Disponible en ligne sur www.sciencedirect.

com

Annales de readaptation et de mdecine physique 51 (2008) 9095

http://france.elsevier.com/direct/ANNRMP/

Original article
Evaluation of the immediate and midterm effects of mobilization in
hot spa water on static and dynamic balance in elderly subjects
L. Berger a,*, C. Klein a, M. Commandeur b
a
Laboratoire de modelisation des activites sportives, universite de Savoie, domaine universitaire, 73376 Le Bourget-du-Lac cedex, France
b
IUP sante kinesitherapie sport, ecole de kinesitherapie, 19A, avenue de Kimberley, 38130 Echirolles, France
Received 31 May 2007; accepted 30 October 2007

Abstract
Objectives. The purpose of this study was to compare one and four weeks of active mobilization during balneotherapy (B) with mobilization on
land (L) in terms of the respective effects on static and dynamic balance in elderly subjects.
Methods. Twelve elderly volunteers (mean age: 65.6  6.3 years) were evaluated for pain (on a visual analogue scale, [VAS]), static balance
(as measured by the centre of foot pressure, [COP]) and dynamic balance (the Timed Up and Go [TUG] test). Six measurement sessions were
performed: three baseline sessions before exercise sessions (PostL, PreB1 and PreB4) and then after a land-based session L and after one and four
weeks (B1 and B4) of balneotherapy at 34 8C (PostL, PostB1 and PostB4, respectively).
Results. A Friedman analysis of variance (ANOVA) revealed significant differences between the six conditions in terms of the TUG and VAS
results ( p < 0.01 and p < 0.05, respectively). The TUG time decreased after the balneotherapy session(s) ( p < 0.01, PreB1 versus PostB1 and
PreB4 versus PostB4). After sessions B1 and B4, a decrease in the VAS score was noted ( p < 0.05, PreB1 versus PostB1 and PreB4 versus PostB4).
In contrast, there were no significant postbalneotherapy changes in any of the postural parameters.
Conclusion. Balneotherapy sessions appear to induce a decrease in pain and an increase in dynamic mobility. This might be explained by
enhanced proprioceptive input during neuromuscular mobilization in water (i.e. with increased resistance but a lower load).
# 2007 Elsevier Masson SAS. All rights reserved.

Keywords: Spa therapy; Balneotherapy; Mobility; Pain; Posture

1. Introduction Balance impairments affect the elderly persons ability to

Ageing is associated with a progressive decrease in muscle
strength and joint flexibility [9]. These decreases may be
compounded by a decline in physical activity. In addition, there
is a progressive decline in visual perception, vestibular function
and somatosensory sensitivity [11,12]. A number of modifi-
cations in the postural system have already been reported
notably changes in the spatiotemporal sequence of the muscles
called into play as a reaction to a loss of balance and decreased
ability to appropriately select sensory information [20]. These
phenomena contribute to the impairment of balance and
mobility.
DOI of original article: 10.1016/j.annrmp.2007.10.006.
* Corresponding author.
E-mail address: laetitia.berger@univ-savoie.fr (L. Berger).
0168-6054/$ see front matter # 2007 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.annrmp.2007.10.007
live independently and comfortably. Consequently, effective
strategies for prevention of falls must be developed. Certain
physical activities (like Tai Chi Chuan [TCC] and aquatic
training) can promote participation by the elderly. Indeed, TCC
training programmes can improve balance and flexibility and
decrease the fear of falls in the over 60s [21]. Likewise, Chin
et al. [4] showed that TCC enhances the strength and endurance
of knee extensor muscles in this population.
Aquatic training (another sensorimotor practice) is opti-
mally performed in a reassuring and relaxing environment [14];
it induces facilitation of some movements [8] and a decrease in
pain [18]. After eight weeks of aquatic exercises, older adults
with arthritis showed improvements in functional physical
fitness and perceived ability to perform activities of daily living
[17]. Recent studies exploring the effects of spa therapy on
postural control have shown a decrease in postural sway (and
consequently, an increase in postural stability) in women with
lower extremity arthritis [16]. In addition, when comparing the
 L. Berger et al. / Annales de readaptation et de mdecine physique 51 (2008) 9095
immediate effects of standard physiotherapy and balneotherapy
on the postural performance of subjects with leg injuries, better
postural control was recorded solely for the balneotherapy
session [1]. A degree of benefit is observed after a short
(10 min) or long (120 min) hot bath, in both young and aged
subjects [2,3].
In order to investigate the possible effects of aquatic
training, we decided to measure the displacements of the centre
of pressure (COP), which correspond to the successive points of
application of the resulting ground reaction forces in an upright,
undisturbed stance. The choice of this sensorimotor task was
justified in that it involves only low-amplitude body movements
and that control of the latter requires only low-intensity forces.
More precisely, maintaining body sway requires body
movement control via activation of muscular reaction forces
through the points of contact with the ground. This means that
postural stability depends on muscle components and visual,
proprioceptive, tactile and vestibular sensory inputs. In a quasi-
static stance, vestibular information is not significant [19].
Although the elderly principally use visual input, propriocep-
tive and tactile inputs also contribute to balance control, and so
we reasoned that aquatic exercises in hot spa water might
influence the latter two inputs. Consequently, evaluation of
postural stability in our subjects was performed with the eyes
closed.
The aim of this study was to compare the following:
 the immediate effects of land-based (L) and aquatic (B)
exercises on mobility and postural control in the elderly;
 the effects of four weeks of aquatic exercises on these same
parameters.
We hypothesized that aquatic exercises might reinforce
sensory inputs due to an increased resistance from water.
If so, this approach could improve dynamic and/or static
balance and could be associated with a subjective pain
decrease. Furthermore, we hypothesized that regular
aquatic training might further enhance these effects. We
used a functional capacity test (the Timed Up and Go test)
and measurements of the posterior chain extensibility to
evaluate changes in dynamic and static balance. In addition,
subjects were asked to rate their pain on a visual analogue
scale.
2. Material and methods
2.1. Subjects
Our elderly subjects were all volunteers who had seen an
announcement in the Wellbeing Centre at the National
Thermal Baths in Aix-les-Bains (France) and were included
after checking that they satisfied the inclusion and
exclusion criteria. The volunteer subjects gave their informed
consent to participation in the experimental procedure, as
required by the Declaration of Helsinki (1964, last amended
in 2000).
The inclusion criteria were as follows:
91
 age over 60;
 participation at all test sessions (weeks one and four);
 performance of one aquatic training session in two and three.
The exclusion criteria were as follows:
 an inability to maintain an upright, quiet stance with the eyes
closed for at least 40 s and/or inability to undergo
balneotherapy;
 presentation with a neurodegenerative disease (Alzheimers
disease, Parkinsons disease, etc.) and/or a personal medical
history of cerebral vascular accidents;
 neuroleptic medication;
 a serious eye diseases;
 a hip replacement less than six months previously or a knee
replacement less than two years previously;
 bodyweight over 120 kg and/or particularly high-arched feet.
Twenty-three subjects (21 females and two males) took part
in the two first sessions but only 12 completed the whole
protocol. The 12 subjects (11 women, one man) had the
following characteristics: age: 65.3  6.6 years; body mass
index (BMI): 23.4  2.9; body weight: 62.8  11.4 kg; height:
163  6.8 cm (i.e. none with a BMI > 30).
2.2. Experimental procedure
All sessions were performed in the National Thermal Baths
(Aix-les-Bains, France). Balneotherapy sessions took place in a
swimming pool (1.3 m in depth) filled with hot spa water at
34 8C.
The order of the measurement sessions was chosen at
random for each subject and each type of measurement session.
Water (B1) and land (L) sessions were strictly identical, with at
least 48 h between sessions. Each session lasted 45 min and
was led by a physiotherapist (Table 1). Subjects then performed
out weekly sessions organized by Wellbeing Centre at the
National Thermal Baths. Over the following three weeks,
subjects participated in a weekly aquatic training session that
was identical to the first, i.e. up to the B4 session in the last
week (see Fig. 1). Clinical and postural tests were performed
before (Pre) and after (Post) the session in the first and last
weeks. Our subjects could not be considered to be sedentary,
since 10 performed regular physical activity (a minimum of 4 h
per week). Furthermore, all declared that they were not taking
any medication whatsoever.
2.3. Tests
2.3.1. Clinical tests
Dynamic balance was assessed with the Timed Up and
Go test. Balance quality can be evaluated by observing the
subjects performance of a complex motor task which demands
significant postural control, such as getting up out of the
armchair, turning round and sitting back down again [12].
Elderly people often display a loss of flexibility and joint
mobility due to a decrease in physical activity. In order to
92 L. Berger et al. / Annales de readaptation et de mdecine physique 51 (2008) 9095

Table 1
Activities performed during land-based and aquatic sessions
Description Exercises Number or time
Warm-up 5 min
Lower limbs exercises To go up on tiptoe 2  10
To raise tips of feet (in support on the heels) 2  10
Movement of the leg in AP axis 1  10
Movement of the leg in ML axis 2  10
Training two arms, two legs be fixed Movement of breast stroke 1  10
Arms abduction, flexion/extension of the elbows 2  10
Flexion to 908 shoulders, extension of the elbows 2  10
Flexion and horizontal adduction of shoulders, extension of the elbows 2  10
Training one leg (right then left) Hip Abduction 2  10
Elevation of the knee 2  10
Unipodal stance 2  10
Relaxation Quiet upright stance, eyes closed 2 min

evaluate these losses, we choose an overall measurement which
was simple and quick to perform: the extensibility of the
posterior chain by evaluating finger-ground distance at the time
of maximal anterior flexion.
Pain frequently limits mobility in the elderly and so it was
important for us to know whether the sessions alleviated or
triggered pain. This evaluation was performed using a visual
analogue scale (VAS). Lastly, we measured the heart rate in
order to evaluate the workload of the sessions.
2.3.2. Postural tasks
Subjects stood barefoot on two force platforms (Equi+,
model PF01) with feet abducted at 308, heels separated by 9 cm,
eyes closed. The signals from the load cells (on which the
platforms were laid) were amplified, converted from analogue
into digital form and then recorded on a computer. The centre of
pressures trajectory under the right (COPr) and left (COPl) feet
was then processed. For all subjects, a preliminary condition of
32 s was tested with eyes closed. Each experimental condition
included three 32 s sessions (sample frequency: 64 Hz) with an
intersession rest period of 32 s. During postural measurements,
subjects were instructed to minimize body movement as much
as possible.

2.3.3. Signal processing

Changes in the COP were determined by analysing the
coordinates of the COPr, COPl and the load distribution of the
feet:
   
COPr  Rr COPl  Rl
COP
Rr Rl Rr Rl

where Rr and Rl are vertical forces under the right and left feet,
respectively and (Rl/Rr + Rl) was the percentage load applied to
the right foot.
The surface area (mm2) covered by the COP trajectory and
the variance of this trajectory in the mediolateral (ML) and
anteroposterior (AP) axes were used to quantify postural sway.
The mean velocity of the COP (mm s1) was used to describe
postural behaviour; it indicates the mean velocity of the COP
over a given period and constitutes a valid index of the amount
of activity required to maintain stability.

2.4. Statistical analysis

In order to compare the different conditions within a group,

Fig. 1. Experimental procedure: the subjects were tested before (Pre) and after we used Friedmans ANOVA. Each parameter was computed
(Post) the sessions in the first and last weeks of the study. for the Pre- and Postconditions and then values were compared
 L. Berger et al. / Annales de readaptation et de mdecine physique 51 (2008) 9095 93

Table 2
Means and standard deviations of the COP parameters: area, mean velocity, variance in the mediolateral (ML) and anteroposterior (AP) axes for each condition in the
Pre- and Post-tests
Area (mm2) ML variance (mm) AP variance (mm) Mean velocity (mm s1)
PreL 86.8  92.6 4.2  3.2 18.4  10.8 9.9  6.2
PostL 119.2  105.9 4.9  5.1 16.8  9.7 9.4  5.6
PreL/PostL NS NS NS **p < 0.01
PreB1 93.1  40.3 2.3  1.3 17.7  8.2 9.9  4.2
PostB1 130.9  128.3 4.0  5.1 19.3  14 10.2  5.8
PreB1/PostB1 NS NS NS NS
PreB4 133.6  133 3.7  5 21.2  13.6 10.4  4.6
PostB4 116.6  78.7 3.4  2.3 18.5  12.8 9.8  5.3
PreB4/PostB4 NS NS NS NS
The statistically significant change in mean velocity (**p < 0.01).

using a paired, non-parametric Wilcoxon T-test. The signifi-
cance threshold was set to p < 0.05.
3. Results
3.1. Postural tests
The same percentage load was applied in all tests (51  2%
of bodyweight distribution on the right foot). No significant
effect of session and/or time was observed.
No significant intercondition effect was observed for the
areas covered by the COP or the variances in either axis
(Table 2).
In contrast, a significant effect of session was noticed for the
mean velocity of the COP ( p < 0.05). In particular, our results
showed that the velocity decreased after the on-land session
( p < 0.05, PreL versus PostL) (Table 2).
3.2. Clinical tests
A significant effect of session was noted for the dynamic
TUG test ( p < 0.01). This means that the time required to
complete the test appears to be lower after the B sessions
( p < 0.01 for PreB1/PostB1 and PreB4/PostB4). It did not
change after the L session (Table 3). Furthermore, this
intersession percentage reduction in test completion appeared
to be greater for the B4 sessions than for the L sessions
( p < 0.01, percentage PreLPostL compared with percentage
PreB4PostB4). However, no significant change between B1
and B4 was noticed for this parameter.
The ANOVA did not reveal any statistical differences
between the various sessions in terms of any of the extensibility
parameters (finger-floor distance). However, seven subjects
(out of 12) consistently presented a zero value for this test
(Table 3).
In addition, a significant difference in the pain VAS score
was observed (Table 3) ( p < 0.05). The results showed that the
mean VAS decreased at the end of session B ( p < 0.05 for
PreB1 versus PostB1 and PreB4 versus PostB4). This decrease
appeared clearly, despite the fact that four subjects stated that
they did not suffer from any pain before or after any of the
sessions.
As seen in Fig. 2, the heart rate data demonstrated that all the
physical training sessions had similar intensities (and thus, very
probably, similar equivalent workloads). The moderate increase
in the heart rate (12%) observed over the duration of a session
indicates that all were of low but constant intensity.

Table 3
Clinical tests (TUG test, pain VAS and extensibility) for all sessions in Pre- and
Postconditions
TUG (s) Extensibility (cm) VAS (cm)
PreL 9  2.4 3.2  6,1 2.8  3
PostL 8,3  2.5 3.6  8 3.1  3.3
PreL/PostL NS NS NS
PreB1 9.6  2.7 3.4  8.2 3  3.7
PostB1 8.4  2 2.9  6.4 1.2  2
PreB1/PostB1 **p < 0.01 NS *p < 0.05
PreB4 8.5  2.6 2.4  6 2.8  3.2
Fig. 2. Heart rate variations during the exercise sessions. Time (t) = 5 min: after
PostB4 7.7  2.8 2  5.4 1.8  2.7
the warm-up period; t = 20 min: after the leg exercises, t = 35 min: after arm
PreB4/PostB4 **p < 0.01 NS *p < 0.05
exercises and t = 45 min: the end of the session (relaxation period) (mean 
The statistically significant changes (*p < 0.05 and **p < 0.01). standard deviation).
94 L. Berger et al. / Annales de readaptation et de mdecine physique 51 (2008) 9095
4. Discussion
It appears that aquatic training in hot spa water can rapidly
improve dynamic balance and induce a decrease in subjective
pain. Furthermore, the effect on dynamic balance appears to be
greater for the last session in water, when compared with the
first training session in land. In contrast, postural control was
not modified by water immersion. However, in our population,
the mean velocity of the sway path after land-based exercises
decreased in a statistically significant manner. A previous study
[1] has reported the same effect in both mobilization conditions
(L and B). The decrease could be linked to a change in postural
control strategy.
Our study confirms previous research [15] showing that
mobility was more improved by aquatic training that by land-
based training. Mobilization in water corresponds to an
environment where constraints due to body weight are least,
the speed of movement is slower and the subjects body image
is probably better. When combined with muscle use, the bodys
lower load in water could reinforce proprioceptive input.
Equally, a warm bath has muscle-relaxant effects [10], which
could increase the contact area under the feet. Consequently,
plantar tactile inputs would be enhanced and would help
improve balance control.
We cannot rule out the possibility that changes in
neuromuscular function observed during aquatic exercises in
general (and especially the reduction in agonist activity
associated with an increased antagonist activity in the knee
flexion-extension exercises [13]) contribute to our results.
Likewise, immersion would also reduce compressive forces in
the joints [7] and thus encourage mobility. In addition, water
immersion induces peripheral vasodilatation that increases the
delivery of oxygen to the muscles and facilitates their activities
[8].
The sessions in hot spa water appears to induce a decrease in
subjective pain, even though the intensity of muscular effort
was equivalent (since the increases in heart rate were similar for
the three sessions). Consequently, this decrease in the VAS pain
score may be attributable to water immersion. This effect has
been observed and documented elsewhere, notably in arthritic
subjects [5,6]. Most of our subjects [7] showed excellent
extensibility at baseline, and so this may explain why we did not
see a statistically significant enhancement of this parameter in
the test group.
However, several limitations should be emphasized. Our
study population was small and predominantly female. Even
though this is the general case for aquatic exercise, it would be
interesting to include more males in the future. Our volunteers
were known to participate in the weekly sessions and all stated
that they were not taking any medication at the time of the
study each subject served as his/her own control. However, it
would be useful to perform studies included a control group
and/or a purely land-based group, although the heterogeneity
of the elderly population would mean that each group would
have to include far more subjects than studied here.
Another limitation concerns the subjects relatively high
level of physical activity. It would also be interesting to evaluate
aquatic exercise programmes in sedentary groups, older groups
and/or subjects with chronic disease, who might present even
greater improvements. Lastly, the relationship between the
effort intensity and any resulting physiological or biome-
chanical effects should be further explored.
5. Conclusion
The immediate and midterm impact of aquatic training was
evaluated in terms of mobility, extensibility, pain and postural
stability. Balneotherapy sessions appeared to improve
dynamic balance (in the TUG test) and decreased perceived
pain, whereas postural stability did not seem to be modified.
Muscle use in water (with lower load) may reinforce the
proprioceptive input and explain the improvement observed
here.
Acknowledgements
We thank for our volunteers for their participation, A.
Vallet for her contribution to the clinical evaluation and the
National Thermal Baths (Thermes Chevallet, Aix-les-Bains,
France) for making the hot spa swimming pool and facilities
available.
References
[1] Berger L, Martinie P, Livain T, Bergeau J, Rougier P. Effets immediats de
seances de reeducation des membres inferieurs par balneotherapie sur le
controle de lequilibre. Ann Readapt Med Phys 2006;49:3743.
[2] Berger L, Demanze-Bernard L. Effets immediats sur le controle postural
de limmersion et de la mobilisation en eau chaude. In: Lacour M,
Rougier P, editors. De Marey a` nos jours. Marseille: Solal; 2006 . p. 255
63.
[3] Effects immediats sur le controle postural de limmersion dans leau
chaude et dans une atmosphere chaude et humide. Berger L, Demanze-
Bernard L, Charles C, et al., editors. Controle postural et representations
spatiales. Marseille: Solal; 2007. p. 34553.
[4] Ching L, Jin-Shin L, Ssu-Yuan C, May-Kuen W. Tai Chi Chuan to improve
muscular strength and endurance in elderly individuals: a pilot study. Ann
Readapt Med Phys 2000;81:6047.
[5] Forestier R, Francon A, Saint-Arromand F, Bertolino C, Guillemot A,
Graber-Duvernay B, et al. Are SPA therapy and pulsed electromagnetic
field therapy effective for chronic neck pain? Randomised clinical
trial First part: clinical evaluation. Ann Readapt Med Phys 2007;50:
1407.
[6] Guillemin F, Constant F, Collin JF, Boulange M. Effectiveness of spa
therapy in chronic low back pain. Br J Rheumatol 1994;33:14851.
[7] Harrison R, Bulstrode R, Percentage S. Weight bearing during partial
immersion in the hydrotherapy pool. Physiother Pract 1987;3:603.
[8] Hall JB, OHare P. The physiology of immersion. Physiotherapy
1990;9(3):20615.
[9] Jette A, Branch L, Berlin J. Musculoskeletal impairments and physical
disablement among the aged. J Gerontol 1990;42:M2038.
[10] Kjellgren A, Sundequist U, Norlander T, Archer T. Effects of flotation
REST on muscle tension pain. Pain Res Manag 2001;6(4):1819.
[11] Manchester D, Woolacott M, Zederbauer-Hyton N, Marin O. Visual,
vestibular and somatosensory contributions to balance control in older
adults. J Gerontol 1989;44:M11827.
[12] Podsiadlo D, Richardson S. The timed Up & Go: a test of basic
functional mobility for frail elderly persons. J Am Geriatr Soc
1991;39(2):1428.
 L. Berger et al. / Annales de readaptation et de mdecine physique 51 (2008) 9095 95

[13] Poyhonen T, Kyrolainen H, Keskinen K, Hautala A, Savolainen J, Malkia [18] Tishler M, Rosenberg O, Levy O, Elias I, Amit-vazima M. The effect of
E. Electromyographic and kinematic analysis of therapeutic knee exer-
cises under water. Clin Biomech 2001;16:496504.
[14] Robiner W. Psychological and physical reactions to whirpool baths. Behav
Med 1990;13(2):15773.
[15] Simmons V, Hansen P. Effectiveness of water exercise on postural
mobility in the well elderly: an experimental study on balance enhance-
ment. J Gerontol 1996;51A(5):M2338.
[16] Suomi R, Koceja D. Postural sway characteristics in women with lower
extremity arthritis before and after aquatic exercices intervention. Arch
Phys Med Rehabil 2000;81:7805.
[17] Suomi R, Collier D. Effects of arthritis exercise programs on functional
fitness and perceived activities of daily living measures in older adults
with arthritis. Arch Phys Med Rehabil 2003;84:158994.
balneotherapy on osteoarthritis. Is an intermittent regimen effective. Eur J
Intern Med 2004;15:936.
[19] Winter DA, Patla AE, Prince F, Ishac M, Gielo-Perczak K.
Stiffness control of balance in quiet standin. J Neurophysiol 1998;
80:121121.
[20] Woolacott M, Shumway-cook A, Nashner L. Aging and posture control:
changes in sensory organization and muscular coordination. Int J Aging
Hum Dev 1986;23:97113.
[21] Zhang J, Ishikawa-Takata K, Yamazaki H, Morita T, Ohta T. The effects of
Tai Chi Chuan on physiological function and fear of falling in the less
robust elderly: an intervention study for preventing falls. Arch Phys Med
Rehabil 2006;42:10716.