Content » Vol 52, Issue 5

Original report


Ferdinand von Walden, MD, PhD1*, Emma Hjalmarsson, PT1,2, Mikael Reimeringer3, Ola Kvist, MD4,5, Peter C. Raffalt, PhD6, Eva Pontén, MD, PhD1,7 and Rodrigo Fernandez-Gonzalo, PhD8

From the 1Division of Pediatric Neurology, Department of Women’s and Children’s Health, Karolinska Institutet, 2Allied Health Professionals Function Medical Unit Occupational Therapy and Physiotherapy, Karolinska University Hospital, 3Motoriklab, Department of Women’s and Children’s Health, Karolinska Institutet, 4Department of Pediatric Radiology, Karolinska University Hospital, 5Clinical Pediatrics Unit, Department of Women’s and Children’s Health, Karolinska Institutet, 6Department of Physical Performance, Norwegian School of Sport Sciences, Oslo, Norway, 7Department of Pediatric Orthopedic Surgery, Karolinska University Hospital, and 8Department of Laboratory Medicine, Division of Clinical Physiology, Karolinska Institutet, and Unit of Clinical Physiology, Karolinska University Hospital, Stockholm, Sweden


Introduction: The development of efficient resistance exercise protocols to counteract muscle dysfunction in cerebral palsy is warranted. Whether individuals with cerebral palsy are able to perform iso-inertial resistance (flywheel) exercise in a comparable manner to typically developed subjects has never been experimentally tested.

Design: A comparative, controlled study.

Subjects: Eight young ambulatory adults with cerebral palsy (mean age 19 years; Gross Motor Function Classification System (GMFCS) I–III) and 8 typically developed control subjects (mean age 22 years).

Methods: Subjects performed acute bouts on the weight-stack and flywheel leg-press device, respectively. Range of motion, electromyography, power, work and muscle thickness (ultrasound) data were collected.

Results: Subjects with cerebral palsy were able to produce a greater eccentric/concentric peak power ratio on the flywheel (p < 0.05 vs ratio in weight-stack), however absolute values were lower (p < 0.05 vs weight-stack). Typically developed subjects produced more power per mm of thigh muscle than the cerebral palsy group, independent of leg, device and action.

Discussion: Subjects with cerebral palsy could not elicit the eccentric overload seen in typically developed subjects. Furthermore, peak power production per mm muscle was markedly reduced in both legs in subjects with cerebral palsy. In conclusion, this comparative study of weight-stack and flywheel exercise does not support the implementation of the current iso-inertial protocol for young adults with cerebral palsy.

Key words: cerebral palsy; resistance exercise; eccentric overload; muscle thickness; ultrasound; electromyography.

Accepted Apr 15, 2020; Epub ahead of print Apr 21, 2020

J Rehabil Med 2020; 52: jrm00060

Correspondence address: Ferdinand von Walden, Division of Pediatric Neurology, Department of Women’s and Children’s Health, Karolinska Institutet, Stockholm, Sweden. E-mail:

Lay Abstract

Cerebral palsy is an umbrella term for a movement disorder in children and adults, which is caused by damage to the immature, developing brain. Individuals with cerebral palsy have weaker, thinner and stiffer muscles compared with individuals without cerebral palsy. Poor muscle function makes everyday activities harder. Efficient training regimes to improve muscle function are therefore needed. This study investigated whether an alternative form of resistance exercise (iso-inertial exercise) was feasible in individuals with cerebral palsy. The results show that iso-inertial exercise was difficult for individuals with cerebral palsy to perform and is probably unsuitable for clinical practice for this specific population.


Cerebral palsy (CP) is the most prevalent motor disorder in children (1). Although initiated by a non-progressive brain lesion before the age of 2–3 years (2), clinical manifestations and functional impairments usually worsen over time. Individuals with CP show weaker, thinner and stiffer skeletal muscle than typically developed (TD) children (3), and have decreased muscle volume in affected limbs (4). Poor muscle quality is believed to contribute substantially to the reduced functional capacity typically seen in this population.

Although resistance exercise (RE) programmes can result in increased strength in individuals with CP (5), evidence of transfer effects on functional mobility are lacking (5, 6). This lack of transfer effects may be explained by the concept of training specificity which incorporates the mismatch of selective neuromuscular control, speed of movement, and loading stimulus of currently employed RE models relative to functional activities. Indeed, Damiano and co-workers (7) have shown that individuals with CP have difficulty activating their muscles during concentric (CON; shortening) compared with eccentric (ECC; lengthening) muscle actions to generate force. Despite this finding, published more than 15 years ago, the vast majority of RE protocols currently used in clinical practice still use traditional isotonic, weight-stack (WS) devices that provide a sub-optimal load during the ECC action (8). This is rather surprising, considering the importance of high-intensity ECC actions to boost efficacy of training (9).

Contrary to gravity-dependent isotonic RE protocols, the flywheel (FW) iso-inertial technology (10) allows for maximal intensity during the entire range of motion of the CON action, and with the appropriate technique (11), brief episodes of ECC overload (i.e. greater force/power production during ECC vs CON). Flywheel RE has been proven effective to counteract the negative effects of unloading on skeletal muscle (12, 13), and, compared with traditional WS, FW RE induces greater muscle hypertrophy in TD subjects (14, 8). In a recent clinical trial with stroke patients, who, similar to individuals with CP, present more preserved ECC vs CON force (15), FW RE training induced robust and positive muscle (i.e. hypertrophy, force, power) adaptations (16). Yet, differences in aetiology of the motor impairment and patient characteristics between stroke and CP warrant specific investigations testing the viability, safety and efficacy of FW RE in individuals with CP.

Individuals with unilateral CP have lower muscle mass and isometric strength on the more-affected side (17), thought to be caused by a combination of the primary cerebral insult and a secondary deterioration of skeletal muscle mass and quality (18). Past studies have shown that muscle mass and the ability to produce isometric force with muscles on the less-affected side is comparatively well preserved and more similar to TD individuals (17). Yet, whether the brain damage hampers power production of larger muscles on the less-affected side during dynamic and explosive actions (e.g. isotonic and iso-inertial RE) needs to be explored to fine-tune RE protocols in individuals with CP.

Therefore, we performed a comparative study of acute RE bouts of isotonic WS-based loading, and iso-inertial FW-based loading in CP and TD adolescents to address the following research questions:

  • Is the difference in ECC/CON peak power ratio between FW and traditional WS RE in CP adolescents comparable to that observed in TD adolescents?
  • Is there a deficit in performance for muscles of both the less- and more-affected sides during dynamic RE in CP, as assessed by peak power (watts) per mm muscle thickness?

General design

Eight teenagers and young adults diagnosed with CP, and 8 age-matched TD subjects completed 4 testing sessions within a period of 3 weeks. The first session consisted of an ultrasound examination of the vastus lateralis and intermedius muscles of the quadriceps group and the medial gastrocnemius muscle, assessment of anthropometry and passive range of motion (pROM) in lower limb, and familiarization on the WS and FW leg press devices, respectively. In the second session a 7-repetition maximum (7RM) test (unilaterally, in both legs) on the WS leg press was performed followed by a second familiarization on the FW device. The third session consisted of one set of 7RM test for each leg, assessing muscle activity (electromyography; EMG), active range of motion (aROM) and power during the test in the WS device. The session ended with a third familiarization on the FW leg press. During the fourth session EMG, aROM, power, and force were recorded during one set (for each leg) of 7 maximal repetitions using the FW leg press. A minimum of 48 h rest was scheduled between sessions for all subjects, and tests were conducted at the same time of the day (± 2 h) for each subject.


Subjects with CP (n = 8) were recruited from the Department of Pediatric Orthopedic Surgery, Astrid Lindgren’s Hospital, Karolinska University Hospital and the Habilitation services (Stockholm County Council). Inclusion criteria were children and young adults between the ages of 15 and 25 years with unilateral or bilateral spastic cerebral palsy and classified as GMFCS I–III according to Gross Motor Function Classification System, i.e. they could all walk with or without aids (19).  Detailed information about type of CP, contractures, spasticity and selective motor control of the foot of all CP subjects can be found in Table SI1. Exclusion criteria included surgical treatment of the knee extensor apparatus within the last 12 months, botulinum toxin treatment within the last 3 months and/or ongoing intrathecal baclofen treatment. A group of 8 TD subjects was recruited in the Stockholm area. Inclusion criteria were an age in the range 15–25 years with no history of neuromuscular diseases. Descriptive characteristics of both groups are shown in Table I. All subjects, and caregivers (when applicable), were informed of the purposes and potential risks associated with the interventions before giving their written informed consent to participate. The study protocol was approved by the Regional Ethical Review Board in Stockholm.

Table I. Characteristics of cerebral palsy and typically developed participants

Seven repetition maximum (7RM)

This test was performed in the WS leg press (NORSK Original Sequence Training System, Scandinavian Rehabilitation Equipment e. K, Borschen, Germany) unilaterally for both legs. Subjects performed 7 repetitions, from ~90° knee flexion to almost full extension (~180°), with a load previously estimated in the familiarization sessions. The load had been set at the second testing occasion, increasing by 5–10 kg in subsequent sets of 7 repetitions until the subject failed to complete the set in 2 successive trials. The 7RM was achieved between 3 and 6 attempts (sets), and trials were interspersed by a 3-min recovery period. Recovery between 7RM in the more-affected/non-dominant leg and the 7RM in the less-affected/dominant leg was 5 min. A standardized warm-up (7 repetitions at 60% of estimated 1RM) was conducted before the 7RM test.

Experimental exercise test on the weight-stack leg press

After the standardized warm-up, one set of 7 repetitions with the load corresponding to the 7RM for each leg was performed (starting with the more-affected/non-dominant leg; 5 min recovery between legs). Active knee ROM during the test was measured through a 3D kinematic system with 8 cameras (Vicon MX40, Oxford, UK) with 18 reflective markers attached to anatomical landmarks on the hips, knees and feet. Peak and mean power, and work were recorded in all repetitions with an encoder (100Hz; SmartCoach™, Stockholm, Sweden), but only the the first 3 repetitions were considered for downstream analysis in order to avoid the influence of potential fatigue. The encoder measures the velocity of every muscle action, and by manually setting the load in the associated software (SmartCoach™), peak and mean power, and work are calculated for each CON and ECC action. Angular velocity was calculated using the range of motion of the first 3 repetitions (°; Vicon) and the time (SmartCoach™) to complete the CON and ECC phases in each repetition. Muscle activity was measured with electromyography (EMG; see below).

Experimental exercise test on the flywheel leg press

Flywheel RE technology utilizes the inertia of a spinning flywheel to generate resistance. One inertia of 0.025 kg·m2 ((YoYo®, Technology AB, Stockholm, Sweden) was used. A standardized warm-up (one set of 7 reps, starting at 50% and progressively increasing the load to ~80% in the last 2 reps) was conducted before the experimental exercise test on the FW leg press. After one initial submaximal repetition to initiate the FW movement, instructions were given to “push” with maximal effort during the entire CON action, as to complete the action as fast as possible. Upon reaching ~170° of knee extension, the FW strap rewound due to inertial forces and the ECC action was initiated. According to our previous studies using this technology, the subjects were instructed during the familiarization sessions to achieve ECC-overload, i.e. they were requested to resist gently during the first third of the ECC action and then to apply maximal “braking” force to stop the movement at approximately 90° knee flexion (20). Then, the next CON action was immediately initiated. CP and TD subjects completed one set of 7 repetitions at maximal intensity for each leg (starting with more-affected/non-dominant leg; recovery time between legs was 5 min). Peak and mean power, work, aROM, and angular velocity were assessed for the first 3 repetitions, as described above for the WS exercise. Muscle activity was recorded using EMG (see below). To investigate whether the CP and TD subjects could produce ECC-overload, the ECC/CON ratio was determined in terms of peak power, i.e. ECC/CON peak power; values >1 indicate the presence of ECC-overload in FW RE execution. These data were compared with the same ratio obtained from the WS exercise bout.


During the experimental tests on both WS and FW devices, electrical activity of the vastus lateralis (VL) and biceps femoris (BF) muscles was assessed by EMG (21, 22). Electrodes were placed according to the Seniam guidelines (23), and raw data (1,000 Hz) was acquired using Vicon system (MX40, Oxford, UK). Data were then transformed to a frequency matching the ROM measurements (i.e. 100 Hz), and peak muscle activation averaged in the 3 first CON and ECC repetitions was calculated. ECC/CON ratio was determined in terms of muscle activity for each muscle group.


Muscle thickness of the thigh (vastus lateralis and intermedius of the quadriceps muscle together) and calf (medial gastrocnemius muscle) was measured in both legs using ultrasound. The examinations were performed by a paediatric radiologist with musculoskeletal ultrasound experience. For the thigh, the length of the upper leg was measured and divided at half length, which was where the probe was placed. For the lower leg, the probe was placed on the most prominent bulge of the medial gastrocnemius muscle. Ultrasound images were captured using a linear transducer with standard musculoskeletal settings on a Siemens Acuson s2000 (Siemens, Erlangen, Germany). The ultrasound settings were standardized and not altered between subjects. To avoid any pressure by the probe on the muscle that could bias the measurement, ample amounts of water-soluble transmission gel was applied. Pictures were considered valid when the normal curved shape of the superficial part of the muscle was clearly visible. Several images (5–8) were taken of each muscle and averaged to ensure consistency of measured values. Muscle thickness of vastus lateralis and intermedius muscles together was defined as the distance between the fascia overlaying the lateral vastus lateralis and the femur. Gastrocnemius muscle thickness was measured as the distance between the superficial and deep fascia of the medial gastrocnemius. The relationship between muscle thickness and power produced was investigated by dividing the power produced (in Watts) by the muscle thickness (in mm), for each leg, action, and device.

Statistical analysis

All data are presented as mean (standard deviation; SD). Normal distribution of data were verified using the Shapiro–Wilk test. Descriptive data were compared between CP and TD using independent Student’s t-test. Muscle activity and peak power ECC/CON ratio was investigated using a 3-way ANOVA with repeated measures for group (CP vs TD), device (FW vs WS) and leg (more-affected/non-dominant vs less-affected/dominant). A 2-way analysis of variance (ANOVA) with repeated measures for leg (more-affected/non-dominant vs less-affected/dominant) and group (CP vs TD) was used to analyse 7RM data and ultrasound (separately for thigh and calf muscles) results. Peak power, mean power, work, angular velocity, and power (Watts) produced per mm of muscle in the thigh were analysed using a 4-way ANOVA with repeated measures for group (CP vs TD), action (CON vs ECC), device (FW vs WS), and leg (more-affected/non-dominant vs less-affected/dominant). When significant interactions were found, simple effect tests were employed. The level of significance was set at 0.05. Cohen’s d was used to assess the effect size based on differences between means for a given variable (EF; 0.2; small, 0.5; medium, 0.8; large, 1.2; very large). All statistics were computed in SPSS version 25, 2017 (IBM Corp., Armonk, N.Y., USA).


Difference in eccentric/concentric peak power ratio between the flywheel and traditional weight-stack resistance exercise

There was a significant interaction leg × group (p = 0.018; F= 7.1), as well as significant main effects of device (p < 0.0005; F= 32.0) and leg (p = 0.035; F= 5.4) (Fig. 1). Thus, the ratio was significantly greater in FW RE than in the WS RE for CP (most-affected leg; p = 0.003, EF= 1.1, less-affected leg; p = 0.207, EF=0.5) and TD (non-dominant leg; p < 0.0005; EF > 1.2, and dominant leg; p = 0.001; EF > 1.2). The TD group achieved mean values for the ECC/CON ratio above 1, demonstrating peak power ECC-overload during FW RE. This was not observed for the CP group. The ECC/CON ratio of averaged peak quadriceps muscle activation (i.e. EMG) was overall higher in TD subjects compared with CP, independent of device and leg (main effect of group; p = 0.038; F= 5.2) (Fig. 2A). No effects or interactions were observed in the ECC/CON ratio for biceps femoris (Fig. 2B).

Fig. 1. Eccentric (ECC)/concentric (CON) peak power ratio (mean ± standard deviation (SD)) during leg press exercise using a flywheel or a weight-stack resistance exercise device in cerebral palsy (CP) or typically developed (TD) individuals. Note that values > 1 indicate eccentric overload. Significant main effects (p < 0.05); ainteraction leg × group; bmain effect of device; cmain effect of leg. Significant simple effects (p < 0.05); *greater than weight-stack within leg and group. Dom: dominant leg; Ndom: non-dominant leg; LaL: less-affected leg; MaL: more-affected leg.

Fig. 2. Eccentric (ECC)/concentric (CON) ratio in muscle activity (mean±SD) for: (A) vastus lateralis and (B) biceps femoris of the more-affected/non-dominant leg and less-affected/dominant leg during the flywheel and weight-stack exercise in cerebral palsy (CP) and typically developed (TD) individuals. amain effect of group; p = 0.038. Dom: dominant leg; Ndom: non-dominant leg; LaL: less-affected leg; MaL: more-affected leg.

Subjects with CP showed significantly lower peak power independent of device, leg and muscle action (interaction device × action × group; p = 0.001; F= 15.6; EFs>1.2 Table II). Both the TD and CP subjects produced higher power during the CON action compared with the ECC action in the WS device. There was no significant differences between CON and ECC peak power during the acute bout on the FW device for either group. In both FW and WS devices subjects with CP produced less peak power with the more-affected leg compared with the less-affected leg (p < 0.04; EFs 0.6-0.9). Finally, only TD subjects showed higher peak power in the ECC action on the FW compared with the WS device (p < 0.001; EF > 0.9 for both legs).

Similar significant results for mean power were found (Table II); (i.e. higher mean power in TD vs CP, higher mean power in less-affected vs more-affected leg within the CP group, and higher mean power in FW vs WS within the TD group only). Work production in the CP group was lower compared with TD, independent of leg, device and muscle action (p < 0.01, Table II), and the more-affected leg of subjects with CP produced less work in both RE devices than the less-affected leg (p < 0.01). Performing RE in the FW device involved a larger range of motion compared with WS RE (main effect p < 0.0005; F= 30.6; EF= 0.6 Table II). Angular velocity was higher in FW compared with WS (main effect device; p < 0.0005; F= 59.5), although such difference was much greater in TD (EF > 1.2) than CP subjects (EF= 0.6) (interaction device × group; p = 0.002; F= 13.0; Table II). Finally, TD showed greater (EF > 1.2) 7RM values in both legs than CP patients (more-affected/non-dominant; p < 0.0005, less-affected/dominant; p = 0.007; Table II). The values for 7RM in CP subjects were larger in the less-affected leg compared with the more-affected leg (p < 0.0005; EF= 0.9; Table II).

Table II. Power and work production during flywheel or weight-stack resistance exercise, angular velocity and 7 repetition maximum (7RM) in the weight-stack device, in cerebral palsy and typically developed participants.

Performance in muscles of less- and more-affected sides during dynamic resistance exercise

An interaction leg × group (p = 0.002; F= 13.6) was found for the thigh muscles thickness (Fig. 3A). TD presented greater values than the CP group in the more-affected/non-dominant leg (p = 0.02; EF > 1.2). Also, CP subjects showed greater thigh muscle thickness in the less-affected leg than in the more-affected one (p < 0.0005; EF= 1.0). Calf muscle thickness was lower in CP patients independent of the leg (main effect of group; p = 0.001; F= 15.6; EF > 1.2; Fig. 3B). TD subjects produced more power (Watts) per mm of muscle in the thigh than the CP group, independent of leg, device and action (main effect of group; p < 0.0005; F= 26.2; EF > 1.2, Figs 3C and 3D).

Fig. 3. (A) Thickness (mm) of the quadriceps muscle in typically developed (TD) and cerebral palsy (CP) subjects (mean±standard deviation (SD)). (B) Thickness (mm) of the gastrocnemius muscle in TD and CP subjects. (C) Peak power (watts) per mm quadriceps muscle in TD and CP on the weight-stack (WS) device. (D) Peak power (watts) per mm quadriceps muscle in TD and CP on the flywheel (FW) device. *Significantly different from TD (p < 0.05). †Significantly different from less-affected leg within CP group (p < 0.05). ‡Significantly different from TD independent of leg, device and action (p < 0.001). Dom: dominant leg; Ndom: non-dominant leg; LaL: less-affected leg; MaL: more-affected leg.


This study examined whether the difference in ECC/CON peak power ratio between FW and traditional WS RE in CP adolescents is comparable to that observed in TD adolescents. Subjects with CP were able to produce a greater ECC/CON peak power ratio while exercising on the FW device, but absolute values were lower compared with the traditional WS device. Individuals with CP can thus use the correct fast deceleration technique during the eccentric phase of FW, and produce a relatively greater peak power than during WS. But, as impaired motor control in CP reduces the rate of force development (24), a lower absolute peak power output is produced during FW compared with WS.

A further potential reason for the difficulties observed with respect to subjects with CP mastering the iso-inertial FW RE device relates to other aspects of poor neuromuscular control. It has been shown that individuals with CP use fewer synergies (weighted groups of muscles) compared with TD individuals during gait (25, 26). As impaired selective neuromuscular control leads to recruitment of fewer synergies (27), less efficient movement patterns could contribute to less power being produced during RE in individuals with CP compared with TD controls. A recent study by Goudriaan and co-workers targets the importance of neural and non-neural contributors to impaired gait in children with Duchenne’s muscle dystrophy (DMD) (non-neural) and CP (neural + non-neural) (28). This study suggests that the non-neural weakness in DMD has only limited influence on the complexity of neuromuscular control during gait and that the decreased synergistic variance in children with CP is mainly related to neural impairments caused by the brain lesion. In the context of the current study, these data infer that the loading complexity (force, speed and acceleration) of the iso-inertial device hampers performance with respect to peak power in adolescents with CP. Muscle composition in CP is altered with reduced muscle mass and increased amount of intramuscular fat and connective tissue (29–32), which probably also contribute to less force and power being produced.

Skeletal muscle strengthening protocols for individuals with CP have been shown to promote gains in isometric force and task-specific performance (33–35). However, weak transfer effects of RE to functional mobility, such as standing up and walking, walking speed, walking in stairs, and movement measured with accelerometers, have highlighted the need for alternative RE protocols, e.g. with respect to intensity, power and loading strategy. In recent years, some studies have described significant improvements in functional outcomes following a period of training to reach specific goals. Van Vulpen and colleagues (36) have performed a study resembling HIIT (high-intensity interval training) with walking/running/climbing exercises 3 times/week for 14 weeks with progressive resistance. They showed significant improvements in muscle strength, walking capacity and achievement of individually tailored walking/running/climbing goals measured with the goal attainment scale (GAS). Likewise, Gillett and co-workers (35) performed a high-intensity 12-week RE training study targeting the lower limbs, with the addition of anaerobic training activities, and observed improvements in functional outcomes (6-min walking test and a combined score for functional strength including 30-s repetition maximum for lateral step-up, sit-to-stand and stand from half-kneel). Moreover, gains in muscle volume were correlated with increased plantar flexor isometric strength. Therefore, it appears that a common denominator for studies demonstrating transfer effects of RE into functional gains in individuals with CP is a high training intensity and volume, as well as aspects of specificity of training, i.e. training actual functional activities.

We thereafter asked whether there is deficit in performance, probably originating from a neural component (voluntary activation) limiting muscles of both the less- and more-affected sides during dynamic RE in CP. Previous studies have investigated how the brain damage affects force production in individuals with CP. Hussain and colleagues (17) investigated a group of 11 men with unilateral spastic CP and 11 TD controls. Skeletal muscle anatomical cross-sectional area (ACSA), as measured with magnetic resonance imaging (MRI), was significantly reduced in the paretic limb (control: 25.9 cm2, paretic: 20.7 cm2, non-paretic 24.3 cm2). In addition, the more-affected limb was clearly the weaker side, as the plantar flexors of the paretic limb produced 52% less isometric torque compared with the control group, and 42% less compared with the non-paretic side. Similar results were observed, with the more-affected side producing significantly less power. The thickness of the vastus intermedius and lateralis muscles were significantly lower on the more-affected side in the CP group, whereas the same muscle in the less-affected leg did not differ from control. Interestingly, when normalizing peak power either during CON or ECC actions with thigh muscle thickness, it becomes evident that performance in both the more- and less-affected sides is markedly impaired. This suggests that the neural component reduces power production on both sides in an equal manner. Thus, even though muscle thickness is approximately 30% greater in the less-affected leg, power per mm muscle is not greater than the more-affected leg. Poor power production will probably hamper exercise intensity and thereby negatively influence training adaptations. A speculative extrapolation of this finding could be that this suggests that the thigh muscles on both the more- and the less-affected sides in the CP group will not receive an optimal training stimulus with respect to mechanical load during either WS or FW exercise when performing a set with a low number of repetitions. To compensate for the lower RE load during low-repetitions, high-intensity routines (due to the neurological deficit), a possible suggestion is that individuals with CP should instead increase training volume and always aim for repetitions until failure (volitional fatigue). Resistance exercise until failure using >7–10 RM loading, rather than fewer repetitions with high intensity, still calls for both type I and type II muscle fiber activation and produces a similar muscle fiber activation pattern compared with a short duration high-intensity protocol (37). Furthermore, data show that chronic RE training with low loads until failure results in hypertrophy of both type I and type II fibres in TD subjects (38–40). Thus, in CP, RE regimes aiming for repetitions until task failure will probably stress the muscle in a more efficient manner, compared with a pre-set number of repetitions at a comparatively higher load.

Study limitations

When interpreting the results of the present study, some limitations should be considered. The number of subjects in each group may be too small to detect all clinically relevant differences in the loading pattern between FW and WS exercises. Also, the variability added by the different degree of disability in the CP group, as well as the differences between groups in terms of age and weight should be considered. Finally, the fact that the order of testing sessions was standardized, and not randomized, should also be taken into account.


In conclusion, FW RE calls for greater muscle activity during the ECC action than WS RE, which is translated into a greater ECC/CON peak power ratio. However, probably due to impaired neuromuscular control, subjects with CP could not elicit the ECC overload or produce higher absolute peak power on the FW device in a comparable manner to TD adolescents. Moreover, peak power production per mm muscle was markedly reduced in both legs in individuals with CP on both the WS and FW device. Thus, our comparative study of WS and FW exercise does not support the implementation of iso-inertial loading regimes for RE training in adolescents with CP, an interpretation specifically supported by the overall lower peak power output observed in CP subjects on the FW device. We therefore suggest that an alternative loading regime, favouring lower load RE performed until failure, could be more suitable in this specific population.


The authors thank students Negar Chamsai, Sara Rindborg, Olav Kvitastein and Johan Persson for their valuable efforts during data collection.

This work was supported by grants from Linnea & Josef Carlsson (FvW) and Stiftelsen Promobilia and RBU research foundation (RFG).

  1. O’Shea TM. Diagnosis, treatment, and prevention of cerebral palsy. Clin Obs Gynecol 2008; 51: 816–828.
    View article    Google Scholar
  2. Rosenbaum P, Paneth N, Leviton A, Goldstein M, Bax M, Damiano D, Dan B, Jacobsson B. A report: the definition and classification of cerebral palsy April 2006. Dev Med Child Neurol Suppl 2007; 109: 8–14.
    View article    Google Scholar
  3. von Walden F, Jalaleddini K, Evertsson B, Friberg J, Valero-Cuevas FJ, Ponten E. Forearm Flexor Muscles in Children with Cerebral Palsy Are Weak, Thin and Stiff. Front Comput Neurosci 2017; 11: 30.
    View article    Google Scholar
  4. Gough M, Shortland AP. Could muscle deformity in children with spastic cerebral palsy be related to an impairment of muscle growth and altered adaptation? Dev Med Child Neurol 2012; 54: 495–499.
    View article    Google Scholar
  5. Taylor NF, Dodd KJ, Baker RJ, Willoughby K, Thomason P, Graham HK. Progressive resistance training and mobility-related function in young people with cerebral palsy: a randomized controlled trial. Dev Med Child Neurol 2013; 55: 806–812.
    View article    Google Scholar
  6. Scholtes VA, Becher JG, Comuth A, Dekkers H, Van Dijk L, Dallmeijer AJ. Effectiveness of functional progressive resistance exercise strength training on muscle strength and mobility in children with cerebral palsy: a randomized controlled trial. Dev Med Child Neurol 2010; 52: e107–e113.
    View article    Google Scholar
  7. Damiano DL, Martellotta TL, Quinlivan JM, Abel MF. Deficits in eccentric versus concentric torque in children with spastic cerebral palsy. Med Sci Sport Exerc 2001; 33: 117–122.
    View article    Google Scholar
  8. Norrbrand L, Fluckey JD, Pozzo M, Tesch PA. Resistance training using eccentric overload induces early adaptations in skeletal muscle size. Eur J Appl Physiol 2008; 102: 271–281.
    View article    Google Scholar
  9. Colliander EB, Tesch PA. Responses to eccentric and concentric resistance training in females and males. Acta Physiol Scand 1991; 141: 149–156.
    View article    Google Scholar
  10. Berg HE, Tesch A, Tesch PA. A gravity-independent ergometer to be used for resistance training in space. Aviat Sp Environ Med 1994; 65: 752–756.
    View article    Google Scholar
  11. Tous-Fajardo J, Maldonado RA, Quintana JM, Pozzo M, Tesch PA. The flywheel leg-curl machine: offering eccentric overload for hamstring development. Int J Sport Physiol Perform 2006; 1: 293–298.
    View article    Google Scholar
  12. Alkner BA, Tesch PA. Knee extensor and plantar flexor muscle size and function following 90 days of bed rest with or without resistance exercise. Eur J Appl Physiol 2004; 93: 294–305.
    View article    Google Scholar
  13. Irimia JM, Guerrero M, Rodriguez-Miguelez P, Cadefau JA, Tesch PA, Cussó R, Fernandez-Gonzalo R. Metabolic adaptations in skeletal muscle after 84 days of bed rest with and without concurrent flywheel resistance exercise. J Appl Physiol 2016; 122: 96–103.
    View article    Google Scholar
  14. Maroto-Izquierdo S, García-López D, Fernandez-Gonzalo R, Moreira OC, González-Gallego J, de Paz JA. Skeletal muscle functional and structural adaptations after eccentric overload flywheel resistance training: a systematic review and meta-analysis. J Sci Med Sport 2018; 20: 943–951.
    View article    Google Scholar
  15. Hedlund M, Sojka P, Lundström R, Lindström B. Insufficient loading in stroke subjects during conventional resistance training. Adv Physiother 2012; 14: 18–28.
    View article    Google Scholar
  16. Fernandez-Gonzalo R, Fernandez-Gonzalo S, Turon M, Prieto C, Tesch PA, García-Carreira MC. Muscle, functional and cognitive adaptations after flywheel resistance training in stroke patients: A pilot randomized controlled trial. J Neuroeng Rehabil 2016; 13: 37.
    View article    Google Scholar
  17. Hussain AW, Onambele GL, Williams AG, Morse CI. Muscle size, activation, and coactivation in adults with cerebral palsy. Muscle and Nerve 2014; 49: 76–83.
    View article    Google Scholar
  18. Von Walden F, Gantelius S, Liu C, Borgstrom H, Björk L, Gremark O, Stål P, Nader GA, Pontén E. Muscle contractures in patients with cerebral palsy and acquired brain injury are associated with extracellular matrix expansion, pro-inflammatory gene expression, and reduced rRNA synthesis. Muscle and Nerve 2018; 58: 277–285.
    View article    Google Scholar
  19. Palisano RJ, Rosenbaum P, Bartlett D, Livingston MH. Content validity of the expanded and revised Gross Motor Function Classification System. Dev Med Child Neurol 2008; 50: 744–750.
    View article    Google Scholar
  20. Martinez-Aranda LMLM, Fernandez-Gonzalo R. Effects of inertial setting on power, force, work, and eccentric overload during flywheel resistance exercise in women and men. J Strength Cond Res 2017; 31: 1653–1661.
    View article    Google Scholar
  21. Alkner BA, Tesch PA, Berg HE. Quadriceps EMG/force relationship in knee extension and leg press. Med Sci Sports Exerc 2000; 32: 459–463.
    View article    Google Scholar
  22. Alkner BA, Bring DK. Muscle Activation During Gravity-Independent Resistance Exercise Compared to Common Exercises. Aerosp Med Hum Perform 2019; 90: 506-512.
    View article    Google Scholar
  23. Hermens HJ, Freriks B, Disselhorst-Klug C, Rau G. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol 2000; 10: 361–374.
    View article    Google Scholar
  24. Moreau NG, Falvo MJ, Damiano DL. Rapid force generation is impaired in cerebral palsy and is related to decreased muscle size and functional mobility. Gait Posture 2012; 35: 154–158.
    View article    Google Scholar
  25. Rozumalski A, Steele KM, Schwartz MH. Muscle synergies are similar when typically developing children walk on a treadmill at different speeds and slopes. J Biomech 2017; 64: 112–119.
    View article    Google Scholar
  26. Shuman B, Goudriaan M, Bar-On L, Schwartz MH, Desloovere K, Steele KM. Repeatability of muscle synergies within and between days for typically developing children and children with cerebral palsy. Gait Posture 2016; 45: 127–132.
    View article    Google Scholar
  27. Steele KM, Rozumalski A, Schwartz MH. Muscle synergies and complexity of neuromuscular control during gait in cerebral palsy. Dev Med Child Neurol 2015; 57: 1176–1182.
    View article    Google Scholar
  28. Goudriaan M, Shuman BR, Steele KM, Van den Hauwe M, Goemans N, Molenaers G, Desloovere K. Non-neural Muscle Weakness Has Limited Influence on Complexity of Neuromuscular control during Gait. Front Hum Neurosci 2018; 12: 5.
    View article    Google Scholar
  29. Booth CM, Cortina-Borja MJF, Theologis TN. Collagen accumulation in muscles of children with cerebral palsy and correlation with severity of spasticity. Dev Med Child Neurol 2001; 43: 314–320.
    View article    Google Scholar
  30. De Bruin M, Smeulders MJ, Kreulen M, Huijing PA, Jaspers RT. Intramuscular connective tissue differences in spastic and control muscle: A mechanical and histological study. PLoS One 2014; 9: e101038.
    View article    Google Scholar
  31. Ito J ichi, Araki A, Tanaka H, Tasaki T, Cho K, Yamazaki R. Muscle histopathology in spastic cerebral palsy. Brain Dev 1996; 18: 299–303.
    View article    Google Scholar
  32. Rose J, Haskell WL, Gamble JG, Hamilton RL, Brown DA, Rinsky L. Muscle pathology and clinical measures of disability in children with cerebral palsy. J Orthop Res 1994; 12: 758–768.
    View article    Google Scholar
  33. Gillett JG, Boyd RN, Carty CP, Barber LA. The impact of strength training on skeletal muscle morphology and architecture in children and adolescents with spastic cerebral palsy: A systematic review. Res Dev Disabil 2016; 56: 183–196.
    View article    Google Scholar
  34. Gillett JG, Lichtwark GA, Boyd RN, Barber LA. Functional Capacity in Adults With Cerebral Palsy: Lower Limb Muscle Strength Matters. Arch Phys Med Rehabil 2018; 99: 900-906.e1.
    View article    Google Scholar
  35. Gillett JG, Lichtwark GA, Boyd RN, Barber LA. Functional anaerobic and strength training in young adults with cerebral palsy. Med Sci Sports Exerc 2018; 50: 1549–1557.
    View article    Google Scholar
  36. van Vulpen LF, de Groot S, Rameckers EA, Becher JG, Dallmeijer AJ. Improved parent-reported mobility and achievement of individual goals on activity and participation level after functional power-training in young children with cerebral palsy: a double-baseline controlled trial. Eur J Phys Rehabil Med 2018; 54: 730–737.
    View article    Google Scholar
  37. Morton RW, Sonne MW, Farias Zuniga A, Mohammad IYZ, Jones A, McGlory C, Keir PJ, Potvin JR, Phillips SM. Muscle fibre activation is unaffected by load and repetition duration when resistance exercise is performed to task failure. J Physiol 2019; 597: 4601-4613.
    View article    Google Scholar
  38. Mitchell CJ, Churchward-Venne TA, West DW, Burd NA, Breen L, Baker SK, Phillips SM. Resistance exercise load does not determine training-mediated hypertrophic gains in young men. J Appl Physiol 2012; 113: 71–77.
    View article    Google Scholar
  39. Morton RW, Oikawa SY, Wavell CG, Mazara N, McGlory C, Quadrilatero J, Baechler BL, Baker SK, Phillips SM. Neither load nor systemic hormones determine resistance training-mediated hypertrophy or strength gains in resistance-trained young men. J Appl Physiol 2016; 121: 129–138.
    View article    Google Scholar
  40. Schoenfeld BJ, Grgic J, Ogborn D, Krieger JW. Strength and hypertrophy adaptations between low- vs. High-load resistance training: A systematic review and meta-analysis. J Strength Cond Res 2017; 31: 3508–3523.
    View article    Google Scholar
Supplementary content
Table SI


Do you want to comment on this paper? The comments will show up here and if appropriate the comments will also separately be forwarded to the authors. You need to login/create an account to comment on articles. Click here to login/create an account.