RESEARCH REPORT PLATFORM PRESENTATION

DOES MIRROR THERAPY IMPROVE UPPER LIMB SENSORY AND MOTOR RECOVERY EARLY POST-STROKE? A RANDOMISED-CONTROLLED TRIAL. Acerra N^1,2, Souvlis T¹, Brauer S¹, Moseley L^1-3; ¹Division of Physiotherapy, The University of Queensland, Brisbane, Australia. ²Physiotherapy Department, Royal Brisbane and Women’s Hospital, Brisbane, Australia. ³Oxford Centre for fMRI of the Brain & Department of Physiology, Anatomy &

PURPOSE: The current study assessed mirror therapy versus sham therapy on early post-stroke upper limb resting pain and sensorimotor recovery. RELEVANCE: During mirror therapy participants perform bilateral limb movements and watch the unaffected hand and its mirror image, while keeping the affected hand hidden from view behind a partition. Mirror therapy is hypothesised to either (i) provide sensory feedback important for sensorimotor integration and recovery, or, (ii) activate the ipsilateral motor cortex to enhance sensorimotor recovery. Mirror therapy was associated with improved motor recovery in a chronic (>6 months) post-stroke population and improved sensation, function and pain intensity in pain patients. However, mirror therapy has not been evaluated to determine its effect on early post-stroke pain and sensorimotor recovery. PARTICIPANTS: Forty acute (<two weeks) stroke patients were recruited from an acute rehabilitation hospital. To increase patient homogeneity, only first-time ischaemic stroke patients participated. All participants met study inclusion/exclusion criteria. METHODS: Blinded assessments included a motor assessment (grip strength and the Motor Assessment Scale (MAS)) and a quantitative sensory assessment (resting pain intensity, sensory detection, pressure-pain threshold, and hot/cold sensation). Participants were assessed prior to treatment, following treatment and one month post-treatment. Participants were randomly assigned into either mirror-therapy or sham-therapy (non-reflective surface) for 20 minutes daily for two weeks, in addition to usual care. Both groups performed bilateral (as able) simple motor tasks and sensory stimulation progressed by complexity. Participants in the intervention group watched the mirror image of the unaffected upper limb during intervention, with the affected limb hidden from view. Sham group participants were unable to view the unaffected limb during intervention. ANALYSIS: Results were analysed on an intention to treat basis. A RM-ANOVA was used to detect differences between groups (2), times (3), and sites (4). RESULTS: There were no between-group differences at baseline. Subjects who underwent mirror therapy demonstrated at each of the three upper limb sites: improved sensory detection (post: p < 0.05, follow-up: p < 0.01), pressure-pain threshold (post: p < 0.05, follow-up: p < 0.01), and hot/cold sensation (post: p < 0.05, follow-up: p < 0.01); improved grip strength (post: p < 0.05, follow-up: p < 0.01) and MAS scores (post: p < 0.05, follow-up: p < 0.01); and decreased resting pain intensity (post: p < 0.01, follow up: p < 0.01) compared with the sham group post-treatment and at follow-up. There was no statistically significant change in sensation at the remote (leg) site. CONCLUSIONS: Mirror therapy was associated with improved early post-stroke upper limb motor recovery, sensory recovery, and resting pain intensity. Mirror therapy had no effect on a remote site in the leg, suggesting that mirror therapy has a local treatment effect. Further research is needed to determine the optimum frequency and duration of mirror therapy and if it is equally effective when un-supervised. IMPLICATIONS: This is the first known RCT of mirror therapy in a post-stroke population. The results of this study support the use of mirror therapy in an acute stroke unit. Mirror therapy was well received by stroke patients and is a simple and cost-effective intervention that can improve the rehabilitation outcome. KEYWORDS: Stroke rehabilitation, somatosensation, sensorimotor integration. FUNDING ACKNOWLEDGEMENTS: NEA is supported by NHMRC Grant ID 409919 and The Royal Brisbane & Women’s Hospital Research, Australia. The current research was supported by an APA Neurology-Tagged Grant. GLM is supported by a Nuffield Oxford Medical Fellowship and is on leave from the School of Physiotherapy, The University of Sydney, Australia. CONTACT: nicoleacerra@hotmail.com

ETHICS COMMITTEE: The Royal Brisbane and Women’s Hospital Human Research Ethics Committee and The University of Queensland Medical Research Ethics Committee.