Epithelial downgrowth into the anterior chamber is an extremely rare
complication of intraocular surgery. Epithelial cells gain access via incisional
defects or by direct implantation at the time of surgery. The tissue can grow
in the form of cysts or as layered sheets. If these sheets are left untreated,
devastating complications such as intractable glaucoma and retinal detachment
result in destruction of the eye.
Historically, the preferred treatment methods are surgical.1 Weiner et al1 suggested
that combining antimetabolites with surgical treatments might be superior
to surgery alone in controlling the disease. There are reported cases of the
use of multiple subconjunctival injections of substantial amounts of 5-fluorouracil
to control the disease.2,3 However,
these failed once the injections were stopped.2,3
A 70-year-old, aphakic white woman underwent penetrating keratoplasty
(PKP) for interstitial keratitis in 1994 in the right eye and 2 years later
in the left eye. In 1997, slitlamp examination of the left eye showed a retrocorneal
membrane without a clear site of origin. Argon laser photocoagulation confirmed
the presence of epithelial downgrowth, and ultrasound biomicroscopy showed
anterior synechiae. Repeat PKP was performed in combination with anterior
vitrectomy, release of anterior synechiae, and cryopexy. Histopathological
examination of the host corneal button showed stratified epithelial tissue
extending along the posterior surface of the cornea. Epithelial downgrowth
recurred in 1998, and a repeat PKP combined with anterior vitrectomy and cryopexy
were done but failed to prevent recurrence. In 1999, cyclocryotherapy and
repeat PKP were performed. Histopathological examination of each of the corneal
buttons showed the typical extensive stratified squamous epithelium along
the posterior corneal surface. The left eye was ultimately lost to epithelial
downgrowth and secondary retinal detachment.
In 1999, the patient experienced graft rejection in the right eye and
a repeat PKP was done. Postoperatively, vitreous incarceration in the superior
aspect of the graft was managed with pars plana vitrectomy. Thereafter, the
intraocular pressure remained elevated despite maximum medical therapy and
was successfully controlled with implantation of an Ahmed valve in October
2000. The patient returned 1 month later with complaints of blurry vision.
Slitlamp examination showed a retrocorneal membrane. The membrane originated
temporally and was thought to be from the site of the paracentesis done at
the time of the seton surgery. Internal cryopexy was performed but failed
to halt extension of the membrane (Figure
1). In December 2000, one anterior chamber injection of a low dose
(0.2 mg) of 5-fluorouracil,4 with a sodium
hyaluronate (Healon-GV; Pharmacia Canada Inc, Mississauga, Ontario) pupillary
plug, was performed in an attempt to retain the 5-fluorouracil in the anterior
chamber and minimize its diffusion into the vitreous. This failed to halt
the extension of the membrane. The injection was repeated with a higher dose
of 5-fluorouracil (1 mg) mixed with chondroitin sulfate–sodium hyaluronate
(Viscoat; Alcon Laboratories, Fort Worth, Tex) after a temporal paracentesis.
The mixture was applied directly to the epithelial membrane, which was viscodissected
from the posterior corneal surface (Figure
2). Viscodissection was achieved by lifting the edge of the membrane
with the cannula and using the viscoelastic–5-fluorouracil mixture to
separate the epithelial membrane free of the posterior corner. The chondroitin
sulfate–sodium hyaluronate was left in the eye, and there were no problems
with elevated intraocular pressure after this treatment. Five months after
this treatment, there was no recurrence of the membrane, but a repeat PKP
was done in May 2001 for graft failure. Histopathological examination of the
host corneal button did not show epithelial downgrowth but showed remnants
of a fibrous network. There was no recurrence after this second treatment
with 5-fluorouracil during 14 months of follow-up.
To our knowledge, there are no previous reports of noncongenital bilateral
epithelial downgrowth and no reports of treatment using anterior chamber injections
of 5-fluorouracil. In the present case, surgical treatment of epithelial downgrowth
in the left eye failed to control the disease, resulting in loss of that eye.
However, in the right eye, 2 anterior chamber injections of a total of 1.2
mg of 5-fluorouracil, the latter mixed with a visoelastic and combined with
viscodissection of the membrane itself, were successful in halting the disease
process and preventing recurrence 14 months after the treatment. The advantages
of intraocular injections include direct delivery of 5-fluorouracil to the
actively proliferating membrane. This allowed use of a substantially smaller
effective dose of 5-fluorouracil, compared with subconjunctival injections,
and potentially decreased the risk of toxic side effects to the cornea.2,3 Finally, the small number of
injections was easily performed and well tolerated. Intraocular injections
of 5- fluorouracil may be a viable treat ment option for epithelial downgrowth
in selected patients.
Corresponding author: Peter J. Kertes, MD, FRCSC, The University of
Ottawa Eye Institute, 501 Smyth Rd, Ottawa, Ontario, Canada K1H 8L6 (e-mail: pkertes@ottawahospital.on.ca).
1.Weiner
MJTrentacoste
JPon
DMAlbert
DM Epithelial downgrowth: a 30-year clinicopathological review.
Br J Ophthalmol. 1989;736- 11
Google ScholarCrossref 2.Loane
MEWeinreb
RN Glaucoma secondary to epithelial downgrowth and 5-fluorouracil.
Ophthalmic Surg. 1990;21704- 706
Google Scholar 3.Weinreb
RN Adjusting the dose of 5-fluorouracil after filtration surgery to minimize
side effects.
Ophthalmology. 1987;94564- 570
Google ScholarCrossref 4.Blumenkranz
MHernandez
EOphir
ANorton
EW 5-Fluorouracil: new applications in complicated retinal detachment
for an established antimetabolite.
Ophthalmology. 1984;91122- 130
Google ScholarCrossref