Technology and Health Care 23 (2015) 443–452 DOI 10.3233/THC-150903 IOS Press
443
Treadmill training with tilt sensor functional electrical stimulation for improving balance, gait, and muscle architecture of tibialis anterior of survivors with chronic stroke: A randomized controlled trial Dal-Yeon Hwanga , Hwang-Jae Leeb , Gyu-Chang Leec and Suk-Min Leea,∗ a Department
of Physical Therapy, College of Health and Welfare, Sahmyook University, Seoul, Korea of Rehabilitation Medicine, Myongji Choonhey Rehabilitation Hospital, Seoul, Korea c Department of Physical Therapy, College of Natural Sciences, Kyungnam University, Changwon-si, Korea b Department
Received 5 September 2014 Accepted 25 January 2015 Abstract. BACKGROUND: Gait training is important for stroke rehabilitation, such as using the treadmill training with functional electrical stimulation (FES). OBJECTIVE: This study was to investigate the effects of the treadmill training with tilt sensor FES on the balance, gait, and muscle architecture of the tibialis anterior in stroke survivors. METHODS: The study was a randomized controlled trial. Thirty-four stroke survivors were recruited and screened eligibility criteria. Thirty-two participants were randomly allocated to two groups using random allocation software: Treadmill training with Tilt Sensor FES (TTSF) group (n = 16) and Treadmill training with Placebo Tilt Sensor FES (TPTSF) group (n = 16). TTSF group performed gait training on treadmill with tilt sensor FES, and TPTSF group performed gait training on treadmill with placebo tilt sensor FES. Two participants were dropped during this study, and 30 participants were included at post-test. Balance and gait were measured using the timed up and go (TUG) test, berg balance scale (BBS), and 10 m walk test (10 mWT). Ultrasound imaging was used to measure the muscle architecture of the tibialis anterior. RESULTS: After intervention, there were significant improvements in the TUG, BBS, and 10 mWT compared to baseline in both groups (p < 0.05). At follow-up, the TUG, BBS, 10 mWT, and muscle architecture of tibialis anterior on the paretic side showed significant improvements in the TTSF group compared to TPTSF group (p < 0.05). CONCLUSIONS: The findings of this study suggest that TTSF can be an effective intervention for improving balance, gait ability, and muscle architecture of tibialis anterior of stroke survivors. Keywords: Stroke, ultrasound imaging, treadmill training, functional electrical stimulation
∗ Corresponding author: Suk-Min Lee, Department of Physical Therapy, Sahmyook University, 26-21, Gongneung2-dong, Nowon-gu, Seoul 139-742, Korea. Tel.: +82 10 3704 5650; Fax: +82 2 3399 1639; E-mail: [email protected].
c 2015 – IOS Press and the authors. All rights reserved 0928-7329/15/$35.00
444
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
1. Introduction Stroke patients generally have impairments such as motor, sensory, emotional, and cognitive dysfunction as well as functional restrictions [1]. As a result of stroke, muscle tension increases abnormally, resulting in a foot drop that makes it difficult to perform ankle dorsiflexion during the swing phase of gait [2]. Foot drop appears in 20% of stroke survivors and mainly impairs balance and gait [3]. To improve the balance and gait of stroke patients, various interventions such as gait training using robots [4], functional electrical stimulation [5], and treadmill gait training have been used [6]. Among these, treadmill gait training constitutes the major method of therapeutic gait training for stroke patients. Furthermore, treadmill gait training encourages a symmetrical gait and reduces spasticity, thereby improving the gait of stroke patients more effectively than simply walking on the ground [7]. Functional electrical stimulation (FES) was first applied to hemiparesis patients suffering from a spastic ankle condition in a study by Liberson [8]. Gait ability has been improved by applying FES to the tibial nerves of patients displaying foot drop as well as those who cannot perform muscle contractions and coordinated movements, and FES has resulted in almost normal movement [9]. Recent studies have focused on active-assist FES as opposed to conventional FES [6]. The active-assist FES is controlled by electrical stimulation of active motion. Specifically, on ground gait training has been used in combination with FES using a tilt sensor that reacts to changes in the knee angle. In a previous study, gait training using a tilt sensor FES showed advantages in improving gait ability, proprioceptive stimulation, facilitation of motor learning, and brain plasticity in subacute stroke patients as compared to gait training performed using foot drop orthoses [10]. However, although active-assist FES may be advantageous in that it can be combined with gait training, there have been few studies investigating the effects of treadmill gait training based on FES using a tilt sensor. The present study was designed to investigate the effects of treadmill gait training combined with FES using a tilt sensor on balance, gait ability, and structural changes of the tibialis anterior as compared with conventional treadmill gait training in stroke survivors.
2. Methods 2.1. Study design, randomization, blinding, and setting This study was a randomly controlled trial. The 32 participants were randomly allocated to either treadmill training combined with tilt sensor FES (TTSF) group (n = 16) or treadmill training combined with placebo tilt sensor FES (TPTSF) group (n = 16) using random allocation software. A research assistant used the random allocation software to divide the patients into groups. The researchers, outcome assessor, and data analyst were blinded to the group assignments. Before and after intervention, the balance, gait, and structure of the tibialis anterior of participants were assessed. Balance was assessed by the Timed Up and Go (TUG) test and Berg Balance Scale (BBS); gait, by the 10-meter walk test (10 MWT); and the structure of the tibialis anterior, by ultrasound. Treadmill gait training combined with tilt sensor FES was performed in the TTSF group, whereas treadmill gait training combined with placebo tilt sensor FES was implemented in the TPTSF group. Both groups received conventional physical therapy. One participant from each of the TTSF and TPTSF groups subsequently dropped out because of discharge from the hospital. Therefore, post-test data were collected for a total of 30 subjects (Fig. 1).
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
445
Fig. 1. Flow chart for procedures in this study.
2.2. Participants We recruited 34 patients who had been diagnosed as having hemiparesis due to stroke and were hospitalized at a rehabilitation center. The participants were screened according to the following inclusion criteria: those who had experienced at least one stroke, those who had had a stroke more than 6 months ago, could walk more than 15 meters independently without a walking aid, scored at least 24 in the Mini-Mental State Examination (MMSE), had no problems with auditory or visual function, were classified as 2nd grade or lower on ankle plantar flexor response on the Modified Ashworth Scale, and were between the 2nd and 4th stages on the Brunnstrom Stages. Participants were excluded whose cardiovascular systems were unstable or those who had other neurological or orthopedic conditions such as fractures. A total of 34 participants qualified for inclusion; one participant was then excluded because of a cardiovascular problem and another was excluded because of hemianopsia. Thus, 32 stroke patients participated in the study. The purpose and procedure of the study were explained to the participants, and informed consent for voluntary participation was received from each. This study was performed after being approved by the Ethics Committee of Sahmyook University. Table 1 presents the characteristics of the participants. 2.3. Sample size calculation The sample size was established by using G-Power 3.1 software. The power and the alpha levels were set at 0.8 and 0.05, respectively, and the effect size was set at 0.95. According to the analysis, we tried to include 19 participants in the each group. However, we included 16 participants in the each group because only 32 participants were eligible.
446
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES Table 1 Characteristics of the participants
TTSF group (n = 15) TPTSF group (n = 15) P value Gender (male/female) 9/6 8/7 0.409 Age (years) 50.00 ± 7.55 49.47 ± 5.01 0.821 Height (cm) 165.92 ± 7.39 164.01 ± 8.25 0.511 Weight (kg) 68.09 ± 7.95 67.33 ± 9.60 0.238 Type of stroke (infarction/hemorrhage) 8/7 6/9 0.715 Affected side (right/left) 4/11 6/9 0.715 Time since onset of stroke (days) 192.53 ± 18.79 194.07 ± 18.95 0.826 MMSE (score) 27.67 ± 1.05 27.40 ± 0.99 0.478 MAS in ankle plantar flexor (1/1+ /2) 5/5/5 3/7/5 NA Brunnstrom stages (2/3/4) 4/7/4 3/8/4 NA Calf length (cm) 33.32 ± 1.80 32.69 ± 1.43 0.299 Values are expressed as mean ± SD or frequency. TTSF, Treadmill training combined with Tilt Sensor Functional electrical stimulation; TPTSF, Treadmill training combined with Placebo Tilt Sensor Functional electrical stimulation; MMSE, MiniMental State Examination; MAS, Modified Ashworth Scale.
Fig. 2. Tilt sensor FES.
2.4. Intervention Treadmill gait training combined with tilt sensor FES was performed in the TTSF group. The WalkAide system (WalkAide model 20-0100, Innovative Neurotronics Inc. USA, 2008) was used to provide the tilt sensor FES, and gait training combined with tilt sensor FES was performed on a treadmill (FITEX T-5050, Seoul, Republic of Korea). The WalkAide system is a combination of the WalkAide stimulator and WalkAnalyst program.
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
447
The WalkAide stimulator is a beeper-sized electronic device composed of a soft cuff that includes the electrode for electrical stimulation, velcro, and a tilt sensor. The tilt sensor senses the angle of knee flexion during walking and transmits electrical stimulation to the muscles and nerves that control flexion of the ankle. In this study, the electrode (Axelgard Manufacturing, Fallbrook, CA) in the WalkAide stimulator was attached to the motor point of the common peroneal nerve between the popliteal fossa and the head of the fibula, and another electrode was attached to the tibialis anterior [11]. Cuffs and fixing bands were used to prevent the electrodes from disconnecting (Fig. 2). The frequency of electrical stimulation was set at 25 Hz, the pulse duration at 150 µs, and the intensity between 60 and 150 V, depending on participant tolerance and the level of stimulation needed to elicit ankle dorsiflexion and foot eversion [11]. The most appropriate stimulus intensity for the participants was determined using the WalkAnalyst program, during which the participants walked while wearing the WalkAide stimulator. The measured value was stored by the WalkAide stimulator. The participants then underwent gait training on a treadmill. When the affected leg of the participants tilted backward in the late stance phase during walking on the treadmill (when the heels are elevated), the tilt sensor of the WalkAide stimulator guided the dorsiflexion motion of the affected ankle. Also, the initial speed of the treadmill was measured while participants maintained a normal gait pattern at a comfortable cadence and stride length for each participant [7]. The mean treadmill speed was 0.6 m/s at the beginning of gait training. If participants were able to walk comfortably at this speed for more than 3 minutes, the speed of the treadmill was increased by 0.2 m/s. When speed was increased and a participant was unable to perform initial contact properly, speed was reduced once again. After 4 weeks, a mean treadmill speed of 1.2 m/s was reached. During treadmill gait training, participants were allowed to hold onto a safety bar on each side. Before starting training, participants were asked to perform an on over-ground walking for 2 minutes in order to become accustomed to the stimulus of the WalkAide. Treadmill training combined with the tilt sensor in FES was performed for 30 minutes, one time per day, for 4 weeks. The TPTSF group also had treadmill gait training combined with tilt sensor FES. The WalkAide system was attached to the subjects, but the stimulator was set in the “off” position. Other than that, the same methods were used as in the TTSF group. Conventional physical therapy was performed based on the neurodevelopmental treatment concept and proprioceptive neuromuscular facilitation [12,13], which are used by physical therapists to improve the function of affected limbs, as well as balance and gait ability, based on motor control and motor learning theory. Conventional physical therapy was performed for 30 minutes, twice per day, for 4 weeks. 2.5. Outcome measures The TUG and BBS were used to measure balance. The inter-rater reliability of the TUG is r = 0.98 while that of intra-rater reliability is r = 0.99 [14]. The BBS also is highly reliable in assessing dynamic balance, as its inter-rater reliability is 0.99 and intra-rater reliability is 0.98 [15]. The 10 MWT was used to evaluate gait ability. The inter-rater and intra-rater reliabilities were as high as r = 0.89 [16]. The TUG and 10MWT were repeated three times each, and the average values were analyzed. Participants had 10 min rest time between each trial. A portable ultrasound scanner (Mysono U5, Samsung Medison, Seoul, Korea) with a 7 MHz linear transducer was used to measure the muscle structure of the tibialis anterior. A linear transducer probe was used in B-mode, and pressure on the skin was minimized by applying an ample amount of ultrasonic gel between the probe and the skin. The probe and the skin were maintained perpendicular to each other
448
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES Table 2 Changes in the TUG, BBS, and the 10-m walk test
TTSF group (n = 15) Pre-test Post-test Changes TUG (sec) 23.31 ± 3.94 15.03 ± 2.75∗∗∗ −8.28 ± 2.76 BBS (score) 37.87 ± 3.94 50.00 ± 1.89∗∗∗ 12.13 ± 3.44 10 m walk test (sec) 28.02 ± 4.05 20.51 ± 2.94∗∗∗ −7.51 ± 2.66
TPTSF group (n = 15) Pre-test Post-test Changes p value 21.85 ± 3.55 16.96 ± 2.74∗∗∗ −4.89 ± 1.81 0.000+++ 38.20 ± 3.97 45.20 ± 2.65∗∗∗ 8.00 ± 2.98 0.002++ 31.23 ± 5.07 25.99 ± 4.35∗∗∗ −5.24 ± 1.81 0.011+
Values are expressed as mean ± SD. TUG, Timed Up and Go; BBS, Berg Balance Scale. ∗ significant differences between pre and post-test. ∗ p < 0.05; ∗∗ p < 0.01; ∗∗∗ p < 0.001. + significant differences between experimental group and control group. + p < 0.05; ++ p < 0.01; +++ p < 0.001.
to ensure constant measurement. The participants were asked to lie down and wear modified ankle orthoses in which their knees were extended and ankle angle fixed at 90◦ . After the digital manual muscle tester (Power Track II, Jtech Medical Industries, Inc., Herber City, UT, USA) was placed on the distal part of the navicular bone on the top side of the foot, the participants were asked to try to contract the ankle dorsiflexor on the affected side as much as possible. When the participants tried to contract their ankle dorsiflexor, the examiner reported the muscle thickness and pennation angles of ankle on the affected side. To measure the thickness of the muscle, the distance, excluding the tendon sheath, between the superior aponeurosis and central aponeurosis of the tibialis anterior was measured. After checking the fascicle between the superior aponeurosis and central aponeurosis to measure the pennation angle, the angle between the fascicle and central aponeurosis was measured. In addition, for pre- and postmeasurement, the examiner measured the muscle structure after palpating the location of distal part of the navicular bone. The location of the linear transducer probe was at a point 20% of the distance from the tip of the lateral malleolus to the head of the fibula, and measuring points were marked by “X” using a pen [17]. To prevent muscle fibers from pressing downward and causing the image to blur, pressure between the linear probe and skin was minimized, and the image was stored and measured when the pennation angle was most clearly seen. The ultrasound scan was performed by an experienced examiner. The test-retest reliability of the ultrasound scan was r = 0.811 [17]. 2.6. Statistical analyses The SPSS Statistical Package, version 18.0 (IBM Inc., Chicago, IL, USA), was used for statistical analysis. Participant demographics were given as mean and standard deviation. Data were analyzed using the paired t-test for changes in values from before to after intervention. Differences between the two groups at baseline and follow-up were analyzed using an independent t-test. The level of statistical significance was set at 0.05. 3. Results After intervention, there were significant improvements in the TUG, BBS, and 10 MWT compared to baseline in both the TTSF and TPTSF groups (p < 0.05). The TUG, BBS, and 10 MWT showed more significant improvements in the TTSF group as compared to TPTSF group at follow-up (p < 0.05) (Table 2). As compared to baseline, at follow-up, both groups had significant improvements (p < 0.05) in the tibialis anterior, the pennation angles in the resting phase on the affected and non-affected sides, pennation angles in the contraction phase on the affected and non-affected sides, muscle thickness in the resting phase on the affected side, and muscle thickness in the contraction phase on the affected side
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
449
Table 3 Changes in the muscle architecture of the tibialis anterior
Resting phase
Contraction phase
TTSF group (n = 15) TPTSF group (n = 15) Pre-test Post-test Changes Pre-test Post-test Changes 10.50 ± 0.92 11.94 ± 0.76∗∗∗ 1.43 ± 1.07 10.78 ± 0.87 11.58 ± 0.65∗∗ 0.80 ± 1.00
Pennation angles of the affected side (angle) Pennation angles 11.86 ± 0.85 12.95 ± 1.19∗∗ 1.09 ± 1.10 12.05 ± 0.68 13.05 ± 0.85∗∗ 1.00 ± 0.90 of the non-affected side (angle) Muscle thickness 1.55 ± 0.13 1.80 ± 0.12∗∗∗ 0.25 ± 0.10 1.51 ± 0.09 1.66 ± 0.06∗∗∗ 0.16 ± 0.08 of the affected side (cm) 0.03 ± 0.68 1.69 ± 0.07 1.72 ± 0.08 0.04 ± 0.07 Muscle thickness 1.68 ± 0.06 1.71 ± 0.09 of the non-affected side (cm) Pennation angles of the affected side (angle) Pennation angles of the non-affected side (angle) Muscle thickness of the affected side (cm) Muscle thickness of the non-affected side (cm)
p value 0.104
0.818 0.009++
0.879
14.03 ± 1.64 17.05 ± 1.40∗∗∗ 3.02 ± 1.47 14.57 ± 0.73 15.00 ± 0.47∗∗ 0.43 ± 0.52
0.000+++
16.12 ± 1.26 16.88 ± 1.31∗
0.720
0.76 ± 1.29 16.90 ± 1.10 17.53 ± 1.02∗∗ 0.62 ± 0.78
1.74 ± 0.06 1.98 ± 0.11∗∗∗ 0.24 ± 0.13
1.72 ± 0.07 1.83 ± 0.11∗∗ 0.10 ± 0.11
0.005++
1.86 ± 0.07 1.95 ± 0.18
1.90 ± 0.07 1.91 ± 0.07
0.056
0.09 ± 0.16
0.01 ± 0.01
Values are expressed as mean ± SD. ∗ significant differences between pre and post-test. ∗ p < 0.05; ∗∗ p < 0.01; ∗∗∗ p < 0.001. + significant differences between experimental group and control group. + p < 0.05; ++ p < 0.01; +++ p < 0.001.
(Table 3). However, there was no significant improvement in muscle thickness in the resting phase on the non-affected side between pre- and post-tests. Muscle thickness in the resting phase on the affected side, pennation angle of contraction phase on the affected side, and muscle thickness in the contraction phase on the affected side significantly improved in the TTSF group as compared to the TPTSF group at follow-up (p < 0.05) (Table 3). 4. Discussion This study investigated the effects of treadmill gait training combined with tilt sensor FES on balance, gait, and structure of the tibialis anterior muscle in stroke survivors. According to the results of the BBS, TUG, and 10MWT, significant improvements were observed in the TTSF group compared to those in the TPTSF group. Previous studies have reported that gait re-education combined with FES significantly improved balance and gait in chronic stroke patients [18]. Specifically, during gait re-education, dorsiflexion of the ankle could be improved after applying tilt sensor FES to the tibial nerve during the swing phase of the gait, resulting in a significantly improved gait and better balance. Lee et al. reported that balance could be significantly improved in participants when FES is applied at the point where the tibialis anterior contracts during treadmill gait training [6]. They reported that FES, which encourages the active participation of the patient, had more positive effects on functional movement and balance ability than conventional FES [6]. In this study, tilt sensor FES controlled by the angle of the knee gradient changes
450
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
was implemented, which encourages active participation of the patient. In the results of this study, gait training combined with active tilt sensor FES was effective for improving balance in stroke patients. When the effects of treadmill gait training combined with FES were compared with those of on ground gait training in 30 stroke patients, Peurala et al. reported that treadmill gait training combined with FES was more effective for improving gait speed and dynamic balance [19]. Their results coincide with the results of this study. Treadmill gait training combined with FES improved the symmetric parameters in stroke patients, which had positive effects on stability, balance for walking, and gait speed. It has been previously reported that repetitive and task-oriented training such as treadmill gait training improves balance in stroke patients through motor relearning by increasing neuroplasticity and reorganization of the brain [20]. The somatosensory response of the ankle to which dorsiflexion was applied is stimulated through FES while brain plasticity is accelerated, which may decrease foot drop. The basic role of tilt sensor FES is to supplement activation of damaged efferent pathways [10]. In sensory-motor integration theory, sensory stimulation provided by the movement of affected limbs directly affects continuous motor output [21]. In other words, impairment of the somatosensory response is responsible for the delayed motor recovery of stroke patients. Therefore, appropriate stimulation of the somatosensory response is essential for improvement in motor recovery [22]. According to previous studies, activation of damaged efferent pathways was most effective when combined with active movements [23]. In addition, increase of somatosensory stimuli during walking can improve the gait patterns of stroke patients [24]. Accordingly, the tilt sensor FES used in this study improved stimulation of the somatosensory response through active knee flexion along with dorsiflexion of the ankle during treadmill gait training, providing additional positive effects in improving gait ability. Currently, ultrasound test equipment is being used to assess rehabilitation to improve muscle function in patients. The ultrasound test used in the rehabilitation field is called rehabilitative ultrasound imaging (RUSI) [25]. According to the results of the RUSI test in this study, treadmill gait training both with and without tilt sensor FES significantly improved the pennation angle between the affected and non-affected parts of the tibialis anterior as well as muscle thickness in resting and contraction phases. However, the treadmill gait training group subjected to tilt sensor FES showed significant improvement in both pennation angle and muscle thickness in the resting and contraction phases on the affected side as compared to the group subjected to placebo tilt sensor FES. For the non-affected side, there were no significant differences between the resting phase and the contraction phase. According to this result, treadmill gait training combined with tilt sensor FES had positive effects on the tibialis anterior of both the affected and non-affected sides, especially the structure of the tibialis anterior on the affected side. Although there is a lack of studies on improving the function of muscles using ultrasound assessment, a previous study showed that the pennation angle increases during ankle dorsiflexion in an analysis of structural changes in the pennation angle during the resting phase as well as during maximal isometric muscle contraction [26]. Bland et al. investigated the relationship between the structure of the tibialis anterior and other muscles related to gait in cerebral palsy patients and found that muscle thickness and length of the muscle fibers are highly correlated with biomechanical elements such as muscular strength and gait speed [27]. In the present study, by treadmill gait training and tilt sensor FES, the tibialis anterior of the affected side was activated during swing phase and thus its structure on the tibilais anterior was changed. Thus, treadmill gait training with tilt sensor FES can be effective in improving balance, gait, and the structure of the tibialis anterior. In particular, tilt sensor FES controlled by the angle of the knee gradient changes can have an advantage in promoting active participation of patients; conventional FES is controlled only by the time set depending on gait cycle. However, tilt sensor FES may have a disadvantage in that it cannot be used by patients who cannot move their knees. In addition, there were some
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
451
limitations in this study. First, the number of participants was small. Second, there was no comparison to conventional FES to confirm that effects of tilt sensor FES were different than those of conventional FES. Therefore, treadmill gait training using tilt sensor FES in this study possibly affected the tendon as well as the structure of the tibialis anterior. Thus, investigations on changes in tendons as well as muscle structure should be performed. Further, since this study used only the 10 WT to assess gait ability, qualitative elements as well as spatiotemporal elements should be assessed.
Acknowledgement This study was supported by Sahmyook University.
Conflict of interest The authors have no conflicts of interest to declare.
References [1] [2]
P.W. Duncan, Stroke disability, Physical Therapy 74 (1994), 399-407. P.T. Cheng, C.L. Chen, C.M. Wang et al., Leg muscle activation patterns of sit-to-stand movement in stroke patients, American Journal of Physical Medicine & Rehabilitation/Association of Academic Physiatrists 83 (2004), 10-16. [3] T.M. Kesar, R. Perumal, A. Jancosko et al., Novel patterns of functional electrical stimulation have an immediate effect on dorsiflexor muscle function during gait for people poststroke, Physical Therapy 90 (2010), 55-66. [4] S. Fisher, L. Lucas and T.A. Thrasher, Robot-assisted gait training for patients with hemiparesis due to stroke, Topics in Stroke Rehabilitation 18 (2011), 269-276. [5] C. Bulley, J. Shiels, K. Wilkie et al., User experiences, preferences and choices relating to functional electrical stimulation and ankle foot orthoses for foot-drop after stroke, Physiotherapy 97 (2011), 226-233. [6] H.J. Lee, K.H. Cho and W.H. Lee, The effects of body weight support treadmill training with power-assisted functional electrical stimulation on functional movement and gait in stroke patients, American Journal of Physical Medicine & Rehabilitation / Association of Academic Physiatrists 92 (2013), 1051-1059. [7] S. Hesse, Locomotor therapy in neurorehabilitation, NeuroRehabilitation 16 (2001), 133-139. [8] W.T. Liberson, H.J. Holmquest, D. Scot et al., Functional electrotherapy: stimulation of the peroneal nerve synchronized with the swing phase of the gait of hemiplegic patients, Archives of Physical Medicine and Rehabilitation 42 (1961), 101-105. [9] T. O’Halloran, M. Haugland, G.M. Lyons et al., An investigation of the effect of modifying stimulation profile shape on the loading response phase of gait, during FES-corrected drop foot: Stimulation profile and loading response, Neuromodulation : Journal of the International Neuromodulation Society 7 (2004), 113-125. [10] G. Morone, A. Fusco, P. Di Capua et al., Walking training with foot drop stimulator controlled by a tilt sensor to improve walking outcomes: A randomized controlled pilot study in patients with stroke in subacute phase, Stroke Research and Treatment 2012 (2012), 523-564. [11] A.R. Lindquist, C.L. Prado, R.M. Barros et al., Gait training combining partial body-weight support, a treadmill, and functional electrical stimulation: effects on poststroke gait, Physical Therapy 87 (2007), 1144-1154. [12] P. Klimkiewicz, A. Kubsik and M. Woldanska-Okonska, NDT-Bobath method used in the rehabilitation of patients with a history of ischemic stroke, Wiadomo´sci lekarskie (Warsaw, Poland: 1960). 65 (2012), 102-107. [13] H.S. Reichel, PNF (proprioceptive neuromuscular facilitation): gait training, Sportverletzung Sportschaden : Organ der Gesellschaft für Orthopädisch-Traumatologische Sportmedizin 10 (1996), A11-20. [14] S.S. Ng, C.W. Hui-Chan, The timed up & go test: its reliability and association with lower-limb impairments and locomotor capacities in people with chronic stroke, Archives of Physical Medicine and Rehabilitation 86 (2005), 1641-1647. [15] S. Downs, J. Marquez and P. Chiarelli, The Berg Balance Scale has high intra- and inter-rater reliability but absolute reliability varies across the scale: A systematic review, Journal of Physiotherapy 59 (2013), 93-99.
452 [16] [17] [18] [19] [20] [21] [22] [23]
[24] [25] [26] [27]
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES G. Scivoletto, F. Tamburella, L. Laurenza et al, Validity and reliability of the 10-m walk test and the 6-min walk test in spinal cord injury patients, Spinal Cord 49 (2011), 736-740. K. McCreesh and S. Egan, Ultrasound measurement of the size of the anterior tibial muscle group: The effect of exercise and leg dominance, Sports Medicine, Arthroscopy, Rehabilitation, Therapy & Technology: SMARTT 3 (2011), 18. J.A. Robertson, J.J Eng and C. Hung, The effect of functional electrical stimulation on balance function and balance confidence in community-dwelling individuals with stroke, Physiotherapy Canada. Physiothérapie Canada 62 (2010), 114-119. S.H. Peurala, I.M. Tarkka, K. Pitkanen et al., The effectiveness of body weight-supported gait training and floor walking in patients with chronic stroke, Archives of Physical Medicine and Rehabilitation 86 (2005), 1557-1564. L.R. Harvey, Improving poststroke recovery: neuroplasticity and task-oriented training, Current Treatment Options in Cardiovascular Medicine 11 (2009), 251-259. D.A. Rosenbaum, R.G. Meulenbroek and J. Vaughan, Planning reaching and grasping movements: theoretical premises and practical implications, Motor Control 5 (2001), 99-115. C.W. Wu, H.J. Seo and L.G. Cohen, Influence of electric somatosensory stimulation on paretic-hand function in chronic stroke, Archives of Physical Medicine and Rehabilitation 87 (2006), 351-357. A. Boyaci, O. Topuz, H. Alkan et al., Comparison of the effectiveness of active and passive neuromuscular electrical stimulation of hemiplegic upper extremities: a randomized, controlled trial, International Journal of Rehabilitation Research. Internationale Zeitschrift für Rehabilitationsforschung. Revue Internationale de Recherches de Réadaptation 36 (2013), 315-322. S.I. Lin, L.J. Hsu and H.C. Wang, Effects of ankle proprioceptive interference on locomotion after stroke, Archives of Physical Medicine and Rehabilitation 93 (2012), 1027-1033. D.S. Teyhen, J.D. Childs and T.W. Flynn, Rehabilitative ultrasound imaging: When is a picture necessary, The Journal of Orthopaedic and Sports Physical Therapy 37 (2007), 579-580. C.N. Maganaris and V. Baltzopoulos, Predictability of in vivo changes in pennation angle of human tibialis anterior muscle from rest to maximum isometric dorsiflexion, European Journal of Applied Physiology and Occupational Physiology 79 (1999), 294-297. D.C. Bland, L.A. Prosser and L.A. Bellini, Tibialis anterior architecture, strength, and gait in individuals with cerebral palsy, Muscle Nerve 44 (2011), 509-517.
443
Treadmill training with tilt sensor functional electrical stimulation for improving balance, gait, and muscle architecture of tibialis anterior of survivors with chronic stroke: A randomized controlled trial Dal-Yeon Hwanga , Hwang-Jae Leeb , Gyu-Chang Leec and Suk-Min Leea,∗ a Department
of Physical Therapy, College of Health and Welfare, Sahmyook University, Seoul, Korea of Rehabilitation Medicine, Myongji Choonhey Rehabilitation Hospital, Seoul, Korea c Department of Physical Therapy, College of Natural Sciences, Kyungnam University, Changwon-si, Korea b Department
Received 5 September 2014 Accepted 25 January 2015 Abstract. BACKGROUND: Gait training is important for stroke rehabilitation, such as using the treadmill training with functional electrical stimulation (FES). OBJECTIVE: This study was to investigate the effects of the treadmill training with tilt sensor FES on the balance, gait, and muscle architecture of the tibialis anterior in stroke survivors. METHODS: The study was a randomized controlled trial. Thirty-four stroke survivors were recruited and screened eligibility criteria. Thirty-two participants were randomly allocated to two groups using random allocation software: Treadmill training with Tilt Sensor FES (TTSF) group (n = 16) and Treadmill training with Placebo Tilt Sensor FES (TPTSF) group (n = 16). TTSF group performed gait training on treadmill with tilt sensor FES, and TPTSF group performed gait training on treadmill with placebo tilt sensor FES. Two participants were dropped during this study, and 30 participants were included at post-test. Balance and gait were measured using the timed up and go (TUG) test, berg balance scale (BBS), and 10 m walk test (10 mWT). Ultrasound imaging was used to measure the muscle architecture of the tibialis anterior. RESULTS: After intervention, there were significant improvements in the TUG, BBS, and 10 mWT compared to baseline in both groups (p < 0.05). At follow-up, the TUG, BBS, 10 mWT, and muscle architecture of tibialis anterior on the paretic side showed significant improvements in the TTSF group compared to TPTSF group (p < 0.05). CONCLUSIONS: The findings of this study suggest that TTSF can be an effective intervention for improving balance, gait ability, and muscle architecture of tibialis anterior of stroke survivors. Keywords: Stroke, ultrasound imaging, treadmill training, functional electrical stimulation
∗ Corresponding author: Suk-Min Lee, Department of Physical Therapy, Sahmyook University, 26-21, Gongneung2-dong, Nowon-gu, Seoul 139-742, Korea. Tel.: +82 10 3704 5650; Fax: +82 2 3399 1639; E-mail: [email protected].
c 2015 – IOS Press and the authors. All rights reserved 0928-7329/15/$35.00
444
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
1. Introduction Stroke patients generally have impairments such as motor, sensory, emotional, and cognitive dysfunction as well as functional restrictions [1]. As a result of stroke, muscle tension increases abnormally, resulting in a foot drop that makes it difficult to perform ankle dorsiflexion during the swing phase of gait [2]. Foot drop appears in 20% of stroke survivors and mainly impairs balance and gait [3]. To improve the balance and gait of stroke patients, various interventions such as gait training using robots [4], functional electrical stimulation [5], and treadmill gait training have been used [6]. Among these, treadmill gait training constitutes the major method of therapeutic gait training for stroke patients. Furthermore, treadmill gait training encourages a symmetrical gait and reduces spasticity, thereby improving the gait of stroke patients more effectively than simply walking on the ground [7]. Functional electrical stimulation (FES) was first applied to hemiparesis patients suffering from a spastic ankle condition in a study by Liberson [8]. Gait ability has been improved by applying FES to the tibial nerves of patients displaying foot drop as well as those who cannot perform muscle contractions and coordinated movements, and FES has resulted in almost normal movement [9]. Recent studies have focused on active-assist FES as opposed to conventional FES [6]. The active-assist FES is controlled by electrical stimulation of active motion. Specifically, on ground gait training has been used in combination with FES using a tilt sensor that reacts to changes in the knee angle. In a previous study, gait training using a tilt sensor FES showed advantages in improving gait ability, proprioceptive stimulation, facilitation of motor learning, and brain plasticity in subacute stroke patients as compared to gait training performed using foot drop orthoses [10]. However, although active-assist FES may be advantageous in that it can be combined with gait training, there have been few studies investigating the effects of treadmill gait training based on FES using a tilt sensor. The present study was designed to investigate the effects of treadmill gait training combined with FES using a tilt sensor on balance, gait ability, and structural changes of the tibialis anterior as compared with conventional treadmill gait training in stroke survivors.
2. Methods 2.1. Study design, randomization, blinding, and setting This study was a randomly controlled trial. The 32 participants were randomly allocated to either treadmill training combined with tilt sensor FES (TTSF) group (n = 16) or treadmill training combined with placebo tilt sensor FES (TPTSF) group (n = 16) using random allocation software. A research assistant used the random allocation software to divide the patients into groups. The researchers, outcome assessor, and data analyst were blinded to the group assignments. Before and after intervention, the balance, gait, and structure of the tibialis anterior of participants were assessed. Balance was assessed by the Timed Up and Go (TUG) test and Berg Balance Scale (BBS); gait, by the 10-meter walk test (10 MWT); and the structure of the tibialis anterior, by ultrasound. Treadmill gait training combined with tilt sensor FES was performed in the TTSF group, whereas treadmill gait training combined with placebo tilt sensor FES was implemented in the TPTSF group. Both groups received conventional physical therapy. One participant from each of the TTSF and TPTSF groups subsequently dropped out because of discharge from the hospital. Therefore, post-test data were collected for a total of 30 subjects (Fig. 1).
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
445
Fig. 1. Flow chart for procedures in this study.
2.2. Participants We recruited 34 patients who had been diagnosed as having hemiparesis due to stroke and were hospitalized at a rehabilitation center. The participants were screened according to the following inclusion criteria: those who had experienced at least one stroke, those who had had a stroke more than 6 months ago, could walk more than 15 meters independently without a walking aid, scored at least 24 in the Mini-Mental State Examination (MMSE), had no problems with auditory or visual function, were classified as 2nd grade or lower on ankle plantar flexor response on the Modified Ashworth Scale, and were between the 2nd and 4th stages on the Brunnstrom Stages. Participants were excluded whose cardiovascular systems were unstable or those who had other neurological or orthopedic conditions such as fractures. A total of 34 participants qualified for inclusion; one participant was then excluded because of a cardiovascular problem and another was excluded because of hemianopsia. Thus, 32 stroke patients participated in the study. The purpose and procedure of the study were explained to the participants, and informed consent for voluntary participation was received from each. This study was performed after being approved by the Ethics Committee of Sahmyook University. Table 1 presents the characteristics of the participants. 2.3. Sample size calculation The sample size was established by using G-Power 3.1 software. The power and the alpha levels were set at 0.8 and 0.05, respectively, and the effect size was set at 0.95. According to the analysis, we tried to include 19 participants in the each group. However, we included 16 participants in the each group because only 32 participants were eligible.
446
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES Table 1 Characteristics of the participants
TTSF group (n = 15) TPTSF group (n = 15) P value Gender (male/female) 9/6 8/7 0.409 Age (years) 50.00 ± 7.55 49.47 ± 5.01 0.821 Height (cm) 165.92 ± 7.39 164.01 ± 8.25 0.511 Weight (kg) 68.09 ± 7.95 67.33 ± 9.60 0.238 Type of stroke (infarction/hemorrhage) 8/7 6/9 0.715 Affected side (right/left) 4/11 6/9 0.715 Time since onset of stroke (days) 192.53 ± 18.79 194.07 ± 18.95 0.826 MMSE (score) 27.67 ± 1.05 27.40 ± 0.99 0.478 MAS in ankle plantar flexor (1/1+ /2) 5/5/5 3/7/5 NA Brunnstrom stages (2/3/4) 4/7/4 3/8/4 NA Calf length (cm) 33.32 ± 1.80 32.69 ± 1.43 0.299 Values are expressed as mean ± SD or frequency. TTSF, Treadmill training combined with Tilt Sensor Functional electrical stimulation; TPTSF, Treadmill training combined with Placebo Tilt Sensor Functional electrical stimulation; MMSE, MiniMental State Examination; MAS, Modified Ashworth Scale.
Fig. 2. Tilt sensor FES.
2.4. Intervention Treadmill gait training combined with tilt sensor FES was performed in the TTSF group. The WalkAide system (WalkAide model 20-0100, Innovative Neurotronics Inc. USA, 2008) was used to provide the tilt sensor FES, and gait training combined with tilt sensor FES was performed on a treadmill (FITEX T-5050, Seoul, Republic of Korea). The WalkAide system is a combination of the WalkAide stimulator and WalkAnalyst program.
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
447
The WalkAide stimulator is a beeper-sized electronic device composed of a soft cuff that includes the electrode for electrical stimulation, velcro, and a tilt sensor. The tilt sensor senses the angle of knee flexion during walking and transmits electrical stimulation to the muscles and nerves that control flexion of the ankle. In this study, the electrode (Axelgard Manufacturing, Fallbrook, CA) in the WalkAide stimulator was attached to the motor point of the common peroneal nerve between the popliteal fossa and the head of the fibula, and another electrode was attached to the tibialis anterior [11]. Cuffs and fixing bands were used to prevent the electrodes from disconnecting (Fig. 2). The frequency of electrical stimulation was set at 25 Hz, the pulse duration at 150 µs, and the intensity between 60 and 150 V, depending on participant tolerance and the level of stimulation needed to elicit ankle dorsiflexion and foot eversion [11]. The most appropriate stimulus intensity for the participants was determined using the WalkAnalyst program, during which the participants walked while wearing the WalkAide stimulator. The measured value was stored by the WalkAide stimulator. The participants then underwent gait training on a treadmill. When the affected leg of the participants tilted backward in the late stance phase during walking on the treadmill (when the heels are elevated), the tilt sensor of the WalkAide stimulator guided the dorsiflexion motion of the affected ankle. Also, the initial speed of the treadmill was measured while participants maintained a normal gait pattern at a comfortable cadence and stride length for each participant [7]. The mean treadmill speed was 0.6 m/s at the beginning of gait training. If participants were able to walk comfortably at this speed for more than 3 minutes, the speed of the treadmill was increased by 0.2 m/s. When speed was increased and a participant was unable to perform initial contact properly, speed was reduced once again. After 4 weeks, a mean treadmill speed of 1.2 m/s was reached. During treadmill gait training, participants were allowed to hold onto a safety bar on each side. Before starting training, participants were asked to perform an on over-ground walking for 2 minutes in order to become accustomed to the stimulus of the WalkAide. Treadmill training combined with the tilt sensor in FES was performed for 30 minutes, one time per day, for 4 weeks. The TPTSF group also had treadmill gait training combined with tilt sensor FES. The WalkAide system was attached to the subjects, but the stimulator was set in the “off” position. Other than that, the same methods were used as in the TTSF group. Conventional physical therapy was performed based on the neurodevelopmental treatment concept and proprioceptive neuromuscular facilitation [12,13], which are used by physical therapists to improve the function of affected limbs, as well as balance and gait ability, based on motor control and motor learning theory. Conventional physical therapy was performed for 30 minutes, twice per day, for 4 weeks. 2.5. Outcome measures The TUG and BBS were used to measure balance. The inter-rater reliability of the TUG is r = 0.98 while that of intra-rater reliability is r = 0.99 [14]. The BBS also is highly reliable in assessing dynamic balance, as its inter-rater reliability is 0.99 and intra-rater reliability is 0.98 [15]. The 10 MWT was used to evaluate gait ability. The inter-rater and intra-rater reliabilities were as high as r = 0.89 [16]. The TUG and 10MWT were repeated three times each, and the average values were analyzed. Participants had 10 min rest time between each trial. A portable ultrasound scanner (Mysono U5, Samsung Medison, Seoul, Korea) with a 7 MHz linear transducer was used to measure the muscle structure of the tibialis anterior. A linear transducer probe was used in B-mode, and pressure on the skin was minimized by applying an ample amount of ultrasonic gel between the probe and the skin. The probe and the skin were maintained perpendicular to each other
448
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES Table 2 Changes in the TUG, BBS, and the 10-m walk test
TTSF group (n = 15) Pre-test Post-test Changes TUG (sec) 23.31 ± 3.94 15.03 ± 2.75∗∗∗ −8.28 ± 2.76 BBS (score) 37.87 ± 3.94 50.00 ± 1.89∗∗∗ 12.13 ± 3.44 10 m walk test (sec) 28.02 ± 4.05 20.51 ± 2.94∗∗∗ −7.51 ± 2.66
TPTSF group (n = 15) Pre-test Post-test Changes p value 21.85 ± 3.55 16.96 ± 2.74∗∗∗ −4.89 ± 1.81 0.000+++ 38.20 ± 3.97 45.20 ± 2.65∗∗∗ 8.00 ± 2.98 0.002++ 31.23 ± 5.07 25.99 ± 4.35∗∗∗ −5.24 ± 1.81 0.011+
Values are expressed as mean ± SD. TUG, Timed Up and Go; BBS, Berg Balance Scale. ∗ significant differences between pre and post-test. ∗ p < 0.05; ∗∗ p < 0.01; ∗∗∗ p < 0.001. + significant differences between experimental group and control group. + p < 0.05; ++ p < 0.01; +++ p < 0.001.
to ensure constant measurement. The participants were asked to lie down and wear modified ankle orthoses in which their knees were extended and ankle angle fixed at 90◦ . After the digital manual muscle tester (Power Track II, Jtech Medical Industries, Inc., Herber City, UT, USA) was placed on the distal part of the navicular bone on the top side of the foot, the participants were asked to try to contract the ankle dorsiflexor on the affected side as much as possible. When the participants tried to contract their ankle dorsiflexor, the examiner reported the muscle thickness and pennation angles of ankle on the affected side. To measure the thickness of the muscle, the distance, excluding the tendon sheath, between the superior aponeurosis and central aponeurosis of the tibialis anterior was measured. After checking the fascicle between the superior aponeurosis and central aponeurosis to measure the pennation angle, the angle between the fascicle and central aponeurosis was measured. In addition, for pre- and postmeasurement, the examiner measured the muscle structure after palpating the location of distal part of the navicular bone. The location of the linear transducer probe was at a point 20% of the distance from the tip of the lateral malleolus to the head of the fibula, and measuring points were marked by “X” using a pen [17]. To prevent muscle fibers from pressing downward and causing the image to blur, pressure between the linear probe and skin was minimized, and the image was stored and measured when the pennation angle was most clearly seen. The ultrasound scan was performed by an experienced examiner. The test-retest reliability of the ultrasound scan was r = 0.811 [17]. 2.6. Statistical analyses The SPSS Statistical Package, version 18.0 (IBM Inc., Chicago, IL, USA), was used for statistical analysis. Participant demographics were given as mean and standard deviation. Data were analyzed using the paired t-test for changes in values from before to after intervention. Differences between the two groups at baseline and follow-up were analyzed using an independent t-test. The level of statistical significance was set at 0.05. 3. Results After intervention, there were significant improvements in the TUG, BBS, and 10 MWT compared to baseline in both the TTSF and TPTSF groups (p < 0.05). The TUG, BBS, and 10 MWT showed more significant improvements in the TTSF group as compared to TPTSF group at follow-up (p < 0.05) (Table 2). As compared to baseline, at follow-up, both groups had significant improvements (p < 0.05) in the tibialis anterior, the pennation angles in the resting phase on the affected and non-affected sides, pennation angles in the contraction phase on the affected and non-affected sides, muscle thickness in the resting phase on the affected side, and muscle thickness in the contraction phase on the affected side
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
449
Table 3 Changes in the muscle architecture of the tibialis anterior
Resting phase
Contraction phase
TTSF group (n = 15) TPTSF group (n = 15) Pre-test Post-test Changes Pre-test Post-test Changes 10.50 ± 0.92 11.94 ± 0.76∗∗∗ 1.43 ± 1.07 10.78 ± 0.87 11.58 ± 0.65∗∗ 0.80 ± 1.00
Pennation angles of the affected side (angle) Pennation angles 11.86 ± 0.85 12.95 ± 1.19∗∗ 1.09 ± 1.10 12.05 ± 0.68 13.05 ± 0.85∗∗ 1.00 ± 0.90 of the non-affected side (angle) Muscle thickness 1.55 ± 0.13 1.80 ± 0.12∗∗∗ 0.25 ± 0.10 1.51 ± 0.09 1.66 ± 0.06∗∗∗ 0.16 ± 0.08 of the affected side (cm) 0.03 ± 0.68 1.69 ± 0.07 1.72 ± 0.08 0.04 ± 0.07 Muscle thickness 1.68 ± 0.06 1.71 ± 0.09 of the non-affected side (cm) Pennation angles of the affected side (angle) Pennation angles of the non-affected side (angle) Muscle thickness of the affected side (cm) Muscle thickness of the non-affected side (cm)
p value 0.104
0.818 0.009++
0.879
14.03 ± 1.64 17.05 ± 1.40∗∗∗ 3.02 ± 1.47 14.57 ± 0.73 15.00 ± 0.47∗∗ 0.43 ± 0.52
0.000+++
16.12 ± 1.26 16.88 ± 1.31∗
0.720
0.76 ± 1.29 16.90 ± 1.10 17.53 ± 1.02∗∗ 0.62 ± 0.78
1.74 ± 0.06 1.98 ± 0.11∗∗∗ 0.24 ± 0.13
1.72 ± 0.07 1.83 ± 0.11∗∗ 0.10 ± 0.11
0.005++
1.86 ± 0.07 1.95 ± 0.18
1.90 ± 0.07 1.91 ± 0.07
0.056
0.09 ± 0.16
0.01 ± 0.01
Values are expressed as mean ± SD. ∗ significant differences between pre and post-test. ∗ p < 0.05; ∗∗ p < 0.01; ∗∗∗ p < 0.001. + significant differences between experimental group and control group. + p < 0.05; ++ p < 0.01; +++ p < 0.001.
(Table 3). However, there was no significant improvement in muscle thickness in the resting phase on the non-affected side between pre- and post-tests. Muscle thickness in the resting phase on the affected side, pennation angle of contraction phase on the affected side, and muscle thickness in the contraction phase on the affected side significantly improved in the TTSF group as compared to the TPTSF group at follow-up (p < 0.05) (Table 3). 4. Discussion This study investigated the effects of treadmill gait training combined with tilt sensor FES on balance, gait, and structure of the tibialis anterior muscle in stroke survivors. According to the results of the BBS, TUG, and 10MWT, significant improvements were observed in the TTSF group compared to those in the TPTSF group. Previous studies have reported that gait re-education combined with FES significantly improved balance and gait in chronic stroke patients [18]. Specifically, during gait re-education, dorsiflexion of the ankle could be improved after applying tilt sensor FES to the tibial nerve during the swing phase of the gait, resulting in a significantly improved gait and better balance. Lee et al. reported that balance could be significantly improved in participants when FES is applied at the point where the tibialis anterior contracts during treadmill gait training [6]. They reported that FES, which encourages the active participation of the patient, had more positive effects on functional movement and balance ability than conventional FES [6]. In this study, tilt sensor FES controlled by the angle of the knee gradient changes
450
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
was implemented, which encourages active participation of the patient. In the results of this study, gait training combined with active tilt sensor FES was effective for improving balance in stroke patients. When the effects of treadmill gait training combined with FES were compared with those of on ground gait training in 30 stroke patients, Peurala et al. reported that treadmill gait training combined with FES was more effective for improving gait speed and dynamic balance [19]. Their results coincide with the results of this study. Treadmill gait training combined with FES improved the symmetric parameters in stroke patients, which had positive effects on stability, balance for walking, and gait speed. It has been previously reported that repetitive and task-oriented training such as treadmill gait training improves balance in stroke patients through motor relearning by increasing neuroplasticity and reorganization of the brain [20]. The somatosensory response of the ankle to which dorsiflexion was applied is stimulated through FES while brain plasticity is accelerated, which may decrease foot drop. The basic role of tilt sensor FES is to supplement activation of damaged efferent pathways [10]. In sensory-motor integration theory, sensory stimulation provided by the movement of affected limbs directly affects continuous motor output [21]. In other words, impairment of the somatosensory response is responsible for the delayed motor recovery of stroke patients. Therefore, appropriate stimulation of the somatosensory response is essential for improvement in motor recovery [22]. According to previous studies, activation of damaged efferent pathways was most effective when combined with active movements [23]. In addition, increase of somatosensory stimuli during walking can improve the gait patterns of stroke patients [24]. Accordingly, the tilt sensor FES used in this study improved stimulation of the somatosensory response through active knee flexion along with dorsiflexion of the ankle during treadmill gait training, providing additional positive effects in improving gait ability. Currently, ultrasound test equipment is being used to assess rehabilitation to improve muscle function in patients. The ultrasound test used in the rehabilitation field is called rehabilitative ultrasound imaging (RUSI) [25]. According to the results of the RUSI test in this study, treadmill gait training both with and without tilt sensor FES significantly improved the pennation angle between the affected and non-affected parts of the tibialis anterior as well as muscle thickness in resting and contraction phases. However, the treadmill gait training group subjected to tilt sensor FES showed significant improvement in both pennation angle and muscle thickness in the resting and contraction phases on the affected side as compared to the group subjected to placebo tilt sensor FES. For the non-affected side, there were no significant differences between the resting phase and the contraction phase. According to this result, treadmill gait training combined with tilt sensor FES had positive effects on the tibialis anterior of both the affected and non-affected sides, especially the structure of the tibialis anterior on the affected side. Although there is a lack of studies on improving the function of muscles using ultrasound assessment, a previous study showed that the pennation angle increases during ankle dorsiflexion in an analysis of structural changes in the pennation angle during the resting phase as well as during maximal isometric muscle contraction [26]. Bland et al. investigated the relationship between the structure of the tibialis anterior and other muscles related to gait in cerebral palsy patients and found that muscle thickness and length of the muscle fibers are highly correlated with biomechanical elements such as muscular strength and gait speed [27]. In the present study, by treadmill gait training and tilt sensor FES, the tibialis anterior of the affected side was activated during swing phase and thus its structure on the tibilais anterior was changed. Thus, treadmill gait training with tilt sensor FES can be effective in improving balance, gait, and the structure of the tibialis anterior. In particular, tilt sensor FES controlled by the angle of the knee gradient changes can have an advantage in promoting active participation of patients; conventional FES is controlled only by the time set depending on gait cycle. However, tilt sensor FES may have a disadvantage in that it cannot be used by patients who cannot move their knees. In addition, there were some
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES
451
limitations in this study. First, the number of participants was small. Second, there was no comparison to conventional FES to confirm that effects of tilt sensor FES were different than those of conventional FES. Therefore, treadmill gait training using tilt sensor FES in this study possibly affected the tendon as well as the structure of the tibialis anterior. Thus, investigations on changes in tendons as well as muscle structure should be performed. Further, since this study used only the 10 WT to assess gait ability, qualitative elements as well as spatiotemporal elements should be assessed.
Acknowledgement This study was supported by Sahmyook University.
Conflict of interest The authors have no conflicts of interest to declare.
References [1] [2]
P.W. Duncan, Stroke disability, Physical Therapy 74 (1994), 399-407. P.T. Cheng, C.L. Chen, C.M. Wang et al., Leg muscle activation patterns of sit-to-stand movement in stroke patients, American Journal of Physical Medicine & Rehabilitation/Association of Academic Physiatrists 83 (2004), 10-16. [3] T.M. Kesar, R. Perumal, A. Jancosko et al., Novel patterns of functional electrical stimulation have an immediate effect on dorsiflexor muscle function during gait for people poststroke, Physical Therapy 90 (2010), 55-66. [4] S. Fisher, L. Lucas and T.A. Thrasher, Robot-assisted gait training for patients with hemiparesis due to stroke, Topics in Stroke Rehabilitation 18 (2011), 269-276. [5] C. Bulley, J. Shiels, K. Wilkie et al., User experiences, preferences and choices relating to functional electrical stimulation and ankle foot orthoses for foot-drop after stroke, Physiotherapy 97 (2011), 226-233. [6] H.J. Lee, K.H. Cho and W.H. Lee, The effects of body weight support treadmill training with power-assisted functional electrical stimulation on functional movement and gait in stroke patients, American Journal of Physical Medicine & Rehabilitation / Association of Academic Physiatrists 92 (2013), 1051-1059. [7] S. Hesse, Locomotor therapy in neurorehabilitation, NeuroRehabilitation 16 (2001), 133-139. [8] W.T. Liberson, H.J. Holmquest, D. Scot et al., Functional electrotherapy: stimulation of the peroneal nerve synchronized with the swing phase of the gait of hemiplegic patients, Archives of Physical Medicine and Rehabilitation 42 (1961), 101-105. [9] T. O’Halloran, M. Haugland, G.M. Lyons et al., An investigation of the effect of modifying stimulation profile shape on the loading response phase of gait, during FES-corrected drop foot: Stimulation profile and loading response, Neuromodulation : Journal of the International Neuromodulation Society 7 (2004), 113-125. [10] G. Morone, A. Fusco, P. Di Capua et al., Walking training with foot drop stimulator controlled by a tilt sensor to improve walking outcomes: A randomized controlled pilot study in patients with stroke in subacute phase, Stroke Research and Treatment 2012 (2012), 523-564. [11] A.R. Lindquist, C.L. Prado, R.M. Barros et al., Gait training combining partial body-weight support, a treadmill, and functional electrical stimulation: effects on poststroke gait, Physical Therapy 87 (2007), 1144-1154. [12] P. Klimkiewicz, A. Kubsik and M. Woldanska-Okonska, NDT-Bobath method used in the rehabilitation of patients with a history of ischemic stroke, Wiadomo´sci lekarskie (Warsaw, Poland: 1960). 65 (2012), 102-107. [13] H.S. Reichel, PNF (proprioceptive neuromuscular facilitation): gait training, Sportverletzung Sportschaden : Organ der Gesellschaft für Orthopädisch-Traumatologische Sportmedizin 10 (1996), A11-20. [14] S.S. Ng, C.W. Hui-Chan, The timed up & go test: its reliability and association with lower-limb impairments and locomotor capacities in people with chronic stroke, Archives of Physical Medicine and Rehabilitation 86 (2005), 1641-1647. [15] S. Downs, J. Marquez and P. Chiarelli, The Berg Balance Scale has high intra- and inter-rater reliability but absolute reliability varies across the scale: A systematic review, Journal of Physiotherapy 59 (2013), 93-99.
452 [16] [17] [18] [19] [20] [21] [22] [23]
[24] [25] [26] [27]
D.-Y. Hwang et al. / Treadmill training with tilt sensor FES G. Scivoletto, F. Tamburella, L. Laurenza et al, Validity and reliability of the 10-m walk test and the 6-min walk test in spinal cord injury patients, Spinal Cord 49 (2011), 736-740. K. McCreesh and S. Egan, Ultrasound measurement of the size of the anterior tibial muscle group: The effect of exercise and leg dominance, Sports Medicine, Arthroscopy, Rehabilitation, Therapy & Technology: SMARTT 3 (2011), 18. J.A. Robertson, J.J Eng and C. Hung, The effect of functional electrical stimulation on balance function and balance confidence in community-dwelling individuals with stroke, Physiotherapy Canada. Physiothérapie Canada 62 (2010), 114-119. S.H. Peurala, I.M. Tarkka, K. Pitkanen et al., The effectiveness of body weight-supported gait training and floor walking in patients with chronic stroke, Archives of Physical Medicine and Rehabilitation 86 (2005), 1557-1564. L.R. Harvey, Improving poststroke recovery: neuroplasticity and task-oriented training, Current Treatment Options in Cardiovascular Medicine 11 (2009), 251-259. D.A. Rosenbaum, R.G. Meulenbroek and J. Vaughan, Planning reaching and grasping movements: theoretical premises and practical implications, Motor Control 5 (2001), 99-115. C.W. Wu, H.J. Seo and L.G. Cohen, Influence of electric somatosensory stimulation on paretic-hand function in chronic stroke, Archives of Physical Medicine and Rehabilitation 87 (2006), 351-357. A. Boyaci, O. Topuz, H. Alkan et al., Comparison of the effectiveness of active and passive neuromuscular electrical stimulation of hemiplegic upper extremities: a randomized, controlled trial, International Journal of Rehabilitation Research. Internationale Zeitschrift für Rehabilitationsforschung. Revue Internationale de Recherches de Réadaptation 36 (2013), 315-322. S.I. Lin, L.J. Hsu and H.C. Wang, Effects of ankle proprioceptive interference on locomotion after stroke, Archives of Physical Medicine and Rehabilitation 93 (2012), 1027-1033. D.S. Teyhen, J.D. Childs and T.W. Flynn, Rehabilitative ultrasound imaging: When is a picture necessary, The Journal of Orthopaedic and Sports Physical Therapy 37 (2007), 579-580. C.N. Maganaris and V. Baltzopoulos, Predictability of in vivo changes in pennation angle of human tibialis anterior muscle from rest to maximum isometric dorsiflexion, European Journal of Applied Physiology and Occupational Physiology 79 (1999), 294-297. D.C. Bland, L.A. Prosser and L.A. Bellini, Tibialis anterior architecture, strength, and gait in individuals with cerebral palsy, Muscle Nerve 44 (2011), 509-517.
