meta-analysis of the effect of strengthening interventions in individuals with cerebral palsy
TRANSCRIPT
Research in Developmental Disabilities 35 (2014) 239–249
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Review article
Meta-analysis of the effect of strengthening interventions in
individuals with cerebral palsyEun-Young Park a,1, Won-Ho Kim b,*a Department of Secondary Special Education, College of Education, Jeonju University, 45 Baengma-gil, Wansan-gu, Jeonju, Jeollabuk-do,
Republic of Koreab Department of Physical Therapy, Ulsan College, PO Box 682-715, 101 Bongsuro, Dong-gu, Ulsan, Republic of Korea
A R T I C L E I N F O
Article history:
Received 21 May 2013
Received in revised form 20 October 2013
Accepted 21 October 2013
Available online
Keywords:
Cerebral palsy
Meta-analysis
Strengthening
A B S T R A C T
This study aimed to investigate the evidence that strengthening interventions can improve
muscle strength and activity in individuals with cerebral palsy. The search focused on
studies that employed strength training for children with cerebral palsy for which six
electronic databases were used to extract literature published from 2001 to 2012. The key
terms used in these searches were combined strength training, strengthening, weight
training, weight lifting, resistance, and cerebral palsy. The quality of each study was
assessed using the PEDro (Physiotherapy Evidence Database) scale. Thirteen randomized
controlled trial studies were selected and divided into categories according to program
type, mode, and outcome measures. The overall effect sizes of each study and types of
strengthening were large. Strengthening exercise improved muscle strength to a greater
degree, when practiced 3 times per week in 40–50 min sessions than in other categories of
session length, and greater improvement was observed in younger children than in older.
The effect size of the activities and variables related to gait, except for gait endurance, were
medium to large. The effect size of individual muscles was large, but the effect sizes for
ankle plantar flexor, hip abductor/adductor, and extensor were insignificant. Strengthen-
ing interventions are useful for increasing muscle strength in individuals with cerebral
palsy, specifically in youth and children, and optimal exercise consisted of 40- to 50-min
sessions performed 3 times per week. Although strengthening interventions may improve
activities, including gait, more studies that are rigorous are needed to determine the
contributions to gross motor function.
� 2013 Elsevier Ltd. All rights reserved.
Contents
1. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
1.1. Study search procedures and inclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
1.2. Quality assessment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
1.3. Coding of studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
1.4. Computation of effect sizes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
1.5. Heterogeneity test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
1.6. Test for publication bias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
1.7. Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
* Corresponding author. Tel.: +82 52 230 0785; fax: +82 52 230 0780.
E-mail addresses: [email protected] (E.-Y. Park), [email protected] (W.-H. Kim).1 Tel.: +82 63 220 3186; fax +82 10 220 2054.
0891-4222/$ – see front matter � 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ridd.2013.10.021
E.-Y. Park, W.-H. Kim / Research in Developmental Disabilities 35 (2014) 239–249240
2. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
2.1. Study characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
2.2. Publication bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
2.3. Overall analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
2.4. Sub-group analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
2.4.1. Type of program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
2.4.2. Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
2.4.3. Mode of program. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
3. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
Although muscle weakness is a common impairment in children with cerebral palsy (CP) (Damiano, Vaughan, & Abel,1995; Wiley & Damiano, 1998), general opinion holds that weakness is not a major problem. Historically, with the advent ofneurodevelopmental therapy, muscle strengthening was not recommended for children with CP in clinical settings becauseit was believed that strengthening therapy would lead to increased spasticity and that children with CP would not benefitfrom resistance training (Damiano et al., 1995; Mockford & Caulton, 2008). Treatment for inhibiting spasticity wasemphasized, and excessive effort was avoided because it was thought that exercise induced spasticity and was ultimatelyassociated with loss of motor control (Bobath, 1990). A recent systematic review (Franki et al., 2012), however, concludedthat muscle strengthening improved the level of body structure and function (i.e., strength), and activity without adverseeffects in children with CP. Now, strengthening is adopted widely by clinicians to improve muscle strength and motorfunction in children with CP (Lee et al., 2013; Scholtes et al., 2010).
Experimental studies have investigated the effect of muscle strengthening in individuals with CP. Researchers havereported that strengthening induces positive changes in strength (Lee, Sung, & Yoo, 2008), gross motor function (Lee et al.,2008; Scholtes et al., 2012), gait (Dodd, Taylor, & Graham, 2003; Scholtes et al., 2012), and spasticity (Engsberg, Ross, &Collins, 2006). Hong et al. (2012) confirmed that knee flexor torques at 608/s and 908/s were predominantly related tobalance (r = 0.167) and strength (r = 0.243) while knee flexor torques at 1208/s predominantly contributed to running speedand agility (r = 0.372). Boyd (2012) reported that muscle strength improved significantly more in the intervention group thanin the control group immediately after the intervention by 14% body weight in school-aged ambulatory children with CP.
Despite the experimental studies that reported the relatively consistent and positive results of strengthening inindividuals with CP, the evidence for the effect of strengthening is insufficient (Franki et al., 2012). There have been reviewsof the studies on strengthening for individuals with CP (Darrah, Fan, Chen, Nunweiler, & Watkins, 1997; Dodd, Taylor, &Damiano, 2002; Mockford & Caulton, 2008; Scianni, Butler, Ada, & Teixeira-Salmela, 2009; Verschuren et al., 2011). Two ofthese reviews are narrative, and three are systematic, analyzing the effect size. Darrah et al. (1997) reviewed seven articlesand reported an improvement in performance after participation in a resistance exercise program. Verschuren et al. (2011)summarized the articles to evaluate the training protocols from the most recent randomized training in relation to thestrength training guidelines of the National Strength and Conditioning Association (NSCA). Through a meta-analysis of 11articles, Dodd et al. (2002) reported that strength-training programs for people with CP could increase strength and mayimprove motor activity without adverse effects. Mockford and Caulton (2008) undertook a systematic review to capture andanalyze the evidence concerning the effects of progressive strength training on function and gait in ambulatory children andadolescents with CP. Scianni et al.’s (2009) meta-analysis of five studies suggest that strengthening interventions are noteffective.
The results of the reviews are controversial, and the effects of exercise remain unclear. Short-term interventions andmethodological shortcomings, such as small sample sizes and inadequate study designs, have prevented researchers fromreaching explicit conclusions (Darrah et al., 1997). Because of this insufficient evidence, caution must be used in interpretingthe results. Reviews have limitations in explaining the comprehensive effect of strengthening for individuals with CP,including the small number and low quality of the studies and their heterogeneous design. The data of five studies showed noeffect of muscle strengthening (Scianni et al., 2009), while the effectiveness of strength training was observed in 11 studiesthat employed a heterogeneous study design (Dodd et al., 2002).
A systematic and comprehensive review of previous research is beneficial to guide future research and practice. There isno doubt that the muscles of individuals with CP are weak, and there is scope for strengthening those (Verschuren et al.,2011). To understand the comprehensive effect of strength training on individuals with CP, it is necessary to determineappropriate methods for their program and to identify exercise prescriptions, such as the length and frequency of exerciseprograms. There is a lack of data on the moderators of the effects of exercise for individuals with CP. The advantage of meta-analysis is that it is possible to compare studies that differ in experimental rigor and other methodological factors (Lipsey,2008).
This study aims to synthesize the effects of strengthening on individuals with CP. The specific research questions are asfollows: (a) what is the magnitude of the effect of strengthening? (b) Which type of strengthening is the most effective? (c)What is the most effective outcome of strengthening? (d) Which mode of strengthening program has the most influence onoutcomes?
E.-Y. Park, W.-H. Kim / Research in Developmental Disabilities 35 (2014) 239–249 241
1. Methods
1.1. Study search procedures and inclusion criteria
This research focused on studies that employed strength training for children with CP; thus, the data were collected fromstudies reported the effects of strengthening. A wide variety of electronic and print resources was screened to identifyarticles for possible inclusion in this study. A literature search was conducted using the following electronic databases:PubMed, Web of Science, PsychINFO, Physiotherapy Evidence Database (PEDro), CINAHL, and Sports Discuss. Thesedatabases were widely employed for previous systematic and meta-analyses. The key terms used in these searches werecombined strength training, strengthening, weight training, weight lifting, resistance, and CP. A supplementary manualsearch was carried out by examining the reference lists of selected articles. The search for studies was not limited by age ofparticipants. As seen in Fig. 1, 980 studies were identified in this initial search of journal articles, and 184 duplicated studieswere excluded. The titles and abstracts of the studies identified by this search strategy were screened using five inclusioncriteria. First, participants must have been identified as having CP. Second, the study must have assessed changes in outcomerelated to strength. Third, the study must have employed a randomized controlled trial (RCT). Fourth, the study must havebeen published in a peer-reviewed journal between 2001 and 2012. Fifth, the study must have provided sufficientinformation to compute the effects, such as mean and standard deviation, of experimental and control groups. The exclusioncriteria were as follows: (a) master’s and doctoral dissertations, (2) reports published in books and conference proceedings,(3) experimental methods including single-subject, case study, quasi-randomized, or controlled trial, and (4) correlation andreview studies. Full paper copies of 62 selected studies were retrieved and read to determine whether they met all inclusioncriteria. Through this reading, 49 non-eligible studies were excluded. In a second step, all non-eligible studies were excludedin the first screening stage. Finally, 13 studies were retained for this meta-analysis.
1.2. Quality assessment
The methodological quality of each study was assessed using the PEDro scale (Verhagen et al., 1998) and two independentreviewers. Inter-rater disagreements were resolved by discussion. The PEDro scale consists of 11 items related to the qualityof the study as follows: (1) specification of eligibility criteria, (2) randomly allocated subjects, (3) concealed allocation, (4)homogeneity of important prognostic indicator in baseline, (5) blinding, (6) blinded therapy, (7) assessor blinding, (8)obtained data at least on key outcome from more than 85%, (9) analysis was ‘‘intention to treat’’ for at least one key outcome,(10) between-group statistical comparison of at least one key outcome, and (11) provided point measure and measures ofvariability data for at least one key outcome. The range of sum score was from 0 to 10 because criterion 1 was related toexternal validity (Verhagen et al., 1998). The PEDro scale scores of 9–10 are considered to be of excellent quality, scores of 6–8 points and 4–5 points to be of good and fair quality, respectively, and scores below 4 to be of poor quality (Foley, Teasell,
[(Fig._1)TD$FIG]
Total 980 searched
∙ 242 identified through PubMed
∙ 332 identified through web of science
∙ 67 identified through PsychINFO
∙ 60 identified through PEDro
∙ 172 identified through CINAHL
∙ 107 identified through Sports Discuss
796 screened
∙ 184 duplicates
734 excluded
∙ 469 not strengthening
∙ 224 not intervention
∙ 41 not group experimental study
62 selected 49 non-eligible studies
· 5 no statistical data
· 14 not strengthening
· 18not RCT
· 2 not children with CP
· 10 not group experimental study design
13 in meta-analysis
Fig. 1. Flowchart of analysis through different phases of the meta-analysis.
E.-Y. Park, W.-H. Kim / Research in Developmental Disabilities 35 (2014) 239–249242
Bhogal, & Speechley, 2003). Previous meta-analysis (Dodd et al., 2002) employed guidelines where a study was considered tobe of low methodological quality, if the PEDro scale score was less than three points. Thus, we also employed criteria belowthree points to screen low quality studies. The mean PEDro score of the studies was 7.23 (range from 6 to 8) in this meta-analysis.
1.3. Coding of studies
The study characteristics were coded to reflect the potential moderating variables for the effect of strengthening onindividuals with CP. These characteristics included the following: (a) author and publication year, (b) type of strengthening,(c) mode of the strengthening program, (d) the sample compositions, and (e) the outcomes of the participants. Based on theframework proposed for reviewing the effects of strengthening in children with CP (Darrah et al., 1997; Dodd et al., 2002;Scianni et al., 2009), we considered the following four outcome categories: activity, gait, strength, and others. For this meta-analysis, two reviewers independently coded all the features of the selected studies according to a pre-developed manual,including information on the study characteristics and data synthesis. Unresolved discrepancies were resolved by discussionwith a third reviewer.
1.4. Computation of effect sizes
Effect size (ES) was calculated using the standardized mean difference (Borenstein, Hedges, Higgins, & Rothstein, 2009).The size of the effect was reported as the weighted mean difference with 95% confidence intervals (CI) (Hedges & Olkin,1985). In accordance with Cooper’s shifting unit of analysis (1998), for this meta-analysis, the unit of analysis for the total ESestimation was a study, whereas the unit of analysis for the sub-group analysis was an ES.
1.5. Heterogeneity test
Overall ES in a meta-analysis is affected by whether or not the results of the individual studies are consistent. If the effectvaries substantially among the studies, the overall ES may shift due to the dispersion. For that reason, to estimate the overallES in a meta-analysis, reviewers should examine whether the ES are homogenous. The results of the heterogeneity testprovide information on model selection for analysis and motivation for the categorical analysis to identify moderatorvariables. Since the homogeneity test was statistically significant, the random effects model was used for the major effectand sub-group analysis, which allowed for generalization to a larger population (Sirin, 2005).
1.6. Test for publication bias
The presence of a publication bias will result in an overestimation of meta-analysis results (Rosenthal, 1979; Rosenberg,2005). To check for publication bias in this research, the Duval and Tweedie’s Trim and Fill, fail-safe N, and Egger’s regressioninterceptors (Egger, Smith, Schneider, & Minder, 1997) were used. The Trim and Fill initially trims the asymmetric studiesfrom the right side to locate the unbiased effect through an iterative procedure and fills the plot by re-inserting the trimmedstudies on the right as well as their imputed counterparts to the left for the mean effect. The fail-safe N is a reference to thenumber of missing studies that would nullify the effect. If a publication bias exists, the p value is below 0.05 in the Egger’sregression intercept test.
1.7. Statistical analysis
Comprehensive Meta-Analysis 2.0 (Biostat, Englewood, NJ, USA) was used to analyze the effect size and statisticalheterogeneity in the selected studies.
2. Results
2.1. Study characteristics
The summary of the study characteristics is presented in Table 1. This synthesis provided ESs for 150 standardized meandifference from 13 primary studies. The results of the homogeneity test were significant (Q = 638.362, df = 12, p< 0.05).Therefore, we also conducted a random effects model analysis.
2.2. Publication bias
There was no evidence of publication bias in this study. The number of studies in the Trim and Fill was zero, and adjustedvalues were the same as the random ES of observed values. The fail-safe N was found to be 1992, which meant that 1992‘‘null’’ studies were needed for the combined two-tailed p-value to exceed 0.05. The results of the Tweedie’s Trim and Fill areshown in Fig. 2. The p-value was 0.99 in Egger’s test of the intercept.
Table 1
Characteristics of the studies included in the analysis.
Study Effect size Program Outcome Sample size Age CP class Times/
week
Duration
(weeks)
Session
length (min)
Chen et al. (2012) 1.363 Virtual cycling training Activity, strength, others E = 13; C = 14 8.7 Spastic di & hemi 3 12 40
Dodd et al. (2003) 0.523 Progressive resistance exercise Activity, strength E = 11; C = 10 13.1 Spastic di 3 6 25
Engsberg et al. (2006) 1.855 Progressive resistance exercise Activity, strength, others E = 9; C = 3 9.9 Spastic di 3 12 N/A
Fowler et al. (2010) 0.269 Cycling training Activity, gait E = 29; C = 29 11.1 Spastic di 3 12 60
Kerr et al. (2006) 1.532 Electrical stimulation Activity, strength E = 17; C = 19 11.0 Not specific 5 16 60
Lee et al. (2008) 1.430 Strengthening program Activity, gait, strength E = 9; C = 8 6.3 Spastic di & hemi 3 5 60
Liao et al. (2007) 0.830 Loaded sit-to-stand
resistance exercise
Activity, gait, strength E = 10; C = 10 7.4 Spastic di 3 6 25
Maeland et al. (2009) 0.422 Progressive resistance exercise Activity, gait, strength E = 6; E = 6 41.2 Spastic di 3 8 N/A
Scholtes et al. (2010) 1.320 Progressive resistance exercise Activity, strength E = 24; C = 23 10.3 Spastic uni & bi 3 12 50
Scholtes et al. (2012) 0.855 Progressive resistance exercise Activity, gait, strength, others E = 24; C = 25 10.4 Spastic uni & bi 3 12 60
Unger et al. (2006) 1.113 Strengthening program Gait, others E = 22; C = 13 15.9 Spastic di & hemi 3 8 50
Unnithan et al. (2007) 1.010 Strength and aerobic exercise Activity, others E = 6; C = 6 15.9 Spastic di 3 12 70
van der Linden et al. (2003) 0.868 Electrical stimulation Activity, gait, strength E = 11; C = 11 8.5 Di, hemi & quadri 6 8 60
Note: E = number of subjects in experimental group; C = number of subjects in control group.
E.-Y
.P
ark
,W
.-H.
Kim
/Resea
rchin
Dev
elop
men
tal
Disa
bilities
35
(20
14
)2
39
–2
49
24
3
[(Fig._3)TD$FIG]
Study name Statistics for each study
Effectsize
Standarderror
Lowerlimit
Upperlimit
Variance Z-Value p-Value
Chen et al(2012) 1.363 0.307 0.094 0.761 1.965 4.439 0.000
Dodd et al(2003) 0.523 0.161 0.026 0.207 0.838 3.246 0.001
Engsberg et al(2006) 1.855 0.428 0.183 1.016 2.694 4.333 0.000
Fowler et al(2010) 0.269 0.154 0.024 -0.032 0.571 1.752 0.080
Kerr et al(2006) 1.532 0.360 0.130 0.827 2.238 4.256 0.000
Lee et al(2008) 1.430 0.169 0.028 1.099 1.760 8.485 0.000
Liao et al(2007) 0.890 0.307 0.094 0.289 1.491 2.902 0.004
Maeland et al(2009) 0.422 0.222 0.049 -0.013 0.858 1.900 0.057
Scholtes et al(2010) 1.320 0.285 0.081 0.760 1.879 4.624 0.000
Scholtes et al(2012) 0.855 0.181 0.033 0.501 1.209 4.733 0.000
Unger et al(2006) 1.113 0.325 0.106 0.476 1.751 3.423 0.001
Unnithan et al(2007) 1.040 0.257 0.066 0.536 1.544 4.047 0.000
van der Linden et al(2003) 0.868 0.241 0.058 0.395 1.341 3.599 0.000
Overall 0.861 0.063 0.004 0.738 0.984 13.759 0.000
-4.00 -2.00 0.00 2.00 4.00
Favors control Favors intervention
Fig. 3. Forest plot of 13 studies.
[(Fig._2)TD$FIG]
-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0
0.0
0.1
0.2
0.3
0.4St
anda
rd E
rror
Std diff in means
Fig. 2. Funnel plot of the Tweedie’s trim and fill.
E.-Y. Park, W.-H. Kim / Research in Developmental Disabilities 35 (2014) 239–249244
2.3. Overall analysis
Fig. 3 shows the forest plots of all 13 studies, including descriptive statistics, such as variance, standard errors, andconfidence intervals to identify the precision of each study. The effect of exercise was 0.861 standard deviation, which isconsidered as large (Cohen, 1988).
2.4. Sub-group analysis
This study performed a sub-group analysis using a random effects model because based on the heterogeneity tests, eachsub-group was heterogeneous. A sub-group analysis was conducted to identify the source of variability, and the moderators
Table 3
The effect sizes by outcomes.
Outcomes Sub-outcomes k �95%CI ES +95%CI SE
Activity GMFM 13 0.260 0.565 0.870 0.156
Other gross motor measures 5 0.286 0.787 1.288 0.256
Sit to stand 3 0.535 1.175 1.814 0.326
Stair climbing 5 0.220 0.552 0.884 0.169
Others 4 0.224 1.858 3.492 0.834
Subtotal 30 0.474 0.668 0.862 0.099
Gait Endurance 1 �0.994 0.139 1.272 0.578
Kinematic 23 1.209 1.671 2.133 0.236
Kinetic 12 0.586 0.916 1.246 0.168
Spatial parameter 11 0.452 0.900 1.348 0.229
Speed 12 0.166 0.475 0.784 0.158
Subtotal 59 0.675 0.858 1.040 0.093
Strength Ankle plantar flexor 3 �0.017 0.349 0.715 0.187
Hip abductor 3 1.408 1.850 2.291 0.225
Hip adductor 1 �0.404 0.568 1.539 0.496
Hip extensor 3 �0.036 1.538 3.112 0.803
Hip flexor 3 0.157 0.772 1.387 0.314
Knee extensor 16 0.983 1.463 1.943 0.245
Knee flexor 5 1.148 1.150 2.152 0.511
Others 5 0.463 1.225 1.988 0.389
Subtotal 39 0.845 1.050 1.255 0.011
Others Cardiopulmonary data 13 0.531 1.018 1.505 0.248
ROM 6 0.453 1.022 1.592 0.291
Spasticity 1 4.220 5.464 6.707 0.635
Self-perception 2 �1.444 1.876 5.915 1.694
Subtotal 22 0.848 1.339 1.830 0.250
Note: GMFM = gross motor function measure; k = number of effect size; CI = confidence interval; ES = effect size; SE = standard error.
Table 2
The effect size according to the type of program.
Type of program k �95%CI ES +95%CI SE
Strength with aerobic exercise 25 0.698 1.043 1.387 0.176
Electrical stimulation 18 0.753 1.174 1.595 0.215
Strengthening exercise 107 0.930 1.105 1.280 0.089
Note: k = number of effect size; CI = confidence interval; ES = effect size; SE = standard error.
E.-Y. Park, W.-H. Kim / Research in Developmental Disabilities 35 (2014) 239–249 245
affecting the direction and amount of relation and difference among sub-groups. The categorical variables were the type ofprogram, outcome, age, and mode, including the duration, frequency, and session length.
2.4.1. Type of program
The ESs according to the categorical analysis by the type of program in the random effects model are shown in Table 2. TheES of electrical stimulation was 1.174. The ESs of strength with aerobic exercise and with only strengthening was 1.043 and1.105, respectively.
2.4.2. Outcomes
The results of the random-effects categorical analysis by outcome are presented in Table 3. The results of strengtheningfor individuals with CP varied. In particular, the ES of activity was 0.668, and for gait and strength, it was 0.858 and 1.050,respectively, and of others, including cardiopulmonary data, range of motion (ROM), spasticity, and self-perception, 1.339.The large ESs above 0.8 of the sub-outcomes are sit-to-stand in the activity outcome; kinematic, kinetic, and parameter ingait; hip abductor, knee extensor, and knee flexor in strength; and spasticity in others. There was an effect of thestrengthening intervention for the endurance and strength of the ankle plantar flexor, hip adductor, and hip extensor, andself-perception, as zero was included in their confidence intervals.
2.4.3. Mode of program
ES of both short-duration (below 8 weeks) and long-duration programs (8–12 weeks) were large. In detail, the ES of theshort-duration program was 1.111, and the ES of the long-duration program was 1.096 (Table 4). The ESs varied according tofrequency of strengthening (d = 0.869–1.532). The ESs of practicing five times per week appeared larger than the ESs ofpracticing three and eight times. In addition, the ESs for the session length ranged from 0.560 to 1.363. The ESs for practicing
Table 4
Effect sizes by subgroup: duration, frequency, session length, and age.
Subgroup Categories k �95%CI ES +95%CI SE
Duration Below 8 weeks 61 0.889 1.111 1.334 0.114
8–12 weeks 89 0.958 1.096 1.287 0.097
Frequency 3 per week 132 0.938 1.094 1.249 0.079
5 per week 8 0.827 1.532 2.238 0.360
6 per week 10 0.395 0.868 1.341 0.241
Session length 25 min 14 0.299 0.560 0.821 0.133
40 min 9 0.761 1.363 1.965 0.307
50 min 22 0.814 1.225 1.636 0.210
60 min 76 0.936 1.139 1.342 0.104
70 min 13 0.536 1.040 1.544 0.257
Age Below 7 years 36 1.099 1.430 1.760 0.169
7–13 years 84 0.857 1.041 1.226 0.094
Above 13 years 30 0.597 0.931 0.597 0.171
Note: k = number of effect size; CI = confidence interval; ES = effect size; SE = standard error.
E.-Y. Park, W.-H. Kim / Research in Developmental Disabilities 35 (2014) 239–249246
40 min per session were larger than the ESs for other session lengths. The ES (d = 1.430) were larger when the participantsaged below seven than for other ages.
3. Discussion
Muscle weakness is considered a common impairment in individuals with CP (Wiley & Damiano, 1998) and is related toactivity and function (Kim & Park, 2011). Although recent studies reported evidence that supported the effect ofstrengthening intervention in individuals with CP (Franki et al., 2012; Scholtes et al., 2010), the question about effectivenessof strengthening remains. Previous meta-analyses did not provide sufficient evidence because they analyzed a small numberof studies and articles, including both RCT and other quasi-experimental studies. Even a systematic review reported nosignificant effect of strengthening (Scianni et al., 2009). This meta-analysis aimed to confirm the effectiveness of musclestrengthening intervention and its modalities on rehabilitation outcomes in individuals with CP. This meta-analysis mayprovide strong evidence for the effect of strengthening as studies of poor quality were excluded through evaluation based onthe PEDro score, and 13 studies of high quality were analyzed, and their effect sizes were computed. Meta-analysis of RCTs isconsidered robust for appraising the outcomes of an intervention because of minimal biases (Harbour & Miller, 2001). Theresults of this meta-analysis showed a large effect (d = 0.861) of strengthening interventions on individuals with CP.Unfortunately, as Dodd et al. (2002), Mockford and Caulton (2008), and Scianni et al. (2009) did not calculate overall ES, theoverall ES of this study could not be directly compared with these studies.
A sub-group analysis was performed using a random effects model because of heterogeneity. The categorical variables ofthe sub-group analysis were the program type, outcome, and mode. In the analysis of the program type, electricalstimulation showed a larger ES (d = 1.174) than strengthening exercise (d = 1.105) and strength with aerobic exercise(d = 1.043). The largest ES of electrical stimulation was interpreted with caution because it was based on two trials, and therange of confidence interval was wide. The fact that the largest number of ESs (k = 107) appeared for strengthening exercises,such as progressive resistant exercise, suggested that it was an accepted method for strengthening in individuals with CP.Although the ES of this meta-analysis could not be compared because previous systematic reviews did not report the overallES, it is clear that strengthening interventions have benefit and large effect on individuals with CP.
Gait was an outcome with one of the greatest numbers (k = 59) and largest magnitudes (d = 0.858) of ES (Table 4). The sub-group analysis indicated an insignificant ES for endurance and median of speed (d = 0.475). Gait receives a particularly highdegree of attention in individuals with CP. Most studies included in Mockford and Caulton’s (2008) review assessed gait(range of ES = 0.010–0.613). Four studies using 3D gait analysis and one using visual gait analysis found improvements insome gait parameters in the study by Mockford and Caulton (2008). This is similar to the large ESs of the kinematic, kinetic,and spatial parameters in this study. Speed showed relatively smaller ESs than other gait parameters except insignificantendurance. Leg strength and walking speed might have a non-linear relationship (Buchner, Larsen, Wagner, Koepaell, & deLateur, 1996), and there is a possibility that an increase of strength would not affect walking speed (Scianni et al., 2009).Walking speed did not affect strength in the review by Scianni et al. (2009), and the timed walking test showed mixed resultsin the Mockford and Caulton (2008) review. Given that strengthening focused on body function (motor impairment), gaitspeed, and activity level endurance might have slightly affected strengthening.
Strength has a lower number of ESs (k = 39), but the ES itself is larger ES (d = 1.050). The reason that muscle strength has asmaller number of ES than gait despite strengthening intervention is thought to be because of an increasing awareness ofactivities and participation arising from enablement models such as the International Classification of Functioning, Disabilityand Health (ICF) (WHO, 2008). The variable of the knee extensor was found to have a large ES (d = 1.463). Previous systematicreviews reported large ESs of strength (Dodd et al., 2002; Mockford & Caulton, 2008). An insignificant and small number of
E.-Y. Park, W.-H. Kim / Research in Developmental Disabilities 35 (2014) 239–249 247
ESs for strength in the ankle plantar flexor, hip adductor, and hip extensor suggested that more studies were needed to clarifythe effect of strengthening intervention in these muscles.
Activity, including gross motor function measure (GMFM), other gross motor measures, sit-to-stand activities, and stairclimbing showed 30 numbers and moderate (d = 0.668) ES. The ES appeared smaller than in the review by Dodd et al. (2002).The total score of GMFM showed no significant effect in the review by Dodd et al. (2002), but GMFM and other measures ofgross motor function in this meta-analysis found moderate ES. Since performance was affected by many factors, such assensory acuity and coordination, strengthening focused on increasing muscle strength might have a smaller ES than musclestrength. Dodd et al. (2002) suggested that an individualized exercise program was needed to draw out better functionaloutcomes of strengthening in individuals with CP. The moderate ES of activity in this meta-analysis indicates that furtherstudies of specific strengthening programs focused on individual activity needs should be performed.
There is little evidence to identify the optimal mode of intervention for strengthening because various modes instrengthening studies have been reported (Verschuren et al., 2007). For the investigation of ES by the characteristics of thestrengthening intervention, this systematic review categorized age of the subject and the frequency, duration, and sessionlength of the intervention (Table 4). The relationship between exercise intensity and effectiveness is a critical issue. Thephysiological response to exercise is dependent on the intensity, duration, and frequency of exercise. Faigenbaum et al.(2009) suggest that resistance training for strength is regarded optimal at 2–3 times per week. In this meta-analysis, themajority of the studies provided training intervention three times per week, except in the case of two trials (Kerr et al., 2006;Linden, Hazlewood, Aitchison, Hillman, & Robb, 2003). Although five times per week showed a larger effect size than threetimes, these trials applied to electrical stimulation for strengthening. As shown by the large ESs of both interventions,strengthening exercise three times a week and electrical stimulation (especially in the case of impossible exercise activity)during spare time, without inducing fatigue, are effective for improving muscle strength.
A session length of 40–50 min revealed a larger ES than other lengths. The session length is somewhat long becausestrengthening intervention is based on functional activities rather than on individual muscular strengthening (Salem &Godwin, 2009; Scholtes et al., 2012). In the case of children (below 13 years old), ESs were large. However, adolescents (above13 years old) had a medium ES. The results of this systematic review coincided with previous study. Franki et al. (2012)stated that a young child experiences greater gains from strength training than an older child does. Duration of strengtheningintervention in this study was 5–12 weeks. Duration of training was categorized as below 8 weeks (6 trials) and 8–12 weeks(7 trials) because physical adaptation of exercise occurs at approximately 6–8 weeks (Kraemer et al., 2002), and most ofstudies used the interval of 6–8 weeks (Verschuren et al., 2007). The ESs of both durations were similar and large. Increase inmuscular strength occurs primarily during earlier weeks of strength training mainly because of neural mechanisms inchildren (Faigenbaum et al., 2009). Because the duration of the reviewed studies was relatively short, the increase ofmuscular strength might have been influenced by neural factors. Further study is necessary to examine the contribution ofmuscular hypertrophic and hormone factors after prolonged training.
Strengthening programs for individuals with CP have traditionally been provided for muscular strength and functionalactivity. This is not applicable only to individuals with spastic CP. However, there are few studies regarding strengthening fordyskinetic and ataxic types of CP. Thus, the subjects of the studies included in this meta-analysis were mostly of the spastictype. The assumed reason for this tendency follows. First, spastic and mixed motor disorder accounts for more than 85% ofchildren with CP, and dyskinetic type is less common (Graham & Selber, 2003). Second, intervention for dyskinetic CP wasprimarily focused on control of involuntary movement (Himmelmann, Hagberg, Wiklund, Eek, & Uvebrant, 2007). Types ofCP should be considered in care and intervention. Although the trials for strengthening in dyskinetic and ataxic CP (Darrah,Wessel, Nearingburg, & O’Connor, 1999) were not included in this meta-analysis due to lack of RCT, more trials should beaccumulated for evidence of strengthening as per type of CP. Recently, Scianni et al. (2009) reported that strengtheninginterventions were neither effective nor worthwhile through a meta-analysis of five RCT studies. Mockford and Caulton(2008) found effectiveness of a strengthening program in a meta-analysis of 13 trials (any experimental and single-groupdesign). In this systematic review of 13 RCT studies, we found positive effects of strengthening interventions on musclestrength and activities.
Walking is closely related to functional outcomes in CP (Sullivan et al., 2007) and is important for independence. Althoughmuscles of the trunk and upper extremities are used during walking, lower extremities take charge of the essentialcomponents of walking activities, such as support and propulsion. Thus, previous studies conducted on strengthening ofchildren with CP focused on the muscles of the lower extremities. No RCT studies on strengthening of the upper extremitieswere included in this study. For these reasons, unfortunately, we could not analyze the effect of strengthening of the upperextremities.
Although this meta-analysis showed large positive effects, some limitations should be noted. First, the number of studiesused in the meta-analysis was not enough, especially while interpreting sub-group analysis because this study limited thesearch to studies published between 2001 and 2012, and extracted articles published in English from six electronicdatabases. Unpublished or other-language studies would affect the results of meta-analysis (Egger et al., 1997). Although it ispossible to do a meta-analysis of as few as two studies, a small number of studies in a meta-analysis might lead to largesampling error in ES. Second, participants of studies included in this meta-analysis were mostly capable of walking with orwithout gait aid. We obtained evidence for positive outcomes of strengthening intervention for individuals with ambulatoryCP but could not determine the effectiveness of strengthening intervention for non-ambulatory CP. Third, intensity andvolume of exercise are important for muscular strength (Verschuren et al., 2011), but we could not analyze because of the
E.-Y. Park, W.-H. Kim / Research in Developmental Disabilities 35 (2014) 239–249248
difficulty of categorizing intensity and volume of exercise in included studies. Although studies applied resistance forstrengthening, each study varied in modality and criteria for progressive load.
More research is needed regarding strengthening in CP. As described above, further research is necessary to confirm theeffectiveness of strengthening for individuals with non-ambulatory CP and other types of CP and in various muscle groups,such as trunk and upper extremities. The effect of specific exercise programs based on individuals’ needs should also beinvestigated. In addition, the ES should be calculated according to the intensity and volume of exercise, establishing astandard category for individuals with CP in the future.
This study was the first trial investigating the ES of strengthening, according to the mode of exercise. Although previousresults have provided clinical implications regarding the mode of exercise, the data were insufficient to determine the mosteffective mode of exercise and the relationship between the mode and effectiveness. It would be possible to determine theoptimal mode of exercise through a meta-regression analysis after accumulating the data from various studies.
4. Conclusions
The effect of strengthening interventions on improving muscle strength and activity in individuals with cerebral palsy hasbeen debated. Meta-analysis could assist in collecting and analyzing relevant evidence-based research. In this study, thirteenRCT studies were selected and analyzed. It was found that strengthening and electrical stimulation interventions are usefulto increase muscle strength in individuals with cerebral palsy, specifically in children and youth. The optimal mode ofexercise consisted of 40–50 min sessions three times per week. Strength of the trained muscle and gait parameter weregreatly influenced through strengthening. Although the number of studies and ESs were not sufficient to make a definitiveconclusion, this meta-analysis provided evidence that strengthening intervention led to positive outcomes in individualswith CP. Further evidence will be accumulated to verify the specific effect of strengthening in individuals with CP.
References2
Bobath, B. (1990). Adult hemiplegia: Evaluation and treatment (3rd ed.). London: William Heinemann Medical Books.Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009). Introduction to meta-anaylsis. West Sussex, UK: Wiley.Boyd, R. N. (2012). Functional progressive resistance training improves muscle strength but walking ability in children with cerebral palsy. Journal of
Physiotherapy, 58(3), 197. http://dx.doi.org/10.1016/S1836-9553(12)70111-XBuchner, D., Larsen, E., Wagner, E., Koepaell, T. D., & de Lateur, B. J. (1996). Evidence for a non-linear relationship between leg strength and gait speed. Age Ageing,
25, 386–391.*Chen, C. L., Hong, W. H., Cheng, H. Y. K., Liaw, M. Y., Chung, Ch. Y. , & Chen, C. Y. (2012). Muscle strength enhancement following home-based virtual cycling
training in ambulatory children with cerebral palsy. Research in Developmental Disabilities, 33(4), 1087–1094. http://dx.doi.org/10.1016/j.ridd.2012.01.017Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.Cooper, H. (1998). Synthesizing research: A guide for literature reviews (3rd ed.). Thousand Oaks, CA: Sage.Damiano, D. L., Vaughan, C. L., & Abel, M. F. (1995). Muscle response to heavy resistance exercise in children with spastic cerebral palsy. Developmental Medicine &
Child Neurology, 37, 731–739.Darrah, J., Fan, J. S., Chen, L. C., Nunweiler, J., & Watkins, B. (1997). Review of the effects of progressive resisted muscle strengthening in children with cerebral
palsy: A clinical consensus exercise. Pediatric Physical Therapy, 9, 12–17.Darrah, J., Wessel, J., Nearingburg, P., & O’Connor, M. (1999). Evaluation of a community fitness programme for adolescents with CP. Pediatric Physical Therapy, 11,
18–23.*Dodd, K. J., Taylor, N. F., & Graham, H. K. (2003). A randomized clinical trial of strength training in young people with cerebral palsy. Developmental Medicine &
Child Neurology, 45(10), 652–657.Dodd, K. J., Taylor, N. F., & Damiano, K. L. (2002). A systematic review of the effectiveness of strength-training program for people with cerebral palsy. Archives
Physical Medicine and Rehabilitation, 83, 1157–1164.Egger, M., Smith, G. D., Schneider, M., & Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. British Medical Journal, 315, 629–634.*Engsberg, J. R., Ross, S. A., & Collins, D. R. (2006). Increasing ankle strength to improve gait and function in children with cerebral palsy: A pilot study. Pediatric
Physical Therapy, 18(4), 266–275.Faigenbaum, A. D., Kraemer, W. J., Blimkie, C. J., Jeffreys, I., Micheli, L. J., Nitka, M., et al. (2009). Youth resistance training: Updated position statement paper from
the national strength and conditioning association. Journal of Strength and Conditioning Research, 23(5 Suppl.), S60–S79. http://dx.doi.org/10.1519/JSC.0b013e31819df407
Floey, N., Teasell, R. Q., Bhogal, S. K., & Speechley, M. R. (2003). Stroke rehabilitation evidence-based review: Methodology. Topics in Stroke Rehabilitation,11(2), 89–101.
*Fowler, E. G., Knutson, L. M., DeMuth, S. K., Siebert, K. L., Simms, V. D., Sugi, M. H., et al. (2010). Pediatric Endurance and Limb Strengthening (PEDALS) for childrenwith cerebral palsy using stationary cycling: A randomized controlled trial. Physical Therapy, 90(3), 367–381. http://dx.doi.org/10.2522/ptj.20080364
Franki, I., Desloovere, K., de Cat, J., Feys, H., Molenaers, G., Calders, P., et al. (2012). The evidence-base for basic physical therapy techniques targeting lower limbfunction in children with cerebral palsy: A systematic review using the International Classification of Functioning, Disability and Health as a conceptualframework. Journal of Rehabilitation Medicine, 44(5), 385–395. http://dx.doi.org/10.2340/16501977-0983
Graham, H. K., & Selber, P. (2003). Musculoskeletal aspects of cerebral palsy. British Journal of Bone and Joint of Surgery, 85(2), 157–166.Harbour, R., & Miller, J. (2001). A new system for grading recommendations in evidence based guidelines. British Medical Journal, 323, 334–336.Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta-analysis. Orlando: Academic Press.Himmelmann, K., Hagberg, G., Wiklund, L. M., Eek, M. N., & Uvebrant, P. (2007). Dyskinetic cerebral palsy: A population-based study of children born between
1991 and 1998. Developmental Medicine & Child Neurology, 49, 246–251.Hong, W. H., Chen, H. C., Shen, I. H., Chen, C. Y., Chen, C. L., & Chung, C. Y. (2012). Knee muscle strength at varying angular velocities and associations with gross
motor function in ambulatory children with cerebral palsy. Research in Developmental Disabilities, 33(6), 2308–2316. http://dx.doi.org/10.1016/j.ridd.2012.07.010
*Kerr, C., McDowell, B., Cosgrove, A., Walsh, D., Bradbury, I., & McDonough, S. (2006). Electrical stimulation in cerebral palsy: A randomized controlled trial.Developmental Medicine & Child Neurology, 48(11), 870–876.
Kim, W. H., & Park, E. Y. (2011). Causal relation between spasticity, strength, gross motor function, and functional outcome in children with cerebral palsy: A pathanalysis. Developmental Medicine & Child Neurology, 53(1), 68–73. http://dx.doi.org/10.1111/j.1469-8749.2010.03777.x
2 References marked with asterisk indicate studies included in the meta-analysis.
E.-Y. Park, W.-H. Kim / Research in Developmental Disabilities 35 (2014) 239–249 249
Kraemer, W. J., Koziris, L. P., Ratamess, N. A., Hakkinen, K., Tripiplett-McBride, N. T., Fry, A. C., et al. (2002). Detraining produces minimal changes in physicalperformance and hormonal variables in recreationally strength-trained men. Journal of Strength and Conditioning Research, 16(3), 373–382.
*Lee, J. H., Sung, I. Y., & Yoo, J. Y. (2008). Therapeutic effects of strengthening exercise on gait function of cerebral palsy. Disability & Rehabilitation, 30(19), 1439–1444.
Lee, J. A., You, J. H., Kim, D. A., Lee, M. J., Hwang, P. W., Lee, N. G., et al. (2013). Effects of functional movement strength training on strength, muscle size, kinematics,and motor function in cerebral palsy: A 3-month follow-up. NeuroRehabilitation, 32(2), 287–295. http://dx.doi.org/10.3233/NRE-130846
*Liao, H. F., Liu, Y. C., Liu, W. Y., & Lin, Y. T. (2007). Effectiveness of loaded sit-to-sand resistance exercise for children with mild spastic diplegia: A randomizedclinical trial. Archives Physical Medicine & Rehabilitation, 88, 25–31.
Lipsey, M. W. (2008). Design sensitivity: Statistical power for applied experimental research. In L. Bickman & D. J. Rog (Eds.), Handbook of applied social researchmethods (pp. 39–68). Thousand Oaks, CA: Sage.
*Maeland, S., Jahnen, R., Opheim, A., Frosile, K. F., Moe-Nilssen, R. M., & Stanghelle, J. K. (2009). No effect on gait function of progressive resistance exercise in adultswith cerebral palsy: A single-blind randomized controlled trial. European Journal of Physiotherapy, 11(4), 227–233.
Mockford, M., & Caulton, J. M. (2008). Systematic review of progressive strength training in children and adolescents with cerebral palsy who are ambulatory.Pediatric Physical Therapy, 20(4), 318–333. http://dx.doi.org/10.1097/PEP.0b013e31818b7ccd
Rosenberg, M. S. (2005). The file-drawer problem revised: A general weighted method for calculating fail-safe numbers in meta-analysis. Evolution, 59(2),464–468.
Rosenthal, R. (1979). The ‘‘file drawer problem’’ and tolerance for null results. Psychological Bulletin, 86, 638–641.Salem, Y., & Godwin, E. M. (2009). Effects of task-oriented training on mobility function in children with cerebral palsy. NeuroRehabilitation, 24(4), 307–313.
http://dx.doi.org/10.3233/NRE-2009-0483*Scholtes, V. A., Becher, J. G., Janssen-Potten, Y. J., Dekkers, H., Smallenbroek, L., & Dallmeijer, A. J. (2012). Effectiveness of functional progressive resistance exercise
training on walking ability in children with cerebral palsy: A randomized controlled trial. Research in Developmental Disabilities, 33(1), 181–188. http://dx.doi.org/10.1016/j.ridd.2011.08.026
*Scholtes, V. A., Becher, J. G., Comuth, A., Dekkers, H., Van Dijk, L., & Dallmeijer, A. J. (2010). Effectiveness of functional progressive resistance exercise strengthtraining on muscle strength and mobility in children with cerebral palsy: A randomized controlled trial. Developmental Medicine & Child Neurology, 52(6),e107–e113. http://dx.doi.org/10.1111/j.1469-8749.2009.03604.x
Scianni, A., Butler, J. M., Ada, L., & Teixeira-Salmela, L. F. (2009). Muscle strengthening is not effective in children and adolescents with cerebral palsy: A systematicreview. Australian Journal of Physiotherapy, 55(2), 81–87.
Sirin, S. R. (2005). Socioeconomic status and academic achievement: A meta-analytic review of research. Review of Educational Research, 75(3), 417–453.Sullivan, E., Barnes, D., Linton, J. L., Calmes, J., Damiano, D., Oeffinger, D., et al. (2007). Relationships among functional outcome measures used for assessing
children with ambulatory CP. Developmental Medicine and Child Neurology, 49(5), 338–344.*Unger, M., Faure, M., & Frieg, A. (2006). Strength training in adolescent learners with cerebral palsy: A randomized controlled trial. Clinical Rehabilitation, 20(6),
469–477.*Unnithan, V. B., Katsimanis, G., Evangelinou, C., Kosmas, C., Kandrali, I., & Kellis, E. (2007). Effect of strength and aerobic training in children with cerebral palsy.
Medical Science Sports Exercise, 39(11), 1902–1909.*van der Linden, M. L., Hazlewood, M. E., Aitchison, A. M., Hillman, S. J., & Robb, J. E. (2003). Electrical stimulation of gluteus maximus in children with cerebral
palsy: Effects on gait characteristics and muscle strength. Developmental Medicine & Child Neurology, 45(6), 385–390.Verhagen, A. P., de Vet, H. C., de Bie, R. A., Kessels, A. G., Boers, M., Bouter, L. M., et al. (1998). The Delphi list: A criteria list for quality assessment of randomized
clinical trials for conducting systematic reviews developed by Delphi consensus. Journal of Clinical Epidemiology, 51, 1235–1241.Verschuren, O., Ketelaar, M., Gorter, J. W., Helders, P. J. M., Uiterwaal, S. P. M., & Takken, T. (2007). Exercise training program in children and adolescents with
cerebral palsy: A randomized controlled trial. Archives of Pediatrics & Adolescent Medicine, 161(11), 1075–1081.Verschuren, O., Ada, L., Maltais, D. B., Gorter, J. W., Scianni, A., & Ketelaar, M. (2011). Muscle strengthening in children and adolescents with spastic cerebral palsy:
Considerations for future resistance training protocols. Physical Therapy, 81(7), 1130–1139. http://dx.doi.org/10.2522/ptj.20100356Wiley, M. E., & Damiano, D. L. (1998). Lower-extremity strength profiles in spastic cerebral palsy. Developmental Medicine & Child Neurology, 40(2), 100–107.World Health Organization. (2008). The International Classification of Functioning, Disability and Health (ICF). Geneva: World Health Organization.