Autorotation test abnormalities of the horizontal

and vertical vestibulo-ocular reflexes in

panic disorder
DONNA 1. HOFFMAN, M4,DENNIS P. OLEARY, PhD, and DENNIS J. MUNJACK, MD. Los Angeles, California

Pattents wlth panic disorder often describe dizziness as a dlsturblng symptom, with more
severe episodes reported than in other psychiatric populations. Nineteen patients
diagnosed as having a panic disorder were tested for vestlbulo-ocular (VOR) abnormalities
with the Vestibular Autorotatlon Test (VAT), a computerlzed test of the high-frequency (2 to
6 Hz) VOR. The patients were unselected for the presence or absence of balance disorders.
Results showed VOR abnormalltles, relathre to a normal population, in the horizontal and/or
vertical VORs of all 19 patients. Vestibuleocular reflex asymmetries were commonly
present. Because the VAT tested the VOR over a frequency range encountered during
common dally activities, the observed abnormalltles could result In a perceptually moving
visual field (oscillopsla). We hypothesize that the resulting experience of a visual-vestibular
disturbance - perhaps In a biologically or psychologlcally predisposed indMdua1- Is
catastrophically misinterpreted, leading to more bodily symptoms and anxiety. These could
then contribute to more misinterpretation In a positive feedback sense, ultimately leading
to a panic attack. [OTOLARYNGOL HEAD NECK SURG 1994;110:259-69.)

An anecdotal association between panic attacks (in
earlier literature referred to as anxiety attacks) and
vestibular symptoms has long been reported,'z2 but
only recently have objective studies been conducted
and criteria for panic disorder been established.
Table 1,section C shows symptoms that are reported
commonly during a panic attack. To meet the cri-
teria for panic disorder, the attacks should be un-
expected, not immediately related to an anxiety-
producing phobia, and not related to another's per-
sonal attention (social phobia). Frequency, duration,
and intensity of symptoms are important. Four at-
From the Departments of Otolaryngology-Head and Neck Sur-
gery (Drs. Hoffman and O'Leary) and Psychiatry (Dr. Mun-
jack), University of Southern California School of Medicine.
Presented at the Midwinter Research Meeting of the Association
for Research in Otolaryngology, St. Petersburg, Fla., Feb. 3-7,
1991.
Received for publication July 23, 1993; revision received Aug. 4,
1993; accepted Aug. 9, 1993.
Reprint requests: Donna L. Hoffman, MA, Department of Oto-
laryngology-Head and Neck Surgery, University of Southern
California, 1200 North State St., Box 795, Los Angeles, CA
90033.
Copyright 0 1994 by the American Academy of Otolaryngology-
Head and Neck Surgery Foundation, Inc.
0194-5998/94/$3.00 + 0 23/1/50518
tacks must occur within a 4-week period or one
attack must be followed by a persistent fear of
another attack over a similar period. At least four of
the symptoms in Table 1 should develop suddenly
and increase in intensity within 10 minutes of the
onset of the first symptom. Finally, an organic factor
should be ruled out as a causal or maintaining agent
(e.g., hyperthyroidism).' Agoraphobia is a common
secondary development, leading to avoidance of
situations in which escape is difficult if panic symp-
toms occur. Commonly avoided situations include
restaurants, shopping centers, public transportation,
automobiles, and crowds.
Barlow4 studied 55 patients with panic disorder
with agoraphobia and reported that 87% experi-
enced dizziness as part of the panic syndrome. Bar-
low' earlier had reported that patients with panic
disorder with or without agoraphobia described
more severe episodes of dizziness than those in
other psychiatric classifications.
Recent studies of vestibular association with
panic disorder have focused on objective evaluations
of the vestibulo-ocular reflex (VOR) and postural
control. Jacob6 tested eight patients with panic dis-
order and 13 patients with panic disorder with ago-
raphobia who were selected for having dizziness or
259

Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016


260 HOFFMAN et 01.
Table 1. Panic disorder
Dlagnostlc crlterla
A. At some time during the disturbance, one or more
panic attacks (discrete periods of intense fear or
discomfort) have occurred that were (1) unexpected;
i.e.,did not occur immediately before or on expo-
sure to a situation that almost always caused anxi-
ety, and (2) not triggered by situations in which the
person was the focus of others' attention.
6. Either four attacks, as defined in criterion A, have
occurred within a 4-week period, or one or more at-
tacks have been followed by a period of at least a
month of persistent fear of having another attack.
C. At least four of the following symptoms developed
during at least one of the attacks:
- 1. shortness of breath (dyspnea) or smothering
sensations
- 2. dizziness, unsteady feelings, or faintness
- 3. palpitations or accelerated heart rate
(tachycardia)
- 4. trembling or shaking
- 5. sweating
- 6. choking
- 7. nausea or abdominal distress
- 8 . depersonalization or derealization
- 9. numbness or tingling sensation (paresthesias)
- 10. flushes (hot flashes) or chills
- 11. chest pain or discomfort
- 12. fear of dying
- 13. fear of going crazy or of doing something
uncontrolled
NOTE Attacks involving four or more symptoms are
panic attacks; attacks involving fewer than four
symptoms are limited symptom attacks (see agora-
phobia without history of panic disorder).
D. During at least some of the attacks, at least four of
the C symptoms developed suddenly and increased
in intensity within 10 minutes of the beginning of the
first C symptom noticed in the attack.
E. It cannot be established that an organic factor initi-
ated and maintained the disturbance (e.g., amphet-
amine or caffeine intoxication, hyper-thyroidism).
NOTE Mitral valve prolapse may be an associated
condition, but does not preclude a diagnosis of
panic disorder.
Printed with permission from the American Psychiatric Association
symptoms of imbalance between episodes of panic.
Four vestibular tests were given: caloric testing,
rotational chair testing, posturography, and posi-
tional testing. Abnormal test results with calorics
were reported in 71% of the patients with panic
disorder and 45% of the patients with panic disorder
and agoraphobia. Positional testing resulted in a
similar finding, with 75% of the patients abnormal in
the uncomplicated panic disorder group and 62% of
the patients abnormal in the panic disorder with
agoraphobic cohort. Posturography and rotational
testing at low frequency (0.2 to 0.8 Hz) showed
Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016
Otolaryngology-
Head and Neck Surgery
March 1994
abnormalities as well. Their criteria for vestibular
abnormality was a test profile of at least two of the
vestibular tests in the abnormal range. Surty-seven
percent of the total group of patients, by this criteria,
were abnormal. Sklare et aL7 used electronystag-
mography (ENG) to evaluate the vestibulo-ocular
responses of 17 patients with panic disorder who
were unselected for the presence or absence of
vestibular complaints. Their results showed that
71% had abnormalities on the basis of established
ENG criteria.
Swinson et al.' compared VOR responses of 15
patients with panic disorder with symptoms of
prominent dizziness with those of a normal control
group, using a computer-controlled rotational chair
driven by pseudorandom stimulation spanning a
frequency range from 0.32 to 3.25 Hz. Panic patients
showed greater differences (higher gains) between
eye and head movements than the normal control
group, which these authors concluded may have
resulted from a hyperactive sense of alertness,
rather than neurologic dysfunction, affecting the
VOR. In addition to pseudorandom VOR testing,
Swinson et al.' tested the same patient and control
groups with routine audiologic, caloric, and vestib-
ular function testing with a standard electronystag-
mography test battery. In contrast to the studies of
Jacob6and Sklare et aL7cited earlier, Swinson et a].'
found no abnormalities in patients with panic dis-
order from audiologic, caloric, and elelctronystag-
mography tests.
The purpose of this study is to test the hypothesis
that panic disorder is associated wtih vestibulo-
ocular abnormalities at higher movement frequen-
cies. The major function of the vestibulo-ocular
reflex is to stabilize the eyes for clear vision during
faster head movements and at higher frequencies
(extending well above 1 Hz), such as occur during
walking.' At these higher frequencies, the VOR is
the primary system for visual stabilization, because
the other ocular movement systems (e.g., smooth
pursuit) are minimally effective.'O Abnormalities of
VOR responses in this higher frequency range could
be particularly relevant to vestibular symptoms re-
ported in patients with panic disorder.
MLTHODS
Patient Characteristics
All patients were referred from the Anxiety Dis-
orders Clinic at the University of Southern Califor-
nia School of Medicine's Department of Psychiatxy.
Twenty patients were evaluated and diagnosed as
having a panic disorder, with or without agorapho-
Otolaryngology-
Head and Neck Surgery
Volume 110 Number 3
bia, by the SCID (Structured Clinical Interview for
DSM-111-R, Patient Version) described in Table 1 . I '
They were medication-free for at least 1 week before
testing, were unselected for specific vestibular symp-
toms such as dizziness or unsteadiness, and were
randomly selected from patients who were consid-
ering participation in a medication trial for their
panic symptoms. Before VAT testing, patients filled
out a questionnaire similar to those in common use
for evaluation of vestibular symptoms.
Eight patients had panic disorder without agora-
phobia and eleven patients had panic disorder with
agoraphobia. All patients had experienced four
panic attacks within a 4-week period and had on-
going panic attacks. The patient group consisted of
eight men and eleven women, with a mean age of
39.1 ycars (range, 29 to 55 years). Patients with
agoraphobia without panic disorder were not in-
cluded in this study because the condition is rare in
clinical populations. They were ill for an average of
9.8 years (range, 6 months to 40 years), with a mean
educational level of 14.2years. They completed their
vcstibular tcsing, as did the panic group, bcforc
being entered into the drug trial.
Patient Inclusion/Exclusion Criteria
1. Patients were of ages 18 to 65 ycars and able to
attend the clinic.
2. They had to meet the DSM-Ill-R criteria for
panic disorder or panic disorder with agora-
phobia. They could not be pregnant, o r have:
a. severe psychiatric disorders including
schizotypal disorder
b. psychotic symptoms
C. substance abuse or alcoholism (within one
year)
d. dementia
e. severe personality disorder
f. clinically unstable disease ( a s dctcrmincd
by history. physical examination, EKG, and
cl i n i ca I I a bo ra t ory t c st s
g. diseases that would interfere with the diag-
nosis and trcatmcnt of panic disorder
Patients could not have received antidepressants
within the last 3 weeks or monoaminc oxidase in-
hibitors, anxiolytics, antipsychotics, o r lithium within
the last 4 weeks. Fifteen of the 19 tested patients
with panic disorder were in a drug trial that allowed
varying degrees of depression. as long as the panic
disorder began first and dominated the clinical pic-
ture. They were given the Montgomery-Asberg De-
pression Rating Scale" with a mean baseline score of
17.36. range of 0 to 42; thc Clinical Anxiety
Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016
HOFFMAN et al. 261
Fig. 1. Instrumented head strap worn by the patient during
the vestibular autorotation test (VAT, Western Systems Re-
search, Inc., Pasadena, Calif.]. The box on the side of the
head strap contains the EOG amplifier for recording eye
movements. The cylinder attached to the back of the hod
contains an electronic sensor that records head rotational
velocity. The cables connect to the computer, which collects
the data, generates an auditory cue, and computes gains
and phases.
with a mean baseline score of 13.55, range of 5 to 22;
and the Clinical Global Impression'" with a mean
baseline score of 4.63, range of 3 to 7. The other
patients were in a drug trial in which no patient
could presently meet the criteria for major depres-
sion disorder. They were given the Hamilton Anxi-
ety Rating Scale'5with a mean baseline score of 20.3,
range 15 to 30; the Hamilton Depression Rating
Scale"' with a mean baseline score of 1 1.67, range 9
to 14; and a Clinical Global Impression'" with a
mean baseline score of 3.67, range 3 to 4. Therefore
in neither group did patients have a primary diag-
nosis of major depression.
VAT Procedures
The horizontal and vertical Vestibulo-ocular Re-
flexes (VORs) of each patient were tested by the
vestibular autorotation test (VAT). The VAT is a
computerized method based on active head move-
ments over a frequency range from 2 to 6 Hz."'.'' At
frcqucncics above 2 Hz, the VORs arc the primary
systems for ocular gaze fixation because other ocular
movement systems (e.g., smooth pursuit) are mini-
mally cffcctivc in this range."' The VAT protocol is
as follows. Seated patients were fitted with conven-
tional electro-oculographic (EOG) electrodes and
lightweight head strap attached to ii rotational vc-
 Waryngdogy-
Head and Neck Surgery
262 HOFFMAN et al. March 1994

- 3 0 0 1~ '2~ '3 " "4 " 5" " 6" "7 " "8 " ~ 13 ' 14
~ '11~ "12' ~
9 ' 10 ~ ' 15
~ 16 17 18

SECONDS

-300~'""""" 1 ' 1 ' 1 ' 1 1 1 ' 1 ' 1 ' 1 ' ' ' 1 ' 1 ' 1 ' '
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
SECONDS

Flg. 2. Head and eye VelOCWy trajectories for patient D13: 4 horlzontal eye Velocity; 8, horizontal
head velocity; C, vertical eye veloclty; and D, vertlcal head velocity. Honiontal ban beneath axes
Indicate 2-second epochs used for EOG calibration. Durlng the latter part of the test epoch,
horizontal eye veloclty amplltudes were smaller than head velocity amplitudes.

1 0 locity sensor and an EOG amplifier (Fig. 1). Hori-
zontal eye movements were recorded with bilateral
electrodes positioned at the outer canthi and a
reference electrode positioned above the bridge of
the nose. Vertical eye movements were recorded
with electrodes placed above and below one eye.
Head velocity was recorded with a calibrated veloc-
ity sensor fixed to the headband worn by the patient.
A computer-generated tone was used as an audible
cue to direct the frequency of head motion while the
computer program swept the frequency from 0.5 to
6.5 Hz during the 18-second test epoch. Two in-
structions were given: (1)"stare at the wall-mounted
target" (1-cm disk) and (2) "move your head
smoothly from side to side in time to the computer-
generated tone." After a 30-second rest, the same
procedure was repeated twice more for a total of
three repetitions with horizontal head movements
and then repeated with a total of three vertical head
movements in a nose up-nose down direction. Be-
havioral anxiety reduction techniques were used if a
panic situation arose during or after the test. Eye
position and head velocity data were amplified and
digitized directly to disk files using an IBM-PC
compatible computer equipped with data acquisi-
tion peripherals. Data from the first six seconds were
used for EOG calibration. Gain and phase were
computed during the last 12 seconds of the test
epoch.
The VAT analysis procedures have been reported
previ~usly."'~~~~'~
In brief, gain is defined as the eye
velocity amplitude divided by the head velocity am-
plitude. Phase is the time lag in degrees of the eye
velocity in relation to the head velocity. Asymmetry
is the amount of drift of the eye towards one side. All
three characteristics are frequency dependent. An
ideal VOR would be expressed as gain = 1 and
phase = 180 degrees with no asymmetry. In pa-

Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016

OtOla~gOlOgY -
Head and Neck Surgery
Volume 110 Number 3 HOFFMAN et al. 263

Fig. 3. Mean ( + I standard deviation] of gains and phases of patlent D13: A, horizontal gain;
B, horizontat phase; C, vertical gains. and D,vertical phase. Filled trianglesindicate patient means;
open clrcles indicate normal subject means; error b a r s indlcate -t I standard devlotion. Horizontal
and vertlcal phase lags were greater than normal.

tients, the lack of the ability of eye velocity to follow
head velocity can indicate pathology, through gains
and phases that differ from normal. Eye drifts to the
right or left could indicate pathology when they
occur systematically toward one side.''
The means and standard deviations from three or
more horizontal and vertical VATS from each pa-
tient with panic disorder were compared with those
from a normal control population (N = 100) tested
previously after screening for absence of vestibular
disorders. A VAT result was considered clinically
abnormal if two or more means and standard de-
viations of gain or phase data points had error bars
that were clearly separable from those of the normal
group in one or more of the four plotted graphs:
horizontal and vertical, gains and phases.
Asymmetry plots were generated from each pa-
tient's data by determining the ratio of the eye
position deviation from straight-ahead position and
the amount of spectral energy at each frequency, as
a percent." This was done by Fourier analysis.
Asymmetry in the VOR suggests that the amount of
neural impulses per unit time contributing to the
extraocular muscles is weaker on one side, which
causes the eye to drift in the orbit to that side during
active head movement.'' Asymmetry suggests the
presence of a unilateral lesion and the direction of
the eye drift is toward the side of that lesion."
For the total number of patients showing abnor-
mal VAT results (described later), the Binomial
Test2' was used to determine the significancelevel of
panic disorder patients with abnormal vestibulo-
ocular tests from the total sample of patients used in
this study.
RESULTS
Data from VAT testing were obtained from 19
patients. One additional patient experienced a panic
attack during the test, damaging the head movement
sensor when he attempted to leave the room. No
data were obtained from this patient. Figure 2 shows
head and eye velocity trajectories for panic patient
D13 from a horizontal and vertical VAT test. The
patient's dizziness questionnaire reported symp-
toms of dizziness, unsteadiness, and veering toward
the right side during walking, lightheadedness, and
marked dizziness while driving. The trajectories
shown were from the second of three horizontal and
three vertical VAT tests administered. During the
latter part of the test epoch of this patient's hori-
zontal eye velocity, amplitudes were slightly smaller
than head velocity amplitudes.

Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016

 OtolaryngoloW-
Wwd and Neck S u m
264 HOFFMAN et al. March I994

HORIZONTAL EYE VELOCITY Pt: D11

1 1 ArrlA

SECONDS

Fig. 4. Head and eye velocity trajectories for patlent D1I : A, 8, C, and D are labeled as descrlbed
for Fig. 2. Doshed p e a k s in C Indicate eye blinks that were removed from the vertical eye records
before gain and phase onalysls. During the latter part of the test epochs,the eye velocity amplitudes
appear slightty larger than head veloclty amplitudes.

Fig3 w Figure 3 shows the results (means & 1 SD) from
the three horizontal and three vertical VAT test
results from panic patient D13, superimposed with
data from normal subjects. Gains were within nor-
mal ranges. Horizontal and vertical phase values
were numerically greater than normal, peaking at
220 degrees at 6 Hz in horizontal phase and 225
degrees at 6 Hz in vertical phase.
Figure 4 shows vigorous eye and head velocity
trajectories from panic patient D11. Eye amplitudes
appear equal to or slightly larger than corresponding
head amplitudes during higher-frequency move-
ments. The patient reported a 6-year history of
dizziness attacks occurring twice a week with 1- to
2-minute durations that could be evoked by rapidly
sitting up.
Figure 5 shows the results (means ? 1 SD) from
three horizontal and two vertical tests for patient
D11. The horizontal mean gains were consistently
greater than 1.1 from 3 to 6 Hz, whereas in normal
controls the mean gains ranged from 0.9 to 1.0 at
these frequencies. The mean horizontal phase lags
were less than the normal ranges, peaking at about
178 degrees at 4.3 Hz. The mean vertical gain
showed high means, increasing to 2.0 at 6.0 Hz,
which was considerably above the normal range. The
mean vertical phase lags were in the low normal
range. The high horizontal gain was accompanied by
a horizontal phase less than normal from 3 to 5 Hz.
It is noteworthy that abnormality in either gain and
phase can cause oscillopsia for movement velocities
that exceed about 3 deg/sec (i.e., a perceptual
threshold). In this patient, the abnormal horizontal
and vertical responses resulted in a relatively rapid
retinal image velocity (retinal slip) in both VOR
systems.
Figure 6, from patient D1, illustrates an example
of normal horizontal and abnormal vertical gains

Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016

Ot~aryngdogy-
Head and N e c k Sugery
Volume 110 Number 3 HOFFMAN et al. 265

Fig. 5. Means [? 1 standard deviation) of gains and phases of patient D11: A, 8, C, and D show
horizontal galns and phases and vertical gains and phases, respectively, as described in Fig. 3
legend. This patient’s mean galns are greater than normal at hlgher frequencles. Mean horizontal
phases are numerically less than normal from 3 to 5 Hz, whereas mean vertlcal phase lags are wlthln
normal ranges.

and phases. The vertical mean gain at frequencies
above 3 Hz was significantly greater than normal. In
contrast, the vertical phase means were numerically
greater than normal at frequencies less than 3 Hz.
Analysis of the mean g a i r and phase plots from
each of the 19 patients showed a striking result: data
from all patients were found to be significant& different
from the normative group in at least one of the four
VOR characteristics: gain or phase in horizontal or
vertical planes. Because abnormalities in any of the
four VOR characteristics are indicative of VOR
dysfunction, all 19 of the patients were therefore
considered abnormal. This result is highly significant
(p c 0.00001) by the Binomial Test,I9 even under
the most conservative assumption of a probability of
0.5 that half of any given population would show
abnormal VAT results.
Table 2 shows the individual test abnormality
ratios and combined percent abnormalities for each
patient. The direction of numerically high or low
values are shown by letters H and L, respectively. A
categorization of individual patient results showed
that 13 patients were abnormal in horizontal gain, 17
were abnormal in horizontal phase, nine were ab-
normal in vertical gain, and 12 were abnormal in
vertical phase.
Responses of the 19 patients with panic disorder
and with agoraphobia on the patient questionnaire
revealed that although these patients represented
consecutive evaluations in which patients were not
selected for symptoms of dizzinesor imbalance, 18
of 19 answered “yes” to the question “Are you
suffering from dizziness?” with an average of 7.9
years of symptoms. Of the 18 patients who reported
dizziness, 15 described the dizziness as occurring
exclusively in attacks, whereas in the three others
the answer was ambiguous. Of these 18,13 indicated
a tendency to fall, four patients denied this tendency,
and one patient’s answer was ambiguous. Of the 19
patients, nine indicated that they experienced ob-
jects spinning or turning around them, six reported
a sensation that they themselves were spinning or
turning, 17 reported lightheadedness, 12 reported a
loss of balance while walking, and six reported full-
ness in one or both ears.
Figure 7, A shows an eye position asymmetry
toward the right side of patient D7, indicating eye
drift in the orbit toward the right during active head
movements. Figure 7, B shows the same patient’s
compensatory head position drift toward the left.
The mean head and eye position were oppositely
directed, as expected from the biomechanics of at-
tempted target fixation during active head move-
ments. Similarly, Fig. 7, C and D show respective eye

Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016

 Otolaryngology-
Head and Neck Surgery
266 HOFFMAN et al. March 4994

Flg. 6. Means [ ? l standard devlotion) of gains and phases of patient D l : A, 8, C, and D show
horizontal gains and phases and vertical gains and phases, respecttvely, as described in Flg. 3
legend. Thls patlent's data show abnormally high vertical mean golns ot frequencles greater than
3 Hz, and abnormally large vertical mean phases at frequencles less than 3 Hz.

Table 2. Test abnormallty ratios for each patient (number of abnormal points/total points in each test]
Patient H-Galn H-Phase V-Galn v-Phase Total %* ArymmeW
~~~~~~

D1 H (3/11) N (1/11) H (7/9) H (3/10) 34 R

02 N (0/8) N (0/8) N (O/lO) H (U10) 6 -
D3 N (OM) L (2/4) N (1/9) H (2/9) 19 R
04 H (8/9) L (9/9) H (5/10) L (410) 68 -
D7 H (3/10) L (10/17) H (7/8) N (OD) 54 R
D8 H (2/7) L (6/7) L (4/9) N (0/9) 38 -
D9 H (6/11) L (4/11) L (a41 L (4/4) 53 L
D10 N (0/9) L (7/9) N (0/10) L (6/10) 34 -
D11 H (6/11) L (3/11) H (3/11) N (0/11) 27 A
D12 L (4/9) H (4/9) N (O/ll) N (0/11) 48 R
D13 N (1/11) H (3/11) N (0/11) H (6/11) 23 -
D14 L (3/11) L (10/11) L (4/4) L W4) 63 L
D15 H (1 1/11) L (1 1/11) L (4/6) H (4/6) 88 L
016 i (4/10) i (9/10) N (0/7) H (4/7) 50 L
D17 L (2/8) L (6/8) N (Oh) N (1/8) 28 L
D18 H (9/11) L (Ull) H (4/5) N (1/5) 50 R
D19 H (4/11) L (8/11) N (0/4) N (0/4) 40 R
D20 N (1/9) L (9/9) N (0/11) L (8111) 45 R
021 N (0/11) L (9/11) N (Oh) L (3/8) 32 R

'Total % = 100% x (Total abnormal pointsflofa1 points in four tests).

t R , Right asymmetry; L, left asymmetry; -, no asymmetry.
H a n d L notations in the gain and phase columns indicate values numericalty higher or lower than normal ranges, respectively; N nolation indicates
fewer than two abnormal points.

and head position data from patient D15, displaying
left eye asymmetry, and compensatory head position
drift toward the right.
According to Ewald's Law," asymmetry in eye
position indicates that the eye drifts toward
the weaker side, and this is often observed to
occur during unilateral labyrinthine pathology.'*
It is noteworthy that VOR asymmetries were
found to be present in 14 of the 19 patients
(Table 2), with nine showing right-deviating asym-

Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016

~~rvneolosv-
Head and N e c k Surgery
Volume 110 Number 3 HOFFMAN el al. 267

A 25
8
a 0
-25
L I I IVL
-50;' I
1
"
2
I
3
' "
4
' 5 6 7
' 6' I
9
I '
10
' 11
' I
12
"
13
' 14
I . ' ' ' ' ' I
' 15 16 17 18

-----

SECONDS

SECONDS

Fig. 7. Eye and head position asymmetries are shown for patient D7 (r and B] and patlent D15
[C and D), plotted as posfflon vs. time. Patient D7 shows an eye asymmetry toward the rlght and a
compensatory head asymmetry toward the left, whereas the results for patient D15 ore oppositely
dlrected.

metries and five showing left-deviating asym-
metries.
DISCUSSION
The most important result of this study is our
finding of high-frequency VOR abnormalities in all
19 patients, diagnosed with panic disorder by estab-
lished DSM 111-R criteria. Our results from high-
frequency VOR testing are in agreement with lower-
frequency rotational chair studies:8,21 which found a
high prevalence of vestibulo-ocular reflex abnor-
malities in a group of similarly diagnosed patients. It
is noteworthy that patients in our study were unse-
lected for the presence or absence of vestibular
symptoms. In contrast, previous rotational chair
studies of patients with panic disorder used selected
patients with vestibular symptoms.6.s.21
Oscillopsia, defined as perceived motion of the
visual field,u can result from gains that are either
high or low, and/or phases that are either advanced
or retarded, relative to normative values. Abnormal
gain or phase values are indicative of the relative
motion of images on the retina, which is perceived as
oscillopsia when the retinal image velocity exceeds a
perceptual threshold of about 3 deg/sec."
The observed VOR abnormalities in this study
were not exclusive to one pattern, but were of
various patterns (e.g., abnormal horizontal and/or
vertical gains or phases relative to a normal control
group). The high incidence of horizontal phase ab-
normality in this population suggests that the com-
pensatory eye movement lagged the head movement
during horizontal movements, resulting in oscillop-
sia for movements that exceeded the 3 deg/sec per-
ceptual threshold.
Visuospatial disturbances are commonly reported
in panic patients." Reported disturbances often in-
clude a feeling of unsteadiness, floating or turning,
and the illusion of veering towards one side while
walking. It is reasonable to speculate that these
visuospatial disturbances could be similar to those
reported by patients having difficulty driving on
certain roadways. Page and Gre~ty,2~ for example,
describe the motorist's vestibular disorientation syn-
drome in which automobile drivers reported an
illusion of veering toward one side and, in an at-

Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016


268 HOFFMAN et al.
tempt to compensate for this illusion, steered their
car out of their correct driving lane. This “veering to
one side” phenomenon and the apparent motion of
surrounding objects has been reported by Jacob et
a1.= and also as a “supermarket syndrome,”whereby
some individuals experience anxiety walking down
supermarket aisles.
In addition to the potentially disabling effects of
oscillopsia (images moving on the retina), the pres-
ence of VOR asymmetries could contribute addi-
tionally to illusory visuospatial disturbances re-
ported by panic patients. A superposition of oscil-
lopsia and VOR asymmetry would result in a
simultaneous jumbling and veering of the visual
field. If oscillopsia contributes to panic symptoms
(nausea, heart palpitations, sweating), it is unlikely
that this occurs directly. Most patients with balance
disorder with visual field instability do not express
panic symptoms. One possible explanation may be
found in the cognitive theory of panic proposed by
Clark.” Clark suggested that a catastrophic misin-
terpretation of any bodily symptom in a biologically
or psychologically predisposed individual can lead to
anxiety, additional bodily symptoms, and more anxi-
ety in an increasingly disturbing spiral, leading to
panic. Perhaps oscillopsia and simultaneous VOR
asymmetries could provide such a bodily symptom
“trigger” for such a spiral. In support of this cogni-
tive catastrophic misinterpretation theory of panic,
Kenardy et al.27reported that vestibular symptoms
contributed to the more catastrophic outcomes pro-
duced by patients with panic disorders than those
with social phobia, in which vestibular symptoms
occur much less frequently.
Because this is the first reported use of the VAT
in evaluating the VOR at higher frequencies in
anxiety disorders, it is useful to consider how the
VAT differs from conventional and previously used
VOR testing methods”:
Lack 07 nystagmus. Both electronystagmogra-
phy and rotational chair testing rely on evoked
nystagmus, whereas nystagmus does not gen-
erally occur at the amplitudes and frequencies
used in the VAT.
Measured characteristics. The high-frequency
VAT results used in this study are a precise
quantification of “retinal image velocity” re-
sulting from natural, active head movements,
analogous to those that occur during locomo-
tion. Rotating chair and ENG testing neces-
sarily overdrive the system by delivering stimuli
that seldom occur naturally.
Horizontal and vertical VOR testing. Both sys-
Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016
Walyngology-
Head and Neck Surgmy
March 1994
tems are tested in the VAT. The vertical VOR
system is generally ignored in other conven-
tional testing methods.
4. Conuption by cortical andlor other ocular move-
ment systems. These are minimally contributory
at frequencies tested by the VAT. The VOR
can be effectively tested in the light without
corruption from other systems. In contrast,
chair and ENG testing must be done in the
dark and with patient-alerting techniques, to
avoid contamination by other systems.
In conclusion, our results describing higher-
frequency vestibulo-ocular abnormalities in our en-
tire sample of patients with panic disorder and panic
disorder with agoraphobia implies a strong correla-
tion of this disorder with the existence of a balance
system disorder. Whether such a balance disorder
can be considered causally related to panic disorder
and whether it can be localized to specific regions of
the peripheral labyrinth and/or the diffuse central
neural projection pathways of the VOR requires
further investigations.
REFERENCES
1. Errera P. Some historical aspects of the concept, phobia.
Psychiatr Q 1962;36:325-6.
2. Guye A. On agoraphobia in relation to ear disease. Laryn-
goscope 1899;6:219-21.
3. American Psychiatric Association. Diagnostic and statistical
manual of mental disorders. 3rd ed. Washington D.C.:
American Psychiatric Association, 1987:235-41.
4. Barlow DH. Anxiety and its disorders: the nature and treat-
ment of anxiety and panic. New York: The Guilford Press
1988:73-111.
5. Barlow DH. The dimensions of anxiety disorders: In: Tuma
AH, Maser JD, eds. Anxiety and the anxiety disorders. New
Jersey: Eilbaum, 1985.
6. Jacob R. Panic disorder and the vestibular system. Psychiatr
Clin North Am 1988;11:361-71.
7. Sklare DA, Stein MB, Pikus AM, Uhde TW.Dysequilibrium
and audiovestibular function in panic disorder: symptom
profiles and test findings. Am J Otol 1990;11:33841.
8. Swinson RP, Cox BJ, Rutka J, Mabel M, Kerr S, Keech K.
Otoneurological functioning in panic disorder patients with
prominent dizziness. Compr Psychiatry 1993;34:127-9.
9. Grossman GE, Leigh JR, Bruce EN, Huebner WP,Lanska
DJ. Performance of the human vestibuleocular reflex during
locomotion. J Neurophysiol 1989;62264-72.
10. OLeary DP, Davis LL. High frequency autorotational testing
of the vestibulo-ocular reflex. Neurol Clin 1990;8:297-312.
11. Spitzer RL, Williams JB, Gibbon M. Structured clinical
interview for DSM-111-R-patient version (SCID-P). New
York: Biornetrics Research Department, New York State
Psychiatric Institute, 1987.
12. Montgomery SA, Asberg M. A new depression scale designed
to be sensitive to change. Br J Psychiatr 1979;134:383-9.
13. Snaith RP, Raugh SJ, Clayden et al. The Clinical Anxiety
Scale: an instrument derived from the Hamilton Anxiety
Scale. Br J Psychiatr 1982;141:518-23.
~~acyngology-
Head and Neck Surgery
Volume I I 0 Number 3
14. Guy W, ed. ECDEU Assessment Manual for Psychopharmo- 22.
cofoa revised. DHEW Pub. No. (ADM)76-338. Rockville,
MD: National Institute of Mental Health, 1976.
15. Hamilton M. The assessment of anxiety states by rating. Br 23.
J Med Psychol 1959;32:50-5.
16. Hamilton M. Development of a rating scale for primary
depressive illness. Br J Med Psychol 1967;6:278-96.
17. Fineberg R, O’Leary DP, Davis LL. Use of active head
movements for computerized vestibular testing. Arch Oto- 24.
laryngol Head Neck Surg 1987;113:1063-5.
18. OLeary DP, Davis LL, Maceri DR. Vestibular autorotation
test asymmetry analysis of acoustic neuromas. OTOIARYNC~L 25.
HEADNECKSURG 1991;104:103-9.
19. Ewald ETR. Physiologische Untersuchungen Uber des Ner-
vus Octavus. Wiesbaden: Bergmann, 1892.
20. Siege1 S. Nonparametric statistics. New York: McGraw Hill, 26.
1956:36.
21. Jacob RB, Moller MB, Turner SM, Wall C. Otoneurological 27.
examination in panic disorder and agoraphobia with panic
attacks: a pilot study. Am J Psychiatr 1985;6:715-20.
Downloaded from oto.sagepub.com at WESTERN MICHIGAN UNIVERSITY on June 5, 2016
HOFFMAN el at. 269
Barnes GR, Smith R. The effect on visual discrimination of
image movement across the stationary retina. Aviat Space
Environ Med 1981;52:466-72.
Udhe T, Roy-Byme P, Vittone Bernard J, Boulenger JP, Post
RN. Phenomenology and neurobiology of panic disorder. In:
Tuma AH, A Hussain, Maser JD, eds. Anxiety and anxiety
disorders. Hillsdale, N.J.: Lawrence Erlbaum Assocs, 1985:
557-76.
Page N, Gresty M. Motorist’s vestibular disorientation
syndrome. J Neurol Neurosurg Psychiatry 1985;48:729-
35.
Jacob RG, Lilienfeld MA, Furman JMR, Durrant JD, Turner
SM. Panic disorder with vestibular dysfunction: further clini-
cal observations and description of space and motion phobic
stimuli. J Anxiety Disord 1989;3:117-30.
Clark DM. A cognitive approach to panic behavior research
and therapy. Behav Res Ther 1986;24:461-70.
Kenardy J, Evans L, Oei TP. The latent structure of anxiety
symptoms in anxiety disorders. Am J Psychiatry 1992149:
1058-61.