Shilpa Desai, MD and Caroline R. Baumal, MD

Retinal vein occlusion (RVO) is the second most common retinal vascular disorder following diabetic retinopathy.¹ The incidence of RVO (branch [BRVO] and central combined [CRVO]) ranges from 0.8% to 1.8%.² RVO has an age- and sex-standardized prevalence of 4.42/1000 for BRVO and 0.80/1000 for CRVO. The prevalence of CRVO increases with age with 0.27 occurring/1000 in those from 0 to 49 years and 5.44 occurring/1000 in those 80 years and older.³ Worldwide, an estimated 16.4 million adults are affected by RVO, with 2.5 million affected by CRVO and 13.9 million affected by BRVO.^4,5

The etiology of RVO can be external compression of the vein by an atherosclerotic artery, intraluminal thrombosis, or inflammation of the vein.^6,7 Systemic risk factors include hypercoagulable state (hazard ratio [HR] 2.92), history of stroke (HR 1.4), hypertension-related end-organ damage (HR 1.92), and diabetes mellitus-related end-organ damage (HR 1.53). Cigarette smoking and both angle closure and primary open-angle glaucoma have also been identified as risk factors for vein occlusion.⁸ Furthermore, in the United States (US), African Americans have a 58% increased risk of CRVO compared to Caucasians (HR 1.58) and women have a 25% decreased risk compared to men (HR 0.75) after adjustment for confounders.⁵ After developing CRVO, there is a 10% chance/year of RVO occurring in the fellow eye.⁹ There is also an increased risk of mortality because of its statistical association with comorbid diabetes and/or cardiovascular disease.¹⁰

The majority of individuals over age 65 have one or more of the following risk factors for RVO:

• Smoking
  
• Hypertension
  
• Diabetes mellitus
  
• Increased age
  
• Elevated lipids/cholesterol

In these patients, optimizing systemic control of these risk factors is part of the therapeutic protocol and no additional systemic workup is required. In younger individuals, studies have shown that systemic hypertension, hyperlipidemia, and increased body mass index are important risk factors for vein occlusions.11 In young patients without these risk factors, however, it is important to assess for hypercoagulable states or inflammatory disease as the underlying cause.

Medicare reimbursement accounting for resource use (fluorescein angiography [FA], optical coherence tomography, intravitreal injection, panretinal laser, vitrectomy) and direct medical costs for CRVO were $11,587 at 1 year and $31,585 at 3 years.¹² These costs excluded calculations for treatment of systemic hypertension and glaucoma.¹³ Using the Beaver Dam Eye Study estimates of prevalence, the total cost to the US Medicare population is expected to be $1.3 billion annually for CRVO.¹² The cost/line of vision saved using therapies for cystoid macular edema (CME) range from $704 to $7611 for intravitreal triamcinolone and ranibizumab, respectively.¹⁴ Patients treated with ranibizumab, the original anti-vascular endothelial growth factor (VEGF) agent tested in clinical trials, have shown improvements in vision-related quality of life at 1 and 6 months after anti-VEGF therapy.¹⁵

Diagnosis of Retinal Vein Occlusion

RVO presents with painless vision loss associated with retinal hemorrhages and retinal venous tortuosity in the quadrant(s) of the affected vein. Additional findings can include cotton wool spots, optic nerve edema, CME, lipid exudation, and concurrent signs of hypertensive retinopathy¹⁶ (Figure 11-1). Optic nerve collateral vessels, neovascularization (NV) of the iris, and sheathed veins may develop after the acute event. RVO is classified as central, hemi-, or branch depending on the distribution of the occlusion. Presentation is typically acute, but may be subacute depending on patient symptoms and the quadrant(s) involved. On imaging, early-phase FA in acute RVO reveals a delay in the arteriovenous transit time (typically longer than 20 seconds); the late phase shows areas of capillary nonperfusion. The ischemic type of CRVO is defined as at least 10 disc areas of retinal capillary nonperfusion on FA,17 and is associated with more retinal venous engorgement, fluorescein leakage, retinal capillary dilation, and capillary nonperfusion than nonischemic CRVO. It accounts for 20% of acute presentations and is associated with worse baseline acuity, afferent papillary defect, and poorer visual outcomes compared to their nonischemic counterparts even after treatment.^10,18–20 FA may not be able to differentiate between ischemic and nonischemic RVO subtypes in up to 40% of cases because of masking of retinal vascular details by extensive retinal hemorrhages, media opacities (vitreous hemorrhage, lens changes), and limited assessment of the peripheral retina.²¹ Furthermore, nonischemic RVO may convert to the ischemic subtype in one-third of cases.¹⁰

Vision loss in RVO can be secondary to ischemia of the macula, CME, ocular NV, or secondary glaucoma.^22,23 CME in RVO is characterized by fluid collection in the intercellular spaces within the outer plexiform layer of the retina. Pockets of fluid may also develop in the subretinal space secondary to breakdown of the capillary endothelium blood-retinal barrier and leakage of fluid from the vasculature.^24,25 In nonischemic CRVO, macular edema is strongly associated with vision loss. In ischemic CRVO, however, visual acuity may not be correlated with macular edema due to ischemia of the retinal ganglion cells (Figure 11-2). In addition, the underlying retinal ischemia in ischemic RVO leads to a large quantity of VEGF being released into the eye.²⁶ This excessively high level of VEGF leads to the proliferation of blood vessels causing NV of the retina and/or optic disc in BRVO and iris in CRVO. The risk of developing ischemia of the macula, macular edema, ocular NV, and secondary glaucoma is 16% in eyes with more than 5.5 disc areas of nonperfusion in contrast to 4.0% of eyes with less than 5.5 disc areas of nonperfusion.²⁷

Natural history studies of RVO reveal that some patients will improve spontaneously without therapy10 (Figure 11-3). In 714 CRVO eyes followed without treatment, 65% of patients with vision better than or equal to 20/40 maintained this level of vision. The degree of improvement varied based on the initial acuity. In patients with vision better than 20/200, 56% showed improvement without treatment. In patients with vision worse than 20/200, only 20% showed improvement without treatment. Overall, patients with nonischemic CRVO spontaneously improve more frequently than those with ischemic CRVO.^23,28–30 In hemi-retinal or BRVO, spontaneous improvement is more likely when the macula is not involved.

Treatment for Central Retina Vein Occlusion and Branch Retinal Vein Occlusion

It is hypothesized that acute vein occlusions also block arterial flow, leading to ischemia of the retina and release of VEGF, which then causes CME and NV of the retina (more common in BRVO) and iris (more common in CRVO). At some point after the acute event, the vein and thrombus become infiltrated with lymphocytes, causing a secondary inflammatory response.⁶ With these concepts in mind, therapies to treat RVO have centered on reducing the release of VEGF (laser, anti-VEGF therapy) or reducing venous inflammation to improve blood flow (corticosteroid therapy). In all cases of RVO, treatment must involve management of the underlying systemic vascular risk factors in conjunction with the primary care provider. Low-dose aspirin (ASA) use is controversial as ASA has been shown to decrease the risk of cerebrovascular accident, but in RVO studies, ASA use was correlated with decreased visual recovery.³¹

Laser

Macular grid laser was introduced as a treatment for RVO as a method to decrease macular edema. The mechanism of grid laser-induced resolution of CME is unknown. Proposed mechanisms include laser-induced scarring of photoreceptors leading to better oxygen perfusion to the inner retina, enhanced proliferation of retinal pigment epithelial and endothelial cells restoring the blood-retina barrier, and laser-induced production of cytokines, which antagonize the effects of VEGF. Initial studies showed macular grid laser photocoagulation improved visual acuity in eyes with CME secondary to BRVO (BVOS)32 but not CRVO (CVOS).^3,10 This may be due to the lower prevalence of ischemia in eyes with BRVO, making them more amenable to visual improvement with grid laser treatment as compared to CRVO eyes. As such, macular grid laser was the mainstay of treatment for CME in BRVO until anti-VEGF agents became available, and is still used in certain situations today.

Panretinal laser photocoagulation (PRP) is used to treat retinal or iris NV secondary to RVO. PRP is thought to work by destroying ischemic retina and reducing the overall amount of VEGF released into the eye. The BVOS study demonstrated that PRP reduces development of retinal NV and vitreous hemorrhage; however, it does not improve visual outcomes.^33,34 Hayreh et al³³ found that in CRVO, for those eyes without retinal NV, only 12% of PRP-treated eyes developed vitreous hemorrhage compared to 22% in the nontreated eyes. For eyes with NV, 29% developed vitreous hemorrhage in the treatment group vs 60% in the nontreated group. Side effects of PRP include peripheral vision loss and decreased night vision.

Corticosteroids

Intravitreal injection with corticosteroids has also been demonstrated to be efficacious in the treatment of CME secondary to RVO. It is thought to work by down-regulating inflammatory cytokines and reducing dysregulation of endothelial tight junction proteins that can lead to CME. Triamcinolone acetonide has also been shown to decrease the amount of VEGF measured in the vitreous.³⁵ The Standard Care vs COrticosteroid for REtinal Vein Occlusion (SCORE) trial evaluated treatment of intravitreal triamcinolone acetonide (IVTA) in RVO and concluded IVTA was better than observation in CRVO but recommended grid laser should still be used in BRVO as there was no difference in percentage of patients who gained 3 or more lines of vision and to avoid the risks of cataract and glaucoma associated with corticosteroid therapy.³⁶ Similarly, Avitabile et al³⁷ found that IVTA improved central macular thickness and vision in the CRVO group but not the BRVO group when compared with grid laser. In addition, the effectiveness of triamcinolone appears to wane by 3 months. Numerous studies have shown that macular edema returns and vision declines at 3 or 6 months post-treatment.^38–43

Extended-release steroid therapy is also used in RVO for the treatment of CME. The extended-release formulation aims to provide a more uniform dose of corticosteroid to the eye and lengthen the time between treatments. The GENEVA trial compared a sustained-release dexamethasone implant (Ozurdex 0.7 mg or 0.35 mg) with sham treatment for CME secondary to BRVO and CRVO. For the CRVO group only, the time to improve 15 letters was significantly faster in the 2 Ozurdex treatment groups. There was a faster improvement in visual acuity and mean decrease in central subfield retinal thickness (CST) through the 90-day point, but this waned by the 180-day timepoint, at which point there was no significant difference in both the visual acuity and mean decrease in CST between treated and nontreated eyes. Ocular hypertension (treated successfully in almost all cases) was noted in 4 and 3.9% in the dexamethasone implant groups, respectively, as compared to 0.7% in the sham group.^44,45 The Retisert implant (Bausch + Lomb) and Iluvien (Alimera Sciences) are alternate intravitreal implants that contain extended-release fluocinolone acetonide. Both have been shown to improve macular edema in diabetes and may have applications in RVO in the future.

Adverse events related to any corticosteroid use include cataract and elevated intraocular pressure (IOP). The risk of these events may be specific to dose and vehicle-related effects. The SCORE study showed a dose-dependent increase in IOP with steroid use (35% of patients in the 4 mg group, 20% of patients in the 1 mg group, and 8% of patients in the observation group needed medication for increased IOP). In addition, the higher the steroid dose, the higher the rate of necessary cataract surgery (21% in the 4 mg group, 3% in the 1 mg group, and 0% in the observation group).22 Suspension formulations of intravitreal corticosteroid (such as IVTA) include the additional risk of sterile endophthalmitis. On the other hand, solid sustained-release implants may migrate into the anterior chamber in aphakic eyes or eyes with an open posterior capsule, causing corneal edema and necessitating removal.

Alternative Treatments

Numerous other treatments of historical interest have been used in an attempt to directly counteract the primary event of the occlusion of the retinal vein rather than treat the secondary consequences of CME and NV. Isovolemic hemodilution was attempted to dilute the blood enough to bypass the RVO; however, this practice was abandoned because of significant systemic risk factors.^46–48 Radial optic neurotomy and pars plana sheathotomy have also been attempted to increase the potential space for venous flow at the level of the occlusion. Randomized studies demonstrated no visual benefit from these interventions.^49,50 Both practices were abandoned because they were considered high risk given the risk of retinal detachment and cataract.^51,52 Laser induction of a chorioretinal anastamosis was attempted to anatomically reestablish adequate venous drainage in CRVO with CME. The Central Retinal Bypass Study found that an anastomosis was created in 76% of treated eyes. The anastomosis group had a mean improvement of 11.7 letters as compared to loss of 8.1 letters in the control group at 18 months post-intervention. However, this therapy was not widely accepted because of procedure-related complications such as fibrovascular traction, macular traction, and vitreous hemorrhage.⁵³ Finally, retinal endovascular lysis was also attempted for CRVO. In this procedure, a fibrinolytic agent is injected into a cannulated retinal vein after pars plana vitrectomy to allow for blood flow through the vein. In the nonrandomized study of this technique in CRVO, there was no improvement in vision at one year and the complication rate was high, including neovascular glaucoma leading to phthisis, retinal detachment, and cataract.⁵⁴

Anti-Vascular Endothelial Growth Factor Therapy

VEGF is produced by Müller cells, endothelial cells, astrocytes, retinal pigment epithelial cells, and ganglion cells in response to decreased oxygen and causes a time- and dose-dependent breakdown in the blood-retinal barrier.²⁵ Studies have shown that vitreous VEGF levels are higher in patients with any disorder that produces ischemia to the retina, including eyes with RVO.⁵⁵ VEGF levels have been correlated with the severity of retinal ischemia and macular edema in RVO.^56,57 Anti-VEGF therapy downregulates the level of VEGF in the vitreous, stabilizing the blood-retinal barrier and reducing intraretinal fluid exudation (Figure 11-4). Anti-VEGF treatment for RVO is typically monthly, treat and extend, or pro re nata (PRN) based on patient symptoms, exam, and physician preference. Commonly used agents include ranibizumab, bevacizumab, and aflibercept.

Stay updated, free articles. Join our Telegram channel

Join