Effectiveness of robot-assisted gait training on motor impairments in people with Parkinson's disease: a systematic review and meta-analysis.

The aim of this systematic review and meta-analysis was to evaluate the effectiveness of robot-assisted gait training (RAGT) on motor impairments in people with Parkinson's disease (PD). A computer-based systematic literature search was performed in six databases according to PRISMA guidelines. Randomized controlled trials (RCTs) that assessed the effects of RAGT on motor impairments in people with PD were included. GRADE approach and PEDro scale were used to determine the studies' quality of evidence. Meta-analyses were performed by calculating the weighted mean difference (WMD) at 95% confidence interval. Seven RCTs (PEDro: 5-8) met the inclusion criteria for systematic review and meta-analyses. The meta-analysis showed significant improvement on Unified Parkinson Disease Rating Scale Part III after intervention [WMD=3.292; 95% confidence interval (CI)=1.378-5.207; P=0.000], and after 1-month follow-up (WMD=5.512; 95% CI=2.396-8.629; P=0.001). Stride length (WMD=9.283; 95% CI=7.153-11.414; P=0.00) and gait speed (WMD=0.166; 95% CI=-0.090 to 0.243; P=0.000) showed significant improvements after RAGT. Balance as measured by Berg Balance Scale was improved significantly after intervention (WMD=3.87; 95% CI=0.374-6.735; P=0.029) and at 1-month follow-up (WMD=3.87; 95% CI=1.324-6.413; P=0.002). The pooled analysis did not detect any significant changes regarding stride time, cadence and functional balance scales. GRADE level of evidence ranged between high and low. The RAGT showed better outcomes than conventional interventions on some motor aspects in PD. However, RAGT did not seem superior to control interventions. Further RCTs that examine the effect of RAGT on more specific outcomes and at different medication statuses are required.