Abstract
Introduction
Stroke is a disease wherein a part of the brain is damaged due to the infarction of blood vessels supplying blood to the brain or due to bleeding (Pendleton & Schultz-Krohn, 2013). After stroke onset, functional damage persists in 40% of patients and severe disability remains in 15–30% of patients (Duncan et al., 2002). The neurological disorders that may occur in stroke patients include motor and sensory disorders, cognitive disorders, perception disorders, language disorders, and emotional disorders, and several patients suffer from permanent disabilities as they do not recover movement and function due to the loss of motor function (Trombly & Wu, 1999). Given that such a loss of motor function leads to difficulties in daily life for stroke patients, a rehabilitation intervention program should be developed for stroke patients that can be performed consistently during the chronic period (Feys et al., 1998).
As most gestures in daily life involve the upper limbs and hands, patients who cannot use their hands not only suffer from severe physical and psychological pain (Pendleton & Schultz-Krohn, 2013), but also encounter difficulties in the activities of daily living that primarily involve upper limb function (Feys et al., 1998). Because upper limbs require fine motor skills and large-scale movement skills, paralysis of the upper limbs is a very important cause of disorders that develop after stroke, and presents a large barrier for hemiplegic patients to overcome in order to lead a normal daily life and return to society (Shumway-Cook & Woollacott, 2007). Stroke patients have various problems such as asymmetrical posture, abnormal body balance, and decreased ability to move weight. The loss of motion element involved on fine functions, and the above-mentioned problems can lead to decreased muscle cooperativity of the lower limbs while walking and may also result in an asymmetric gait due to imbalance in the ability to perform exercise (Rodriguez & Aruin, 2002). Human gait is reportedly accompanied by the active movement of the upper limbs, and hence, movement of the upper limbs affects the gait of the patient (Stephenson, Lamontagne, & De Serres, 2009). The trunk and upper limbs have been found to move simultaneously (Hingtgen, McGuire, Wang, & Harris, 2006), and such harmony in movement is reported to play a major role in gait (Stephenson et al., 2009).
More than 85% of stroke patients develop hemiplegia immediately after stroke onset, and 55–75% of survivors present continuous disability such as exercise deficiency associated with decreased quality of life (Nichols-Larsen, Clark, Zeringue, Greenspan, & Blanton, 2005). Due to such persistent disorders, most stroke patients suffer from depression, problems in interpersonal relationships and social life, and a general lower quality of life (Forkan et al., 2006). Hence, the focus of rehabilitation in stroke patients should be to maximize the ability to perform exercise or functional activities and help them return and adapt to society by enhancing their independence with regard to the activities of daily living (Skilbeck, Wade, Hewer, & Wood, 1983). Accordingly, balance should be improved, and suitable functional movement should be achieved (Walker, Brouwer, & Culham, 2000). Therefore, treatment should be performed while emphasizing on the gait that is most related to patient quality of life (Kim, Warren, Madill, & Hadley, 1999). Recent studies on the treatment of stroke patients reported that approaches such as more intensive and repetitive training as compared to conventional general and passive intervention (French et al., 2009), training related to reality (Nudo, Wise, SiFuentes, & Milliken, 1996), intervention involving motivation and active participation (Maclean, Pound, Wolfe, & Rudd, 2000) and forced induction exercise (Page, Levine, & Leonard, 2005), visual exercise feedback (Quaney, He, Timberlake, Dodd, & Carr, 2010), auditory exercise feedback (Malcolm, Massie, & Thaut, 2009), purpose-oriented training (Trombly & Wu, 1999), and task-oriented training (Michaelsen, Dannenbaum, & Levin, 2006) are more effective in promoting function after stroke onset.
Studies on boxing exercise that included a jab, cross, and hook combination training program in patients with Parkinson disease indicated that agility, speed, and backward walking improved (King & Horak, 2009), whereas other studies in patients with Parkinson disease included a boxing exercise program involving stretching of the trunk and limbs and breathing as warm-up for the first 20 minutes, followed by regular circulation training, endurance training, and punching motion for 45–60 minutes, with a 3-minute exercise period and 1-minute rest interval. Punching gestures were performed by hitting a goal using sand bags or mitts in various combinations. Boxing training including traditional stretching, muscular strength training, and duration training for 90 minutes, 24–36 times over 12 weeks, was considered to be effective for improved functional stretching, dynamic balance, walking speed, and quality of life (Combs et al., 2011). These positive results are obtained by the involvement of the whole body in the boxing program, including the footwork of the lower limbs and punch gestures of the upper limbs in various directions; in particular, the punch gesture rotates the truck and combines fast arm action with expected posture control (King & Horak, 2009). In the previous studies, boxing programs have been conducted for patients with Parkinson disease (Combs et al., 2011), and several virtual reality studies have comprehensively applied various sports events other than boxing by using the Wii fit platform from Nintendo (Taylor et al., 2012). However, only a few studies have applied an actual boxing program in stroke patients. Hence, by using a boxing program involving hitting mitts and a sand bag and not virtual reality settings, we aimed to achieve a more realistic setting, generate greater interest, and increase participation and motivation In the present study, we aimed to determine the effects of an actual boxing program on the changes in upper limb function, balance, gait, and quality of life in chronic strokepatients.
Methods
Subjects
In the present study, we examined 30 patients diagnosed with hemiplegia due to stroke who were receiving physical therapy in E rehabilitation care hospital located in Seoul; all the patient understood the reason for the study and actively participated in it. The selection criteria for the subjects of this study included individuals with hemiplegia due to stroke, stroke onset within 6 months to minimize the possibility of natural recovery, a score of >21 points on the Mini Mental State Examination-Korea (MMSE-K), ability to independently walk 10 m, and the ability to understand the research purpose and agree to participate in the study.
The exclusion criteria for the subjects of this study were individuals who had participated in a similar experiment in the past 6 months, individuals with complaints of back and shoulder pain, and individuals who could not walk 10 m.
During a pre-test before the experiment, upper limb function, balance, gait, and quality of life were recorded, and 15 people each were allocated to the boxing program group and conventional physical therapy group through a random draw in order to minimize any bias.
Each exercise program was performed over 6 weeks, and the patients were repeatedly educated regarding the training method 1 week before the experiment so that the subjects could understand and participate in the boxing program. The boxing program group underwent the boxing program and conventional physical therapy for 30 minutes over 6 weeks (3 times/week), whereas the conventional physical therapy group inly underwent conventional physical therapy for 30 minutes over 6 weeks (3 times/week). Two people, including 1 case of voluntary dropout and 1 case of discharge, were excluded from the boxing program group. Moreover, 2 individuals were discharged in the conventional physical therapy group, and were excluded from the analysis. Hence, 13 individuals each were included in the boxing program group and conventional physical therapy group (Fig. 1). Moreover, subjects exhibiting a training participation rate of <80% over the study period were excluded from the analysis. The experiment was carried out with the approval of Sahmyook University Research Ethics Committee.
Measures
Upper limb function was evaluated by using the manual functional test (MFT). The MFT is used to assess the damage in motor function of the upper limbs. This evaluation tool can record minor changes in upper limb movement during the neurologic recovery period. The MFT consists of 4 items related to upper limb movement, 2 items related to identification, and 2 items related to finger operation. Thus, it comprises a total of 8 items, wherein 1 point is assigned for an item that can be performed and 0 points are recorded when an item cannot be performed. The sum total of points for all the items is 32 points (Kim & Mi-young, 1994). The reliability between testers administering the MFT is reported to be sufficiently high (affected side: r = 0.99, non-affected side: r = 0.84; p < 0.01) (Nakamura, 2000). Hand grip was used to measure hand grip strength. The Jamar hand dynamometer (Sammons Preston, Canada) was used for grip evaluation, and the average of 3 measurements with the shoulder joint and elbow joint in the neutral position was used.
Dynamic balance was measured by using the Berg Balance Scale (BBS). It is a functional balance test method that considers 3 aspects, such as posture maintenance, postural control, and response to external agitation, and is divided into a total of 14 items. The minimum score is 0 points, whereas the maximum score is 4 points, for each item, and the total score of 14 items is calculated. The items comprise sitting, standing up, standing, moving, stretching arm, picking up things, looking back, rotating, and crossing feet; the total score is 56 points, and higher scores represent good balance. The intra- and inter-examiner reliability is r = 0.99 and r = 0.98, which indicates high reliability and internal validity (Berg, Wood-Dauphinee, & Williams, 1995; Thorbahn & Newton, 1996).
The Good Balance system (Metitur Ltd, Finland, 2008) was used for measurement of static balance. This equipment is widely used for measuring balance in the elderly and stroke patients (Era et al., 2006; Sihvonen, Sipilä, & Era, 2004). In the test-retest method, the correlation coefficient (ICC = 0.83) was found to be >0.83, which indicates a high reliability. In this method, subjects were asked to spread their legs by about 4 inches, with their eyes open and gaze at a dot located about 3 cm and 15° ahead; this position had to be maintained for about 30 seconds, and the static balance ability was recorded. This test was performed 3 times, and the average value of static balance ability was measured.
To assess walking ability, the 10-m walk test (10 MWT) was used. This test is the most representative method to identify the degree of damage to walking ability, in cases of walking disability due to nerve damage (Pohl, Mehrholz, Ritschel, & Rückriem, 2002; Visintin, Barbeau, Korner-Bitensky, & Mayo, 1998). The time taken to cover 10 m, excluding 2 m from the starting point and arrival point, while walking a total of 14 m was measured. The 10 m walking time was recorded as the maximum walking speed while walking at full speed, and was measured 2 times; the higher value was used in the analysis, and the time taken was recorded at intervals of 0.1 second. The intra- and inter-examiner reliability were both high (r = 0.089–1.00) (Steffen, Hacker, & Mollinger, 2002).
To measure the quality of life of stroke patients, we used the Stroke-Specific Quality of Life questionnaire (SS-QOL) developed in this study (Williams, Weinberger, Harris, Clark, & Biller, 1999). This tool was specifically developed for stroke patients and included areas that are easily overlooked in the general quality of life assessment; it consists of 49 items over 12 areas (energy, family, language skills, movement, feeling, personality, self-help activities, social role, thinking skills, upper extremity function, vision, and work). The scores are recorded on a 5-point scale, and higher values indicated better quality of life. A study on the SS-QOL indicated high reliability (Cronbach’s α≥0.73).
Procedures
During a pre-test before the experiment, upper limb function, balance, gait, and quality of life were measured, and 15 people each were allocated to the boxing program group and conventional physical therapy group through a random draw to minimize bias. Each exercise program was performed over 6 weeks, and the patients were repeatedly educated regarding the training method 1 week before the experiment so that the subjects could understand and participate in the boxing program. The boxing program group underwent the boxing program and conventional physical therapy for 30 minutes over 6 weeks (3 times/week), whereas the conventional physical therapy group inly underwent conventional physical therapy for 30 minutes over 6 weeks (3 times/week). Two people, including 1 case of voluntary dropout and 1 case of discharge, were excluded from the boxing program group. Moreover, 2 individuals were discharged in the conventional physical therapy group, and were excluded from the analysis. Hence, 13 individuals each were included in the boxing program group and conventional physical therapy group.
Interventions
Boxing exercise program
The program started with a warm-up session involving breathing and stretching of the trunk and limbs for 5 minutes over 6 weeks (3 times/week). The program then included mitt hitting and sand bag hitting for 10 minutes, with a 2-minute rest period. Thereafter, stretching of the trunk and limbs was performed for 5 minutes, similar to the warm-up. Before the experiment, patients were trained to wear gloves in the sitting position, and hit mitts and a sand bag in various directions (up, down, left, and right) by applying gestures including jab, straight, one two, and a combination of these. The training was conducted in the sitting position in the first and second week. In the third and fourth week, the training was conducted after below the hips. Subsequently, during the fifth and sixth week, the subjects were asked to hit the target while sitting and standing (Table 1).
Conventional physical therapy
Conventional physical therapy involves resistance exercises and gait training performed by a physical therapist. During the 30-minute conventional physical therapy session, neurodevelopmental treatment (NDT) and proprioceptive neuromuscular facilitation (PNF) were performed for 15 minutes each.
Statistical analysis
All the variables are presented as mean and standard deviation, and the statistical analysis was conducted using SPSS ver. 19.0. To assess the general characteristics of the subjects, descriptive statistics was used. The paired t-test was performed to compare the variables before and after the intervention within each group. To compare the differences between the dependent variables according to the treatment method between the groups, independent sample t-tests were used. The statistical significance level of the data was set at 0.05.
Results
Changes in upper limb function
The upper limb function on the non-affected side significantly increased from 25.77 points before the experiment to 29.38 points after the experiment in the boxing program group (p < 0.05), and also significantly increase by 1.62 points from 24.08 points before the experiment to 25.69 after the experiment in the conventional physical therapy group (p < 0.05). Moreover, a significant difference was noted between the changes between the two groups (p < 0.05). In addition, the upper limb function on the affected side increased significantly from 13.23 points before the experiment to 16.85 points after the experiment in the boxing program group (p < 0.05), and significantly increased by 1 point from 13.15 points before the experiment to 14.15 points after the experiment in the conventional physical therapy group (p < 0.05). Furthermore, a significant difference was noted between the changes between the two groups (p < 0.05) (Table 2).
Grip
The grip on the non-affected hand significantly increased by 2.85 kg from 18.69 kg before the experiment to 21.54 kg after the experiment in the boxing program group (p < 0.05), and also significantly increased by 1.46 kg from 26.15 kg before the experiment to 27.62 kg after the experiment in the conventional physical therapy group (p < 0.05). Moreover, a significant difference was noted between the changes between the two groups (p < 0.05). The grip of the affected hand significantly increased by 1.08 kg from 4.31 kg before the experiment to 5.38 kg after the experiment in the boxing program group (p < 0.05), and also significantly increased by 0.85 kg from 4.15 kg before the experiment to 5 kg after the experiment in the conventional physical therapy group (p < 0.05). However, no significant difference was observed between the changes between the two groups (p < 0.05) (Table 2).
Dynamic balance
Dynamic balance significantly increased by 6.08 points from 30.15 points before the experiment to 36.23 points after the experiment in the boxing program group (p < 0.05), and also showed significantly increased by 1.9 points from 26.62 points before the experiment to 28.54 points after the experiment in the conventional physical therapy group (p < 0.05). Moreover, a significant difference was not between the changes between the two groups (p < 0.05) (Table 3).
Static balance
Eye open condition
The front and rear postural sway velocity, mediolateral postural sway velocity, and velocity moment in the eye open condition significantly increased in the boxing program group (p < 0.05). In the conventional physical therapy group, a significant increase (p < 0.05) was noted in mediolateral postural sway velocity, but not in the front and rear postural sway velocity and velocity moment. Between the groups, the front and rear posture sway velocity and mediolateral postural sway velocity did not significantly differ, although the velocity moment showed a significant difference (p < 0.05) (Table 4).
Eye close condition
In the eye close condition, the front and rear postural sway velocity, mediolateral postural sway velocity, and velocity moment significantly increased in the boxing program group (p < 0.05). However, not all these parameters showed a significant increase in the conventional physical therapy group. Moreover, no significant difference was observed in the comparison between the groups (Table 4).
Walking ability
Walking ability significantly increased by –5.78 seconds from 34.88 seconds before the experiment to 29.09 seconds after the experiment in the boxing program group (p < 0.05), and increased (not significantly) by –0.96 from 36 seconds before the experiment to 35.04 seconds after the experiment in the conventional physical therapy group. A significant difference was noted between the changes between the two groups (p < 0.05) (Table 5).
Quality of life
The quality of life significantly increased by 17.31 points from 146.46 points before the experiment to 163.77 points after the experiment in the boxing program group (p < 0.05), and significantly increased by 5.31 points from 136.92 points before the experiment to 142.23 points after the experiment in the conventional physical therapy group (p < 0.05). Moreover, a significant difference was noted between the changes between the two groups (p < 0.05) (Table 6).
Discussion
The deterioration in upper extremity exercise ability makes it difficult to perform essential daily life activities related to removal movements, eating, and personal hygiene, and becomes the major factor responsible for reducing functional independence. Upper limb function disorders are the most fundamental cause of disorders that clinically occur in stroke patients, and generates a significant barrier that needs to be overcome by hemiplegic patients to perform daily living activities and return to society (Page, Levine, Sisto, Bond, & Johnston, 2002). Hence, specific tasks related to activities of daily living are believed to be essential in exercise and training in stroke patients.
Ku et al. (2003) stated that upper limb function ability improved by applying tactile feedback with tactile gloves in virtual reality (Ku et al., 2003). However, the score on the non-affected and affected side improved by 1.62 points and 1 point, respectively, after measuring upper limb function in the conventional physical therapy group, which was significantly different (p < 0.05). However, the boxing program group exhibited an even higher score. To assess hand grip on the affected and non-affected side, Broeren et al. (2004) administered a program involving 3-D computer games for 90 minutes a day over 4 weeks (3 times/week) to improve upper limb function; they noted an increase in the average hand grip of the affected hand by 124N (12.649 kg), from 52N (5.303 kg) to 176N (17.952 kg) (Broeren, Rydmark, & Sunnerhagen, 2004).
In the present study, upper extremity ability in the conventional physical therapy group also improved, which was believed to be due to the NDT and PNF provided for 30 minutes to all the patients; however, this improvement was lower than that noted in the boxing program group. The findings of previous studies employing virtual reality are this consistent with those in the present study.
The sitting posture is basic position for safely performing activities of daily living or various movements, and is an important indicator for exercise and functional recovery after stroke onset (Morishita et al., 2009; Verheyden et al., 2004). Balance can be divided into static balance and dynamic balance. Static balance refers to the ability to stand without shaking on a fixed supporting surface, whereas dynamic balance refers to the balance maintained while the supporting surface is moving, an external stimulus is given, or the person is moving by himself (Horak, Henry, & Shumway-Cook, 1997). Evaluating static and dynamic balance in the sitting position is important to determine precise information regarding the damage in stroke patients (Lanzetta, Cattaneo, Pellegatta, & Cardini, 2004). Static balance measuring equipment can more accurately evaluate balance, as the postural sway and weight support rate can be calculated through the average velocity in the X-axis and Y-axis of the center of gravity (John, Cherian, & Babu, 2010; Mononen, Viitasalo, Konttinen, & Era, 2003; Salminen et al., 2009). however, it is considered to be difficult to indicate the increasing effect of static balance in functional movement tasks as the main muscles of stroke patients are affected by spasticity.
Kim et al. (2010) used a program involving tennis and boxing games on the Nintendo Wii for 30 minutes/day for 5 weeks (5 times/week), and noted that the dynamic balance increased by 9 points, from 39 points to 48 points (Kim, Kang, & Lee, 2010). Dynamic boxing exercise in the present study is considered to help enhance the static balance ability, and can significantly affect the dynamic balance ability; however, it may not have a greater effect as compared to that produced by conventional physical therapy.
The improvement of walking ability in the process of recovery of function in stroke patients is the main goal of physical therapy because walking is an important requirement in achieving functional independence (Perry, Garrett, Gronley, & Mulroy, 1995), and walking is an essential element for functional activities and daily activities (Broeren et al., 2004). Mirelman et al. (2009) trained 18 chronic stroke patients for 1 hour/day over 4 weeks (3 times/week) with a force feedback system on the virtual reality system, and noted an significant increase in the average walking speed by 17.1%, from 0.53 m/s to 0.63 m/s (p < 0.05).
Walking in stroke patients shows an inefficient gait pattern, because of compensation motion in each part of the body due to the reduction in balance; therefore, stroke patients consume more energy than normal people (Granat, Maxwell, Ferguson, Lees, & Barbenet, 1996) and balance and gait are hence associated with each other. In particular, balance in the sitting position in stroke patients reportedly has a greater effect on walking than muscular strength perceptual function of the paralyzed lower limb (Reisman et al., 2013). The present study indicated that the boxing program improves the walking ability by improving the balance ability. Stroke, a chronic disease, decreases the quality of life of patients of at least 80% of patients in 4 years after onset (Niemi, Laaksonen, Kotila, & Waltimo, 1988), and also has a significant effect on the quality of life of the families of the patients (Nichols, 1997).
Salbach et al. (2006) indicated that improvement of both balance and walking ability (physical items), as well as mental items may be important indicators for the quality of life in stroke patients (Salbach et al., 2006). Disorders due to stroke make it difficult for patients to lead an independent life (Son & Park, 2005). Hence, exercise learning programs should be provided to stroke patients by sufficiently motivating them and inducing active participation (Maclean et al., 2000).
Accordingly, the quality of life can be increased by improving the upper limb function and balance, and such improvements in walking ability could further improve the ability to perform activity of daily living.
Conclusions
In the present study, we noted that the boxing program group showed increases in upper limb function, balance, walking ability, and quality of life. Hence, boxing exercise can be considered to be a helpful for use as an active exercise with general exercise to promote functional recovery. However, further studies should be performed to develop various programs to enhance the motivation and interest of patients, while focusing on the objectives of rehabilitation and promoting functional recovery.
Conflict of interest
Financial disclosure statements have been obtained, and no conflicts of interest have been reported by the authors or by any individuals in control of the content of this article.