H. Nedergård1, A. Arumugam2, M. Sandlund1, A. Bråndal1, C.K. Häger1
1Umeå University, Department of Community Medicine and Rehabilitation – Physiotherapy Section, Umeå, Sweden, 2College of Health Sciences, University of Sharjah, Department of Physiotherapy, Sharjah, United Arab Emirates
Background: Robotic-Assisted Gait Training (RAGT) may enable an intensive, task-specific and repetitive gait training post-stroke. Earlier reviews have shown that RAGT, in combination with conventional physiotherapy, has a similar or slightly better effect on ambulation when compared to conventional gait training alone. However, there is no comprehensive review yet to summarize the effects of RAGT on objective biomechanical measures of gait post-stroke.
Purpose: To summarize the level of evidence for the effects of RAGT on gait movement patterns, quantified with objective biomechanical measures, in persons post-stroke.
Methods: PubMed, Web of Science, CINAHL, AMED, Academic Search Elite, SPORTDiscus Scopus, ProQuest (Sports Medicine & Education Index) and Cochrane databases were searched using database-specific search terms from their inception to 01st May 2020. Only randomized controlled trials (RCTs) investigating the effects of RAGT (e.g. using exoskeletons/end-effectors), employed as a monotherapy or in combination with other therapies, on spatiotemporal, kinematic and kinetic parameters among adults suffering from any stage of stroke were included. RCTs published as peer-reviewed full-text papers in English were included. Screening, data extraction and judgement of risk of bias (using the Cochrane Risk of bias 2 tool) were performed independently by two reviewers. The certainty of evidence for biomechanical measures of interest was evaluated using the Grading of Recommendations Assessment Development and Evaluation (GRADE) criteria.
Results: Twelve studies including a total of 393 individuals (mean age: 52 to 69 years; 249 males) in a subacute or chronic phase post-stroke were found. The intervention groups received RAGT as either a monotherapy (n=10) or combined with non-robotic gait training methods (n=2) like those received by the comparator group. Biomechanical gait analysis was performed using a gait mat (n=7), a 3D motion capture system (n=3), an accelerometer (n=1), or an in-shoe plantar pressure analyzer (n=1). While all included studies analyzed at least one temporal gait parameter, less than half of them reported spatial or kinematic parameters. Kinetic parameters were not reported at all. There was a high risk of bias (n=5), some concerns (n=6) or a low risk of bias (n=1) associated with the included studies. Meta-analyses using a random-effects model on gait speed and cadence revealed insignificant mean differences between the RAGT and comparator groups (statistical heterogeneity, I2≥85%). Due to serious weaknesses with almost all GRADE domains (risk of bias, consistency, directness, or precision of the findings) across studies for biomechanical measures, the certainty of evidence found was very low for RAGT post-stroke to improve gait biomechanics.
Conclusion(s): The twelve included studies using different RAGT parameters and biomechanical measures revealed a very low certainty in current evidence for employing RAGT to improve gait post-stroke. Thus, further high-quality, robust trials on RAGT that complement clinical data with biomechanical data are warranted to substantiate the effects of such interventions on gait biomechanics.
Implications: This review revealed a lack of significant effects of RAGT on gait biomechanics in persons post-stroke, primarily because of (clinical) heterogeneity of participants, intervention parameters and outcome measures, based on a very low certainty of evidence.
Funding, acknowledgements: Umeå University Foundation for Medical Research; Region Västerbotten; The Foundation for Stroke Research in Norrland; King-Gustaf-V and Queen-Victoria’s Foundation-of-Freemasons
Keywords: Stroke, Robotics, Gait biomechanics
Topic: Innovative technology: robotics
Did this work require ethics approval? No
Institution: N/A
Committee: N/A
Reason: This study is a systematic review and meta-analysis
All authors, affiliations and abstracts have been published as submitted.
