[Backward disequilibrium in the elderly: review of symptoms and proposition of a tool for quantitative assessment]
Purpose: Although there is currently no epidemiological data on backward disequilibrium, this disturbance of posture does not seem to be rare in frail elderly. Actualities and strong points: Backward disequilibrium is characterized by the following criteria: the location of buttocks on the anterior side of the seat while the trunk rested at the back of the armchair in the sitting position; an inadequate forward of the trunk and a backward projection of the trunk outside the base of support during sit-to-stand; and a posterior projection of the center of mass outside the base of support in the standing position. Several pathological situations either somatic (degenerative, ischemic and traumatic brain lesions), psycho-somatic (psychomotor disadaptation syndrome, extended bed confining, non-use) or psychological (depression) affections can entail backward disequilibrium. Falls, loss of autonomy and the risk of the vicious circle with its causes are the main consequences of backward disequilibrium. Prospects and projects: Although the geriatrician is familiarized with backward disequilibrium, there is no scale to quantify it. In this paper we review causes, consequences and management of backward disequilibrium, and in order to assess it, we propose a semi-quantitative scale, based on some activities of everyday living which are sitting position, sit-to-stand, back-to-sit and standing position. So, a backward disequilibrium score could be determined.
Is backward disequilibrium in the elderly caused
by an abnormal perception of verticality? A pilot study
, France Mourey
, Pierre Pﬁtzenmeyer
Jacques Van Hoecke
, Dominic Pe
INSERM/ERIT-M 0207 Motricite
de Bourgogne, Dijon, France
Service de Me
decine Interne Ge
riatrie CHU Dijon, France
Service de Re
ducation Neurologique, CHU Dijon, France
Accepted 18 November 2006
Objective: We hypothesised that backward disequilibrium (BD), deﬁned by a posterior position of the centre of mass with respect to the
base of support, could be caused by a backward tilt in the perception of verticality.
Methods: The relationship between BD, the perception of verticality, and the history of falls in 25 subjects aged 84.5 ± 7.4 years was
analysed. An original ordinal scale, the BD scale (BDS), was used to quantify BD. Postural (PV) and haptic verticals (HV) were mea-
sured in sagittal plane.
Results: BDS scores closely correlated with the number of falls (r = 0.81, p =10
). The more the PV was tilted backward, the greater
the BDS scores (r = 0.95, p <10
), with a huge backward tilt of about 15 in 4 subjects with severe BD. In these subjects, the tilt in
perception of verticality was transmodal since a severe backward HV tilt was also found.
Conclusions: This transmodality suggested high-order cognitive disruption in the construction of the subjective vertical used in postural
control by subjects showing BD, which conﬁrmed our hypothesis.
Signiﬁcance: This study clearly shows that perception and action with respect to gravity are closely related and brings a new insight
about fall mechanisms in the elderly.
2006 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Keywords: Backward disequilibrium; Elderly; The perception of verticality; Subjective vertical
A high prevalence of falls, evaluated at 50% after 80–85
years (Tinetti and Williams, 1997), is a major public health
problem because of the social and economic repercussions
on health and morbi-mortality associated with falls (Tinetti
et al., 1994). Approximately 10% of falls engender physical
or psychological consequences (Nevitt et al., 1989; Sattin,
1992; Tinetti et al., 1995; Moreland et al., 2003). Most
mechanisms put forward to explain falls, particularly in
elderly people, are based on defective postural and balance
controls. These approaches mainly concern body stabilisa-
tion, coordination between posture and motion, attention
demand for performing multiple tasks, lessening of muscu-
lar strength, sensory deﬁcits (visual, vestibular, somaesthet-
ic) or cerebral integration of a range of information
(Brocklehurst et al., 1982; Aniansson et al., 1986; Woolla-
cott, 1993; Lopez et al., 1997; Choy et al., 2003). However,
these a pproaches do not explain all of the mechanisms of
falls, and despite the numerous postural tests available to
evaluate the risk of falling, these tests cannot accurately
predict this risk (Mathias et al., 1986; Tinetti, 1986; Wolf-
son et al., 1990; Anacker and Di Fabio, 1992; Berg et al.,
1992; Lundin-Olsson et al., 1997). Falls might also be
1388-2457/$32.00 2006 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Corresponding author. Tel.: +33 3 80 29 39 70; fax: +33 3 80 29 36 21.
E-mail address: firstname.lastname@example.org (P. Manckoun-
Clinical Neurophysiology 118 (2007) 786–793
caused by an abnormal perception/construction of vertical-
ity, a very undersearched avenue in the literature, regarding
postural disorders in the elderly (Lord and Webster, 1990).
Somebody’s perception of verticality or subjective vertical
corresponds to his/her own speciﬁc representation of the
vertical. It can be measured in the frontal or sagittal planes
according to diﬀerent sensory modalities: haptic vertical
(HV), postural vertical (PV) and visual vertical (VV) (Wit-
kin and Asch, 1948; Mann et al., 1949; Bauermeister et al.,
1964; Tobis et al., 1981). The subjective vertical is built up
in the parieto-insular cortex (Brandt et al., 1994), mainl y in
the right hemisphere (Kerkhoﬀ and Zoelch, 1998; Pe
et al., 1998). Vestibular, somesthetic and visceral gravicep-
tors contrib ute to the updating of the subjective vertical
(Mittelstaedt, 1992). Golgi tendon organs might monitor
the forces exerted by the muscles to compete against grav-
ity (Dietz et al., 1992 ). Several authors have shown abnor-
mal perception of verticality in patients with a brain lesion
(Brandt et al., 1994; Bisdorﬀ et al., 1996; Kerkhoﬀ, 1999;
rennou et al., 2000; Yelnik et al., 2002; Saj et al.,
2005). In this kind of patient, Bonan et al. (2006) have
recently reported a relationship between a lateral tilt in
the VV and postural disorders. In the elderly, the relation-
ship between the history of falls and abnormal perception
of verticality was pointed out 20 years ago by Tobis et al.
(1981, 1985) who found that errors in the visual perception
of verticality were signiﬁcantly greater in fallers than in
non-fallers. It was also shown that the visual perception
of the vertical is more inﬂuenced by the visual surroundings
in elderly fallers than in non-fallers (Tobis et al., 1985;
Lord and Webster, 1990). These three studies analysed
the visual perception of the vertical in the frontal plane.
To our knowledge, the perception of verticality in the
elderly has never been analysed in the sagittal plane and
the PV has never been investigated in elderly subjects show-
ing spontaneous backward disequilibrium (BD). Deﬁned
by a posterior position of the centre of mass (CoM) with
respect to the base of support, spontaneous BD is abno r-
mal postural behaviour, usually characterized by a poster-
ior trunk tilt in standing and sitting positions, which
exposes subjects to backward falls. While rising from a
chair, subjects with BD move their CoM insuﬃciently for-
ward to be able to stand up without the risk of falling
(Mourey et al., 2004). Little is known about this disorder,
especially regarding its prevalence and the underlying
mechanisms; this is partly due to the absence of an appro-
priate scale to assess spontaneous BD. So far, most clinical
tools proposed to quantify BD focused either on BD in
reaction to posterior destabilisation (Wolfson et al., 1986;
Fahn et al., 1987; Duncan et al., 1990 ) or on the limit of
stability of subjects required to lean backward (Newton,
2001). In this study, we were more interested in the detec-
tion and the quantiﬁcation of spontaneous BD in the elder-
ly and designed a novel clinical scale with this aim in mind.
From a theoretical point of view, two explanations may
be put forward to explain the mechan isms underlyin g spon-
taneous BD. First, BD could be caused by an exaggeration
of the postural tone of extensor muscles due to a disorder
in the excitability of the corresponding descending path-
ways (Porter and Lemon, 1993 ). Such a disorder is diﬃcult
to quantify. The second explanation is less reﬂexive and
more cognitive. Indeed, it is possible to hypothesise that
people with BD could implicitly align their erect posture
with a backward-tilted reference vertical (biological verti-
cal). This is the hypothesis we tested in the present study.
Given the frequency of falls and their somatic, psycho-
logical and economic consequences, it was also interesting
to analyse the possible relationship between the existence
of spontaneous BD behaviour and the history of falls in
In accordance with the inclusion and exclusion criteria,
we enrolled 25 subjects. Seventeen people, aged from 74 to
94 years (mean 86.3 ± 6.3), were recrui ted in a geriatric
falls clinic and 8 healthy control subjects, aged from 70
to 91 years (mean 80.6 ± 8.3), were recruited among their
relatives. All these control subjects had a normal gait.
Elderly people admitted to the falls clinic were mostly
referred by their general practitioner for the assessment
and treatment of balance disorders. The unique inclusion
criterion was age 70 and older. All subjects were clinically
examined by a senior physician. A Mini Mental State
Examination score (MMSE) (Folstein et al., 1975) was sys-
tematically performed in each subject and the medical his-
tory was collected. Extrapyramidal (tremor, akinesia,
hypertonia and freezing), pyramidal (Babinski sign) and
vestibular symptoms (vertigo, nystagmus, lateral falling
and gait lateral deviation) were systemat ically sought, but
not objectively quantiﬁed. A simple examination of their
vision including visual ﬁeld and acuity was performed. A
simple osteoarticular examination was also performed in
order to check the absence of severe skeleton deformation
or joint stiﬀness. The muscul ar strength of the main mus-
cles (4 limbs) was manually tested according to the British
Medical Council (Kendall and Kendall-Mc Creary, 1983).
To estimate the tone and the stiﬀness of calf muscles, the
amplitude of the passive dorsal ﬂexion of the ankle was
quantiﬁed, the subject being laid on the back the knees in
extension. The amplitude of ﬂexion over 90 was measured
with a goniometer (accuracy 1) at the right and the left
sides, and the average value considered for the statistical
analysis. This measurement was only performed in 18 sub-
jects (1 control, and 17 patients with balance disorders). As
has been previously proposed, the somatosensory threshold
was assessed by investigation of pressure sensitivity using
rennou et al.,
1998, 2001a). In the absence of any attested neurological
disease (see exclusion criteria), we considered that sensitiv-
ities of the two sides of the body were similar, and pressure
sensitivity was tested only at the pulp of the right big toe.
P. Manckoundia et al. / Clinical Neurophysiology 118 (2007) 786–793 787
As recommended for this technique (Semmes et al., 1960),
we used a set of 20 calibrated nylon ﬁlaments equal in
length but varying in diameter, each implanted at one
end of a plastic rod. The force applied to the skin needed
for a given subject to perceive the stimulus was subjected
to a Log transform with a view to obtaining a 20-point lin-
ear scale (values increasing with hypoesthesia).
Subjects with an attested neurological, vestibular, ortho-
paedic, psychiatric disease and subjects with a MMSE <24,
a severe skeleton deformation, a joint stiﬀness or a muscu-
lar deﬁcit were not included in this study, nor were those
who had a severe visual or auditory impairment. Subjects
who took any psychotropic medication that would have
altered the study were also excluded. In the 25 subjects
(10 females and 15 males) included in this study, the mean
age was 84.5 ± 7.4 years (minimum 70 and maximum 94)
and the muscular stre ngth was always 5/5. The level of
hypoesthesia was 4.15 (minimum 2.83 and maximum
6.45), which corresponded to a mean score of 8 on a 20-
point linear scale ranged from 1 (excellent pressure detec-
tion) to 20 (anesthesia). The mean amplitude of the passive
dorsal ﬂexion of the ankle was 4.2 (minimum 0 and maxi-
mum 12.5). Clinically, subjects with BD had a tendency to
fall backward or presented adaptive patterns characterized
by forward trunk bending, knee ﬂexion and toe clenchi ng
in the standing position, and a backward position of the
trunk in the sitting position. Prior to the experiment, all
subjects had given their informed consent. The study was
approved by the local Ethics Committee.
2.2. Assessment of falls
The number of falls during the previous six months was
assessed by a structured interview of the subject and rela-
tives and by asking the general practitioner. A fall was
deﬁned as an unintentional event that results in a person
coming to rest on the ground, ﬂoor, or other lower level
(Campbell et al., 1997). Subjects were classiﬁed into 3
groups: those who did not report any falls (non-fallers),
those who reported 1 or 2 falls (occasional-fallers), and
those who reported more than 2 falls (iterative or recur-
rent-fallers). This study was not an epidemiological study
designed to explain the risk of falling in elderly people
and we did not analyse fall circumstances.
2.3. Assessment of backward disequilibrium
We designed a speciﬁc assessment tool, namely the BD
scale (BDS), whose purpose was to diﬀerentiate BD, a spe-
ciﬁc postural behaviour, from postural imbalance. We set
up an ordinal scale based on the following 5 items which
corresponded to the main daily postural tasks: maintaining
the sitting position, maintaining the standing position feet
together, with open and then closed eyes, sit-to-stand,
and stand-to-sit. This scale assessed the severity of the
BD, not trunk orientation. Indeed, even in the presence
of severe BD, some subjects show no backward tilt of the
trunk because of adaptive patterns: knee ﬂexion, forward
trunk bending, toe clenching. Each subject was instructed
to perform each task 3 times while being observed by one
investigator familiarised with this assessment, and who
did not know the history of falls or the measur ements of
the perception of verticality. The subject was required to
perform the tasks 3 times with a view to improving the scal-
ing accuracy. The investigator determined the score for
each task after the subjects had completed the 3 trials. Each
item was scored from 0 to 3. Zero means no BD i.e. a con-
stantly upright trunk. One means slight BD with no diﬃ-
culty to perform the task. Two means moderate and/or
intermittent BD with diﬃculty to perform the task;
although the subject could perform the task without help,
there was a risk of falling due to the backward body tilt;
and 3 means severe BD compromising the task. The BDS
total score ranged from 0 to 15. A score 62 indicated that
the subject could show slight BD only in one or two postur-
al tasks. The dynamic postural task i.e. sit-to-stand and
stand-to-sit, however, was generally performed without
any diﬃculty. Consequently, normality was deﬁned as a
BDS score 62. A BDS score P3 indicated either that a
slight BD was found for the majority of items, or that
the subject showed a moderate and/or intermittent BD
with diﬃculty in performing the task in at least 1 item. A
BDS score P8 indicated that the subject showed a majority
of items scored 2 (moderate and/or intermittent BD with
diﬃculties in performing the task) or a combination of 2
and 3 (severe BD compromising the task). Therefore sub-
jects with a BDS score 62 were considered as normal, those
with a BDS score between 3 and 7 were considered as hav-
ing moderate BD, and those whose BDS score over 7 were
considered as having severe BD.
In order to test BDS sensitivity and speciﬁcity to detect
recurrent-fallers, subjects were classiﬁed either BD – (BDS
scores 0, 1 or 2) or BD + (BDS scores P3). The BDS intra-
judge repeatability was analys ed through a preliminary
assessment of 9 subjects tested 2 days apart: Spearman cor-
relation coeﬃcient = 0.99 (p <10
). The BDS inter-judge
repeatability was also high as attested by a preliminary
assessment in 8 subjects tested the same day by 2 indepen-
dent investigators: Spearman correlation coeﬃcient
between BDS total scores = 0.89 (p = 0.003).
2.4. Perception of verticality
The perception of verticality was measured by one of the
authors who did not know the history of falls or the BDS
score. We measured both PV and HV in a completely dark
room. The subjects were positioned in an apparatus, eyes
closed and entirely masked to avoid any bias during the
reading of the inclinometer. The apparatus used was
derived from that designed by Pe
rennou et al. to measure
the lateral tilt of PV in stroke patients (Pe
rennou et al.,
rennou and Bronstein, 2005). The subjects were
strapped in the sitting position to a framework within the
180 cm diameter · 80 cm deep drum constructed from
788 P. Manckoundia et al. / Clinical Neurophysiology 118 (2007) 786–793
welded steel tubes, with head, trunk, thighs and legs
restrained by webbing and pads and aligned upright, and
the feet strapped on a plane support. The rotation axis of
the drum was approximately at the level of the navel
(Fig. 1a). Measurement of the PV and HV was carried
out according to a random order after several practice tri-
als to familiarise the subject with the procedure. No feed-
back was given to subjects about their performance
before the whole assessment was completed.
For the PV, the subjects were randomly tilted to a given
position (backward or forward at 10,15,20,25 or 30),
then the wheel was manual ly rolled in the opposite direc-
tion, as steadily as possible at an approximate velocity of
1.5 per second, until subjects reported reaching an upright
position. Small adjustments around this position were then
performed if needed until the subjects were satisﬁed that
they were perfectly vertical. Ten unpredictable trials were
performed, 5 from front to back (starting position: 10,
15,20,25 or 30), and 5 from back to front (starting posi-
tion: 10, 15, 20, 25 or 30). Body inclination for
each trial was measured using an inclinometer (accuracy
0.5), and the PV obtained by averaging the 10 values.
For the HV, a rod (25 cm long) pivoting around a central
axis was manually set to vertical. The device was centred on
the trunk at a height adjusted for each sub ject. The orienta-
tion of the rod was pseudo-randomly tilted (balanced
between forward and backward) and subjects instructed
to manually set the rod to vertical using the dominant hand
(Fig. 1b). Since there were no visual clues, a tactile explora-
tion of the rod was encouraged. Ten trials were performed,
their average giving the HV (accuracy 0.5).
A positive angle of the PV and HV corresponded to a
forward tilt and a negative angle to a backward tilt. The
repeatability of PV and HV measurements was assessed
by calculating the angular dispersion (standard deviation)
of the 10 trials perfor med by each subject.
2.5. Statistical analysis
First, the 17 subject s recruited from the geriatric falls
clinic were compared to the 8 healthy subjects as concerns
their mean age using Wilcoxon Rank test.
After distribution of subjects into three groups according
to their BDS scores, the mean of the following parameters
was calculated for each group: age, number of falls, PV
and HV. We also determined the sex ratio for each group.
Diﬀerences between the groups of subjects were tested
using non-parametric tests and multiple comparison post
hoc tests (Bonferroni test) and ANCOVA when required.
Regarding angular values, diﬀerences from the vertical
(zero) were tested using sample t-tests. The possible rela-
tionships between BDS score and: the PV, HV, the number
of falls, the amplitude of the passive dorsal ﬂexion of the
ankle, the level of hypoesthesia, and age were analysed with
a Spearman test, followed by a multiple regression in order
to master possible confounding factors between these vari-
ables. Groups were compared with regard to the sex-ratio
using the v
test. Data are presented as means ± standard
deviation. Sensitivity and speciﬁcity of BDS to detect
recurrent-fallers were also calculated.
3.1. Age comparison between subjects recruited from the fall
clinic and control subjects
As concerns mean age, the stat istical analysis did not
ﬁnd a signiﬁcant diﬀerence between the 17 subjects recruit-
Fig. 1. Apparatus to measure the two modalities of biological vertical, i.e. postural vertical (a) and haptic vertical (b), in the sagittal plane.
P. Manckoundia et al. / Clinical Neurophysiology 118 (2007) 786–793 789
ed from the geriatric falls clinic and the 8 healthy subjects
recruited in their relatives (p = 0.1).
3.2. Scores on the backward disequilibrium scale
The BDS scores were well distributed along the whole
scale from 0 to 15, which was a ﬁrst indication that this
scale has satisfactory clinimetric properties (Fig. 2a).
According to BDS scores, 16 subjects showed no BD,
5 moderate BD, and 4 severe BD. The 4 subjects show-
ing severe BD were not able to stand and to walk with-
out help at the time of the examination. Age, sex ratio,
and the number of falls are given in Table 1 for each
group. Subjects without BD were younger than subjects
with either moderate or severe BD; there was no diﬀer-
ence with regard to age for the latter. The BDS scores
correlated with age (r = 0.52, p = 0.0077). Since females
were older than males (88.3 ± 6.1 vs. 82 ± 7.3 years;
p = 0.03), females and males were compared using
ANCOVA with age as a covariable. No diﬀerence was
found as concerns BDS scores (p = 0.16). There was no
correlation between BDS scores and the amplitude of
the passive dorsal ﬂexion of the ankle (r = 0.08,
p = 0.74) while the level of hypoesthesia correlated with
BDS scores (r = 0.59, p = 0.003). The deeper the hypoes-
thesia, the more severe BD.
Among the 25 sub jects, there were 9 recurrent-fallers (3–
6 falls during the previous six months, mean = 4 falls), 8
occasional-fallers and 8 non-fallers. All 4 subjects who
showed severe BD on BDS were recurrent-fallers. Among
subjects who showed moderate BD there were 3 recur-
rent-fallers and 2 occasional-fallers. Finally, there were 2
recurrent-fallers, 6 occasional-fallers and 8 non-fallers
among subjects who showed no BD.
Subjects without BD according to the BDS fell less often
than those with moderate or severe BD (p = 0.007) (Table
1). A close correlatio n was found between BDS scores and
the number of falls (r = 0.81, p =10
The sensitivity and the speciﬁcity of BDS to detect
recurrent-fallers were, respectively, 0.78 and 0.87.
3.4. Perception of verticality
Four su bjects were not able to perform the HV task,
whereas they performed the PV task without any diﬃculty.
The repeatability of PV and HV measurements was satis-
factory, as attested by the low angular dispersion, on aver-
age in the 25 (for PV) or 21 (for HV) subjects: 3.9 and 4.9
for the PV and HV, respectively.
Distributions of the PV and HV are given in Figs.
2b and c. Interestingly PV values were signiﬁcantly
negative (3.7 ±4.1, T = 4.5, p = 0.00007) whereas
HV values did not diﬀer from zero (0.05 ± 0.04,
T =0.07, p = 0.95). Thus, the PV was more often tilt-
ed backward than was the HV (p = 0.000 3). Thes e
ﬁndings indicate that postural perception of the verti-
cal was either normal or tilted backward whereas their
tactile perception of the vertical was distributed
The perception of verticality according to BD severity
on the BDS is analysed in Table 1. A gradient in the back-
ward tilt was found for the PV, with a huge backward tilt in
subjects with severe BD: the PV correlates closely with
BDS scores (r = 0.95, p <10
)(Figs. 2a, b and 3). In
contrast, for the HV, only subjects with severe BD dis-
played a clear backward tilt and no correlation was found
between BDS scores and the HV (r = 0.40, p = 0.074)
(Figs. 2a and c).
The level of hypoesthesia correlated with the PV
(r = 0.40, p = 0.05).
Fig. 2. Backward disequilibrium scale (BDS) scores (a, 25 subjects),
postural vertical (PV) (b, 25 subjects) and haptic vertical (HV) (c, 21
subjects). On (a), lines indicate limits between groups of subjects: No
backward disequilibrium (BD), moderate BD and severe BD. A positive
angle of the PV and HV corresponded to a forward tilt and a negative
angle to a backward tilt. (s) identiﬁes non-fallers, (j) occasional-fallers
and (m) recurrent-fallers.
790 P. Manckoundia et al. / Clinical Neurophysiology 118 (2007) 786–793
The purpose of this pilot study was to understand BD.
We used two speciﬁc measur ement tools which are suited
to the clinical context: an ordinal scale, namely BDS,
which quantiﬁes BD, and an apparatus dedicated to the
measurement of the PV in the sagittal plane in elderly
people. The whole assessment for each subject did not last
more than 30 min. The distribution of BDS scores was
satisfactory. This scale was convenient, fast and easy to
use, and above all relevant since it closely correlated with
the PV, a biological measurement of the perception of
verticality. In addition, its inter-judge repeatability
checked in a low number of subjects was high. Together
with the fact that this ﬁrst study with few subjects showed
satisfactory BDS sensitivity and speciﬁcity to detect a risk
of falling, these ﬁndings suggest that the BDS would be
an interesting assessment tool that should be validated
further by studies which analyse BDS intra and inter-
judge repeat ability, internal consistency, sensitivity, and
predictive validity. In our study, PV measurement was
more feasible than HV, since all subjects completed the
PV task whereas 4 subjects were not able to perform
the HV task, maybe because the HV is not a common
way to estimate the vertical in everyday life. Performing
the HV task might involve a representation of the direc-
tion of the vertical in which the hands set up the pivoting
rod. This repres entation might induce an additional diﬃ-
culty in some elderly subjects who are not able to imagine
(represent) the direction of the rod in space, or to imagine
(represent) the direction of the vertical, although they
seemed to understand the instruction and had a
MMSE P 24. On the contrary, the PV is typically percep-
tive, that is to say easier to understand and execu te, thus
ﬁnally more feasible and relevant than the HV to detect
abnormal perception of verticality as a possible cause of
postural disorders in elderly people.
The present study has some limitations. The relatively
low number of subjects does not allow us to draw any
conclusion on BD frequency. Although spontaneous BD,
also termed retropulsion or backward sway, is reported
as a common disorder observed in clinical practice (Pﬁtzen-
-16 -14 -12 -10 -8 -6 -4 -2 0 2
Postural Vertical (˚)
Fig. 3. Representation of the scores of BDS according to the postural vertical of each subject. Lines indicate limits between groups of subjects: No
backward disequilibrium (BD), moderate BD and severe BD. (s) identiﬁes non-fallers, (j) occasional-fallers and (m) recurrent-fallers.
Comparison between the 3 groups regarding age, sex ratio, the number of falls during the previous 6 months, postural vertical (PV) and haptic vertical
Variable No BD
(n = 16)
Age (years) 81.7 ± 1.6
88.2 ± 2.9 91.2 ± 3.3 0.023
Sex ratio F/M 4/12 3/2 3/1 0.09
Number of falls 1 ± 0.4
3.4 ± 0.6 3.5 ± 0.7 0.007
PV () 1.27 ± 0.41
5.42 ± 0.74
11.45 ± 0.82 0.0002
HV () 0.72 ± 0.97
1.95 ± 1.89
5.6 ± 1.89 0.036
Data presented are means ± SD (except sex ratio). The p-values correspond to the comparisons of the 3 groups, either by Kruskal–Wallis one-way
ANOVAs on Ranks for continue variables, or by a v
test for gender.
Signiﬁcant diﬀerence between no BD and moderate BD.
Signiﬁcant diﬀerence between no BD and severe BD.
Signiﬁcant diﬀerence between moderate BD and severe BD. BD, Backward disequilibrium.
P. Manckoundia et al. / Clinical Neurophysiology 118 (2007) 786–793 791
meyer et al., 1999; Visser et al., 2003; Dimitrova et al.,
2004; Mourey et al., 2004; Brondel et al., 2005), its preva-
lence remains to be determined. The relationship between
BD according to BDS scores, the perception of verticality
and the number of falls was obtained from interviews
about falls history. This retrospective analysis is less valid
than a prospective survey. This study was not an epidemi-
ological study designed to explain the risk of falling in
elderly people and we did not analyse fall circumstances.
However, we hypothesised that the existence of spontane-
ous BD could mechanically increase loss of balance and
therefore falling, and thus compared fall history for
subjects with and without spontaneous BD. We found that
all subjects with severe BD on the BDS were recurrent-
fallers. This means that the existence of severe BD compro-
mises pos tural stability in daily life and suggests that
spontaneous BD could be an independent risk of falling
in elderly peo ple. This could be investigated further by epi-
demiological studies. Finally, although the assessment
tools were novel and their clinimetric properties not yet
extensively analysed, the present data have provided initial
results about their validity. It is unlikely that the main ﬁnd-
ings revealed in this paper could result from a bias in the
The close correlatio n between BDS scores and the
backward tilt of the PV (see Fig. 3) demonstrates that
BD can have a cognitive origin, and conﬁrms that sub-
jects with BD aligned their erect posture with an errone-
ous referential of verticality. The fact that subjects
presenting severe BD on the BDS sho wed a backward
tilt of the two modalities of the biological vertical (we
tested both the PV and HV) strengthened this cognitive
interpretation (see Fig. 2). Only the PV was altered in
subjects with moderate BD, which indicates that the
trouble predominated on the somaesthetic contribution
to graviception. A correlation between BDS scores and
the degree of the pressure hypoesthesia on the one hand,
and between the PV and the degree of the pressure hypo-
esthesia on the other hand was found in the present
study. This strongly suggests that the backward tilt in
perception of the PV and thus BD could be caused by
some degree of somatosensory deaﬀerentation. Whether,
this alteration in the somesthetic gravitoception may be
itself at the origin of the trouble or alternatively just
facilitate the clinical manifestation of a more pronounced
dysfunction bearing on the vestibular graviception that
would no longer be compensated will have to be further
investigated in the future. Both vestibular and somaes-
thetic explorations, which we did not do in the present
study, will be required. Interestingly, while the HV was
distributed around the physical vertical, the PV was
always close to the physical vertical or tilted backward
with a continuum between normality and an extreme
backward tilt (10–15). This tendency to perceive the
PV backward even in subjects without BD on the BDS
will have to be further investigated in the future, in order
to better understand whether or not it is an adaptation
mechanism due to ageing. These further studies will also
have to give normative data for BDS, the PV and HV in
The present study brings a new insight into BD. In our
study, it was unlikely that backward tilt of the PV resulted
from the history of falls itself. In contrast, alterations in the
perception of verticality may explain BD. This cogn itive
origin of BD is not in opposition with the clinical data in
the literature about the phobia of the upright position
and the fear of the void in front of the subject linked to
the fear of falling in the context of post-fall syndrome for
example (Murphy and Isaacs, 1982; Pﬁtzenmeyer et al.,
1999; Mourey et al., 2004). Moreover, our study does not
allow us to draw co nclusions about other causes of BD
such as pathological hypertonia of extensor muscles. How-
ever, our study was a pilot study with relatively few sub-
jects and we cannot exclude the possibility that there are
several mechanisms underlying BD. Interestingly, the fear
of falling may concern people with no history of falls (Leg-
ters, 2002). This is another argument for a possible causal
link between BD and falls, and not always the revers e.
Although we did not analyse fear of falling, our data pre-
dict that a person, who perceives his or her body upright
when tilted 15 backward, would feel uncomfortable when
he or she is forced to be 15 more forward. Could this part-
ly cause the fear of falling?
This study was supported by ‘‘La Fondation de l’Ave-
nir’’. We are grateful to M. D Antoine, M. Ph Bastable
and Mrs. F. Panne tier-Manckoundia.
Anacker SL, Di Fabio RP. Inﬂuence of sensory inputs on standing
balance in community-dwelling elders with a recent history of falling.
Phys Ther 1992;72:575–81.
Aniansson A, Hedberg H, Henning GB, Grimby G. Muscle morphology,
enzymatic activity, and muscle strength in elderly men: a follow-up
study. Muscle Nerve 1986;9:585–91.
Bauermeister M, Werner H, Wapner S. The eﬀect of body tilt on tactual-
kinesthetic perception of verticality. Am J Psychol 1964;77:451–6.
Berg KO, Wood-Dauphinee SL, Williams JI, Maki B. Measuring balance
in the elderly: validation of an instrument. Can J Public Health
Bisdorﬀ AR, Wolsley CJ, Anastasopoulos D, Bronstein AM, Gresty MA.
The perception of body verticality (subjective postural vertical) in
peripheral and central vestibular disorders. Brain 1996;119:1523–34.
Bonan IV, Guettard E, Leman MC, Colle MF, Yelnik AP. Subjective
visual vertical perception relates to balance in acute stroke. Arch Phys
Med Rehabil 2006;87:642–6.
Brandt T, Dieterich M, Daneck A. Vestibular cortex lesions aﬀect the
perception of verticality. Ann Neurol 1994;35:403–12.
Brocklehurst JC, Robertson D, James-Groom P. Clinical correlates of
sway in old age-sensory modalities. Age Age 1982;11:1–10.
Brondel L, Mourey F, Mischis-Troussard C, d’Athis P, Pitzenmeyer P.
Energy cost and cardiorespiratory adaptation in the ‘‘Get-Up-and-
Go’’ test in frail elderly women with postural abnormalities and in
controls. J Gerontol A Biol Sci Med Sci A 2005;60:M98–M103.
792 P. Manckoundia et al. / Clinical Neurophysiology 118 (2007) 786–793
Campbell AJ, Robertson MC, Gardner MM, Norton RN, Tilyard MW,
Buchner DM. Randomised controlled trial of a general practice
pro-gramme of home based exercise to prevent falls in elderly women.
Brit Med J 1997;315:1065–9.
Choy NL, Brauer S, Nitz J. Changes in postural stability in women aged
20 to 80 years. J Gerontol A Biol Sci Med Sci A 2003;58:M525–30.
Dietz V, Gollhofer A, Kleiber M, Trippel M. Regulation of bipedal
stance: dependency on ‘‘load’’ receptors. Exp Brain Res 1992;89:
Dimitrova D, Horak FB, Nutt JG. Postural muscle responses to
multidirectional translations in patients with Parkinson’s disease. J
Duncan PW, Studenski S, Chandler J, Bloomfeld R, LaPointe LK.
Electromyographic analysis of postural adjustments in two methods of
balance testing. Phys Ther 1990;70:88–96.
Fahn S, Elton RL, members of the UPDRS. Development Committee
Uniﬁed Parkinson’s Disease Rating Scale. In: Fahn S, Marsden CD,
Goldstein M, Clane DB, editors. Recent developments in Parkinson’s
disease. Florham Park (NJ): Macmillan Healthcare Information,
Folstein MF, Folstein SE, McHugh PR. ‘‘Mini-mental state’’. A practical
method for grading the cognitive state of patients for the clinician. J
Psychiatr Res 1975;12:189–98.
Kendall FP, Kendall-Mc Creary E. Muscles. Testing and function. Bal-
timore: The Williams & Wilkins Company; 1983.
Kerkhoﬀ G. Multimodal spatial orientation deﬁcits in left-sided visual
neglect. Neuropsychologia 1999;37:1387–405.
Kerkhoﬀ G, Zoelch C. Disorders of visuospatial orientation in the frontal
plane in patients with visual neglect following right or left parietal
lesions. Exp Brain Res 1998;122:108–20.
Legters K. Fear of falling. Phys Ther 2002;82:264–72.
Lopez I, Honrubia V, Baloh RW. Aging and the human vestibular
nucleus. J Vest Res 1997;7:77–85.
Lord SR, Webster IW. Visual ﬁeld dependence in elderly fallers and non-
fallers. Int J Aging Hum Dev 1990;31:267–77.
Lundin-Olsson L, Nyberg L, Gustafson Y. ‘‘Stops walking when talking’’
as a predictor of falls in elderly people. Lancet 1997;349:617.
Mann C, Berthelot-Berry N, Dauterive H. The perception of the vertical:
I. Visual and non-labyrinthine cues. J Exp Psychol 1949;39:
Mathias S, Nayak US, Isaacs B. Balance in elderly patients: the ‘‘get-up
and go’’ test. Arch Phys Med Rehabil 1986;67:387–9.
Mittelstaedt H. Somatic versus vestibular gravity reception in man. Ann N
Y Acad Sci 1992;656:124–39.
Moreland J, Richardson J, Chan DH, O’Neill J, Bellissimo A, Grum RM,
et al. Evidence-based guidelines for the secondary prevention of falls
in older adults. Gerontology 2003;49:93–116.
Mourey F, Manckoundia P, Martin-Arveux I, Tavernier-Vidal B,
Pﬁtzenmeyer P. Psychomotor disadaptation syndrome. A new clinical
entity in geriatric patients. Geriatrics 2004;59:20–4.
Murphy J, Isaacs B. The post-fall syndrome. A study of 36 elderly
patients. Gerontology 1982;28:265–70.
Nevitt MC, Cummings SR, Kidd S, Black D. Risk factors for recurrent
nonsyncopal falls. A prospective study. JAMA 1989;261:2663–8.
Newton RA. Validity of the multi-directional reach test: a practical
measure for limits of stability in older adults. J Gerontol A Biol Sci
Med Sci A 2001;56:M248–52.
rennou D, Bronstein A. Balance disorders and vertigo after stroke:
assessment and rehabilitation. In: Bogousslavsky J, Barnes M, Dobkin
B, editors. Recovery After Stroke. Cambridge: Cambridge University
Press; 2005. p. 320–98.
rennou DA, Amblard B, Leblond C, Pelissier J. Biased postural vertical
in humans with hemispheric cerebral lesions. Neurosci Lett
rennou DA, Leblond C, Amblard B, Micallef JP, Rouget E, Pelissier J.
The polymodal sensory cortex is crucial for controlling lateral postural
stability: evidence from stroke patients. Brain Res Bull
rennou DA, Leblond C, Amblard B, Micallef JP, Herisson C, Pelissier
JY. Transcutaneous electric nerve stimulation reduces neglect-related
postural instability after stroke. Arch Phys Med Rehabil
rennou D, Playford D, Guerraz M, Mazibrada G, Gresty M, Bronstein
AM. Dissociation in the verticality perception after a stroke. In:
Duysens J, Smits-Engelsman B, Kingma H, editors. Control of Posture
and Gait. Amsterdam: Elsevier; 2001b. p. 686–8.
Pﬁtzenmeyer P, Mourey F, Tavernier B, Camus A. Psychomotor
desadaptation syndrome. Arch Gerontol Geriatr 1999;28:217–25.
Porter R, Lemon R. Corticospinal Function and Voluntary Move-
ment. Oxford: Clarendon Press; 1993.
Saj A, Honore
J, Bernati T, Coello Y, Rousseaux M. Subjective visual
vertical in pitch and roll in right hemispheric stroke. Stroke
Sattin RW. Falls among older persons: a public health perspective. Annu
Rev Public Health 1992;13:489–508.
Semmes J, Weinstein S, Ghent L, Teuber H. Somatosensory Changes after
Penetrating Brain Wounds in Man. Cambridge: Harvard University
Tinetti ME. Performance-oriented assessment of mobility problems in
elderly patients. J Am Geriatr Soc 1986;34:119–26.
Tinetti ME, Williams CS. Falls, injuries, and the risk of admission to a
nursing home. N Engl J Med 1997;337:1279–84.
Tinetti ME, Baker DI, McAvay G, Claus EB, Garrett P, Gottschalk M,
et al. A multifactorial intervention to reduce the risk of falling
among elderly people living in the community. N Engl J Med
Tinetti M, Doucette J, Claus E, Marottoli R. Risk factors for serious
injury during falls by older persons in the community. J Am Geriatr
Tobis JS, Nayak L, Hoehler F. Visual perception of verticality and
horizontality among elderly fallers. Arch Phys Med Rehabil
Tobis JS, Reinsch S, Swanson JM, Byrd M, Scharf T. Visual perception
dominance of fallers among community-dwelling older adults. J Am
Geriatr Soc 1985;33:330–3.
Visser M, Marinus J, Bloem BR, Kisjes H, van den Berg BM, van
Hilten JJ. Clinical tests for the evaluation of postural instability in
patients with Parkinson’s disease. Arch Phys Med Rehabil
Witkin HA, Asch SE. Studies in space orientation. III. Perception of the
upright in the absence of a visual ﬁeld. J Exp Psychol 1948;38:603–14.
Wolfson LI, Whipple R, Amerman P, Kleinberg A. Stressing the postural
response. A quantitative method for testing balance. J Am Geriatr Soc
Wolfson L, Whipple R, Amerman P, Tobin JN. Gait assessment in the
elderly: a gait abnormality rating scale and its relation to falls. J
Woollacott, M.H. Age-related changes in posture and movement. J
Gerontol 1993;48 (Spec. No.): pp. 56–60.
Yelnik AP, Lebreton FO, Bonan IV, Colle FM, Meurin FA, Guichard JP,
et al. Perception of verticality after recent cerebral hemispheric stroke.
P. Manckoundia et al. / Clinical Neurophysiology 118 (2007) 786–793 793