Vertigo and Imbalance: Clinical Neurophysiology of the Vestibular System
Handbook of Clinical Neurophysiology, Vol. 9
S.D.Z. Eggers and D.S. Zee (Vol. Eds.)
# 2010 Elsevier B.V. All rights reserved
CHAPTER 25
Vestibular neuritis
Michael Strupp* and Thomas Brandt
Department of Neurology, University of Munich, Klinikum Grosshadern, D-81377 Munich, Germany
25.1. Introduction
Vestibular neuritis (VN), also known as acute unilat-
eral vestibular paralysis or vestibular neuronitis, is
the third most common cause of peripheral vestibular
vertigo. VN accounts for 8% of the patients presenting
in a dizziness unit (Brandt et al., 2005) and has an
incidence of about 3.5/100,000 population (Sekitani
et al., 1993). It was first described by Rutin in 1909
(Ruttin, 1909) and later by Nylen in 1924 (Nylen,
1924). Key signs and symptoms of vestibular neuritis
are an acute onset of sustained rotatory vertigo,
postural imbalance with Romberg fall (toward the
affected ear), horizontal-torsional spontaneous nys-
tagmus (toward the unaffected ear), and nausea. The
head-impulse test (Halmagyi and Curthoys, 1988)
and caloric testing invariably show a deficit of the
vestibulo-ocular reflex (VOR).
In the past, either inflammation of the vestibular
nerve (Ruttin, 1909; Nylen, 1924; Dix and Hallpike,
1952) or labyrinthine ischemia (Lindsay and Hemen-
way, 1956) were proposed to be the cause of vestibu-
lar neuritis. Nowadays a viral etiology is favored.
The evidence, however, remains circumstantial
(Nadol, 1995; Brandt, 1999; Baloh, 2003). Vestibular
nerve histopathology in cases of vestibular neuritis
was similar to that in single cases of herpes zoster oti-
cus (Schuknecht and Kitamura, 1981). Herpes simplex
virus type (HSV I) DNA was detected in about two-
thirds of autopsied human vestibular ganglia by poly-
merase chain reaction (PCR) (Furuta et al., 1993;
Arbusow et al., 1999). This indicates that the vestibu-
lar ganglia like other cranial nerve ganglia are latently
infected by HSV-1 (Nahmias and Roizman, 1973;
*
Correspondence to: Michael Strupp, MD, Department of
Neurology, University of Munich, Klinikum Grosshadern,
Marchioninistrasse 15, D-81377 Munich, Germany.
Tel.: þ49-(0)89-7095-3678; fax: þ49-(0)89-7095-6673.
E-mail: Michael.Strupp@med.uni-muenchen.de (M. Strupp).
315
Theil et al., 2001, 2003). A similar etiology is also
assumed for Bell’s palsy and strongly supported by
the demonstration of HSV-1 DNA in the endoneurial
fluid of the facial nerve of the affected subjects
(Murakami et al., 1996).
VN is most likely a partial rather than a complete
vestibular paresis, with predominant involvement of
the horizontal and anterior semicircular canals
(the posterior semicircular canal is spared) and the
utricle. This is likely to be due to anatomical differ-
ences between the bony channels of the superior
and the inferior canal. It was shown that the lateral
bony channel of the superior vestibular nerve is
seven times longer than the inferior vestibular chan-
nel. Further, there are a larger percentage of bony
spicules occupying the superior vestibular compared
with the inferior vestibular or singular channels
(Gianoli et al., 2005). Therefore, the superior nerve
is more susceptible to swelling. The condition mainly
affects adults, ages 30–60, and has a spontaneous
course of gradual recovery within 1–6 weeks. Recov-
ery from vestibular neuritis is due to a combination
of: (a) peripheral restoration of labyrinthine function,
which is usually incomplete (Okinaka et al., 1993)
but can be improved by early treatment with corti-
costeroids (Strupp et al., 2004); with this treatment
the recovery is 62% within 12 months; (b) mainly
somatosensory and visual substitution; and (c) central
compensation, which can be improved by vestibular
exercise (Strupp et al., 1998b).
The diagnosis of VN is based on the simple
assessment of an acute vestibular tone imbalance
associated with a unilateral peripheral vestibular loss
(head-impulse test and caloric testing) after clinical
exclusion of other peripheral as well as central
vestibular disorders that may mimic VN, namely
“vestibular pseudoneuritis” due to a lesion in the
root entry zone of the VIIIth nerve or vestibulo-
cerebellum.
In this chapter, the clinical syndrome, sponta-
neous course, findings of laboratory examinations,
316
etiology and pathophysiology, differential diagnoses,
and, finally, the management will be described.
25.2. Clinical syndrome
Key signs and symptoms of VN (Fig. 1) are an acute/
subacute onset of sustained (1) rotational vertigo
(contraversive) with pathological adjustments of the
subjective visual vertical and perceived straight-
ahead; (2) postural imbalance with Romberg fall
and past-pointing (ipsiversive); (3) horizontal sponta-
neous nystagmus (contraversive) with a torsional
component associated with oscillopsia; and (4) nau-
sea and vomiting. Short attacks of rotatory vertigo
occasionally precede the onset of manifest loss of
function by a few days.
Ocular motor evaluation reveals an incomplete
ocular tilt reaction (see below), apparent horizontal
saccadic pursuit, gaze-evoked nystagmus toward the
fast phase of the spontaneous nystagmus (obeying
Alexander’s law), and a directional preponderance
of optokinetic nystagmus (contraversive to the
lesion). All of these symptoms are secondary to
spontaneous nystagmus, which indicates vestibular
L 30⬚ C
nystagmus
vertigo
falling tendency
eye torsion
Subjective visual vertical
Subjective straight ahead
Fig. 1. Ocular signs, perception (vertigo, subjective visual vertical,
and subjective straight ahead), and posture in the acute stage of
right-sided vestibular neuritis. Spontaneous vestibular nystagmus
is always horizontal-rotatory away from the side of the lesion (best
observed with Frenzel’s glasses). The initial perception of apparent
body motion (vertigo) is also directed away from the side of the
lesion, whereas measurable destabilization (Romberg fall) is
always toward the side of the lesion. The latter is the compensatory
vestibulo-spinal reaction to the apparent tilt.
M. STRUPP AND T. BRANDT
tone imbalance in the yaw (horizontal) and roll (tor-
sional) planes. Hearing loss is not found in VN, and
the detection of any neurological deficit besides the
above-indicated signs and symptoms should raise
doubts about the diagnosis of VN.
A suspected diagnosis is supported by demon-
strating a unilateral deficit of the VOR by the head-
impulse test (Halmagyi and Curthoys, 1988) and a
hypo- or unresponsiveness in bithermal caloric
testing (Fig. 2; horizontal semicircular canal paresis
of the labyrinth opposite to the fast-phase of the
spontaneous nystagmus). There is, however, no
R
L
time
44⬚ C 44⬚ C
30⬚ C
B
Fig. 2. Eye signs in the acute stage of right VN. (A) Spontaneous
VN is always horizontal-rotatory away from side of the lesion. It is
best observed with Frenzel’s glasses, since fixation largely sup-
presses nystagmus (top). During lateral gaze and fixation of a
stationary target, spontaneous nystagmus is inhibited when gaze
is directed toward the affected ear and increased when gaze is
directed toward the unaffected ear. (B) Caloric irrigation of the
external auditory canal (caloric test) demonstrates unresponsive-
ness of the affected right horizontal semicircular canal but normal
responses in the left horizontal semicircular canal (bottom). Spon-
taneous vestibular nystagmus to the left causes a directional bias of
the recorded eye movements during caloric irrigation (modified
from Brandt, 1999).
DISEASES AND TREATMENTS
pathognomonic test or sign for VN as a clinical entity.
In a strict sense, only an acute unilateral peripheral
vestibular hypofunction with horizontal semicircular
canal paresis can be diagnosed by the proposed pro-
cedure. Differential diagnosis (see section 25.6)
may be difficult, especially the exclusion of “vestib-
ular pseudoneuritis”.
25.2.1. Eye movements
The nystagmus is always horizontal, but it has a tor-
sional component (beating counterclockwise-left or
clockwise-right from the patient’s point of view).
The nystagmus is typically reduced in amplitude by
fixation (visual fixation suppression) and enhanced
by eye closure, Frenzel’s (high plus) lenses, and dur-
ing convergence. According to Alexander’s law,
amplitude and slow-phase velocity are increased with
gaze shifts toward the direction of the fast phase, and
decreased with gaze shifts toward the direction of the
slow phase of the nystagmus. This may mimic unilat-
eral gaze-evoked nystagmus in a patient with moder-
ate spontaneous nystagmus that is completely
suppressed by visual fixation straight ahead but still
present with the gaze directed toward the fast phase.
Using a motor-driven 3-D rotating chair, Fetter and
Dichgans (1996) studied 3-D properties of the vestib-
ulo-ocular reflex in 16 patients in the acute stage of
VN. Their measurements support the view that VN
is a partial rather than a complete unilateral vestibu-
lar lesion (see below) (Büchele and Brandt, 1988)
and that this partial lesion affects the superior divi-
sion of the vestibular nerve including the afferents
from the horizontal and anterior semicircular canals
and the utricle (Fetter and Dichgans, 1996): “In all
patients, spontaneous nystagmus axes clustered
between the direction expected with involvement of
just one horizontal semicircular canal and the direc-
tion expected with combined involvement of the hor-
izontal and anterior semicircular canals on one side.
Likewise, dynamic asymmetries were found only
during rotations about axes which stimulated the ipsi-
lesional horizontal or ipsilesional anterior semicircu-
lar canals. No asymmetry was found when the
ipsilesional posterior semicircular canal was stimu-
lated.” This analysis was based on physiological data
showing that electrical stimulation of single semicir-
cular canal nerves elicits eye movements in the plane
of the canal and that combinations of canal lesions
should result in a direction of spontaneous nystagmus
reflecting the weighted vector sum of the axes of the
317
involved canals. A significant directional preponder-
ance of optokinetic nystagmus (OKN) may be
another consequence of the peripheral lesion rather
than from involvement of the brainstem or cerebel-
lum. These vestibular-induced differences in OKN
slow-phase velocity can be as large as 70% and are
due to enhancement toward the side of the lesion
and depression in the opposite horizontal direction
(Brandt et al., 1978). The interaction is not purely
additive or subtractive; a feedforward optokinetic
gain control of the vestibular component (multi-
plication) is involved before the two signals are
combined.
An incomplete ocular tilt reaction with ocular
torsion (Fig. 3) and perceived tilts of the subjective
visual vertical has been described in most patients
with VN (Böhmer and Rickenmann, 1995). Although
mentioned in the literature before (Safran et al.,
1994; Vibert et al., 1996), patients with VN do not
have a skew deviation. This typically occurs in ves-
tibular pseudoneuritis (Cnyrim et al., 2008) and can
also be found in a complete de-afferentation as
occurs in zoster oticus (Arbusow et al., 1997). These
signs indicate a vestibular tone imbalance in the roll
plane induced by involvement of the anterior semi-
circular canal, otolith function, or both. The superior
division of the vestibular nerve innervates not only
the cristae of the horizontal and anterior semicircular
canals but also the maculae of the utricle and the
anterosuperior part of the saccule. It is possible that
a lesion of only the superior division results in ocular
torsion and tilts of the subjective visual vertical,
whereas a lesion of both the superior and the inferior
divisions of the VIIIth nerve results in ocular torsion,
tilts of the subjective visual vertical, and skew devia-
tion. As mentioned above, we have seen evidence for
the latter in a patient with herpes zoster oticus, which
manifested with a complete ocular tilt reaction,
Fig. 3. Fundus photograph of the left eye in a patient with vestib-
ular neuritis on the left side showing 20 cyclorotation (OT)
toward the left (excyclotropia, torsion of the papilla-fovea line
clockwise from the viewpoint of the observer) on day 3 after
symptom onset. On day 30 the ocular torsion was within the nor-
mal range (x  2 SDs: 1 to 11.5 ).
318
including skew deviation (ipsilesional eye under-
most) and showed a contrast enhancement of the
superior and inferior parts of the VIIIth nerve on
MRI (Arbusow et al., 1997).
25.2.2. High-frequency defect of VOR in
permanent peripheral vestibular lesion
It is possible to demonstrate a permanent, direction-
specific, high-frequency defect of the VOR in
patients whose semicircular canal function has not
been restored. This is predicted by Ewald’s second
law (Ewald, 1892), which states that horizontal canal
function has a directional asymmetry, with ampullo-
petal stimulation (cupula deflection toward the utri-
cle) being more effective than ampullofugal
stimulation (cupula deflection away from the utricle).
Electrophysiological studies of primary vestibular
afferents in the monkey during constant angular
accelerations have confirmed the law (Goldberg and
Fernandez, 1982). Gain asymmetries have been
demonstrated in humans after acute unilateral periph-
eral vestibular lesions, showing rotation toward the
side with the lesion (ampullopetal stimulation of the
remaining intact labyrinth) resulted in lower gain
than rotation away from the side with the lesion.
Unpredictable, passive rotational head impulses with
accelerations up to 4000 /s2 demonstrated consider-
able asymmetries in VOR gain even 1 year after a
unilateral peripheral vestibular lesion (Halmagyi
et al., 1990). That there is no central compensation
of the directional asymmetry of high-frequency canal
function was also demonstrated by Halmagyi and
Curthoys (1988) using a simple VOR bedside test
(Fig. 4). When the head was rapidly rotated toward
the side with the lesion all 12 patients who had
undergone unilateral vestibular neurectomy made
clinically evident, oppositely directed, compensatory
re-fixation saccades. This indicates a unilateral high-
frequency deficiency of the VOR, which is produced
by functional asymmetry of the remaining labyrinth.
Furthermore, the well-known clinical method of
provoking spontaneous nystagmus by head-shaking
with Frenzel’s glasses (Kamei, 1975) reveals a uni-
lateral labyrinthine loss even if it is apparently com-
pensated centrally. It was also shown that horizontal
head-shaking in yaw elicits horizontal nystagmus
with slow phases that are initially directed toward
the side of the lesion and upward (fast phases
directed toward the unaffected ear) (Hain et al.,
1987). They assume that head-shaking nystagmus is
M. STRUPP AND T. BRANDT
generated by the combination of a central velocity-
storage mechanism, which perseverates peripheral
vestibular signals, and Ewald’s second law, which
states that high-velocity vestibular excitatory inputs
are more effective than inhibitory inputs.
25.2.3. Vertigo and posture
In VN the fast phase of the spontaneous rotational
nystagmus (Fig. 1) and the initial perception of
apparent body motion are directed away from the
side of the lesion, and the postural reactions initiated
by vestibulo-spinal reflexes are usually opposite to
the direction of vertigo. These result in both the
Romberg fall and in past-pointing toward the side
of the lesion. Patients with this type of vertigo often
make confusing and contradictory statements about
the directionality of their symptoms. In actual fact,
there are two sensations of opposite directions, and
the patient may be describing one or the other. The
first is a purely subjective sense of self-motion in
the direction of the nystagmus fast phases, which is
not associated with any measurable body sway. The
second is the compensatory vestibulo-spinal reaction
resulting in objective, measurable destabilization and
a possible Romberg fall in the direction opposite to
the fast phases (Brandt and Daroff, 1980). Subjective
straight-ahead and tilts of the perceived vertical can
be determined psychophysically as the perceptual
consequence of vestibular tone imbalance in yaw
(horizontal semicircular canal) and roll (anterior
semicircular canal). The direction of pathological
deviation – as adjusted by the patient – and the
Romberg fall are ipsiversive to the affected ear.
Since the severity of tone imbalance can be measured
in degrees, it is possible to assess quantitatively the
recovery of spatial disorientation during the course
of the illness. Furthermore, the efficacy of physical
and medical treatment of the condition can also be
measured in this way (Strupp et al., 1998b).
25.3. Epidemiology, spontaneous course,
recurrences, and complications
25.3.1. Epidemiology
VN has an incidence of about 3.5 per 100,000 popula-
tion (Sekitani et al., 1993). In our dizziness unit VN is
the third most common cause of peripheral vestibular
disorders (first Benign paroxysmal positional vertigo
BPPV, second Ménière’s disease). It accounts for
 (1)

HEAD ROTATION HEAD ROTATION

(2)

EYE MOVEMENT EYE MOVEMENT

(3)

SACCADE

(3)

amplitude
L L
(2) (3)
(1) (2) (1)
R R
time
A B

Fig. 4. Clinical examination of the horizontal vestibulo-ocular reflex (VOR) by the head-impulse test (Halmagyi and Curthoys, 1988). To test
the horizontal VOR, the examiner holds the patient’s head between both hands, asks her to fixate a target in front of her eyes, and rapidly and
arbitrarily turns the patient’s head horizontally to the left and then to the right. This rotation of the head in a healthy subject causes rapid com-
pensatory eye movements in the opposite directions (A). In cases of unilateral labyrinthine loss the patient is not able to generate the VOR-
driven fast contraversive eye movement and has to perform a corrective (catch up) saccade to re-fixate the target. Part B explains the findings
in a patient with a loss of the right horizontal canal. During rapid head rotations toward the affected right ear, the eyes move with the head to
the right and the patient has to perform a re-fixation saccade to the left; the latter can be easily detected by the examiner (C).
320
about 8% of the patients (Brandt et al., 2005). Its usual
age of onset is between 30 and 60 years (Depondt,
1973), and age distribution plateaus between 40 and
50 years (Sekitani et al., 1993). There is no significant
sexual difference. VN is relatively rare in children, but
it has repeatedly been reported to occur in children
aged 3–15 years (Tahara et al., 1993). VN in children
seems to differ from VN in adults by: (1) a higher fre-
quency of preceding upper respiratory tract infections;
(2) a more rapid recovery from vertigo and nystagmus;
and (3) a better prognosis as to the recovery rate of
labyrinthine function assessed by follow-up caloric
testing (Shirabe, 1988; Sekitani et al., 1993; Tahara
et al., 1993).
25.3.2. Spontaneous course
There is usually a sudden onset of the disease (some-
times preceded by a short vertigo attack hours or
days earlier) with rotational vertigo, oscillopsia,
impaired fixation, postural imbalance, nausea, and
often vomiting. Patients feel severely ill and prefer
to stay immobilized in bed. They avoid head move-
ments, which exaggerate symptoms, until the vertigo,
postural imbalance, and nausea subside, usually after
1–3 days. After 5–7 days spontaneous nystagmus is
largely suppressed by fixation in the primary posi-
tion, although – depending on the severity of the
canal palsy – it is still present for 2–3 weeks with
Frenzel’s glasses and on lateral gaze directed away
from the lesion. After recovery of peripheral vestibu-
lar function, spontaneous nystagmus transiently
reverses its direction in some patients (“Erholungs-
nystagmus”), i.e., when the centrally compensated
lesion regains function. “Erholungsnystagmus” then
reflects a tone imbalance secondary to compensation.
Bechterew’s phenomenon, a reversal of post-
unilateral labyrinthectomy spontaneous nystagmus
occurring after contralateral labyrinthectomy in ani-
mals or humans (Zee et al., 1982; Katsarkas and
Galiana, 1984), is produced by a similar mechanism.
After 1–6 weeks most of the patients feel symp-
tom-free, even during slow body movements, but
actual recovery depends on whether and how quickly
functional restitution of the vestibular nerve occurs
during “central compensation” (Brandt et al., 1997b)
and possibly on how much physical exercise the
patient has done. Rapid head movements, how-
ever, may still cause slight oscillopsia of the visual
scene and impaired balance for a second in those
who do not regain normal labyrinthine function (see
M. STRUPP AND T. BRANDT
above). This explains why only 34 (57%) of 60
patients with VN reported complete relief from sub-
jective symptoms at long-term follow-up (Okinaka
et al., 1993), a figure that roughly corresponds to
the 50–70% complete recovery rate of labyrinthine
function assessed by caloric irrigation (Meran and
Pfaltz, 1975; Ohbayashi et al., 1993; Okinaka et al.,
1993).
25.3.3. Recurrence rate
In a recent long-term follow-up study (5.7–20.5
years, mean 9.8 years) on a total of 103 patients with
VN, only two patients (1.9%) had developed a
second VN, 29–39 months after the first (Huppert
et al., 2006). It affected the contralateral ear in both
and caused less severe, distressing vertigo and pos-
tural imbalance. Thus, unlike Bell’s palsy and sudden
hearing loss, a relapse in the same ear does not occur.
25.3.4. Complications
In 10–15% of patients with vestibular neuritis a typi-
cal, benign paroxysmal positioning vertigo develops
in the affected ear within a few weeks (Büchele and
Brandt, 1988; Huppert et al., 2006). It is possible that
the otoconia loosen during the additional inflamma-
tion of the labyrinth (HSV-1 DNA was also found
in the human labyrinth (Arbusow et al., 2000)), and
this eventually results in canalolithiasis. Patients
should be warned about this possible complication,
because there are therapeutic liberatory maneuvers
that can quickly free the patient of his complaints.
The second important complication is that vestibular
neuritis can develop into a somatoform phobic pos-
tural vertigo (Brandt and Dieterich, 1986; Brandt,
1996). The traumatic experience of a persisting
organic rotatory vertigo leads to fearful introspection
resulting in a somatoform, fluctuating, and persistent
postural vertigo, which is reinforced in specific situa-
tions and culminates in a phobic behavior of
avoidance.
25.4. Laboratory examinations
25.4.1. Caloric testing
The principal diagnostic marker of the disease is an
initial paresis of the horizontal semicircular canal
on the affected side; this can be demonstrated
by caloric tests (Fig. 2). Since there is a large
DISEASES AND TREATMENTS
intersubject variability of the nystagmus induced by
caloric irrigation and a small intraindividual variabil-
ity of the response of the right and the left labyrinths
in healthy subjects, Jongkees’s “vestibular paresis
formula” (Jongkees et al., 1962; Honrubia, 1994):
[((R30 þ R44 )  (L30 þ L44 ))/(R30 þ R44 þ
L30 þ L44 )  100] should be used to determine ves-
tibular paresis, where, for instance, R30 is the mean
peak slow-phase velocity during caloric irrigation with
30 C water. Vestibular paresis is usually defined as
> 25% asymmetry between the two sides (Honrubia,
1994). This formula allows a direct comparison of
the function of the horizontal semicircular canals of
both labyrinths, which is important due to the large
inter-individual variability of caloric excitability.
Jongkees’s “vestibular paresis formula” is also useful
for treatment studies in VN (Strupp et al., 2004).
Meran and Pfaltz (1975) reported that 2 weeks
after onset of vestibular neuritis, 66% of patients
did not respond to thermal irrigation of the external
auditory canal, and 34% showed reduced responses.
Two years later, however, 72% had normal reactions,
12% showed reduced responses, and 16% did not
respond. They found complete recovery of semicir-
cular canal function in two-thirds of the patients.
Okinaka et al. (1993) found that caloric responses
normalized in only 25 (42%) of 60 patients. Horizon-
tal semicircular canal paresis was found in about
90% 1 month after onset, and in 80% after 6 months.
The different results in these two studies may be due
to different criteria for defining a pathological unilat-
eral hyporesponsiveness.
25.4.2. Vestibular-evoked myogenic potentials
and galvanic stimulation
In response to loud clicks, “vestibular-evoked myo-
genic potentials” (VEMPs) can be recorded from
the sternocleidomastoid muscles (Murofushi et al.,
1996; Colebatch et al., 2000). There is good evidence
that VEMPs originate in the medial (striola) area of
the saccular macula (Murofushi et al., 2000). VEMPs
allow examination of the function of the sacculus
and, thereby, of the inferior vestibular nerve. In
VN, VEMPs are preserved in two-thirds of the
patients (Colebatch, 2000). This is due to the sparing
of the inferior part of the vestibular nerve in most
patients (see below), which supplies the sacculus
and posterior canal, among others. In one study on
two patients the recovery of abnormal VEMPs and,
thereby, of the inferior vestibular nerve was found
321
in one of them (Ochi et al., 2003). More recently,
recovery was found in 5 of 13 patients with VN
and pathological VEMPs. It took 6 months to 2 years
to return to the normal range. On the other hand,
caloric responses returned to the normal range in
only one patient (Murofushi et al., 2006). It has also
been demonstrated that skull tap stimulation evokes
responses similar to VEMPs. In a study on 18
patients with VN, abnormal skull tap VEMPs were
found in 10 (56%), but only 4 of 18 (22%) showed
asymmetry in the click-induced VEMPs. The high
percentage of abnormal skull tap VEMP differences
in adjustment of the subjective visual horizontal
between the two groups suggests that this response
is not only dependent on the inferior but also superior
vestibular nerves, namely fibers that innervate the
utricle (Brantberg et al., 2003).
25.4.3. MR imaging
High resolution MRI has become increasingly impor-
tant for detecting labyrinthine or VIIIth nerve disor-
ders (Jäger et al., 1997; Casselman, 2002) such as
vestibular schwannoma, Cogan’s syndrome, vestibu-
lar paroxysmia, leptomeningeal carcinomatosis, or
meningitis. It has become clinically more relevant
for the exclusion of vestibular pseudoneuritis by
brainstem or cerebellar lesions (MS plaque or
stroke). Due to recent MRI advances it is now possi-
ble to demonstrate facial nerve enhancement in
Bell’s palsy and cochlear enhancement in sudden
hearing loss. However, all attempts to image lesions
of the vestibular nerve or ganglion in patients with
cryptogenic VN have so far failed (Fig. 5) (Hasuike
et al., 1995; Strupp et al., 1998c).
25.5. Pathophysiology and etiology
25.5.1. Pathophysiology
Normal vestibular end organs generate an equal
resting-firing frequency, which is the same on both
sides. This continuous excitation (resting discharge
rate in monkey 100 Hz (Goldberg and Fernandez,
1971); 1800 vestibular afferents for each labyrinth
(Bergstrom, 1973)) is transmitted to the vestibular
nuclei via vestibular nerves. Pathological processes
affecting an end organ alter its firing frequency, thereby
creating a tone imbalance. This imbalance causes most
of the manifestations of the vertigo syndrome: percep-
tual, ocular motor, postural, and vegetative (nausea).
322 M. STRUPP AND T. BRANDT

Fig. 5. MRI of a patient with a left vestibular neuritis (and persisting caloric unresponsiveness) which were performed 20 days after symp-
toms began. (A) Axial T1-weighted 2-D FLASH (fast low angle shot) sequence with gadolinium-DTPA: the facial nerve (short big arrow),
the superior part of the vestibular nerve (large big arrow), the cochlear nerve (long small arrow), and the inferior part of the vestibular nerve
(short small arrow) can be delineated. No contrast enhancement is visible (from Hasuike et al., 1995; Strupp et al., 1998c). (B) Herpes zos-
ter neuritis: coronal T1-weighted 2-D FLASH sequence reveals gadolinium enhancement of the pars superior and inferior of the right ves-
tibular nerve (white arrows) (from Arbusow et al., 1997).

As distinct from bilateral vestibulopathy, unilat-
eral semicircular canal paresis can be largely substi-
tuted for by the redundant canal input from the
unaffected labyrinth. Angular head accelerations
are detected by three pairs of semicircular canals,
and linear head accelerations by two pairs of
otoliths. These sensors induce compensatory eye
movements (slow phases) in the opposite direction
to head acceleration and transduce the sensation of
head motion. Sensorimotor transformation proceeds
via canal planes to the planes of eye movements
so that the neurons always contact their two
respective extraocular muscles. This means that a
lesion of a single semicircular canal induces a spon-
taneous nystagmus with slow phases in the “off-
direction” of that canal. If multiple canals are
lesioned, the slow phases should be in a direction
that is a weighted vector sum of the axes of the
involved canals [8]. The direction of head rotation
is sensed by corresponding on-and-off modulation
of the resting activity (on: 100!500 Hz; off:
100!0 Hz) of the right and left canals, corroborat-
ing in pairs for the particular plane of motion (yaw
¼ horizontal semicircular canals, right and left).
Loss of function in the on-direction (head rotation
to the right with right horizontal semicircular canal
paresis) is still sensed by the opposite canal, which
is stimulated (inhibited) in its off-direction. Modu-
lation of the neural activity in the off direction is
limited, and as the speed of head rotations increases,
the firing rate of the neurons will reach zero; this is
also called Ewald’s second law (Ewald, 1892) (see
also Section 25.2.2.). Vertigo, spontaneous nystag-
mus with oscillopsia, and postural imbalance in
VN are the appropriate stimuli to promote central
vestibular compensation and vestibular substitution
by visual and somatosensory input. Since vestibular
compensation is less perfect than generally
believed, especially for dynamic conditions, further
mechanisms such as sensory substitution by vision
or proprioception in part replace the missing vestib-
ular input (Brandt et al., 1997b). There is, for exam-
ple, a measurably increased influence of cervical
proprioception on spatial orientation and gaze in
space ipsilateral to a peripheral vestibular lesion
(Strupp et al., 1998a). Vestibular exercises and
pharmacological substances may facilitate these
processes (Strupp et al., 2001).
25.5.2. Vestibular neuritis – not a total but
a partial unilateral vestibular loss
Does VN produce a complete or a partial unilateral
vestibular paralysis? The coexistence of VN and
benign paroxysmal positional vertigo in the same
individual, in the same ear, at the same time seems
impossible, for this implies the simultaneous function
and loss of function of one labyrinthine structure.
The repeated clinical observation of this apparently
paradoxical coincidence led us to draw the following
conclusions (Büchele and Brandt, 1988). VN affects
only part of the vestibular nerve trunk, usually the
superior vestibular nerve (innervating the horizontal
and anterior semicircular canals, the maculae of the
DISEASES AND TREATMENTS
utricle, and the anterosuperior part of the sacculus),
which has its own path and ganglion (Lorente de
Nó, 1933; Sando et al., 1972), whereas the inferior
vestibular nerve (innervating the posterior semicircu-
lar canal and the posteroinferior part of the sacculus)
is spared (Fig. 6). This hypothesis of partial involve-
ment of the vestibular nerve is supported by findings
of temporal bone pathology (Schuknecht and Kita-
mura, 1981) and also by the histopathology of a case
of herpes zoster oticus (Proctor et al., 1979). In the
latter case the otolith apparatus and the posterior
semicircular canal remained intact. An analysis of
3-D properties of the VOR in patients with VN
clearly demonstrated that the vectors of the spontane-
ous nystagmus clustered between the expected direc-
tions for lesions either of the horizontal or of a
combined lesion of the horizontal plus the anterior
AC
AVA
HC
PC
Vasculature
AC
VN
HC
PC
Innervation
Fig. 6. Vestibular neuritis, a partial labyrinthine lesion with hori-
zontal semicircular canal paresis: vascular or viral etiology? A the-
oretical explanation is that only the anterior vestibular artery
(AVA) is affected, sparing the posterior branch which supplies
the posterior canal (top). The more likely explanation is a viral eti-
ology affecting parts of the vestibular nerve (VN), in particular the
horizontal ampullary nerve, but sparing the inferior division, the
functioning of which is necessary for the occurrence of simulta-
neous vestibular neuritis and benign paroxysmal positioning nys-
tagmus in the same ear. AC ¼ anterior canal; HC ¼ horizontal
canal; PC ¼ posterior canal (modified from Brandt, 1999).
323
semicircular canals (Fetter and Dichgans, 1996). This
is supported by head-impulse tests and the recording
of the thereby elicited eye movements in patients
with VN (Aw et al., 2000). All in all, these data
and the above-mentioned findings in VEMPs support
an isolated lesion of the superior vestibular nerve; the
inferior vestibular nerve is spared in most patients
with VN.
25.5.3. Etiology
In the past, two main causes were proposed: either
inflammation of the vestibular nerve (Ruttin, 1909;
Nylen, 1924; Dix and Hallpike, 1952) or vascular
disturbance, which could be due to labyrinthine
ischemia (Lindsay and Hemenway, 1956) or even
microvascular disturbances caused by infection
(Meran and Pfaltz, 1975). The histologic findings in
single cases (Hilding et al., 1968) suggest infectious
pathogenesis. On the basis of a few autopsy studies,
in which the pathological findings were similar
to those occurring with known viral disorders,
Schuknecht and Kitamura (1981) deduced that typi-
cal VN is in fact a viral neuritis of the superior ves-
tibular nerve (Fig. 7).
The most popular theory is that of viral etiology,
but the evidence for it remains circumstantial (Nadol,
1995; Brandt, 1999; Baloh, 2003). The following
arguments are presented to support a viral etiology:
(1) Vestibular nerve histopathology in cases of VN
(Schuknecht and Kitamura, 1981) is similar to that
seen in single cases of herpes zoster oticus, when
temporal bone histopathology was available. (2) An
animal model of VN was developed by inoculating
HSV-1 into the auricle of mice (Hirata et al., 1995);
postural deviation was observed in 5% of the mice
6–8 days after the inoculation. Degeneration of
Scarpa’s ganglion and HSV-1 antigens were found
only in symptomatic animals. Vestibular symptom-
atology can be induced by intraperitoneal, intracere-
bral, intralabyrinthine, or intracutaneous inoculation
of various viral agents (Davis, 1993). (3) HSV-1
DNA was repeatedly detected in about two-thirds of
autopsied human vestibular ganglia by polymerase
chain reaction (PCR) (Furuta et al., 1993; Arbusow
et al., 1999); further the “latency associated tran-
script” was found in about 70% of human vestibular
ganglia (Theil et al., 2002) (Fig. 8). All these find-
ings indicate that the vestibular ganglia like other
cranial nerve ganglia are latently infected by HSV-1
(Nahmias and Roizman, 1973; Theil et al., 2001,
324 M. STRUPP AND T. BRANDT

RIGHT

Total atrophy of
ampullary nerve fibers Atrophy of crista

A
LEFT

Normal crista Normal ampullary

B nerve fibers

Fig. 7. Histopathology of a patient with a right VN who died 12 years after symptom onset. (A) Right ear. This view shows degeneration of
the ampullary branch of the superior division of the vestibular nerve as well as the lateral canal crista, including the sensory epithelium
(x 39). (B) Left ear. The ampullary nerve and lateral canal crista, including the sensory epithelium, appear normal (x 39) (from Schuknecht
and Kitamura, 1981).

Geniculate
ganglion

Superior Anterior, horizontal SCC

vestibular
nerve Utricle
a
Facial r Saccule
nerve ula c
s tib is
-ve os
a cio stom Posterior SCC
Intermediate
F na
a a b
Inferior
nerve vestibular
Vestibular nerve b
ganglion
Vestibular
nerve

Fig. 8. Schematic drawing of the vestibular and facial nerves, the facio-vestibular anastomosis, the geniculate ganglion, and different
sections of the vestibular ganglion (a, stem; b, inferior portion; and c, superior portion). Right: Longitudinal cryosection of a human ves-
tibular ganglion, in which the individual portions are separated. Using PCR HSV-1 DNA was found in about 60% of the examined human
vestibular ganglia. Moreover, the double innervation of the posterior canal, which led to the preservation of its function during vestibular
neuritis, is visible (from Arbusow et al., 1999).

2003). A similar etiology is also assumed for Bell’s If HSV is the most likely candidate, it can be
palsy and strongly supported by the demonstration assumed to reside in a latent state in the vestibular
of HSV-1 DNA in the endoneurial fluid of affected ganglia, e.g., in the ganglionic nuclei as has been
subjects (Murakami et al., 1996). reported in other cranial nerves (Theil et al., 2000,
DISEASES AND TREATMENTS
2001, 2004). As a result of intercurrent factors, it
suddenly replicates and induces an inflammation
and edema and, thereby, secondary cell damage of
the vestibular ganglion cells and axons in the bony
canals.
25.6. Differential diagnosis and clinical problems
When based on careful history taking and clinical
evaluation, the differential diagnosis is determined
by two elementary questions: (1) Is the clinical syn-
drome compatible with peripheral vestibular loss
only or are there any central neurological deficits
incompatible with VN? (2) Are there any signs,
symptoms, or clinical indications for a specific etiol-
ogy of an acute unilateral, partial, or complete vestib-
ular loss? Topographically, dysfunctions or lesions
in the brainstem and/or cerebellum (so-called ves-
tibular pseudoneuritis) as well as other peripheral
vestibular disorders may mimic VN.
25.6.1. Central lesions mimicking vestibular
neuritis
There is a small area in the lateral medulla including
the root entry zone of the vestibular nerve and the
medial and superior vestibular nuclei in which a
lesion may cause confusion with peripheral vestibu-
lar nerve or labyrinthine lesions. We have seen sev-
eral patients with multiple sclerosis who have
pontomedullary plaques or small lacunar strokes
(Fig. 9) (Thomke and Hopf, 1999) at the root entry
zone of the VIIIth nerve. This leads to “fascicular
nerve lesion”, which mimics VN: so-called vesti-
bular pseudoneuritis. The differential diagnosis
A B
Fig. 9. Fascicular and nuclear lesion of the vestibular nerve due to an MS plaque (A) and vascular lesion (B), mimicking vestibular neuritis
(T2-weighted MR images).
325
between central and peripheral causes of unilateral
vestibular loss is simple, if the patient presents with
obvious additional brainstem signs. If this is not the
case differential diagnosis is indeed difficult. In a
recent study we, therefore, correlated clinical signs
to differentiate vestibular neuritis (40 patients) from
central “vestibular pseudoneuritis” (43 patients) in
the acute situation, and the final diagnosis was
assessed by neuroimaging. Skew deviation was the
only specific, but non-sensitive (40%) sign for ves-
tibular pseudoneuritis. None of the other isolated
signs (head-thrust test, saccadic pursuit, gaze- evoked
nystagmus, subjective visual vertical) were reliable;
however, multivariate logistic regression increased
their sensitivity and specificity to 92% (Cnyrim
et al., 2008).
Cerebellar infarction may also mimic VN,
namely in the territory of the posterior inferior cere-
bellar artery (PICA) (Duncan et al., 1975; Huang
and Yu, 1985; Magnusson and Norrving, 1991,
1993; Norving et al. 1995). It may also cause incom-
plete ocular tilt reaction (Mossman and Halmagyi,
2000) (see Section 25.2.1), which may make the dif-
ferential diagnosis even more difficult.
In a recent series of 240 consecutive cases of
isolated cerebellar infarction in the territories of the
cerebellar arteries diagnosed by brain MRI, 25
patients (10.4%) with isolated cerebellar infarction
were identified who had clinical features suggesting
VN (Lee et al., 2006). Twenty-four of these patients
presented with isolated spontaneous prolonged ver-
tigo, and imbalance was the initial manifestation of
cerebellar infarction; 15 of them had a spontaneous
nystagmus and 17 of them a pathological Romberg
with falls toward the affected side. MRI showed that
326
the medial branch of the PICA was affected in 24 of
the 25. The medial branch of the PICA supplies the
nodulus and uvula, both are key structures of the ves-
tibulo-cerebellum with strong connections to the ipsi-
lateral vestibular nuclei (Voogd et al., 1996; Fushiki
and Barmack, 1997; Lee et al., 2003). Further, if
the infarction is small, limb ataxia may be minimal
or even absent.
Infarction in the territory of the anterior inferior
cerebellar artery (AICA) may also mimic VN, but it
is most often associated with AICA unilateral
hearing loss (due to cochlear ischemia) and addi-
tional brainstem signs (Lee et al., 2002, 2006).
All in all, cerebellar infarction may cause isolated
vertigo and a pathological Romberg sign, but clinical
examination and testing of hearing will allow its dif-
ferentiation from VN and vestibular pseudoneuritis.
However, further studies are necessary on this
important issue in order to correlate clinical vestibu-
lar and ocular motor signs with MRI lesions in the
cerebellum.
Acute attacks of vestibular migraine may also
mimic VN, because they may be associated with a
rotatory vertigo and horizontal-torsional nystagmus.
Accompanying symptoms and the course of the dis-
ease help to differentiate between the two entities
(see also chapter on migrainous vertigo, Chapter 33).
25.6.2. Peripheral vestibular lesions
The differential diagnosis of peripheral labyrinthine
and vestibular nerve disorders mimicking VN
includes numerous rare conditions. Nevertheless,
extensive laboratory examinations, lumbar puncture,
and CT/MR imaging are not part of the routine diag-
nostics of VN for two reasons: (1) the rareness of
these disorders, and (2) typical additional signs and
symptoms indicative of other disorders. For example,
the combination of vestibular with auditory symp-
toms suggests herpes zoster oticus, if the ear is pain-
ful and blisters are observed in the external auditory
canal; or Cogan’s syndrome, if inflammatory eye
symptoms are found. Thus, any further diagnostic
procedures in patients with VN are usually prompted
and guided by the unusual presentation of the syn-
drome, an atypical course, or additional signs and
symptoms.
An initial monosymptomatic vertigo attack in
Me´nie`re’s disease or a short attack in vestibular par-
oxysmia can be confused with VN in a patient admit-
ted to the hospital in an acute stage. The shortness of
M. STRUPP AND T. BRANDT
the attack and the patient’s rapid recovery, however,
allow differentiation. During the course of the dis-
ease almost all patients with Ménière’s disease
develop hypoacusis, tinnitus, or aural fullness in the
affected ear, which also allows differentiation (see
Chapter 28). An initially burning pain and blisters
as well as hearing disorders and facial paresis are
typical for herpes zoster oticus (Ramsay–Hunt syn-
drome) (in such cases acyclovir or valacyclovir is
indicated). It has to be pointed out that there may
be a skew deviation (see Section 25.6.1.) in herpes
zoster oticus due to the complete unilateral periph-
eral vestibular deficit, i.e., of the pars superior and
inferior of the vestibular nerve (Arbusow et al.,
1997) and – contrary to VN – a contrast enhancement
of the eighth cranial nerve (Fig. 5B). Very rare
causes of acute unilateral peripheral vestibular
lesions are toxically serous types of labyrinthitis
which accompany a middle-ear inflammation and
are usually painful (here antibiotic treatment is indi-
cated); the acute suppurative form of labyrinthitis
and mastoiditis, which is also characterized by pain
and frequently by hearing disorders as well (these
require in addition operative decompression and
drainage); and tuberculous labyrinthitis, which is
more frequently a complication of tuberculous men-
ingitis than of tuberculous middle-ear inflammation.
Cogan’s syndrome (often overlooked) is a severe
autoimmune disease accompanied by interstitial ker-
atitis and audiovestibular symptoms (hearing disor-
ders are very prominent). It occurs most often in
young adults and responds, in part only temporarily,
to the very early administration of high doses of cor-
ticoids (1000 mg per day for 3–5 days, then slowly
tapered off) (Vollertsen et al., 1986; Vollertsen,
1990) or – like other autoimmune diseases of the
inner ear – to a combination of steroids and cyclo-
phosphamide (McCabe, 1991; Orsoni et al., 2002).
Vestibular schwannomas produce such a gradual
reduction in vestibular brainstem input from the end
organ on the side of the tumor that central compensa-
tion is capable of preventing vertigo. However, acute
rotational vertigo and semicircular canal paresis are
rarely the first manifestations of a rapidly growing
and larger tumor of the cerebellopontine angle. At
that time the critical site of the lesion is peripheral,
even though larger tumors compress the brainstem
and the flocculus.
Rare variants of VN have been described, e.g.,
inferior vestibular neuritis (here there is a selective
deficit of the posterior canal combined with sparing
DISEASES AND TREATMENTS
of the lateral and anterior canals) (Halmagyi et al.,
2002) and a form in which a dysfunction of the
posterior canal is combined with one of the cochlea.
The latter probably does not have a viral but rather a
vascular etiology, since both structures have a com-
mon area of vascular supply. Despite the absence of
click vestibular-evoked myogenic potentials, normal
galvanic vestibular-evoked myogenic potentials were
found in some patients with vestibular neuritis; this
finding is compatible with the presence of labyrin-
thine lesions (Murofushi et al., 2003). Acute unilat-
eral disorders of labyrinthine function can also be
caused by ischemia as a result of labyrinthine infarc-
tions with or without hearing disorders (see above)
(supply region of the A. labyrinthi or AICA).
Recently, vestibular and auditory functions were
examined in 29 patients who had suffered from uni-
lateral sensorineural hearing loss. A vestibular lesion
was found in 45% of these patients, and 53% of them
had a combined impairment of the cochlea and the
ipsilateral posterior semicircular canal. The authors
suggested that this possibly reflects a vascular dis-
ease in the common cochlear artery, leading to com-
bined vestibular and auditory dysfunction (Rambold
et al., 2005).
Vestibular schwannomas, which arise in the mye-
lin sheaths of the vestibular part of the VIIIth nerve,
only cause vertigo, a tendency to fall, and nystagmus
when the pontomedullary brainstem and the floccu-
lus are compressed and the increasing peripheral tone
difference can no longer be neutralized by central
compensation. The main symptom is a slowly pro-
gressive unilateral reduction of hearing without any
identifiable otological cause which is combined with
a caloric hypoexcitability or non-excitability. In rare
cases there is also a loss of hearing as well as acute
vertigo in cases of a purely intracanalicular dilata-
tion, which can be confirmed by MRI and treated
early by microsurgery or with the “gamma knife”.
Other tumors that can cause vertigo are meningioma
of the cerebellopontine angle, epidermoid cysts, car-
cinomas, or glomus tumors of the vagus or
glossopharyngeus.
25.7. Management
The management of VN involves: (1) symptomatic
treatment with antivertiginous drugs (e.g., dimenhy-
drinate, scopolamine) to attenuate vertigo, dizziness
and nausea/vomiting; (2) “causal” treatment with
corticosteroids to improve recovery of peripheral
327
vestibular function; and (3) physical therapy (vestib-
ular exercises and balance training) to improve
central vestibular compensation. Treatment with
measures to improve circulation (vasodilators, low-
molecular weight dextrans, hydroxyethyl starches,
local anesthetics, or inhibitors of the ganglion stella-
tum) is ineffective.
25.7.1. Symptomatic treatment
During the first 1–3 days, when nausea is pro-
nounced, vestibular sedatives such as antihistamine
dimenhydrinate (Dramamine) 50–100 mg every 6 h
or the “anticholinergic” scopolamine (Transderm
scop) 0.6 mg can be administered parenterally for
symptomatic relief. Their major side effect is general
sedation. Transdermal application of scopolamine
hydrobromide avoids some of the side effects of the
conventional means of administration. The most
probable sites of primary action are the synapses of
the vestibular nuclei, which exhibit a reduced dis-
charge and diminished neural reaction to body rota-
tion. These drugs should not be given for longer
than 3 days, because they prolong the time required
to achieve central compensation (Zee, 1985;
Curthoys and Halmagyi, 2000).
25.7.2. Causal treatment
Based on the assumption that VN is caused by the reac-
tivation of a latent HSV-1 infection, a prospective, ran-
domized, double-blind trial was conducted to
determine whether steroids, antiviral agents, or a com-
bination of the two might improve outcome in VN
(Strupp et al., 2004). Patients with acute VN were ran-
domized to a placebo, methylprednisolone, valacyclo-
vir, and methylprednisolone plus valacyclovir groups.
Vestibular function was determined by caloric irriga-
tion, using Jongkees’s vestibular paresis formula (see
Section 25.4.1) within 3 days after symptom onset
and 12 months later. A total of 141 patients were
included in the study and randomized: 38 to the pla-
cebo; 35 to methylprednisolone (initially 100 mg/day
and then tapered off by 20 mg every 3 days), 33 to vala-
cyclovir (1 g tid for 7 days), and 35 to methylpredniso-
lone plus valacyclovir groups. Follow-up examination
showed that vestibular function improved: in the pla-
cebo group from 78.9  24.0 percentage points (mean
 SD) to 39.0  19.9, in the methylprednisolone group
from 78.7  15.8 to 15.4  16.2, in the valacyclovir
group from 78.4  20.0 to 42.7  32.3, and in the
328 M. STRUPP AND T. BRANDT

Control Methylprednisolone
100 100 100 100
90 90 90 90
Vestibular Paresis (%)

Vestibular Paresis (%)

80 80 80 80
70 70 70 70
60 60 60 60
50 50 50 50
40 40 40 40
30 30 30 30
20 20 20 20
10 10 10 10
0 0 0 0
Onset Follow-up Onset Follow-up Onset Follow-up Onset Follow-up

Valacyclovir Methylprednisolone plus Valacyclovir

100 100 100 100
90 90 90 90
Vestibular Paresis (%)

80 80 80 80

Vestibular Paresis (%)

70 70 70 70
60 60 60 60
50 50 50 50
40 40 40 40
30 30 30 30
20 20 20 20
10 10 10 10
0 0 0 0
Onset Follow-up Onset Follow-up Onset Follow-up Onset Follow-up

Fig. 10. Unilateral vestibular failure within 3 days after symptom onset and after 12 months. Vestibular function was determined by caloric
irrigation, using the “vestibular paresis formula” (which allows a direct comparison of the function of both labyrinths) for each patient in the
placebo (upper left), methylprednisolone (upper right), valacyclovir (lower right), and methylprednisolone plus valacyclovir (lower left)
group. Also shown are box plot charts for each group with the mean (▪)  SD, and 25% and 75% percentile (box plot) as well as the
1% and 99% range (x). A clinically relevant vestibular paresis was defined as > 25% asymmetry between the right-sided and the left-sided
responses (Honrubia, 1994). Follow-up examination showed that vestibular function improved in all four groups: in the placebo group from
78.9  24.0 (meanSD) to 39.0  19.9, in the methylprednisolone group from 78.7  15.8 to 15.4  16.2, in the valacyclovir group from
78.4  20.0 to 42.7  32.3, and in the methylprednisolone plus valacyclovir group from 78.6  21.1 to 20.4  28.4. Analysis of variance
revealed that methylprednisolone and methylprednisolone plus valacyclovir caused significantly more improvement than placebo or vala-
cyclovir alone. The combination of both was not superior to steroid monotherapy (from Strupp et al., 2004).

methylprednisolone plus valacyclovir group from
78.6  21.1 to 20.4  28.4 (Fig. 10). Analysis of
variance revealed that methylprednisolone and meth-
ylprednisolone plus valacyclovir caused significantly
more improvement than placebo or valacyclovir
alone.
This study shows that monotherapy with steroids
suffices to significantly improve the peripheral vestib-
ular function in patients with vestibular neuritis; there
was no evidence for synergy between methylpredniso-
lone and valacyclovir. Glucocorticoids (methylpred-
nisolone) should be given within 3 days after
symptom onset and for 3 weeks (initially 100 mg/day
and then tapered off by 20 mg every 3 days). As in
Bell’s palsy, the benefit of steroids might be explained
by their anti-inflammatory effects, which reduce
the swelling and cause a mechanical compression of
the vestibular nerve within the temporal bone. Thus,
steroids but not antiviral agents can be recommended
as a treatment for acute vestibular neuritis, for they
cause a significant functional improvement. These
findings are also supported by an ongoing trial in
Sweden (Michael Karlberg, personal communication).
25.7.3. Physical therapy
So far the most important principle of therapy is to
promote central compensation by means of physical
therapy. This so-called central compensation is not
a uniform process. It involves various neural and
structural mechanisms that operate in different loca-
tions (vestibulospinal or vestibulo-ocular areas)
DISEASES AND TREATMENTS
45
40
35
n = 5⬘
Sway path (m/min)
30 n = 4⬘
25
20
15
10
5
0
0 5
Fig. 11. Time course of the changes in SP values for the control and physiotherapy groups: vestibular exercises improved central vestibulo-
spinal compensation. For postural control we measured the SP values (m/min, mean  SD) of patients with eyes closed and standing on a
compliant foam-padded posturography platform. There was a significant difference (ANOVA, P < 0.001) between the two groups at the
statistical endpoint (day 30 after symptom onset). The dotted line indicates the normal range. During the first days after symptom onset
not all patients could stand long enough (> 20 sec) on the platform to permit accurate quantitative measurement of the SP values (from
(Strupp et al., 1998b).
within different time courses, have various limited
possibilities, and cause incomplete results, especially
as regards head oscillations at high frequencies
(Halmagyi et al., 1992; Brandt et al., 1997a). Central
compensation of a unilateral labyrinthine lesion is
enhanced and increased if movement stimuli trigger
inadequate and incongruent afferent signals that
provoke a sensory mismatch. Animal experiments
have proven the efficacy of exercises to promote
central compensation of spontaneous nystagmus and
to counteract the tendency to fall after unilateral
labyrinthine lesions (Igarashi et al., 1981). A pro-
spective, randomized, controlled study (Strupp
et al., 1998b) has demonstrated that intensive physio-
therapy significantly improved vestibulo-spinal
compensation in patients with vestibular neuritis
(Fig. 11).
Vestibular exercises, first recommended by
Cawthorne (1944) and later modified according to
new knowledge of the vestibular function (Hamann,
1988; Herdman, 1997; Strupp et al., 1998b), include
the following: (1) voluntary eye movements and
fixation to improve impaired gaze stabilization; (2)
active head movements to recalibrate the vestibulo-
ocular reflex; and (3) balance training, goal-directed
329
Postural control
control
vestibular exercises
p < 0.001
n = 20
n = 19
10 15 20 25 30
Time (day after symptom onset)
movements, and gait exercises to improve the vestibu-
lospinal regulation of posture and goal-directed motion.
Pharmacological and metabolic studies in animals
suggest that alcohol, phenobarbital, chlorpromazine,
diazepam, and ACTH antagonists retard central com-
pensation, whereas caffeine, amphetamines, and
ACTH accelerate it (survey in Zee, 1985; Curthoys
and Halmagyi, 2000). So far no clinical studies have
been performed on this subject (overview in Strupp
et al., 2001).
References
Arbusow, V, Dieterich, M, Strupp, M, Jäger, L and Brandt,
T (1997) Herpes zoster neuritis involving superior and
inferior parts of the vestibular nerve causes ocular tilt
reaction. Akta. Neurol., 24: S94.
Arbusow, V, Schulz, P, Strupp, M, Dieterich, M, Von
Reinhardstoettner, A, Rauch, E and Brandt, T (1999)
Distribution of herpes simplex virus type 1 in human
geniculate and vestibular ganglia: implications for ves-
tibular neuritis. Ann. Neurol., 46: 416–419.
Arbusow, V, Theil, D, Strupp, M, Mascolo, A and Brandt,
T (2000) HSV-1 not only in human vestibular ganglia
but also in the vestibular labyrinth. Audiol. Neurootol.,
6: 259–262.
330
Aw, ST, Fetter, M, Cremer, PD, Karlberg, M and Halma-
gyi, GM (2000) Individual semicircular canal function
in superior and inferior vestibular neuritis. Neurology,
57: 768–774.
Baloh, RW (2003) Clinical practice. Vestibular neuritis.
New Engl. J. Med., 348: 1027–1032.
Bergstrom, B (1973) Morphology of the vestibular nerve. I.
Anatomical studies of the vestibular nerve in man. Acta
Otolaryngol. (Stockh.), 76: 162–172.
Böhmer, A and Rickenmann, J (1995) The subjective visual
vertical as a clinical parameter of vestibular function in
peripheral vestibular diseases. J. Vestib. Res., 5: 35–45.
Brandt, T (1996) Phobic postural vertigo. Neurology, 46:
1515–1519.
Brandt, T (1999) Vertigo; Its Multisensory Syndromes, 2nd
ed. Springer, London.
Brandt, T and Daroff, RB (1980) The multisensory physio-
logical and pathological vertigo syndromes. Ann.
Neurol., 7: 195–203.
Brandt, T and Dieterich, M (1986) Phobischer Attacken-
Schwankschwindel, ein neues Syndrom. Münch. Med.
Wschr., 128: 247–250.
Brandt, T, Allum, JH and Dichgans, J (1978) Computer
analysis of optokinetic nystagmus in patients with spon-
taneous nystagmus of peripheral vestibular origin. Acta
Otolaryngol. (Stockh.), 86: 115–122.
Brandt, T, Strupp, M, Arbusow, V and Dieringer, N
(1997a) Plasticity of the vestibular system: central com-
pensation and sensory substitution for vestibular defi-
cits. Adv. Neurol., 73: 297–309.
Brandt, T, Strupp, M, Arbusow, V and Dieringer, N
(1997b) Plasticity of the vestibular system: central
compensation and sensory substitution for vestibular
deficits. Adv. Neurol., 73: 297–309.
Brandt, T, Dieterich, M and Strupp, M (2005) Vertigo and
Dizziness – Common Complaints. Springer, London.
Brantberg, K, Tribukait, A and Fransson, PA (2003) Ves-
tibular evoked myogenic potentials in response to skull
taps for patients with vestibular neuritis. J. Vestib. Res.,
13: 121–130.
Büchele, W and Brandt, T (1988) Vestibular neuritis, a hor-
izontal semicircular canal paresis? Adv. Otorhinolaryn-
gol., 42: 157–161.
Casselman, JW (2002) Diagnostic imaging in clinical
neuro-otology. Curr. Opin. Neurol., 15: 23–30.
Cawthorne, T (1944) The physiological basis for head
exercises. J. Chart. Soc. Physiother., 106–107.
Cnyrim, CD, Newman-Toker, D, Karch, C, Brandt, T and
Strupp, M (2008) Bedside differentiation of vestibular
neuritis from central “vestibular pseudoneuritis”. J.
Neurol. Neurosurg. Psychiatry, 79: 458–460.
Colebatch, JG (2000) Vestibular evoked potentials. Curr.
Opin. Neurol., 14: 21–26.
Colebatch, JG, Halmagyi, GM and Skuse, NF (2000)
Myogenic potentials generated by a click-evoked
M. STRUPP AND T. BRANDT
vestibulocollic reflex. J. Neurol. Neurosurg. Psychiatry,
57: 190–197.
Curthoys, IS and Halmagyi, GM (2000) Vestibular com-
pensation: a review of the oculomotor, neural, and clin-
ical consequences of unilateral vestibular loss. J. Vestib.
Res., 5: 67–107.
Davis, LE (1993) Viruses and vestibular neuritis: review of
human and animal studies. Acta Otolaryngol. Suppl.
(Stockh.), 503: 70–73.
Depondt, M (1973) Vestibular neuronitis. Vestibular paral-
ysis with special characteristics. Acta Otorhinolaryngol.
Belg., 27: 323–359.
Dix, MR and Hallpike, CS (1952) The pathology, symp-
tomatology, and diagnosis of certain common disorders
of the vestibular system. Ann. Otol., 61: 987–991.
Duncan, GW, Parker, SW and Fisher, CM (1975) Acute
cerebellar infarction in the PICA territory. Arch.
Neurol., 32: 364–368.
Ewald, R (1892) Physiologische Untersuchungen über das
Endorgan des Nervus octavus. Bergmann, Wiesbaden.
Fetter, M and Dichgans, J (1996) Vestibular neuritis spares
the inferior division of the vestibular nerve. Brain, 119:
755–763.
Furuta, Y, Takasu, T, Fukuda, S, Inuyama, Y, Sato, KC
and Nagashima, K (1993) Latent herpes simplex virus
type 1 in human vestibular ganglia. Acta Otolaryngol.
Suppl. (Stockh.), 503: 85–89.
Fushiki, H and Barmack, NH (1997) Topography and
reciprocal activity of cerebellar Purkinje cells in the
uvula-nodulus modulated by vestibular stimulation. J.
Neurophysiol., 78: 3083–3094.
Gianoli, G, Goebel, J, Mowry, S and Poomipannit, P
(2005) Anatomic differences in the lateral vestibular
nerve channels and their implications in vestibular neu-
ritis. Otol. Neurotol., 26: 489–494.
Goldberg, JM and Fernandez, C (1971) Physiology of
peripheral neurons innervating semicircular canals of
the squirrel monkey. I. Resting discharge and response
to constant angular accelerations. J. Neurophysiol., 34:
635–660.
Goldberg, JM and Fernandez, C (1982) Eye movements
and vestibular-nerve responses produced in the squirrel
monkey by rotations about an earth-horizontal axis.
Exp. Brain Res., 46: 393–402.
Hain, TC, Fetter, M and Zee, DS (1987) Head-shaking nys-
tagmus in patients with unilateral peripheral vestibular
lesions. Am. J. Otolaryngol., 8: 36–47.
Halmagyi, GM and Curthoys, IS (1988) A clinical sign of
canal paresis. Arch. Neurol., 45: 737–739.
Halmagyi, GM, Curthoys, IS, Cremer, PD, Henderson, CJ,
Todd, MJ, Staples, MJ and D’Cruz, DM (1990) The
human horizontal vestibulo-ocular reflex in response
to high-acceleration stimulation before and after unilat-
eral vestibular neurectomy. Exp. Brain Res., 81:
479–490.
DISEASES AND TREATMENTS
Halmagyi, GM, Aw, ST, Cremer, PD, Todd, MJ and
Curthoys, IS (1992) The human vertical vestibulo-
ocular reflex in response to high acceleration stimulation
after unilateral vestibular neurectomy. Ann. N.Y. Acad.
Sci., 656: 732–738.
Halmagyi, GM, Aw, ST, Karlberg, M, Curthoys, IS and
Todd, MJ (2002) Inferior vestibular neuritis. Ann. N.Y.
Acad. Sci., 956: 306–313.
Hamann, KF (1988) [Rehabilitation of patients with vestib-
ular disorders]. HNO, 36: 305–307.
Hasuike, K, Sekitani, T and Imate, Y (1995) Enhanced
MRI in patients with vestibular neuronitis. Acta Otolar-
yngol. Suppl. (Stockh.), 519: 272–274.
Herdman, SJ (1997) Advances in the treatment of vestibu-
lar disorders. Phys. Ther., 77: 602–618.
Hilding, DA, Kanda, T and House, WF (1968) Vestibular
neuronitis and small acoustic neuroma: electron micro-
scopic observations. Otol. Clin. N. Am., 112: 305–318.
Hirata, Y, Gyo, K and Yanagihara, N (1995) Herpetic ves-
tibular neuritis: an experimental study. Acta Otolaryn-
gol. Suppl. (Stockh.), 519: 93–96.
Honrubia, V (1994) Quantitative vestibular function tests
and the clinical examination. In: SJ Herdman (Ed.),
Vestibular Rehabilitation. F.A. Davis, Philadelphia,
pp. 113–164.
Huang, CY and Yu, YL (1985) Small cerebellar strokes
may mimic labyrinthine lesions. J. Neurol. Neurosurg.
Psychiatry, 48: 263–265.
Huppert, D, Strupp, M, Theil, D, Glaser, M and Brandt, T
(2006) Low recurrence rate of vestibular neuritis: a
long-term follow-up. Neurology, 67: 1870–1871.
Igarashi, M, Levy, JK, O-Uchi, T and Reschke, MF (1981)
Further study of physical exercise and locomotor bal-
ance compensation after unilateral labyrinthectomy in
squirrel monkeys. Acta Otolaryngol. (Stock.), 92:
101–105.
Jäger, L, Strupp, M, Brandt, T and Reiser, M (1997) Imag-
ing of the labyrinth and vestibular nerve: clinical rele-
vance for differential diagnosis of vestibular disorders.
Nervenarzt, 68: 443–458.
Jongkees, LB, Maas, J and Philipszoon, A (1962) Clinical
electronystagmography: a detailed study of electronys-
tagmography in 341 patients with vertigo. Pract. Oto-
rhinolaryngol. (Basel), 24: 65–93.
Kamei, T (1975) The two-phase occurrence of head-shak-
ing nystagmus. Arch. Otorhinolaryngol., 209: 59–67.
Katsarkas, A and Galiana, HL (1984) Bechterew’s phe-
nomenon in humans. A new explanation. Acta Otolar-
yngol. Suppl. (Stockh.), 406: 95–100.
Lee, H, Sohn, SI, Jung, DK, Cho, YW, Lim, JG, Yi, SD,
Lee, SR, Sohn, CH and Baloh, RW (2002) Sudden
deafness and anterior inferior cerebellar artery infarc-
tion. Stroke, 33: 2807–2812.
Lee, H, Yi, HA, Cho, YW, Sohn, CH, Whitman, GT,
Ying, S and Baloh, RW (2003) Nodulus infarction
331
mimicking acute peripheral vestibulopathy. Neurology,
60: 1700–1702.
Lee, H, Sohn, SI, Cho, YW, Lee, SR, Ahn, BH, Park, BR
and Baloh, RW (2006) Cerebellar infarction presenting
isolated vertigo: frequency and vascular topographical
patterns. Neurology, 67: 1178–1183.
Lindsay, JR and Hemenway, WG (1956) Postural vertigo
due to unilateral sudden partial loss of vestibular func-
tion. Arch. Otolaryngol., 65: 692–706.
Lorente de Nó, R (1933) Vestibulo-ocular reflex arc. Arch.
Neurol. Psychiat., 30: 245–291.
Magnusson, M and Norrving, B (1991) Cerebellar infarc-
tions as the cause of ‘vestibular neuritis’. Acta Otolar-
yngol. Suppl. (Stockh), 481: 258–259.
Magnusson, M and Norrving, B (1993) Cerebellar infarc-
tions and “vestibular neuritis”. Acta Otolaryngol. Suppl.
(Stockh.), 503: 64–66.
McCabe, BF (1991) Autoimmune inner ear disease: results
of therapy. Adv. Otorhinolaryngol., 46: 78–81.
Meran, A and Pfaltz, CR (1975) The acute vestibular paral-
ysis. Arch. Otorhinolaryngol., 209: 229–244.
Mossman, S and Halmagyi, GM (2000) Partial ocular tilt
reaction due to unilateral cerebellar lesion. Neurology,
49: 491–493.
Murakami, S, Mizobuchi, M, Nakashiro, Y, Doi, T, Hato,
N and Yanagihara, N (1996) Bell palsy and herpes sim-
plex virus: identification of viral DNA in endoneurial
fluid and muscle. Ann. Intern. Med., 124: 27–30.
Murofushi, T, Halmagyi, GM, Yavor, RA and Colebatch,
JG (1996) Absent vestibular evoked myogenic poten-
tials in vestibular neurolabyrinthitis. An indicator of
inferior vestibular nerve involvement. Arch. Otolaryn-
gol. Head Neck Surg., 122: 845–848.
Murofushi, T, Curthoys, IS, Topple, AN, Colebatch, JG
and Halmagyi, GM (2000) Responses of guinea pig pri-
mary vestibular neurons to clicks. Exp. Brain Res., 103:
174–178.
Murofushi, T, Monobe, H, Ochiai, A and Ozeki, H (2003)
The site of lesion in “vestibular neuritis": study by gal-
vanic VEMP. Neurology, 61: 417–418.
Murofushi, T, Iwasaki, S and Ushio, M (2006) Recovery of
vestibular evoked myogenic potentials after a vertigo
attack due to vestibular neuritis. Acta Otolaryngol.,
126: 364–367.
Nadol, JB, Jr (1995) Vestibular neuritis. Otolaryngol. Head
Neck Surg., 112: 162–172.
Nahmias, AJ and Roizman, B (1973) Infection with herpes-
simplex viruses 1 and 2. II. New Engl. J. Med., 289:
719–725.
Norrving, B, Magnusson, M and Holtas, S (1995) Isolated
acute vertigo in the elderly: vestibular or vascular dis-
ease? Acta Neurol Scand, 91(1): 43–48.
Nylen, CO (1924) Some cases of ocular nystagmus due to
certain positions of the head. Acta Otolaryngol.
(Stockh.), 6: 106–137.
332
Ochi, K, Ohashi, T and Watanabe, S (2003) Vestibular-
evoked myogenic potential in patients with unilateral
vestibular neuritis: abnormal VEMP and its recovery.
J. Laryngol. Otol., 117: 104–108.
Ohbayashi, S, Oda, M, Yamamoto, M, Urano, M, Harada,
K, Horikoshi, H, Orihara, H and Kitsuda, C (1993)
Recovery of the vestibular function after vestibular neu-
ronitis. Acta Otolaryngol. Suppl. (Stockh.), 503: 31–34.
Okinaka, Y, Sekitani, T, Okazaki, H, Miura, M and Tahara,
T (1993) Progress of caloric response of vestibular neu-
ronitis. Acta Otolaryngol. Suppl. (Stockh.), 503: 18–22.
Orsoni, JG, Zavota, L, Pellistri, I, Piazza, F and Cimino, L
(2002) Cogan syndrome. Cornea, 21: 356–359.
Proctor, L, Perlman, H, Lindsay, J and Matz, G (1979)
Acute vestibular paralysis in herpes zoster oticus. Ann.
Otol. Rhinol. Laryngol., 88: 303–310.
Rambold, H, Boenki, J, Stritzke, G, Wisst, F, Neppert, B and
Helmchen, C (2005) Differential vestibular dysfunction in
sudden unilateral hearing loss. Neurology, 64: 148–151.
Ruttin, B (1909) Zur Differentialdiagnose der Labyrinth- und
Hörnerverkrankungen. Z. Ohrenheilkunde., 57: 327–333.
Safran, AB, Vibert, D, Issoua, D and Hausler, R (1994)
Skew deviation after vestibular neuritis. Am. J. Ophthal-
mol., 118: 238–245.
Sando, I, Black, FO and Hemenway, WG (1972) Spatial
distribution of vestibular nerve in internal auditory
canal. Ann. Otol., 81: 305–319.
Schuknecht, HF and Kitamura, K (1981) Vestibular neuri-
tis. Ann. Otol., 90(Suppl. 78): 1–19.
Sekitani, T, Imate, Y, Noguchi, T and Inokuma, T (1993) Ves-
tibular neuronitis: epidemiological survey by questionnaire
in Japan. Acta Otolaryngol. Suppl. (Stockh.), 503: 9–12.
Shirabe, S (1988) Vestibular neuronitis in childhood. Acta
Otolaryngol. Suppl. (Stockh.), 458: 120–122.
Strupp, M, Arbusow, V, Dieterich, M, Sautier, W and
Brandt, T (1998a) Perceptual and oculomotor effects
of neck muscle vibration in vestibular neuritis. Ipsilat-
eral somatosensory substitution of vestibular function.
Brain, 121: 677–685.
Strupp, M, Arbusow, V, Maag, KP, Gall, C and Brandt, T
(1998b) Vestibular exercises improve central vestibu-
lospinal compensation after vestibular neuritis. Neurol-
ogy, 51: 838–844.
Strupp, M, Jager, L, Muller, LU, Arbusow, V, Reiser, M and
Brandt, T (1998c) High resolution Gd-DTPA MR imaging
of the inner ear in 60 patients with idiopathic vestibular neu-
ritis: no evidence for contrast enhancement of the labyrinth
or vestibular nerve. J. Vestib. Res., 8: 427–433.
Strupp, M, Arbusow, V and Brandt, T (2001) Exercise and
drug therapy alter recovery from labyrinth lesion in
humans. Ann. N.Y. Acad. Sci., 942: 79–94.
M. STRUPP AND T. BRANDT
Strupp, M, Zingler, VC, Arbusow, V, Niklas, D, Maag, KP,
Dieterich, M, Bense, S, Theil, D, Jahn, K and Brandt, T
(2004) Methylprednisolone, valacyclovir, or the combi-
nation for vestibular neuritis. N. Engl. J. Med., 351:
354–361.
Tahara, T, Sekitani, T, Imate, Y, Kanesada, K and Okami,
M (1993) Vestibular neuronitis in children. Acta Otolar-
yngol. Suppl. (Stockh.), 503: 49–52.
Theil, D, Derfuss, T, Paripovic, I, Herberger, S, Meinl, E,
Schueler, O, Strupp, M, Arbusow, V and Brandt, T
(2000) Latent herpesvirus infection in human trigeminal
ganglia causes chronic immune response. Am. J.
Pathol., 163: 2179–2184.
Theil, D, Arbusow, V, Derfuss, T, Strupp, M, Pfeiffer, M,
Mascolo, A and Brandt, T (2001) Prevalence of HSV-
1 LAT in human trigeminal, geniculate, and vestibular
ganglia and its implication for cranial nerve syndromes.
Brain Pathol., 11: 408–413.
Theil, D, Derfuss, T, Strupp, M, Gilden, DH, Arbusow, V
and Brandt, T (2002) Cranial nerve palsies: herpes sim-
plex virus type 1 and varizella-zoster virus latency. Ann.
Neurol., 51: 273–274.
Theil, D, Derfuss, T, Paripovic, I, Herberger, S, Meinl, E,
Schueler, O, Strupp, M, Arbusow, V and Brandt, T
(2003) Latent herpesvirus infection in human trigeminal
ganglia causes chronic immune response. Am. J.
Pathol., 163: 2179–2184.
Theil, D, Horn, AK, Derfuss, T, Strupp, M, Arbusow, V and
Brandt, T (2004) Prevalence and distribution of HSV-1,
VZV, and HHV-6 in human cranial nerve nuclei III, IV,
VI, VII, and XII. J. Med. Virol., 74: 102–106.
Thomke, F and Hopf, HC (1999) Pontine lesions mimick-
ing acute peripheral vestibulopathy. J. Neurol. Neuro-
surg. Psychiatry, 66: 340–349.
Vibert, D, Hausler, R, Safran, AB and Koerner, F (1996) Dip-
lopia from skew deviation in unilateral peripheral vestibu-
lar lesions. Acta. Otolaryngol. (Stockh.), 116: 170–176.
Vollertsen, RS (1990) Vasculitis and Cogan’s syndrome.
Rheum. Dis. Clin. N. Am., 16: 433–439.
Vollertsen, RS, McDonald, TJ, Younge, BR, Banks, PM,
Stanson, AW and Ilstrup, DM (1986) Cogan’s syn-
drome: 18 cases and a review of the literature. Mayo
Clin. Proc., 61: 344–361.
Voogd, J, Gerrits, NM and Ruigrok, TJ (1996) Organiza-
tion of the vestibulocerebellum. Ann. N.Y. Acad. Sci.,
781: 553–579.
Zee, DS (1985) Perspectives on the pharmacotherapy of
vertigo. Arch. Otolaryngol., 111: 609–612.
Zee, DS, Preziosi, TJ and Proctor, LR (1982) Bechterew’s
phenomenon in a human patient [letter]. Ann. Neurol.,
12: 495–496.