IMPACT OF GAIT PARAMETER BASED OPTIMAL STIFFNESS SETTING OF ANKLE-FOOT ORTHOSIS ON GAIT IN PATIENTS WITH STROKE

M. Yamamoto^1,2,3, Y. Ishige¹, K. Shimatani⁴, Y. Murakami⁵, D. Matsuura⁵, H. Takemura¹

¹Tokyo University of Science, Faculty of Science and Technology, Noda, Japan, ²Hiroshima University, Graduate School of Advanced Science and Engineering, Higashi-Hiroshima, Japan, ³Fukuyama Memorial Hospital, Department of Rehabilitation, Fukuyama, Japan, ⁴Prefectural University of Hiroshima, Faculty of Health and Welfare, Mihara, Japan, ⁵Brain Attack Center Ota Memorial Hospital, Department of Rehabilitation, Fukuyama, Japan

Background: Ankle–foot orthosis (AFO) is one of the representative treatment methods for stroke rehabilitation. AFO improves gait speed and kinematics parameters in patient with stroke. AFO with plantarflexion resistance (PFR) mainly assists rocker function and weight acceptance on post stroke gait as well as improve foot clearance. One of the important AFO characteristics is magnitude of stiffness. Varying the magnitude of PFR stiffness of AFO influenced ankle joint angle and moment. Therefore, magnitude of PFR should be adjusted to each patient with stroke. However, the effect of stiffness magnitude on post stroke gait is still unclear. In particular, optimal adjusted stiffness based on temporo-spatial and kinematics parameters on gait function of patient with stroke has not been investigated sufficiently.

Purpose: We aimed to investigate the effect of gait parameter based optimal PFR stiffness on post-stroke gait.

Methods: Nineteen stroke patients (67.7 ± 9.2 years) participated in this study. The participants were asked to walk with self-selected speed under five conditions: no-AFO condition and AFO with four different PFR stiffness conditions (PFR1 to 4). PFR1 was most flexible condition, whereas PFR4 was most rigid condition. Single RGB camera was positioned on the participants’ affected side for 2D pose estimation system using OpenPose. Openpose estimates body keypoints by two branch multi-stage convolution neural network. Peak lower limb joint angles, gait speed, and stride during gait in each condition were calculated by the Openpose body keypoints. One-way repeated measures analysis of variance was used to compare these gait parameters in the five conditions and best condition, followed by Shaffer’s modified sequentially rejective Bonferroni procedure as a post-hoc test. The best condition was defined as the condition with fastest gait speed or highest cross-correlation coefficients with lower limb joint angles during gait on healthy subjects among the five conditions for each participant. Statistical significance was set at p < 0.05.

Results: Gait speed in the best condition (0.60 ± 0.24 m/s) significantly increased compared with no-AFO, PFR1 and PFR2 condition(0.50 ± 0.26 m/s, 0.53 ± 0.23 m/s, and 0.54 ± 0.22 m/s, respectively). In addition, stride in best condition (0.87 ± 0.21°) significantly increased than that of PFR1 condition (0.79 ± 0.21°). In contrast, most of kinematic parameter in the best condition were not significantly different compared with other the five conditions. Dorsiflexion angle of initial contact in PFR3 condition (-0.4 ± 6.2°) significantly increased compared with PFR1 condition (-2.0 ± 6.2°).

Conclusions: Varying magnitude of PFR influenced lower limb joint angles, whereas the results also suggested that optimal PFR settings such as the best condition improved temporo-spatial parameters compared with other different stiffness conditions. Magnitude of PFR may need to be adjusted depending on which gait parameters of each patient would like to improve.

Implications: This research could contribute to stroke rehabilitation and improvement of treatment effect using AFO based on stiffness optimal settings.

Funding acknowledgements: This work was supported in part by JSPS KAKENHI under Grant 22K18240.

Keywords:
Stroke
Orthosis
Marker-less gait analysis

Topics:
Neurology: stroke
Disability & rehabilitation
Neurology

Did this work require ethics approval? Yes

Institution: Tokyo University of Science

Committee: Ethics Committee of the Tokyo University of Science

Ethics number: 20022