Two-Month Individually Supervised Exercise Therapy Improves Walking Speed, Step Length, and Temporal Gait Symmetry in Chronic Stroke Patients: A before-after Trial

Abstract

Gait asymmetry is common after stroke and is a major risk factor for falls. In particular, temporal gait asymmetry often remains in the chronic stage of stroke. However, health insurance does not cover rehabilitation for patients with chronic stroke in many countries. Accordingly, it is undetermined whether individually supervised exercise therapy has beneficial effects on chronic hemiparetic gait. Patients with stroke (n = 25) more than 6 months after onset performed 70 min of individually supervised exercise twice weekly for 2 months in 16 sessions with qualified personnel. The intervention significantly reduced the pre-swing phase on the paretic side (mean = 91.8%, 95%CI, 84.8−98.8). In addition, there was a significant improvement in pre-swing phase symmetry in those with great asymmetry prior to the intervention (p = 0.022). Step length significantly increased after the intervention on both sides (non-paretic, p = 0.029; paretic, p = 0.0055). Walking time at both comfortable and maximum speeds was significantly shortened (comfortable, p = 0.0041; maximum, p < 0.0001). Our findings suggest that there remains scope to improve gait ability with individually supervised exercise therapy in patients with chronic stroke, whose functional recovery is often considered unlikely. This type of intervention may be a simple and effective option to improve gait parameters, including temporal asymmetry, even in patients with chronic stroke.

Keywords: chronic phase; exercise therapy; gait asymmetry; hemiparesis; stroke.

Figure 1

Flow chart of inclusion and exclusion process.

Figure 2

Pre- and post-intervention comparison of gait parameters during comfortable walking speed: (a) comfortable walking speed, (b) length of a gait cycle, (c) non-paretic step length, and (d) paretic step length. The connected plots show the pre- and post-intervention data of a subject. Bar graphs show the mean value. * p < 0.05, ** p < 0.01, error bar = SD, n = 25.

Figure 3

Proportion of gait cycle prior to the intervention and changes in gait cycle after the intervention. (a) Illustration shows the gait cycle (stance phase and swing phase). The pre-swing phase (double support phase) is the time spent with both feet in contact with the floor. The graph shows the pre-intervention gait cycle (mean ± SD, * p < 0.05, **** p < 0.0001, n = 25). (b) Percent change in gait phase after the intervention. * p < 0.05, n.s. not significant; Error bar = 95%; confidence interval, n = 25. (c) Fold change in pre-swing asymmetry (paretic/non-paretic) in a subgroup with >1.1-fold asymmetry at the baseline (pre-intervention). The connected plots show the pre- and post-intervention data of a subject. Bar graphs show the mean value. * p < 0.05, error bar = SD, n = 15.

Figure 4

Pre-/post-comparison of 5 m maximum-speed walking time. Each plot shows the data of a subject, and the pre- and post-intervention data are connected by lines. The bar graph shows the mean value. The graph on the right is a magnified view of the area enclosed by the dotted line. **** p < 0.0001, error bar = SD, n = 25.