Early robot-assisted gait retraining in non-ambulatory patients with stroke: a single blind randomized controlled trial

BACKGROUND: Restoration of walking function is a primary concern of neurorehabilitation with respect to the aspired social and vocational reintegration. To date, the best practice for improving gait early after stroke is still object of debate. On one hand, repetitive task-specific approaches with higher intensities of walking have been observed to result in greater improvements of gait after stroke. Conversely there is some evidence that conventional gait training would be more effective for facilitating walking ability after stroke. AIM: To compare the effects of an early treatment protocol of add-on robot-assisted gait training with add-on conventional overground physiotherapy for improving locomotion in non-ambulatory adult stroke patients. DESIGN: Single-blind randomized controlled trial. SETTING: Neurorehabilitation hospital. POPULATION: Seventy-four subacute patients with first-ever ischemic stroke. METHODS: The patients were randomized into two groups. The training program consisted of forty, 2-hour sessions (including 45 minutes basic training, 45 minutes add-on training plus rest periods), 5 days a week, for 8 consecutive weeks. Patients allocated to the add-on robot-assisted gait training were treated by means of the Lokomat. Patients allocated to the add-on conventional overground gait training aimed at improving postural control during gait, body weight transfer, stability during the stance phase, free swing phase, adequate heel contact and gait pattern. Primary outcome was the modified Emory Functional Ambulation I'rofile. Secondary outcomes were the Riverniead Motor Index, the Mobility Milestones and the Hochzirl Walking Aids Profile. RESULTS: No significant difference was observed between groups with regards to age (P=0.661), time from stroke onset (P=41.413) and the primary outcome (P=0.854) at baseline evaluation. As to the primary outcome, no significant differences were found between groups at the end of the study. As During the 8-week training, within-group comparisons showed significant improvements of mean modified Emory Functional Ambulation Profile in both groups (P<0.001). CONCLUSIONS: Our results support the hypothesis that an early treatment protocol of robot-assisted gait retraining is not superior to add-on conventional gait training intervention for improving locomotion in non-ambulatory stroke patients. CLINICAL REHABILITATION IMPACT: This study might help to better understand the role of robot-assisted gait training in early phase stroke rehabilitation.