See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/283993339

Clinical changes in vestibular function with time after unilateral vestibular

loss

Article · January 2000

CITATIONS READS

33 72

2 authors, including:

Ian S Curthoys
The University of Sydney
323 PUBLICATIONS   12,212 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Transient and sustained vestibular systems View project

The video head impulse test of semicircular canal function View project

All content following this page was uploaded by Ian S Curthoys on 18 August 2018.

The user has requested enhancement of the downloaded file.

 COMMONWEALTH OF AUSTRALIA
Soufce:
Copyright Regulations 1969
Chaptef 7 in S. J. Hefdman(2000).
'iJ e:s:tib ul af Rehab iIi tati on: WARNING
Phi Iadelphi a : Davi :s: pp.172-194. This material has been reproduced and communicated to
you by or on behalf of the University of Sydney pursuant
to Part VB of the Copyright Act 1968 (the Act).

The material in this communication may be subject to

copyright under the Act. Any further reproduction or
communication of this material by you may be the subject
of copyright protection under the Act.

------1-l-""'""'""'ve this notice

CHAPTER 7
Clinical Changes in Vestibular
Function With Time Mter
Unilateral Vestibular Loss
Ian S. Curthoys, PhD and G. Michael Halmagyi, MD

It has been known for a more than a century that acute total loss of a previously intact
labyrinth, unilateral vestibular loss (or unilateral vestibular deafferentation [uVD]) invari-
ably results in a stereotyped pattern of disruption of equilibrium in species as diverse as
fish, amphibians, birds, rodents, monkey, and man. Although the loss is usually perma-
nent, the sensory and motor disruption it causes is usually only temporary. The process of
recovery from the neurologic effects of acute u VD is termed vestibular compensation. I-ll
Unilateral vestibular loss is a common clinical event. It occurs as a result of dis-
ease, such as acute vestibular neuritis, or as a result of surgical treatment, such as uni-
lateral labyrinthectomy or unilateral vestibular neurectomy. Although most patients
recover well from uVD, some do not, and the cause of the poor compensation of those
few patients is a question of great research interest. The aim of this chapter is to de-
scribe the pattern of onset and resolution of the sensory and motor components after
uVD, and the pathophysiological basis of the uVD syndrome in humans.

CONSEQUENCES OF UNILATERAL VESTIBULAR LOSS

IN HUMANS

Immediately after uVD, humans invariably experience intense dysequilibrium. This

dysequilibrium has both sensory and motor components that can be further categorized
as static or dynamic. Static components are those present when the person is at rest,
whereas the dynamic components are those evident only during movement; that is dur-
ing stimulation of the sole remaining labyrinth by angular or linear acceleration. Some of
these perceptual abnormalities are temporary, whereas others appear to be permanent.

172
CHAPTER 7 CLINICAL CHANGES IN VE~~~ 173

It is important to emphasize that the syndrome that develops following acute loss
of a normal labyrinth is totally stereotyped in pattern and duration in each species.
Some species recover within a few hours (goldfish), 9 whereas in other species recovery
takes hours, days, or weeks. Certain behavioral components are invariably present and
in any particular species; each component changes over time with a characteristic tem-
poral profile.
To measure precisely the pattern and resolution of the uVD syndrome in humans,
it is best to study the same patients before and after surgical deafferentation of one in~
tact labyrinth. Although such cases and the facilities for studying them are few, some
long~term quantitative data on the precise sensory and motor consequences of uVD in
humans has recently become available 1 il- 12 In particular, some patients with Meniere's
disease probably have normal or near normal labyrinthine function between attacks of
vertigo and some of these patients have been studied before and after unilateral vestib-
ular neurectomy.

Static Sensory Components

After surgical uVD, immediately on recovering from the anesthesia, patients expe-
rience two different false spatial sensations (i.e., illusions). Both occur at rest, and both
resolve within 1 to 2 days. One is an illusion of angular motion in yaw and the other is
an illusion of linear tilt in roll. The angular motion illusion is a false sense of angular
rotation, either of self or of the world. With eyes closed, the illusion is of self-motion
with the body turning about its long axis toward the uVD side. With eyes open, the il-
lusion is of world-motion, now in the opposite direction. These false sensations of rota-
tion are called vertigo. The linear tilt illusion is a false sense of body roll-tilt about the
naso-occipital (or roll) axis, again toward the side of the uVD.
Both illusions are probably due to the asymmetry in resting neural activity be-
tween the two vestibular nuclei. This asymmetry occurs as a result of the sudden pro-
found decrease in resting activity in the deafferented vestibular nucleus produced by
the uVQ.B.I4 These illusions occur because it seems that whenever the level of neural
activity in one vestibular nucleus exceeds the level of activity in the other, the imbal-
ance is interpreted by the brain as rotation or tilt toward the side generating the higher
level of activity. Such an imbalance occurs during a real angular acceleration or a real
linear tilt in roll, but also after uVD. The relatively higher level of neural activity in the
medial vestibular nucleus (MVN) on the intact side, as compared to the level of activity
on the lesioned side, is interpreted by the brain as rotation toward the intact side. Sim-
ilarly the relatively higher level of resting neural activity in the lateral vestibular nu-
cleus (LVN) of the intact side, is interpreted by the brain as roll-tilt toward the intact
side. In both cases the imbalance of neural activity produced by unilateral loss is simi-
lar to the imbalance produced by natural stimulation.

Dynamic Sensory Components

Deficits in the perception of angular and linear acceleration stimulation have been
found after uVD. Although both deficits improve with time after uVD, in some pa~
tients some deficit appears to be permanent.
174 VESTIBULAR REHABILITATION

ANGULAR ACCELERATION PERCEPTION

To evaluate the precision of yaw angular acceleration perception, subjects seated
in a rotating chair were asked in a recent study to counter-rotate themselves following
a passive rotation in the dark. Brookes et aJls showed that normal subjects passively ro-
tated through an angle of between 30° and 180° (at a velocity of 80° per second) can,
when given control over chair rotation, return the chair precisely back to its starting
position. Some chronic uVD patients consistently under-responded to rotations to-
ward the u VD side whereas they were able to respond correctly to rotations toward the
intact side. Other uVD patients were able to respond correctly to rotations to either
side. These results indicate that perception of angular rotation toward the lesioned side
is impaired following uVD, and that this impairment compensates with time in some
patients.

LINEAR ACCELERATION PERCEPTION

A normal subject on earth is continuously stimulated by the linear acceleration of
gravity. If the person is upright then the direction of the gravitational vector corre-
sponds to the longitudinal axis of his body. If the person is tilted in his coronal (i.e., roll)
plane, about his nasa-occipital axis, then the angle between his body long axis and the
gravitational linear acceleration vector is called the roll-tilt angle. In our terms, the per-
son is said to be experiencing roll-tilt stimulation. Roll-tilt stimulation activates the lin-
ear acceleration sensors of the inner ear: the otolithic receptors in the utricle and the sac-
cule of each labyrinth (see Chapter 8).
Another method of producing roll-tilt stimulation is centrifugation. If a subject,
seated upright on a centrifuge with his interaural axis parallel to the centrifuge arm, is
rotated at a constant angular velocity, he will be subject to a centrifugal linear accelera-
tion (but no angular acceleration). For convenience we refer to centrifugal rather than
to centripetal accelerations because the otolithic receptor hair cells are bent in the di-
rection of the centrifugal acceleration. During constant velocity rotation the centrifugal
linear acceleration and the gravitational linear acceleration will sum to produce a re-
sultant gravitoinertial force. This resultant force is in the coronal plane and is directed
away from the body long axis. It is also a roll-tilt stimulus. If the subject is now cen-
trifuged in total darkness to exclude visual cues to verticality, he will experience an ir-
resistible sensation of being tilted in the roll plane.
Roll-tilt perception is the subjective conscious sensation of body tilt produced by
roll-tilt stimulation. If during centrifugation the subject views a luminous, rotatable
bar attached to the centrifuge chair and aligned with his interaural axis, the bar will
appear to him to be tilted (i.e., rolled) with his body, by the same amount as his body
appears to have rolled. This tilt of the subjective visual horizontal with respect to the
perceived gravitational horizontal during centrifugation is called the oculogravic "illu-
sion."16.17 A subject's ability to sense this roll-tilt stimulus during centrifugation can be
accurately measured by requiring the subject to return the small visible bar so that it is
aligned with respect to the perceived direction of the gravitational horizontal with re-
spect to his interaural axis. The following example shows how this can be done.
Consider a subject sitting upright on a centrifuge, 1 meter from the rotation axis,
with his left ear directed toward the rotation axis (Fig. 7-1). The centrifuge rotates at a
constant velocity of 30 rpm (equal to 180° per second) in darkness. Because gravita-
tional and linear acceleration are identical physical forces, the subject now cannot but
CHAPTER 7 CLINICAL CHANGES IN VESTIBULAR FUNCTION 175

n
,
·+·o X~
[ X [ ·-1·-·

~: ~:
A. B.
iHF%\%\\1~&
SEES BAR SEES BAR SETS BAR
AT REST DURING DURING
ROTATION ROTATION
FIGURE 7-l. Measurement of linear acceleration perception using the oculogravic illusion. (A)
With the centrifuge at rest in an otherwise darkened room, a normal subject (shown from behind),
seated with his interaural axis parallel to the long axis of the centrifuge arm and with his left ear to-
wards the axis of rotation, vie\VS a dimly illuminated, gravitationally horizontal, rotatable bar shown
here v.'ith interrupted lines. At rest he correctly pcrcei\'tS the bar as aligned not only with the gravi-
tational horizontal, but also \Vith his interaural axis. {B) VVhen the centrifuge with its 1-meter arm
turns at a constant velocity of 1800 per second the gravito-inertial resultant tOrce will be directed
45° from the gravitational vector, towards the subject's right labyrinth. In the otherwise darkened
room the subject can now only assume that the resultant force is rl1e gravitational force and there-
fore perceives that his own body long axis has rotated 45° to his right; he will sense a 45° roll-tilt to
his right. The illuminated bar is, however, still physically aligned with his own interaural axis and not
with the resultant force, so that he perceives the bar as also having rolled with him, 45° to the right.
(C) 'When required to set the bar to the perceived gravitational horizontal, the subject does so by
rotating the bar 45° to his left (i.e., counter clockwise) and so accurately aligns the bar normal to the
resultant fOrce. (From Dai, et aPO).

regard the gravitoinertial resultant as his subjective gravitational vertical. Because the
resultant is now directed from his left to his right, the only percept that will accord
with his sensations is that he has been tilted onto his right side.
The angle of resultant linear acceleration is simple to calculate; in this case it is 45'.
A normal subject will sense the direction of this resultant accurately, and will therefore
perceive that his body has been roll-tilted 45' to his right. If he now views a dimly illu-
minated, rotatable light bar that is attached to the centrifuge chair, and is physically
aligned parallel with his interaural axis (and therefore also normal to the gravitational
vertical) it will appear that the bar has tilted with him to the right, also by 45'. He can
now indicate the precise angle of perceived body tilt by rotating the bar to his left until
it is set to where he perceives the gravitational horizontal to be. Normal subjects cen-
trifuged in darkness can indicate their perception of roll-tilt accurately; they can set a
light bar to within 5° of the gravitational horizontal on 95 percent of attempts. These
176 VESTIBULAR REHABILITATION

observations show that roll-tilt perception can be accurately measured by nulling the
oculogravic illusion and raise the possibility of using roll-tilt perception to measure
otolith function clinically. Because roll-tilt perception partly depends on otolithic sen-
sory input and normal subjects have accurate roll-tilt perception, it is possible to test
otolith function clinically by means of the oculogravic illusion.
In a study of 30 patients before and after uVO,JO we found that 1 week after opera-
tion all 30 patients showed a loss of sensitivity when given a roll-tilt stimulus toward
their affected ear. They showed an asymmetry of roll-tilt perception in that they had a
smaller oculogravic illusion for roll-tilt stimuli directed to their affected ear than they
had before the operation (Fig. 7-2). This loss of sensitivity became increasingly evident
with increasing roll-tilt stimulus angles. Furthermore, even when tested 6 months after
uVD, these patients still showed a significant loss of sensitivity to linear accelerations
that were medially directed with respect to the single functioning utricle (Fig. 7-3).
Asymmetries of roll-tilt perception have been shown during simple roll-tilts to the left
and right on a tilting chairiB-20
These results indicate that total u VD causes a deficit in roll-tilt perception of linear
accelerations directed toward the lesioned side and, although this deficit compensates
over time, this compensation is incomplete, and a small but detectable deficit in roll-tilt
perception toward the lesioned side is a permanent legacy of uVD.

B
OPERATED~
EAR~
I
/··'·''"''"''' ,. '\
MEDIALLY DIRECTED LATERALLY DIRECTED
ROLL-TILT STIMULUS ROLL-TILT STIMULUS

FIGURE 7-2. Medial versus lateral roll tilt stimulation following unilateral vestibular loss.
Schematic representation of roll-tilt stimulation of the remaining intact right labyrinth of a patient
who has had a left vestibular loss. The patient, viewed from behind, sits across the centrifuge arm so
that his interaural axis is colinear \Vith the centrifugal linear acceleration. If his intact right labyrinth
is positioned away toward the rotation axis (A), then its otoconia! membrane \viii be displaced medi-
ally, toward the center ofthe head, and he is said to be subject to a medially-directed roll-tilt stimu-
lus. If his intact right labyrinth is positioned away from the rotation axis (B) 1 its otoconia! membrane
will be displaced laterally, away from the center of the head, and he is said to be subject to a laterally-
directed roll-tilt stimulus. The direction of the centrifugal acceleration indicates the direction of
bending of the hair cells. The resultant (R) has nvo components: Fy1 the roll-shear component
which acts in the interaural axis, across the mean utricular plane, and Fz the component due to grav-
itational acceleration which acts as a compressive force in the body longitudinal axis. Fz is constant
at lg during centrifugation. (From Dai, et aJIO).
 ROLL - TILT PERCEPTION (DEG)
40 I- I
1 WEEK 3 WEEK
PRE Ul-' 12 WEEK 24 WEEK
)g

,
30

~ ~
,,.'"
'""' .::::·

=~(:·
20 .

~
.
10 tf.~ ~ _?~
::=== :/~ =====v :/
0 •

ROLL - TILT PERCEPTUAL

DIFFERENCE (DEG)
3Q 1- I
PRE Ul-' 1 WEEK 3 WEEK 12 WEEK 24 WEEK

~
20

10

0
'""'~"' ::::::::::: :::=:::::::::::;:: ,~,, ~
::::::::: :::::;::::::::::::
~
. I I I I I
-10 ' '
0 10 20 30 0 10 20 30 0 10 20 30 0 10 20 30 0 10 20 30

ROLL - TILT STIMULUS ANGLE (DEG)

FIGURE 7-3. Linear acceleration perception befOre and after unilateral vestibular loss. Summary of results of linear acceleration perception from 33
patients compared to 31 normal subjects at l oo, 17°, and 26° roll-tilt stimulus angle, before and at various times from l to 24 weeks after unilateral ves-
tibular loss. Shaded areas show two-tailed confidence intervals for normal subjects. (Top row.) Absolute values of perceived roll-tilt angle: open symbols
show settings for linear accelerations directed towards the intact labyrinth (i.e., laterally directed roll-tilt stimulation); tilled symbols show settings fix
..... linear accelerations directed toward the operated labyrinth (i.e., medially directed roll-tilt stimulation). The large interaural difference present I week af-
'-l ter operation has decreased by 3 weeks but is still abnormal at all stimulus levels even 24 weeks after operation. (Bottom row.) lntcraural differences in
'-l
roll-tilt perception remain abnormally high, perhaps indefinitely, after unilateral vestibular loss. (From Dai, et al,lO).
178 VESTIBULAR REHABILITATION

The clinical significance of these results is that oculogravic tests of roll-tilt percep-
tion may prove a useful means of detecting severe unilateral loss of dynamic otolith
function and may provide a way of monitoring a sensory component of vestibular
compensation. Others who have conducted careful tests of roll-tilt perception at a
number of angles using a simple tilting chair also confirm the clinical value of such a
perceptual test. IS

Static Motor Components

The static motor components of the u VD syndrome all reflect a motor offset or
bias towards the deafferented side. These components reflect both canal and otolithic
imbalances.

SPONTANEOUS NYSTAGMUS
A spontaneous horizontal nystagmus is invariably present immediately after
uVD. The slow phases are always directed toward the lesioned side. Movements that
are clinically apparent are quick phases, and these are directed away from the lesioned
side. An observer sees a horizontal eye movement pattern beating away from the le-
sioned ear. The essential characteristics of uVD nystagmus are the same as that of any
other peripheral vestibular nystagmus: it is largely horizontal, unidirectional, and sup-
pressed by visual fixation. Visual fixation suppression may be so effective that the nys-
tagmus will only be apparent when visual fixation is completely excluded, emphasiz-
ing the need to check for nystagmus in the absence of visual fixation in all patients
with suspected uVD (Fig. 7-4). Clinically, it is possible to exclude visual fixation by us-
ing Frenzel glasses or by using an ophthalmoscope to view the fundus of one eye while
the other eye is covered.2I The presence of nystagmus in the absence of visual fixation,
combined with the absence of nystagmus in the presence of visual fixation, is defini-
tive evidence that there is a peripheral vestibular lesion, on the side from which the
quick phases are beating.
Typically the patient will have primary position (second degree) nystagmus even
with visual fixation for the first day after uVD, and then first degree gaze-evoked nys-
tagmus until the end of the first week. Even after a month there will still be a low-
velocity (2" to 3" per second) first degree gaze-evoked nystagmus present only in the
absence of visual fixation. This appears to be a permanent legacy of uVD.

FIGURE 7-4. Peripheral vestibular nystagmus. Oculographic recording shows a left-beating pri-
mary-position nystagmus that is obvious only \vhen visual fixation is removed (open arrow), and is
quickly suppressed again when visual fixation is permitted (filled arrow). Peripheral vestibular nys-
tagmus can be detected clinically by viewing the fundus of one eye while occluding the other. The
patient had a right vestibular neurectomy the previous day. Upward deflections indicate rightward
eye movements, downward deflections indicate leftward eye movements. Bar= 10° and 1 sec.
CHAPTER 7 CLINICAL CHANGES IN VESTIBULAR FUNCTION 179

Recent studies suggest that spontaneous nystagmus is due to the loss of resting
activity in type I secondary horizontal semicircular canal (HSCC) neurons in the
medial vestibular nucleus on the same side as the lesion (called here the "ipsilesional"
side). The intensity of the spontaneous nystagmus is held to be an accurate index
of the relative resting rates of ipsilesional and contralesional type I HSCC canal neu-
rons (see discussion of "Neural Activity in the Vestibular Nuclei during Vestibular
Compensation").
Because the two vertical semicircular canals on one side are also deafferented by a
uVD, it is surprising that the torsional component of unilateral nystagmus is so much
Jess apparent than the horizontal component. In fact a prominent torsional component
usually indicates that the lesion is in the vestibular nucleus rather than the labyrinth or
vestibular nerve. There appears to be no entirely satisfactory explanation for these ob-
servations, probably because so few objective measurements of the three-dimensional
components of spontaneous nystagmus component following uVD have been made.

OCULAR TILT REACTION

Following uVD there is a tonic ocular tilt reaction consisting of head tilt, conjugate
eye torsion, and skew deviation, all directed to the lesioned side. 22 Although head tilt
and skew deviation are usually very small, conjugate ipsilesional ocular torsion ap-
pears to be invariably presen\. 11 .22-25 The direction of the torsion is always ipsilesional:
the 12 o'clock meridians of both eyes are rotated toward the side of the u VD. The mag-
nitude of the ocular torsion can be measured objectively by ocular fundus photogra-
phy (Fig. 7-5). One week after uVD, there is up to 15° of ocular torsion (on average 9');
one month after uVD the ocular torsion has diminished to about half of the 1-week
value. A small but statistically significant ocular torsion (2' to 3') also appears to be a
permanent legacy of u VD.
This change in ocular torsional position can be readily detected in perceptual tests,
which show a bias in settings of the subjective visual vertical or horizontal toward the
uVD side (see Chapter 8). A normal subject sitting upright in an otherwise totally dark-
ened room can accurately align a dimly illuminated bar to within 2' of the gravita-
tional vertical or horizontal. Thirty patients studied before and after uVD showed that,
although their preoperative settings of the visible bar to the perceived gravitational
horizontal were reasonably accurate, 1 week after operation they invariably set the bar
so that it was tilted down on the side of the uVD, in some cases by up to 15°.11 It must
be emphasized that these patients reported that while seated upright they perceived
themselves to be upright and they can set a nonvisual indicator (a somatosensory bar)
to the perceived horizontal accurately. The settings of the visual bar correspond very
closely to the size of the ocular torsionn and our more recent studies have confirmed
the very close relationship between ocular torsional position and the visually per-
ceived orientation of short lines.26
Although the setting of the visual bar returned toward the horizontal with time, it
was still tilted by a mean of 4' 6 months or more after uVD. In all cases the magnitude
of the tilting of the visual horizontal was closely correlated with the magnitude of the
ocular torsion. It appears therefore that an ipsilesional tilting of the visual horizontal is
a permanent legacy of uVD. This result has been confirmed by others18·"·''
The clinical significance of these findings is that careful standardized measure-
ment of the visual horizontal using a dim light bar in an otherwise totally darkened
room gives valuable diagnostic information about vestibular (mainly otolithic) func-
tionl1.29 (see Chapter 8). A significant tilting of the visual horizontal indicates vestibu-
180 VESTIBULAR REHABILITATION

BEFORE RIGHT VESTIBULAR NEURECTOMY

AFTER RIGHT VESTIBULAR NEURECTOMY

FIGURE 7-5. Fundus photographs of the left and right eye of a patient befOre (top) and l week
after (bottom) right vestibular neurectomy. After operation there is tonic rightward torsion of the 12
o'clock meridian of each eye toward the patient's right side. The change in torsion angle measures }8°
in the right eye and 16° in the left eye. \Vhen the patient \Vas asked to set a luminous bar to the per-
ceived visual horizontal in an othenvisc darkened room he set the bar tilted toward his right side by
14.2° when \'it\Ying with the right eye and 15.1° when viewing with the left. (From Curthoys, et all!).

Jar, probably otolithic, hypofunction on the side to which the patient tilts the bar. It is
reasoned to be otolithic because semicircular canal activation or loss does not generate
a tonically maintained eye position but rather a changing eye position (nystagmus).
Although it appears that the tilting of the visual horizontal is due to ocular torsion, the
mechanism of the ocular torsion itself is speculative. It is most likely similar to the
mechanism of the spontaneous nystagmus that occurs after uVD and reflects de-
creased resting activity in otolithic secondary vestibular neurons in the ipsilesional
vestibular nucleus, owing to loss of input from primary otolithic neurons.30
CHAPTER 7 CLINICAL CHANGES IN VESTIBULAR FUNCTION 181

LATEROPULSION
Following u VD there is a position offset or "lateropulsion" of limb and body pos-
ture toward the operated side evident in the absence of visual fixation. Many clinical
tests can be used to demonstrate this offset including the Barany past pointing test, the
Fukuda vertical writing test, or the Unterberger stepping test. Although these are all
positive in the first postoperative week, they return to normal after about a month.

POSTURAL DYSEQUILIBRIUM
Despite the apparent return' to normal posture on clinical tests, posturographic
tests using a standardized posture platform test often show a permanent deficit in pos-
tural equilibrium following uVD.' 1 This deficit becomes evident if the movements of
either the platform alone, or of the platform and the visual surround together, are ref-
erenced to the body sway. Static posturography on the other hand yields paradoxical
results with those patients who were visually dependent before uVD becoming less so
and vice versa." The relationship of these posturographic abnormalities to other tests
of vestibular function and to the overall clinical state after u VD is not yet clear. There
are many unanswered questions. For example, is a permanent posturographic deficit
after u VD simply the result of the UL, or does it imply a subclinical abnormality in the
remaining labyrinth be detected before u VD? What is the relationship between an ab-
normal posturogram and symptomatic chronic vestibular insufficiency? These ques-
tions impact both the physical and surgical treatment of vestibular disorders, and these
matters are addressed in detail in other chapters.

Dynamic Motor Components

HORIZONTAL VESTIBULO-OCULAR REFLEXES

The changes which occur in the horizontal vestibulo-ocular reflex (VOR) after
uVD have been investigated in many species, including monkey and human. There-
sults obtained depend on the stimulus used. In response to low-frequency (1 Hz) low-
acceleration (100° I sec') symmetrical (i.e., sinusoidal) horizontal angular acceleration
immediately after uVD, there is a severe and asymmetrical horizontal VOR (HVOR)
deficit. The deficit persists for about a month in humans, and then improves so that
by 1 year after u VD the HVOR (in response to this type of stimulus) is normal or near
normaJ.33-35 This improvement in the low-frequency, low-acceleration HVOR is com-
monly used as an index of vestibular compensation.
If angular accelerations in a natural range are used, a different result is obtained.
In monkeys, immediately after uVD there is a profound deficit of both the ipsilesional
and contralesional HVOR, but some recovery is already apparent by the second post-
operative day. The VOR deficit is most apparent with the fastest head movements. In
monkeys, Fetter and Zee36 have shown, using a test stimulus consisting of constant
125' I sec2 angular acceleration stimulation lasting for 2 to 3 seconds, that 3 months af-
ter operation the ipsilesional HVOR still has a mean gain that is only 60% of normal
and the contralesional HVOR has a gain that is 80% of normal. Using abrupt passive
unpredictable head rotations with an angular acceleration up to 3000° per second, we
have shown in humans that the ipsilesional HVOR gain up to 2 years after operation is
on average still only 25% of normaJ12.37.38 (Fig. 7-6 and Fig. 7-7). Some patients may
182 VESTIBULAR REHABILITATION

a HEAD & EYE VELOCTIY

b
HEAD & EYE VELOCTIY
-E -E
0~0 H 00 H
too%

TIME 1()) msec TIME 100 msec

FIGURE 7-6. Single head impulses from a patient 3 years after unilateral vestibular neurectomy.
Head velocity is shown in interrupted lines, eye velocity in continuous lines. Eye velocity more or
less mirrors head velocity in response to the ampullopetal excitation produced by head rotation to-
ward the intact side. In contrast eye velocity lags head velocity from the onset of head rotation in re-
sponse to ampullofugal disfacilitation produced by head rotation tmvard the deafferented side.
(From Halmagyi, et all2).

show some recovery of gain during this time, but the gain is always less than is physi-
ologically adequate for stabilizing the retinal image during head movement. From this
one can presume that the high-acceleration response of the HVOR is never restored to
normal after uVD. It appears that dynamic equilibrium, just like static equilibrium, is
permanently impaired by uVD.
Two signs of HVOR asymmetry can be detected clinically: head-shaking nystag-
mus" and the head-impulse sign.40 Head-shaking nystagmus is horizontal nystagmus
with quick phases directed toward the normal ear that appears for 3 to 10 seconds after
20 seconds or so of active horizontal head shaking. Like any other type of peripheral
vestibular nystagmus, it is absent in the presence of visual fixation and present only in
the absence of visual fixation. It is most readily observed using Frenzel glasses. Head-
shaking nystagmus is a direct result of the inherent right-left sensory asymmetry of
each horizontal semicircular canal (Ewald's second law), signaling to a brainstem
neural network, which normally perseverates the peripheral vestibular input (the
velocity storage integrator). Head-shaking nystagmus requires a properly functioning
velocity storage integrator and can be absent after uVD.'l The head-impulse sign con-
sists of a compensatory saccade toward the intact ear during or immediately after a
rapid passive unpredictable horizontal head rotation toward the affected ear. The sign
is always positive in patients with severe unilateral loss of HSCC function. The head-
impulse test also depends on Ewald's second law and the ability of rapid head im-
pulses to drive the afferents of the single remaining HSCC to silence.

VERTICAL VESTIBULO-OCULAR REFLEXES

Normal subjects have a symmetric near-unity gain for their pitch VOR in response
to 0.4 to 1.6 Hz active sinusoidal head oscillation both in upright and in onside posi-
cHAPTER 7 CLINICAL CHANGES IN VESTIBULAR FUNCTION 183

200
,i:
EYE (deg/sec) .. ~\~·:·
~."":.

' ..~
&!<.""
.:-l:i.,-~'··
.
. . :.fi) .·
100 ...;."t·.
• !).·,.

-:J..~·
.'.4,,'

200 .·:~_j{''"'
'· .... . ..
· · ~· 1oo rJo·;ot;·"·
-- ..• ..9.
0
/" 100
1
200
t~· f3~"~"' ... _.
< - ' - ..

HEAD (deg/s ec)

"100

AF-HVOR AF-HVOR

-200
FIGURE 7-7. Horizontal eye velocity plotted as a function of horizontal head velocity for 20
horizontal head impulses in a patient \vho had undergone a left vestibular neurectomy 3 years previ-
ously. There is a profound HVOR deftcit in response to head impulses directed toward the deafter-
ented side-the ampullofugal (AF) HVOR. In contrast the HVORin response to head impulses di-
rected toward the intact side is normal-the ampullopctal (AP) HVOR. (From Halmagyi, et al 12 ).

tions.4 2 In response to high-acceleration passive head impulses in the upright position,

normal subjects have a symmetrical pitch and roll VOR. The gain of the pitch VOR
is close to 1, but the gain of the roll VOR is around 0.7. Following u VD there is a
bidirectional pitch VOR deficit, and a unidirectional deficit for ipsilesional roll head
rotations.43--48

DYNAMIC RESPONSES TO OTOLITHIC STIMULATION

During simple roll-tilt of the head, if the head is maintained in a rolled head posi-
tion (e.g., rolled so that it is toward the left shoulder), the action of the linear acceler-
ation of gravity on the inner ear is different to that with head vertical. The otolithic
sensors in the inner labyrinth sense this linear acceleration and cause a reflexive re-
sponse-a torsion or roll of both eyes, the visual axes in the direction opposite to the
direction of the roll-tilt--<:alled ocular counterrolling (OCR). OCR is one of the few
accepted measures of otolith operation,49.so but it has been shown that uVD does not
184 VESTIBULAR REHABILITATION

appear to produce consistent changes in the amplitude of OCR during roll-tilt towards
the affected or to the intact ear. There have been few reports of comparable measure-
ments of OCR before and after uVD.
Brief impulsive linear accelerations directed along an interaural axis also activate
otolithic receptors and generate a horizontal eye velocity response (an otolith-ocular
reflex). This response is equal in magnitude for accelerations directed to either side.
There have been reports that there is a smaller response for linear accelerations di-
rected to the intact ear than accelerations directed to the operated ear in post-uVD pa-
tients, but the response asymmetry declines within a few weeks.stsz

Summary of Changes in Vestibular Function

With Time after Lesions
• Angular rotation illusion: maximum during first few hours, completely resolved by
the third day after uVD.
• Roll tilt illusion: maximum during first few hours, completely resolved by third day.
• Angular acceleration perception deficit: maximum during first week; in some cases
completely resolved at 1 year but in other cases deficit still present; data incomplete.
• Linear acceleration perception deficit: maximum during the first week; largely but in-
completely resolved within 1 year.
• Spontaneous nystagmus: maximum during the first 2 days; improved at 1 week;
largely but incompletely resolved within 1 year.
• Ocular torsion: maximum during the first week; largely but incompletely resolved
within 1 year.
• Lateropulsion: maximum during the first 3 days; improved by 1 week; completely
resolved within 1 year.
• Postural dysequilibrium: maximum during first week; incompletely resolved within
1 year.
• Horizontal VOR deficit: maximum during the first week; partial recovery at 1
month; near-complete recovery within 1 year in response to low acceleration stimu-
lation; no significant recovery at all in response to high-acceleration stimulation.
• Vertical VOR deficit: persistent deficit at 1 year in response to high acceleration
stimulation; data incomplete.
• Ocular counterrolling: no long-lasting asymmetry of ocular counterrolling during
roll-tilt stimulation. Whether there is an asymmetry early after uVD remains to be
established conclusively.

FACTORS INFLUENCING THE RESTORATION

OF STATIC AND DYNAMIC EQUILIBRIUM
There is little reliable quantitative information regarding the effects of any physi-
cal or chemical interventions on the rate or extent of vestibular compensation in hu-
mans. The data that are available come from studies on cats, monkeys, and guinea
pigs; much of it is incomplete, inconclusive, or contradictory (for a recent review see
Smith and Curthoys,' Curthoys and Halmagyi,s Dieringer,s Vidal et al.' One reason for
these contradictions could be that the large number of different inputs to the vestibular
CHAPTER 7 CLINICAL CHANGES IN VESTIBULAR FUNCTION 185

nuclei may all directly or indirectly affect the activity of vestibular nucleus neurons. It
should be noted that in general the restoration of static equilibrium (i.e., static compen-
sation) is remarkably robust; very little appears to hasten or hinder it. That robustness
is in contrast to the restoration of dynamic equilibrium. Dynamic compensation, which
appears to depend at least in part in intact visual, vestibular, and proprioceptive sen-
sory inputs, is usually incomplete.

Visual Inputs

Studies in animals have shown that visual deprivation has no effect on the resolu-
tion of the static motor component, spontaneous nystagmus53a,S3b but such deprivation
impedes the recovery of roll head tilt,54• which is part of the ocular tilt reaction. Al-
though bilateral occipital lobectomy has no effect on the resolution of spontaneous
nystagmus, it does impede the recovery of the HVOR to low-acceleration stimu-
lation.53 Visual inputs do augment the diminished muscle responses to linear accelera-
tion54b and the deficient righting reflexes" that occur after uVD. Visual motion depri-
vation delays recovery of locomotor equilibrium. 56

Vestibular Inputs

There are only scant data on the effects of vestibular stimulation or deprivation
on static or dynamic compensation. In frogs, otolithic stimulation hastens whereas
otolithic deprivation delays static compensation of head tilt.57 In cat low-frequency,
combined visual-vestibular stimulation helps reverse the symmetrical deficit in HVOR
gain, which occurs in response to low-frequency stimulation following uVD," but has
no effect on the asymmetry of the HVOR. There are no data on the effects of vestibular
deprivation on vestibular compensation in mammals.

Proprioceptive Inputs

Cervical proprioceptive input could be important in static compensation because

head restraint retards resolution of head tilt and spontaneous nystagmus." So-
matosensory proprioceptive deprivation appears to retard static compensation,54b
whereas somatosensory proprioceptive stimulation appears to facilitate the restoration
of dynamic postural equilibrium.6o Acute spinal lesions can produce a temporary de-
compensation of static postural symptoms61.62

Medications

The restoration of near-normal levels of spontaneous activity in the neurons of the

ipsilesional vestibular nucleus in the absence of reinnervation could have a neuro-
chemical basis. Investigations so far have not revealed any changes in glutamate, do-
pamine, norepinephrine, acetylcholine, histamine, or serotonin receptors, which could
account for the restoration of spontaneous activity (for recent reviews see Smith and
Darlington,63 de Waele et al.' In several species treatment with an ACTH fragment ac-
186 VESTIBULAR REHABILITATION

celerates static compensation.64-07 In cat, amphetamine and trimethobenzamide may

facilitate both static and dynamic compensation.68b

Lesions
Data on the effects of lesions of the cerebellum or its connections on vestibular
compensation are contradictory. Whereas some cerebellar lesion studies68a show a
marked delay in the resolution of spontaneous nystagmus, others69 show no effect. Al-
though bilateral occipital lobectomy has no effect on the resolution of spontaneous
nystagmus, it does impede the recovery of the HVOR to low-acceleration stimula-
tion.s3 Lesions of the brain stem7Uor transcerebellar vestibular commissures71 do not
impede static compensation, at least in mammals. This suggests that input from the
contralesional (intact) vestibular nucleus is not essential for static compe!-t'sation. Sec-
tion of the brainstem commissures might, however, abolish the HVOR.SO
It is important to note that in patients with fluctuating vestibulopathies such as
Meniere's disease, attacks of vertigo are brief compared with the time required for
compensation. In humans compensation takes 3 to 5 days to get under way and a
month or more to achieve a functionally useful level. Vestibular compensation cannot
help the patient with recurrent or paroxysmal vertigo; the process is too slow. Com-
pensation does, however, help the patient to recover after a permanent uVD.

Chronic Vestibular Insufficiency Following

Unilateral Vestibular Loss
Chronic vestibular insufficiency (CVI) is a clinical syndrome consisting of gait
ataxia and oscillopsia. The gait ataxia is always most evident when visual and proprio-
ceptive inputs are disrupted, for example, when the patient tries to walk on uneven
ground in the dark. The oscillopsia is only evident during head movement, for exam-
ple, when the patient walks or runs, or when he looks rapidly from side-to-side while
driving or crossing a road.72 CVI can be due to central or peripheral vestibular lesions.
It invariably occurs in patients who have severe bilateral loss of vestibular function as
evidenced by absent HVOR responses to rapid accelerations and to ooc caloric stimu-
lation. A common cause of severe bilateral loss of vestibular function is aminoglyco-
side ototoxicity.73 Although the CVI may be asymptomatic in some patients with
severe bilateral vestibular loss during the activities of daily living, the symptoms and
abnormalities can always be demonstrated under certain provocative conditions such
as rapid head movements and eye closure on a soft surface.
We have found that certain uVD patients also experience symptomatic CVI. 74 Con-
sidering that recent data shows permanent and in some cases severe deficits of hori-
zontal and vertical VORs and postural equilibrium following u VD, this is not entirely
surprising,12.3t,45-4S What is surprising is that most uVD patients do not experience
symptoms of CVI, even though their VOR and posturographic results are apparently
indistinguishable from the results of those patients who do. From a therapeutic view-
point, it is important to determine what differences there are between patients who do
and those who do not experience symptoms of CVI following uVD. Do the sympto-
matic uVD patients have a subtle defect in the contralesional sole functioning
CHAPTER 7 CLINICAL CHANGES IN VESTIBULAR FUNCTION 187

labyrinth, or do they have some defect in the compensation process? This question is
under investigation.
The clear result from our research on uVDs over the last 12 years is that when pas-
sive head movements with angular accelerations in the natural range are used, uVD
patients show long-lasting and probably permanent deficits of their horizontal, ver-
tical and torsional VORs 4 4-48 These deficits contrast sharply with the relative ease
with which VOR gain can be changed in normal healthy subjects in response to appar-
ently comparable challenges.75.76 Many studies have shown how the gain of the hori-
zontal VOR (measured in darkness) can be changed as a result of wearing magnify-
ing or minifying lenses for even just a few hours. The VOR gain changes in a direction
so as to compensate for this visually induced "challenge." Why, then, does this well-
established adaptive plasticity of the VOR of normal healthy subjects not act to restore
the gains of the horizontal vertical and torsional VORs of post u-VD patients to a sym-
metrical value of 1.0?
There are many factors that may prevent mechanisms of VOR adaptive plasticity
from operating after u VD. In u VDs the entire vestibular afferent input from one side
has been removed, whereas in normal subjects both labyrinths project information to
the brainstem and any head movement results in stimulation of both labyrinths sym-
metrically. To achieve a stable retinal image after uVD, the VOR would have to gener-
ate different VOR gains for rotations to each side. VOR plasticity studies to date have
required increased or decreased VOR for both directions of rotation. Unilateral vestib-
ular loss disrupts central processing of vestibular information-the so-called velocity
storage integrator-and this may disrupt the neural substrates responsible for VOR
modification (see Wade et al." for a recent review). Our recent measures have shown
that after uVD the axis of eye rotation is not appropriate in that it is not parallel to the
axis of head rotation.45.46 .The axis of eye rotation changes during head rotation with
the result that the image must be smeared across the retina because of inadequate eye
velocity and inappropriate (and changing) axis of eye rotation. Both eye velocity and
axis deficits are larger for·head rotations toward the lesioned side.
Our present hypothesis is that patients who compensate well for u VD most proba-
bly learn to use other responses to minimize the effect of the eye velocity and eye axis
deficits. The result is that they probably do not experience the retinal smear resulting
from their permanently inadequate VOR. One way appears to be by eliminating
smear; a blink during the head movement will effectively prevent the retinal image
from being smeared because during the blink there will be no retinal image. Our cur-
rent measures have confirmed that during natural head movements blinks are com-
mon in normal healthy subjects. Experimental measures require subjects to keep their
eyes open during gaze shifts; in other words, the measures require subjects to suppress
their natural blink response. Patients with u VD may simply be using this natural re-
sponse mechanism to block out the retinal smear to which their doubly inadequate
VOR would otherwise expose them.

NEURAL ACTIVITY IN THE VESTIBULAR NUCLEI

DURING VESTIBULAR COMPENSATION
To appreciate the mechanisms of the uVD syndrome and vestibular compensa-
tions it is useful to look at the changes in neural activity underlying these changes in
188 VESTIBULAR REHABILITATION

behavior (for recent reviews see Goldberg and Fernandez,78 Smith and Curthoys,' and
Cirelli et al. 79

Normal Medial Vestibular Nucleus Activity

Two types of HSCC-driven neurons have been found in the medial vestibular
nuclei of monkeys, cats, and guinea pigs. Both types of vestibular nucleus neurons,
like primary vestibular neurons, discharge spontaneously (at rest) at rates sometimes
in excess of 80 impulses per second. The discharge rate of type I neurons increases
when the head acceleration is ipsilateral and decreases when the head acceleration is
contralateral. The reverse applies to type II neurons, which increase their discharge
rate in response to contralateral head accelerations and decrease their discharge rate in
response to ipsilateral head accelerations (Fig. 7-8). The reason that type I and type II
neurons respond oppositely is that, whereas type I neurons are excited by ipsilateral
HSCC primary afferent neurons and are inhibited by ipsilateral type II neurons, type II
neurons themselves are excited by contralateral type I neurons via commissural path-
ways (see Fig. 7-8). Motor and sensory equilibrium requires equal resting activity of
type I neurons in the two medial vestibular nuclei. Type I neurons drive the HVOR by
excitatory projections to abducens motoneurons and interneurons in the contralateral
abducens nucleus.

Ipsilesional Medial Vestibular Nucleus Activity

Immediately after uVD there are changes in the activity of both type I and type II
neurons in the medial vestibular nucleus on the operated side,l4,Slb,S2b,S3a,S4a,ssa The rest-
ing activity of type I neurons is decreased whereas the resting activity of type II neu-
rons is increased. The decrease in resting activity of type I neurons reflects the loss of
excitatory drive by HSCC primary afferent neurons. The increase in resting activity of
type II neurons may reflect increased excitatory drive by contralesional type I neurons,
which have become disinhibited by the decrease in the activity of contralesional type II
neurons, which are themselves normally excited by ipsilesional type I neurons (see Fig.
7-8). As well as showing a reduced resting discharge rate, immediately after uVD ip-
silesional type I neurons show a decrease in sensitivity to angular acceleration. The
sensitivity of type II neurons to angular acceleration remains unchanged. In the days
and weeks that follow, a remarkable series of changes occur in the resting activity of ip-
silesional medial vestibular nucleus neurons. The resting discharge rates of both type I
and type II neurons is restored to normal even though the medial vestibular nucleus
no longer receives any afferent drive from its labyrinth. Data so far, mainly from the
guinea pig and the gerbil, have shown a limited restoration of sensitivity of type I neu-
rons to angular accelerations. This restoration of resting activity in type I neurons
could also underlie the recovery of humans from the disabling consequences of u VD
and the restoration of normal static equilibrium.

Contralesional Medial Vestibular Nucleus Activity

Immediately after unilateral loss there is an increase in the resting activity of con-
tralesional type I neurons without much change in their sensitivity.B,S2b,S3a,S4a,S5a There
CHAPTER 7 CLINICAL CHANGES IN VESTIBULAR FUNCTION 189

/----c. Slow Phases

HEAD LEFT RIGHT
ROTATION EYE ~~,EYE
ccw Midline

o- Excitatory

.__Inhibitory

FIGURE 7~8. Schematic simplified representation of the responses in some of the identified con-
nections ofthe normal HVORpathways, in response to a counter-clockwise (i.e., left\vard) head ac-
celeration. Neurons from the left HSCC are shown in blue; neurons from the right HSCC are
shown in yellow. Primary HSCC neurons, excitatory type I medial vestibular nucleus neurons, ab-
ducens motoneurons, abducens interneurons and medial rectus motoneurons are shmvn by open
hexagonal symbols. Inhibitory type II medial vestibular nucleus neurons are shown by filled hexago-
nal symbols.

is also a decrease in the sensitivity of contralesional type II neurons without much

change in resting activity. This increase in resting activity of type I neurons is due to
decreased inhibition by type II neurons, which are themselves normally excited by ip-
silesional type I neurons now silenced. In the following days and weeks, the resting ac-
tivity of contralesional type I neurons is restored to normal and the resting activity of
contralesional type II neurons increases to above normal. These changes in the resting
190 VESTIBULAR REHABILITATION

activity of contralesional medial vestibular nucleus neurons occur despite the fact that
the ipsilesional vestibular nucleus remains isolated from its labyrinth. However, there-
markable restoration of resting activity in ipsilesional type I neurons described above
can account for the changes in activity of contralateral medial vestibular nucleus neu-
rons. The restoration of contralesional type I resting activity to normal is presumably
the result of the increased inhibition by contralesional type II neurons, now excited by
the restored resting activity of ipsilesional type I neurons. Together with the decrease
of contralesional type I resting activity to normal, there is a late decrease in contrale-
sional type I sensitivity, whereas contralesional type II sensitivity remains low.

Normal Lateral Vestibular Nucleus Activity

Primary otolithic neurons project to secondary vestibular neurons mainly in the

lateral (and descending) vestibular nuclei. The predominant response of lateral vestib-
ular nucleus neurons is an increase in firing rate in response to ipsilateral tilts (i.e., lat-
erally directed linear accelerations), the alpha response. The commissural connections
between secondary otolithic neurons are poorly understood. Unlike the commissural
connections of the HSCC secondary neurons in the medial vestibular nucleus, which
are direct and functionally inhibitory, it appears that the commissural connections be-
tween the secondary otolithic neurons in the lateral vestibular nucleus are indirect and
functionally excitatory. There are also interconnections between the lateral and medial
vestibular nuclei and some medial vestibular nucleus neurons respond to both semicir-
cular canal and otolithic stimulation. The changes that occur in the lateral vestibular
nucleus after uVD vary between the rostroventral and dorsocaudal areas of the
nucleus, which project to the cervicothoracic and lumbosacral segments of the spinal
cord, respectively.

Ipsilesional Lateral Vestibular Nucleus Activity

There is a decrease in the proportion of roll-tilt responsive neurons in the rostra-

ventral area, but not in the dorsocaudal area, as well as an overall decrease in the
average resting activity of neurons.Bia In contrast, there are increases in the number of
position-sensitive neurons, in the tilt sensitivity of dorsocaudal neurons, and in the
number of beta responses (increased firing with medially directed linear acceleration).
With compensation there is little recovery in the resting activity of either alpha or beta
neurons. The proportion of neurons in the rostroventral areas responsive to roll-tilt in-
creases to normal whereas the sensitivity remains normal. The sensitivity of dorsocau-
dal neurons decreases to normal. The proportion of position sensitivity neurons and
beta responses does not change.

Contralesional Lateral Vestibular Nucleus Activity

The proportion of roll-tilt-sensitive neurons is normal. The overall resting activity

is slightly reduced. As in the ipsilesionallateral vestibular nucleus, there is an increase
in position-sensitive neurons and beta responses, and a decrease in the roll-tilt sensitiv-
ity of neurons in the rostroventral areas. There are scanty data on the changes with
cHAPTER 7 CLINICAL CHANGES IN VESTIBULAR FUNCTION 191

compensation, but there appear to be few differences in the contralesional neuronal ac-
tivity in normal and uncompensated cats.s2a

SUMMARY

"Balance" is a common term used to describe the function of the vestibular system
itself. Coincidentally, it also applies to the neural mechanism of vestibular operation.
Unilateral loss or disease causes a massive disruption of this nicely balanced neural
system and the behavioral symptoms are the manifestations of this imbalance. As the
balance in neural activity between the two vestibular nuclei returns, behavioral symp-
toms, such as nystagmus, disappear. During this time there are significant anatomic
changes occurring in the vestibular nuclei.S,S3b,S4b,SSb,S6 Concurrently, other response
mechanisms probably substitute for those compensatory responses permanently af-
fected by the aberrant vestibular function. Neural equilibrium is disrupted by u VD in
both the MVN and the LVN with consequences for static head and eye position, and
for the dynamic response of both the canal and the otolithic systems to imposed stim-
uli. The major question still unanswered is the cause of the return of firing of neurons
in the ipsilesional vestibular nucleus. The answer will likely be found in studies of the
neurochemistry of the vestibular nuclei using detailed electrophysiology of the brain
slice or isolated whole brain preparations."

ACKNOWLEDGMENTS
This work was supported by the Australian National Health and Medical Re-
search Council, the Garnett Passe and Rodney Williams Memorial Foundation, and by
the RPA Hospital Neurology Department Trustees. We owe thanks to Ann Burgess for
her meticulous proofreading of the manuscript.

REFERENCES
1. Schaefer, KP, and Meyer, DL: Compensation of vestibular lesions. In Kornhuber, HH (ed): Handbook of
Sensory Physiology, Vol VI. Berlin, Springer Verlag, 1974, pp 463-490.
2. Precht, W: Recovery of some vestibuloocular and vestibulospinal functions following unilateral
labyrinthectomy. Prog Brain Res 64:381, 1986.
3. Precht, W, and Dieringer, N: Neuronal events paralleling functional recovery (compensation) following
peripheral vestibular lesions. Rev Oculomot Res 1:251,1985.
4. Smith, PF, and Curthoys, IS: Mechanisms of recovery following unilateral labyrinthectomy: A review.
Brain Res Rev 14:155,1989.
5. Curthoys, IS, and Halmagyi, GM: Vestibular compensation: A review of the oculomotor, neural, and clin-
ical consequences of unilateral vestibular loss. J Vestib Res 5:67, 1995.
6. Dieringer, N: "Vestibular compensation": Neural plasticity and its relations to functional recovery after
labyrinthine lesions in frogs and other vertebrates. Prog in Neurobiol46:97, 1995.
7. de Waele, C, eta!: Neurochemistry of central vestibular pathways: A review. Brain Res Rev 20:24, 1995.
8. Vidal, PP, et al: Vestibular compensation revisited. Otolaryngol Head Neck Surg 119:34, 1998.
9. Weissenstein, L, et al: Vestibular compensation in the horizontal vestibula-ocular reflex of the goldfish.
Behav Brain Res 75:127, 1996.
10. Dai, MJ, et al: Linear acceleration perception in the roll plane before and after unilateral vestibular
neurectomy. Exp Brain Res 77:315,1989.
11. Curthoys, IS, et al: Human ocular torsional position before and after unilateral vestibular neurectomy.
Exp Brain Res 85:218, 1991.
192 VESTIBULAR REHABILITATION

12. Halmagyi, GM, et a!: The human horizontal vestibula-ocular reflex in response to high-acceleration
stimulation before and after unilateral vestibular neurectomy. Exp Brain Res 81:479, 1990.
13. Smith, PF, and Curthoys, IS: Neuronal activity in the contralateral medial vestibular nucleus of the
guinea pig following unilateral labyrinthectomy. Brain Res 444:295, 1988.
14. Smith, PF, and Curthoys, IS: Neuronal activity in the ipsilateral medial vestibular nucleus of the guinea
pig following unilateral labyrinthectomy. Brain Res 444:308,1988.
15. Brookes, GB, et al: Sensing and controlling rotational orientation in normal subjects and patients with
loss of labyrinthine function. Am J Otol14:349, 1993.
16. Graybiel, A: The oculogravic illusion. AMA Arch Ophthalmol48:605, 1952.
17. Graybiel, A, and Clark, B: Validity of the oculogravic illusion as a specific indicator of otolith function.
Aerospace Med 36:1173,1965.
18. Bergenius, J, et al: The subjective horizontal at different angles of roll-tilt in patients with unilateral ves-
tibular impairment. Brain Res Bul140:385, 1996.
19. Arshi, A, et al: Roll-tilt perception using a somatosensory bar task (abstract) Association for Research in
Otolaryngology Meeting, St. Petersburg, Florida, February 5, 1998.
20. Tribukait, A, et al: The subjective visual horizontal for different body tilts in the roll plane-Characteri-
zation of normal subjects. Brain Res Bull40:375, 1996.
21. Zee, DS: Ophthalmoscopy in the examination of patients with vestibular disorders. Ann Neural 3:373,
1978.
22. Halmagyi, GM, et al: Ocular tilt reaction with peripheral vestibular lesion. Ann Neurol6:80, 1979.
23. Wolfe, GI, et al: Ocular tilt reaction resulting from vestibuloacoustic nerve surgery. Neurosurg 32:417,1993.
24. Vibert, D, et al: Ocular tilt reaction associated in sudden idiopathic unilateral peripheral cochleovestibu-
lar loss. J Otorhinol Relat Spec 57:310, 1995.
25. Safran, AB, et al: Skew deviation following vestibular neuritis. Am J Ophthalmol118:238, 1994.
26. Wade, S, and Curthoys, IS: The effect of ocular torsional position on perception of the roll-tilt of visual
stimuli. Vision Research 37:1071, 1997.
27. Vibert, D, et al: Diplopia from skew deviation in unilateral peripheral vestibular lesions. Acta Otolaryn-
gol116,170, 1996.
28. Tabak, S, et al: Deviation of the subjective vertical in long-standing unilateral vestibular loss. Acta Oto-
laryngol117:1, 1997.
29. Friedmann, G: The judgement of the visual vertical and horizontal with peripheral and central vestibu-
lar lesions. Brain 93:313, 1970.
30. Pompeiano, 0, et al: Central compensation of vestibular deficits. II. Influences of roll tilt on different-size
lateral vestibular neurons after ipsilateral labyrinth deafferentation. J Neurophysiol52:18, 1984.
31. Black, FO, et al: Effects of unilateral loss of vestibular function on the vestibula-ocular reflex and pos-
tural control. Ann Otol Rhinal Laryngol98:884, 1989.
32. Lacour, M, et al: Sensory strategies in human postural control before and after unilateral vestibular neu-
rotomy. Exp Brain Res 115:300, 1997.
33. Jenkins, HA: Long-term adaptive changes of the vestibula-ocular reflex in patients following acoustic
neuroma surgery. Laryngoscope 95:1224,1985.
34. Paige, GD: Nonlinearity and asymmetry in the human vestibula-ocular reflex. Acta Otolaryngol (Stockh)
108,1, 1989.
35. Takahashi, M, eta!: Recovery of vestibula-ocular reflex and gaze disturbance in patients with unilateral
loss of labyrinthine function. Ann Otol Rhinal Laryngol93:170, 1984.
36. Fetter, M, and Zee, DS: Recovery from unilateral labyrinthectomy in rhesus monkey. J Neurophysiol
59370,1988.
37. Tabak, S, et al: Gain and delay of human vestibula-ocular reflexes to oscillation and steps of the head by
a reactive torque helmet. I. Normal subjects. Acta Otolaryngol. 117:785,1997.
38. Tabak, S, et al: Gain and delay of human vestibula-ocular reflexes to oscillation and steps of the head by
a reactive torque helmet. II. Vestibular-deficient subjects. Acta Otolaryngol117:796, 1997.
39. Hain, TC, et al: Head-shaking nystagmus in patients with unilateral peripheral vestibular lesions. Am J
Otolaryngol8:36, 1987.
40. Halmagyi, GM, and Curthoys, IS: A clinical sign of canal paresis. Arch Neuro145:737, 1988.
41. Fetter, M, et al: Head-shaking nystagmus during vestibular compensation in humans and rhesus mon-
keys. Acta Otolaryngol (Stockh) 110:175,1990.
42. Baloh, RW, and Derner, J: Gravity and the vertical vestibula-ocular reflex. Exp Brain Res 83:427, 1991.
43. Allum, JH, eta!: Long-term modifications of vertical and horizontal vestibula-ocular reflex dynamics in
man. I. After acute unilateral peripheral vestibular paralysis. Acta Otolaryngol (Stockh) 105:328,1988.
44. Aw, ST, et al: Unilateral vestibular deafferentation causes permanent impairment of the human
vestibula-ocular reflex in the pitch plane. Exp Brain Res 102:121, 1994.
45. Aw, ST, et al: Three dimensional kinematics of the human vestibuloocular reflex response during high-
acceleration head rotations after unilateral vestibular deafferentation and semicircular canal occlusion. J
Neurophysiol76:4009, 1996.
46. Aw, ST, et al: Three dimensional kinematics of the human vestibuloocular reflex response during high~
acceleration head rotations: Speed gain and misalignment angle. J Neurophysiol76:4021, 1996.
CHAPTER 7 CLINICAL CHANGES IN VESTIBULAR FUNCTION 193

47. Aw, ST, et al: Compensation of the human vertical vestibula-ocular reflex following occlusion of one
vertical semicircular canal is incomplete. Exp Brain Res 103:471, 1995.
48. Cremer, PD, eta!: Semicircular canal plane head impulses detect absent function of individual semicir-
cular canals. Brain 121:699, 1998.
49. Diamond, SG, and Markham, CH: Binocular counterrolling in humans with unilateral labyrinthectomy
and in normal controls. Ann NY Acad Sci 374:69, 1981.
50. Diamond, SG, and Markham, CH: Ocular counterrolling as an indicator of vestibular otolith function.
Neurology 33:1460,1983.
51. Bronstein, AM, eta!: Compensatory otolithic slow phase eye movement responses to abrupt linear head
motion in the lateral direction. Findings in patients with labyrinthine and neurological lesions. Acta
Otolaryngol Suppl (Stockh) 481:42, 1991.
52. Lempert, T, et al: Effect of otolith dysfunction-impairment of visual acuity during linear head motion
in labyrinthine defective subjects. Brain 120:1005,1997.
53a. Fetter, M, et al: Effect of lack of vision and of occipital lobectomy upon recovery from unilateral
labyrinthectomy in rhesus monkey. J :--Jeurophysiol 59:394, 1988.
53b. Smith, PF, et al: The effect of visual deprivation on vestibular compensation in the guinea pig. Brain Res
364:195, 1986.
54a. Putkonen, PT, et al: Compensation of postural effects of hemilabyrinthectomy in the cat. A sensory sub-
stitution process? Exp Brain Res 28:249, 1977.
54b. Lacour, M, et al: Compensation of postural reactions to fall in the vestibular neurectomized monkey.
Role of the remaining labyrinthine afferences. Exp Brain Res 37:563, 1979.
55. Igarashi, M, and Gutierrez, 0: Analysis of righting reflex in cats with unilateral and bilateral
labyrinthectomy. Otorhinolaryngology 445:279, 1983.
56. Xerri, C, and Zennou, Y: Sensory, functional and behavioural substitution processes in vestibular com-
pensation. In Lacour, M, et al (eds): Vestibular Compensation: Facts, Theories and Clinical Perspectives.
Paris, Elsevier, 1989, pp 35-58.
57. Flohr, H, et al: Concepts of vestibular compensation. In Flohr, H, and Precht, W (eds): Lesion-induced
Neuronal Plasticity in Sensorimotor Systems. Berlin, Springer-Verlag, 1981.
58. Maioli, C, and Precht, W: On the role of vestibula-ocular reflex plasticity in recovery after unilateral pe-
ripheral vestibular lesions. Exp Brain Res 59:267,1985.
59. Pettorossi, VE, and Petrosini, L: Tonic cervical influences on eye nystagmus following labyrinthectomy:
Immediate and plastic effects. Brain Res 324:11, 1984.
60. Igarashi, M: Physical exercise and the acceleration of vestibular compensation. In Lacour, M, et al (eds):
Vestibular Compensation: Facts Theories and Clinical Perspectives. Elsevier, Paris, 1989 pp 131-144.
61. Jensen, DW: Reflex control of acute postural asymmetry and compensatory symmetry after a unilateral
vestibular lesion. Neuroscience 4:1059, 1979.
62. Jensen, DW: Vestibular compensation: tonic spinal influence upon spontaneous descending vestibular
nuclear activity. Neuroscience 4:1075,1979.
63. Smith, PF, and Darlington, CL: Neurochemical mechanisms of recovery from peripheral vestibular le-
sions (vestibular compensation). Brain Res Review 16:117, 1991.
64. Flohr, H, and Luneburg, U: Influence of melanocortin fragments on vestibular compensation. In Lacour,
M, et al (eds): Vestibular Compensation: Facts Theories and Clinical Perspectives. Elsevier, Paris, 1989.
pp 161-174.
65. Gilchrist, DP, et al: Effects of flunarizine on ocular motor and postural compensation following periph-
eral vestibular deafferentation in the guinea pig. Pharmacal Biochem Behav 44:99,1993.
66. Gilchrist, DP, et al: Evidence that short ACTH fragments enhance vestibular compensation via direct ac-
tion on the ipsilateral vestibular nucleus. Neuroreport 7:1489,1996.
67. Gilchrist, DP, et al: ACTH(4-10) accelerates ocular motor recovery in the guinea pig following vestibular
deafferentation. Neurosci Lett 118:14,1990.
68a. Courjon, JH, et al: The role of the flocculus in vestibular compensation after hemilabyrinthectomy.
Brain Res 239:251, 1982.
68b. Peppard, SB: Effect of drug therapy on compensation from vestibular injury. Laryngoscope 96:878,1986.
69. Haddad, GM, et al: Compensation of nystagmus after VIIIth nerve lesions in vestibulo-cerebellec-
tomized cats. Brain Res 135:192, 1977.
70. Smith, PF, et al: Vestibular compensation without brains tern commissures in the guinea pig. Neurosci
Lett 65:209, 1986.
71. Newlands, SD, and Perachio, AA: Effects of commisurotomy on vestibular compensation in the gerbil.
Soc Neurosci Abstr 12:254, 1986.
72. Halmagyi, GM, and Henderson, CJ: Visual symptoms of vestibular disease. Aust N Z J Ophthalmol
16:177,1988.
73. Halmagyi, GM, et al: Gentamicin vestibulotoxidty. Otolaryngol Head Neck Surg 111:571, 1994.
74. Halmagyi, GM: Vestibular insufficiency following unilateral vestibular deafferentation. Aust J Oto-
laryngol1:510, 1994.
75. Berthoz, A, and Melvill Jones, G (eds): Adaptive Mechanisms in Gaze Control: Facts and Theories. Else-
vier, Amsterdam, 1985.
 194 VESTIBULAR REHABILITATION

76. Lisberger, SG: The neural basis for learning of simple motor skills. Science 242:728, 1988.
77. Wade, S, et al: Time constant of nystagmus slow phase velocity to yaw-axis rotation as a function of the
severity of unilateral caloric paresis. Am J Otol20, 1999.
78. Goldberg, JM, and Fernandez, C: The vestibular system. In: Handbook of Physiology. The Nervous Sys-
tem. Sensory Processes. Bethesda, MD, Physiological Society, 1981. pp 977-1022.
79. Cirelli, C, et al: c-fos Expression in the rat brain after unilateral labyrinthectomy and its relation to the
uncompensated and compensated stages. Neuroscience 70:515, 1996.
80. Precht, W, et a!: A mechanism of central compensation of vestibular function following hemil-
abyrinthectomy. J Neurophysiol29:996, 1966.
81a. Xerri, C, et al: Central compensation of vestibular deficits. I. Response characteristics of lateral vestibu-
lar neurons to roll tilt after ipsilateral labyrinth deafferentation. J Neurophysiol50:428, 1983.
81b. Markham, CH, et al: The contribution of the contralateral labyrinth to second order vestibular neuronal
activity in the cat. Brain Res 138:99, 1977.
82a. Lacour, M, et al: Central compensation of vestibular deficits. III. Response characteristics of lateral ves-
tibular neurons to roll tilt after contralateral labyrinth deafferentation. J Neurophysiol54:988, 1985.
82b. Ris, L, et al: Neuronal activity in the ipsilateral vestibular nucleus following unilateral labyrinthectomy
in the alert guinea pig. J Neurophysiol74:2087, 1995.
83a. Ris, L, et al: Dissociations between behavioural recovery and restoration of vestibular activity in the
unilabyrinthectomized guinea-pig. J Physiol500:509, 1997.
83b. Gacek, RR, et al: Ultrastructural changes in vestibula-ocular neurons following vestibular neurectomy
in the cat. Ann Otol Rhinal Laryngol97:42, 1988.
84a. Newlands, SO, and Perachio, AA: Compensation of horizontal canal related activity in the medial ves-
tibular nucleus following unilateral labyrinth ablation in the decerebrate gerbil. I. Type I neurons. Exp
Brain Res 82:359, 1990.
84b. Gacek, RR, et al: Ultrastructural changes in contralateral vestibulo-ocular neurons following vestibular
neurectomy in the cat. Acta Otolaryngol Suppl (Stockh) 477:1, 1991.
85a. Newlands, SO, and Perachio, AA: Compensation of horizontal canal related activity in the medial ves-
tibular nucleus following unilateral labyrinth ablation in the decerebrate gerbil. II. Type II neurons. Exp
Brain Res 82:373,1990.
85b. Gacek, RR, et al: Morphologic correlates of vestibular compensation in the cat. Acta Otolaryngol Suppl
(Stockh) 462,1, 1989.
86. Gacek, RR, and Schoonmaker, JE: Morphologic changes in the vestibular nerves and nuclei after
labyrinthectomy in the cat: a case for the neurotrophin hypothesis in vestibular compensation. Acta
Otolaryngologica 117:244, 1997.
87. Vibert, N, et al: The vestibular system as a model of sensorimotor transformations. A combined in vivo
and in vitro approach to study the cellular mechanisms of gaze and posture stabilization in mammals.
Prog Neurobiol51:243, 1997.

View publication stats