FES in Young Children With Perinatal Stroke

Functional Electrical Stimulation to Improve Upper Extremity Function in Young Children With Perinatal Stroke: A Proof of Concept Study

Every year about 1 out of every 1,600-5,000 infants has a stroke around the time of birth. Many of these children will have lifelong physical problems. For example, the arm muscles are often paralyzed. This makes every day activities, like reaching and grasping objects, very difficult. To date there are few effective treatments for the paralyzed arm of young children with stroke. The main objective of this study is to test whether a new kind of treatment, known as functional electrical stimulation (FES), is able to improve arm function in children with stroke. FES involves applying electrical currents to weak or paralyzed muscles. This enables movements, such as reaching and grasping, which can then be practiced. The investigators will compare the effectiveness of FES treatment to standard arm rehabilitation in children aged 3-6 years who had a stroke early in life. They will measure the effectiveness using a number of clinical measures of arm function. Other objectives of this project are to test how well children adhere to the treatment schedule, and to evaluate parent and child satisfaction with FES treatment.

Primary Aim: To compare, in young children with perinatal stroke, the change in UE function between two treatment approaches: 1) FES treatment, and 2) conventional UE training. Secondary Aim 1: To compare the intensity (i.e., number of repetitions performed per treatment session) between FES treatment and conventional UE training. Secondary Aim 2: To compare adherence to the two treatment methods (FES treatment and conventional training). Secondary Aim 3: To evaluate parent and child satisfaction with FES treatment. This is a proof-of-concept, mixed methods, pilot study that will use a single-blind, randomized clinical trial design. The study will involve two sites: the University of Alberta in Edmonton, and the Toronto Rehabilitation Institute - University Health Network. The total study duration is three years, while the time commitment of each child is about 9 months. Participants will complete two baseline assessments spaced 1-2 weeks apart, and will then be randomized into one of two groups (FES treatment or conventional UE training). The treatment period will last 12 weeks, and will be followed by a 6 month follow-up period. Participants Twenty-four children (12 per treatment group) will participate. To reach the target of 24 children, a total of 30 children with perinatal stroke will be recruited. This allows for 20% attrition. As this is a pilot study, the sample size is based on a conservative estimate of the number of children the investigators anticipate being able to recruit and enroll over a three year period in the two participating cities. Participants will be recruited through numerous means, including two main clinical sites: the Glenrose Rehabilitation Hospital (Edmonton) and the Holland Bloorview Kids Rehabilitation Hospital (Toronto). Following recruitment, potential participants will be screened through a telephone interview by the Research Coordinator. Assessments Assessments of UE function and spasticity will be completed seven times throughout the study: pre-treatment (2 assessments spaced 1-2 weeks apart), after 18 treatment sessions, after 36 treatment sessions (i.e., the completion of treatment), 6 weeks post-training, 12 weeks post-training and 6 months post-training. The primary outcome measure will be the Melbourne Assessment 2 (MA2). Secondary measures will be the Assisting Hand Assessment (AHA), the Modified Tardieu Scale, and accelerometry to assess UE use in daily life. A single occupational or physical therapist at each site will complete the assessments. The assessing therapists will be blind to group allocation and the study objectives. The parents of children who received FES treatment will be asked to participate in a semi-structured interview within one week of their child completing FES treatment. The interview will be completed over the phone by one of the study investigators. The interview will last 45-60 minutes. The Researcher will ask a series of open-ended questions concerning the family's experience with, and impressions of, FES treatment. The data collected from the semi-structured interviews will allow us to gauge parent and child satisfaction with FES treatment. The interview will be digitally recorded and transcribed verbatim off-line. The transcribed dialogue will be shared with the parents, who will be asked to indicate whether it is an accurate reflection of the interview (i.e., member check). Treatment Methods Participants will be randomized into one of two treatment groups, FES treatment or conventional UE training, using a matched pairs design. The children will be matched in pairs based on their Manual Ability Classification System level. This randomization method, which has been used in other studies working with a similar population, will remove bias in the comparison of the two groups by ensuring that a potential confounding variable (i.e., UE impairment) is equally distributed across groups. A Research Assistant who is not involved in this study will randomize the participants by picking a label out of a box. Both treatment methods will be provided by a single physical or occupational therapist at each site. The dosage of treatment will be the same for the two treatment groups: 3 sessions/week with each session lasting 1 hour for a total of 36 sessions or hours of training. Participants will be asked to repeat the same task >10 times during a single treatment session. Treatment sessions will be video recorded at regular intervals (i.e., at the first session, and every 9 sessions thereafter) for two reasons. First the video data will allow the investigators to accurately quantify the intensity of each treatment method over time. The video will be viewed offline to document the number of repetitions performed of each type of grasp, direction of reach (forwards, lateral, across midline), etc. Second, the video data will be reviewed to ensure that the treatment being delivered adheres to the study protocol. FES treatment and conventional UE training will follow principles of motor learning and neuroplasticity, such as practicing many repetitions of salient tasks. The only difference between the two approaches will be that for the FES group, electrical stimulation will assist the children's movements. During a one hour training session, roughly half of the time will be spent focusing on unilateral activities with the affected UE (similar to CIMT), while the remaining time will be spent engaged in bimanual activities (similar to intensive bimanual training). The tasks practiced will be customized to the child's ability and goals, and will be embedded in play. Both proximal and distal muscles will be targeted, with no requirement of achieving proximal control prior to training distal movements. Previous work suggests that targeting impairments in the forearm, wrist and thumb with therapy also leads to improvements in proximal arm muscles. The treating therapist may provide manual assistance as needed to the child as he/she attempts a task. As motor control returns, the difficulty level of the tasks will be progressed by withdrawing assistance (manual or FES), and increasing the complexity of the tasks. If a child is able to complete a task 5-7 times without assistance and a near-normal movement pattern, the treating therapists will progress the task. The FES treatment group will receive electrical stimulation while participating in the functional training described above. At each session up to eight pairs of self-adhesive surface electrodes will be attached to the child's skin. A variety of toys will be used to keep children occupied during donning and doffing, which takes 5-10 minutes. If a child finds removing the electrodes uncomfortable, the electrodes can be dampened for easier removal. In adults, stimulation is more effective if it is combined with active movement. Thus, prior to facilitating a movement with FES, the children will be asked to attempt the movement themselves for about 10 seconds. After this time, FES will be triggered through a button pushed by the therapist to help execute the movement. Two commercially available FES devices will be used. One is the MyndMove FES System (MyndTec Inc., Mississauga, ON), which is a fully programmable FES device with eight channels. This system allows motor practice of 17 different reaching and grasping protocols. For example, children can practice pinch grip, lumbrical grasp, palmar grasp, lumbrical hand opening, tripod grasp, etc., with and without reaching. The second FES device that will be used is the Odstock 2-channel Stimulator (FES Mobility Ltd., Vancouver, BC). The Odstock device may be used if the child is practicing reaching movements while walking outdoors or over longer distances indoors. Throughout the training period and at the completion of training, the treating therapist will provide parents and children in both treatment groups with education on how to promote use of the affected arm in daily activities at home. Integrating improved UE function into family routines is believed to be important for the retention of motor skill and transfer to everyday environments. Parents/guardians will be given a diary, in which they can record 1) activities involving the weak arm that their child practices at home, including the approximate length of time in minutes, and 2) new activities involving the weak arm that they observe their child doing. Participants will wear two accelerometers (wGT3X-BT wireless activity monitor, Actigraph), one on each wrist, for the duration of every treatment session. This will enable us to record the amount or frequency of arm movements elicited with each type of treatment. Data Analysis Data from the accelerometers will be uploaded and analyzed using ActiLife 6 software (Actigraph). For each child, change scores on the clinical measures will be calculated as follows: score after 18 or 36 training sessions subtract the mean of the two baseline scores. A retention index will also be calculated for each child's performance on the measures as follows: the score at the 6-week or 12-week or 6-month follow-up divided by the score at the end of training (i.e., retention index of 1 = perfect retention). Group raw scores, change scores and retention indices on the outcome measures will be reported as mean + 1 standard deviation. For statistical tests, the alpha level will be set at 0.05. To test the assumption of normality, the Shapiro-Wilk test will be used. Non-parametric tests will be used when this assumption is not met. To address our primary aim, within-group and between-group changes on the primary and secondary outcome measures will be evaluated using a two-way repeated measures ANOVA, with the Bonferroni correction for post-hoc comparisons. The non-parametric Scheirer-Ray-Hare test will be used if appropriate. An intention-to-treat analysis will be used. Video footage of the treatment sessions will be reviewed to extract the total number of movement repetitions/session (intensity) for each child. Five sessions will be recorded/child (at the start of treatment and every 9 sessions thereafter). To compare the intensity of training between groups (Secondary aim 1), a two-way repeated measures ANOVA, or Scheirer-Ray-Hare test, will be used to compare the mean number of repetitions completed over the course of training between groups. The total number of hours trained (i.e., dosage) will be calculated for each child from the treating therapists' records. The percentage of children in each treatment group who complete >35 hours of training in 12 weeks will be reported. To compare the adherence to the two treatment methods (Secondary Aim 2), the total number of hours trained will be compared between groups using independent t-tests, or Mann-Whitney U tests. Following the member check (described above), the transcribed dialogue from the semi-structured interviews will be imported into NVivo 11 (QSR International) for data management, coding and analysis. An inductive, grounded approach will be taken for coding the interviews. The coding process will involve two researchers and one graduate student. First, the three researchers will independently review 3-4 interviews and take note of ideas and terms that reflect the content of the text. Next, the researchers will discuss the content and jointly develop a coding scheme (i.e., codes with definitions). This coding scheme will then be applied to the interviews by each of the researchers, with discrepancies being reconciled along the way. Lastly, together the researchers will group the identified codes into larger categories, and the relationships between these categories will be considered.

The MyndMove FES System (MyndTec Inc., Mississauga, ON) is a fully programmable FES device with eight channels. This system allows motor practice of 17 different reaching and grasping protocols. The muscles stimulated may include the anterior, middle and posterior deltoid, biceps and triceps brachii, flexor digitorum superficialis and profundus, extensor digitorum, thenar eminence, lumbricals, and interossei. The sophistication of this system is unique. In contrast, many of the MyndMove stimulation protocols involve activation of small muscles that are used for fine motor control. The second FES device that will be used is the Odstock 2-channel Stimulator (FES Mobility Ltd., Vancouver, BC). While less sophisticated than the MyndMove system, it is smaller and more portable.

Conventional training involves repetitive practice of upper extremity movements with manual assistance provided by a therapist as needed.

The Melbourne Assessment 2 (MA2) evolved from the original Melbourne Assessment of Unilateral Upper Limb Function. It assesses the quality of arm movement in children as young as two years of age. It consists of four subscales (range of motion, dexterity, accuracy, and fluency) and involves performing 14 tasks, such as reaching, grasping and manipulating objects. It is scored from a video recording or live observation of a child playing with standardized objects. The MA2 is valid, and has high interrater reliability and good test-retest reliability in children with neurological conditions. Published standard error of measurement values for the raw Melbourne scores were used to calculate the minimally important difference for each subscale. Calculated MID values were as follows: range of motion = 4, accuracy = 4, dexterity = 2, and fluency = 4.

2 baseline assessments, after 18 treatment sessions, after 36 treatment sessions, 6 weeks post-treatment, 12 weeks post-treatment, 6 months post-treatment

The Assisting Hand Assessment (AHA) was designed for use in children with a hemiplegic arm aged 18 months - 5 years. It measures how well children use their affected arm in bimanual activities, such as spontaneous handling of toys during play. Like the MA2, the AHA is routinely scored from a video recording. The validity, reliability and responsiveness of the AHA has been demonstrated in children with hemiplegic cerebral palsy. Raw AHA scores will be converted to a logit-based 0-100 scale in AHA units. Five AHA units represent the smallest detectable change.

2 baseline assessments, after 18 treatment sessions, after 36 treatment sessions, 6 weeks post-treatment, 12 weeks post-treatment, 6 months post-treatment

The Modified Tardieu Scale is a measure of spasticity that considers resistance to passive stretch of a muscle at slow and fast speeds. This scale has been used in children with cerebral palsy as young as 3 years of age, and has adequate interrater reliability in this population. To perform this assessment, the child's affected elbow and wrist will be passively moved through the available range of motion at slow and fast speeds. The angle of the 'catch' in the passive joint movement (i.e., R1), and the full passive range of motion (i.e., R2) will be noted. For this measure, a researcher will secure electrogoniometers over the affected elbow and wrist, and place electromyography recording electrodes over the triceps, biceps, wrist extensors and wrist flexors on the affected side. We expect this to increase the robustness of this clinical measure in children, and have previously found this instrumented approach to be useful.

2 baseline assessments, after 18 treatment sessions, after 36 treatment sessions, 6 weeks post-treatment, 12 weeks post-treatment, 6 months post-treatment

Accelerometry will be used to assess UE movement in daily life. Participants will be asked to wear a wGT3X-BT wireless activity monitor (Actigraph) on each wrist for seven consecutive days, with the aim of achieving a total of 24 hours of wear time (monitor removed during treatment sessions). Accelerometry is a valid method of assessing UE activity in adults with stroke and a MSc student in Dr. Musselman's lab is currently validating the tool in children with hemiparesis. Raw activity counts for the two arms will be collected, and the ratio of impaired to unimpaired UE activity will be calculated.

2 baseline assessments, after 18 treatment sessions, after 36 treatment sessions, 6 weeks post-treatment, 12 weeks post-treatment, 6 months post-treatment

Inclusion Criteria: be aged 3-6 years have experienced a unilateral perinatal stroke show focal brain injury on an MRI be rated a level II, III or IV on the Manual Ability Classification System by a physical or occupational therapist (this a 5-point ordinal scale with levels II, III and IV described as "Handles most objects but with somewhat reduced quality and/or speed of achievement", "Handles objects with difficulty; needs help to prepare and/or modify activities" and "Handles a limited selection of easily managed objects in adapted situations", respectively) be able to maintain an unsupported sitting position for >5 minutes. Exclusion Criteria: any other disease, injury or condition that affects their UE motor function contractures of the hand, wrist or elbow an implanted electronic device peripheral nerve damage in the affected UE surgical hardware in the UE a history of epilepsy a skin rash or wound at a potential electrode site received an injection of botulinum toxin to the UE within the past 6 months. Children entered into the study may be withdrawn at some point during the study if their skin is irritated by the FES treatment they begin a new physical or pharmacological intervention during the treatment period of the study, and it affects UE function.