Effectiveness of functional progressive resistance exercise strength training on muscle strength and mobility in children with cerebral palsy: a randomized controlled trial

Authors


Dr Annet J Dallmeijer, at VU University Medical Center, Department of Rehabilitation Medicine, PO Box 7057, 1007 MB Amsterdam, the Netherlands. E-mail: a.dallmeijer@vumc.nl

Abstract

Aim  To evaluate the effectiveness of functional progressive resistance exercise (PRE) strength training on muscle strength and mobility in children with cerebral palsy (CP).

Method  Fifty-one children with spastic uni- and bilateral CP; (29 males, 22 females; mean age 10y 5mo, SD 1y 10mo, range 6y 0mo–13y 10mo; Gross Motor Function Classification System levels I–III) were randomized to the intervention group (n=26) or the control group (n=25, receiving usual care). The intervention group trained for 12 weeks, three times a week, on a five-exercise circuit, which included a leg-press and functional exercises. The training load progressively increased based on the child’s maximum level of strength, determined by the eight-repetition maximum. Muscle strength (measured with hand-held dynamometry and a six-repetition maximum leg-press test), mobility (measured with the Gross Motor Function Measure, two functional tests, and a mobility questionnaire), and spasticity (measured by the appearance of a catch) were evaluated before, during, directly after, and 6 weeks after the end of training by two blinded research assistants.

Results  Directly after training, there was a statistically significant effect (p<0.05) on muscle strength (knee extensors +12% [0.56N/kg; 95% confidence interval {CI} 0.13–0.99]; hip abductors +11% [0.27N/kg; 95% CI 0.00–0.54]; total +8% [1.30N/kg; 95% CI 0.56–2.54]; six-repetition maximum +14% [14%; 95% CI 1.99–26.35]), but not on mobility or spasticity. A detraining effect was seen after 6 weeks.

Interpretation  Twelve weeks of functional PRE strength training increases muscle strength up to 14%. This strength gain did not lead to improved mobility.

List of Abbreviations
LSU

Lateral Step-Up test

PRE

Progressive resistance exercise

STS

Sit-To-Stand test

Cerebral palsy (CP) describes a group of disorders in the development of movement and posture, causing activity limitation, that are attributed to non-progressive disturbances that occurred in the developing brain.1 It is the most common cause of movement disability in childhood. Children with CP may experience a variety of impaired muscle functions, such as spasticity, muscle weakness, and loss of selective motor control. Although all impaired muscle functions limit the performance of daily life activities and participation in a child with CP, a recent study has shown that muscle weakness showed a stronger association with mobility limitations in children with CP than spasticity.2 Strength training for these children is, therefore, expected to improve or maintain their mobility.

Until recently, strength training in children with CP was discouraged as it was assumed that it would increase spasticity. However, this was not supported by the results of earlier uncontrolled studies which showed that strength training can improve lower-limb muscle strength in children with CP without increasing spasticity.3,4 Although studies have shown there to be sufficient evidence for its effectiveness on muscle strength, these effects are probably overestimated because of the low methodological quality of the studies.5,6 The few uncontrolled studies that have evaluated the effectiveness of strength training on mobility outcomes in children with CP have reported limited effectiveness.3,7 Three recently published randomized clinical trials8–10 evaluated both muscle strength and mobility in children with CP, but conflicting results were reported.11,12 One of the explanations for these conflicting results could be the high variability in training characteristics, such as type of training, training intensity, and training period. To be successful, strength training must be individualized, and involve a progressive increase in intensity, thereby stimulating strength gains that are greater than those associated with normal growth and development (i.e. ‘overload’). This is known as progressive resistance exercise (PRE). According to the US National Strength and Conditioning Association, children should be trained at an 8- to 15-repetition maximum, which is the number of repetitions that can be completed before fatigue.13 To stimulate strength progression, the amount of resistance should be increased as strength increases. Most previous strength training studies did not present full details of their training programme, or used insufficient intensities. Adequate interpretation of study results is not possible when key principles of the PRE are not followed. Further, earlier reported training programmes lasted 5 to 8 weeks but some have argued that a 6-week training programme may not be long enough to induce functional adaptations.14 Therefore, the purpose of this study was to evaluate the effectiveness of a sufficiently long and intensive 12-week functional PRE strength training programme, by applying current guidelines.13 We hypothesize that this training would improve muscle strength and, subsequently, lead to improved mobility in children with CP, without increasing muscle spasticity.

Method

Participants

Ambulatory children with spastic unilateral or bilateral CP1 were pre-selected by a paediatric physiatrist. The inclusion criteria were (1) age between 6 and 13 years, (2) able to accept and follow verbal instructions, (3) able to walk independently indoors, with or without walking aids (Gross Motor Function Classification System [GMFCS] levels I–III), and (4) able to participate in a group training programme. The exclusion criteria were unstable seizures, any treatment for spasticity or surgical procedures from 3 months (for botulinum toxin type A injections) to 6 months (for surgery) before the study (or planned during the study period), any change in medication expected during the study period, and suffering from any other diseases that interfered with physical activity.

Study design

This study, performed between April 2007 and May 2008, had a single-blind randomized controlled design. It was conducted at three special schools for children with physical disabilities in the Netherlands (Amsterdam, Wijk aan Zee, Valkenburg). The study protocol was approved by the medical ethics committee of the VU University Medical Center in Amsterdam. Full written informed consent was obtained from all parents and 12-year-old children before enrolment. Randomization was performed for each school separately by one independent physiatrist. The children were pre-stratified according to three stratification variables: sex, GMFCS level (I, II–III), age (youngest: 6–9y; oldest 10–13y), and subsequently randomized to one of two groups using sealed envelopes. Parents, therapists, and schoolteachers were notified about the randomization group by letter. Two independent research assistants performed all assessments and data entry. They were blinded for treatment allocation. Assessments were at baseline (T0), during training (T1; a subset of outcomes), directly after training (T2), and at the 6-week follow-up (T3). The measurement points of main interest were T2 and T3.

Intervention

The control group continued their conventional physical therapy programme. Children in the control group received one to three sessions a week. The training group followed a 12-week progressive functional strength-training programme for the lower extremities. This replaced their conventional physical-therapy programme. Training was given three times a week for 45 to 60 minutes at school in small groups (four or five children). Each training session consisted of one exercise on a child-adapted leg-press (Enraf Nonius, Delft, the Netherlands) and three functional exercises (sit-to-stand, lateral step-up, half knee-rise), loaded with a weighted vest. During the training, intensity progressively increased, based on repeated estimation of the eight-repetition maximum. A detailed description and characteristics of the PRE training programme can be found elsewhere.15

Outcome measures

The primary outcome, ‘mobility’, was measured with the 66-item version of the Gross Motor Function Measure (GMFM-66), the Sit-To-Stand (STS) test, and the Lateral Step-Up (LSU) test. The GMFM-66 is a standardized observational instrument, based on interval scaling.16 It reports on the child’s level of capacity, and has been internationally validated for evaluating change in the gross motor activities of children with CP. The STS test assesses the number of repetitions (full movement from standing up to sitting down) that the child can perform in 30 seconds on a child-sized chair with a height-adaptable seat (no backrest, no armrest). The LSU test assesses the number of repetitions (full movement from stepping up to stepping down) that the child can perform in 30 seconds on a 21cm (GMFM I–II) or 11cm (GMFCS III) step. Subjectively reported mobility was assessed with the 28-item version of the Dutch mobility questionnaire (MobQues-28), which measures caregiver-reported mobility limitations. It was found to be valid and reliable in a recent study in a Dutch population.17

The secondary outcome, ‘muscle strength’, was measured with (1) a six-repetition maximum test on a leg-press and (2) isometric strength tests of the hip flexor/abductor, knee flexor/extensor, and ankle plantar flexor muscles of the most affected leg, using a hand-held dynamometer (MicroFet; Biometrics, Almere, the Netherlands). At T1 only a subset of these outcomes were tested: the STS, LSU, and six-repetition maximum.

The control outcome measure was spasticity, quantified on a scale of 0 to 5, calculated as the sum of a catch in response to a fast muscle stretch (0, no catch; 1, catch) in five leg muscles (adductor, rectus femoris, hamstrings, soleus, gastrocnemius). Scholtes et al.15 give detailed descriptions of all outcome measures and measurement protocols.

Muscle soreness was scored weekly on a Likert scale (no; mild; moderate; severe; extremely severe).

Statistical analysis

Baseline characteristics were analysed with the Student’s t-test for continuous normally distributed data, a χ2 test for dichotomous data, and a Mann–Whitney U test for ordinal data. A two-tailed value of p<0.05 was considered to be statistically significant.

To investigate the effect of the strength training, we used generalized estimating equations for longitudinal analysis of continuous outcomes and log linear generalized estimating equation analysis for ordinal outcomes. This method takes into account the dependency of within-patient observations and the fact that not all children are assessed at each measurement point (missing data). All analyses were performed in SPSS, version 15 (SPSS Inc., Chicago, IL, USA). In the primary generalized estimating equation model, the outcome variable (e.g. GMFM-66, LSU, STS, hand-held dynamometer, six-repetition maximum, MobQues-28, or spasticity) was analysed as a dependent variable, with treatment allocation (1, training; 0, control) as the key independent variable adjusted for time. To evaluate whether the two groups differed in change over time, the interaction term (group*time) was added as an independent variable. We had four measurement points, so time was included as a dummy variable (reference was baseline), and three interaction terms were analysed (group 1*T1; group 1*T2; group 1*T3). All models were corrected for the school attended. We also analysed whether sex (1, male; 0, female), age, body mass index, limb distribution (1, unilateral; 0, bilateral), GMFCS level (dummies for levels I–III; reference, I), puberty onset (Tanner stages, dummies for early [stages 1–2], middle [stages 3–4], or late adolescence [stage 5]; reference, early), and problem behaviour (Strength and Difficulties Questionnaire,18 dummies for normal [scores 0–10], borderline [scores 11–13], or abnormal behaviour [scores 14–40]; reference, normal) were confounders of the training effect. A two-tailed value of p<0.05 was considered statistically significant.

Study power

This study had a power of 87% in detecting a difference in improvement of 5% on the GMFM-66 between the intervention and the control group, with an alpha of 0.05.

Results

Sixty-one patients were asked to participate; four turned out to be ineligible. Of the remaining 57 eligible patients, six did not give consent. The remaining 51 children were randomized. Of these, one dropped out before T0 (GMFCS level III, female, age 13y 1m, training) owing to a hip injury that made pretesting and training impossible, and one was lost to follow-up at T1 (GMFCS level II, female, age 12y 1mo, training) owing to an unexpected long-term stay abroad.

Analyses were performed for 49 participants (Fig. 1). Personal characteristics of the participants are presented in Table I. There was no statistically significant difference between the groups for personal characterstics.

Figure 1.

 Patient flow chart.

Table I.   Personal characteristics of participants (n =24)
Personal characteristicsIntervention group (n=24)Control group (n=25)p value95% CI
  1. Data are number of patients, unless otherwise indicated. aDerived from the Tanner stages of sexual maturation. bDerived from the Strength and Difficulties Questionnaire. CI, confidence interval.

Sex (females/males)8/1612/130.30 
Age (y:mo), mean (SD)10:4 (1:10)10:3 (2:3)0.96−1.2 to 1.2
Height (cm), mean (SD)141.4 (13.6)139.1 (13.1)0.55−9.9 to 5.4
Body weight (kg), mean (SD)38.8 (11.2)34.49 (10.4)0.1710.5 to 1.9
Limb distribution (unilateral/bilateral)7/1710/150.43 
Gross Motor Function Classification System Level (I/II/III)13/8/312/9/40.64 
Orthoses (yes/no)12/1214/110.67 
Puberty onseta (early/middle/late adolescence)16/7/016/7/20.57 
Problem behaviourb (normal/borderline/abnormal)10/5/76/9/70.46 

Training compliance and intensity

Of the 36 scheduled training sessions, three to six sessions were cancelled because of school-related activities. After correction for these, the mean compliance was 92.3% (range 71–100%). A mean of 32 training sessions (range 30–33) were attended. Reasons for absence were illness (41.4%), medical appointment (8.6%), vacation (6.9%), or other/unknown (43.1%).

Every 2 weeks, starting from week four, eight-repetition maximum tests were performed for leg-press and loaded STS exercises. Based on these eight-repetition maximum test values the training load progressed. During the last 8 weeks of training, the mean eight-repetition maximum strength increased from 116 to 149% on the leg-press, and from 28 to 41% on the loaded STS.

Each week, one to six children reported mild to moderate muscle soreness.

Outcome measures

Table II presents the mean values at all measurement points for all outcome measures. Some children could not be motivated to complete all tests at each occasion; therefore, a variable number of observations is shown.

Table II.   Estimated effect of intervention on mobility, muscle strength and spasticity according to generalized estimating equation analyses
 TValues at measurement pointsEffect of intervention
InterventionControlRegression coefficient (95% CI)p value
nMean (SD)nMean (SD)
  1. aStatistically significant baseline difference between the two groups. T, measurement point; T0, baseline; T1, 6 weeks; T2, 12 weeks (end of training); T3, 18 weeks (follow-up); n, number of individuals; GMFM, Gross Motor Function Measure; STS, Sit-to-Stand test; LSU, Lateral Step-Up test; reps, repetitions; MobQues-28, mobility questionnaire; CI, confidence interval.

Primary outcomes
 GMFM-66 (0–100)T02476.1 (12.8)2371.8 (12.5)  
T22476.1 (11.8)2473.1 (12.4)−0.56 (−2.11 to 0.99)0.48
T32476.6 (13.0)2472.7 (12.8)0.26 (−1.23 to 1.76)0.73
 STS test (reps)T02412.9 (2.8)2310.8 (3.0)a  
T12413.3 (3.2)2412.1 (4.2)−0.47 (−2.28 to 1.33)0.61
T22413.6 (3.0)2412.7 (4.3)−0.75 (−2.21 to 0.72)0.32
T32414.3 (2.9)2312.7 (4.6)0.03 (−1.43 to 1.49)0.97
 LSU test (reps)T02315.6 (4.0)2313.3 (5.4) 0.95
T12415.2 (3.9)2413.7 (4.3)−0.05 (−1.87 to 1.77)0.63
T22417.0 (5.1)2315.4 (4.3)0.48 (−1.45 to 2.40)0.9
T32417.5 (4.8)2415.8 (6.6)0.13 (−1.84 to 2.10) 
Secondary outcomes
 Isometric muscle strength (N/kg)
  Knee extensorsT0244.78 (1.12)234.36 (1.05)  
T2245.39 (1.10)244.48 (1.21)0.56 (0.13 to 0.99)0.01
T3235.20 (1.04)244.46 (1.20)0.35 (−0.16 to 0.85)0.16
  Knee flexorsT0242.73 (0.79)232.25 (0.96)  
T2242.76 (0.75)242.27 (1.02)0.05 (−0.25 to 0.36)0.71
T3242.67 (0.86)242.33 (0.90)−0.10 (−0.43 to 0.24)0.58
  Hip flexorsT0243.96 (0.75)233.76 (0.99)  
T2244.43 (0.99)244.12 (0.99)0.16 (−0.22 to 0.55)0.41
T3244.46 (0.90)244.43 (0.86)−0.12 (−0.50 to 0.27)0.55
  Hip abductorsT0242.66 (0.76)222.41 (0.74)  
T2242.78 (0.85)242.28 (0.70)0.27 (0.00 to 0.54)0.05
T3242.90 (0.99)242.45 (0.94)0.23 (−0.10 to 0.56)0.17
  Ankle plantar flexorsT0243.90 (1.43)233.11 (0.87)a  
T2244.53 (2.16)233.54 (0.94)0.23 (−0.47 to 0.93)0.51
T3245.01 (2.26)244.38 (1.49)−0.16 (−1.01 to 0.69)0.72
  TotalT02418.04 (3.52)2215.94 (3.57)a  
T22419.88 (4.13)2316.65 (4.11)1.30 (0.56 to 2.54)0.04
T32320.39 (4.49)2417.80 (4.01)0.40 (−1.02 to 1.83)0.58
Six-repetition maximum on leg-press (% body weight)T023112.78 (21.28)2393.76 (20.18)a  
T124119.38 (26.61)23100.80 (23.72)1.97 (−8.45 to 12.41)0.71
T223135.63 (31.87)22102.88 (26.76)14.17 (1.99 to 26.35)0.02
T324129.90 (32.15)23111.99 (26.17)3.42 (−8.62 to 15.46)0.58
MobQues-28 (0–100)T02268.42 (20.93)2264.77 (26.26)  
T22367.51 (24.58)2366.43 (25.93)−0.46 (−6.00 to 5.07)0.87
T32170.03 (23.49)1767.49 (20.33)−2.22 (−7.88 to 3.43)0.44
Control outcome  Median (SD) Median (SD)  
  Spasticity (0–5)T0241.00 (1.32)232.00 (1.32)  
T1232.00 (1.41)222.00 (1.39)0.02 (−0.99 to 1.02)0.97
T2232.00 (1.11)241.50 (1.10)0.46 (−0.34 to 1.26)0.26
T3221.00 (0.87)241.00 (0.87)−0.22 (−0.92 to 0.49)0.55

At baseline, the intervention group achieved statistically significantly more STS repetitions, had statistically significant greater ankle plantarflexion and total isometric muscle strength, and had statistically significant greater six-repetition maximum leg-press strength than the control group.

The estimated intervention effect, corrected for baseline differences, is presented in Table II. The intervention had no statistically significant effect on the primary outcome mobility, measured with the GMFM-66, STS, LSU, and MobQues-28 (Fig. 2a–d). There was a statistically significant effect on the secondary outcome for the total isometric (1.30N/kg, 8%), knee extensor (0.56N/kg, 12%), and hip abductor (0.27N/kg, 11%) muscle strength, and the six-repetition maximum strength (14%; Fig. 2e,f). There was no effect on spasticity.

Figure 2.

 Estimated marginal means and standard errors of (a) Gross Motor Function Measure, 66 items (GMFM-66), (b) Sit-to-Stand, (c) Lateral Step-Up, (d) Dutch mobility questionnaire (MobQues-28), (e) hand-held dynamometer (HHD)– total, and (f) six-repetition maximum (6RM) scores at measurement points 0 (baseline), 6, 12, and 18 weeks for the intervention and the control groups. aStatistically significant baseline difference. bStatistically significant difference in change between intervention and control groups, corrected for baseline differences (p<0.05).

Confounders

None of the investigated variables confounded on the training effect. Table II and Figure 2 present the results of the final regression models.

Discussion

This study hypothesized that a 12-week functional PRE strength training programme, in which appropriate guidelines for increasing muscle strength are applied,13 improves muscle strength in children with CP, and subsequently leads to improved mobility, without any adverse effect on muscle spasticity.

This 12-week functional PRE strength training resulted in statistically significant larger improvements (8%) in total hand-held isometric muscle strength in the training group than in the control group. Isolated significant improvements were found in the knee extensors (12%) and hip abductors (11%), but not in the knee, hip, and ankle plantar flexors. This seems to underline the importance of specificity in strength training; i.e. the exercise must be specific to the working of the muscle. Because the circuit training specifically targeted strengthening knee and hip extensors (with the leg-press, STS, forward step-up, and half-knee-rise exercises) and hip abductors (with the lateral step-up exercise), improvements in these target muscles were, therefore, to be expected. Unfortunately, isometric hip extension strength was not assessed, owing to this assessment’s low reliability, as found in a pre-pilot study. Two other randomized studies have evaluated isometric muscle strength, and found similar results, either in total muscle strength9 or isolated (targeted) muscle strength.8 The training also resulted in a statistically significant improvement (14%) in six-repetition maximum leg-press strength, compared with the control group. To our knowledge, no other study has assessed this outcome measure in a similar patient group, so it is impossible to compare these results with others. However, one randomized study10 did use the one-repetition maximum STS as an outcome. When adjusted for body weight, the strength increase in favour of the intervention group was comparable (approximately 15%).

The results of this study show that training with a sufficient intensity, e.g. with a training load at which no more than 8 to 12 repetitions can be completed before muscular fatigue, results in increases in muscle strength. With this result we hope to fill the gap that was recently addressed in a literature discussion19 on the effectiveness of strength training on muscle strength, and stressing the need for new randomized controlled trials that apply appropriate guidelines. We feel that our study adds to this discussion and the current evidence20 that PRE strength training is effective in increasing leg muscle strength in this group of patients.

There was a detraining effect, resulting in loss of muscle strength at 6 weeks after the training ended, which was expected, and has also been observed in healthy children.21 Surprisingly, this effect was not observed in the few comparable (mostly uncontrolled) studies in children with CP3,9,12 with a follow-up assessment. Based on the findings of our randomized controlled trial, we recommend that strength training should be included in a regular exercise routine to maintain increased strength levels.22

Unexpectedly, the 8 to 14% strength improvements were not accompanied by mobility improvements, according to the GMFM-66, the functional strength tests, and the mobility questionnaire. Two out of three comparable randomized controlled trials showed that their strength training resulted in small, but significant, functional improvements on the GMFM-66 in favour of the training group.8–10 So, despite the current popularity of strength training, the evidence from current randomized controlled trials is inconclusive on the effectiveness of strength training to improve mobility in children with CP.

Possible explanations for the lack of mobility improvement in our study might be that the 8 to 14% improvements were too small to improve mobility, or that the number of individual muscles gaining strength was too limited. Therefore, future studies should focus on optimizing the training programme to reach maximum strength gains in children with CP. Besides this, we need to gain a better understanding of the longitudinal relation between muscle strength and mobility. Daily activities require only a certain amount (e.g. lowest threshold) of muscle strength. Increases above these lowest threshold values may be accompanied by increases in mobility,23 but there might also be a point (e.g. highest threshold) at which further strength increases provide no additional advantage in mobility improvements.24 In that case, strength training would not be the appropriate choice of treatment when the aim is to improve mobility. Here, other components, such as balance and coordination, might influence mobility improvement to a greater degree than muscle strength alone. This should also be a subject for future research. In addition, the lack of effectiveness on mobility improvement found in this study might also be explained by the lack of variation in task contexts and individually tailored exercises. It is said that functional improvement is unlikely to occur unless the task to be learned is practised under various different contexts.25 Even though our training contained functional exercises that resembled daily mobility activities (e.g. rising from a chair), the actual context of these exercises (i.e. chair height) remained unchanged throughout the training period. The lack of effectiveness on mobility improvement might also be explained by the non-individual specificity of the exercises. Selection of individually tailored mobility exercises with a more goal-oriented approach might improve the effectiveness of strength training on mobility outcomes.26

There was no change in spasticity (in either group) during, directly after, or 6 weeks after the training, confirming the results of similar studies.3 This supports the current belief that strength training for patients with spasticity is not contra-indicated. Children were informed that they could experience a mild to moderate degree of muscle soreness the day after the training. Muscle soreness was occasionally reported. This muscle soreness, rated as mild or moderate, was transient and never interfered with their training or daily life. Therefore, this cannot be interpreted as a contra-indication. Moreover, it might confirm that training intensity was sufficient to generate muscle overload.

This study has some limitations. First, the groups differed slightly statistically, although non-significantly, in personal characteristics, but none of these were found to be confounders. Secondly, the intervention group performed better at baseline on all outcome measures, reaching statistical significance in the number of STS repetitions, isometric ankle plantarflexion, and total muscle and six-repetition maximum strength. Therefore, outcome measures were controlled for baseline differences. Another limitation was the unknown frequency and content of the usual physiotherapy of the control group. However, the therapists of the control group did not know the PRE training principles.

In conclusion, 12 weeks of functional PRE strength training is effective in increasing isometric muscle strength of the knee extensors and hip abductor by 11 to 12%, and increasing six-repetition maximum leg-press strength by 14%. However, this does not result in an increase in mobility. Consequently, functional PRE strength training is indicated for a child with CP when the aim is to improve leg muscle strength. It could also be included in a more extensive treatment regime, addressing several elements of fitness including muscle strength, or used as a target treatment specifically anticipating temporary muscle weakness, such as before or after botulinum-toxin-A or surgical treatment.

What this paper adds

  • • Insight into the true effectiveness of strength training in children with CP, using a blinded randomized controlled study.
  • • It evaluates the effectiveness of a strength training program in which appropriate guidelines for strength improvements are applied, based on the key elements of PRE.
  • • It shows that PRE based strength training is effective on muscle strength improvement, but not on mobility improvement.

Acknowledgements

This study was supported financially by a grant from the Johanna Kinder-Fonds (2005/0123-357), the Adriaanstichting, and the Phelps Stichting (2006016).We are indebted to all the children and their parents for their participation. We are also grateful for the contributions of members of the Popeye study group: Linda Smallenbroek, Maartje Hensen (Mytyl and Tyltylschool and Rehabilitation Center Amsterdam), Eugene Rameckers, Yvonne Jansen-Potten, and Lucianne Speth (Rehabilitation Foundation Limburg, location Franciscusoord, Valkenburg), Els Tempelaars, Nienke Bergsma, and Han Houdijk (Rehabilitation Center Heliomare, Wijk aan Zee). Thanks are due to Enraf Nonius and Noppe Orthopedie for their support in providing training equipment.

Ancillary