Effects of inclined treadmill training on inadequate ankle control during walking in individuals after stroke: A pilot randomized controlled trial

Abstract

BACKGROUND:

Inadequate ankle control influences walking ability in people after stroke. Walking on inclined surface activates ankle muscles and movements. However, the effect of inclined treadmill training on ankle control is not clear.

OBJECTIVE:

To investigate the effects of inclined treadmill training on ankle control in individuals with inadequate ankle control after chronic stroke.

METHODS:

This was a randomized single-blinded study. Eighteen participants were randomly assigned to receive 12 sessions of 30 min inclined (n = 9) or regular (n = 9) treadmill training and 5 min over-ground walking training. The outcomes included ankle control during walking, muscle strength of affected leg, walking performance, and stair climbing performance.

RESULTS:

Inclined treadmill training significantly improved ankle dorsiflexion at initial contact (p = 0.002), increased tibialis anterior activities (p = 0.003 at initial contact, p = 0.006 in swing phase), and decreased dynamic plantarflexors spasticity (p = 0.027) as compared with regular treadmill training. Greater improvements were also shown in stair climbing with affected leg leading (p = 0.006) and affected knee extensors strength (p = 0.002) after inclined treadmill training.

CONCLUSIONS:

Inclined treadmill training was proposed to improve inadequate ankle control after chronic stroke. Inclined treadmill training also improved the stair climbing ability accompanied with increased muscle strength of the affected lower extremity.

Keywords

Rehabilitation ankle inclined treadmill training stroke walking

1 Introduction

Stroke is a common neurological disease due to a cerebrovascular disorder causing motor impairments, such as poor motor control, and could lead to a long-term disability (Mozaffarian et al., 2015). Impaired motor control on ankle joint after stroke was characterized by decreased ankle plantarflexion at toe off, and decreased dorsiflexion at both initial contact and swing phase during walking (Chen, Patten, Kothari, & Zajac, 2005; Lin, Yang, Cheng, & Wang, 2005; Olney & Richards, 1996). Previous studies have demonstrated abnormal muscle activation, spasticity, reduced muscle strength, and altered sensation after stroke, which may attribute to inadequate motor control for gait ability (Chen et al., 2003; Olney & Richards, 1996; Patterson et al., 2008). Specifically, the decreased activation of the affected tibialis anterior (TA) were observed at swing phase, which may further impact gait speed, step length and the postural stability during walking (Laufer, Dickstein, Chefez, & Marcovitz, 2001; Sousa, Silva, & Santos, 2015). Muscle strength of dorsiflexor was also suggested to play a role in ankle control during walking, and was found to correlate with walking speed and temporal symmetry (Dorsch, Ada, Canning, Al-Zharani, & Dean, 2012; Lin et al., 2005). In addition, spasticity and/or stiffness of plantarflexors was proposed to result in the inadequate ankle dorsiflexion during gait cycles, and correlated negatively with walking speed and spatial symmetry (Lamontagne, Malouin, & Richards, 2001; Lin et al., 2005). The activation of plantarflexors at stance phase, which associated with vertical support and forward propulsion, was decreased in patients with stroke (Kitatani et al., 2016; Neptune, Kautz, & Zajac, 2001; Turns, Neptune, & Kautz, 2007). Therefore, gait ability relies on normal functioning of ankle control, and impaired motor control after stroke could result in significant impact on walking performance.

Walking on the inclined surface has been suggested to potentially induce better motor control and improve gait performance. The ankle movements and gait patterns could adapt to inclination levels of the surface in healthy adults. Increased step length and reduced cadence was observed when healthy individuals were walking uphill (Phan, Blennerhassett, Lythgo, Dite, & Morris, 2013). When the walking surface became steeper, people would perform greater dorsiflexion during stance phase and greater plantarflexion during toe off (Lay, Hass, & Gregor, 2006; McIntosh, Beatty, Dwan, & Vickers, 2006). On the other hand, patients with stroke have exhibited increasing angle of ankle dorsiflexion, hip and knee flexion during swing phase as well as at initial contact in response to the increasing incline angle (Moreno, Mendes, & Lindquist, 2011). In addition to decreased cadence and increased stride length, inclined walking also led to symmetry in swing time, and increased stance time on the affected limb in individuals with stroke (Moreno et al., 2011; Werner, Lindquist, Bardeleben, & Hesse, 2007). However, the response of affected muscle activities to the inclination was inconsistent. One study has shown increased activity of affected TA at stance phase (Phadke, 2012), while the others have indicated no significant change in any muscle on the affected lower limb during gait cycles in people with stroke (Mohammadi & Phadke, 2017; Mohammadi, Talebian, Phadke, Yekaninejad, & Hadian, 2016; Werner et al., 2007).

Inclined treadmill training has been suggested to provide beneficial effects on walking. Carda et al. has demonstrated that treadmill training with 5% inclination improved gait speed and endurance in people with chronic stroke (Carda, Invernizzi, Baricich, Cognolato, & Cisari, 2013). The addition of inclination at 10% enhanced the effects of treadmill training with or without partial body weight support on gait performance and balance in this population (Gama et al., 2015; Yoon & Kang, 2016). However, it has not yet been explored whether inclined treadmill training could improve gait in people with inadequate ankle control after stroke, not to mention its effect on motor control of the ankle joint specifically. Therefore, this study aimed to investigate the effects of inclined treadmill training on inadequate ankle control during walking in people with chronic stroke. We hypothesized that a 12-session of inclined treadmill training would result in better motor control on the ankle joint due to the challenge of inclination as compared with regular treadmill training in individuals with inadequate ankle control after stroke. In addition, the improvement in ankle control would be accompanied by improvement in gait and stair-climbing performance.

2 Methods

2.1 Participants

Participants were recruited from a district hospital in Taiwan between December 2017 and December 2019. The inclusion criteria were as follows: (1) hemiparesis due to single stroke for at least 6 months, (2) ability to walk independently with or without an assistive device, (3) with inadequate ankle control as indicated by active ankle dorsiflexion less than 0° at initial contact during walking and passive range of motion (PROM) of dorsiflexion at least to 10° in supine (Yang et al., 2018), and (4) Mini-Mental State Examination (MMSE) score≥24. Exclusion criteria were: (1) any comorbidity or disability that would preclude walking training; (2) any uncontrolled health condition for which exercise was contraindicated; and (3) any neurological or orthopedic disease or condition that might interfere with the study. The study protocol was approved by the Institutional Review Board of the Cardinal Tien Hospital and the Institutional Review Board of National Yang-Ming University. This study was registered at http://www.clinicaltrials.in.th/ (TCTR20171224001) on December 24, 2017 and conformed to the CONSORT checklist. All participants were informed about the study protocol and provided the written consent.

2.2 Experimental design

This was a single-blinded, randomized controlled trial with pre- and post- measurements. The basic characteristic data, including age, gender, the type of stroke, the affected side, duration after stroke, the assistive device used and the MMSE scores, were obtained by the blinded assessor. The participants were randomly assigned to either the experimental group or the control group according to a computer-generated randomization sequence. The sequence was generated by a person who was not involved in this study, and was concealed from the researcher who assigned the allocation until the pre-test assessment was done. Participants in the experimental and control group received two 15 min of inclined treadmill training and regular treadmill training respectively, followed by 5 min of over-ground walking training for 12 sessions (2-3 times per week in 4– 6 weeks). All participants maintained their routine activities and/or regular treatment. The outcome measurements were done by another physical therapist (assessor) who was blinded to the group assignment.

2.3 Intervention protocol

A standard treadmill (RTM600 Treadmill, Biodex Medical Systems, Inc.) was used for all the treadmill training sessions conducted by the same physical therapist. The treadmill is equipped with front and side handrails and an emergency stop button for safety consideration. All participants were asked to walk without any ankle-foot orthosis during the training sessions. The therapist provided verbal instructions emphasizing the active control of the ankle joints during walking, and asked participants to focus on the movements of their lower legs.

2.3.1 Inclined treadmill training

In the experimental group, the training speed was set at the comfortable speed on the treadmill for each participant. The initial inclination was set at 0% and was increased by 5% (2.85°) every 5 mins as tolerated. There was a 5 min rest between the two inclined treadmill training periods, and a 5 min ambulation training was conducted after the treadmill training. The training speed in each session and the maximal inclination reached by each participant were recorded.

2.3.2 Regular treadmill training

Participants in the control group received two 15 min of regular treadmill training followed by 5 min of over-ground walking training. The initial training speed was set at the comfortable speed determined by each participant, and the inclination was set at 0% throughout the training. In order to provide a similar training intensity as in the experimental group, the treadmill speed was increased by 0.1 mph (0.045 m/s) every 5 mins as tolerated during regular treadmill training. The speed used in each training session was recorded.

2.4 Outcome measurements

2.4.1 Primary outcomes

The primary outcomes were the ankle control during walking, including angle and coefficient of variance (CV) of ankle dorsiflexion at initial contact, muscle activities of dorsiflexors at initial contact and swing phase and plantarflexors during stance, and dynamic spasticity of plantarflexors during stance.

Participants were asked to walk on bare feet at a comfortable speed for two walking trials. The walkway was 10 meters long and the middle three gait cycles of each trial were used for analysis by BIOPAC system (MP 150WMW, BIOPAC system, INC, USA). Two footswitches were placed under the heel and toe of the affected leg to determine the swing and stance phase during walking. In addition, twin-axis electronic goniometers (SG110, Biometics Ltd, UK) were attached to the affected knee and ankle joint to measure the range of motion (ROM) during walking. The axis center of electronic goniometer for the ankle was positioned at the lateral malleolus, with one endblock placed in line with fibula and the other placed along the line between the lateral malleolus and the fifth metatarsal head. The axis center of electronic goniometer for the knee was aligned with the knee joint, with one endblock placed alongside the lateral aspect of femur and the other placed along the line between the lateral epicondyle and the lateral malleolus (Winter & Scott, 1991). Bipolar Ag-AgCl electrodes were placed on the muscle belly of tibialis anterior (on the one third of the line between fibular head and medial malleolus) and medial gastrocnemius (on 5 finger-widths distal to the popliteal fossa and 2 cm medial to the midline) of the affected leg according to the Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles (SENIAM) guidelines (Hermens, Freriks, Disselhorst-Klug, & Rau, 2000).

Ankle dorsiflexion movement at initial contact were measured by the electronic goniometer. The signals were recorded using AcqKnowledge software version 3.7.5 with BIOPAC System and were stored for offline analysis. The sampling rate was set at 1000 Hz. At initial contact, the positive value from the goniometer indicated ankle dorsiflexion and the negative value indicated ankle plantarflexion. The CV of the dorsiflexion angle represented the movement variations, and the smaller CV indicated better ankle control. The CV was calculated by the following formula (Cheng, Yang, Cheng, Lin, & Wang, 2010): $\begin{matrix} CV & = (standard deviation of ankle dorsiflexion / \\ mean of ankle dorsiflexion) \times 100 % \end{matrix}$ (1)

Muscle activities of TA and medial gastrocnemius (MG) during gait cycles were measured by electromyography (EMG). The signals were acquired at a sampling frequency of 1000 Hz. The band-stop Butterworth filter at 60 Hz and the high-pass Butterworth filter at 20 Hz were used to minimize the effects of noise and movement artifacts. The participants were seated, and the electrographic activities were recorded during the maximal isometric contractions of dorsiflexors and plantarflexors against manual resistance provided by the assessor. The signals were calculated into root mean square (RMS), and the data during walking were normalized by following formula (Wakasa & Fukuda, 2013): $\begin{matrix} % RMS EMG & = (walking RMS EMG / \\ maximal RMS EMG) \times 100 % \end{matrix}$ (2)

The dynamic spasticity of plantarflexors was represented by spasticity index, defined as the value of the electromyography-lengthening slope of MG during the stance phase of gait cycles. The positive slope indicated the velocity-sensitive muscle activation suggesting the presence of dynamic spasticity during gait (Lamontagne et al., 2001). The muscle-lengthening velocity of MG was calculated with the model developed by Winter and Scott, through measuring the displacements of knee and ankle joints by two electronic goniometers. Slope relationship was analyzed by linear regression between EMG activities and muscle-lengthening velocity of MG (Winter & Scott, 1991).

2.4.2 Secondary outcomes

Secondary outcomes included muscle strength of the lower extremity, walking performance, and stair climbing performance. The muscle strength of hip extensors, knee extensors, ankle dorsiflexors, and ankle plantarflexors of the affected leg were measured by handheld dynamometer (Power Track II; Jtech Medical Industries Inc., Herber City, UT) (Cooper, Alghamdi, Alghamdi, Altowaijri, & Richardson, 2012; Martins et al., 2016). To measure the strength of hip extensor, the participants lied in supine position with hip and knee flexed to 90° and waist stabilized on bed. The handheld dynamometer was placed on the distal third of posterior thigh. For knee extensors, the participants were seated with hip and knee flexed to 90° and thigh stabilization. The dynamometer was held against the anterior surface of lower leg. The ankle plantarflexors were measured in supine position with hip and knee flexion at 90°, and the lower leg was supported by a wooden block. The testing position for dorsiflexors was supine with extended legs, and waist and shin were stabilized on bed. The handheld dynamometer was held perpendicularly over the tarsal bones for ankle dorsiflexors and over the metatarsal joints for plantarflexors. Participants were asked to exert a maximal force against the handheld dynamometer for 5 seconds during each trial. Three trials for each muscle group were averaged for analysis.

Fig. 1

Study flowchart.

Two wearable Physilog movement inertial sensors (Gait Up, Lausanne, Switzerland) were used to measure the walking performance. Participants were asked to walk 10 meters at a comfortable speed in their own shoes without any orthosis. The velocity (cm/s), cadence (steps/min), and stride length (cm) were recorded twice, and were averaged for further analysis.

To assess the stair climbing performance, participants were asked to climb up four 18 cm steps “safely as fast as possible” for a total of 4 trials, with the affected leg leading and the unaffected leg leading for two trials, respectively. The complete time for stair climbing was recorded, and the results in two trials were averaged.

2.5 Sample size

The effect size was not available for inclined treadmill training in improving ankle control after stroke up to present. Thus, we set the effect size of 0.5 for ankle control improvement, power of 0.8 and two-tailed alpha level of 0.05 to calculate the sample size in present study. A total sample size was required to be 26 (13 per group) calculated by G^*power v3.1.9.7. However, we only recruited 18 participants in this study due to difficulty in scheduling the training sessions into the routine services of the rehabilitation department of a local hospital. In particular, we had to stop the recruitment due to the breakout of COVID-19 at the end of year 2019.

2.6 Statistical analysis

All data were analyzed with the SPSS 24.0 software (SPSS Inc., USA). Descriptive statistics were generated and presented in median (interquartile, IQR) for all variables. Non-parametric test was used for comparisons due to the heterogeneity of the study sample (Campbell & Gardner, 1988). The demographic characteristics and baseline data were analyzed using Mann-Whitney U test for quantitative variables, and chi-square test for categorical variables. The Wilcoxon signed-rank test was used for within-group comparisons. Change values were calculated by subtracting pre-training data from the post-training data, and were analyzed using Mann-Whitney U-test for between group comparisons. Statistical significance level was set at 0.05. Effect size for Mann-Whitney U-test was calculated by dividing the Z-score by the square root of N (N = the total number of observations), as 0.1 indicating a small effect, 0.3 a medium effect, and 0.5 a large effect (Tomczak & Tomczak, 2014).

3 Results

Twenty-two individuals with chronic stroke were recruited in this study, and 18 of them met the criteria and agreed to participate (n = 9 for each group). One participant in the control group failed to finish the study due to health problems, and thus 17 participants completed the study protocol. Intention-to-treat analysis as imputing the missing data with mean values of the group was used. The study flowchart is shown in Fig. 1. There were no significant group differences in demographic data (Table 1). The median (IQR) of the maximal treadmill training speed was 58.5 (19.5) cm/s (72.9 % of the comfortable speed when walking on the ground) in the experimental group, and 42.5 (10.0) cm/s (70.8 % of the comfortable speed when walking on the ground) in the control group. Two participants in the experimental group could not tolerate the final target inclination, and the average maximal inclination of treadmill training was 9.3±1.4 % (ranging from 6% to 10%).

Table 1
Demographic characteristics of experimental and control group

Experimental group (N = 9) Control group (N = 9) p value

Age (year) 54.0 (4.0) 46.0 (21.0) 0.718

Height (cm) 172.0 (6.0) 176.0 (10.0) 0.570

Weight (kg) 75.0 (12.0) 76.0 (30.0) 0.918

Post-onset duration (mos) 12.2 (2.5) 12.5 (38.4) 0.503

MMSE 28.0 (2.0) 29.0 (1.0) 0.339

Gender

Male 9 8 0.299

Female 0 1

Type of stroke

Infarction 7 4 0.100

Hemorrhage 2 5

Side of stroke

Right 3 2 0.353

Left 6 7

Assistive device

AFO 1 1 1.000

Quadricane 1 3 0.214

	Experimental group (N = 9)	Control group (N = 9)	p value
Age (year)	54.0 (4.0)	46.0 (21.0)	0.718
Height (cm)	172.0 (6.0)	176.0 (10.0)	0.570
Weight (kg)	75.0 (12.0)	76.0 (30.0)	0.918
Post-onset duration (mos)	12.2 (2.5)	12.5 (38.4)	0.503
MMSE	28.0 (2.0)	29.0 (1.0)	0.339
Gender
Male	9	8	0.299
Female	0	1
Type of stroke
Infarction	7	4	0.100
Hemorrhage	2	5
Side of stroke
Right	3	2	0.353
Left	6	7
Assistive device
AFO	1	1	1.000
Quadricane	1	3	0.214

Values are median (interquartile range) or frequency. Abbreviations: mos: months; MMSE: mini-mental state examination; AFO: ankle-foot orthosis.

After 12 sessions of intervention, the movement angle of the ankle was significantly increased toward dorsiflexion at initial contact in the experimental group (– 14.3° to – 10.7°, p = 0.021). There was a significant between-group difference in the angle change at initial contact (p = 0.002, effect size: 0.74). However, no significant change in CV of ankle dorsiflexion at initial contact was shown in both groups (Table 2). The EMG activities of the affected TA were significantly increased at initial contact (p = 0.008) and during the swing phase (p = 0.038) in the experimental group. In addition, there were significant decreases in the EMG activities of the affected MG during the stance phase after 12 sessions of inclined treadmill training (p = 0.021). Furthermore, those changes in the EMG activities in the experimental group were significantly more than in the control group (TA at initial contact: p = 0.003, effect size: 0.69; TA during swing phase: p = 0.006, effect size: 0.64; MG during stance phase: p = 0.020, effect size: 0.55). The experimental group also showed significant decrease in the spasticity index after training (p = 0.038), while there was no significant change in the control group (p = 0.859). Significant between-group difference was observed in the changes in the spasticity index (p = 0.027, effect size: 0.52; Table 2).

Table 2

The effects on ankle control in both groups. Ankle control as the primary outcomes, including ankle dorsiflexion movement, muscle activities of dorsiflexors and plantarflexors, and dynamic spasticity of plantarflexors during walking. Table shows the results of ankle control during walking measured before and after intervention in the experimental and control group

	Experimental group (N = 9)			Control group (N = 9)			p value	ES
	Pre	Post	p value	Pre	Post	p value
Ankle movement during gait
Dorsiflexion angle at IC (°) changes	– 14.3 (7.5)	– 11.2 (3.7)	0.021^*	– 14.9 (7.3)	– 15.8 (2.2)	0.086
		3.1 (5.0)			– 1.2 (1.0)		0.002#	0.74
CV of dorsiflexion at IC (%) changes	10.5 (3.1)	11.8 (1.4)	0.676	13.8 (7.6)	15.1 (7.3)	0.515
		0.8 (4.0)			1.3 (7.6)		0.440	0.18
Muscle activities during gait
TA activation at IC (%) changes	44.0 (8.9)	56.4 (34.7)	0.008^*	44.0 (34.3)	39.4 (12.2)	0.374
		21.9 (22.9)			– 6.1 (23.6)		0.003#	0.69
TA activation in swing phase (%) changes	44.0 (19.0)	54.2 (9.1)	0.038^*	48.3 (35.3)	40.7 (2.1)	0.260
		12.0 (12.7)			– 7.6 (29.7)		0.006#	0.64
MG activation in stance phase (%) changes	213.3 (141.3)	129.0 (91.4)	0.021^*	123.9 (90.3)	129.5 (61.2)	0.441
		– 67.3 (54.0)			46.6 (71.7)		0.020#	0.55
Dynamic spasticity
Spasticity index changes	2.5 (13.1)	– 2.2 (4.8)	0.038^*	0.8 (3.0)	1.5 (2.1)	0.859
		– 4.8 (4.9)			0.2 (2.2)		0.027#	0.52

Values are median (interquartile range). Abbreviations: ES: effect size; IC: initial contact; CV: coefficient of variation; TA: tibialis anterior; MG: medial gastrocnemius. ^*, p < 0.05 for within group comparison; #, p < 0.05 for between group comparison.

The results showed significant increases in the strength of knee extensors (p = 0.011), ankle dorsiflexors (p = 0.011) and ankle plantarflexors (p =0.036) of the affected leg after the inclined treadmill training, and significantly improved strength of hip extensors (p = 0.015), ankle dorsiflexors (p = 0.013) and ankle plantarflexors (p = 0.011) of the affected leg after the regular treadmill training. Only the knee extensors muscle strength improved more in the experimental group as compared with the control group (p = 0.002, effect size: 0.72; Table 3).

Table 3

The effects on muscle strength of the affected lower extremity in both groups. The table shows the results of muscle strength assessed before and after intervention in the experimental and control group

	Experimental group (N = 9)			Control group (N = 9)			p value	ES
	Pre	Post	p value	Pre	Post	p value
Muscle strength
Hip extensors strength (N) changes	52.8 (12.0)	62.3 (10.3)	0.110	50.6 (7.8)	57.2 (6.6)	0.015^*
		5.9 (2.9)			3.9 (3.7)		0.354	0.22
Knee extensors strength (N) changes	100.2 (18.1)	108.3 (15.3)	0.011^*	103.5 (21.7)	106.3 (19.9)	0.374
		14.0 (8.3)			0.14 (7.0)		0.002#	0.72
Ankle dorsiflexors strength (N) changes	33.4 (28.3)	39.8 (32.7)	0.011^*	47.0 (13.2)	56.7 (16.1)	0.013^*
		6.3 (4.6)			13.9 (10.3)		0.080	0.41
Ankle plantarflexors strength (N) changes	49.2 (22.0)	57.2 (25.7)	0.036^*	41.1 (26.4)	55.9 (11.9)	0.011^*
		3.67 (6.0)			9.45 (14.7)		0.134	0.35

Values are median (interquartile range). Abbreviation: ES: effect size. ^*, p < 0.05 for within group comparison; #, p < 0.05 for between group comparison.

As for the walking performance, no significant within-group differences were observed in gait velocity and stride length in both groups. Only the cadence was significantly decreased in experimental group (p = 0.021), and a significant between-group difference was shown in the changes in cadence (p = 0.015, effect size: 0.53; Table 4). Both the inclined and regular treadmill training improved the ability to climb stairs. Participants in both groups walked faster in stair climbing with the unaffected leg leading (control: p = 0.015; experimental: p = 0.044) and with the affected leg leading (control: p = 0.011; experimental: p = 0.008). There was a significant between-group difference when climbing stairs with the affected leg leading (p = 0.006, effect size: 0.64; Table 4).

Table 4

The effects on walking and stair climbing performance in both groups. The table shows the results of gait parameters and timed stair climbing performance before and after intervention in the experimental and control group

	Experimental group (N = 9)			Control group (N = 9)			p value	ES
	Pre	Post	p value	Pre	Post	p value
Walking performance
Velocity (cm/s) changes	66.0 (18.0)	67.0 (10.0)	0.594	62.0 (40.0)	59.0 (34.0)	0.953
		1.0 (6.5)			1.0 (5.5)		0.796	0.06
Cadence (steps/min) changes	100.0 (16.5)	96.8 (15.7)	0.021^*	79.2 (15.1)	78.9 (10.0)	0.953
		– 4.5 (4.3)			– 0.1 (3.7)		0.015#	0.53
Stride length (cm) changes	79.0 (10.2)	80.0 (4.8)	0.075	79.0 (43.5)	84.3 (21.5)	0.110
		6.0 (7.6)			2.0 (4.7)		0.797	0.06
Stair climbing
Unaffected leg leading (sec) changes	7.4 (1.5)	6.9 (1.6)	0.044^*	7.6 (0.8)	6.8 (1.1)	0.015^*
		– 0.3 (0.4)			– 0.6 (0.5)		0.257	0.27
Affected leg leading (sec) changes	7.4 (1.4)	6.7 (0.5)	0.008^*	8.1 (2.3)	7.8 (2.4)	0.011^*
		– 0.7 (0.4)			– 0.2 (0.2)		0.006#	0.64

Values are median (interquartile range). Abbreviation: ES: effect size. ^*, p < 0.05 for within group comparison; #, p < 0.05 for between group comparison.

4 Discussion

This study demonstrated that inclined treadmill training improved ankle control, including increasing TA activities and ankle dorsiflexion angle at initial contact, increasing TA activities during swing phase, and decreasing the dynamic spasticity of MG during stance phase in individuals with inadequate ankle control after stroke. In addition, such effects on ankle control were significantly better than regular treadmill training. We also noted that inclined treadmill training enhanced stair climbing abilities especially with the affected leg leading.

Motor control of ankle dorsiflexion during walking is influenced by activation of dorsiflexors and spasticity and/or stiffness of plantarflexors after stroke (Lamontagne, Malouin, Richards, & Dumas, 2002; Lin et al., 2005; Olney & Richards, 1996). In present study, the ankle movement improved at initial contact, which may result from increased TA activation and decreased MG dynamic spasticity after the inclined treadmill training. It was shown that TA activated more during inclined walking than during level walking to avoid tripping (Moreno et al., 2011; Phadke, 2012). We further demonstrated the positive effects of multiple sessions of inclined treadmill training on overground walking performance. Lamontagne et al. suggested that an increased TA activation may overcome the stiffness of plantarflexors, which allowed for normal dorsiflexion angle during swing phase (Lamontagne et al., 2002). On the other hand, the dynamic spasticity of ankle plantarflexors also influenced ankle dorsiflexion movement during walking (Cheng et al., 2010). In this study, the dynamic ankle plantarflexors spasticity indicated by spasticity index was reduced, accompanied with improved ankle control of dorsiflexion during walking. However, Gama et al. demonstrated no significant changes in angular variables of the ankle movement after twelve 20-min inclined treadmill training with 10% of inclination in individuals with chronic stroke (Gama et al., 2015). In their study, up to 30% of partial body weight support was used during inclined treadmill training, which may reduce the challenge for ankle control. Moreover, we only included participants with inadequate ankle control during walking, while other studies did not specify this impairment in their participants. Our results suggest that inclined treadmill training could be a treatment option, especially in people with stroke who demonstrated inadequate ankle control, due to its better effects on ankle control and stair climbing performance compared with regular treadmill training.

Several studies demonstrated increased walking speed after treadmill training in people with stroke (Mehrholz, Pohl, & Elsner, 2014; Nascimento, Boening, Galli, Polese, & Ada, 2021). However, we did not find any beneficial effect on walking speed in neither group. The inconsistent results may be caused by the treadmill training speed. In this study, participants were asked to focus on ankle control during inclined or regular treadmill walking which slowed down the training speed. The median (IQR) treadmill speed of experimental group and control group were 52.2 (22.8) cm/s and 37.5 (10.0) cm/s respectively. On the other hand, previous studies applied relatively fast speed during treadmill training in order to promote walking ability in individuals with stroke (Kuys, Brauer, & Ada, 2011; Pohl, Mehrholz, Ritschel, & Rückriem, 2002). Tyrell et al. reported that fast-speed walking improved gait deviations observed on hip and knee joints during treadmill walking (Tyrell, Roos, Rudolph, & Reisman, 2011). However, the effect of different speed used for treadmill walking on ankle control was not clear. We may suggest that the emphasis on ankle control instead of speed during inclined treadmill training exerted the specific effect for people with inadequate ankle control after stroke. Furthermore, improved ankle control during walking may possibly reduce the number of falls and may also decrease energy costs for carrying out the task (Chen et al., 2003; Patterson et al., 2008).

Previous studies found that the rectus femoris activated more during walking uphill (Lay, Hass, Richard Nichols, & Gregor, 2007), and multiple sessions of treadmill training with 5° inclination significantly increased muscle contraction of knee extensor during sit-to-stand (Kim, 2012). Choi et al. indicated that improvement of stair climbing ability was related to increased knee extensor strength (Choi, Yoo, Shin, & Lee, 2015). In present study, the inclined treadmill training resulted in greater effects than regular treadmill training on stairs with the affected leg leading, which may be due to the significant increase in knee extensors strength. However, we found that the beneficial effect of our inclined treadmill training (up to about 5.7°) on ankle dorsiflexors strength was not superior to the effect of regular treadmill training. Kim et al. indicated that although both their training programs with 5° and 10° of inclination increased the muscle contraction of tibialis anterior, only 10° inclined treadmill training resulted in better improvement than regular treadmill training in people with chronic stroke (Kim, 2012). Therefore, the effect of inclined treadmill training on muscle strength, especially the ankle, may be influenced by the level of treadmill inclination during training.

There are several limitations in this study. First, the sample size was relatively small. Despite the relatively small sample size, the effect size (0.52– 0.74) is rather large for ankle control. A larger, randomized controlled trial is needed to validate the reported benefits of the inclined treadmill training protocols established in current study. Second, this study did not measure the follow-up assessment, and we could not demonstrate whether the training effects could be maintained. Third, the therapist was not blinding to group assignment which may introduce bias. Fourth, present study included people with stroke who demonstrated inadequate ankle control, however, the heterogeneity of our sample population, such as type and side of stroke, post-stroke duration, should be noted.

5 Conclusions

Our results demonstrated the inclined treadmill training exerted greater effects on angular variable of ankle movement, muscle activation and dynamic spasticity during walking, compared with regular treadmill training. The inclined treadmill training program reported in current study could be a potential treatment specifically for the people with inadequate ankle control after stroke.

Footnotes

Acknowledgments

This work was supported by grant from the Ministry of Science and Technology (MOST 106-2314-B-010-040-MY3) and National Health Research Institutes (NHRI-EX109-10913PI). The authors would like to thank all participants in this study.

Conflict of interest

None of the authors have any conflict of interest to report.

References

Campbell,

M. J.

, & Gardner,

M. J.

(1988) Calculating confidence intervals for some non-parametric analyses. British Medical Journal (Clinical research ed.), 296(6634), 1454–1456. https://doi.org/10.1136/bmj.296.6634.1454

Carda,

, Invernizzi,

, Baricich,

, Cognolato,

, & Cisari,

(2013) Does altering inclination alter effectiveness of treadmill training for gait impairment after stroke? A randomized controlled trial. Clinical Rehabilitation, 27(10), 932–938. https://doi.org/10.1177/0269215513485592

Chen,

C. L.

, Chen,

H. C.

, Tang,

S. F.

, Wu,

C. Y.

, Cheng,

P. T.

, & Hong,

W. H.

(2003) Gait performance with compensatory adaptations in stroke patients with different degrees of motor recovery. American Journal of Physical Medicine & Rehabilitation, 82(12), 925–935. https://doi.org/10.1097/01.phm.0000098040.13355.b5

Chen,

, Patten,

, Kothari,

D. H.

, & Zajac,

F. E.

(2005) Gait differences between individuals with post-stroke hemiparesis and non-disabled controls at matched speeds. Gait & Posture, 22(1), 51–56. https://doi.org/10.1016/j.gaitpost.2004.06.009

Cheng,

J. S.

, Yang,

Y. R.

, Cheng,

S. J.

, Lin,

P. Y.

, & Wang,

R. Y.

(2010) Effects of combining electric stimulation with active ankle dorsiflexion while standing on a rocker board: a pilot study for subjects with spastic foot after stroke. Archives of Physical Medicine and Rehabilitation, 91(4), 505–512. https://doi.org/10.1016/j.apmr.2009.11.022

Choi,

, Yoo,

, Shin,

, & Lee,

(2015) The effects of stepper exercise with visual feedback on strength, walking, and stair climbing in individuals following stroke. Journal of Physical Therapy Science, 27(6), 1861–1864. https://doi.org/10.1589/jpts.27.1861

Cooper,

, Alghamdi,

G. A.

, Alghamdi,

M. A.

, Altowaijri,

, & Richardson,

(2012) The relationship of lower limb muscle strength and knee joint hyperextension during the stance phase of gait in hemiparetic stroke patients. Physiotherapy Research International: The Journal for Researchers and Clinicians in Physical Therapy, 17(3), 150–156. https://doi.org/10.1002/pri.528

Dorsch,

, Ada,

, Canning,

C. G.

, Al-Zharani,

, & Dean,

(2012) The Strength of the Ankle Dorsiflexors Has a Significant Contribution to Walking Speed in People Who Can Walk Independently After Stroke: An Observational Study. Archives of Physical Medicine and Rehabilitation, 93(6), 1072–1076. https://dx-doi-org.web.bisu.edu.cn/10.1016/j.apmr.2012.01.005

Gama,

G. L.

, de Lucena Trigueiro,

L. C.

, Simao,

C. R.

, de Sousa,

A. V.

, de Souza,

E. S. E. M.

, Galvao,

E. R.

, & Lindquist,

A. R.

(2015) Effects of treadmill inclination on hemiparetic gait: controlled and randomized clinical trial. American Journal of Physical Medicine & Rehabilitation, 94(9), 718–727. https://doi.org/10.1097/phm.0000000000000240

10.

Hermens,

H. J.

, Freriks,

, Disselhorst-Klug,

, & Rau,

(2000) Development of recommendations for SEMG sensors and sensor placement procedures. Journal of Electromyography and Kinesiology: Official Journal of the International Society of Electrophysiological Kinesiology, 10(5), 361–374. https://doi.org/10.1016/s1050-6411(00)00027-4

11.

Kim,

S.-G.

(2012) Effect of treadmill gradient training on lower limb muscle activity in chronic stroke patient. Journal of the Korea Academia-Industrial Cooperation Society, 13(1), 220–226.

12.

Kitatani,

, Ohata,

, Aga,

, Mashima,

, Hashiguchi,

, Wakida,

, ... Yamada,

(2016) Descending neural drives to ankle muscles during gait and their relationships with clinical functions in patients after stroke. Clinical Neurophysiology, 127(2), 1512–1520. https://doi.org/10.1016/j.clinph.2015.10.043

13.

Kuys,

S. S.

, Brauer,

S. G.

, & Ada,

(2011) Higher-intensity treadmill walking during rehabilitation after stroke in feasible and not detrimental to walking pattern or quality: a pilot randomized trial. Clinical Rehabilitation, 25(4), 316–326. https://doi.org/10.1177/0269215510382928

14.

Lamontagne,

, Malouin,

, & Richards,

C. L.

(2001) Locomotor-specific measure of spasticity of plantarflexor muscles after stroke. Archives of Physical Medicine and Rehabilitation, 82(12), 1696–1704. https://doi.org/10.1053/apmr.2001.26810

15.

Lamontagne,

, Malouin,

, Richards,

C. L.

, & Dumas,

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

16.

Laufer,

, Dickstein,

, Chefez,

, & Marcovitz,

(2001) The effect of treadmill training on the ambulation of stroke survivors in the early stages of rehabilitation: a randomized study. Journal of Rehabilitation Research and Development, 38(1), 69–78.

17.

Lay,

A. N.

, Hass,

C. J.

, & Gregor,

R. J.

(2006) The effects of sloped surfaces on locomotion: a kinematic and kinetic analysis. Journal of Biomechanics, 39(9), 1621–1628. https://doi.org/10.1016/j.jbiomech.2005.05.005

18.

Lay,

A. N.

, Hass,

C. J.

, Richard Nichols,

, & Gregor,

R. J.

(2007) The effects of sloped surfaces on locomotion: An electromyographic analysis. Journal of Biomechanics, 40(6), 1276–1285. https://doi.org/10.1016/j.jbiomech.2006.05.023

19.

Lin,

P. Y.

, Yang,

Y. R.

, Cheng,

S. J.

, & Wang,

R. Y.

(2005) The Relation Between Ankle Impairments and Gait Velocity and Symmetry in People With Stroke. Archives of Physical Medicine and Rehabilitation, 87(4), 562–568. https://doi.org/10.1016/j.apmr.2005.12.042

20.

Martins,

J. C.

, Aguiar,

L. T.

, Lara,

E. M.

, Moura,

J. B.

, Teixeira-Salmela,

L. F.

, & Faria,

C. D. C. d. M.

(2016) Assessment of the strength of the lower limb muscles in subjects with stroke with portable dynamometry: a literature review. Fisioterapia em Movimento, 29(1), 193–208.

21.

McIntosh,

A. S.

, Beatty,

K. T.

, Dwan,

L. N.

, & Vickers,

D. R.

(2006) Gait dynamics on an inclined walkway. Journal of Biomechanics, 39(13), 2491–2502. https://doi.org/10.1016/j.jbiomech.2005.07.025

22.

Mehrholz,

, Pohl,

, & Elsner,

(2014) Treadmill training and body weight support for walking after stroke. The Cochrane Database of Systematic Reviews(1), Cd002840. https://doi.org/10.1002/14651858.CD002840.pub3

23.

Mohammadi,

, & Phadke,

C. P.

(2017) Effects of treadmill incline and speed on peroneus longus muscle activity in persons with chronic stroke and healthy subjects. Gait & Posture, 54, 221–228. https://doi.org/10.1016/j.gaitpost.2017.03.006

24.

Mohammadi,

, Talebian,

, Phadke,

C. P.

, Yekaninejad,

M. S.

, & Hadian,

M.-R.

(2016) Effects of Treadmill Incline and Speed on Ankle Muscle Activity in Subjects After a Stroke. Archives of Physical Medicine and Rehabilitation, 97(3), 445–453. https://doi.org/10.1016/j.apmr.2015.10.088

25.

Moreno,

C. C.

, Mendes,

L. A.

, & Lindquist,

A. R.

(2011) Effects of treadmill inclination on the gait of individuals with chronic hemiparesis. Archives of Physical Medicine and Rehabilitation, 92(10), 1675–1680. https://doi.org/10.1016/j.apmr.2011.05.016

26.

Mozaffarian,

, Benjamin,

E. J.

, Go,

A. S.

, Arnett,

D. K.

, Blaha,

M. J.

, Cushman,

, ... Turner,

M. B.

(2015) Heart disease and stroke statistics–2015 update: a report from the American Heart Association. Circulation, 131(4), e29–322. https://doi.org/10.1161/cir.0000000000000152

27.

Nascimento,

L. R.

, Boening,

, Galli,

, Polese,

J. C.

, & Ada,

(2021) Treadmill walking improves walking speed and distance in ambulatory people after stroke and is not inferior to overground walking: a systematic review. Journal of Physiotherapy, 67(2), 95–104. https://doi.org/10.1016/j.jphys.2021.02.014

28.

Neptune,

R. R.

, Kautz,

S. A.

, & Zajac,

F. E.

(2001) Contributions of the individual ankle plantar flexors to support, forward progression and swing initiation during walking. Journal of Biomechanics, 34(11), 1387–1398. https://doi.org/10.1016/s0021-9290(01)00105-1

29.

Olney,

S. J.

, & Richards,

(1996) Hemiparetic gait following stroke. Part I: Characteristics. Gait & Posture, 4(2), 136–148. https://dx-doi-org.web.bisu.edu.cn/10.1016/0966-6362(96)01063-6

30.

Patterson,

K. K.

, Parafianowicz,

, Danells,

C. J.

, Closson,

, Verrier,

M. C.

, Staines,

W. R.

, ... McIlroy,

W. E.

(2008) Gait asymmetry in community-ambulating stroke survivors. Archives of Physical Medicine and Rehabilitation, 89(2), 304–310. https://doi.org/10.1016/j.apmr.2007.08.142

31.

Phadke,

C. P.

(2012) Immediate effects of a single inclined treadmill walking session on level ground walking in individuals after stroke. American Journal of Physical Medicine & Rehabilitation, 91(4), 337–345. https://doi.org/10.1097/PHM.0b013e31823cabe3

32.

Phan,

P. L.

, Blennerhassett,

J. M.

, Lythgo,

, Dite,

, & Morris,

M. E.

(2013) Over-ground walking on level and sloped surfaces in people with stroke compared to healthy matched adults. Disability and Rehabilitation, 35(15), 1302–1307. https://doi.org/10.3109/09638288.2012.729646

33.

Pohl,

, Mehrholz,

, Ritschel,

, & Rückriem,

(2002) Speed-Dependent Treadmill Training in Ambulatory Hemiparetic Stroke Patients. Stroke, 33(2), 553–558. https://doi.org/10.1161/hs0202.102365

34.

Sousa,

A. S.

, Silva,

, & Santos,

(2015) Ankle anticipatory postural adjustments during gait initiation in healthy and post-stroke subjects. Clinical Biomechanics (Bristol, Avon), 30(9), 960–965. https://doi.org/10.1016/j.clinbiomech.2015.07.002

35.

Tomczak,

, & Tomczak,

(2014) The need to report effect size estimates revisited. An overview of some recommended measures of effect size.

36.

Turns,

L. J.

, Neptune,

R. R.

, & Kautz,

S. A.

(2007) Relationships between muscle activity and anteroposterior ground reaction forces in hemiparetic walking. Archives of Physical Medicine and Rehabilitation, 88(9), 1127–1135. https://doi.org/10.1016/j.apmr.2007.05.027

37.

Tyrell,

C. M.

, Roos,

M. A.

, Rudolph,

K. S.

, & Reisman,

D. S.

(2011) Influence of systematic increases in treadmill walking speed on gait kinematics after stroke. Physical Therapy, 91(3), 392–403. https://doi.org/10.2522/ptj.20090425

38.

Wakasa,

, & Fukuda,

(2013) The Time-dependent Changes in EMG Recorded from the Medial Gastrocnemius of Young and Elderly Amputees. Journal of Physical Therapy Science, 25(11), 1511–1513. https://doi.org/10.1589/jpts.25.1511

39.

Werner,

, Lindquist,

A. R.

, Bardeleben,

, & Hesse,

(2007) The influence of treadmill inclination on the gait of ambulatory hemiparetic subjects. Neurorehabilitation and Neural Repair, 21(1), 76–80. https://doi.org/10.1177/1545968306289958

40.

Winter,

D. A.

, & Scott,

S. H.

(1991) Technique for interpretation of electromyography for concentric and eccentric contractions in gait. Journal of Electromyography and Kinesiology: Official Journal of the International Society of Electrophysiological Kinesiology, 1(4), 263–269. https://doi.org/10.1016/1050-6411(91)90013-u

41.

Yang,

Y. R.

, Mi,

P. L.

, Huang,

S. F.

, Chiu,

S. L.

, Liu,

Y. C.

, & Wang,

R. Y.

(2018) Effects of neuromuscular electrical stimulation on gait performance in chronic stroke with inadequate ankle control - A randomized controlled trial. PLoS ONE, 13(12), e0208609. https://doi.org/10.1371/journal.pone.0208609

42.

Yoon,

S. K.

, & Kang,

S. H.

(2016) Effects of inclined treadmill walking training with rhythmic auditory stimulation on balance and gait in stroke patients. Journal of Physical Therapy Science, 28(12), 3367–3370. https://doi.org/10.1589/jpts.28.3367