Effects of water-based and land-based exercises on walking and balance functions of patients with hemiplegia

Abstract

BACKGROUND:

After the stroke, a number of changes occur in the neuromuscular system functions.

OBJECTIVE:

To determine whether the water based exercise (WBE) program applied in combination with the land-based exercises (LBE) compared to LBE alone contributes to the stroke patients’ motor functions, walking, balance functions and quality of life (QoL).

METHODS:

In total, 60 patients participated in this study. Patients were randomly divided into two groups. WBE therapy (3/week) + LBE (2/week) combination was applied to the patients in the study group (n = 30) for six weeks. LBE was applied to the control group (n = 30) 5/week for six weeks. Patients were evaluated before and after the treatment. Functional independence measurement, Berg balance scale, timed up and go test, and short form (SF) -36 assessment questionnaire were performed.

RESULTS:

Posttreatment results showed significant improvements in all of the parameters (except SF - 36 pain parameter) in both groups. The improvement in the vitality parameter of SF-36 was higher in the study group (p < 0.05), and improvement in the BBS was significantly higher in the LBE group than the WBE group (p < 0.05).

CONCLUSION:

Applying WBE together with the LBE (except SF-36 vitality sub-parameter) in patients with hemiplegia did not make any additional contribution to the application of LBE alone.

Keywords

Stroke aquatic rehabilitation balance quality of life

1 Introduction

Stroke is the most common cause of morbidity in the world and is the second culprit of the mortality following the ischemic heart disease (Donnan et al., 2008). In recent years, the decrease in the stroke mortality and increase in the life expectancy have been recorded as a result of the improvements in treatment methods applied in acute period after stroke. However, the number of stroke patients who need rehabilitation has also been rapidly increasing (Coleman et al., 2017).

Difficulty in walking and balance restricting the ambulatory activity due to the limitation of motor function and mobility is the most encountered problems among the clinical presentations after the stroke (Langhorne et al., 2011). Although the methods that involve conventional land-based exercises (LBE) programs are frequently used as a basis for walking and balance difficulty, aquatic therapy, which is applied as an alternative, is another treatment option (Mehrholz et al., 2011).

Water, which is a suitable training environment for exercise, facilitates the functional movements, provides somatosensory inputs from hydrostatic pressure and temperature, enhances cortical processing of sensory and motor information, and thus, improves functional activity. In the current studies, which aimed to improve the neuromuscular function of stroke patients using these physical properties of water, aquatic therapy has been shown to have many positive effects, such as balance control, reaching sufficient speed in independent walking ability and increasing functionalities in daily activities (Zhang et al., 2016; Park et al., 2011; Montagna et al., 2014).

When we reviewed the relevant literature, there are insufficient numbers of randomized controlled trials comparing the efficiency of water-based exercise (WBE) and LBE in stroke patients, which remained to be investigated. In this study, to contribute to the literature, we aimed to determine whether the aquatic exercise program applied in combination with the LBE contributes to the motor functions, walking, balance functions and quality of life (QoL) of the stroke patients compared to LBE alone.

2 Methods

2.1 Partipicants and study design

This study was a prospective, randomized, controlled study. 60 patients with stroke diagnosis between the ages of 50 and 85 who admitted to our Physical Medicine and Rehabilitation Clinic between June 2016 and January 2017 participated in our study. Patients were randomly divided into two groups by envelope method: n = 30 WBE + LBE group and n = 30 LBE group.

Inclusion criteria of the patients were determined as having stroke at least six months ago before admitted to the hospital, being diagnosed with hemiplegia by a neurologist, being able to comply with the two-stage verbal commands, being medically stable (i.e., no history of myocardial infarction, no musculoskeletal problems), and having a mini mental test (MMT) score >24 points.

Exclusion criteria were determined as having cognitive impairment which causes lack of communication, having an aphasia, having a severe heart disease (e.g., aortic stenosis, angina, hypertrophic cardiomyopathy, arrhythmia, cardiac pacemaker), having uncontrolled diseases, such as hypertension and epilepsy, having contraindications to receive WBE therapy, and having other accompanying neurological disorders, such as Parkinson’s disease.

All participants provided written informed consent form and written permission from their physician allowing their participation and the university ethics committee had approved the study protocol.

2.2 Interventions

2.2.1 Exercise program

Exercises were performed under the supervision of a physiotherapist during the treatment program in both WBE group and LBE group. The 40-minute training program (i.e., range of motion [ROM] exercises, strengthening exercises, trunk mobility exercises, balance exercises and walking training) was applied for the LBE group (n: 30) for a total of six weeks, including an hour per day and five sessions per week starting with 10 minutes light warm-up exercises, including stretching exercises for all major joint groups on the hemiplegic side. The exercise program was ended with 10 minutes light cool-down exercises, including stretching exercises for all major joint groups.

In the LBE group, strengthening exercises were applied as isometric and isotonic exercises (with 500 g weight/flexible exercise bands) in three sets of 15 reps. Trunk mobility exercises consisted of trunk flexion-extension, trunk lateral flexion, bridging exercise in the crook-lying position, crawling exercise, upright kneeling position and half upright kneeling position. Balance exercises were performed as follows: sitting on an exercise ball and then standing up, sitting on the exercise ball and then leaning face-down, and changing the upper extremity position by standing on the balance board. All of the activities were performed with eyes open first and then, eyes closed when control was developed. Walking exercises consisted of ceremony walking on flat ground, heel-to-toe walking on a straight line, side walking, hurdle walkovers, climbing up and down stairs of 10–15 cm.

In the WBE group, the program consisted of 18 sessions, 3×per week for six weeks in a swimming pool at 33°C. The program was applied in three stages: 10 minutes warm-up, 40 minutes aquatic exercise program and 10 minutes cool-down exercises. During the exercise, the intensity was gradually increased so as not to force the body. Strengthening exercises and balance coordination exercises were applied within the 40-minute program. Strengthening exercises included three sets of 15 reps of hip flexion and extension by holding a place near the pool, hip abduction and adduction, knee flexion and extension, and cycling. Balance coordination exercises applied in this 40-minute program included moving the opposite arm and leg while being balanced on one leg, weight transfer from one side to another in the standing position, longitudinal/transversal/sagittal rotational control movements, walking in the pool, climbing up and down stairs, squatting in the pool while holding a place near the pool. The LBE program described in the above paragraph was applied to patients in WBE group for six weeks with two sessions per week and an hour per day.

None of the participants experienced any significant adverse treatment-related events, and all of the patients in the intervention groups were fully compliant.

2.3 Outcomes

The treatment responses were evaluated at the beginning (T0) and at the sixth week (T2) after the initial examination of the patients. The demographic (age, gender, marital status, education level) and clinical characteristics (affected side, dominant side, stroke type, duration of the stroke, lower, upper extremity and hand stages of the Brunnstrom) were recorded at the beginning of the treatment.

Berg balance scale (BBS), Functional independence measurement (FIM), short form (SF)-36 QoL assessment questionnaire, Sportkat Balance Device Balance Index measurements and maximum peak torque levels of the isokinetic quadriceps and hamstring muscles were used for assessment. Time up and go (TUG) test was performed to determine the walking speed of the patients.

The BBS is used to evaluate balance abilities and fall risk. This scale consisted of 14 items. Level of competence of the activity for each item was indicated with five points (0–4): 0 means “Not able to do it” 4 means “able to do it independently and safely”. Maximum score of the test was 56, and higher scores showed better balance (Alonso et al., 2014).

The FIM focuses on six functional areas: self-care, sphincter control, mobility, locomotion, communication and social cognition. Each item consisted of 18 items evaluated using a 7-point scale, and the total score was 126 (Kidd et al., 1995).

SF-36; There are two main titles that assessed the physical and mental health dimensions of the survey population. The increase in score of each health dimension of the SF-36 questionnaire with positive scores indicated that the health-related quality of life increased (Wang et al., 2008).

2.3.1 Balance index measurement with the Sportkat balance device

Two Balance Indices, static and dynamic, were obtained with the Sportkat 4000 Balance Device. The patient standing during static measurement tried to keep the cursor at center by moving the weight forward, backward, right or left by following a cross cursor that represents the center of the platform on the monitor screen. During the dynamic measurement, the patient tried to follow the moving object by drawing a circle and moving his/her weight towards the platform. Scores were recorded (Hansen et al., 2000).

2.3.2 Maximal peak torque measurement of isokinetic quadriceps and hamstring

An isokinetic device (IsoMed 2000, isokinetic dynamometer, Germany) was used to evaluate the isokinetic strength of the knee flexion and extension of all patients. Evaluation was made at two different speeds, 10 reps at 90°/sec and 10 reps at 120°/sec as concentric- concentric contraction to measure the maximal isokinetic forces of the knee flexor (hamstring) and knee extensor (quadriceps) muscles. The knee was evaluated at 90°/sec first, and then at 120°/sec to avoid exhaustion. The measurements were repeated 10 times to improve reliability. The first movement and the tenth movement, which had the lowest patient compliance, were not considered to increase the reliability (Tsepis et al., 2004).

The TUG test evaluates the activities related to walking, including dynamic stability. When an individual was ordered to “get up from the chair you are sitting”, the patient was asked to walk to the marked spot three meters ahead, and then, to come back and sit the chair. While the individual was sitting on the chair in a position that his/her back in contact with the back of the chair, he/she was asked to get up from the chair. The time elapsed from the moment that the patient stood up to the moment that he/she sat down again was recorded in seconds (Bonnyaud et al., 2015).

3 Statistics

Statistical analysis was performed using SPSS for Windows version 22.0 software (IBM Corporation, Armonk, NY, USA). The Kolmogorov-Smirnov test was used to evaluate the distribution of data. Categorical data and demographic characteristics of groups were compared with Chi-square test. In the comparison of pretreatment baseline data and percentage change scores of the groups before the treatment (posttreatment value - pretreatment value/pretreatment value×100), Independent T-Test was used for the analysis of normally distributed numerical data, and Mann Whitney U Test was used for the analysis of not normally distributed numerical data. Paired samples T-test was used to compare normally distributed numerical data before and after the treatment. Wilcoxon signed rank test was used to compare not normally distributed non-numerical data before and after the treatment. Pearson correlation coefficient was used to evaluate the relationship between parameters. The results were analyzed at 95% confidence interval, and a p-value < 0.05 was accepted to be statistically significant.

4 Results

Five of the 72 patients, who had a stroke at least 6 months ago and planned to be included in the study, were excluded from the study because they did not meet the inclusion criteria, and two of them refused to participate in this study stating that they were not able to continue to come to the controls. 65 patients were randomly divided into two groups (n: 33, WBE group and n: 32, LBE group). Flowchart of the study was given in Fig. 1.

Fig.1

Flowchart of the study.

The comparison of the pretreatment baseline demographic and clinical characteristics of the groups is given in Table 1. No significant difference was observed between the groups regarding baseline demographic data. However, there was a significant difference between two groups regarding FIM, BBS and TUG measurements, peak torque values (PT120EQ, PT120EH, PT120SH and PT90EQ, PT90EH, PT90SH) of the isokinetic measurements, and physical function, social function and general health sub-parameters of SF-36.

Table 1

Baseline demographic and clinical characteristics of the patients

Variables	Study group (n:30)	Control group (n:30)	P
Age (years) (mean±SD)	58.5±6.27	58.3±5.43	0.913
Type of stroke (i/h) (n)	27/3	23/7	0.166
Gender (f/m) (n)	12/18	17/13	0.196
Affected side (right/left) (n)	12/18	16/14	0.201
Times since injury (months) (mean±SD)	23.4±15.1	24.2±15.1	0.853
SBI (mean±SD)	1331.7±276.1	1446.9±644.0	0.192
DBI (mean±SD)	2310.3±716.1	2492.2±437.6	0.305
BBS (mean±SD)	39.6±7.1	30.3±10.9	0.001
FIM (mean±SD)	107±6.3	99.96±10.4	0.003
TUG (mean±SD)	18.5±6.3	26.5±10.5	< 0.001
PT120AH (mean±SD)	22.8±11.1	14.3±6.7	0.001
PT120AQ (mean±SD)	25.2±14.1	16.6±11.0	0.01
PT120UH (mean±SD)	33.0±12.3	24.3±8.6	0.002
PT120UQ (mean±SD)	48.4±20.8	41.5±21.3	0.206
PT90AH (mean±SD)	21.4±10.9	13.5±6.8	0.001
PT90AQ (mean±SD)	27.5±14.6	17.7±12.3	0.007
PT90UH (mean±SD)	34.2±14.3	23.9±8.4	0.001
PT90UQ (mean±SD)	55.3±25.5	44.4±22.8	0.088
SF 36 (mean±SD)
PF	34.3±19.2	20.0±18.2	0.004
RL	32.0±20.1	24.1±19.1	0.125
RLEP	32.2±18.5	22.2±22.0	0.062
V	57.6±19.1	55.0±15.5	0.554
GMH	57.2±14.6	53.3±15.2	0.318
SF	63.7±18.9	48.7±17.5	0.002
P	85.0±19.5	79.1±21.4	0.274
GH	65.5±14.9	56.8±16.4	< 0.037
Brunstroom (Median)(min-max)
Upper	4(1–5)	3(1–6)	0.129
Hand	2(1–6)	2(1–5)	0.273
Lower	5(3–5)	4(3–5)	0.152

SD: Standard deviation. I/H: ischemic/haemorrhagic, F/M; female/male, SBI: Statik balance index, DBI: Dinamik balance index, BBS: Berg Balance Scale, FIM: Functional Independence Measure, TUG: Time Up and Go, PT120AH: The affected side 120-degree/second peak torque value of the hamstring muscle, PT120AQ: The affected side 120-degree/second peak torque value of the quadriceps muscle, PT120UH: The unaffected side 120-degree/second peak torque value of the hamstring muscle, PT120UQ: The unaffected side 120-degree/second peak torque value of the quadriceps muscle, PT90AH: The affected side 90-degree/second peak torque value of the hamstring muscle, PT90AQ: The affected side 90-degree/second peak torque value of the quadriceps muscle, PT90UH: The unaffected side 90-degree/second peak torque value of the hamstring muscle, PT90UQ: The unaffected side 90-degree/second peak torque value of the quadriceps muscle, SF-36: short form 36 health survey, PF: physical function, RL: role limitations due to physical functioning, RLEP: role limitations due to emotional problems, V: vitality, GMH: general mental health, SF: social functioning, P: pain, GH: general health, WBR: water based exercise, LBE: land based exercise.

Comparison of the pretreatment and posttreatment values of two groups is given in Table 2. After the treatment, significant improvements were observed in both groups in clinical parameters (BBS, FIM, TUG), static and dynamic kinesthetic balance device parameters, and isokinetic measurements of quadriceps and hamstring muscles at 90 and 120 degrees/second angular velocity.

Table 2

Comparison of Pre-Treatment and Post-Treatment Evaluation Parameters of the WBE and LBE Group

	Study Group (mean±SD)			Control Group (mean±SD)
	Pre-treatment	Post-treatment	P	Pre-treatment	Post-treatment	P
BBS	39.6±7.1	45.1±6.7	< 0.001	30.2±10.8	36.7±10.2	< 0.001
FIM	107±6.3	110±6.2	0.003	99.9±10.4	103.8±10.7	0.002
TUG	18.5±6.3	14.7±3.8	< 0.001	26.2±10.5	20.0±8.6	< 0.001
SBI	1331.6±276.1	1011±272.4	0.001	1446.9±389.4	1083±296.8	0.001
DBI	2310.2±716.7	1760.5±499.6	< 0.001	2492.2±644.7	2009.0±593.8	< 0.001
PT120AH	22.8±11.1	27.7±12.1	< 0.001	14.3±6.7	18.7±7.6	< 0.001
PT120AQ	25.2±14.1	34.2±18.9	< 0.001	16.5±11.0	24.4±14.1	< 0.001
PT120UH	33.1±12.3	38.8±13.2	< 0.001	24.3±8.6	31.3±12.3	< 0.001
PT120UQ	48.4±20.8	57.5±21.4	< 0.001	41.5±21.3	51.2±25.1	< 0.001
PT90AH	21.4±10.9	26.6±12.6	< 0.001	13.5±6.8	18.2±7.9	< 0.001
PT90AQ	27.5±14.6	36.2±20.0	< 0.001	17.7±12.3	26.2±15.6	< 0.001
PT90UH	34.2±14.3	38.0±14.5	< 0.001	23.9±8.4	31.2±13.1	< 0.001
PT90UQ	55.3±25.5	64.7±26.9	< 0.001	44.4±22.8	56.5±26.1	< 0.001
SF 36 PF	34.3±19.2	48.5±16.4	< 0.001	20.0±18.2	30.8±16.5	<0.001
SF 36 RL	32.0±20.1	47.5±18.9	0.003	24.1±19.1	32.5±17.6	0.027
SF 36 RLEP	32.2±18.5	50.0±21.0	0.190	22.2±22.0	31.0±21.3	0.314
SF 36 V	57.6±19.1	72.0±16.0	0.001	55.0±15.5	58.5±14.7	<0.001
SF 36 GMH	57.2±14.6	70.0±11.9	0.003	53.3±15.2	61.7±14.5	<0.001
SF 36 SF	63.7±18.9	79.5±18.3	< 0.001	48.7±17.5	58.7±20.3	<0.001
SF 36 P	85.0±19.5	102±57.7	0.09	79.1±21.4	88.9±15.7	<0.001
SF 36 GH	65.5±14.9	76.5±12.1	< 0.001	56.8±16.4	69.6±16.7	<0.001
Brunstroom upper	4(1–5)	4(1–5)	0,005	3(1–6)	4(2–6)	< 0.001
Brunstroom hand	2(1–6)	2(1–6)	0,007	2(1–5)	2(1–6)	0.005
Brunstroom lower	5(3–5)	5(3–6)	0,003	4(3–5)	5(3–6)	< 0.001

SD: Standard deviation, BBS: Berg Balance Scale, FIM: Functional Independence Measure, TUG: Time Up and Go, SBI: Statik balance index, DBI: Dinamik balance index, PT120AH: The affected side 120-degree/second peak torque value of the hamstring muscle, PT120AQ: The affected side 120-degree/second peak torque value of the quadriceps muscle, PT120UH: The unaffected side 120-degree/second peak torque value of the hamstring muscle, PT120UQ: The unaffected side 120-degree/second peak torque value of the quadriceps muscle, PT90AH: The affected side 90-degree/second peak torque value of the hamstring muscle, PT90AQ: The affected side 90-degree/second peak torque value of the quadriceps muscle, PT90UH: The unaffected side 90-degree/second peak torque value of the hamstring muscle, PT90UQ: The unaffected side 90-degree/second peak torque value of the quadriceps muscle, SF-36: short form 36 health survey, PF: physical function, RL: role limitations due to physical functioning, RLEP: role limitations due to emotional problems, V: vitality, GMH: general mental health, SF: social functioning, P: pain, GH: general health.

In the the WBE exercise group, significant improvement was observed after the treatment regarding physical function, physical role limitation, vitality, social function, mental health and general health parameters included in SF- 36, and significant improvement was observed in the comparison of pre and post treatment values of the control group regarding physical function, physical role limitation, vitality, social function, mental health, general health and pain parameters (Table 2).

The mean values and p-values of the percentage change scores of the inter-group clinical parameters are shown in Table 3. Control group was observed to have higher improvement in the BBS according to the results of the indicated percentage change scores. In the comparison of percentage changes of dynamic balance index (DBI), no statistically significant difference was observed between two groups. However, a higher improvement was observed in WBE group regarding DBI values. In the inter-group comparison, no significant difference was found between the two groups regarding static balance index (SBI), FIM, TUG and all isokinetic peak torque measurements. According to the results of the percentage change scores of the SF-36 sub-parameters used inter-group comparison, the improvement observed in the vitality sub-parameter was found to be higher in WBE group than the control group (p = 0.015). No significant difference was observed between other sub-groups of SF-36 (Table 3).

Table 3

Comparison of Percent Change and Difference Scores of Clinical Parameters Between Groups

	Study Group	Control Group	P
	(mean±SD)	(mean±SD)
BBS	0.14±0.09	0.27±0.31	0.040
FIM	0.04±0.02	0.04±0.03	0.551
TUG	0.19±0.07	0.22±0.10	0.106
SBI	0.22±0.19	0.23±0.16	0.963
DBI	0.22±0.13	0.18±0.12	0.290
PT90AQ	0.34±0.36	0.63±0.82	0.087
PT120AQ	0.22±0.19	0.29±0.34	0.194
PT90UQ	0.20±0.17	0.32±0.34	0.091
PT120UQ	0.22±0.19	0.29±0.34	0.379
PT90AH	0.28±0.22	0.43±0.41	0.088
PT120AH	0.27±0.29	0.45±0.68	0.202
PT90UH	0.13±0.12	0.31±0.35	0.187
PT120UH	0.19±0.15	0.29±0.25	0.072
SF 36 PF	0.83±0.78	0.98±0.96	0.511
SF 36 RL	0.98±0.90	0.90±0.61	0.061
SF 36 RLEP	0.42±0.60	0.05±0.55	0.057
SF 36 V	0.37±0.61	0.08±0.16	0.015
SF 36 GMH	0.29±0.40	0.19±0.29	0.304
SF 36 SF	0.34±0.57	0.26±0.42	0.588
SF 36 P	0.22±0.57	0.17±0.26	0.678
SF 36 GH	0.20±0.26	0.26±0.26	0.440

SD: Standard deviation, BBS: Berg Balance Scale, FIM: Functional Independence Measure, TUG: Time Up and Go, SBI: Statik balance index, DBI: Dinamik balance index, PT90AQ: The affected side 90-degree/second peak torque value of the quadriceps muscle, PT120AQ: The affected side 120-degree/second peak torque value of the quadriceps muscle, PT90UQ: The unaffected side 90-degree/second peak torque value of the quadriceps muscle, PT120UQ: The unaffected side 120-degree/second peak torque value of the quadriceps muscle, PT90AH: The affected side 90-degree/second peak torque value of the hamstring muscle, PT120AH: The affected side 120-degree/second peak torque value of the hamstring muscle, PT90UH: The unaffected side 90-degree/second peak torque value of the hamstring muscle, PT120UH: The unaffected side 120-degree/second peak torque value of the hamstring muscle, SF-36: short form 36 health survey, PF: physical function, RL: role limitations due to physical functioning, RLEP: role limitations due to emotional problems, V: vitality, GMH: general mental health, SF: social functioning, P: pain, GH: general health.

5 Discussion

In this study, the goals of stroke rehabilitation, including motor improvement, improvement of balance, increased walking speed, and improvement of the QoL, were achieved in both treatment groups. When the comparison between the two groups was examined, the improvement observed in the vitality parameter of SF-36 was found to be higher in the WBE group, whereas the improvement in BBS was found to be higher in the conventional rehabilitation group.

Since performing an exercise inside pool is not exhausting, despite the number of repetitions, as it supports body weight and patients have less fear of falling, it is an appropriate choice to achieve success by increasing balance, coordination, flexibility and strenght. The use of this exercise has been increasing in many areas, such as in many orthopedic problems, osteoarthritis, rheumatoid arthritis, fibromyalgia, postmenopausal osteoporosis, Parkinson’s disease, multiple sclerosis, spinal cord injuries, and hemiplegia (Becker, 2009).

In a study comparing WBE and LBE and including 24 children diagnosed with cerebral palsy in 2014, WBE group achieved more improvement in terms of motor function and physical activity (Lai et al., 2015). In a study with patients with multiple sclerosis, where the motor function was greatly affected, and the risk of fall increased, the effects of 10-week aquatic therapy program on muscle strength, stability and strength were examined. An increase was observed in each angular velocity evaluated in the isokinetic maximum peak torque measurements, and the findings showed that there could be positive changes in muscle strength, fatigue and strength (Gehlsen et al., 1984).

During postural and proprioceptive balance training, water provides a comfortable environment with its hydrostatic pressure effect and allows to create resistance and a stable support environment through its buoyancy effect. In a randomized controlled study by Tripp et al. (2014) with 30 patients, 2-day LBE was combined with WBE (Halliwick method) of 3 days/week for two weeks, like our study, whereas LBEs were applied for control group for five days. In the WBE group, increase in BBS was shown to be statistically higher than the LBE group. In a study carried out by Zhu et al. (2016) on 28 stroke patients, 10-minute underwater treadmill training was added to the balance exercise program of 5 days/week for four weeks in WBE group, whereas only LBEs were applied in control group for five days. In both groups, significant improvements were observed in the BBS, functional stretch test (FST), and TUG results of the patients after treatment and WBE group had more significant improvements in FST. The findings suggested that functional mobility could be improved by water-based exercise program. However, no significant difference was observed between two groups, which is consistent with our findings. A Cochrane review states ‘no significant improvement of postural control for the people after stroke’ following aquatic therapy in comparison to standard treatment when analyzing the pooled mean difference of the BBS (Mehrholz et al., 2011). This conclusion was based on a study population of independent ambulatory patients at least one-year post stroke like the patients who participated in our study.

In this study, when the pre- and post-treatment values of the WBE group were compared, significant improvement was observed in all isokinetic measurements of the clinical parameters (Brunnstrom motor stage, BBS, FIM, and TUG), static and dynamic kinesthetic balance device parameters, and 90 and 120 degrees/second angular velocity of quadriceps and hamstring muscles, which were compatible with the literature. The same improvements were observed in the LBE group. The improvement in the BBS was found to be statistically significant in the LBE group in the inter-group comparison. We think that because patients were randomly selected for our study, they had different pre-treatment BBS values, which might have been effective in this result. As the BBS value of the LBE group was lower before the treatment, the improvement in the clinical measurements might have been more effective in this group even a little. Balance training program was applied non-stop for five days in the conventional treatment program, whereas it was applied for three days with intervals in the water and two days on the land in WBE group. We think that this situation affected the reinforcement of the patients’ improvements negatively. In addition to this, we have performed an evaluation objectively with the Sportkat kinesthetic balance device, and the findings showed that the WBE group was also effective in SBI and DBI measurements as well as BBS in the intra-group comparison. We revealed the improvements of the patients, which were not seen in the BBS, through objective numerical data.

In the literature review, studies on WBE are generally concerning the balance or extremity function, and there are very few studies about the effect of these exercises on muscle power (Montagna et al., 2014; Tripp & Krakow 2014; Zhu et al., 2016). In a randomized controlled trial conducted by Chu et al. (2004) in 2004 with 12 chronic stroke patients, an aquatic endurance program that consisted of aerobic exercises in water was applied three times a week for eight weeks. Before and after the treatment, isokinetic flexor and extensor muscular strength (at 60 degrees/sec) of the knee and hip joints were measured. In the group, in which the aquatic endurance program was applied, significant improvement was achieved regarding paretic lower limb muscle strength compared to the control group. According to the speeds of 90 and 120 degrees/sec used in our study, measurement at a low speed of 60 degrees/sec, on the one hand, provided true isokinetic conditions for a longer time, and on the other hand, can cause the dynamometer arm to accelerate and slow down in a shorter time and to be out of control as well as it may have negative effects on the results.

In this study, the improvement in the isokinetic parameters of the WBE group was shown at both 90 degrees/sec and 120 degree/sec measurements. It is known that the data received when the measurement rate is increasing, is the data showing the stability. Noh et al. (2008) evaluated muscle strength of the stroke patients after aquatic therapy. In a randomized controlled trial with 25 ambulatory patients with chronic stroke, patients were divided into two groups, which were WBE group and control group, as we did in our study. For the study group, 8-week aquatic therapy of 3 times/week based on a combination of Halliwick and Ai Chi methods was applied, whereas LBEs of 3 times/week were applied for the control group. 10 people from each group, in total 20 people, completed the study. Isokinetic flexor and extensor muscular strength of the affected and not-affected knee at the 60 degrees/sec angular velocity, and of the back at the 30 degrees/sec were measured. The maximum peak torque values were measured with the Biodex isokinetic device.

Consistent with our findings, a statistically significant improvement was observed when all pretreatment and posttreatment isokinetic parameters of both groups were compared. In the literature, the angular velocity of 60 degrees/sec used to evaluate the flexor and extensor muscles of the knee was reported to be low for ideal measurements. The angular velocity of 90 and 120 degrees/sec used in our study provided evaluation both at two different speeds and closer to ideal speed. In Noh et al.’s (Noh, 2008) study, only increase in knee flexor maximum peak torque values was found to be significantly higher than the control group, whereas an increase in the knee extensors and flexors on the affected side was observed in the WBE group in our study. This finding could be associated with that strengthening exercises were applied with the aquatic therapy of 3 days/week, to which 2-day LBE was added, for the patients in WBE group.

The improvement in walking, motor strength and balance parameters achieved by rehabilitation methods applied to stroke patients increases the patient’s participation in social life. Patients are expected to have better cognitive function and better QoL. We also evaluated the patients with the SF-36 QoL questionnaire in this respect. In Aidar et al.’s (Aidar, 2007) study a 12-week aquatic exercise program was applied to the WBE group, and SF-36 results were compared with the control group. Significant differences were found between WBE group before treatment and after the treatment in the intra-group evaluation. Significant differences were also found in inter-group evaluation in regarding functional capacity, physical properties, pain, general health status, vitality, social function, and mental health. In our study, in line with previous studies, significant improvement was observed in SF-36 subheadings, including physical function, physical role limitation, vitality, social function, mental health, and general health parameters, in both groups after the 8-week treatment in the intra-group evaluation. The improvement in pain parameters of the WBE group was not statistically significant, whereas a significant improvement was observed in the control group. An explanation for the difference may be because randomization reason and the pretreatment SF-36 questionnaire values of the groups were not the same. Thus, percentage change scores were used in the inter-group comparison. According to the results, the recovery in pain was not significant in inter-group comparison. There was no statistically significant difference between the groups regarding physical function, physical role limitation, social function, mental health, and general health parameters, which were found to be significant in both groups. These results suggest that regular exercise done by in patients with hemiplegia in any environment is effective on physical and mental component scales of QoL. The improvement in the vitality sub-parameter was higher in the WBE groups compared to the control group according to the percentage change scores. Doing physical exercises in the water creates an excellent supporting environment by providing an independent movement environment thanks to the buoyancy of water, which could make the patients more energetic in their daily lives as being a fun and motivating activity regarding the patient (Zhu et al., 2016).

In our study, commonly used balance and walking measures were included, which are also used in other studies (Montagna et al., 2014; Tripp & Krakow 2014; Zhu et al., 2016). SBI and DBI scores of Sportkat kinesthetic balance device were also included in addition to the BBS. In a study by Günendi et al. (2010) the scores obtained by the Sportkat balance device were found to be correlated with other balance measures and were shown to be reliable. It has been recommended that Sportkat balance device should be used widely in the evaluation of balance ability. Our findings showed that WBE group received very low scores (Low score refers to a better balance. Zero means perfect balance, and impossible to reach) from both static and dynamic balance measurements compared to control group. Despite the excellent balance increase, no significant superiority was achieved in TUG test measuring walking speed compared to the control group. One of the remarkable findings of our study is that our study is in line with the rehabilitation practices whose efficacy was proven, and it is possible to have better results with increasing the duration and frequency of exercises done in combination with WBE therapy. Since significant improvements were achieved in both groups, there is a need for comprehensive studies to evaluate these parameters with more patients.

6 Limitations

One of the most important limitations of this study is the patients’ follow-up period. Future research may be useful in which the patients should be followed up on the third and 12th months. Another limitation was that 60 patients participated in this study. However, we should note that when we examine the relevant studies on WBE in the relevant literature, our study has the largest number of patients and measuring the largest number of parameters. We think future research with larger numbers patients could provide valuable insights into the literature. Another limitation is the statistically significant difference in the comparison of pre-treatment data of groups regarding some parameters, such as BBS, FIM, Brunnstrom stages, TUG and SF-36. This difference could be due to the random selection of the groups. Thus, we used percentage change scores in the comparison of two groups.

7 Conclusion

Our findings suggest that WBEs are as effective as conventional rehabilitation regarding increasing muscle strength and improving QoL although there is no statistically significant difference between the evaluation parameters of the study group and the control group. Adding the aquatic rehabilitation to conventional rehabilitation program may contribute to the vitality of the patient and continuity of rehabilitation since it is enjoyable, motivating and enhances the QoL. The improvement in walking speed and balance functions provided by conventional LBE, of which efficacy has been proven, still maintains its importance.

Conflict of interest

The authors declare that they have no conflicts of interest.

References

Aidar

F. J.

, Silva

A. J.

, Reis

V. M.

, Carneiro

, Carneiro-Cotta

(2007). A study on the quality of life in ischaemic vascular accidents and its relation to physical activity. Revista de Neurologia, 45, 518–522.

Alonso

A. C.

, Luna

N. M.

, Dionísio

F. N.

, Speciali

D. S.

, Leme

L. E. G.

, Greve

J. M. D.

(2014). Functional Balance Assessment: Review. Med Express, 1, 298–301.

Becker

B. E.

(2009). Aquatic Therapy: Scientific Foundations and Clinical Rehabilitation Applications. PM&R, 1, 859–872.

Bonnyaud

, Pradon

, Bensmail

, Roche

(2015). Dynamic stability and risk of tripping during the timed up and go test in hemiparetic and healthy subjects. PLoS One, 10, 1–14.

Chu

K. S.

, Eng

J. J.

, Dawson

A. S.

, Harris

J. E.

, Ozkaplan

, Gylfadóttir

(2004). Water-based exercise for cardiovascular fitness in people with chronic stroke: A randomized controlled trial. Arch Phys Med Rehabil, 85, 870–874.

Coleman

E. R.

, Moudgal

, Lang

, Hyacinth

H. I.

, Awosika

O. O.

, Kissela

B. M.

, Feng

(2017). Early Rehabilitation After Stroke: a Narrative Review. Curr Atheroscler Rep, 7, 59.

Donnan

G. A.

, Fisher

, Macleod

, Davis

S. M.

(2008). Stroke. Lancet, 371, 1612–1623.

Gehlsen

G. M.

, Grigsby

S. A.

, Winant

D. M.

(1984). Effects of an Aquatic Fitness Program on the Muscular Strength and Endurance of Patients with Multiple Sclerosis. Phys Ther, 64, 653–657.

Günendi

, Özyemisçi

T. Ö.

, Uzun

M. K.

, Öztürk

G. T.

, Demirsoy

(2010). Reliability of quantitative static and dynamic balance tests on kinesthetic ability trainer and their correlation with other clinical balance tests. J PMR Sci, 13, 1–5.

10.

Hansen

M. S.

, Dieckmann

, Jensen

, Jakobsen

B. W.

(2000). The reliability of balance tests performed on the kinesthetic ability trainer (KAT 2000). Knee Surg Sports Traumatol Arthrosc, 8, 180–185.

11.

Kidd

, Stewart

, Baldry

, Johnson

, Rossiter

, Petruckevitch

, Thompson

A. J.

(1995). The Functional Independence Measure: A comparative validity and reliability study. Disabil Rehabil, 17, 10–14.

12.

Lai

C-J.

, Liu

W-Y.

, Yang

T-F.

, Chen

C-L.

, Wu

C-Y.

, Chan

R-C.

(2015). Pediatric aquatic therapy on motor function and enjoyment in children diagnosed with cerebral palsy of various motor severities. J Child Neurol, 30, 200–208.

13.

Langhorne

, Bernhardt

, Kwakkel

(2011). Stroke rehabilitation. Lancet, 377, 1693–1702.

14.

Mehrholz

, Kugler

, Pohl

(2011). Water-based exercises for improving activities of daily living after stroke. Cochrane Database Syst Rev, 19, CD008186.

15.

Montagna

J. C.

, Santos

B. C.

, Battistuzzo

C. R.

, Loureiro

A. P. C.

(2014). Effects of aquatic physiotherapy on the improvement of balance and corporal symmetry in stroke survivors. Int J Clin Exp Med, 7, 1182–1187.

16.

Noh

D. K.

, Lim

J. Y.

, Shin

H. I.

, Paik

N. J.

(2008). The effect of aquatic therapy on postural balance and muscle strength in stroke survivors — a randomized controlled pilot trial. Clin Rehabil, 22, 966–976.

17.

Park

, Lee

, Yoon

, Lee

, Choi

, Roh

(2011). Comparison of the Effectsof Exercise by Chronic Stroke Patients in Aquatic and Land Environments. J Phys Ther Sci, 23, 821–824.

18.

Tripp

, Krakow

(2014). Effects of an aquatic therapy approach (Halliwick-Therapy) on functional mobility in subacute stroke patients: A randomized controlled trial. Clin Rehabil, 28, 432–439.

19.

Tsepis

, Giakas

, Vagenas

, Georgoulis

(2004). Frequency content asymmetry of the isokinetic curve between ACL deficient and healthy knee. J Biomech, 37, 857–864.

20.

Wang

, Wu

, Zhao

, Yan

, Ma

, Wu

, . (2008). Health related quality of life measured by SF-36: a population-based study in Shanghai, China. BMC Public Health, 8, 292.

21.

Zhang

, Wang

Y-Z.

, Huang

L-P.

, Bai

, Zhou

, Yin

M-M.

, Zhao

, Zhou

X. N.

, Wang

H. T.

(2016). Aquatic Therapy Improves Outcomes for Subacute Stroke Patients by Enhancing Muscular Strength of Paretic Lower Limbs Without Increasing Spasticity: A Randomized Controlled Trial. Am J Phys Med Rehabil, 11, 1–10.

22.

Zhu

, Cui

, Yin

, Yu

, Zhou

, Wang

, . (2016). Hydrotherapy vs. conventional land-based exercise for improving walking and balance after stroke: A randomized controlled trial. Clin Rehabil, 30, 587–593.