Strabismus after Deep Lateral Wall Orbital
Decompression in Thyroid-Related
Orbitopathy Patients Using
Automated Hess Screen
Guy J. Ben Simon, MD, Ahmad M. Syed, MD, Seongmu Lee, BS, Debbie Y. Wang, BS,
Robert M. Schwarcz, MD, John D. McCann, MD, PhD, Robert A. Goldberg, MD
strabismus in thyroid-related orbitopathy (TRO) patients using automated Hess screen (AHS).
Prospective nonrandomized clinical study.
Eleven TRO patients (19 surgeries) operated on at the Jules Stein Eye Institute from January,
2004, through December, 2004.
Automated Hess screen testing was performed in all patients before surgery and 3 months after
surgery; all patients received surgery in the nonactive phase of the disease.
Main Outcome Measures:
Amplitude of horizontal and vertical deviations (prism diopters) in all standard
positions of gaze.
Eleven TRO patients (7 females; mean age, 47 years) were included in the study; 8 patients
underwent bilateral surgery. After surgery, exophthalmos decreased an average (?standard deviation) of 2.7 mm
(?2.5 mm; P ? 0.003). Before surgery, 7 patients (63%) reported primary gaze diplopia, whereas only 2 patients
(18%) showed diplopia in primary gaze after surgery (P ? 0.03, chi-square analysis). Orbital decompression had
no statistically significant effect on horizontal and vertical ocular deviations measured by AHS. Mean amplitude
of deviation in primary gaze was 1.2 prism diopters (PD) esotropia and 0.07 PD hypotropia before surgery, and
2.5 PD exotropia with 0.6 PD hypertropia after surgery (? ? 3.7 PD for horizontal deviation and ?0.7 for vertical
deviation; P ? 0.051, paired samples t test for horizontal difference and P not significant for vertical difference).
Nonsignificant P values were obtained in all 9 positions of gaze. Most patients had periocular numbness that
resolved spontaneously 2 to 6 months after surgery.
Deep lateral wall orbital decompression with intraconal fat debulking had no statistically
significant effect on horizontal and vertical deviations measured by the AHS. Patients may demonstrate small
angle exotropia shift, but this finding was not clinically significant. Ophthalmology 2006;113:1050–1055 © 2006
by the American Academy of Ophthalmology.
To evaluate the effect of deep lateral wall orbital decompression with intraconal fat debulking on
Orbital decompression is an important cornerstone in the
surgical rehabilitation of thyroid-related orbitopathy (TRO)
patients. It is performed to treat disfiguring proptosis,1along
with corneal exposure and optic neuropathy.2,3
Several surgical options are considered in orbital decom-
pressions, including removing the medial, inferior, or deep
lateral orbital wall, with or without intraconal fat debulk-
ing.4–8Surgery is tailored to each patient, and often patients
with more severe proptosis will require additional and ex-
tensive tissue removal, with removal of bone, intraconal
fat, or both. Surgery is performed only during the non-
active phase of the disease, initially with orbital decom-
pression, followed by eye muscle surgery and some type
of eyelid retraction procedures with or without upper eyelid
One of the more common complications associated with
orbital decompression is development or worsening of dip-
lopia, with primary or downgaze diplopia being most cum-
bersome.8,11–14Medial wall and floor decompression surger-
ies are associated with up to 30% of new-onset strabismus,4
whereas deep lateral wall decompression may be associated
with a much lower rate of new-onset diplopia (2%–
15%).6,15–17Recently, we published results that showed a
new-onset primary gaze diplopia rate of only 2.6% in pa-
tients with mild to moderate TRO undergoing deep lateral
wall decompression with intraconal fat debulking.15
To understand the biomechanics of this surgical tech-
nique and to analyze its influence on ocular deviations, we
Originally received: July 5, 2005.
Accepted: February 13, 2006.
From the Jules Stein Eye Institute and Department of Ophthalmology,
David Geffen School of Medicine at the University of California Los
Angeles, Los Angeles, California.
Correspondence to Guy J. Ben Simon, MD, Goldschleger Eye Institute,
Sheba Medical Center, Tel Hashomer, Ramat Gan, Israel 52621. E-mail:
Manuscript no. 2005-600.
© 2006 by the American Academy of Ophthalmology
Published by Elsevier Inc.
ISSN 0161-6420/06/$–see front matter
used an automated Hess screen (AHS)18before surgery and
after surgery in TRO patients undergoing deep lateral wall
decompression. The purpose of the current study was to
describe surgical effect on ocular deviations as assessed by
Patients and Methods
We evaluated in a prospective fashion 11 consecutive TRO pa-
tients who were scheduled to undergo deep lateral wall orbital
decompression with intraconal fat debulking at the Jules Stein Eye
Institute from January, 2004, through December, 2004. Compre-
hensive eye examination was performed in all patients. Automated
Hess screen testing was performed 2 to 4 weeks before surgery. All
patients were operated on during the nonactive phase of the dis-
ease. The AHS test was repeated 3 months after surgery, and
amplitudes of horizontal and vertical deviations in all 9 positions
of gaze were recorded. The study was approved by the local
institutional review board. The indications for decompression sur-
gery in these patients were mostly cosmetic, although some had
pressure pain associated with thyroid eye disease.
Automated Hess Screen
We used the AHS testing by Lifelearn Eyecare (School of Optom-
etry, University of Waterloo, Waterloo, Canada; Thompson Soft-
ware Solutions, City University, London, UK). Testing was per-
formed as follows. The patient was seated with his/her head
restrained 30 cm from a 17-inch color monitor and viewed the
screen through red–blue goggles.18A red target (small circle) was
displayed on the screen, and the patient was instructed to move a
blue stimulus (larger circle) using a mouse control until the target
appeared to be centered on the stimulus. When the patient pressed
the left mouse button, the computer registered the relative position
of the target and stimulus. This procedure was repeated for a range
of target locations (9 positions of gaze), and the entire routine then
was repeated with the other eye fixing, by reversing the color of the
target and stimulus. The amplitude of horizontal and vertical
deviation in each position of gaze of was calculated and displayed
on the monitor along with the AHS results. Patients were given 1
to 2 practice tests to acclimate to the use of the AHS testing. Figure
1 shows an example of AHS testing in a normal participant.
Each point on the AHS was numbered 1 through 9 to match
preoperative and postoperative points. Esodeviation was marked as
positive horizontal deviation (?), and exodeviation was marked as
negative horizontal deviation (?). Similarly, hypertropia (right
over left for the right eye and left over right for the left eye) was
assigned as a positive vertical deviation, whereas hypotropia was
marked as a negative (?) vertical deviation. We calculated the
difference between horizontal values (before surgery minus after
surgery) in all 9 positions of gaze and in vertical values (before
surgery minus after surgery) in all 9 positions of gaze. Cyclode-
viations were not measured in the current study.
The surgical technique was described previously.5The orbital
surface of the sphenoid bone was exposed through an eyelid crease
incision. Using a high-speed neurosurgical drill, cortical bone was
removed from the lacrimal gland fossa, the marrow space of the
sphenoid between the superior and inferior orbital fissure, and the
zygomatic marrow space on the anterior rim of the inferior orbital
fissure. The extent of bone removal was individualized: patients
with substantial proptosis (for example, more than 26 mm) under-
went maximal bone removal from each of the 3 areas, but patients
with lesser degrees of proptosis were treated with a more conser-
vative bone removal. In all patients, the maximal available intra-
conal fat, located between the lateral and inferior rectus muscle,
was dissected bluntly out of the muscle cone and was excised. The
volume of fat removed ranged from 1.5 to 3 ml.
Figure 1. Output of automated Hess screen (AHS) used in the current study in a healthy participant. Left image shows field of the left eye (right eye
fixating), and right image shows field of the right eye (left eye fixating). Horizontal and vertical strabismus measurements (amplitude of deviations) in prism
diopters are shown in all 9 positions of gaze. ESO ? esotropia; EXO ? exotropia; LHYPER ? left hypertropia; RHYPER ? right hypertropia.
Ben Simon et al ? Strabismus after Deep Lateral Wall Orbital Decompression
The Wilcoxon signed-rank test was used to compare preoperative
and postoperative values of visual acuity, intraocular pressure, and
exophthalmos. A paired samples t test was used to calculate the
difference in ocular deviations measured by AHS testing; one
sample t test was used to calculate ? values of amplitudes of
deviations. Cross tabs and chi-square analysis was used to calcu-
lated proportion difference for patients with primary gaze diplopia
before surgery and after surgery. Pearson bivariate correlation was
used to evaluate preoperative and postoperative amplitudes of
deviations in all positions of gaze. Conversion of Snellen acuity to
a logarithm of the minimum angle of resolution value was per-
formed. Statistical analysis was carried out using Microsoft Excel
2003 (Microsoft Corporation, Redmond, WA) and SPSS software
version 13.0 (SPSS, Inc., Chicago, IL).
Eleven TRO patients (7 females; mean age, 47 years) who under-
went 19 deep lateral wall orbital decompressions with intraconal
fat debulking surgeries were included in the study; 8 patients
underwent bilateral surgery. After surgery, exophthalmos de-
creased an average (?standard deviation) of 2.7 mm (?2.5 mm; P
? 0.003, Wilcoxon signed-rank test). Visual acuity remained
unchanged after surgery (? logarithm of the minimum angle of
resolution visual acuity, 0.03; P ? 0.2, one-sample t test; Table 1).
Before surgery, 7 patients (63%) reported primary gaze diplo-
pia, whereas only 2 patients (18%) showed diplopia in primary
gaze after surgery (P ? 0.03, chi-square analysis). Only 2 patients
had clinically significant postoperative strabismus and required
eye muscle surgery after undergoing orbital decompression.
Overall, patients showed a very low degree of strabismus
before and after surgery with amplitudes of deviations ranging
from 1.6 PD exotropia to 1.2 PD esotropia and 1.1 PD hypertropia
to 2.2 PD hypotropia before surgery and 2.5 PD exotropia to 0.7
PD esotropia with 1.2 PD of hypertropia to ?2.8 PD of hypotropia
after surgery. Positive correlation was found between preoperative
and postoperative vertical deviations (R ? 0.87; P ? 0.002,
Pearson bivariate correlation), but not in horizontal deviations (Fig
Orbital decompression had no statistically significant effect on
horizontal and vertical ocular deviations measured by automated
Hess screen. Mean amplitudes of deviations in primary gaze were
1.2 PD esotropia and 0.07 PD hypotropia before surgery, and 2.5
PD exotropia with 0.6 PD hypertropia after surgery (? ? 3.7 PD
for horizontal strabismus and ?0.7 for vertical strabismus; P ?
0.051, paired samples t test for horizontal difference and P not
significant for vertical difference). Nonsignificant P values were
obtained in all 9 positions of gaze (Table 2); the lack of statistically
significant influence of surgery on postoperative horizontal and
vertical strabismus is reflected by low ? values and nonsignificant
P values as shown in Table 2 (paired samples t test).
Interestingly, the largest ? value was measured in horizontal
strabismus in primary gaze (3.7-PD exotropia shift); this small
angle deviation, however, does not represent clinically significant
strabismus. Subgroup analysis of patients with no preoperative
primary gaze diplopia showed similar results, with no effect of
surgery on horizontal and vertical strabismus in all positions of
gaze; none of the P values were significant (data not shown). No
correlation was found between the changes in ocular deviation and
preoperative exophthalmos (Pearson bivariate correlation).
No severe complications such as visual loss, cerebrospinal fluid
leak, or death occurred in this group of patients. Most patients
reported periocular sensory anesthesia that resolved spontaneously
2 to 6 months after surgery.
The current study shows that deep lateral wall orbital de-
compression with intraconal fat debulking had no statistical
effect on horizontal and vertical ocular deviations measured
by AHS testing. These results support our previous report15
and our assumption that minimal to no ocular shift occurs
with this type of orbital decompression.
Automated Hess screen testing using a personal com-
puter was found to be a viable alternative to electronic Hess
screen testing and showed similar amplitudes of deviation in
patients with a wide range of oculomotor problems.18We
used AHS testing in the current study for its simplicity to
operate, and it can be performed by the treating physician as
a part of a comprehensive eye and orbital evaluation of the
TRO patients. The average time for each test was 5 to 10
minutes, and this time diminishes with repeated tests. We
believe that the AHS is a good test for follow-up of TRO
patients with ocular motility disturbances; however, this
was not assessed in the current study.
A recent study19examined the mechanism of ocular
motility disturbances after orbital decompression surgery.
The investigators used magnetic resonance imaging to mea-
sure the positions and the displacements of the anterior
rectus muscle paths. In general, postoperative muscle path
positions were similar to preoperative positions and to those
of normal participants. However, centrifugal displacement
of the inferior rectus and the medial rectus muscles was
noted. The amount of muscle displacement (inferior rectus
or medial rectus) was directly correlated with severity of
vertical and horizontal diplopia. The authors concluded that
the anterior orbital connective tissue is capable of keeping
the rectus muscle path aligned after orbital decompression
surgery except for the inferior rectus and medial rectus
muscles. Their results may be explained by the type of
decompression surgery used in their study, which included
translid (2 walls: medial and inferior) or coronal (bilateral 3
walls: medial, inferior, and deep lateral) approaches. Inter-
estingly, patients in our study displayed more exotropia
after surgery. This may reflect a small lateral displacement
or lateral rotation of the globe after tissue has been removed
from the deep lateral wall, more specifically the marrow
space between the superior and inferior orbital fissures and
Table 1. Preoperative and Postoperative Data from 11 Thyroid-
Related Orbitopathy Patients Undergoing Deep Lateral Wall
Orbital Decompression at the Jules Stein Eye Institute,
January, 2004, through December, 2004
Primary gaze diplopia
IOP ? intraocular pressure; NS ? not significant.
Volume 113, Number 6, June 2006
intraconal fat between the lateral and inferior recti muscles.
This is only a small angle deviation that may not imply any
clinically evident strabismus.15Automated Hess screen test-
ing is inexpensive and can be implemented easily in routine
ocular examinations of TRO patients. It seems to provide
accurate measurements of ocular deviations and should be
considered as a routine test in TRO patients, especially
those with ocular motility disturbances.
Orbital decompression surgery involving removal of the
medial and inferior orbital walls may be associated with a
relatively high rate of postoperative new onset strabis-
mus.4,20,21This can be lowered by preservation of the
Figure 2. Scatter plots of preoperative and postoperative (A) horizontal and (B) vertical strabismus in 11 thyroid-related orbitopathy patients who
underwent deep lateral wall orbital decompression. Positive correlation was found between preoperative and postoperative vertical strabismus. Amplitudes
of deviations were obtained using automated Hess screen and are shown in prism diopters (PD). Negative values imply exotropia or hypotropia, whereas
positive values imply esotropia or hypertropia.
Ben Simon et al ? Strabismus after Deep Lateral Wall Orbital Decompression
maxilloethmoidal strut or by performing a balanced decom-
pression of the medial and deep lateral orbital walls.8,16,17
Patients with mild to moderate proptosis may benefit from
deep lateral wall orbital decompressions with or without
intraconal fat debulking. This surgery was shown to be
associated with a relatively low rate of new-onset primary
gaze diplopia.15Orbital decompression may be the first step
in the rehabilitation of TRO patients with eye muscle sur-
gery, if needed and performed at a later stage. Interestingly,
one study described eye muscle surgery performed at the
same session of orbital decompression to reduce the rate of
postoperative diplopia.22Eyelid retraction surgery is usually
reserved as the last step (with or without concurrent bleph-
aroplasty) because large recessions of vertical muscles may
alter eyelid positions. We recently showed that eyelid re-
traction surgery may be performed simultaneously with
deep lateral wall orbital decompression without affecting
functional and cosmetic outcomes.10
We do not claim that AHS testing is a better method for
evaluating postoperative strabismus than testing the field of
binocular vision. It is a relatively simple test, is inexpensive,
and can be performed at the clinic by the treating ophthal-
mologist or by an ophthalmic technician. It may offer a
more accurate way of following up ocular deviations in
patients rather than using an arbitrary scale, as is commonly
used in many ophthalmic plastic clinics. In addition, it is
less time consuming than measuring ocular deviations using
prisms. One of the main purposes of the manuscript was to
examine the effect of deep lateral wall orbital decompres-
sion on ocular deviations; AHS was used for this purpose.
We do not claim by any means that it is superior to standard
prism testing. It adds to our armamentarium of evaluating
TRO patients, and we believe that should be implemented in
the routine ocular examination of these patients.
In summary, we suggest the use of AHS testing to
monitor ocular motility disturbances and ocular deviations
after orbital decompression in TRO patients. Using this test,
we showed that deep lateral wall orbital decompression with
intraconal fat debulking, although proving effective in prop-
tosis reduction, is not associated with horizontal and vertical
ocular deviations in all positions of gaze. Primary gaze
small-angle exotropia may ensue after deep lateral wall
decompression, but this may not be clinically apparent.
1. Lyons CJ, Rootman J. Orbital decompression for disfiguring
exophthalmos in thyroid orbitopathy. Ophthalmology 1994;
2. Kazim M, Trokel SL, Acaroglu G, Elliott A. Reversal of
dysthyroid optic neuropathy following orbital fat decompres-
sion. Br J Ophthalmol 2000;84:600–5.
3. Trobe JD, Glaser JS, Laflamme P. Dysthyroid optic neurop-
athy. Clinical profile and rationale for management. Arch
4. Nunery WR, Nunery CW, Martin RT, et al. The risk of
diplopia following orbital floor and medial wall decompres-
sion in subtypes of ophthalmic Graves’ disease. Ophthal Plast
Reconstr Surg 1997;13:153–60.
5. Goldberg RA, Kim AJ, Kerivan KM. The lacrimal keyhole,
orbital door jamb, and basin of the inferior orbital fissure:
three areas of deep bone in the lateral orbit. Arch Ophthalmol
6. Paridaens DA, Verhoeff K, Bouwens D, van Den Bosch
WA. Transconjunctival orbital decompression in Graves’
ophthalmopathy: lateral wall approach ab interno. Br J Oph-
7. Cruz AA, Leme VR. Orbital decompression: a comparison
between trans-fornix/transcaruncular inferomedial and coro-
nal inferomedial plus lateral approaches. Ophthal Plast Recon-
str Surg 2003;19:440–5.
8. White WA, White WL, Shapiro PE. Combined endoscopic
medial and inferior orbital decompression with transcutaneous
lateral orbital decompression in Graves’ orbitopathy. Ophthal-
9. Shorr N, Neuhaus RW, Baylis HI. Ocular motility problems
after orbital decompression for dysthyroid ophthalmopathy.
10. Ben Simon GJ, Mansury AM, Schwarcz RM, et al. Simulta-
neous orbital decompression and correction of upper eyelid
retraction versus staged procedures in thyroid-related or-
bitopathy. Ophthalmology 2005;112:923–32.
11. Garrity JA, Saggau DD, Gorman CA, et al. Torsional diplopia
after transantral orbital decompression and extraocular muscle
Table 2. Preoperative and Postoperative Automated Hess
Screen for 11 Thyroid-Related Orbitopathy Patients Who
Underwent 19 Orbital Decompression Surgeries
5 (primary gaze)
H ? horizontal; NS ? not significant; V ? vertical; ? ? prism diopters.
Data represent average amplitudes of deviations for all patients.
*Nine measurements were obtained by automated Hess test, in each point
horizontal (H) and vertical (V) deviations, are shown. Positive values refer
to esodeviation, whereas negative values refer to exodeviation; for vertical
strabismus, positive values refer to hypertropia, whereas negative values
refer to hypotropia depending on the tested eye. Pair 5 represents primary
gaze strabismus, pair 2 represents upgaze, and pair 8 represents downgaze.
Volume 113, Number 6, June 2006