Upper Limb Somatosensory Discrimination Therapy and Dose-matched Motor Therapy in Children and Adolescents With Unilateral Cerebral Palsy

Upper Limb Somatosensory Discrimination Therapy and Dose-matched Motor Therapy in Children and Adolescents With Unilateral Cerebral Palsy

The Effectiveness of an Intensive Upper Limb Therapy Program Focused on Somatosensation in Children and Adolescents With Unilateral Cerebral Palsy

A randomized, controlled, and evaluator-blinded trial will be carried out to investigate the effects of 8 weeks of upper limb somatosensory discrimination therapy compared to an equal dose of motor therapy on sensorimotor outcomes in children and adolescents with spastic unilateral cerebral palsy aged 7 to 15 years old. Additionally, the potential role of clinical and neurological baseline characteristics on treatment response will be explored, including the extent of the brain lesion, age and baseline somatosensory function. The researchers primarily expect that the bimanual performance will improve equally in children and adolescents from both intervention groups, as measured immediately after the intervention has ended. The researchers hypothesize, however, that there will be a better retention effect in children and adolescents that received somatosensory discrimination therapy, resulting in differences between both groups in bimanual performance as measured at 6-months follow-up. The researchers further expect larger improvements in somatosensory function for children and adolescents in the somatosensory discrimination group and this both immediately after the intervention and at 6-month follow-up.

BACKGROUND: Although cerebral palsy has primarily been defined as a motor disorder, research has shown that up to 90% of children and adolescents with unilateral cerebral palsy also present with somatosensory deficits in the upper limb. Despite the fact that somatosensation has shown to be crucial for coordinated movements and even motor learning, therapy programs designated to improve somatosensory deficits is scarce. Nevertheless, previous studies provided a neurological basis for somatosensory interventions in individuals with unilateral cerebral palsy, showing that the somatosensory system is disorganized and potentially treatment responsive. Hence, a clinical reasoning framework for somatosensory interventions was recently developed. Somatosensory discrimination therapy, also known as Sense for Kids therapy, was one of the recommended approaches. This therapy was found to improve tactile perception in chronic stroke patients. Also in children and adolescents with unilateral cerebral palsy, a first study reported promising results to improve functional hand use as well as somatosensory function. Yet, the full clinical potential and long-term effects of this therapy still remain to be explored. AIM: The aim of this study is to investigate the effects of Sense for Kids therapy compared to an equal dose of motor therapy on upper limb sensorimotor outcomes in children and adolescents with unilateral cerebral palsy. Additionally the potential role of clinical and neurological baseline characteristics on treatment response will be explored, including the anatomical characterization of the brain lesion, age and baseline somatosensory function. METHODS/DESIGN: A randomized, controlled, and evaluator-blinded randomized controlled trial will be carried out in 50 children and adolescents with spastic unilateral cerebral palsy aged between 7 and 15 years old. To determine the number of required participants, a sample size calculation was performed for the primary research objective, being better bimanual performance at 6 months follow-up for children and adolescents that received Sense for Kids therapy compared to upper limb motor therapy. An effect size of 0.9 was calculated based on the results of earlier studies on intensive upper limb therapy in children with unilateral cerebral palsy. Combined with an alpha-level of 0.05 and a statistical power of 0.80, this results in a required sample size of 21 children and adolescents in each group to detect a difference equal or larger than the smallest detectable difference between groups. This number will be increased to 25 participants per group to account for inevitable dropouts and to maximize the study sample for prediction purposes. Participants will be randomized to the intervention or active control group through randomization by minimization with an allocation ratio of 1:1. Participants will be stratified based on age (2 levels: <11 years and ≥11 years), manual ability (3 Manual Ability Classification Scale: I, II and III) and tactile impairment (2 levels: mild or moderate-severe). The probability will be set at 80% to diminish allocation prediction. Participants in both groups will receive three session of 45 minutes upper limb therapy per week for a period of 8 weeks, resulting in a total therapy intensity of 18 hours. The intervention group will receive Sense for Kids therapy, which consists of component-based training and occupation-based training. During the component-based training participants will perform graded and structured practice of tactile object recognition, texture discrimination and proprioception. During the occupation-based practice, participants will repetitively perform self-chosen activities of daily living while attention is denoted to somatosensory aspects necessary for successful task completion (e.g., position of the upper limb, texture of materials, localization of materials based on touch). Participants in the control group will receive an equal amount of motor training, consisting of unimanual motor tasks and bimanual goal directed training of self-chosen activities. Outcomes will be assessed at three different points in time, being within two weeks prior to therapy commencement (baseline), within one week following therapy completion (post-intervention) and 6 months after the therapy has ended (follow-up). Outcome assessments include qualitative and quantitative measures of upper limb sensorimotor function. The primary outcome measure is the Assisting Hand Assessment and is used to score the (spontaneous) use of the more impaired hand during bimanual activities. The medical imaging protocol, which will be only acquired at baseline, includes structural Magnetic Resonance Imaging and diffusion weighted images. The time effects of the intervention on upper limb sensorimotor outcomes will be objectified using linear and/or generalized mixed models. Behavioral and neurological variables will further be included in the model as covariates to study their predictive value on the therapy response.

Upper extremity[MeSH], Physiotherapy[MeSH], Occupational therapy[MeSH], Somatosensory discrimination therapy, Upper limb motor therapy, Randomized controlled trial[MeSH], Child[MeSH], Adolescent[MeSH], Somatosensation[MeSH], Bimanual performance, Goal performance, Robotics[MeSH], Neuroimaging[MeSH]

Randomization will be performed by an independent person who is not involved in the selection procedure and will not have access to any additional clinical information about the participants.

The assessments of sensorimotor upper limb function at all three time points will be performed by a blinded assessor who is not involved in the conception of the study or the provision of therapy to the participants. Assessments that were video-recorded (Assisting Hand Assessment and Goal Attainment Scaling) will be scored afterwards by another evaluator who will be blinded to group allocation and time point of assessment. Since the robotic measurements and three-dimensional motion analysis are fully automated, these assessments will be performed by a physiotherapist not blinded to group allocation. Participants and parents will be blinded to the study hypotheses. The treating therapists will not be blinded to group allocation.

Participants will receive three 45 minute sessions of upper limb somatosensory discrimination therapy per week for a period of 8 weeks. Each session includes 30 minutes of component-based training, including graded practice of three components of somatosensation (i.e., tactile object recognition, texture discrimination and proprioception). Next, 15 minutes of occupation-based practice will be completed to enhance transfer to a functional context. This includes repetitive and structured practice of self-chosen activities of daily living, while attention is designated to the somatosensory aspects necessary for successful task completion (e.g., identification of button based on touch).

Participants in the control group will receive an equal amount of motor training, including 30 minutes of unimanual motor tasks and 15 minutes of bimanual goal directed training. The unimanual motor tasks will be selected based on the main difficulties of the child (e.g., grip strength, supination). Bimanual goal directed training includes whole task practice of self-chosen activities of daily living. All therapy sessions will be performed under supervision of a trained physical and/or occupational therapist. Therapy adherence and progression will be closely monitored by the therapist.

The Assisting Hand Assessment (AHA) and Adolescent Assessment Hand Assessment (Ad-AHA) evaluate the spontaneous use of the more impaired upper limb during bimanual activities. For these assessments a semi-structured play session with standardized toys and materials, that require the use of both hands, is performed and video-recorded. Subsequently, 20 items are scored on a four-point Likert scale ranging from 1 ('does not do') to 4 ('effective use') and a logit-based total score in AHA unit is calculated, ranging from 0 to 100 AHA units. A highers score represents better bimanual performance. The AHA and Ad-AHA have shown to be a reliable and valid assessment for children and adolescents with unilateral cerebral palsy. Depending on the age of the participants, the AHA or Ad-AHA will be selected at each time point.

The second version of the Children's Hand-use Experience Questionnaire (CHEQ 2.0) is an online questionnaire that is used to capture the child's experience of using the more impaired upper limb during 29 daily life activities for which usually two hands are needed. For each bimanual activity the parents are asked to indicate if the child or adolescent uses one or both hands to perform the activity or if assistance is needed. Followed by three sub-questions that are scored on a four-point Likert scale, indicating the (1) efficacy of the more impaired hand, (2) time needed to perform the activity compared to peers and (3) if the child feels bothered by the more impaired hand when performing the activity. For each sub-scale a total score will be calculated ranging from 0 to 100, with a higher score representing better performance.

3D motion analysis will be used to evaluate spatiotemporal aspects of bimanual coordination during the functional box-opening task. Parameter of interest are total movement time and goal synchronization, with higher score indicating worse bimanual coordination.

Bimanual coordination will further be objectified using the ball-on-bar task on the Kinarm Exoskeleton robot. Parameters of interest are determined based on previous research in children and adolescents with unilateral cerebral palsy and include bar tilt standard deviation, hand path length bias, reaction time difference and hand speed difference. For all parameters, a higher score indicated worse bimanual coordination.

For the perceptual boundary task, the slope of a psychometric curve will be calculated. The slope describes how sensitive someone can discriminate angular displacements during reaching movements, with lower values indicating better performance.

The Kinarm Exoskeleton robot will be used to measure proprioception of the proximal upper limb joints (i.e., shoulder and elbow) during three uni- and bimanual proprioceptive assessments: (1) the contralateral position matching task, (2) the perceptual boundary task and (3) the indicating location task. The contralateral position matching task results in four parameters; absolute error, variability, systematic shift and contraction/expansion. For the first three parameters a higher value represents worse task performance. For the last parameters (contraction/expansion), a value of 0 indicates perfect alignment while a lower or higher score indicates contraction or expansion, respectively.

For the indicating location task, where participants will be asked to indicate the perceived location of the tip of their index finger, the outcome parameter is the number of correct responses. This score ranges from 0 to 24, with a higher score indicating better performance.

Distal proprioception of the metacarpophalangeal joint of the index finger will be investigated using a position sense task on the ETH MIKE robot. This results in two parameters of interest: the mean absolute error between the actual and the indicated position across the 11 trials and the mean variability over these trials. For both parameters a higher score indicates worse performance.

Tactile registration will be determined by identifying a threshold value using the Semmes Weinstein monofilaments. Tactile registration will be classified as normal (0.008-0.07 gram), diminished light touch (0.16-0.4 gram), diminished protective sensation (0.6-2 gram) and loss of protective sensation (4.19-300 gram).

Static two-point discrimination will be measured with the aesthesiometer by identifying the smallest distance that can correctly be sensed as two separate points. This score will range from 2 till 10 millimeters, with a higher score indicating worse performance.

Tactile perception will further be objectified using the Tactile Discrimination Test, which is a three-alternative forced choice response test where participants are guided to feel several triplets of textures and are asked to identify the texture that is different from the others. The final score will be transformed to an converted to an Area Under the Curve threshold score with a higher score indicating better performance.

Three different components of tactile perception will be assessed: (1) stereognosis or tactile object recognition, (2) two-point discrimination and (3) texture discrimination.Tactile object recognition will be assessed by identifying six familiar objects based on touch. The final score will be the total amount of objects that were correctly identified, ranging from 0 to 6 with a higher score indicating better performance.

Unimanual motor skills will be assessed using the instrumented Tyneside Pegboard Test, during which the time needed to transfer 9 medium-sized pegs from one board to an adjacent one is measured. A higher score, meaning a longer task duration, indicates worse unimanual motor skills.

The Canadian Occupational Performance Measure (COPM) will be used to assess the children's self-perceived occupation performance over different categories, including self-care, leisure and productivity. The parents or caregiver rates performance and satisfaction for each functional goal on a ten point ordinal scale, where 1 = "not able to do it all"/"not satisfied at all" and 10 = "able to do it extremely well"/"extremely satisfied".

Goal attainment will be objectified using the Goal Attainment Scale (GAS) for at least two individualized therapy goals. The GAS enables the conversion of goal attainment on a 5-point scale into t-scores. The score for each functional outcome ranges from -3 to 2. The participant's baseline performance is represented by a score of -2. Improvements in functional goal performance correspond to scores ranging from -1 to +2, with score 0 being the expected outcome. Deterioration in functional goal performance results in a score of -3.

The Child Behavioral Checklist (CBCL) will be completed by the parents to describe their children's behavioral and emotional problems, of which attentional functioning will primarily be considered in this study. The raw score on the attention sub scale ranges from 0 to 22, with a higher score indicating more attention difficulties.

Next, lesion location and extent will be evaluated in more detail using the semi-quantitative MRI (sqMRI) scale, showing a score for the brain damage in resulting in different global and sub regions of the brain, separately for both sides and different depths. For the total score and all sub scores, a higher score indicates more brain damage.

The imaging protocol consists of a T1-weighted anatomical image using three-dimensional T1-weighted turbo field echo with the following parameters: Echo time (TE)/Repetition time (TR) 4.2/9.1 milliseconds (ms), Field of view (FOV) 256 Foot/Head (FH) x180 Right/Left (RL) x242 Anterior/posterior (AP) mm³, 0.90x0.90x0.90 mm3 voxel size, acquisition time 3:31. In addition, 3D Fluid Attenuated Incovery Images will be acquired with the following parameters: TE/TR/ Inversion time (TI) 283/4800/1650 ms, FOV 250FHx200RLx250AP mm³, 1x1x1 mm³ voxel size, acquisition time 4:48. Also, a 3D T2-image will be collected with the following parameters: TE/TR 280/3000 ms, FOV 256FHx198RLx256AP mm³, 1x1x1 mm³ voxel size, acquisition time 4:48. The MRI classification system (MRICS) will be completed in order to classify the brain lesions according to the timing of the lesion and predominant lesion pattern, resulting in five main groups of lesion types.

Diffusion weighted images will also be acquired using the following parameters: single-shot spin-echo Echo Planar Imaging; slice thickness 2.3 mm, TE/TR 93/3765 ms, AP phase encoding direction with a SENSE acceleration factor of 1.5, FOV 136FHx240RLx240AP mm³, voxel size 2.22x2.22x2.2 mm³, acquisition time 8:04, 124 uniformly distributed diffusion directions. Implemented b values are 1000 and 2500 s/mm², applied in 50 and 74 uniformly distributed directions, respectively. Constraint spherical deconvolution will be applied to delineate the motor and somatosensory tracts for both hemispheres (e.g., corticospinal tract, thalamic radiations, medial lemniscus). This results in three parameters of interest, being fiber density, fiber cross-section and fiber density-cross-section.

Inclusion Criteria: Congenital or acquired, predominantly spastic unilateral cerebral palsy; Aged 7 to 15 years; Minimal ability to actively grasp and hold an object (modified House Functional Classification System ≥4); Having a confirmed impairment in tactile function as objectified using a clinical assessment battery containing assessments of tactile registration and tactile perception; and Sufficient cooperation to comprehend and complete the test and therapy procedures; Exclusion Criteria: Botulinum toxin-A injections six months prior to testing; A history of upper limb surgery one year prior to testing; Severe comorbidities hindering test assessments (e.g., severe cognitive problems); and Inability to communicate in Dutch.