CASE PRESENTATION
A 58-year-old Moroccan female of Arab and Berber
descent was referred to the Yale Eye Center in New
Haven, Connecticut, for uncontrolled, increasing IOP in
her right eye while on maximum medical therapy.
Approximately 3 years before her visit, the patient had
experienced mild pain and lost vision in her left eye within
a period of several weeks. At the time, she was living in
Morocco and did not have access to good medical care. She
was diagnosed with breast cancer approximately 6 months
before her visit. The need for improved cancer treatment
was the primary reason her son had temporarily relocated
her to the United States. As treatment, she received
carboplatin-paclitaxel and bevacizumab to be supplemented
by radiation therapy in the future.
In the months leading up to her visit, her IOP had been around 30 mm Hg, and it was 45 mm Hg for 4 weeks immediately before she presented to the center. Because the patient spoke no English, her son served as her translator. She had seen three general ophthalmologists in the United States in the past 3 months who had administered medical glaucoma therapy without success. She had noticed decreasing vision in her only seeing, right eye that seemed to be accelerating. She had also experienced pain in her right eye for which she was prescribed Combigan (timolol-brimonidine; Allergan, Inc., Irvine, CA) and pilocarpine by an outside physician. The patient felt nauseated from taking 500 mg of Diamox (Wyeth Pharmaceuticals, Philadelphia, PA) twice a day for about 4 weeks. She felt physically exhausted because of the many drops she was instructed to use. At the time of her visit, she was taking Cosopt (timolol-dorzolamide; Merck & Co., Inc., Whitehouse Station, NJ) twice a day, Xalatan (latanoprost; Pfizer Inc., New York, NY) at bedtime, Combigan twice a day, and pilocarpine 2% four times a day in both eyes. It was obvious from a well-written instruction sheet given to her by the physician who treated her that she had not simply accumulated samples. When we asked if laser peripheral iridotomy for angle closure had been recommended, she responded that she had not heard of it.
Besides the diagnosis of breast cancer, the review of systems, past medical history, systemic medications, and family and social history were not contributory. The examination revealed that finger-counting confrontational visual field tests were severely constricted. Both pupils were not reactive to light. The UCVA in her right eye was 20/60 without improvement using a pinhole and no light perception in her left eye.
The patient's IOP by applanation was 45 mm Hg OD and 48 mm Hg OS. The central corneal thickness was 620 μm OD and 580 μm OS, and axial length measured 22.19 mm OD and 22.15 mm OS. She had deep-set orbits and small lid fissures. Slit-lamp examination showed trace corneal edema in both eyes. The van Herrick test revealed a very shallow anterior chamber with obvious grade 0 peripheral closure that could not be opened with compression gonioscopy. The patient also had a middilated, nonreactive pupil, a congested iris without any signs of attempted peripheral laser iridotomy, and mild cataracts in both eyes (nuclear sclerosis grade 3, cortical cataract grade 1, posterior capsular cataract grade 1 on LOCS III). The right optic nerve measured 1.5 mm in vertical diameter with a deep, nearly absolute cup, whereas the vessels, macula, and periphery were unremarkable. The examination of the left eye reflected that of the right.
We diagnosed end-stage, angle-closure glaucoma with an urgent need for surgical control of the pressure in the patient's right eye and absolute angle-closure glaucoma in the left eye.
HOW WOULD YOU PROCEED?
1. Would you attempt a laser peripheral iridotomy?
2. Given the advanced stage of her glaucoma, how would you intervene?
3. How does the stage and treatment of the patient's cancer influence your assessment of the pathogenesis, outlook, and management of the angle closure?
SURGICAL COURSE
We performed a trabeculectomy with mitomycin C
and phacoemulsification with IOL implantation 3 days
after the initial examination. To improve the surgeon's
control of ocular motility and pain, a retrobulbar block
was administered using 2% lidocaine and 0.75% bupivacaine
with a volume of 2.5 mL. It was delivered with a
retrobulbar Atkinson needle (BD Visitec, Franklin Lakes, NJ).
The patient was very fearful to undergo surgery on her
only seeing eye. She was agitated and cooperated poorly
despite the presence of a translator.
We used a novel small-incision trabeculectomy technique developed by Robert Weinreb, MD. The technique does not require traction sutures or bleb compression and does not usually require an iridectomy.1 The surgeon created a small 3-mm limbal conjunctival incision at the 12-o'clock position. After opening the sub-Tenon's space with a Blumenthal Conjunctival Dissector (Katena Products, Inc., Denville, NJ), the surgeon applied mitomycin C-soaked (0.4 mg/mL) sponges for 3 minutes. He fashioned a shallow, triangular scleral flap, and advanced the dissection with a bent crescent knife barely into the clear cornea. The surgeon then repositioned temporally to perform the cataract extraction. Phacoemulsification and the IOL's insertion were without complications but they were challenging due to an extremely shallow anterior chamber and corneal edema. The surgeon then repositioned to the superior position to complete the trabeculectomy. Although repositioning adds to the case time, the surgeon felt that initial fashioning of the scleral flap in a well-pressurized eye allowed for the best control of the flap in this difficult setting. After penetration of the anterior chamber underneath the scleral flap, the surgeon squared off the incision with a sideport knife. He performed three punches with a Kelly Punch (ASICO, LLC, Westmont, IL). An iridectomy was not necessary. The scleral flap was tied down with three 10–0 nylon sutures, the knots were buried, and the conjunctiva was closed with two 10–0 nylon wing sutures.
OUTCOME
The postsurgical course was initially unremarkable
with a quiet, diffuse bleb (Figure 1). By 1 day postoperatively,
the IOP was 28 mm Hg and adjusted by laser
suture lysis to 20 mm Hg through postoperative day 7,
10 mm Hg by postoperative day 14, and 5 mm Hg by 21 days
postoperatively. The patient received two injections of
5-fluorouracil (volume 0.15 mL, concentration 50 mg/mL)
throughout the postoperative management. Her UCVA
remained unchanged at 20/60 for 1 week. Thirteen days
after surgery, the patient noticed that her vision had
become very blurry, which coincided with a cycle of
chemotherapy. Upon examination on the next day, her
visual acuity had dropped to counting fingers at 5 feet,
and the IOP measured 10 mm Hg. The posterior pole
showed a reversal of cupping, normally perfused vessels,
and a normal, flat, dry macula. Her visual acuity decreased to hand motions 1 week later on postoperative
day 21, but the remainder of the examination,
including the refraction, was unremarkable. The IOP
was 5 mm Hg.
Imaging on postoperative day 21 with the Stratus OCT (Carl Zeiss Meditec, Inc., Dublin, CA) showed a complete loss of the retinal nerve fiber layer in the patient's right eye, with the exception of a severely thinned papillomacular bundle and some superior and nasal fibers (Figure 2). Her left eye had no retinal nerve fiber layer, with the exception of the nasal rim. Optical coherence tomography also demonstrated a dramatic change in the optic disc architecture of the right eye, with a profound reversal of cupping (Figure 3) that had been very similar preoperatively to the nearly absolute cup in the left eye. The patient was completely dependent on a caretaker and devastated by the visual loss, as she felt her fears about the results of surgery had materialized. At her 3-month visit, she was in unusually good spirits and reported a dramatic improvement in vision that had started 2 weeks prior to her visit. Remarkably, her UCVA had improved to 20/100 that was not better with pinhole. The remainder of her examination was unchanged, with an IOP of 6 mm Hg and the previously seen reversed cupping of the optic nerve.
DISCUSSION
Severe visual loss after uncomplicated trabeculectomy
in advanced glaucoma is a rare but feared outcome that
occurs in up to 6% of patients.2 The various names for
this complication—idiopathic vision loss, snuff out, and
wipe out—imply that no cause can be identified. A careful
examination often reveals that factors such as cystic
macular edema,3 hypotony maculopathy,4 corneal
decompensation,5 or a high refractive error6 may be
responsible.
In this case, none of these findings was present. The possibility of optic nerve injury from retrobulbar block is unlikely, as the patient's vision remained good for 2 weeks after surgery. Rather, two mechanisms may have contributed to her visual loss. First, sustained high preoperative IOP may have triggered apoptosis that took several weeks to fully manifest. Second, a profound change in the optic nerve head's architecture with an anterior shift of the lamina cribrosa may have damaged the remaining axons and thus presented a substrate of true snuff out solely caused by a drop in IOP. In that regard, it is noteworthy that we gradually reduced the IOP with a target of 10 mm Hg. This goal was overshot to 5 mm Hg but without any pathology from hypotony.
None of the chemotherapeutics or medications that this patient was on were implicated in acute optic neuropathy. To the contrary, her cancer therapy likely helped to control the postoperative fibrosis of the bleb and reactive gliosis that occurs after optic nerve damage.7
As unsettling as the severe, temporary drop in visual acuity is the fact that apparently two of three general ophthalmologists did not perform gonioscopy, which would have revealed the complete angle closure. Instead, the patient was treated for several months by combining topical and systemic medications with antagonizing or identical mechanisms.
Nils A. Loewen, MD, PhD, is an assistant professor in the Department of Ophthalmology and Visual Science at the Yale University School of Medicine in New Haven, Connecticut. He acknowledged no financial interest in the products or companies mentioned herein. Dr. Loewen may be reached at (203) 737-5746; nils.loewen@yale.edu.
