RETINA OPHTHALMIC PEARLS
Diagnosis and Management of Central Retinal Vein Occlusion
etinal vein occlusion (RVO) has Ocular conditions. Open-angle glau- 1 a prevalence of 0.5%, making coma is a major ocular risk factor for Rit the second most-common CRVO. retinal vascular disorder after diabetic In addition, individuals with CRVO retinopathy.1 RVO is classified accord- in 1 eye are at higher risk of developing ing to the anatomic level of the occlu- CRVO in the fellow eye.2 In the Central sion, with 3 major distinct entities: Vein Occlusion Study (CVOS), 4% of • Central retinal vein occlusion patients presented with bilateral CRVO (CRVO): occlusion of the central reti- at study enrollment, and a further 5% nal vein at the level of, or posterior to, had evidence of previous CRVO in the the lamina cribrosa (Fig. 1) fellow eye at baseline. In the remaining • Hemiretinal vein occlusion (HRVO): subjects, 1.4% developed CRVO in the occlusion at the disc, involving either fellow eye during 3 years of follow-up. ACUTE CRVO. Classic “blood and thun- the superior or inferior hemiretina Other ocular risk factors include der” fundus appearance of a patient • Branch retinal vein occlusion retrobulbar external compression of the presenting acutely with central retinal (BRVO): occlusion of a tributary vein, central retinal vein, as occurs in thyroid vein occlusion of the right eye. typically at the site of an arteriovenous orbitopathy, or compression by intra crossing; thought to be caused by com- orbital space-occupying lesions. may be absent. In subacute or late pression from an overlying atheroscle- presentations in which disc swelling rotic arteriole Clinical Presentation has resolved (with or without collateral This article will focus on diagnosis Patients with CRVO typically present vessel formation), the flame-shaped and management of the first entity, with a history of unilateral acute, pain hemorrhages clear first, leaving deeper CRVO. less visual loss. Visual impairment may dot/blot hemorrhages that may be be severe, ranging from better than difficult to distinguish from a severe Risk Factors 20/40 to worse than 20/200. A relative microangiopathic retinopathy such as Systemic disorders. Systemic risk factors afferent pupillary defect may be present diabetic retinopathy (Fig. 2). Fluores- for CRVO include increasing age, diabetes in the affected eye. cein angiography (FA) may help to mellitus, and hypertension. In selected Fundus findings.Dilated fundus confirm the diagnosis of CRVO. cases, hypercoagulable states, including examination reveals unilateral disc Other key aspects. As part of the hyperhomocysteinemia and factor V swelling with peripapillary intraretinal examination, the clinician should note Leiden mutation, or local vessel factors hemorrhages, dilated tortuous veins, the intraocular pressure and cup-to-disc such as vasculitis are also associated and intraretinal dot, blot, and flame ratio, which may suggest concurrent with increased risk of CRVO. The hemorrhages in all quadrants, result- glaucoma, as well as any sequelae, such literature also contains case reports of ing in the classic “blood and thunder” as rubeosis iridis. Undilated gonioscopic many other systemic conditions possi- fundus appearance (Fig. 1). The macula examination is important to rule out bly associated with the development of may be edematous. neovascularization of the angles. CRVO. In less severe cases, disc swelling Types of CRVO Clinically, CRVO may be divided into 2 BY SIEHYEAN KIEW, MD, AND DANIEL S.W. TING, MD, PHD. EDITED BY SHA- major subtypes: ischemic and nonisch-
Kasi Sandhanam and Joseph Ho, Singapore National Eye Centre Eye National Singapore Sandhanam and Joseph Ho, Kasi RON FEKRAT, MD, AND INGRID U. SCOTT, MD, MPH. emic.
EYENET MAGAZINE • 33 2A 2B 2C
Ischemic. The CVOS investigators CHANGES OVER TIME. Same eye as shown in Fig. 1 at (2A) 1 month, (2B) 4 months, defined ischemic CRVO as evidence and (2C) 1 year following initial presentation, demonstrating evolution of the clinical of more than 10 disc areas of capillary picture. Disc edema resolves first, then the flame hemorrhages, and finally the dot nonperfusion on 7-field fundus FA (al- and blot hemorrhages, with development of collateral vessels at the optic disc. though investigators are reassessing this definition in light of recent advances in in a patient with existing nonischemic often performed in patients with CRVO widefield angiography). CRVO should, therefore, prompt further and is directed by the patient’s age, Ischemic CRVO may be identified by assessment for development of ischemic coexisting risk factors, and medical his- the following characteristics: CRVO. tory. Assessment should be performed • Poor visual acuity (>90% had VA of Of the eyes that remained non- in conjunction with an internist, as <20/200) ischemic, approximately 30% showed patients with RVO may be at higher • Presence of a relative afferent pupil- resolution of macular edema within risk of cardiovascular disease and cere- lary defect in the affected eye 15 months. Occurrence of subsequent brovascular accidents. • Presence of extensive dark, deep neovascular complications is rare in There are no clear guidelines on intraretinal hemorrhages nonischemic eyes. systemic testing, but it generally begins • Presence of multiple cotton-wool with a dilated funduscopic examination spots Workup in clinic, along with a detailed medical • Greater than 10 disc areas of retinal A thorough initial workup can provide history to identify risk factors; further capillary nonperfusion on 7-field FA useful information to guide clinical assessment includes blood pressure and • Reduced b-wave amplitude, reduced decision making. serum glucose, complete blood count, b:a ratio, and prolonged b-wave implic- Optical coherence tomography. and erythrocyte sedimentation rate. it time on electroretinography OCT is useful to confirm and quantify In young patients without clear risk In ischemic CRVO, visual acuity the severity of macular edema, assess factors, additional testing should be remains poor, often decreasing further the integrity of the ellipsoid zone/ considered to exclude a hematologic over time. Causes of visual loss include photoreceptor layers, and monitor or vasculitic etiology. chronic macular edema, macular isch- response to treatment. In clinical prac- emia, peripheral/global ischemia with tice, OCT measurements often guide Treatment secondary vitreous hemorrhage, and treatment decisions. All patients should optimize control neovascular glaucoma. Fluorescein angiography. Fea- of systemic risk factors, with the help Approximately 23% of eyes with tures of CRVO on FA include delayed of their internist. Management of the ischemic CRVO develop iris neovas- arm-to-retina time, prolonged arterio ocular manifestations may be divided cularization over 15 months; in the venous transit time (markedly so in into the following areas. CVOS, 44% of eyes that presented with ischemic CRVO), late staining along Macular edema. Both laser and vision worse than 20/200 subsequently vessel walls, capillary dropout with medical therapies have been used in the developed iris neovascularization.2 pruning of the vessels in areas of treatment of macular edema. Some patients may develop retinal ischemia, and late leakage in a petal- Laser. Studies assessing grid-pattern neovascularization. loid pattern in the presence of macular laser photocoagulation for treatment of Nonischemic. In the CVOS, 34% of edema (Fig. 3). macular edema in CRVO showed ana- eyes that initially presented with non- Clinically, FA allows evaluation of tomic improvement without improve- ischemic CRVO underwent conversion the extent of capillary nonperfusion ment in visual acuity.2 to an ischemic perfusion status over 3 and the degree of macular ischemia and Anti–vascular endothelial growth years2; conversion is heralded by rapid enables differentiation between collater- factor. Intravitreal anti-VEGF therapy visual deterioration in the affected al vessels and new vessels. is currently the gold standard of treat-
eye. Sudden decrease in visual acuity Systemic. Systemic evaluation is ment for macular edema associated Centre Eye National Singapore Sandhanam and Joseph Ho, Kasi
34 • JANUARY 2018 (mean improve- Anti-VEGF agents are antiangiogen- 3A 3B ment of 18.6 ic and may be useful adjuncts to PRP vs. 18.9 ETDRS in the management of patients with letters, respec- CRVO and associated anterior segment tively).3 neovascularization, particularly when Corticoste- the view of the fundus is not sufficient- roids. Cortico- ly clear to permit adequate PRP. steroids reduce Venous outflow.A number of alter- retinal capillary native therapies focused on improving permeability retinal blood flow have been described. 3C 3D and inhibit the These include the use of antiplatelet expression and agents (e.g., ticlopidine),4 hemodilu- metabolic path- tion,5 and thrombolytic agents de- way of VEGF. The livered systemically, intravitreally, or SCORE-CRVO directly into a retinal vein during pars trial demonstrat- plana vitrectomy. ed that intravitre- Techniques to alleviate a possible al triamcinolone compartment syndrome, with optic acetonide was nerve sheath decompression through FLUORESCEIN FINDINGS. FA at 4 time points shows (3A) superior to obser- an orbital approach or radial optic masking from intraretinal hemorrhages, (3B) delayed arte- vation for visual neurotomy via a pars plana approach, riovenous transit time, (3C) leakage at the swollen optic disc, loss associated have been tried. However, these are no and (3D) late staining of the vessel walls. with CRVO-relat- longer used because of their limited ed macular ede- benefit and significant risks. with CRVO. There is increasing evi- ma. The GENEVA trial evaluated the Creation of a laser chorioretinal dence that anti-VEGF therapy results use of a sustained-release intravitreal venous anastomosis (L-CRA) to by- in lower risk of visual loss, higher rates dexamethasone implant (Ozurdex) and pass the occluded central retinal vein of visual gains, greater reduction in demonstrated improvements in visual has been reported to be beneficial in central retinal thickness, and reduced acuity and macular thickness compared nonischemic CRVO, with improvement risk of progression to iris neovascular- with both sham and laser-treated groups. in visual acuity and reduced rates of ization. More recently, the Clinical Efficacy ischemic progression,6 but less so in For example, the CRUISE study and Safety of Ranibizumab Versus eyes with the ischemic type of disease. reported that intravitreal ranibizumab Dexamethasone for Central Retinal The failure of anastomosis was most significantly improved best-corrected Vein Occlusion (COMRADE C) trial likely due to endothelial cell damage visual acuity (BCVA) at 6 and 12 compared intravitreal ranibizumab secondary to ischemia.7 months compared with sham injec- 0.5 mg (monthly for at least 3 months, tions. In the open-label extension followed by as-needed dosing) to 1 Klein R et al. Arch Ophthalmol. 2008;126(4): HORIZON trial, the eyes initially a single injection of Ozurdex. This 513-518. treated with sham and subsequently trial reported similar efficacy between 2 Central Vein Occlusion Study Group. Arch Oph- treated with ranibizumab showed im- ranibizumab and Ozurdex but found thalmol. 1997;115(4):486-491. [Erratum in Arch provement in BCVA but did not catch a higher incidence of adverse effects in Ophthalmol. 1997;115(10):1275.] up to the visual outcomes attained by the group receiving Ozurdex. 3 Scott IU et al. JAMA. 2017;317(20):2072-2087. the group that received ranibizumab at Retinal ischemia. Current evidence 4 Yamamoto T et al. Am J Ophthalmol. 2004;138(5): enrollment. This finding suggests that recommends regular monitoring of 809-817. delaying treatment for macular edema patients with ischemic CRVO for 5 Wolf S et al. Graefes Arch Clin Exp Ophthalmol. has adverse effects on visual outcomes. development of iris or angle neovas- 1994;232(1):33-39. Aflibercept, a VEGF-trap mole- cularization, for which panretinal laser 6 McAllister IL et al. Ophthalmology. 2010;117(5): cule, has also been shown to improve photocoagulation (PRP) remains the 954-965. BCVA compared with sham and laser mainstay of treatment. 7 Kwok AK et al. Br J Ophthalmol. 2003;87(8): treatment in the COPERNICUS and There is currently no evidence to 1043-1044. GALILEO trials. recommend prophylactic treatment More recently, SCORE2, a random- prior to the development of new ves- Dr. Kiew is an ophthalmology resident at the ized noninferiority trial including eyes sels. However, in circumstances where Singapore National Eye Centre. Dr. Ting is an with CRVO or HRVO, demonstrated regular follow-up is impractical and the associate consultant at the Singapore National that bevacizumab was noninferior to degree of ischemia is severe (high risk Eye Centre and assistant professor at Duke-Na- aflibercept in terms of visual acuity of progression to neovascularization), tional University Singapore. Relevant financial
Kasi Sandhanam and Joseph Ho, Singapore National Eye Centre Eye National Singapore Sandhanam and Joseph Ho, Kasi gain at month 6 compared to baseline prophylactic PRP may be appropriate. disclosures: None.
EYENET MAGAZINE • 35