Effects of an aquatic therapy approach (Halliwick-Therapy) on functional mobility in subacute stroke patients: a randomized controlled trial

Abstract

Objective:

To evaluate the effects of an aquatic physiotherapy method (Halliwick-Therapy) upon mobility in the post-acute phase of stroke rehabilitation.

Design:

Randomized controlled trial.

Setting:

Hospital for neurological rehabilitation.

Participants:

Adult patients after first-ever stroke in post-acute inpatient rehabilitation at least two weeks after the onset of stroke (n = 30).

Interventions:

In the Halliwick-Therapy group (n = 14) the treatment over a period of two weeks included 45 minutes of aquatic therapy three times per week and a conventional physiotherapeutic treatment twice a week. Subjects in the control group (n = 16) received conventional physiotherapeutic treatment over a period of two weeks five times per week.

Outcome measures:

The primary outcome variable was postural stability (Berg Balance Scale). Secondary outcome variables were functional reach, functional gait ability and basic functional mobility.

Results:

Compared to the control group, significantly more subjects in the Halliwick-Therapy group (83.3% versus 46.7%) attained significant improvement of the Berg Balance Scale (P < 0.05). Improvement of the functional gait ability was significantly higher in the Halliwick-Therapy group (mean (SD) 1.25(0.86)) than in the control group (mean (SD) 0.73 (0.70)) (P < 0.1). The mean differences of improvements in functional reach and basic functional mobility were not statistically significant between groups.

Conclusions:

This study indicates that Halliwick-Therapy is safe and well tolerated in stroke patients in post-acute rehabilitation and has positive effects upon some aspects of mobility.

Keywords

Stroke aquatic therapy mobility balance Halliwick

Introduction

Many stroke survivors remain dependent on assistance after the acute phase of treatment due to restrictions of mobility.¹ The present knowledge about motor recovery after stroke and successful therapeutic interventions in motor rehabilitation emphasizes the need for treatment strategies with early active and repetitive training which focuses on realizing the remaining motoric function in a meaningful and motivating context. Nevertheless, the required active input for motor relearning is difficult to realize for patients and therapists when facing severe motor impairments. Therefore patients in the early rehabilitation phase are often forced into inactivity or ineffective compensations. In order to enable patients to be more active at an early stage of rehabilitation, ‘modifying the task or environment’² is required. The specific environmental conditions of water influencing physiological processes and motor activity are described as density, buoyancy, hydrostatic pressure, viscosity and thermodynamics.³ In this context the training environment in water may facilitate motor activity, as training of paretic muscles is promoted, while their use might be restricted under conditions of gravity outside the water.

Halliwick-Therapy, which this paper focuses on, was developed and first described by James McMillan and is characterized as a problem-solving approach with systematic intervention.^4–9 Halliwick-Therapy focuses on postural stability and controlled mobility as a basis for safe and coordinated movement in and out of water. In a 10-point programme skills for being safe and independent in the water are gradually acquired. Phase 1 of the programme includes adjustment to the environment, while the second phase focuses on learning three-dimensional active balance control utilizing dynamic and static tasks. The goal of the third phase of movement is an independent swimming technique adjusted to individual physical disability. In the course of neurological rehabilitation phase 3 is being conducted less frequently, as the goal of the therapeutic use of the Halliwick concept is not primarily independent swimming, but to enhance general motor independence both in and out of water. The 10-point programme is therefore complemented by water-specific therapy, a system for creating individual exercises which can be adapted and increased in difficulty by carefully selecting a combination of hydromechanical effects, the starting position and the task.

Although previous studies found improvements in postural stability,¹⁰ cardiovascular fitness,¹¹ gait speed¹¹ and strength of the hemiparetic leg^10,11 in chronic stroke patients, a recent Cochrane review¹² could neither endorse nor reject aquatic physiotherapy for stroke patients, as the authors described a ‘lack of hard evidence for water-based exercises after stroke’.

In particular, to our knowledge, so far no data about the effectiveness of aquatic therapy in the subacute phase of stroke rehabilitation in an inpatient setting are available. Therefore the aim of this study was to examine in this patient group the effects of aquatic therapy in reference to postural stability, ambulation and basic functional mobility in comparison to standard physiotherapy within a randomized controlled trial.

Methods

In this study we included adult patients after first-ever stroke in post-acute inpatient rehabilitation at least two weeks after the onset of stroke. It was determined that as a minimum motor requirement for inclusion, transferring from one chair to another had to be possible independently or with assistance of not more than one person. Excluded from participation were patients with independent gait ability indoors. Further exclusion criteria were an instable general condition, infectious diseases, severe heart failure, tracheostomy, sores, incontinence, lack of ability to cooperate, rejection of aquatic therapy, perforation of the tympanic membrane and the presence of additional other neurological disorders.

All patients classified as suitable after a screening utilizing inclusion and exclusion criteria received a detailed written and verbal explanation of the structure and content of the study. After informed consent was given, patients were randomly allocated to one of the treatment groups by a person not involved with the process of treatment, testing and evaluation. The randomization was conducted using numbered pre-filled envelopes which contained an equal number of lots for both treatment groups. As the number of study participants was not exactly predictable before the onset of the study, we conducted a blocked randomization with the aim of achieving a similar number in both study groups.¹³ Initially 20 lots were randomized and further 14 at a later date.

For the Halliwick-Therapy group it was determined that the physiotherapeutic treatment consisted of a combination of three Halliwick-Therapy sessions and two conventional physiotherapy sessions per week over a period of two weeks, with all sessions having a treatment time of 45 minutes. The Halliwick-Therapy sessions were conducted by physiotherapists who had participated in a certified Halliwick-Therapy course.

Included in the aquatic therapy sessions was the time for entering and exiting the pool, time for preparations and post-processing, such as towelling oneself off and getting dressed in a bathrobe. The pure treatment time in the pool was approximately 35 minutes, of which about 5 minutes were for exercises in water familiarization and mental adaption (learning phase 1 of the Halliwick concept)⁸ and about 15 minutes for exercising rotational control (learning phase 2 of the Halliwick concept).⁸ In the remaining 15 minutes locomotion under various disturbances and in changing water depths was primarily exercised.

The control group received five sessions of standard physiotherapy per week over a period of two weeks and treatment time was 45 minutes as in the Halliwick-Therapy group. The contents of the standard physiotherapy sessions were not defined and therefore the treatment consisted of an individual mix of different treatment concepts, task-specific exercising of various tasks in the area of mobility and possibly treadmill training.

As the aim of this study was to investigate effects on functional mobility, we decided to assess different aspects of mobility that are meaningful for safety and independence in the everyday mobility of stroke patients.

As primary outcome measure postural stability was determined and assessed by the Berg Balance Scale.¹⁴ With this tool, performance as well as dependency on assistance is scored with 0–4 points in 14 different items in the area of mobility, so a maximum of 56 points can be achieved on an interval scale. Improvement of postural stability is described as the most important prognostic factor in stroke patients for achieving independent gait ability, while improving the strength of the hemiparetic leg is less associated with the improvement of walking ability.¹⁵ Patients with a Berg Balance Scale score of less than 45 are known to have an increased risk of falling¹⁶ and changes of more than six points indicate clinically relevant changes.¹⁷

For the secondary outcome measure, the measurements of functional reach,¹⁸ functional gait ability and basic functional mobility were determined. Whereas functional reach measures the safe and stable weight shifting ability while standing, Functional Ambulation Categories (FAC)¹⁹ describe the dependency on assistance for gait on a 5-point scale. The Rivermead Mobility Index²⁰ assesses independent performance of 15 tasks in the area of everyday mobility.

In addition to the outcome measurements, descriptive data were used to define the study groups. This data consisted of age, gender, time since stroke, side of lesion and type of stroke (ischaemic or haemorrhagic). Furthermore, the Barthel Index²¹ was collected to determine and compare the patients’ functional status at the start of intervention.

All tests were conducted observer-blind and the assignment of subjects to Halliwick-Therapy group or control group was not made available to the evaluating person until after the study period.

We conducted a descriptive analysis of the baseline data and performed a chi-square test of the nominally scaled data for different distributions between study groups at baseline. Furthermore we tested the mean values of the quantitative baseline data for differences with a t-test.

For both study groups we calculated median and mean values of the differences between baseline and follow-up measurements of all outcome variables. Group means were compared with t-tests for unconnected samples to determine significant differences. Additionally we determined for each group the number of participants with clinically relevant changes of the Berg Balance Scale, Functional Ambulation Categories and Rivermead Mobility Index. Analysis of the collected data was performed using BiAS. for Windows 10.0 and G * Power 3.²²

Results

During the study period 30 stroke patients met the inclusion criteria and were randomized to one of the study groups (Figure 1). Fourteen patients were assigned to the Halliwick-Therapy group and 16 to the control group and baseline measurements were conducted with all 30 patients. Two participants in the Halliwick-Therapy group and one participant in the control group dropped out during the course of the study. Aside from these drop-outs, which occurred independently of the interventions in this study, no other complications or negative side-effects of the Halliwick-Therapy were observed. The follow-up measurements after the study period were conducted with 27 patients, so that complete data of 12 participants of the Halliwick-Therapy group and 15 participants of the control group could be evaluated.

Figure 1.

Flow diagram.

Table 1 shows the participants’ characteristics. The characteristics of the participants were similar in both study groups and there were no statistically significant differences before the treatment period with regard to the investigated outcome measures.

Table 1.

Characteristics of the participants.

	Halliwick-Therapy group n = 14	Control group n = 16
Gender (male/female)^a	9/5	10/6
Age (years)^a	64.8 ± 15.0	65.0 ± 15.1
Side of lesion (left/right)^a	4/10	6/10
Type of stroke (ischaemic/haemorrhagic)^a	12/2	15/1
Days since stroke^a	51.9 ± 37.7	39.0 ± 27.9
Barthel Index at baseline^a	48.57 ± 15.74	54.38 ± 17.21

Difference between the study groups not significant (P > 0.05).

Table 2 shows significant improvements (P < 0.01) of the primary outcome measure (Berg Balance Scale) as well as the secondary outcome measures (functional reach, Functional Ambulation Categories, Rivermead Mobility Index) in both the Halliwick-Therapy group and the control group. We tested the mean improvements of the outcome measures for group differences with a t-test for independent samples. The t-test shows the mean improvement of the Functional Ambulation Categories was significantly higher in the Halliwick-Therapy group than in the control group (P < 0.1). The differences in mean values of the improvements in the Berg Balance Scale, functional reach and the Rivermead Mobility Index were not statistically significant.

Table 2.

Outcome measures at baseline and after two weeks of intervention.

	Halliwick-Therapy group				Control group
	Baseline	Follow-up	Median difference (90% CI)	Mean (SD) difference	Baseline	Follow-up	Median difference (90% CI)	Mean (SD) difference
BBS	27.08 (13.46)^a	38.08 (15.95)	10.5 (6–15)	11 (6.25)^b	28.73 (14.93)^a	37.60 (11.01)	5 (2–14)	8.87 (9.08)^b
FR (cm)	13.66 (10.70)^a	21.08 (11.75)	4.5 (1–15)	7.42 (7.36)^b	10.86 (9.75)^a	16.86 (8.52)	4 (2–12)	6 (6.19)^b
FAC^c	1.50 (1.24)^a	2.75 (1.65)	1 (1–2)	1.25 (0.86)^b	2.20 (1.47)^a	2.93 (1.22)	1 (0–1)	0.73 (0.70)^b
RMI	5.08 (2.42)^a	7.50 (3.55)	2 (0–5)	2.42 (2.35)^b	5.86 (2.61)^a	7.93 (2.49)	2 (1–3)	2.07 (1.33)^b

Difference between the study groups at baseline not significant (P > 0.05).

Significant difference between baseline and follow-up (P < 0.01).

Significant difference between improvements of the study groups (P < 0.1).

BBS, Berg Balance Scale; FR, functional reach; FAC, Functional Ambulation Categories; RMI, Rivermead Mobility Index; CI, confidence interval.

With regard to the Berg Balance Scale, improvements greater than 6 points are described as clinically relevant.¹⁷ Likewise this applies to changes of at least 1 point in the Functional Ambulation Categories²³ or at least 2 points in the Rivermead Mobility Index.²⁰ Table 3 shows a higher number and proportion of participants with clinically relevant improvements after the treatment period in the Halliwick-Therapy group. The different distributions were statistically significant (P < 0.05) in a chi-square test concerning the Berg Balance Scale (83.3% (n = 10) versus 46.7% (n = 7)) in favour of interventional treatment. The different distributions in the evaluation of the results of the Functional Ambulation Categories and the Rivermead Mobility Index were not statistically significant.

Table 3.

Participants with clinically relevant outcome improvement.

		Halliwick-Therapy group (n = 12)	Control group (n = 15)
BBS	Mean (SD) difference	11 (6.25)	8.87 (9.08)
	Improvement >6	n = 10 (83.3%)^a	n = 7 (46.7%)^a
	Improvement ≤6	n = 2 (16.7%)^a	n = 8 (53.3%)^a
FAC	Mean (SD) difference	1.25 (0.86)^b	0.73 (0.70)^b
	Improvement ≥1	n = 10 (83.3%)	n = 9 (60%)
	Improvement <1	n = 2 (16.7%)	n = 6 (40%)
RMI	Mean (SD) difference	2.42 (2.35)	2.07 (1.33)
	Improvement ≥2	n = 5 (41.7%)	n = 5 (33.3%)
	Improvement <2	n = 7 (58.3%)	n = 10 (66.7%)

Significant difference between study groups (P < 0.05).

Significant difference between mean improvements of the study groups (P < 0.1).

BBS, Berg Balance Scale; FAC, Functional Ambulation Categories; RMI, Rivermead Mobility Index.

Discussion

The results of this study support the hypothesis that aquatic therapy may have positive effects on functional mobility in subacute stroke patients. This study has shown that significantly more participants of the Halliwick-Therapy group achieved clinically relevant improvements of the Berg Balance Scale compared to standard treatment. Furthermore the improvement in functional gait ability was significantly greater in the Halliwick-Therapy group than in the control group. We did not observe any negative side-effects of the application of aquatic therapy after a minimum of two weeks post stroke.

Considering the limited number of participants and the short intervention period of two weeks in this study, the power to detect significant differences between standard treatment and aquatic therapy was limited. A power analysis conducted post hoc showed that with respect to 30 included patients, the relatively short study period of two weeks achieved only a 30% probability of detecting a false null hypothesis in relation to the primary outcome measure. Therefore the results need to be interpreted with caution. This also applies to generalizability of the results of this study to the stroke population, as we do not know what proportion of subacute stroke patients in principle meets the inclusion criteria to take part in aquatic therapy.

This study did not investigate any long-term effects of an aquatic treatment. Therefore it is unknown if the observed improvements in favour of interventional treatment were a functional advantage for the patients at a later point in time.

With regard to the effect on functional mobility, the Rivermead Mobility Index, which we chose due to its reliability and fast performance, did not have the sensitivity to detect minor changes over the short treatment period in a group of severely impaired patients.

The results of the Berg Balance Scale of our study can be compared with those of the study of Noh et al.,¹⁰ which is, as far as is known to us, the only published study investigating the effects of Halliwick-Therapy on stroke patients. At baseline Noh et al. found a mean Berg Balance Scale score of 43.3 in the aquatic therapy group, which is considerably higher than the mean Berg Balance Scale score of the Halliwick-Therapy group at baseline in our study (mean (SD) 27.08 (13.46)). However, the mean improvement of postural stability after two weeks of intervention was higher, with a mean (SD) of 11.0 (6.25) points on the Berg Balance Scale than the mean (SD) improvement of 7.6 (6.2) points in the study of Noh et al. after an eight-week intervention. This greater effect after a shorter intervention period can be explained by the considerably different functional status at baseline and a different potential for motor recovery due to the difference in time between stroke and the start of the study (2.8 years vs. 51.93 days).

A Cochrane review¹² states ‘no significant improvement of postural control for people after stroke’ following aquatic therapy in comparison to standard treatment when analysing the pooled mean difference of the Berg Balance Scale. However, unlike our investigation, the two studies^10,11 included in this comparison described chronic stroke patients who were able to walk independently. Furthermore, in this Cochrane review Mehrholz et al.¹² found the ability to walk was not significantly improved by water-based exercises, but again this conclusion was based on a study population of independently ambulatory patients at least one year post stroke.¹¹

The different findings on the effect of aquatic therapy upon postural stability and gait ability in our study compared with previous studies about chronic, independently ambulatory patients may imply a higher effectiveness of water-based exercises earlier post stroke and with patients who are not independently ambulatory.

In addition to significant improvements of all measured outcome variables no negative side-effects of the aquatic therapy occurred in this study. Therefore this study indicates a relatively safe application of the Halliwick-Therapy in post-acute stroke rehabilitation when utilizing the defined inclusion and exclusion criteria and a minimum of two weeks between stroke and beginning of aquatic treatment. With regard to the safety of stroke patients with impaired mobility who are likely to have an increased risk of falling, Halliwick-Therapy differs from other aquatic therapy approaches in that it offers a systematic structure beginning with water familiarization and breath control. In the course of the therapy sessions, the goal is to learn to organize motor control in order to keep or regain balance while performing different tasks. The therapist uses fluid mechanics to facilitate and challenge motor activity, while the buoyancy and viscosity of the water give the patient more time to develop coordinated motor strategies to accomplish the task. It is important to note that the findings in this study suggest that motor relearning in the water utilizing the Hallwick concept has positive effects on mobility outside the water, however it remains unclear if the overall effect on mobility is greater than with standard land-based physiotherapy.

This study did not investigate whether a longer intervention period increases the observed effects of Halliwick-Therapy on postural stability and functional gait ability, or whether a follow-up study would show any differences between the study groups at a later date. As the highest potential for motor recovery can be observed in the first months after a stroke, early improvements in the area of functional mobility by aquatic therapy could positively influence the further outcome of independent mobility. Thus, the improvement of postural stability is the most important factor for achieving independent gait ability.¹⁵ Further studies with larger sample sizes, a longer intervention period and follow-up measurements are therefore necessary.

Clinical messages

Aquatic physiotherapy with application of the Halliwick-Therapy in combination with conventional physiotherapy may effectively improve postural stability and gait ability in stroke patients in a subacute rehabilitation phase.

This study indicates a safe application of the Halliwick-Therapy after a minimum of two weeks post stroke.

Footnotes

Acknowledgements

The authors would like to acknowledge the contributions of Stephanie Humbert, Ann-Kathrin Bruhn and Gisela Göser for organizing the participants’ appointments and the support of Joy Coulton, Barbara Pudlo and Kristina Pudlo for translation of this manuscript. We would also like to thank the team of physiotherapists at the Asklepios Neurological Hospital Falkenstein for supporting the study.

Conflict of interest

The authors declare that there is no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Roth

. Trends in Stroke Rehabilitation. Eur J Phys Rehabil Med 2009; 45: 247–254.

Carr

Shepherd

. Stroke rehabilitation: guidelines for exercise and training to optimize motor skill. London: Butterworth-Heinemann, 2003, p. 22.

Becker

. Aquatic therapy: scientific foundations and clinical rehabilitation applications. Phys Med Rehabil 2009; 9: 859–872.

McMillan

. The role of water in rehabilitation. Fysioterapeuten 1978; 45: 43–46, 87–90, 236–240.

Martin

. The Halliwick Method. Physiotherapy 1981; 67: 288–291.

Paeth

. Schwimmtherapie “Halliwick-Methode” nach McMillan bei erwachsenen Patienten mit neurologischen Erkrankungen. Z Krankengymn 1984; 36: 100–112.

Gamper

. Wasserspezifische Bewegungstherapie und Training. Stuttgart: Gustav-Fischer-Verlag, 1991.

Lambeck

. Das Halliwick-Konzept, 2001. http://www.halliwick.de/downloads (accessed 17 November 2009).

Schick

Lambeck

Hulselmans

. Evidente Wassertherapie in der Rehabilitation. Halliwick-Konzept nach McMillan. Pt Z Physiother 2011; 5: 30–35.

10.

Noh

Lim

Shin

Paik

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

11.

Chu

Eng

Dawson

Harris

Ozkaplan

Gylfadóttir

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

12.

Mehrholz

Kugler

Pohl

. Water-based exercises for improving activities of daily living after stroke. Cochrane Database Syst Rev 2011; (1): CD008186.

13.

Schulz

Altman

Moher

for the CONSORT Group. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomized trials. BMJ 2010; 340: 698–702.

14.

Berg

Wood-Dauphinee

Williams

. Clinical and laboratory measures of postural balance in an elderly population. Physiother Can 1989; 41: 304–311.

15.

Kollen

Van de Port

Lindeman

Twisk

Kwakkel

. Predicting improvement in gait after stroke: a longitudinal prospective study. Stroke 2005; 36: 2676–2680.

16.

Andersson

Kamwendo

Seiger

Appelros

. How to identify potential fallers in a stroke unit: Validity indexes of four test methods. J Rehabil Med 2006; 38: 186–191.

17.

Stevenson

. Detecting change in Patients with stroke using the Berg Balance Scale. Austr J Physiother 2001; 47: 29–38.

18.

Duncan

Weiner

Chandler

Studenski

. Functional reach: a new clinical measure of balance. J Gerontol 1990; 45: M192–M197.

19.

Holden

Gill

Magliozzi

Nathan

Piehl-Baker

. Clinical gait assessment in the neurologically impaired. reliability and meaningfulness. Phys Ther 1984; 1: 35–40.

20.

Collen

Wade

Robb

Bradshaw

. The Rivermead Mobility Index. A further development of the Rivermead Motor Assessment. Int Disabil Stud 1991; 2: 50–54.

21.

Schädler

Kool

Lüthi

. Assessments in der Rehabilitation, Vol 1: Neurologie, second edition. Bern: Verlag Hans Huber, 2009, pp. 76–78.

22.

Faul

Erdfelder

Lang

Buchner

. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Meth 2007; 2: 175–191.

23.

Mehrholz

Wagner

Rutte

Meissner

Pohl

. Predictive validity and responsiveness of the Functional Ambulation Category in hemiparetic patients after stroke. Arch Phys Med Rehabil 2007; 88: 1314–1319.