Mirror Box Therapy for Upper Limb Function With Stroke

Introduction. Stroke is a major cause of mortality in the United Kingdom with around 111,000 people per year being newly diagnosed. Of the survivors, 50% will be left with significant, long-term effects including residual deficits of the upper limb. Rehabilitation of upper limb impairment rests with the OT who plays a vital role in enabling stroke patients to self-manage their condition and live optimally independent lives. Mirror box therapy (MBT) is a relatively new therapeutic intervention that is gaining recognition within OT for the potential it offers in rehabilitation of upper limb function in stroke patients. Based upon mirror visual feedback originally used in the treatment of phantom limb pain after amputation, neural recovery in the brain can be stimulated using mirrored movements of the non-affected upper limb. It is thought that visual feedback helps recruit dormant motor pathways that replace the damaged pathways and encourage the return of movement, thus mirror box therapy is thought to improve upper limb function through both movement and mental stimulation. However, existing studies are limited due to non- consistent participant post-stroke delay (ranging from 3-12 months). Few studies have included patients in the sub- acute period post stroke (0-3 months), a population considered most likely to benefit from this therapy at the early recovery stage.

Aims and Objectives. The aim of this pilot study is therefore to explore the feasibility of conducting a fully powered randomized controlled trial of mirror box therapy for upper limb rehabilitation within a sub-acute stroke population.

The objectives of the study are to:

1. Evaluate the feasibility of patient recruitment within an in-patient sub-acute single setting;
2. Assess the feasibility of delivering MBT as a component of OT treatment in the sub-acute in-patient population;
3. Evaluate the sensitivity of the outcome measures for use in a fully powered trial and conduct a power calculation;
4. Conduct a preliminary analysis of the data to identify potential treatment gains within and between the 2 groups;
5. Pilot the collection of data to enable cost-consequence analysis to be undertaken as an output of the main RCT.

Sample: 50 participants will be recruited and randomized into two groups (treatment n=25; control n=25) over a 2 year period of 1 January 2015 - 31st Dec 2016.

Sample Size Justification: This sample size will allow us to estimate a standard deviation for the primary outcome, and allow us to estimate participation rate with a precision of +/-12.5% if, as we expect, the rate is in the vicinity of 75%.

Randomization: Block randomization will be undertaken using a computer generated randomization list. Each block is estimated to run over a 16 week period. This will allow for recruitment of between 1-2 new subjects per week and assumes an average inpatient stay of 6 weeks. Group allocation will be concealed in consecutively numbered, opaque sealed envelopes.

Intervention: Participants in both groups shall receive their standard OT treatment for upper limb rehabilitation for the duration of their in-patient stay, which is 3-5 sessions per week of approximately 45 minutes duration. This classic rehabilitation treatment is based upon neurodevelopmental theory using the Bobath approach of postural control and repetitive task training. Participants in the treatment group will be additionally required to perform two 20-minute sessions of mirror box therapy, five days/week for the duration of their in-patient stay. Also based upon neurodevelopmental theory, this treatment creates the illusion of perfect bilateral synchronization of repetitive task training by concealing the affected arm in a mirrored box that reflects the repetitive upper arm movements conducted by the unaffected limb.

Control Group Intervention: Participants will receive standard Occupational Therapy intervention for this population in the sub-acute rehabilitation setting, delivered by members of the OT stroke team. This follows the documented protocol used within the Health and Social Care Trust and progresses through 8 phases from assisted to unassisted movements, gross upper limb movements to wrist and fine finger movement, using remedial and functional activities as well as ward-level rehabilitation.

Treatment Group: Participants in the intervention group will be required to perform two 20 minute sessions of MBT, five days/week for the duration of their in-patient stay carried out under the direction of members of the OT stroke team. Sessions will be conducted at the patient's bedside or in the OT Department. Participants will be seated in a comfortable high chair and positioned in front of an adjustable height table. The mirror box will be positioned on the table in front of the participant. The participant will place or be assisted by the therapist to place the affected arm into the open end section of the nylon box; the mirror section will face the patient's non affected side.

Follow-up data period: We will follow up the initial blocks at both 3 and 6 monthly intervals in order to collect longer term data for use on sustained functional gain as well as for use in economic analysis. We will attempt to follow up as many subjects as possible in the latter blocks at the 3 \& 6 monthly intervals.

Analysis: Participation rates to both the complete set of assessments, and to the paired baseline and discharge assessments will be estimated and reported. If compliance to the complete set is similar to compliance with baseline and discharge, then assessment every two weeks will be considered for the main trial. However, if compliance with baseline and discharge falls below 60% this will question the value of conducting a larger study. Differences from baseline at discharge will be analysed using ANCOVA, with baseline assessment as the covariate. The upper 90% limit of the estimated sd will be used in future power calculation. The data from multiple assessments will be analysed using repeated measures ANOVA, and the estimated within and between patient sd used in future calculation. Confidence intervals will be presented for treatment effects, and the upper 95% limits used to inform future planning. The qualitative analysis of the patient exit questionnaire will consist of thematic analysis and synthesis.

Economic Analysis: If the subsequent main RCT demonstrates effectiveness of MBT, then analysis of relative costs and outcomes of the intervention will be demonstrable through a cost-consequence analysis using cost and outcome data gathered through the EQ-5D-5L and information relating to discharge destination and discharge care plans.

Ethics and Data Protection: Ethical approval will be obtained from ORECNI and full research governance approvals before commencement of this project.