Acta Otolaryngol 2000; 120: 151 155

Postural Stability Using Different Neck Positions in Normal Subjects and

Patients with Neck Trauma
A. KOGLER, J. LINDFORS, L. M. O8 DKVIST and T. LEDIN
From the Department of ENT, Uni6ersity Hospital, Linkoping, Sweden

Kogler A, Lindfors J, O8 dkvist LM, Ledin T. Postural stability using different neck positions in normal subjects and patients
with neck trauma. Acta Otolaryngol 2000; 120: 151155.
Subjects with neck problems, such as whiplash injuries, often complain of disturbed equilibrium and, in some instances,
provocation of the neck position can elicit such problems. The importance of neck proprioceptors for maintaining balance
is gaining increased interest, moreover the function or malfunction of the otoliths may disturb equilibrium in certain head
positions. The aim of the study was to create a reference material for postural control and its dependence on head position
in healthy subjects and to compare this with a set of patients with known neck problems and associated vertiginous
problems. A total of 32 healthy subjects (16 men, 16 women; age range 21 58 years) as well as 10 patients age range
27 62 years (mean 44 years) with neck problems and associated balance problems since a whiplash injury were tested for
postural control using the EquiTest dynamic posturographic model. The normal subjects were initially split into four age
groups in order to estimate the effects of age on performance. The postural stability was evaluated for dependence of
support surface conditions (stable or sway-referenced), visual input (eyes open or closed) and head position (neutral, left
rotated, right rotated, extended backwards or flexed forward) using analysis of variance (ANOVA) with Tukeys post hoc
test in case of a significant factor effect. As expected, visual cues as well as stable support surface improve postural
stability (p B 0.001). Postural stability is statistically different in the head extended backwards condition compared with
the other four head positions (pB0.001 in all cases) in both patients and controls. Eliminating this test condition from
the analysis, only a slight (pB0.05) difference between head forwards and head turned left remained. This pattern of
results remained if the normal subjects were only split into two age groups instead of four. Finally, the patient group
exhibited significantly lower postural performance than all the groups of normal subjects (p B0.01), but none of the
normal groups differed significantly from each other. It is concluded that the postural control system is significantly
challenged in the head extended backwards condition in both normal subjects and patients with previous whiplash injury
and persistent neck problems. The patient group differed statistically from all groups of normal subjects. This suggests
that neck problems impair postural control, and that the head extended position is a more challenging task for the
postural system to adapt to. Whether this is due to utricular malpositioning, central integrative functions or cervical
proprioceptive afferents is not within the scope of this study to answer. Key words: neck position, neck proprioceptors,
postural control, whiplash injury.

INTRODUCTION
The human posture is the result of positioning and
orientation of the body and limbs in equilibrium with
motion and gravitation. Postural compensations fulfil
three behavioural tasks. First, the support of head
and body against gravitation forces. Secondly, the
maintenance of the body mass balanced over the base
of support. Thirdly, the stabilization of the support-
ive parts of the body while others are in motion.
Postural adjustments are based on visual, vestibular
and somatosensory input integrated in a complex
feedback regulatory system (1).
The vestibular system detects linear and angular
acceleration of the head. Signals from this system
enable a stable visual image on the retina by the
induction of compensatory head and eye movements.
Information from the vestibular organs also con-
tribute to balance by adjustments of posture via the
vestibulocollic and vestibulospinal reflexes (2).
The somatosensory system is composed of muscle
spindles, the Golgi tendon organs and joint receptors,
all of which are proprioceptive, and the exteroceptive
receptors, such as mechano-receptors in subcuta-
neous tissue. The muscle spindles are essential for the
sense of muscle length and the velocity of length
change, and the Golgi tendon organs monitor the
muscle tension. Mechanical joint receptors register
the movements in the joint capsule (1, 3).
The neck balances and moves the head and its
weight. Furthermore, it protects the spinal cord and
the vessels passing through it. In this study, one of
the most interesting functions is the necks proprio-
ceptive contribution to the CNS, which enables the
CNS to locate the heads orientation relative to the
trunk (4). The neck muscles, along with the small
muscles of the hand and the eye, contain one of the
highest densities of muscle spindles compared with
other muscles in the body. It is indicated that muscle
spindles are the main contributors to neck proprio-
ception and that the joint receptors play a minor role
(2). Vestibular and neck reflexes cooperate in the
positioning and stabilizing of the head. These reflexes
synergize in the neck, but antagonize in the limbs (1).
The clinical implications are often severe in the
case of lesions to the neck, and it has been shown by
Oosterveld et al (5) that whiplash injuries produce

2000 Taylor & Francis. ISSN 0001-6489

152 A. Kogler et al.
cervical vertigo in many patients. In some instances
provocation of the neck position can elicit problems
with imbalance and vertigo, especially in subjects
with neck pain and subjective postural stability prob-
lems (6). Moreover the functions of the otoliths may
disturb equilibrium in certain head positions, and
must be accounted for (7). As the importance of the
neck proprioceptors for maintaining balance is gain-
ing increased interest, is was decided to carry out a
study in normal subjects and subjects with neck prob-
lems, in order to improve the understanding of the
physiological importance of different neck positions
on equilibrium.
The study thus evaluated postural control in five
different head positions in conditions with both sta-
ble and sway-referenced support surface, as well as
with and without vision. A group of normal subjects
divided into four age categories was studied and, in
addition, a group of patients who had experienced a
whiplash trauma with subsequent unsteadiness prob-
lems. Thus we could evaluate both whether age
played a significant role and whether the patients had
significantly impaired postural control compared with
healthy subjects.
MATERIALS AND METHODS
Normal subjects
A total of 32 healthy subjects (16 men, 16 women;
mean age 39 years, age range 21 58 years), who were
free of medication, with no history of previous neck
injury or vertigo were selected. They were subdivided
into four equal age-groups; 21 29 years, 3039
years, 4049 years and 50 58 years. In another
analysis, the two younger groups were treated as one
group, and the two older groups as one.
Patients
Ten patients aged 27 62 years (mean 44 years) with
neck problems and associated balance problems since
a whiplash injury on average 2.7 years ago (range
1 5 years). They used non-narcotic painkillers, but
otherwise were not under medication. All had
whiplash injuries of type WAD 2 3 (8), i.e. no radio-
logical findings, but subjective complaints and, in
instances, abnormalities on advanced vestibulo-ocu-
lomotor testing.
All subjects were required not to take any alcoholic
beverages for at least 24 h before the study. The
vestibular function of the subjects was clinically nor-
mal, as tested negative for spontaneous nystagmus,
gaze nystagmus and headshake nystagmus in Frenzel
glasses. Rombergs test was normal. Hearing was con-
sidered normal according to a one frequency Schwa-
bach test. All subjects were given detailed infor-
Acta Otolaryngol 120
mation on the test conditions, but not of the purpose
of the study. All gave informed consent.
Methods
EquiTest dynamic posturography (Neurocom Int Inc,
Clackamas, Oregon, USA) was used in a partly
modified way. In the tests the subject stands on a
forceplate, which is surrounded by a visual environ-
ment. The subject stands with feet parallel and com-
fortably apart. Shoes are not allowed. A safety
harness is applied to the subject to catch them if they
fall. Both the forceplate and the surround can be
made movable exactly following the subjects antero-
posterior (AP) sway. The forceplate registers the
forces between feet and ground as well as horizontal
AP shear forces, thereby allowing estimation of posi-
tion of centre of body mass (i.e. equilibrium) as well
as the sway pattern in terms of hip or ankle compen-
satory movements (i.e. strategy). Physiologically the
highest possible sway range is 12.5. A score of 100
indicates absolutely no sway and a zero score means
a fall or a sway of 12.5. A strategy score expresses
the amount of horizontal shear force (a pure ankle
strategy means no shear forces and thus a score of
100, maximum hip movements means large shear
forces, giving a zero score).
The standardized investigation is composed of six
test conditions (SO 16), varying support surface
condition (stable or sway referenced) and visual input
(open eyes and looking into the stable visual sur-
round, closed eyes and open eyes looking at a sway-
referenced visual surround). Fig. 1 shows the test
conditions.
Procedure
All 32 normal subjects were evaluated by either of the
two first authors, and all patients were evaluated by a
skilled vestibular laboratory assistant.
Fig. 1. The six different sensory conditions of EquiTest
dynamic posturography.
Acta Otolaryngol 120 Postural stability in different neck positions 153

EquiTest dynamic posturography was performed in

a modified manner. First, a full standard test se-
quence (SO1-6) was performed with the head in the
neutral position. Then four additional head positions
were tested with eyes closed as well as open, and on
stable as well as sway-referenced support surface.
Thus a total of 6+4 4 = 22 tests were conducted.
The whole test lasted approximately 50 min per sub-
ject. None of the subjects considered the test stressful
or laborious.
The four tested neck positions were maximum flex-
ion forwards, maximum extension backwards, and
maximum rotation (without tilting) of the neck to the
left and right side. No other movement of the trunk
was allowed. Each SO was performed 2 20 s, and a
third repetition was performed in occurrence of a fall Fig. 2. Equilibrium scores in patients (lowest line) and four
or \25% variation, in order to achieve stable values. subgroups of controls (upper four lines, not significantly
Only the mean of the two best results were registered deviating from each other). The plot shows on the x-axis
the five different head positions (in order from the left:
in the statistical material. The neck positions were neutral, backwards, forwards, right and left). Note that the
statically held during the 20 s of each SO test, but the average of the four conditions formed by stable and sway-
subjects were instructed to relax in the head neutral referenced support surface conditions, in addition to the
position in between each test repetition. Unfortu- two visual conditions, are shown in the figure. As the group
nately, a bias was introduced in the test schedule as factor was without significant interactions, this group pat-
tern of results holds true for these four conditions.
both neck position, visual conditions and support
surface conditions were not randomized.
added, pB 0.05). Finally, the patient group exhibited
Statistics significantly lower postural performance than all the
The four (and later on two) different age groups as groups of normal subjects (pB 0.01, at least), but
none of the normal groups differed significantly from
well as the patient group constituted the first factor in
each other (Fig. 2). The group factor in the analysis
the analysis. In addition, the dependence of support
was free of interactions with the other factors, mak-
surface conditions (stable or sway-referenced), visual
ing the achieved results hold for all tested groups in a
input (eyes open or closed) and head position (neu-
uniform manner. The data in Fig. 3 serves as a more
tral, left rotated, right rotated, extended backwards
clinical description of the effects of head extension
or flexed forward) was evaluated using analysis of
for control subjects (all 32 pooled into 1 group) vs
variance (ANOVA) with Tukeys post-hoc test in case
the patient group. This grouping of the controls was
of a significant factor effect. A probability level of 5%
justified, as there was no age group dependence of the
was considered significant. Normality of data was
results in the controls. It is easily seen that the
ensured using the Kolmogorov-Smirnov test.
differences on the stable support surface are quite
small (though highly significant), whereas the differ-
RESULTS ences on the sway referenced support surface are
Equilibrium scores much larger and display larger variation.
As expected, visual cues as well as stable support
surface improve postural stability compared with Equilibrium strategies
their counterparts (p B 0.001). Postural stability was Again, visual cues and a stable support surface were
statistically different in the head extended backwards beneficiary (pB0.001 for both). The head backwards
condition compared with the other four head posi- extension test differed from all other positions (pB
tions (pB0.001 in all cases, Fig. 2) in both patients 0.001). The neutral head position differed slightly vs
and controls. Eliminating the extended head position both left and right torsion tests (pB 0.01 and pB
test from the analysis, only a slight (pB 0.05) differ- 0.05, respectively). The patient group differed vs the
ence between head flexed forwards and head turned three youngest age groups (pB0.01, pB 0.01, pB
left positions remained. This pattern of results re- 0.05), but not vs the oldest normal group (p B 0.08).
mained if the normal subjects were only split into two Splitting only into two normal subject age groups, in
age groups instead of four (a weak significance be- spite of this, confirmed the above findings to almost
tween head forwards and the neutral position was every detail (p values slightly better in a couple of
154 A. Kogler et al. Acta Otolaryngol 120

tests). The group factor in the analysis was also free
of interactions with the other factors.
DISCUSSION
This study has shown that the postural control sys-
tem is significantly challenged in the head extended
backwards condition in both normal subjects and
patients with previous whiplash injury and persistent
neck problems. Moreover, the patient group differed
statistically from all groups of normal subjects. This
suggests that neck problems impair postural control,
and that the head extended position is a more chal-
lenging task for the postural system to adapt to.
Whether this is due to utricular malpositioning, cen-
tral integrative deficits or disturbed cervical proprio-
ceptive afferents is not within the scope of this study
to answer.
Animal studies have shown the importance of cer-
vical sensory information for postural control. Stud-
ies in humans are however more ambiguous (9).
Hamid (10) has argued that head position is of no
large importance for equilibrium performance, how-
ever those results were based on mere case observa-
tions and do not appear to be well founded. The
explanation of the results may, according to Hamid,
be the presence of a mechanism that compensates for
different head positions and adapts balance reflexes,
which results in a minimal postural sway.
The results of our study however indicate that neck
extension backwards, compared with the other four
neck positions, impairs postural balance, as evidenced
by a significant increase of AP-sway in normal sub-
jects as well as patients with neck trauma evaluated
by dynamic posturography. The other three neck
provocations (forward flexion, left and right rotation)
did not differ significantly from each other or the
neutral position. This is supported in other reports
(11, 12, 7). Barin et al. (11) proposed that head
extension increases postural sway in normal subjects.
This pattern, according to the authors, indicates that
head extension causes deterioration of both visual
and vestibular cues and increases the dependence of
the proprioceptive input. Brandt et al. (12) showed
that an increase in sway amplitudes was present in
normal subjects in the condition of head extension
and with eyes closed while standing on a slab of foam
rubber. The difference seems to decline with both
short- and long-term practice with the abnormal head
position. Interestingly, they did not obtain any sig-
nificant increase in postural sway with eyes open.
Jackson and Epstein (7) tested the effect of 55
backwards head extension in normal subjects. The
effect was a significant increase in sway both with

Fig. 3. (A D) Four graphs showing the equilibrium scores in the total group of control subjects (n= 32, the two bars on
the left in each subgraph), compared with the patient group (the two right bars in each group). Means and standard
deviations for both head in the neutral position (neu) as well as extended backwards (ext) are shown. All the four
combinations of support surface condition (stable, sway referenced) and visual conditions (eyes open and closed) are shown.
Acta Otolaryngol 120
eyes closed and open in SO 4 5. They found the
largest increase in the condition eyes closed. They
suggest that this increase is attributable to the un-
usual positioning of the utricular otoliths, which
places these organs out of their working range.
Three hypotheses have been postulated to explain
the mechanisms causing vertigo in abnormal head
positions. The 6ascular hypothesis: that vertigo is due
to compression of the vertebral artery causing inter-
mittent ischaemia of the brain-stem and/or the inner
ear (13). Several studies (11, 12), however, have been
able to exclude vascular insufficiency as a cause of
cervical vertigo. Karlberg (4) states that the possibil-
ity of this ischaemia theory must although be con-
sidered in elderly patients with known
arteriosclerotic disease. The neuro-6ascular hypothesis
implicates that cervical vertigo is a consequence of
vasoconstriction in the vertebrobasilar system fol-
lowing irritation of the cervical sympathetic nerve
plexus (14). The somatosensory input hypothesis has
its basis in an abnormal sensory input from cervical
receptors (15). It has also been proposed that cervi-
cal vertigo might spring from a combination of all
these different pathogenic mechanisms (16). Karl-
berg (4) means that both animal and human studies
point to the last hypothesis as the most likely, al-
though there are at present no reliable methods to
isolate the effects of proprioceptive disturbances.
It is thus concluded that the neck backwards posi-
tion, in healthy adults as well as in patients with
neck problems due to whiplash injuries, significantly
influences equilibrium performance. Patients with a
whiplash injury and subjective neck complaints have
impaired postural control compared with control
subjects. We conclude that it would be of value to
evaluate postural control in the extended backwards
position.
ACKNOWLEDGMENTS
We thank Mrs Lisbeth Noaksson for valuable technical
and administrative assistance.
REFERENCES
1. Guez C. Posture. In: Kandel ER, Schwartz JH, Jessel
TM, eds. Principles of neural science. New York: El-
sevier, 1991: 596607.
2. Willis Jr WD. The auditory and vestibular system. In:
.
Postural stability in different neck positions 155
Berne RM, Levy MN, eds. Physiology. St Louis: Mosby-
Year Book Inc, 1992: 166 88.
3. Enbom H. Vestibular and somatosensory contribution
to postural control. Dissertation. Lund: University Hos-
pital, 1990.
4. Karlberg M. The neck and human balance. Dissertation.
Lund: University Hospital, 1995.
5. Oosterveld WJ, Kortschot HW, Kingma GG, de Jong
HAA, Saatci MR. Electronystagmographic findings fol-
lowing cervical whiplash injuries. Acta Otolaryngol
(Stockh) 1991; 111: 201 5.
6. A, lund M, Ledin T, O8 dkvist LM, Larsson S-E, Moller
C. Dynamic posturography among patients with com-
mon neck disorders a study of 15 cases with suspected
cervical vertigo. J Vest Res 1993; 3: 383 9.
7. Jackson RT, Epstein CM. Effect of head extension on
equilibrium in normal subjects. Ann Otol Rhinol Laryn-
gol 1991; 100: 63 7.
8. Spitzer WD, Skovron ML, Salmi LR. Scientific mono-
graph of the Quebec task force on whiplash-associated
disorders: redefining whiplash and its management.
Spine 1995; 20: 1 73.
9. de Jong JM, Bles W. Cervical dizziness and ataxia. In:
Bles W, Brandt T, eds. Disorders of posture and gait.
Amsterdam: Elsevier Science Publisher BV (Biomedical
division), 1986: 185 206.
10. Hamid MA. Effect of head position on posture during
altered visual and proprioceptive orientation: Prelimi-
nary report. J Vestib Res 1994; 6: 481 3.
11. Barin K, Curtis MS, Welling DB. Effect of head orien-
tation on the diagnostic sensitivity of posturography in
patients with compensated unilateral lesions. Head Neck
Surg 1992; 106: 355 61.
12. Brandt T, Krafczyk S, Malsbenden I. Postural imbalance
with head extension: improvement by training as model
for ataxia therapy. Ann NY Acad Sci 1981; 374: 63649.
13. Sheehan S, Bauer RB, Meyer JS. Vertebral artery
compression in cervical spondylosis. Neurology 1960; 10:
968 86.
14. Barre M. Sur un syndrome sympathique cervical pos-
terieur et sa cause frequente: larthrite cervicale. Rev
Neurol 1926; 33: 1246 8.
15. Ryan GM, Cope S. Cervical vertigo. Lancet 1955; 31:
1355 8.
16. Pfaltz CR. Vertigo in disorders of the neck. In: Dix MR,
Hood JD, eds. Vertigo. New York: John Wiley and Sons
Ltd, 1984: 179 97.
Address for correspondence:
Torbjorn Ledin, MD, PhD
Department of ENT
University Hospital
SE-581 85 Linkoping
Sweden
Tel: +46 13 222000
Fax: + 46 13 222504
E-mail: torbjorn.ledin@inr.liu.se
Copyright of Acta Oto-Laryngologica is the property of Taylor & Francis Ltd and its content may not be copied
or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.