Effectiveness of Motor Imagery Training for Lower Extremity Motor Function in Patients With Stroke: A Systematic Review and Meta-Analysis

Abstract

<h3>Objective(s)</h3> Motor imagery training (MIT) is considered an emerging and non-invasive rehabilitation technique that patients could easily master to improve functional outcomes. To evaluate the effectiveness of MIT combined with conventional rehabilitation therapy (CRT) in treatment of lower extremity motor functions (LEMFs) in patients with stroke. <h3>Data Sources</h3> PubMed, Embase, Cochrane Library, Web of Science, ClinicalTrials, Chinese Biomedical Literatures database, Chinese National Knowledge Infrastructure, Wan-Fang Database, and Chinese Scientific Journals Database were researched from the earliest records to December 1, 2020. <h3>Study Selection</h3> Randomized controlled trials (RCTs) that compared MIT associated with CRT versus a control CRT in patients with lower extremity motor dysfunction after stroke, were selected. Among the 781 records screened, 15 trials, in which motor imagination questionnaires clearly performed, were included with 761 eligible patients. <h3>Data Extraction</h3> Pooled outcome data were completed by two researchers independently. Primary evaluation index was the Fugl-Meyer assessment scale-Lower extremity (FMA-LE). The second evaluation indices were Functional Ambulation Category (FAC) and Berg Balance Scale (BBS). Meanwhile, the methodological quality of included studies was conducted via the Physiotherapy Evidence Database scale and the Cochrane risk of bias criteria tools. <h3>Data Synthesis</h3> Meta-analysis showed that there existed statistically significant improvements in FMA-LE (WMD=3.65, 95%CI=2.52, 4.78, P < 0.00001), FAC (WMD=0.80, 95% CI=0.22, 1.38, P=0.006) and BBS (WMD=6.15, 95% CI=2.98, 9.32, P=0.0001) in the MIT associated with CRT group versus the control group after an intervention of more than 3 weeks. <h3>Conclusions</h3> MIT associated with CRT had better effectiveness in LEMFs than CRT alone. <h3>Author(s) Disclosures</h3> None.