This work aimed to update and summarize the existing evidence on the effectiveness of robot-assisted training (RAT) in adults with Parkinson's disease (PD). We conducted a systematic review with meta-analysis,reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (PROSPERO CRD42022371124). Seven databases and two trial registries were searched for randomized-controlled trials (RCTs) addressing RAT alone or in addition to other treatments in adults with PD up to January 2024. Primary outcomes were disease-specific motor impairment, balance, mobility, freezing of gait, falls, number of people who fell at least once, and adverse events. Meta-analysis using a random-effects model was performed. Risk of bias (RoB) and certainty of the evidence for the primary outcomes were assessed using the Cochrane RoB Tool and the Grades of Recommendation Assessment, Development, and Evaluation (GRADE) approach, respectively. Fifteen RCTs (629 randomized adults with PD) were included. Our results show that the evidence is very uncertain about the effectiveness of any kind of RAT, either focused on gait, balance or upper limb impairment, compared to any comparator (treadmill training, overground gait training, exercises without the exoskeleton, conventional physical therapy, balance training, and no treatment), mainly because of RoB, inconsistency in individual studies results, and very limited number (less than 200) of participants considered in each comparison. In light of the aforementioned very low certainty evidence, clinical considerations should be drawn very carefully. High-quality studies are thus highly needed to investigate potential benefits, risks, and cost/benefit ratio of RAT in adults with PD.
Read full abstract