European Journal of Obstetrics & Gynecology and Reproductive Biology 147 (2009) 25–28

Contents lists available at ScienceDirect

European Journal of Obstetrics & Gynecology and

Reproductive Biology
journal homepage: www.elsevier.com/locate/ejogrb

Postural sway changes during pregnancy: A descriptive study using stabilometry§

Liliam F. Oliveira a,*, Taian M.M. Vieira b, Adriana R. Macedo a, David M. Simpson c, Jurandir Nadal d
a
Sport and Physical Education School, Federal University of Rio de Janeiro, 21941-590 Rio de Janeiro, Brazil
b
Laboratorio di Ingegneria del Sistema Neuromuscolare, Dipartimento di Elettronica, Politecnico di Torino, Italy
c
Institute of Sound and Vibration Research, University of Southampton, UK
d
Biomedical Engineering Program, Federal University of Rio de Janeiro, Brazil

A R T I C L E I N F O A B S T R A C T

Article history: Objective: This study aims to analyse changes in body sway over the course of pregnancy.
Received 17 September 2008 Study design: This is a descriptive study in which stabilometric tests were applied at three stages of
Received in revised form 29 May 2009 pregnancy and with a combination of different visual conditions (eyes open/closed) and support base
Accepted 30 June 2009
configuration (feet together/apart). Twenty healthy pregnant women participated in the study. Changes
in postural control with pregnancy were analysed via the elliptical area of the stabilograms and spectral
Keywords: analysis of the displacements of the centre of pressure (COP) along the lateral and anterior/posterior
Stabilometry
directions.
Pregnancy
Power-spectral analysis
Results: The elliptical area encompassing the COP significantly increased over the course of the
Sway pregnancy for the feet apart and eyes closed test protocols. The spectral analysis revealed a significant
Balance increase of COP oscillations along the anterior–posterior direction when subjects stood with the eyes
open/feet together and feet apart. A reduction (significant) of the lateral oscillations of COP was observed
for the eyes open/feet together protocol.
Conclusion: Pregnancy induced significant changes in the postural control when pregnant women stood
with a reduced support base or with eyes closed.
ß 2009 Elsevier Ireland Ltd. All rights reserved.

1. Introduction
From a biomechanical perspective, progressive alterations in
body shape, weight distribution and thus the average position of
body centre of gravity (COG) during the course of the gestational
period are documented [1,2]. The upright [3–5] and sitting
postures [3,6] as well as joint moments and foot plantar pressure
[7,8] during pregnancy are topics discussed in the literature.
Nevertheless, the adaptations in the postural control at upright
stance emerging throughout pregnancy have been marginally
investigated so far [9,10].
Human postural control during quiet standing involves the
integration of sensory information from body periphery, in
particular from mechano receptors on the foot soles [11], and
from specialised receptors coding body position and orientation
with respect to the gravitational acceleration, the environment and
the body segments [12]. Such sensory information is coordinated
with precise modulations in ankle torque, possibly through fine
adjustments in the length of plantar flexors [13], to compensate for
§
This study was conducted in Rio de Janeiro, Brazil.
* Corresponding author. Tel.: +55 21 2562 6824; fax: +55 21 2562 6826.
E-mail addresses: liliam@eefd.ufrj.br, liliam.oliva@gmail.com (L.F. Oliveira).
the continuous and spontaneous sways of the body during quiet
standing. In addition, passive elements (e.g. muscle–tendon
connective tissues) seem to provide substantial contribution for
the tonic ankle stiffness [14].
While the postural control mechanisms seem to be unaffected
during pregnancy, the increased and asymmetric distribution of
body mass and the posterior tilt observed over the pregnancy time
course [7,8] could play an important role in modulating body
sways amplitude and frequency, reflecting specific strategies for
maintaining upright standing posture. On one hand, the increased
mass in the anterior pelvic region in pregnant women would likely
be compensated for with the increase of tonic activity of ankle
plantar flexors and with the augment of ankle stiffness, under the
inverted pendulum framework [15,16]. On the other hand, the
time constant of body sways would also increase (e.g. the duration
of body sways increases – [17]). Notwithstanding the larger degree
of ankle tonic activity, women in the later stages of pregnancy
would possibly compensate for the body sways with modulations
of ankle torque at lower frequencies.
Stabilometry is a reliable method to quantify the position of
body centre of pressure (COP – defined as the coordinates of the
resultant force applied through the feet on the force-plate). Even
though this method has been extensively used with normal
subjects and patients, establishing normality ranges for classic

0301-2115/$ – see front matter ß 2009 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.ejogrb.2009.06.027
26 L.F. Oliveira et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 147 (2009) 25–28
stabilometric descriptors [18–20], only two studies seem to have
focused on the assessment of changes in postural sways with
pregnancy. By using temporal parameters of the stabilometric
signal to assess the postural control, Butler et al. [9] observed a
decline in balancing ability with pregnancy, which persisted for 6–
8 weeks after delivery, and an increased reliance on visual cues to
maintain balance during pregnancy. However, the effects of
support base configuration on the stabilometric descriptors were
not accounted for by these authors. Recently, Jang et al. [10]
analysed the postural sways of pregnant women along both the
medio-lateral (ML) and anterior–posterior (AP) directions, report-
ing an increased postural instability in the sagittal plane over the
course of pregnancy. The lack of changes in COP sways along ML
direction was attributable to the self-selection of support base
configuration [10]. After all, the identification of postural strategies
based on the interpretation of stabilometric data may be biased
when anthropometric and biomechanical effects on stabilometric
descriptors are overlooked [21,22].
The present study aims to detect and analyse possible changes
in body sways, in both time and frequency domains, over the
course of pregnancy in a sample of young and healthy women,
accounting for possible effects of reducing the support base and
suppressing the visual inputs. The following research questions are
addressed in this study: (1) Are there changes in body sway during
pregnancy when subjects stand with feet comfortably apart? (2)
What are the effects induced by the reduction of support base and
the suppression of visual input on the postural control with the
course of pregnancy? The rationale for applying eyes open/closed
and feet apart/together protocols is to investigate how the postural
control of pregnant women adapts in response to the exposition to
hazardous situations (e.g. standing in environments with scarce
luminosity or standing on unstable supporting surface) throughout
pregnancy.
2. Materials and methods
Twenty healthy pregnant women (28.7  6.2 years,
158.2  5.6 cm height) participated in this descriptive study. The
subjects were attending pre-natal clinics at the Institute of Child
Health of the Federal University of Rio de Janeiro, did not develop
complications during their pregnancies and did not report a history of
neurological or orthopaedic pathology. All subjects volunteered and
gave informed consent for the study.
The experimental protocol consisted in the application of 30
stabilometric tests in each of the following standing conditions: (1)
eyes open with feet comfortably apart (EO/FA); (2) eyes closed
with feet comfortably apart (EC/FA); (3) eyes open with feet
together (EO/FT); and finally (4) eyes closed with feet together (EC/
FT). Two-minute rest periods were interposed between each test.
Stabilometric exams were applied during the first (to be referred to
as G1), second (G2) and third (G3) trimesters of pregnancy.
Subjects were allowed to choose their preferred foot position over
the force-plate during the protocols with feet comfortably apart,
and the size of their support base, defined as the distance between
the mid-points of the long axis of each foot (from hallux to the
middle of the heel), was measured. For the three exams, Table 1
shows mean and standard deviation of the parameters body mass,
abdominal girth and the size of the support base over the course of
pregnancy.
The deflections of the centre of pressure (COP) along the lateral
(x) and anterior/posterior (y) axes were recorded (sampling rate:
50 Hz) with a custom force platform, developed in accordance with
the specifications provided by the French Association of Postur-
ology [23] and Bizzo et al. [24]. The area (area) encompassed by the
COP was analytically evaluated as an ellipse (ca. 85% of COP
samples are within the ellipse), whose principal axes were
Table 1
Description of the sample (mean  standard deviation, N = 20 subjects) over the three
trimesters of pregnancy.
Trimester G1 G2 G3
Time of exam (week into pregnancy) 15.1  1.8 24.0  2.4 34.5  2.5
Mass (kg) 59.8  8.8 64.9  8.2 69.1  9.1*
Abdominal girth (cm) 86.1  9.0 95.4  6.7 102.2  5.3**
Size of support base (cm) 14.2  3.4 13.5  2.6 13.9  2.1
Statistically significant differences among the three trimesters *p = 0.01, **p = 0.00.
estimated through PCA – Principal Component Analysis [18,25].
Spectral analysis was applied for each stabilometric test in both x
and y direction by using Burg’s autoregressive method [19,26],
with order 100 and decimation to 10 Hz. The total power (TP) was
calculated separately for each direction, AP and ML, to obtain a
global measure of COP sways, which corresponds to its variance.
Only the frequencies encompassed in the band 0–2 Hz were
included in TP calculation, since very little activity was observed
above this limit.
To investigate the variations in area and TP with the gestational
period, a 3  2  2 multivariate analysis of variance (MANOVA)
design was applied (gestational trimesters  eyes open/clo-
sed  feet apart/together). Orthogonal contrasts were used for
post hoc pair-wise comparisons and results were considered
significant at p < 5%.
3. Results
The relative gain in body mass from the first to the third
trimester was 16.2  7.9% with an increase of abdominal girth of
15.4  8.6% (mean  standard deviation). In contrast, the size of
preferred configuration of the support base did not change
significantly with pregnancy (Table 1).
Although the elliptic area of COP sways progressively increased
with pregnancy, except for the eyes open and feet apart condition,
it reached statistical significance only in the second (p = 0.018) and
third (p = 0.003) trimesters, in comparison to the early stage of
pregnancy (Fig. 1).
Significant changes in the postural control with development of
pregnancy were observed in EC/FA and EO/FT conditions. The total
power of COP sways along ML direction significantly increased
when pregnant women stood with feet apart and eyes closed
during the second and third gestational periods, with respect to the
same standing protocol applied within the first 3 months of
pregnancy (Fig. 2a). Interestingly, TP in the frontal plane decreased
Fig. 1. Mean area of the stabilograms (in mm2) and standard errors (error bars) over
the three trimesters of pregnancy (G1, G2 and G3) for each of the four test protocols.
The brackets indicate statistically significant change over the course of pregnancy.
EO = eyes open; EC = eyes closed; FA = feet apart; FT = feet together.
 L.F. Oliveira et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 147 (2009) 25–28
Fig. 2. Total power (mean and standard error) in the medial/lateral (a) and anterior/
posterior (b) directions over the three trimesters of pregnancy for each of the four
test protocols. The brackets indicate differences between trimesters. EO = eyes
open; EC = eyes closed; FA = feet apart; FT = feet together.
for all standing conditions when comparing the third with the
second trimester of gestation (Fig. 2a). Such decrease is significant
only for EO/FT condition (p = 0.009). This same trend was not
observed for TP descriptor along the AP direction, which increased
during the last 3 months of gestation for all conditions, being
significant only when standing with eyes closed/feet apart and
with eyes open/feet together (p = 0.017 and 0.013, respectively).
In general, the stabilometric descriptors used in this study were
sensitive to pregnancy only in eyes closed/feet apart and eyes
open/feet together protocols.
4. Comment
Postural equilibrium during pregnancy has not been exten-
sively explored [9,10], in particular regarding the application of
different protocols. Nevertheless, understanding the adaptations
in the postural control of pregnant women throughout their
gestation, with different combinations of available sensory inputs
and the support base configurations, could provide remarkable
insights for the development and application of specific ther-
apeutic and preventive approaches.
5. Effect of visual input with pregnancy
According to sensory organisation theory, as firstly reported by
Nasher [27] and further developed by Kawakami et al. [15], the
central nervous system (CNS) possibly codes redundant informa-
tion provided by visual, vestibular and somatosensory receptors.
Albeit the effects of absence of visual input on the control of
upright stance in normal subjects are debatable [28], the outcomes
of this study indicate that pregnant subjects tend to compensate
27
for the lack of vision by increasing the frequency of COP sways
along both AP and ML directions throughout pregnancy, when
standing with the feet comfortably apart (Fig. 2a and b – EC/FA
protocol). The increase in COP area reported in Fig. 1 during
pregnancy, for this protocol, is in agreement with the results
obtained with a smaller sample of 12 subjects investigated by
Butler et al. [9]. By comparing pregnant women with a control
group, these authors observed a significant increase in the COP
average radial displacement for the first, second and third
trimester of gestation, when subjects stood with eyes closed.
Such an increase was not significant for the first trimester of
gestation, with the eyes open. Although Butler et al. [9] did not
explicitly describe the position of the feet, the configuration of the
support base applied by these authors seems to be equivalent to
our feet apart protocol (see their Fig. 1). Our data thus confirm their
findings and indicate that subjects rely heavily on the visual input
for balance control during later stages of pregnancy. Curiously, the
elliptic area did not increase (G1 = G3; p = 0.448) with gestation
when subjects stood without visual input and with the feet
together (Fig. 1, EC/FT protocol), possibly because the effect
induced by the smaller base of support masked the influence of
pregnancy time course on the size of COP sways (Fig. 1).
6. Variations of COP total power with gestation
Considering the inverted pendulum assumption [16,17] and
given the increase in mass distribution with pregnancy, pre-
dominantly along the anterior–posterior direction [1,2,4,5,7], it
would not seem surprising that such an increased moment of
inertia of the body, if not compensated for, could lead to larger
amplitude of COP sways, likely resulting in greater muscle activity.
Interestingly, pregnant women seem to have been unable to
compensate for the inhomogeneous increase in body mass only
with the suppression of visual input or the reduction of the support
base in the late stages of pregnancy (Fig. 2). In all protocols, the
third trimester showed the highest power for COP sways
frequencies along the AP direction (though differences only
reached significance in EC/FA and EO/FT). It is worth mentioning
that the total power descriptor relates to the global amplitude of
COP sways in the time domain, separately for both AP and ML
direction [29]. In contrast, Jang et al. [10] observed a similar
increase in COP sways when analysing a group of 15 pregnant
women during the last months of pregnancy with eyes open and
feet apart. Whether the divergence between studies results from
methodological issues or from the sample recruited remains
unknown.
On the other hand, the decrease in the total power of COP
medial–lateral displacements, observed in the third trimester for
the eyes open/feet together protocol, was less intuitive. Given that
the stabilisation of quiet standing posture depends on precise and
fine modulations in ankle torque [13], the changes in COP position
is linked to changes in ankle torque [11,17] and the ankle stiffness
is affected by the frequency of ankle torque [30], one may argue
that increases in COP frequencies in the sagittal plane with
pregnancy may be related to increased ankle stiffness. The
reduction of the support base in the frontal plane would likely
elicit the participation of other muscles, like those around the hip
[16,17], rather than the plantar flexors. Consequently, subjects
would shift from ankle to hip strategy for stabilisation of their
quiet standing posture, with smaller movements of their body.
In general, the postural control of pregnant women significantly
changed throughout pregnancy only when the visual input was
suppressed or the support base was reduced. Therefore, the
accuracy of visual information and the comfortable support base
could be critical for preserving balancing ability in the later stages
of pregnancy. These factors must be considered when applying
28 L.F. Oliveira et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 147 (2009) 25–28
clinical postural tests to pregnant women. Exercises focusing on
the variation of the size of support base and the application of
different visual stimuli could possibly posit an effective interven-
tion to minimise the effects of gestation on the postural control
during quiet standing. After all, whether the changes in postural
control induced by gestation could be associated with the
development of undesired side effects, like back pain, adapted
gait pattern and eventually risk of falls, should be the focus of
forthcoming investigations.
Acknowledgements
This work was supported by the Brazilian Ministry of Science
and Technology (CNPq and PRONEX) and José Bonifácio Founda-
tion (FUJB).
References
[1] Jensen RK, Doucet S, Treitz T. Changes in segment mass and mass distribution
during pregnancy. J Biomech 1996;29:251–6.
[2] Fries EC, Hellebrandt FA. The influence of pregnancy on the location of the
center of gravity, postural stability, and body alignment. Am J Obstet Gynecol
1943;46:374–80.
[3] Gilleard WL, Crosbie J, Smith R. Static trunk posture in sitting and standing
during pregnancy and early postpartum. Arch Phys Med Rehabil
2002;83:1739–44.
[4] Dumas GA, Heid JG, Wolfe LA, Griffin MP, McGrath MJ. Exercise, posture and
back pain during pregnancy. Part 1. Exercise and posture. Clin Biomech
1995;10:104–9.
[5] Nicholls JA, Grieve DW. Posture, performance and discomfort in pregnancy.
Appl Ergon 1992;23:128–32.
[6] Lee YH, Her LL, Tsuang YH. A comparison of sitting posture adaptations of
pregnant and non-pregnant females. Int J Indust Ergon 1999;23:391–6.
[7] Paul JA, Sallé H, Frings-Dresen MHW. Effect of posture on hip joint moment
during pregnancy, while performing a standing task. Clin Biomech
1996;11:111–5.
[8] Nyska M, Sofer D, Porat A, Howard CB, Levi A, Meizner I. Plantar foot pressures
in pregnant women. Isr J Med Sci 1997;33:139–46.
[9] Butler EB, Cólon I, Druzin ML, Rose J. Postural equilibrium during pregnancy:
decreased stability with a increased reliance on visual cues. Am J Obstet
Gynecol 2006;195:1104–8.
[10] Jang J, Hsiao KT, Hsiao-Wecksler ET. Balance (perceived and actual) and
preferred stance width during pregnancy. Clin Biomech 2008;23:468–76.
[11] Morasso PG, Schieppati M. Can muscle stiffness alone stabilize upright stand-
ing? J Neurophysiol 1999;82:1622–6.
[12] Horak FB. Postural control. In: Binder MD, Hirokawa N, Windhorst U, editors.
Encyclopedia of neuroscience. Berlin, Heidelberg: Springer-Verlag; 2009.
[13] Loram ID, Maganaris CM, Lakie M. Human postural sway results from frequent,
ballistic bias impulses by soleus and gastrocnemius. J Physiol 2005;564:
295–311.
[14] Loram ID, Lakie M. Direct measurement of human ankle stiffness during quiet
standing: the intrinsic mechanical stiffness is insufficient for stability. J Physiol
2002;545:1041–53.
[15] Kawakami O, Sudoh H, Koike Y. Control of upright standing posture during
low-frequency linear oscillation. Neurosci Res 1998;30:333–42.
[16] Winter DA, Patla EA, Prince F, Ishac M, Gielo-Perczak K. Stiffness control of
balance in quiet standing. J Neurophysiol 1998;80:1211–21.
[17] Loram ID, Gawthrop PJ, Lakie M. The frequency of human, manual adjustments
in balancing an inverted pendulum is constrained by intrinsic physiological
factors. J Physiol 2006;577:417–32.
[18] Oliveira LF, Simpsom DM, Nadal J. Calculation of area of stabilometric
signals using principal components analysis. Physiol Meas 1996;17:
305–12.
[19] Oliveira LF, Simpsom DM, Nadal J. Autoregressive spectral analysis of stabilo-
metric signals. In: Proceedings of the 16th Ann. Int. Conf. IEEE Engineering in
Medicine and Biology; 1994.
[20] Mello RGT, Oliveira LF, Nadal J. Anticipation mechanism in body sway
control and the influence of muscle fatigue. J Electromyogr Kinesiol 2007;
17:739–46.
[21] Kirby R, Price NA, Macleod DA. The influence of foot position on standing
balance. J Biomech 1987;20:423–7.
[22] Chiari L, Rocchi L, Cappello A. Stabilometric parameters are affected by
anthropometry and foot placement. Clin Biomech 2002;17:666–7.
[23] French Association of Posturology, Normes 85 – Editées par l’’Association
Française de Posturologie. 4. Avenue de Corbéra, 75012 Paris, France; 1985.
[24] Bizzo G, Guillet N, Patat A, Gagey PM. Specifications for building a vertical force
platform designed for clinical stabilometry. Med Biol Eng Comput
1985;23:474–6.
[25] Morrison DF. Multivariate statistical methods, 2nd ed., Auckland: McGraw-
Hill; 1978.
[26] Bräuer D, Seidel H. The autoregressive structure of postural sway. Agressol
1980;21E:101–4.
[27] Nasher LM. Sensory, neuromuscular and biomechanical contributions to
human balance. In: Proceedings of the APTA Forum; 1989. p. 4–12.
[28] Vieira TMM, Oliveira LF, Nadal J. An overview of age-related changes in
postural control during quiet standing tasks using classical and modern
stabilometric descriptors. J Electromyogr Kinesiol 2008. doi: 10.1016/j.jelekin.
2008.10.007.
[29] Shiavi R. Introduction to applied statistical signal analysis, 2nd ed., San Diego,
CA: Academic Press; 1999.
[30] Loram ID, Kelly S, Lakie M. Human balancing of an inverted pendulum: is sway
size controlled by ankle impedance? J Physiol 2001;532:879–91.