As we search for new therapies to treat retinal vein occlusions, it is important to recall what we know about the mechanisms that are in play. An ischemic CRVO occurs when venous thrombosis is closer to the optic nerve head and anterior to most preexisting collateral veins, causing venous pressure to increase to the point where arterial flow is compromised. 1Whether this occurs at the lamina cribrosa or more posteriorly is inconsequential. What is important is the location of the occlusion in relation to the naturally occurring collaterals. With a more posterior occlusion within the CRVO as it passes through the optic nerve, collateral veins allow limited flow, so there is still arterial pressure, creating a nonischemic CRVO. According to Hayreh, the outlook is poor for ischemic CRVOs: 70% develop rubeosis; 50% develop neovascular glaucoma.2

DIFFERENTIATING CRVO FROM ISCHEMIC CRVO
For the Central Vein Occlusion Study (CVOS),3 researchers primarily used fluorescein angiography to differentiate ischemic from nonischemic CRVOs. They preferred the term nonperfused because the nonperfusion on the angiogram does not differentiate if the retina is ischemic or infarcted. The CVOS defined a nonperfused central vein occlusion as one with at least 10 disc areas of nonperfusion. Substantial hemorrhage often exists in a CRVO, so determining if areas on angiography are nonperfused can be difficult. If hemorrhage obscured the view, a CRVO was termed indeterminate. The CVOS did determine, however, that most indeterminate central vein occlusions have a natural history similar to nonperfused vein occlusions. In summary, angiography can be used to differentiate perfused from nonperfused central vein occlusions, and if there is too much blood to make this differentiation, the vein occlusion should be considered nonperfused.

In natural history studies, Hayreh has shown that several prognostic tests are sensitive and specific identifiers for ischemic CRVO (Figure 1).1 The CVOS found that if one takes into account the visual acuity and extent of hemorrhage, then additional data from pupil responses to electroretinogram provide further prognostic information as to whether the eye will progress from nonperfusion to neovascularization.

Hayreh reported the cumulative risk of developing iris neovascularization, angle neovascularization, and neovascular glaucoma is highest during the first 90 days. After 90 to 180 days, venous-venous collaterals often enlarge enough to decrease venous pressure, and improvement can occur. By about 9 months, the curve is flat (Figure 2).

PROMISING INTERVENTIONS
The list of CRVO interventions is littered with good intentions and bad results. Laser anastomosis has been shown to work in case series, but usually an adequate anastomosis cannot be created.4 Radial optic neurotomy and optic nerve sheath fenestration have not been shown convincingly to alter visual acuity outcomes or progression to neovascularization in a way that differs from the natural history. Ongoing studies, such as the Standard Care vs Corticosteroid for Retinal Vein Occlusion (SCORE) study,5 should tell us how steroid therapy compares with the natural history. And thus far, intravitreal anti-VEGF therapy has shown promising short-term results in uncontrolled studies.

Is there a rationale for using an anti-VEGF agent? Pe'er looked at neovascular glaucoma enucleation specimens going back to the 1920s and found messenger RNA upregulation of VEGF.6 In another study, Boyd and colleagues performed anterior chamber taps in patients with neovascular glaucoma before and after laser treatment. They found neovascularization occurred when aqueous VEGF concentrations were 849 pg/mL to 1569 pg/mL and regressed fully when they fell below 550 pg/mL.7

We also have good studies showing increased VEGF in proliferative diabetic retinopathy and vein occlusions.8,9 These data confirm that VEGF correlates to neovascularization.

RAVE TRIAL
In the ongoing Rubeosis Anti-VEGF (RAVE) Trial for Ischemic CRVO,10 monthly intravitreal injections of ranibizumab (Lucentis, Genentech, Inc.) were administered over 9 months. Optical coherence tomography, wide-field and conventional angiography, and Goldmann perimetry were performed monthly. The goal was to prevent neovascular glaucoma and visual field loss related to panretinal photocoagulation, and to learn more about nonperfused CRVO.

Preliminary analysis of the first 10 subjects suggests that one went from marked retinal thickening to resolution of edema and even loss of normal retinal thickness (Figure 3). Despite these large improvements in retinal thickening, most subjects had only slightly improved visual acuity, usually at very low (poor) levels (Figure 4). Figure 5 shows best-corrected visual acuity changes using the 20-minute ETDRS refraction: 60% had a 4-line gain with monthly ranibizumab to month 9. Although some of these follow-up visual acuities are only in the 20/100 to 20/200 range, visual acuity appears definitively better than the initial acuity and may have some value to the patient.

The next question was: What would happen when treatment was stopped? Would regression of effect occur because the treatment had led to upregulation of VEGF?

In about two-thirds of the eyes, the swelling did not return (Figure 6). In about one-third, the swelling returned immediately, and those subjects lost the visual acuity they had gained (Figure 7). Some of the gains from baseline to month 9 were lost by not treating from month 9 to month 12 (Figure 8). Thus far, it appears much of the edema in these cases is VEGF-driven and not necessarily related only to venous pressure, because if you give enough anti-VEGF, the edema resolves, and you can prevent it if you keep giving anti-VEGF.

QUESTIONS REMAIN
Questions remain on the management of nonperfused CRVO. Why is there a disconnect between edema and neovascularization? Why do some patients with or without edema get neovascularization? We should have more definitive data in the next year or so from trials that are currently recruiting, such as SCORE, BRAVO (A Study of the Efficacy and Safety of Ranibizumab Injection in Patients With Macular Edema Secondary to Branch Retinal Vein Occlusion), CRUISE (A Study of the Efficacy and Safety of Ranibizumab Injection in Patients With Macular Edema Secondary to Central Retinal Vein Occlusion), and the Branch Retinal Vein Occlusion or Central Vein Occlusion with Macular Edema Posurdex Implant Study, which should put us closer to answering some of these questions. ■

1. Hayreh SS, Klugman MR, Beri M, Kimura AE, Podhajsky P. Differentiation of ischemic from non-ischemic central retinal vein occlusion during the early acute phase. Graefes Arch Clin Exp Ophthalmol. 1990;228:201-217.
2. Hayreh, SS; Rojas, P; Podhajsky, P; Montague, P; Woolson, RF. Ocular neovascularization with retinal vascular occlusion-III. Incidence of ocular neovascularization with retinal vein occlusion. Ophthalmology. 1983;90:488–506.
3. The Central Vein Occlusion Study Group. A randomized clinical trial of early panretinal photocoagulation for ischemic central vein occlusion. The Central Vein Occlusion Study Group N report. Ophthalmology. 1995;102:1434–1444.
4. Fekrat S, Goldberg MF, Finkelstein D. Laser-induced chorioretinal venous anastomosis for nonischemic central or branch retinal vein occlusion. Arch Ophthalmol. 1998;116:43-52.
5. Scott IU, Ip MS. It's time for a clinical trial to investigate intravitreal triamcinolone for macular edema due to retinal vein occlusion: the SCORE study. Arch Ophthalmol. 2005;123:581-582.
6. Pe'er J, Folberg R, Itin A, Gnessin H, Hemo I, Keshet E. Vascular endothelial growth factor upregulation in human central retinal vein occlusion. Ophthalmology. 1998;105;412-416.
7. Boyd SR, Zachary I, Chakravarthy U, et al. Correlation of increased vascular endothelial growth factor with neovascularization and permeability in ischemic central vein occlusion. Arch Ophthalmol. 2002;120:1644-1650.
8. Aiello LP, Avery RL, Arrigg PG, et al. Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med. 1994;331:1519-1520.
9. Augustin AJ, Keller A, Koch F, Jurklies B, Dick B. Effect of retinal coagulation status on oxidative metabolite and VEGF in 208 patients with proliferative diabetic retinopathy. Klin Monatsbl Augenheilkd. 2001;218:89-94.
10. Brown DM. Rubeosis anti-VEGF (RAVE) trial for ischemic CRVO: 1-year data. Paper presented at: 40th Annual Scientific Meeting of the Retina Society; September 2007; Boston, MA.