Exploring the Impact of Backward and Forward Locomotor Treadmill Training in Chronic Stroke Survivors with Severe Post-Stroke Walking Impairment: A Single-Center Pilot Randomized Controlled Trial.

Abstract

Background: Defined as a self-selected speed of <0.4 m/s, chronic stroke survivors falling in this category are classified as "severe", usually homebound and sedentary, and they experience worse outcomes. Limited rehabilitation strategies are available to improve walking speed and related outcomes in this subgroup, and questions regarding effective rehabilitation options remain. The objective of this study was to determine the effects of backward (BLTT) and forward (FLTT) locomotor treadmill training on overground walking speed, spatiotemporal symmetry, and dynamic postural stability.

Methods: In this single-center, assessor-blinded, randomized controlled pilot trial, 14 stroke survivors with severe waking impairment underwent 12 sessions of BLTT (n = 7) or FLTT (n = 7). The primary outcome was the proportion of participants reaching clinically meaningful important difference (MCID) on the 10-meter walk test following training completion. Secondary outcomes were between-group differences in walking speed, spatiotemporal symmetry, and completion time on the 3-meter timed up and go (3M TUG) at 24 h, 30 days, and 90 days POST.

Results: Two subjects in the BLTT group (28.6%) and one (14.3%) in FLTT achieved MCID following training; however, most subjects did not, with significant variability in response. At 24 h POST, the median (IQR) percent change in walking speed was 28.9 (9.01-36.7) and 17.4 (12.6-39.7) with BLTT and FLTT, respectively; however, no between-group differences were seen (p = 0.80) at this time point or at 30 (p > 0.99) and 90 (p > 0.99) days follow up. Likewise, there were no significant between-group differences in spatiotemporal symmetry and the 3M TUG across time points.

Conclusions: While preliminary, this study found that 12 training sessions did not lead to group-level achievement of MCID for walking speed in our cohort and found no significant between-group differences in walking capacity or dynamic postural stability. Future well-powered dosing trials and mechanistically driven studies are needed to optimize and identify predictors of training response.