Maylon Hsu, MD, Michael Nolan, Amy Lin, MD Loyola University Medical Center Department of Ophthalmology FA CTORS AFFECTING THE RATE OF DISSOLUTION OF TEMPORARY SUTURELESS AMNIOTIC MEMBRANE APPLIED TO THE OCULAR SURFACE The authors have no financial interest in the subject matter of this e-poster.

INTRODUCTION Amniotic membrane (AM) transplantation to the ocular surface has been performed for a variety of conditions, including ocular burns, persistent epithelial defects, acute Stevens- Johnson syndrome, limbal stem cell deficiency, keratitis, and surgical defects after conjunctival excision. 1-6 AM is a biologic tissue which is secured to the ocular surface to not only fill gaps in corneal and conjunctival epithelium, but also to promote epithelial healing with proposed anti-inflammatory, anti-fibrotic, anti-angiogenic, and antimicrobial properties. 1 ProKera (Bio-Tissue Inc., Miami, FL) is a device which allows for temporary sutureless application of cryopreserved AM. The device consists of AM clipped to a polycarbonate ring, which is easily inserted and removed from the eye. When the device is placed on the ocular surface, the ring rests on the bulbar conjunctiva, encircling the cornea. The device is well- tolerated with minimal side effects. 4 AM has been noted to dissolve naturally over time. To date, there have not been any studies describing the rate of AM dissolution. It is hypothesized that higher degrees of ocular inflammation may lead to faster dissolution. This study investigates the factors which affect the rate of AM dissolution in ProKera devices.

METHODS Retrospective chart review of all patients who had insertion of a 15 or 16 mm ProKera device from June 2005 – February Non-healing corneal epithelial defect was the primary indication for ProKera. The etiology of corneal epithelial defects and exam findings at the time of placement were recorded including: acute toxic epidermal necrolysis syndrome (TENS), chronic TENS, exposure keratopathy, neurotrophic ulcer, and infectious ulcer. Medications used during treatment with ProKera and the clinical outcomes were recorded. The median AM dissolution rates for these groups were compared by Kruskal-Wallis and Dunn’s tests.

RESULTS This study involved 29 ProKera membranes used in 13 patients. 20 membranes remained in place until dissolution, and 9 devices were removed while the membranes were still intact. Of the different disease categories, devices placed in eyes with epithelial defects secondary to corneal exposure dissolved the quickest (4 days), while devices in eyes with neurotrophic ulcers (19.5 days) and chronic TENS (10 days) lasted the longest. The median number of days until dissolution was significantly longer with versus without concurrent topical steroid treatments (10 vs. 5 days respectively, p = ). There was no statistical correlation between degree of conjunctival inflammation and rate of dissolution.

TABLE 1: CLINICAL CHARACTERISTICS IDAgeEyeReason for ProKeraDissolution time (days)Outcome 1.111ODAcute TENS, epithelial defect4Dissolved 1.211OSAcute TENS, epithelial defect4Dissolved 2.169ODneurotrophic ulcer27Removed 3.17ODChronic TENS10removed partially dissolved 3.27OSChronic TENS10removed partially dissolved 3.37ODAcute TENS7Dissolved 3.47OSChronic TENSStill intactRemoved – replaced day OSChronic TENS15Dissolved 3.67OSAcute TENS8Dissolved 3.77ODcorneal ulcer8Dissolved 4.18ODAcute TENS, epithelial defect7Dissolved 5.152OSAlkaline burn - epithelial defectStill intactfell out day ODAcute TENS14Dissolved 6.279OSAcute TENS17Dissolved 6.379ODAcute TENSStill intactdeceased day OSAcute TENSStill intactdeceased day ODEpi defect 2/2 medial canthus fistula6Dissolved 7.278ODEpi defect 2/2 medial canthus fistula3Dissolved 7.378ODEpi defect 2/2 medial canthus fistula4Dissolved 7.478ODEpi defect 2/2 medial canthus fistulaStill intactremoved – day ODperiorbital rad, neurotrophic ulcer12Dissolved 9.123ODExposure Keratopathy 2/2 thermal burn eyelid5Dissolved 9.223OSExposure Keratopathy 2/2 thermal burn eyelid2Dissolved 9.323OSExposure Keratopathy 2/2 thermal burn eyelid5Dissolved OSAcute TENSStill intactremoved 3 – for AM transplantation OSK ulcer, h/o SJSStill intactremoved 4 – healed ODpersistent epithelial defect s/p vitrectomyStill intactfell out same day ODpersistent epithelial defect s/p vitrectomyStill intactremoved 7 – healed 1349OSneurotrophic ulcer from Bell’s palsy3Removed-for AM transplantation

TABLE 2. PROKERA DISSOLUTION BY DISEASE CATEGORY Disease CategoryTotal Prokera (n)Prokera w/ Dissolution (n) Median days until dissolution Acute TENS/ epithelial defect Chronic TENS Exposure epithelial defect874.0 Neurotrophic ulcer Infectious Ulcer218

TABLE 3. DURATION OF AMNIOTIC MEMBRANE WITH ASSOCIATED FACTORS Duration of Prokera Total eyes: LagophthalmosConjunctival Injection Corneal Neovascularization Topical Steroid Systemic Steroid 1-6 days days > 14 days404132

TABLE 4. TOPICAL STEROID USE AND DURATION OF AMNIOTC MEMBRANE Topical SteroidNumber of EyesMedian Duration (days) P Value* measures by Dunn’s Test No115 Yes

Photographs of a ProKera device in an eye with a diabetic non-healing neurotrophic ulcer (A), also taken under cobalt blue light (B). The device was removed after 27 days showing a small amount of initial central breakdown of the AM (C), with a resolved epithelial defect (C). Figure 1. Photographs of Slow Dissolving ProKera Device

Photographs of a ProKera device in an eye with a neurotrophic ulcer with 11 mm of lagopthalmos due to Bell’s palsy from an invasive parotid gland adenocarcinoma (A). After three days of ProKera insertion, there was 70% AM dissolution (B). Temporal and medial tarsorrhaphies (medial one was removed for photo) were done during the time of ProKera insertion. Despite prior upper lid gold weight placement and the tarsorrhaphies, there was residual 1-2 mm of central lagophthalmos. After 3 days of ProKera, the cornea had begun to re-epithelialize, with a persistent central corneal epithelial defect (C). Figure 2. Photographs of Fast Dissolving ProKera Device

CONCLUSIONS This study provides the first analysis of factors that influence the rate of dissolution of ProKera AM. Patients were divided into 6 different disease categories. Patients with acute TENS/ corneal epithelial defect were the largest group in our study. AM dissolved the most quickly in eyes where exposure was the primary reason for epithelial defects. This finding suggests that desiccation may lead to more rapid break down of the membranes. AM in eyes receiving topical steroids had a significantly longer duration than those that did not. Topical steroids may aid in limiting inflammatory processes that lead to amniotic membrane dissolution. Our study is limited by the small number of subjects, many who had several ProKera insertions. The study failed to correlate increased conjunctival inflammation to faster dissolution times. More studies are needed to further characterize this association.

REFERENCES 1. Dua HS et al. The amniotic membrane in ophthalmology. Surv Ophthalmol 2004;49: Gregory D. The ophthalmologic management of acute Stevens-Johnson syndrome. The Ocular Surface 2008;6: Kheirkhah A et al. Temporary sutureless amniotic membrane patch for acute alkaline burns. Arch Ophthalmol 2008;126: Pachigolla G et al. Evaluation of the role of ProKera in the management of ocular surface and orbital disorders. Eye & Contact Lens 2009;4: Shammas MC et al. Management of acute Stevens-Johnson syndrome and toxic epidermal necrolysis utilizing amniotic membrane and topical corticosteroids. Am J Ophthalmol 2010;149: Sheha H et al. Sutureless amniotic membrane transplantation for severe bacterial keratitis. Cornea 2009;28: