File
S.G. Lazzarini1, S. Campagnini2,3, C. Cordani4, C. Kiekens5, M.C. Carrozza2, S. Negrini4,6, C. Arienti1
1IRCCS Fondazione Don Carlo Gnocchi, Milan, Italy, 2Scuola Superiore Sant’Anna, The Biorobotics Institute, Pisa, Italy, 3IRCCS Fondazione Don Carlo Gnocchi, Florence, Italy, 4IRCCS Istituto Ortopedico Galeazzi, Milan, Italy, 5IRCCS MultiMedica, Milan, Italy, 6University La Statale, Department of Biomedical, Surgical and Dental Sciences, Milan, Italy
Background: Considering the growing evidence in stroke rehabilitation and the rapid development of robot-assisted training (RAT) devices, there is a need to signpost the rehabilitation stakeholders to more detailed evidence on RAT for stroke rehabilitation.
Purpose: The purpose of our overview of systematic reviews (SRs) was to summarise existing evidence on RAT in adults with stroke and signpost readers to the best quality evidence available.
Methods: We conducted an overview of SRs, including all SRs that addressed the effectiveness of RAT alone or in addition to any other intervention, compared with other interventions, for improving motor function and reducing motor impairments in adults with stroke in any stage. Eight electronic databases were searched up to December 2021. We considered SRs published in the last five years to include up-to-date evidence and try to limit the degree of overlap. The methodological quality of included reviews was assessed using AMSTAR-2. We used the WHO-ICF framework to summarise the outcomes. The degree of overlap was calculated using the Corrected Covered Area. We extracted the GRADE certainty of the evidence for each relevant comparison and outcome as reported within systematic reviews, or if this was not provided, we assessed this ourselves. PROSPERO registration number: CRD42020160859.
Results: We included 22 SRs (207 primary studies and 8913 participants), in which all participants were adults with stroke in all stages, and analysed 96 comparisons. 36% of them showed high or moderate certainty of evidence.
Our findings show that RAT alone improves muscle power functions, while we found discordant results on movement functions, hand and arm use and fine hand use and on activities and participation in adults with stroke in any stage immediately after the treatment (high certainty of evidence). There is probably a promising effect of RAT alone on maintaining body position and gait patterns (stride length and temporal symmetry) (moderate certainty of evidence). Moreover, RAT associated with physiotherapy improves walking (in terms of independent walking) (high certainty of evidence) and probably improves maintaining a body position when associated with body weight support and conventional therapy (moderate certainty of evidence). None of the effects lasts at follow-up. No study reported serious adverse events and neither RAT alone nor probably additional RAT increases the number of dropouts, thus it seems to be safe for people with stroke. We found moderate overlap across the reviews and very high overlap between some outcomes.
Our findings show that RAT alone improves muscle power functions, while we found discordant results on movement functions, hand and arm use and fine hand use and on activities and participation in adults with stroke in any stage immediately after the treatment (high certainty of evidence). There is probably a promising effect of RAT alone on maintaining body position and gait patterns (stride length and temporal symmetry) (moderate certainty of evidence). Moreover, RAT associated with physiotherapy improves walking (in terms of independent walking) (high certainty of evidence) and probably improves maintaining a body position when associated with body weight support and conventional therapy (moderate certainty of evidence). None of the effects lasts at follow-up. No study reported serious adverse events and neither RAT alone nor probably additional RAT increases the number of dropouts, thus it seems to be safe for people with stroke. We found moderate overlap across the reviews and very high overlap between some outcomes.
Conclusions: The high degree of overlap between the results suggests research waste and requires cautiously interpreting the findings. Further high-quality research will have to identify new ways of evidence synthesis to avoid producing overwhelming numbers of SRs that probably confuse rather than support clinical decisions related to the usability of RAT in stroke rehabilitation.
Implications: Robot-assisted training may be considered an option for the rehabilitation of adults with stroke at the end of the intervention, used alone or in combination with other interventions, but further research is needed to reduce discrepancies and improve the certainty of the evidence.
Funding acknowledgements: The authors declare no conflict of interest except for Prof. Carrozza, who has interests in organizations with financial interests.
Keywords:
Robot-assisted training
Stroke
Rehabilitation
Robot-assisted training
Stroke
Rehabilitation
Topics:
Innovative technology: robotics
Neurology: stroke
Research methodology, knowledge translation & implementation science
Innovative technology: robotics
Neurology: stroke
Research methodology, knowledge translation & implementation science
Did this work require ethics approval? No
Reason: The work is an overview of systematic reviews and thus does not require ethics approval because it is a secondary paper based on the results of other papers and not directly working on humans.
All authors, affiliations and abstracts have been published as submitted.
