Share this article on:

Changes in Endurance and Walking Ability Through Functional Physical Training in Children with Cerebral Palsy

Gorter, Hetty PT; Holty, Lian PT; Rameckers, Eugène E.A. PT, MRes; Elvers, Hans J.W.H. RI, MSc; Oostendorp, Rob A.B. Prof Dr

Pediatric Physical Therapy: April 2009 - Volume 21 - Issue 1 - p 31-37
doi: 10.1097/PEP.0b013e318196f563
Research Report

Purpose: To investigate the feasibility and effect of a functional physical training program on aerobic endurance and walking ability of children with cerebral palsy.

Methods: Thirteen children (8–13 years, Gross Motor Function Classification System level I or II, with normal intelligence or mild retardation) participated in this study. A functional physical training program addressing aerobic endurance, walking distance, walking velocity, and ambulation, consisted of a circuit with 4 stations and lasted 30 minutes twice weekly for 9 weeks. The Bruce, 6-minute-run test, Timed Up and Down Stairs Test, and Ambulation Questionnaire were administered 2 weeks before the start, immediately after, and 11 weeks after the intervention.

Results: Significant improvement in aerobic endurance, walking distance, and ambulation were observed immediately after the intervention. Maximum treadmill time had improved significantly at 11 weeks.

Conclusion: A functional physical training improves the aerobic endurance and the functional walking ability of children with cerebral palsy.

The authors demonstrate that a functional physical training program can be used to improve aerobic endurance and functional walking ability of children with cerebral palsy.

Roessingh Rehabilitation Centre Enschede (H.G., L.H.), Enschede, The Netherlands; Stichting Revalidatie Limburg (E.E.A.R.), Franciscusoord Valkenburg, The Netherlands; Dutch Institute of Allied Health Care (J.W.H.E., R.A.B.O.), Amersfoort, The Netherlands; Methodological Health-skilled Institute (J.W.H.E.), Beuningen, The Netherlands; Radboud University Nijmegen Medical Centre, Department of Public Health and Research (J.W.H.E, R.A.B.O.); Centre for Allied Health Sciences, Department of Quality of Care Research (J.W.H.E, R.A.B.O.), Nijmegen, The Netherlands

Address correspondence to: Hetty Gorter, PT, Roessinghsbleekweg 33, 7522AH Enschede, The Netherlands. E-mail:

Back to Top | Article Outline


Children with cerebral palsy (CP) have a decreased level of daily physical activity in comparison with their healthy peers.1 Despite physical training and therapy, they do not use their physical reserve sufficiently during the day to achieve optimal levels of daily physical activity. They have to train almost 2.5 hours a day to reach the same level of daily physical activity as their healthy peers.

In addition, individuals with CP need to retain a higher level of physical fitness than people without impairments to resist the natural process of decline because of both age and the restricted endurance related to CP.2 Reduced endurance is the main factor in the decline in walking ability for individuals with CP.3

van den Berg-Emons4 demonstrated that peak aerobic power does not increase without a specific training program. In a study by van den Berg-Emons,1 physical training for children with CP involved both a long training period and training time as well as a high frequency. This demands a lot of motivation and time of the children to maintain the program. A combination of strength training and training focused on aerobic and sometimes anaerobic endurance was given in the above-mentioned study1 and in other studies.2,4 It is not clear which effect can be reached through aerobic training only.

The maximum aerobic capacity, VO2max, is an internationally acknowledged measure of the fitness of the cardiorespiratory system.5 Baquet et al6 demonstrated that children can increase their VO2max through performing a short intermittent training program twice weekly for 30-minutes each session across 7 weeks. The assumption is that it is the same result that would hold for children with CP.

The main problems, which parents often report during consultative meetings regarding their children, are the children’s inability to walk for longer periods, sufficient for family trips or shopping, and their slow pace. These children tend to have busy school programs (which include swimming, gymnastics, and/or physiotherapy), leaving them with limited time for extra training during school hours. Moreover, as the rehabilitation and/or education center serves a large region, many children’s travel times are too long for training after school hours, so this is not an option.

This has led our Rehabilitation center to launch a modified form of physical training. Aimed primarily at improving endurance, this 30-minute training is given twice a week during school hours, based on the study of physical training in healthy preadolescents of Baquet.6

The aim of this study was to investigate differences in aerobic endurance and walking capacity before and after participating in a functional physical training program, fitting in the total school and/or therapy program.

Back to Top | Article Outline



The study is a test–retest repeated measures design. Baseline tests were carried out 2 weeks before training (T0), the actual training took 9 weeks (T1), postintervention tests were carried out 11 weeks after the training was completed (T2) (Table 1).



Back to Top | Article Outline


The participants consisted of students at the school for disabled affiliated with the Rehabilitation Centre during the school year 2004–2005. The children were included in the study if they had a diagnosis of CP, level I or II according to the Gross Motor Function Classification System (GMFCS),7 aged between 8 and 13 years at the start of this study and with normal intelligence or mild retardation. The limit of mental retardation was set at IQ of 60, as determined by Wechsler.8 All children of the school population who met inclusion criteria were invited to the study.

Children were excluded if they had had orthopedic surgery or botox injections within 3 months before this study or had genetic or neurological abnormalities in addition to CP, or if they had serious behavior problems. Children were also excluded if they had been involved in condition training less than 6 months before this study.

The age range from 8 to13 years was chosen because of children below the age of 8 are generally too small for the training apparatus that is used. Children above the age of 13 were excluded because of the differences that appear between boys and girls in muscle power and peak aerobic endurance as a result of hormonal influences during adolescence.9

Approval of the Medical Ethics Committee was given because the intervention (training) is already part of the rehabilitation program and no invasive therapy took place.

All parents and children signed an Informed Consent.

Back to Top | Article Outline


The children trained for half an hour twice weekly for 9 weeks. The training consisted of a form of circuit training with 4 stations (Fig. 1). Each workstation consisted of repetitive exercises. Because the training was focused on aerobic endurance, the schedule was 3 minutes of effort and 3 minutes of rest.9 The training took place once a week in the gym and once a week in the fitness room. A group was composed of maximum of 8 children.

Fig. 1.

Fig. 1.

The stations in the training session included jumping on a trampoline, climb/bend/jump/run combinations, basketball circuit, and a run circuit. The other training session consisted of treadmill walking horizontally, treadmill walking on an incline, steps and exercises on a bicycle. For this circuit training, the children were divided into fixed pairs with one instructor for each pair. Every team member had a personal card with information about the order, intensity of training, and where he or she should be trained (Fig. 2). The outcomes of the Bruce test (HR6) aided in defining the intensity of training in the fitness room.

Fig. 2.

Fig. 2.

Every other week the children wore heart rate meters, to determine whether they trained with an intensity of 60% to 70% of their maximum heart rate. The intensity of exercises increased with the same heart rate. In the gym, the children were required to talk during the exercises. By choice of the exercises, the children could see if they achieved improvement.

The group of instructors consisted of 1 physical therapist, 1 gym teacher, and 1 or 2 assistants. Both physical therapists and gym teachers have experience working with children and were well instructed to work with the training protocol.

During the period of this study, the children followed their normal therapy program including sport.

For practical reasons, a 9-week period of intervention was chosen for the study. This period fits in between two school holidays, so the children could train uninterrupted. The training time of 30 minutes is in conformity with the usual therapy time in the centre. The two sessions of training are chosen in consult with the school and take into account both school programs and the availability of the fitness room. To compensate absence there was the possibility to train at another session.

Back to Top | Article Outline


The Bruce treadmill test10 was used to determine the maximum aerobic capacity and the 6-minute-run test11 was used to determine walking distance and walking velocity. To measure functional mobility (ambulation), Timed Up and Down Stairs test (TUDS)12 and the Ambulation Questionnaire13 (MoVra) were used. The MoVra focuses on the home situation of the child.

The child wore his usual walking aids if needed, such as orthoses, during the Bruce treadmill test, 6-minute-run test, and the TUDS. All children walked on the treadmill during a week of baseline measurement to get used to the treadmill.

Back to Top | Article Outline

Bruce Treadmill Test

The Bruce treadmill test10 is often used for measuring aerobic endurance. In this study, it was decided to use the “half Bruce” protocol because it was considered that children with CP could adapt better to the smaller steps of this protocol. The normative values are not relevant to children with CP. The test was used only to compare the subsequent performances of the child.

HR6 and the Tmax were used as outcome measurements. The measures were not converted into VO2max because an error of 10% to 15% could occur in the calculations.14 Studies have proven the existence of a linear relation between heart rate and oxygen uptake, so the rate can been taken as a proxy of the energy cost in normal children and children with CP.15

Decrease of HR6 is related to a higher oxygen uptake and increase of Tmax shows a higher level of sustained endurance.

Back to Top | Article Outline

Run Test

This test is part of the Motor Performance (Moper) fitness test.16 Jackson et al11 demonstrated that the 6-minute run test is a reliable and construct valid test for estimating walking distance. To avoid weather disturbances, the course was indoors. There are normal values available for children aged 9 to 17 years.16 The normative values were not considered to be relevant because in this study the 6-minute-run test was only used to compare the child’s performance with his previous one.

Back to Top | Article Outline

Timed Up and Down Stairs Test

The TUDS is a quick, low budget, reliable, and valid test to measure the ambulation (functional mobility) of children aged from 8 to 14 years with and without CP.12

The test is suitable as evaluative measuring-instrument. The validity, compared with the Timed Up and Go, the Functional Reach Test, and the Timed One Leg Stance is moderate to high (r s = 0.78, −0.57, −0.77, respectively). A shorter time for going up and downstairs is related to a higher level of ambulation.

Back to Top | Article Outline

MoVra (Version 1.3)

MoVra is a questionnaire for parents and is developed to measure the extent of limitation in ambulation of children aged from 2 to 12 year with CP. The MoVra was developed within the Bolien research project.13 It is a part of an ongoing national study consisting of the scalability, reliability, and validity of the MoVra. Consent was obtained from the makers of the test to use the questionnaire in connection with this study. The questionnaire in the research version (version 1.3) contains 47 questions about activities of daily life, divided into “activities indoors” (such as getting out of bed, walking up and downstairs) and “activities outdoors” (such as walking on uneven ground, running, walking outside for half an hour). Parents are asked the extent of difficulty the child experiences in performing the tasks independently. The normal 9-point scale of version 1.3 was reduced to a 5-point scale for the purpose of this study. To process the measurement, every scale was given a number to facilitate the processing of the measurement.

Development of MoVra is still incomplete. Information regarding the reliability, validity, and responsiveness was not available at the time of this study. In the future, there will be a modification toward a 5-point scale. In this study, the scores of the different questions were counted together. The difference in scores per child between T0 and T1 was examined. A higher score is related to a lower level of ambulation. The total questionnaire (MoVra-overall) as well the subdivisions such as activities indoors (MoVra-indoor) and activities outdoors (MoVra-outdoor) were considered. From the questions of activities outside, those questions that referred to the specific training activities (MoVra-training) were considered separately.

The parents were only asked to fill in MoVra at the beginning and straight after the period of the training to avoid overburdening them. All parents returned the completed form.

Together with the MoVra, the parents were asked questions about other hobbies, sport activities, and finally events concerning the child.

Back to Top | Article Outline

Statistical Analysis

Statistix 8.0 was used to process the data.17 Statistical analysis was carried out using descriptive statistics (mean and standard deviation, median and range), the Shapiro-Wilks (test of normality), the paired t test (Bruce and run test), and the Wilcoxon signed-rank Test (TUDS* and MoVra). The differences between T1 and T0, between T2 and T1 and between T2 and T0 with α = 0.05 were also examined.18,19 Analysis of variance (ANOVA) for repeated measures design was used to check if the difference between the results at the three measurement times was significant.

Back to Top | Article Outline


Study Group

Fourteen children were invited to participate in the study; all parents gave their signed informed consent. This was the maximum number of children that could be considered in the study. The results from one child were excluded from the study because the child underwent a medical treatment soon after he had participated in the intervention.

Thirteen children participated in the study: There were 8 boys and 5 girls. The mean age was 9.9 years (SD = 1.15). The children had normal intelligence (n = 1) to mild mental retardation (n = 12). There were 12 Caucasian children and 1 with Chinese parents. All children had a diagnosis of CP and at GMFCS level 1 (n = 12) or level 2 (n = 1).

Back to Top | Article Outline

Results Measures

In the 9 weeks of the intervention period, 12 children trained 18 times and 1 child 16 times.

All measurement sessions took place by protocol.

Back to Top | Article Outline

Aerobic Endurance (Bruce)

One child could not sustain exercise for 6 minutes at T0. The results are summarized in Tables 2 and 3. ANOVA for repeated measures design Bruce HR6, p = 0.02 revealed a significant difference between the 3 measurement sessions as a result of time. ANOVA for repeated measures design Bruce Tmax, p = 0.00 demonstrated a significant difference between the 3 measurement sessions as a result of time.





The reduction in the HR6 at T1 with regard to T0 was significant (p = 0.01). There was no significant difference in HR6 between T2 and T1 (p = 0.60) and the HR6 at T2 was significantly lower than at T0 (p = 0.01). There was no significant difference between Tmax at T1 and T0 (p = 0.12), at T2 the Tmax significantly increased with regard to T1 (p = 0.00). The Tmax at T2 was significantly higher than at T0 (p = 0.00).

Back to Top | Article Outline

Walking Distance and Walking Velocity (6-Minute-Run Test)

The results are summarized in Tables 2 and 4. ANOVA for repeated measures design for the run test (p = 0.06) revealed the difference between the 3 measurement sessions as a result of time was not significant. But when looking at the results separately, the differences between both T1 and T0 as T2 and T0 were significant. The walking distance at T1 was significantly more than at T0 (p = 0.03); there was no significant difference in distance between T2 and T1 (p = 0.22). The walking distance at T2 was significantly more than at T0 (p = 0.03).



Back to Top | Article Outline

Ambulation (TUDS, MoVra)

The results are summarized in Tables 2, 4, and 5. The results of the TUDS were not normally distributed. The differences of the 3 outcomes of 1 child are not included in the calculation of ANOVA, because these are significantly different from the other outcomes.



ANOVA for repeated measures design TUDS (n = 12; p = 0.00) revealed a significant difference in time between the 3 measurement sessions as a result of time. The time at T1 was significantly decreased with regard toT0 (p = 0.00); the difference in time between T2 and T1 was not significant (p = 0.94). The time at T2 significantly decreased with regard to T0 (p = 0.00). The manner of walking up the stairs (using the banisters or not) did not change.

The difference in scores on the MoVra-overall (p = 0.38), the MoVra-indoor (p = 0.32) as well as by the MoVra-outdoor (p = 0.17) were not significant (Table 5). The score at the MoVra-training was significantly lower (p = 0.03). There were no changes in hobbies or sport activities.

Back to Top | Article Outline


The study was executed according to the protocol. The high level of participation of the children gives an impression of the feasibility of the program within our organization. No problems were found as well in training or during measuring.

The size of the study group was the maximum possible under the inclusion criteria. One child was excluded from the study because he underwent a medical treatment during the follow-up period but he did participate in the intervention. Twelve of the 13 children from the study group were classified at GMFCS level I. They had the most functional abilities within the population of children with CP. Hence, caution is required during interpretation of the data for children with a GMFCS level II.

Looking at the results of this pilot study of changes in aerobic endurance an improvement in VO2max was found and this improvement is attributed to the intervention, based on the controlled findings from the study by van den Berg-Emons.4 The results of the tests immediately after the intervention showed an improvement in aerobic endurance (Bruce HR6), walking distance, and ambulation measured with the TUDS. This improvement was still present 11 weeks after the intervention. A possible explanation is that the children also begin to function at a higher level of activity in their home situations. The most common remark of the parents was that the children could play outside with other children for a longer time and that they could walk for a longer period.

The result of the MoVra, when specifically considering the activities that are trained (MoVra-training), makes this plausible. Those activities become more part of their daily routines, so that the level of daily activities of the children increased. This finding was not seen in the results of MoVra-overall, MoVra-indoor, or MoVra-outdoor. Only the MoVra-training activities demonstrated a significant difference. This means that specific training is required depending on what needs improving.

The maximum treadmill time (Tmax) did not improve immediately after the intervention. It is possible that the children need time to get used to longer periods at higher levels of exertion. This could be an explanation for the fact that after 11 weeks there was a demonstrable improvement in the ability to maintain longer periods of exertion.

Ambulation, measured with the TUDS, improved immediately after the intervention, although the ambulation at home, measured with the MoVra-overall, did not improve. A feasible explanation for this is that parents and child are habituated to a certain measure of assistance and that they could not change this pattern of behavior in this short period.

During this period of intervention, there was no exchange of any information with the parents. In the future, this deserves more attention to bring about use of the benefits of the training during daily routines. Parents were not asked to fill in the MoVra questionnaire at the follow-up at 11 weeks after the intervention. With hindsight we regret that these data are not available.

In this study, 3 measure periods were used. Future studies should include more measurements during the period of intervention to have a better look at the individual growth of increasing aerobic endurance.

The period of follow-up was relatively short, we cannot be sure what this means for the long-term effects. The benefit of the training is not equal in all children because of the standard nature of the training. To realize optimum results for every individual, the training will have to be individualized.

Back to Top | Article Outline


The organization of all training and measurements was successful, but much adjustment between school and center was needed (availability of fitness room, trainers and testers, school program related to moments of training, and so on). Twelve of the 13 children from the study group had mild mental retardation. These children needed much support and stimulation during the training and measurements, but all the children were able to complete the intervention. A sufficient number of instructors were always available. Because the children could compensate for absences most children trained 18 times. The measurements are easy to obtain and the combination of these give sufficient information about the endurance. Only 1 child could not complete the treadmill test for 6 minutes at the T0 measure.

The children enjoyed the training and stimulated each other by working in fixed pairs. They could see their improvement by choice of exercises. The response from parents to fill in the MoVra was 100%.

Back to Top | Article Outline

Clinical Implications

This intervention program is feasible within the school program; however, much adjustment and a sufficient number of coworkers are required. If this training will be continued, this will impact the number of staff needed. All the children enjoyed participating in this training. They stimulated each other and had a lot of fun. The results of this study have led to better communication and adjustment within the organization.

Back to Top | Article Outline


The research question that was asked in this study, “Does endurance training with a mean frequency of 2 × 30 minutes weekly, in the form of circuit training, for 9 weeks, improve the aerobic endurance and the functional walking ability of children, aged between 8 and 13 years with cerebral palsy, who have independent walking function?” could be answered positively. At the end of the period of the study VO2max increased by 9% (estimated from HR6), maximal treadmill time by 23%, walking distance by 7%, and ambulation by 21%. The improvement in ambulation is partially noticeable in the home situation. The training frequency of 2 times weekly for 30 minutes each makes it possible to contain the endurance training within the school program. Because of the short period of training more children could profit from the endurance training within the school facility of the Rehabilitation Centre.

Back to Top | Article Outline


1. van den Berg-Emons RJ, Saris WH, Barbanson de DC, et al. Daily physical activity of schoolchildren with spastic diplegia and of healthy control subjects. J Pediatr.1995;127:578–584.
2. Rimmer JH. Physical fitness levels of persons with CP. Dev Med Child Neurol. 2001;43:208–212.
3. Damiano DL. Strength, endurance and fitness in CP. Quebec Abstr. 2003;8–10.
4. van den Berg-Emons RJ, Baak van MA, Speth L, et al. Physical training of school children with spastic CP: effects on daily activity, fat mass and fitness. Int J Rehab Res. 1998;21:179–194.
5. Macsween A. The reliability and validity of the Åstrand nomogram and linear extrapolation for deriving VO2max from submaximal exercise data. J Sports Med Phys Fitness 2001;41:312–317.
6. Baquet G, Bethoin S, Dupont G, et al. Effects of high intensity intermittent training on peak VO(2) in prepubertal children. Int J Sports Med. 2002;23:439–444.
7. Palisano R, Gocha Marchese V, Westcott S. Development and reliability of a system to classify gross motor function in children with CP. Dev Med Child Neurol. 1997;39:214–223.
8. Wechsler D. Wechsler Intelligence Scale for Children (WISC-III). 3rd ed. London: The Psychological Corporation; 1991.
9. Fox EL, Bowers RW, Foss ML. Fysiologie voor lichamelijke opvoeding, sport en revalidatie. Maarssen: Elsevier Gezondheidszorg; 2004.
10. Cumming GR, Everatt D, Hastman L. Bruce treadmill test in children: normal values in a clinic population. Am J Cardiol. 1978;41:69–75.
11. Jackson AS, Coleman AE. Validation of distance run tests for elementary school children. Res Q. 1976;47:86–94.
12. Zaino CA, Gocha Marchese V, Westcott SL. Timed up and down stairs test: Preliminary reliability and validity of a new measure of functional mobility. Ped Phys Ther. 2004;90–98.
13. Scholtes V. MOVRA version 1.3. In press.
14. Nederlandse Hartstichting, Brochure Maximale inspanning door kinderen; referentiewaarden voor 6–18 jarige meisjes en jongens, 1992.
15. Rose J, Gamble JG, Medeiros J. Energy cost of walking in normal children and in those with cp: comparison of heart rate and oxygen uptake. J Pediatr Orth. 1989;9:276–279.
16. Leyten C, Kemper HCG, Verschuur R. MOPER fitnesstest; Handleiding en Prestatieschalen voor 9 t/m 11 jarigen. Haarlem: Uitgeverij De Vrieseborch;1982.
17. Analytical Software. Statistix 7 User’s Manual. Tallahassee: Analytical Software; 2000. ISBN 1–881789–05–5.
18. Elvers JWH. Inleiding tot de beschrijvende statistiek. Beuningen: Gezondheidskundig Methodologisch Instituut; 2003.
19. Elvers JWH, Oostendorp RAB. Syllaby Scholing en Wetenschap I, II, en III, Nederlands: Paramedisch Instituut; 2005.

* Data from the TUDS were not normally distributed.
Cited Here...


aerobic exercise; cerebral palsy; child; exercise therapy; human movement system; physical endurance; physical therapy; walking

© 2009 Lippincott Williams & Wilkins, Inc.