Effects of robot-assisted walking training on balance,motor function,and ADL depending on severity levels in stroke patients

Abstract

BACKGROUND:

Despite the explosive increase in interest regarding Robot-Assisted Walking Training (RAWT) for stroke patients, very few studies have divided groups according to the severity levels of patients and conducted studies on the effects of RAWT.

OBJECTIVE:

The purpose of this study was to present a definite basis for physical therapy using the robot-assisted walking device through a more detailed comparison and analysis and to select the optimal target of RAWT.

METHODS:

This study was designed as a prospective and randomized controlled trial to investigate the effect of RAWT on balance, motor function, and Activities of Daily Living (ADL) depending on severity levels in stroke patients. 100 participants were randomly divided into study and control groups in equal numbers. The study group was 49 and the control group was 47. One from the study group and three from the control group were eliminated. The study period is four weeks in total, and RAWT is performed five times a week for 40 minutes only for study group. During the same period, all group members had 30 minutes of Conventional Physiotherapy (CP) five times a week.

RESULTS:

The results of this study clearly confirmed that RAWT combined with CP produces more significant improvement in patients with stroke than the CP alone. And they indicated that RAWT had a more considerable effect in the poor or fair trunk control group for trunk balance and in the high fall risk group for balance. In motor function, RAWT showed its value in the severe and marked motor impairment group. The total or severe dependence group in ADL experienced more improvements for RAWT.

CONCLUSION:

This study can be concluded that the lower the level of physical functions, the more effective it responds to RAWT. As demonstrated in the results of this study, the potential of current robotic technology appears to be greatest at very low functional levels of stroke patients. Patients with low functional levels among stroke patients may benefit from robot rehabilitation.

Keywords

ADL balance motor function robot-assisted walking training stroke

1. Introduction

Stroke, one of the most common brain lesion diseases, causes many disorders in daily living activities due to motor dysfunction, cognitive dysfunction, perceptual disorder, and speech problems [1].

After a stroke, 51% of patients initially cannot walk for themselves, 12% of them can walk with support and only 37% can walk independently in a hospital setting [2]. Even though the walking ability is improved, the walking speed is naturally lower than that of a normal person, so there are not many cases of doing community ambulation without problem [3]. Therefore, the recovery of walking ability becomes the most interesting treatment for patients, family members, and therapists after a stroke [4]. Conventional Physiotherapy (CP) for walking remains the most commonly used therapeutic intervention [5]. CP may include maintaining or increasing the range of joint motion, stimulating neuromuscular reeducation, and producing positive alignment of the trunk through maintaining sitting and standing posture. It also includes various ways of walking training.

However, most stroke patients with more than a certain degree of damage may not be able to recover their normal walking patterns even with regular conventional walking training [6]. CP has a limitation in that the therapist treats a large number of patients at a constant intensity every day. The level and competence of the therapist play a significant effect in the treatment, leading to the inconsistency of therapy along with the intensity of work given to each therapist. For these reasons, we desperately need new and advanced physical therapeutic approaches that can help stroke patients improve their walking ability and upgrade their quality of life [7] Under this circumstance, there is a growing interest in the robot therapeutic system. They can be the one of useful ways to enhance the walking function of stroke patients [8, 9]

Recently, many studies on Robot Assisted Walking Training (RAWT) have been conducted in various ways for patients with stroke [10, 11, 12, 13, 14, 15]. Wu et al. [10], and Baronchelli et al. [12] revealed that RAWT showed improvements in the balance function of stroke patients more in comparison with the results of general conventional physical therapy. Mehrholz et al. [16] compared and analyzed the walking ability after RAWT and neurological physical therapy intervention for patients with hemiplegia. This study revealed that stroke patients who received walking training based on robotic devices improved their walking ability compared to those who did not. In addition, Mustafaoglu et al. [15] indicate that mobility, Activities of Daily Living (ADL), and quality of life were further improved in a group of stroke patients performed with a combination of RAWT and CP.

According to previous studies [8, 9, 10, 11, 12, 13, 14, 15, 17], RAWT has been reported to be effective for stroke patients in balance, mobility, ADL, and walking, therefore, physical therapy using a robotic system has received a lot of attention. However, despite considering these broad and diverse outcomes, it is still elusive whether the positive effects of RAWT outweigh the positive effects of CP in stroke patients. Some studies have also shown that CP alone is more beneficial for patients’ walking, arguing that RAWT supports walking training only in fixed and linear two-dimensional directions, while therapists assist patients in walking in various three-dimensional directions [18, 19].

In the studies above, we also found that some studies had favorable results for CP, others had favorable results for RAWT, and another found no significant difference in functional gait improvement between the two types of treatment. The fact that there was no clear demonstration of the superiority of RAWT draws frustration to physical therapists and health professionals. Despite the advantages of robot devices, the promise of training concepts and techniques does not appear to have been fulfilled yet.

The objective of this study is to (1) compare the effect of RAWT plus CP with the effect of CP alone on functional outcomes, including balance, motor function of extremities, and ADL. Through a more detailed comparison and analysis of each of the variables representing the functional outcomes listed above, this study aims to (2) present a definite basis for physical therapy using a robot-assisted walking device. In addition, the participants are classified according to severity levels, and the effect of RAWT on the classified individual groups is compared and analyzed to (3) figure out the optimal target of RAWT.

The hypothesis of this study is that there will be different results in balance and motor function, ADL according to the severity levels of patients within and between study and control group. After measuring and comparing the mentioned variables, the participants are classified according to severity levels, followed by the comparison and analysis of the effect of RAWT on the classified individual groups to achieve the purpose of the study. With these outcomes, this study aims to present a definite basis for physical therapy using a robot-assisted walking device.

2. Methods

This study was designed as a prospective and randomized controlled trial to investigate the effect of RAWT on balance, motor function, and ADL depending on severity levels in stroke patients. In this study, out of a total of 100 stroke patients from a single medical institution, 96 patients who met the inclusion criteria were randomly assigned to study and control groups. This assignment was done using a random number generator, specifically the “RANDBETWEEN” function in Microsoft Excel, which selected numbers 1 and 2. This freely available software enables both simple and blocked random allocation. To enhance the objectivity of the assessments, all measurement procedures were conducted by assessors who were blinded to the intervention status of the participants. The study was conducted as a double-blind trial for both researchers and subjects. It adhered to the Consolidated Standards of Reporting Trials (CONSORT) guidelines [20]. The framework of this study is depicted in Fig. 1. All study protocols were approved and reviewed by the institutional review committee of Yong In University (2-1040966-AB-N-01-2210- HSR-277-2). The study period is four weeks in total (as shown in Fig. 1). Demographic details of the stroke survivors are listed in Table 1. All stroke patients mentioned above were hospitalized at S rehabilitation hospital in the Seoul metropolitan area.

Table 1
General characteristics of the subjects ( $n=$ 96)

	Study ( $n=$ 49)	Control ( $n=$ 47)	$x^{2}/$ t	$p$
Gender (M/F, %)	31/18 (63/37)	28/19 (59/41)	0.14^b	0.833
Diagnosis (H/I)	30/19 (61/39)	29/18 (62/38)	0.01	0.962
Affected side (R/L, %)	22/27 (44/56)	18/29 (38/62)	0.43	0.541
Age (year)	65.53 $\pm$ 12.17^a	71.08 $\pm$ 12.07	1.24^c	0.872
Height (cm)	165.00 $\pm$ 8.87	163.72 $\pm$ 9.55	0.67	0.491
Weight (kg)	61.04 $\pm$ 10.05	60.83 $\pm$ 10.40	0.09	0.923
Duration (month)	7.36 $\pm$ 8.63	7.23 $\pm$ 6.52	0.08	0.934
MMSE	21.46 $\pm$ 4.79	21.12 $\pm$ 5.90	0.31	0.771

M $=$ Male; F $=$ Female; H $=$ Hemorrhage; I $=$ Infarction; R $=$ Right; L $=$ Left. ^aMean $\pm$ standard deviation, ^bChi-square test, ^cIndependent $t$ -test.

Figure 1.

Visual representation following the CONSORT guidelines for the participant enrollment process in this study. Abbreviations: TIS, trunk impairment scale; BBS, berg balance scale; FMA, Fugl-Meyer Assessment; MBI, Modified Barthel Index.

A physical therapist screened all subjects according to the inclusion and exclusion criteria. As a result, all met the following inclusion criteria: no prior stroke, no other neurologic or orthopedic disorder, independent ambulation before the stroke, and no severe medical illnesses [21]. Subjects were excluded if they had evidence of multiple strokes, chronic white matter disease on magnetic resonance imaging, congestive heart failure, peripheral artery disease with intermittent claudication, cancer, pulmonary or renal failure, unstable angina, dementia (Mini-Mental State Exam $<$ 15), severe aphasia, orthopedic conditions affecting the legs or the back, or cerebellar signs such as ataxia [22].

This study compares the effect of RAWT plus CP with the effect of CP alone by measuring balance, motor function of extremities, and ADL. Trunk Impairment Scale (TIS) and Berg Balance Scale (BBS) were used for the evaluation of balance, and Fugl-Meyer Assessment (FMA) was used to evaluate motor function of extremities. ADL was assessed using Modified Barthel Index (MBI).

2.1 Experimental procedure

Between November 2022 and April 2023, 100 participants participated in the study. Subjects were divided into the study and control group in equal numbers. One from the study group and three from the control group were eliminated.

The sample size was determined using G*Power software (version 3.1.9.6; Heinrich Heine University, Düsseldorf, Germany) [23, 24] under the following configurations: t-tests, a calculated size effect (1.43) based on a previous study [25, 26], a significance level ( $\alpha$ ) of 0.05, and a desired statistical power of 0.80. The determined sample size was 96 participants, evenly distributed, with 49 individuals in the study group and 47 individuals in the control group. Statistical tools were comprised of descriptive analysis of baseline characteristics using the mean and standard deviation. Patients who received RAWT plus CP were defined as the study group, and patients who received CP alone were defined as the control group.

The control group received 30 minutes of CP based on neurodevelopmental treatment 5 times a week for 4 weeks. CP was conducted based on neurophysiological treatment concept [27]. Depending on the patient’s physical functional level, walking training focused on the promotion of trunk alignment, weight-bearing, shifting and transfer in standing posture, and walking under supervision by the therapist. If necessary, walking training might be conducted using parallel bars, and further assistance from other therapists might be needed [28]. In principle, CP sessions should last for 30 minutes.

Figure 2.

RAWT with Lokomat V6 (Hocoma AG, Zurich, Switzerland).

The RAWT was carried out using Lokomat (Lokomat, Hocoma AG, Zurich, Switzerland) (as shown in Fig. 2), a running device that executes robot walking, which includes a treadmill, a patient weight support system with harnesses, and two robotic drives attached to the patient’s legs. Several robotic joints are attached to the patient’s leg to facilitate hip, knee, and ankle movements. When a patient is lifted by a body weight support device on a treadmill for walking, the degree of weight support can be adjusted. Although slight differences depend on the patient’s condition, RAWT is initially started by supporting about 50% of the patient’s weight. The degree of body weight support is gradually reduced as the therapeutic intervention progresses. Guidance force starts at 100% and gradually decreases depending on the patient’s walking ability, encouraging stroke patients to walk normally. In addition, this enhances patient motivation and concentration [18]. RAWT sessions were incorporated into CP for only study group. RAWT was conducted 5 times a week for 4 weeks, and 40 minutes per session. If the patients complained fatigue or pain during RAWT, they were immediately stopped and rested. For safety, physical therapist and assistant stood by patients.

Dunsky et al. [29] stated that the subjects of the group who used robotic system to conduct walking training through visual feedback stimulation and imagination training for the stroke patients showed better results in walking speed and stride length than those of the group who did not. During sessions, biofeedback combined with a virtual reality system can be used. Virtual reality avatars visualized in the monitor help patients walk actively. In virtual reality, a patient can perform various functional movement tasks such as avoiding obstacles, catching an animal, and participating in sports games. Patients can perform repetitive task-oriented training beyond simple repetitive training through these accurate tasks. The selection of virtual reality applications during RAWT is important successful rehabilitation, because these applications may have varying effects on balance and walking ability [11].

2.2 Assessment tools

2.2.1 Balance abilities

Trunk Impairment Scale (TIS)

The Trunk Impairment Scale (TIS) was developed by Verheyden et al. in 2004, which is used to evaluate the stability, balance, and coordination of the trunk for stroke patients in a sitting position. Compared to traditional balance assessments conducted based on standing posture, it has an advantage when applied to patients with relatively low physical functional levels [30]. Its score ranges from a minimum of 0 to a maximum of 23. The more perfectly you perform the required movement, the higher the score. Therefore, the higher the score, the better the trunk balance. Recently, the TIS version (version 2.0), excluding static balance checks, is used as a way to exclude the ceiling effect of the inspection, which has a score range of 0 to 16 points [31]. The previous version was used in this study. Scores of 20–23 are considered good trunk control, 11–19 as fair trunk control, and 0–10 as poor trunk control [32].

TIS, which is primarily used for trunk evaluation in stroke patients, has sufficient reliability, internal consistency, and validity to be used in research related to clinical practice and balance ability [31]. In this study the criteria for evaluating balance depending on the severity levels of patients were classified into TIS1 (0–5), TIS2 (6–10), and TIS3 (11–19) according to the participants’ TIS score distribution.

Berg Balance Scale (BBS)

Berg Balance Scale (BBS) was initially developed to evaluate the balance ability for the elderly and analyze the risk of falling based on the results [33, 34]. This measurement has been widely used to evaluate the balance of patients with central nervous system disorders such as stroke and traumatic brain injury due to its excellent reliability and validity in stroke patients [35]. It has 14 items to measure balance in sitting or standing positions. The measurement score ranges from a minimum of 0 to a maximum of 56.

The higher the score, the better the balance ability. The total score must be at least 45 for safe and independent walking [36]. Scores of 41–56 out of 56 are considered low fall risk, 21–40 as medium fall risk, and 0–20 as high fall risk [33, 34, 37]. In old adults and patients with stroke, this scale has higher test-retest and inter-tester reliability [38]. In this study, the criteria for evaluating balance depending on the severity levels of patients were classified into BBS1 (0–5), BBS2 (6–20), and BBS3 (21–40) according to the participants’ BBS score distribution.

2.2.2 Motor function of extremities

Fugl-Meyer Assessment (FMA)

The Fugl-Meyer Assessment (FMA) is a scale to assess motor function suitable for stroke patients. It is designed to evaluate motor function, balance, sensation, and joint function of the patient’s hemiplegic body [39]. Evaluation using FMA helps health professionals determine the severity of patient disease, predict movement recovery, and make a treatment plan [40]. The motor function of upper extremity test consists of 33 items, with a perfect score of 66 points, while the lower extremity test consists of 17 items, with a perfect score of 34 points.

The total possible scale score of motor function is 100. Scores are grouped according to the various level of impairment, which are as follows; 96–99 points $=$ slight motor impairment, 85–95 points $=$ moderate motor impairment, 50–84 points $=$ marked motor impairment, and $<$ 50 points $=$ severe motor impairment [41]. This scale has higher test-retest and inter-tester reliability [42]. In this study, the criteria for evaluating motor function of extremities depending on severity levels of patients were classified into FMA1 (0–20), FMA2 (21–49), and FMA3 (50–84) according to the participants’ FMA score distribution.

2.2.3 ADL

Modified Barthel Index (MBI)

The Bartel Index (BI) consists of a total of 10 items to evaluate the independence of activities of daily living which includes personal hygiene, bathing self, feeding, toilet, stair climbing, dressing, bladder control, bowel control, ambulation or wheelchair, and chair-to-bed transfer. BI was amended to MBI in 1989 to improve the agility and reliability of the evaluation tool [43]. Contrary to the original purpose, it was widely used as an important indicator for evaluating the degree of functional improvement and deterioration in stroke patients [44].

Evaluation scores range from a minimum of 0 to a maximum of 100, and higher scores mean that patients do not need help from others in their daily lives. A score of less than 24 out of 100 tends to predict total dependence in motor skills and difficulty with other basic skills. Scores of 24–49 are considered severe dependence, 50–74 as moderate dependence, 75–90 as mild dependence, and 91–99 as slight dependence. A score of 100 indicates that the patient is independent of assistance from others [43]. In some cases, the evaluation criteria are separated into less than 40 points and more than 60 points [45]. This scale has higher test-retest and inter-tester reliability for stroke patients [46]. In this study, the criteria for evaluating independence in ADL depending on severity levels of patients were classified into MBI1 (0–23), MBI2 (24–49), and MBI3 (50–74) according to the patients’ MBI score distribution.

2.3 Data analysis

The general characteristics of the study subjects were surveyed on all subjects before the experiment. Data obtained from the experiment was processed using SPSS version 18.0 statistical package (IBM SPSS, Armonk, NY, USA) and all statistical processes were tested at 0.05 significance level. Shapiro-Wilk test was performed to verify the normality of the dependent variable, and a parametric test was used because the measured values satisfied the normal distribution. Continuous variables are presented as the mean and standard deviation (SD), whereas nominal variables are presented as frequencies and percentages.

Paired $t$ -test was performed to compare the difference between before and after the experiment within group. Before comparing between groups, Chi-Squared test and Independent $t$ -test were used for the test of homogeneity between groups. In order to compare the study with control groups, Independent $t$ -test was used. Kruskal-Wallis test was used for comparison between three groups which were divided into the severity, and then Mann-Whitney U test as a post-hoc test was implemented one after another.

3. Results

3.1 General characteristics of the subjects

The Table 1 shows the subject’s general characteristics.

3.2 Clinical outcomes

3.2.1 Comparison by severity levels on balance

Comparison by severity levels within the group on TIS

In the comparison between the three groups divided by severity levels in the study group, there were significant differences between TIS2 and TIS1 (TIS2 $>$ TIS1) ( $p<$ 0.001), between TIS3 and TIS1 (TIS3 $>$ TIS1) ( $p<$ 0.01), and between TIS2 and TIS3 (TIS2 $>$ TIS3) ( $p<$ 0.05). The response to RAWT might be more effective in the order of TIS2, TIS3, and TIS1 (Table 2).

Table 2
Comparison by severity levels within and between the group on TIS ( $n=$ 96)

Variables

Study (

n=

49)

Control (

n=

47)

(

n=

SG/CG)

Pre

Post

Change (post-pre)

Pre

Post

Change (post-pre)

t

p

TIS1 (

n=

26/23)

1.46

\pm

1.55^a

3.65

\pm

2.86

2.19

\pm

1.81^b***

1.42

\pm

2.01

2.38

\pm

3.04

0.95

\pm

1.74^*

-

2.512

0.016^c

TIS2 (

n=

13/10)

6.23

\pm

1.09

11.61

\pm

1.89

5.38

\pm

1.75^***

6.23

\pm

2.68

9.38

\pm

3.66

3.15

\pm

0.62^**

-

1.369

0.182

TIS3 (

n=

10/14)

14.30

\pm

3.09

18.10

\pm

3.63

3.80

\pm

1.13^***

17.07

\pm

4.34

19.00

\pm

3.65

1.92

\pm

2.21^**

-

2.592

0.017

x^{2}/

20.184

9.097

p

00.000^d

0.011

Post hoc

(Mann-

Whitney)

TIS2

>

TIS1

(

p<

0.001)

TIS3

>

TIS1

(

p<

0.01)

TIS2

>

TIS3

(

p<A

0.05)

TIS2

>

TIS1 (

p

< 0.01)

TIS $=$ Trunk Impairment Scale; TIS1 $=$ score 0–5 group; TIS2 $=$ score 6–10 group; TIS3 $=$ score 11–19 group; SG $=$ Study Group; CG $=$ Control Group. ^aMean $\pm$ standard deviation, ^bPaired $t$ -test, ^cIndependent $t$ -test, ^dKruskal-Wallis test, $p<$ 0.05^*, $p<$ 0.01^**, $p<$ 0.001^***.

In the comparison between the three groups divided by severity levels in the control group, there were significant differences only between TIS2 and TIS1 (TIS2 $>$ TIS1) ( $p<$ 0.01). The response to CP might be more effective in TIS2 than others.

Comparison by severity levels between the groups on TIS

There were significant changes in TIS1 ( $p<$ 0.001), TIS2 ( $p<$ 0.001), and TIS3 ( $p<$ 0.001) in study group before and after intervention. There were also significant changes in TIS1 ( $p<$ 0.05), TIS2 ( $p<$ 0.01), and TIS3 ( $p<$ 0.01) in control group before and after intervention (Table 2). After the intervention, the measurement of TIS in both the study and control group was statistically better than before the intervention.

In the comparison between the study and control group, there were significant differences in TIS1 ( $p<$ 0.05), and TIS3 ( $p<$ 0.05). These findings implied that TIS1, and TIS3 groups responded more effectively to RAWT than TIS2.

Comparison by severity levels within the group on BBS

In the comparison between the three groups divided by severity levels in the study group, there were no significant differences between three groups. The response to RAWT might be similar in BBS1, BBS2, and BBS3 (Table 3).

Table 3

Comparison by severity levels within and between the group on BBS ( $n=$ 96)

Variables

Study (

n=

49)

Control (

n=

47)

(

n=

SG/CG)

Pre

Post

Change (post-pre)

Pre

Post

Change (post-pre)

t

p

BBS 1 (

n=

23/20)

1.82

\pm

1.40^a

14.08

\pm

9.61

12.26

\pm

8.69^b***

1.80

\pm

1.60

4.28

\pm

4.37

2.47

\pm

3.51**

-

4.808

0.000^c

BBS 2 (

n=

17/11)

9.52

\pm

4.50

23.00

\pm

10.80

13.47

\pm

10.88^***

10.15

\pm

4.61

18.69

\pm

7.16

8.53

\pm

6.37^***

-

1.448

0.159

BBS 3 (

n=

9/16)

29.22

\pm

7.83

38.88

\pm

9.63

9.66

\pm

6.70^**

34.69

\pm

8.07

43.61

\pm

9.66

8.92

\pm

8.03^**

-

0.997

0.340

x^{2}/

0.423

9.682

p

0.809^d

0.008

Post hoc

(Mann-

Whitney)

BBS2

>

BBS1 (

p<

0.01) BBS3

>

BBS1 (

p<

0.05)

BBS $=$ Berg Balance Scale; BBS1 $=$ score 0–5 group; BBS2 $=$ score 6–20 group; BBS3 $=$ score 21–40 group; SG $=$ Study Group; CG $=$ Control Group. ^aMean $\pm$ standard deviation, ^bPaired $t$ -test, ^cIndependent $t$ -test, ^dKruskal-Wallis test, $p<$ 0.05^*, $p<$ 0.01^**, $p<$ 0.001^***.

In the comparison between the three groups divided by severity levels in the control group, there were significant differences between BBS2 and BBS1 (BBS2 $>$ BBS1) ( $p<$ 0.01) and between BBS3 $>$ BBS1 ( $p<$ 0.05). The response to CP might be more effective in BBS2 and BBS3 than BBS1.

Comparison by severity levels between the groups on BBS

There were significant changes in BBS1 ( $p<$ 0.001), BBS2 ( $p<$ 0.001), and BBS3 ( $p<$ 0.01) in study group before and after intervention. There were also significant changes in BBS1 ( $p<$ 0.01), BBS2 ( $p<$ 0.001), and BBS3 ( $p<$ 0.01) in control group before and after intervention (Table 3). After the intervention, the measurement of BBS in both the study and control group was statistically better than before the intervention.

In the comparison between the study and control group, there were significant differences only in BBS1 ( $p<$ 0.001). These findings implied that BBS1 group responded more effectively to RAWT than other.

3.2.2 Comparison by severity levels on motor function

Comparison by severity levels within the group on FMA

In the comparison between the three groups divided by severity levels in the study group, there were significant differences between FMA1 and FMA2 (FMA1 $>$ FMA2) ( $p<$ 0.001), and between FMA1 and FMA3 (FMA1 $>$ FMA3) ( $p<$ 0.001). The response to RAWT might be more effective in FMA1 than others (Table 4).

Table 4
Comparison by severity levels within and between the group on FMA ( $n=$ 96)

Variables

Study (

n=

49)

Control (

n=

47)

(

n=

SG/CG)

Pre

Post

Change (post-pre)

Pre

Post

Change (post-pre)

t

p

FMA 1 (

n=

27/18)

11.00

\pm

3.82^a

42.92

\pm

8.77

31.92

\pm

8.49^b***

12.05

\pm

5.09

17.55

\pm

13.52

5.50

\pm

11.20

-

8.816

0.000^c

FMA 2 (

n=

14/15)

34.42

\pm

9.24

54.64

\pm

10.74

20.21

\pm

7.85^***

34.73

\pm

8.13

40.53

\pm

9.19

5.80

\pm

4.78^***

-

6.015

0.000

FMA 3 (

n=

8/14)

62.00

\pm

11.47

77.75

\pm

4.13

15.75

\pm

7.32^***

71.64

\pm

10.60

79.14

\pm

10.89

7.50

\pm

7.49^**

-

3.072

0.006

x^{2}/

20.619

3.124

p

00.000^d

0.210

Post hoc

(Mann-

Whitney)

FMA1

>

FMA2

(

p<

0.001)

FMA1

>

FMA3

(

p<

0.001)

FMA $=$ Fugl-Meyer Assessment; FMA1 $=$ score 0–20 group; FMA2 $=$ score 21–49 group; FMA3 $=$ score 50–84 group; SG $=$ Study Group; CG $=$ Control Group. ^aMean $\pm$ standard deviation, ^bPaired t-test, ^cIndependent t-test, ^dKruskal-Wallis test, $p<$ 0.05^*, $p<$ 0.01^**, $p<$ 0.001^***.

In the comparison between the three groups divided by severity levels in the control group, there were no significant differences between three groups. The response to CP might be similar in FMA1, FMA2, and FMA3.

Comparison by severity levels between the groups on FMA

There were significant changes in FMA1 ( $p<$ 0.001), FMA2 ( $p<$ 0.001), and FMA3 ( $p<$ 0.001) in study group before and after intervention. There were also significant changes in FMA2 ( $p<$ 0.001), and FMA3 ( $p<$ 0.01) in control group before and after intervention (Table 4). After the intervention, the measurement of FMA in both the study and control group was statistically better than before the intervention except for FMA1 of control group.

In the comparison between the study and control group, there were significant differences in FMA1 ( $p<$ 0.001), FMA2 ( $p<$ 0.001) and FMA3 ( $p<$ 0.01). In particular, the difference was clear in FMA1 and FMA2. These findings implied that all FMA groups responded effectively to RAWT.

3.2.3 Comparison by severity levels on ADL

Comparison by severity levels within the group on MBI

In the comparison between the three groups divided by severity levels in the study group, there were significant differences between MBI1 and MBI3 (MBI1 $>$ MBI3) ( $p<$ 0.01), and between MBI2 and MBI3 (MBI2 $>$ MBI3) ( $p<$ 0.01). The response to RAWT might be more effective in MBI1 and MBI2 than MBI3 (Table 5).

Table 5
Comparison by severity levels within and between the group on MBI ( $n=$ 96)

Variables

Study (

n=

49)

Control (

n=

47)

(

n=

SG/CG)

Pre

Post

Change (post-pre)

Pre

Post

Change (post-pre)

t

p

MBI 1 (

n=

15/16)

16.00

\pm

5.61^a

42.40

\pm

15.65

26.40

\pm

14.24^b***

7.12

\pm

6.17

11.31

\pm

6.75

4.18

\pm

4.33^**

-

5.955

0.000^c

MBI 2 (

n=

22/17)

35.95

\pm

6.89

56.86

\pm

11.35

20.90

\pm

11.18^***

34.64

\pm

6.77

42.41

\pm

16.72

7.76

\pm

13.35^*

-

3.344

0.002

MBI 3 (

n=

12/14)

62.66

\pm

12.85

69.91

\pm

6.55

7.25

\pm

11.12^*

63.21

\pm

9.99

68.14

\pm

7.83

4.92

\pm

7.83^*

0.217

0.830

x^{2}/

12.941

2.958

p

00.002^d

0.228

Post hoc

(Mann-

Whitney)

MBI1

>

MBI3

(

p<

0.01)

MBI2

>

MBI3

(

p<

0.01)

MBI $=$ Modified Barthel Index; MBI1 $=$ score 0–23 group; MBI2 $=$ 24–49 group; MBI3 $=$ 50–74 group; SG $=$ Study Group; CG $=$ Control Group. ^aMean $\pm$ standard deviation, ^bPaired $t$ -test, ^cIndependent $t$ -test, ^dKruskal-Wallis test, $p<$ 0.05^*, $p<$ 0.01^**, $p<$ 0.001^***.

Comparison by severity levels between the groups on MBI

There were significant changes in MBI1 ( $p<$ 0.001), MBI2 ( $p<$ 0.001), and MBI3 ( $p<$ 0.05) in study group before and after intervention. There were also significant changes in MBI1 ( $p<$ 0.01), MBI2 ( $p<$ 0.05) and MBI3 ( $p<$ 0.05) in control group before and after intervention (Table 5). After the intervention, the measurement of MBI in both the study and control group was statistically better than before the intervention.

In the comparison between the study and control group, there were significant differences in MBI1 ( $p<$ 0.001), and MBI2 ( $p<$ 0.01). In particular, the difference was clear in MBI1. These findings implied that MBI1 and MBI2 groups responded more effectively to RAWT than MBI3.

4. Discussion

Through recent technological advances, RAWT is being applied more actively than other methods. One of the reasons why this is possible is that RAWT can reduce the therapist’s workforce intervention, automatically control the amount and intensity of training, and support the patient’s weight to safely train walking [47].

Until now, the effectiveness of RAWT have been limited to whether stroke patients are able to walk independently after RAWT. Additionally, they only focused on improving walking ability indicators such as adjustable parameters by the robotic device, which only concerns microscopic factors, including walking speed, duration, distance, percentage of body weight support, and guidance. Moreover, most of the studies were simply focused on comparing the effectiveness of RAWT with CP [5, 17, 18, 19, 28, 48, 49, 50]. The existing studies lack a comprehensive evaluation of the effectiveness of RAWT compared to CP for rehabilitation. Therefore, the purpose of this study is to fill this gap.

Chen et al. [5] studied the effect of robot-assisted gait training on chronic stroke patients. In their study, BBS, Functional Ambulation Category (FAC), Motricity Index (MI), Barthel Index (BI), and Mini-Mental State Examination (MMSE) were used as evaluation tools. Their study found that robot-assisted gait orthosis combined with CP produced more functional outcomes than did CP alone. There have also been studies showing the effectiveness of RAWT using various measurement variables [5, 17, 28, 51]. Their study also found that robot-assisted gait orthosis combined with CP produced more functional outcomes than did CP alone. In this study, TIS, BBS, FMA, and MBI were used as evaluation tools. There have been many different studies results related RAWT, but as expected in this study showed that a combination of RAWT and CP produced better improvement than the CP alone.

For more detailed analysis of this study, each of four evaluation tools that showed between-group difference were classified into three groups according to the score distribution of the study participants. In other words, they were classified into three groups depending on the severity levels of the study participants. By comparing the study and control group, this study tried to find the optimal target of RAWT [52, 53]. In this study, the level of evaluation tools was classified as follows due to the low score distribution of study participants. TIS was classified into TIS1, TIS2, and TIS3, BBS was classified into BBS1, BBS2, and BBS3, FMA was classified into FMA1, FMA2, and FMA3, and MBI was classified into MBI1, MBI2, and MBI3.

In the comparison between the study and control group, there were significant differences in TIS1 ( $p<$ 0.05), TIS3 ( $p<$ 0.05), BBS1 ( $p<$ 0.001), FMA1 ( $p<$ 0.001), FMA2 ( $p<$ 0.001), FMA3 ( $p<$ 0.01), MBI1 ( $p<$ 0.001), and MBI2 ( $p<$ 0.01). Simply reinterpreting the results, the response and effect of RAWT on stroke patients were more pronounced when the patient’s functional levels were as follows; trunk control was poor or fair, balance was a high-risk fall, motor function was severe or moderate impairment, and ADL was a total or severe independent level.

In the comparison between the three groups divided by severity levels in the study group, there were significant differences between TIS2 and TIS1 (TIS2 $>$ TIS1) ( $p<$ 0.001), TIS3 and TIS1 (TIS3 $>$ TIS1) ( $p<$ 0.05), TIS2 and TIS3 (TIS2 $>$ TIS3) ( $p<$ 0.05), FMA1 and FMA2 (FMA1 $>$ FMA2) ( $p<$ 0.001), FMA1 and FMA3 (FMA1 $>$ FMA3) ( $p<$ 0.001), MBI1 and MBI3 (MBI1 $>$ MBI3) ( $p<$ 0.01), and MBI2 and MBI3 (MBI2 $>$ MBI3) ( $p<$ 0.01). The results of this study implied that RATW had a more considerable effect in the poor or fair trunk control group for trunk balance and in the high fall risk group for balance [49]. In motor function, the training showed its value in the severe and marked motor impairment group. The total or severe dependence group in ADL experienced more improvements.

Taken together, it can be thought that the lower the physical function the more effective it is for RAWT [50]. Of course, it may be unreasonable to think that simply an increase in scores provide its effectiveness because the difference in difficulty of the functional level of each score is very large. In other words, the higher the functional level, the higher the difficulty of the functional level per point. In order to perform RAWT, the patient needs to be raised first and keep on standing relatively for a long time. The lower the functional level of the patient, the more challenging this process will be. On the other hand, it may be a very important opportunity and a therapeutic process to experience standing up against gravity. For that reason, in order to clarify the results of this study, it is necessary to conduct research using other standing aids and then compare and analyze the results. However, even considering these concerns, this study is meaningful in that it presented and further expanded the research direction for the future RAWT.

5. Conclusion

Even though the explosive increase in interest regarding treatment using robot assisted walking system as a treatment method to improve the motor function of stroke patients, very few studies have divided groups according to the severity levels of patients and conducted studies on the effects of RAWT. Researchers in this study tried to find the optimal target suitable for RAWT through this study.

Of course, there are several limitations to this study. Firstly, since the sample size is relatively small, the results should be considered as preliminary findings that may not indicate all patients receiving RAWT. Therefore, further prospective studies are needed to verify the current results. Secondly, because RAWT and CP interventions are performed simultaneously during the period of this study, it is hard to conclude that the research results are simply the effect of RAWT. Thirdly, this study only examined the effectiveness of interventions immediately after training. The long-term effects of RAWT may vary in stroke survivors and further investigation is required. In addition, some difficulties in controlling the patient’s personal lives can affect the study.

In conclusion, the most meaningful aspect of this study was to measure various variables such as balance ability, motor function of extremities, and ADL using each measuring tool. Based on these results, the study compared the effect of combination RAWT and CP with CP alone. Another is that it tried to find the suitable target for RAWT by classifying the evaluation tools in detail and comparing and analyzing them. Thus, the study attempted to present and further expanded the research direction for the future RAWT by complementing the previous studies’ limitations. These results do not seem to speak of the superiority of robotic therapy, but support the view that robots are complementary rather than replacing therapists [54].

Ethics approval and consent to participate

All study protocols were approved and reviewed by the institutional review committee of Yong In University (2-1040966-AB-N-01-2210- HSR-277-2).

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Footnotes

Acknowledgments

Nones.

Conflict of interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Anderson

. Rehabilitation of patient with complete stroke. In Kottke

, eds. Krusen’s handbook of physical medicine and rehabilitation, 4th ed. Philadelphia, WB Sounder’s Company. 1990; 656-78.

Jorgensen

Nakayama

Raaschou

Olsen

. Recovery of walking function in stroke patients: the copenhagen stroke study. Archives of Physical Medicine and Rehabilitation. 1995; 76(2): 27-32.

Goldie

Matyas

Evans

. Deficit and change in gait velocity during rehabilitation after stroke. Archives of Physical Medicine and Rehabilitation. 1996; 77(3): 1074-82.

Bohannon

Andrew

Smith

. Rehabilitation goals of patient with hemiplegia. International Journal of Rehabilitation Research. 1988; 11(5): 181-3.

Chen

Kang

Peng

Hsu

Lin

Lai

. Adjustable parameters and the effectiveness of adjunct robot-assisted gait training in indivisuals with chronic stroke. International Journal of Environmental Research and Public Health. 2022; 19: 8186.

Dohring

Daly

. Automatic synchronization of functional electrical stimulation and robotic assisted treadmill training. IEEE Transaction on Neural Systems and Rehabilitation Engineering. 2008; 16(4): 310-3.

Conesa

Costa

Morales

. An observational report of intensive robotic and manual gait training in sub-acute stroke. Journal of NeuroEngineering and Rehabilitation. 2012; 9(13): 241-5.

Riener

Lunenburger

Jezernik

Anderschitz

Colombo

Dietz

. Patient-cooperative strategies for robot-aided treadmill training: first experimental results. IEEE Transaction on Neural Systems and Rehabilitation Engineering. 2005; 13(1): 380-94.

Sale

De Pandis

Vimercati

. Robot-assisted walking training for individuals with parkinson’s disease: a pilot randomized controlled trial. BMC Neurology. 2013; 13(1): 13-50.

10.

. The effect of the lokomat robotic-orthosis system on lower extremity rehabilitation in patients with stroke: A systematic review and meta-analysis. Frontiers in Neurology. 2023; 14: 1260652.

11.

Akıncı

Burak

Yaşar

Kılıç

. The effects of robot-assisted gait training and virtual reality on balance and gait in stroke survivors: A randomized controlled trial. Gait Posture. 2023; 103: 215-22.

12.

Baronchelli

Zucchella

Serrao

Intiso

Bartolo

. The effect of robotic assisted gait training with lokomat on balance control after stroke: Systematic review and meta-analysis. Frontiers in Neurology. 2021; 12: 661815.

13.

Manuli

Maggio

Stagnitti

Aliberti

Cannavò

Casella

Milardi

Bruschetta

Naro

Calabrò

. Is intensive gait training feasible and effective at old age? A retrospective case-control study on the use of lokomat free-d in patients with chronic stroke. Journal of Clinical Neuroscience. 2021; 92: 159-64.

14.

Hobbs

Artemiadis

. A review of robot-assisted lower-limb stroke therapy: Unexplored paths and future directions in gait rehabilitation. Frontiers in Neurorobotics. 2020; 14: 19.

15.

Mustafaoglu

Erhan

Yeldan

Gunduz

Tarakci

. Does robot-assisted gait training improve mobility, activities of daily living and quality of life in stroke? A single-blinded, randomized controlled trial. Acta Neurologica Belgica. 2020; 120(2): 335-44.

16.

Mehrholz

Thomas

Werner

. Electromechanical-assisted training for walking after stroke. Cochrane Database of Systematic Reviews. 2017; 5(5): CD006185. Printed Online.

17.

Mayr

Kofler

Quirbach

. Prospective, blinded, randomized crossover study of gait rehabilitation in stroke patients using the Lokomat gait orthosis. Neurorehabilitation and Neural Repair. 2007; 21(4): 307-14.

18.

Hornby

Campbell

Kahn

Demott

Moore

Roth

. Enhanced gait-related improvements after therapist-versus robotic-assisted locomotor training in subjects with chronic stroke; a randomized controlled study. Stroke. 2008; 39: 1786-92.

19.

Husemann

Muller

Krewer

Heller

Koenig

. Effects of locomotion training with assistance of a robot-driven gait orthosis in hemiparetic patients after stroke; a randomized controlled pilot study. Stroke. 2007; 38; 349-54.

20.

Schulz

Altman

Moher

. CONSORT Group. CONSORT 2010 Statement: Updated guidelines for reporting parallel group randomized trials. British Medical Journal. 2010; 8: 18.

21.

Kim

. Effects of robot-assisted therapy on lower limb in patients with subacute stroke. Journal of the Korea Academia-Industrial Cooperation Society. 2016; 17: 459-66.

22.

Sirivastava

Kao

Kim

. Assist-as-needed Robot-adided Gait Training Improves Walking Function in Individuals Following Stroke. IEEE Transaction on Neural Systems and Rehabilitation Engineering. 2015; 23(6): 956-63.

23.

Faul

Erdfelder

Lang

, et al. G*power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods. 2007; 39(2): 175-91.

24.

Faul

Erdfelder

Buchner

Lang

. Statistical power analyses using g*power 31: Tests for correlation and regression analyses. Behavior Research Methods. 2009; 41(4): 1149-60.

25.

Kim

Yang

. The effects of robot assisted gait training on kinematic factors of the stroke patients. Journal of The Korean Society of Integrative Medicine. 2022; 10(1): 91-9.

26.

Park

. The effects of robot-assisted gait training for the patient with post stroke: A meta-analysis. Physical Therapy Korea. 2015; 22(2): 30-40.

27.

Lennon

Baxter

Ashburn

. Physiotherapy based on the Bobath concept in stroke rehabilitation: a survey within the UK.. Disability and Rehabilitation. 2001; 23(2): 254-62.

28.

Dundar

Toktas

Solak

Ulasli

Eroglu

. A comparative study of conventional physiotherapy versus robotic training combined with physiotherapy in patients with stroke. Topics in Stroke Rehabilitation. 2014; 21(6): 453-61.

29.

Dunsky

Dickstein

Marcovitz

. Home-based motor imagery training for gait rehabilitation of people with chronic poststroke hemiparesis. Archives of Physical Medicine and Rehabilitation. 2008; 89(8): 1580-8.

30.

Verheyden

Nuyens

Nieuwboer

Asch

Ketelaer

Weerdt

. Reliability and validity of trunk assessment for people with multiple sclerosis. Physical Therapy. 2006; 86(1): 66-76.

31.

Verheyden

Kersten

. Investigating the internal validity of the Trunk Impairment Scale (TIS) using Rasch analysis: The TIS 20.. Disability and Rehabilitation. 2010; 6(2): 1-11.

32.

Alhwoaimel

Turk

Hughes

Ferrari

Burridge

Wee

Verheyden

Warner

. Instrumented trunk impairment scale (iTIS): a reliable measure of trunk impairment in the stroke population. Topics in Stroke Rehabilitation. 2021; 23(1): 456-63.

33.

Berg

Maki

Williams

Holliday

Wood-Dauphinee

. Clinical and laboratory measures of postural balance in an elderly population. Archives of Physical Medicine and Rehabilitation. 1992; 73(11): 1073-80.

34.

Berg

Wood-Dauphinee

Williams

Maki

. Measuring balance in the elderly; validation of an instrument. Canadian Journal of Health Technologies. 1992; 2: 7-11.

35.

Berg

Wood-Dauphinee

Williams

. The Balance Scale: Reliability assessment with elderly residents and patients with an acute stroke. Scandinavian Journal of Rehabilitation Medicine. 1995; 27(4): 27-36.

36.

Bogle

Newton

. Use of the berg balance test to predict falls in elderly persons. Physical Therapy. 1996; 76(6): 576-83.

37.

Lusardi

. Functional performance in community living older adults. Journal of Geriatric Physical Therapy. 2004; 26(3): 14-22.

38.

Beninato

Portney

Sullivan

. Using the international classification of functioning, disability and health as a framework to examine the association between falls and clinical assessment tools in people with stroke. Physical Therapy. 2009; 89(2): 816-25.

39.

Fugl-Meyer

Jääskö

Leyman

Olsson

Steglind

. The post-stroke hemiplegic patient. 1. a method for evaluation of physical performance. Scandinavian Journal of Rehabilitation Medicine. 1975; 17(1): 13-31.

40.

Gladstone

Danells

Black

. The Fugl-Meyer assessment of motor recovery after stroke: a critical review of its measurement properties. Neurorehabilitation and neural repair. 2002; 16(3): 232-40.

41.

Fugl-Meyer

. Post-stroke hemiplegia: assessment of physical properties. Scandinavian Journal of Rebabilitation Medicine Supplement. 1980; 7(2): 85-93.

42.

Sanford

Moreland

Swanson

Stratford

Gowland

. Reliability of the Fugl-Meyer assessment for testing motor performance in patients following stroke. Physical Therapy. 1993; 73(7): 447-54.

43.

Shah

Vanclay

Cooper

. Improving the sensitivity of the Barthel Index for stroke rehabilitation. Journal of Clinical Epidemiology. 1989; 42(8): 703-9.

44.

Quinn

Langhorne

Stott

. Barthel index for stroke trials: development, properties, and application. Stroke. 2011; 2: 1146-51.

45.

Granger

Albrecht

Hamilton

. Outcome of comprehensive medical rehabilitation: measurement by PULSES profile and the barthel index. Archives of Physical Medicine and Rehabilitation. 1979; 60(4): 145-54.

46.

Leung

Chan

Shah

. Development of a chinese version of the modified barthel index-validity and reliability. Clinical Rehabilitation. 2007; 21(10): 912-22.

47.

Poli

Morone

Rosati

Masiero

. Robotic technologies and rehabilitation: New tools for stroke patients’ therapy. BioMed Research International. 2013; 2013: 153872.

48.

Hidler

Nichols

Pelliccio

Brady

Campbell

Kahn

Hornby

. Multicenter randomized clinical trial evaluating the effectiveness of the Lokomat in subacute stroke. Neurorehabilitation and Neural Repair. 2009; 23(1): 5-13.

49.

Schwartz

Sajin

Fisher

Neeb

Shochina

Katz-Leurer

Meiner

. The effectivness of locomotor therapy using robotic-assisted gait training in subacute stroke patients. A randomized controlled trial. American Academy of physical Medicine and Rehabilitation. 2009; 1: 516-23.

50.

Westlake

Patten

. Pilot study of Lokomat versus manual-assisted treadmill training for locomotor recovery post-stroke. Journal of NeuroEngineering and Rehabilitation. 2009; 6: 18.

51.

Chang

Kim

Huh

. Effects of robot-assisted gait training on cardiopulmonary fitness in subacute stroke patients: a randomized controlled study. Neurorehabilitation of Neural Repair. 2012; 26(4): 318-24.

52.

Dennis

Janice

. Berg Balance Scale score at admission can predict walking suitable for community ambulation at discharge from inpatient stroke rehabilitation. Journal of Rehabilitation Medicine. 2018; 50: 34-7.

53.

Kim

Lee

Hwang

Pyo

Yang

Lim

Park

Kim

. Clinical characteristics of proper robot-assisted gait training group in non-ambulatory subacute stroke patients. Annals of Rehabilitation Medicine. 2016; 40(2): 183-9.

54.

Hesse

. Treadmill training with partial body weight support after stroke: a review. NeuroRehabilitation. 2008; 23(2): 55-65.

Effects of robot-assisted walking training on balance,motor function,and ADL depending on severity levels in stroke patients

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Methods

Table 1 General characteristics of the subjects ( n = 96)

2.2.1 Balance abilities

2.2.2 Motor function of extremities

2.2.3 ADL

2.3 Data analysis

3. Results

3.1 General characteristics of the subjects

3.2 Clinical outcomes

3.2.1 Comparison by severity levels on balance

Table 2 Comparison by severity levels within and between the group on TIS ( n = 96)

Table 4 Comparison by severity levels within and between the group on FMA ( n = 96)

Table 5 Comparison by severity levels within and between the group on MBI ( n = 96)

5. Conclusion

Ethics approval and consent to participate

Funding

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
General characteristics of the subjects ( $n=$ 96)

Table 2
Comparison by severity levels within and between the group on TIS ( $n=$ 96)

Table 4
Comparison by severity levels within and between the group on FMA ( $n=$ 96)

Table 5
Comparison by severity levels within and between the group on MBI ( $n=$ 96)