Robotic treatment of the upper limb in chronic stroke and cerebral neuroplasticity: a systematic review

Stroke is the second cause of mortality and the third cause of long-term disability worldwide. Deficits in upper limb (UL) capacity persist at 6 months post-stroke in 30-66% of hemiplegic stroke patients with major limitations in activity of daily living (ADL), thus making the recovery of paretic UL function the main rehabilitation goal. Robotic rehabilitation plays a crucial role since it allows to perform a repetitive, intensive, and task-oriented treatment, adaptable to the patients' residual abilities, necessary to facilitate recovery and the rehabilitation of the paretic UL. It has been proposed that robot-mediated training may amplify neuroplasticity by providing a major interaction of proprioceptive and/or other sensory inputs with motor outputs, with significant modifications in functional connectivity (coherence) within the fronto-parietal networks (inter- and intra-hemispheric functional connectivity) related to processes of movement preparation and execution. However, the neurophysiological mechanisms underlying this reorganization are not entirely clear yet. Therefore, the aim of this study is to revise the literature, which assesses the effect of robotic treatment in the recovery of UL deficits measured in terms of neuroplasticity in patients affected by chronic stroke. This systematic review was conducted using PubMed, PEDro, Cinahl (EBSCOhost), Scopus and Cochrane databases. The research was carried out until February 2020 it included articles written in English language, published between 2009 and 2020, and the outcomes considered were neuroplasticity assessments. We included 23 studies over 6145 records identified from the preliminary research. The selected studies proposed different methods for neuroplasticity assessment (i.e. transcranial direct current stimulation (tDCS), EEG-Based Brain Computer Interface (BCI) and Neuroimaging (fMRI)), and different Robotic Rehabilitation treatments. These studies demonstrated a positive correlation between changes in central nervous circuits and post-treatment clinical outcomes. Our study has highlighted the effectiveness of robotic therapy in promoting mechanisms that facilitate re-learning and motor recovery in patients with post-stroke chronic disabilities. However, future studies should overcome the limitations of heterogeneity found in the current literature, by proposing a greater number of high-level RCTs, to better understand the mechanisms of robot-induced neuroplasticity, follow the clinical progress, estimate a prognosis of recovery of motor function, and plan a personalized rehabilitative programme for the patients.