Feasibility and Outcomes of Gait Training with a Powered Exoskeleton in Individuals with Multiple Sclerosis

Background: Walking limitations are a common consequence of multiple sclerosis (MS), and have a major impact on the lives of affected individuals. There are a limited number of interventions for gait training with demonstrated effectiveness in MS, particularly for patients with severely limited ambulation. Powered exoskeletons (PES) are utilized for gait training in patients with spinal cord injury and post-stroke hemiparesis, but they are not indicated for MS in the United States. There is a gap in evidence to guide the use of PES in this patient population.

Objectives: In this study we sought to describe the parameters of gait training with a PES in an MS neurorehabilitation clinic, to assess the feasibility and safety of PES use for gait training with a physical therapist, and to analyze clinical outcomes.

Methods: We conducted a retrospective chart review of all adult patients with MS who participated in at least 3 sessions of gait training with a PES in our neurorehabilitation clinic between June 2019 and July 2021. Training descriptors, programmed PES parameters, and performance on clinical walking and mobility tests without PES were compared between the first and the last training session, using the paired t-test.

Results: 21 patients were included in the analysis (mean (SD) age 56 (10) years, 76% women, 81% progressive MS, 76% used a walker and 33% a wheelchair as their primary means of indoor mobility). The median number of sessions was 11 (IQR 6,22), with an mean (SD) of 517.7 (119.3) steps/session, and 21 (4) minutes of walk time per session. Between the initial and final gait training session, there was a statistically significant increase in walk time (p=0.008), number of steps (p=0.011), and programmed step length (p<0.001). Clinical outcomes were available on a subset of patients. Within-group analysis showed no significant change on the Timed 25 Foot Walk, with 3/15 (20%) patients exhibiting a greater than 20% increase in walking speed. There was a statistically significant increase in the time to complete the Timed Up and Go (p=0.029). One adverse event was reported in 5 patients.

Conclusions: Our retrospective analysis of real-world data suggests that gait training with a PES is safe and feasible in individuals with MS and severe walking limitations. A training progression was demonstrated between beginning and end of treatment. Change on clinical outcomes between first and last treatment session was highly variable between patients. Further studies are needed to determine optimal training paradigms and predictors of response to treatment.