High-Resolution Magnetic Resonance

Imaging Demonstrates Varied Anatomic

Abnormalities in Brown Syndrome
Rahul Bhola, MD,a Arthur L. Rosenbaum, MD,a Maria C. Ortube, MD,a and
Joseph L. Demer, MD, PhDa,b

Inroduction: Although Brown syndrome classically is considered to be limited to the SO tendon sheath and
trochlea, it does not always respond to SO surgery. We investigated mechanisms of Brown syndrome by magnetic
resonance imaging (MRI). Methods: Three patients with congenital and 8 with acquired Brown syndrome were
compared with matched normal subjects under a prospective protocol of high-resolution, multipositional orbital
MRI using surface coils. Muscle size and contractility were determined using digital image analysis. Results: Five
of 8 patients with acquired Brown syndrome had a history of trauma or surgery and demonstrated extensive
scarring, avulsion, or fracture of the trochlea. One of the 8 had a cyst in the SO tendon. One congenital and one
acquired case demonstrated inferior displacement of the lateral rectus (LR) pulley in adduction, with a normal SO
tendon–trochlear complex. Such cases of Brown syndrome responded to surgical stabilization of the LR pulley.
Two congenital cases had clinical findings of ipsilateral SO palsy confirmed on MRI by atrophy or absence of the
SO belly. In congenital absence of the SO belly, the anterior tendon was present but terminated directly on the
trochlea. Conclusion: High-resolution MRI demonstrates a variety of abnormalities in patients presenting with
Brown syndrome, including atrophy or absence of the SO belly. Management in Brown syndrome should be
tailored to the pathophysiology of the individual patient. (J AAPOS 2005;9:438-448)

rown syndrome was first characterized in 1950 by mechanisms for Brown syndrome besides pathology of the

B Harold Whaley Brown as a restrictive limitation to

elevation in adduction.1 On the basis of surgical
findings, Brown implicated a short SO tendon sheath as
the cause of this syndrome. Most subsequent reports have
alternatively proposed an abnormality in the trochlear–SO
tendon complex as the cause of restriction to elevation in
adduction.
Although SO surgery, including tenotomy, recession,
or lengthening by a variety of techniques, has been used
principally as surgical treatment for moderate-to-severe
Brown syndrome, in our experience, not all patients re-
spond to this surgery. Such surgical failures suggest other
a
From the Jules Stein Eye Institute, University of California, Los Angeles and the
b
Department of Neurology, University of California, Los Angeles, California
Presented at the 30th annual meeting of the American Association for Pediatric Ophthal-
mology and Strabismus, Washington, DC, March 27-31, 2004.
Supported by USPHS NIH EY08313 & Research to Prevent Blindness. J.L.D. received an
unrestricted award from Research to Prevent Blindness and is the Leonard Apt Professor of
Ophthalmology at UCLA. A.L.R. is a recipient of a Research to Prevent Blindness
Physician-Scientist Merit Award.
Submitted March 9, 2004.
Revision accepted June 17, 2005.
Reprint requests: Joseph L. Demer MD, PhD, Jules Stein Eye Institute, 100 Stein Plaza,
David Geffen School of Medicine at UCLA, Los Angeles, CA 90095-7002 (e-mail:
jld@ucla.edu).
Copyright © 2005 by the American Association for Pediatric Ophthalmology and
Strabismus.
1091-8531/2005/$35.00 ⫹ 0
doi:10.1016/j.jaapos.2005.07.001
SO tendon and trochlea.
Knowledge of orbital and extraocular muscle (EOM)
anatomy has progressed since the time of Harold Brown.
Motivated by efforts to develop computer simulations of
EOM behavior to guide strabismus surgery, researchers
recently have conducted anatomical re-examinations of
orbital anatomy using immunohistochemistry of serially
sectioned orbits,2-4 computer reconstructions in virtual
reality,5 and multipositional magnetic resonance imaging
(MRI) with spatial resolution enhanced by use of surface
coils6,7 and tissue resolution enhanced with paramagnetic
contrast8 to study the functional anatomy of the
EOMs.7,9-12 From these investigations has emerged a de-
tailed picture of the vital role of EOM paths in determi-
nation of EOM function.13In brief, the action exerted by
any EOM on the globe is determined not only by the
magnitude of tension it exerts but also on the pulling
direction of that tension. For any EOM, pulling direction
is determined by its scleral insertion and the location of its
functional origin. The functional origin of each EOM is at
a connective tissue structure called a pulley. Harold Brown
recognized the importance of the SO pulley—the classic
trochlea—to SO function. We now know that all of the
other EOMs have less rigid, but still essential, pulleys.3-5,8
The pathology of these pulleys causes incomitant strabis-
mus that can clinically simulate oblique EOM dysfunction.
For example, instability of the LR pulley has been dem-

438 October 2005 Journal of AAPOS

Journal of AAPOS
Volume 9 Number 5 October 2005
onstrated by multipositional MRI to simulate Brown syn-
drome.14 It was observed that the instability of the pulleys
was accentuated in particular gazes and was responsible for
the incomitance of the deviation.
Although clinical tests like alignment measurements,
forced duction, forced generation, and saccadic velocity
testing may be helpful in evaluating Brown syndrome, they
may not define the pathophysiology accurately. Multipo-
sitional MRI can provide valuable additional clinical infor-
mation to demonstrate a variety of different anatomical
mechanisms in patients presenting clinically with Brown
syndrome. We performed this study to explore the spec-
trum of clinical abnormalities in Brown syndrome.
MATERIALS AND METHODS
Participating patients were selected from an ongoing, pro-
spective study of orbital imaging in strabismic patients.
Eleven consecutive cases of unilateral Brown syndrome
were identified based on clinical findings, including 3 pa-
tients with congenital and 8 patients with acquired disor-
ders. All the patients complained of constant or intermit-
tent vertical diplopia and had limitation to elevation in
adduction in one eye. A complete ophthalmic evaluation,
including a detailed ocular motility examination, was per-
formed in all the cases. Binocular alignment was measured
using prism and alternate cover test both for distance and
near in the cardinal gaze directions and with head tilt to
each side, and with the Hess screen test in 21 fixation
positions over a 30 degree field for each eye. Ductions and
versions were quantified using a 9-point scale, with 0
suggesting a normal movement, – 4 signifying an inability
to move the eye past midline, and ⫹4 signifying maximum
observable overrotation.
After obtaining written informed consent according to
a protocol confirming to the Declaration of Helsinki and
approved by the Institutional Review Board, each subject
underwent multipositional high-resolution T1-weighted
MRI with a 1.5-T General Electric Signa (Milwaukee,
WI) scanner using methods previously described in de-
tail.6,7 Depending on the clinical indication, T2 imaging
was used occasionally in addition. Each subject’s head was
carefully stabilized in a supine position with the nose
aligned to the longitudinal and the pupils to the transverse
light projection references of the scanner.
Imaging of the orbits was performed in multiple gaze
positions using a phased array of 4 surface coils deployed
in a mask-like enclosure held strapped to the face. An
adjustable array of monocular, afocal, illuminated fixation
targets at 9 diagnostic positions of gaze was secured in
front of each orbit with the center target in subjective
central position for each eye. Head movement of the
subjects was minimized by secure stabilization to the sur-
face coil facemask and judicious use of padded restraints.
Multiple contiguous quasicoronal digital images of 2-mm
slice thickness were then obtained using a 256 ⫻ 256
matrix over an 8-cm2 field, giving pixel resolutions of 313
Bhola et al 439
␮m. Imaging was then repeated in multiple gaze positions.
The trochlea and reflected SO tendon also were imaged
using axial image planes of the same resolution. In appro-
priate cases, the inferior oblique muscle was imaged in
quasisagittal image planes parallel to the long axis of the
orbit. Digital MRIs were transferred to Macintosh com-
puters (Apple Computer, Cupertino, CA) and converted to
8-bit tagged image file format and were quantitatively
analyzed using the program NIH Image (W. Rasband,
National Institute of Health; available by ftp from zippy.
nimh.nih.gov or on floppy disk from NTIS, Springfield,
VA, part number PB95-500195GEI).
We studied the size of the EOMs in different gaze
positions of both the eyes in all subjects. Change in EOM
cross section from relaxed to contracting gaze positions
was used as an index of contractility. Anatomical charac-
teristics of the SO muscle and tendon and its relationship
with the trochlea was evaluated in multiple gaze positions
where clinically appropriate.
Subject data were compared with the existing data of
the normal rectus pulley positions relative to the globe
center as determined by Clark et al15 in 22 orbits of 11
normal adults using the same NIH Image program and a
similar prospective protocol of high-resolution, multipo-
sitional orbital MRI using surface coils.
RESULTS
Twenty-two orbits of 11 subjects with Brown syndrome
were analyzed. Clinical information of the subjects is sum-
marized in Table 1. According to the anatomic abnormal-
ities as elicited by MRI, 4 distinct mechanisms of Brown
syndrome were identified: (1) trochlear damage; (2) SO
tendon abnormalities; (3) abnormalities of rectus EOM
pulleys; and (4) congenital abnormalities of SO muscle.
Cases were segregated by mechanism.
Trochlear Damage
Case 1. This 17-year-old boy presented with a history
of vertical binocular diplopia in upgaze after a bilateral
frontal sinus surgery for bacterial sinusitis 3 months before
his presentation. Uncorrected visual acuity was 20/15 in
each eye. External examination was notable for healed
Lynch incisions bilaterally (Figure 1) with a palpable scar
web over the region of the left trochlea but without troch-
lear tenderness. There was marked limitation to supraduc-
tion of the right eye in adduction (Figure 1). The patient
was orthotropic in all diagnostic fields, including head tilt,
with the exception of upgaze, in which there was a large
right hypotropia. High-resolution multipositional MRI
revealed an extensive scarring in the region of both troch-
leas. Imaging with contrast demonstrated continuity of a
cicatrix from the right frontoethmoidal sinus region
through the trochlea entering the SO tendon (Figure 2).
Case 2. This 36-year-old woman presented with a
history of vertical binocular diplopia in up gaze, subse-
quent to a trauma on the left orbit from a softball. She
 Journal of AAPOS
440 Bhola et al Volume 9 Number 5 October 2005

TABLE 1 Clinical profile of patients

Subject Age(years)/sex Onset Central gaze deviation MRI findings
1. 17/M Acquired: frontal sinus surgery Orthotropic Scarring of right trochlea from the frontal sinus
2. 36/F Acquired: trauma 4 PD Exophoria Adhesion bands left trochlea, SOT thickening
3. 7/M Acquired: dog bite 18 PD LHT 7° Excy. OS Disorganization of left tendon/trochlear
complex with scarring
4. 21/F Acquired: automotive accident 30 PD RHT 4° Excy. OD Right trochlear avulsion; left trochlear
disruption
5. 64/F Acquired: blepheroplasty Orthotropic Possible suture in reflected SOT
6. 36/F Acquired: idiopathic 3 PD RH(T) Cyst in SOT posterior to trochlea
7. 11/F Congenital 20 PD LHoT 10 PD XT Left LR pulley instability
8. 69/M Acquired: idiopathic 8 PD RHT 7° Excy. OD Left LR pulley instability
9. 53/F Acquired: orbital decompression 20 PD LHoT 10° Incyc. OS Left LR pulley instability and adhesion to orbit
10. 14/M Congenital 35 PD LHT Hypoplastic left SOM, fibrous band from sclera
to trochlea
11. 8/M Congenital 14 PD RHT 8 PD ET Attenuation of right SOM, fibrous band from
sclera to trochlea
PD, prism diopters; RHT, right hypertropia; LHT, left hypertropia; RH(T), intermittent right hypertropia; LHoT, left hypotropia; XT, exotropia; F, female. Excy, excycotorsion. LR,
lateral rectus muscle; M, male; SOM, superior oblique muscle; SOT, superior oblique tendon.

FIG 1. Versions of subject 1 in 9 diagnostic positions of gaze, demonstrating underelevation on adduction of the right eye. Note scarring from
the Lynch incisions over both trochleas.

complained of left trochlear pain, trochlear tenderness,
and an occasional clicking sound from the left trochlear
region when looking upward. Uncorrected visual acuity
was 20/20 in each eye. There was limitation to elevation of
the left eye in both abduction and adduction that was
worse in adduction. At distance, there was 4⌬ of exophoria
in primary gaze and in right gaze. She was orthotropic in
left gaze and down gaze, but in up gaze exhibited a left
hypotropia of 8⌬. At near gaze she had 8⌬ exophoria.
High-resolution MRI demonstrated an abnormality in the
left reflected SO tendon with adhesion bands in the troch-
lear region. The patient was administered steroid injec-
tions in the left trochlea without resolution of her symp-
toms. Subsequently, a surgical exploration of the left
trochlea and SO was performed, which revealed an en-
largement of the left SO tendon, which was found to be
folded on itself with fibrous adhesions extending back to
the trochlea. The adhesions were lysed surgically, and the
left trochlea was infiltrated with long and short acting
steroids. Symptoms recurred after initial improvement.
SO tenotomy with a compensatory surgery on the ipsilat-
eral IO is planned.
Case 3. This 7-year-old boy developed a vertical bin-
ocular diplopia and spontaneous right head tilt after being
bitten by a dog above the left eye 5 months before his
presentation. The clinical and examination findings of the
patient have been previously reported.16 Multipositional
MRI scans of the left orbit showed marked scarring in the
area of the left trochlea along with SO tendon sheath
thickening. There was marked disorganization of the ten-
don/trochlear complex.
Case 4. This 21-year-old woman developed a vertical
binocular diplopia after severe head trauma in an automo-
tive collision. Visual acuity was 20/20 with a myopic cor-
Journal of AAPOS
Volume 9 Number 5 October 2005
FIG 2. Coronal T1-weighted, contrast-enhanced MRI image of the
right orbit of subject 1 showing a cicatrix from the right frontoeth-
moidal sinus region to the trochlea Note enhancement of the sinus
mucosa.
rection. There was moderate overelevation in adduction of
the right eye, with marked underelevation and downshoot
in adduction of the left eye (Figure 3). There was mild
limitation to infraduction of the right eye, particularly in
adduction. At distance, there was 30⌬ right hypertropia in
primary position, decreasing in right gaze and increasing
in left gaze. The deviation increased in upgaze and dimin-
ished slightly in downgaze. Forced head tilt test revealed
50⌬ of right hypertropia with right head tilt diminishing
to 15⌬ in left head tilt. At near there was 25 ⌬ right
hypertropia with 6⌬ exotropia. Double Maddox rod test-
ing demonstrated 3-4o right excyclotorsion. High-
resolution MRI revealed right trochlear avulsion along
with disengagement of the SO tendon and a trochlear
disruption in the left eye (Figure 4). The patient under-
went left SO tenotomy, a 10-mm right IO recession, and
a 3-mm left inferior rectus recession on adjustable suture.
At 3-week postoperative follow-up, she presented with a
small consecutive left hypertropia that was treated success-
fully with a small spectacle prism.
Superior Oblique Tendon Abnormalities
Case 5. This 64-year-old woman complained of verti-
cal binocular diplopia immediately after undergoing bilat-
eral upper eyelid blepharoplasty 9 months before. Uncor-
rected visual acuity was 20/20 in each eye. There was
marked limitation of elevation of the left eye in adduction
(Figure 5). She was orthotropic in primary position but
had 25⌬ left hypotropia in upward gaze to the right.
Motion of a nodule could be palpated in the region of the
left trochlea during attempted up and down gaze. Forced
duction testing showed marked restriction to elevation of
Bhola et al 441
the left eye in adduction. Axial (Figure 6A) and coronal
(Figure 6B) MRI revealed a nodular structure in the left
reflected SO tendon just lateral to the trochlea, and mul-
tiple nodular structures in the anterior superomedial right
orbit. These structures demonstrated a low intensity both
on T1- and T2-weighted imaging, with a high-intensity
rim suggestive of suture material or cautery effect from the
blepharoplasty surgery interfering with the reflected SO
tendon.
Case 6. This 36 year-old woman noticed progressively
increasing intermittent pain over the right trochlea with
intermittent vertical diplopia beginning18 months before
presentation. She also noticed a palpable click in the re-
gion of the right trochlea on shift from up- to downgaze.
Details of this patient have been previously described in a
case report.16 MRI showed a spherical structure in the SO
tendon that was presumed to be a cyst. This cystic struc-
ture was seen to change anteroposterior position with
vertical gaze shifts. In supraduction, the SO tendon was
passively stretched and the cyst was pulled anteriorly. In
primary position the cyst was posterior and began to abut
the trochlea but apparently did not enter it. In infraduc-
tion the SO muscle contracted and the cyst retracted,
distant from the trochlea. The patient’s symptoms can be
explained by interference of the cyst with the trochlea; the
cyst must be dragged through the trochlea on excursion
from down- to upgaze. This causes a temporary restric-
tion, which gives way with a palpable click as the bulk of
the cyst moves anterior to the trochlea in upgaze. The SO
tendon was explored surgically. The abnormal cystic struc-
ture surrounding the tendon was excised, and the sur-
rounding area was infiltrated with steroids. The patient
was asymptomatic on the first postoperative day.
Abnormalities of Rectus Extraocular Muscle Pulleys
Case 7. This 11-year-old girl presented with a history
of anomalous head posture, inability to elevate the left eye
in adduction, and left hypotropia since the age of 3
months. Uncorrected acuity was 20/20 in each eye. The
patient had a 20° right face turn with a 5° chin elevation.
There was marked underelevation of the left eye in adduc-
tion and moderate elevation limitation in abduction.
There was a moderate overdepression of the left eye in
adduction. In forced primary position at distance there was
20⌬ left hypotropia and 10⌬ exotropia. At near there was
20⌬ left hypotropia and 16⌬ exotropia. Intraoperative
forced duction testing revealed marked restriction to su-
praduction of the left eye in adduction. Surgical explora-
tion showed the anatomical position, fibers and tension of
the SO tendon to be normal. Left SO tenotomy with a
7-mm silicon spacer failed to immediately relieve restric-
tion to supraduction. The spacer was replaced with a
suture and the posterior end of the SO tendon was allowed
to retract approximately 15 mm. Postoperative examina-
tion on the first day and 6 months later showed minimal
improvement in the anomalous head posture and ocular

442 Bhola et al
FIG 3. Versions of subject 4 in 9 diagnostic positions of gaze, demonstrating severe underelevation of the left eye on adduction with moderate
underelevation on abduction.
FIG 4. case 4. Top row, Coronal T1 MRI showing dislocation of the
right SO tendon after avulsion of the right trochlea. Damage to the
left trochlea was evident, along with medial wall blow out fractures
larger on the right than left. IO, IO muscle; MR, medial rectus
muscle. Bottom row, Axial T1 MRI images showing posterior dis-
placement of the path of the right SO tendon, in relation to the
normal location of the reflected left SO tendon. SR, SR muscle.
rotations, with 16⌬ left hypotropia and 8⌬ exotropia in
forced primary position. High-resolution MRI at this
point revealed inferior displacement of the left LR pulley,
a defect that was exaggerated in adduction. There was no
such abnormality demonstrated in the right orbit. In view
Journal of AAPOS
Volume 9 Number 5 October 2005
of the anatomical abnormality on MRI, a second surgery
was performed, consisting of a left LR recession and su-
perior insertional transposition, with posterior reteroequa-
torial myopexy to prevent inferior slip of the LR pulley.
Intraoperative forced duction testing was immediately
completely relieved of restriction. Three months postop-
eratively the patient had a minimal left hypotropia in
primary gaze and no anomalous head posture.
Case 8. This 69 year-old men presented with a 1-year
history of sudden onset of vertical diplopia worse in up-
gaze to the right. Corrected visual acuity was 20/20 in both
eyes. He had a 5° chin elevation. Cover test revealed a
Y-pattern exotropia with 8⌬ right hypertropia in primary
position, which increased in upgaze to 16⌬ but was gen-
erally comitant in other gaze positions. The hypertropia
slightly increased on forced left head tilt. Double Maddox
rod test revealed right excyclotorsion of 7°. There was
limitation of supraduction of the left globe, worse in ad-
duction. Forced duction testing revealed restriction to
supraduction of the left globe, greatest in adduction.
High-resolution MRI revealed no anatomic abnormality
of the left SO muscle, tendon or trochlea. Instability of the
left LR pulley was seen on coronal imaging, with down-
ward shift of the left LR on adduction.
Case 9. This 53-year-old woman had a 5-year history
of thyroid ophthalmolopathy. Immediately after undergo-
ing left orbital decompression via a superotemporal
approach 4 months before presentation, she noticed bin-
ocular vertical diplopia with relative tilting of images,
greatest in right gaze and minimized with a left head tilt
and right face turn. She also felt tightness inferotemporally
in the left eye socket while looking to the right. Corrected
visual acuity was 20/16 in the right and 20/50 in the left
eye. There was severe limitation to elevation of the left eye
in adduction, but with extreme elevating effort there was a
Journal of AAPOS
Volume 9 Number 5 October 2005
FIG 5. Versions of case 5 in nine diagnostic gaze positions showing underelevation of the left eye in adduction.
snap-like, sudden upward rotation of the left eye nearly
normalizing supraduction in adduction (Figure 7). There
was a marked overdepression of the left eye in adduction.
There was no limitation to elevation of the left eye in
abduction. Forced duction testing revealed a marked re-
sistance to passive supraduction in adduction, which was
suddenly relieved with a prolonged effort to look up by the
patient. At distance, there was 20⌬ left hypotropia in
primary gaze that diminished to 4⌬ in left gaze, and
increased to 40⌬ in right gaze. The left hypotropia was
30⌬ in upward gaze and was 10⌬ in downward gaze. At
near, there was 25⌬ left hypotropia with 6⌬ exotropia.
There was 25⌬ left hypotropia in right head tilt diminish-
ing to 10⌬ in left head tilt. Double Maddox rod testing
demonstrated 10° left incyclotropia. High-resolution MRI
revealed postsurgical cicatrisation around the left LR mus-
cle, with apparent adhesion of the inferior pole of that
EOM to the orbital wall (Figure 8). There was no abnor-
mality of trochlea and SO on either side. Instability of the
left LR pulley was suspected. This instability was con-
firmed by intraoperative demonstration of obvious down-
ward slip of the LR muscle path when rotating the eye
from abduction to adduction using a muscle hook engaged
at the LR insertion. The LR path was stabilized by joining
its superior margin to the underlying sclera using a non-
absorbable suture; the left IR muscle was recessed 3.5 mm.
Six months after surgery, the patient had a marked im-
provement of ocular versions, and was orthotropic in pri-
mary position. The ocular torticollis was eliminated.
Congenital Abnormalities of the SO Muscle
Case 10. This 14-year-old boy presented with inter-
mittent vertical diplopia that had existed since early child-
hood. Visual acuity in each eye was 20/20. He spontane-
ously adopted a 10° head tilt to the right. There was down
shoot of the left eye in adduction with marked inability to
elevate the left globe in adduction and reduced ability to
depress the left globe in adduction. There was mild limi-
tation to full upward rotation of the left globe in straight
up position. In forced primary position, cover testing re-
vealed 35⌬ left hypertropia that increased to 40⌬ in down
gaze. The left hypertropia was reduced to 14⌬ in levover-
Bhola et al 443
sion, but was 25⌬ in dextroversion. The left hypertropia
was reduced in upgaze to 6⌬. Forced head tilt testing
showed a left hypertropia of 25⌬ on left head tilt reducing
to 6⌬ on right head tilt. The increase of left hypertropia in
down gaze and left head tilt clinically suggested coexistent
SO paresis. High-resolution coronal MRI suggested com-
plete absence of the left SO muscle belly (Figure 9A). The
reflected tendon of the left SO was present, but apparently
terminated on the trochlea. Axial MRI showed an attenu-
ated remnant of the left SO belly, in contrast with a
normal sized right SO belly (Figure 9B). The right SO was
of normal size and demonstrated robust contraction from
supraduction to infraduction. Intraoperative forced duc-
tion testing revealed marked restriction to elevation of the
left eye in adduction. Surgical exploration revealed a dense
fibrous band that attached to the sclera at the nasal border
of the left superior rectus (SR) muscle, incorporating some
of its fibers. The band was traced posteriorly for a distance
of 8 mm to the same area as the reflected SO tendon, but
normal tendon fibers could not be visualized. After divid-
ing this fibrous band the left IO muscle was recessed, and
the left IR was resected. At the time of last follow-up, 4
years postoperatively, the patient was orthotropic in pri-
mary gaze and free of ocular torticollis. The limitation to
elevation in adduction was also markedly improved.
Case 11. This 8-year-old boy presented with a lifelong
left head tilt. Uncorrected visual acuity was 20/30 in the
right and 20/25 in the left eye. Cycloplegic refraction
revealed hyperopia of ⫹ 3.50 in the right eye and ⫹ 4.25
in the left eye. On examination, there was a left head tilt of
5°. There was limitation to elevation in adduction of the
right eye (Figure 10). In forced primary position, there was
14⌬ right hypertropia and 8⌬ esotropia. The right hyper-
tropia increased to 20⌬ in downgaze, was 8⌬ in dextro-
version, and 14⌬ in levoversion. There was no hypertropia
in the upgaze. Esotropia was eliminated by hyperopic
correction. High-resolution MRI revealed a attenuation of
the right SO muscle belly (Figure 11), with size less the
95% confidence limit of normal17 with the anterior tendon
terminating directly on the trochlea similar to case 10. The
left SO muscle was normal in size and had normal con-

444 Bhola et al
FIG 6. Case 5. A, Axial T1 MRI showing a nodule in the reflected
tendon of the left SO muscle, with irregularity around the reflected
tendon. B, Coronal T2 MRI image of the left orbit the nodule in the
reflected tendon of the SO muscle.
tractile thickening from supraduction to infraduction. In-
traoperative forced duction testing revealed a marked re-
striction to elevation of the right globe in adduction.
Exploration revealed a fibrous band adhesion to the sclera
at the nasal border of the SR. The band appeared to be in
the same area as the SO tendon, but the tendon fibers
could not be visualized. This fibrous band was divided,
which immediately relieved the restriction to forced duc-
tion. The left IO muscle was recessed. At the time of last
follow-up 5 year, 2 month postoperatively the patient was
orthophoric in primary position, and free of anomalous
head posture.
DISCUSSION
Multipositional, high-resolution MRI has enhanced the
understanding of incomitant strabismus by revealing mul-
tiple anatomical and functional abnormalities of the ex-
Journal of AAPOS
Volume 9 Number 5 October 2005
traocular muscles and their pulleys.6,7,9,10-12,16 Surgical
success in incomitant strabismus often depends on patho-
logic mechanism, so that treatment can be directed to its
correction.
Brown syndrome is a narrowly defined, restrictive, in-
comitant strabismus whose clinical features are neverthe-
less nonspecific to etiology. In 1950, Brown reported 8
cases having a restrictive limitation to elevation in adduc-
tion, a condition that he called the “SO tendon sheath
syndrome.”1 This restriction to elevation was believed to
be caused by secondary shortening of the anterior sheath
of the SO tendon from congenital palsy of the “ipsilateral
IO.” Pathology of the IO muscle is no longer considered
important in Brown syndrome, but numerous other etiol-
ogies have been proposed. The inconsistent response to
SO relaxation and tenotomy procedures in some of these
cases inspired us to study the anatomical abnormalities
associated with this syndrome using a multipositonal,
high-resolution MRI.
We here report 11 patients with Brown syndrome and
ipsilateral restrictive limitation to elevation in adduction.
Despite the typical clinical features, we found four distinct
abnormalities of functional anatomy. Six of the 8 acquired
cases developed Brown syndrome secondary to surgical or
accidental trauma. Although all these cases had a similar
clinical presentation with acute onset, multipositional
MRI revealed varied abnormalities. Four of these trau-
matic cases demonstrated trochlear damage restricting free
passage of the SO tendon through the trochlea. Case 1
developed Brown syndrome after a frontal sinus surgery.
Blanchard and Young in 198418 and Rosenbaum and
Astle19 in 1985 reported Brown syndrome after frontal
sinus surgery. Improper surgical technique, such as place-
ment of the surgical incision too low or excessive perios-
teal stripping near the trochlea, can contribute to the SO
damage or restriction. In addition, a cicatrisation of the
SO to the globe, bone or into the sinus scar tissue defect
may restrict eye movement.19 High-resolution imaging
with contrast explicitly demonstrated continuity of a ci
catrix from the right frontal sinus region through the
trochlea entering the SO tendon in case 1, thus revealing
the cause of restrictive incomitance.
Direct nonsurgical trauma to the superomedial orbit
can disturb the function of the SO muscle, causing the
clinical appearance of SO palsy, Brown syndrome, or
both.20 Restricted elevation in adduction secondary to
trauma can be a manifestation of either direct injury to
trochlear-SO tendon complex the result of scarring from
the inflammation surrounding the SO tendon.21 High-
resolution MRI can be useful in such cases to define the
exact anatomical abnormality. Three of our cases devel-
oped a Brown syndrome secondary to accidental trauma
and showed varied abnormalities on MRI. Case 2 showed
scarring in the area of the trochlea along with tendon
sheath thickening. Case 3 demonstrated abnormalities in
the reflected SO tendon anterior to the trochlea, with
Journal of AAPOS
Volume 9 Number 5 October 2005
FIG 7. Versions of case 9 in 9 diagnostic positions of gaze, demonstrating underelevation and downshoot of the left eye in adduction.
FIG 8. case 9. Top- Axial images of the left orbit show normal SO
tendon trochlear complex. Bottom-Coronal MRI of the left orbit in
central gaze showing cicatrization near the inferior pole of the LR
pulley after orbital decompression surgery. The orbitotomy site is
visible superolaterally in the left orbit. LR, lateral rectus muscle; SO,
SO muscle.
tendon thickening and adhesions. Case 4 revealed a troch-
lear disruption in one eye and a trochlear avulsion in the
contralateral eye, presenting clinically as Brown syndrome
with contralateral SO palsy.
Incarceration of the SO tendon, with fat, orbicularis,
levator aponeurosis, and septum has been reported after
Bhola et al 445
blepharoplasty.22 Case 5 developed vertical binocular dip-
lopia immediately after undergoing bilateral upper eyelid
blepharoplasty. High-resolution MRI scan in this patient
suggested that deeply placed blepharoplasty sutures, or
possibly cautery, had included the reflected SO tendon.
Impediment of free movement of the SO tendon in the
trochlear tendon complex has been proposed as a cause of
Brown syndrome. Inflammation by such etiologies as ju-
venile rheumatoid arthritis, lupus, and Sjogren’s syn-
drome,23-26 trauma or other restrictive phenomena have
been proposed to compromise function. In 1971 Roper-
Hall27 in their series described “SO ‘click’ syndrome”
presumed to be an inflammatory form of Brown syn-
drome. The clinical picture can alternate between Brown
syndrome and SO palsy, depending on the direction in
which the tendon movement through the trochlea is im-
peded. One of our acquired cases presented with an inter-
mittent picture of Brown syndrome with a palpable click
over the trochlear region on change from supraduction to
infraduction. Although the patient had no clinical evidence
of any inflammatory or autoimmune disorder, a multipo-
sitional MRI showed the location and movement of a
spherical structure in the SO tendon that correlated with
the patient’s symptoms. This spherical structure was sur-
gically confirmed to be a cyst that interfered with normal
transit of the SO tendon through the trochlea.
An acute onset of Brown syndrome was observed in case
9 after an orbital decompression surgery for thyroid oph-
thalmopathy that completely avoided the trochlear region.
High-resolution MRI revealed evidence of damage to the
left LR pulley. This indication of LR pulley instability was
confirmed intraoperatively by demonstrating the shift in
LR path during passive adduction of the eye. Supraplace-
ment and stabilization of the LR path by superior myopexy
corrected the limitation to elevation of the eye in adduc-
tion.

446 Bhola et al
FIG 9. T1 MRI in case 10. A, Coronal images showing apparent
absence of left SO muscle (SO) but presence of its reflected tendon
inserting on the trochlea. IO, IO muscle; LPS, levator palpebrae
superioris muscle; ON, optic nerve; SR, SR muscle. B, Axial images
showing a normal right SO belly, trochlea, and reflected tendon, but
marked attenuation of the left SO posterior to the trochlea.
One congenital (case 7) and one acquired case (case 8)
demonstrated a vertical instability of the LR pulley in
adduction. In both patients there was an inferior displace-
ment of the LR pulley of the involved eye, a defect that
was exaggerated in adduction. This LR side slip causes
restrictive limitation to elevation in adduction. Oh et al14
reported that instability of the pulleys redirects the passive
elastic forces of the EOMs in certain gaze positions,
thereby producing incomitant strabismus. It is presumed
that LR pulley instability may develop or progress as the
Journal of AAPOS
Volume 9 Number 5 October 2005
result of degeneration or trauma to the connective tissue
suspensions of the LR pulley.
The pulleys of the rectus EOMs normally minimize
sideslip relative to the orbit of posterior EOM paths dur-
ing globe rotation and determine the effective pulling
direction of each EOM. MRI has demonstrated that these
tissue sleeves stabilize EOM bellies relative to the orbit,
permitting only the insertional ends of the EOMs to move
with the globe rotation. Radial displacements of pulleys
from the orbital center do not appear to influence binoc-
ular alignment. However, vertical displacement of the pul-
leys to their planes of action may influence the risk of
strabismus by altering EOM pulling direction.15 Thus,
heterotopy of the pulleys and pulley instability during
ocular rotation may result in a redistribution of the elastic
passive forces of the rectus EOMs in turn producing in-
comitant strabismus.14,28 The exaggerated malposition of
the LR muscle was demonstrated during strabismus sur-
gery in one of our cases and was found to be consistent
with the MRI. Progressive LR heterotopy might occur
because of gradual or abrupt dehiscence of the LR pulley
suspension caused by connective tissue degeneration anal-
ogous to levator tendon dehiscence in aponeurotic blepha-
roptosis.2 Lack of a contribution of the SO to Brown
syndrome in cases of LR pulley instability is indicated by
failure of SO tenotomy to relieve restriction to elevation in
adduction. Repositioning and stabilization of the malposi-
tioned LR was effective instead.
Intraoperative traction testing also may be helpful in
distinguishing mechanisms of Brown syndrome. In classic
Brown syndrome with restricted travel of the SO tendon
through the trochlea, limitation to passive supraduction in
adduction should be increased by globe retropulsion, while
the opposite effect should occur with LR pulley instability.
If preoperative MRI is unavailable, intraoperative minimal
exploration of the ipsilateral LR could confirm if LR
instability is the cause of Brown syndrome. A muscle hook
placed under the LR insertion, and held lightly by its
handle end, can be used to adduct the eye. Inferior shift of
the LR pulley is then obvious from inferior shift of the LR
path and globe extorsion as the eye is passively rotated into
adduction. Although repetition of intraoperative forced
duction testing after SO tenotomy can be used as another
diagnostic marker for a non-SO cause of restriction to
elevation in adduction, this is a significantly invasive ma-
neuver that would better be avoided if the diagnosis can be
reached by less traumatic means. Even if the diagnostically
divided SO tendon were immediately re-anastamosed, the
resulting scarring near the trochlea could easily impede
tendon motion through it, and induce classic Brown syn-
drome. Surgery to stabilize the posterior unstable LR
completely relieved the restriction and significantly cor-
rected the vertical tropia in cases of Brown syndrome due
to unstable LR pulleys.
Few cases of Brown syndrome caused by restrictive
bands with a normal SO tendon trochlear complex have
Journal of AAPOS
Volume 9 Number 5 October 2005
FIG 10. Versions of case 11 in 9 diagnostic positions of gaze, demonstrating underelevation of the right eye in adduction.
FIG 11. Coronal T1 MRI of case 11 demonstrating hypoplasia of the
right SO muscle.
been reported. Romaine29 suggested that in cases of
Brown syndrome where no abnormality is found in the
SO, then areas of Lockwood’s ligament, IO, and IR muscle
should be surgically explored. According to Romaine, re-
stricting bands near the IR or SO muscles could cause
Brown syndrome. No such bands were encountered in the
present series, and the availability of modern MRI now
makes this noninvasive diagnostic option attractive in
comparison to blind surgical exploration of the inferior
orbit. Two of the congenital cases reported here demon-
strated fibrous adhesion bands extending from the troch-
lear region to the nasal aspect of the SR muscle. One of
these cases demonstrated an absence of the SO belly, and
the other had a grossly atrophic SO muscle. Both these
cases had a combined ipsilateral Brown syndrome with SO
palsy, and so might not be regarded as typical Brown
syndrome in this respect. The fibrous band was responsi-
ble for a restriction to elevation in adduction of the af-
fected eye. This fibrous band could be a remnant of the
anterior tendon of a congenitally maldeveloped SO mus-
cle.
The various distinct functional anatomical abnormali-
ties demonstrated in both congenital and acquired cases of
Brown syndrome highlight the importance of obtaining a
specific etiological diagnosis in such patients preopera-
tively. Brown syndrome is only a syndrome, not a specific
mechanism, and as such has a differential list of causes.
Bhola et al 447
High-resolution, multipositional MRI can define in great
detail the pathological anatomical abnormalities causing
Brown syndrome, allowing rational individualization of
surgical management without reliance on extensive explor-
atory surgery or a trial and error series of operations.
High-quality orbital MRI is currently within the technical
capabilities of many North American medical centers, but
for this purpose must be requested in some specific detail
by the strabismologist. Where adequate imaging is un-
available, knowledge of the possible mechanisms described
here can aid in interpretation of intraoperative traction
testing to localize some of the specific mechanisms respon-
sible for Brown syndrome.
References
1. Brown HW.Congenital structural muscle anomalies. In: Allen JH,
editor. Strabismus Ophthalmic Symposium. St Louis: CV Mosby,
1950, pp. 205-36.
2. Oh SY, Poukens V Demer JL. Quantitative analysis of extraocular
muscle layers in monkey and human. Invest Ophthalmol Vis Sci.
2001;42:10-6.
3. Demer JL. The orbital pulley system—a revolution in concepts of
orbital anatomy. Ann NY Acad Sci 2002;956:17-32.
4. Kono R, Poukens V, Demer JL. Quantitative analysis of the structure
of the human extraocular muscle pulley system. Invest Ophthalmol
Vis Sci 2002;43:2923-32.
5. Miller JM, Demer JL, Poukens V, Pavlowski DS, Nguyen HN, Rossi
EA. Extraocular connective tissue architecture. J Vision 2003;2:12-23.
6. Demer JL, Miller JM.Orbital imaging in strabismus surgery. In: Rosen-
baum AL, Santiago P, editors. Clinical strabismus management: prin-
ciples and surgical techniques. New York: Mosby; 1999, 84-98.
7. Demer JL, Clark RA, Kono R, Wright W, Velez F, Rosenbaum AL.
A 12-year, prospective study of extraocular muscle imaging in com-
plex strabismus. J AAPOS. 2002;6:337-47.
8. Demer JL, Oh SY, Clark RA, Poukens V. Evidence for a pulley of the
inferior oblique muscle. Invest Ophthalmol Vis Sci. 2003;44:3856-65.
9. Demer JL, Miller JM, Koo EY, Rosenbaum AL. Quantitative mag-
netic resonance morphometry of the extra- ocular muscles: a new
diagnostic tool in paralytic strabismus. J Ped Opththalmol Strabis-
mus 1994;31:177-88.
10. Kono R, Clark RA, Demer JL. Active pulleys: magnetic resonance
imaging of rectus muscle paths in tertiary gazes. Invest Ophthalmol
Vis Sci 2002;43:2179-88.
11. Demer JL, Kono R, Wright W. Magnetic resonance imaging of
human extraocular muscles in convergence. J Neurophysiol 2003;9:
2072-85.
 Journal of AAPOS
448 Bhola et al Volume 9 Number 5 October 2005

12. Kono R Demer JL. Magnetic resonance imaging of the functional
anatomy of the inferior oblique muscle in superior oblique palsy.
Ophthalmology 2003;110:1219-29.
13. Demer JL. Pivotal role of orbital connective tissues in binocular
alignment and strabismus. The Friedenwald lecture. Invest Ophthal-
mol Vis Sci 2004;45:729-38.
14. Oh SY, Clark RA, Velez F, Rosenbaum AL, Demer JL. Incomitant
strabismus associated with instability of rectus pulleys. Invest Oph-
thalmol Vis Sci 2002;43:2169-78.
15. Clark RA, Miller JM, Demer JL. Three-dimensional location of
human rectus pulleys by path inflections in secondary gaze positions.
Invest Ophthalmol Vis Sci 2000;41:3787-97.
16. Siegel LM, DeSalles NL, Rosenbaum AL, Demer JL. Magnetic
resonance imaging features of two cases of acquired Brown syn-
drome. Strabismus 1998;6:19-29.
17. Demer JL, Miller JM. Magnetic resonance imaging of the functional
anatomy of the superior oblique muscle. Invest Ophthalmol Vis Sci
1995;36:906-13.
18. Blanchard CL, Young LA. Acquired inflammatory superior oblique
tendon sheath (Brown syndrome). Report of a case following frontal
sinus surgery. Arch Otolaryngol 1984;110:120-122
19. Rosenbuam AL, Astle WF. Superior oblique and inferior rectus
muscle injury following frontal and intranasal sinus surgery. J Pediatr
Ophthalmol Strabismus 1985;22:194-202.
20. Wilson EM, Eustis SH, JR., Parks MM. Brown syndrome Surv
Ophthalmol 1989;34:153-72.
21. Wright KW, Silverstein D, Marrone AC, Smith RE. Acquired in-
flammatory superior oblique tendon sheath syndrome—a clino-
pathologic study. Arch Ophthalmol 1982;100:1752-4.
22. Levine MR, Boynton J, Tenzel RR, Miller GR. Complications of
blepheroplasty. Ophthalmic Surg 1975;6:47-53.
23. Barnette JA, Griffiths JC, West RH. ,Acquired Brown syndrome .
Ann Rheum Dis; 1993;52:835. (letter comment)
24. Hermann JS. Acquired Brown syndrome of inflammatory origin.
Arch Ophthalmol 1978;96:1228-32.
25. Whitefield L, Isenberg DA, Brazier DJ, Forbes J. Acquired Brown
syndrome in systemic lupus erythematosus. Br J Rheum 1995;34:
1092-4.
26. Brahma AK, Hay E, Sturgess DA, Morgan LH. Acquired Brown
syndrome and primary Sjogren’s syndrome . Br J Ophthalmol 1995;
79:89-90. (letter)
27. Roper-Hall MJ, Roper-Hall G. The superior oblique “click” syn-
drome in Orthoptics. Proc Second Int Orthoptic Congress Amster-
dam, May 1971:11-13.
28. Clark RA, Miller JM, Rosenbaum AL, Demer JL. Heterotopic mus-
cle pulleys or oblique muscle dysfunction? J AAPOS 1998;2:17-25.
29. Romaine HH. Motility surgery. NY State J Med 1963;63:1511-4.

An Eye on the Arts – The Arts on the Eye

THE LAST RAYS of evening light are filtering through the window when
Vishnu sees the image. A man is standing over his body on the landing down
below. He kneels besides him, and pulls back the sheet. With one hand, the man
touches Vishnu’s cheek; with the other, he presses the forehead and brushes the
wisps of hair off the eyes. Fingertips trace across Vishnu’s lips, then down his
chin, and to his chest, where they rub against his heart.
The man has his eyes closed. His neck is arched, head tilted upwards, lips
reciting silent words. Vishnu has seen this silhouette before, he knows he should
recognize the crouching figure.
The man’s eyes open. Their whiteness reaches through the dark. They are
large and milky, staring up through the air, through the ceiling, through the
stone, at some point outside in the sky. Vishnu looks at them and is unsure if
they are filled with reverence or fear.
The eyes blink, the fingers caress the tufts of chest hair, the lips open and
close. Soft words float slowly up from the upturned face. Vishnu sees the gray
hair, sees the bulbous nose, sees the pockmarks on the cheeks. Recognition
floods in finally. He peers down at Mr. Jalal on the landing, crouching next to
his body, staring up through the darkness towards heaven.—Manil Suri (from
The Death of Vishnu, HarperCollins)