The effects of task-oriented mental practice on upper limb function and coordination in chronic stroke patients—Randomized controlled trial design

Introduction

Stroke is a central nervous system damage disease that causes various motor and sensory disorders, and many other problems in the ability to perform daily activities, ultimately undermining the functional independence of the subject (Brewer et al., 2013). After a stroke, the lower limb function tends to recover relatively quickly, but the recovery of the upper limb is very difficult and proceeds slowly, so most patients have a large limitation in their daily life due to difficult functional movements such as stretch and pinch (Wade et al., 1983). The impairment of upper limb function not only causes problems in performing fine motor skills closely related to daily life such as eating, personal hygiene, and writing but also in gross motor skills such as walking, balancing, and protective reflex response (Gowland et al., 1993). Therefore, the restoration of upper limb function in the stroke patients is considered essential in improving not only their daily living ability but also their work ability and overall body ability.

In recent years, a lot of effective therapeutic approaches for restoring upper limb function and daily living ability of stroke patients have been studied. For example, modified forced induction therapy, task-oriented training, motion observation training, biofeedback, mirror therapy, mental practice, etc. have been suggested as interventions to improve the upper limb function in stroke patients (Mangold et al., 2009; Platz et al., 2005). Among them, the task-oriented training is a concept developed in motion science and motor learning. It is defined as a training or treatment in which the patient performs specific motor tasks in a given environment and receives feedback. Instead of using a rewarding strategy to complete the task, this rehabilitation approach focuses on the correction of motor deficits and the recovery of motor skills (Schmidt et al., 2018; Teasell et al., 2008).

Task-oriented training is a method in which the difficulty level of task is gradually raised according to the progress of physical training which is repeatedly executed with the functional tasks, and the positive and timely feedback is provided to the patient according to the patient’s training situation (Kleim and Jones, 2008; Takeuchi and Izumi, 2013). In the previous studies for animals, the task-oriented approach turned out to restore function of an undamaged brain region adjacent to the brain injury site (Nudo et al., 1996), and to mobilize the complementary areas of the system (Nudo et al., 2000). This task-oriented training attempts the patient-centered intervention rather than the conventional therapist-centered approach, considering tasks that are meaningful to patients as a core element of intervention. As it selects closely related to the daily life of the subjects when selecting the actual tasks, it has advantages in terms of clinical economics because there is no need to additionally purchase the expensive equipment in the clinic, and it is a form of physical training for patients with neurological injury that has been applied in many recent studies (Rensink et al., 2009).

Previous studies related to motor learning have emphasized not only the physical training of patients with nervous system damage but also their cognitive training, which is to learn the movement of the body by imagining the movement of the body in the head, for the smooth functional recovery. Many studies have reported that the most effective functional recovery effect can be obtained when these two types of training are combined (Guillot et al., 2008; Jackson et al., 2001). A representative form of cognitive intervention for patients with such nervous system injury is mental practice. The mental practice is a rehabilitation training to which the priming technique, which is a form of inner learning that induces behavioral change through the preceding stimulation, is applied (Stoykov and Madhavan, 2015). It is a kind of cognitive training that acquires motor skills and induces improvement of motor performance by imagining body movements without direct movement and practicing the tasks repeatedly in mind (Braun et al., 2006). In general, mental practice is applied to neurologically damaged patients along with physical practice to maximize its therapeutic effect (Barclay et al., 2020; Saruco et al., 2019). According to the research studies, when applying mental practice interventions, it is essential to select the movements of imagination and physical practice which are so closely related to the actual situation of patients that they can increase their motivation (Maclean et al., 2002). This mental practice is advantageous in that it can be applied easily and safely in clinical practice, and can obtain a high level of therapeutic effect at a low cost because it does not require a separate high-cost device (Maring, 1990), but it also has a disadvantage that high concentration is required by the patients for the intervention and they can be easily bored, leading to insufficient active participation of patients or lower level of motivation (Mulder et al., 2004).

In the respect of the previous research results that the combined application of physical and cognitive trainings is most suitable for functional recovery of stroke patients (Guillot et al., 2008; Jackson et al., 2001), it seems more positive in the intervention effect to concurrently use the imagination training, a form of cognitive practice, along with the existing task-oriented training, a form of physical practice, than to just use the conventional task-oriented training. In addition, since both interventions do not require additional purchase of special expensive equipment in clinical practice, the parallel application of them is judged to have advantages in terms of economic feasibility in clinical practice.

Therefore, this study is carried out to investigate the effect of task-oriented mental training, which combines the task-oriented practice with the characteristics of physical training that has advantages in terms of patient’s motivation and the imagination practice with the characteristics of cognitive training, on the upper limb function and the coordination of chronic stroke patients.

Methods

Experimental procedures and Interventions

In this study, the training program applied to the experimental group and the control group was conducted 5 times a week for a total of 3 weeks. The task-oriented mental practice (TOMP) was applied to the experimental group of this study. First, mental practice for the task was conducted for 5 min, and then practical task training for the task was conducted with the therapist for 10 min. In the control group in this study, the task-oriented training using the same tasks as those applied to the experimental group was applied for 15 min each session. During this study, four occupational therapists with more than 3 years of therapy experience participated in the intervention process for the training of the subjects. In addition, two occupational therapists with 5 years of therapy experience participated in the study evaluation. Prior to the intervention, the therapists who conducted the intervention were fully instructed on the intervention method and, only after cross-checking what they had been instructed, they could participate in the intervention. In addition, both the experimental and the control groups were pre-evaluated before the intervention, as well as the post-intervention evaluation after the intervention. To secure the objectivity of the study, the therapists and evaluators who led the training were blinded so that they could not distinguish the experimental group and the control group.

Measurements

BTS FreeEMG 300(BTS, Inc., Milan, Italy)

To measure the muscle activity, a surface wireless EMG measurement system, BTS FreeEMG 300 (BTS, Inc., Milan, Italy) was used. In measuring the muscle EMG related to the movement of the upper limb, the signal-noise ratio (S/N) was 96 dB and the Common Mode Rejection Ratio was 123 dB. Using a sampling frequency of 1 KHz and four out of eight channels, the measurement was carried out on the anterior deltoid, posterior deltoid, biceps brachii, and triceps brachii of the upper limb which were suggested as appropriate muscles to measure the upper limb function of stroke patients by Molteni et al., (2008).

For the muscle activity measurement, the subject was seated comfortably on a chair with a table in front of the subject. While the subject stretches his/her hand to grab a cup placed on the table at the distance of his/her arm and place it at the end of the table close to the subject, the measurement was carried out. It was set and measured by dividing the entire process into two steps: step 1 is the process from the original position to holding the cup and step 2 is the process from picking up the cup to returning to the original position to place the cup in front of his/her body.

A total of three measurements were made before and after the experiment, and the average value was used. The measured muscle activity signal was processed using the root mean square (RMS) method, and the coordination power of their movements was calculated to analyze the co-coordination ratio (CCR) when the subjects performed their movements. The CCR is calculated as the RMS of the antagonist muscle to the measured RMS of the agonist muscle in the motions made during the specified activity. And the reduced CCR after intervention means more improved coordination power and smoother movement (Busse et al., 2005). The calculation formula for the CCR value is as follows (Figure 1):

OPEN IN VIEWER

Figure 1.

Co-cordination ratio.

Data analysis

For statistical analysis of this study, SPSS 18.0 and Excel program were used. As a result of the Kolmogorove–Smirnove test for the normality verification of each measurement item, it was found that all items were normally distributed. For the homogeneity check of the target groups, the ײ test and the independent sample t-test were used and there was no significant difference between the two groups. Based on the measured FMA value, the paired samples t-test was carried out to compare the changes in upper limb function before and after the intervention within the group, and the independent sample t-test to compare the differences between groups. Changes in CCR for each group before and after the intervention through muscle activity were calculated using the Excel program.

Result

Figure 2 shows a flowchart of this study and Table 1 shows the general and medical characteristics of the subjects of this study. There were no significant differences between the experimental group and the control group in all items of gender, age, height, weight, stroke type, onset period, and cognitive level (Figure 2) (Table 1).

OPEN IN VIEWER

Figure 2.

Flow diagram of the study process.

OPEN IN VIEWER

Table 1.

General and medical characteristics of subjects.

Variable	EG (n = 17)	CG (n = 17)	p
Sex (male/female)	8/9	8/9	1.00
Age (years)	61.42 ± 8.12a	62.12 ± 10.20	0.37
Height (cm)	162.49 ± 7.23	161.32 ± 8.54	0.71
Weight (kg)	65.85 ± 7.23	67.02 ± 7.29	0.74
Stroke type (ischemic/hemorrhage)	12/5	11/6	0.66
Affected side (Lt/Rt)	2/15	1/16	0.98
Month from stroke onset	69.45 ± 21.51	72.22 ± 20.37	0.18
MMSE-K	26.10 ± 2.18	26.98 ± 2.53	0.27

EG: experimental group; CG: control group; MMSE-K: Korean version of Mini-Mental State Examination.

mean±SD.

In the FMA of motor function evaluated before the intervention, there was no significant difference between the experimental and control groups in all items of shoulder, wrist, hand, upper limb, and total scores (p > 0.05). When upper limb functions were compared before and after the intervention in each experimental and control group, significant differences were found in all items of experimental group (p < 0.05). In the control group, there was no significant difference in the total score of wrist, hand, and upper limbs (p > 0.05), but there was a significant change in the shoulder item (p < 0.05). The difference in upper limb function between the two groups after the intervention was not significant in all measured variables except for the total score (p > 0.05) (Table 2).

OPEN IN VIEWER

Table 2.

Change in FMA score.

		EG (n = 17)	CG (n = 17)	t
FMA
Shoulder	pre	24.40 ± 7.48	23.75 ± 4.49	−0.34
	post	31.25 ± 6.97	28.60 ± 5.14	−1.43
	t	−2.84*	−1.56*
Wrist	Pre	4.70 ± 2.62	5.15 ± 2.05	−0.72
	Post	9.75 ± 2.26	6.20 ± 3.20	−1.91
	T	−2.41*	−0.89
Hand	Pre	10.40 ± 3.13	10.73 ± 2.58	−0.63
	Post	14.10 ± 2.28	11.41 ± 2.57	−1.63
	T	−2.84*	−0.85
Total	Pre	41.71 ± 12.39	41.02 ± 6.18	−1.19
	Post	48.70 ± 11.03	43.00 ± 6.56	−2.69*
	T	−3.93*	−1.21

When CCR values in step 1 were compared, the experimental group showed the CCR change of 14.3 in the shoulder bend and −0.1 in the elbow extension, which means that it was negative for the shoulder but positive for the elbow. In the control group, there was only a negative change in shoulder flexion of 2.2 and elbow extension of 19.7 (Table 3).

OPEN IN VIEWER

Table 3.

Changes in CCR values during step 1.

CCR	Group	pre	post	Percent change
Shoulder flexion [DP/DA]	EG	0.42±0.15	0.48±0.40	14.3
	CG	0.46±0.27	0.47±0.23	2.2
Elbow extension [BI/TRI]	EG	1.92±1.10	1.90±0.66	−0.1
	CG	1.27±2.07	1.52±0.85	19.7

When CCR values in step 2 were compared, the experimental group showed the CCR change of −12.3 in the shoulder extension and −18.2 in the elbow bend, which means that it was positive in both the shoulder and elbow of the experimental group. It is judged that appears. In contrast, the control group showed the CCR change of 7.7 in the shoulder extension and 9.8 in the elbow bend after the intervention, indicating negative changes (Table 4).

OPEN IN VIEWER

Table 4.

Changes in CCR values during step 2.

CCR	Group	pre	Post	Percent change
Shoulder extension [DA/DP]	EG	1.88 ± 0.89	1.65 ± 1.19	−12.3
	CG	1.42 ± 0.37	1.53 ± 1.04	7.7
Elbow flexion [TRI/BI]	EG	0.44 ± 0.22	0.36 ± 0.19	−18.2
	CG	0.52 ± 0.43	0.57 ± 0.91	9.8

Discussion

In this study, mental practice was combined with the conventional task-oriented practice and applied to chronic stroke patients to investigate the effect of TOMP on their upper limb function and coordination.

According to Dunsky et al. (2006) who recommended the appropriate application time of mental practice to be within 10 min for patients with nervous system damage, the application time of mental practice was set as 5 min. And a relaxation time of 1–2 min was given to the subjects before starting their imagination according to Lamirand and Rainey that the subjects need a relaxation process for a certain amount of time before they can fully imagine (Lamirand and Rainey, 1994). In general, the mental practice can be divided into the visual imagery that the subject imagines his/her body movements from a third-person perspective, and the motor imagery that one becomes the subject of movement and imagines the movement from an inner point of view (Dunsky et al., 2008). In this study, based on the research results of Rangantan et al. that first-person motion imagery is more effective in motor learning of neurological injured patients than third-person visual imagery, the subjects’ mental practice was guided so that it could be performed from a first-person perspective (Ranganathan et al., 2002).

Comprehensively speaking about the results of this study, the control group, to which just the conventional task-oriented training was applied, showed a significant functional improvement only at the shoulder item in the FMA measured after the intervention, but the experimental group showed significant functional improvements at all items of shoulder, wrist, hand, and total scores in the FMA. However, in the comparison of FMA after intervention between the two groups, there were no statistically significant differences in all items except the total score. In this study, to check the subjects’ smooth use of the upper limbs, the CCR for each group was calculated with RMS (root mean square) values measured by the BTS FreeEMG 300 (BTS, Inc., Milan, Italy). The measurement was carried out by the process that the object takes a cup on the table at the distance of the arm by stretching his/her arm and brings it to the end of the table in front of his/her body, dividing it into two steps: step 1 is the process “from the start position to holding the cup” and step 2 is the process “from picking up the cup to placing it in front of his/her body.” In the case of the experimental group for the TOMP, the CCR of the shoulder bend at step 1 got higher after the intervention than before the intervention, showing a negative change in the coordination movement of the upper limb. However, the CCR of the elbow extension had a negative value, showing a slightly positive change after the intervention. At step 2, the CCR of poststretch restored posture had negative values after the intervention in both the subjects’ shoulder extension and elbow bending, indicating that there was a positive change. Meanwhile, in the control group for the TOMP, the CCRS in all movement items were positive, showing negative changes.

These results show that, in the antero-posterior comparison of each individual group through the FMA score, the experimental group achieved more statistically significant functional improvements than the control group, and that the experimental group also showed more positive CCR results than the control group in terms of the task performance and the joint movement. In these respects, the TOMP applied to the experimental group was more effective in improving the upper limb function of the subjects than that applied to the control group. It is consistent with the precedent research results reporting that the combination of physical practice and mental training may be more effective in the motor learning of neurological injured patients than simple physical practice (Guillot et al., 2008). In addition, the tasks of this study selected by combining the task-oriented training method were so related to subjects’ daily life that they are consistent with the concept suggested in previous studies that the tasks for the subjects to imagine and execute should be directly related to daily life to motivate them (Maclean et al., 2002). These could have made the TOMP applied to the experimental group more effective than that applied to the control group.

However, in the comparison between the groups after FMA intervention, no significant difference was found between them in all items except for the total score. In the CCR value, the bending movement of the shoulder had a positive value during the step 1 movement of the experimental group, showing that there was a negative change in the coordination of the upper limb function after the intervention. According to a systematic review study on mental practice published in 2013, in the 11 studies that obtained statistically significant improvement results among a total of 16 studies that met the selection criteria for the study, the intervention period was at least 3 weeks, and the average intervention time was 790.00 ± 386.23 min (Braun et al., 2013). When compared with the intervention time of 225 min for 3 weeks in this study, the time above can be said to have very high intervention intensity. Therefore, the intervention of the TOMP applied in this study might have been too insufficient intensity to represent a statistically significant difference in upper extremity function using the FMA from the control group after the intervention or, in terms of CCR values, it was not enough to create positive changes of coordination in all analyzed movements of the experimental group during the entire process of steps 1 and 2.

From the above results, this study can be judged to have confirmed the possibility that the TOMP applied to the subjects in this study can help improve the upper limb function and the coordination in chronic stroke patients. In addition, through the results of this study, it was confirmed once again that the combination of physical and mental training can be more effective in improving the upper limb function of patients with nervous system damage than the single application of the existing physical practice. In terms of task selection for mental practice, the introduction and application of the concept of task-oriented training can be more effective in improving the concentration and motivation of the subjects who actually perform the intervention.

The limitations of this study are the difficulty in generalizing the results of this study to all stroke patients due to the small number of samples and the insufficient intervention period and time to confirm a significant difference from the control group in terms of the overall intensity of intervention. In addition, there was no follow-up investigation in this study because of which it was not possible to confirm whether the effect in the experimental group shown in this study persists or not. And the facts that the actual level of upper limb function in the daily living environment was not investigated and that the onset period of the subjects was not variously adjusted but limited to 6 months or longer were also regretful in this study. In future studies, these aspects shall be supplemented with more detailed studies on TOMP.

Conclusion

In order to investigate the effect of TOMP on the upper limb function and coordination for chronic stroke patients, this study conducted the evaluation with FMA and BTS FreeEMG 300 (BTS, Inc., Milan, Italy) for 17 subjects in the experimental group to which the TOMP was applied and 17 in the control group to which just the task-oriented training was applied, analyzing their upper limb function changes. As a result of comparison before and after using FMA, the experimental group showed significant changes in all items, but the control group showed a significant change only in the shoulder item. In addition, after the intervention, the two groups also showed significant differences in the total score.

In the comparison of CCR values, the group using BTS FreeEMG 300 showed positive changes in three of the four measured upper limb coordination items, but the control group showed no positive changes in all items. Through this, it was confirmed the possibility that TOMP can have a positive effect on upper limb function and coordination of chronic stroke patients, and that the task-oriented practice combined with the mental training could be more effective than the conventional task-oriented training mainly composed of physical activities in improving upper limb function and coordination in chronic stroke patients.