OP-BN05 Gait I: Effect of home-based robotic-assisted gait training on spatiotemporal gait parameters in patients with chronic stroke

Introduction: A large majority of stroke survivors rank improving their ability to walk among their top rehabilitation goals (1). As health care delivery shifts away from traditional hospital settings and increasingly into the community (2), there is a growing need and interest in the efficacy of home-based rehabilitation interventions. Traditional gait rehabilitation therapies can be labor intensive, often requiring multiple therapists, however the capacity of robots to deliver training with high intensity and repeatability makes them a valuable tool (3). The evaluation of patients’ performance and progress throughout interventions is usually subjective because few adequate objective measures are available (4). This study aims to objectively assess the effectiveness of a home-based robotic-assisted gait training (RAGT) programme in patients with chronic stroke.

Methods: Forty stroke patients (age: 62.2±14.1y; time since stroke: 29.9±19.1months) were randomised into a RAGT programme or control group (CON) while continuing their regular rehabilitation therapy. Participants receiving RAGT performed 10-weeks of home-based rehabilitation using a lower limb robotic device (Alter G Bionic Leg, USA) and were required to wear the device for a minimum of 30mins per day. Robotic settings were adjusted every 2 weeks to induce progressive overload. Participants in the CON group were advised to engage in a minimum of 30mins of physical activity per day. Before and after the 10 week programme, spatiotemporal gait parameters (G-Walk, USA) were assessed during repeated 10m walks. Data was analysed using Two-way ANOVA (Time [Pre/Post] x Condition [RAGT/Con).

Results: The following results are from a preliminary cohort of participants (n=16). A significant Time*Condition interaction was observed for the stance phase (P<0.05; ƞp2=0.304), with the RAGT group demonstrating a significant reduction in the duration of the stance phase (pre: 67.6±8.2%; post: 59.3±6.1%), compared to CON (pre: 72.5±5.9%; post: 73.9±5.2%). Although non- significant, similar findings were observed for speed, with increases seen in the RAGT group (pre: 0.61±0.33m/s; post:0.81±0.38m/s), but not for CON (pre: 0.43±0.17m/s; post: 0.47±0.18m/s) (P=0.07; ƞp2=0.212). No significant interactions were seen for RAGT or CON for cadence or gait cycle duration (both P>0.05).

Conclusion: A 10-week home-based RAGT programme improved gait outcomes in patients with chronic stroke. Results showed a decrease in time spent in the stance phase, leading to a more typical non-pathological gait (60% stance, 40% swing), and an increase in speed. The present results demonstrate encouraging evidence supporting the use of RAGT in a home-based environment.