Abstract
Background
Virtual reality-augmented therapist-delivered exercise-based training has promise for enhancing upper limb motor recovery after stroke. However, the neurophysiological mechanisms are unclear.
Objective
To find if neurophysiological changes are correlated with or accompany a reduction in motor impairment in response to virtual reality-aided exercise-based training.
Data sources
Databases searched from inception to August 2020: MEDLINE, AMED, EMBASE, PUBMED, COCHRANE, CINHAL, PROQUEST and OPEN GREY.
Eligibility criteria
Studies that investigated virtual reality-augmented exercise-based training for the upper limb in adults with stroke, and, measured motor impairment and neurophysiological outcomes. Studies that combined VR with another technology were excluded.
Data extraction and synthesis
Using pre-prepared proformas, three reviewers independently: identified eligible studies, assessed potential risk-of-bias, and extracted data. A critical narrative synthesis was conducted. A meta-analysis was not possible because of heterogeneity in participants, interventions and outcome measures.
Results
Of 1387 records identified, four studies were eligible and included in the review. Overall, included studies were assessed as having high potential risk-of-bias. The VR equipment, and control interventions varied between studies. Two studies measured motor impairment with the Fugl-Meyer Assessment but there was no commonality in the use of neurophysiological measures. One study found improvement in neurophysiological measures only. The other three studies found a reduction in motor impairment and changes in neurophysiological outcomes, but did not calculate correlation coefficients.
Conclusion
There is insufficient evidence to identify the neurophysiological changes that are correlated with, or accompany, reduction in upper limb motor impairment in response to virtual reality-augmented exercise-based training after stroke.
Systematic Review Registration Number PROSPERO 2017 CRD42017071312