Ophthalmol Clin N Am 17 (2004) xi – xii
Preface
Anterior Segment Complications of Posterior
Segment Surgery
Dante J. Pieramici, MD
Guest Editors
Our ability to manage increasingly complex
vitreoretinal pathology has been tremendously accel-
erated by recent technologic advances in vitreoreti-
nal surgery. Whereas only two decades ago giant
retinal tears, macular holes, detachments associated
with retinopathy of prematurity, advanced prolifer-
ative diabetic retinopathy, and proliferative vitreo-
retinopathy were for the most part inoperable, today
these pathologic processes are routinely managed.
With improvements in outcomes have come height-
ened expectations, and it is no longer the goal to
simply save some basic level of vision or the
anatomic globe but instead to restore higher levels
of visual function. To this end, we as physicians are
constantly attempting to bfine tuneQ our surgical
skills and reduce complications, because their occur-
rence may limit visual outcomes. A zero percent
complication rate is impossible, and despite impec-
cable surgical technique, vision-limiting side effects
are anticipated. This is especially true today with
modern vitreous surgical techniques, as many of the
clinical adjuncts that allow us to treat posterior seg-
ment pathology have potentially deleterious effects
0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.07.006
Allesandro Castellarin, MD
on the anterior segment. As scientists we strive to
develop newer techniques and adjuncts that have
fewer unwanted side effects; as clinicians we wel-
come the current technology and work within its
limitations to strive for the best outcomes possible.
In this issue of the Ophthalmology Clinics of
North America, we have assembled an international
group of scholars to review and elaborate upon
complications affecting the anterior segment follow-
ing the management of posterior segment pathology.
There have been many reviews describing posterior
segment complications of anterior segment proce-
dures, but we think the converse has been rarely
discussed in detail. The information included in this
issue should be useful for the anterior as well as
posterior segment specialist, because it is often our
anterior segment colleagues who are called upon to
manage these complications. In each section of this
issue, the authors provide detailed descriptions of the
associated complications, with an emphasis on the
pathophysiologic mechanisms involved. Highlighting
mechanisms permits rational approaches to preven-
tion and treatment of these complications.
ophthalmology.theclinics.com
xii D.J. Pieramici, A. Castellarin / Ophthalmol Clin N Am 17 (2004) xi – xii

We appreciate the efforts and expertise of all the Dante J. Pieramici, MD

contributors featured in this issue. We also acknowl- Allesandro Castellarin, MD
edge the work of so many who have pioneered the California Retina Consultants
development and refinement of vitreoretinal tech- 515 East Micheltornea, Suite C
niques in the past few decades. Despite the short- Santa Barbara, CA 91303, USA
comings described in this issue, these techniques E-mail addresses: dpieramici@yahoo.com
allow us to salvage and restore vision to hundreds of (D.J. Pieramici)
thousands of patients worldwide who without this aacastellarin@yahoo.com (A. Castellarin)
technology would be blind.
 Ophthalmol Clin N Am 17 (2004) 495 – 506

Strabismus following posterior segment surgery

Maria B. Yadarola, MD, Megan Pearson-Cody, DO, David L. Guyton, MD*
The Krieger Children’s Eye Center at the Wilmer Institute, 233, The Johns Hopkins University School of Medicine,
600 North Wolfe Street, Baltimore, MD 21287-9028, USA

Although the reported incidence of muscle imbal- Causes

ance after retinal reattachment surgery varies among
the existing reports, overall, it has ranged from 3% to
as high as 60% [1 – 23]. Interestingly, the incidence
of strabismus (tropias) following scleral buckling
procedures performed under general anesthesia has
been reported to range from 4% to 11% [4,6,13,
16,20], whereas when local anesthesia has been used,
the incidence of strabismus has been reported to
range from 15% to 43% [2,4,21 – 23]. This difference
suggests that local anesthetic myotoxicity may be an
important cause of strabismus after posterior seg-
ment surgery.
Even though strabismus immediately following
retinal reattachment surgery seems to be relatively
common, it usually resolves within 3 to 6 months, and
only 5% to 25% of patients experience persistent
strabismus [1,3,4,6 – 8,10,11,13 – 16,18,24,25]. Sev-
eral mechanisms have been proposed to explain
permanent strabismus. Mechanical causes include
adhesions formed between extraocular muscles and
the sclera or between these tissues and the periorbita,
exoplants placed beneath extraocular muscles, inser-
tional changes when a muscle is removed and
reattached, and redirection of force vectors caused
by the scleral buckle. Other mechanisms include
direct muscle injury, foveal misalignment, surgically
induced anisometropia, and sensory disruption.
Mechanical causes
Adhesions
Adhesions may form between extraocular muscles
and orbital connective tissue or sclera and are likely
the major cause of mechanical restrictions [1,26].
These adhesions are thought to be the result of
Tenon’s capsule violation that exposes extraconal fat,
leading to a fibrotic inflammatory reaction. This
fibrotic reaction, termed the fat adherence syndrome,
causes scarring, contraction, and restriction, creating
a restrictive strabismus [27]. Surgical attempts to
remove such restrictions are difficult and often un-
successful because of recurrent inflammation and
further scarring [1,3,27].
Insertional changes
If an extraocular muscle is detached during reti-
nal reattachment surgery, it may inadvertently be
recessed, advanced, or transposed when attempting
to reattach it to its original insertion site [1,3]. On
the other hand, it has been suggested that muscle
disinsertion is useful to gain exposure and may avoid
prolonged stretching that can cause muscle damage
[28,29]. The authors recommend that extraocular
muscle disinsertion be avoided whenever possible.
If it is necessary, the muscle should generally be
replaced at its original insertion site.

Exoplant
* Corresponding author. Several studies implicate the scleral buckle in
E-mail address: dguyton@jhmi.edu (D.L. Guyton). restrictive strabismus after posterior segment surgery.

0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.005 ophthalmology.theclinics.com
496 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506
Its type, size, and location, and the fibrotic reaction of
adjacent tissues can have an effect on the incidence of
postoperative strabismus [1,6,7,10,15 – 17,19]. When
the exoplant is placed beneath the extraocular muscle,
a large sponge can effectively tighten the muscle as
well as limit its action. Furthermore, the buckling
element may change the direction of forces if it in-
volves the oblique muscles and may lead to vertical
strabismus and ocular torsion. Anterior displacement
of the superior oblique tendon by an encircling band
can significantly decrease its depressing effect and
intort the globe. Anteriorization of the inferior
oblique muscle may also cause vertical strabismus as
well as extorsion [1,3,25,26,30]. Scleral buckle sur-
gery may cause muscle weakness from associated
damage to the muscle, muscle shortening with a
resection-like effect, or a change in the muscle’s field
of action. Circumferential scleral buckles have been
reported to cause more postoperative ocular motility
restriction than local, radial, or oblique sponges [2,5,
6,15,18,22,31,32].
In vitrectomy surgeries performed without placing
an exoplant, one might be tempted to assume that
the incidence of postoperative strabismus is lower.
Nevertheless, a British study found that the incidence
of muscle imbalance following posterior segment
surgery was similar in patients undergoing vitrectomy
alone or with a scleral buckling procedure [33].
Muscle/nerve injury
The extraocular muscles can be injured directly
during posterior segment surgery owing to stretching,
rupture, and anesthetic myotoxicity. Excessive muscle
stretching can cause muscle fibrosis, which may lead
to restrictive strabismus. Excess traction applied to the
muscle during surgery can cause avulsion or rupture.
There have been several reports of muscle rupture
after presumed damage from aggressive cryotherapy
[1,3,34 – 36]. Moreover, if a muscle has to be removed
to gain more exposure during retinal reattachment
surgery, the nerves innervating these muscles may
accidentally be damaged, causing muscle paresis
[1,3]. The incidence of strabismus following scleral
buckling procedures performed under general anes-
thesia is approximately 4% to 11% [4,6,13,16,20],
whereas when local retrobulbar or peribulbar anes-
thesia is used, the incidence rises to 15% to 43%
[2,4,21 – 23]. This difference suggests that myotox-
icity caused by local anesthetic agents may have a role
in restrictive strabismus following scleral buckling
surgery. Retrobulbar or peribulbar anesthesia can
cause strabismus after surgery via muscle trauma
from the anesthetic needle, trauma to the primary
motor nerve, or by a fibrosing inflammatory reaction
representing myotoxicity from the anesthetic agent
itself [4,30]. Many local anesthetic agents have been
shown to be myotoxic, but restrictive strabismus has
been reported more often since the introduction of
long-acting local anesthetic agents such as bupiva-
caine [4].
The injection of an anesthetic agent directly into
an extraocular muscle may lead to initial paresis and
later fibrosis [3,37,38]. The end result of myotoxicity
is muscle shortening owing to localized muscle fibro-
sis and scarring, as evidenced by orbital CT and MRI
scans as well as histology [4,39,40]. Carlson and
colleagues [41] noticed that local anesthetic myotox-
icity to monkey extraocular muscles was more
extensive when the agent was injected into the
muscle than when it was infused surrounding the
muscle. In other words, direct injection is probably
necessary to cause significant changes. The most
common presentation is hypotropia and limited
elevation, usually with a V pattern and extorsion of
the hypotropic eye. This hypotropia is caused by the
frequent involvement of the inferior rectus muscle
during retrobulbar or peribulbar anesthesia. Actually,
any of the extraocular muscles can be reached by a
1.5-inch (38 mm) retrobulbar needle from an infero-
temporal approach, as shown in a cadaver study by
Capó and colleagues [42] in 1996.
Many patients, especially older ones with poor
muscle regenerative ability, present initially with an
apparent paresis of the involved extraocular muscle
that evolves several weeks to 2 months later into a
pattern of strabismus consistent with overaction of
the involved muscle [43]. This overaction pattern is
most likely explained by segmental fibrotic short-
ening of the affected muscle at the site of anesthetic
damage. This shortening leads to stretching of the
remaining actin and myosin myofilament interdigita-
tions, decreasing the active force of the muscle. When
the shortened muscle moves the eye into the field of
action of the affected muscle, increased interdigita-
tion of the myofilaments occurs, and the muscle
becomes stronger, creating an overaction pattern. In
this mechanism of the overaction pattern, the fibrotic
shortening must occur in only one segment of the
muscle, otherwise a restrictive pattern is seen [43].
An additional consideration involves hyaluroni-
dase (Wydase), an enzyme frequently added to local
anesthetic agents for retrobulbar or peribulbar blocks.
Clusters of cases of diplopia have been observed after
periocular anesthesia without hyaluronidase [44 – 46].
Hyaluronidase reportedly decreases the risk of local
anesthetic myotoxicity because it speeds the onset of
anesthesia and akinesia, reducing the amount of anes-
 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506
thetic needed. It also promotes more rapid disper-
sion of the anesthetic agent away from the injection
site [47].
Foveal misalignment
Foveal misalignment caused by macular reposi-
tioning surgery, or simply by retinal disease that
produces traction on the fovea, is another cause of
diplopia and strabismus. Macular repositioning sur-
gery for choroidal neovascularization secondary to
age-related macular degeneration was first described
in 1983 by Lindsey and colleagues [48]. In 1993,
Machemer and Steinhorst [49] developed a technique
that involved detaching the entire retina from the
retinal pigment epithelium, performing a peripheral
360-degree retinotomy, and rotating this around the
optic nerve. This technique, as can be expected,
caused diplopia in nearly 100% of patients. Limited
translocation was developed in an attempt to simplify
macular translocation and to decrease postoperative
complications. In this procedure, the macula is
translocated by using scleral imbrication without the
need for a 360-degree retinotomy [50,51]. In a study
by Buffenn and colleagues [52] performed in 2001 at
the authors’ center, only 5.2% of patients undergoing
limited macular translocation complained of diplopia,
with most successfully treated with prisms.
Foveal misalignment caused by retinal disease
producing traction on the fovea is another important
cause of diplopia. Although this sometimes occurs
after retinal reattachment surgery, with the retina re-
attaching in a distorted manner, it can also occur from
direct foveal disease processes. In 1980, Burgess and
colleagues [53] described binocular central diplopia
associated with subretinal neovascular membranes. In
1984, Bixenman and Joffe [54] discussed this finding
in association with retinal wrinkling.
The authors have seen approximately 80 patients
over the past 15 to 20 years with a small angle,
comitant diplopia, almost always with a small vertical
component, that has been impossible to correct with
prisms. In each of these patients, the Amsler grid test
was abnormal in at least one eye. In patients who
underwent Lancaster red-green testing, a nonspecific,
small, comitant, usually vertical misalignment was
documented. All of these patients had retinal disease
producing traction on the fovea, which caused the
fovea in one eye to be dragged out of alignment with
the fovea in the other eye. The peripheral fusion in
these patients maintained single vision in the periph-
ery, but foveal fusion was not strong enough to
compete with peripheral fusion, and central double
vision was the result. Although prisms may momen-
497
tarily relieve central diplopia as the foveas are re-
aligned, peripheral diplopia occurs, peripheral fusion
mechanisms overwhelm central fusion, and central
diplopia recurs. In 90% of the authors’ patients, best-
corrected visual acuity was better than 20/40 in the
affected eye. A possible explanation is that patients
with moderate to severe visual impairment are less
likely to notice central diplopia. The authors have
named this syndrome the bdragged-fovea diplopia
syndromeQ [55].
In the authors’ experience, the most common
cause of this problem has been a parafoveal epiretinal
membrane. Other causes are scars near the fovea,
subretinal neovascular membranes causing foveal
distortion, central serous choroidopathy, and a pre-
viously detached retina that reattaches in a distorted
manner. Patients with retinopathy of prematurity with
macular dragging also have distortion of large por-
tions of the retina but generally do not have diplopia
because of longstanding suppression. In a small
percentage of patients, complaints of diplopia second-
ary to epiretinal membranes did not occur until the
postoperative period. This observation was attributed
to the improvement of visual acuity, making the
second image more noticeable to the patient.
The authors have developed a useful test to
demonstrate this competition between peripheral
versus central fusion, called the bsmall-field central
fusion testQ or, more simply, the blights on/off test.Q
The test consists of asking the patient to view a single
white letter, about 20/70 in size, on a black monitor
screen. With the room lights on, the single letter is
double as seen by patients with the dragged-fovea
diplopia syndrome. All lights in the room are then
suddenly turned off, eliminating peripheral fusion be-
cause of the absence of peripheral stimuli. If the dou-
ble letter becomes single within 2 to10 seconds, the
result is labeled a positive small-field central fusion
test, or positive blights on/off Q test, practically pathog-
nomonic for the dragged-fovea diplopia syndrome.
Although diplopia has been described in several
studies as one of the symptoms of epiretinal mem-
branes or retinal disease, it has not previously been
recognized as a common finding in patients with
these problems. Neither the incidence nor the pre-
valence has been reported. The authors see it fre-
quently, but our patient population is biased toward
diplopia problems.
Other causes
Any outcome that alters fusion after retinal
reattachment surgery can lead to a bsensoryQ form
of strabismus. Fusion loss can result from poor vision
498 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506
or from the anisometropia and aniseikonia secondary
to aphakia, the use of silicone oil, or simply from
myopia induced by the scleral buckle [1,3].
Evaluation
Careful evaluation should be performed before
any attempted correction of strabismus occurring
after posterior segment surgery. A thorough review
of past surgical records is important; a history of
multiple surgical procedures performed on the af-
fected eye is associated with a poorer prognosis for
restitution of binocularity [56].
Assessing best-corrected visual acuity helps rule
out the ghost images of monocular diplopia, usually
caused by irregular corneal astigmatism or early
cataracts. Poor visual acuity does not eliminate the
possibility of diplopia from strabismus, because some
patients may still be aware of peripheral diplopia [1].
Ocular motor evaluation should include measur-
ing primary and secondary deviations using prisms
and alternate cover testing, in the nine diagnostic
positions of gaze, with special attention given to the
reading position. Ductions and versions are per-
formed to evaluate and quantify the presence or
absence of limitations and overactions. If limitation is
found, forced duction testing is needed to distinguish
restriction from paralysis, often performed more
easily at surgery than in the office.
Ocular torsion is the term used for an abnormal
rotation of the eye about the line of sight. Anatomic
torsion and subjective torsion often differ because
of sensory adaptations to cyclodeviations, and both
should be assessed [57]. To estimate or measure ana-
tomic torsion, one can use the indirect ophthalmo-
scopic fundus examination or fundus photography. To
measure subjective torsion, one may use double
Maddox rod testing, an amblyoscope, or the Lancas-
ter red-green test [57].
The Lancaster red-green test is fast, conveys
torsional and vertical deviations accurately in the
nine diagnostic positions of gaze, and is easy to
interpret. The patient wears a pair of red-green filter
goggles with the red filter over the right eye. The
examiner then takes the hand-held streak projector of
the color corresponding to the eye that is going to be
the fixing eye and directs that light streak to the wall
in each of the nine positions of gaze in turn. The
patient is asked to superimpose his or her light streak
on the examiner’s light streak. The examiner then
switches lights with the patient, so as to switch the
eye that fixates in each direction of gaze, and repeats
the procedure. The results are recorded on a grid.
From this diagram, one can easily determine patterns
of horizontal and vertical deviation, as well as pat-
terns of ocular torsion (Fig. 1) [57].
To detect ocular torsion by means of indirect
ophthalmoscopy, the position of the fovea is judged
relative to the optic nerve head. Because the indirect
ophthalmoscopic image is inverted and reversed, if
the image appears extorted, the eye is extorted. To
estimate the amount of torsion, a practical grading
system is applied using the optic disk diameter as a
guide. The grading system is scaled from trace to 4+
torsion, with the fovea normally appearing level with
the upper third of the disk in the indirect ophthalmo-
scopic view. If the fovea is higher than this, there is
extorsion; if it is lower, there is intorsion (Fig. 2) [57].
It is helpful to evaluate the patient’s potential for
fusion, because even if orthotropia is achieved with
surgical treatment, double vision will recur if fusion
is not possible. Fusion assessment is performed by
optically aligning the eyes and testing for evidence of
fusion, usually performed by measuring fusional ver-
gence amplitudes with prisms or a haploscope [1,3].
Postoperative deviations are usually classified as
comitant or incomitant. Comitant deviations are
generally the result of sensory deprivation secondary
to any disorder that interferes with fusion. If the
deviation is small and comitant, and especially if it is
only intermittently responsive to prisms, the dragged-
fovea diplopia syndrome should be considered. The
lights on-off test and Amsler grid testing should be
performed, as well as a careful slit-lamp fundus
examination looking for foveal abnormalities such as
an epiretinal membrane. If the deviation is incomi-
tant, special attention should be focused on ductions,
versions, and measurement of the horizontal, vertical,
and torsional deviations in all fields of gaze. In this
case, the strabismus is most likely caused by mechani-
cal limitations, muscle injury, or insertional change.
Management
Surgical treatment
Before surgery is performed for strabismus fol-
lowing posterior segment surgery, it is wise to discuss
alternative treatments and possible outcomes with the
patient, because the results of surgery in such cases
are less predictable than in other strabismus cases,
and standard tables for strabismus surgery are not as
applicable [3,58].
Unless there are contraindications because of
health, the authors prefer to perform these surgeries
 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506 499

under general anesthesia. Forced ductions are more
reliable, and exposure of the operative site is better
without the tissue swelling produced by the local
anesthetic. Forced duction testing is done at all stages
of the surgery, that is, before and after muscle dis-
insertion, after lysis of adhesions, and after reposi-
tioning of the muscles.
There is much controversy regarding whether to
remove the scleral buckle. The presence of the scleral
buckle usually complicates the strabismus surgery,
but removing the exoplant has led to retinal rede-
tachment in 4% to 33% of cases [3,59]. In addi-
tion, excessive manipulation to remove the buckle
can induce even more scarring and restriction. For

Fig. 1. (A – D) Lancaster red-green plots. Solid red line = right eye; dashed green line = left eye. The small black dots are
separated by 15 PD. Large black dots indicate fixation target for the nine diagnostic positions of gaze. The examiner holds
the green flashlight (left eye fixing) in A, B, and D and the red flashlight (right eye fixing) in C. (A) A typical pattern from a
patient with the right inferior rectus muscle affected by anesthetic myotoxicity following cataract surgery. The right hypotropia
and esotropia increase in downgaze. The extorsion of the right eye is worst in gaze down and to the left, in the torsional field of
action of the inferior rectus muscle. (B) A similar pattern from right inferior rectus muscle anesthetic myotoxicity following
repair of a retinal detachment. The telltale extorsion increases on gaze down and to the left, as seen in A. (C) Plot of a patient
after retinal detachment repair with myotoxicity of the left superior rectus muscle. The large left hypertropia increases on upgaze
as does the intorsion, especially notable in up and right gaze, the torsional field of action of the left superior rectus muscle.
(D) Plot of a patient with a tight restricted right inferior rectus muscle following repair of a retinal detachment. The right
hypotropia increases on upgaze in this case.
500 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506

Fig. 1 (continued).

this reason, the authors usually leave the buckle in
place, unless it is found to be the direct cause of the
muscle imbalance.
In comitant deviations secondary to sensory stra-
bismus, standard surgical procedures are performed
using recession or resection of indicated muscles. The
authors prefer to operate on the poorer eye to avoid
risk to the better eye. In incomitant strabismus caused
by adhesions, freeing the affected muscle from
adhesions and recessing it with adjustable sutures
has usually given good results. Resections should be
done with caution and are a poor choice as a single
procedure in patients with a restrictive component
[1]. Forced duction testing at the conclusion of the
surgery is important to confirm elimination of any
significant restriction.
The authors perform most strabismus surgery
through small radial cul-de-sac incisions through
the conjunctiva and Tenon’s capsule. If exposure
proves to be inadequate, the radial incision is
extended to the limbus, and the approach is converted
to a limbal incision. It is often difficult to isolate the
rectus muscle anteriorly in the area of the buckle
capsule, and it is often hooked with a muscle hook
posterior to the buckle. If the buckle is tight, and the
buckle capsule is well formed, the posterior insertion
of the muscle onto the buckle capsule may be treated
as a secondary insertion of the muscle, and any re-
cession or resection of the muscle is done with
respect to that secondary insertion. If the buckle is
loose, and the buckle capsule is not well formed,
isolating the muscle sometimes exposes the buckle,
 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506 501

Fig. 1 (continued).

but even in these cases, no problems have occurred
from infection or buckle extrusion. The wound is
often irrigated with antibiotic solution when the
buckle becomes exposed as a precaution against
these complications.
If the muscle is isolated anterior to the buckle and
is to be recessed in such a way that the end of the
muscle will hang backward over the exoplant and
will not have the opportunity to attach to the globe,
the authors either suspend the muscle on an adjust-
able suture over the exoplant using a permanent
suture material such as 6-0 polyester, or the muscle
sutures are fished under the exoplant from behind so
that the cut muscle end contacts sclera rather than
lying on top of the silicone exoplant material. A
similar procedure is used for resections. If the cut end
of the muscle, suspended on the adjustable suture, is
in danger of lying over the silicone exoplant material,
the suture ends are fished under the exoplant from be-
hind so that the cut end of the muscle contacts sclera.
Surgery on the superior oblique tendon may be
difficult, especially if the buckle has been placed
posterior to the superior oblique tendon, displacing it
forward. If this complication is identified, careful
consideration should be given to removing the buckle
to restore the normal anatomic course of the tendon.
The buckle can be excised only locally in such cases
and even sutured back together if necessary.
Damage to the superior oblique tendon or its
insertion is not uncommon after scleral buckle sur-
502 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506

Fig. 1 (continued).

gery. Commonly, such eyes are extorted and hyper-
tropic, with a pattern of superior oblique paresis. A
modified Harada-Ito procedure, consisting of lateral-
izing the anterior third to half of the superior oblique
tendon on an adjustable suture, is often helpful in
such cases [1,3].
Strabismus caused by local anesthetic myotoxicity
usually responds well to a single large recession or a
recession-resection procedure, best done with adjust-
able sutures. When multiple surgeries have been per-
formed on the affected eye, with significant scarring
and fibrosis, definitive muscle surgery on the non-
affected eye is often best, with a higher chance of
success than reoperating on the affected eye. If the
conjunctiva is significantly scarred or shortened, con-
junctival recession is often used to avoid creating new
restrictions [60].
Strabismus following retinal reattachment surgery
can sometimes be improved by injection of botu-
linum toxin (Botox) into a shortened or tight muscle.
This injection paralyzes the muscle for 1 to 3 months,
allowing the opposing muscle to shorten permanently
somewhat, improving the deviation [61]. In the
presence of scar tissue, this technique is usually not
successful. One must wait several months for a sta-
ble effect, and multiple treatments are often needed.
Nevertheless, benefit rates have been reported to
range from 15% to 85% when Botox has been used as
primary treatment [61,62], and it can also be used as
an adjunct to other strabismus surgery [1].
 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506
Fig. 2. Grading system for estimating abnormal torsion by
indirect ophthalmoscopy. The fovea is normally level within
the upper third of the optic disk in the indirect ophthalmo-
scopic view, as illustrated here for a left fundus. (From
Guyton DL. Clinical assessment of ocular torsion. Am
Orthoptic J 1983;33:7 – 15; with permission.)
Nonsurgical treatment
Nonsurgical management of strabismus following
posterior segment surgery includes the use of prisms,
optical blur, and partial occlusion.
Patients presenting with small (less than 12 –
15 PD) comitant deviations can often be treated with
prisms ground into their glasses. If the deviation is
greater, the patient can often benefit from plastic
Fresnel prisms, which range from 0.5 to 40 PD and
are relatively inexpensive. They are easily applied on
the rear surface the patient’s current glasses with just
water, usually over the nonpreferred eye. Fresnel
prisms have the disadvantage of reducing visual
acuity if the power of the prism is high. Prisms cannot
be used to correct torsional deviations [1,3].
A trial of prism correction in the office is often
useful before a prescription is given. If the patient
beats upQ the prism, one should be wary of the
dragged-fovea diplopia syndrome, for which prism
correction is futile. An advantage of Fresnel prisms is
that they can be added to a portion of the lens. If the
patient only has trouble reading, the prism can be
placed on the lower portion of the lens, leaving the
rest of the lens prism free. Prisms are often used as a
preoperative trial. If the patient can gain good fusion
with the prism correction, it is likely that he or she
will fuse postoperatively. Prisms can also be useful
after surgery if small deviations remain [1].
If the patient has reasonably good vision in each
eye, a plus lens can be used to blur one eye optically
503
for distance, converting the patient to a bmonovisionQ
situation using one eye for distance and one eye for
near. In some cases, ordinary monovision is enough
to eliminate bothersome diplopia. In other cases, the
blurring in the eye is not enough to eliminate the
diplopia, and an even stronger plus lens may be tried
to blur that eye entirely for distance and near.
Many adults learn to suppress or ignore the
second image after several months. If this is not
possible, monocular occlusion is a last resort for
patients in whom binocular vision cannot be restored.
Options are to patch one eye or to blur the current
lens using Bangerter foils, clear fingernail polish, or
Scotch Satin tape (3M Consumer Stationary Division,
St. Paul, Minnesota). The authors prefer the use of a
piece of Scotch Satin tape placed on the rear surface
of the patient’s lens. This material immediately
eliminates the diplopia. The benefit of this type of
tape is that it is scatters light enough to relieve the
diplopia while being clear enough to not be obvious
to others. Another option is an opaque contact lens.
Preventive measures
The best approach to the problem of strabismus
after posterior segment surgery is prevention of the
problem in the first place. If local anesthesia is used,
the authors recommend sub-Tenon’s administration
of the local anesthetic with a blunt cannula rather than
by a retrobulbar or peribulbar route with a sharp
needle to avoid local anesthetic myotoxicity.
Sub-Tenon’s administration of local anesthesia is
performed under sedation. A topical anesthetic drop
of proparacaine hydrochloride (0.5%) is followed by
a drop of antiseptic solution, and then a small incision
is made through the conjunctiva and Tenon’s tissue in
the fornix, oriented inferonasally to avoid the oblique
muscles. The anesthetic agent is infused into the sub-
Tenon’s space using a blunt cannula (Fig. 3) [3]. This
Fig. 3. Professor P. Muthusamy’s Sub-Tenon Cannula
(24 gauge) from Zabby’s Ophthalmic Instruments (www.
zabbys.itgo.com).
504 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506
technique is a simple and safe way to administer local
anesthesia and provides akinesia comparable with
that of retrobulbar or peribulbar approaches with
just 3 to 4 mL of anesthetic mixture [63,64]. Compli-
cations of this technique are chemosis and conjunc-
tival hemorrhage [65,66], but the authors have never
seen local anesthetic myotoxicity as a result of
this procedure.
During the posterior segment surgery, the surgeon
should be careful not to violate the orbital fat pads to
avoid scarring from the fat adherence syndrome.
Also, one should avoid excessive dissection around
the extraocular muscles and excessive tension on
traction sutures placed beneath the muscles. Care
should be taken when placing the buckle in the area
of the superior oblique tendon. The buckle should
always pass inferior to the tendon.
Summary
Persistent strabismus following posterior segment
surgery has a reported incidence ranging from 5% to
25%. Many mechanisms have been proposed to
explain this strabismus, with some being preventable.
Careful dissection and a gentle approach to extra-
ocular muscles during posterior segment surgery can
prevent a significant number of complications. The
use of general anesthesia or sub-Tenon’s local
anesthesia rather than retrobulbar or peribulbar
anesthetic blocks will avoid strabismus from local
anesthetic myotoxicity. Although strabismus after
posterior segment surgery is often challenging to
treat, surgical intervention or nonsurgical adjuncts
such as prisms or partial tape occlusion can often
succeed in relieving symptoms.
References
[1] Seaber JH, Buckley EG. Strabismus after retinal
detachment surgery: etiology, diagnosis, and treatment.
Semin Ophthalmol 1995;10:61 – 73.
[2] Spencer AF, Newton C, Vernon SA. Incidence of
ocular motility problems following scleral buckling
surgery. Eye 1993;7:751 – 6.
[3] Farr AK, Guyton DL. Strabismus after retinal detach-
ment surgery. Curr Opin Ophthalmol 2000;11:207 – 10.
[4] Salama H, Farr AK, Guyton DL. Anesthetic myotox-
icity as a cause of restrictive strabismus after scleral
buckling surgery. Retina 2000;20:478 – 82.
[5] Cooper LL, Harrison S, Rosenbaum AL. Ocular tor-
sion as a complication of scleral buckle procedures for
retinal detachments. J Am Assoc Pediatr Ophthalmol
Strabismus 1998;2:279 – 84.
[6] Fison PN, Chignell AH. Diplopia after retinal detach-
ment surgery. Br J Ophthalmol 1987;71:521 – 5.
[7] Mets MB, Wendell ME, Geiser RG. Ocular deviation
after retinal detachment surgery. Am J Ophthalmol
1985;99:667 – 72.
[8] Amemiya T, Yoshida H, Harayama K, Miki M, Koi-
zumi K. Long-term results of retinal detachment sur-
gery. Ophthalmologica 1978;177:64 – 9.
[9] Arruga A. Motility disturbances induced by operations
for retinal detachment. Mod Probl Ophthalmol 1977;
18:408 – 14.
[10] Kutchera E, Antlanger H. Influence of retinal detach-
ment surgery on eye motility and binocularity. Mod
Probl Ophthalmol 1979;20:354 – 8.
[11] Maillete de Buy Wenninger-Prick L, Van Mourik-
Nordeenbos A. Diplopia after retinal detachment sur-
gery. Doc Ophthalmol 1988;70:237 – 42.
[12] Peduzzi M, Campos EC, Guerrieri F. Disturbances
of ocular motility after retinal detachment surgery.
Doc Ophthalmol 1984;58:115 – 8.
[13] Price RL, Pederzolli A. Strabismus following retinal
detachment surgery. Am Orthoptic J 1982;32:9 – 17.
[14] Roth AM, Sypnicki BA. Motility dysfunction follow-
ing surgery for retinal detachment. Am Orthoptic J
1975;25:118 – 21.
[15] Sewell JJ, Knobloch WH, Eifrig DE. Extraocular
muscle imbalance after surgical treatment for retinal
detachment. Am J Ophthalmol 1974;78:321 – 3.
[16] Theodossiadis G, Nikolakis S, Apostolopoulus M.
Immediate postoperative muscular disturbance in reti-
nal detachment surgery. Mod Probl Ophthalmol 1979;
20:367 – 72.
[17] Wolff SM. Strabismus after retinal detachment surgery.
Trans Am Ophthalmol Soc 1983;81:182 – 92.
[18] Kanski JJ, Elkington AR, Davis MS. Diplopia after
retinal detachment surgery. Am J Ophthalmol 1973;76:
38 – 40.
[19] Waddell E. Retinal detachment and orthoptics. Br
Orthop J 1983;40:5 – 12.
[20] Klainguti G, Castella A, Chamero J, Gonvers M.
Extraocular muscle complications in retinal detach-
ment surgery. In: Kaufmann H, editor. Transactions
of the 19th meeting of the European Strabismological
Association. 1991. p. 125 – 30.
[21] Metz HS, Norris A. Cyclotorsional diplopia following
retinal detachment surgery. J Pediatr Ophthalmol Stra-
bismus 1987;24:287 – 90.
[22] Smiddy WE, Loupe D, Michels RG, Enger C, Glaser
BM, deBustros S. Extraocular muscle imbalance af-
ter scleral buckling surgery. Ophthalmology 1989;96:
1485 – 9.
[23] Berk AT, Saatci AO, Kir E, Durak I, Kaynak S.
Extraocular muscle imbalance after scleral buckling.
Strabismus 1996;4:69 – 75.
[24] Arruga A. Binocularity after retinal detachment sur-
gery. Doc Ophthalmol 1973;34:41 – 5.
[25] Muñoz M, Rosembaum AL. Long term strabismus
complications following retinal detachment surgery.
J Pediatr Ophthalmol Strabismus 1987;24:309 – 14.
 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506
[26] Wright KW, Hwang J. Diplopia and strabismus after
retinal and glaucoma surgery. Am Orthopt J 1994;44:
26 – 30.
[27] Wright KW. The fat adherence syndrome and stra-
bismus after retinal surgery. Ophthalmology 1986;93:
411 – 5.
[28] Cibis PA. General discussion. In: Schepens CL, Regan
CDJ, editors. Controversial aspects of the management
of retinal detachment. Boston7 Little, Brown; 1965.
p. 222 – 3.
[29] Pischel DK. Complications of retinal detachment
surgery. In: Fasanella RM, editor. Management of
complications in eye surgery. Philadelphia7 Saunders;
1965. p. 299 – 345.
[30] Kushner BJ. Unexpected cyclotropia simulating dis-
ruption of fusion. Arch Ophthalmol 1992;110:1415 – 8.
[31] Pearlman JT, Christensen RE. Motility problems
following retinal detachment surgery. Am Orthoptic J
1972;22:64 – 7.
[32] Maurino V, Kwan A, Khoo B, Gair E, Lee JP. Ocular
motility disturbances after surgery for retinal detach-
ment. J Am Assoc Pediatr Ophthalmol Strabismus
1998;2:285 – 92.
[33] Wright LA, Cleary M, Barrie T, Hammer HM. Motility
and binocularity outcomes in vitrectomy versus scleral
buckling in retinal detachment surgery. Graefes Arch
Clin Exp Ophthalmol 1999;237:1028 – 32.
[34] Bell FC, Pruett RC. Effects of cryotherapy upon
extraocular muscle. Ophthalmic Surg 1977;8:71 – 5.
[35] Hamlet YJ, Goldstein JH, Rosenbaum JD. Dehiscence
of lateral rectus muscle following intrascleral buckling
procedure. Ann Ophthalmol 1982;14:694 – 7.
[36] McPherson A. Complications during surgery for retinal
detachment. Highlights Ophthalmol 1969;12:43 – 50.
[37] Guyton DL. Strabismus complications from local
anesthetics. In: Long DA, editor. Anterior segment
and strabismus surgery. Transactions of the New
Orleans Academy of Ophthalmology. New York7
Kugler; 1996. p. 243 – 51.
[38] Carlson BM, Emerick S, Komorowski TE, Rainin EA,
Shepard BM. Extraocular muscle regeneration in
primates: local anesthetic-induced lesions. Ophthal-
mology 1992;99(4):582 – 9.
[39] Hamed LM, Mancuso A. Inferior rectus muscle con-
tracture syndrome after retrobulbar anesthesia. Oph-
thalmology 1991;98:1506 – 12.
[40] Porter JD, Edney DP, McMahon EJ, Burns LA.
Extraocular myotoxicity of the retrobulbar anesthetic
bupivacaine hydrochloride. Invest Ophthalmol Vis Sci
1988;29:163 – 74.
[41] Rainin EA, Carlson BM. Postoperative diplopia and
ptosis: a clinical hypothesis based on the myotoxic-
ity of local anesthetics. Arch Ophthalmol 1985;103:
1337 – 9.
[42] Capó H, Roth E, Johnson T, Muñoz M, Siatkowski
RM. Vertical strabismus after cataract surgery. Oph-
thalmology 1996;103:918 – 21.
[43] Capó H, Guyton DL. Ipsilateral hypertropia after
cataract surgery. Ophthalmology 1996;103:721 – 30.
505
[44] Brown SM, Brooks SE, Mazow ML, Avila CW,
Braverman DE, Greenhaw ST, et al. Cluster of diplopia
cases after periocular anesthesia without hyaluroni-
dase. J Cataract Refract Surg 1999;25:1245 – 9.
[45] Brown SM, Coats DK, Collins MLZ, Underdahl JP.
Second cluster of strabismus cases after periocular
anesthesia without hyaluronidase. J Cataract Refract
Surg 2001;27:1872 – 5.
[46] Jehan FS, Hagan JC, Whittaker TJ, Subramanian M.
Diplopia and ptosis following injection of local
anesthesia without hyaluronidase. J Cataract Refract
Surg 2001;27:1876 – 9.
[47] Rowly SA, Hale JE, Finlay RD. Sub-Tenon’s local
anesthesia: the effect of hyaluronidase. Br J Ophthal-
mol 2000;84:435 – 6.
[48] Lindsey P, Finkelstein D, D’Anna S. Experimental
retinal rotation [ARVO abstract]. Invest Ophthalmol
Vis Sci 1983;24(suppl):242.
[49] Machemer R, Steinhorst UH. Retinal separation,
retinotomy and macular relocation: a surgical approach
for age-related macular degeneration? Graefes Arch
Clin Exp Ophthalmol 1993;231:635 – 41.
[50] Lewis H, Kaiser PK, Lewis S, Estafanous M. Macular
translocation for subfoveal choroidal neovasculariza-
tion in age-related macular degeneration: a prospective
study. Am J Ophthalmol 1999;128:135 – 46.
[51] Pieramici DJ, De Juan E, Fujii GY, Reynolds SM,
Melia M, Humayun MS, et al. Limited inferior macular
translocation for the treatment of subfoveal choroidal
neovascularization secondary to age-related macular
degeneration. Am J Ophthalmol 2000;130:419 – 28.
[52] Buffenn AN, De Juan E, Fujii G, Hunter DG. Diplopia
after limited macular translocation surgery. J Am
Assoc Pediatr Ophthalmol Strabismus 2001;5:388 – 94.
[53] Burgess C, Roper-Hall G, Burde R. Binocular diplopia
associated with subretinal neovascular membranes.
Arch Ophthalmol 1980;98:311 – 7.
[54] Bixenman WW, Joffe L. Binocular diplopia associated
with retinal wrinkling. J Pediatr Ophthalmol Strabis-
mus 1984;21:215 – 9.
[55] Guyton DL. The dragged-fovea diplopia syndrome. In:
Balkan RJ, Ellis Jr GS, Eustis HS, editors. At the
crossings: pediatric ophthalmology and strabismus.
Proceedings of the 52nd Annual Symposium of the
New Orleans Academy of Ophthalmology, New
Orleans, LA, February 2003. The Hague7 Kugler
Publications, 2004. p. 169 – 72.
[56] Maurino V, Kwan A, Khoo B, Gair E, Lee JP. Ocu-
lar motility disturbances after surgery for retinal de-
tachment. J Pediatr Ophthalmol Strabismus 1998;2:
285 – 92.
[57] Guyton DL. Clinical assessment of ocular torsion. Am
Orthoptic J 1983;33:7 – 15.
[58] Mallette RA, Kwon JY, Guyton DL. A technique for
repairing strabismus after scleral buckling surgery.
Am J Ophthalmol 1988;106:364 – 5.
[59] Schwartz PL, Pruett RC. Factors influencing retinal
detachment after removal of buckling elements. Arch
Ophthalmol 1977;95:804 – 7.
506 M.B. Yadarola et al / Ophthalmol Clin N Am 17 (2004) 495 – 506
[60] Flanders M, Wise J. Surgical management of strabis-
mus following scleral buckling procedures. Can J
Ophthalmol 1984;19:17 – 20.
[61] Petitto VB, Buckley EG. Use of botulinum toxin in
strabismus after retinal detachment surgery. Ophthal-
mology 1991;98:509 – 13.
[62] Scott AB. Botulinum treatment of strabismus follow-
ing retinal detachment surgery. Arch Ophthalmol 1990;
108:509 – 10.
[63] Friedberg MA, Spellman FA, Pilkerton AR, Perraut E,
Stephens RF. An alternative technique of local anes-
thesia for vitreoretinal surgery. Arch Ophthalmol 1991;
109:1615 – 6.
[64] Rous MS. Simplified sub-Tenon’s anesthesia: mini-
block with maxiblock effect. J Cataract Refract Surg
1999;25:10 – 5.
[65] Olitsky SE, Juneja RG. Orbital hemorrhage after the
administration of sub-Tenon’s infusion anesthesia.
Ophthalmic Surg Lasers 1997;28:145 – 6.
[66] Bellucci R. Anesthesia for cataract surgery. Curr Opin
Ophthalmol 1999;10:36 – 41.
 Ophthalmol Clin N Am 17 (2004) 507 – 512
Elevated intraocular pressure following vitreoretinal surgery
Anastasios P. Costarides, MD, PhD*, Phil Alabata, DO, Chris Bergstrom, MD
Emory Eye Center, Emory University School of Medicine, 1365B Clifton Road, Glaucoma Section, Atlanta, GA 30322, USA
Elevated intraocular pressure (IOP) is common
following vitreoretinal surgery [1]. Approximately
35% of patients have IOP of at least 30 mm Hg
within 48 hours of pars plana vitrectomy [2]. This rise
in IOP may be the result of an open- or closed-angle
mechanism. Angiogenic, inflammatory, steroid-
induced, and blood-mediated mechanisms may also
be involved. The use of scleral buckling, silicone oil,
or intraocular gas in conjunction with pars plana
vitrectomy produces unique diagnostic and therapeu-
tic challenges in managing the IOP in these patients.
Pars plana vitrectomy
Simple pars plana vitrectomy alone has been
shown to cause significant pressure elevation within
the first 2 hours postoperatively [3]. Several periopera-
tive conditions may increase the risk of acute
pressure elevation following pars plana vitrectomy.
Patients with proliferative vitreoretinopathy under-
going pars plana vitrectomy are five times more likely
to have an acute postoperative pressure elevation than
are those undergoing pars plana vitrectomy for
macular hole repair [2]. Pars plana vitrectomy in
conjunction with placement of a scleral buckle,
intraoperative scatter endophotocoagulation, or per-
formance of pars plana lensectomy also increase the
risk of elevated postoperative pressures. Fibrin for-
* Corresponding author.
E-mail address: a_costarides@emoryhealthcare.org
(A.P. Costarides).
0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.007
mation postoperatively is another risk factor for acute
IOP elevation [2].
In a prospective study by Han and coworkers,
approximately 60% of 222 consecutive pars plana
vitrectomy patients had an IOP increase of at least
5 mm Hg within 48 hours of surgery [2]. Most of
these patients had a secondary open-angle mechanism
accounting for their IOP elevation. In order of
decreasing frequency, these mechanisms included
gas expansion without angle closure, inflammation,
silicone oil without pupillary block, corticosteroid
response, and blood-mediated mechanisms. Angle-
closure mechanisms accounted for elevated IOP in
approximately 20% of these patients. Pupillary block
mediated by intraocular gas, silicone oil, fibrin, or an
intraocular lens was the predominant angle-closure
mechanism. Ciliary body edema and iridocorneal
apposition were also significant mechanisms for
angle closure. Despite the multiple etiologies for
pressure elevation, this study found no significant
difference between preoperative and late postopera-
tive IOP. Nevertheless, a study by Aaberg and Van
Horn [4] found that late postvitrectomy glaucoma
occurred in 26% of 206 eyes owing to neovascular
and open-angle mechanisms. A possible reason for
this difference was the greater number of patients
with diabetic retinopathy.
Most patients with postvitrectomy IOP elevation
are controlled with medical therapy. Rarely, surgical
intervention may be required. The appropriate surgi-
cal procedure depends on the mechanism of IOP
elevation. Han and coworkers showed that 11% of
patients in their study required surgical intervention.
These procedures consisted of anterior chamber
paracentesis, laser iridotomy, laser iridoplasty, and
laser membranectomy. None of the patients required a
ophthalmology.theclinics.com
508 A.P. Costarides et al / Ophthalmol Clin N Am 17 (2004) 507 – 512
trabeculectomy or placement of a glaucoma drainage
implant [2].
Scleral buckling procedures
The rate of detectable angle narrowing following
scleral buckling procedures in nonvitrectomized eyes
has been reported to be as high as 50% in the first
postoperative week [5]. Resolution typically occurs
over the next several weeks [6]; however, angle-
closure glaucoma may occur in as many as 4% of
these patients [7 – 10].
The cause of angle narrowing following buckling
procedures is most likely multifactorial. The predomi-
nant mechanism is detachment of the ciliary body.
Experimental and clinical evidence suggests that
uveal congestion owing to impaired drainage from
the vortex veins is a significant contributor to this
condition [6]. Occlusion of the vortex veins in
nonhuman primates results in ciliary body and
choroidal congestion with shifting of the lens-iris
diaphragm anteriorly and angle closure [11]. Ultra-
sound biomicroscopy has shown that effusions are
often present after buckling procedures even in the
absence of clinically detectable choroidal detachment
or anterior chamber angle narrowing [6]. Pupillary
block may also be present, but it is usually not the
predominant mechanism. Other risk factors for angle-
closure development in patients after scleral buckling
include older patient age [9], high myopia [9], a pre-
existing narrow angle [7], and placement of the
encircling band anterior to the equator [1,7].
In most cases, angle closure resolves spontane-
ously following scleral buckling procedures. Uveal
effusion may take days to weeks to resolve. Therapy
should include cycloplegics, corticosteroids, and
aqueous suppressants. Laser iridoplasty may be
helpful in cases uncontrolled by medical therapy
[1]. Laser iridotomy is rarely required, because
pupillary block is not a predominant mechanism for
angle closure in this setting.
Silicone oil
Silicone oil injection is used for long-term retinal
tamponade in the treatment of complex retinal detach-
ments. An advantage of silicone oil is that it allows the
patient useful vision while providing retinal tam-
ponade without the need for prolonged prone-head
positioning. Like gas bubbles, there is a tendency for
secondary IOP elevation that is often mild and
transient. These pressure elevations are often managed
conservatively with observation or medical therapy.
Nevertheless, vision-threatening pressure elevations
can occur and may require surgical intervention.
Emulsified oil in the anterior chamber, pupillary
block from silicone oil, angle closure without pupil-
lary block, and idiopathic open-angle glaucoma
without silicone oil in the anterior chamber have all
been demonstrated to be potential glaucoma compli-
cations following silicone oil tamponade [12 – 14].
Risk factors for postoperative pressure elevation after
silicone oil injection include a history of glaucoma,
diabetes mellitus, and a high IOP on the first post-
operative day. The most common mechanisms for
postoperative pressure elevation during the first few
weeks after silicone oil injection are a closed inferior
iridectomy or large silicone oil bubbles in the anterior
chamber. Late causes of postoperative pressure ele-
vation are silicone oil emulsification in the anterior
chamber and pre-existing glaucoma [14]. In addition
to glaucoma, hypotony may complicate silicone oil
injection. Risk factors for hypotony include preopera-
tive hypotony and aphakia [14,15].
In a large study of patients undergoing pars plana
vitrectomy and various adjunctive retinal procedures,
silicone oil – related glaucoma without pupillary
block occurred in 3.6% of eyes, and silicone oil –
related pupillary block occurred in 0.9% of eyes [2].
The incidence of silicone oil – related pupillary block
was low owing to the placement of a prophylactic
inferior iridectomy at the time of surgery in pseudo-
phakic and aphakic eyes. Jackson and coworkers [16]
demonstrated that silicone oil could produce signifi-
cant glaucomatous morbidity even in the presence of
a patent inferior iridectomy in phakic and pseudo-
phakic eyes. They examined seven phakic patients
and one pseudophakic patient who presented 1 to
90 days post vitrectomy and silicone oil injection
with IOP ranging from 36 to 70 mm Hg. Treatment of
a blocked iridectomy or creation of an iridotomy with
a neodymium: yttrium-aluminum-garnet (Nd: YAG)
laser provided temporary relief for some patients.
Nevertheless, all of the patients required eventual
removal of silicone oil for recurrent or persistent ele-
vated IOP. In five of the patients, angle closure was
attributed to an observed or potential weakness of the
iris-lens diaphragm. Navas and coworkers [17]
treated seven phakic patients with pupillary block
from silicone oil. These patients presented 1 to
60 days post vitrectomy and silicone oil injection
with IOP ranging from 45 to 70 mm Hg. A large
iridectomy along with viscoelastic washout of resid-
ual silicone oil provided long-term IOP control.
 A.P. Costarides et al / Ophthalmol Clin N Am 17 (2004) 507 – 512
Open-angle glaucoma is a common complication
of intravitreal silicone oil [6]. Silicone oil more
commonly enters the anterior chamber in aphakic
and pseudophakic patients. Less frequently, it can
enter the anterior chamber of phakic patients over time
[18,19]. Using ultrasound biomicroscopy, Avitabile
and coworkers demonstrated a high correlation
between the incidence of a high IOP and the quantity
of emulsified silicone oil in the anterior chamber [12].
They performed ultrasound biomicroscopy on 49 eyes
in 49 patients who underwent vitrectomy with silicone
oil as an adjunctive procedure. Thirteen eyes were
phakic, 23 eyes were pseudophakic, and 13 eyes were
aphakic. Scarce to no oil emulsification was found in
the anterior chamber of the phakic eyes. All of these
patients had normal IOP. Approximately 61% of the
pseudophakic eyes had high IOP that was associated
with anterior chamber oil emulsification. All of the
aphakic eyes had a high IOP with an abundant amount
of anterior chamber oil emulsification.
Management of elevated IOP associated with
silicone oil is multifaceted and depends on the clinical
presentation. Treatment may include topical and oral
antiglaucoma medications alone or in conjunction
with laser or incisional surgical interventions. A laser
peripheral iridotomy, or reopening of a closed pe-
ripheral iridectomy, may be curative for pupillary
block glaucoma in some instances. Removal of sili-
cone oil may be necessary in eyes resistant to other
interventions [6]. Diode cyclophotocoagulation may
be an option in poorly sighted eyes.
Intravitreal gas
Intraocular gas has been used to tamponade retinal
tears and detachments for decades. A survey con-
ducted in 1991, polling members of the Vitreous
Society and Retina Society, showed that 100% of
respondents considered the use of long-acting intra-
ocular gas to be an acceptable standard of care when
clinically indicated, even though its use had not yet
received final approval from the Food and Drug
Administration (FDA) [20]. Elevated IOP (typically
defined as 22 to >30 mm Hg) is a well-known
complication of intraocular gas, and its occurrence
ranges from 6% to 67% depending on the concen-
tration and type of gas used [21 – 23]. The only two
long-acting gases that have received FDA approval
are sulfur hexafluoride (SF6) and perfluoropropane
(C3F8). When injected into the eye undiluted, these
gases rapidly expand to approximately two times and
four times their original volume, respectively, poten-
509
tially leading to severe pressure elevations. Reports of
central retinal artery occlusion [23,24] have occurred
following the injection of intraocular gas. Nonexpan-
sile concentrations have been determined to be 20%
for SF6 and 12% for C3F8 [25,26]; however, even
nonexpansile concentrations have been associated
with elevated pressure [27].
Elevated IOP following intravitreal gas can occur
from open-angle or closed-angle mechanisms. Risk
factors associated with elevated IOP include expansile
gas concentrations, use of C3F8, increasing patient
age, photocoagulation, lensectomy, a circumferential
scleral buckle, and fibrin in the anterior chamber
[2,27]. Following injection, expansion of the intra-
ocular gas bubble occurs as oxygen, carbon dioxide,
and nitrogen from surrounding tissues diffuse into the
bubble more rapidly than the injected gas can diffuse
out. Open-angle pressure elevation occurs when
expansion of the bubble exceeds outflow of intra-
ocular fluid, or when the gas volume exceeds the
capacity of the vitreous cavity without compromise of
the anterior chamber angle [28].
Closed-angle mechanisms comprise gas-induced
pupillary block, anterior displacement of the lens-iris
diaphragm, and iridocorneal apposition. Pupillary
block occurs when expansion of the gas bubble
displaces the lens-iris diaphragm anteriorly, blocking
the pupillary space and resulting in iris bombé and
secondary angle closure. Angle closure can arise
secondary to anterior displacement of the lens-iris
diaphragm without pupillary block. Iridocorneal
apposition tends to occur in aphakic eyes with a
large, long-acting gas bubble and moderate post-
operative fibrin formation [29]. In this mechanism, the
expansile and buoyant forces from the gas bubble
push the iris anteriorly into apposition with the cornea,
closing the angle.
Preoperative gonioscopy should be performed be-
fore injection of intravitreal gas, especially in patients
with glaucoma or narrow angles. Patients with
narrow angles, significant peripheral anterior syne-
chiae (PAS), or angle neovascularization should be
observed closely for subsequent pressure elevation.
Intraocular pressure measurement in gas-filled
eyes is best assessed using a low-displacement
tonometer, such as the Goldmann applanation to-
nometer or Perkins tonometer [30,31]. The Tono-Pen
is increasingly used to measure postoperative IOP. Its
light weight and portability make it convenient for
checking postoperative IOP while patients are in the
hospital. It is effective for patients with corneal
epithelial irregularities or lid edema, often seen after
surgery. Studies investigating the accuracy of Tono-
Pen (Mentor Inc, Norwell, Massachusetts) readings in
510 A.P. Costarides et al / Ophthalmol Clin N Am 17 (2004) 507 – 512
gas-filled eyes have shown good correlation with
manometric readings in normal ranges; however,
underestimations occur in pressures above 30 mm
Hg [32]. Pneumatic tonometers have also been shown
to underestimate IOP at high pressure levels [32,33].
High displacement tonometers, which are seldom
used anymore, significantly underestimate IOP owing
to compressibility of the intraocular gas [31,33].
Elevated pressure after the use of intraocular gas
usually occurs within the first 24 hours. Simple pars
plana vitrectomy alone has been shown to cause
significant pressure elevation within the first 2 hours
postoperatively [3]. Mild elevations usually respond
to topical aqueous suppressants or oral carbonic
anhydrase inhibitors within 24 to 72 hours. Prophy-
lactic treatment with carbonic anhydrase inhibitors
intraoperatively has been shown to decrease IOP in
the early postoperative period but shows little
protective effect on the first postoperative day [34].
Similarly, the use of topical aqueous suppressants at
the end of surgery has been shown to prevent
pressure elevation within the first 4 to 6 hours
postoperatively [35]. Severe pressure elevations
unresponsive to medications or threatening ocular
perfusion may require aspiration of a portion of the
intraocular gas to normalize pressure. Patients in
pupillary block require an inferiorly placed laser
iridotomy, whereas iridocorneal apposition should be
treated promptly with an anterior chamber-deepening
procedure before permanent adhesions occur. This
procedure can often be accomplished at the slit lamp
using sodium hyaluronate on a 30-gauge cannula to
separate the iris from the endothelium.
Glaucoma is a well-known complication of intra-
ocular gas, and different etiologies exist. Careful
preoperative evaluation can help identify patients at
risk for elevated pressure following intravitreal gas
injection. Early recognition of the mechanism in-
volved leads to appropriate management and de-
creased morbidity.
Surgical management
Most patients with postoperative pressure eleva-
tions are managed medically. Ophthalmologists are
confronted with significant challenges when surgical
intervention is warranted. Scarring and recession of
the conjunctiva makes standard filtration surgery less
attractive. Standard outflow-enhancing procedures in
the presence of silicone oil or intraocular gas may
allow egression of these compounds into the subcon-
junctival space [36]. The presence of a scleral buckle
makes placement of a glaucoma drainage implant
more difficult. Cyclodestructive procedures are gen-
erally less attractive owing to the unpredictability of
the outcome and risk of visual loss [1].
Despite these limitations, surgical intervention is
often successful. Filtration surgery with an antime-
tabolite has been reported to be successful [5,37] in
approximately 80% of cases. This outcome parallels
the success rates of filtration surgery with antime-
tabolites in other high-risk eyes [5]. Retinal surgeons
should be encouraged to close peritomy wounds
carefully at the limbus so that recession of the con-
junctiva does not take place, thereby increasing the
likelihood of successful filtration surgery.
Several case series describe the use of glaucoma
drainage implants in eyes with pre-existing scleral
buckles. In a study by Smith and coworkers [38],
modified Baerveldt or Krupin-Denver aqueous tube
shunts were used in eyes with encircling bands. In
seven eyes, the valved Krupin-Denver tube without
the plate was inserted into the encapsulated band to
prevent postoperative hypotony. In four other eyes,
trimmed 200 or 250 mm2 Baerveldt implants were
inserted underneath the encircling band. Success was
defined as an IOP of 21 mm Hg with or without
medications, no loss of visual acuity, and no need for
further glaucoma surgery. Life-table success rates
were 83% and 72% at 1 and 2 years, respectively. Two
of the eyes with the modified Krupin-Denver shunt
required surgical revision owing to occlusion of the
tube distally. In the study by Scott and coworkers [39],
Baerveldt implants, 250 or 350 mm2, were placed
over or behind the encircling band in 16 eyes. An
attempt was made to excise the capsule over the band
so that it would allow for contiguous encapsulation
between the implant and the encircling element. IOP
was controlled (>5 mm Hg and 21 mm Hg) in all of
the eyes with or without medications, with follow-up
ranging from 19.1 to 45.5 months. Complications
included choroidal effusions (31%) and hyphema
(25%), which resolved spontaneously in all patients
within 2 weeks; vitreous hemorrhage (16%), which
resolved within 1 month; and recurrent retinal detach-
ment (6%), which required surgical repair. Persistent
hypotony, tube occlusion, diplopia, implant migration
or exposure, and scleral buckle migration and
exposure did not occur in this case series.
Surgical management of secondary glaucoma after
silicone injection for complex retinal detachment may
be successful. Budenz et al studied 43 eyes in
43 patients over a 9-year period on which incisional
surgery was performed for secondary glaucoma after
pars plana vitrectomy with silicone oil injection [12].
The cumulative success (IOP  21 mm Hg and
 A.P. Costarides et al / Ophthalmol Clin N Am 17 (2004) 507 – 512
5 mm Hg with or without medication but without
surgical reoperation for glaucoma) was 69%, 60%,
56%, and 48% at 6, 12, 24, and 36 months, re-
spectively. Removal of silicone oil alone (32 patients),
silicone oil removal in conjunction with glaucoma
surgery (8 patients), and glaucoma surgery alone
(3 patients) were effective in reducing IOP. Thirty-
four percent of the patients who underwent silicone oil
removal alone failed treatment because of elevated
(>21 mm Hg) intraocular pressure. Five of these
patients subsequently underwent glaucoma drainage
implant placement. IOP was eventually controlled in
80% of these patients. One patient who underwent
silicone oil removal alone failed treatment because of
persistent hypotony. Thirty-eight percent of patients
who underwent silicone oil removal and glaucoma
surgery simultaneously failed treatment because of
hypotony. One of the three patients who underwent
glaucoma surgery alone also failed treatment because
of persistent hypotony. It was concluded that patients
who underwent silicone oil removal alone were more
likely to have a persistent elevation in IOP requiring
further glaucoma surgery, whereas those who under-
went concurrent silicone oil removal and glaucoma
surgery were more likely to have hypotony.
Transscleral cyclophotocoagulation has been used
to treat eyes with persistently elevated IOP following
vitrectomy with silicone oil injection. This treatment
option is of greater use when the risk of redetachment
with silicone oil removal is greater than the risks
associated with cyclophotocoagulation. Successful
pressure control has been achieved, but the number
of cases reported has been small. In one series
including 11 eyes in 11 patients with intravitreal
silicone oil and medically uncontrolled IOP, trans-
scleral diode cyclophotocoagulation was a qualified
success (IOP <21 mm Hg with or without medicine)
in 82% of patients at 1 year [40]. Caution is
warranted when this procedure is used, because
hypotony may result from overly aggressive treat-
ment, whereas inadequate treatment may lead to a
persistently elevated IOP. The therapeutic window is
narrow, because these eyes typically have a limited
outflow facility [41]. Aggressive treatment may be
necessary to lower the IOP. Aggressive intervention
predisposes these eyes to significant pain, inflamma-
tion, macular edema, and phthisis.
Summary
Elevated IOP is a common occurrence following
vitreoretinal surgery. Open- and closed-angle mecha-
511
nisms may be responsible. Understanding the etiology
of the pressure elevation is vital to control its level
effectively. Fortunately, most pressure elevations are
transient and controlled medically. The use of intra-
vitreal silicone oil and gas in conjunction with pars
plana vitrectomy increases the likelihood of post-
operative pressure spikes. Removal of these com-
pounds from the eye may be necessary to control eye
pressure. Surgical options, including standard filtra-
tion surgery, glaucoma drainage implants, and cyclo-
destructive procedures, may be successful in lowering
IOP in these complicated eyes but may be associated
with significant ocular morbidity. Treatment options
should balance the risk of the intervention with the
likelihood of success and should be tailored to the
needs of the individual patient.
References
[1] Gedde SJ. Management of glaucoma after retinal de-
tachment surgery. Curr Opin Ophthalmol 2002;13(2):
103 – 9.
[2] Han DP, Lewis H, Lambrou FH, Mieler WF. Mecha-
nisms of intraocular pressure elevation after pars plana
vitrectomy. Ophthalmology 1989;96:1357 – 62.
[3] Desai UR, Alhalel AA, Schiffman RM, Campen TJ,
Sundar G, Muhich A. Intraocular pressure elevation
after simple pars plana vitrectomy. Ophthalmology
1997;104:781 – 6.
[4] Aaberg TM, Van Horn DL. Late complications of
pars plana vitreous surgery. Ophthalmology 1978;85:
126 – 40.
[5] Hartley RE, Marsh RJ. Anterior depth changes after
retinal detachment. Br J Ophthalmol 1973;57:546 – 50.
[6] Wand M. Glaucoma after vitreoretinal surgery. In: Ep-
stein DL, Allingham RR, Schuman JS, editors. Chan-
dler and Grant’s glaucoma. 4th edition. Baltimore7
William and Wilkins; 1997. p. 327 – 33.
[7] Sebestyen JG, Schepens CL, Rosenthal ML. Retinal
detachment and glaucoma. I. Tonometric and gonio-
scopic study of 160 cases. Arch Ophthalmol 1962;67:
736 – 45.
[8] Smith TR. Acute glaucoma developing after scleral
buckling procedures. Am J Ophthalmol 1967;63:
1807 – 8.
[9] Krieger AE, Hodgkinson BJ, Frederick AR, Smith TR.
The results of retinal detachment surgery. Arch
Ophthalmol 1971;86:385 – 94.
[10] Perez RN, Phelps CD, Burton TC. Angle closure
glaucoma following scleral buckling procedures. Trans
Am Acad Ophthalmol Otolaryngol 1976;81:247 – 52.
[11] Hayreh SS, Baines JAB. Occlusion of the vortex
veins: an experimental study. Br J Ophthalmol 1973;
57:217 – 38.
[12] Avitabile T, Bonfiglio V, Cicero A, Torrisi B, Reibaldi
A. Correlation between quantity of silicone oil emul-
512 A.P. Costarides et al / Ophthalmol Clin N Am 17 (2004) 507 – 512
sified in the anterior chamber and high pressure in
vitrectomized eyes. Retina 2002;22:443 – 8.
[13] Budenz DL, Taba KE, Feuer WJ, Eliezer R, Cousins S,
Henderer J, et al. Surgical management of secondary
glaucoma after pars plana vitrectomy and silicone oil
injection for complex retinal detachment. Ophthalmol-
ogy 2001;108:1628 – 32.
[14] Jonas JB, Knorr HLJ, Rank RM, Budde WM. In-
traocular pressure and silicone oil endotamponade.
J Glaucoma 2001;10:102 – 8.
[15] Henderer JD, Budenz DL, Flynn Jr HW, Schiffman JC,
William J, Murray TG. Elevated intraocular pressure
and hypotony following silicone oil retinal tamponade
for complex retinal detachment: incidence and risk
factors. Arch Ophthalmol 1999;117:189 – 95.
[16] Jackson TL, Thiagarajan M, Murthy R, Snead MP,
Wong D, Williamson TH. Pupil block glaucoma in
phakic and pseudophakic patients after vitrectomy with
silicone oil injection. Am J Ophthalmol 2001;132:
414 – 6.
[17] Navas F, Boyer DS, Thomas EL, Novak RL, Chu TG,
Gallemore RP. Management of pupillary block glau-
coma in phakic patients after vitrectomy with silicone
oil injection. Am J Ophthalmol 2002;134:634 – 5.
[18] Valone J, McCarthy M. Emulsified anterior chamber
silicone oil and glaucoma. Ophthalmology 1994;101:
1908 – 12.
[19] Chan C, Okun E. The question of ocular tolerance to
intravitreal liquid silicone: a long term analysis. Oph-
thalmology 1986;93:651 – 60.
[20] Ai E, Gardner TW. Current patterns of intraocular gas
use in North America. Arch Ophthalmol 1993;111:
331 – 2.
[21] The Silicone Study Group. Vitrectomy with silicone
oil or perfluoropropane gas in eyes with severe pro-
liferative vitreoretinopathy: results of a randomized
clinical trial. Silicone Study Report 2. Arch Ophthal-
mol 1992;110:780 – 92.
[22] The Silicone Study Group. Vitrectomy with silicone
oil or sulfur hexafluoride gas in eyes with severe
proliferative vitreoretinopathy: results of a randomized
clinical trial. Silicone Study Report 1. Arch Ophthal-
mol 1992;110:770 – 9.
[23] Abrams GW, Swanson DE, Sabates WI, Goldman AI.
The results of sulfur hexafluoride gas in vitreous sur-
gery. Am J Ophthalmol 1982;94:165 – 71.
[24] Abe T, Nakajima H, Ishikawa M, Sakuragi S. Intra-
ocular pressure during pneumatic retinopexy. Ophthal-
mic Surg Lasers 1998;29:391 – 6.
[25] Fineberg E, Machemer R, Sullivan P, Norton EW,
Hamasaki D, Anderson D. Sulfur hexafluoride in owl
monkey vitreous cavity. Am J Ophthalmol 1975;79:
67 – 76.
[26] Peters MA, Abrams GW, Hamilton LH, Burke JM,
Schrieber TM. The nonexpansile, equilibrated concen-
tration of perfluoropropane gas in the eye. Am J
Ophthalmol 1985;100:831 – 9.
[27] Chen PP, Thompson JT. Risk factors for elevated
intraocular pressure after the use of intraocular gases in
vitreoretinal surgery. Ophthalmic Surg Lasers 1997;28:
37 – 42.
[28] Chang S, Lincoff HA, Coleman DJ, Fuchs W, Farber
ME. Perfluorocarbon gases in vitreous surgery. Oph-
thalmology 1985;92:651 – 6.
[29] Han DP, Lewis H, Williams GA. Management of
complete iridocorneal apposition after vitrectomy. Am
J Ophthalmol 1987;103:108 – 9.
[30] Hines MW, Jost BF, Fogelman KL. Oculab Tono-Pen,
Goldmann applanation tonometry, and pneumatic
tonometry for intraocular pressure assessment in gas-
filled eyes. Am J Ophthalmol 1988;106:174 – 9.
[31] Poliner LS, Schoch LH. Intraocular pressure assess-
ment in gas-filled eyes following vitrectomy. Arch
Ophthalmol 1987;105:200 – 2.
[32] Lim JI, Blair NP, Higginbotham EJ, Farber MD,
Shaw WE, Garretson BR. Assessment of intraocular
pressure in vitrectomized gas-containing eyes: a clini-
cal and manometric comparison of the Tono-Pen to
the pneumotonometer. Arch Ophthalmol 1990;
108:684 – 8.
[33] Del Priore LV, Michels RG, Nunez MA, Smiddy W,
Glaser BM, de Bustros S. Intraocular pressure mea-
surement after pars plana vitrectomy. Ophthalmology
1989;96:1353 – 6.
[34] Ruby AJ, Grand MG, Williams D, Thomas MA.
Intraoperative acetazolamide in the prevention of intra-
ocular pressure rise after pars plana vitrectomy with
fluid-gas exchange. Retina 1999;19:185 – 7.
[35] Mittra RA, Pollack JS, Dev S, Han DP, Mieler WF,
Pulido JS, et al. The use of topical aqueous suppres-
sants in the prevention of postoperative intraocular
pressure elevation after pars plana vitrectomy with
long-acting gas tamponade. Ophthalmology 2000;107:
588 – 92.
[36] Nazemi PP, Chang LP, Varma R, Burnstine MA.
Migration of intraocular silicone oil into the subcon-
junctival space and orbit through an Ahmed glaucoma
valve. Am J Ophthalmol 2001;132(6):929 – 31.
[37] Ophir A, Ticho U. Trabeculectomy with 5-fluorouracil
subsequent to circular buckling operation and cataract
extraction. Ann Ophthalmol 1992;24:386 – 90.
[38] Smith MF, Doyle JW, Fannous MM. Modified aqueous
drainage implants in the treatment of complicated
glaucomas in eyes with pre-existing episcleral bands.
Ophthalmology 1998;105:2237 – 42.
[39] Scott IU, Gedde SJ, Budenz DL, et al. Baerveldt
drainage implants in eyes with pre-existing scleral
buckle. Arch Ophthalmol 2000;118:1509 – 13.
[40] Han SK, Park KH, Kim DM, et al. Effect of diode laser
trans-scleral cyclophotocoagulation in the management
of glaucoma after intravitreal silicone injection for
complicated retinal detachments. Br J Ophthalmol
1999;83:713 – 7.
[41] Schuman JL. Cyclodestruction. In: Epstein DL, Alling-
ham RR, Schuman JL, editors. Chandler and Grant’s
glaucoma. 4th edition. Baltimore7 William and Wil-
liams; 1997. p. 484 – 94.
 Ophthalmol Clin N Am 17 (2004) 513 – 520
Corneal and conjunctival changes after posterior
segment surgery
J. Bradley Randleman, MD*, Susanne M. Hewitt, MD, C. Diane Song, MD
Department of Ophthalmology, Emory University, 1365B Clifton Road NE, Suite 4500, Atlanta, GA 30322, USA
Vitreoretinal surgical techniques have evolved
markedly over the past 40 years. With each advance,
the indications for surgery have expanded to younger
patients and relatively healthier eyes with greater
visual potential; therefore, complications of these
procedures, especially subtle changes to other parts of
the eye, become more critical to identify, and one
should attempt to prevent or minimize their occur-
rence. The cornea is particularly susceptible to
intraoperative compromise after vitreoretinal surgery.
Corneal damage during pars plana vitrectomy can
manifest as damage to the epithelium or to the
endothelium. Intentional removal of the epithelium,
when the need for improved intraoperative visual-
ization arises, or incidental trauma may complicate
postoperative recovery. Damage to the endothelium
can occur as a primary mechanical process or
secondary to chemical toxicity. Preoperative lens
status and the specific procedure undertaken with
regard to intraoperative manipulation of the crystal-
line lens directly correlate with corneal endothelial
damage. The use of adjunctive substances during the
vitrectomy procedure, such as various irrigating
solutions or multiple types of intraocular gasses with
varying degrees of corneal toxicity, can further
compromise the endothelium.
This article was supported in part by Research to Prevent
Blindness, New York, New York, and by National Institutes
of Health core grant P30 EYO6360, Bethesda, Maryland.
* Corresponding author. 1365 B Clifton Road NE, Suite
4500, Atlanta, GA 30322.
E-mail address: Jrandle@emory.edu (J.B. Randleman).
0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.001
Although scleral buckling procedures generally
induce less direct trauma to the cornea, significant
problems can arise as a result of buckle placement to
the ocular surface and conjunctiva, including infec-
tion and other reactions related to the location and
potential migration of the buckle elements.
Corneal epithelium compromise after pars plana
vitrectomy
Damage to the corneal epithelium has been
recognized as one of the most frequent early post-
operative complications of bclosedQ pars plana
vitrectomy procedures [1 – 5]. Intraoperative epithe-
lial defects can be classified into two general forms—
those that occur incidentally and those that are
created to improve visualization. Intraoperative epi-
thelial defects result in delayed visual recovery and
predispose patients to infection, persistent epithelial
irregularity, persistent corneal edema, and corneal
scarring, and can result in chronic nonhealing or
recurrent epithelial defects postoperatively [4]. A
reduction in the frequency of or avoidance of
intended and unintended defects has great potential
benefit to improve postoperative outcomes.
Reported risk factors for the development of
intraoperative epithelial defects include the duration
of the surgical procedure, diabetes mellitus, previous
vitrectomy, intraoperative debridement, and the type
of viewing lens system used for surgery [2 – 4,6 – 9].
The diabetic cornea has been shown subjectively
and objectively to behave differently during surgery.
Mandelcorn and colleagues reported a subjective
poor adhesion of corneal epithelium that manifested
ophthalmology.theclinics.com
514 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 513 – 520
as ease of epithelial removal [1]. Brightbill and
colleagues reported serious corneal complications in
34% of their patients; all of the complications
occurred in diabetic patients [4]. Perry and colleagues
reported an overall incidence of corneal complica-
tions of 43% compared with an incidence of 83%
among diabetic patients [3]. In a series reported by
Foulks and colleagues, 93% of corneal complications
occurred in diabetic patients, and 78% of these
complications were epithelial [2]. A later report by
Chung and colleagues demonstrated a reduction in
the overall complication rate; however, diabetes was
still a significant risk factor, with 19.9% of diabetic
patients and 10.4% of nondiabetic patients experi-
encing corneal complications [7]. Diabetic corneas
predispose to epithelial defects not only directly but
also indirectly, because these corneas become edema-
tous and compromise the surgeon’s view, necessitat-
ing epithelial debridement at a greater rate [2 – 4].
The unique properties of the diabetic cornea
warrant special consideration, because complications
from diabetes have been and continue to represent
one of the most common indications for pars plana
vitrectomy [10 – 13]. Many reports have demonstrated
that diabetes is the most significant predictor for
intraoperative epithelial defects [2 – 4,7,8]. The dia-
betic corneal epithelium has abnormal accumulation
of glycogen in the basal cell layer with concomitant
localized edema [14 – 17]. Induction of the sorbitol
pathway in the diabetic cornea has also been
implicated in epithelial adhesion abnormalities
[2,14,16,17]. Kenyon and colleagues [18] demon-
strated morphologic abnormalities in the basement
membrane of diabetic corneas. These biochemical
and structural abnormalities in the diabetic cornea
manifest clinically as decreased sensation, defective
epithelial adhesions, and decreased re-epithelializa-
tion [14], all of which increase the rate and potential
severity of intraoperative epithelial defect formation.
Interestingly, studies have failed to demonstrate
any correlation between patient age or gender and the
rate of epithelial defect formation [3,7]. One might
expect to see an association with age, because
basement membrane abnormalities in diabetic cor-
neas predispose to defect formation, and age-related
changes in the basement membrane of otherwise
normal corneas have been demonstrated [19]. The
occurrence of epithelial defects has been demon-
strated to increase with age in patients undergoing
laser in situ keratomileusis (LASIK) [20 – 22].
The viewing lens systems used during vitrectomy
have been related to the epithelial defect rate [8,23].
Virata and colleagues found that hand-held infusion
lenses were associated with the highest epithelial
defect rate, followed by sew-on contact lenses, with
noncontact lenses having the lowest epithelial defect
rate [8]. Friberg and colleagues also found a
significantly higher epithelial defect rate when irri-
gating contact lens systems were used; however,
they found no difference between sew-on and non-
contact lens systems [23]. Because, by definition,
noncontact lenses cannot directly cause epithelial
defects or affect corneal clarity, epithelial debride-
ment when using these lens systems must be
necessitated by a non – lens-related reduction in
corneal clarity.
From its inception, corneal clarity has been
paramount to the success of pars plana vitrectomy
procedures; therefore, the epithelium is often
removed intraoperatively when epithelial edema
reduces posterior segment visualization. Risk factors
for intraoperative epithelial edema include diabetes,
increased intraocular pressure, mechanical trauma,
increased intraocular inflammatory mediators, and
corneal toxicity from agents used on the corneal
surface and intraocularly [1 – 3,7,9,24]. Although the
rate of epithelial debridement, as high as 50% to 60%
in early reports [1 – 4], has been reduced [7], a recent
survey by Friberg and colleagues found that epithelial
debridement was still frequently performed to pro-
vide good visualization [23]. In their 2003 study, the
average debridement rate was 17.4% among the
surgeons surveyed, with a range of 0% to 90% for
cases in which vitrectomy was performed for
complications from diabetes [23].
Garcia-Valenzuela and colleagues have reported
strategies to reduce epithelial debridement rates
[24]. The use of GenTeal gel instead of Goniosol
as a surface lubricant reduced the epithelial debride-
ment rate from 54% to 14%. This difference was
related to the preservatives found in Goniosol,
especially benzalkonium chloride, which has proven
corneal toxicity.
In addition to epithelial defects, ocular surface and
tear film disruption can occur, occasionally resulting
in the formation of corneal dellen [25]. Insler and
colleagues [25] reported a case in which a desceme-
tocele resulted in an area of persistent corneal dellen
that occurred after vitrectomy.
Great care should be taken to avoid intentional
or incidental epithelial defects intraoperatively,
because they may significantly delay visual recovery
and predispose to other postoperative corneal com-
plications, such as chronic epithelial defects, pain,
bacterial keratitis, and delayed visual recovery.
Special attention should be given to the diabetic
cornea, which, because of structural and biochemical
alterations, is uniquely susceptible to epithelial
 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 513 – 520
defects. Diabetic patients may also have delayed
healing of the cornea and are at increased risk for
postoperative infections. The use of noncontact
or sew-on viewing lens systems may reduce epithelial
defect rates, and gentler surface lubricants may
be beneficial. Other ocular surface disruptions,
although less common, can occur with occasionally
serious sequelae.
Corneal endothelium compromise after pars plana
vitrectomy
When compared with intraoperative epithelial
damage, induced changes to the corneal endothelium
may manifest more subtly, with less acute but
potentially more significant long-term manifestations.
The cause of postoperative endothelial cell dysfunc-
tion is multifactorial. Intraocular irrigating solutions,
preoperative lens status, anterior capsule integrity,
and the use of adjunctive intraocular gas mixtures are
contributing factors.
Early studies reported significant proportions of
patients with corneal edema following vitrectomy
[1,3,4,10,26]. This edema was often severe, resulting
in symptomatic bullous keratopathy and necessitating
penetrating keratoplasty in some instances [4,26].
These early reports almost exclusively represented
severely compromised diabetic eyes that required
extensive surgery, and they often predated the use of
more protective intraocular irrigating solutions. Work
by McCarey and colleagues [27 – 31] and Edelhauser
and colleagues [27 – 31] demonstrated the potential
deleterious effects of intraocular irrigating solutions
on the corneal endothelium. Subsequently, more
protective irrigating solutions were developed using
bicarbonate as a buffer, and these solutions have
been shown to reduce postoperative endothelial cell
dysfunction after vitrectomy [32,33].
The preoperative lens status also influences
endothelial cell loss. With the natural crystalline lens
in place, endothelial cell loss is minimal. Buettner
and Bourne [34] reported a less than 3% rate of cell
loss after pars plana vitrectomy in phakic patients.
Other investigators have reported similar results
[35,36] and hypothesized that an intact crystalline
lens or anterior capsule provides a barrier to flow,
thereby limiting mechanical endothelial trauma [36].
Mittl and colleagues [37] demonstrated that pseu-
dophakia provides a similar benefit. In contrast, in
aphakic eyes, endothelial cell loss can be significant.
Increased cell loss rates, often greater than 12%, have
been reported in aphakic eyes [7,35,38]. Concomitant
lensectomy also increases endothelial cell loss rates
515
[7,35,38]. The barrier function of the lens and
anterior capsule seems to have an even greater impact
when adjunctive intraocular gas mixtures are used.
Intraocular gasses are used in vitreoretinal surgery
to provide a tamponade force to facilitate retinocho-
roidal adhesion formation after retinal detachment
repair. Expansile gasses are also used for pneumatic
retinopexy procedures. Substances used for this pur-
pose in a temporary or longer-term fashion include air,
sulfur hexafluoride (SF6), perfluoropropane (C3F8),
and perfluorocarbon liquids.
Early work in rabbits found that the effects of
SF6 on the corneal endothelium were similar to the
effect of air [39]. Similar animal studies consistently
demonstrated equivalent endothelial toxicities of
SF6, C3F8, and air, all of which exhibited greater
toxicity than balanced salt solutions or aqueous
humor [40,41]. Two clinical studies have demon-
strated moderate toxic effects of fluid-gas exchange
with SF6 or C3F8 [35,38]. In both studies, the toxic
effects were noted when gas exchange was performed
in the presence of an open anterior capsule, allowing
direct access of gas to the anterior chamber.
More significant toxicity has been reported with
the use of perfluorocarbons. Han and colleagues
[42] demonstrated in a rabbit model that when direct
corneal contact was avoided, perfluorocarbons ex-
hibited minimal effects on the cornea. In contrast,
other clinical studies have demonstrated the potential
for significant toxicity limiting corneal function when
retained perfluorocarbon substances are left in the
anterior chamber for as little as 1 month [43,44].
The corneal endothelium can be compromised
significantly during pars plana vitrectomy. This
damage may be clinically apparent only transiently.
Nevertheless, it may have significant long-term
ramifications, especially in younger patients and those
with compromised corneas preoperatively. The use of
high-quality intraocular irrigating solutions is critical
in limiting this damage. The presence of the natural
crystalline or an intraocular lens seems protective,
whereas aphakia exacerbates endothelial cell loss. An
intact anterior capsule also seems protective, espe-
cially when adjunctive intraocular gas mixtures are
employed. The perfluorocarbons seem to have the
greatest potential for significant toxicity.
Special circumstances
Silicone oil use during vitrectomy
The corneal changes induced by silicone oil
deserve special attention owing to the widespread
516 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 513 – 520
use of this substance and the range of corneal
changes that can be encountered. Silicone oil was
introduced in 1962 as a method of replacing vitreous
humor to achieve tamponade of retinal tears and de-
tachments [45]. Silicone oil was chosen as a tam-
ponade owing to its high surface tension with water,
transparency, stability, and relatively low toxicity
[46]; however, corneal decompensation can occur
with silicone oil use, more commonly in aphakic
eyes [47].
Corneal compromise after silicone oil use
Corneal complications associated with silicone oil
include band keratopathy and other epithelial irregu-
larities, persistent stromal edema, vascularization,
and endothelial cell damage [47,48]. When silicone
oil is used, the rate of corneal decompensation is
increased relative to that with pars plana vitrectomy
alone. When ten eyes with previous silicone oil
injection and inferior iridotomy were evaluated in one
study, the endothelial cell count averaged 70% lower
than in the unoperated fellow eye [49]. In that study,
silicone oil had been placed 2 to 48 months before the
endothelial counts were obtained, and the operated
eye endothelial counts ranged from undetectable to
1750 cells per square millimeter [49]. Seven of these
eyes had gross corneal edema or band keratopathy.
Several other studies have corroborated these find-
ings [50,51].
Preoperative and postoperative risk factors for the
development of corneal abnormalities with silicone
oil use noted in the Silicone Study Report 7 [48]
included aphakia or pseudophakia, iris neovascular-
ization, reoperation, corneal touch with silicone oil,
no fluid-gas exchange, and anterior chamber cells.
The presence of silicone oil in the anterior chamber,
more common in aphakic or pseudophakic eyes, is
associated with a higher rate of corneal endothelial
decompensation owing to increased contact of oil
with the endothelium (relative risk, 3.1 at 24 months
in the Silicone Study). It is postulated that the sili-
cone oil creates a barrier between the endothelial cell
and the nutritive aqueous humor; however, in the
Silicone Study, the rate of keratopathy was equal with
the use of C3F8 gas or silicone oil and was higher
when SF6 gas was used [48]. The eyes with corneal
abnormalities in this study had poor visual acuity,
with 53% of eyes with corneal abnormalities having
visual acuities less than 5/200 as compared with 26%
of those without corneal abnormalities. The 2-year
incidence rate of corneal decompensation with
silicone oil use was 26% in this study, despite the
placement of inferior iridotomies in each case. This
risk is also increased in eyes that receive silicone oil
after a previous corneal refractive procedure such as
photorefractive keratectomy [52].
On specular microscopy, although the endothelial
cells appear generally normal, there are numerous
minute bubbles on the endothelium, representing
emulsified silicone oil [46]. The amount of emulsi-
fied oil in the anterior chamber is related to the time
since silicone oil injection, and bubbles can be noted
in the anterior chamber, infiltrating the iris stroma
and the superior angle, and in the posterior chamber.
Near these bubbles, some abnormal endothelial cells
with pleomorphic changes have been noted. In an ani-
mal study, silicone oil injection resulted in a 40% re-
duction in endothelial cell counts [47]. On electron
microscopy of cats with silicone oil injected into the
eye, a fine filamentous material was found to lie on
top of enlarged endothelial cells.
In humans, diffuse edema and band keratopathy
are noted clinically; histopathologically, subepithelial
bullae are noted near an intact Bowman’s layer [53].
In some cases, the underlying stroma has increased
cellularity. Endothelial cell counts are markedly
decreased. In eight of ten eyes, specimens demon-
strated a thickened layer on the posterior corneal
surface, either collagenous and acellular or associ-
ated with fibroblasts [53]. On electron microscopy
of these eyes, the endothelial cell borders were ir-
regular, and there were fibrous layers posterior to
Descemet’s membrane.
Wherever possible, silicone oil should be removed
to prevent or slow the development of the complica-
tions mentioned previously. If necessitated by band
keratopathy or corneal decompensation, penetrating
keratoplasty can be performed simultaneously. In the
Silicone Study, five eyes randomized to receive
silicone oil and four randomized to receive gas
underwent penetrating keratoplasty, and the grafts
remained clear in five of these eyes at 2 years [48]. In
another study by Karel et al [54], 13 patients
underwent penetrating keratoplasty owing to silicone
oil keratopathy. Ten of these eyes had the oil removed
before or during the keratoplasty, and six of the grafts
remained clear during the follow-up period (average,
32 months). In the three eyes in which silicone oil
remained after penetrating keratoplasty, all grafts
failed. The factors contributing to failure in this
group included contact of the graft with silicone oil.
The failure occurred within 2 to 8 months.
Prevention of complications related to silicone oil
The risk of keratopathy may be reduced by
avoiding silicone oil overfill, which can be pre-
vented by filling the anterior chamber with air
before injecting the silicone oil. Keeping the intra-
 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 513 – 520
ocular pressure under control also minimizes the risk
of keratopathy.
To prevent complications when silicone oil is
used, supine positioning of the patient with aphakia
should be avoided to reduce the appearance of oil
in the anterior chamber. Similarly, an inferior iridec-
tomy is important to allow aqueous to flow from
the posterior to anterior chamber if the pupil is
blocked by the oil bubble, reducing the potential
contact of silicone oil with the endothelium [55].
This iridotomy may close owing to scarring, neces-
sitating a repeat laser or surgical procedure to main-
tain its patency. If silicone is discovered in the
anterior chamber, surgical removal is often indicated.
In these cases, repeat retinal examination often re-
veals retinal detachment owing to inadequate tam-
ponade after oil escapes into the anterior chamber. If
the oil has emulsified, and bubbles are noted in
the anterior chamber, it should be removed if the
retina has stabilized and replaced if the retina has
not stabilized.
LASIK flap dehiscence after vitreoretinal surgery
LASIK flap dislocation during vitrectomy [56] or
scleral buckling procedures [57] has been reported.
LASIK flaps most likely never completely heal;
therefore, these patients are at lifelong risk for
traumatic dislocation during surgical procedures. As
the prevalence of LASIK surgery increases, vitreo-
retinal surgeons will increasingly treat corneas
with LASIK flaps, and great care must be taken to
avoid dislocations that have the potential to limit
postoperative visual acuity if not recognized and
treated effectively.
Corneal burns after posterior segment laser therapy
Corneal burns have been reported during intra-
operative laser indirect ophthalmoscope treatment
[58]. Corneal stromal opacities were noted during
treatment. Increased corneal epithelial edema was
thought to facilitate laser uptake by the cornea.
Corneal and conjunctival complications related to
scleral buckling procedures
The use of scleral buckling elements for retinal
detachment surgery is associated with specific poten-
tial corneal and ocular surface complications. Ele-
ments may disrupt the ocular surface through
concomitant extrusion and infection, with a range of
symptoms and presentations related to element type.
517
Extrusion of silicone buckling elements has
been recognized as a potentially frequent complica-
tion of retinal detachment repair [59 – 65]. Silicone
sponges necessitate removal more frequently than
do silicone rubber explants [66,67]. Extruding ele-
ments can cause local foreign body symptoms and
pain and disruption of adjacent ocular tissues,
resulting in strabismus [68,69] or corneal deforma-
tion [70]. They can also provide a nidus for infection
[59 – 65,71 – 74].
Reported infection rates of scleral buckle elements
range from 0.1% [73] to 24% [59]; however, the
actual rate is more likely closer to 1% to 2%. The
organisms most frequently isolated include Staphy-
lococcus epidermidis and Staphylococcus aureus
[59,63,75]; however, atypical organisms have also
been reported [76,77]. Work by Holland and col-
leagues [75] and Asaria and colleagues [72] has
demonstrated the presence of biofilms, glycocalix
structures that support and protect bacteria, on the
surfaces of exposed and covered scleral explants.
These biofilms facilitate the growth and survival of
bacteria and may explain why some explants exhibit
signs of infection without prior element exposure.
Because extruding scleral buckling elements can
promote infection, and infection can stimulate ele-
ment extrusion, the causal relationship between
these two processes remains undetermined [72].
In the early 1980s, a new material, the hydrogel
implant, became available for scleral buckling pro-
cedures. This material absorbs tissue fluids and
thereby enlarges, creating an expansile buckling
element. Early clinical studies were favorable [78];
however, longer follow-up demonstrated problematic
continued element swelling with resultant complica-
tions [66,67,79 – 81]. Use of this material was dis-
continued in the mid 1990s (around 1994 to1995).
Nevertheless, many patients still have this material
in place, and complications from its use continue
to arise.
Complications associated with the various ele-
ment types generally present differently. Silicone
element complications typically present much earlier
with signs of acute infection, whereas hydrogel
implant complications generally present much later
with space-occupying mass effects [66,67,81]. Both
complication profiles usually necessitate element re-
moval. When removed, silicone elements frequently
have positive bacterial cultures, whereas hydrogel
implants are only occasionally culture positive [66].
Among the other complications following scleral
buckling procedures, reduced corneal sensitivity has
been reported in as many as 85% of operated eyes
and can last up to 5 years. It may be secondary to
518 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 513 – 520
damage of the long ciliary nerves caused by the
buckle indentation and can lead to recurrent erosions.
Conjunctival chemosis and inclusion cysts have also
been reported following scleral buckle surgery and
can be a source of peripheral corneal drying and
dellen formation [82].
Scleral buckle elements can induce corneal and
ocular surface complications through local mechani-
cal forces, element extrusion, and infection. These
complications usually present earlier with the place-
ment of silicone sponge and rubber elements,
whereas complications associated with hydrogel
implants usually present much later.
Summary
Significant corneal compromise, that is, epithelial
abnormalities and endothelial cell loss, can occur
subsequent to vitreoretinal surgery. Planned epithelial
debridement and inadvertent epithelial defects occur
frequently, especially in diabetic corneas. Noncontact
or sew-on lenses create less epithelial problems than
do hand-held infusion lenses. Newer corneal lubri-
cants without benzalkonium chloride preservatives
may prove beneficial.
Risk factors for endothelial cell damage include
the use of irrigating solutions without appropriate
buffers, preoperative aphakia, a disrupted anterior lens
capsule intraoperatively, and the use of adjunctive
intraocular gasses or silicone oil. In addition to corneal
endothelial cell compromise, silicone oil use can
result in band keratopathy, persistent stromal edema,
and corneal vascularization; therefore, attempts
should be made to remove the oil when possible.
Corneal, conjunctival, and ocular surface compli-
cations related to scleral buckling procedures usually
occur as a result of concomitant element extrusion
and infection. Hydrogel implant elements, although
no longer in use, may continue to cause problems.
These elements have demonstrated continued swel-
ling, enlargement, and fragmentation, which may ne-
cessitate removal.
References
[1] Mandelcorn MS, Blankenship G, Machemer R. Pars
plana vitrectomy for the management of severe diabetic
retinopathy. Am J Ophthalmol 1976;81(5):561 – 70.
[2] Foulks GN, et al. Factors related to corneal epithelial
complications after closed vitrectomy in diabetics.
Arch Ophthalmol 1979;97(6):1076 – 8.
[3] Perry HD, et al. Corneal complications after closed
vitrectomy through the pars plana. Arch Ophthalmol
1978;96(8):1401 – 3.
[4] Brightbill FS, Myers FL, Bresnick GH. Postvitrec-
tomy keratopathy. Am J Ophthalmol 1978;85(5 Pt 1):
651 – 5.
[5] Faulborn J, Conway BP, Machemer R. Surgical
complications of pars plana vitreous surgery. Ophthal-
mology 1978;85(2):116 – 25.
[6] Schachat AP, et al. Complications of vitreous surgery
for diabetic retinopathy. II. Postoperative complica-
tions. Ophthalmology 1983;90(5):522 – 30.
[7] Chung H, et al. Reevaluation of corneal complications
after closed vitrectomy. Arch Ophthalmol 1988;106(7):
916 – 9.
[8] Virata SR, Kylstra JA, Singh HT. Corneal epithelial
defects following vitrectomy surgery using hand-held,
sew-on, and noncontact viewing lenses. Retina 1999;
19(4):287 – 90.
[9] Oyakawa RT, et al. Complications of vitreous surgery
for diabetic retinopathy. I. Intraoperative complica-
tions. Ophthalmology 1983;90(5):517 – 21.
[10] Michels RG, Ryan Jr SJ. Results and complications of
100 consecutive cases of pars plana vitrectomy. Am J
Ophthalmol 1975;80(1):24 – 9.
[11] Aaberg TM, Abrams GW. Changing indications and
techniques for vitrectomy in management of compli-
cations of diabetic retinopathy. Ophthalmology 1987;
94(7):775 – 9.
[12] Machemer R, Blankenship G. Vitrectomy for prolifera-
tive diabetic retinopathy associated with vitreous
hemorrhage. Ophthalmology 1981;88(7):643 – 6.
[13] Smiddy WE, Flynn Jr HW. Vitrectomy in the manage-
ment of diabetic retinopathy. Surv Ophthalmol 1999;
43(6):491 – 507.
[14] Friend J, Thoft RA. The diabetic cornea. Int Oph-
thalmol Clin 1984;24(4):111 – 23.
[15] Friend J, Ishii Y, Thoft RA. Corneal epithelial
changes in diabetic rats. Ophthalmic Res 1982;14(4):
269 – 78.
[16] Fukushi S, et al. Reepithelialization of denuded cor-
neas in diabetic rats. Exp Eye Res 1980;31(5):611 – 21.
[17] Kinoshita JH, et al. Aldose reductase in diabetic com-
plications of the eye. Metabolism 1979;28(4 Suppl 1):
462 – 9.
[18] Kenyon KRW, Michels Z, Conway R, Tolentino BF.
Corneal basement membrane abnormality in diabetes
mellitus. Invest Ophthalmol Vis Sci 1978;17:245.
[19] Alvarado J, Murphy C, Juster R. Age-related changes
in the basement membrane of the human corneal
epithelium. Invest Ophthalmol Vis Sci 1983;24(8):
1015 – 28.
[20] Jabbur NS, O’Brien TP. Incidence of intraoperative
corneal abrasions and correlation with age using the
Hansatome and Amadeus microkeratomes during laser
in situ keratomileusis. J Cataract Refract Surg 2003;
29(6):1174 – 8.
[21] Tekwani NH, Huang D. Risk factors for intraoperative
epithelial defect in laser in-situ keratomileusis. Am J
Ophthalmol 2002;134(3):311 – 6.
 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 513 – 520
[22] Hewitt SR, Couser JM, Stulting RD. Risk factors for
epithelial defects after laser in situ keratomileusis.
Invest Ophthalmol Vis Sci 2003;44:2646.
[23] Friberg TR, et al. Frequency of epithelial debridement
during diabetic vitrectomy. Am J Ophthalmol 2003;
135(4):553 – 4.
[24] Garcia-Valenzuela E, et al. Reduced need for corneal
epithelial debridement during vitreo-retinal surgery
using two different viscous surface lubricants. Am J
Ophthalmol 2003;136(6):1062 – 6.
[25] Insler MS, Tauber S, Packer A. Descemetocele
formation in a patient with a postoperative corneal
dellen. Cornea 1989;8(2):129 – 30.
[26] Kenyon KR, Stark WJ, Stone DL. Corneal endothelial
degeneration and fibrous proliferation after plana
vitrectomy. Am J Ophthalmol 1976;81(4):486 – 90.
[27] McCarey BE, Edelhauser HF, Van Horn DL. Func-
tional and structural changes in the corneal endothe-
lium during in vitro perfusion. Invest Ophthalmol
1973;12(6):410 – 7.
[28] Edelhauser HF, et al. Comparative toxicity of intra-
ocular irrigating solutions on the corneal endothelium.
Am J Ophthalmol 1976;81(4):473 – 81.
[29] Edelhauser HF, et al. Intraocular irrigating solutions:
their effect on the corneal endothelium. Arch Oph-
thalmol 1975;93(8):648 – 57.
[30] Edelhauser HF, et al. Effect of thiol-oxidation of glu-
tathione with diamide on corneal endothelial function,
junctional complexes, and microfilaments. J Cell Biol
1976;68(3):567 – 78.
[31] Edelhauser HF, Gonnering R, Van Horn DL. Intra-
ocular irrigating solutions: a comparative study of
BSS Plus and lactated Ringer’s solution. Arch Oph-
thalmol 1978;96(3):516 – 20.
[32] McDermott M, et al. Effects of intraocular irrigants on
the preserved human corneal endothelium. Cornea
1991;10(5):402 – 7.
[33] Benson WE, Diamond JG, Tasman W. Intraocular
irrigating solutions for pars plana vitrectomy: a pro-
spective, randomized, double-blind study. Arch Oph-
thalmol 1981;99(6):1013 – 5.
[34] Buettner H, Bourne WM. Effect of trans pars plana
surgery on the corneal endothelium. Dev Ophthalmol
1981;2:28 – 34.
[35] Friberg TR, Doran DL, Lazenby FL. The effect of
vitreous and retinal surgery on corneal endothelial cell
density. Ophthalmology 1984;91(10):1166 – 9.
[36] Diddie KR, Schanzlin DJ. Specular microscopy in
pars plana vitrectomy. Arch Ophthalmol 1983;101(3):
408 – 9.
[37] Mittl RN, et al. Endothelial cell counts following pars
plana vitrectomy in pseudophakic and aphakic eyes.
Ophthalmic Surg 1989;20(1):13 – 6.
[38] Mitamura Y, Yamamoto S, Yamazaki S. Corneal en-
dothelial cell loss in eyes undergoing lensectomy with
and without anterior lens capsule removal combined
with pars plana vitrectomy and gas tamponade. Retina
2000;20(1):59 – 62.
[39] Van Horn DL, et al. In vivo effects of air and sulfur
519
hexafluoride gas on rabbit corneal endothelium.
Invest Ophthalmol 1972;11(12):1028 – 36.
[40] Foulks GN, et al. The effect of perfluoropropane on the
cornea in rabbits and cats. Arch Ophthalmol 1987;
105(2):256 – 9.
[41] Lee DA, et al. The ocular effects of gases when
injected into the anterior chamber of rabbit eyes.
Arch Ophthalmol 1991;109(4):571 – 5.
[42] Han DP, et al. Evaluation of anterior segment tolerance
to short-term intravitreal perfluoron. Retina 1994;
14(3):219 – 24.
[43] Scott IU, et al. Outcomes and complications associated
with perfluoro-n-octane and perfluoroperhydrophenan-
threne in complex retinal detachment repair. Ophthal-
mology 2000;107(5):860 – 5.
[44] Wilbanks GA, et al. Perfluorodecalin corneal toxicity:
five case reports. Cornea 1996;15(3):329 – 34.
[45] Cibis PB, Okun E, et al. The use of liquid silicone in
retinal detachment surgery. Arch Ophthalmol 1962;68:
590 – 9.
[46] Azuara-Blanco A, Dua HS, Pillai CT. Pseudo-endo-
thelial dystrophy associated with emulsified silicone
oil. Cornea 1999;18(4):493 – 4.
[47] Sternberg Jr P, et al. The effect of silicone oil on the
cornea. Arch Ophthalmol 1985;103(1):90 – 4.
[48] Abrams GW, et al. The incidence of corneal abnor-
malities in the Silicone Study: Silicone Study Report 7.
Arch Ophthalmol 1995;113(6):764 – 9.
[49] Friberg TR, Guibord NM. Corneal endothelial cell
loss after multiple vitreoretinal procedures and the use
of silicone oil. Ophthalmic Surg Lasers 1999;30(7):
528 – 34.
[50] Lemmen KD, et al. [Keratopathy following vitrectomy
with silicone oil injection]. Fortschr Ophthalmol 1989;
86(6):570 – 3.
[51] Pang MP, Peyman GA, Kao GW. Early anterior
segment complications after silicone oil injection.
Can J Ophthalmol 1986;21(7):271 – 5.
[52] Buch H, Vesti Nielsen N. Keratopathy and pachymetric
changes after photorefractive keratectomy and vitrec-
tomy with silicone oil injection. J Cataract Refract
Surg 2000;26(7):1078 – 81.
[53] Foulks GN, et al. Histopathology of silicone oil
keratopathy in humans. Cornea 1991;10(1):29 – 37.
[54] Karel I, Kalvodova B, Kuthan P. Results of penetrating
keratoplasty in bullous silicone oil keratopathy. Graefes
Arch Clin Exp Ophthalmol 1998;236(4):255 – 8.
[55] Ando F. Intraocular hypertension resulting from
pupillary block by silicone oil. Am J Ophthalmol
1985;99(1):87 – 8.
[56] Chaudhry NA, Smiddy WE. Displacement of corneal
cap during vitrectomy in a post-LASIK eye. Retina
1998;18(6):554 – 5.
[57] Sakurai E, et al. Late-onset laser in situ keratomileusis
(LASIK) flap dehiscence during retinal detachment
surgery. Am J Ophthalmol 2002;134(2):265 – 6.
[58] Irvine WD, Smiddy WE, Nicholson DH. Corneal and
iris burns with the laser indirect ophthalmoscope. Am J
Ophthalmol 1990;110(3):311 – 3.
520 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 513 – 520
[59] Russo CE, Ruiz RS. Silicone sponge rejection: early
and late complications in retinal detachment surgery.
Arch Ophthalmol 1971;85(6):647 – 50.
[60] Ulrich RA, Burton TC. Infections following scleral
buckling procedures. Arch Ophthalmol 1974;92(3):
213 – 5.
[61] Flindall RJ, et al. Reduction of extrusion and infection
following episcleral silicone implants and cryopexy in
retinal detachment surgery. Am J Ophthalmol 1971;
71(4):835 – 7.
[62] Hilton GF, Wallyn RH. The removal of scleral buckles.
Arch Ophthalmol 1978;96(11):2061 – 3.
[63] Hahn YS, et al. Infection after sponge implantation for
scleral buckling. Am J Ophthalmol 1979;87(2):180 – 5.
[64] Lindsey PS, Pierce LH, Welch RB. Removal of scleral
buckling elements: causes and complications. Arch
Ophthalmol 1983;101(4):570 – 3.
[65] McMeel JW, et al. Acute and subacute infections
following scleral buckling operations. Ophthalmology
1978;85(4):341 – 9.
[66] Roldan-Pallares M, et al. Long-term complications of
silicone and hydrogel explants in retinal reattachment
surgery. Arch Ophthalmol 1999;117(2):197 – 201.
[67] Le Rouic JF, et al. Late swelling and removal of
Miragel buckles: a comparison with silicone inden-
tations. Retina 2003;23(5):641 – 6.
[68] Lanigan LP, Wilson-Holt N, Gregor ZJ. Migrating
scleral explants. Eye 1992;6(Pt 3):317 – 21.
[69] Maguire AM, Zarbin MA, Eliott D. Migration of solid
silicone encircling element through four rectus
muscles. Ophthalmic Surg 1993;24(9):604 – 7.
[70] Osman Saatci A, et al. An extruded encircling band
straddling the cornea and corneal groove formation.
Ophthalmic Surg Lasers 1998;29(12):991 – 2.
[71] Schwartz PL, Maris PJ, Maris CE. Silastic sponge
implants in retinal detachment surgery. Ann Ophthal-
mol 1981;13(9):1089 – 92.
[72] Asaria RH, et al. Biofilm on scleral explants with and
without clinical infection. Retina 1999;19(5):447 – 50.
[73] Buettner H, Goldstein BG, Anhalt JP. Infection
prophylaxis with silastic sponge explants in retinal
detachment surgery. Dev Ophthalmol 1981;2:71 – 6.
[74] Arribas NP, et al. Preoperative antibiotic soaking of
silicone sponges: does it make a difference? Ophthal-
mology 1984;91(12):1684 – 9.
[75] Holland SP, et al. Biofilm and scleral buckle-associated
infections: a mechanism for persistence. Ophthalmol-
ogy 1991;98(6):933 – 8.
[76] Smiddy WE, Miller D, Flynn Jr HW. Scleral buckle
infections due to atypical mycobacteria. Retina 1991;
11(4):394 – 8.
[77] Callanan D, Rubsamen PE. Moraxella infection of a
scleral buckle. Am J Ophthalmol 1992;114(5):637 – 8.
[78] Tolentino FI, et al. Hydrogel implant for scleral
buckling: long-term observations. Retina 1985;5(1):
38 – 41.
[79] Marin JF, et al. Long-term complications of the MAI
hydrogel intrascleral buckling implant. Arch Ophthal-
mol 1992;110(1):86 – 8.
[80] Hwang KI, Lim JI. Hydrogel exoplant fragmentation
10 years after scleral buckling surgery. Arch Ophthal-
mol 1997;115(9):1205 – 6.
[81] Kearney JJ, et al. Complications of hydrogel explants
used in scleral buckling surgery. Am J Ophthalmol
2004;137(1):96 – 100.
[82] Ambati J, Arroyo JG. Postoperative complications of
scleral buckling surgery. Int Ophthalmol Clin 2000;
40(1):175 – 85.
 Ophthalmol Clin N Am 17 (2004) 521 – 526
Refractive changes after posterior segment surgery
J. Bradley Randleman, MD*, Susanne M. Hewitt, MD,
R. Doyle Stulting, MD, PhD
Department of Ophthalmology, Emory University, 1365B Clifton Road NE, Suite 4500, Atlanta, GA 30322, USA
Surgical correction of retinal detachment has been
attempted since the early 1900s. As the anatomic
success of this and newer procedures has improved,
greater attention has been directed toward maximiz-
ing postoperative visual function, including the
correction of induced refractive error. Many variables
have a role in determining the effect of retinal surgery
on postoperative refraction, including the procedure
type, the surgical technique, patient age, and the use
of adjunctive measures such as silicone oil.
Refractive changes after retinal surgery may in-
clude a hyperopic or myopic shift, as well as induced
regular or irregular astigmatism. The significance of
these changes has been debated and may be related in
part to surgical technique. With improved technology,
the ability to detect more subtle alterations in corneal
shape has increased.
The eyes of infants and adolescents respond
differently to retinal surgery than do those of adults.
Retinal surgery in infants and children may have
profound effects on ocular development, producing
extreme alterations in postoperative refraction.
Silicone oil introduces its own unique set of
refractive changes and complicates biometric mea-
surements used to determine the appropriate power
of intraocular lens implants for subsequent cata-
ract surgery.
This article was supported in part by Research to Prevent
Blindness, New York, New York, and by National Institutes
of Health core grant P30 EYO6360, Bethesda, Maryland.
* Corresponding author. 1365 B Clifton Road NE, Suite
4500 Atlanta, GA 30322.
E-mail address: Jrandle@emory.edu (J.B. Randleman).
0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.002
Refractive changes caused by alterations in axial
length after retinal surgery
Early reports focused on alterations in axial length
and the resultant spherical changes induced after
retinal detachment repair, first using scleral resection
and later using scleral buckling techniques. Scleral
resection reduces axial length and produces a hyper-
opic shift [1 – 3]. Rosenthal reported the develop-
ment of marked hyperopia and astigmatism up to 5 D
within the first postoperative week after scleral re-
section [3].
Theoretically, scleral buckling should increase the
axial length of the eye, producing a myopic shift;
however, reported postoperative outcomes have been
variable. Grupposo observed significant spherical
changes of more than 5 D in 1965 [4]. In 1977,
Burton and colleagues reported the results in
12 aphakic and 18 phakic eyes after scleral buckling
procedures [5]. Within the first 6 weeks postopera-
tively, 60% of the eyes had a significant change
in axial length; however, this change was unpre-
dictable. Approximately half of the eyes had signifi-
cant lengthening, and half had significant shortening
after surgery.
In a postmortem study of eyes that had previously
undergone scleral buckling procedures, Rubin [6]
found that myopic shifts were usually related to
buckle height; however, high indentations para-
doxically produced a hyperopic shift. Larsen and
Syrdalen [7] reported a consistent axial elongation
of nearly 1 mm in all eyes examined, with a resultant
refractive change of
2.5 D. Elongation occurred in
the vitreous cavity, with minimal changes in anterior
chamber depth or lens thickness. They concluded
that differing surgical techniques could be responsible
ophthalmology.theclinics.com
522 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 521 – 526
for the variations in postoperative refractive out-
comes reported. Work by Harris and colleagues [8]
performed on eye bank eyes supports this concept.
They found that the type of buckle element and
technique directly affected postoperative refraction.
Progressive horizontal shortening with a thin solid
silicone band (no. 41) increased buckle height and
axial length. Thicker elements (no. 287 tire, no.
505 sponge, no. 507 sponge) increased axial length
when horizontally shortened but decreased axial
length when applied by invagination techniques.
Smiddy and colleagues [9] also found a correlation
between buckle type and axial lengthening, with
encircling scleral bands increasing axial length by an
average of 0.99 mm and producing
2.75 D of re-
fractive change, whereas nonencircling procedures
increased axial length by 0.26 mm and induced

0.31 D of refractive change.
Induced astigmatism after retinal surgery
Highly unpredictable alterations in keratometric
power and corneal curvature can be produced by
scleral buckling procedures. It is not clear from the
literature whether these astigmatic changes are
transient or permanent. In 1958, Givner and Karlin
[10] described a case in which scleral buckling
induced 3 D of astigmatism immediately postopera-
tively, which decreased to 0.5 D at 7 months post-
operatively. Other investigators have also reported
small transient alterations in postoperative corneal
astigmatism [11,12].
Burton [13] reported the induction of severe
persistent irregular astigmatism after scleral buckling
with episcleral silicone sponges. Goel and colleagues
[14] compared radial buckles with circumferential
buckles and found that radial elements were signifi-
cantly more likely to produce astigmatic errors
greater than 2 D. Furthermore, they found that,
although small amounts of astigmatism usually
represented transient changes, larger amounts of
induced astigmatism (>3 D) usually persisted.
Smiddy and colleagues also found moderate amounts
of induced astigmatism but did not find a correlation
with the use of radial versus nonradial elements [9].
Slusher and colleagues [15] reported seven cases
of severe persistent astigmatism after pars plana
vitrectomy. Most of these cases required lysis of the
suture at the vitrectomy site to relieve the astigma-
tism. They determined that repeat pars plana vitrec-
tomy and total air or gas tamponade are risk factors
for high astigmatism owing to the tight wound
closure required in these situations.
Induced astigmatism is typically mild and tran-
sient, but significant alterations can occur, particu-
larly after pars plana vitrectomy with total air or gas
tamponade. Use of the 25-gauge sutureless vitrec-
tomy system is likely to reduce astigmatism induced
by scleral suturing. This system may result in more
rapid visual recovery in these patients.
Refractive changes from alterations in corneal
topography after retinal surgery
Recently, potentially more significant alterations
in corneal shape have been evaluated by corneal
topography. Although keratometry detects global
changes in corneal shape, topography can identify
irregular astigmatism – focal alterations in corneal
curvature that may not be translated to the entire
corneal meridian but nonetheless can diminish vi-
sual acuity. Additional descriptors have been created
to quantify the amount of irregular astigmatism.
These descriptors include the SRI, or surface regu-
larity index, and the SAI, or surface asymmetry
index [16]. As reported by Wilson and Klyce, the
SRI is a measure that describes the regularity
and optical quality of the central cornea and is
correlated with best spectacle-corrected visual
acuity [16]. The SAI measures surface asymme-
try; this value is less significant but still impor-
tant in determining the optical performance of
the cornea.
Weinberger and colleagues [17] found that circu-
lar buckles alone induced only 0.4 D steepening over
the entire cornea but over 2 D in the central cornea by
the first postoperative week. The addition of partial
elements increased steepening over the entire cornea
but did not affect the measured curvature of the
central cornea. Vitrectomy had a minimal effect on
overall corneal topography; however, it induced 1 to
1.5 D of central corneal steepening, which correlated
with the location of the entry port. These changes
were transient; they became apparent by the first
postoperative week but normalized by week twelve in
all cases [17]. Ornek and colleagues [18] also found a
transiently increased central steepening and recom-
mended a delay in prescribing spectacles until 6 or
more months postoperatively.
Topographic analysis has shown two patterns of
corneal steepening after the placement of scleral
buckles. Encircling buckles produce either uniform
central steepening or coupled steepening and flat-
tening in the opposite regions, whereas local or
segmental buckles produce steepening in the corre-
sponding meridian [19 – 21]. The SRI and SAI values
 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 521 – 526
increase, indicating increased surface irregularity and
asymmetry [19].
Wirbelauer and colleagues [22] found that suture
diameter, wound closure, and cautery were related to
corneal curvature changes after pars plana vitrectomy.
They found astigmatic axis shifts in the meridians
corresponding to sclerotomy sites and recommended
using small diameter absorbable suture, avoiding
excess suture tightening during closure, and mini-
mizing cautery to lessen these effects.
Some loss of best spectacle-corrected visual
acuity after scleral buckling procedures or pars
plana vitrectomy, previously attributed to macular
dysfunction, may, in fact, be due to irregular cor-
neal astigmatism.
Refractive changes in children undergoing retinal
surgery
Although most of the literature focuses on
induced refractive changes after retinal surgery in
adults, infants and children are most susceptible to
large, variable, and fluctuating postoperative refrac-
tive results owing to effects on ocular development.
Scleral buckling has a profound effect on the
development and emmetropization of the eyes in
young rabbits [23,24], human infants [25], and
adolescents [26]. In a rabbit model, Choi and
colleagues found that scleral buckles induced axial
myopia and prevented normal emmetropization [23],
whereas Moshfeghi and colleagues demonstrated that
noncontinuous encircling bands exerted less delete-
rious effects on ocular growth and development [24].
In human infants, Chow and colleagues found an
induced myopia greater than 11 D with scleral
buckling performed for stage 4 retinopathy of
prematurity [25]. This induced myopia was reduced
by one half after division of the buckle [25]. Sato and
colleagues [26] found that in 7- to 15-year-old
patients, scleral buckles actually reduced the pro-
gression of myopia in a comparison with the non-
buckled eye. They concluded that a scleral buckle
prevented normal ocular development, and that
children under the age of 10 years were particularly
susceptible to this effect.
Refractive and biometric changes related to
silicone oil use
The use of silicone oil as an adjunctive agent in
vitrectomy provides a unique set of variables that
may influence postoperative visual acuity and re-
523
fractive error. Silicone oil was introduced in 1962
as a method of replacing vitreous humor to achieve
tamponade of retinal tears and detachments [27].
Before the advent of contemporary vitrectomy tech-
niques, its efficacy was low. With the advent
of these techniques, its use has become popular in
cases of complex retinal detachments involving
proliferative vitreoretinopathy and penetrating inju-
ries. Silicone oil was chosen as a tamponade owing
to its high surface tension, transparency, stability,
and relatively low toxicity; however, it induces
refractive changes when it occupies the vitreous
cavity [28 – 30], and it creates difficulties in ob-
taining accurate measurements for intraocular
lens calculations after it is injected into the eye
[31,32].
Refractive changes
When present, silicone oil has a variable effect
on refraction dependent on the phakic status of the
eye [28,29]. Aphakic eyes become less hyperopic
by an average of 6 to 7 D [28,29], whereas phakic
eyes become more hyperopic by an average of
5.5 to 7.6 D [28,29]. Silicone oil has an index of
refraction of 1.405 compared with that of vitreous
(1.336). In the aphakic state, silicone oil forms a
convex anterior surface relative to the corneal en-
dothelium. This surface decreases the overall hy-
peropia by acting as a plus lens. Conversely, in the
phakic eye, the silicone oil forms a concave surface
behind the natural lens, which acts as a minus lens
and increases the overall level of hyperopia. In pseu-
dophakic eyes, silicone oil negates the effect of the
intraocular lens (if the refractive index is similar to
that of silicone), causing a myopic shift like that of
an aphakic eye [29].
The effect silicone oil has on refraction can also
be affected by head position [29,30]. This effect is
more pronounced in aphakia, presumably related to
the increased volume in which silicone can shift in
the aphakic eye. When patients’ refractions were
measured in the supine position and then remeasured
in the head-down position, the spherical equivalent
increased by approximately 6 D in aphakic eyes. This
difference was most likely caused by a shift in the
position of the oil bubble. In the head-down position,
the anterior surface is less convex than in the supine
position, resulting in more hyperopia in the head-
down position. In the same study, the cylinder axis
was observed to shift 11.5 degrees on average in
aphakic eyes and 10.1 degrees on average in phakic
eyes with changing head position, but the cylindrical
power did not change. These changes can be noted
524 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 521 – 526
by the patient, who may complain of fluctuations in
vision with various activities.
In addition, more variation in refraction can occur
with incomplete silicone oil fill of the vitreous cavity,
owing to the interactions of light at the oil – water
interface. When the oil fill is btopped upQ in these
cases, a myopic shift is noted owing to bulging of the
anterior surface of the oil bubble and increased plus
lens power at that interface.
Silicone oil has been noted to decrease the
accommodation of the lens in phakic patients,
and, typically, a +2.00 to 2.50 D bifocal is required
in phakic patients when the fellow eye requires
none [29].
Biometry of the silicone oil – filled eye
To determine the axial length and intraocular lens
calculation in a silicone oil – filled eye, A-scan
ultrasonic biometry requires an adjustment for the
slower speed of sound in oil. Sound waves travel at
987 m/s in silicone oil of viscosity of 1000 centi-
stokes versus 1532 m/s in vitreous [31]. Silicone oil
also causes poor penetration of the sound wave,
complicating measurements further [33]. Because
the A-scan uses time in microseconds from the
cornea to the retina to determine the length of the
eye, if no adjustment is made for the oil, an
artificially high axial length is obtained, resulting in
the placement of a lower power intraocular lens
than necessary and a hyperopic result. To adjust for
this effect, the axial length obtained with silicone
oil in place can be multiplied by a correction factor
of 0.71 [32]. When the A-scan measures the vit-
reous cavity depth separately, that figure can be
multiplied by a correction factor of 0.64 and added
to the anterior chamber depth and lens thickness
to obtain a true axial length [32,34].
Another method that may be more predictable has
also been proposed. This method is based on a
velocity conversion equation, which provides for
the correction of an erroneous measurement that
results from the use of an incorrect sound velocity
setting. To determine the true axial length (TAL)
using the equation, the correct sound velocity (Vc)
should be divided by the incorrect sound velocity
setting used for the measurement (Vm) and then
multiplied by the incorrect (apparent) axial length
reading (AAL) [35].
TAL ¼ Vc=Vm  AAL
Owing to shifting of the oil bubble, the patient
should be measured upright so the oil fills the eye
from lens to macula. The transducer should be
oriented with the beam perpendicular to the globe
to reduce refraction [33,36]. If the axial length is out
of the range of the A-scan used, the B-scan may be
used instead, although this is known to underestimate
the true length and may induce more error in long
eyes [32].
In a study in which A-scan adjustments were
performed before cataract extraction/silicone oil
removal in 27 patients, the final refraction differed
from that predicted by 1.81 ± 2.16 D, with a range of
+3.25 to
9.00 D [33]. Of these measurements,
51.9% deviated from predicted values by less than
1 D, and 74.1% differed by 2 D or less. Posterior
staphylomas were found in the eyes with the greatest
deviation from predicted values, and it has been
suggested that leaving these eyes aphakic, at least
initially, may be the best strategy.
In the future, newer technologies may obviate the
need to use ultrasonic A-scanning to determine axial
length and may allow more accurate measurements to
be obtained [37].
Post cataract surgery, with silicone oil retained
If silicone oil is retained for several months, a
cataract is likely to develop owing to inhibition of
lens metabolism [33,38,39]. In some eyes, it is
preferable to preserve the silicone oil in the vitreous
cavity for the purpose of long-term retinal tampon-
ade. Because the silicone oil, with its increased re-
fractive index compared with that of vitreous, will
remain in direct contact with the posterior capsule, it
becomes difficult to predict the postoperative refrac-
tion. Owing to the difficulties in obtaining accurate
axial length measurements, many authorities have
suggested that A-scan ultrasonic biometry be per-
formed before silicone oil injection [39]; however, if
an encircling band is placed during the retinal
surgery, these measurements will be inaccurate owing
to an increase in axial length.
In a study by Grinbaum and colleagues [40] in
which the A-scan had been completed before oil
injection and after scleral buckling, extracapsular cata-
ract extraction and biconvex intraocular lens implan-
tation were successfully completed in eight cases.
The actual postoperative refraction differed from the
predicted value by an average of +4.08 D; however,
when the oil was removed in two cases, a myopic shift
resulted in a final refraction within 0.25 and 0.80 D
from the original predicted values, respectively. These
findings suggest that if silicone oil removal is
planned eventually, this technique may result in
acceptable postoperative refractive outcomes.
 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 521 – 526
Refractive changes following retinal cryotherapy
and panretinal photocoagulation
Loss of accommodation, transient myopia, or both
have been reported following retinal cryotherapy and
panretinal photocoagulation. Loss of accommodation
can be secondary to the retrobulbar block owing to
mechanical injury to the ciliary ganglion, its roots, or
the short ciliary nerves. Accommodative paresis and
myopia have also been reported without retrobulbar
block following cryotherapy, laser retinopexy of
retinal tears, and panretinal photocoagulation. These
symptoms are usually transient and tend to resolve
within 5 weeks without treatment. A possible
explanation could be damage to the short ciliary
nerves during treatment [41].
Summary
Retinal surgery can induce significant refractive
errors. These errors include spherical changes caused
by alterations in axial length after scleral buckle
placement and astigmatic changes induced by scleral
buckling or pars plana vitrectomy. Focal alterations
in corneal curvature can significantly limit post-
operative visual acuity when axial length and
keratometry values seem relatively normal. Surgical
technique may also influence the induction of cor-
neal surface irregularity, especially in highly symp-
tomatic cases. These refractive errors are usually
transient, but suture lysis and buckle transection may
occasionally be indicated. In very young patients,
retinal surgery not only affects refractive outcomes
but also alters the course of normal ocular develop-
ment. The adjunctive use of silicone oil can impose
alterations directly, by the oil’s interaction with the
other refractive elements of the eye, and indirectly,
through its effects on intraocular lens power calcu-
lations for subsequent cataract surgery.
References
[1] Borley W. The scleral resection (eyeball shortening)
operation. Trans Am Ophthalmol Soc 1949;47:462.
[2] Shaplund C. Changes in the refraction of the eye
following the operation of lamellar scleral resection.
Trans Ophthalmol Soc UK 1953;73:205.
[3] Rosenthal M. A method of scleral resection for retinal
detachment. In: Schepens CL, editor. Importance of the
vitreous body in retina surgery with special emphasis
on re-operation. St. Louis7 CV Mosby; 1960. p. 165 – 6.
[4] Grupposo S. Visual results after sceral buckling with
silicone implant. In: Scheppens RC, editor. Contro-
525
versial aspects of the management of retinal detach-
ment. Boston7 Little Brown; 1965. p. 354 – 63.
[5] Burton TC, Herron BE, Ossoinig KC. Axial length
changes after retinal detachment surgery. Am J Oph-
thalmol 1977;83(1):59 – 62.
[6] Rubin ML. The induction of refractive errors by retinal
detachment surgery. Trans Am Ophthalmol Soc 1975;
73:452 – 90.
[7] Larsen JS, Syrdalen P. Ultrasonographic study on
changes in axial eye dimensions after encircling
procedure in retinal detachment surgery. Acta Oph-
thalmol (Copenh) 1979;57(3):337 – 43.
[8] Harris MJ, et al. Geometric alterations produced by
encircling scleral buckles: biometric and clinical
considerations. Retina 1987;7(1):14 – 9.
[9] Smiddy WE, et al. Refractive changes after scleral
buckling surgery. Arch Ophthalmol 1989;107(10):
1469 – 71.
[10] Givner I, Karlin D. Alterations in refraction and their
clinical significance. Ear Nose Throat J 1958;37(10):
676 – 8.
[11] Jampel HD, et al. Corneal astigmatic changes after
pars plana vitrectomy. Retina 1987;7(4):223 – 6.
[12] Tomidokoro A, Oshika T, Kojima T. Corneal astigma-
tism after scleral buckling surgery assessed by Fou-
rier analysis of videokeratography data. Cornea 1998;
17(5):517 – 21.
[13] Burton TC. Irregular astigmatism following episcleral
buckling procedure with the use of silicone rubber
sponges. Arch Ophthalmol 1973;90(6):447 – 8.
[14] Goel R, Crewdson J, Chignell AH. Astigmatism
following retinal detachment surgery. Br J Ophthalmol
1983;67(5):327 – 9.
[15] Slusher MM, Ford JG, Busbee B. Clinically significant
corneal astigmatism and pars plana vitrectomy. Oph-
thalmic Surg Lasers 2002;33(1):5 – 8.
[16] Wilson SE, Klyce SD. Quantitative descriptors of
corneal topography: a clinical study. Arch Ophthalmol
1991;109(3):349 – 53.
[17] Weinberger D, et al. Corneal topographic changes af-
ter retinal and vitreous surgery. Ophthalmology 1999;
106(8):1521 – 4.
[18] Ornek K, et al. Corneal topographic changes after
retinal detachment surgery. Cornea 2002;21(8):803 – 6.
[19] Hayashi H, et al. Corneal shape changes after scleral
buckling surgery. Ophthalmology 1997;104(5):831 – 7.
[20] Domniz YY, Cahana M, Avni I. Corneal surface
changes after pars plana vitrectomy and scleral
buckling surgery. J Cataract Refract Surg 2001;27(6):
868 – 72.
[21] Azar-Arevalo O, Arevalo JF. Corneal topography
changes after vitreoretinal surgery. Ophthalmic Surg
Lasers 2001;32(2):168 – 72.
[22] Wirbelauer C, et al. Corneal shape changes after pars
plana vitrectomy. Graefes Arch Clin Exp Ophthalmol
1998;236(11):822 – 8.
[23] Choi MY, Yu YS. Effects of scleral buckling on
refraction and ocular growth in young rabbits. Graefes
Arch Clin Exp Ophthalmol 2000;238(9):774 – 8.
526 J.B. Randleman et al / Ophthalmol Clin N Am 17 (2004) 521 – 526
[24] Moshfeghi AA, et al. The effects of scleral buckling
on young rabbit eyes. Arch Ophthalmol 2004;122(4):
473 – 6.
[25] Chow DR, Ferrone PJ, Trese MT. Refractive changes
associated with scleral buckling and division in
retinopathy of prematurity. Arch Ophthalmol 1998;
116(11):1446 – 8.
[26] Sato T, et al. Refractive changes following scleral
buckling surgery in juvenile retinal detachment. Retina
2003;23(5):629 – 35.
[27] Cibis PB, Okun E, et al. The use of liquid silicone in
retinal detachment surgery. Arch Ophthalmol 1962;68:
590 – 9.
[28] Stefansson E, et al. Refractive changes from use of
silicone oil in vitreous surgery. Retina 1988;8(1):20 – 3.
[29] Smith RC, Smith GT, Wong D. Refractive changes in
silicone filled eyes. Eye 1990;4(Pt 1):230 – 4.
[30] Dick HB, et al. Effect of head position on refraction in
aphakic and phakic silicone-filled eyes. Retina 1997;
17(5):397 – 402.
[31] Hoffer KJ. Ultrasound velocities for axial eye length
measurement. J Cataract Refract Surg 1994;20(5):
554 – 62.
[32] Murray DC, et al. Biometry of the silicone oil-filled
eye. Eye 1999;13(Pt 3a):319 – 24.
[33] Ghoraba HH, et al. The problems of biometry in
combined silicone oil removal and cataract extraction:
a clinical trial. Retina 2002;22(5):589 – 96.
[34] Larkin GF, Leaver PK. Phacoemulsification and
silicone oil removal through a single corneal incision.
Ophthalmologe 1998;105:2023 – 7.
[35] Hoffer KJ. Ultrasound velocities for axial eye length
measurement. J Cataract Refract Surg 1994;20(5):
554 – 62.
[36] Shugar JK, et al. Ultrasonic examination of the sili-
cone-filled eye: theoretical and practical considera-
tions. Graefes Arch Clin Exp Ophthalmol 1986;224(4):
361 – 7.
[37] Tehrani M, et al. Evaluation of the practicality of opti-
cal biometry and applanation ultrasound in 253 eyes.
J Cataract Refract Surg 2003;29(4):741 – 6.
[38] Borislav D. Cataract after silicone oil implantation.
Doc Ophthalmol 1993;83(1):79 – 82.
[39] Assi AW, Gotzaridis E, et al. Combined phacoemulsi-
fication and transpupillary drainage of silicone oil:
results and complications. Br J Ophthalmol 2001;85:
942 – 5.
[40] Grinbaum A, Treister G, Moisseiev J. Predicted and
actual refraction after intraocular lens implantation in
eyes with silicone oil. J Cataract Refract Surg 1996;
22(6):726 – 9.
[41] Lerner BC, Lakhanpal V, Schocket SS. Transient
myopia and accommodative paresis following retinal
cryotherapy and panretinal photocoagulation. Am J
Ophthalmol 1984;97(6):704 – 8.
 Ophthalmol Clin N Am 17 (2004) 527 – 537
Anterior segment inf lammation and hypotony after
posterior segment surgery
J. Fernando Arevalo, MD*, Reinaldo A. Garcia, MD, Carlos F. Fernandez, MD
Clı́nica Oftalmológica Centro Caracas, Centro Caracas PH-1, Av. Panteon, San Bernardino, Caracas 1010, Venezuela
There are many indications for posterior segment
surgery, including rhegmatogenous retinal detach-
ment, nonresolving vitreous opacities, proliferative
retinopathies, a posteriorly dislocated crystalline lens
or intraocular lens (IOL), epiretinal membranes,
vitreomacular traction syndrome, vitreopapillary trac-
tion, macular holes, subretinal neovascularization
or hemorrhage, diagnostic vitrectomy, ocular trauma,
and endophthalmitis.
Recent advances in the instrumentation and sur-
gical techniques for posterior segment surgery (es-
pecially pars plana vitrectomy [PPV]) permit better
anatomic and functional outcomes. Nevertheless,
these procedures may be associated with potential
complications involving the anterior segment, includ-
ing cataracts, keratopathies, elevated intraocular
pressure or glaucoma, endophthalmitis, noninfectious
postoperative inflammation, and hypotony. This
article discusses anterior segment inflammation and
hypotony after posterior segment surgery.
Postoperative inflammation after posterior
segment surgery
Intraocular inflammation can occur after any
ocular surgical procedure, including posterior seg-
This article was supported in part by the Fundación
Arevalo-Coutinho para la Investigación en Oftalmologı́a
(FACO), Caracas, Venezuela (JFA, RAG, CFF), and the
Pan-American Ophthalmological Foundation (CFF).
* Corresponding author.
E-mail address: areval1@telcel.net.ve (J.F. Arevalo).
0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.004
ment procedures. Most cells in the anterior chamber
are erythroclasts released from the vitreous lamella or
from intraoperative or postoperative bleeding. If no
destructive retinopexy is performed, PPV results in
little inflammation. Iris surgery in conjunction with
vitreous surgery does result in inflammation and
should be minimized. Persistent lens material has a
role in postoperative inflammation as well.
All cases of iris neovascularization or rubeosis
iridis and most cases of retinal detachment are
associated with protein release into the anterior
chamber, which is visible as flare. Treatment of the
basic disease process by reattachment of the retina or
panretinal photocoagulation is far more effective than
steroids in preventing this inflammation. Any fibrin
appearing in the frontal plane as a result of severe
inflammation may result in the development of a
cyclitic membrane and phthisis bulbi in severe cases.
Steroids do not seem to retard healing of any of the
ocular structures and should be used to suppress all
inflammation [1]. One must differentiate between the
infectious and noninfectious causes of postoperative
inflammation to implement prompt treatment with
specific antimicrobial therapy.
Noninfectious uveitis associated with intraocular
surgery is often low grade and self-limited. Never-
theless, patients with pre-existing uveitis often have
an exacerbation of intraocular inflammation after
surgery. The flare up of the uveitis usually occurs 3 to
7 days after surgery and may occur earlier in patients
who do not receive proper preoperative prophylaxis
with steroids. The uveitis can be severe, presenting
with pain and hypopyon, and can be misdiagnosed as
infectious endophthalmitis.
ophthalmology.theclinics.com
528 J.F. Arevalo et al / Ophthalmol Clin N Am 17 (2004) 527 – 537
Possible etiologies of postsurgical intraocular
inflammation according to its time of appearance
include the following [2]:
Intraocular inflammation occurring from post-
operative day 1 to 30: bacterial endophthal-
mitis, sterile endophthalmitis, recurrence or
increased activity of previous uveitis, phaco-
genic (lens related) uveitis, reactions to an IOL,
responses to a laser procedure, new onset of
idiopathic or previously unrecognized uveitis
Intraocular inflammation occurring from post-
operative day 15 to years later: fungal endoph-
thalmitis, Propionibacterium acnes or other
anaerobic endophthalmitis, low virulence aero-
bic bacterial endophthalmitis, phacogenic (lens
related) uveitis, sympathetic ophthalmia, reac-
tions to an IOL, iris ciliary body irrita-
tion related to physical contact with an IOL,
new onset of idiopathic or previously unrecog-
nized uveitis
In many cases, topical prednisolone 1% may be
sufficient to suppress inflammation. In others, peri-
ocular steroid injection or oral prednisolone may be
required. A topical cycloplegic agent, such as 1%
atropine sulfate or 0.1% cyclopentolate, is often used
in conjunction with corticosteroids. Cycloplegia helps
relieve ocular discomfort secondary to inflammation
and prevents the formation of pupillary synechiae. In
some cases, antiglaucoma medications may be
employed to control ocular hypertension.
Noninfectious inflammation
Conjunctival complications
Conjunctival inclusion cysts (Fig. 1A) are ac-
quired mobile cystic lesions lined with stratified
squamous epithelium or goblet cells that may appear
Fig. 1. (A) Large translucent cyst with pseudohypopyon and vascularized surface overlying a previous sclerotomy incision.
(B) Silicone oil conjunctival cyst. (From Bourcier T, Monin C, Baudrimont M, Larricart P, Borderie V, et al. Conjunctival
inclusion cyst following pars plana vitrectomy. Arch Ophthalmol 2003;121:1067; with permission.)
after posterior segment surgery. These cysts may re-
sult from amalgamations of mucosal folds in inflam-
matory conjunctival conditions or, more commonly,
from implantation of epithelium in the conjunctival
stroma secondary to PPV [3]. Postoperative inflam-
mation can stimulate the development of cystic
inclusions that progressively grow to form a rela-
tively large cyst. The cysts can also appear as an
inflammatory reaction to the scleral suture material
[4]. If the cyst is small and does not bother the
patient, it should be left alone. When indicated,
careful surgical removal, without rupture of the cyst
wall if possible, should be performed. A silicone oil
conjunctival cyst may develop owing to high intra-
ocular pressure and sclerotomy leak (Fig. 1B). This
type of cyst should be drained.
Corneal complications
Dellen are painless spontaneous areas of thinned
cornea located near the limbus, generally at the 3 or
9 o’clock position. They are caused by edema of the
limbal conjunctiva that prevents appropriate distribu-
tion of the tear film to the limbal cornea. They are
treated by increasing the moisture to the cornea with
lubricants or patching [5].
Stromal and epithelial edema results from decom-
pensation of the corneal endothelium and may be
seen as part of anterior segment ischemia. Corneal
striae may indicate hypotony, poor aqueous flow,
compromised endothelium, inflammation, or ische-
mia. If corneal edema persists, corneal penetrating
keratoplasty may be necessary [5].
Precipitates on the corneal endothelium may be
evidence of inflammation, but they also can represent
pigment released from the iris pigment epithelium
and erythroclasts.
Regarding keratopathy after posterior segment
surgery, there is no difference between silicone oil
and C3F8 gas with respect to corneal complications.
 J.F. Arevalo et al / Ophthalmol Clin N Am 17 (2004) 527 – 537
Fig. 2. Band keratopathy in a silicone oil – filled eye.
In the Silicone Study, keratopathy developed in both
groups (silicone oil and C3F8) in the first year of
follow-up in 27% of cases [6]. In younger patients,
corneal damage frequently is manifested as band
keratopathy (Fig. 2); in older age groups, diffuse
bullous keratopathy is more common. The Silicone
Study [6] reported factors increasing the likelihood of
postoperative corneal abnormalities after surgery for
severe proliferative vitreoretinopathy. These factors
included preoperative aphakia or pseudophakia, pre-
operative rubeosis iridis, the absence of a fluid/gas
exchange, corneal touch by silicone oil, postoperative
aqueous cells or aqueous flare, and the need for
reoperation [6].
Management strategies for minimizing the inci-
dence of postoperative keratopathy include maintain-
ing a patent inferior peripheral iridectomy and early
removal of silicone oil. In the Silicone Study, 33% of
the aphakic eyes developed closure of the inferior
peripheral iridectomy, and 80% of eyes with closure
of the inferior peripheral iridectomy developed
forward migration of the silicone oil. Tissue plas-
minogen activator (tPA) can be used to avoid closure
of the inferior peripheral iridectomy at a dose of
25 mg in 0.1 mL. Corneal chelation is sometimes
helpful for patients with band keratopathy. When
severe keratopathy develops, penetrating keratoplasty
with or without silicone oil removal may be
considered. The frequency of graft failure is much
Fig. 3. (A) Fibrin pupillary membrane in a patient after PPV. (B) Injection of tPA (25 mg/0.1 mL) into the eye through the limbus
into the anterior chamber lysed the pupillary fibrin membrane and improved visual acuity in 45 minutes.
529
lower when silicone oil has previously been removed
or is removed at the time of penetrating keratoplasty
as compared with when silicone oil is retained [7].
Anterior chamber complications
After surgery for the management of severe
proliferative vitreoretinopathy, fibrin occasionally
develops in the anterior chamber, pupillary area,
and anterior vitreous postoperatively and may inter-
fere with surgical success and visual outcome. The
predisposing factors are severe flare, the presence of a
previously placed scleral buckle, poor preoperative
visual acuity, and intraoperative anterior epiretinal
membrane dissection [8]. Postoperative fibrin for-
mation in the anterior segment may not only obstruct
postoperative examination of the posterior segment
but may also contribute to iridocorneal apposition [9],
pupillary block glaucoma [10], and anterior vitreo-
retinopathy [11]. Injection of tPA (25 mg/0.1 mL)
into the eye through the limbus into the anterior cham-
ber may lyse the papillary fibrin membrane and im-
prove visual acuity (Fig. 3).
Fibrinous postoperative uveitis may be accompa-
nied by sterile hypopyon. In a retrospective review of
30 patients undergoing PPV and silicone oil place-
ment, fibrinous postoperative uveitis with hypopyon
developed in two cases (Fig. 4) [12]. Panretinal
photocoagulation and impurities in the oil were
considered possible contributing factors. Neverthe-
less, the exclusion of infectious endophthalmitis is of
primary importance in these patients, and they must
be closely monitored. Patients should be treated
aggressively with topical steroids. A response should
be expected within 48 to 72 hours with resolution by
1 week in noninfectious cases [12].
Iridocorneal apposition can develop in an aphakic
eye in which intraocular gas has been injected early
after PPV. The incidence of iridocorneal apposition is
3% according to Konishi et al [9]. Female patients
and those who undergo PPV for the management of
proliferative diabetic retinopathy and proliferative
530 J.F. Arevalo et al / Ophthalmol Clin N Am 17 (2004) 527 – 537
Fig. 4. Sterile hypopyon after posterior segment surgery.
vitreoretinopathy are more prone to this postoperative
complication. It is produced owing to inappropriate
positioning of the patient’s head at night after surgery.
Injection of tPA into the eye through the pars plana
may lyse the papillary fibrin membrane and release
the apposition, avoiding corneal endothelial damage
that may occur if mechanical repositioning is at-
tempted. Mechanical repositioning with viscoelastic
material may be required in the case of tPA treatment
failure [9].
Fibrin formation and hypopyon may occur after
laser treatment. The laser shock waves can cause
alteration of the hematoretinal barrier, particularly in
patients with more heavily pigmented irides [13]. A
significant increase in aqueous flare can be found
10 days after panretinal photocoagulation [14]. In ad-
dition, fibrin and hypopyon after laser photocoagula-
tion can be seen as part of anterior segment ischemia.
Transient or sustained elevation of intraocular
pressure is a frequent complication after PPV and
intravitreous injection of silicone oil for complicated
retinal detachment [15]. Early postoperative intra-
ocular pressure elevation has been seen in 7% to
48% of eyes. The related causes are pupillary block,
Fig. 5. (A) Silicone oil in the anterior chamber. The silicone oil is emulsified and forms a hyperoleon in a phakic eye. (B)
Glaucoma developed after silicone oil emulsification in this patient. The silicone oil is emulsified and forms a hyperoleon in a
pseudophakic eye.
inflammation, pre-existing glaucoma or angle pathol-
ogy, and migration of silicone oil into the anterior
chamber with consequent mechanical impediment to
filtration [15]. Leaver et al [16] reported histopatho-
logic evidence of the presence of silicone-laden
macrophages within the trabecular meshwork without
evidence of structural damage to the collagen fibers
and the trabecular endothelium. This finding strongly
suggests obstruction as the cause of glaucoma.
Emulsification of silicone oil (binverse hypopyonQ or
bhyperoleonQ) (Fig. 5) is a significant risk factor for
glaucoma. Postoperative iris neovascularization is an
independent risk factor and increases the risk of
glaucoma tenfold. Peripheral anterior synechiae and
intraocular inflammation did not contribute signifi-
cantly to the development of glaucoma. Prolonged
postoperative steroid use (beyond 6 weeks) also was
not a significant risk factor for the elevation of
intraocular pressure. In the postoperative period,
intraocular pressure should be checked at regular
intervals, especially if the patient has a risk factor.
Treatment options are medical therapy, early silicone
oil removal (the removal of silicone oil in the setting
of emulsification and glaucoma may not necessarily
result in intraocular pressure reduction), trabe-
culectomy, cyclodestructive procedures, and shunt
procedures [15].
Sympathetic ophthalmia is a bilateral diffuse granu-
lomatous uveitis occurring after previous penetrat-
ing ocular injury (trauma or, more rarely, ocular
surgery). Contemporary concepts regarding the patho-
genesis of sympathetic ophthalmia suggest that drain-
age of ocular antigens, melanin, or a soluble fraction
from outer segments of photoreceptor cells into the
lymphatic system in conjunction with adjuvant
activity of molecules entering through the perforating
wound elicit a delayed bilateral cellular immune
response [17]. The incidence of sympathetic ophthal-
 J.F. Arevalo et al / Ophthalmol Clin N Am 17 (2004) 527 – 537
mia after routine surgical procedures is approximately
1 case per 10,000 eyes [18]. The incidence of post-
vitrectomy sympathetic ophthalmia is 0.007% to
0.01% [18,19]. The onset of symptoms in postsurgical
sympathetic ophthalmia usually occurs between
3 weeks and 6 months post injury to the inciting
eye. Initial prodrome symptoms include photophobia,
transient obscuration of vision, and lacrimation. The
inciting eye often exhibits an inflammatory response.
Sympathetic ophthalmia presents in a variety of ways,
including an anterior chamber inflammatory reaction,
granulomatous keratic precipitates on the corneal en-
dothelium (Fig. 6), vitreous inflammation, optic nerve
involvement, and small white nodular subretinal
pigment epithelium deposits (Dalen-Fuchs nodules)
or recurrent multifocal choroiditis. Treatment options
are enucleation (there is some controversy regarding
the role of enucleation therapy) [20], oral prednisone,
and topical steroids plus cycloplegics. In prednisone-
intolerant patients, azathioprine, chlorambucil, or
cyclosporine can be used. Persistent or atypical uveitis
following PPV should alert the surgeon to the
possibility of the development of sympathetic oph-
thalmia [21].
Iris complications
The development of iris neovascularization
shortly after diabetic PPV may indicate the presence
of a peripheral retinal detachment or anterior hya-
loidal fibrovascular proliferation [22]. In nondiabetic
patients who undergo surgery for proliferative vitreo-
retinopathy, iris neovascularization is often associ-
ated with peripheral retinal detachment or residual
anterior retina after retinectomy [22]. Progressive
neovascular glaucoma and intraocular hemorrhage
after the development of iris neovascularization has
Fig. 6. Anterior chamber inflammatory reaction with
granulomatous keratic precipitates on the corneal endothe-
lium in sympathetic ophthalmia.
531
Fig. 7. Rubeosis iridis after PPV.
been reported [22]. Management should be directed
at suppressing the associated inflammation, because
there is no specific treatment for this form of iris
neovascularization if the retinal detachment is in-
operable. Panretinal photocoagulation is unlikely to
benefit patients with progressive complications of
rubeosis iridis (Fig. 7) associated with peripheral
retinal detachment, because there is no clinical or
angiographic evidence of posterior retinal ischemia in
those eyes [23]. The angiogenic stimulus probably
results from local retinal ischemia caused by the
peripheral retinal detachment or from anterior pro-
liferative vitreoretinopathy. Successful reattachment
of the peripheral retina may result in regression or
resolution of iris neovascularization, reduction of
related complications, and stabilization of the eye.
When a circumferential retinotomy needs to be per-
formed, it should include retinectomy of the residual
anterior retina to decrease the risk of iris neovascula-
rization [22].
Posterior synechiae can cause a markedly dilated,
nonreactive pupil that is cosmetically undesirable
after combined pars plana lensectomy and PPV for
complicated forms of retinal detachment. A combined
limbal and pars plana approach can be used to lyse
the posterior synechiae, resulting in partial pupillary
constriction [24].
Lens complications
Lens material after incomplete pars plana lens
removal can cause intraocular inflammation or me-
chanical damage to other tissues. Swollen cortical
material usually absorbs spontaneously, and the eye is
treated with steroids by topical application or subcon-
junctival depot injection. Every effort is made to re-
move these fragments during the initial operation [25].
Perfluorocarbon liquids (PFCL) have gained wide
acceptance in the surgical management of compli-
cated retinal detachment because of their chemical
and physical properties [1,2]. Although their high
surface tension allows closure of retinal breaks and
532 J.F. Arevalo et al / Ophthalmol Clin N Am 17 (2004) 527 – 537
prevents PFCL from flowing into the subretinal
space, particularly during intraocular manipulation,
these liquids may gain entry into the subretinal space,
resulting in residual amounts remaining in the eye.
The removal of PFCL is recommended at the end of
surgery. Singh et al [26] reported a case of intravitreal
inflammation and unusual deposits on the posterior
surface of the crystalline lens capsule and lens
epithelium (Fig. 8), which are not normally present
on the posterior surface of the lens, after PPV in
which perfluorodecalin was used as a short-term
tamponade. Intravitreal residual perfluorodecalin was
present during the course of the inflammatory
reaction, and the inflammation resolved following
the removal of this material. It was postulated that
emulsification of PFCL may have enhanced access to
tissue planes such as the retrolental space and
subretinal space, bringing the PFCL into close tissue
contact, altering the inert nature of PFCL, and
inducing a foreign body type inflammatory reaction.
Anterior segment ischemia
Anterior segment ischemia is another cause of
postoperative inflammation. The incidence is about
3% in the general retinal detachment population. This
complication should be suspected after encircling
procedures with high buckles, particularly if more
than two rectus muscles were disinserted. The anterior
ciliary arteries should be kept intact by avoiding
unnecessary tenotomy of the rectus muscles. Exces-
sive pulling on traction sutures placed around muscle
tendons or the handling of tissue should be avoided.
Extensive photocoagulation or cryotherapy must be
avoided in the area of the long ciliary arteries. Kaiser
and Trese [27] reported five cases of retinopathy of
prematurity in which anterior segment ischemia
developed after photocoagulation. Sometimes there
is a greater risk of ocular ischemia owing to a systemic
Fig. 8. (A) Slit-lamp biomicroscopic examination shows precipitates on the posterior lens capsule. (B) Photomicrograph shows
monolayer of epithelial cells (arrowhead), foam cells (F), and lens capsule (c) (hematoxilin and eosin, 300 before reduction).
(Modified from Singh J, Ramaesh K, Wharton SB, Cormack G, Chawla HB. Perfluorodecalin-induced intravitreal inflammation.
Retina 2001;21:247 – 51; with permission.)
condition of the patient, such as sickle cell anemia,
carotid disease, and vascular disease that affects blood
flow, including diabetes mellitus.
In anterior segment ischemia, a massive exudation
may occur in the anterior segment with an associated
inflammatory response. Clinically, the syndrome is
recognized by pain, corneal edema, folds in Desce-
met’s membrane, proteinaceous flare, cells in the
anterior chamber, keratic precipitates, transillumina-
tion defects of the iris pigment epithelium, posterior
and anterior synechiae, cataracts, and increased or
decreased intraocular pressure.
The pathogenesis remains unclear. The association
of anterior segment ischemia with disinsertion of
rectus muscles or damage to posterior ciliary arteries
suggests that it results from impaired blood flow in
the long posterior ciliary arteries and anterior ciliary
arteries to the ciliary body and iris. Treatment of the
ischemia is not often effective; therefore, the best
management is prevention. Mild cases respond to
medical treatment with topical or systemic steroids
and cycloplegics; in severe cases, it may be necessary
to cut the encircling band [5].
Infectious inflammation
Scleral buckling infections
The various implant materials and sutures used in
retinal detachment surgery are foreign bodies and can
become a nest for infection [28]. The organisms
responsible are frequently coagulase-positive staphy-
lococci but may be gram-negative bacteria. The major
clinical clue of infection is pain, which should be
considered a symptom of infection until proven
otherwise. Other signs to look for in scleral buckling
infection (Fig. 9) are recurrent subconjunctival hem-
orrhages, conjunctival swelling, purulent discharge,
or buckle erosion. For infections diagnosed after the
 J.F. Arevalo et al / Ophthalmol Clin N Am 17 (2004) 527 – 537
Fig. 9. Conjunctival erosion and scleral buckle infection.
first postoperative month, treatment consists of re-
moval of the scleral buckling, identification of the
organisms, and appropriate antibiotics. For extraocu-
lar infections occurring within the first postoperative
month, medical treatment may be administered to
suppress the infection and to gain sufficient time for
the retina to reattach firmly before the scleral buckling
is removed. Any evidence of intraocular infection
demands immediate removal of the scleral buckling.
When the retina has been completely reattached,
removal of the scleral buckling infrequently leads
to redettachment. If required for recurrent retinal
detachment, a repeat scleral buckling may be under-
taken as soon as all evidence of residual infection
has disappeared.
Endophthalmitis
Endophthalmitis is a rare complication after pos-
terior segment surgery. Despite major advances in
asepsis, surgical technique, and antibiotic therapy, it
remains a major concern for any ocular surgeon.
Endophthalmitis has been reported in 0.02% (2 of
10,000) of scleral buckling procedures [29] and is
believed to be associated with bacterial access
secondary to subretinal fluid drainage or air infection.
The incidence of post-PPV endophthalmitis (Fig. 10)
has been reported to be 0.07% (9 of 12,216) [30].
Endophthalmitis after pneumatic retinopexy has been
reported in 0.05% of procedures (1 of 198) [31].
Sterile technique is advised in all surgical cases,
along with the use of 5% to 10% povidone-iodine
preoperatively in the conjunctival cul-de-sac. Initial
symptoms include blurred vision and a red or painful
eye. Although thought to be a major diagnostic sign,
pain can be absent in as many as 25% of cases [32].
Poor visual acuity, hypopyon, decreased media
clarity, and poor fundus visualization are the main
clinical signs.
Three forms of clinical presentation can be dis-
tinguished. The acute form, usually fulminant, occurs
2 to 4 days postoperatively and is most frequently
caused by streptococci, Staphylococcus aureus, or
533
gram-negative organisms. The delayed form, moder-
ately severe, occurs 5 to 7 days postoperatively and is
frequently caused by Staphylococcus epidermidis,
coagulase-negative cocci, or, more rarely, fungal
species. The third form, chronic infection, can occur
as early as 1 month postoperatively. Propionibacte-
rium acnes, S epidermidis, or fungal infections are
the most common causes found in this late onset
group. If infection is strongly suspected, empiric
intravitreous antibiotics may be given after the culture
is taken. Cases with a positive aqueous tap and a
negative vitreous tap have been described, showing
that a combined vitreous and aqueous tap culture is
required [33].
The Endophthalmitis Vitrectomy Study [32]
reported that patients with initial visual acuity of
hand motions or better received no benefit from early
PPV. Patients with initial visual acuity of light
perception were three times as likely to have a final
visual acuity of 20/40 or better if they had early PPV.
Intravitreal injections of antibiotic recommended in
suspected cases of bacterial endophthalmitis include
vancomycin, 1 mg [34], plus ceftazidime, 2.25 mg
[35]. In suspected fungal endophthalmitis, intravitreal
amphotericin B, 5 to 10 mg, should be used [36].
Intravitreal dexamethasone, 0.4 mg, can also be used
with the intravitreal antibiotics if fungal endophthal-
mitis is not suspected [37].
Postoperative hypotony after posterior segment
surgery
Hypotony is a natural occurrence, symptom, and
complication of surgical treatment. With more sophis-
ticated and aggressive techniques, postsurgical hypot-
ony (Fig. 11) has been given increased attention as an
obstacle to success in retinal detachment surgery.
Hypotony (5 mm Hg or less) can be defined as the low
pressure (acute, transient, chronic, or permanent) in an
individual eye that leads to functional changes
Fig. 10. Hypopyon in endophthalmitis after PPV.
534 J.F. Arevalo et al / Ophthalmol Clin N Am 17 (2004) 527 – 537
Fig. 11. Choroidal folds and macular edema in postsur-
gical hypotony.
(asymptomatic or symptomatic) and structural changes
(reversible or irreversible). Depending on its duration
and degree, postsurgical hypotony produces tissue
changes that often are modified by, but separate from,
the tissue changes caused by an underlying disease or
its surgical treatment.
In the Silicone Study, 58 of 241 eyes (24%)
developed chronic hypotony [38]. Factors associated
with chronic hypotony were retinal detachment, a
history of ocular trauma, a history of previous PPV,
heavy photocoagulation of the retina during retinal
laser or cryotherapy treatment, scleral buckle erosion,
and proliferative vitreoretinopathy of C3 or greater
[38]. Even in eyes with attached retinas, hypotony
was a significant complication [38]. Chronic hypot-
ony was significantly associated with a poor visual
outcome when compared with that in eyes with
normal intraocular pressure (84% versus 51% with a
visual acuity of 5/200 or less) in the Silicone Study
[38]. In addition to poor vision, hypotony can result
in a painful cosmetically unacceptable eye [39].
Etiology
The cause of hypotony in the postoperative eye is
probably multifactorial. Hyposecretion of aqueous
humor occurs in most instances, but increased
uveoscleral outflow, particularly with large retinec-
tomies, may also be an important factor. Decreased
aqueous production may be a function of multiple
vitreoretinal procedures, with damage to the aqueous
production mechanisms, because hypotony is more
common in eyes requiring reoperations. The precise
cause of hypotony in eyes after successful reattach-
ment of the retina is chronic traction of the anterior
vitreous base on the ciliary body, resulting in low
detachment and hyposecretion [38]. The release of
this traction and the restoration of normal ciliary body
anatomy may result in an increase in the intraocular
pressure and an improvement in ocular function [40].
Pathophysiology
Excessive filtration
Nonhealing or persistent wound leakage may be
encountered in sclerotomy sites after conventional
PPV. In general, leaking wounds must be repaired
surgically, because they have a low rate of sponta-
neous closure and represent a significant risk for
endophthalmitis. Exceptions to this rule include a
small (eg, typically less than 2 mm long), linear,
sharp, clean wound, which may close spontaneously
with medical management alone, or a wound too
posterior for safe closure [41].
Rhegmatogenous retinal detachment can result in
relative or absolute hypotony. In these eyes, it has
been postulated that reduced intraocular pressure
results from the rerouting of aqueous outflow to the
absorbing compartment of the retinal pigment epi-
thelium and choriocapillaris [42]. Reattachment alone
can result in restoration of normal intraocular
pressure. Through a similar mechanism of rhegma-
togenous retinal detachment, retinectomy can result
in lowering of the intraocular pressure. It has even
been proposed as a means of controlling intraocular
pressure in intractable glaucoma [43]. A cyclodialysis
cleft is another so-called binternal fistulaQ allowing
posterior exit of aqueous to have a direct access to
the suprachoroidal space. Repair of a cyclodialysis
cleft by photocoagulative or other surgical techniques
can result in restoration of normal intraocular pres-
sure [40].
Decreased aqueous production
Hypotony has been a substantial problem in eyes
undergoing repair of complex retinal detachment with
silicone oil. Preoperative and postoperative anterior
proliferative vitreoretinopathy has been shown to be a
risk factor for a poor anatomic outcome [39] and a
higher incidence of hypotony [44]. A possible factor
in the development of hypotony may be the develop-
ment of anterior proliferative vitreoretinopathy or
posterior recurrent retinal detachment [45].
Active intraocular inflammation can result in
reduced ciliary body secretion and increased uveoscl-
eral outflow leading to reduce intraocular pressure
[42]. The mechanism is thought to be a prostaglan-
din-mediated phenomenon. Certainly, damage to
the ciliary body by chemical (eg, exposure to a toxic
medication), thermal (eg, cyclocryodestructive pro-
cedures), or mechanical (eg, repeated surgery) means
can result in hypotony.
An important cause of hypotony is chronic trac-
tion of the anterior vitreous base, resulting in low de-
tachment of the ciliary body and hyposecretion [38].
 J.F. Arevalo et al / Ophthalmol Clin N Am 17 (2004) 527 – 537
Coleman [46] proposed that this condition of trac-
tional hypotony be called bproliferative vitreociliopa-
thy or iridociliopathy,Q especially because it may be
present in the absence of proliferative vitreoreti-
nopathy. In his experience using high-resolution
ultrasonography to examine the anterior segment,
intraocular pressure reduction usually occurred when
at least two clock hours of tractional ciliary body
detachment were present. A multivariate analysis of
eyes in the Silicone Study lends further support to
this concept of tractional hypotony. Diffuse contrac-
tion of the retina anterior to the equator, and
presumably involving the ciliary body, remained an
independent factor prognostic of chronic hypotony
regardless of whether the retina was attached post-
operatively [38]. The Silicone Study also showed that
chronic hypotony was more prevalent in the eyes that
had received C3F8 gas tamponade versus silicone oil
(31% versus 18%) [38]. This finding may be a result
of greater epiciliary proliferation occurring in eyes
that have gas tamponade [40,44].
Management
The effective management of postoperative hypot-
ony is problematic. If hypotony develops early post-
operatively, revision of the vitrectomy with peeling of
membranes from the ciliary processes might be of
value. In addition, subconjunctival injections of a
long-acting (depot) steroid may be given. Anterior
subconjunctival injections as opposed to posterior
sub-Tenon injections are more likely to produce an
elevation (normalization) of intraocular pressure.
Once hypotony is well established, no treatment is
effective in reversing it. Some cases with silicone
oil in place and borderline intraocular pressure (5 to
10 mm Hg) become hypotonus when silicone oil is
removed. In these cases, silicone oil removal can
precipitate phthisis bulbi. The authors recommend
that one not remove silicone oil if the intraocular
pressure is under 5 mm Hg. In addition, silicone oil
can be useful to avoid phthisis bulbi in selected cases.
Although some investigators have found that laser
ablation of the trabecular meshwork to decrease the
outflow has some benefit, when patients have
peripheral anterior synechiae and closed angles, this
intervention is impossible. An endoscope can assist in
epiciliary membrane dissection. The endoscope obvi-
ates the need for scleral indentation and allows
visualization even when significant media opacity,
such as corneal edema, precludes the use of coaxial
illumination of the surgical microscope. Use of the
endoscope requires a period of training, because most
ophthalmic surgeons are unaccustomed to the lack of
535
stereopsis and the need to look at a television monitor
while performing intraocular manipulations [40].
Summary
Many indications exist for posterior segment sur-
gery, including rhegmatogenous retinal detachment,
nonresolving vitreous opacities, proliferative retino-
pathies, a posteriorly dislocated crystalline lens or
IOL, epiretinal membranes, vitreomacular traction
syndrome, vitreopapillary traction, macular holes,
subretinal neovascularization or hemorrhage, diagnos-
tic vitrectomy, ocular trauma, and endophthalmitis.
Recent advances in the instrumentation and
surgical techniques for posterior segment surgery
(especially vitrectomy) permit improved anatomic
and functional outcomes. Nevertheless, these proce-
dures may be associated with potential complications
involving the anterior segment, including postopera-
tive inflammation and hypotony.
Postoperative inflammation can produce frequent
associated complications. One must differentiate in-
fectious from noninfectious inflammation to initiate
appropriate therapy promptly. Hypotony is a frus-
trating problem because it is difficult to reverse. The
exclusion of eyes with other known causes of hypo-
tony is important before establishing the putative
mechanism of tractional ciliary body detachment
caused by epiciliary proliferative tissue.
References
[1] Charles S. Management of complications. In: Vitreous
microsurgery. 2nd edition. Baltimore7 Williams &
Wilkins; 1987. p. 207 – 14.
[2] Nussenblatt RB, Whitcup SM, Palestine AG. Postsur-
gical uveitis. In: Nussenblatt RB, Whitcup SM, Pales-
tine AG, editors. Uveitis: fundamentals and clinical
practice. 3rd edition. St. Louis7 Mosby; 2004. p. 264 – 9.
[3] Bourcier T, Monin C, Baudrimont M, Larricart P,
Borderie V, Laroche L. Conjunctival inclusion cyst
following pars plana vitrectomy. Arch Ophthalmol
2003;121(7):1067.
[4] Dadeya S, Ms K. Strabismus surgery: fibrin glue
versus vicryl for conjunctival closure. Acta Ophthal-
mol Scand 2001;79(5):515 – 7.
[5] Hartnett ME. Complications. In: Schepens CL, Hart-
nett ME, Hirose T, editors. Schepens’ retinal detach-
ment and allied diseases. 2nd edition. Woburn (MA)7
Butterworth-Heinemann; 2000. p. 721 – 40.
[6] Abrams GW, Azen SP, Barr CC, Lai MY, Hutton WL,
Trese MT, et al. The incidence of corneal abnormalities
in the Silicone Study: Silicone Study Report 7. Arch
Ophthalmol 1995;113(6):764 – 9.
536 J.F. Arevalo et al / Ophthalmol Clin N Am 17 (2004) 527 – 537
[7] Noorily SW, Foulks GN, McCuen BW. Results of
penetrating keratoplasty associated with silicone oil
retinal tamponade. Ophthalmology 1991;98(8):1186 – 9.
[8] Jaffe GJ, Schwartz D, Han DP, Gottlieb M, Hartz A,
McCarty D, et al. Risk factors for postvitrectomy fibrin
formation. Am J Ophthalmol 1990;109(6):661 – 7.
[9] Konishi KI, Kondo H, Oshima K. Iridocorneal appo-
sition after vitrectomy and gas injection. Retina 2000;
20(5):549 – 50.
[10] Jaffe GJ, Lewis H, Han DP, Williams GA, Abrams
GW. Treatment of postvitrectomy fibrin pupillary
block with tissue plasminogen activator. Am J Oph-
thalmol 1989;108(2):170 – 5.
[11] Vidaurri-Leal JS, Glaser BM. Effect of fibrin on
morphologic characteristics of retinal pigment epithe-
lial cells. Arch Ophthalmol 1984;102(9):1376 – 9.
[12] Ramsay A, Lightman S. Hypopyon uveitis. Surv Oph-
thalmol 2001;46(1):1 – 18.
[13] Moriarty AP, Spalton DJ, Shilling JS, Ffytche TJ,
Bulsara M. Breakdown of the blood-aqueous barrier
after argon laser panretinal photocoagulation for pro-
liferative diabetic retinopathy. Ophthalmology 1996;
103(5):833 – 8.
[14] Larsson LI, Nuija E. Increased permeability of the
blood-aqueous barrier after panretinal photocoagula-
tion for proliferative diabetic retinopathy. Acta Oph-
thalmol Scand 2001;79(4):414 – 6.
[15] Honavar SG, Goyal M, Majji AB, Sen PK, Naduvilath
T, Dandona L. Glaucoma after pars plana vitrectomy
and silicone oil injection for complicated retinal
detachments. Ophthalmology 1999;106(1):169 – 76.
[16] Leaver PK, Grey RH, Garner A. Silicone oil injection
in the treatment of massive preretinal retraction. II.
Late complications in 93 eyes. Br J Ophthalmol 1979;
63(5):361 – 7.
[17] Chan CC, Nussenblatt RB, Fujikawa LS, Palestine
AG, Stevens Jr G, Parver LM, et al. Sympathetic
ophthalmia: immunopathological findings. Ophthal-
mology 1986;93(5):690 – 5.
[18] Liddy L, Stuart J. Sympathetic ophthalmia in Canada.
Can J Ophthalmol 1972;7(2):157 – 9.
[19] Gass JD. Sympathetic ophthalmia following vitrec-
tomy. Am J Ophthalmol 1982;93(5):552 – 8.
[20] Marak Jr GE. Sympathetic ophthalmia. Ophthalmol-
ogy 1982;89(11):1291 – 2.
[21] Pollack AL, McDonald HR, Ai E, Green WR, Halpern
LS, Jampol LM, et al. Sympathetic ophthalmia
associated with pars plana vitrectomy without ante-
cedent penetrating trauma. Retina 2001;21(2):146 – 54.
[22] Barile GR, Chang S, Horowitz JD, Reppucci VS,
Schiff WM, Wong DT. Neovascular complications
associated with rubeosis iridis and peripheral retinal
detachment after retinal detachment surgery. Am J
Ophthalmol 1998;126(3):379 – 89.
[23] van Meurs JC, Bolt BJ, Mertens DA, Peperkamp E, De
Waard P. Rubeosis of the iris in proliferative vitreo-
retinopathy. Retina 1996;16(4):292 – 5.
[24] Michels RG, Glaser BM. Surgical treatment of post-
operative dilated, nonreactive pupil in eyes after
combined pars plana lensectomy and vitrectomy.
Retina 1990;10(3):205 – 8.
[25] Michels RG. Complications in vitreous surgery: manuals
program. San Francisco (CA)7 American Academy of
Ophthalmology; 1982. p. 108 – 21.
[26] Singh J, Ramaesh K, Wharton SB, Cormack G,
Chawla HB. Perfluorodecalin-induced intravitreal in-
flammation. Retina 2001;21(3):247 – 51.
[27] Kaiser RS, Trese MT. Iris atrophy, cataracts, and
hypotony following peripheral ablation for threshold
retinopathy of prematurity. Arch Ophthalmol 2001;
119(4):615 – 7.
[28] Hilton GF, McLean EB, Chuang EL. Complications
in retinal detachment: ophthalmology monographs.
5th edition. San Francisco (CA)7 American Academy
of Ophthalmology; 1989. p. 139 – 53.
[29] Ho PC, McMeel JW. Bacterial endophthalmitis after
retinal surgery. Retina 1983;3(2):99 – 102.
[30] Cohen SM, Flynn Jr HW, Murray TG, Smiddy WE.
Endophthalmitis after pars plana vitrectomy: the
Postvitrectomy Endophthalmitis Study Group. Oph-
thalmology 1995;102(5):705 – 12.
[31] Tornambe PE, Hilton GF. Pneumatic retinopexy: a
multicenter randomized controlled clinical trial com-
paring pneumatic retinopexy with scleral buckling.
The Retinal Detachment Study Group. Ophthalmology
1989;96(6):772 – 83; discussion, 784.
[32] Results of the Endophthalmitis Vitrectomy Study. A
randomized trial of immediate vitrectomy and of
intravenous antibiotics for the treatment of postopera-
tive bacterial endophthalmitis: Endophthalmitis Vitrec-
tomy Study Group. Arch Ophthalmol 1995;113(12):
1479 – 96.
[33] Donahue SP, Kowalski RP, Jewart BH, Friberg TR.
Vitreous cultures in suspected endophthalmitis: biopsy
or vitrectomy? Ophthalmology 1993;100(4):452 – 5.
[34] Pflugfelder SC, Hernandez E, Fliesler SJ, Alvarez J,
Pflugfelder ME, Forster RK. Intravitreal vancomycin:
retinal toxicity, clearance, and interaction with genta-
micin. Arch Ophthalmol 1987;105(6):831 – 7.
[35] D’Amico DJ, Caspers-Velu L, Libert J, Shanks E,
Schrooyen M, Hanninen L, et al. Comparative toxicity
of intravitreal aminoglycoside antibiotics. Am J Oph-
thalmol 1985;100(2):264 – 75.
[36] Axelrod AJ, Peyman GA, Apple DJ. Toxicity of
intravitreal injection of amphotericin B. Am J Oph-
thalmol 1973;76(4):578 – 83.
[37] Sunaric-Megevand G, Pournaras CJ. Current approach
to postoperative endophthalmitis. Br J Ophthalmol
1997;81(11):1006 – 15.
[38] Barr CC, Lai MY, Lean JS, Linton KL, Trese M,
Abrams G, et al. Postoperative intraocular pressure
abnormalities in the Silicone Study: Silicone Study
Report 4. Ophthalmology 1993;100(11):1629 – 35.
[39] Lewis H, Aaberg TM, Abrams GW. Causes of failure
after initial vitreoretinal surgery for severe prolifera-
tive vitreoretinopathy. Am J Ophthalmol 1991;111(1):
8 – 14.
[40] O’Connell SR, Majji AB, Humayun MS, de Juan Jr E.
 J.F. Arevalo et al / Ophthalmol Clin N Am 17 (2004) 527 – 537
The surgical management of hypotony. Ophthalmology
2000;107(2):318 – 23.
[41] Kuhl D, Mieler WF. Ciliary body. In: Kunh F,
Pieramici DJ, editors. Ocular trauma: principles and
practice. New York7 Thieme; 2002. p. 157 – 67.
[42] Schubert HD. Postsurgical hypotony: relationship to
fistulization, inflammation, chorioretinal lesions, and
the vitreous. Surv Ophthalmol 1996;41(2):97 – 125.
[43] Kirchhof B. Retinectomy lowers intraocular pressure
in otherwise intractable glaucoma: preliminary results.
Ophthalmic Surg 1994;25(4):262 – 7.
537
[44] Lewis H, Verdaguer JI. Surgical treatment for chronic
hypotony and anterior proliferative vitreoretinopathy.
Am J Ophthalmol 1996;122(2):228 – 35.
[45] Henderer JD, Budenz DL, Flynn Jr HW, Schiffman JC,
Feuer WJ, Murray TG. Elevated intraocular pressure
and hypotony following silicone oil retinal tamponade
for complex retinal detachment: incidence and risk
factors. Arch Ophthalmol 1999;117(2):189 – 95.
[46] Coleman DJ. Evaluation of ciliary body detachment
in hypotony. Retina 1995;15(4):312 – 8.
 Ophthalmol Clin N Am 17 (2004) 539 – 543

Anterior ischemia after posterior segment surgery

Guri Bronner, MD, Marco A. Zarbin, MD, PhD, Neelakshi Bhagat, MD*
Institute of Ophthalmology and Visual Science, New Jersey Medical School, 90 Bergen Street, Suite 6168,
Newark, NJ 07103, USA

Anterior segment ischemia is an infrequent but
potentially serious sequela of posterior segment
surgery that can lead to severe visual loss in an
otherwise uncomplicated case. The cause is compro-
mised blood flow to the anterior segment. Any pro-
cess that reduces perfusion pressure in the eye can
predispose to this condition. Factors that promote an-
terior segment ischemia include interruption of blood
flow through the anterior ciliary arteries or the long
posterior ciliary arteries and compression of the vor-
tex veins [1 – 3]. Underlying systemic conditions such
as generalized atherosclerosis, dysthyroid ophthalmo-
pathy, and hematologic disorders including anemia,
increased hypercoagulability, or sickle cell disease/
trait can increase the risk for anterior segment is-
chemia [3 – 6]. Intraoperative hypotension and the use
of sympathomimetic agents can also contribute to an
increased risk for anterior segment ischemia.
Vascular anatomy of the anterior segment
Primarily the long posterior and short anterior
ciliary arteries provide blood flow to the anterior
segment. Two long posterior ciliary arteries arise from
the ophthalmic artery and course toward the anterior
This article was supported by Research to Prevent
Blindness, the New Jersey Lions Eye Research Foundation,
and the Eye Institute of New Jersey.
* Corresponding author. IOVS-NJMS, Department
of Ophthalmology, Suite 6168, 90 Bergen Street, Newark,
NJ 07103.
E-mail address: bhagatne@umdnj.edu (N. Bhagat).
segment within the sclera along the horizontal
meridians. These vessels supply approximately 30%
of the blood flow to the anterior segment [7,8]. The
anterior ciliary arteries, which arise from the muscular
branches of the ophthalmic artery, lie within the
bellies of the rectus muscles. There are two anterior
ciliary arteries per muscle, the sole exception being
the lateral rectus, which carries only one. The one
anterior ciliary artery to the lateral rectus is also
unique in that it often arises from the lacrimal artery
[9]. These arteries supply the remainder of the arterial
flow to the anterior segment. Primate studies suggest
that the anterior ciliary arterial system supplies 70% to
80% of the anterior segment blood supply [7]. Just
posterior to the muscle insertions, the anterior ciliary
arteries deviate from the muscle belly and penetrate
the sclera to anastomose with the long posterior ciliary
arteries [5]. These vessels then form three distinct
circulations: the episcleral circle, the intramuscular
circle of the iris, which sits within the ciliary muscle,
and the major arterial circle. Even a small disruption
in the anterior segment circulation in some patients
can give rise to anterior segment ischemia [5]. The
conjunctival perilimbal Tenon’s plexus may also
contribute to the anterior segment blood supply [10].
The venous drainage of the anterior segment is via
the four vortex veins, one in each quadrant [11].
Obstruction of venous blood flow (eg, obstruction of
vortex veins that reduces uveal blood flow) can also
predispose to anterior segment ischemia.
Clinical presentation
The clinical presentation of anterior segment is-
chemia is varied and can be subtle. Patients usually

0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.003 ophthalmology.theclinics.com
540 G. Bronner et al / Ophthalmol Clin N Am 17 (2004) 539 – 543
present with mild pain and reduced visual acuity
beginning 1 or 2 days after surgery. Early signs in-
clude a poorly reactive pupil, episcleral vessel di-
lation, conjunctival chemosis, corneal edema with or
without striate keratopathy, mild iritis, a fibrinoid
reaction, hypopyon, rubeosis iridis, and neovascu-
larization of the anterior chamber angle [12]. In two
thirds of eyes with anterior segment ischemia and
neovascularization of the angle, the intraocular
pressure may be low or normal, even if 360 degrees
of peripheral anterior synechiae are present [13]. In
that setting, decreased aqueous production secondary
to ciliary body hypoperfusion tends to prevent
elevation of intraocular pressure. Corneal thickening
may result in shallowing of the anterior chamber.
Sloughing of a portion of the anterior sclera may be
present [12].
Anterior segment ischemia after posterior seg-
ment surgery can be difficult to detect, because many
of the clinical signs are among the expected post-
operative findings [14]. Furthermore, iris ischemia
causing segmental defects of reactivity, possibly
with peaking, as well as sluggish reactivity can be
masked by the postoperative use of cycloplegic
agents [13]. Late findings include posterior synechiae
and iris atrophy, chronic iridocyclitis, keratopathy,
anterior cortical cataract, hypotony, and, finally, phthi-
sis [13,15,16].
Predisposing conditions
To avoid precipitating anterior segment ischemia,
one must recognize underlying systemic medical
conditions that predispose the patient to this con-
dition. Older age, generalized atherosclerosis, blood
dyscrasias, hematologic disorders that increase serum
viscosity, dysthyroid ophthalmopathy, carotid artery
disease, and carotid cavernous sinus fistula can place
patients at risk for anterior segment ischemia [10].
Patients with sickle cell disease, sickle cell trait, or
sickle thalassemia have a particularly increased risk
for anterior segment ischemia. Intravascular sickling
results in hemostasis and thrombosis and can be
precipitated or exacerbated by hypoxia, hypotension,
or acidosis [6]. Anterior segment ischemia is a com-
plication of retinal reattachment surgery in individ-
uals with sickle cell disease and rarely in those with
sickle cell trait [3,17]. The precipitating surgery usu-
ally involves placement of an encircling band, sig-
nificant intraocular pressure elevation, or some other
mechanical disruption of ciliary blood flow (eg, dis-
insertion of the rectus muscles).
Predisposing ophthalmic surgical procedures
The ophthalmic surgical procedures described in
the following sections and that are performed during
the course of posterior segment surgery have been
associated with anterior segment ischemia.
Rectus muscle surgery
After strabismus surgery, the incidence of anterior
segment ischemia is approximately 1 case per 13,000
procedures [18]. The classic situation in which
anterior segment ischemia develops involves strabis-
mus surgery in which three or more rectus muscles
are disinserted, leaving collateral circulation to sup-
ply the devascularized tissues [4,5,8,10,12,19]. Al-
though it is a feared consequence of strabismus
surgery, anterior segment ischemia is an extremely
rare complication of pediatric strabismus surgery
regardless of the number of muscles operated. In
contrast, adults with certain risk factors are predis-
posed to this condition following surgery on two
muscles and, rarely, even after surgery on a single
muscle [4,5,19,20]. Patients who have undergone
two-muscle vertical rectus manipulation (particularly
if thyroid eye disease is present), operations involv-
ing three or more extraocular muscles, or strabismus
surgery following encircling scleral buckling surgery
are also more likely to experience significant anterior
segment ischemia [4,5]. It is necessary to manipulate
and occasionally disinsert the rectus muscles during
placement of an encircling scleral buckle [1,12,21].
Previous rectus muscle surgery may predispose
patients to anterior segment ischemia during sub-
sequent retinal detachment repair. Closure of the
anterior ciliary arteries during strabismus surgery
presumably confers increased importance to the long
posterior ciliary arteries. These vessels are particu-
larly susceptible to damage during posterior seg-
ment surgery.
Scleral buckling
Scleral buckling procedures can compromise
anterior segment circulation in several ways. Manipu-
lation and disinsertion of the rectus muscles may
reduce anterior ciliary artery perfusion of the ante-
rior segment. The long posterior ciliary arteries are
susceptible to occlusion owing to diathermy or
transscleral cryopexy [22]. Compromise of both cir-
culatory systems is more likely to result in anterior
segment ischemia [5]. In the past, scleral buckling
procedures often included disinsertion of one or more
rectus muscles [9,11,12]. Even when contemporary
 G. Bronner et al / Ophthalmol Clin N Am 17 (2004) 539 – 543
surgical techniques are used, anterior segment ische-
mia following scleral buckle surgery can arise from
occlusion of the vortex veins, which causes vascular
congestion [21]. In an experimental primate model,
cauterization of the vortex veins produced the con-
stellation of findings of anterior segment ischemia
[2]. In that model, one, two, three, or all four vortex
veins were occluded, with the resultant increasingly
severe signs.
Scleral buckle application alone can cause changes
in the arterial circulation as well. In an experimental
model using albino rabbits, cerclage decreased perfu-
sion of the iris and ciliary body [23]. The perfusion
was measured using injections of radioactive labeled
microspheres in the operated and control eyes.
Panretinal photocoagulation
Panretinal photocoagulation can cause closure of
the choriocapillaris [24]. The larger choroidal vessels
(presumably including the long posterior ciliary
arteries) usually remain patent after panretinal photo-
coagulation [24]; however, an additional insult, such
as intraoperative hypotension, may precipitate ante-
rior segment ischemia [6]. Hypotension can facilitate
closure of larger choroidal vessels in the same way
that increased intraocular pressure facilitates photo-
coagulation-induced feeder vessel closure [6].
Relatively low systemic blood pressure combined
with increased intraocular pressure during scleral
depression may facilitate closure of the long posterior
ciliary arteries in patients undergoing panretinal
photocoagulation with the indirect laser ophthalmo-
scope [25]. In one series, confluent treatment of
avascular retina for retinopathy of prematurity with
either diode or argon photoablation rarely resulted in
anterior segment ischemia [25]. Diathermy when
applied to the long posterior ciliary arteries in rabbits
can induce anterior segment necrosis [22]; therefore, it
is postulated that laser treatment of the retina over-
lying these vessels can be associated with an increased
risk for anterior segment ischemia. Nevertheless, the
rabbit does not have anterior ciliary arteries, and the
uvea is supplied by the posterior circulation [23].
Pressure from the contact lens was suggested as
the precipitating factor for anterior segment ischemia
in a patient with systemic lupus erythematosus
undergoing panretinal photocoagulation [26]. Slow
blood flow promotes laser-induced heat absorption by
erythrocytes and vascular thrombosis. Pre-existing
ischemia and the added stress of direct pressure on
the anterior ciliary vessels might have been sufficient
to surpass the threshold for laser-induced arterial
thrombosis in this patient’s eye.
541
Panretinal photocoagulation may be more likely to
cause anterior segment ischemia if orbital pressure is
high, because it may cause compression of the vortex
veins. A particularly vigorous inflammatory response
occurred following panretinal photocoagulation for
proliferative diabetic retinopathy in a patient who had
a history of scleral buckle repair for retinal detach-
ment. In that case, the use of a retrobulbar block may
have caused an increase in orbital pressure with
vortex vessel tamponade [14].
Anterior segment ischemia has also been reported
as a complication of cyclocryotherapy in patients
with neovascular glaucoma. In the reported cases,
cryotherapy was applied with a retinal probe (2.5 mm
in diameter) for 12, 1-minute applications ( 60°
degrees or 80°C) over the entire 360-degree
circumference at the globe [27].
Vitrectomy
Anterior segment ischemia is an unusual compli-
cation of vitrectomy performed for eyes with sickle
cell disease [6,28]. In one case, panretinal photo-
coagulation was applied to seal retinal breaks and to
induce regression of peripheral neovascularization
[6,29]. Slow choroidal blood flow from the combined
effects of intraoperative hypotension and scleral
depression, photocoagulation in the meridians of the
long posterior ciliary arteries, and the tendency of
erythrocytes with sickle cell hemoglobin C to induce
thrombosis most likely precipitated anterior segment
ischemia [6]. If the anterior segment perfusion is
compromised (eg, from previous rectus muscle
surgery) and ocular perfusion is compromised during
surgery (eg, from hypotension or unduly increased
intraocular pressure or orbital pressure), sickling and
anterior segment ischemia might result in a patient
with sickle cell hemoglobinopathy.
Reducing the risk of anterior segment ischemia
Risk assessment and careful planning can mini-
mize the risk for anterior segment ischemia after
ophthalmic surgery. Anterior segment ischemia is a
rare complication, and one should not compromise
necessary treatment; however, modifications in the
treatment plan may be justified to reduce the risk for
this condition.
Patients with vasculopathic risk factors who have
strabismus may benefit from alternatives to conven-
tional rectus muscle surgery. These strategies include
staged botulinum toxin injections and staged surgery,
which avoids disinsertion of multiple muscles in
542 G. Bronner et al / Ophthalmol Clin N Am 17 (2004) 539 – 543
1 day. Another technique is anterior ciliary sparing
rectus surgery with careful dissection and preserva-
tion of the vessels [5].
When treating patients who are prone to anterior
segment hypoperfusion (eg, those with severe carotid
occlusive disease), extra care must be taken to prevent
ischemia. Direct pressure by a contact lens during
panretinal photocoagulation should be minimized in
such patients. The use of a retrobulbar block has been
implicated as an additional precipitating factor in
patients with multiple risk factors for anterior segment
ischemia and should be avoided [14]. In addition, in
patients at risk, laser photocoagulation should be
staged or cryotherapy used to treat the horizontal
meridian, because freezing causes less damage to
ocular vessels than does photocoagulation [30].
Procedures known to lower anterior segment
perfusion, such as encircling scleral buckles, should
be avoided if possible. In such cases, direct internal
relief of vitreoretinal traction or segmental buckles
might be a better alternative. One should consider
sparing the 3 o’clock and 9 o’clock meridians during
panretinal photocoagulation in susceptible patients
(eg, those with sickle cell hemoglobin C disease).
Preoperative use of sympathomimetic agents should
be excluded if possible. The intraoperative blood
pressure and oxygen saturation should be monitored
carefully and maintained in a normal range through-
out the procedure.
In patients at relatively high risk for anterior
segment ischemia (eg, those with sickle cell hemo-
globin C disease undergoing vitrectomy with pan-
retinal photocoagulation and scleral buckling), one
should consider aggressive postoperative manage-
ment in addition to intraoperative precautions. For
example, such patients can be admitted to the hospital
after surgery and treated with face-mask oxygen and
eye-mask oxygen delivery systems as well as
frequent topical corticosteroid drops [6,31,32].
Treatment
Often, anterior segment ischemia resolves over
several weeks with minimal sequelae [5,33]. Never-
theless, permanent complications such as cataract
and, rarely, phthisis can occur. Although the value of
steroid treatment is unproved, topical and systemic
steroids are often used in the acute phase of anterior
segment ischemia [18]. Cycloplegic topical agents are
used for patient comfort and to prevent posterior
synechiae. Hyperbaric oxygen therapy has been used
[32]. In patients with significant signs of anterior
segment ischemia, treatment with face-mask oxygen
and eye-mask oxygen delivery systems may be
helpful [31,32]. The latter device increases oxygen
tension in the anterior chamber via diffusion across
the cornea. The value of blood transfusion in patients
with sickle cell hemoglobinopathy is unclear unless
the hemoglobin level is abnormally low. Additional
steps in treatment vary with the predisposing factor.
Patients who present following placement of an
encircling scleral buckle may benefit from cutting
or removal of the explant [23].
Summary
Anterior segment ischemia after posterior segment
surgery is a rare complication with a broad spectrum
of presentations. Most often, it follows a mild self-
limited course. More severe cases are usually the re-
sult of a co-incidence of precipitating factors. Mindful
consideration in high-risk patients may result in
avoidance or minimization of anterior segment ische-
mia. Management includes nonspecific steps, such
as topical medications (ie, corticosteroids and cyclo-
plegics) and face-mask and eye-mask oxygen deliv-
ery, as well as specific steps related to the underlying
cause (eg, cutting of an encircling scleral buckle).
References
[1] Kwartz J, Charles S, McCormack P, Jackson A, Lavin
M. Anterior segment ischaemia following segmental
scleral buckling. Br J Ophthalmol 1994;78:409 – 10.
[2] Hayreh SS, Baines JAB. Occlusion of the vortex
veins. Br J Ophthalmol 1973;57:217 – 38.
[3] Cartwright MJ, Blair CJ, Combs JL, Stratford TP.
Anterior segment ischemia: a complication of retinal
detachment repair in a patient with sickle cell trait.
Ann Ophthalmol 1990;22:333 – 4.
[4] Wolf E, Wagner RS, Zarbin MA. Anterior segment
ischemia and retinal detachment after vertical rectus
muscle surgery. Eur J Ophthalmol 2000;10(1):82 – 7.
[5] Saunders RA, Bluestein EC, Wilson E, Berland JE.
Anterior segment ischemia after strabismus surgery.
Surv Ophthalmol 1994;38(5):456 – 66.
[6] Leen JS, Ratnakaram R, Del Priore LV, Bhagat N,
Zarbin MA. Anterior segment ischemia after vitrec-
tomy in sickle cell disease. Retina 2002;22(2):216 – 9.
[7] Wilcox LM, Keough EM. The contribution of blood
flow by the anterior ciliary arteries to the anterior seg-
ment in the primate eye. Exp Eye Res 1980;30:167 – 74.
[8] Virdi PS, Hayreh SS. Anterior segment ischemia after
recession of various recti: an experimental study. Oph-
thalmology 1987;94:1258 – 71.
[9] Boniuk M, Zimmerman LE. Necrosis of uvea, sclera,
 G. Bronner et al / Ophthalmol Clin N Am 17 (2004) 539 – 543
and retina following operations for retinal detachment.
Arch Ophthalmol 1961;66:671 – 93.
[10] Saunders RA, Phillips MS. Anterior segment ische-
mia after three rectus muscle surgery. Ophthalmology
1988;95(4):533 – 7.
[11] Boniuk M, Zimmerman LE. Necrosis of the iris, ciliary
body, lens and retina following scleral buckling op-
erations with circling polyethylene tubes. Trans Am
Acad Ophthalmol Otolaryngol 1961;65:671 – 93.
[12] Girard LJ, Beltranena F. Early and late complications
of extensive muscle surgery. Arch Ophthalmol 1960;
64:128 – 36.
[13] Mills RP. Anterior segment ischemia secondary to
carotid occlusive disease. J Clin Neuro-ophthalmol
1989;9:200 – 4.
[14] Lee BL, von Heuven WAJ. Hypopyon uveitis follow-
ing panretinal photocoagulation. Ophthalmic Surg
Lasers 1997;28(6):505 – 7.
[15] Young LHY, Appen RE. Ischemic oculopathy. Arch
Neurol 1981;38:358 – 61.
[16] Knox DL. Ischemic ocular inflammation. Am J Oph-
thalmol 1965;60:995 – 1001.
[17] Ryan SJ, Goldberg MF. Anterior segment ischemia
following scleral buckling in sickle cell hemoglo-
binopathy. Am J Ophthalmol 1971;72(1):35 – 50.
[18] France TD, Simon JW. Anterior segment ischemia
syndrome following muscle surgery: the AAPO&S
experience. J Pediatr Ophthalmol Strabismus 1986;23:
87 – 91.
[19] Hiatt RL. Production of anterior segment ischemia.
Trans Am Ophthalmol Soc 1977;125:87 – 102.
[20] Wilson WA, Irvine SR. Pathologic changes following
disruption of blood supply to iris and ciliary body.
Trans Am Acad Ophthalmol Otolaryngol 1955;59:
501 – 2.
[21] Ambati J, Arroyo JG. Postoperative complications of
scleral buckling surgery. Int Ophthalmol Clin 2000;
40(1):175 – 85.
[22] Freeman HM, Hawkins WR, Schepens CL. Anterior
543
segment necrosis—an experimental study. Arch Oph-
thalmol 1966;75:644 – 50.
[23] Diddie KR, Ernest T. Uveal blood flow after 360 de-
gree constriction in the rabbit. Arch Ophthalmol 1980;
98:729 – 30.
[24] Wilson DJ, Green WR. Argon laser panretinal photo-
coagulation for diabetic retinopathy: scanning electron
microscopy of human choroidal vascular casts. Arch
Ophthalmol 1987;105:239 – 42.
[25] Lambert SR, Capone A, Cingle KA, Drack AV.
Cataract and phthisis bulbi after laser photoablation
for threshold retinopathy of prematurity. Am J Oph-
thalmol 2000;129(5):585 – 91.
[26] Jost BF, Olk RJ, Patz A, Fine SL, Murphy RP. Anterior
segment ischemia following laser photocoagulation
in a patient with systemic lupus erythematosus. Br J
Ophthalmol 1988;72:11 – 6.
[27] Krupin T, Johnson MF, Becker B. Anterior segment
ischemia after cyclocryotherapy. Am J Ophthalmol
1977;84:426 – 8.
[28] Pulido JS, Flynn HW, Clarkson JG, Blankenship GW.
Pars plana vitrectomy in the management of compli-
cations of proliferative sickle retinopathy. Arch Oph-
thalmol 1988;106:1553 – 7.
[29] Farber MD, Jampol LM, Fox P. A randomized clinical
trial of scatter photocoagulation of proliferative sickle
cell retinopathy. Arch Ophthalmol 1991;109:363 – 7.
[30] Lincoff HA, McLean JM, Nano H. Cryosurgical
treatment of retinal detachment. Trans Am Acad
Ophthalmol Otolaryngol 1964;68:412 – 32.
[31] Jampol LM, Orlin C, Cohen SB. Hyperbaric and trans-
corneal delivery of oxygen to the rabbit and monkey
anterior segment. Arch Ophthalmol 1988;106:825 – 30.
[32] de Smet MD, Carruthers J, Lepawski M. Anterior
segment ischemia treated with hyperbaric oxygen. Can
J Ophthalmol 1987;22:381 – 3.
[33] Olver JM. Recovery of anterior segment circulation
after strabismus surgery in adult patients. Ophthalmol-
ogy 1992;99:305 – 15.
 Ophthalmol Clin N Am 17 (2004) 545 – 556
Anterior segment complications associated with
scleral buckling
Alexander Charonis, MD, Tom S. Chang, MD*
Doheny Retina Institute, Keck – USC School of Medicine, 1450 San Pablo Street, Los Angeles, CA 90033, USA
Since its early introduction by Custodis and
Schepens and their colleagues, scleral buckling has
been an effective surgical technique for the repair of
rhegmatogenous retinal detachment. Even after the
introduction of alternative techniques, each with dis-
tinct advantages and disadvantages, scleral buckling
remains the standard by which overall reattachment
success must be judged. This article concentrates on
the anterior segment complications of scleral buck-
ling in the early and late postoperative period.
Early complications
High intraocular pressure—glaucoma
Glaucoma occurs in a high percentage of patients
with retinal detachment. Phelps and Burton [1]
diagnosed glaucoma in 9.5% of their patients with
retinal detachment. In 7.3% of the patients, glaucoma
was present before the retinal detachment. Primary
open-angle glaucoma was the type most frequently
encountered, occurring in 4% of patients. Becker [2]
detected evidence of glaucoma in 12.3% of 530 pa-
tients with retinal detachment. These percentages
are higher than the reported prevalence of glaucoma
in the general population. Even though the presence
of glaucoma does not seem to affect the rate of retinal
reattachment, it does compromise the final visual
outcome [1]. One should always keep in mind
how often these two conditions occur together. For
* Corresponding author.
E-mail address: tomchang@hsc.usc.edu (T.S. Chang).
0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.008
the purpose of this review, the focus is on the im-
mediate intraocular pressure elevation induced by
scleral buckling.
Angle-closure glaucoma without pupillary block
Angle closure following scleral buckling has been
well established in the literature. Boniuk and Zimmer-
man [3] reported the presence of angle closure in 14 of
204 enucleated eyes after scleral buckling. Holland
and Smith [4] reported an incidence of 4.8% of angle
closure after buckling. Phelps and Burton [1] found
that 2.1% of their buckle cases resulted in angle
closure, whereas Perez et al [5] reported an overall
incidence of 1.4%. The occurrence of anterior cham-
ber shallowing without angle closure is estimated to
be considerably higher (14.4% to 50%) [6,7].
Angle closure after scleral buckling is caused by
forward rotation of the ciliary body about the scleral
spur. Perez et al [5] postulated that this forward
rotation was caused by ciliary body edema induced
by a surgical compromise in the venous outflow.
Choroidal detachments, noted in most but not all of
their patients, seemed to be contributing to this for-
ward displacement of the ciliary body. There was
a noticeable absence of iris bombe, and there was
often a space between the lens and the iris, excluding
a pupillary block component. In an animal model
using large anterior buckles to induce angle closure,
Berler and Goldstein [8] consistently demonstrated
this forward rotation of the ciliary body and found no
evidence of pupillary block. Choroidal detachment
and suprachoroidal hemorrhage rotated the ciliary
body in a similar manner, but closure was possible
even without such choroidal changes. It was con-
cluded that in this bexaggeratedQ animal model, a
ophthalmology.theclinics.com
546 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556
large buckle could cause flattening of the sclera over
the ciliary body, creating forward rotation in the
absence of choroidal congestion. Predisposing factors
for this type of angle closure include hyperopia,
aphakia, a tight encircling band, excessive cryother-
apy, and rapid decompression during subretinal fluid
drainage [9,10]. Perez et al reported that the type of
scleral buckle (implant versus exoplant) was also
important. Angle closure occurred in 3.4% of eyes
with exoplants compared with 0.6% of eyes with
implants [5].
Acute angle closure is recognized during the first
week after surgery, often as early as the first or
second postoperative day [5,10]. Diagnosis can be
difficult, because the central anterior chamber appears
deep to cursory examination, and the intraocular pres-
sure is only moderately elevated. The choroidal de-
tachment – aqueous hyposecretion cycle often results
in lower intraocular pressures than anticipated in
eyes with complete angle closure. Corneal edema is
almost always present, even when the intraocular
pressure is below 30 mm Hg. Careful slit-lamp ex-
amination and gonioscopy reveal a shallow anterior
chamber without iris bombe. In fact, the midperiph-
eral iris often falls away from the peripheral angle [5].
Occasionally, a space can be observed between the
lens and the iris. Engorged ciliary processes may be
seen. A choroidal detachment is often present, even
with a high intraocular pressure. Prolonged apposi-
tional closure of the angle can lead to permanent
synechial closure and chronic glaucoma.
Treatment is directed at deepening the anterior
chamber with cycloplegics and lowering the intra-
ocular pressure with aqueous suppressants. In addi-
tion, frequent topical corticosteroids can be used in an
attempt to prevent peripheral anterior synechiae.
Medical therapy may be attempted for up to 5 to
7 days. If this is unsuccessful, a surgical approach is
necessary. Burton and Folk [11] described a tech-
nique of peripheral iris retraction using argon laser
photocoagulation performed on 12 patients with acute
angle closure without pupillary block after scleral
buckling. Using this approach, they achieved imme-
diate reopening of the angle of some degree in all
patients. Even though the retraction produced by this
laser gonioplasty or iridoplasty is transient, one must
keep in mind that the whole condition is transient;
therefore, this temporary solution may be enough
to prevent chronic sequelae. If laser retraction is un-
successful, drainage of the suprachoroidal space with
anterior chamber reformation is recommended [12].
Synechiolysis might be necessary depending on the
chronicity of angle closure. Excessive buckle height
should also be considered, and the possibility of
loosening or cutting the encircling band entertained,
especially if repeat suprachoroidal drainage and an-
terior chamber reformation become necessary.
The prognosis is generally favorable. Prompt rec-
ognition with a high index of suspicion is critical for
optimal results.
Angle-closure glaucoma with pupillary block
This variant of angle closure has also been rec-
ognized during the postoperative period after retinal
detachment repair surgery. Frequently, it is associated
with perioperative mydriasis, with or without an
increased systemic sympathetic drive. It can be bi-
lateral. The examination is typical for a pupillary
block, with prominent iris bombe configuration in the
affected eye and an occludable angle in the fellow
eye. In contrast to the former variant of angle closure,
pupillary block acute glaucoma responds well to
conventional iridotomy. When a pupillary block
component cannot be ruled out in a case of acute
glaucoma after scleral buckling, an iridotomy is
recommended in conjunction with the institution of
medical therapy [10].
Other causes of postoperative intraocular pressure
elevation
Other causes of postoperative intraocular pressure
elevation after scleral buckling surgery include in-
flammation-induced ocular hypertension, steroid hyper-
tensive response, and, rarely, expansile intraocular gas
tamponade. These causes should be differentiated from
the aforementioned angle-closure glaucoma and should
be managed in a conventional manner.
Infection
Scleral buckling materials are foreign bodies and
are at risk for infection at any time during the
postoperative course. Early infections result from
bacterial contamination from the skin, lid margins,
and conjunctiva at the time of surgery [13]. Late
infections are most likely caused by bacteria entering
through an erosion of the overlying conjunctiva [9].
The clinical features of infected scleral buckles differ
substantially between cases employing diathermy and
scleral implants and those using cryotherapy and
scleral exoplants. Infection in the setting of diathermy
and scleral implantation is an acute clinical syndrome
manifesting within the first 4 to 9 days postopera-
tively with acute pain, proptosis, scleral abscess
formation, and severe intraocular inflammation [13].
Unless the implant is removed promptly, perforating
scleral necrosis occurs with fulminant endophthalmi-
 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556
Fig. 1. Early scleral buckle infection with conjunctival hy-
peremia and localized mucopurulent discharge. (From Regillo
CD, Benson WE. Postoperative complications. In: Regillo
CD, Benson WE, editors. Retinal detachment. Philadelphia:
Lippincott-Raven; 1998. p. 175 – 93; with permission.)
tis. Diathermy-induced scleral necrosis at the bed of
the implant is thought to be the responsible factor for
this distinct syndrome. Infections after cryotherapy
and scleral exoplants manifest as a more subacute
clinical syndrome with primarily extraocular conse-
quences (Fig. 1). The absence of scleral necrosis
protects the eye from early scleral abscess formation.
The cumulative incidence of bacterial infection
after scleral buckling has been reported to range from
0.5% to 5%. Lincoff et al [13] reported an infection
rate of 3.5%. Smiddy et al [14] reported an infection
rate of 1.1% in a retrospective review of 3000 cases.
In a prospective randomized trial, Arribas et al [15]
reported an overall infection rate of 1%, with an ad-
ditional sevenfold decrease by preoperatively soak-
ing buckling materials in antibiotic solution. Some
investigators have found the rate of infection to be
higher when silicone sponges are used rather than
solid silicone [16] and when exoplants are used rather
than implants [17]. A higher incidence of infection
has been reported in buckling reoperations [18], pro-
longed surgery [19], and when multiple buckling ele-
ments [15,18] are used.
Staphylococcus is the most common infecting
organism [13 – 15,20]. Coagulase-positive S aureus
infections appear earlier and are more severe. Oshima
et al [21] reported a high incidence of acute scleral
buckle infections from methicillin-resistant S aureus
among patients with atopic dermatitis. Coagulase-
negative S epidermidis and S albus infections mani-
fest later with a chronic granulomatous inflammatory
response leading often to extrusion. Gram-negative
organisms are more virulent and may cause early
scleral necrosis independent of whether previous
diathermy has been employed [9].
Usually, infections after cryotherapy with exo-
plants appear 2 weeks to 2 months postoperatively
547
[13]. More acute presentations have been reported.
Hitchings et al [22] reported 10 cases of exoplant
infection presenting acutely within 2 weeks of sur-
gery. In this series, pain was the earliest indicator of
infection, accompanied by marked lid edema, che-
mosis, injection, and mucopurulent discharge. Of the
seven patients who underwent intraoperative drainage
of subretinal fluid, all experienced significant intra-
ocular inflammation. Later presentations do not mani-
fest significant pain or intraocular inflammation. The
patients usually complain of a chronic irritable eye
with mild discharge and recurrent subconjunctival
hemorrhage. The infection becomes evident with
the formation of a fistula in the line of conjunctival
closure, the appearance of an external granuloma, or a
subconjunctival hemorrhagic tumor. Progression of
the infection may lead to increasing pain, chemosis,
proptosis, and, eventually, signs of intraocular inflam-
mation [13].
Treatment of infected scleral buckles requires
removal of the buckling element with its sutures
[13]. Systemic and topically administered antibiotics
provide only temporary remission in the signs and
symptoms. These agents may be employed to delay
removal of an exoplant in special circumstances.
Lincoff et al suggest that a delay to obtain better
chorioretinal adhesion is not indicated if more than
2 weeks have elapsed since the operation [13]. The
need for complete removal of the buckling materials
is underscored by an electromicroscopic study of
explanted scleral buckles. In that study, a glycocalyx
layer enveloping the exoplant was demonstrated. This
biofilm may hinder the penetration of antibiotics,
effectively sequestering a reservoir of pathogenic
microorganisms [23]. The reported risk of redetach-
ment after scleral buckle removal ranges from 4% to
33% [16,17]. The risk of redetachment is highest for
eyes with persistent vitreous traction, proliferative
vitreoretinopathy, acute scleral buckle infection as
opposed to extrusion without infection, and eyes with
buckles in place for a shorter duration [10]. If the
retina does redetach, a reoperation can be performed
within 5 to 7 days after buckle removal, by which
time the orbit will have sterilized [13].
Anterior segment ischemia—necrosis
Necrosis of anterior segment structures has been
observed after scleral buckling surgery. This ischemic
syndrome comprises a set of severe complications of
retinal detachment repair surgery and can nullify an
otherwise successful scleral buckling procedure. Its
pathogenesis has been attributed to several factors,
including disruption of the anterior ciliary arteries,
548 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556

damage to the long posterior ciliary arteries, and

placement of an encircling buckle compromising the
choroidal venous outflow [24,25]. A postmortem
study of 49 eyes after retinal detachment repair re-
vealed some evidence of anterior segment necrosis
in 4 eyes (8.2%) [26]. Zimmerman and Boniuk [3]
found evidence of ischemic necrosis of the iris and
ciliary body in 25 of 204 (12.3%) eyes enucleated
after retinal detachment repair surgery. The incidence
of clinically relevant anterior segment necrosis seems
to be significantly lower with current scleral buckling
techniques. Predisposing factors include tight poste-
rior encircling buckles [27], intraoperative disinser-
tion of multiple recti muscles, diathermy over the long
posterior ciliary arteries [28], and systemic micro-
vascular risk factors, such as in patients with sickle
cell anemia or related hemoglobinopathies [24,29].
In severe cases, the earliest finding is striate
keratopathy, manifesting 2 to 5 days postoperatively
[24]. Descemet’s folds with associated epithelial
edema are present. Marked chemosis is character-
istic. The patient may complain of eye ache. A pro-
nounced anterior uveitis is present, often with large
keratic precipitates. At this stage, this syndrome may Fig. 2. Moderately severe anterior segment necrosis with
mimic the appearance of early postoperative infec- diffuse iris atrophy and posterior synechiae. Transillumina-
tious endophthalmitis. Later on, iris necrosis devel- tion defects are present along the inferior iris. (From Regillo
ops, manifested by irregular dilatation of the pupil and CD, Benson WE. Postoperative complications. In: Regillo
shrinkage toward the area of greatest necrosis (Fig. 2). CD, Benson WE, editors. Retinal detachment. Philadelphia:
White flakes floating in the anterior chamber or Lippincott-Raven; 1998. p. 175 – 93; with permission.)
deposited on the lens capsule are diagnostic. Cataract
formation begins first on the side of greatest hypoxia
and is usually complete by 3 months. Hypotony is
often marked. The anterior chamber remains shallow, tion by the encircling buckle is illustrated by the
possibly in association with ciliary body and choroidal beneficial effect of buckle division in such cases.
detachment. Histopathologically, thrombosis of the
major arterial circle of the iris is prominent [29]. Pupillary abnormalities
Patients with sickle cell hemoglobinopathy are at
greatest risk for anterior segment ischemia, even In addition to dysfunction caused by compromised
when a nonencircling procedure is performed. Ryan blood flow to the iris and the ciliary body, diathermy,
and Goldberg [24] reported an incidence of anterior cryotherapy, and photocoagulation can independently
segment necrosis of 71% among patients with sickle cause direct damage to the ciliary nerves, resulting in
cell hemoglobinopathy in comparison with an esti- abnormalities in the sympathetic and parasympathetic
mated incidence of 3% among a general retinal innervation of the pupil [30]. Direct surgical trauma to
detachment population. In the same study, they were the inferior oblique can also induce parasympathetic
unable to correlate any of the different surgical denervation to the iris and the ciliary body. Newsome
techniques used and the development of postopera- and Einaugler [31] reported the occurrence of a tonic
tive ischemia. Nevertheless, a series of preventive pupil following retinal detachment repair surgery,
measures were introduced by these investigators, who which they attributed to a postganglionic parasympa-
underscored the efficacy of partial exchange trans- thetic denervation injury related to the operation.
fusion preoperatively. Kronfeld [32] described five cases with segmental
Mild cases of anterior segment necrosis respond to sympathetic dysfunction demonstrated by an impaired
topical and systemic corticosteroids. Severe cases reflex dilation to sensory stimulation. His patients did
require loosening or division of the encircling band show segmental sensitivity to adrenergic agents.
[9,10]. The pathogenetic role of venous obstruc- Ciliary nerve dysfunction can also manifest with a
 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556
decrease in accommodation of up to 7 D [33];
however, this change seems to be transient.
Blepharoptosis
Ptosis occurring after conventional scleral buck-
ling usually resolves within 6 months. Usually, it is
caused by injury to the neural supply of the levator
palpebrae superioris, or injury to the levator’s attach-
ments to the superior rectus muscle [34]. Care should
be taken when isolating the superior rectus muscle,
especially when stripping the intermuscular septae,
because this maneuver may cause damage to the
levator’s attachments. In any event, corrective sur-
gery is not recommended earlier than 6 months after
the original surgery.
Eyelid malpositioning
Ectropion after scleral buckling usually manifests
on the premise of senile lower lid laxity. Postoperative
soft tissue edema within the lower cul-de-sac,
symblepharon with reduced size of the lower cul-
de-sac, and surgical trauma to the attachment of the
lower tarsus to the lateral palpebral ligament can inde-
pendently result in ectropion. Early postoperatively,
conservative management with aggressive ocular lu-
brication is recommended, especially if most of the
eyelid malposition can be attributed to pronounced
postoperative soft tissue edema or chemosis. If, on the
other hand, the tarsal attachment to the palpebral
ligament has been severed, or the ectropion persists,
surgical correction may become necessary.
Entropion occurs rarely after scleral buckling and,
again, is seen in conjunction with senile lower lid
laxity. Postoperative eyelid edema and symblepharon
have been implicated in the pathogenesis of entropion
after retinal detachment repair. If it persists, surgical
correction is advised.
Symblepharon—conjunctival inclusion epithelial
cysts
A compromised conjunctiva, often in the setting
of multiple operations, as well as poor conjunctival
wound closure can result in symblepharon formation
after scleral buckling. This complication manifests
as blepharoptosis with some degree of lagophthal-
mos if it occurs in the upper lid, and as ectropion in
the lower lid. Symblepharon with severe shortening
of the cul-de-sac can cause restrictive strabismus. At-
tention to the surgical principles of conjunctival clo-
sure is of critical importance for the prevention of
this complication.
549
Conjunctival inclusion cysts form by direct im-
plantation of conjunctival epithelium and may occur
within weeks to months after surgery. They are trans-
lucent cystic lesions demonstrating minimal growth.
They may become symptomatic, causing intermittent
ocular surface irritation, especially when situated an-
teriorly. Conjunctival inclusion cysts should be differ-
entiated from suture-related granulomas. The latter
often display more prominent inflammatory signs
earlier postoperatively. No treatment is usually rec-
ommended for relatively asymptomatic conjunctival
inclusion cysts. Should surgical removal be attempted,
complete excision should be performed without rup-
ture of the cyst wall to prevent recurrences.
Late complications
Refractive errors induced by scleral buckling
Geometric alterations of scleral buckling
Conventional scleral buckling surgery for retinal
detachment alters the shape of the globe. The changes
in the geometry of the eye may induce secondary
effects, some of which are clinically important to the
patient. These changes include alterations in the axial
length, corneal shape, anterior chamber depth, ocular
volume and rigidity, and ocular growth after place-
ment of the scleral buckle. The specific type of
buckling material, the location of the scleral sutures,
and the presence of an encircling band critically affect
these parameters [35].
Radial soft silicone sponges have little effect on
the axial length of the eye. In contrast, encircling
scleral buckles can alternate the axial length of the
eye. Ultrasonographic measurements on eyes with
retinal detachment treated with an encircling buckle
have shown clinically relevant changes in the axial
length. Elongation and shortening of the anteropos-
terior diameter of the eye have been reported in the
literature [36,37]. The net effect on the axial length
depends primarily on the material, the location of the
buckle, the presence of mattress sutures invaginating
the sclera, and the degree of horizontal shortening of
the encircling band. If an encircling band is tightened
around the equator of the eye, the first effect is to
decrease the circumference of the eye in a coronal
cross-section. The eye assumes an elliptical shape on
horizontal and sagittal cross-section, with an increase
in its anteroposterior axial length. If the band is
tightened further, the eye will assume a dumbbell
configuration. Part of the scleral circumference in the
horizontal and sagittal cross-sections is used to create
this dumbbell indentation, with a resultant decrease in
550 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556
Fig. 3. (A) Normal spherical shape of the eye. (B) The
anteroposterior axial length of the eye increases at moderate
low buckles. (C) At very high encircling buckles, the eye
acquires a dumbbell shape. The axial length of the eye de-
creases. (From Thompson JT. The effects and action of scle-
ral buckles in the treatment of retinal detachment. In: Ryan
SJ, editor. Retina. St Louis: Mosby; 2001. p. 1994 – 2009;
with permission.)
the axial length of the eye (Fig. 3) [35]. Clinical [38]
and laboratory studies [38,39] have verified this
paradoxical response, with initial globe elongation
reverting to axial shortening with bhigherQ equatorial
indentations. Using an in vitro model of scleral
buckling, Harris et al [39] showed that scleral in-
dentation with mattress sutures spaced around broad
silicone exoplants caused a direct decrease in the axial
length of the eye, provided there was no concomitant
horizontal shortening. In this model, the mattress
sutures pull together the anterior and posterior
anchoring scleral sites, producing axial length short-
ening. The elongating effect of circumferential tight-
ening of the encircling buckle is counterbalanced by
the shortening effect of scleral indentation, perhaps
limiting the net axial length change in this model. Low
to moderate encircling buckles produce an increase in
the axial length. A mean elongation of 0.98 mm
(± 0.20 mm) was reported by Larsen et al [37] in a
small clinical series using an encircling 2-mm band.
Smiddy et al [40] in a prospective series of 75 eyes
reported a mean elongation of 0.99 mm using a variety
of techniques. High encircling buckles create axial
shortening. Rubin [38] in an in vitro model reported a
mean shortening of 0.35 mm using a high encircling
2-mm band.
Scleral buckling can also alter the corneal shape,
inducing clinical or subclinical astigmatic aberrations.
Keratometric studies have shown changes in the
corneal curvature after scleral buckling [40,41], even
though some investigators regard them as clinically
insignificant [7,38], at least in the majority of cases.
Anterior bhighQ radial sponges and segmental circum-
ferential buckles cause irregular indentation of the
anterior sclera, which is often transmitted to the
cornea secondary to the inelasticity of both tissues
[35]. Although some of these changes are transient,
others persist beyond 6 months, requiring therapeutic
intervention. Early studies used conventional kera-
tometry in an attempt to quantify the effects of scleral
buckling, perhaps underestimating the full spectrum
of the often irregular and peripheral astigmatic
aberrations induced.
Hayashi et al [42] used computer-assisted video-
keratography and studied separately four different
techniques of circumferential buckling. All four
techniques produced prolonged (beyond 6 months)
irregular and asymmetric corneal shape changes, with
distinguishing patterns based on the individual
techniques (Fig. 4). It was concluded that bminimalQ
buckling produced the least corneal distortion,
avoiding excessively tight and asymmetrically posi-
tioned encircling bands as well as high segmental
buckles. Kinoshita et al [43] used vector analysis of
 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556 551

Fig. 4. (A) Periodic changes in the differential map after local buckling. Steepening of the lateral cornea corresponding to the
local buckle (white arc) is observed 1 month postoperatively. This steepening decreased gradually but persisted 6 months
postoperatively. (B) Marked flattening of the middle-to-peripheral cornea with focal central steepening 1 month after encircling
(white circle). This peripheral flattening decreased considerably at 6 months. (C) Flattening of the lower cornea after encircling
with vitrectomy. (D) Marked steepening of the lateral-inferior cornea corresponding to the meridians of the segmental tire (white
arc) seen 1 and 6 months after surgery (encircling/segmental). (From Hayashi H, Hayashi K, Nakao F, Hayashi F. Corneal shape
changes after scleral buckling surgery. Ophthalmology 1997;104:831 – 37; with permission.)

corneal astigmatism after circumferential buckling
surgery. They concluded that segmental buckles of
one to less than two quadrants produced significantly
greater astigmatic changes (1.67 D) than those of less
than one quadrant (0.88 D) and those spanning two
quadrants or more (mean, 1.09 D). Although early
studies underscored the role of bhighQ anterior radial
sponges in inducing irregular, often permanent
astigmatism [41], other studies do not report higher
induced astigmatism after radial scleral buckling,
even with anterior placement of the radial sponges
[40]. There is no conclusive evidence that the overall
central corneal power changes in a clinically relevant
manner after scleral buckling [38].
Rubin reported an anterior shift in the lens-iris
diaphragm after encircling buckling in 26 eyes using
anterior chamber pachymetry and the fellow unop-
erated eyes as controls [38]. He also observed a
deeper anterior chamber preoperatively among eyes
with a retinal detachment when compared with
controls. In his retrospective study of 1477 eyes
treated with an encircling procedure, he reported a
larger mean myopic shift in the phakic eyes
( 1.70 D) when compared with the aphakic eyes
( 0.91 D), implicating the anterior displacement of
the lens as the causative factor for this phenomenon.
Similarly, other investigators have reported shallow-
ing of the anterior chamber after encircling buckling,
even though some concluded that this was a tran-
sient event [6,7]. In contrast, Larsen et al [37] did not
find any changes in the anterior chamber depth in
10 phakic eyes.
552 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556
Indentation of the eye wall from a scleral buckle
also displaces fluid from the vitreous cavity, causing a
reduction in the vitreous volume. The amount of
volume displacement can be substantial for broad and
encircling buckles. This volume change is associated
with a concomitant decrease in ocular rigidity [35].
Scleral buckling may impede ocular growth in
children, resulting in a decreased myopic shift when
compared with that in the fellow unoperated eye. In a
study of 35 eyes treated with scleral buckling at a
mean age of 11.5 years, Sato et al [44] reported
decreased myopia in a comparison with the fellow
eye, especially among children operated on at youn-
ger than 10 years of age and treated with a buckle
extending beyond two quadrants.
Refractive errors after scleral buckling
As outlined previously, conventional scleral buck-
ling can induce spherical and astigmatic refractive
errors. In general, encircling procedures produce the
greatest refractive shifts, whereas segmental and radial
buckles can cause regular or irregular astigmatism. The
extent and direction of refractive change depends on
the surgical technique. Most of the spherical refractive
shift relates to the changes in the axial length and
anterior chamber depth. The astigmatic shift is
associated with the changes in corneal curvature. The
refractive changes induced by scleral buckling usually
stabilize by 2 to 3 months postoperatively.
Most eyes that undergo encircling buckling have a
myopic refractive shift induced by axial elongation.
This myopic shift is more pronounced among phakic
eyes as a result of a forward displacement of the lens-
iris diaphragm. In a subjective assessment of buckle
height after encircling with a 2-mm band in 1477 eyes,
Rubin [38] reported that low and moderate buckles
resulted in a mean myopic shift of 1.56 D and
2.24 D in phakic eyes and 0.74 D and 1.41 D in
aphakic eyes, respectively. High buckles induced a
hyperopic shift of +0.35 D and +0.59 D in phakic and
aphakic eyes, respectively, because high encircling
buckles cause axial length shortening. In a prospective
study of 75 eyes and using a variety of exoplants,
Smiddy et al [40] reported a mean myopic shift of
2.75 D. Induced astigmatism was also reported
(mean change, 0.28 D), but was not related to the use
of a radial sponge. In the same study, a myopic shift of
0.31 D was found among nonencircled eyes.
Nonencircling buckles may cause astigmatism.
The induced changes in corneal curvature are
asymmetric and irregular and are often underesti-
mated by conventional keratometry. Vector analysis
[43] and videokeratography [42] have shown patterns
of corneal distortion, some of which persist beyond
6 months postoperatively. High anterior radial full-
thickness sponges are particularly associated with
permanent irregular astigmatism, requiring hard con-
tact lenses and scleral buckle removal. Burton [41]
and Rubin [38] in separate reports indicated that
removal of anterior radial sponges could reverse the
irregular astigmatism and restore visual function.
In addition to distance refractive changes, loss of
accommodation has been reported after scleral buck-
ling [33]. This change is usually transient. Refractive
shifts do occur post scleral buckling. Rarely, they can
cause significant visual morbidity necessitating fur-
ther interventions. Recent reports have indicated that
keratorefractive surgery may be safe and effective for
the correction of refractive errors induced by buck-
ling [45,46].
Strabismus—muscle imbalance
Strabismus has been reported in variable frequency
after conventional scleral buckling. Retrospective
studies have shown that as many as 14% of patients
demonstrate symptomatic motility imbalance after
retinal detachment repair surgery [47,48]. Prospective
studies have shown an even higher level, with as many
as 73% of patients demonstrating some degree of
duction limitation during their postoperative course
[49,50]. A high proportion of these patients are not
symptomatic. In addition, the extraocular muscle
imbalance is usually of a transient nature, resolving
in a period of 3 to 6 months. Persistent heterotropia
in primary gaze may be a significant problem after
scleral buckling and is estimated to occur in 4% to
10% of cases.
Sewell et al [48] reported that the placement of
large exoplants beneath muscles and reoperations
were the most important factors associated with
persistent diplopia. Kutschera and Antlanger [51]
reported that diplopia was more often associated with
encircling elements, buckling operations involving
temporary detachment of recti muscles, pre-existent
phorias, and reduced central vision. The role of
encircling elements was also underscored by Smiddy
et al [50] in a prospective study of 76 eyes. These
investigators found postoperative deviations to be
associated with encircling buckles but not with radial
buckles. This finding is in contrast to earlier reports
in which radial buckles, especially if placed anteriorly
and proximal to recti muscles, frequently resulted in
persistent diplopia [52]. Most investigators agree that
restrictive scarring is the major cause of permanent
strabismus after scleral buckling. Wright [53] identi-
 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556
fied a specific type of restrictive scarring, the fat
adherence syndrome, caused by surgical violation of
Tenon’s capsule and exposure of extraconal fat
forming adhesions to the muscle insertion or sclera.
Even though the etiology of persistent strabismus
after buckling remains multifactorial, meticulous
surgical technique in regards to applying bminimalQ
and batraumaticQ buckling to the extraocular muscles
and their surrounding tissues seems to be critical for
the prevention of this complication.
Initial management of postoperative heterotropia
remains conservative, because, in most cases, this
will spontaneously resolve within 3 to 6 months.
Prisms can be employed in an attempt to promote
fusion in primary position. Fison and Chignell [54]
reported an initial restoration of fusion with prisms in
11 (73%) of their cases, 5 of which required no
additional treatment. If heterotropia persists, one
could consider surgical intervention, bearing in mind
that secondary to anatomic alterations induced by
scleral buckling, standard strabismus surgical princi-
ples may not be applicable. Individualized preopera-
tive planning is necessary for optimal outcomes.
Munoz and Rosenbaum [55] underscored the impor-
tance of forced duction testing in an attempt to
differentiate between restrictive and paretic strabis-
mus. They concluded that when mechanical restric-
tions cause severe rotational limitation around the
primary positions, a surgical approach to the buckled
eye is necessary. In their experience, this mechanical
restriction was more common among patients with
vertical tropias. Eighty percent of their patients
achieved single binocular vision in primary gaze
using an adjustable suture technique. Fison and
Chignell [54] in their series employed a stepwise
approach using scleral buckle removal before any
strabismus operation. Buckle removal alone restored
binocular vision in six patients (40%), whereas
additional strabismus surgery achieved single binocu-
lar vision in three patients (20%). In their series,
20% of patients did not achieve fusion. The most
consistent and predictable results followed strabismus
surgery in the unbuckled fellow eye. Wright [53]
reported cosmetically satisfactory yet functionally
poor results in a series of seven patients with the fat
adherence variety of restriction.
Strabismus surgery after scleral buckling remains
somewhat unpredictable and technically demanding.
Several investigators have reported on the use of
botulinum toxin injections as a means of reducing
the deviation enough so that, by slowly increasing
fusional vergence amplitudes, the patient maintains
an acceptable motor alignment, perhaps avoiding
further strabismus surgery. In three case series
553
combined, 47 patients (70%) obtained significant
benefit using this approach [56,57].
Cataracts
Cataracts may develop or progress more rapidly
after retinal detachment repair [58]. A true cause-and-
effect relationship between scleral buckling and
cataract formation or acceleration when no other
complications (ie, lens touch, anterior segment ne-
crosis) are present is unclear. Retinal detachment,
per se, especially if long-standing and associated with
inflammation or ocular hypotony, has been reported to
have a cataractogenic effect [59]. Nevertheless, in a
multicenter controlled clinical trial comparing pneu-
matic retinopexy with scleral buckling for the
management of selected retinal detachments, the
incidence of cataract progression after buckling was
found to be 47% (21 of 44 eyes) versus 19% (10 of
53 eyes) after pneumatic retinopexy ( P < .05) [60]. In
the same study, only two patients (4%) in the pneu-
matic retinopexy group compared with eight patients
(18%) in the scleral buckle group required cataract
surgery in the 2-year follow-up period ( P < .05).
Interestingly, all of the eyes in the scleral buckle/
cataract extraction group had an encircling element.
The investigators hypothesized that this encircling
procedure may have caused subclinical anterior seg-
ment ischemia accelerating cataractogenesis. Never-
theless, in most cases, subsequent cataract surgery
improved the visual acuity by at least one line.
Phacoemulsification surgery can be safe and can
achieve good results in eyes with previous scleral
buckling. In a study of 47 eyes with a mean follow-up
of 2.3 years, Kerrison et al [61] reported no retinal
redetachments after phacoemulsification surgery.
Postoperative vision of 20/40 or better was achieved
in 72.3% of eyes. It is prudent to allow time for
maximal chorioretinal adhesion to form around
breaks, for the refractive status of the eye to stabilize,
and for all ocular inflammation to resolve before
considering cataract surgery in these eyes [9].
Buckle extrusion or intrusion
An infrequent but potentially serious complication
of buckling is exposure of the elements caused by
breakdown of the overlying conjunctiva and Tenon’s
layer. The rate of buckle extrusion leading to surgical
removal ranges from 1.2% to 24% in the literature and
differs significantly between studies with a variable
period of follow-up [13,16 – 18]. Nevertheless, there
seems to be an agreement of higher rates of exposure
necessitating removal when silicone sponges are used
554 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556
Fig. 5. Extruding silicone sponge exoplants. (From Regillo
CD, Benson WE. Postoperative complications. In: Regillo
CD, Benson WE, editors. Retinal detachment. Philadelphia:
Lippincott-Raven; 1998. p. 175 – 93.)
versus solid silicone exoplants (Fig. 5) [16]. Exposure
causes persistent ocular irritation and discomfort and
is often associated with a concomitant buckle infec-
tion. Pain, mucopurulent discharge, and subconjunc-
tival hemorrhage are some of the cardinal signs of
buckle exposure with or without an infection [9]. Even
if an infection is not suspected clinically, it is usually
necessary to remove the exposed elements, because
attempts at covering the defect using conjunctival
suturing, autogenous fascial grafts, or processed
pericardium often fail. The rate of redetachment after
buckle removal ranges from 4% to 33% [16,17];
therefore, patients should be observed closely in the
immediate postoperative period.
Erosion of the scleral buckle into the globe
(intrusion) is much less common in the current era of
cryotherapy and exoplants. This event is a potentially
devastating complication and has been described in
greater frequency with the use of diathermy and
encircling polyethylene tubes [62]. Predisposing
factors include multiple reoperations, thin sclera, and
tight encircling buckles. Choroidal, subretinal, or
vitreous hemorrhage with recurrent retinal detachment
and migration of the buckle into the globe are the
cardinal signs of this feared complication. Removal of
an intruding buckle without a concomitant infection
should be performed only when absolutely necessary.
Preoperative planning is of utmost importance,
because globe rupture can occur intraoperatively,
necessitating immediate intervention.
References
[1] Phelps CD, Burton TC. Glaucoma and retinal detach-
ment. Arch Ophthalmol 1977;95:418 – 22.
[2] Becker B. [In discussion, Smith JL. Retinal detachment
and glaucoma]. Trans Am Acad Ophthalmol Otolar-
yngol 1963;67:731 – 2.
[3] Boniuk M, Zimmerman LE. Pathological anatomy of
complications. In: Schepens CL, Regan CDJ, editors.
Controversial aspects of the management of retinal
detachment. Boston7 Little, Brown; 1965. p. 286.
[4] Holland PM, Smith TR. Broad scleral buckle in the
management of retinal detachments with giant tears.
Am J Ophthalmol 1977;83:518 – 25.
[5] Perez RN, Phelps CD, Burton TC. Angle-closure
glaucoma following scleral buckling operations. Trans
Am Acad Ophthalmol Otolaryngol 1976;81:247 – 52.
[6] Hartley RE, Marsh RJ. Anterior chamber depth
changes after retinal detachment. Br J Ophthalmol
1973;57:546 – 50.
[7] Fiore Jr JV, Newton JC. Anterior segment changes
following the scleral buckling procedure. Arch Oph-
thalmol 1970;84:284 – 7.
[8] Berler DK, Goldstein B. Scleral buckles and rotation of
the ciliary body. Arch Ophthalmol 1979;97:1518 – 21.
[9] Regillo CD, Benson WE. Postoperative complications.
In: Regillo CD, Benson WE, editors. Retinal detachment.
Philadelphia7 Lippincott-Raven; 1998. p. 175 – 93.
[10] Williams GA, Aaberg TM. Techniques of scleral
buckling. In: Ryan SJ, editor. Retina. St Louis7 Mosby;
2001. p. 2010 – 46.
[11] Burton TC, Folk JC. Laser iris retraction for angle-
closure glaucoma after retinal detachment surgery.
Ophthalmology 1988;95:742 – 8.
[12] Simmons RJ. Angle-closure glaucoma after scleral
buckle operations for detached retina. In: Epstein DL,
editor. Chandler and Grant’s glaucoma. Philadelphia7
Lea & Febinger; 1986. p. 279 – 83.
[13] Lincoff H, Nadel A, O’Connor P. The changing
character of the infected scleral implant. Arch Oph-
thalmol 1970;84:421 – 6.
[14] Smiddy WE, Miller D, Flynn Jr HW. Scleral buckle
removal following retinal reattachment surgery: clinical
and microbiological aspects. Ophthalmic Surg 1993;
24:440 – 5.
[15] Arribas NP, Olk RJ, Schertzer M, Okun E, Johnston
GP, Boniuk I, et al. Preoperative antibiotic soaking of
silicone sponges: does it make a difference? Ophthal-
mology 1984;91:1684 – 9.
[16] Hilton GF, Wallyn RH. The removal of scleral buckles.
Arch Ophthalmol 1978;96:2061 – 3.
[17] Ulrich RA, Burton TC. Infections following scle-
ral buckling procedures. Arch Ophthalmol 1974;92:
213 – 5.
[18] Hahn YS, Lincoff A, Lincoff H, Kreissig I. Infection
after sponge installation for scleral buckling. Am J
Ophthalmol 1979;87:180 – 5.
[19] Hadden OB. Infection after retinal detachment surgery.
Aust N Z J Ophthalmol 1986;14:69 – 71.
[20] Wiznia RA. Removal of solid silicone rubber exoplants
after retinal detachment surgery. Am J Ophthalmol
1983;95:495 – 7.
[21] Oshima Y, Ohji M, Inoue Y, Harada J, Motokura M,
Saito Y, et al. Methicillin-resistant Staphylococcus
 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556
aureus infections after scleral buckling procedures for
retinal detachments associated with atopic dermatitis.
Ophthalmology 1999;106:142 – 7.
[22] Hitchings RA, Levy IS, Chignell AH. Acute infection
after retinal detachment surgery. Br J Ophthalmol
1974;58:588 – 90.
[23] Holland SP, Pulido JS, Miller D, et al. Biofilm and
scleral buckle-associated infections: a mechanism for
persistence. Ophthalmology 1991;98:933 – 8.
[24] Ryan SJ, Goldberg MF. Anterior segment ischemia
following scleral buckling in sickle cell hemoglobi-
nopathy. Am J Ophthalmol 1971;72:35 – 50.
[25] Hayreh SS, Baines JAB. Occlusion of the vortex
veins: an experimental study. Br J Ophthalmol 1973;
57:217 – 38.
[26] Wilson DJ, Green WR. Histopathologic study of the
effect of retinal detachment surgery on 49 eyes
obtained post mortem. Am J Ophthalmol 1987;103:
167 – 79.
[27] Diddie KR, Ernest JT. Uveal blood flow after 360
degrees constriction in rabbit. Arch Ophthalmol 1980;
98:729 – 30.
[28] Freeman HM, Hawkins WR, Schepens CL. Anterior
segment necrosis: an experimental study. Arch Oph-
thalmol 1966;75:644 – 50.
[29] Eagle RC, Yanoff M, Morse PH. Anterior segment
necrosis following scleral buckling in hemoglobin SC
disease. Am J Ophthalmol 1973;75:426 – 33.
[30] Lobes LA, Bourgon P. Pupillary abnormalities induced
by argon laser photocoagulation. Ophthalmology
1985;92:234 – 6.
[31] Newsome DA, Einaugler RB. Tonic pupil following
retinal detachment surgery. Arch Ophthalmol 1971;
86:233 – 4.
[32] Kronfeld PC. Segmental impairment of pupillary
motility after operations for retinal detachment. Trans
Am Ophthalmol Soc 1961;59:239 – 51.
[33] Lerner BC, Lakhanpal V, Schocket SS. Transient
myopia and accommodative paresis following retinal
cryotherapy and panretinal photocoagulation. Am J
Ophthalmol 1984;97:704 – 8.
[34] Hartnett ME. Complications. In: Schepens CL, Hart-
nett ME, Hirose T, editors. Schepens’ retinal detachment
and allied diseases. Boston7 Butterworth-Heinemann;
2000. p. 721 – 40.
[35] Thompson JT. The effects and action of scleral buckles
in the treatment of retinal detachment. In: Ryan SJ,
editor. Retina. St Louis7 Mosby; 2001. p. 1994 – 2009.
[36] Burton TC, Herron BE, Ossoinig KC. Axial length
changes after retinal detachment surgery. Am J
Ophthalmol 1977;83:59 – 62.
[37] Larsen JS, Syrdalen P. Ultrasonographic study on
changes in axial length dimensions after encircling
procedure in retinal detachment surgery. Acta Oph-
thalmol 1979;57:337 – 43.
[38] Rubin ML. The induction of refractive errors by retinal
detachment surgery. Trans Am Ophthalmol Soc 1975;
73:452 – 90.
555
[39] Harris MJ, Blumenkrantz MS, Wittpenn J, Levada A,
Brown R, Frazier-Byrne S. Geometric alterations
produced by encircling scleral buckles: biometric and
clinical considerations. Retina 1987;7:14 – 9.
[40] Smiddy WE, Loupe DN, Michels RG, Enger C,
Glaser BM, deBustros S. Refractive changes after
scleral buckling surgery. Arch Ophthalmol 1989;107:
1469 – 71.
[41] Burton TC. Irregular astigmatism following episcleral
buckling procedure with the use of silicone rubber
sponges. Arch Ophthalmol 1973;90:447 – 8.
[42] Hayashi H, Hayashi K, Nakao F, Hayashi F. Corneal
shape changes after scleral buckling surgery. Ophthal-
mology 1997;104:831 – 7.
[43] Kinoshita M, Tanihara H, Negi A, Kawano SI,
Ishigouoka H, Ueda Y, et al. Vector analysis of corneal
astigmatism after scleral buckling surgery. Ophthal-
mologica 1994;208:250 – 3.
[44] Sato T, Kawasaki T, Okuyama M, Ideta H. Refractive
changes following scleral buckling surgery in juvenile
retinal detachment. Retina 2003;23:629 – 35.
[45] Sforza PD, Saffra NA. Laser in situ keratomileusis as
treatment for anisometropia after scleral buckling
surgery. J Cataract Refract Surg 2003;29:1042 – 4.
[46] Belda JI, Ruiz-Moreno JM, Perez-Santoya JJ, Alio JL.
Laser in situ keratomileusis to correct myopia after
scleral buckling for retinal detachment. J Cataract
Refract Surg 2003;29:1231 – 5.
[47] Kanski JJ, Elkington AR, Davies MS. Diplopia from
retinal detachment surgery. Am J Ophthalmol 1973;
76:38 – 40.
[48] Sewell JJ, Knobloch WH, Eifrig DE. Extraocular
muscle imbalance after surgical treatment for retinal
detachment. Am J Ophthalmol 1974;78:321 – 3.
[49] Mets MB, Wendell ME, Gieser RG. Ocular deviation
after retinal detachment surgery. Am J Ophthalmol
1985;99:667 – 72.
[50] Smiddy WE, Loupe D, Michels RG, Enger C, Glaser
BM, deBustros S. Extraocular muscle imbalance after
scleral buckling surgery. Ophthalmology 1989;96:
1485 – 90.
[51] Kutshera E, Antlanger H. Influence of retinal detach-
ment surgery on eye motility and binocularity. Mod
Probl Ophthalmol 1979;20:354 – 8.
[52] Wolff SM. Strabismus after retinal detachment surgery.
Trans Am Ophthalmol Soc 1983;81:182 – 92.
[53] Wright KW. The fat adherence syndrome and strabis-
mus after retina surgery. Ophthalmology 1986;93:
411 – 5.
[54] Fison PN, Chignell AH. Diplopia after retinal detach-
ment surgery. Br J Ophthalmol 1987;71:521 – 5.
[55] Munoz M, Rosenbaum AL. Long-term strabismus
complications following retinal detachment surgery.
J Pediatr Ophthalmol Strabismus 1987;24:309 – 14.
[56] Scott AB. Botulinum treatment of strabismus follow-
ing retinal detachment surgery. Arch Ophthalmol 1990;
108:509 – 10.
[57] Petitto VB, Buckley EG. Use of botulinum toxin in
556 A. Charonis, T.S. Chang / Ophthalmol Clin N Am 17 (2004) 545 – 556
strabismus after retinal detachment surgery. Ophthal-
mology 1991;98:509 – 13.
[58] AAO. The repair of retinal detachments. Ophthalmol-
ogy 1996;103:1313 – 24.
[59] Scott JD. Lens changes and retinal detachment. Trans
Ophthalmol Soc U K 1979;99:241.
[60] Tornambe PE, Hilton GF, Brinton DA, Flood TP,
Green S, Grizzard W, et al. A two-year follow-up study
of the multicenter clinical trial comparing pneumatic
retinopexy with scleral buckling. Ophthalmology
1991;98:1115 – 23.
[61] Kerrison JB, Marsh M, Stark WJ, Haller JA. Phaco-
emulsification after retinal detachment surgery. Oph-
thalmology 1996;103:216 – 9.
[62] Yoshizumi MO, Friberg T. Erosion of implants in
retinal detachment surgery. Ann Ophthalmol 1983;15:
430 – 4.
 Ophthalmol Clin N Am 17 (2004) 557 – 568
Cataracts associated with posterior segment surgery
Giacomo Panozzo, MD*, Barbara Parolini, MD
Teclo, Vitreoretinal Service, Via del Perlar 2, Verona 37135, Italy
Cataract formation is the most frequent compli-
cation of pars plana vitrectomy, even without the use
of air, gas, or silicone oil. The formation of lens
opacities may occur intraoperatively, precluding
adequate visualization during the last phases of a
long procedure, or early postoperatively, making it
difficult or impossible to complete fundus examina-
tion. Alternatively, a slow but progressive lens
opacification following vitrectomy can be a major
cause of underestimation of visual potential.
Although cataract is widely described among the
complications of vitrectomy, the precise relationship
between vitrectomy and the development of lens
opacity is not completely understood. The following
text presents an extensive literature review and the
authors’ personal experience with this complication.
Pathophysiology
The cause of nuclear sclerotic cataracts after
vitrectomy is unknown. Anterior segment surgical
procedures such as penetrating keratoplasty or
trabeculectomy are associated cataract formation
[1,2], and many investigators suggest that the natural
aging of the crystalline lens is accelerated by surgical
manipulation of the eye.
The crystalline lens responds to the cataractogenic
stimulus of vitrectomy in the acute and chronic
phases. The typical intraoperative or early post-
operative feathered subcapsular opacities usually
disappear completely. The formation of permanent
posterior subcapsular cataracts after vitrectomy is
infrequently observed [3 – 5]. More commonly, per-
* Corresponding author.
E-mail address: g.panozzo@iol.it (G. Panozzo).
0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.009
manent changes occur slowly weeks to months
following the procedure and involve the nuclear
layer. The reasons for these lenticular changes are
unclear. It is possible that contact of the posterior
capsule with the vitreous substitute (BSS [balanced
saline solution] at high flow during the procedure, air/
gas, silicone oil) produces acute damage in the
permeability of the posterior capsule, with accumu-
lation of fluid in the posterior subcapsular fibers
(a phase called blens edemaQ). This accumulation is
then reabsorbed with relative restoration of blens
balanceQ following the formation of a stable and
increasing meniscus of fluid behind the lens (the
absence of this meniscus after the use of silicone oil is
probably the reason for the frequent formation of
posterior subcapsular cataracts). A similar hypothesis
has been suggested by Hsuan et al [5]. The
permeability of the posterior capsule is altered, and,
slowly, an accumulation/denaturation of proteins in
the nucleus causes the formation of nuclear sclerotic
cataracts. Other factors that have been advocated to
contribute to lens damage after vitrectomy are
described in the following sections.
Intense light exposure
Lens exposure to the ultraviolet light of the
microscope or to the reflected light from the fiber-
optic probe may be partially responsible for cataract
formation. The mechanism may involve light inacti-
vation of the antioxidant enzyme catalase and
subsequent oxidation of lens proteins [6].
Fluid infusion into the vitreous cavity
High levels of glucose cause nonenzymatic
glycation of lens proteins and eventual cataract for-
mation [7]. This experimental finding led to the
ophthalmology.theclinics.com
558 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568
theory that lens opacities after vitrectomy could be
caused by an excessive glucose concentration in the
vitreous cavity from the infusion bottle. Nevertheless,
this hypothesis is not confirmed by the clinical
reports of de Bustros et al [4] and Cherfan et al [8]
who found no differences in cataract formation in
two groups of patients with varying glucose con-
centrations in the infusion solution. Moreover, in a
well-conducted study, Haimann and Abrams [9]
demonstrated that, although a low glucose concentra-
tion (BSS Plus, glucose = 100 mg/dL) in the infusion
bottle was related to intraoperative lens opacification,
a more elevated glucose concentration (glucose-
fortified BSS Plus, glucose = 400 mg/dL) prevented
intraoperative lens changes.
In a similar fashion, high concentrations of
oxygen in the irrigating solution could contribute to
oxidation of lens proteins, leading to an increase in
fluorescent derivates and the development of nuclear
cataracts [4].
Vitreous gel removal
Removal of the vitreous gel is probably the most
important factor influencing cataract formation. This
hypothesis is strongly supported by the fact that
vitreoretinal procedures without vitreous gel removal
seem to avoid the development of cataracts. Saito and
associates [10] reported no cataract formation in
21 eyes followed up for 6 to 24 months (mean,
9.7 months) after nonvitrectomizing vitrectomy (two-
port access to the vitreous cavity and membrane
peeling without infusion and without vitreous gel re-
moval) for idiopathic epiretinal membranes. These
results were confirmed by the same investigators [11]
in a larger series of eyes (41) with a longer follow-up
(mean, 22 months) and using Scheimpflug photo-
graphs to measure the progression of nuclear scle-
rosis. The pneumatic retinopexy procedure also
seems to have a low risk of cataract formation.
Mougharbel et al [12] observed for 2 years a group of
33 patients treated in one eye with pneumoretinopexy
(SF6 20%) for superior retinal detachment. They
compared using Scheimpflug photographs the lens
transparency of both eyes in each patient, finding no
statistical differences and no nuclear sclerotic cataract
formation in the operated eye.
Why vitreous gel removal is related to nuclear
sclerosis is unclear. Is is possible that the presence of
the vitreous gel is essential to maintain the lens
transparency, and that contact with a fluid media
alters the permeability of the posterior capsule,
modifying the lens metabolic exchange and leading
to the accumulation of insoluble proteins and pig-
ments responsible for the eventual color changes in
the nucleus. This hypothesis is supported by the
observation that vitreous gel liquefaction is often
followed by cataract formation in many physiologic
or pathologic conditions such as aging, degenerative
myopia, chronic posterior uveitis, and Wagner-Stick-
ler syndrome, among others.
Age
The development or progression of nuclear
sclerotic cataracts after vitrectomy is greatly influ-
enced by age. In 1991, Cherfan et al [8] were the first
to report a much higher incidence of cataracts in
patients older than 50 years following vitrectomy for
macular pucker. Of 100 eyes observed for an average
of 29 months, 9% of patients aged less than 50 years
experienced a significant nuclear sclerosis compared
with 68% of patients aged more than 50 years
(P < .0003). The lower incidence of cataract forma-
tion in young patients was confirmed by Melberg and
Thomas [13], even in cases of gas tamponade. In
their study, three independent masked observers
compared the progression of lens opacities (Lens
Opacities Classification System [LOCS] III) after
vitrectomy and fluid-gas exchange in two sets of
28 patients followed for a mean of 25.4 months.
Only 7% of patients younger than 50 years experi-
enced significant lens opacity in the surgical eye
when compared with the nonsurgical eye versus
79% of patients older than 50 years (P < .001). The
same results were reported by Ogura et al [14]
who prospectively evaluated lens changes after
vitrectomy on 55 patients with vitreoretinal interface
syndrome by fluorophotometric measurement of
lens autofluorescence.
Thompson [3] used linear regression analysis to
evaluate (LOCS II grading scale) the development of
lens opacity as a function of patient age in 301 eyes
observed for a mean of 2.1 years after vitrectomy. His
study not only demonstrated that the risk for nuclear
sclerotic cataracts was sixfold greater after the age of
50 years but also that, after this age, the progression
rate of cataracts was similar in patients 60, 70, or
80 years of age independent of the grade of pre-
operative lens opacity.
Incidence
Vitrectomy without tamponade
Although the development of nuclear sclerotic
cataracts is the most frequent complication of
 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568 559

Table 1 population consisted of 301 consecutive eyes. Lens

Percentage of nuclear sclerotic cataract after simple vitrec- opacities were graded on a scale from 0 to 4.0
tomy without use of vitreous tamponade (LOCS II) before and after vitrectomy and compared
Percentage with the findings in a similar group of nonoperated
Study Eyes Follow-up of cataract eyes. Follow-up ranged from 0.5 to 9.2 years (mean,
Margherio [18] 184 12 – 92 Months 12.5 15 months). Using linear regression analysis, Thomp-
McDonald [19] 33 — 39.0 son did not report an absolute value of incidence of
De Bustros [4] 75 14 Months 47.0 cataract but found a progression rate of lens opacity
Cherfan [8] 100 29 Months 80.0 that was sixfold to sevenfold greater in the operated
Helbig [15] 306 5 Years 75.0 eyes when compared with the nonoperated eyes.
Berger [20] 63 6 Months 30.0
Intravitreal gas bubbles were associated with a
Hsuan [5] 10 — 30.0
nuclear sclerosis increase of approximately 60% in
Panozzo, 305 3 Years 42.3
Parolinia
a comparison with eyes without use of a gas bubble.
a
Hsuan and coworkers [5] conducted a prospective
Unpublished data.
study to evaluate morphologic changes of the
crystalline lens by digital Scheimpflug image analysis
vitrectomy, the reported incidence is extremely in 33 consecutive phakic patients after vitrectomy. A
variable in the literature, ranging from 20% to 80%. transient posterior subcapsular cataract developed in
This variability reflects many factors—the method 89% (17 of 19) of eyes with gas tamponade within
used to define and grade the lens opacity, the length 24 hours of surgery compared with 9% (1 of 11) of
of follow-up, the mean age of patients, and the use of the nontamponated eyes. Nuclear opacity developed
vitreous tamponade (Tables 1 and 2) [15 – 27]. in 67% (12 of 18) of the tamponated cases versus
De Bustros and associates studied the relationship 30% (3 of 10) of the nontamponade cases. Eighteen
between vitrectomy without the use of tamponade percent (2 of 11) of the nontamponade cases and
and the development of cataracts. They reported the 67% (14 of 21) of the tamponade cases had cata-
development of nuclear sclerotic cataracts in 47% ract surgery after 10.7 months and 12.4 months of
of 75 eyes (35 eyes) observed for an average of follow-up, respectively.
14 months [4]. In a second report on more eyes A longer tamponade is correlated to a higher
(100) with a longer follow-up (mean, 29 months), incidence of cataract. Mulhern et al [28] compared
this percentage increased to 80% [8]. When this the use of SF6 and C3F8 for macular hole surgery,
value is compared with the 23% of nonoperated focusing on the development of lens opacities in a
eyes that experienced similar cataracts, vitrectomy short-term study with 3 months of follow-up. They
increased about four times the risk for significant found an incidence of cataract 1.5 times greater in the
lens opacities after 2 years. The development of group of eyes with long-standing tamponade. Pro-
posterior subcapsular opacity was not relevant in gression of lens opacities occurred in 55% of cases
these two studies. with C3F8 compared with 37% of cases with SF6.
A similar high incidence was reported by Helbig
et al [15] who studied the time course of cataract
formation following vitrectomy for diabetic retinopa-
Table 2
thy. The course of 306 consecutive eyes in which the Incidence of cataract after vitrectomy and use of gas
lens was retained during vitrectomy was analyzed for tamponade
subsequent cataract surgery. The proportion of eyes
Follow – up Percentage
that underwent cataract surgery after vitrectomy Study Eyes (months) of cataract
increased nearly linearly with time, approaching
Pournaras [21] 12 3.0 42.0
75% after 5 years.
Ezra [22] 46 10.3 65.0
Blumenkranz [23] 99 11.0 33.0
Following gas tamponade Tabandech [24] 62 12.0 45.0
Scott [17] 74 13.9 83.4
Gas tamponade seems to have a significant effect Ezra [25] 124 24.0 58.1
on cataract formation, increasing from three to six Thompson [26] 21 24.0 76.0
times the incidence of lens opacities after vitrectomy. Haritoglu [16] 86 32.0 84.0
Thompson [3] evaluated the rate of nuclear sclerosis Kalvodova [27] 84 36.0 67.9
in eyes as a function of patient age and the use of Panozzo, Parolinia 22 36.0 95.0
a
intravitreal gas at the time of vitrectomy. The study Unpublished data.
560 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568
The incidence of cataracts after vitrectomy with
gas tamponade increases with time and eventually
approaches 100% (Table 2) [16,17,21 – 27]. Harito-
glou et al [16] studying the long-term results of
vitrectomy for macular holes in 86 eyes reported
cataract extraction in 72 eyes (84%) 32 months after
surgery. Of a cohort of 99 eyes that underwent a
combined phaco/vitrectomy procedure, 91 were
pseudophakic after 3 years of follow-up.
Contrary to most of the literature, Helbig et al [15]
found no significant cataractous effect of intravitreal
gas when compared with balanced salt solution in a
group of 306 eyes with diabetic retinopathy followed
up for 5 years.
Following silicone oil injection
Intravitreal silicone oil increases the incidence of
cataracts following vitrectomy, but the percentage
varies greatly in the literature, ranging from 60%
[29] to 100% [30]. Histopathologic evaluation has
not demonstrated silicone oil in the crystalline lens,
and the cataract may be caused by the inhibition
of normal metabolic exchange by the silicone
bubble [31].
The cataract formation seems to be independent
from the viscosity of silicone used, from partial or
complete filling, from the postoperative positioning,
and even from the time of removal.
Perfluorocarbon
Liquid perfluorocarbon is a heavy fluid used in
vitreoretinal surgery as an intraoperative tool and
as early postoperative vitreoretinal tamponade for
complicated retinal detachments. Because this agent
is used in complex cases and is removed at the end
of surgery or soon after, its role in the formation
of postoperative cataracts is difficult to establish
and ranges from a rate of 18% to 92% at 1 year
[17,32 – 35].
Personal experience
To determine the incidence of cataract extraction
after vitrectomy in surgical practice, the authors
retrospectively reviewed the data from surgical
procedures performed over 3 years (2000 to 2002).
Excluding combined phaco/vitrectomies and proce-
dures for complex cases such as retinal detachments
and diabetic or other proliferative retinopathies, a
total of 327 eyes that underwent vitrectomy for
macular pathologies were identified. Of these,
305 procedures were performed without the use of
gas tamponade and 22 with the use of gas (the
low percentage of gas tamponade was due to the
authors’ preference of combining phaco/pars plana
vitrectomy in these cases). Ninety-three of the
patients were over 50 years of age. Among the
305 eyes undergoing simple vitrectomy, after 3 years,
42.3% (129 eyes) had cataract extraction compared
with 95% of eyes undergoing vitrectomy with gas
tamponade (21 of 22 eyes).
Cataracts and other vitreoretinal procedures
Episcleral surgery
Although there is no direct relationship between
episcleral surgery and the crystalline lens, an en-
circling buckle may cause anterior segment ische-
mia [36], possibly leading to cataract formation.
The incidence of cataracts after episcleral surgery
is unknown. Tornambe et al [37], comparing pneu-
matic retinopexy with episcleral surgery, reported
nuclear sclerotic cataract requiring surgery after
2 years of follow-up in 18% of eyes treated with
scleral buckling (4% incidence in the pneumatic
retinopexy group).
Pneumatic retinopexy
The insertion of a small gas bubble into the vit-
reous cavity probably does not significantly interfere
with the lens metabolism and leads to cataract
formation much more infrequently than does a full
gas fill procedure. Mugharbel et al [12] monitored for
2 years with a Scheimpflug camera the lens trans-
parency of both eyes of 33 patients who underwent
pneumatic retinopexy in one eye for a localized
superior retinal detachment. They found no differ-
ences in lens opacity in treated versus fellow eyes.
The 4% of cataracts reported by Tornambe et al [37]
in the multicenter study on pneumatic retinopexy may
have been related to the normal incidence of cataract
in that group of patients. Inadvertent injection of a
gas bubble into the lens may lead to cataract
formation; however, this complication is uncommon.
Management of cataracts after vitreoretinal
procedures
A few special modifications must be made to
contemporary cataract extraction procedures in eyes
with previous vitreoretinal surgery. Current phaco-
emulsification techniques with a small corneal inci-
 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568
sion, maintenance of a regular anterior chamber
during the procedure, and insertion of the intraocular
lens in the bag increase the safety of the extraction in
eyes with a history of previous posterior segment
work. The cataract surgeon should be prepared for
potentially challenging factors relative to the proce-
dure itself or to the correct intraocular lens selection
and power calculation.
Cataract extraction after previous episcleral surgery
No significant modifications in the cataract sur-
gical technique are necessary in eyes with previous
retinal reattachment surgery with scleral buckling.
The cataract can usually be extracted by an extra-
capsular technique or phacoemulsification [38].
The overall complication rate is low. Because
cryotherapy or marked intraoperative scleral inden-
tation can cause focal zonular dialysis, excessive
traumatic maneuvers on the lens during phacoemul-
sification or intraocular insertion should be avoided.
A high encircling scleral buckle can cause an
increase in corneal curvature and anterior displace-
ment of the lens with shallowing of the anterior
chamber [39]. Intraocular pressure is not altered, but
intraoperative anterior chamber loss can be prevented
using a small corneal incision. The use of visco-
elastics helps protect corneal endothelium.
Although some postoperative complications
occur, they do not seem to be related to previous
retinal surgery and usually do not prevent good vi-
sion. Immediate postoperative complications include
corneal edema, anterior chamber hyphema, vitreous
hemorrhage, and fibrinous pupillary membrane. Late
complications include isolated cases of posterior
capsular opacity, herpetic corneal ulcer, and iris pos-
terior synechia [40].
Cataracts after vitrectomy
Timing
The ideal timing of cataract removal after vitrec-
tomy is unclear. The cataract should be removed
when visual improvement is expected to be good, or
when visualization of the retinal surface is precluded.
There is no evidence that cataract extraction is a risk
factor for recurrence of the retinal pathology in eyes
with a history of vitrectomy surgery.
Numerous reports have suggested that cataract
extraction leads to visual improvement in the vast
majority of eyes with a history of previous vitreous
surgery for macular hole formation, and no rela-
tionship with hole reopening has been reported
[16,25,27]. Similar results are reported in cases of
561
epiretinal membrane removal [4]. Although there is
no supportive evidence in the literature, in the
authors’ experience, cataract extraction performed
early after vitrectomy may have an increased inflam-
matory response, with subsequent cystoid macular
edema (Irvine-Gass syndrome); therefore, cataract ex-
traction is not recommended in the first 3 months
following vitrectomy unless absolutely necessary.
Procedure
The best way to manage cataract extraction in the
vitrectomized eye was controversial in previous
decades, and many different techniques were pro-
posed [41 – 43]. Phacoemulsification and in-the-bag
intraocular lens implantation is now considered the
safest technique [44 – 47]. Nevertheless, this proce-
dure can be challenging, and the surgeon must be
prepared for some unusual surgical findings.
Absence of the central vitreous gel following
vitrectomy eliminates some of the natural support
for the crystalline lens. This deficiency can lead to
the formation of an abnormally deep anterior cham-
ber and is a frequent, if not constant, finding during
the forced fluid infusion of phacoemulsification. In
these cases, the position of the surgical instruments is
unusually tilted posteriorly and can generate intra-
operative complications. The self-sealing corneal
tunnel may remain open and leak, increasing the risk
of capsular rupture. A small corneal incision, atten-
tion to avoid pressure on the corneal side of the
tunnel, a lower infusion bottle height, and complete
lens hydrolysis may guard against these risks.
Other surgical challenges include loose zonular
fibers, a small pupil size, sudden changes in anterior
chamber depth, unusual mobility and flaccidity of
the posterior capsule [48,49], and posterior capsule
fibrosis or plaque [44,46,47]. These complications
are rare in eyes with previous limited core vitrec-
tomy [50].
Control of intraocular fluid flow is crucial to avoid
intraoperative hypotony. Some authorities suggest
performing cataract extraction (both extracapsular
and phaco) under posterior irrigation with an infusion
cannula [43]. In the authors’ experience, this is not
necessary, and infusion from the vitreous chamber
may increase the risk of anterior chamber loss during
surgical maneuvers.
When feasible, phacoemulsification should imme-
diately be followed by implant of the intraocular lens
in the capsular bag. The use of different techniques of
intraocular lens implantation depending on the post-
vitrectomy anterior segment anatomy is not discussed
in this review.
562 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568
Table 3
Visual results in eyes undergoing cataract surgery following
vitreous surgery
Percentage of eyes with
Study final vision  20/40
Chang (2002) 77.4
Grusha (1998) 72.7
Pinter (1999) 46.0
McDermott (1997) 50.0
Results
The anatomic and visual results of cataract
extraction after vitrectomy depend in large part on
previous retinal pathology but are generally satisfy-
ing, with most patients experiencing improvement.
The percentage of patients with prior vitrectomy who
improve to 20/40 or better after phacoemulsification
ranges from 46% to 70% (Table 3) [41,42,48,51].
In one study by Ahfat et al [50], Snellen visual
acuity improved in 84.6% of eyes previously treated
for a macular hole, in 85.7% of eyes treated for
bmacula-onQ retinal detachment, in 66.7% of eyes
treated for bmacula-offQ retinal detachment, and in
57.1% of eyes treated for diabetic retinopathy.
Complications
The occurrence and rate of early and late
complications are not significantly different when
compared with the outcome of cataract extraction
in nonvitrectomized eyes [42,43]. Postoperative
posterior capsule opacification is more common in
postvitrectomy than in control eyes (51% versus
21%; P = .002), especially if expandable gas or
silicone oil is used at the time of vitrectomy. Rubeosis
iridis and secondary glaucoma are rare postoperative
complications when vitrectomy is performed in
diabetic eyes.
Combined procedure of cataract extraction and
vitrectomy
Since the advent of vitreous surgery, the crystal-
line lens has been removed during vitrectomy to
allow a better view of the surgical field or to facilitate
more complete access to the vitreous, which may be
required in complicated proliferative vitreoretinopa-
thy cases [52 – 54], in infants [55 – 57], or in trauma
cases [58 – 61]. Nevertheless, the term combined
surgery mainly refers to the surgical choice of
combining, in a single surgical procedure, a conven-
tional cataract extraction with artificial lens implan-
tation and vitrectomy in cases when lens extraction is
not strictly necessary for the retinal procedure. This
approach has been proposed since the early 1990s
and is now widely used when initial lens opacity is
already present, or its evolution is expected after
vitrectomy. Combined surgery has the advantage of
single surgical intervention, cost reduction, and is less
troublesome, especially for elderly patients [62,63].
The authors have reviewed more than 40 articles
concerning combined surgery of cataract extraction
and vitrectomy published since 1991 [59,62 – 98] and
herein report a summary of our personal experience
of more than 200 cases.
Technique
Phacoemulsification and posterior chamber intra-
ocular lens implantation followed by three-port
vitrectomy (phacovitrectomy) is the preferred tech-
nique for cataract surgery. Chen and Zhang [73] have
investigated the combined operation of pars plana
lensectomy-vitrectomy, preserving the lens anterior
capsule and implanting the intraocular lens into the
ciliary sulcus. They believe that this technique is a
valuable option when cataract extraction becomes
necessary in the course of a vitrectomy. When the
necessity of lens extraction is obvious preoperatively,
the combination of phacoemulsification and in-the-
bag intraocular lens insertion is preferable. The
addition of phacoemulsification does not prolong
vitreoretinal operative time notably nor increase the
risk of intraoperative and postoperative complications
significantly [86]. A small limbal incision in phaco-
emulsification allows better control during the vitrec-
tomy procedure and ensures a watertight wound. In
addition, with the limbal approach, the posterior lens
capsule is maintained, with all the attendant advan-
tages [64].
Most surgeons simply perform the two procedures
as independent steps, beginning with cataract extrac-
tion. Possible variants are limited and depend on the
surgeon’s preference. These variants include position-
ing of the posterior infusion cannula before cataract
extraction to improve anterior chamber stability,
filling of the anterior chamber with viscoelastic fluid
until the end of vitrectomy, and intraocular lens
implantation in the bag as the last step (but before the
use of any vitreal tamponade). Reported techniques
that may also simplify surgery and reduce complica-
tions include careful curvilinear capsulorrhexis, the
use of intraocular lenses with larger optics, suturing
of cataract wounds before vitrectomy, the use of
miotics, and the avoidance of long-acting dilating
drops in patients with intravitreal gas [95].
 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568
When the scleral approach is compared with a
clear corneal incision, either technique seems safe.
Eyes with smaller clear corneal incisions and foldable
intraocular lenses have less postoperative inflamma-
tion and posterior capsule opacification, but may
experience corneal endothelial dysfunction more
frequently [97].
The authors’ technique is a normal two-step
procedure, with a superior (not temporal) corneal
self-sealing incision. A corneal suture is not used
unless necessary. The intraocular lens is positioned
before vitrectomy, and, in the authors’ experience,
even a heavy peripheral indentation to remove
vitreous base does not create anterior chamber loss
nor dislocate the intraocular lens if the capsulorrhexis
overlaps with the intraocular lens optic and a scleral
tunnel incision is used.
Complications
Intraoperative
Because phacovitrectomy can be considered as
two separate procedures, the intraoperative compli-
cations and their management do not differ from
those reported for either step separately and are not
reported herein. Accidental rupture of the posterior
capsule does not preclude proceeding with vitrectomy
if safe implantation of an intraocular lens in the bag
can still be performed. Major ruptures with lens
fragments in the vitreous chamber are usually
managed by the vitreoretinal surgeon with conven-
tional techniques. In this event, the authors recom-
mend completion of the vitrectomy and consideration
of postponing implantation of the artificial lens,
with performance of one of the multiple options
available (anterior chamber, sulcus, scleral fixation)
in a second step, because the use of vitreous tam-
ponade, already programmed or unexpected, can be
followed by lens dislocation and significant anterior
segment complications.
Postoperative
The incidence of postoperative complications is
higher and the need for additional operations greater
in eyes that require tamponade, corresponding with
the severity of each case and the complexity of
the surgical procedures [76]. Complications that
may be increased with combination surgery include
the following:
Neovascular glaucoma [64,67,89,91]. This com-
plication is reported to be more frequent in
diabetic patients undergoing combined sur-
gery. Unless it is strictly necessary, in cases of
563
advanced retinopathy, the authors recommend
postponing cataract surgery for at least for
3 months following vitrectomy.
Pupillary capture [79,94]. The incidence of
pupillary capture after phacovitrectomy and
the injection of long-acting gas is high. This
complication can be minimized by creating a
smaller capsulorrhexis, avoiding long-acting
dilating drops after surgery, face-down posi-
tioning, securing wound closure, and injecting
an air bubble into the anterior chamber to push
the iris-lens diaphragm posteriorly.
Intraocular lens dislocation [79].
Cystoid macular edema. Cystoid macular edema
is a recognized complication of cataract
surgery that does not seem to be enhanced
by the combination with vitrectomy. Ando and
coworkers [69], measuring with laser flare
and cellometry the anterior chamber level of
proteins, demonstrated that pars plana vitrec-
tomy alone created little trauma to the blood-
aqueous barrier, and that the postoperative
flare in pars plana vitrectomy alone was close
to preoperative levels. The association of pars
plana vitrectomy and lens surgery should not
be more traumatizing to the eye than lens
surgery alone. A slightly higher occurrence of
postoperative macular edema following pha-
covitrectomy when compared with vitrectomy
alone has been reported by el Aouni [75] and
Sheidow [78] and their colleagues, but this is
probably related to cataract extraction and not
to the combination of the two procedures.
Posterior capsular fibrosis. This complication
seems to be more frequent when tamponade is
used [77,89,90,94,95]. The capsular fibrosis
is presumably caused by an accumulation of
fibrin and the proliferation of stimulating fac-
tors in the narrow space between the intra-
ocular lens and air/SF6 gas bubble.
Myopic shift. Use of a gas tamponade may
increase the myopic change owing to slight
anterior displacement of the intraocular
lens [67].
Cataracts and silicone oil
Cataract extraction in silicone oil – filled eyes
Cataract formation is the most common compli-
cation of intraocular silicone oil. Preventing cataract
formation by removing the silicone oil in the first
postoperative weeks is hampered by an increased risk
564 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568
of recurrence of vitreoretinal pathology. If cataract
surgery becomes necessary, it should be performed
maintaining the silicone oil tamponade [99,100].
Although extracapsular cataract extraction has
been performed [101], the most suitable technique
is phacoemulsification with in-the-bag intraocular
lens implantation, avoiding damage to the posterior
capsule and zonules [100]. An uncomplicated pha-
coemulsification using an anterior chamber maintain-
ing system and a self-sealed corneal tunnel incision
followed by in-the-bag intraocular lens implantation
is usually not followed by anterior chamber migration
of silicone oil [101]. Silicone intraocular lenses
should be avoided.
The main postoperative complication is increased
intraocular pressure [102]. Posterior capsule opacifi-
cation occurs in all eyes within a few months and
may be resolved with a neodymium: yttrium-alumi-
num-garnet laser capsulotomy [100,101].
Combined cataract and silicone oil extraction
Combining cataract surgery with silicone oil
removal offers the advantages of a single surgery
and faster visual rehabilitation [98]. The most
common technique requires cataract extraction with
phacoemulsification and silicone oil removal via pars
plana sclerotomies, but different techniques have
been proposed [98,99,103 – 109].
Investigators have compared the pars plana
technique with transpupillary removal of the silicone
oil [98,108,109]. In view of a decreased frequency of
postoperative vitreous hemorrhage, a reduced rate of
secondary cataract, and a shorter duration of surgery
and visual rehabilitation, transpupillary drainage of
silicone oil through a planned posterior capsulotomy
compares favorably with removal of silicone oil
through pars plana sclerotomies. The frequency of
retinal redetachment did not differ significantly
between the two groups and usually occurred within
the first 3 postoperative months.
Combined cataract extraction and silicone oil
removal poses a problem when trying to determine
the power of intraocular lens to be used. Silicone oil
leads to the following artifacts in ultrasonography
[110 – 112]:
Longer eye due to slower sound speed. The
sound velocity in normal vitreous is 1532 m/s,
whereas in silicone oil – filled eyes, the veloc-
ity is 987 m/s. The sound velocity also
depends on the degree of filling of vitreous
cavity by silicone oil and whether the eye is
phakic (1139 m/s) or aphakic (1052 m/s), on
the length of the eye, and on the viscosity of
silicone oil used.
Multiple fluid interfaces.
bLighthouse artifact.Q This artifact resembles a
slightly divergent cone of light from a light-
house and is caused by the accumulation of
foreign body substances within silicone oil.
Different methods have been proposed to calcu-
late biometry in combined procedures:
Conversion factors multiplied to the axial
length measured in the presence of silicone
oil. The conversion factor is different based
on the viscosity of silicone oil. Some in-
vestigators report 0.71 for 1300 centistokes,
0.62 for 1000 centistokes, and 0.30 for
5000 centistokes [113].
Changing the velocity of the beam in a silicone
oil – filled vitreous cavity to 987 m/s. This
method is not accurate in highly myopic eyes.
Intraoperative biometry right after silicone oil
removal. This is surely the most valuable
intraocular lens measurement and is the
method that the authors use in clinical practice,
but it requires a wide availability of all powers
of intraocular lens implants.
Fellow eye calculation.
References
[1] Clarke M, Vernon S, Sheldrick J. The development of
cataract following trabeculectomy. Eye 1990;4:577 – 83.
[2] Martin T, Reed J, Legault L, et al. Cataract formation
and cataract extraction after penetrating keratoplasty.
Ophthalmology 1994;101:113 – 9.
[3] Thompson JT. The role of patient age and intraocular
gas use in cataract progression after vitrectomy for
macular holes and epiretinal membranes. Am J
Ophthalmol 2004;137:250 – 7.
[4] de Bustros S, Thompson JT, Michels RG, Enger C,
Rice TA, Glaser BM. Nuclear sclerosis after vitrec-
tomy for idiopathic epiretinal membranes. Am J
Ophthalmol 1988;105:160 – 4.
[5] Hsuan JD, Brown NA, Bron AJ, Patel CK, Rosen PH.
Posterior subcapsular and nuclear cataract after
vitrectomy. J Cataract Refract Surg 2001;27:437 – 44.
[6] Mitchell RAI. Catalase photoinactivation. Science
1965;150:74.
[7] Stevens VJ, Rouzer CA, Monnier VM, Cerami A.
Diabetic cataract formation: potential role of glyco-
sylation of lens crystalline. Proc Natl Acad Sci USA
1978;75:2918 – 22.
[8] Cherfan GM, Michels RG, de Bustros S, Enger C,
 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568
Glaser BM. Nuclear sclerotic cataract after vitrectomy
for idiopathic epiretinal membranes causing macular
pucker. Am J Ophthalmol 1991;111:434 – 8.
[9] Haimann MH, Abrams GW. Prevention of lens
opacification during diabetic vitrectomy. Ophthal-
mology 1984;91:116 – 21.
[10] Saito Y, Lewis JM, Park I, et al. Nonvitrectomizing
vitreous surgery: a strategy to prevent postoperative
nuclear sclerosis. Ophthalmology 1999;106:1541 – 5.
[11] Sawa M, Saito Y, Hayashi A, Kusaka S, Ohji M, Tano
Y. Assessment of nuclear sclerosis after nonvitrecto-
mizing vitreous surgery. Am J Ophthalmol 2001;132:
356 – 62.
[12] Mougharbel M, Koch FH, Boker T, Spitznas M. No
cataract two years after pneumatic retinopexy. Oph-
thalmology 1994;101:1191 – 4.
[13] Melberg NS, Thomas MA. Nuclear sclerotic cataract
after vitrectomy in patients younger than 50 years of
age. Ophthalmology 1995;102:1466 – 71.
[14] Ogura Y, Takanashi T, Ishigooka H, Ogino N.
Quantitative analysis of lens changes after vitrectomy
by fluorophotometry. Am J Ophthalmol 1991;111:
179 – 83.
[15] Helbig H, Kellner U, Bornfeld N, Foerster MH.
Cataract surgery and YAG-laser capsulotomy follow-
ing vitrectomy for diabetic retinopathy. Ger J
Ophthalmol 1996;5:408 – 14.
[16] Haritoglou C, Gass CA, Schaumberger M, Gandorfer
A, Ulbig MW, Kampik A. Long-term follow-up after
macular hole surgery with internal limiting membrane
peeling. Am J Ophthalmol 2002;134:661 – 6.
[17] Scott IU, Moraczewski AL, Smiddy WE, Flynn Jr
HW, Fueer WJ. Long-term anatomic and visual acuity
outcomes after initial anatomic success with macular
hole surgery. Am J Ophthalmol 2003;135:633 – 40.
[18] Margherio RR, Cox Jr MS, Trese MT, Murphy PL,
Johnson J, Minor LA. Removal of epimacular
membranes. Ophthalmology 1985;92:1075 – 83.
[19] McDonald HR, Verre WP, Aaberg TM. Surgical man-
agement of idiopathic epiretinal membranes. Oph-
thalmology 1986;93:978 – 83.
[20] Berger AS, Conway M, Del Priore LV, Walker RS,
Pollack JS, Kaplan HJ. Submacular surgery for
subfoveal choroidal neovascular membranes in
patients with presumed ocular histoplasmosis. Arch
Ophthalmol 1997;115:991 – 6.
[21] Pournaras CJ, Donati G. [Treatment of full-thickness
macular holes by vitreoretinal surgery]. Klin Mon-
atsbl Augenheilkd 1996;208:311 – 4.
[22] Ezra E, Aylward WG, Gregor ZJ. Membranectomy
and autologous serum for the retreatment of full-
thickness macular holes. Arch Ophthalmol 1997;115:
1276 – 80.
[23] Blumenkranz MS, Ohana E, Shaikh S, et al. Adjuvant
methods in macular hole surgery: intraoperative
plasma-thrombin mixture and postoperative fluid-
gas exchange. Ophthalmic Surg Lasers 2001;32:
198 – 207.
[24] Tabandeh H, Smiddy WE, Mello M, Flynn HW,
565
Feuer WJ. Outcomes of bilateral macular hole
surgery. Ophthalmic Surg Lasers 2002;33:9 – 15.
[25] Ezra E, Gregor ZJ. Surgery for idiopathic full-
thickness macular hole: two-year results of a ran-
domized clinical trial comparing natural history,
vitrectomy, and vitrectomy plus autologous serum.
Morfields Macular Hole Study Group Report no. 1.
Arch Ophthalmol 2004;122:224 – 36.
[26] Thompson JT, Glaser BM, Sjaarda RN, Murphy RP.
Progression of nuclear sclerosis and long-term visual
results of vitrectomy with transforming growth factor
beta-2 for macular holes. Am J Ophthalmol 1995;
119:48 – 54.
[27] Kalvodova B, Karel I, Dotrelova D, Stepankova J,
Moravcova Z, Diblik P. [Cataract surgery in eyes
treated with vitrectomy for idiopathic macular holes].
Cesk Slov Oftalmol 2001;57:75 – 9.
[28] Mulhern MG, Cullinane A, Cleary PE. Visual and
anatomical success with short-term macular tampon-
ade and autologous platelet concentrate. Graefes Arch
Clin Exp Ophthalmol 2000;238:577 – 83.
[29] Pang MP, Peyman GA, Kao GW. Early anterior
segment complications after silicone oil injection.
Can J Ophthalmol 1986;21:271 – 5.
[30] Karia N, Laidlaw A, West J, Ezra E, Gregor MZ.
Macular hole surgery using silicone oil tamponade.
Br J Ophthalmol 2001;85:1320 – 3.
[31] Leaver PK, Grey RH, Garner A. Silicone oil injec-
tion in the treatment of massive preretinal retraction.
II. Late complications in 93 eyes. Br J Ophthalmol
1979;63:361 – 7.
[32] Blinder KJ, Peyman GA, Desai UR, Nelson Jr NC,
Alturki W, Paris CL. Vitreon, a short-term vitreo-
retinal tamponade. Br J Ophthalmol 1992;76:525 – 8.
[33] Scott IU, Murray TG, Flynn Jr HW, Feuer WJ,
Schiffman JC. Outcomes and complications associ-
ated with giant retinal tear management using per-
fluoro-n-octane. Ophthalmology 2002;109:1828 – 33.
[34] Brazitikos PD, Androudi S, D’Amico DJ, et al.
Perfluorocarbon liquid utilization in primary vitrec-
tomy repair of retinal detachment with multiple
breaks. Retina 2003;23:615 – 21.
[35] Kertes PJ, Wafapoor H, Peyman GA, Calixto Jr N,
Thompson H. The management of giant retinal tears
using perfluoroperhydrophenanthrene: a multicenter
case series. Vitreon Collaborative Study Group.
Ophthalmology 1997;104:1159 – 65.
[36] Wilson DJ, Green WR. Histopathologic study of the
effect of retinal detachment surgery on 49 eyes
obtained post mortem. Am J Ophthalmol 1987;103:
167 – 79.
[37] Tornambe PE, Hilton GF, Brinton DA, et al. Pneu-
matic retinopexy: a two-year follow-up study of the
multicenter clinical trial comparing pneumatic reti-
nopexy with scleral buckling. Ophthalmology 1991;
98:1115 – 23.
[38] Smiddy WE, Michels RG, Stark WJ, Maumenee AE.
Cataract extraction after retinal detachment surgery.
Ophthalmology 1988;95:3 – 7.
566 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568
[39] Panozzo G, Parolini B. Relationships between vitreo-
retinal and refractive surgery. Ophthalmology 2001;
108:1663 – 8; discussion, 1668 – 9.
[40] Kang YH, Lee JH. Phacoemulsification and poste-
rior chamber intraocular lens implantation after scle-
ral buckling, vitrectomy, or both. Ophthalmic Surg
Lasers 1998;29:23 – 7.
[41] Saunders DC, Brown A, Jones NP. Extracapsular
cataract extraction after vitrectomy. J Cataract Refract
Surg 1996;22:218 – 21.
[42] Ruellan YM, Hamard H, Fu WL, Ullern M, Auclin F.
[Cataract and implantation in the vitrectomized eyes].
J Fr Ophtalmol 1993;16:315 – 9.
[43] Bao Y, Jiang Y, Li X. [Cataract extraction and
intraocular lens implantation after vitrectomy].
Zhonghua Yan Ke Za Zhi 1997;33:331 – 3.
[44] Grusha YO, Masket S, Miller KM. Phacoemulsifica-
tion and lens implantation after pars plana vitrectomy.
Ophthalmology 1998;105:287 – 94.
[45] Pinter SM, Sugar A. Phacoemulsification in eyes
with past pars plana vitrectomy: case-control study.
J Cataract Refract Surg 1999;25:556 – 61.
[46] Chang M. Outcome of phacoemulsification after
pars plana vitrectomy. Ophthalmology 2002;109:
948 – 54.
[47] Biro Z, Kovacs B. Results of cataract surgery in
previously vitrectomized eyes. J Cataract Refract
Surg 2002;28:1003 – 6.
[48] Diaz Lacalle V, Orbegozo Garate FJ, Martinez Alday
N, Lopez Garrido JA, Aramberri Agesta J. Phaco-
emulsification cataract surgery in vitrectomized eyes.
J Cataract Refract Surg 1998;24:806 – 9.
[49] Braunstein R, Ariani S. Cataract surgery results after
pars plana vitrectomy. Curr Opin Ophthalmol 2003;
14:150 – 4.
[50] Ahfat FG, Yuen CH, Groenewald CP. Phacoemulsi-
fication and intraocular lens implantation following
pars plana vitrectomy: a prospective study. Eye 2003;
17:16 – 20.
[51] McDermott ML, Puklin JE, Abrams GW, Eliott D.
Phacoemulsification for cataract following pars
plana vitrectomy. Ophthalmic Surg Lasers 1997;28:
558 – 64.
[52] Liu W, Wang J, Chen S. [Lenticular-vitreoretinal
surgery for complicated retinal detachment]. Zhong-
hua Yan Ke Za Zhi 1997;33:207 – 9.
[53] Jacobi FK, Pavlovic S. Combined temporal phaco-
emulsification and pars plana vitrectomy for the
treatment of cataract and giant retinal tear in a
buphthalmic eye. Ophthalmic Surg Lasers 1999;30:
687 – 8.
[54] Scott IU, Flynn Jr HW, Murray TG, Feuer WJ.
Outcomes of surgery for retinal detachment asso-
ciated with proliferative vitreoretinopathy using
perfluoro-n-octane: a multicenter study. Am J Oph-
thalmol 2003;136:454 – 63.
[55] Maguire AM, Trese MT. Lens-sparing vitreoretinal
surgery in infants. Arch Ophthalmol 1992;110:
284 – 6.
[56] Mittra RA, Huynh LT, Ruttum MS, et al. Visual
outcomes following lensectomy and vitrectomy for
combined anterior and posterior persistent hyper-
plastic primary vitreous. Arch Ophthalmol 1998;116:
1190 – 4.
[57] Ahmadieh H, Javadi MA, Ahmady M, et al. Primary
capsulectomy, anterior vitrectomy, lensectomy, and
posterior chamber lens implantation in children:
limbal versus pars plana. J Cataract Refract Surg
1999;25:768 – 75.
[58] Ounnas N, Ruellan YM, Camboulives D, Rozenbaum
JP. [Indications for vitrectomy in the extraction of
intraocular foreign bodies]. J Fr Ophtalmol 1983;6:
815 – 22.
[59] Lam DS, Tham CC, Kwok AK, Gopal L. Combined
phacoemulsification, pars plana vitrectomy, removal
of intraocular foreign body (IOFB), and primary
intraocular lens implantation for patients with IOFB
and traumatic cataract. Eye 1998;12(Pt 3a):395 – 8.
[60] Chaudhry NA, Belfort A, Flynn Jr HW, Tabandeh H,
Smiddy WE, Murray TG. Combined lensectomy,
vitrectomy and scleral fixation of intraocular lens
implant after closed-globe injury. Ophthalmic Surg
Lasers 1999;30:375 – 81.
[61] Batman C, Cekic O, Totan Y, Ozkan SS, Zilelioglu O.
Combined phacoemulsification, vitrectomy, foreign-
body extraction, and intraocular lens implantation.
J Cataract Refract Surg 2000;26:254 – 9.
[62] Alexandrakis G, Chaudhry NA, Flynn Jr HW, Murray
TG. Combined cataract surgery, intraocular lens
insertion, and vitrectomy in eyes with idiopathic
epiretinal membrane. Ophthalmic Surg Lasers 1999;
30:327 – 8.
[63] Foster RE, Lowder CY, Meisler DM, Zakov ZN,
Meyers SM, Ambler JS. Combined extracapsular
cataract extraction, posterior chamber intraocular lens
implantation, and pars plana vitrectomy. Ophthalmic
Surg 1993;24:446 – 52.
[64] Mamalis N, Teske MP, Kreisler KR, Zimmerman PL,
Crandall AS, Olson RJ. Phacoemulsification com-
bined with pars plana vitrectomy. Ophthalmic Surg
1991;22:194 – 8.
[65] Pagot V, Gazagne C, Galiana A, Giraud MA, Male-
caze F, Mathis A. [Extracapsular cataract extraction
and implantation in the capsular sac during vitrec-
tomy in diabetics]. J Fr Ophtalmol 1991;14:523 – 8.
[66] Leyland MD, Schulenburg WE. Combined phaco-
emulsification – vitrectomy surgery: technique, indica-
tions and outcomes. Eye 1999;13(Pt 3a):348 – 52.
[67] Suzuki Y, Sakuraba T, Mizutani H, Matsuhashi H,
Nakazawa M. Postoperative complications after
simultaneous vitrectomy and cataract surgery. Oph-
thalmic Surg Lasers 2001;32:391 – 6.
[68] Suzuki Y, Sakuraba T, Mizutani H, Matsuhashi H,
Nakazawa M. Postoperative refractive error after
simultaneous vitrectomy and cataract surgery. Oph-
thalmic Surg Lasers 2000;31:271 – 5.
[69] Ando A, Nishimura T, Uyama M. Surgical outcome
on combined procedures of lens extraction, intra-
 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568
ocular lens implantation, and vitrectomy during re-
moval of the epiretinal membrane. Ophthalmic Surg
Lasers 1998;29:974 – 9.
[70] Sourdille P. Lensectomy-vitrectomy indications and
techniques in cataract surgery. Curr Opin Ophthalmol
1997;8:56 – 9.
[71] Malinowski SM, Mieler WF, Koenig SB, Han DP,
Pulido JS. Combined pars plana vitrectomy-lensec-
tomy and open-loop anterior chamber lens implanta-
tion. Ophthalmology 1995;102:211 – 6.
[72] Ullern M, Nicol JL, Ruellan YM, Boureau C, Morel
C. [Phacoemulsification by the anterior approach
combined with vitreoretinal surgery]. J Fr Ophtalmol
1993;16:320 – 4.
[73] Chen G, Zhang X. [Clinical observation on pars plana
lensectomy-vitrectomy preserving lens anterior cap-
sule]. Zhonghua Yan Ke Za Zhi 1997;33:444 – 6.
[74] Girard P, Saade G. A triple procedure: phacoemulsi-
fication, intraocular lens implantation, and scleral
buckling surgery. Retina 1997;17:502 – 6.
[75] el Aouni A, Behar-Cohen F, Chauvaud D. [Combined
phacoemulsification/membrane resection and macular
edema]. J Fr Ophtalmol 1998;21:403 – 7.
[76] Honjo M, Ogura Y. Surgical results of pars plana
vitrectomy combined with phacoemulsification and
intraocular lens implantation for complications of
proliferative diabetic retinopathy. Ophthalmic Surg
Lasers 1998;29:99 – 105.
[77] Scharwey K, Pavlovic S, Jacobi KW. [Early poste-
rior capsule fibrosis after combined cataract and
vitreoretinal surgery with intraocular air/SF6 gas
tamponade]. Klin Monatsbl Augenheilkd 1998;212:
149 – 53.
[78] Sheidow TG, Gonder JR. Cystoid macular edema
following combined phacoemulsification and vitrec-
tomy for macular hole. Retina 1998;18:510 – 4.
[79] Johnston RL, Charteris DG, Horgan SE, Cooling RJ.
Combined pars plana vitrectomy and sutured poste-
rior chamber implant. Arch Ophthalmol 2000;118:
905 – 10.
[80] Kotecha AV, Sinclair SH, Gupta AK, Tipperman R.
Pars plana vitrectomy for macular holes combined
with cataract extraction and lens implantation. Oph-
thalmic Surg Lasers 2000;31:387 – 93.
[81] Lam DS, Tam BS, Chan WM, Bhende P. Combined
cataract extraction and submacular blood clot evac-
uation for globe perforation caused by retrobulbar
injection. J Cataract Refract Surg 2000;26:1089 – 91.
[82] Lam DS, Young AL, Rao SK, Cheung BT, Yuen CY,
Tang HM. Combined phacoemulsification, pars plana
vitrectomy, and foldable intraocular lens implanta-
tion. J Cataract Refract Surg 2003;29:1064 – 9.
[83] Simcock PR, Scalia S. Phaco-vitrectomy for full-
thickness macular holes. Acta Ophthalmol Scand
2000;78:684 – 6.
[84] Simcock PR, Scalia S. Phacovitrectomy without
prone posture for full thickness macular holes. Br J
Ophthalmol 2001;85:1316 – 9.
[85] Jackson TL, Larsson J, Tanner V, Williamson TH.
567
Combined phaco-emulsification cataract extraction
and pars plana vitrectomy without intraocular lens
implantation. Ophthalmologica 2001;215:271 – 5.
[86] Jun Z, Pavlovic S, Jacobi KW. Results of combined
vitreoretinal surgery and phacoemulsification with
intraocular lens implantation. Clin Exp Ophthalmol
2001;29:307 – 11.
[87] Amino K, Tanihara H. Vitrectomy combined with
phacoemulsification and intraocular lens implantation
for diabetic macular edema. Jpn J Ophthalmol 2002;
46:455 – 9.
[88] Chaudhry NA, Flynn HW, Murray TG, Belfort A.
Combined cataract surgery and vitrectomy for break-
through vitreous hemorrhage from age-related macu-
lar degeneration. Ophthalmic Surg Lasers 2002;33:
16 – 8.
[89] Chung TY, Chung H, Lee JH. Combined surgery and
sequential surgery comprising phacoemulsification,
pars plana vitrectomy, and intraocular lens implanta-
tion: comparison of clinical outcomes. J Cataract
Refract Surg 2002;28:2001 – 5.
[90] Gottlieb CC, Martin JA. Phacovitrectomy with
internal limiting membrane peeling for idiopathic
macular hole. Can J Ophthalmol 2002;37:277 – 82;
discussion, 282.
[91] Hui Y, Wang L, Huang W, Han Q, Wang Y.
[Simultaneous intraocular lens implantation during
vitreous surgery for treatment of proliferative diabetic
retinopathy]. Zhonghua Yan Ke Za Zhi 2002;38:
598 – 602.
[92] Lahey JM, Francis RR, Fong DS, Kearney JJ, Tanaka
S. Combining phacoemulsification with vitrectomy
for treatment of macular holes. Br J Ophthalmol
2002;86:876 – 8.
[93] Lahey JM, Francis RR, Kearney JJ. Combining pha-
coemulsification with pars plana vitrectomy in pa-
tients with proliferative diabetic retinopathy: a series
of 223 cases. Ophthalmology 2003;110:1335 – 9.
[94] Rahman R, Rosen PH. Pupillary capture after
combined management of cataract and vitreoretinal
pathology. J Cataract Refract Surg 2002;28:1607 – 12.
[95] Demetriades AM, Gottsch JD, Thomsen R, et al.
Combined phacoemulsification, intraocular lens
implantation, and vitrectomy for eyes with coexisting
cataract and vitreoretinal pathology. Am J Ophthal-
mol 2003;135:291 – 6.
[96] Douglas MJ, Scott IU, Flynn Jr HW. Pars plana
lensectomy, pars plana vitrectomy, and silicone oil
tamponade as initial management of cataract and
combined traction/rhegmatogenous retinal detach-
ment involving the macula associated with severe
proliferative diabetic retinopathy. Ophthalmic Surg
Lasers Imaging 2003;34:270 – 8.
[97] Heiligenhaus A, Holtkamp A, Koch J, Schilling H,
Bornfeld N, Losche C, et al. Combined phacoemulsi-
fication and pars plana vitrectomy: clear corneal
versus scleral incisions. Prospective randomized
multicenter study. J Cataract Refract Surg 2003;29:
1106 – 12.
568 G. Panozzo, B. Parolini / Ophthalmol Clin N Am 17 (2004) 557 – 568
[98] Jonas JB, Budde WM, Panda-Jonas S. Cataract sur-
gery combined with transpupillary silicone oil removal
through planned posterior capsulotomy. Ophthalmol-
ogy 1998;105:1234 – 7; discussion, 1237 – 8.
[99] Tanner V, Haider A, Rosen P. Phacoemulsification
and combined management of intraocular silicone oil.
J Cataract Refract Surg 1998;24:585 – 91.
[100] Grewing R, Mester U. [Therapeutic possibilities in
lens opacity after silicone oil tamponade]. Klin
Monatsbl Augenheilkd 1992;200:30 – 2.
[101] Weinberger D, Kremer I, Lichter H, Axer-Siegel R,
Yassur Y. Extracapsular cataract extraction and intra-
ocular lens implantation in eyes filled with silicone
oil. J Cataract Refract Surg 1996;22:403 – 6.
[102] Bilinska E, Nawroki J. [Evaluation of the results of
cataract surgery in eyes after silicone oil removal].
Klin Okzna 2003;205:17 – 20.
[103] Boscia F, Recchimurzo N, Cardascia N, Sborgia L,
Ferrari TM, Sborgia C. Phacoemulsification with
transpupillary silicone oil removal and lens implanta-
tion through a corneal incision using topical anes-
thesia. J Cataract Refract Surg 2003;29:1113 – 9.
[104] Krepler K, Mozaffarieh M, Biowski R, Nepp J,
Wedrich A. Cataract surgery and silicone oil re-
moval: visual outcome and complications in a com-
bined vs. two step surgical approach. Retina 2003;
23:647 – 53.
[105] Korobelnik JF, Alietta A, Hannouche D, Hoang-Xuan
T. [Value of posterior capsulorrhexis during combined
phacoemulsification and silicone oil removal]. J Fr
Ophtalmol 1998;21:649 – 53.
[106] Larkin GB, Flaxel CJ, Leaver PK. Phacoemulsifica-
tion and silicone oil removal through a single corneal
incision. Ophthalmology 1998;105:2023 – 7.
[107] Budde WM, Jonas JB, Papp A. [Cataract extraction
combined with trans-pupillary silicone oil drainage by
planned posterior capsulorrhexis]. Klin Monatsbl
Augenheilkd 1999;215:345 – 8.
[108] Assi A, Woodruff S, Gotzaridis E, Bunce C, Sullivan
P. Combined phacoemulsification and transpupillary
drainage of silicone oil: results and complications. Br
J Ophthalmol 2001;85:942 – 5.
[109] Dada VK, Talwar D, Sharma N, Dada T, Sudan R,
Azad RV. Phacoemulsification combined with sili-
cone oil removal through a posterior capsulorrhexis.
J Cataract Refract Surg 2001;27:1243 – 7.
[110] Fledelius HC, Scherfig E. The acoustic lighthouse
effect: an ultrasonic response met in eyes after
vitreoretinal surgery with silicone oil. Acta Ophthal-
mol Scand 2000;78:89 – 92.
[111] Hoffer KJ. Ultrasound velocities for axial eye length
measurement. J Cataract Refract Surg 1994;20:
554 – 62.
[112] Suger J, de Juan Jr E, et al. Ultrasonic examination
of the silicone-filled eye: theoretical and practical
considerations. Graefes Arch Clin Exp Ophthalmol
1986;114:361 – 7.
[113] Ghoraba HH, El-Dorghamy AA, Atia AF, Ismail
Yassin Ael A. The problems of biometry in combined
silicone oil removal and cataract extraction: a clinical
trial. Retina 2002;22:589 – 96.
 Ophthalmol Clin N Am 17 (2004) 569 – 576
Anterior segment complications related to
vitreous substitutes
Rubin W. Kim, MD, Caroline Baumal, MD, FRCSC*
Department of Vitreoretinal Surgery, New England Eye Center, Tufts University School of Medicine, 750 Washington Street,
Box 450, Boston, MA 02111, USA
Vitrectomy is the most common surgical tech-
nique performed in eyes with vitreoretinal disease.
Contemporary vitreous surgery techniques were
largely developed by Robert Machemer in the
1970s. Advances in surgical instrumentation and
adjuncts to the surgical procedure have refined this
surgical technique. The development and widespread
use of vitreous substitutes has revolutionized vitreo-
retinal surgery and improved anatomic and visual
results. The three most common types of vitreous
substitutes available in North America include sili-
cone oil, intraocular gas, and perfluorocarbon liquid.
Each of these agents has unique properties, allowing
them to have different roles in vitreoretinal surgery.
The physical properties, clinical indications, and po-
tential anterior segment complications of these agents
are described herein.
Silicone oil
Silicone oil was introduced by Paul Cibis for use
in retinal reattachment surgery in the early 1960s.
Initially, silicone oil was used without vitrectomy to
separate epiretinal membranes from the retina; how-
ever, currently, it is used primarily as a vitreous
substitute in vitrectomy [1 – 3]. Silicone oil is useful
to provide long-term or permanent tamponade to
maintain the retinal anatomy after repair of a
complicated retinal detachment or in other situations
in which chronic retinal apposition forces are re-
* Corresponding author.
E-mail address: cbaumal@tufts-nemc.org (C. Baumal).
0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.011
quired. Silicone oil can tamponade retinal breaks and
detachments based on its buoyant force and hydraulic
space-occupying properties. Nevertheless, intraocular
silicone oil can be associated with significant com-
plications, and its use is reserved for more compli-
cated surgical cases in which the benefits outweigh
the potential for side effects.
Silicone oil and silicone rubber share a similar
chemical formula and general chemical structure
(Fig. 1); however, the polymer chains of silicone oil
are shorter than those of silicone rubber and are not
chemically cross-linked, resulting in a liquid rather
than the more solid synthetic rubber product. The
main physical properties of silicone oil to consider
are its specific gravity, surface tension, and viscosity.
The specific gravity of silicone oil is slightly less than
1.0, which enables it to float when immersed in water
or saline [4]. Because silicone oil is immiscible with
water, surface tension is created at the interface of
these two liquids. This surface tension is responsible
for the tamponade effect of silicone oil. The surface
tension of silicone oil to saline is less than that of gas
to saline; therefore, silicone oil will pass through
retinal breaks that are under traction more easily than
will gas. One common cause of confusion is the
relationship between surface tension and viscosity.
Surface tension changes little with marked differences
in viscosity [5]. The two main varieties of silicone oil
used in vitreous surgery in North America are
classified based on their viscosity. Oil of the lower
viscosity of 1000 centistokes was used in the Sili-
cone Oil Study, whereas oil with the higher viscos-
ity of 5000 centistokes was the first variety approved
in 1996 by the Food and Drug Administration for
use in the United States [6]. Viscosity is determined
ophthalmology.theclinics.com
570 R.W. Kim, C. Baumal / Ophthalmol Clin N Am 17 (2004) 569 – 576
Fig. 1. Chemical structure of silicone oil. The number of
elements {n} determines the length of the chain.
by the average molecular weight of the chains. Two
samples of silicone oil may have the same viscos-
ity but differing concentrations of short- and long-
chain molecules.
Silicone oil emulsification occurs when small
droplets of silicone oil separate from the large central
reservoir of silicone oil. This phenomenon seems to
be related to the proportion of low molecular weight
molecules. Two samples of 1000 centistoke silicone
oil may have different propensities to emulsify
depending on the concentration of short-chain mole-
cules. The surface tension has minimal influence on
the rate of emulsification when one considers that
surface tension changes little with different viscosi-
ties. Biologic substances such as fibrinogen, fibrin,
serum, gamma globulins, and balanced salt solutions
may also contribute to emulsification [7]. Emulsifi-
cation is thought to be an important contributor to
some of the anterior segment complications of sili-
cone oil, but its exact mechanism is uncertain. Al-
though silicone oil emulsification can occur at any
time, it is more common after prolonged intraocular
retention of silicone oil [8,9].
Surface tension is the primary factor that accounts
for the tamponade effect of silicone oil on retinal
breaks. To a lesser degree, the flotation force and
volumetric displacement contribute to its apposition
of retina to the eye wall. Tamponade of the retinal
breaks allows the subretinal fluid to be absorbed by
the retinal pigment epithelium (RPE), resulting in
retinal reattachment [10]. Silicone oil will not keep
the retina attached in areas with residual tractional
forces on the retina.
Indications for silicone oil
The use of silicone oil to provide longer-term
internal retinal tamponade must be weighed against
its potential to produce significant complications. Its
use is reserved for more severe vitreoretinal surgical
cases and for specific indications. Typically, it is used
as a vitreous substitute at the completion of surgery.
A second surgical procedure to remove the silicone
oil for visual rehabilitation may be considered 3 to
6 months after the silicone oil placement. In some in-
stances, such as in the patient with a severely necrotic
retina, it may be preferable to leave the silicone oil
tamponade indefinitely as long as complications
related to silicone oil do not occur. Tamponade with
silicone oil may be preferable to intraocular gas when
postoperative travel by airplane or alteration of the
elevation above sea level is planned. It is also useful
when difficulty with postoperative positioning is
anticipated, such as in children or physically impaired
persons. Potential surgical cases that may benefit from
silicone oil use include complicated retinal detach-
ment, proliferative vitreoretinopathy, giant retinal tear
with detachment, infectious retinitis, trauma-related
ocular damage, severe diabetic retinopathy, combined
rhegmatogenous-tractional retinal detachment, and
selected macular hole cases.
Complications of silicone oil use
Lens
There are various anterior segment considerations
and potential complications related to silicone oil.
Many of these complications are related to the lens;
therefore, the lens status should be considered when
silicone oil use is anticipated during vitrectomy. Sili-
cone oil may be used in the surgical repair of eyes
with membranes secondary to proliferative vitreo-
retinopathy. The decision to remove the crystalline
lens at the time of surgery largely depends on the
location and extent of the membranes. Removal of
the natural lens may facilitate or be required for
dissection of anterior proliferative vitreoretinopathy,
whereas the natural lens may be retained if the
pathology is located posteriorly.
The anterior segment anatomy and intraocular
lens material are considerations with intraoperative
silicone oil use. An intact posterior capsule is useful
to maintain the silicone oil segregated within the
posterior segment. Placement of silicone oil in an eye
with a silicone intraocular lens and an open posterior
capsule can be particularly problematic, because sili-
cone oil droplets may adhere to the silicone lens. This
adherence can create difficulty with visualization of
the retina during surgery. Silicone oil droplets also
may remain adherent to the silicone intraocular lens
after silicone oil removal. Currently, there is no ef-
fective solution to eliminate these silicone droplets
other than removal of the intraocular lens. In eyes that
are aphakic with pupillary miosis, the small pupil can
cause the oil to protrude partially through the pu-
pillary opening, causing a convex refractive surface,
 R.W. Kim, C. Baumal / Ophthalmol Clin N Am 17 (2004) 569 – 576
which can result in intraoperative distortion of the
surgeon’s view. Filling the entire anterior chamber
with oil to reduce the plus lens effect of the convex
protrusion or stretching the pupil surgically can
reduce this distortion effect. If the anterior chamber
is filled with oil, the oil should be removed from the
chamber before completion of surgery.
The development of a postoperative cataract re-
lated to silicone oil tamponade is extremely common
(Fig. 2) [11 – 14]. Even when silicone oil is removed
promptly, the risk for a cataract is high. Proposed
mechanisms for silicone oil – related cataract devel-
opment include impaired metabolic exchange across
the posterior lens capsule and possible toxicity of
the oil itself. Surgical removal of a visually signifi-
cant cataract may be indicated in a silicone-filled eye.
Ideally, the silicone oil should be removed before
or concurrently with cataract surgery. If intraocular
silicone oil tamponade needs to be maintained, mi-
gration of oil into the anterior segment can be prob-
lematic during cataract surgery. Phacoemulsification
is preferred for cataract removal in this situation be-
cause of the small incision size and improved ante-
rior chamber stability. An intact posterior capsule
and the use of posterior chamber intraocular lenses
are advantageous to maintain the silicone oil in the
posterior segment when silicone oil is to be retained.
Preoperative measurements for lens calculations must
take into account maintenance of the silicone oil fill
[15 – 18]. The speed of sound in silicone oil is 986 m/s
versus 1552 m/s in vitreous-filled eyes. A correction
factor must be used to calculate the axial length in
eyes filled with silicone oil, or an error of up to
40% can occur. Ideally, the measurement should be
taken with the patient in the upright position. If
the patient is in the supine position, a layer of aque-
ous interface may be present in the preretinal space,
causing errors in measurements. Use of a silicone
Fig. 2. Development of a cataract secondary to silicone
oil use.
571
Fig. 3. Silicone oil migration into the anterior chamber. The
silicone oil was used for retinal tamponade after surgical
repair of a grade D proliferative vitreoretinopathy retinal
detachment in an eye that had initially sustained severe
ocular trauma with rupture. The eye is surgically aphakic,
and the pupil is distorted from initial trauma. The ocular bar-
riers that act to prevent silicone oil migration are damaged.
intraocular lens is contraindicated if there is any
possibility of future vitreoretinal surgery with silicone
oil tamponade, owing to potential adherence of
silicone oil droplets to silicone intraocular lens.
Cornea and anterior chamber
Silicone oil migration into the anterior chamber
can cause a variety of problems (Fig. 3). Silicone
oil contact with the corneal endothelium can lead
to corneal decompensation and band keratopathy
[19 – 22]. The 2-year incidence of corneal abnormali-
ties was 26% in eyes randomly assigned to silicone
oil tamponade in the Silicone Oil Study [23]; how-
ever, this rate was not significantly different from
the 28% incidence of corneal abnormalities in eyes
assigned to perfluoropropane intraocular gas tam-
ponade. Preoperative aphakia, pseudophakia, or iris
neovascularization, as well as postoperative aque-
ous flare and multiple vitreoretinal surgeries, increase
the likelihood of postoperative corneal abnormalities
with silicone oil tamponade for severe proliferative
vitreoretinopathy. Limiting intraocular inflammation,
cornea-oil touch, and reoperations should decrease
the risk for corneal problems. Corneal chelation can
be performed to treat eyes that develop band kera-
topathy related to silicone oil use. Penetrating
keratopathy is an option for severe keratopathy. It is
preferable to remove silicone oil before or at the time
of penetrating keratopathy to reduce the rate of cor-
neal graft failure to approximately 25% versus the
67% rate of failure in eyes with chronic silicone oil
tamponade [24]. Fig. 4 demonstrates a breverseQ or
binverted hypopyonQ appearance of emulsified bub-
bles of silicone oil that have migrated into the anterior
572 R.W. Kim, C. Baumal / Ophthalmol Clin N Am 17 (2004) 569 – 576
Fig. 4. Silicone oil emulsification and migration into the
anterior chamber produces an binvertedQ or breverseQ hy-
popyon. The layered silicone oil migrates superiorly in the
anterior chamber as it floats on top of the aqueous layer.
chamber. This appearance develops as the silicone
oil rises above the aqueous fluid.
Silicone migration into the subconjuctival space
and upper eyelid is an uncommonly reported com-
plication [25]. One rare case of intraocular silicone oil
migration out of the eye along the intracranial portion
of the optic nerve into the lateral ventricles of the
brain has been reported [26].
Intraocular pressure
Glaucoma can occur through a variety of mecha-
nisms related to silicone oil use. The reported in-
cidence of glaucoma in studies using silicone oil to
treat proliferative vitreoretinopathy has ranged from
2% to 40% [12,14,19,22,27 – 29]. Pupillary block
glaucoma occurs when silicone oil occludes the
pupillary space and prevents the flow of aqueous
from the ciliary processes through the pupil into the
anterior chamber. This block produces misdirection
of aqueous fluid, leading to shallowing of the anterior
chamber and elevated intraocular pressure. Pupillary
block glaucoma may occur at anytime but is more
likely early in the postoperative course [30 – 32]. This
type of glaucoma can be prevented by the creation of
an inferior peripheral iridectomy in aphakic eyes to
create a pathway for aqueous to flow into the anterior
chamber [30,33,34]. Although a patent peripheral
iridectomy may have been created at the time of
surgery, fibrin, blood, or a residual posterior capsule
can postoperatively block this opening. Injection of
tissue plasminogen activator and yttrium-aluminum-
garnet laser iridotomy have been used to open the
iridectomy in these postoperative situations.
Overfilling the vitreous cavity with silicone oil
can produce elevated intraocular pressure acutely.
Shallowing of the anterior chamber with or without
silicone oil in the anterior chamber is a sign that
overfilling may be present. Choroidal swelling may
result in a relative oil overfill, but this typically will
resolve with medical management and resolution of
the swelling. Surgical intervention to remove some of
the oil may be necessary when an actual overfill is
present. Silicone oil overfill can usually be prevented
by ensuring that the intraocular pressure is low to
normal and that the anterior chamber is deep and
devoid of silicone oil at the completion of surgery.
Emulsification of the silicone oil is thought be a
contributor to ocular hypertension or glaucoma in
some silicone-filled eyes [14,35,36]. The emulsified
oil may directly block the trabecular meshwork or
produce inflammation within this meshwork [37].
Removal of silicone oil alone does not usually relieve
the chronic elevation of intraocular pressure second-
ary to trabecular meshwork scarring. Although medi-
cal management can be effective initially, glaucoma
surgery may eventually be necessary.
Glaucoma surgery in eyes with silicone oil can
be problematic. Ideally, the silicone oil should be
removed before glaucoma surgery. When retention of
silicone oil is necessary for retinal reattachment,
migration of oil into glaucoma valve implants and
obstruction has been reported [38 – 41]. Although
inferior placement of the glaucoma valve theoreti-
cally reduces the potential for problems because
silicone oil is lighter than aqueous, migration into the
valves has still been reported.
In addition to issues concerning elevated intra-
ocular pressure, chronic hypotony can be a compli-
cation associated with silicone oil. In the Silicone
Oil Study, the prevalence of hypotony at 36 months
was lower in silicone-filled eyes (18%) when com-
pared with C3F8 gas-filled eyes (31%) [29]. Hypo-
tony was more likely to occur in eyes with diffuse
contraction of the retina anterior to the equator and
with postoperative retinal redetachment. The exact
mechanism for the development of postoperative hy-
potony is not understood. Management of hypotony
can be problematic. If it develops early postopera-
tively, vitrectomy with peeling of membranes from
the ciliary processes may be useful [42]. Subcon-
junctival injection of depot steroids may elevate intra-
ocular pressure in some cases. In eyes with chronic
postoperative hypotony, silicone oil may be retained
to prevent development of phthisis.
Intraocular gases
In 1911, Ohm treated two patients with retinal
detachment by injecting air after drainage of sub-
retinal fluid [43]. Since then, many different gases
 R.W. Kim, C. Baumal / Ophthalmol Clin N Am 17 (2004) 569 – 576
have been investigated for use in vitreoretinal sur-
gery. The main effect of the gas bubble within the
eye is based on its tamponade, producing func-
tional closure of retinal breaks [44,45]. Intraocular
gases have the highest surface tension currently
available of any of the vitreous replacements at
approximately 70 dynes/cm. The surface tension of
the gas bubble prevents it from passing through the
retinal break, and blocks fluid from the vitreous
cavity from entering the subretinal space. With this
tamponade in place, the RPE can absorb the residual
subretinal fluid to re-appose the retina against the eye
wall. A secondary effect of the gas bubble is flotation
from the buoyancy of the gas bubble that pushes the
retina toward the eye wall. The force is greatest at the
apex of the arc of contact of the gas bubble.
The two gases most commonly used clinically
in North America are sulfur hexafluoride (SF6) and
perfluoropropane (C3F8). SF6 is colorless, odorless,
and five times lighter than air (Fig. 5) [46]. It has
been used extensively in ophthalmology and seems to
be nontoxic within the eye [47]. C3F8 is also
colorless, odorless, nontoxic, and is six times lighter
than air (Fig. 6). Both of these gases have expansile
properties that allow long-term tamponade, which
makes them useful for vitreoretinal surgery (Table 1).
A bubble of pure SF6 expands by 2 to 2.5 times its
original volume [48 – 52]. SF6 reaches its maximally
expanded volume by 24 to 78 hours [51]. Pure C3F8
expands four times its original volume and reaches its
maximal volume at 72 to 96 hours [53 – 55]. The
nonexpansile concentration of SF6 is 20% and that
of C3F8, 12% to 14%, respectively [44,45]. These
nonexpansile concentrations are often used when
the entire vitreous cavity is to be filled with gas at the
end of vitreoretinal surgery. Expansile concentrations
of gas are more often used when only part of the
posterior segment is filled with gas. For example, in
the office-based procedure of pneumatic retinopexy
to repair a rhegmatogenous retinal detachment, a
concentration of 100% SF6 may be used, but in a
much smaller volume in the range of 0.3 to 0.5 mL.
The half-life of each of these gases is also unique and
Fig. 5. Chemical structure of SF6.
573
Fig. 6. Chemical structure of C3F8.
depends on multiple factors, including the initial
concentration and bubble volume, the volume of the
vitreous cavity, the presence or absence of vitreous,
and whether the eye is phakic or aphakic.
The most serious complication when using these
potentially expansile intraocular gases is marked
elevation of the intraocular pressure, which could
lead to central retinal artery occlusion and permanent
visual loss. The longer-acting gases and the ones with
greater expansile properties are more likely to cause
elevation of intraocular pressure. Patients with
compromised outflow, especially from peripheral
anterior synechiae or neovascularization, are more
likely to have the complication of elevated intraocular
pressure. Topical or systemic antiglaucoma medica-
tions may be administered at the end of surgery
prophylactically if issues with intraocular pressure are
anticipated. In most cases, intraocular pressure can be
controlled medically. On occasion, removal of some
intraocular gas from the posterior segment may be
necessary early in the postoperative period, especially
when there is an overfill of gas in the posterior
segment, or if there is overexpansion of intraocular
gas. Pupillary block glaucoma can occur if the
intraocular gas bubble pushes the lens diaphragm
forward. Proper positioning and avoidance of the
supine position after vitreous surgery helps to reduce
the chance of pupillary block glaucoma.
Gas-induced cataracts are common. Lens opacities
appear as feathery linear changes or vacuoles in the
posterior subcapsular region and are most frequently
located in the superior portion of the lens. These lens
opacities may disappear within 24 hours if contact
between the gas bubble and the lens is avoided. This
goal can be accomplished by altering the positioning
of the patient’s head to decrease lens contact with
intraocular gas. Permanent lens opacities are more
likely with longer-acting gases, and more than 50%
of these patients progress to permanent lens opacifi-
cation [56].
Gas bubbles in the anterior chamber may cause
corneal damage leading to corneal edema and bullous
keratopathy. The gas is not thought to be toxic to the
endothelium; rather, the contact between the gas
bubble and the endothelium seems to interfere with
574 R.W. Kim, C. Baumal / Ophthalmol Clin N Am 17 (2004) 569 – 576
Table 1
Comparison of the properties of SF6 and C3F8 intraocular gas
Gas Molecular weight Expansivity Nonexpansile concentration
SF6 146 2 to 2.5  20%
C3F8 188 4 12% – 14%
corneal access to nutrients [57]. Proper postoperative
positioning can usually alleviate this problem.
An important postoperative consideration when
using intraocular gas includes the avoidance of high
altitudes, such as with travel in commercial airplanes.
Rapid decompression of cabin pressure can produce
an increase in the size of the gas bubble, which can
elevate intraocular pressure [58 – 60]. This elevated
intraocular pressure can produce central retinal artery
occlusion and result in loss of vision. Patients should
avoid air travel until the gas bubble has been replaced
by intraocular fluid. Lincoff and colleagues have
reported that volumes of certain intraocular gases up
to 1.0 mL may be tolerated with air travel, although
this is not recommended [59]. Another consideration
is the interaction of intraocular gas with nitrous oxide.
Nitrous oxide is a commonly used inhalation anes-
thetic agent for general anesthesia. Nitrous oxide is
highly soluble and rapidly diffuses into intraocular
gas bubbles, leading to expansion of the gas bubble
and elevation of intraocular pressure. The maximum
rise in intraocular pressure occurs 15 to 20 minutes
after the administration of nitrous oxide [61]. Once
the nitrous oxide is discontinued, the intraocular
pressure will lower as the nitrous oxide diffuses out
of the body by pulmonary exchange. When nitrous
oxide is used as a general anesthetic agent during eye
surgery, it should be discontinued at least 15 minutes
before anticipated intraocular gas injection. This in-
terval allows the inhaled nitrous oxide to be com-
pletely removed systemically, and the intraocular gas
can be safely administered without excessive expan-
sion of the bubble [62]. Careful patient instruction on
these possible complications of intraocular gas is
important. The use of a wristband to notify medical
personnel regarding the presence of a gas bubble in
Fig. 7. Bracelet worn by patient postoperatively after intraocular gas use.
Time to maximal expansion Duration
24 to 78 hours 1 – 2 weeks
72 to 96 hours 4 – 6 weeks
the case of an emergency should be standard practice
(Fig. 7).
Perfluorocarbon liquid
Perfluorocarbon liquids were initially designed
to be a blood substitute because of their capacity
to carry oxygen and their relatively inert nature.
Perfluorocarbon liquids are used as an adjunct in
vitreoretinal surgery for retinal detachment repair,
removal of retained lens fragments, and to assist with
peeling of preretinal membranes. Their use has
markedly improved the repair of giant retinal tears
and prevents posterior slip of the retina during sur-
gery. Multiple agents have been used and approved
in different countries, including perfluoro-n-octane,
perfluoroperhydrophenanthrene (Vitreon), and per-
fluorodecalin. Perfluoro-n-octane is the agent pri-
marily used in North America.
Perfluorocarbon liquids have a specific gravity of
1.9; therefore, they have a higher density than water
[63]. This property causes them to sink onto the retina
and displace any subretinal fluid anteriorly. If there is
no significant traction on the retina, the perfluo-
rocarbon liquids will push the retina flat on the eye
wall, making these liquids useful during vitreoretinal
surgery. Because the liquids are usually removed at
the end of the procedure, anterior segment complica-
tions are uncommon. Potential anterior segment
complications of perfluorocarbon liquids have been
investigated in animal studies and may occur in
aphakic patients when retained perfluorocarbon
liquid is in direct contact with the corneal endothe-
lium [64]. Intravitreal perfluorocarbon liquid does not
usually place perfluorocarbon liquids in touch with
 R.W. Kim, C. Baumal / Ophthalmol Clin N Am 17 (2004) 569 – 576
the endothelium, and intravitreal placement of these
liquids has not been shown to be toxic to the cornea
[65,66].
Summary
Vitreoretinal surgery has been revolutionized with
the introduction of a variety of vitreous substitutes.
Each of these substitutes has unique properties, in-
dications, and potential complications. Anterior seg-
ment complications may develop intraoperatively or
postoperatively, and may occur acutely or remotely
after vitreoretinal surgery. New developments and
applications for vitreous substitutes will continue to
improve anatomic and visual outcomes in vitreoreti-
nal surgery.
References
[1] Scott JD. Treatment of the detached immobile retina.
Trans Ophthalmol Soc U K 1972;92:351 – 7.
[2] Scott JD. A rationale for the use of liquid silicone.
Trans Ophthalmol Soc U K 1977;97:235 – 7.
[3] McCuen III BW, de Juan Jr E, Machemer R. Silicone
oil in vitreoretinal surgery. Part 1. Surgical techniques.
Retina 1985;5:189 – 97.
[4] Haut J, Larricart JP, Van Effenterre G, et al. Some
of the most important properties of silicone oil to ex-
plain its action. Ophthalmologica 1985;191:150 – 3.
[5] Crisp A, de Jr JE, Tiedeman J. Effect of silicone oil
viscosity on emulsification. Arch Ophthalmol 1987;
105:546 – 50.
[6] The Silicone Study Group. Vitrectomy with silicone oil
or perfluoropropane gas in eyes with severe prolifera-
tive vitreoretinopathy: results of a randomized clinical
trial. Silicone Study Report 2. Arch Ophthalmol 1992;
110:780 – 92.
[7] Heidenkummer HP, Kampik A, Thierfelder S. Emulsi-
fication of silicone oils with specific physicochemical
characteristics. Graefes Arch Clin Exp Ophthalmol
1991;229:88 – 94.
[8] Scott JD. The treatment of massive vitreous retraction
by the separation of pre-retinal membranes using liq-
uid silicone. Mod Probl Ophthalmol 1975;15:185 – 90.
[9] Gabel VP, Kampik A, Gabel C, et al. Silicone oil with
high specific gravity for intraocular use. Br J Oph-
thalmol 1987;71:262 – 7.
[10] Gonvers M. Temporary use of intraocular silicone oil
in the treatment of detachment with massive periretinal
proliferation: preliminary report. Ophthalmologica
1982;184:210 – 8.
[11] Gonvers M. Temporary silicone oil tamponade in the
management of retinal detachment with proliferative
vitreoretinopathy. Am J Ophthalmol 1985;100:239 – 45.
[12] McCuen BW, Landers III MB, Machemer R. The use
575
of silicone oil following failed vitrectomy for retinal
detachment with advanced proliferative vitreoreti-
nopathy. Ophthalmology 1985;92:1029 – 34.
[13] Zilis JD, McCuen BW, de Jr JE, et al. Results of sili-
cone oil removal in advanced proliferative vitreoreti-
nopathy. Am J Ophthalmol 1989;108:15 – 21.
[14] Federman JL, Schubert HD. Complications associated
with the use of silicone oil in 150 eyes after retina-
vitreous surgery. Ophthalmology 1988;95:870 – 6.
[15] Ghoraba HH, El Dorghamy AA, Atia AF, et al. The
problems of biometry in combined silicone oil re-
moval and cataract extraction: a clinical trial. Retina
2002;22:589 – 96.
[16] Murray DC, Durrani OM, Good P, et al. Biometry of
the silicone oil-filled eye. II. Eye 2002;16:727 – 30.
[17] Murray DC, Potamitis T, Good P, et al. Biometry of
the silicone oil-filled eye. Eye 1999;13(Pt 3a):319 – 24.
[18] Bolger J. Biometry of the silicone oil-filled eye. Eye
2000;14(Pt 1):118 – 20.
[19] Okun E. Intravitreal surgery utilizing liquid silicone:
a long term followup. Trans Pac Coast Otoophthalmol
Soc Annu Meet 1968;52:141 – 59.
[20] Cockerham WD, Schepens CL, Freeman HM. Silicone
injection in retinal detachment. Arch Ophthalmol 1970;
83:704 – 12.
[21] Grey RH, Leaver PK. Results of silicone oil injection
in massive preretinal retraction. Trans Ophthalmol
Soc U K 1977;97:238 – 41.
[22] Leaver PK, Grey RH, Garner A. Silicone oil injection
in the treatment of massive preretinal retraction. II. Late
complications in 93 eyes. Br J Ophthalmol 1979;63:
361 – 7.
[23] Abrams GW, Azen SP, Barr CC, et al. The incidence
of corneal abnormalities in the Silicone Study: Silicone
Study Report 7. Arch Ophthalmol 1995;113:764 – 9.
[24] Noorily SW, Foulks GN, McCuen BW. Results of
penetrating keratoplasty associated with silicone oil
retinal tamponade. Ophthalmology 1991;98:1186 – 9.
[25] Quintyn JC, Genevois O, Ranty ML, et al. Silicone
oil migration in the eyelid after vitrectomy for retinal
detachment. Am J Ophthalmol 2003;136:540 – 2.
[26] Eller AW, Friberg TR, Mah F. Migration of silicone
oil into the brain: a complication of intraocular sili-
cone oil for retinal tamponade. Am J Ophthalmol 2000;
129:685 – 8.
[27] Haut J, Ullern M, Chermet M, et al. Complications
of intraocular injections of silicone combined with
vitrectomy. Ophthalmologica 1980;180:29 – 35.
[28] Sell CH, McCuen BW, Landers III MB, et al. Long-
term results of successful vitrectomy with silicone oil
for advanced proliferative vitreoretinopathy. Am J
Ophthalmol 1987;103:24 – 8.
[29] Barr CC, Lai MY, Lean JS, et al. Postoperative intra-
ocular pressure abnormalities in the Silicone Study:
Silicone Study Report 4. Ophthalmology 1993;100:
1629 – 35.
[30] Ando F. Intraocular hypertension resulting from pupil-
lary block by silicone oil. Am J Ophthalmol 1985;99:
87 – 8.
576 R.W. Kim, C. Baumal / Ophthalmol Clin N Am 17 (2004) 569 – 576
[31] Burk LL, Shields MB, Proia AD, et al. Intraocular
pressure following intravitreal silicone oil injection.
Ophthalmic Surg 1988;19:565 – 9.
[32] Zborowski-Gutman L, Treister G, Naveh N, et al.
Acute glaucoma following vitrectomy and silicone
oil injection. Br J Ophthalmol 1987;71:903 – 6.
[33] Beekhuis WH, Ando F, Zivojnovic R, et al. Basal
iridectomy at 6 o’clock in the aphakic eye treated with
silicone oil: prevention of keratopathy and secondary
glaucoma. Br J Ophthalmol 1987;71:197 – 200.
[34] Bartov E, Huna R, Ashkenazi I, et al. Identification,
prevention, and treatment of silicone oil pupillary
block after an inferior iridectomy. Am J Ophthalmol
1991;111:501 – 4.
[35] Gao RL, Neubauer L, Tang S, et al. Silicone oil in the
anterior chamber. Graefes Arch Clin Exp Ophthalmol
1989;227:106 – 9.
[36] Avitabile T, Bonfiglio V, Cicero A, et al. Correlation
between quantity of silicone oil emulsified in the
anterior chamber and high pressure in vitrectomized
eyes. Retina 2002;22:443 – 8.
[37] Champion R, Faulborn J, Bowald S, et al. Peritoneal
reaction to liquid silicone: an experimental study.
Graefes Arch Clin Exp Ophthalmol 1987;225:141 – 5.
[38] Parwar BL, Coleman AL, Small KW. Silicone oil
migration through an Ahmed valve. Retina 2002;22:
657 – 8.
[39] Hyung SM, Min JP. Subconjunctival silicone oil drain-
age through the Molteno implant. Korean J Ophthal-
mol 1998;12:73 – 5.
[40] Senn P, Buchi ER, Daicker B, et al. Bubbles in the
bleb – troubles in the bleb? Molteno implant and
intraocular tamponade with silicone oil in an aphakic
patient. Ophthalmic Surg 1994;25:379 – 82.
[41] Nazemi PP, Chong LP, Varma R, et al. Migration of
intraocular silicone oil into the subconjunctival space
and orbit through an Ahmed glaucoma valve. Am J
Ophthalmol 2001;132:929 – 31.
[42] Zarbin MA, Michels RG, Green WR. Dissection of
epiciliary tissue to treat chronic hypotony after sur-
gery for retinal detachment with proliferative vitreo-
retinopathy. Retina 1991;11:208 – 13.
[43] Chang S. Vitreous surgery. 2nd edition. Baltimore7
Williams & Wilkins; 2004.
[44] de Juan Jr E, McCuen B, Tiedeman J. Intraocular
tamponade and surface tension. Surv Ophthalmol
1985;30:47 – 51.
[45] McCuen BW, Azen SP, Stern W, et al. Vitrectomy with
silicone oil or perfluoropropane gas in eyes with severe
proliferative vitreoretinopathy: Silicone Study Report 3.
Retina 1993;13:279 – 84.
[46] Matheson Gas Products. Matheson gas book. 5th edi-
tion. Milwaukee (WI)7 Matheson Gas Products; 2004.
[47] Norton EW. Intraocular gas in the management of se-
lected retinal detachments. Trans Am Acad Ophthal-
mol Otolaryngol 1973;77:OP85 – 98.
[48] Fineberg E, Machemer R, Sullivan P, et al. Sulfur
hexafluoride in owl monkey vitreous cavity. Am J
Ophthalmol 1975;79:67 – 76.
[49] Fineberg E, Machemer R, Sullivan P. SF6 for retinal
detachment surgery: a preliminary report. Mod Probl
Ophthalmol 1974;12:173 – 6.
[50] Killey FP, Edelhauser HF, Aaberg TM. Intraocular
sulfur hexafluoride and octofluorocyclobutane: effects
on intraocular pressure and vitreous volume. Arch
Ophthalmol 1978;96:511 – 5.
[51] Abrams GW, Edelhauser HF, Aaberg TM, et al.
Dynamics of intravitreal sulfur hexafluoride gas. Invest
Ophthalmol 1974;13:863 – 8.
[52] Tenney SM, Carpenter FG, Rahn H. Gas transfers in
a sulfur hexafluoride pneumoperitoneum. J Appl Phy-
siol 1953;6:201 – 8.
[53] Lincoff H, Coleman J, Kreissig I, et al. The perfluo-
rocarbon gases in the treatment of retinal detachment.
Ophthalmology 1983;90:546 – 51.
[54] Lincoff A, Haft D, Liggett P, et al. Intravitreal
expansion of perfluorocarbon bubbles. Arch Ophthal-
mol 1980;98:1646.
[55] Peters MA, Abrams GW, Hamilton LH, et al. The
nonexpansile, equilibrated concentration of perfluoro-
propane gas in the eye. Am J Ophthalmol 1985;100:
831 – 9.
[56] Chang S, Lincoff HA, Coleman DJ, et al. Perfluo-
rocarbon gases in vitreous surgery. Ophthalmology
1985;92:651 – 6.
[57] Van Horn DL, Edelhauser HF, Aaberg TM, et al. In
vivo effects of air and sulfur hexafluoride gas on rabbit
corneal endothelium. Invest Ophthalmol 1972;11:
1028 – 36.
[58] Dieckert JP, O’Connor PS, Schacklett DE, et al. Air
travel and intraocular gas. Ophthalmology 1986;93:
642 – 5.
[59] Lincoff H, Weinberger D, Stergiu P. Air travel with
intraocular gas. II. Clinical considerations. Arch
Ophthalmol 1989;107:907 – 10.
[60] Lincoff H, Weinberger D, Reppucci V, et al. Air travel
with intraocular gas. I. The mechanisms for compen-
sation. Arch Ophthalmol 1989;107:902 – 6.
[61] Smith RB, Carl B, Linn Jr JG, et al. Effect of nitrous
oxide on air in vitreous. Am J Ophthalmol 1974;78:
314 – 7.
[62] Stinson III TW, Donlon Jr JV. Interaction of intraocular
air and sulfur hexafluoride with nitrous oxide: a
computer simulation. Anesthesiology 1982;56:385 – 8.
[63] Chang S, Ozmert E, Zimmerman NJ. Intraoperative
perfluorocarbon liquids in the management of prolif-
erative vitreoretinopathy. Am J Ophthalmol 1988;106:
668 – 74.
[64] Moreira H, de Jr QJ, Liggett PE, et al. Corneal toxicity
study of two perfluorocarbon liquids in rabbit eyes.
Cornea 1992;11:376 – 9.
[65] Blinder KJ, Peyman GA, Desai UR, et al. Vitreon, a
short-term vitreoretinal tamponade. Br J Ophthalmol
1992;76:525 – 8.
[66] Blinder KJ, Peyman GA, Paris CL, et al. Vitreon, a new
perfluorocarbon. Br J Ophthalmol 1991;75:240 – 4.
 Ophthalmol Clin N Am 17 (2004) 577 – 582
Retinopathy of prematurity and anterior
segment complications
Michael S. Ibarra, MD, Antonio Capone, Jr, MD*
Associated Retinal Consultants, William Beaumont Hospital, 3535 West 13 Mile Road, Suite 632, Royal Oak, MI 48073, USA
Retinopathy of prematurity (ROP) is a prolifer-
ative vascular disorder occurring in premature infants
and is a leading cause of childhood blindness. The
advancement of neonatology in recent decades has
allowed infants with extremely low birth weights and
young gestational ages to survive and potentially
develop ROP. A widely recognized screening protocol
is to screen all infants weighing less than 1500 g
and aged less than 32 weeks’ postmenstrual age or
4 weeks after birth if less than 31 weeks’ postmen-
strual age. Although most infants in whom ROP de-
velops have spontaneous regression, approximately
7% progress to threshold ROP [1]. Threshold disease
is defined as stage 3+ disease involving five con-
tiguous or eight cumulative clock-hours in zone 1 or
2 and requires ablative therapy of the avascular retina
with either cryotherapy or laser photoablation [2,3].
Despite timely and appropriate ablative therapy, one
fifth of infants who reach threshold ROP have
progression of disease [1]. Advanced stages of ROP
(stages 4 and 5) causing partial to total retinal de-
tachment are treated with more invasive techniques,
including vitrectomy and scleral buckling procedures.
The management of ROP has been successful in
reducing the incidence of an unfavorable structural
outcome and improving functional outcome [4];
however, as is true for any therapy in medicine, there
are associated complications. This article reviews the
anterior segment complications of cryotherapy, laser
No author has any proprietary interests.
* Corresponding author.
E-mail address: acaponejr@yahoo.com (A. Capone, Jr).
0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.006
photoablation, and vitreoretinal surgery for the treat-
ment of ROP.
Cryotherapy
In 1988, the Cryotherapy for Retinopathy of
Prematurity Cooperative Group published the first
of several reports on a multicenter trial of cryotherapy
for ROP. This report demonstrated the effectiveness
of cryotherapy in preventing the progression of
threshold disease [1]. Subsequent reports, along with
other publications, have clearly shown that cryother-
apy is effective but not without complications.
The most common treatment complications related
to cryotherapy are eyelid edema, conjunctival chemo-
sis, conjunctival hyperemia, conjunctival hemorrhage,
and subconjunctival hemorrhage [1,2,5 – 7]. The
multicenter trial of cryotherapy for ROP reported
conjunctival or subconjunctival hemorrhage in 11.7%
of treated eyes [2]. McNamara et al reported that
all 24 eyes treated with cryotherapy in their study
developed lid edema, conjunctival hyperemia, and
chemosis [5]. Likewise, inadvertent conjunctival
lacerations and corneal clouding have been reported
[1,6]. These treatment complications are believed
to be caused by mechanical manipulation of the
cryoprobe or freezing of the conjunctiva and cornea
during treatment and generally resolve a few days
after treatment.
A more severe complication of cryotherapy is
anterior segment ischemia, in which treated eyes
develop hypotony, iris atrophy, and cataracts. No large
series of this complication have been reported in the
literature; therefore, its frequency cannot be accurately
ophthalmology.theclinics.com
578 M.S. Ibarra, A. Capone, Jr / Ophthalmol Clin N Am 17 (2004) 577 – 582
estimated. Kaiser and Trese [8] reported one case in
which, 4 days after cryotherapy was performed, the
eye showed signs of anterior segment ischemia and
subsequently underwent lensectomy and vitrectomy.
Kaiser and Trese noted confluent retinal ablation on
examination and proposed a mechanism by which
anterior segment ischemia may develop after ablation
of the entire avascular retina from the edge of the pars
plicata to the anterior edge of the ridge of prolifer-
ation. This area covers the entire circumference of the
peripheral retina, including the horizontal meridians
where the long posterior ciliary arteries traverse
through the suprachoroidal space and anastomose
with the anterior ciliary arteries. These arteries
constitute the vascular supply for the anterior segment
of the eye. It is theorized that ablation of retinal tissue
through which the long posterior ciliary arteries travel
could impair blood flow and cause ischemia to the
anterior segment. The additional possible compres-
sion of the long posterior ciliary arteries during scleral
depression could also contribute to impaired blood
flow and anterior segment ischemia.
Laser photoablation
Following the encouraging results of the multi-
center trial of cryotherapy for ROP in 1988 [1], the
portability of the indirect ophthalmoscopic laser led
physicians in the early 1990s to begin using this
instrument to treat ROP. Numerous reports have
shown laser photoablation to be an effective method
of treating avascular retina when patients reach
threshold ROP [5,6,9 – 15]; however, as is true for
cryotherapy, potential complications are associated
with laser treatment.
Similar to cryotherapy, transient conjunctival hy-
peremia can occur [5,16]. This complication may be
caused less by the actual laser therapy than by the
globe manipulation required when performing laser
treatment to the entire avascular retina. Conjunctival
hyperemia may be more pronounced with transscleral
laser photoablation than with indirect ophthalmo-
scopic laser delivery [16].
Corneal and iris defects are widely known po-
tential complications of laser photoablation [10,11,
17 – 19]. Irvine et al described small white opacities
estimated to be 10% excavations of the anterior cornea
following indirect laser photoablation in adult patients
[17]. They postulated that these superficial corneal
defects or burns may be secondary to laser energy
absorbed by the edematous corneal epithelium or from
light scattering at the corneal surface [17]. Corneal
epithelial edema can occur and may be secondary to
epithelial absorption of laser energy as well as
epithelial exposure during treatment. Removal of the
corneal epithelium helps maintain clarity during
treatment [11]. Iris burns have been described and
are probably caused by inadvertent laser treatment to
the iris surface and the absorption of energy by iris
pigment. A subsequent mild iridocyclitis can occur, as
can posterior synechiae [17]. Some authorities advo-
cate the use of transscleral laser delivery to prevent
injury to the cornea and iris [16]. In addition, anterior
chamber hyphema has been reported as a complication
of laser photoablation [13,14,20,21]. These hyphemas
most likely occur secondary to inadvertent lasering of
the iris and iris vessels.
Angle-closure glaucoma is a known complication
after laser photoablation and has been described after
panretinal photocoagulation for proliferative diabetic
retinopathy [22,23]. After widespread photocoagula-
tion, an exudative retinal or choroidal detachment can
develop, causing forward rotation of the ciliary body
with subsequent closure of the angle. Despite nu-
merous reports in the literature describing use of the
laser for treatment of ROP, only Lee et al [24] have
reported a patient in whom angle-closure glaucoma
developed after laser treatment for ROP that required
surgical iridectomies.
Cataract formation has been reported as a com-
plication of laser photoablation in elderly patients.
Early reports theorized that the aged-lens changes,
with possible pigmentation, made the lens susceptible
to laser energy absorption [25,26]. The clear lenses of
preterm infants have also been shown to develop
cataracts secondary to laser therapy; these occur-
rences are well described [10,12,14,20,27 – 34]. One
report indicated that cataracts occurred in as many as
6% of eyes treated for ROP [34].
Two types of cataracts have been described. Both
are believed to result from the thermal effects of the
laser. The first type appears as punctated or vacuolated
focal opacities occurring at the capsular or subcap-
sular level. These focal opacities are believed to be
visually insignificant and may spontaneously resolve
over a period of 2 to 6 weeks [31,32]. Capone et al
have suggested that the preterm infant lens is
secondarily affected by the laser owing to factors
such as its proximity to the iris, iris pigment within
broken iris synechiae to the anterior lens capsule, or
the tunica vasculosa lentis, all of which may absorb
energy and heat the lens [9]. In theory, there is a
reduced potential of thermal-induced cataract forma-
tion in infants with a prominent tunica vasculosa lentis
with a diode laser (810 nm) when compared with an
argon laser (514 nm), because the diode laser wave-
length is less readily absorbed by blood vessels [5,14].
 M.S. Ibarra, A. Capone, Jr / Ophthalmol Clin N Am 17 (2004) 577 – 582
The second type of cataract is visually significant and
may appear immediately following or weeks after
laser treatment [27]. The cataract appears as a dense
opacity, sometimes total in nature, which obscures
visualization of the fundus and requires extraction by
a vitreoretinal surgeon because of possible vitreous
adhesions. Although the etiology is of this type of
cataract is unclear, Lambert et al have proposed
several explanations [27]. First, microperforations in
the lens capsule may allow a phacoantigenic response
leading to cataract formation. This explanation is
supported by the findings of Lambert et al of posterior
synechiae, iris atrophy, and pigment on the anterior
lens capsule, suggesting an inflammatory component.
A second explanation is that these forms of cataracts
represent anterior segment ischemia.
As discussed previously, ablation of the avascular
retina, especially along the horizontal meridians at the
3 and 9 o’clock positions, can theoretically damage or
destroy the long posterior ciliary arteries and cause
anterior segment ischemia. Despite limited discussion
in the cryotherapy literature, several articles describe
laser ablation leading to anterior segment ischemia
[8,10,27]. Lambert et al reported on a series of eight
patients (10 eyes) in whom signs of anterior segment
ischemia developed, including cataracts, corneal
edema, iris atrophy, pupillary membranes, and
shallow anterior chambers, within 1 to 4 weeks after
confluent treatment with the argon or diode laser.
Fifty percent of the eyes developed retinal detach-
ments before cataract surgery. On follow-up after
cataract surgery, 9 of 10 eyes became phthisical [27].
Kaiser and Trese reported anterior segment ischemia
occurring in eight eyes in five patients following
confluent treatment of cryotherapy or laser therapy.
Fallaha et al reported on 2 of 87 eyes (2.3%) that
underwent confluent laser photoablation for ROP and
subsequently developed anterior segment ischemia.
One possible way to prevent this severe complication,
suggested by Lambert et al and Kaiser and Trese, is to
treat lightly (with laser or cryotherapy) the horizontal
meridians where the long posterior ciliary arteries are
found [8,27].
Vitreoretinal surgery
The goal of ROP screening is to monitor carefully
infants at risk for ROP and to treat those patients with
threshold disease using laser photoablation or cryo-
therapy. If advanced ROP occurs and the retina
becomes detached (stage 4 or 5), more invasive therapy
such as vitrectomy or scleral buckling is required.
579
Although many reports in the literature demon-
strate the effectiveness of vitrectomy for advanced
ROP [35 – 39], few address the potential ocular
complications of the intervention. Quinn et al did
report ocular complications at 5.5 years following
vitrectomy for stage 5 ROP. These complications in-
cluded glaucoma, shallow anterior chambers, corneal
clouding, and iris synechiae. Nevertheless, the dif-
ference in the rate of complications compared with
that in nonvitrectomized eyes was not statistically
significant [35].
Other potential complications of vitrectomy in the
treatment of ROP are extrapolated from complica-
tions of vitrectomy in adults. Corneal complications
may include the inadvertent loss of corneal epithe-
lium and corneal edema owing to high intraocular
pressure [40]. Poor postoperative healing of the
corneal epithelium may lead to recurrent erosions
and corneal edema. Corneal endothelial loss may lead
to cornea decompensation, the risk of which is
highest during fluid-gas exchange [40 – 42].
The literature is limited in the description of
cataract formation following vitrectomy performed
specifically for ROP. In the initial reports of lens-
sparing vitrectomy for infants, Maguire and Trese [43]
reported on 2 of 10 eyes developing lens opacities.
Similarly, Ferrone et al reported a 15% rate of cataract
formation after lens-sparing vitrectomy among pedi-
atric patients [36]. Studies of vitrectomy in adults
have provided several theories for the formation of
cataracts in the pediatric population. First, manipu-
lation of instruments may cause inadvertent mechan-
ical trauma to the lens [36]. Second, the mechanical
dissection of vitreoretinal adhesions from the poste-
rior lens surface may lead to lens opacities [43]. Third,
prolonged contact of the lens with intraocular gas can
result in posterior subcapsular opacities. These opac-
ities may clear with repositioning of the patient [40].
Glaucoma has not been reported extensively
following vitrectomy for ROP. Capone and Trese
reported the development of glaucoma in 2 of 40 eyes
that underwent lens-sparing vitrectomy for ROP [39].
The mechanism for glaucoma may be inflammation
from the surgery leading to scarring of the trabecular
meshwork. Likewise, as seen in adult patients,
postoperative bleeding may lead to degenerated red
blood cells and clogging of the aqueous outflow
system, that is, bghost cell glaucomaQ [44 – 46]. An
intraocular pressure rise has also been found follow-
ing the use of viscoelastic compounds [47] and gas
bubble intervention.
Scleral buckling has been shown to be an effective
treatment for eyes with stage 4 and stage 5 ROP
disease [48 – 51]. Few reports of complications fol-
580 M.S. Ibarra, A. Capone, Jr / Ophthalmol Clin N Am 17 (2004) 577 – 582
lowing scleral buckling procedures for ROP are found
in the literature. As is true for vitrectomy complica-
tions, complications from scleral buckling can only be
extrapolated from literature on retinal detachment
surgery in adults. A high encircling band has been
demonstrated to cause angle-closure glaucoma sec-
ondary to choroidal detachment and anterior rotation
of the ciliary body [52,53]. Another complication that
can follow scleral buckling is anterior segment
ischemia. The mechanism of this complication is not
completely understood but may be related to the
compromised anterior segment blood flow of the
anterior ciliary arteries when muscle disinsertion is
required [54]. Another proposed mechanism is that the
encircling band may cause obstruction of venous
outflow from the ciliary body [55]. Because scleral
buckling materials are foreign bodies, infection or
extrusion is a potential complication. The incidence of
infection or extrusion in adult surgery has been
reported to be approximately 1% [56]. Neither buckle
infection nor extrusion has been reported in the
ROP literature.
With any invasive procedure in ophthalmology,
there is always a risk of endophthalmitis. There are
no reports in the literature of endophthalmitis after
vitrectomy or scleral buckling for ROP. A limited
number of reports exist regarding endophthalmitis
after vitreoretinal surgery in adults. From these
reports, endophthalmitis has been estimated to occur
in 0.14% to 0.15% of adult vitrectomy cases [57,58].
Summary
The management of ROP has changed over the last
20 years. Screening protocols and recommendations
have helped ophthalmologists establish points of
treatment for this potentially blinding condition. The
fact that most infants who are screened for ROP never
develop threshold disease is indicative of the progress
made in identifying and treating the disease. For those
who do go on to develop ROP, effective therapies
exist. Although these therapies have inherent risks of
complications, they are often successful in treating the
disease. Further research must be conducted to better
understand the pathogenesis of ROP and to identify
the most effective ways of minimizing the risks of
complications in ROP treatments.
References
[1] Cryotherapy for Retinopathy of Prematurity Coopera-
tive Group. Multicenter trial of cryotherapy for reti-
nopathy of prematurity: preliminary results. Arch
Ophthalmol 1988;106(4):471 – 9.
[2] Cryotherapy for Retinopathy of Prematurity Coopera-
tive Group. Multicenter trial of cryotherapy for reti-
nopathy of prematurity: three-month outcome. Arch
Ophthalmol 1990;108(2):195 – 204.
[3] The Committee for the Classification of Retinopathy
of Prematurity. An international classification of reti-
nopathy of prematurity. Arch Ophthalmol 1984;102(8):
1130 – 4.
[4] Multicenter Trial of Cryotherapy for Retinopathy of
Prematurity. Ophthalmological outcomes at 10 years.
Arch Ophthalmol 2001;119(8):1110 – 8.
[5] McNamara JA, Tasman W, Vander JF, Brown GC.
Diode laser photocoagulation for retinopathy of pre-
maturity: preliminary results. Arch Ophthalmol 1992;
110(12):1714 – 6.
[6] Hunter DG, Repka MX. Diode laser photocoagulation
for threshold retinopathy of prematurity: a randomized
study. Ophthalmology 1993;100(2):238 – 44.
[7] Robinson R, O’Keefe M. Cryotherapy for retinopathy
of prematurity—a prospective study. Br J Ophthalmol
1992;76(5):289 – 91.
[8] Kaiser RS, Trese MT. Iris atrophy, cataracts, and hypo-
tony following peripheral ablation for threshold reti-
nopathy of prematurity. Arch Ophthalmol 2001;
119(4):615 – 7.
[9] Capone Jr A, Diaz-Rohena R, Sternberg Jr P, Mandell
B, Lambert HM, Lopez PF. Diode-laser photocoagu-
lation for zone 1 threshold retinopathy of prematurity.
Am J Ophthalmol 1993;116(4):444 – 50.
[10] Fallaha N, Lynn MJ, Aaberg Jr TM, Lambert SR.
Clinical outcome of confluent laser photoablation
for retinopathy of prematurity. J AAPOS 2002;6(2):
81 – 5.
[11] Landers 3rd MB, Toth CA, Semple HC, Morse LS.
Treatment of retinopathy of prematurity with argon
laser photocoagulation. Arch Ophthalmol 1992;110(1):
44 – 7.
[12] McGregor ML, Wherley AJ, Fellows RR, Bremer DL,
Rogers GL, Letson AD. A comparison of cryotherapy
versus diode laser retinopexy in 100 consecutive
infants treated for threshold retinopathy of prematurity.
J AAPOS 1998;2(6):360 – 4.
[13] Paysse EA, Lindsey JL, Coats DK, Contant Jr CF,
Steinkuller PG. Therapeutic outcomes of cryotherapy
versus transpupillary diode laser photocoagulation for
threshold retinopathy of prematurity. J AAPOS 1999;
3(4):234 – 40.
[14] Seiberth V, Linderkamp O, Vardarli I, Knorz MC,
Liesenhoff H. Diode laser photocoagulation for
threshold retinopathy of prematurity in eyes with tu-
nica vasculosa lentis. Am J Ophthalmol 1995;119(6):
748 – 51.
[15] Laser ROP Study Group. Laser therapy for retinopa-
thy of prematurity. Arch Ophthalmol 1994;112(2):
154 – 6.
[16] Seiberth V, Linderkamp O, Vardarli I. Transscleral vs
transpupillary diode laser photocoagulation for the
 M.S. Ibarra, A. Capone, Jr / Ophthalmol Clin N Am 17 (2004) 577 – 582
treatment of threshold retinopathy of prematurity. Arch
Ophthalmol 1997;115(10):1270 – 5.
[17] Irvine WD, Smiddy WE, Nicholson DH. Corneal and
iris burns with the laser indirect ophthalmoscope. Am J
Ophthalmol 1990;110(3):311 – 3.
[18] Hunt L. Complications of indirect laser photocoagula-
tion. Insight 1994;19(4):24 – 5.
[19] Lobes Jr LA, Bourgon P. Pupillary abnormalities
induced by argon laser photocoagulation. Ophthalmol-
ogy 1985;92(2):234 – 6.
[20] Simons BD, Wilson MC, Hertle RW, Schaefer DB.
Bilateral hyphemas and cataracts after diode laser
retinal photoablation for retinopathy of prematurity.
J Pediatr Ophthalmol Strabismus 1998;35(3):185 – 7.
[21] Rundle P, McGinnity FG. Bilateral hyphaema follow-
ing diode laser for retinopathy of prematurity. Br J
Ophthalmol 1995;79(11):1055 – 6.
[22] Prendiville PL, McDonnell PJ. Complications of laser
surgery. Int Ophthalmol Clin 1992;32(4):179 – 204.
[23] Doft BH, Blankenship GW. Single versus multiple
treatment sessions of argon laser panretinal photo-
coagulation for proliferative diabetic retinopathy.
Ophthalmology 1982;89(7):772 – 9.
[24] Lee GA, Lee LR, Gole GA. Angle-closure glaucoma
after laser treatment for retinopathy of prematurity. J
AAPOS 1998;2(6):383 – 4.
[25] Lakhanpal V, Schocket SS, Richards RD, Nirankari
VS. Photocoagulation-induced lens opacity. Arch Oph-
thalmol 1982;100(7):1068 – 70.
[26] McCanna R, Chandra SR, Stevens TS, Myers FL,
de Venecia G, Bresnick GH. Argon laser-induced cata-
ract as a complication of retinal photocoagulation.
Arch Ophthalmol 1982;100(7):1071 – 3.
[27] Lambert SR, Capone Jr A, Cingle KA, Drack AV.
Cataract and phthisis bulbi after laser photoablation for
threshold retinopathy of prematurity. Am J Ophthalmol
2000;129(5):585 – 91.
[28] Christiansen SP, Bradford JD. Cataract following diode
laser photoablation for retinopathy of prematurity.
Arch Ophthalmol 1997;115(2):275 – 6.
[29] Pogrebniak AE, Bolling JP, Stewart MW. Argon laser-
induced cataract in an infant with retinopathy of
prematurity. Am J Ophthalmol 1994;117(2):261 – 2.
[30] Gold RS. Cataracts associated with treatment for
retinopathy of prematurity. J Pediatr Ophthalmol
Strabismus 1997;34(2):123 – 4.
[31] Capone Jr A, Drack AV. Transient lens changes after
diode laser retinal photoablation for retinopathy of
prematurity. Am J Ophthalmol 1994;118(4):533 – 5.
[32] Drack AV, Burke JP, Pulido JS, Keech RV. Transient
punctate lenticular opacities as a complication of argon
laser photoablation in an infant with retinopathy of
prematurity. Am J Ophthalmol 1992;113(5):583 – 4.
[33] Campolattaro BN, Lueder GT. Cataract in infants
treated with argon laser photocoagulation for threshold
retinopathy of prematurity. Am J Ophthalmol 1995;
120(2):264 – 6.
[34] Christiansen SP, Bradford JD. Cataract in infants
treated with argon laser photocoagulation for threshold
581
retinopathy of prematurity. Am J Ophthalmol 1995;
119(2):175 – 80.
[35] Quinn GE, Dobson V, Barr CC, Davis BR, Palmer EA,
Robertson J, et al. Visual acuity of eyes after vitrectomy
for retinopathy of prematurity: follow-up at 5 1/2 years.
The Cryotherapy for Retinopathy of Prematurity Co-
operative Group. Ophthalmology 1996;103(4):595 – 600.
[36] Ferrone PJ, Harrison C, Trese MT. Lens clarity after
lens-sparing vitrectomy in a pediatric population.
Ophthalmology 1997;104(2):273 – 8.
[37] Zilis JD, deJuan E, Machemer R. Advanced reti-
nopathy of prematurity: the anatomic and visual results
of vitreous surgery. Ophthalmology 1990;97(6):821 – 6.
[38] Kono T, Oshima K, Fuchino Y. Surgical results and
visual outcomes of vitreous surgery for advanced
stages of retinopathy of prematurity. Jpn J Ophthalmol
2000;44(6):661 – 7.
[39] Capone Jr A, Trese MT. Lens-sparing vitreous surgery
for tractional stage 4A retinopathy of prematurity
retinal detachments. Ophthalmology 2001;108(11):
2068 – 70.
[40] Stern WH. Complications of vitrectomy. Int Ophthal-
mol Clin 1992;32(4):205 – 12.
[41] Friberg TR, Doran DL, Lazenby FL. The effect of
vitreous and retinal surgery on corneal endothelial
cell density. Ophthalmology 1984;91(10):1166 – 9.
[42] Mittl RN, Koester CJ, Kates MR, Wilkes E. Endothe-
lial cell counts following pars plana vitrectomy in
pseudophakic and aphakic eyes. Ophthalmic Surg
1989;20(1):13 – 6.
[43] Maguire AM, Trese MT. Lens-sparing vitreoretinal
surgery in infants. Arch Ophthalmol 1992;110(2):
284 – 6.
[44] Weinberg RS, Peyman GA, Huamonte FU. Elevation
of intraocular pressure after pars plana vitrectomy.
Albrecht Von Graefes Arch Klin Exp Ophthalmol
1976;200(2):157 – 61.
[45] Campbell DG, Simmons RJ, Tolentino FI, McMeel
JW. Glaucoma occurring after closed vitrectomy. Am J
Ophthalmol 1977;83(1):63 – 9.
[46] Campbell DG, Simmons RJ, Grant WM. Ghost cells
as a cause of glaucoma. Am J Ophthalmol 1976;81(4):
441 – 50.
[47] Sihota R, Saxena R, Agarwal HC. Intravitreal sodium
hyaluronate and secondary glaucoma after complicated
phacoemulsification. J Cataract Refract Surg 2003;
29(6):1226 – 7.
[48] Trese MT. Scleral buckling for retinopathy of prema-
turity. Ophthalmology 1994;101(1):23 – 6.
[49] Ricci B, Santo A, Ricci F, Minicucci G, Molle F.
Scleral buckling surgery in stage 4 retinopathy of
prematurity. Graefes Arch Clin Exp Ophthalmol 1996;
234(Suppl 1):S38 – 41.
[50] Greven C, Tasman W. Scleral buckling in stages 4B
and 5 retinopathy of prematurity. Ophthalmology 1990;
97(6):817 – 20.
[51] Chuang YC, Yang CM. Scleral buckling for stage 4
retinopathy of prematurity. Ophthalmic Surg Lasers
2000;31(5):374 – 9.
582 M.S. Ibarra, A. Capone, Jr / Ophthalmol Clin N Am 17 (2004) 577 – 582
[52] Perez RN, Phelps CD, Burton TC. Angle-closure glau-
coma following scleral buckling operations. Trans Am
Acad Ophthalmol Otolaryngol 1976;81(2): 247 – 52.
[53] Berler DK, Goldstein B. Scleral buckles and rotation
of the ciliary body. Arch Ophthalmol 1979;97(8):
1518 – 21.
[54] Freeman HM, Hawkins WR, Schepens CL. Anterior
segment necrosis: an experimental study. Arch Oph-
thalmol 1966;75(5):644 – 50.
[55] Hayreh SS, Baines JA. Occlusion of the vortex veins:
an experimental study. Br J Ophthalmol 1973;57(4):
217 – 38.
[56] Smiddy WE, Miller D, Flynn Jr HW. Scleral buckle
removal following retinal reattachment surgery: clini-
cal and microbiologic aspects. Ophthalmic Surg 1993;
24(7):440 – 5.
[57] Bacon AS, Davison CR, Patel BC, Frazer DG, Ficker
LA, Dart JK. Infective endophthalmitis following
vitreoretinal surgery. Eye 1993;7(Pt 4):529 – 34.
[58] Ho PC, Tolentino FI. Bacterial endophthalmitis after
closed vitrectomy. Arch Ophthalmol 1984;102(2):
207 – 10.
 Ophthalmol Clin N Am 17 (2004) 583 – 590
Anterior segment complications following periocular and
intraocular injections
Allesandro Castellarin, MD, Dante J. Pieramici, MD*
California Retina Consultants and Research Foundation, 515 East Micheltorena Street, Suite C, Santa Barbara, CA 93103, USA
Periocular and intraocular injections have always
been a part of routine procedures in a retina practice.
Recently, the advent of new drugs has broadened
the therapeutic armamentarium, with subsequent in-
creased indications for such injections. As is true for
any invasive procedure, potential complications
deserve consideration. This article reviews the post-
operative complications following periocular and
intraocular injections, with particular attention to
complications of the anterior segment.
Pneumatic retinopexy
Pneumatic retinopexy is a retina reattachment
procedure that uses cryopexy or laser photocoagula-
tion in combination with intravitreal injection of an
expandable gas. Following the procedure, correct
positioning of the patient is required. The gas slowly
reabsorbs and works as a tamponade to the retinal
break until a permanent scar forms [1 – 6]. Compli-
cations of pneumatic retinopexy can be intraoperative
or postoperative and may involve several anterior
segment structures.
Cataract
Intraoperative lens touch has been reported during
pneumatic retinopexy [7] and usually occurs during
Financial support for this article was provided by The
California Retina Research Foundation.
* Corresponding author.
E-mail address: dpieramici@yahoo.com (D.J. Pieramici).
0896-1549/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ohc.2004.06.010
the insertion of the needle or during the paracentesis.
This event can easily be prevented by aiming the
needle posteriorly and slightly inferiorly into the
midvitreous. Reversible changes in lens transparency
have been reported after gas injection in the vitreous
cavity [8]. It is reasonable to hypothesize that gas
may have a biochemical or mechanical influence on
the metabolism of the lens with subsequent cataract
formation. Lincoff et al postulated that perfluorocar-
bon gas-induced cataract resulted from mechanical
interference with transport across the posterior
capsule. Lenticular opacities are uncommon after
pneumatic retinopexy. In the review of the literature
by Hilton et al [9], cataract was not mentioned as a
late complication of pneumoretinopexy. In a pro-
spective study to examine the effect on the lens
caused by pneumatic retinopexy with SF6, significant
lens opacifications were not found 2 years after
surgery [10]. In another series with 12 years of
follow-up, the incidence of cataract after pneumatic
retinopexy was estimated to be less than 1% [11].
Intraocular pressure elevation
Acutely following injection, intraocular pressure
can rise significantly up to 180 mm Hg. Hilton and
Grizzard reported an elevation of intraocular pressure
up to 80 mm Hg immediately following the injection,
with temporary closure of the central retina artery [1].
There has been one reported case of irreversible
central artery occlusion after pneumatic retinopexy
[12]. Postoperative intraocular pressure should be
controlled. If intraocular pressure higher than 55 mm
Hg or occlusion of the central artery occurs, para-
centesis should be considered.
ophthalmology.theclinics.com
584 A. Castellarin, D.J. Pieramici / Ophthalmol Clin N Am 17 (2004) 583 – 590
Spikes in intraocular pressure are unlikely after
1 day, and the pressure ranges between 7 and 30 mm
Hg. Postoperative angle-closure glaucoma is rare. It
is usually secondary to pupillary block or can be
caused by aqueous misdirection, especially in
aphakic eyes.
Sub – pars plana gas entrapment
Inadvertent injection of gas in the suprachoroidal
space, under the pars plana epithelium, in the anterior
hyaloidal space, and within the pseudophakic lens
capsule has been reported. Sub – pars plana gas
entrapment is usually identified as a sausage-shaped
bubble posterior to the lens periphery. It occurs if the
needle is not completely within the vitreous and can
be avoided by passing the needle deep into the
vitreous cavity, approximately 7 mm. Small bubbles
in the sub – pars plana space can be observed. Larger
trapped bubbles may cause peripheral cataract and
potentially incite vitreoretinal proliferation; therefore,
some surgeons have attempted evacuation [13].
Anterior chamber gas entrapment has also been
reported in pseudophakic and phakic patients. A
small gas bubble in the anterior chamber is rarely
a significant problem and generally resolves with-
out sequelae.
Subconjunctival gas
Subconjunctival gas is an inconsequential effect
resulting from the escape of gas from the globe into
the subconjunctival space. It can be reduced by
placing a sterile cotton tip applicator on the injection
site [14].
Iris and vitreous incarceration
Vitreous incarceration has been reported at the
needle tract following injection. This event is
prevented in a similar manner as described previously
to prevent escape of subconjunctival gas. Vitreous
and iris incarceration have also been reported at the
paracentesis site of aphakic eyes or in presence of an
open capsule.
Uveitis
Uveitis is a rare complication that has been
described in few cases. Usually, it is mild, and all
of the cases reported have cleared spontaneously.
Prophylactic treatment with topical steroids is rarely
necessary but may be reasonable, particularly if
extensive cryotherapy is performed.
Refractive changes
In a multicenter randomized controlled clinical
trial comparing pneumatic retinopexy and scleral
buckling, 3% of patients undergoing pneumatic
retinopexy had a myopic shift of around 1 D
compared with 68% of patients treated with a scleral
buckle [3]. The cause of the refractive change in
pneumatic patients is not known but may be
secondary to nuclear sclerotic changes.
Wound dehiscence
A clear corneal incision from recent cataract
surgery may dehisce during the gas injection. Two
cases have been reported. In both cases, the pneu-
matic procedure was performed less than 3 weeks
after the cataract surgery [15]. Particular attention
should be taken in such cases and a pre-injection
paracentesis performed.
Vitreous aspiration
Vitreous aspiration is a relatively safe diagnostic
procedure, often providing an adequate sample to
differentiate infectious, malignant, and inflammatory
causes of uveitis in most patients. Specimens are
obtained by biopsy with a 20-gauge vitrectomy
cutting instrument or by needle tap with a 22- to
27-gauge needle. Despite being a relatively straight-
forward procedure, potential anterior segment com-
plications include lens touch, lens dislocation, iris and
vitreous prolapse, hypotony or elevated intraocular
pressure, and hyphema. In the Endophthalmitis
Vitrectomy Study, intraoperative hyphema occurred
in two needle tap eyes (3%) and three vitrectomy eyes
(2%) [16].
Intraocular tissue plasminogen activator
Tissue plasminogen activator (tPA) is a fibrino-
lytic agent and was initially used to treat acute
myocardial infarction. Intraocular injection of tPA has
been employed in ophthalmology to treat conditions
such as postvitrectomy fibrinoid syndrome [17 – 19],
to lyse fibrin occluding inferior peripheral iridec-
tomy in eyes with silicone oil, and to treat submacular
hemorrhage [20,21]. Recently, it has also been
employed for the treatment of postcataract fibrin-
ous membranes.
Retinal toxicity and endophthalmitis [22] are the
main concerns following the procedure. Among the
 A. Castellarin, D.J. Pieramici / Ophthalmol Clin N Am 17 (2004) 583 – 590
anterior segment complications, the most frequent is
hyphema. A temporary increase in intraocular pres-
sure, iritis, and posterior synechiae also have been
reported. Another concern is the potential for corneal
toxicity. Fibrin degradation products are chemotactic
for leukocytes and can cause uveitis and corneal
decompensation. Transient clouding of the corneal
endothelium or corneal stromal is a common finding
in patients treated with tPA. These complications
have been reversible, except for irreversible endothe-
lial decompensation in one eye with Fuchs’ corneal
dystrophy [23].
Steroid injections
Corticosteroids have been used for years in
ophthalmology. They inhibit prostaglandin and leu-
kotriene pathways, producing an anti-inflammatory
effect, stabilizing blood vessels and the blood-retinal
barrier. Periocular and intravitreal injections have
minimized the onset of systemic complications while
allowing high drug levels at the necessary therapeu-
tic location.
Periocular injections
Periocular injections of steroids have been used
for the past 50 years. Indications may vary, but the
main reasons for these injections are the presence of
intermediate or posterior uveitis and cystoid macular
edema. Drug formulation can vary, and steroids can
be injected in the form of methylprednisolone or as
depot (triamcinolone acetonide). There are various
types of periocular injections, including subconjunc-
tival, anterior sub-Tenon, posterior sub-Tenon, and
retrobulbar. Associated complications include ocular
perforation [24], injection into the choroid or retinal
circulation [25], glaucoma, and cataracts [26]. Blepha-
roptosis, proptosis, orbital granulomas, orbital fat
atrophy or herniation, strabismus, delayed hyper-
sensitivity reactions, conjunctival changes, and infec-
tions have also been reported [27].
Increased intraocular pressure
Increased intraocular pressure can occur with
topical steroid or periocular steroid injections. Corti-
costeroids induce an elevation in intraocular pressure
through an increased resistance to aqueous humor
outflow. Although many theories exist, the exact
mechanism of decreased outflow is not known. The
exact frequency of glaucoma is not well known, and
the increased pressure may exacerbate pre-existing
585
glaucoma. The incidence of increased intraocular
pressure reported by different investigators ranges
from 1.8% to 50% [27,28]. It varies according to the
drug used (depot versus short acting), the location of
the injection, and the number of injections. The onset
of increased intraocular pressure also varies. A spike
in intraocular pressure usually occurs at around
3 weeks from treatment, and there seems to be an
association between the number of injections and the
risk of increased intraocular pressure [28]. Posterior
sub-Tenon injection of steroids seems to be less likely
than anterior sub-Tenon injection to cause an increase
of the intraocular pressure [29].
If repository corticosteroids (triamcinolone aceto-
nide) are employed, increased intraocular pressure
has a delayed onset and tends to last longer. Kalina
[30] reported that after subconjunctival injection of
repository steroids, the onset of increased intraocular
pressure was noted at a mean of 7.1 weeks, with a
mean duration of 3 months. In a different series, the
increased in pressure was noted at a mean of
9.4 weeks, with a duration of 3.2 months [31]. In-
tractable glaucoma, unresponsive to maximum medi-
cal therapy, has also been reported 13 months after
sub-Tenon injection of triamcinolone. In these
patients, the intraocular pressure decreased only after
excision of the sub-Tenon’s corticosteroid depot,
and biochemically active drug could be detected in
the excised tissue [32].
Cataract
Cataract is also one of the most common anterior
segment complications from periocular steroids. The
mechanism of steroid-induced cataract is controver-
sial, but it is hypothesized that glucocorticoids are
covalently bound to lens proteins, with subsequent
destabilization of the protein structure allowing
further oxidation leading to cataract. Others consider
the corticosteroid inhibition of the Na-K ATPase
pump as a mechanism for posterior subcapsular
cataract formation [33]. Its exact frequency is not
known and depends on the injections given and the
age of the patient. Helm and coauthors, in a ret-
rospective review of patients who received posterior
sub-Tenon injection of triamcinolone for the treat-
ment of intermediate uveitis, reported the develop-
ment of significant cataracts in 4 (36.4%) of 11 phakic
eyes 10 months to 4 years after treatment [28].
There also seems to be a genetic predisposition
with an individual susceptibility. An ethnic suscepti-
bility has also been shown, with Hispanics more
predisposed to posterior subcapsular changes than
whites or African Americans [34].
586 A. Castellarin, D.J. Pieramici / Ophthalmol Clin N Am 17 (2004) 583 – 590
Cosmetic complications
Among the complications of periocular injections,
blepharoptosis is common and usually associated
with superior injections [35]. Herniation of the lower
eyelid orbital fat has also been described following
multiple orbital floor injections of triamcinolone
through the inferior eyelid [36]. Orbital fat atrophy
has been described following periocular injections
through the eyelid [37]. Proptosis and orbital lip-
omatosis have been reported following retrobulbar
injections of steroids [38]. It has been speculated that
corticosteroid diffusing anteriorly may weaken the
orbital septum, facilitating a herniation of orbital
fat. An immune-mediated reaction and multiple
steroid component interaction may also have a role
in the excessive deposition of adipose tissue. Ana-
phylactic and delayed hypersensitivity reactions have
been reported secondary to local triamcinolone
acetonide use.
More recently, conjunctival necrosis has been de-
scribed following subconjunctival injection of methyl-
prednisolone acetate suspension and triamcinolone
acetonide [39]. A possible explanation could be
localized toxic reaction to the drug or vehicle, or an
underlying autoimmune disease. More rarely, con-
junctival granulomas and orbital granulomas have
been reported to be associated with periocular steroids
[40]. Long-acting steroids should be avoided in
patients with scleritis, because they may inhibit
collagen cross-linking and cause staphyloma and
perforation of the globe.
Strabismus
Although rare, strabismus and diplopia have been
reported and can be secondary to inadvertent injec-
tion into the muscle. These complications may be
temporary or permanent.
Intravitreal injections
Recently, intravitreal steroid injections have be-
come popular for several conditions, including
refractory cystoid macular edema, clinically signifi-
cant diabetic macular edema [41], choroidal neo-
vascularization, proliferative diabetic retinopathy,
chronic prephthisical ocular hypotony, and chronic
uveitis [41 – 43].
Increased intraocular pressure
As mentioned previously, one of the ocular side
effects of steroids is elevation of intraocular pressure
that can lead to open-angle glaucoma. The incidence
of increased intraocular pressure following intra-
vitreal steroid injection is varied. In a series of
26 eyes, secondary ocular hypertension was observed
in 4% of the eyes treated with a single dose of
intravitreal triamcinolone (4 mg/0.1mL) for exudative
age-related macular degeneration [44]. In a random-
ized clinical trial using intravitreal triamcinolone for
exudative age-related macular degeneration, Danis
et al [45] found intraocular pressure elevation in
25% of treated patients. Another randomized clinical
trial of a single dose of intravitreal triamcinolone
acetonide for neovascular age-related macular de-
generation found an intraocular pressure elevation
in 41% of the treated eyes versus 4% of the placebo
group [46,47]. In a small series of intravitreal
triamcinolone for cystoid macular edema, the inci-
dence of increased intraocular pressure was as high
as 83% [48]. Using higher concentrations of triam-
cinolone (25 mg), elevation of intraocular pressure
was recorded in 50% of the eyes, starting between
1 and 2 months from the injection [49].
The duration of an intravitreal injection of tri-
amcinolone is not known, but it is likely that sig-
nificant levels of triamcinolone may persist inside
the eye for at least 4 months. In most studies, the
pressure spike occurred 2 months from the injection
and normalized in about 6 months, and only a few
cases did not respond to medical therapy. The
decision to treat elevated intraocular pressure should
be based on the degree of elevation, the amount of
cupping of the optic nerve head, and whether there
is a history or family history of glaucoma.
Cataract
Steroid-induced cataract is a well-known compli-
cation of topical steroids. In a recent randomized
study of triamcinolone injection for exudative age-
related macular degeneration, by 2 years, 8 (24.2%)
of 33 triamcinolone-treated eyes had progression of
cataract compared with 0 of 22 placebo-treated eyes
(P = .02), and cataract surgery was performed in
16 (28.6%) of 56 treated eyes [47]. In another ran-
domized clinical trial of a single dose of intravitreal
triamcinolone acetonide for neovascular age-related
macular degeneration, cataract was reported in 4% of
the treated eyes versus none of the placebo group
after 12 months, but the difference was not statisti-
cally significant [44]. Martidis et al reported cataract
progression in 1 of 16 eyes treated with intravitreal
triamcinolone for diabetic macular edema [42]. In
another small series, subcapsular cataract developed
 A. Castellarin, D.J. Pieramici / Ophthalmol Clin N Am 17 (2004) 583 – 590
in two of six eyes following triamcinolone injection
for cystoid macular edema in uveitis [50].
Hypopyon and pseudohypopyon
One of the most devastating complications of an
intraocular injection is infection. In a recent study, the
overall incidence of infectious endophthalmitis was
0.87% in the first 6 weeks following intravitreal
triamcinolone acetonide injection [51]. Most of the
patients with infectious endophthalmitis presented
1 week following the injection. The most common
clinical findings were pain, iritis, vitritis, hypopyon,
and decreased vision. All culture-positive cases lost
vision in a comparison of pre-injection visual acuity.
Identified predisposing risk factors included diabetes
mellitus, the use of multi-injection bottles, pre-
existing filtering blebs, and blepharitis [51].
Along with infection, there has been a rising
number of reports of patients presenting with a clini-
cal presentation simulating endophthalmitis. Most of
the patients with presumed sterile endophthalmitis
have presented acutely within 2 days from the
injection. Patients usually present with an anterior
chamber reaction, hypopyon, and various grades of
vitritis. In a series of 440 injections, the reported
incidence of noninfectious endophthalmitis was 1.6%
[52]. Most of the patients with presumed sterile
endophthalmitis recovered their visual acuity to the
pre-injection level. In patients with bacterial endoph-
thalmitis, visual acuity recovery took much longer.
It has been speculated that presumed noninfec-
tious or sterile endophthalmitis may be secondary to
an inflammatory reaction to the preservative present
in the triamcinolone. Its commercial form, Kenalog,
contains benzyl alcohol, carboxymethylcellulose
sodium, polysorbate, and sodium hydroxide, which
can trigger a potent inflammatory reaction. In
addition, the vials may contain bacterial toxins,
which can also trigger inflammation once injected.
In other cases, especially in pseudophakic patients,
it has been speculated that triamcinolone material
may migrate in the anterior chamber, resembling a
true hypopyon.
Amphotericin B injections
Amphotericin was the first polyene effective in
treating systemic mycosis, and it is one of the most
important drugs for systemic and intraocular fungal
infections. Because of the poor ocular penetration and
the high incidence of nephrotoxicity when used sys-
temically, intravitreal administration has been recom-
587
mended in the treatment of fungal endophthalmitis.
No severe complications have been observed clini-
cally with anterior or posterior segment therapeutic
injections in doses of 10 mg or less; however, the re-
ported electroretinographic changes in a few pa-
tients have raised the speculation of possible retinal
toxicity [53]. Regarding anterior segment toxic
effects, conflicting data exist. In a rabbit model,
Foster et al [54] found that anterior chamber injection
of as much as 50 mg of amphotericin B failed to cause
corneal or lenticular toxicity. Other investigators
found cataract formation in the majority of eyes of
albino rabbits receiving 75mg or more [55].
Recent clinical data have failed to disclose clini-
cal evidence of corneal or lenticular toxicity [56].
Fibrinoid iritis was occasionally observed in patients
after intraocular injection of 10mg of amphotericin B
[57]. This finding was also confirmed in animal
studies and raises the concern for potential ocular
toxicity of amphotericin B.
Antiviral injections
Intravitreal injections of antiviral agents have been
advocated for several conditions, including cytome-
galovirus (CMV) retinitis, acute retinal necrosis, and
progressive outer retinal necrosis. Ganciclovir and
foscarnet are among the most commonly used drugs.
Both drugs are virostatic agents and do not eliminate
viral infections. Ganciclovir is a nucleoside analogue
of deoxyguanosine, and its structure is similar to
acyclovir. Foscarnet is an inhibitor of viral DNA
polymerase as well as a reverse transcriptase. Among
the posterior segment complications of intravitreous
injections, retinal detachment, vitreous hemorrhage,
and endophthalmitis have been reported [58]. Re-
ported anterior segment complications include scleral
induration and iritis. Mild iris atrophy with subse-
quent heterochromia has also been reported with
high doses of intravitreal ganciclovir and foscarnet.
Intravitreal cidofovir, an acyclic nucleoside ana-
logue, has also been employed for the treatment of
CMV retinitis. Owing to the high incidence of hypo-
tony and uveitis, it is no longer recommended for
local therapy.
Anti-angiogenic agents
Anti-vascular endothelial growth factor (VEGF)
therapy is a promising new avenue for the treatment
of neovascular diseases of the eye, including exuda-
tive macular degeneration and diabetic retinopathy.
588 A. Castellarin, D.J. Pieramici / Ophthalmol Clin N Am 17 (2004) 583 – 590
Recently, phase III study results were released for
the use of intravitreal anti-VEGF aptamer (Macugen)
as a treatment for subfoveal choroidal neovascu-
larization secondary to age-related macular degenera-
tion. Among the posterior adverse effects, vitreous
floaters, vitreous haze, and vitreous prolapse were
reported. Among the anterior segment adverse
effects, anterior chamber inflammation, ptosis, lid
edema, subconjunctival hemorrhage, conjunctival
injection, increased intraocular pressure, dry eye,
and corneal abrasion were reported, and most of
these complications resolved within 24 hours [59].
All of these side effects were transient and did not
require intervention. Future studies and longer
follow-up must address the long-term safety of such
therapy, which is likely to become in the near fu-
ture the routine treatment for exudative age-related
macular degeneration.
Antibiotic injections
The most commonly used intraocular antibiotics
for the treatment of bacterial endophthalmitis are
ceftazidime and vancomycin and, rarely, the amino-
glycoside amikacin. Potential side effects can derive
from the injection itself or can be related to the
toxicity of the agent once injected. Amikacin, in
doses higher than 500 mg, and vancomycin, at doses
greater than 10 mg, may cause retinal toxicity. With
doses at the therapeutic range, no toxic effects to the
lens or the corneal endothelium were reported [60].
Summary
Intraocular and periocular injections have become
a part of routine procedures in retina practice. The
number of intravitreal injections has surged owing to
the advent of new treatments, such as the use of
intravitreal triamcinolone, and the advent of new
antiangiogenic drugs. Any intervention has risks, and
the recent propagation of such invasive procedures
has generated a new wave of complications.
Complications may be intraoperative, related to
the surgical procedure, or postoperative. A careful
preoperative examination and an appropriate surgical
technique are essential prerequisites to prevent com-
plications. Especially in the setting of an intraocular
or periocular injection, one must explain the risks of
the procedure so that informed consent can be ob-
tained. Patient selection is cardinal, and compliance
with postoperative care and follow-up helps to mini-
mize or prevent postoperative complications.
References
[1] Hilton GF, Grizzard WS. Pneumatic retinopexy:
a two-step outpatient operation without conjunctival
incision. Ophthalmology 1986;93(5):626 – 41.
[2] Tornambe PE, Hilton GF, Kelly NF, Salzano TC, Wells
JW, Wendel RT. Expanded indications for pneumatic
retinopexy. Ophthalmology 1988;95(5):597 – 600.
[3] Tornambe PE, Poliner LS, Hilton GF, Grizzard WS.
Comparison of pneumatic retinopexy and scleral buckl-
ing in the management of primary rhegmatogenous reti-
nal detachment. Am J Ophthalmol 1999;127(6):741 – 3.
[4] Hilton GF, Kelly NE, Salzano TC, Tornambe PE,
Wells JW, Wendel RT. Pneumatic retinopexy: a col-
laborative report of the first 100 cases. Ophthalmology
1987;94(4):307 – 14.
[5] Han DP, Mohsin NC, Guse CE, Hartz A, Tarkanian
CN. Comparison of pneumatic retinopexy and scleral
buckling in the management of primary rhegmatoge-
nous retinal detachment. Southern Wisconsin Pneu-
matic Retinopexy Study Group. Am J Ophthalmol
1998;126(5):658 – 68.
[6] Tornambe PE, Hilton GF, Brinton DA, Flood TP,
Green S, Grizzard WS, et al. Pneumatic retinopexy:
a two-year follow-up study of the multicenter clinical
trial comparing pneumatic retinopexy with scleral
buckling. Ophthalmology 1991;98(7):1115 – 23.
[7] Tornambe PE, Hilton GF. Pneumatic retinopexy:
a multicenter randomized controlled clinical trial
comparing pneumatic retinopexy with scleral buckling.
The Retinal Detachment Study Group. Ophthalmology
1989;96(6):772–83.
[8] Lincoff H, Mardirossian J, Lincoff A, Liggett P,
Iwamoto T, Jakobiec F. Intravitreal longevity of three
perfluorocarbon gases. Arch Ophthalmol 1980;98(9):
1610 – 1.
[9] Hilton GF, Tornambe PE, Brinton DA, Flood TP,
Green S, Grizzard WS, et al. The complication of
pneumatic retinopexy. Trans Am Ophthalmol Soc
1990;88:191 – 207; discussion, 207 – 10.
[10] Mougharbel M, Koch FH, Boker T, Spitznas M. No
cataract two years after pneumatic retinopexy. Oph-
thalmology 1994;101(7):1191 – 4.
[11] Tornambe PE. Pneumatic retinopexy: the evolution of
case selection and surgical technique. A twelve-year
study of 302 eyes. Trans Am Ophthalmol Soc 1997;
95:551 – 78.
[12] Abe T, Nakajima A, Nakamura H, Ishikawa M,
Sakuragi S. Intraocular pressure during pneumatic reti-
nopexy. Ophthalmic Surg Lasers 1998;29(5):391 – 6.
[13] Hilton GF, Tornambe PE. An analysis of intraoperative
and postoperative complications: the Retinal Detach-
ment Study Group. Retina 1991;11(3):285 – 94.
[14] Taher RM, Haimovici R. Anterior chamber gas entrap-
ment after phakic pneumatic retinopexy. Retina 2001;
21(6):681 – 2.
[15] Jun AS, Pieramici DJ, Bridges WZ. Clear corneal
cataract wound dehiscence during pneumatic retino-
pexy. Arch Ophthalmol 2000;118(6):847 – 8.
 A. Castellarin, D.J. Pieramici / Ophthalmol Clin N Am 17 (2004) 583 – 590
[16] Han DP, Wisniewski SR, Kelsey SF, Doft BH, Barza
M, Pavan PR. Microbiologic yields and complication
rates of vitreous needle aspiration versus mechanized
vitreous biopsy in the Endophthalmitis Vitrectomy
Study. Retina 1999;19(2):98 – 102.
[17] Williams GA, Lambrou FH, Jaffe GA, Snyder RW,
Green GD, Devenyi RG, et al. Treatment of post-
vitrectomy fibrin formation with intraocular tissue
plasminogen activator. Arch Ophthalmol 1988;106(8):
1055 – 8.
[18] Williams DF, Bennett SR, Abrams GW, Han DP,
Mieler WF, Jaffe GJ, et al. Low-dose intraocular tissue
plasminogen activator for treatment of postvitrectomy
fibrin formation. Am J Ophthalmol 1990;109(5):606 – 7.
[19] Boldt HC, Abrams GW, Murray TG, Han DP, Mieler
WF. The lowest effective dose of tissue plasminogen
activator for fibrinolysis of postvitrectomy fibrin.
Retina 1992;12(3 Suppl):S75 – 9.
[20] Heriot WJ. Intravitreal gas and tPA: an outpatient
procedure for submacular hemorrhage. Presented at
the AAO annual vitreoretinal update. Chicago, Octo-
ber 1996.
[21] Ibanez HE, Williams DF, Thomas MA, Ruby AJ,
Meredith TA, Boniuk I, et al. Surgical management of
submacular hemorrhage: a series of 47 consecutive
cases. Arch Ophthalmol 1995;13(1):62 – 9.
[22] Hassan AS, Johnson MW, Schneiderman TE, Regillo
CD, Tornambe PE, Poliner LS, et al. Management of
submacular hemorrhage with intravitreous tissue
plasminogen activator injection and pneumatic dis-
placement. Ophthalmology 1999;106(10):1900 – 6; dis-
cussion, 1906 – 7.
[23] Rehfeldt K, Hoh H. Therapeutic and prophylactic
application of TPA (recombinant tissue plasminogen
activator) into the anterior chamber of the eye.
Ophthalmologe 1999;96(9):587 – 93.
[24] Giles CL. Bulbar perforation during periocular injec-
tion of corticosteroids. Am J Ophthalmol 1974;
77(4):438 – 41.
[25] Morgan CM, Schatz H, Vine AK, Cantrill HL, Da-
vidorf FH, Gitter KA, et al. Ocular complications
associated with retrobulbar injections. Ophthalmology
1988;95(5):660 – 5.
[26] Schlaegel Jr TF, Weber JC. Treatment of pars planitis.
II. Corticosteroids. Surv Ophthalmol 1977;22(2):
125 – 30.
[27] O’Connor GR. Periocular corticosteroid injections:
uses and abuses. Eye Ear Nose Throat Mon 1976;
55(3):83 – 8.
[28] Helm CJ, Holland GN. The effects of posterior
subtenon injection of triamcinolone acetonide in
patients with intermediate uveitis. Am J Ophthalmol
1995;120(1):55 – 64.
[29] Herschler J. Increased intraocular pressure induced by
repository corticosteroids. Am J Ophthalmol 1976;
82(1):90 – 3.
[30] Kalina RE. Increased intraocular pressure following
subconjunctival corticosteroid administration. Arch
Ophthalmol 1969;81(6):788 – 90.
589
[31] Herschler J. Increased intraocular pressure induced by
repository corticosteroids. Am J Ophthalmol 1976;
82(1):90 – 3.
[32] Akduman L, Kolker AE, Black DL, Del Priore LV,
Kaplan HJ. Treatment of persistent glaucoma second-
ary to periocular corticosteroids. Am J Ophthalmol
1996;122(2):275 – 7.
[33] Urban Jr RC, Cotlier E. Corticosteroid-induced cata-
racts. Surv Ophthalmol 1986;31(2):102 – 10.
[34] Loredo A, Rodriguez RS, Murillo L. Cataracts after
short-term corticosteroid treatment. N Engl J Med
1972;286(3):160.
[35] Riordan-Eva P, Lightman S. Orbital floor steroid
injections in the treatment of uveitis. Eye 1994;
8(Pt 1):66 – 9.
[36] Smith JR, George RK, Rosenbaum JT. Lower eyelid
herniation of orbital fat may complicate periocular
corticosteroid injection. Am J Ophthalmol 2002;
133(6):845 – 7.
[37] Nozik RA. Orbital rim fat atrophy after repository
periocular corticosteroid injection. Am J Ophthalmol
1976;82(6):928 – 30.
[38] Gupta OP, Boynton JR, Sabini P, Markowitch Jr W,
Quatela VC. Proptosis after retrobulbar corticosteroid
injections. Ophthalmology 2003;110(2):443 – 7.
[39] Agrawal S, Agrawal J, Agrawal TP. Conjunctival ul-
ceration following triamcinolone injection. Am J Oph-
thalmol 2003;136(3):539 – 40.
[40] Abel AD, Carlson JA, Bakri S, Meyer DR. Sclerosing
lipogranuloma of the orbit after periocular steroid
injection. Ophthalmology 2003;110(9):1841 – 5.
[41] Martidis A, Duker JS, Greenberg PB, Rogers AH,
Puliafito CA, Reichel E, et al. Intravitreal triamcino-
lone for refractory diabetic macular edema. Ophthal-
mology 2002;109(5):920 – 7.
[42] Martidis A, Duker JS, Puliafito CA. Intravitreal triam-
cinolone for refractory cystoid macular edema second-
ary to birdshot retinochoroidopathy. Arch Ophthalmol
2001;119(9):1380 – 3.
[43] Penfold PL, Gyory JF, Hunyor AB, Billson FA. Ex-
udative macular degeneration and triamcinolone: a pi-
lot study. Aust N Z J Ophthalmol 1995;23:293 – 8.
[44] Challa JK, Gillies MC, Penfold PL, Gyory JF, Hunyor
AB, Billson FA. Exudative macular degeneration and
intravitreal triamcinolone: 18 month follow up. Aust
N Z J Ophthalmol 1998;26(4):277 – 81.
[45] Danis RP, Ciulla TA, Pratt LM, Anliker W. Intravitreal
triamcinolone acetonide in exudative age-related macu-
lar degeneration. Retina 2000;20:244 – 50.
[46] Gillies MC, Simpson JM, Luo W, Penfold P, Hunyor
AB, Chua W, et al. A randomized clinical trial of a
single dose of intravitreal triamcinolone acetonide for
neovascular age-related macular degeneration: one-
year results. Arch Ophthalmol 2003;121(5):667 – 73.
[47] Gillies MC, Simpson JM, Billson FA, Luo W, Penfold
P, Chua W, et al. Safety of an intravitreal injection of
triamcinolone: results from a randomized clinical trial.
Arch Ophthalmol 2004;122(3):336 – 40.
[48] Young S, Larkin G, Branley M, Lightman S. Safety
590 A. Castellarin, D.J. Pieramici / Ophthalmol Clin N Am 17 (2004) 583 – 590
and efficacy of intravitreal triamcinolone for cystoid
macular oedema in uveitis. Clin Exp Ophthalmol 2001;
29(1):2 – 6.
[49] Jonas JB, Kreissig I, Degenring R. Intraocular pressure
after intravitreal injection of triamcinolone acetonide.
Br J Ophthalmol 2003;87(1):24 – 7.
[50] Young S, Larkin G, Branley M, Lightman S. Safety
and efficacy of intravitreal triamcinolone for cystoid
macular oedema in uveitis. Clin Exp Ophthalmol 2001;
29(1):2 – 6.
[51] Moshfeghi DM, Kaiser PK, Scott IU, Sears JE, Benz
M, Sinesterra JP, et al. Acute endophthalmitis follow-
ing intravitreal triamcinolone acetonide injection.
Am J Ophthalmol 2003;136(5):791 – 6.
[52] Nelson ML, Tennant MT, Sivalingam A, Regillo CD,
Belmont JB, Martidis A. Infectious and presumed
noninfectious endophthalmitis after intravitreal triam-
cinolone acetonide injection. Retina 2003;23(5):
686 – 91.
[53] Stern WH, Tamura E, Jacobs RA, Pons VG, Stone RD,
O’Day DM, et al. Epidemic postsurgical Candida
parapsilosis endophthalmitis: clinical findings and
management of 15 consecutive cases. Ophthalmology
1985;92(12):1701 – 9.
[54] Foster JBT, Almeda E, Littman ML, et al. Some intra-
ocular and conjunctival effects of amphotericin B in
man and the rabbit. Arch Ophthalmol 1958;60:555 – 64.
[55] Souri EN, Green WR. Intravitreal amphotericin B
toxicity. Am J Ophthalmol 1974;78(1):77 – 81.
[56] Pflugfelder SC, Flynn HW, Zwicky TA, Forster RK,
Tsiligianni A, Culbertson WW, et al. Exogenous fungal
endophthalmitis. Ophthalmology 1988;95:19 – 30.
[57] Foster CS, Lass JH, Moran-Wallace K, Giovanoni R.
Ocular toxicity of topical antifungal agents. Arch
Ophthalmol 1981;99(6):1081 – 4.
[58] Heinemann MH. Long-term intravitreal ganciclovir
therapy for cytomegalovirus retinopathy. Arch Oph-
thalmol 1989;107(12):1767 – 72.
[59] Eyetech Study Group. Anti-vascular endothelial growth
factor therapy for subfoveal choroidal neovasculariza-
tion secondary to age-related macular degeneration:
phase II study results. Ophthalmology 2003;110(5):
979 – 86.
[60] Kattan H, Pflugfelder SC. Complications of intraocu-
lar antimicrobial agents. Int Ophthalmol Clin 1989;
29(3):188 – 94.