Abstract T P116: Increased Maintenance of Motor Gains Following a Novel Stroke Therapy

Abstract

Introduction: Action observation therapy (AOT), a novel approach based on the physiology of the mirror neuron system, uses observation/imitation of movements to facilitate recovery from stroke. We previously showed that AOT confers significant improvement in motor performance. More importantly, these gains are maintained up to 1 year compared to a classic therapy, ND Bobath (NDBT) (Falcon et al 2013). In this study we sought to determine concomitant brain function associated with motor recovery after these therapies and with maintenance of motor gains. Methods: Seventeen stroke subjects received either AOT or NDBT for 2 hours/day for 4 weeks. Motor outcome measures and MRI (T1-weighted and fMRI during a bimanual motor task) were acquired pre-treatment (Pre-Tr), post-treatment (Post-Tr) and maintenance after 6-12 months (MiNT). MRI data was processed in AFNI using a general linear model. Network analysis assessing effective connectivity was done via structural equation models (SEM). MRI from 20 healthy controls was included. Results: 1. fMRI: Pre-Tr: Compared to controls, all stroke subjects showed increased activation, mainly in parietal and frontal regions. Post-Tr: Activation decreased in these regions in all stroke subjects. MiNT: Activation was further decreased in the AOT group, while the NDBT group showed increased activity reminiscent of Pre-Tr levels. 2. SEM: Pre-Tr: Both group networks involved nodes in parietal and frontal regions with strong connections in both hemispheres. Post-tr: The weight of connections in the non-dominant hemisphere after AOT decreased compared to Pre-Tr levels. After NDBT, it increased. MiNT: The weight of connections in the non-dominant hemisphere after AOT continued to decrease. Conclusions: Although decreased brain activation in both groups was associated with better motor performance Post-Tr, brain connectivity was discordant; effective connectivity decreased after AOT and increased after NDBT. This indicates differing mechanisms associated with stable (gains maintained) and poor (gains lost) recovery after AOT and NDBT respectively. These results may highlight relevant functional mechanisms for developing potential therapies.