Gait impairment in post-stroke patients increases the risk of falls, but the role of ground reaction force variability (GRF variability) in controlling gait stability remains unclear. The objectives of this study were (1) to clarify the differences in GRF variability during walking between post-stroke patients and age-matched controls and (2) to identify the differences in GRF variability between post-stroke patient fallers and non-fallers. Sixteen post-stroke patients (age: 72.19 ± 8.54, six female, four fallers: age: 71.75 ± 11.32, twelve non-fallers: age: 72.33 ± 8.03) and nineteen age-matched controls (age: 68.63 ± 5.73, nine female) participated. GRF variability was measured using shoe sensors during walking. After adjusting for walking speed, the anterior–posterior (AP) GRF variability on the paretic side in the 91–100% stance phase was significantly lower in the post-stroke patients (F = 3.721, p = 0.038). This phase’s AP GRF variability was not correlated with Berg Balance Scale scores. Furthermore, the faller group in stroke patients showed the AP GRF variability on the paretic side was lower in the 41–50% (W = 17, p = 0.045), 51–60% (W = 16, p = 0.045), 61–70% (W = 16, p = 0.045), and 91–100% (W = 23, p = 0.045) sub-stance phases. After adjusting for sex and orthosis, the sensitivity analysis showed no significant intergroup difference. This suggested an adaptive control mechanism for maintaining gait and avoiding falls in post-stroke patients.
Although falls commonly occur in post-stroke patients during turning, the characteristics of trunk movement during turning in individuals at a high risk of falling remain unclear. The aim of this study was to clarify the characteristics of trunk translational and rotational movements during turning in post-stroke patients with a high risk of falling. Trunk acceleration and angular velocity were measured using the inertial measurement unit of an iPhone in the timed up and go test and compared among 13 post-stroke patients with a high risk of falling (age: 69.38 ± 12.44 years, Berg Balance Scale (BBS) < 45), 18 post-stroke patients with a low risk of falling (age: 71.22 ± 8.50 years, BBS 45), and 10 age-matched healthy controls (age: 65.90 ± 11.57). We examined the differences in trunk movement during turning between groups and the relationships between the BBS score and trunk movement. The high-risk group exhibited the longest completion time (χ2 = 31.21, p < 0.001) and the lowest maximum of trunk angular velocity along the vertical axis among groups (χ2 = 28.51, p < 0.001). Furthermore, the high-risk group showed a higher minimum (absolute value) of trunk angular velocity along the mediolateral axis compared to the low-risk group (χ2 = 9.80, p = 0.007). The maximum trunk angular velocity along the vertical axis (r = 0.66, p < 0.001) and the minimum trunk angular velocity along the mediolateral axis (r = 0.51, p = 0.003) were significantly correlated with the BBS score. We found that post-stroke patients with a high risk of falling exhibited slower trunk rotation angular velocity and faster trunk flexion angular velocity during turning compared to low-risk groups. Our findings suggest that despite the decrease during turning speed due to poor balance control, post-stroke patients with a high risk of falling exhibit a greater disturbance in the sagittal plane.
Gait impairment in post-stroke patients increases the risk of falls, but the role of g round r eaction f orce variability (GRF variability) in controlling gait stability remains unclear. This study aimed to characterize GRF variability during walking in post-stroke patients. 16 post-stroke patients (age: 72.19 ± 8.54, 6 female, 4 fallers: age: 71.75 ± 11.32, 12 non-fallers: age: 72.33 ± 8.03) and 19 age-matched healthy controls (age: 68.63 ± 5.73, 9 female) participated. GRF variability was measured using shoe sensors during walking. After adjusting for walking speed, the a nterior- p osterior (AP) GRF variability on the paretic side in the 91 – 100% stance phase was significantly lower in the post-stroke patients (p = 0.038). This phase’s AP GRF variability was not correlated with Berg Balance Scale scores. Furthermore, the faller group in stroke patients showed the AP GRF variability on the paretic side was lower in the 11 – 20% (p = 0.045), 41 – 50% to 61 – 70% stance phases (p = 0.045, p = 0.034, p = 0.034) after adjusting for sex and orthosis. This suggests that AP GRF variability on the paretic side is involved in the maintenance of walking stability in post-stroke patients.
Background: Gait speed is an important indicator of walking ability in post-stroke patients. Although gait speed is associated with anterior–posterior reaction force (AP RF), how AP RF is distributed across the plantar surface and whether this distribution is associated with gait speed remain unclear. Methods: Fifteen post-stroke patients and 18 healthy controls were included. AP RF was measured during walking using a sensor shoe system equipped with four triaxial force sensors located at the heel, first metatarsal (MT1), fifth metatarsal (MT5), and toe regions. AP RF distribution was quantified as the relative contribution of each plantar region and compared among the paretic, non-paretic, and control sides across the stance phase using linear mixed-effects models. Associations between AP RF distribution and gait speed were examined using Spearman’s rank correlation coefficients, with p-values adjusted for multiple testing using the Benjamini–Hochberg false discovery rate procedure. Results: Significant group × stance phase interactions were observed for the relative contributions of MT1 and MT5. After correction for multiple comparisons, the MT1 contribution at 81–90% of the stance phase was lower on both the paretic and non-paretic sides than on the control side, whereas the MT5 contribution at 1–10% was higher on both sides than on the control side. Total AP RF was also lower on the paretic and/or non-paretic sides than on the control side during several stance phases. On the non-paretic side, the MT5 contribution at 1–10% was negatively correlated with gait speed (rs = −0.66, BH-FDR adjusted p = 0.04). Conclusions: Post-stroke patients showed altered AP RF distribution during the early and late stance phases compared with healthy controls. Greater MT5 contribution on the non-paretic side during early stance was associated with slower gait speed, suggesting that AP RF distribution may provide additional information for characterizing gait impairment after stroke beyond total AP RF.
Although falls are a commonly occur in post-stroke patients during turning, the characteristics of trunk movement during turning in those at high-risk falling remains unclear. The aim of this study was to clarify the characteristics of trunk translational and rotational movements during turning in post-stroke patients with high-risk falling. Trunk acceleration and angular velocity were measured using the inertial measurement unit of iPhone in the Timed up and go test and compared among 13 high-risk falling post-stroke patients (Berg Balance Scale (BBS) < 45), 18 low-risk falling post-stroke patients (BBS ≥ 45) and 10 age-matched healthy controls. We examined the differences in the trunk movement during turning between groups and the relationships between BBS score and trunk movement. The high-risk group exhibited the longest completion time (χ2 = 31.21, p < 0.001) and the lowest maximum of trunk angular velocity about the vertical axis among groups (χ2 = 28.51, p < 0.001). Furthermore, the high-risk group showed a higher minimum (absolute value) of trunk angular velocity about the medio-lateral axis compared to the low-risk group (χ2 = 9.80, p = 0.007). The maximum of trunk angular velocity about vertical axis (r = 0.66, p < 0.001) and the minimum of trunk angular velocity about the medio-lateral axis (r = 0.51, p = 0.003) were significantly correlated with BBS score. We found that the post-stroke patients with high-risk falling exhibited the slower trunk rotation angular velocity and faster trunk flexion angular velocity during turning compared to low-risk groups. Our findings suggest that despite of decrease during turning speed due to poor balance control, post-stroke patients with high-risk falling exhibited a greater disturbance in the sagittal plane.