome using the Gross Motor Function Measure (GMFM). Ten studies studied the D dimension; 11 studied the E dimension. The sum of dimensions D and E only was measured in three studies, the total GMFM Score in seven studies. GMFM-88 was used mostly while GMFM-66 was used in four studies.
Regarding strength training and their characteristics, results varied across studies. With regards to the training dose, programs lasted between 4 and 12 weeks (short-term effects), with exercises carried out between 2 and 5 days a week, for a duration ranging from 20 to 60 minutes. Each exercise had between 3 and 20 repetitions with 1 or 3 series of repetitions. In addition, these exercises were usually performed without resistance or with the use of specific equipment such as balls, weight machines, backpacks, cuff weights, body vests, or sand bags, (Table A4 in the Supplemental Appendix).
With respect to gait parameters, 13 measured temporal-spatial parameters such as speed with cadence and stride length parameters additionally reported in four trials.
Notably, eight trials measured balance with three studies utilizing the Timed Up and Go test, two with the Pediatric Berg Balance Scale, one with the Bruininks Oseretsky Test of Motor Proficiency, one with Biodex Balance System Test, and one with the Functional Reach Test.
Seven studies measured spasticity. Four of them through the modified Ashworth scale and three through the presence of a spastic catch in a rapid stretch of the lower limb musculature. All studies demonstrated that strength training did not change or have adverse effects on spasticity, (Table A2 in the Supplemental Appendix).
Finally, the following secondary outcomes were also reported in various studies: ability to climb stairs (Timed Stair Test), symmetric weight bearing capacity, morphological properties of the muscle (cross-sectional area), participation, quality of life, attainment of individualized goals, and satisfaction with the level of attainment (Canadian Occupational Performance Measure), self-concept, activities of daily living, Sit to Stand and Lateral Step up and energy expenditure (beats/minute and Net Nondimensional oxygen cost – NNcost.
Risk of bias
For randomized clinical trials conducted by intention to treat and assessed with the Cochrane Collaboration’s tool for assessing risk of bias (RoB2), 100% of the studies show a low risk of bias, with a total of three studies in this risk of bias assessment group (Figure A1 in the Supplemental Appendix). For randomized clinical trials performed by protocol and evaluated with Cochrane Collaboration’s tool for assessing risk of bias (RoB2), 4.2% studies show a low risk of bias, 62.5% studies show some concerns with the risk of bias and 33.3% of studies demonstrated at least one high risk of bias category (mainly due to the evaluation process, the randomization process or not being able to blind the interventions to the target population), with a total of 24 studies in this risk of bias assessment group (Figure A1 in the Supplemental Appendix).
Meta-analyses grouped by outcome measures
Of the 27 studies included in the qualitative synthesis, 24 studies were used, since three were excluded because they did not report data that could be analyzed.18,32,35 We also did not include studies that presented data on the same sample already reported to prevent a replication bias as in the studies by Scholtes et al.30,31
Muscle strength at the knee joint
Figure A2 of the Supplemental Appendix summarizes the six trials evaluating the effects of strength training compared to other interventions for increasing muscle strength of the knee flexors, which showed a small but greater positive effect, but with considerable heterogeneity. The Figure A3 in Supplemental Appendix summarizes the 11 trials evaluating the increase in muscle strength of the knee extensors which minimal positive effects in favor of the strengthening with no heterogeneity.
Plantarflexors muscle strength
Figure A4 of the Supplemental Appendix summarizes the four trials evaluating the effects of strength training compared to other interventions for increasing muscle strength of the plantarflexors, which showed a small to moderate positive effect, but with substantial heterogeneity.
Maximum resistance
Regarding maximum resistance, Figure A5 of Supplemental Appendix summarizes the seven trials evaluating maximal force specifically, which show a small to moderate effect with moderate heterogeneity for strengthening versus comparators for improving resistance maximum.
Balance
The Figure A6 in Supplemental Appendix summarizes the eight trials evaluating strength training force specifically, being more effective than other interventions to improve standing balance; with an effect very strong effect with substantial heterogeneity.
Gait speed
Regarding the variable gait speed, Figure A7 of the Supplemental Appendix summarizes the 13 trials evaluating training force specifically, which exhibits a small effect with no heterogeneity compared to other interventions.
GMFM
The Figure 2 summarizes the five trials evaluating strength training specifically with respect to other interventions for improving the GMFM scores, presenting a small but positive effect without heterogeneity. The Figure 3 summarizes the trials evaluating strength training specifically regarding the other interventions for improving the GMFM dimension D (9 trials) and for Dimension E (10 trials) (Figure 4), showing a small positive effect and no heterogeneity for both dimensions.

Forest plots of pooled effect size for GMFM total score improvement.

Forest plots of pooled effect size for GMFM dimension D score improvement.

Forest plots of pooled effect size for GMFM dimension E score improvement.
Spasticity
Regarding the variable spasticity, Figure A8 of the Supplemental Appendix summarizes the four trials evaluating training force specifically, which exhibits a small to moderate effect with no heterogeneity compared to other interventions.
Discussion
The results from this review indicate a weak effect on the strength of the knee flexors and extensors, in addition to the plantar flexors. They further demonstrated an increase in gait speed and gross motor function, without increasing spasticity; through the implementation of strength programs for children and adolescents with cerebral palsy. These findings indicate an increase in the strength of the muscle groups of the lower extremities, which implies the possibility of improving the workload capacity and the resistance of said muscles, which are weak due to the neurological injury and resultant decreased activity. In addition, the speed of walking is increased, which implies that children and adolescents will be able to have better functional mobility. Also, gross motor function, that is, the ability of each individual to perform certain voluntary movements and tasks (e.g. run or jump), is increased. Finally, a large effect was obtained in terms of improvement in balance.
One of the most important results of this research is that strength training does not change or have adverse effects on spasticity; with some authors providing data showing no worsening of spasticity; while other authors merely state at the outset that spasticity will not be increased by strength training.
Based on the Quality of Evidence Assessment, a strong recommendation is made for improving the knee extensors muscle strength, gait speed, dimension E of the GMFM, and for not increasing spasticity. In addition, a moderate recommendation is made for the improvement of the knee flexors muscle strength, the total score and the score of dimension D of the GMFM. In contrast, there is a weak recommendation for the improvement of plantarflexors muscle strength, maximum resistance and balance.
Our results are in line with those made by Dodd et al., 4 where significant results were obtained for the strength and function variables, but not for activity or participation. In relation to Mockford and Caulton, 6 positive effects were also similar to ours in terms of function and gait, but the studies included in their review were not of as high a level of evidence. Finally, the publication by Park and Kim 5 obtained positive effects for gait and strength in randomized clinical trials where all cerebral palsy ages were included: on the other hand, our results are in contradiction to those obtained by Scianni et al.; 7 where these researchers concluded that strength training does not have positive effects on any variable in the cerebral palsy population. Our results on spasticity are similar that obtained by Dodd et al., 4 Mockford and Caulton, 6 and Scianni et al. 7 reporting no adverse effects from strength training in cerebral palsy.
The results for the specific domains of the International Classification of Functioning, Disability, and Health show an improvement in different aspects related to structure and function, such as energy expenditure during walking, improvement in balance, joint ranges, and the lack of increase in spasticity. Within the activity domain, a favorable result has been obtained for the first time from a systematic Review on strength training programs in children and adolescents with cerebral palsy, shown here by a positive effect on GMFM scores, especially in the GMFM domains on standing, and walking, running and jumping. In addition, improvements were also obtained in increased the distance walked per unit of time, which again indicates an improvement in the ability to walk. Regarding participation, the results found are inconsistent since some studies do show benefits on this domain and others do not, so more research is needed in terms of participation for strength training programs, because it has been seen that certain interventions may have an improvement of all the domains of the International Classification of Functioning, Disability and Health. 41
When searching the literature on strength training programs for other neurological pathologies, we found that for stroke, 42 the evidence generally indicates that muscle strengthening programs generate improvements in gait,42,43 muscle strength,42,44 muscle function, 42 postural control, 42 balance, 42 quality of life, 42 and independence; 42 without increasing spasticity 42 and improving activity.42,45 In addition, within this pathology, improvements were found in in both the lower and upper extremities. 42 However, some studies still do not show improvements in certain aspects of gait and balance; 44 and even suggest that perhaps this should not be done during the first three months after stroke. 46 For those with Parkinson’s disease we found that strength training programs can improve muscular strength,47,48 balance, 48 and some of the typical motor symptoms 48 of this pathology; but no improvement was obtained in terms of walking, 48 the patients’ confidence in balance, 48 and their quality of life; 48 compared with a control group or no intervention. 48 But it is necessary to provide other types of interventions in order to complement and improve the functionality of these patients, 47 since by itself strengthening does not seem to improve functional gait 49 or balance. 49
In summary strength training programs appear to improve the strength of trained muscles, balance, gait, and motor function in CP; without increasing spasticity; but this type of intervention may demonstrate different effects in different neurological conditions.
After the analysis of the different strength training protocols in this study (Table A4 in the Supplemental Appendix) and the review of some guidelines about strength training in children and adolescents, 50 the minimum recommendations to obtain positive effects with strength training in children and adolescents with cerebral palsy are summarized in Table A5 in the Supplemental Appendix.
Several limitations to this evidence summary were identified. (1) The assessment of risk of bias showed that about half of the studies had some concerns or a moderate risk of bias, while the other half of the studies showed a high risk of bias. It should be noted that the main reasons behind the higher risk of bias in the included studies were errors when evaluating the results and problems when blinding (which is possible to do with blind assessors); while the main reason behind the moderate risk of bias studies was certain deviations during the treatment period, due to the number of sample losses. (2) The studies did not analyze follow-up or long-term effect. (3) The vast majority of studies analyze the benefits of strength programs for lower limbs and trunk, where antigravity strength is critically important for gross motor function and mobility; but little research has been done on the benefits of training over upper limbs (where coordination movements are most important). (4) Despite the favorable result in terms of improved function with strength training, more research is needed in this field, particularly on its potential effects on participation, which has hardly been studied with respect to strength training programs. (5) Future studies should analyze the effects of combining strength training and botulinum toxin type A injections in antagonistic muscle pairs or strength training combined with task related motor skill training. (6) The analyzed sample was from levels I, II, and III of the Gross Motor Function Classification System, so the results cannot be extrapolated to levels IV and V, (their motor control and involuntary movements may limit their ability to strengthen specific muscles and to do it safely, so they may need to do it with others modalities such as aquatic strengthening exercise or neuromuscular electrical stimulation). (7) Many of the studies do not present a program structured according to the recommended exercise and activity guidelines, which could influence the results of each of the studies, 51 so both under and over dosing can lead to poorer outcomes. (8) More research about strength training programs and their effects in spasticity in a wider range of patients is needed.
Conclusions
In summary, our study shows that strength training in cerebral palsy subjects in the pediatric and adolescents age group has positive effects in terms of muscle strength in lower limbs, maximum work resistance, gait speed, and standing balance and gross motor function, without affecting spasticity; but these positive effects have a short-term duration, making it necessary for children with cerebral palsy to do high-intensity strength training regularly to maintain and ideally accumulate benefits over time. Finally, strength training is only one aspect of physical training and should be part of a larger program that includes task-related motor skill training as well as endurance training for optimal results.
Clinical messages
Strength training programs resulted in gains in muscle strength, maximum resistance, balance, gait, and gross motor function in children and adolescents with cerebral palsy without effects on spasticity.
Strength training programs improved structure, function and activity in International Classification of Functioning, Disability, and Health domains, but did not improve the participation domain.
More research is needed on strength training programs in pediatric neurological conditions.
Supplemental Material
sj-docx-1-cre-10.1177_02692155211040199 – Supplemental material for Effect of muscle strength training in children and adolescents with spastic cerebral palsy: A systematic review and meta-analysis
Supplemental material, sj-docx-1-cre-10.1177_02692155211040199 for Effect of muscle strength training in children and adolescents with spastic cerebral palsy: A systematic review and meta-analysis by Javier Merino-Andrés, Agustín García de Mateos-López, Diane L Damiano and Alberto Sánchez-Sierra in Clinical Rehabilitation
Footnotes
Author contributions
Conception and design: AGM-L and AS-S. Collection and assembly of data: AGM-L and JM-A. Analysis and interpretation of data: All authors. Drafting of the article: All authors. Critical revision of the article for important intellectual content: JM-A and DM. All authors take responsibility for the integrity of that data analysis.
PROSPERO database,registration number ID
CRD42020193535.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
