A 36-year-old woman (patient 15) after autologous fat injections in the glabella and nasolabial fold. A, Fundus photograph obtained at the initial visit shows diffuse retinal edema and multiple segmented retinal arteries with fat emboli. B and C, Retinal and choroidal perfusion is severely diminished. D, Electroretinography (ERG) shows markedly reduced cone and rod responses. E, Multifocal acute infarction in the left middle cerebral artery territories on diffusion-weighted brain magnetic resonance imaging. F, Three months later, a retinal fibrous scar associated with optic atrophy was observed. The initial and final visual acuity in the left eye was no light perception.
A 27-year-old woman (patient 44) after hyaluronic acid injections in the glabella and nasal dorsum for rhinoplasty. A, Fundus photograph obtained at the initial visit shows normal retina and optic disc appearance. B, A fluorescein angiogram shows normal retinal perfusion. C, An indocyanine green angiogram shows normal choroidal perfusion. D, Three months later, the optic disc has a temporal pallor. E, Optical coherence tomography reveals normal retinal structure. F, Electroretinographic (ERG) findings are normal in both eyes and symmetric. G, Goldmann perimetry reveals a small remnant island in the nasal area. H, Visual evoked response (VER) shows delayed latency and reduced amplitude in the right eye. I, No definitive evidence of intraorbital trauma (eg, orbital wall fracture, hemorrhage) or acute brain parenchyma lesions are found on T2-weighted magnetic resonance imaging. J, Diffusion-weighted magnetic resonance imaging shows a focal high–signal intensity lesion in the retrobulbar optic nerve of the right eye (yellow arrowhead).
Injected filler material (yellow droplet) is presumed to access the ophthalmic artery retrogradely via the supratrochlear, supraorbital, or dorsal nasal artery. Ophthalmic artery occlusion (OAO) is likely caused by complete proximal ophthalmic artery obstruction by a large filler bolus that migrated backward from the high injection pressure. It may also be that small particles migrated back to the central retinal artery and posterior ciliary artery origins and dispersed anterogradely into each branch as injection pressure decreased. This would cause a diffuse obstruction. Generalized posterior ciliary artery occlusion (GPCAO) or central retinal artery occlusion (CRAO) may occur depending on the extent of central retinal artery or posterior ciliary artery obstruction. When only the medial short posterior ciliary artery is involved, localized posterior ciliary artery occlusion (LPCAO) involving only the nasal choroid occurs. When only a branch of the central retinal artery is occluded, a branch retinal artery occlusion (BRAO) occurs. The mechanism of posterior ischemic optic neuropathy (PION) remains uncertain. The pial vascular plexus supplies blood to the intraorbital posterior optic nerve, and some vessels responsible for the pial plexus, which usually arise directly from the ophthalmic artery, might also be involved in these cases. Last, some particles may have accessed the internal carotid artery, causing a brain infarction. SNUBH indicates Seoul National University Bundang Hospital.
eTable 1. Summary of Total Patients With Iatrogenic Occlusion of the Ophthalmic Artery and Its Branches After Cosmetic Facial Filler Injections
eTable 2. Comparison of Clinical Characteristics Between Autologous Fat–Injected Group and Hyaluronic Acid–Injected Group
eFigure 1. Generalized Posterior Ciliary Artery Occlusion With Relative Central Retinal Artery Sparing
eFigure 2. Central Retinal Artery Occlusion
eFigure 3. Localized Posterior Ciliary Artery Occlusion
eFigure 4. Branch Retinal Artery Occlusion
Park KH, Kim Y, Woo SJ, Kang SW, Lee WK, Choi KS, Kwak HW, Yoon IH, Huh K, Kim JW, for the Korean Retina Society. Iatrogenic Occlusion of the Ophthalmic Artery After Cosmetic Facial Filler InjectionsA National Survey by the Korean Retina Society. JAMA Ophthalmol. 2014;132(6):714-723. doi:10.1001/jamaophthalmol.2013.8204
Iatrogenic occlusion of the ophthalmic artery and its branches is a rare but devastating complication of cosmetic facial filler injections.
To investigate clinical and angiographic features of iatrogenic occlusion of the ophthalmic artery and its branches caused by cosmetic facial filler injections.
Design, Setting, and Participants
Data from 44 patients with occlusion of the ophthalmic artery and its branches after cosmetic facial filler injections were obtained retrospectively from a national survey completed by members of the Korean Retina Society from 27 retinal centers. Clinical features were compared between patients grouped by angiographic findings and injected filler material.
Main Outcomes and Measures
Visual prognosis and its relationship to angiographic findings and injected filler material.
Ophthalmic artery occlusion was classified into 6 types according to angiographic findings. Twenty-eight patients had diffuse retinal and choroidal artery occlusions (ophthalmic artery occlusion, generalized posterior ciliary artery occlusion, and central retinal artery occlusion). Sixteen patients had localized occlusions (localized posterior ciliary artery occlusion, branch retinal artery occlusion, and posterior ischemic optic neuropathy). Patients with diffuse occlusions showed worse initial and final visual acuity and less visual gain compared with those having localized occlusions. Patients receiving autologous fat injections (n = 22) had diffuse ophthalmic artery occlusions, worse visual prognosis, and a higher incidence of combined brain infarction compared with patients having hyaluronic acid injections (n = 13).
Conclusions and Relevance
Clinical features of iatrogenic occlusion of the ophthalmic artery and its branches following cosmetic facial filler injections were diverse according to the location and extent of obstruction and the injected filler material. Autologous fat injections were associated with a worse visual prognosis and a higher incidence of combined cerebral infarction. Extreme caution and care should be taken during these injections, and physicians should be aware of a diverse spectrum of complications following cosmetic facial filler injections.
Increasing interest in aesthetics and rejuvenation has led to a recent dramatic increase in the number of aesthetic procedures performed. This is especially true for cosmetic nonsurgical injection procedures because of the development of numerous novel filling materials, the ease of the procedure, and a favorable safety profile.1,2
Various materials are used for facial filling, with autologous fat, collagen, hyaluronic acid, and biosynthetic polymers the most common.3 Although these products generally have favorable safety profiles, with only minor adverse effects (eg, bruising, redness, and local swelling), in rare cases more severe or devastating adverse effects have occurred, including blindness.4- 15
Blindness after cosmetic facial filler injections is a rare complication, with only fragmented case reports on this issue. Our group previously reported 12 cases of iatrogenic retinal artery occlusion associated with cosmetic filler injections.13 Clinical features and the presumed mechanism of retrograde occlusion were investigated. However, because of the small sample size, clinical features by occlusion degree and location and by filler type could not be fully evaluated. Lazzeri et al12 examined 32 cases of blindness after cosmetic facial injections reported in the literature. However, that study was not uniformly designed and did not represent a comprehensive summary of clinical features, especially regarding angiographic findings and visual outcomes. In the present study, we investigated clinical and angiographic features of iatrogenic occlusion of the ophthalmic artery and its branches caused by cosmetic facial filler injections.
The study adhered to the tenets of the Declaration of Helsinki and was approved by the institutional review board of Seoul National University Bundang Hospital. Between December 1, 2012, and May 30, 2013, we conducted a national survey of the members of the Korean Retina Society on their personal experience with iatrogenic retinal artery occlusion caused by cosmetic facial filler injections. There are approximately 60 referral retinal clinics in Korea, with most being university hospitals. We obtained reports of 44 cases from 27 retinal centers, including most of the major centers in Korea, of which 19 cases (43%) had already been published in the literature.4,6,7,10,11,13- 15 We collected survey case report forms and available ocular imaging, including fundus photographs, fluorescein angiography, indocyanine green angiography, and optical coherence tomography, as well as examination data (visual field tests, electroretinography, etc). Data obtained with survey case report forms were clinic type, injection site, injected filler material, basic demographic information (age, sex, and laterality of the involved eyes), and initial symptoms and signs (ocular pain, ophthalmoplegia, strabismus, ptosis, corneal edema, anterior chamber inflammation, pupillary abnormality, neurologic symptoms, and skin lesion), as well as examination results, including brain magnetic resonance (MR) imaging, treatment methods, visual outcomes, and final sequelae.
Ophthalmic artery occlusion (OAO) was classified into the following types according to the presumed location of occlusion based on fundus photographs and angiographic findings: (1) OAO, (2) generalized posterior ciliary artery occlusion (PCAO) with relative central retinal artery sparing, (3) central retinal artery occlusion (CRAO), (4) localized PCAO, (5) branch retinal artery occlusion, and (6) posterior ischemic optic neuropathy (PION). Clinical features were compared between patients grouped by angiographic findings and injected filler material.
For statistical analyses, Snellen best-corrected visual acuity (BCVA) measurements were converted to the logarithm of the minimal angle of resolution (logMAR) scale. For visual acuity worse than 20/400, logMAR values of 2.0, 2.3, 2.6, and 2.9 were assigned for counting fingers, hand motion, light perception (LP), and no LP (NLP), respectively.16 The means of continuous variables were compared with Mann-Whitney test and Kruskal-Wallis test, and categorical variables were compared with χ2 test and Fisher exact test. Statistical analyses were performed using available software (PASW, version 18.0; SPSS, Inc), and P < .05 was considered statistically significant.
Demographic and clinical characteristics examined in this study are summarized in Table 1 and in eTable 1 in the Supplement. Briefly, most patients with iatrogenic occlusion of the ophthalmic artery after cosmetic facial filler injections were young women (mean age, 35.8 years; 93% [41 of 44] female). The most common types of filler injected were autologous fat (50% [22 of 44]) and hyaluronic acid (30% [13 of 44]). Filler was most commonly injected in the glabella area (59% [26 of 44]) and was used at multiple sites in 8 patients (18%). Eighty-four percent of patients sought medical attention within 24 hours of symptom onset, which occurred immediately after filler injections. Patients were most often managed with observation (30% [13 of 44]) or anterior chamber paracentesis (25% [11 of 44]). Thirty-one patients (71%) underwent brain MR imaging, of whom 12 (39%) showed focal or multifocal brain infarctions. Six patients also had corresponding neurologic symptoms, including contralateral hemiplegia (4 patients) and dysarthria (2 patients). Visual prognosis was poor, with 27 patients (61%) having a final visual acuity of NLP. In 21 patients with at least 6 follow-up months, iris atrophy (2 patients [10%]), strabismus (7 patients [33%]), and skin lesions (2 patients [10%]) persisted until the last follow-up visit. One patient (5%) progressed to phthisis bulbi.
In this study, we classified OAO into 6 types according to angiographic findings and fundus photographs. Clinical characteristics across groups were then compared (Table 2).
Seventeen patients had OAO, the most severe occlusion form, with both retinal and choroidal insufficiency. This type is characterized by diffuse retinal whitening and severely compromised retinal and choroidal filling as viewed on fluorescein angiography (Figure 1A-C). Nine of 15 patients (60%) who underwent brain MR imaging exhibited focal or diffuse brain infarction (Figure 1E), representing 75% of all patients with a brain lesion. Visual prognosis was poor. The BCVA after a mean (SD) follow-up period of 14.4 (15.4) months was NLP in all patients. Most OAO cases had received autologous fat injections (71% [12 of 17]) and had high amounts of ocular pain (65% [11 of 17]) and ophthalmoplegia or strabismus (77% [13 of 17]).
Three patients showed choroidal ischemia with spared retinal circulation. Initially, only focal retinal edema was present, and optical coherence tomography revealed normal retinal architecture (eFigure 1A and E in the Supplement). The retinal perfusion remained intact, despite severely compromised choroidal perfusion (eFigure 1B-D in the Supplement). However, within 3 days, the inner retina became compromised (eFigure 1F in the Supplement), and optic atrophy eventually ensued (eFigure 1G in the Supplement). Visual outcomes were poor, with a final BCVA of NLP in all 3 patients.
Eight patients showed delayed and compromised retinal circulation and reduced choroidal perfusion, but choroidal ischemia was not as prominent as it was in the OAO cases (eFigure 2B and C in the Supplement). Severe inner retinal edema was detected on optical coherence tomography, and a cherry-red spot was observed (eFigure 2A and D in the Supplement). The CRAO cases were also associated with autologous fat injections (75% [6 of 8]), and visual prognosis was poor, with a final BCVA of NLP in 6 of 8 patients.
All 4 cases of localized PCAO showed nasal choroidal ischemia, with no compromising of the retinal or remaining choroidal circulations (eFigure 3 in the Supplement). Because the macular region was unaffected, visual prognosis was good in this group, except in patient 30, who also had optic disc edema, likely resulting from ischemic optic nerve head damage. Two patients (29 and 31) initially had anterior segment ischemia, as evidenced by corneal edema and anterior chamber reaction, which resolved during the follow-up period. These 2 patients also showed a dramatic visual acuity improvement (from 20/63 to 20/20 in patient 29 and from hand motion to 20/25 in patient 31), suggesting that the initial visual impairment was caused by corneal edema and anterior chamber reaction, with visual acuity then improving as these anterior segment ischemic changes resolved. In contrast to the more diffusely occluded cases (OAO, generalized PCAO, and CRAO), none of the localized PCAO cases were associated with autologous fat injections. Three patients had received hyaluronic acid filler, and one patient had received poly-l-lactic acid filler.
The 10 patients with branch retinal artery occlusion had localized retinal edema in the involved arterial domain (eFigure 4A-C in the Supplement). Visual acuity was good in most cases, and 5 patients reported only a visual field defect (eFigure 4D in the Supplement). Fewer cases had received autologous fat filler (2 patients [20%]) than in the diffuse occlusion group.
We encountered 2 cases of suspicious PION. A 27-year-old woman (patient 44 in Figure 2) underwent hyaluronic acid injections in the glabella and nasal dorsum for rhinoplasty. The cannula was inserted from the nasal tip to the root. When the cannula tip reached the nasal root, hyaluronic acid was gently and continuously injected as the cannula was slowly removed, leaving a filler trail in its place. Within 5 minutes, the right eyeball deviated laterally, and the patient reported diplopia, blurry vision, and eyeball pain. The physician immediately injected hyaluronidase using the same method as hyaluronic acid injections and massaged around the right orbit. When the patient visited the emergency department 4 hours later, visual acuity was LP OD and 20/25 OS. She also demonstrated 30–prism diopter exotropia and had limitations in supraduction, infraduction, and adduction in the right eye. No definitive abnormalities were observed on fundus examination, and retinal and choroidal perfusion was normal based on fluorescein angiography and indocyanine green angiography (Figure 2A-C). Electroretinography findings were normal and symmetric (Figure 2F), but the visual evoked potentials revealed delayed latency and reduced signal amplitude in the right eye (Figure 2H). No evidence of intraorbital trauma (eg, orbital wall fracture, intraorbital hemorrhage) was found (Figure 2I), but a focal high–signal intensity lesion was observed in the retrobulbar optic nerve of the right eye (Figure 2J). This suggested ischemic optic neuropathy, and 3 months later the optic disc had a temporal pallor (Figure 2D). Patient 43 also had normal fundus appearance initially, but 6 months later optic atrophy ensued. Both patients had poor visual outcome, with visual acuities of LP in patient 44 and NLP in patient 43.
We divided OAOs into 2 major categories of diffuse and localized. Diffuse occlusions included OAO, generalized PCAO, and CRAO. All of these cases resulted in generalized decreases in retinal or choroidal perfusion and involved the retina and choroid as a whole. Localized occlusions included localized PCAO, branch retinal artery occlusion, and PION and showed only focal vascular compromise in the retina or choroid. Uninvolved areas were normal in perfusion and in appearance. Compared with localized occlusions, diffuse occlusions had higher rates of autologous fat injections (68% [19 of 28] vs 19% [3 of 16], P = .002), brain lesions on MR imaging (50% [12 of 24] vs 0% [0 of 7], P = .02), and ocular pain (54% [15 of 28] vs 13% [2 of 16], P = .01). Visual prognosis was also poor in the diffuse occlusion group, with a worse initial and final BCVA, as well as a worse visual gain (Table 2).
Differences in clinical characteristics between the 2 most commonly injected substances (autologous fat and hyaluronic acid) were examined. Patients who received autologous fat injections were more likely to have diffuse occlusions (86% [19 of 22] vs 39% [5 of 13], P = .007) and consequently a worse visual outcome (mean [SD] final BCVA, 2.6 [0.8] vs 1.4 [1.4]; P = .01), a lower visual gain (mean [SD], −0.1 [−0.2] vs 0.3 [0.6]; P = .02), and a higher rate of long-term (≥6 months) visual loss (100% [9 of 9] vs 43% [3 of 7], P = .02). Patients injected with autologous fat also had a higher prevalence of brain lesions on MR imaging (46% [10 of 22] vs 8% [1 of 13], P = .03). In contrast, hyaluronic acid injections were associated with a higher prevalence of anterior segment ischemia, which manifested as corneal edema (39% [5 of 13] vs 5% [1 of 22], P = .02) and anterior chamber inflammation (54% [7 of 13] vs 0% [0 of 22], P < .001). No significant differences were observed between the groups in the number of injections or the site of injection (eTable 2 in the Supplement).
In this study, we summarized clinical and angiographic characteristics of 44 patients experiencing iatrogenic occlusion of the ophthalmic artery and its branches after cosmetic facial filler injections. This is the largest case series reported to date on ocular complications following cosmetic facial filler injections and includes a comprehensive analysis of clinical characteristics of patients according to their angiographic findings and injected filler material.
Figure 3 shows a schematic drawing of the ophthalmic artery and its branches. Possible obstruction points responsible for various clinical features are shown. Many researchers have proposed that retrograde embolic mechanisms are responsible for the development of arterial occlusion.5- 8,12,13,17,18 An OAO seems to be caused by a large bolus of filler material that has migrated in a retrograde fashion to the ophthalmic artery origin, creating a complete obstruction. Small particles can also migrate back to the central retinal artery and posterior ciliary artery origins, and this may result in particle dispersion into each arterial branch, causing a multifocal obstruction. The latter mechanism could apply to other diffuse occlusions, including generalized PCAO with central retinal artery sparing and CRAO. This results in a clinical spectrum of obstructions. In contrast, localized occlusion cases are related to focal obstructions, and uninvolved arteries have completely normal perfusion.
Two cases of presumed PION were observed in this study. They had an initially normal-appearing optic disc and fundus, but optic atrophy or temporal pallor developed during the follow-up period. In addition, brain MR imaging in patient 44 revealed a high–signal intensity lesion in the retrobulbar optic nerve without any signs of intraorbital trauma, suggesting ischemic damage. This is the first report to date associating PION with cosmetic facial filler injections. Unfortunately, the pathogenic mechanism remains a mystery. Considering that the patients were healthy before cosmetic facial filler injections and that it is not a complex surgical procedure requiring general anesthesia or resulting in arterial hypotension (the causes of surgical PION19), it is reasonable to postulate that small injected particles directly occluded the vasculature responsible for perfusing the posterior part of the optic nerve. The posterior optic nerve is supplied by the pial plexus,19- 22 and we theorize that injected material approaches the pial plexus vessel origins in a retrograde manner that usually arises directly from the ophthalmic artery,21 but the precise mechanism remains unknown. However, when sudden visual loss following cosmetic facial filler injections is accompanied by a normal-appearing fundus and optic disc, PION should be considered. Visual evoked potentials or brain imaging may also be helpful in making or ruling out the diagnosis.
We also found that the injection material was strongly associated with the occlusion type, visual outcomes, and combined cerebral infarction incidence. The use of autologous fat was related to the occurrence of diffuse OAO, a worse visual prognosis, and a higher combined cerebral infarction incidence compared with hyaluronic acid. In a previous study,13 our group postulated that size variation in injected particles leads to different clinical occlusion features (eg, small-sized hyaluronic acid is less likely to completely block the ophthalmic artery). In addition, because of high resorption,23 fat is usually injected in a substantial amount using a large-bore needle (14-gauge to 18-gauge), aiming to overcorrect the cosmetic defect.24 The high volume and high pressure of this injection seem responsible for multifocal and complete obstruction of the ophthalmic artery and its branches after autologous fat injections.
Cerebral infarction was more frequently observed in autologous fat–associated retinal artery occlusions. Of 12 patients with a cerebral parenchymal lesion on MR imaging, 10 had received autologous fat filler. The remaining 2 patients had received hyaluronic acid or collagen filler. When the cosmetic facial filler injection literature was reviewed for related cerebral infarctions, it was found that most cases occurred after autologous fat injections.5,17,18,25- 27 It seems that fat particles are directly injected into the internal carotid artery system in a retrograde fashion rather than into the systemic circulation because most fat particles and emboli are filtered out of circulation by the lungs.28 In contrast to autologous fat, which resorbs substantially, hyaluronic acid attracts an influx of water and for the most part retains its injected volume.24 This may affect hyaluronic acid particle migration properties and prevent diffuse propagation.
Our study was limited by its retrospective design. First, we could not directly calculate the ocular complication incidence following cosmetic facial filler injections. However, most Korean general ophthalmologists usually refer complicated cases to referral centers because such complications may lead to legal problems that are difficult to handle in private clinics. Therefore, we believe that most of the cosmetic facial filler–associated blindness cases were included in our national survey. According to the International Society of Aesthetic Plastic Surgery, more than 20 000 cases of autologous fat injections and more than 90 000 cases of hyaluronic acid injections were performed in 2011 in South Korea.29 Considering the large number of procedures performed and the few cases collected, despite a national survey, the rate of ophthalmic complications following cosmetic facial filler injections seems extremely low. Second, the study was not designed to identify risk factors that might mitigate complications, such as the type of needle used (sharp vs blunt cannula), the gauge of the instruments, and the depth or total volume of the injection. However, to decrease the risk of intravascular injection, many physicians suggest slow injection with low pressure of a small fractionated dose instead of a bolus injection, as well as the use of a blunt cannula, particularly with the injection of larger volumes.12,30
Clinical features of cosmetic facial filler injection–associated iatrogenic occlusion of the ophthalmic artery branches varied according to the location and extent of OAO and the injected filler material. Autologous fat injections were associated with a higher diffuse occlusion rate, a worse visual prognosis, and a greater incidence of combined cerebral infarction compared with hyaluronic acid injections. Although this is a rare complication of cosmetic facial filler injections considering the large number performed each year, it is usually devastating, and physicians and patients should be aware of a possible ophthalmic complication.
Submitted for Publication: August 22, 2013; final revision received November 26, 2013; accepted December 19, 2013.
Corresponding Author: Jong Woo Kim, MD, PhD, Department of Ophthalmology, Kim’s Eye Hospital, Konyang University College of Medicine, 156 4-Ga Yeoungdeungpo-Dong, Yeoungdeungpo-Gu, Seoul 150-034 Korea (email@example.com).
Published Online: March 27, 2014. doi:10.1001/jamaophthalmol.2013.8204.
Author Contributions: Drs Huh and J. W. Kim had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Park and Y.-K. Kim contributed equally to this article and should be considered co-first authors.
Study concept and design: Park, Y.-K. Kim, Woo.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Park, Y.-K. Kim, Woo.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Park, Y.-K. Kim.
Study supervision: Park, Huh, J. W. Kim.
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported by the Korean Retina Society and by grant 2013R1A2A2A04015829 from the National Research Foundation of Korea (Drs Park and Woo) for the design and conduct of the study and the analysis of the data.
Role of the Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Group Information: The Korean Retina Society members were Hee Seung Chin, MD, Department of Ophthalmology and Inha Vision Science Laboratory, Inha University School of Medicine, Incheon, Korea; Hum Chung, MD, and Jang Won Heo, MD, Department of Ophthalmology, Seoul National University College of Medicine, Seoul National University, Seoul, Korea; Hyun Woong Kim, MD, Department of Ophthalmology, Busan Paik Hospital, Inje University College of Medicine, Busan, Korea; Jae Suk Kim, MD, Department of Ophthalmology, Sanggye Paik Hospital, Inje University College of Medicine, Busan, Korea; Jeeyun Ahn, MD, Department of Ophthalmology, Seoul National University College of Medicine, Seoul Metropolitan Government–Seoul National University, Boramae Medical Center, Seoul, Korea; Joo Eun Lee, MD, Department of Ophthalmology, Haeundae Paik Hospital, Inje University College of Medicine, Busan, Korea; Ju Byung Chae, MD, Department of Ophthalmology, Chungbuk National University School of Medicine, Cheongju, Korea; June Gone Kim, MD, and Joo Yong Lee, MD, Department of Ophthalmology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea; Kyu Seop Kim, MD, Department of Ophthalmology, Seoul St Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea; Moo Hwan Chang, MD, Department of Ophthalmology, Dankook University Medical College, Cheonan, Korea; Moon Key Lee, MD, Purun Eye Hospital, Gwangju, Korea; Sang Jin Kim, MD, Department of Ophthalmology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea; Seong-Woo Kim, MD, and Jaeryung Oh, MD, Department of Ophthalmology, Korea University College of Medicine, Seoul, Korea; Seung-Young Yu, MD, Department of Ophthalmology, Kyung Hee University Medical Center, College of Medicine, Kyung Hee University, Seoul, Korea; Sung Pyo Park, MD, Department of Ophthalmology, Hallym University College of Medicine, Seoul, Korea; Sung Soo Kim, MD, and Christopher Seung Kyu Lee, MD, Department of Ophthalmology and Institute of Vision Research, Yonsei University College of Medicine, Seoul, Korea; Woohyok Chang, MD, Department of Ophthalmology, Yeungnam University College of Medicine, Daegu, Korea; Yong Seop Han, MD, Department of Ophthalmology, Gyeongsang National University School of Medicine, Jinju, Korea; Yong Sok Ji, MD, Department of Ophthalmology, Chonnam National University Medical School, Gwangju, Korea; Young Hoon Lee, MD, Department of Ophthalmology, Konyang University College of Medicine, Daejeon, Korea; Young Hoon Ohn, MD, Department of Ophthalmology, Soonchunhyang University Bucheon Hospital, Soonchunhyang University College of Medicine, Bucheon, Korea; Yun Taek Kim, MD, Department of Ophthalmology, School of Medicine, Ewha Womans University Mokdong Hospital, Ewha Womans University, Seoul, Korea; and Young Joon Jo, MD, Department of Ophthalmology, Chungnam National University School of Medicine, Daejeon, Korea.