Neuroscience Letters 366 (2004) 63–66

Proprioceptive weighting changes in persons with low back

pain and elderly persons during upright standing
Simon Brumagne a,∗ , Paul Cordo b , Sabine Verschueren c
a Department of Rehabilitation Sciences, Faculty of Physical Education and Physiotherapy, K.U. Leuven, Tervuursevest 101, B-3001 Leuven, Belgium
b Neurological Sciences Institute, Oregon Health and Science University, West Campus, 505 N.W. 185th Avenue, Beaverton, OR 97006, USA
c Department of Kinesiology, Faculty of Physical Education and Physiotherapy, K.U. Leuven, Leuven, Belgium

Received 21 January 2004; received in revised form 8 April 2004; accepted 8 May 2004

Abstract

The purpose of this study was to examine whether postural instability observed in persons with spinal pain and in elderly persons is due to
changes in proprioception and postural control strategy. The upright posture of 20 young and 20 elderly persons, with and without spinal pain,
was challenged by vibrating ankle muscles (i.e. tibialis anterior, triceps surae) or paraspinal muscles. Center of pressure displacement was
recorded using a force plate. Persons with spinal pain were more sensitive to triceps surae vibration and less sensitive to paraspinal vibration
than persons without spinal pain. Elderly persons were more sensitive to tibialis anterior vibration than young healthy persons. These results
suggest that spinal pain and aging may lead to changes in postural control by refocusing proprioceptive sensitivity from the trunk to the ankles.
© 2004 Elsevier Ireland Ltd. All rights reserved.

Keywords: Postural control; Proprioception; Vibration; Low back pain; Aging; Inverted pendulum; Somatosensory integration

Postural stability is achieved, in part, through input from
the proprioceptive, visual, and vestibular systems, which
the central nervous system (CNS) must weight relative to
one another depending on the immediate conditions. An im-
portant property of the postural control system is its ability
to gate sensory input in accordance with the internal repre-
sentation of the current posture, so as to avoid undesirable
responses triggered by external or internal perturbations [8].
In some circumstances, it might be advantageous for the
CNS to be able to reweight sensory input based on location
of origin, rather than modality. For example, if the quality
of input from a particular body location decreases due to
injury, disease, or normal aging, the CNS might increase
the weighting of input from other locations that provide in-
formation useful for maintaining a stable posture. Without
spatial reweighting of sensory input, reliance on impover-
ished input from a particular body location might lead to
loss of balance and falling.
Proprioceptive input from the muscles of the legs and
trunk plays an important role in maintaining postural stabil-
∗ Corresponding author. Tel.: +32 16 329121; fax: +32 16 329197.
E-mail address: simon.brumagne@flok.kuleuven.ac.be (S. Brumagne).
ity [1], suggesting that sensory deficits from either location
might result in instability. Postural instability has been ob-
served in patients with low back pain [11] as well as in
elderly persons [12]. Patients with low back pain have also
been attributed reduced lumbosacral proprioception [3],
which might be a causative factor in their instability.
The study described in this paper investigated whether
the weighting of proprioceptive input in the back and an-
kles changes in persons with low back pain and persons
who are elderly. Tendon and muscle vibration was used as
an experimental probe to quantify the weighting of propri-
oceptive input from the areas of the lumbar back and the
ankles, at least as far as it influences stability during quiet
standing. Vibration is a potent stimulus for muscle spindle
Ia afferents [2], and applied to the leg muscles during quiet
standing, it can induce postural instability, sometimes to
the point of falling (e.g., [5]). Mechanical vibrations were
applied, during quiet standing, to the triceps surae, tibialis
anterior, or lumbar paraspinal muscles in young and elderly
persons with low back pain (LBP) and age-matched healthy
persons during quiet standing.
Twenty young adults (10 healthy, 10 LBP; mean age
= 25 years) and 20 elderly persons (10 healthy, 10 LBP;
mean age = 63 years) participated in this study. The patients

0304-3940/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.neulet.2004.05.013
64 S. Brumagne et al. / Neuroscience Letters 366 (2004) 63–66

were recruited from the Department of Physical Medicine

and Rehabilitation, University Hospitals of Leuven. Most
of these patients did not have a more specific diagnosis than
mechanical LBP. They had an average score of 20/100 on the
Oswestry Disability Index, version 2 [6]. A 6-channel force
plate (Bertec, OH, USA) recorded the moments of force
around the frontal (Mx ) and sagittal (My ) axes and the ver-
tical ground reaction force (Fz ). Force plate data were sam-
pled at 500 Hz using a Micro1401 data acquisition system
and Spike2 software (Cambridge Electronic Design, UK)
and low pass filtered (fc = 5 Hz). The subjects stood bare-
foot on the force plate with the feet separated by the width of
the hips, the arms hanging loosely at the sides. The subjects
Fig. 1. Anteroposterior sways for the four conditions and subgroups. Mean
were instructed to remain still and relaxed. Each subject was RMS values and standard deviations.
tested under four experimental conditions: (1) control (no
vibration); (2) bilateral vibration of the triceps surae ten-
dons; (3) bilateral vibration of the tibialis anterior tendons; larger than that in the young individuals (4.8 ± 2.8 cm)
and (4) bilateral vibration of the paraspinal muscle bellies. (F(1, 39) = 6.9, P < 0.05). In addition, the CoP displace-
In all four conditions, vision was occluded by means of ment in the persons with low back pain (7.0 ± 3.1 cm)
liquid-crystal goggles (Translucent Technologies, Canada). was significantly larger than that in the healthy individuals
Trials lasted 60 s. Tendon and muscle vibration (60 pps, (5.0 ± 3.0 cm) (F(1, 39) = 7.22, P < 0.05). When vibra-
≈0.5 mm) started 15 s after the start of the trial and lasted tion was applied bilaterally to the paraspinal muscles, the
for 15 s. Displacement of the center of pressure (CoP) was persons with low back pain and the elderly produced little
estimated from the force plate data using the equation: CoP or no displacement of the CoP, whereas the CoP of healthy
= Mx /Fz , and the CoP was averaged over the 15 s periods individuals were displaced anteriorly by an average of 3.5
prior to and during vibration to quantify the directional effect ± 2.4 cm (F(1, 39) = 4.23, P < 0.05) (see Fig. 2). In in-
of vibration. In control trials, the root mean square (RMS) dividuals with low back pain, both young and elderly, and
value of the CoP was also calculated to quantify the baseline in the healthy elderly as well, these differences in CoP dis-
variability in each group of subjects. Postural recovery times placement produced by vibration may reflect a refocusing of
were determined as the total time from cessation of vibra- proprioceptive control of balance away from proximal and
tion until CoP displacement of the subject was again within axial proprioception input to that derived from receptors in
the limits of two standard deviations of the mean position, the ankle muscles.
that was determined over the 15 s period prior to vibration. Following the cessation of vibration, the vibration-evoked
During all conditions, minor force variations were measured displacement of the CoP diminished over a period of
in the frontal plane. Hence, for clarity, the analysis will only time, although the length of time depended on the subject
be carried out in the sagittal plane and no data relative to group. For example, following the cessation of the triceps
the frontal plane will be given. Differences in CoP measure- surae vibration, the persons with low back pain had signif-
ments between control conditions and vibration conditions, icantly longer postural recovery times (16.2 ± 8.6 s) than
and between patient group and healthy control group, and the healthy individuals (7.9 ± 4.4 s) (F(1, 79) = 15.55, P <
between the young and elderly persons were based on F-test 0.001). When age is taken into account, the young healthy
analysis of variance, using one-way and two-way proce- persons recovered their postural stability significantly faster
dures with repeated measures on one factor (Statistica, OK,
USA). The significance level was set at P < 0.05.
In the absence of vibration, the RMS CoP displace-
ment in the anterior–posterior direction was quite small
(0.5–1.0 cm), but significantly greater in the persons
with low back pain compared to the healthy individuals
(F(1, 39) = 4.33, P < 0.05) (see Fig. 1).
When vibration was applied bilaterally to the triceps
surae, all subjects shifted the CoP posteriorly (see Fig. 2),
but much larger sways were seen in the persons with low
back pain (10.4 ± 4.1 cm) compared to the healthy indi-
viduals (5.9 ± 5.2 cm) (F(1, 39) = 8.78, P < 0.01). When
vibration was applied to the tibialis anterior muscles, all
subjects shifted the CoP anteriorly (see Fig. 2), but the CoP Fig. 2. Anteroposterior sways for the muscle vibration conditions and
displacement in the elderly (7.2 ± 3.2 cm) was significantly subgroups. Group mean values and standard deviations.
 S. Brumagne et al. / Neuroscience Letters 366 (2004) 63–66
Fig. 3. Anteroposterior sways of a representative healthy person (A) and
a person with low back pain (B) during the vibration trials: (1) bilateral
triceps surae muscles vibration; (2) lumbar paraspinal muscles vibration;
(3) bilateral tibialis anterior muscles vibration. Note the absence of postural
recovery after cessation of ankle muscles vibration in the person with
low back pain.
after vibration stopped (4.9 ± 1.9 s) than the elderly healthy
persons (10.9 ± 4.1 s) (F(1, 39) = 16.93, P < 0.001).
Fig. 3 displays the anteroposterior CoP displacement during
the muscle vibration conditions of a representative healthy
young person (panel A) compared to a representative young
person with LBP (panel B).
Healthy subjects, both young and elderly, proved sen-
sitive to muscle vibration at the ankles and lumbar trunk.
This sensitivity manifested as a change in the average dis-
placement in the CoP away from the center of support, in
the forward direction to vibration of tibialis anterior and
paraspinal muscles and posterior to vibration of triceps surae
muscles. Our observation that elderly healthy persons are as
sensitive to muscle vibration as young individuals are not in
agreement with previous studies that reported little [13] or
no sensitivity to muscle vibration [12] in elderly persons. A
possible explanation for this disparity could be that the age
of the elderly in these previous studies was much higher
(i.e., 82 years) than in our study. An alternative explanation
might be that only triceps surae muscles vibration was used
in these studies (and not tibialis anterior muscle vibration).
Moreover, a recent study on dynamic position sense at the
ankle in elderly persons showed that the age-related de-
cline in position sense performance was not due to muscle
spindle changes of tibialis anterior muscle [14].
Our results suggest that both LBP and aging alter pro-
prioceptive sensitivity in different parts of the body, at least
65
as far as proprioception influences standing posture. In our
comparison of the vibration responses of individuals with
and without LBP, sensitivity to vibration of the paraspinal
muscles was lower in persons with LBP. In contrast, per-
sons with LBP were more sensitive to vibration of the
triceps surae muscles compared to the healthy persons. In
our comparison of young and elderly persons, tibialis an-
terior vibration induced a larger displacement of the CoP
in elderly individuals. In addition, after cessation of mus-
cle vibration, healthy persons recover their normal postural
orientation after the offset of vibration much faster than
persons with LBP, and young individuals much faster than
their elderly counterparts. These changes in sensitivity to
vibration in the lower back and ankles produced by aging
and back pain could have resulted from a decrease in the
sensitivity of paraspinal muscle spindles or to changes in
the central processing of this afferent information [3].
In healthy persons, it has been shown that the interpre-
tation and use of sensory information can be altered in
accordance with the internal representation of the current
posture [8]. Similarly, sensorimotor deficits may result from
pathology or normal aging, as observed in patients with low
back pain [3]. One way to compensate for these propriocep-
tive deficits is to increase reliance on other sensory systems
such as vision, for instance, in the control of posture [9,12].
Alternatively, individuals can shift the focus of attention
from loci with deficient sensory input to others with intact
input. The hypothesized sensory deficiency might be due to
specific proprioceptive deficits from a particular part of the
body or to the individual’s unwillingness to move that part
of the body, thereby generating proprioceptive information,
because movement causes pain or it induces a fear of injury.
For example, patients with LBP might stiffen the trunk
and pelvis and hyperextend the knees in order to decrease
the number of degrees of freedom. Similar observations
have been reported in deafferented patients and individuals
with Parkinson disease, who seem to stiffen their trunk and
pelvis, possibly to compensate for defective information
about body position [16]. Fear of falling, for example, in
the elderly, decreases the CoP excursion due to increased
muscle activity and axial stiffness [4]. In patients with LBP,
anteroposterior sway increases during complex conditions
or tasks, but not in quiet stance [11]. In elderly individuals,
who are generally less flexible in the axial parts of the body,
the CoP moves less during quiet stance than in young adults.
It can be argued that this kind of postural control strategy,
i.e., inverted pendulum model of postural control [17], seems
to be sufficient during quiet stance, resulting sometimes
in even a smaller CoP displacement compared to healthy
persons, but in more complex conditions or tasks, such as
unstable support surface which decreases the reliance on
the ankle proprioception [8], this postural control strategy
seems to be inadequate resulting in postural instability and
even falls. The advantage of controlling fewer degrees of
freedom is offset by loss of flexible responses to dampen
external perturbations. Instead, perturbations are transferred
66 S. Brumagne et al. / Neuroscience Letters 366 (2004) 63–66
more equally to all body segments in stiff persons. Postural
chain mobility seems to be necessary for postural stability,
in particular, in the pelvic and lumbar spine regions [7,10].
Whether the observed reduced proprioceptive sensitivity
of the lower back is a result of or a cause of LBP remains
an intriguing question to be solved. In either case, the ob-
served change in proprioceptive sensitivity in persons with
LBP might be an underlying cause of the high recurrence
rate seen in patients with low back pain. Similarly, the al-
tered proprioceptive sensitivity seen in the elderly group
could partially explain the higher risk for falling in complex
postural activities. Flexibility and efficient central process-
ing might be a key factor in postural stability, as previously
shown in expert gymnasts [15], and such flexibility could
be important for preventing LBP and falls.
In conclusion, elderly individuals and persons with LBP
have altered proprioceptive sensitivity. Reweighting of pro-
prioceptive input by increasing the gain at the ankle joints
seems to take place in both groups of individuals. The
adapted postural control strategy might be effective in sim-
ple conditions, but perhaps less so when postural demands
increase.
Acknowledgements
This work was partially supported by grants PDM/99/121
(Research Council K.U. Leuven) and 1.5.104.03 (Fund for
Scientific Research—Flanders).
References
[1] B.R. Bloem, J.H. Allum, M.G. Carpenter, F. Honegger, Is lower
leg proprioception essential for triggering human automatic postural
responses? Exp. Brain Res. 130 (2000) 375–391.
[2] M.C. Brown, I. Engberg, P.B.C. Matthews, Relative sensitivity to
vibration of muscle receptors of the cat, J. Physiol. 192 (1967) 773–
800.
[3] S. Brumagne, P. Cordo, R. Lysens, S. Swinnen, S. Verschueren, The
role of paraspinal muscle spindles in lumbosacral position sense in
individuals with and without low back pain, Spine 25 (2000) 989–
994.
[4] M.G. Carpenter, J.S. Frank, C.P. Silcher, G.W. Peysar, The influence
of postural threat on the control of upright stance, Exp. Brain Res.
138 (2001) 210–218.
[5] G. Eklund, General features of vibration-induced effects on balance,
Ups. J. Med. Sci. 77 (1972) 112–124.
[6] J.C.T. Fairbank, P.B. Pynsent, The Oswestry Disability Index, Spine
25 (2000) 2940–2953.
[7] P.W. Hodges, V.S. Gurfinkel, S. Brumagne, T.C. Smith, P.J. Cordo,
Coexistence of stability and mobility in postural control: evidence
from postural compensation for respiration, Exp. Brain Res. 144
(2002) 293–302.
[8] Y.P. Ivanenko, I.A. Solopova, Y.S. Levik, The direction of postural
instability affects postural reactions to ankle muscle vibration in
humans, Neurosci. Lett. 292 (2000) 103–106.
[9] K.E. Jones, J. Wessberg, A. Vallbo, Proprioceptive feedback is re-
duced during adaptation to a visuomotor transformation: preliminary
findings, Neuroreport 12 (2001) 4029–4033.
[10] E. Kantor, L. Poupard, S. Le Bozec, S. Bouisset, Does body stability
depend on postural chain mobility or stability area? Neurosci. Lett.
308 (2001) 128–132.
[11] M.I. Mientjes, J.S. Frank, Balance in chronic low back pain patients
compared to healthy people under various conditions in upright
standing, Clin. Biomech. 14 (1999) 710–716.
[12] I. Pyykkö, P. Jantti, H. Aalto, Postural control in elderly subjects,
Age Ageing 19 (1990) 215–221.
[13] C. Quoniam, L. Hay, J.P. Roll, F. Harlay, Age effects on reflex
and postural responses to propriomuscular inputs generated by ten-
don vibration, J. Gerontol. A. Biol. Sci. Med. Sci. 50 (1995) 155–
165.
[14] S.M.P. Verschueren, S. Brumagne, S.P. Swinnen, P.J. Cordo, The
effect of aging on dynamic position sense at the ankle, Behav. Brain
Res. 136 (2002) 593–603.
[15] N. Vuillerme, N. Teasdale, V. Nougier, The effect of expertise in
gymnastics on proprioceptive sensory integration in human subjects,
Neurosci. Lett. 311 (2001) 73–76.
[16] N. Wijnberg, Posture in Parkinson patients: a proprioceptive
problem?, in: J. Duysens, B.C.M. Smits-Engelsman, H. Kingma
(Eds.), Control of Posture and Gait, Maastricht, 2001, pp. 758–
761.
[17] D.A. Winter, A.E. Patla, F. Prince, M. Ishac, K. Gielo-Perczak,
Stiffness control of balance in quiet standing, J. Neurophysiol. 80
(1998) 1211–1221.