Examining the effect and clinical relevance of targeted memory reactivation during sleep for stroke rehabilitation interventions

Stroke poses significant healthcare costs and leaves patients with significantly lowered quality of life (Department of Health, 2014). Movement impairments are a major contributor to long-term disability. Commonly used movement therapies are time-consuming and expensive and yield variable outcomes. Thus, there is growing interest in development of more effective and scalable treatment of movement impairment. In particular, directly targeting the underlying neural mechanisms involved in rehabilitation might be predicted to result in better outcomes.
Stroke rehabilitation requires motor learning. Motor learning depends not only on performance gains made during practice but also on consolidation - performance gains after practice. Such memory consolidation is a key function of sleep and is associated with neural reactivation of the movement representation. Approaches are emerging that target processes of motor consolidation directly during sleep.
One such approach is targeted memory reactivation (TMR). TMR involves repeatedly pairing a memory, such as button presses, with sensory cues, such as sounds, during initial learning. If the sensory cues are re-presented during subsequent slow wave sleep this elicits reactivation of the motor representation and thereby enhances behavioural consolidation (i.e. better performance of a learnt motor sequence after a night of TMR compared to control auditory cues).
However, although a large-scale meta-analysis suggests robust effects of TMR in the healthy brain (Hu et al., 2020), TMR has yet to be tested in a clinical context. I propose to test whether TMR is effective in improving motor consolidation after stroke. I also aim to take TMR out of the laboratory and into patients' homes by adapting low-cost portable technology. In my current role as an RA at NDCN, I am adapting the Dreem headband for this purpose, in collaboration with Tim Denison.
Research Plan
Stage 1: Pilot at-home TMR in healthy populations (n=30) to establish optimal protocols for boosting memory consolidation. Participants will first train at home on a motor sequence task, where movement elements are paired with auditory cues. During the night, participants will wear the EEG headband and will be delivered the same auditory cues (TMR group) or different cues (control group), during slow-wave sleep. Behavioural outcomes (reaction time) and transfer to untrained tasks will be measured.
Stage 2: Test of feasibility of home-based TMR system in stroke patients (n=10) using similar design to Stage 1 but with a reaching task that is more relevant to rehabilitation. Semi-structured interviews and questionnaires will assess feasibility and inform study modifications.
Stage 3: Test of TMR as an adjunct therapy for stroke patient rehabilitation (n=40). Patients will perform the motor training from Stage 2 daily for 9 days with overnight EEG plus TMR/control auditory cues. We will compare overnight motor consolidation, and changes in clinical scores, for TMS versus control group.
Conclusion
This project will provide evidence on effects of home-based TMR as an adjunct to motor training in stroke survivors. This will help inform larger-scale RCTs that could consider efficacy and cost-effectiveness of a novel adjunctive rehabilitative therapy for motor impairment after stroke.