Do powered over-ground lower limb robotic exoskeletons affect outcomes in the rehabilitation of people with acquired brain injury?

Purpose: To assess the effects of lower limb robotic exoskeletons on outcomes in the rehabilitation of people with acquired brain injury.Materials and methods: A systematic review of seven electronic databases was conducted. The primary outcome of interest was neuromuscular function. Secondary outcomes included quality of life, mood, acceptability and safety. Studies were assessed for methodological quality and recommendations were made using the GRADE system.Results: Of 2469 identified studies, 13 (n = 322) were included in the review. Five contained data suitable for meta-analysis. When the data were pooled, there were no differences between exoskeleton and control for 6-Minute Walk Test, Timed Up and Go or 10-Meter Walk Test. Berg Balance Scale outcomes were significantly better in controls (MD = 2.74, CI = 1.12–4.36, p = 0.0009). There were no severe adverse events but drop-outs were 11.5% (n = 37). No studies reported the effect of robotic therapy on quality of life or mood. Methodological quality was on average fair (15.6/27 on Downs and Black Scale).Conclusions: Only small numbers of people with acquired brain injury had data suitable for analysis. The available data suggests no more benefit for gait or balance with robotic therapy than conventional therapy. However, some important outcomes have not been studied and further well-conducted research is needed to determine whether such devices offer benefit over conventional therapy, in particular subgroups of those with acquired brain injury.Implications for RehabilitationThere is adequate evidence to recommend that powered over-ground lower limb robotic exoskeletons should not be used clinically in those with ABI, and that use should be restricted to research.Further research (controlled trials) with dependent ambulators is recommended.Research of other outcomes such as acceptability, spasticity, sitting posture, cardiorespiratory and psychological function, should be considered.