Anophthalmic Sockets in Retinoblastoma: A Single Center Experience

Transcription

1 ORIGINAL CLINICAL STUDY Anophthalmic Sockets in Retinoblastoma: A Single Center Experience Preethi Jeyabal, MBBS, and Gangadhara Sundar, DO, FRCSEd, FAMS, AB(USA) Purpose: To evaluate outcomes of anophthalmic sockets in retinoblastoma at a tertiary care center in Singapore. Design: A retrospective study. Methods: Patients who underwent enucleation as sole/part of treatment for retinoblastoma were reviewed at our center from Details including demographics, grouping and staging, adjuvant therapy, surgery, implant, and complications were collected. Results: Of 42 patients with retinoblastoma managed over the period, the anophthalmic sockets of 31 patients who underwent enucleation were analyzed. Mean age at enucleation was 2 years. Twenty-three enucleations were performed at our institution and 8 enucleations had been performed elsewhere. Seventeen patients (52%) had porous polypropylene, 9 patients (27%) had polymethylmethacrylate, 1 patient (3%) had glass implant, and 3 (9%) had dermis fat graft. The sizes of implants varied from 10 to 20 mm. Twelve patients had attempts at globe salvation before enucleation. Out of 28 patients with primary orbital implants, 3 had implant exposure. The rates of repeat surgery among patients with and without primary implant were 66.67% (2 out of 3) and 10.7% (3 out of 28), respectively. One patient had postenucleation socket syndrome with stock eye. Five patients referred for enucleation were conservatively managed. Conclusions: Anophthalmic sockets in retinoblastoma have long-term implications if the primary procedure is not performed well. While the majority had good outcomes (structural and esthetic), a minority had complications requiring intervention. Ophthalmologists managing retinoblastoma must be aware of these. Primary implant had favorable outcome with minimal complications. Key Words: anophthalmia, retinoblastoma, enucleation, anophthalmic socket, socket management (Asia-Pac J Ophthalmol 2018;7: ) The most important steps in managing patients with retinoblastoma is complete treatment of the primary tumor(s), avoiding extraocular spread, preventing complications, and when necessary, managing them. Among the various treatment modalities available for retinoblastoma, despite numerous advances, ocular enucleation is still the treatment of choice for advanced retinoblastoma for otherwise unsalvageable eyes. There are various From the Department of Ophthalmology, National University of Singapore, Singapore. Received for publication March 15, 2018; accepted June 12, The authors have no funding or conflicts of interest to declare. Presented at the 33rd Asia-Pacific Academy of Ophthalmology Congress; February 7 11, 2018; Hong Kong. Reprints: Gangadhara Sundar, DO, FRCSEd, FAMS, AB(USA), National University Hospital, 1E Kent Ridge Road, National University Health System Tower Block, Department of Ophthalmology, Level 7, Singapore E mail: Gangadhara_SUNDAR@nuhs.edu.sg. Copyright 2018 by Asia Pacific Academy of Ophthalmology ISSN: DOI: /APO techniques and modifications of anophthalmic socket reconstruction with a wide variety of orbital implants (ie, materials, size and shape, including dermis fat grafts), various wrapping materials, and varied techniques of their placement within the orbit, each with its own advantages and disadvantages. Once the socket has been reconstructed, a customized ocular prosthesis becomes the next essential component of both socket and total patient rehabilitation. The most common challenges faced in anophthalmic sockets include dealing with complications such as discharge, implant exposure, migration, extrusion, socket contracture, vigilantly watching for recurrence of tumor, and finally poor cosmesis. It is therefore essential for all ophthalmologists to understand the management of anophthalmic patients to prevent, recognize and manage complications. We herewith describe a 12-year experience in the assessment, management and follow up of anophthalmic sockets in retinoblastoma at a tertiary care centre in Southeast Asia including complications encountered. MATERIALS AND METHODS This was a retrospective study of all patients who underwent enucleation as sole or part of management of retinoblastoma who were primarily or secondarily managed at our center over a 12-year period ( ). Details of the patients including their demographics, grouping and staging of retinoblastoma, type of adjuvant/neoadjuvant therapy received, type and technique of surgery, type of implant, wrapping materials, and complications were collected and statistical analyses were performed. RESULTS Of 42 patients with retinoblastoma managed over a 12-year period whose complete records were available, the anophthalmic sockets of 31 patients who underwent enucleation were analyzed. Twenty patients (64.5%) had unilateral retinoblastoma and 11 patients (35.4%) had bilateral retinoblastoma. Nineteen patients underwent primary ocular enucleation and 12 patients underwent enucleation after attempts at globe preservation in the form of systemic chemotherapy with or without local treatment (n = 10) and intra-arterial chemotherapy (n = 2). All patients, including those with bilateral retinoblastoma, had unilateral enucleation only. The second eyes of all bilateral patients (n = 11) were salvaged, with visual preservation in all. Sixteen of 31 patients who underwent enucleation had Group E retinoblastoma, 12 patients had Group D, and 3 had Group C disease. Overall, the mean age at presentation was 1.7 years (range, 3 weeks 81 months). Average age at presentation for bilateral retinoblastoma was 1.5 years (range, 1 28 months) and for unilateral retinoblastoma was 1.8 years (range, 3 weeks 81 months). Mean age at enucleation was 2 years (range, 2 81 months). Average follow up after enucleation was 93 months (range, 0 26 years). All enucleations were performed under general anesthesia Asia-Pacific Journal of Ophthalmology Volume 7, Number 5, September/October

2 Jeyabal and Sundar Asia-Pacific Journal of Ophthalmology Volume 7, Number 5, September 2018 after a detailed examination under anesthesia, determined unsalvageable with or without globe conserving attempt. Enucleation was performed after intraoperative verification and confirmation of the need for the procedure after a modified peribulbar local anesthetic block [lidocaine with adrenaline and bupivacaine (50:50)] administered cautiously, to avoid accidental globe penetration. After completion of ocular enucleation with as long an optic nerve segment as possible, a thin slice from the distal cut end of the optic nerve was sent for intraoperative histopathological confirmation to ensure it was free of tumor infiltration. Twenty-three enucleations were performed at our center and 8 enucleations had been performed elsewhere with patients being referred for further management. A successful enucleation was one where there was complete and uncomplicated removal of the globe with primary socket rehabilitation and without early or late postoperative complication and a satisfactory cosmetic outcome. Orbital implants used were as follows. Seventeen patients (52%) had porous polypropylene (Medpor), 9 patients (27%) had a polymethylmethacrylate (PMMA) implant, 1 patient (3%) had a glass implant (Mules sphere), and 3 (9%) had dermis fat graft. Three patients (9%) did not have primary implant (1 was performed at our center and the other 2 performed elsewhere). Bovine pericardium was the most commonly used wrapping material in 12 patients followed by donor sclera in 6 patients. The following algorithm is used at our center to decide the nature of the implant. If postoperative pegging is not planned, when enucleation is performed less than 18 months of age and if enucleation is planned for an eye with suspected pathologically high risk characteristics, eg, Group E retinoblastoma, a nonporous (acrylic) implant is used, for easy monitoring of the socket clinically and radiologically. On the other hand, when a definitive implant is required, especially in an older child (more than 18 months of age) with potential to consider pegging of prosthesis at later date with no plan for implant exchange, a porous implant was considered. The length of the optic nerve stump in enucleated eyes ranged from 2 10 mm. Infiltration along the entire length of the optic nerve (cut end positive) was found on histopathology of one eye with group E retinoblastoma, discussed later. Two patients had focal choroidal invasion, and two patients had tumor presence upto but not beyond the lamina cribrosa, both without pathologic high-risk characteristics. No other high-risk characteristics like massive choroidal/scleral invasion/anterior segment invasion were present. The sizes of orbital implants placed varied from 10 mm to 20 mm (mean, mm). The shapes of implants included spheres (87%), cones (3%), and Quad motility implant (10%). Twelve out of 17 children (70.5%) who had enucleation performed beyond 18 months of age received a 20 mm implant. The average age of children who received implants smaller than 20 mm (18, 16, 12, and 10 mm) were 19 months, 21 months, 18 months, and 24 months respectively. One patient referred at a later date had undergone a 10 mm spherical implant placement with a dermis fat graft for a late contracted socket management. Patients who had unilateral disease had no further treatment after enucleation except for a customized ocular prosthesis fitting. For patients with bilateral disease, the contralateral eye required varying combinations of chemoreduction, laser thermotherapy and/or cryotherapy. Twelve patients had attempts at globe salvage (10 patients with systemic chemoreduction, 2 patients with intraarterial chemotherapy) before enucleation. Of these patients, eight eyes had residual viable tumor on histological examination and four had complete necrosis of tumor with dystrophic calcification. Out of 28 sockets with primary orbital implants, three had implant exposure. The first patient was a 4-month-old child who had familial and bilateral retinoblastoma (Group D) whose mother was also blind from childhood bilateral enucleations for retinoblastoma. The child had failed attempted systemic chemoreduction and local consolidation therapy prior to enucleation. At 7 months of age after failure of tumor control, an enucleation of the left eye was performed with placement of a bovine pericardium (Tutopatch) wrapped 18 mm spherical PMMA implant. The child had missed follow up with poor socket hygiene from inability of her mother to monitor the socket or apply lubricant eye drops. Implant exposure was noted on subsequent follow up requiring an implant removal with secondary dermis fat graft placement and satisfactory outcome until last follow up at 6 years of age (Fig. 1). The second patient had unilateral Group D retinoblastoma for which enucleation was performed at 2 years of age with 16 mm glass spherical implant placed by the myoconjunctivalization technique. She was referred at 7 years of age for an implant exposure and blepharoptosis of unknown duration, which was managed by an implant exchange with 20 mm Medpor implant wrapped in bovine pericardium followed by ptosis correction. She subsequently developed localized recurrent pyogenic granuloma, which was excised followed by an exposure that was subsequently repaired with a donor sclera patch graft at the age of 16 years (Fig. 2). She has remained incident free for 3 years until last follow-up. The third patient also had advanced unilateral (Group D) retinoblastoma which was enucleated at 1 year of age with the placement of a 16 mm spherical PMMA implant by a myoconjunctivalisation technique. The primary implant was exchanged with 20 mm Medpor implant wrapped in bovine pericardium at 2 years of age as the child grew, with a subsequent exposure at 3 years of age. She subsequently underwent an implant exchange A B A B FIGURE 1. A, Exposure in acrylic implant wrapped in bovine pericardium. B, Picture taken after dermis fat graft. FIGURE 2. A, Exposure in Medpor implant wrapped in Duraguard. B, Picture taken after repair of exposure with donor sclera Asia-Pacific Academy of Ophthalmology

3 Asia-Pacific Journal of Ophthalmology Volume 7, Number 5, September/October 2018 Anophthalmic Sockets in Retinoblastoma with 18 mm PMMA implant wrapped in donor sclera and has remained symptom free at last follow up at 23 years of age. Two of the three patients without a primary implant underwent repeat surgery. The first patient had undergone enucleation at 1 year of age for group E unilateral disease with cut end of the optic nerve positive for tumor. She developed orbital recurrence of tumor 6 months later for which additional excision was performed with adjuvant systemic chemotherapy without further recurrences. At 2 years of age she had a 10 mm PMMA implant placement with dermis fat graft. At 8 years of age she underwent an implant exchange with an 18 mm conical polypropylene (unwrapped) implant followed by attempted enophthalmos correction with an inferior subperiosteal Medpor enophthalmos wedge implant placement at 18 years. She was referred at 23 years of age with a large implant exposure of 2 years duration. which required an implant removal with an autologous dermis fat graft an acceptable result (Fig. 3). The second patient had bilateral retinoblastoma and underwent enucleation of the left eye without primary implant when she was 5 years old. She had received neoadjuvant systemic chemotherapy before enucleation. The second surgery involved reconstruction of the anophthalmic socket and placement of a 20 mm Medpor implant wrapped in donor sclera at 7 years of age with good long-term results. Only three of 28 patients with primary implant required repeat surgery. Notably, the rates of repeat surgery among patients with and without primary implant were 66.67% (2 out of 3) and 10.7% (3 out of 28), respectively. One patient who underwent enucleation elsewhere with small sized implants had signs of post-enucleation socket syndrome (PESS) with ill-fitting stock eyes, sunken deformity and excessive discharge (Figs. 4, 5). One of our patients with mild discharge without implant exposure despite having a customized prosthesis over a 20 mm porous implant was managed conservatively. Five patients referred to us for enucleation were conservatively managed. Globe preserving strategies employed included systemic chemoreduction, intraarterial chemotherapy, transpupillary thermotherapy, cryotherapy, and periocular carboplatin. Twelve patients received neoadjuvant therapy, seven received adjuvant therapy, and three received additional treatment both before and after enucleation. Of the patients who received additional treatments, only one patient had discharge. No other significant complications were noted in patients undergoing additional treatment modalities. DISCUSSION The basic principle of management of retinoblastoma is to achieve oncological medical cure, preserving the affected globe(s) and vision whenever possible without posing a threat to systemic spread and life. Various options of globe A FIGURE 3. A, Exposure of Medpor implant. B, After dermis fat graft. preservation include either systemic or selective ophthalmic artery chemotherapy with local consolidation by transpupillary thermotherapy, cryotherapy and more recently intravitreal and intracameral chemotherapy for vitreous seeding and anterior chamber seeding when the primary tumor has been eradicated. Occasionally plaque brachytherapy may also be indicated for suitable lesions, where available. When indicated, other options include periocular chemotherapy and plaque brachytherapy. Despite these globe-preserving modalities, ocular enucleation with primary orbital implant placement of the appropriate size within the posterior orbit either by a baseball technique or myoconjunctivalization, remains a relatively simple, brief, and definitive therapeutic procedure especially for otherwise unsalvageable eyes and in parts of the world where other modalities or trained professionals are scarce. Thus, when enucleation is indicated, optimal structural, cosmetic, and thus psychosocial rehabilitation is essential for the afflicted children so that they may lead stigma-free normal lives into adulthood acceptable personally and socially. Where histopathological high-risk characteristics that pose a threat for systemic spread are present, adjuvant treatment with chemotherapy or orbital radiotherapy may be indicated. As documented in many earlier studies, 1 3 we also found that bilateral retinoblastoma presented much earlier than unilateral retinoblastoma. When ocular enucleation is considered facilitating normal orbital growth is an important consideration in children. Nearly 80% of orbital growth happens before five years of age. 4 It has been reported that children who have their globes removed with incomplete rehabilitation less than three years of age had significant orbital growth retardation and facial asymmetry. 5 In our study, 85% of enucleations were performed before three years of age. Although most sockets were event-free, a small proportion did have residual morbidity either related to the primary procedure (poor technique, inadequate implant sizing or lack of implant placement) or patient characteristics. It should be remembered that owing to the developmental and structurally damaging effects and long-term risk of second malignant neoplasms in the field of radiation, external beam radiation should be avoided whenever possible. 6 The following are general principles in relation to anophthalmic sockets. These include ensuring absolute necessity for globe removal, removal of intact globe with as long an optic B A B C D FIGURE 4. Migrated acrylic implant in a patient with PESS with ill-fitting stock eye (A) and abrasions of surface of prosthesis due to ill fit (B). Note the presence of acquired epiblepharon in the lower lid secondary to ill-fitting stock eye prosthesis (C) and final outcome after she was fitted with customized best fit prosthesis (D) Asia-Pacific Academy of Ophthalmology 309

4 Jeyabal and Sundar Asia-Pacific Journal of Ophthalmology Volume 7, Number 5, September 2018 FIGURE 5. Poor motility of prosthesis in the right eye due to ill-fitting prosthesis in sunken socket with PESS. nerve stump as possible, obtaining a frozen section of cut end of the optic nerve to ensure intraoperative clearance of margins, harvesting tumor tissue for genetic studies when possible, choice and placement of an orbital implant of the appropriate size, shape, and material either wrapped (baseball technique) or unwrapped (myoconjunctivalisation). 7 When dealing with anophthalmic sockets in children, attention is paid to restoring the orbital architecture and volume with an appropriately sized implant. In general, adult-sized implants, usually around 20 mm, can be placed in children beyond 18 months of age as demonstrated by our study and many previous ones, 8,9 usually as a definitive implant. However, it must be borne in mind that implants that are too large, poorly covered, or placed too anteriorly may be associated with increased risk of exposure with or without extrusion and may provide inadequate space for the prosthesis. On the contrary, implants that are too small may fail to restore the lost orbital volume, thus resulting in a sunken socket and other features of PESS. A larger prosthesis as required in these sockets often results in complications including vertical descent, lower eyelid laxity, poor fit, limited motility, and increased discharge. 13 Yet another factor that results in socket discomfort and increased discharge is prior orbital irradiation from effects on the conjunctival mucosa and the lacrimal gland secretory process, which often increases with age. 13 Amongst various materials, porous implants are generally preferred in older children and when placed as definitive implants, preferably wrapped in donor sclera. While there is no difference in implant and prosthesis motility between porous and non-porous implants, porous implants however may facilitate pegging if and when indicated at a later date. However, in children less than 1 year of age, when an implant of ideal size cannot be placed, a non-porous implant may be placed in anticipation of easier implant exchange in the future. Likewise, in patients at high risk of orbital disease recurrence or where cost considerations play a role, non-porous may be preferred. 14,15 A summary of the advantages and disadvantages of various types of implants is listed in Table 1. An alternative to alloplastic orbital implants, especially when the socket surface is also compromised, is autologous dermis fat graft. This may also be considered where alloplastic implants may not be readily available. Although it is an ideal choice in non-irradiated contracted sockets especially after recurrent implant exposures, it runs the risk of increasing volume from fat hypertrophy with age and weight gain, thus rarely requiring debulking. 16,17 Dermis fat grafts can also be used to replace volume deficit after removal of exposed implants, for contracted socket, and after secondary reconstruction of orbits without primary implant, 18 as reported in our study without complications. As their survival is dependent on orbital vascularity, they are contraindicated in patients who have undergone external beam radiation. Wrapping of orbital implants not only serves as a protective barrier for exposure of implants, but also a means of anchoring extraocular muscles, both retaining the implant within the orbit and also imparting implant motility, which in turn may contribute to ocular prosthetic motility. Although implants placed behind the posterior Tenon s layer may remain secure, non-porous implants may migrate if poorly secured and porous implants may be exposed if overlying soft tissue cover is inadequate from poor layered closure or from an ill-fitting ocular prosthesis. Various wrapping materials and their characteristics are shown in Table 2. In the authors experience, donor sclera remains the wrapping material of choice owing to ready availability from most eye banks, ease of use, and minimizing implant exposure. In concordance with previous studies, our study also demonstrates a higher rate of implant exposure with bovine pericardium wrapping compared with donor sclera. The apparent higher rate of exposure of porous implants in our study can be attributed to factors like unwrapped implants or their secondary TABLE 1. Advantages and Disadvantages of Different Types of Orbital Implants Implant Type Example Advantages Disadvantages Non-porous Porous Polymethyl methacrylate (PMMA), silicone Hydroxyapatite Medpor (synthetic porous polyethylene) Aluminum oxide (bioceramic implant) Lightweight (acrylic) Low rate of exposure Cost-effective Regular and complete system of interconnecting pores Secure attachment of extraocular muscles Suitable for peg-sleeve system Integrates within orbit. Extrusion rare More biocompatible Lightweight Less expensive Protein coating minimizes socket inflammation Implant migration (imbrication) Sagging within orbit (silicone) Abrasive surface Conjunctival thinning High rate of exposure Infections with exposures Pyogenic granuloma formation Expensive Higher exposure rate (if unwrapped, anteriorly placed, poor layered closure) Less biocompatible than hydroxyapatite Asia-Pacific Academy of Ophthalmology

5 Asia-Pacific Journal of Ophthalmology Volume 7, Number 5, September/October 2018 Anophthalmic Sockets in Retinoblastoma TABLE 2. Characteristics of Various Types of Orbital Implant Wrapping Materials Wrapping Material Donor sclera Bovine pericardium Autologous grafts (temporal fascia, fascia lata) Vicryl mesh Characteristics Reduced implant exposure Potential risk of transmission of diseases from donor Long shelf life Higher chances of exposure than donor sclera Need for second operative site Prolonged operative time Less risk of infection/less inflammation Does not require second surgical site Inexpensive Allows suturing directly to implant Allows 360-degree entry of fibrovascular tissue Higher rate of exposure (possibly technique related) placements after multiple surgeries. Thus, primary implant placement certainly seems to be more advantageous than secondary implantation 22 with both socket and implant-related complications. Factors attributed to complications with secondary orbital implantation include disrupted orbital anatomy, collapsed orbital contents, underlying orbital fat loss, socket contracture with fibrosis and finally scarring of the extraocular muscles and intermuscular septa. 22 Customized ocular prosthesis complements a wellreconstructed anophthalmic socket with good physical and cosmetic rehabilitation of the sockets in these children with retinoblastoma. In conclusion, anophthalmic sockets in retinoblastoma pose long-term implications when primary rehabilitation is incomplete or performed poorly. A close collaboration between the ophthalmologist and ocularist with the placement of an implant of optimal size and position within the socket is required. The goals of management include a well-centered and -positioned orbital implant primarily with sufficient soft tissue and epithelial cover with a well-fitting customized ocular prosthesis, thus delivering good physical, social, and psychological rehabilitation. REFERENCES 1. El Zomor H, Nour R, Alieldin A, et al. Clinical presentation of intraocular retinoblastoma; 5-year hospital-based registry in Egypt. J Egypt Natl Canc Inst. 2015;27: Chawla B, Hasan F, Azad R, et al. Clinical presentation and survival of retinoblastoma in Indian children. Br J Ophthalmol. 2016;100: Subha L, Reddy AS, Ramyaa. A clinical study of retinoblastoma. J Pharm Bioallied Sci. 2015;7(Suppl 1):S2 S3. 4. Lin HY, Liao SL. Orbital development in survivors of retinoblastoma treated by enucleation with hydroxyapatite implant. Br J Ophthalmol. 2011;95: Oatts JT, Robbins JA, de Alba Campomanes AG. The effect of enucleation on orbital growth in patients with retinoblastoma. J AAPOS. 2017;21: Shildkrot Y, Kirzhner M, Haik BG, et al. The effect of cancer therapies on pediatric anophthalmic sockets. Ophthalmology. 2011;118: Shields JA, Shields CL, De Potter P. Enucleation technique for children with retinoblastoma. J Pediatr Ophthalmol Strabismus. 1992;29: Choi YJ, Park C, Jin HC, et al. Outcome of smooth surface tunnel porous polyethylene orbital implants (Medpor SST) in children with retinoblastoma. Br J Ophthalmol. 2013;97: Kirzhner M, Shildkrot Y, Haik BG, et al. Pediatric anophthalmic sockets and orbital implants: outcomes with polymer-coated implants. Ophthalmology. 2013;120: Schmitzer S, Simionescu C, Alexandrescu C, et al. The anophthalmic socket reconstruction options. J Med Life. 2014;7: Krishna G. Contracted sockets -I (aetiology and types). Indian J Ophthalmol. 1980;28: Aggarwal H, Singh SV, Kumar P, et al. Prosthetic rehabilitation following socket reconstruction with Blair-Brown graft and conformer therapy for management of severe post-enucleation socket syndrome a clinical report. J Prosthodont. 2015;24: Mourits DL, Hartong DT, Budding AE, et al. Discharge and infection in retinoblastoma post-enucleation sockets. Clin Ophthalmol. 2017;11: Baino F, Perero S, Ferraris S, et al. Biomaterials for orbital implants and ocular prostheses: overview and future prospects. Acta Biomater. 2014;10: Schellini SA, El Dib R, Limongi RM, et al. Anophthalmic socket: choice of orbital implants for reconstruction. Arq Bras Oftalmol. 2015;78: Mitchell KT, Hollsten DA, White WL, et al. The autogenous dermis-fat orbital implant in children. J AAPOS. 2001;5: Essuman VA, Tagoe NN, Ndanu TA, et al. Dermis-fat grafts and enucleation in Ghanaian children: 5 years experience. Ghana Med J. 2014;48: Raizada K, Shome D, Honavar SG. Management of an irradiated anophthalmic socket following dermis-fat graft rejection: a case report. Indian J Ophthalmol. 2008;56: Arat YO, Shetlar DJ, Boniuk M. Bovine pericardium versus homologous sclera as a wrapping for hydroxyapatite orbital implants. Ophthal Plast Reconstr Surg. 2003;19: Mourits DL, Moll AC, Bosscha MI, et al. Orbital implants in retinoblastoma patients: 23 years of experience and a review of the literature. Acta Ophthalmol. 2016;94: Custer PL, Trinkaus KM. Porous implant exposure: Incidence, management, and morbidity. Ophthal Plast Reconstr Surg. 2007;23: Smit TJ, Koornneef L, Mourits MP, et al. Primary versus secondary intraorbital implants. Ophthal Plast Reconstr Surg. 1990;6: Asia-Pacific Academy of Ophthalmology 311