Skip to content

The Effect of Perturbation Training on Reactive Balance in Children With Cerebral Palsy

The Effect of Perturbation Training on Reflex Modulation, Reactive Balance and Walking in Children With Spastic Cerebral Palsy.

Status
Completed
Phases
Unknown
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT07818928
Enrollment
24
Registered
2026-09-14
Start date
2021-07-26
Completion date
2025-08-29
Last updated
2026-09-15

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Cerebral Palsy (CP), Typically Developing Children

Keywords

Spastic cerebral palsy, Reactive balance, Perturbation training, Balance performance, muscle co-activation

Brief summary

The aim of this project was to evaluate whether perturbation training influences reflex modulation in children with spastic CP and whether changed reflex modulation leads to improved reactive balance and walking. Reactive balance was trained using rotational perturbations. Reactive balance performance and reactive muscle activity during trained and untrained perturbation conditions were assessed in children with cerebral palsy and typically developing children.

Detailed description

Background: With an incidence of 2 to 3 per 1000 live births, cerebral palsy (CP) is the most common cause of physical disability in children. CP is caused by a non-progressive lesion to the developing brain. The resulting inability to selectively control muscles, spasticity, and muscle weakness undermine the child's mobility and balance. Spasticity is the most common impairment in CP and is clinically characterized by increased resistance to imposed stretch to the muscle. During clinical tests of spasticity, muscles are stretched with the purpose of assessing resistance. Yet, similar stretches occur on a daily basis when balance during standing or walking is perturbed (e.g. standing on a departing bus or walking on uneven terrain). Therefore, spasticity might contribute to balance impairments that are common in CP. When standing balance is perturbed, intrinsic mechanical properties depending on muscle tone (\> 0ms), spinal reflexes leading to small changes in muscle activity (\> 50ms), and balance correcting changes (\>100-120ms after onset) contribute to restoring posture with different delays. The effectiveness of those mechanisms will depend on the ability to modulate muscle tone and reflex gains. Children with CP have increased muscle tone and reflex hyper-excitability, which might contribute to decreased balance capacities. Furthermore, the ability to modulate reflex activity is decreased in children with spastic CP compared to healthy children. This might explain why children with CP respond differently to a perturbation of standing balance than typically developing (TD) children. For instance, children with CP adopt a hip strategy at lower perturbation magnitudes than TD children. However, there is evidence that spinal reflexes can be trained in both healthy individuals and individuals with neurological disorders and that such reflex conditioning improves function, such as walking. Perturbation training has been shown to improve balance in children with CP . However, the underlying mechanisms of these improvements are not well understood. Therefore, the aim of this project was to evaluate whether perturbation training influences reflex modulation in children with spastic CP and whether changed reflex modulation leads to improved reactive balance and walking. Methods: An intervention study was performed in which children with spastic CP followed perturbation training during a 3-week period. The training consists of 9 sessions (3 sessions/week) of 30 minutes. During these sessions, balance will be trained using rotational perturbations with different magnitudes. The participant is asked to maintain balance without stepping and stay as upright as possible. Difficulty level will increase every training adapted to the subject. Before and after training, the response to muscle stretches, reactive balance (strategies) and walking performance was assessed. During the pre- and post-measurement sessions, kinematics (joint angles), ground reaction forces (GRFs) and muscle activity (electromyography, EMG) were recorded.

Interventions

OTHERBalance training on a moving platform

The training intervention consisted of nine sessions (spread over 3 weeks), which took about 30 minutes each. The first and last session immediately followed and preceded respectively the pre- and post-intervention assessments. During one session, three identical blocks of 32 toe-up rotational perturbations were administered. Each block consisted of perturbations with two different difficulty levels (16 trials each). The administered perturbation levels were selected based on the participant's performance in the previous training session (for training sessions two to nine) and pre-intervention assessment (for the first training session). If the participant was able to maintain balance without stepping in at least 80% of the trials, a more difficult level was offered in the next training. When the participant was able to perform all levels, the training was continued by combining both lower and higher levels, in order to train the complete range of perturbation levels.

Sponsors

Universitaire Ziekenhuizen KU Leuven
Lead SponsorOTHER
KU Leuven
CollaboratorOTHER

Study design

Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
TREATMENT
Masking
NONE

Intervention model description

Both groups (CP and TD) followed the perturbation training (3 weeks - 9 sessions)

Eligibility

Sex/Gender
ALL
Age
5 Years to 17 Years
Healthy volunteers
Yes

Inclusion criteria

* Diagnosis of cerebral plasy * Spasticity as defined by a clinical assessment, in the gastrocnemius muscle * Aged 5-17 years old * GMFCS level 1-2

Exclusion criteria

* Presence of ataxia or dystonia * Cognitive problems that impede measurements * Severe co-morbidities * Irritated skin or open wounds where sensors will be placed * Botox or surgery in the lower limbs in the last six months * Not be able to stand or walk independently

Design outcomes

Primary

MeasureTime frameDescription
Balance performanceBaseline and after finishing the nine training sessions (3 weeks)Balance performance was evaluated as the number of levels the participant was able to complete before and after training. A level was considered successful if the participant was able to maintain balance without stepping in at least five out of eight trials.
Co-contraction indexBaseline and after finishing the nine training sessions (3 weeks)The co-contraction index (CCI), a measure of muscle co-activation, was computed as the common area below tibialis anterior and respectively lateral gastrocnemius, medial gastrocnemius, and soleus filtered and scaled EMG trajectories averaged over time (equation 1) during rotational and translational perturbations of standing balance. CCI = (∑〖min⁡(〖EMG\_PF,〖EMG\_TA)〗)/(# frames) (Equation 1) with PF referring to lateral gastrocnemius, medial gastrocnemius, or soleus; and # frames the number of data points in the analyzed time interval. The CCI was calculated for each trial over an interval from 0.5s before perturbation onset until 1.5s after perturbation onset. Trials where the participant took a step were excluded. The average CCI across all non-stepping trials was calculated for each level.

Secondary

MeasureTime frameDescription
Average muscle activity within three time zonesBaseline and after finishing the nine training sessions (3 weeks)Average muscle activity was compute for lateral and medial gastrocnemius, soleus, and tibialis anterior in three time bins during rotational and translational perturbations of standing balance. Reactive muscle activity within each time bin was computed as the average filtered and scaled EMG within the time bin from which baseline activity (average EMG during 100ms preceding perturbation onset) was subtracted. The first time bin started at platform onset and ended 150 ms later. The second time bin lasted from 150 ms to 250 ms after platform onset. The third time bin (Z3) lasted from 250 ms to 400 ms after perturbation onset.

Countries

Belgium

Contacts

PRINCIPAL_INVESTIGATORKaat Desloovere, Prof. dr.

Department of Rehabilitation Sciences, KU Leuven, Belgium

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Sep 16, 2026