Effectiveness of a wearable ankle-tubing gait training on ankle kinematics and motor control in hemiparetic stroke

Abstract

BACKGROUND:

While excessive ankle plantarflexion is a common neuromuscular impairment resulting from insufficient coordination of selective ankle neuromotor control and kinematics during gait. We recently developed a wearable, inexpensive and sustainable wearable ankle-tubing gait training (WAGT) aimed at improving selective ankle motor control and kinematic coordination.

OBJECTIVE:

We investigated the effects of WAGT on tibialis anterior (TA) and gastrocnemius (GCM) muscle electromyography (EMG) activity, TA: GCM muscle imbalance ratio, and ankle joint kinematics during gait in hemiparetic stroke patients.

METHODS:

A convenience sample of 33 participants (15 non-stroke healthy adults and 18 hemiparetic stroke patients) underwent standardized electromyography and kinematic biomechanical tests under conventional gait training (CGT) and WAGT conditions. Analysis of variance (ANOVA) was used to determine the significance of differences in the TA: GCM muscle activation, muscle imbalance ratio, and ankle joint kinematics before and after the intervention and between the two groups at P < 0.05.

RESULTS:

WAGT was more effective than CGT in improving TA muscle activation (P < 0.01), TA: GCM muscle imbalance ratio (P < 0.01), and kinematic movement (P < 0.01) in adults with or without hemiparetic stroke.

CONCLUSIONS:

This study demonstrated that WAGT is relatively ease to design, wear and affordable to most clinicians and patients, hence it is suitable for many health care applications to correct gait-related movement abnormalities presented in the hemiparetic stroke patients.

Keywords

Stroke ankle-tubing gait wearable electromyography joint kinematics

1 Introduction

Excessive ankle plantarflexion is the most common gait limitation in hemiparetic stroke and is often associated with insufficient coordination between selective ankle neuromotor control and kinematic movement (Cioni et al., 2006; Deltombe et al., 2007; Foley et al., 2010). The conceptual neuromechanical basis of the development of the wearable ankle-tubing gait training (WAGT) approach was derived from the restoration of neuromuscular imbalance between the agonist tibialis anterior (TA) and antagonist gastrocnemius (GCM) muscle activations. Normally, the TA muscle is activated during the initial contact and swing phases of the gait cycle, whereas the GCM muscle is reciprocally activated during the loading-terminal stance phases (Di Giulio et al., 2009). In our patients with hemiparetic stroke, the altered neuromuscular balance between the agonist TA and antagonist GCM is manifested by the increased synergistic activation of GCM and reciprocally inhibited TA during the initial contact and swing phases of the gait cycle (Buurke et al., 2008; Di Giulio et al., 2009; Ghédira et al., 2021). Such altered neuromuscular control often results in plantarflexion or circumduction gait. To restore neuromuscular imbalance between the agonist TA and antagonist GCM muscle activations during gait, the WAGT approach was designed to facilitate the underactive TA and reciprocally inhibit overactive GCM (Buurke et al., 2008; Lamontagne et al., 2002), which further assists ankle dorsiflexion when combined with treadmill walking. The specific tubing application parameters, including tension, direction, and line of force or pull, were adjusted and curtailed based on each patient’s neuromuscular response to generate balanced neuromuscular activation control between the agonist TA and antagonist GCM during training. Theoretically, as the treadmill moves forward, the participant’s TA is reflexively activated by stepping forward in response to the anterior displacement perturbation induced by the treadmill. Knee tubing generates hip-knee-ankle triple flexion during the initial contact and pre-swing to terminal swing phases, whereas ankle tubing provides tracking resistance to ankle dorsiflexors to further augment active TA activation and associated ankle dorsiflexion movements. The ankle plantarflexor contributes to eccentric contraction by decelerating the tibial forward momentum around the axis of the talus from the terminal stance (TS) to the pre-swing phases, while the tibialis anterior (TA) concentrically activates to accelerate dorsiflexion (Richie Jr, 2021). The mesencephalic locomotor region regulates neuromuscular coactivation between the agonist TA and antagonist GCM to stabilize the ankle joint during midstance and reciprocally inhibit the antagonist plantarflexor electromyography (EMG) muscle activation (5%) during the TS phase in chronic hemiparetic stroke (Levin et al., 2000; Mindy F Levin et al., 2000). However, in hemiparetic stroke patients, coordinated neuromuscular control results in plantarflexion gait, increasing compensatory circumduction or altered toe clearance and the associated risk of falls in hemiparetic stroke (Ardestani et al., 2019; Giannotti et al., 2015). Such joint decentration damages the biomechanical properties of the capsular and ligamentous structures in the medial aspect of the ankle, causing degenerative changes and equinovarus feet.

Current gait training approaches, including partial body weight-supported treadmill training (PWTT) and robot-assisted gait training (RAGT), have been widely used to address excessive ankle plantarflexion in hemiparetic stroke gait (Daly et al., 2004; Duncan et al., 2011; Mao et al., 2015). However, outcome studies have failed to demonstrate the efficacy of selective ankle motor control and associated kinematics (Combs et al., 2010; Mao et al., 2015; McCain et al., 2008). PWTT studies have focused on functional limitations rather than targeting the primary neuromuscular imbalance or impairment to restore gait dysfunction in hemiparetic stroke patients (Combs et al., 2010; Duncan et al., 2011; Hassid et al., 1997; Yagura et al., 2006). The inconsistencies associated with PWTT mandate a more effective locomotor conditioning paradigm (Mulroy et al., 2010). Robot-assisted training provides a well-programmed impedance control to generate optimal gait trajectory, but it is extremely expensive, which is not readily affordable in most clinical settings (Hidler et al., 2009; Park et al., 2020; Riener et al., 2010). A majority of RAGT tend to be ‘a robot in charge of a person rather than a person in charge of a robot’, which warrants for the development of interactive robot-human gait training system (Schwartz & Meiner, 2015).

To overcome the shortcomings of conventional gait training (CGT), we recently developed a wearable, inexpensive and sustainable wearable ankle-tubing gait training (WAGT) aimed at improving selective ankle motor control and kinematic coordination (Lee & You, 2017; Lee et al., 2017; Shin et al., 2014). In WAGT, a durable elastic band mounted with a wearable velcro suit is used and adjusted to simulate manual assistance or a combination of assistance and resistance, improving muscle coordination. There is a need to develop an effective and sustainable intervention for selective ankle motor control and kinematic patterns in hemiparetic stroke patients. This study aimed to compare the effects of CGT and WAGT on TA: GCM EMG muscle activity and imbalance ratio as well as ankle joint kinematics during treadmill gait in hemiparetic stroke patients.

2 Methods

2.1 Participants

A convenience sample of 33 participants [18 hemiparetic stroke patients (8 females, aged 39.2±16.8 years) and 15 non-stroke healthy adults (6 females, aged 26.3±2.6 years)] from Myongji Hospital were enrolled (No. 2015-01-066). All participants provided written informed consent. The study was approved by the Ethics Committee of Myongji Hospital. The inclusion criteria for hemiparetic stroke patients were as follows: (1) patients with a stroke onset within the past 4–7 months were recruited because this criterion represents our sample population in our hospital setting, where natural recovery is relatively stabilized. (2) chronic cortical/subcortical ischemic stroke (4–7 months post-stroke onset), (3) first stroke, (4) Korean Mini-Mental State Examination (MMSE-K) score > 24, (5) passive ankle dorsiflexion > 0°, (6) Modified Ashworth Scale (MAS) score < 2 on ankle plantarflexion, (7) Berg Balance Scale (BBS) score > 40, and (8) ankle plantarflexion gait beyond the neutral anatomical position during the stance phase (Kinsella & Moran, 2008; Manca et al., 2014). The exclusion criteria were as follows: (1) a history of lower extremity surgery within the past 6 months, (2) no musculoskeletal disease or other neurological disease, (3) current seizures and related medications interfering with the experiment, and (4) a history of botulinum toxin A injection in the GCM muscle. The control group included healthy participants without any medical problems. The participants’ demographic and clinical characteristics are shown in Table 1.

Table 1
Demographic and clinical characteristics of the subjects (N = 33)

Parameters Control group (n = 15) Stroke group (n = 18)

Sex (male/female) 9/6 10/8

Age (years) 26.3±2.6 39.2±16.8

Height (cm) 171.3±5.7 163.1±8.8

Weight (kg) 61.5±7.8 65.3±11.0

Post-stroke duration (month) N/A 6.7±2.9

Type (hemorrhage/infarct) N/A 10/8

Paretic side (right/left) N/A 9/9

MMSE-K 30 27.6±2.0

MAS (L/E) 0 1∼1+

Gait speed (km/h) 1.7±0.2 0.65±0.1

BBS 55.1±0.8 43.9±2.3

MMT (lower extremity) Rt. Lt. Rt. Lt.

Hip flexors 5 5 3.9±1.1 3.8±1.3

Hip extensors 5 5 3.7±1.1 4.1±1.2

Knee flexors 5 5 3.7±1.1 4.1±1.2

Knee extensors 5 5 3.7±1.1 4.1±1.2

Ankle dorsiflexors 5 5 3±1.6 3.5±1.7

Ankle plantar flexors 5 5 3±1.6 3.6±1.6

Parameters	Control group (n = 15)	Stroke group (n = 18)
Sex (male/female)	9/6	10/8
Age (years)	26.3±2.6	39.2±16.8
Height (cm)	171.3±5.7	163.1±8.8
Weight (kg)	61.5±7.8	65.3±11.0
Post-stroke duration (month)	N/A	6.7±2.9
Type (hemorrhage/infarct)	N/A	10/8
Paretic side (right/left)	N/A	9/9
MMSE-K	30	27.6±2.0
MAS (L/E)	0	1∼1+
Gait speed (km/h)	1.7±0.2	0.65±0.1
BBS	55.1±0.8	43.9±2.3
MMT (lower extremity)	Rt.	Lt.	Rt.	Lt.
Hip flexors	5	5	3.9±1.1	3.8±1.3
Hip extensors	5	5	3.7±1.1	4.1±1.2
Knee flexors	5	5	3.7±1.1	4.1±1.2
Knee extensors	5	5	3.7±1.1	4.1±1.2
Ankle dorsiflexors	5	5	3±1.6	3.5±1.7
Ankle plantar flexors	5	5	3±1.6	3.6±1.6

Data are presented as means±standard deviations (SD), unless otherwise specified; N/A, not applicable; MMSE-K, Korean version of the Mini-Mental Status Examination; MAS, Modified Ashworth Scale; BBS, Berg Balance Scale; MMT, manual muscle testing; Rt, right; Lt, left.

2.2 Clinical tests

This was a case-control study. A procedural checklist was developed to ensure a consistent experimental protocol. Clinical tests included the MMSE-K, MAS, BBS, manual muscle testing, and ankle joint range of motion (ROM).

A consistent experimental procedure using standardized tests was implemented, which included assessment of the MMSE-K, MAS, BBS, ankle joint ROM, and sensory function. The reliability and validity of all of the tests are well established. The MMSE-K was used to determine cognitive orientation to time and place, storage register, recall force, attention, computational power, and language function. The test included 12 items, with a total score ranging from 0 to 30 (≤17, severe cognitive impairment; 18–23, mild cognitive impairment;≥24, normal) (Lee et al., 1999). Reliability and validity were reported to be r = 0.95 and intraclass correlation coefficient (ICC)_3,k = 0.998, respectively (Lee et al., 1999).

The MAS was used to measure the degree of ankle plantar flexor spasticity during passive soft-tissue stretching in patients with central nervous system lesions or neurological disorders (Charalambous, 2014). The grading scale was as follows: 0, no increase in muscle tone; 1, slight increase in tone, followed by minimal resistance; 2, more marked increase in tone, but the affected part moved easily; 3, considerable increase in tone and passive movement difficult; and 4, rigid affected part in flexion or extension. Reliability and validity were ascertained, with an ICC_3,k and r of 0.87 and 0.90, respectively (Mutlu et al., 2008; Sloan et al., 1992).

The BBS was used to measure static and dynamic balance and functional performance among older individuals with balance impairments. It is an acceptable instrument for evaluating intervention efficacy and quantitatively describing functional performance (Major et al., 2013). The test comprised 14 balance tasks, with each task given a score from 0 (unable) to 4 (intendent), and the total score ranging from 0 to 56 points (≤20, wheelchair users; > 20–40, walking with assistance; > 40–56, independent). Reliability and validity were verified, with an ICC_3,k and r of 0.89 and 0.957, respectively (Viveiro et al., 2019).

The midstance ankle joint kinematic angle was measured to assess the extent of gait on full knee extension at 0° using a captured video image (Samsung, Seoul, Republic of Korea) and ImageJ (NIH, Bethesda, MD, USA) (Hicks et al., 2008). Touch, pressure, and position sense of the lower extremities were also assessed (Fillmore, 1999). Additionally, a post-intervention survey on kinesthesia and proprioception was conducted to determine the qualitative influence of the tubing gait training.

2.3 EMG measurements

Surface EMG (LAXTHA Inc., Daejeon, Republic of Korea) was used to determine the lower extremity muscle activity and imbalance. Prior to placement of the EMG electrode, the skin was shaved, cleaned with an alcohol pad, and abraded. Active 1.8 cm electrodes were attached parallel to the muscle zone as follows: for each TA muscle, a pair on the anterolateral aspect of the tibia at the proximal one-quarter to one-third of the distance between the knee and ankle, 2 cm apart; and for each GCM muscle, one electrode at the proximal muscle area for general recordings, and a pair distal to the knee, 2 cm lateral to the midline, and 2 cm apart (Criswell, 2010).

Reference voluntary contraction (RVC) was used to normalize the TA and GCM muscle activity. The mean root mean square (RMS) of the RVC was calculated for each muscle when the subjects were in a comfortable standing position to normalize the EMG amplitude data. For the data analysis, we used 3 s of the 5 s of EMG data, excluding the initial and final 1 s. All EMG data were expressed as percentages of the RVC (% RVC). The mean value of the three trials for each muscle activity was recorded as the RVC. Before testing, all participants underwent two to five sets of 10-min CGT and 10-min WAGT once the ideal tubing tension was identified to become familiarized with the experimental testing procedure. The participants’ normal gait speed was evaluated by having them walk on an even surface. The participants practiced on a treadmill, with a suspension harness around the chest to ensure safety. The participants were randomly tested on either CGT or WAGT based on the flipping coin method. To minimize experimental bias associated with the participants’ expectations, they were blinded to any potentially influencing information until after study completion.

The EMG signals were collected at a sampling rate of 1024 Hz along with a 60 Hz notch filter. The bandpass filter was between 20 and 450 Hz. Each participant performed three repetitions of 10 s treadmill walking at a self-selected speed to determine the RVC (Knutson et al., 1994). The participants practiced CGT and WAGT for 10 min to familiarize themselves with the experimental process. EMG data were then recorded for 5 min, and recordings of the last 10 gait cycles were stored for further analysis. The raw EMG data were analyzed using Telescan version 3.06 (LAXTHA Inc., Daejeon, Republic of Korea) to determine TA and GCM amplitudes. The muscle imbalance ratio between the TA and GCM muscle (TA:GCM muscle imbalance ratio) was computed by measuring the average peak RMS values of the TA over the GCM of the entire gait cycle (Hwang et al., 2019).

2.4 Kinematic measurements

Kinematic data were analyzed with a camcorder and ImageJ to determine gait-associated ankle joint angular changes. A reflective marker was placed on each of the following areas of the paretic leg: the fibular lower end over the lateral malleolus, superior to the lateral femoral epicondyle, and lateral to the fifth metatarsal head. Kinematic video images were used to define the spatiotemporal gait stages. Kinematic data were recorded at 15 Hz and synchronized with EMG data using a customized foot switch applied under the fifth metatarsal head, which lit up the lamplight upon every foot contact with the ground during the gait cycle. A 15-min rest interval was provided to avoid fatigue. Video images captured during the midstance phase were analyzed using ImageJ to determine the ankle joint kinematics (Fig. 1). The location of the video camcorder and marker placements were consistently implemented for all subjects to avoid any potential confounding factors or bias associated with this measurement. Furthermore, we have previously established the validity and reliability of the Image J measurement, which yielded excellent accuracy (R² = 0.976, r = 0.988) and consistency (Cronbach’s alpha = 0.994) (Lee & You, 2016).

Fig. 1

Foot switch customized for identifying the gait cycle.

2.5 WAGT technique

As illustrated in Fig. 2, the WAGT technique involved walking with two elastic tubes (Thera-Band; Hygenic Corp., Akron, OH, USA) to generate resistive coupling. WAGT is a form-fitting garment with different attachment points for the tubing band that offer the wearer assistance and resistance to ankle movement. It consists of a tubing strap and specially connected hooks and rings; the pieces of the garment are connected and adjusted to optimally position the ankle joint (Jakobsen et al., 2012; Lee & You, 2017). Specifically, the proximal tube was applied to the popliteal fossa, while the distal tube was applied 2 cm above the medial malleolus, to generate anterior and posterior tangential forces, respectively. The distal tubing tension was approximately 1.5 times greater than the proximal tubing tension, resisting knee joint extension in both the stance and swing phases. Elastic tube resistance was determined in accordance with the manufacturer’s guidelines (Page, 2000; Patterson et al., 2001). Force elongation in pounds was computed using the normalized length-tension equation [(final length − resting length)/resting length]×100 (Lee & You, 2017). In this experiment, approximately 1.8–3.6 kg of elastic tubing was used as resistance and adjusted to accommodate each participant’s impairment level, creating a more normalized locomotor pattern. The order of gait training methods was randomized to minimize any potential ordering effects (Fig. 3).

Fig. 2

Tubing gait training for preventing ankle plantarflexion.

Fig. 3

Experimental procedure.

2.6 Statistical analysis

Data are expressed as means and standard deviations. A power analysis using G-Power software (Franz Faul, Kiel, Germany) was conducted to assess the sample size requirement (n = 30) based on our previous study, which demonstrated the effect size (eta squared, η2 = 0.6) and power (1-β= 0.8) (Lee & You, 2017). Two-group, two-intervention, repeated-measures analysis of variance (ANOVA) was used to determine the statistical significance of differences in TA and GCM activation, muscle imbalance ratio, and ankle joint kinematics before and after the intervention between adults with and without hemiparetic stroke. Significant differences between the control and hemiparetic stroke groups, which indicated patient–tubing interaction, were subjected to Tukey’s post hoc test. Statistical significance was set at P < 0.05. Statistical Package for the Social Sciences for Windows version 25.0 (Chicago, IL, USA) was used for all statistical analyses.

3 Results

An independent t-test was performed to determine the difference in baseline demographic and clinical characteristics of the 33 participants, which revealed no significant difference, indicating homogeneity in terms of male sex, age, height, and weight.

3.1 EMG activation

Repeated-measures ANOVA showed significant effects of both WAGT and CGT on TA and GCM muscle activity (P < 0.01) and a significant difference in TA muscle activity between the two groups (P < 0.01) (Table 2). A paired t-test revealed significant differences in amplitude between the pretest and post-test of the hemiparetic stroke (P = 0.004) groups. The control group showed no significant differences (P = 0.184). Independent t-test revealed a significant difference in TA amplitude between the two groups after WAGT (P = 0.005). However, post hoc analysis using Tukey’s test revealed no significant differences between the groups (f < 4.05) (Table 5).

Table 2
EMG activation (% RVC) and EMG muscle imbalance ratio between the control and hemiparetic stroke groups

Control (mean±SD) Hemiparetic stroke (mean±SD) P-value

Pretest Post-test Pretest Post-test Time effect (CGT, WAGT) Time×Group Between groups

TA 97.3±11.0 115.5±21.3 98.8±4.4 189.2±13.7 0.000^ 0.000^ 0.000^

GCM 99.8±23.6 89.2±24.4 112.6±18.4 106.6±15.2 0.003^ 0.124 0.263

TA:GCM 1.0±0.2 1.3±0.4 0.8±0.1 1.8±0.2 0.000^ 0.000^ 0.293

	Control (mean±SD)	Hemiparetic stroke (mean±SD)	P-value
TA	97.3±11.0	115.5±21.3	98.8±4.4	189.2±13.7	0.000^**	0.000^**	0.000^**
GCM	99.8±23.6	89.2±24.4	112.6±18.4	106.6±15.2	0.003^**	0.124	0.263
TA:GCM	1.0±0.2	1.3±0.4	0.8±0.1	1.8±0.2	0.000^**	0.000^**	0.293

TA, tibialis anterior; GCM, gastrocnemius; CGT, conventional gait training; WAGT, wearable ankle-tubing gait training. ^**P < 0.01.

3.2 Muscle imbalance ratio between groups

Repeated-measures ANOVA showed significant effects of WAGT and CGT on the TA: GCM muscle imbalance ratio and training time of both groups (P < 0.01) (Table 2). The paired t-test revealed significant differences in the TA: GCM ratio between the pretest and post-test of the control (P = 0.027) and hemiparetic stroke (P = 0.015) groups. Independent t-test and Tukey’s post hoc test showed no significant difference in the WAGT TA: GCM ratio between the control and hemiparetic stroke groups (P = 0.906, f < 4.05) (Table 4).

Table 4
Tukey post hoc analysis

TA activity (P < 0.05)

Groups M₁-M₂ sqrt(MS_w(1/n)) Tukey’ HSD F score

M1 vs. M2 9.080667 8.624126365 1.05293757 < 4.05

M1 vs. M3 19.50656 8.624126365 2.26185874 < 4.05

M1 vs. M4 10.711 8.624126365 1.24198087 < 4.05

M2 vs. M3 28.587222 8.624126365 3.314796307 < 4.05

M2 vs. M4 19.791667 8.624126365 2.294918445 < 4.05

M3 vs. M4 8.7955556 9.147267352 0.961550069 < 4.05

TA:GCM ratio (P < 0.05)

Groups M₁-M₂ sqrt(MS_w(1/n)) Tukey’ HSD F score

M1 vs. M2 0.422667 0.330227195 1.27992689 < 4.05

M1 vs. M3 0.233111 0.330227195 0.70591131 < 4.05

M1 vs. M4 0.262444 0.330227195 0.79473904 < 4.05

M2 vs. M3 0.6557778 0.330227195 1.985838201 < 4.05

M2 vs. M4 0.1602222 0.330227195 0.485187849 < 4.05

M3 vs. M4 0.4955556 0.350258833 1.41482672 < 4.05

TA activity (P < 0.05)
Groups	M₁-M₂	sqrt(MS_w(1/n))	Tukey’ HSD	F score
M1 vs. M2	9.080667	8.624126365	1.05293757	< 4.05
M1 vs. M3	19.50656	8.624126365	2.26185874	< 4.05
M1 vs. M4	10.711	8.624126365	1.24198087	< 4.05
M2 vs. M3	28.587222	8.624126365	3.314796307	< 4.05
M2 vs. M4	19.791667	8.624126365	2.294918445	< 4.05
M3 vs. M4	8.7955556	9.147267352	0.961550069	< 4.05
TA:GCM ratio (P < 0.05)
Groups	M₁-M₂	sqrt(MS_w(1/n))	Tukey’ HSD	F score
M1 vs. M2	0.422667	0.330227195	1.27992689	< 4.05
M1 vs. M3	0.233111	0.330227195	0.70591131	< 4.05
M1 vs. M4	0.262444	0.330227195	0.79473904	< 4.05
M2 vs. M3	0.6557778	0.330227195	1.985838201	< 4.05
M2 vs. M4	0.1602222	0.330227195	0.485187849	< 4.05
M3 vs. M4	0.4955556	0.350258833	1.41482672	< 4.05

3.3 Ankle joint kinematics

Repeated-measures ANOVA showed significant effects of WAGT and CGT in ankle joint kinematics of both groups (P < 0.01) (Table 3). The paired t-test confirmed significant differences in the ankle joint kinematics between the pretest and post-test of the control (P = 0.037) and hemiparetic stroke (P < 0.01) groups. The independent t-test revealed a significant difference in ankle joint kinematics between the two groups after WAGT (P = 0.005).

Table 3
Ankle joint dorsiflexion and plantarflexion

Control (mean±SD) Hemiparetic stroke (mean±SD) P-value

Pretest (degrees) Post-test (degrees) Pretest (degrees) Post-test (degrees) Time effect (CGT, WAGT) Time×Group Between groups

Ankle joint 112.3±2.49 110.6±41.5 121.1±1.45 114.2±3.8 0.000^^ 0.002^^ 0.000^^

	Control (mean±SD)	Hemiparetic stroke (mean±SD)	P-value
Ankle joint	112.3±2.49	110.6±41.5	121.1±1.45	114.2±3.8	0.000^^	0.002^^	0.000^^

CGT, conventional gait training; WAGT, wearable ankle-tubing gait training. ^**P < 0.01.

3.4 Post-intervention survey results

The post-intervention survey to a therapist and patients showed that WAGT reduced the therapist’s physical demand to guide or assist the limb movement while it increased the easiness of the ankle-knee-hip movement in patients during gait training.

4 Discussion

This clinical research investigated the immediate effects of WAGT on TA and GCM activity, TA: GCM muscle imbalance ratio, and associated ankle joint kinematics in healthy adults and hemiparetic stroke patients presenting with excessive ankle plantarflexion. As anticipated, WAGT helped restore the TA-GCM activation coordination in both groups. Muscle activation changes were more pronounced and thus more favorable in the hemiparetic stroke group. These changes improved the ankle joint dorsiflexion and plantarflexion during gait. To the best of our knowledge, no previously published data are available for comparison with our clinical data in order to optimize guidance therapy.

EMG activation analysis demonstrated that WAGT more effectively increased deactivated TA (72.8%) while decreasing overactive GCM (−5.3%) in the hemiparetic stroke group compared with the levels in the control group. This finding was in parallel with previous EMG results (Ferris et al., 2006; Swank et al., 2020). Swank et al. (2020) observed increased TA activation (9.5%) and decreased GCM activation (−4.2%) after 4 weeks of robotic exoskeleton gait training (EKSO) in six hemiparetic stroke patients (Swank et al., 2020). Ferris et al. reported that powered ankle-foot orthosis improved both TA (10%) and soleus (53%) activation compared with passive orthosis during gait in a chronic hemiparetic stroke patient (Ferris et al., 2006). Furthermore, additional EMG activation imbalance ratio analysis revealed a more improved TA: GCM imbalance ratio in the hemiparetic stroke group, which improved to near normal during the application of the WAGT (from 0.8 to 1.8). This result corroborates previous EMG findings of TA: GCM imbalance ratio improvement after EKSO (from 1.0 to 1.6) in six hemiparetic stroke patients. These results may support the importance of an accurate guidance force and proprioceptive feedback, which influence the supraspinal regulation of the central pattern generator during gait training (Ferris et al., 2005). Recently, Kang et al. examined the immediate effect of tethered pelvic control robot-assisted gait training on a 10% body weight-loaded treadmill condition on the soleus and GCM activation, kinematics, and kinetics in six individuals with cerebral palsy. Initially, the GCM and soleus muscles were activated too early and hampered ankle dorsiflexion and knee extension (Kang et al., 2017); however, they were enhanced during application of the pelvic control robot-assisted gait training. This finding suggests that robot-assisted ankle gait training may have facilitated ankle proprioception and kinesthesia, which plays an important role in “reference correction” during locomotor re-training. Similarly, a possible underlying rationale is that corrective ankle guidance during WAGT enhanced ankle joint movement sensation and awareness required for selective, coordinated neuromotor control of active TA facilitation and reciprocal inhibition of GCM during the gait training (Jakobsen et al., 2012). Possible mechanisms underlying such improvement in the muscle imbalance ratio are increased inhibition of the GCM and concurrent facilitation of the TA muscle as well as the associated lengthening of the GCM muscle belly, aponeurosis, or Achilles’ tendon.

Ankle kinematic analysis revealed a greater improvement in selective ankle dorsiflexion (5.70%) during WAGT than during CGT in both groups. The present kinematic findings were similar to those in previous studies utilizing a variety of different therapeutic modalities including Walkbot robot-assisted gait training (RAGT), Adeli suit therapy, and Thera suit. Recently, Park et al. (2021) showed significantly enhanced selective ankle dorsiflexion (24.5%) and plantarflexion (20.6%) kinematics during the application of the ankle-knee-hip Walkbot RAGT than during the application of RAGT without an ankle actuator in three participants who had suffered left hemiparetic stroke (PARK et al., 2021). In addition, Ko et al. (2014) reported substantially improved ankle dorsiflexion (50%) and plantarflexion (30%) angular kinematics after 6 weeks of Adeli suit training in patients with diplegic cerebral palsy (Ko et al., 2015). Similarly, Martins et al. (2019) found increased ankle dorsiflexion (10%) kinematics at the initial contact and swing phase during the application of the Thera suit in seven participants with CP (Martins et al., 2019). A possible explanation for this is that the tubing band created therapeutic tension to provide deep proprioceptive and kinesthetic pressure sense via guidance force as well as resistance force for strengthening the muscles during WAGT. Abnormal muscle alignment is corrected by adjusting the tension of the tubing bands to simulate the flexor and extensor synergetic activation pattern of the ankle-knee-hip joint musculature, which assisted normalized coordinated interlimb ankle-knee-hip joint movement (Lee, 2013). Post-intervention survey to the therapists and patients reported that the WAGT reduced the therapists’ physical burden that is needed to mobilize the patient whereas it increased the easiness of the ankle-knee-hip movement in the patients during gait training.

Taken together with present findings, the gait assisted guidance force utilizing WAGT was more effective in improving selective ankle muscle and movement control than CGT. Moreover, the current study provides new clinical insights that WAGT can help mitigate the impaired coordinated, selective ankle motor control in gait abnormalities including foot drop, toe-dragging excessive plantarflexor gait in individuals with hemiparetic stroke (Shin et al., 2014). WAGT is relatively ease to design, wear and affordable to most clinicians and patients, hence it is suitable for many health care applications to correct gait-related movement abnormalities presented in the hemiparetic stroke patients.

The current study has a couple of limitations that need to be considered in future clinical trials. One limitation is that the present case-control study examined only the immediate effects of WAGT. Prospective investigation of the long-term effects of WAGT in ankle plantarflexion gait is warranted. The other limitation is that although the EMG and kinematic data showed promising effects, it was difficult to quantify the optimal tubing tension to be applied. Further studies should be conducted to quantify the tubing tension appropriate for each patient’s locomotor impairment, neuromuscular control, and functional performance. Our present findings suggest that the innovative WAGT is an affordable, effective augmented treadmill-based locomotor rehabilitation approach in which guidance (assistance) and resistive force can be adjusted to restore neuromuscular control and coordination between the TA and GCM muscles in hemiparetic stroke patients with plantarflexion and circumduction gait.

5 Conclusion

The present clinical study validated the effectiveness of tubing gait training in restoring ankle joint muscle coordination and kinematics in hemiparetic stroke patients presenting with excessive ankle plantarflexion. Our results showed that WAGT significantly improved initial muscle coordination and TA: GCM activity imbalance ratio to normal ranges, reducing excessive ankle plantarflexion. These findings suggest that WAGT is effective in restoring muscle coordination and improving kinematic function in hemiparetic stroke patients, providing clinical insights into locomotor neurorehabilitation of these patients.

Conflict of interest

None to report.

References

Ardestani,

M. M.

, Kinnaird,

C. R.

, Henderson,

C. E.

, & Hornby,

T. G.

(2019). Compensation or Recovery? Altered Kinetics and Neuromuscular Synergies Following High-Intensity Stepping Training Poststroke. Neurorehabilitation and Neural Repair, 33(1), 47–58. https://doi.org/10.1177/1545968318817825

Buurke,

J. H.

, Nene,

A. V.

, Kwakkel,

, Erren-Wolters,

, IJzerman,

M. J.

, & Hermens,

H. J.

(2008). Recovery of gait after stroke: what changes? Neurorehabilitation and Neural Repair, 22(6), 676–683.

Charalambous,

C. P.

(2014). Interrater reliability of a modified Ashworth scale of muscle spasticity. In Classic Papers in Orthopaedics (pp. 415–417). Springer.

Cioni,

, Esquenazi,

, & Hirai,

(2006). Effects of botulinum toxin-A on gait velocity, step length, and base of support of patients with dynamic equinovarus foot, American Journal of Physical Medicine & Rehabilitation, 85(7), 600–606.

Combs,

S. A.

, Dugan,

E. L.

, Passmore,

, Riesner,

, Whipker,

, Yingling,

, & Curtis,

A. B.

(2010). Balance, balance confidence, and health-related quality of life in persons with chronic stroke after body weight–supported treadmill training, Archives of Physical Medicine and Rehabilitation, 91(12), 1914–1919.

Criswell,

(2010). Cram’s introduction to surface electromyography. Jones & Bartlett Publishers.

Daly,

J. J.

, Roenigk,

K. L.

, Butler,

K. M.

, Gansen,

J. L.

, Fredrickson,

, Marsolais,

E. B.

, Rogers,

, & Ruff,

R. L.

(2004). Response of sagittal plane gait kinematics to weight-supported treadmill training and functional neuromuscular stimulation following stroke. Journal of Rehabilitation Research & Development, 41(6a), 807–820. https://doi.org/10.1682/jrrd.2003.08.0120

Deltombe,

, Gustin,

, De Cloedt,

, Vandemeulebroecke,

, & Hanson,

(2007). The treatment of spastic equinovarus foot after stroke.in Physical and Rehabilitation Medicine, Critical Reviews™ , 19(3).

Di Giulio,

, Maganaris,

C. N.

, Baltzopoulos,

, & Loram,

I. D.

(2009). The proprioceptive and agonist roles of gastrocnemius, soleus and tibialis anterior muscles in maintaining human upright posture, The Journal of Physiology, 587(10), 2399–2416.

10.

Duncan,

P. W.

, Sullivan,

K. J.

, Behrman,

A. L.

, Azen,

S. P.

, Wu,

S. S.

, Nadeau,

S. E.

, Dobkin,

B. H.

, Rose,

D. K.

, Tilson,

J. K.

, & Cen,

(2011). Body-weight–supported treadmill rehabilitation after stroke, New England Journal of Medicine, 364(21), 2026–2036.

11.

Ferris,

D. P.

, Czerniecki,

J. M.

, & Hannaford,

(2005). An ankle-foot orthosis powered by artificial pneumatic muscles, Journal of Aplied Biomechanics, 21(2), 189–197.

12.

Ferris,

D. P.

, Gordon,

K. E.

, Sawicki,

G. S.

, & Peethambaran,

(2006). An improved powered ankle–foot orthosis using proportional myoelectric control, Gait & Posture, 23(4), 425–428.

13.

Fillmore,

(1999). Muscle and Sensory Testing, Physical Therapy, 79(11), 1100.

14.

Foley,

, Murie-Fernandez,

, Speechley,

, Salter,

, Sequeira,

, & Teasell,

(2010). Does the treatment of spastic equinovarus deformity following stroke with botulinum toxin increase gait velocity? A systematic review and meta-analysis, European Journal of Neurology, 17(12), 1419–1427.

15.

Ghédira,

, Albertsen,

I. M.

, Mardale,

, Loche,

C.-M.

, Vinti,

, Gracies,

J.-M.

, Bayle,

, & Hutin,

(2021). Agonist and antagonist activation at the ankle monitored along the swing phase in hemiparetic gait, Clinical Biomechanics, 89, 105459.

16.

Giannotti,

, Merlo,

, Zerbinati,

, Longhi,

, Prati,

, Masiero,

, & Mazzoli,

(2015). Early rehabilitation treatment combined with equinovarus foot deformity surgical correction in stroke patients: safety and changes in gait parameters, European Journal of Physical and Rehabilitation Medicine, 52(3), 296–303.

17.

Hassid,

, Rose,

, Commisarow,

, Guttry,

, & Dobkin,

B. H.

(1997). Improved gait symmetry in hemiparetic stroke patients induced during body weight-supported treadmill stepping, Journal of Neurologic Rehabilitation, 11(1), 21–26.

18.

Hicks,

J. L.

, Schwartz,

M. H.

, Arnold,

A. S.

, & Delp,

S. L.

(2008). Crouched postures reduce the capacity of muscles to extend the hip and knee during the single-limb stance phase of gait, Journal of Biomechanics, 41(5), 960–967.

19.

Hidler,

, Nichols,

, Pelliccio,

, Brady,

, Campbell,

D. D.

, Kahn,

J. H.

, & Hornby,

T. G.

(2009). Multicenter randomized clinical trial evaluating the effectiveness of the Lokomat in subacute stroke. Neurorehabiitation andl Neural Repair, 23(1), 5–13. https://doi.org/10.1177/1545968308326632

20.

Hwang,

H. C.

, Kim,

S. H.

, & Kim,

T. H.

(2019). Effects of Aquatic and Ground Obstacle Training on Balance and Muscle Activity in Patients With Chronic Stroke, Physical Therapy Korea, 26(4), 35–41.

21.

Jakobsen,

M. D

, Sundstrup,

, Andersen,

C. H.

, Bandholm,

, Thorborg,

, Zebis,

M. K.

, & Andersen,

L. L.

(2012). Muscle activity during knee-extension strengthening exercise performed with elastic tubing and isotonic resistance, International Journal of Sports Physical Therapy, 7(6), 606.

22.

Kang,

, Martelli,

, Vashista,

, Martinez-Hernandez,

, Kim,

, & Agrawal,

S. K.

(2017). Robot-driven downward pelvic pull to improve crouch gait in children with cerebral palsy, Science Robotics, 2(8).

23.

Kinsella,

, & Moran,

(2008). Gait pattern categorization of stroke participants with equinus deformity of the foot, Gait & Posture, 27(1), 144–151.

24.

Knutson,

L. M.

, Soderberg,

G. L.

, Ballantyne,

B. T.

, & Clarke,

W. R.

(1994). A study of various normalization procedures for within day electromyographic data, Journal of Electromyography and Kinesiology, 4(1), 47–59.

25.

Ko,

M. S.

, Lee,

J. A.

, Kang,

S. Y.

, & Jeon,

H. S.

(2015). Effect of Adeli suit treatment on gait in a child with cerebral palsy: a single-subject report. Physiotherapy Theory and Practice, 31(4), 275–282. https://doi.org/10.3109/09593985.2014.996307

26.

Lamontagne,

, Malouin,

, Richards,

, & Dumas,

(2002). Mechanisms of disturbed motor control in ankle weakness during gait after stroke, Gait & Posture, 15(3), 244–255.

27.

Lee,

D. R.

(2013). Effects of dynamic tubing gait training on knee joint neuromuscular control and kinematics during locomotion in children with cerebral palsy Graduate School, Yonsei University].

28.

Lee,

D. Y.

, Yoon,

J. C.

, Lee,

K. U.

, Jhoo,

J. H.

, Kim,

K. W.

, Lee,

J. H.

, & Woo,

J. I.

(1999). Reliability and validity of the Korean version of Short Blessed Test (SBT-K) as a dementia screening instrument, Journal of Korean Neuropsychiatric Association, 38(6), 1365–1375.

29.

Lee,

J. J.

, & You,

J. S. H.

(2017). Effects of Novel Guidance Tubing Gait on Electromyographic Neuromuscular Imbalance and Joint Angular Kinematics During Locomotion in Hemiparetic Stroke Patients. Archives Physical Medicine and Rehabilitation, 98(12), 2526–2532. https://doi.org/10.1016/j.apmr.2017.04.018

30.

Lee,

J. J.

, & You,

S. H.

(2016). Movement measurement validity and reliability of the image J program for kinematic analysis, Journal of Mechanics in Medicine and Biology, 16(08), 1640028.

31.

Lee,

S. M.

, Cynn,

H. S.

, Yi,

C. H.

, Yoon,

T. L.

, & Lee,

J. H.

(2017). Wearable tubing assistive walking device immediately enhances gait parameters in subjects with stroke: A randomized controlled study. NeuroRehabilitation, 40(1), 99–107. https://doi.org/10.3233/nre-161394

32.

Levin,

M. F.

, Selles,

R. W.

, Verheul,

M. H.

, & Meijer,

O. G.

(2000). Deficits in the coordination of agonist and antagonist muscles in stroke patients: implications for normal motor control. Brain Research, 853(2), 352–369. https://doi.org/10.1016/s0006-8993(99)02298-2

33.

Levin,

M. F.

, Selles,

R. W.

, Verheul,

M. H.

, & Meijer,

O. G.

(2000). Deficits in the coordination of agonist and antagonist muscles in stroke patients: implications for normal motor control, Brain Research, 853(2), 352–369.

34.

Major,

M. J.

, Fatone,

, & Roth,

E. J.

(2013). Validity and reliability of the Berg Balance Scale for community-dwelling persons with lower-limb amputation, Archives of Physical Medicine and Rehabilitation, 94(11), 2194–2202.

35.

Manca,

, Ferraresi,

, Cosma,

, Cavazzuti,

, Morelli,

, & Benedetti,

(2014). Gait patterns in hemiplegic patients with equinus foot deformity. BioMed Research International, 2014.

36.

Mao,

Y. R.

, Lo,

W. L.

, Lin,

, Li,

, Xiao,

, Raghavan,

, & Huang,

D. F.

(2015). The Effect of Body Weight Support Treadmill Training on Gait Recovery, Proximal Lower Limb Motor Pattern, and Balance in Patients with Subacute Stroke. Biomedical Research International, 2015, 175719. https://doi.org/10.1155/2015/175719

37.

Martins,

, Cordovil,

, Oliveira,

, Pinho,

, Diniz,

, & Vaz,

J. R.

(2019). The Immediate Effects of a Dynamic Orthosis on Gait Patterns in Children With Unilateral Spastic Cerebral Palsy: A Kinematic Analysis. Frontier Pediatrics, 7, 42. https://doi.org/10.3389/fped.2019.00042

38.

McCain,

K. J.

, Pollo,

F. E.

, Baum,

B. S.

, Coleman,

S. C.

, Baker,

, & Smith,

P. S.

(2008). Locomotor treadmill training with partial body-weight support before overground gait in adults with acute stroke: a pilot study. Archives Physical Medicne and Rehabilitation, 89(4), 684–691. https://doi.org/10.1016/j.apmr.2007.09.050

39.

Mulroy,

S. J.

, Klassen,

, Gronley,

J. K.

, Eberly,

V. J.

, Brown,

D. A.

, & Sullivan,

K. J.

(2010). Gait parameters associated with responsiveness to treadmill training with body-weight support after stroke: an exploratory study, Physical Therapy, 90(2), 209–223.

40.

Mutlu,

, Livanelioglu,

, & Gunel,

M. K.

(2008). Reliability of Ashworth and Modified Ashworth scales in children with spastic cerebral palsy, BMC Musculoskeletal Disorders, 9(1), 1–8.

41.

Page,

(2000). Clinical force production of Thera-Band^∧(! R) elastic bands. A-A, Journal of Orthopaedic & Sports Physical Therapy, 30, 48.

42.

Park,

, Hwang,

, & YOU,

J. S. H.

(2021). Comparative effectiveness of robot-interactive gait training with and without ankle robotic control in patients with brain damage. Journal of Mechanics in Medicine and Biology, 2140035.

43.

Park,

, Oh-Park,

, Dohle,

, Bialek,

, Friel,

, Edwards,

, Krebs,

H. I.

, & You,

J. S. H.

(2020). Effects of innovative hip-knee-ankle interlimb coordinated robot training on ambulation, cardiopulmonary function, depression, and fall confidence in acute hemiplegia. NeuroRehabilitation, 46(4), 577–587. https://doi.org/10.3233/nre-203086

44.

Patterson,

R. M.

, Stegink Jansen,

C. W.

, Hogan,

H. A.

, & Nassif,

M. D.

(2001). Material properties of thera-band tubing, Physical Therapy, 81(8), 1437–1445.

45.

Richie

D. H.

Jr (2021). HumanWalking: The Gait Cycle. In Pathomechanics of Common Foot Disorders (pp. 45–61). Springer.

46.

Riener,

, Lünenburger,

, Maier,

I. C.

, Colombo,

, & Dietz,

(2010). Locomotor Training in Subjects with Sensori Motor Deficits: An Overview of the Robotic Gait Orthosis Lokomat. Journal of Healthcare Engineering, 1, 517674. https://doi.org/10.1260/2040-2295.1.2.197

47.

Schwartz,

, & Meiner,

(2015). Robotic-assisted gait training in neurological patients: who may benefit? Annals of Biomedical Engineering, 43(5), 1260–1269.

48.

Shin,

Y. K.

, Lee,

D. R.

, Kim,

D. H.

, Lee,

J. J.

, You,

S. J.

, Yi,

C. H.

, & Jeon,

H. S.

(2014). Effects of novel tubing gait on neuromuscular imbalance in cerebral palsy. NeuroRehabilitation, 35(3), 587–596. https://doi.org/10.3233/nre-141154

49.

Sloan,

, Sinclair,

, Thompson,

, Taylor,

, & Pentland,

(1992). Inter-rater reliability of the modified Ashworth Scale for spasticity in hemiplegic patients, International Journal of Rehabilitation Research, 15(2), 158–161.

50.

Swank,

, Almutairi,

, Wang-Price,

, & Gao,

(2020). Immediate kinematic and muscle activity changes after a single robotic exoskeleton walking session post-stroke, Topics in Stroke Rehabilitation, 27(7), 503–515.

51.

Viveiro,

L. A. P.

, Gomes,

G. C. V.

, Bacha,

J. M. R.

, Junior,

N. C.

, Kallas,

M. E.

, Reis,

, Jacob Filho,

, & Pompeu,

J. E.

(2019). Reliability, validity, and ability to identity fall status of the Berg Balance Scale, Balance Evaluation Systems Test (BESTest), Mini-BESTest, and Brief-BESTest in older adults who live in nursing homes. E-E, Journal of Geriatric Physical Therapy, 42(4), 54.

52.

Yagura,

, Hatakenaka,

, & Miyai,

(2006). Does therapeutic facilitation add to locomotor outcome of body weight–supported treadmill training in nonambulatory patients with stroke? A randomized controlled trial, Archives of Physical Medicine and Rehabilitation, 87(4), 529–535.