Finger's Amniotic Membrane Buffer Technique: Protecting the Cornea During Radiation Plaque Therapy

Objective  To use amniotic membranes as a buffer between the cornea and radioactive eye plaques.

Methods  Six melanomas were treated with ophthalmic plaque radiation therapy. Plaque-tumor localization required that a portion of the gold plaque touch the cornea during treatment. To enhance patient comfort and protect the cornea, an (0.1-mm-thick) amniotic membrane was interposed between the metal plaque edge and the cornea.

Results  Minimal ocular discomfort was noted during plaque radiation therapy. On a scale of 1 (none) to 10 (severe), all 6 patients reported pain levels of 1. As a tissue equivalent and because the mean thickness was only 0.1 mm, amniotic membranes had no significant effect on radiation dose calculations. No adverse effects, infections, or abrasions were noted.

Conclusion  The amniotic membrane buffer technique improves patient comfort and protects the cornea during ophthalmic plaque radiation therapy.

Radioactive plaque therapy has been used for the treatment of iris, iridociliary, and ciliary body malignant neoplasms.^1-4 In this therapy, the surgeon must affix the plaque to cover the tumor's base plus a free margin of normal-appearing tissue; tumors that extend to the iris root or are in the iris require that the metal plaque edge rest on the cornea. In these cases, postoperative findings of corneal epithelial ridges and abrasions are common transient findings. Plaque-cornea touch (during treatment) is also uncomfortable and occasionally painful.

Amniotic membrane patching was introduced in ophthalmology to treat extreme chemical burns of the conjunctiva and cornea, to manage persistent corneal epithelial defects, and for reconstruction of the conjunctival fornices.^5-11 The membranes are available as freeze-dried or frozen rectangles of tissue that have been placed, sewn, and glued on recipient sites. In this study, an amniotic membrane graft (AMG) was used to act as a buffer between the gold rim and the cornea during ophthalmic plaque radiation therapy.

This study conforms to the tenets of the Declaration of Helsinki and the Health Insurance Privacy and Accountability Act of 1996. Patients signed consent forms for the investigational use of AMGs, and approval was obtained from The New York Eye Cancer Center's Internal Review Board, New York.

Six patients with biopsy-proved anterior uveal melanoma were screened for metastatic disease and found suitable for ophthalmic plaque radiation therapy (Table 1). Each tumor involved the iris, the ciliary body, or both (iridociliary). Coverage of the entire tumor and a 2-mm tumorfree margin required that the gold plaque override and thereby contact the cornea.

Tumor localization and episcleral plaque insertion involved measuring the anterior margins as visible through the cornea, found to extend along the anterior chamber angle on gonioscopy and within the ciliary body (as revealed by high-frequency ultrasonography).¹² The posterior tumor margins were also visible on transillumination or obscured by the ciliary body band. In such cases, the posterior margin of the ciliary body band was used as the posterior tumor margin. In all cases, the plaque was placed to cover the entire tumor plus a 2- to 3-mm tumor-free margin. This required that the plaques be placed on the corneal surface.

A 360° conjunctival peritomy was performed around the corneal scleral limbus. A curved Stevens scissors was used to open Tenon's capsule in all 4 quadrants. Depending on the tumor's shape, either a custom-made anterior segment plaque or a standard round plaque was sewn to the eye to cover the episcleral and epicorneal markings.

In this series, we used six 1 × 1.5-cm frozen AMGs (Bio-Tissue, Miami, Florida). The tissue used has been reported to be a mean of 0.1 mm thick.¹³ Each AMG was placed sticky-side up on the cornea (Figure 1). This approach typically required some gentle teasing of the AMG to become flattened on the corneal surface. While the assistant gently raises the posterior edge of the plaque, the surgeon slides the AMG beneath the gold plaque (Figure 1). The AMG should cover the affected cornea but not be folded beneath the plaque. Folding of the graft (beneath the plaque) should be avoided because it might cause significant plaque displacement (away from the tumor). Extra AMG can be left on the unaffected cornea (Figure 2). Last, the posterior aspect of the plaque was covered by the patient's conjunctiva to shield it from eyelid trauma. Once irradiation was completed, the conjunctiva was opened and the plaque and AMG were removed.

Preoperative comparative dosimetry was performed on all patients before plaque insertion. Because of the availability of 2 low-energy radionuclides in our center, palladium Pd 103 (¹⁰³Pd) was compared with iodine I 125 sources in gold plaques. As a result of these preoperative comparative dosimetry comparisons, all patients were treated with ¹⁰³Pd (Table 2).

The choice of ¹⁰³Pd marginally increased the mean radiation dose to the subjacent sclera (4%) and lens (2%) and decreased the mean radiation dose to the patient's macula (foveal dose) by 48%. These numbers are important for several reasons. The scleral dose comparison revealed a slight (questionably significant) increase in irradiation of the tumor. The lens doses were both cataractogenic and relatively equivalent. In contrast, the macula was the farthest point measured from the plaque and most closely represents the total organ dose (that received by the eye as a whole). In this study, the eyes received up to 48% less radiation because of the use of ¹⁰³Pd.^14,15

The mean 0.1-mm thickness of the AMG and its potential effect on the tumor dose were not considered significant enough to include in our calculations. Like other discounted variables, AMG thickness is similar to the standard error of ultrasonography tumor measurements (0.1 mm) and the thickness of the retained conjunctiva at the limbus (also typically disregarded during dosimetric calculations). In addition, this thickness is at least partially compensated for by the 4% increase in base dose provided by the use of ¹⁰³Pd (instead of iodine 125). In this study, patients were prescribed a mean dose of 83 Gy to the deepest measured point the tumor extended into the eye (Table 2).

All patients received one 7-day plaque radiation course that started at insertion and continued until the prescribed dose was delivered to the point of deepest intraocular tumor extension (as measured by high-frequency ultrasonography after dilation). Continuous postoperative mydriasis and cycloplegia were prescribed to immobilize the tumor, minimize its size, and reduce the need for anterior plaque extension during radiation therapy. Postoperative topical antibiotic steroid eyedrops were also placed on the eye 4 times daily.

Amniotic membranes were placed between the cornea and epicorneal radioactive plaques in 6 patients (Figure 2). By the American Joint Committee on Cancer–International Union Against Cancer Criteria (AJCC-UICC) staging system, there was 1 T1a iris tumor (fewer than 3 clock hours), 4 T2 iridociliary tumors (confluent with or extending into the ciliary body and/or choroid), 1 T1c ciliary body tumor, and 1 T2 ciliary body tumor (>10 mm in largest basal dimension).¹⁶ Each was in contact with or extended anteriorly to the corneal scleral limbus, thus requiring anterior plaque placement (Table 1).

On the day after plaque insertion until the day before explantation surgery, patients were asked if they were experiencing eye pain. On a subjective scale from 1 (none) to 10 (severe), all 6 patients noted a pain level of 1 (Table 2). During that same interval, no patients required narcotic pain medication.

In no case did the amniotic membrane graft require an interruption of treatment. No infections, allergic reactions, or corneal abrasions occurred (during or after plaque radiation therapy). Amniotic membranes were totally and easily removed at the time of plaque explantation. No perioperative complications could be attributed to the use of the amniotic membrane buffer technique.

Iridectomy or iridocyclectomy had been the procedures of choice for select iris and ciliary body melanomas.^17-19 However, ophthalmic plaque radiation therapy offers the advantages of larger treatment margins and retained iris function.² In contrast to intraocular surgery, extraocular plaque radiation therapy carries little risk of hyphema, endophthalmitis, or retinal detachment.^2,4 Iris retention reduces postoperative symptoms of glare. Although the incidence of secondary radiation cataract is high, little maculopathy has been noted.¹⁴ In 2001, I published an article² about the first use of ophthalmic plaque radiation therapy for resectable iris and iridociliary melanoma. Now widely used, plaque radiation therapy has been found to be both safe and effective for most anterior uveal tumors.^4,12,20 However, many patients are uncomfortable and some experience significant pain when the plaque's edge rests on the cornea. Acute corneal findings can include temporary epithelial ridges and abrasions. No corneal infections, ulcerations, dystrophies, or opacities have been reported.^2-4,12

E. Rand Simpson, MD (oral communication, August 2007), and others have described methods to enhance patient comfort during anterior plaque radiation therapy. These methods include full conjunctival cover techniques (as described in the “Methods” section), superior rectus muscle disengagement (for inferior plaque positions), and customized nonpressure dressings. Although these techniques address exposure of the anterior surface of the plaque (between the eyelids), my amniotic membrane buffer technique creates a buffer between the posterior surface of the plaque and the cornea. No short-term complications were noted, all patients were comfortable throughout treatment.

Correspondence: Paul T. Finger, MD, The New York Eye Cancer Center, 115 E 61st St, New York, NY 10065 (pfinger@eyecancer.com).

Submitted for Publication: September 4, 2007; final revision received September 17, 2007; accepted September 17, 2007.

Financial Disclosure: None reported.

Funding/Support: This study was supported by The EyeCare Foundation Inc, New York, NY (http://eyecarefoundation.org).