Chapter 115 – Venous Obstructive Disease of the Retina
MICHAEL G. MORLEY
JEFFREY S. HEIER
Central Retinal Vein Obstruction
Branch Retinal Vein Obstruction
• Obstruction of the central retinal vein at the lamina cribrosa.
• Retinal hemorrhages in all four quadrants.
• Dilated, tortuous veins in all four quadrants.
• Optic disc edema.
• Macular edema.
• Cotton-wool spots.
• Capillary nonperfusion.
• Neovascularization of the iris, retina, or optic disc.
• Neovascular glaucoma.
• Optic disc venous–venous collateral vessels (opticociliary shunt vessels).
• Exudative retinal detachment in severe cases.
• Obstruction of a branch retinal vein.
• Retinal hemorrhages in the distribution of the obstructed branch retinal vein.
• Macular edema.
• Retinal neovascularization.
• Vitreous hemorrhage.
• Dilated, tortuous retinal vein.
• Capillary nonperfusion.
• Cotton-wool spots.
• Venous–venous retinal collateral vessels.
• Sheathing of vessel.
• Lipid exudates.
• Microvascular changes including microaneurysms and collateral vessels.
• Pigmentary macular disturbances.
• Subretinal fibrosis.
CENTRAL RETINAL VEIN OBSTRUCTION
Venous obstructive disease of the retina is a relatively common retinal vascular disorder, second only to diabetic retinopathy in incidence. It typically affects patients who are 50 years of age or older. Usually, retinal vein obstructions are recognized easily and treatment options have been investigated thoroughly using large, multicenter, randomized clinical trials.
Retinal vein obstructions are classified according to whether the central retinal vein or one of its branches is obstructed. Central retinal vein obstruction and branch retinal vein obstruction differ with respect to pathophysiology, underlying systemic associations, average age of onset, clinical course, and therapy.
Central retinal vein obstructions can be divided further into ischemic and nonischemic varieties. This distinction among central retinal vein obstructions, although somewhat arbitrary, is important because up to two thirds of patients who have the ischemic variety develop iris neovascularization and neovascular glaucoma.
EPIDEMIOLOGY AND PATHOGENESIS
Central retinal vein obstruction is found most commonly in individuals over 50 years old.   Diabetes mellitus, systemic arterial hypertension, and atherosclerotic cardiovascular disease are the most frequently associated underlying medical diseases; however, their direct relationship to pathogenesis remains speculative.     Completely normal medical and laboratory evaluation results are found in about one fourth of patients.  A significant inverse association with central retinal vein obstruction, which represents decreasing risk, is present with alcohol consumption, education, physical activity and, in women, exogenous estrogen use.
Open-angle glaucoma is a relatively common finding in patients who have central retinal vein obstruction. Patients who have a history of glaucoma are about 5 times more likely to have central retinal vein obstruction than those who do not, presumably because of structural alterations of the lamina cribrosa induced by elevated intraocular pressure. Acute angle-closure glaucoma may precipitate central retinal vein obstruction.
The precise pathogenesis of central retinal vein obstruction remains obscure. The obstruction is believed to be the result of a thrombus in the central retinal vein at, or posterior to, the lamina cribrosa. Arteriosclerosis of the neighboring central retinal artery that causes turbulent venous flow and then endothelial damage often is implicated. Also, endothelial cell proliferation has been suggested. An alternative theory is that thrombosis of the central retinal vein is an end-stage phenomenon, induced by a variety of primary lesions. Such lesions could include compressive
or inflammatory optic nerve or orbital problems, structural abnormalities in the lamina cribrosa, or hemodynamic changes.
Because the retinal venous circulation represents a relatively high-resistance, low-flow system it is particularly sensitive to hematological factors. Along with an elevated erythrocyte sedimentation rate and antithrombin III levels, other studies indicate that an elevated hematocrit level, elevated homocysteine level, elevated fibrinogen level, increased blood viscosity, the presence of a lupus anticoagulant or another antiphospholipid antibody, and a deficiency in activated protein C may be associated with retinal venous disease.    Whether these hematological factors alone can initiate a central retinal vein obstruction or whether their role is to function as cofactors remains unknown.
Both types of central retinal vein obstruction, ischemic and nonischemic, share similar findings—dilated, tortuous retinal veins and retinal hemorrhages in all four quadrants. The distinction between the two varieties is important, because it assists the clinician in the following:
• Prediction of the risk of subsequent ocular neovascularization
• Identification of patients who have poorer visual prognosis
• Determination of the likelihood of spontaneous visual improvement
• Decision as to appropriate follow-up intervals
The distinction between the two types of vein obstructions remains somewhat arbitrary and is based on the total area of nonperfusion on fluorescein angiography. Most investigators accept that nonischemic and ischemic central retinal vein obstructions represent varying severity of the same underlying disease continuum. Other investigators suggest, however, that these are two distinct clinical entities with different pathogenesis. The ischemic variety is associated with concurrent, severe retinal arterial disease, while the milder, nonischemic type results from a thrombosis located more distally, behind the lamina cribrosa.
Nonischemic Central Retinal Vein Obstruction
Alternative names include partial, incomplete, imminent, threatened, incipient, or impending vein obstruction, as well as venous stasis retinopathy. Of the patients who have central retinal vein obstruction, 75–80% can be classified as having this milder form. Patients usually have mild to moderate decreased visual acuity, although this can vary from normal to as poor as difficulty with finger counting. Intermittent blurring or transient visual obscuration also may be a complaint. Pain is rare.
Figure 115-1 Nonischemic central retinal vein obstruction. Fundus view of diffuse retinal hemorrhages, optic nerve head edema, dilated and tortuous veins, and a cotton-wool spot.
Pupillary testing rarely reveals an afferent defect which, if present, is only slight. Ophthalmoscopy reveals a variable number of dot and flame retinal hemorrhages, present in all four quadrants ( Fig. 115-1 ). Optic nerve head swelling is common, and engorgement and tortuosity of the retinal veins are characteristic. Cotton-wool spots, if present, are few in number and located posteriorly. When vision is decreased, this is usually the result of macular hemorrhage or edema, which may be in the form of cystoid macular edema, diffuse macular thickening, or both.
Neovascularization of either the anterior or posterior segment is rare in a true nonischemic central retinal vein obstruction (less than 2% incidence), although conversion from an initially nonischemic vein obstruction to the ischemic variety is fairly common. The Central Vein Occlusion Study Group noted that 34% of nonischemic central retinal vein occlusions (CRVOs) progressed to become ischemic within 3 years, and 15% of the study group converted within the first 4 months.
Many or all of the pathological retinal findings may resolve over the 6–12 months following diagnosis. Retinal hemorrhages can resolve completely. The optic nerve may appear normal, but opticociliary collateral vessels are common. Macular edema also may resolve, to leave a normal appearance. However, persistent cystoid macular edema can linger and result in permanent visual loss, often leading to pigmentary changes, epiretinal membrane formation, or subretinal fibrosis.
Ischemic Central Retinal Vein Obstruction
Ischemic central retinal vein obstructions are referred to as severe, complete, or total vein obstruction, and hemorrhagic retinopathy ; they account for 20–25% of all central retinal vein obstructions. Acute, markedly decreased visual acuity is the usual initial complaint. Vision usually ranges from 20/200 (6/60) to hand-motion acuity. A prominent afferent pupillary defect is typical. Pain at the time of evaluation may occur if neovascular glaucoma has developed.
The ophthalmoscopic picture of an ischemic central retinal vein obstruction may be confused with other entities, but rarely. It is characterized by extensive retinal hemorrhages in all four quadrants, most notably centered in the posterior pole ( Fig. 115-2 ). Hemorrhages can be so extensive that the retinal and choroidal details are obscured. Bleeding may break through the internal limiting membrane, which results in vitreous hemorrhage. The optic disc usually is edematous, and the retinal veins are markedly engorged and tortuous. Cotton-wool spots are usually present and may be numerous. Macular edema is often severe but may be obscured by hemorrhage. Massive lipid exudation in the
Figure 115-2 Ischemic central retinal vein obstruction. Fundus view of extensive retinal hemorrhages, venous dilation and tortuosity, and scattered cotton-wool spots.
macular region can occur, especially in patients who have elevated triglyceride levels. Exudative retinal detachment may develop and is associated with a poor visual prognosis. Secondary, non-neovascular angle-closure glaucoma may occur.
The incidence of anterior segment neovascularization in ischemic central retinal vein obstruction is 60% or higher and has been documented as early as 9 weeks after onset.  Neovascularization of the angle and neovascular glaucoma may occur within 3 months of disease onset (90-day glaucoma), and it can result in intractably elevated pressure. Neovascularization of the optic disc and retinal neovascularization may be seen as well, but they are less common. As with nonischemic central retinal vein obstruction, the findings may decrease or resolve 6–12 months after diagnosis.
During the resolution phase, the optic nerve shows pallor and opticociliary collateral vessels more often than it does in the mild form of central retinal vein obstruction. Permanent macular changes can develop that include pigmentary changes, epiretinal membrane formation, and subretinal fibrosis that resembles disciform scarring. Macular ischemia may be present, as well.
HEMICENTRAL RETINAL VEIN OBSTRUCTION.
In about 20% of eyes, the central retinal vein enters the optic nerve as two separate branches, the superior and inferior, prior to merging as a single trunk posterior to the lamina cribrosa. In these eyes, obstruction of one of the dual trunks within the substance of the optic nerve results in a hemicentral retinal vein obstruction. Although only one half of the retina is involved, these obstructions act more like central retinal vein obstructions than a branch retinal vein obstruction in terms of visual outcome, risk of neovascularization, and response to laser treatment.
Papillophlebitis or Optic Disc Vasculitis
Some mild central retinal vein obstructions in patients younger than 50 years have been referred to as papillophlebitis or optic disc vasculitis—terms that suggest a benign course. An inflammatory optic neuritis or vasculitis is hypothesized as the cause. These eyes tend to have optic disc edema out of proportion to the retinal findings, cotton-wool spots that ring the optic disc, and occasionally cilioretinal artery obstructions or even partial central retinal artery obstructions. Although spontaneous improvement is common, the course is not always benign. Up to 30% of these patients may develop the ischemic type of occlusion, a final visual acuity of 20/200 (6/60) or worse occurs in nearly 40%, and neovascular glaucoma has been reported.
DIAGNOSIS AND ANCILLARY TESTING
The diagnosis of an ischemic central retinal vein obstruction is based on the characteristic fundus findings:
• Widespread retinal hemorrhages
• Retinal venous engorgement and tortuosity
• Cotton-wool spots
• Macular edema
• Optic disc edema
Rarely is this clinical picture confused with other entities. However, the clinical picture of a nonischemic central retinal vein obstruction can be far more subtle. Although retinal hemorrhages usually are present in all four quadrants, they may be scant. If the eye is observed several months after disease onset, the hemorrhages may have resolved. Cotton-wool spots, optic nerve edema, and macular edema tend to be absent. Venous engorgement and tortuosity, which may be mild, are usually present.
Fluorescein angiography is the most useful ancillary test for the evaluation of the two most serious, debilitating and, unfortunately, common complications of central retinal vein obstruction—anterior segment
Figure 115-3 Ischemic central retinal vein obstruction. Fluorescein angiography reveals marked hypofluorescence secondary to widespread capillary nonperfusion. The venous system shows marked dilation with focal areas of constriction, and the vessel walls stain in areas of ischemia.
neovascularization and macular edema. Studies suggest that eyes with 10 disc areas or greater of nonperfusion noted on fluorescein angiography are at increased risk for the development of anterior segment neovascularization and, therefore, should be classified as ischemic.  The Central Vein Occlusion Study found the greatest risk was in patients with worse than 20/200 (6/60) visual acuity or 30 or more disc areas of nonperfusion.  Electroretinography is used occasionally to help determine the prognosis of a CRVO.  
Fluorescein angiography in ischemic central retinal vein obstruction may show marked hypofluorescence ( Fig. 115-3 ), which may be secondary to blockage from extensive hemorrhages or to retinal capillary nonperfusion. When extensive hemorrhages are present, little information is gained from the angiogram. However, as the hemorrhages clear over several months, the degree of capillary nonperfusion may become apparent. Most eyes (80%) that have this degree of hemorrhage eventually are classified as ischemic.
Macular edema is the most common cause of visual loss in central retinal vein obstruction. It is present almost universally in ischemic cases and frequently is severe. It may manifest as large cystoid spaces or diffuse leakage on fluorescein angiography. Macular edema may be obscured by hemorrhage, but as the hemorrhage and edema resolve, macular ischemia may become apparent. Angiography also reveals optic nerve head leakage and perivenous staining. In the late stages of the disease, generalized extensive retinal capillary nonperfusion, arteriovenous collateral vessels, and microaneurysms are seen. The macular region shows persistent edema or pigmentary degeneration.
With a nonischemic central retinal vein obstruction, fluorescein angiography reveals staining along the retinal veins, microaneurysms, and dilated optic nerve head capillaries. Retinal capillary nonperfusion ( Fig. 115-4 ) is minimal or absent. As the nonischemic central retinal vein obstruction resolves, angiography may become normal. If macular edema persists, or if pigmentary changes occur, these become evident.
The Central Vein Occlusion Study Group reported that 37% of ischemic central retinal vein obstructions demonstrated anterior segment (iris or angle or both) neovascularization at or before the 4-month follow-up. Although the results of fluorescein angiography help to differentiate patients at high risk for the development of neovascularization, visual acuity alone is a more powerful, less expensive, and less invasive measurement by which to determine the prognosis and appropriate follow-up.
A general medical evaluation, to include extensive medical history and physical examination with blood pressure evaluation, is recommended ( Box 115-1 ). Laboratory evaluation may include a complete blood count, glucose tolerance test, lipid profile, serum protein electrophoresis, chemistry profile, and syphilis serology. Additional testing, based upon the above findings,
Figure 115-4 Nonischemic central retinal vein obstruction. Fluorescein angiography shows marked venous dilation and tortuosity, optic nerve head edema, and staining of vessel walls. Capillary nonperfusion is absent.
Medical and Ophthalmic Work-Up for Central Retinal Vein Obstruction and Branch Retinal Vein Obstruction
CENTRAL RETINAL VEIN OBSTRUCTION
Complete history and physical examination
Complete ophthalmic examination
Gonioscopy to look for iris and/or angle neovascularization
Complete blood count
Partial thromboplastin time
Serum protein electrophoresis
Erythrocyte sedimentation rate
BRANCH RETINAL VEIN OBSTRUCTION
Complete history and physical examination
Complete ophthalmic examination
may be necessary. If a history of systemic clotting diathesis exists, further hematological tests such as lupus anticoagulant level, anticardiolipin antibody, and protein S and protein C levels should be considered. Diagnosis and treatment of an associated disease is not expected to improve the visual outcome in the affected eye, but it may help to prevent subsequent obstruction in the fellow eye.
As stated above, rarely is the full-blown picture of an ischemic central retinal vein obstruction confused with other disease entities. However, nonischemic or long-standing central retinal vein obstructions can appear similar to the retinopathy of carotid occlusive disease—the ocular ischemic syndrome(s). A great deal of confusion existed in the past over these two entities, not only because of their similar clinical pictures, but also because each has been referred to in the literature as venous stasis retinopathy.  Both conditions are associated with blurred vision, and both may have transient visual loss. Blurring of vision when a darker room is entered after being in a brighter area is suggestive of carotid artery disease. Although disc edema always is present in ischemic central retinal vein obstruction, and may be present in nonischemic central retinal vein obstruction, it is quite rare in carotid occlusive disease. Although the veins are engorged in both diseases, they are generally not tortuous in the ocular ischemic syndrome. The retinal hemorrhages seen in carotid disease tend to localize to the midperiphery, instead of the posterior pole as seen in central retinal vein obstruction.
Hyperviscosity syndromes may produce a bilateral retinopathy similar to central retinal vein obstruction and may, in fact, induce a true central retinal vein obstruction with thrombus formation. Simultaneous bilateral disease is an unusual finding in central retinal vein obstructions but occurs more commonly in hypercoagulable and hyperviscous states. Diseases such as sickle cell disease, polycythemia vera, leukemia, and multiple myeloma are but a few of the possibilities. When a patient seeks treatment for bilateral central retinal vein obstructions, especially simultaneous, the medical and laboratory evaluation should include a search for evidence of hyperviscous and hypercoagulable syndromes. Improvement in the affected eye is possible when a hyperviscosity syndrome is responsible and plasmapheresis is performed. Severe anemia with thrombocytopenia can masquerade as a central retinal vein obstruction, and it is differentiated from a central retinal vein obstruction by a complete blood count with platelets. In addition, acute hypertensive retinopathy with disc edema may resemble bilateral central retinal vein obstruction.
Central retinal vein obstruction has been associated with systemic vascular disease such as hypertension, diabetes mellitus, and cardiovascular disease; blood dyscrasias such as polycythemia vera, lymphoma, and leukemia; paraproteinemias and dysproteinemias including multiple myeloma and cryoglobulinemia; vasculitis of syphilis and sarcoidosis; and autoimmune disease such as systemic lupus erythematosus.  Blood dyscrasias and dysproteinemias result in hyperviscosity syndromes, which may appear similar to central retinal vein obstruction but possibly represent curable disease (as discussed under differential diagnosis above). Oral contraceptive use in women may be associated with both thromboembolic disease and central retinal vein obstruction.
Green et al. evaluated histological sections of 29 eyes in 28 patients who had central retinal vein obstruction. All 29 eyes had the formation of a fresh or recanalized thrombus at or just posterior to the lamina cribrosa. Within the thrombi, a mild lymphocytic infiltration with prominent endothelial cells was seen. Loss of the inner retinal layers consistent with inner retinal ischemia was a common finding.
Alterations in blood flow, hyperviscosity, and vessel wall abnormalities may produce central retinal vein obstructions by enabling a thrombus of the central retinal vein to form. Local factors can predispose to central retinal vein obstruction. Glaucoma has been associated with central retinal vein obstruction. It has been hypothesized that glaucoma causes stretching and compression of the lamina cribrosa, which results in vessel abnormalities, increased resistance to flow and, ultimately, thrombosis. 
No treatment has been proven to reverse the pathology seen in central retinal vein obstruction. Aspirin; systemic anticoagulation with coumarin, heparin, and alteplase; local anticoagulation with intravitreal alteplase; corticosteroids; anti-inflammatory agents; isovolemic hemodilution; plasmapheresis; and optic nerve sheath decompression all have been advocated but without definitive proof of efficacy. Certain complications of central retinal
Treatment Guidelines for Patients Who Have Central Retinal Vein Obstruction
NO PROVED EFFECTIVE TREATMENT
Panretinal photocoagulation if intraocular neovascularization present
Lower intraocular pressure if elevated
Treat underlying medical conditions
Macular edema generally does not respond to grid laser
vein obstruction may be preventable or reversible, however ( Box 115-2 ).
Neovascular glaucoma is a devastating complication of ischemic central retinal vein obstruction. Intractable glaucoma, blindness, and pain that culminates in enucleation can occur. The Central Vein Occlusion Study Group determined whether prophylactic panretinal photocoagulation (PRP) was an effective method with which to prevent the development of iris neovascularization or angle neovascularization in patients who had ischemic central retinal vein obstruction, or whether it was more appropriate to apply PRP after the development of anterior segment neovascularization.  The study found that prophylactically treated ischemic eyes developed iris neovascularization less frequently than ischemic eyes that were followed (20% in the early treatment group versus 35% in the no-early-treatment group), although the difference was not statistically significant. However, PRP is more likely to result in prompt regression of neovascularization of the iris in the previously untreated group versus the prophylactically treated group (56% versus 22%, respectively, after 1 month). As a result, for ischemic central retinal vein obstructions, frequent follow-up examinations during the early months and prompt PRP if iris neovascularization develops is the recommended treatment strategy.
Treatment should be applied in all four quadrants to give medium-white burns of diameter 400–500?µm (a total of 1000–2000 burns). Identification of early iris neovascularization at the pupillary border is critical—examination of the undilated pupil is recommended. Routine gonioscopy also is suggested, because angle neovascularization can occur without iris neovascularization.
Macular edema and subsequent permanent macular dysfunction occur in virtually all patients with ischemic central retinal vein obstruction, and in many patients with nonischemic central retinal vein obstruction. The Central Vein Occlusion Study evaluated the efficacy of macular grid photocoagulation in patients with central retinal vein obstruction and macular edema. Patients with both ischemic and nonischemic central retinal vein obstruction were studied. Although macular grid laser treatment conclusively reduced angiographic macular edema, the study did not find a difference in visual acuity between the treated and untreated eyes at any stage of the follow-up period. As a result, currently it is not recommended that macular grid photocoagulation be employed in the setting of central retinal vein obstruction. Intravitreal triamcinolone is being studied as a treatment for cystoid macular edema.
COURSE AND OUTCOME
The prognosis for visual recovery is highly dependent upon the subtype of central retinal vein obstruction. In general, the visual prognosis can be predicted from the visual acuity during evaluation. Patients who have nonischemic central retinal vein
Follow-Up for Patients Who Have Central Retinal Vein Obstruction
PRESENTING VISUAL ACUITY OF 20/40 (6/12) OR BETTER
Examinations every 1–2 months for 6 months after diagnosis
Annual examinations as the patient’s condition stabilizes
PRESENTING VISUAL ACUITY BETWEEN 20/50 (6/15) AND 20/200 (6/60)
Examinations monthly to bimonthly (at the physician’s discretion, based on which end of the spectrum the visual acuity lies) for the first 6 months after diagnosis
Examinations every 6 months to yearly afterward
PRESENTING VISUAL ACUITY OF 20/200 (6/60)
Examinations every month for the initial 6 months
Then every 2 months until 8 months after presentation
Then every 4 months until 2 years after presentation
obstructions may experience a complete recovery of vision, although this occurs in less than 10% of cases. Although patients with nonischemic disease may retain acuity of 20/60 (6/18) or better, as many as 50% deteriorate to levels of 20/200 (6/60) or worse.  Conversion of nonischemic to ischemic occlusions is seen in about one third of cases and typically occurs during the first 6–12 months after evaluation.    Of the patients who have ischemic central retinal vein obstructions, more than 90% have a final visual acuity of 20/200 (6/60) or worse.
As many as 7% of patients with central retinal vein obstruction develop a nonsimultaneous venous occlusion of the fellow eye within 2 years. Contralateral branch retinal vein and retinal arterial obstructions also may be seen. The risk of any vascular occlusion in the fellow eye is estimated to be 0.9% per year.
Based on the Central Vein Occlusion Study, the recommended follow-up examinations in patients who have central retinal vein obstruction are given in Box 115-3 .  
If visual acuity deteriorates to less than 20/200 (6/60) at any time during the disease course, the patient should be treated as a patient with a new CRVO who has an acuity of that level and assessed monthly.
BRANCH RETINAL VEIN OBSTRUCTION
Branch retinal vein obstruction is a common retinal vascular disorder of the elderly. Visual loss from a branch retinal vein occlusion usually is caused by macular edema, macular ischemia, or vitreous hemorrhage. In some patients, laser treatment can help stabilize or even improve vision.
EPIDEMIOLOGY AND PATHOGENESIS
Branch retinal vein obstructions occur approximately 3 times more commonly than central retinal vein obstructions. Men and women are affected equally, with the usual age of onset between 60 and 70 years. Most epidemiological and histopathological evidence implicates arteriolar disease as the underlying pathogenesis. Branch retinal vein obstruction almost always occurs at an arteriovenous crossing, where the artery and vein share a common adventitial sheath. The artery nearly always is anterior (innermost) to the vein. It is postulated that a rigid, arteriosclerotic artery compresses the retinal vein, which results in turbulent blood flow and endothelial damage, followed by thrombosis and obstruction of the vein. Most branch retinal vein obstructions occur superotemporally, probably because this is where the highest concentration of arteriovenous crossings lie.
Figure 115-5 Branch retinal vein obstruction. Fundus view of extensive retinal hemorrhages in segmental distribution of a superotemporal retinal vein. Dilated, tortuous veins, cotton-wool spots, and macular edema also can be seen.
Rarely, local ocular diseases, especially of an inflammatory nature, can result in a secondary branch retinal vein obstruction. This has been reported in diseases such as toxoplasmosis, Eales’ disease, Behçet’s syndrome, and ocular sarcoidosis. Also, macroaneurysms, Coats’ disease, retinal capillary hemangiomas, and optic disc drusen are linked to branch retinal vein obstruction. Glaucoma is also a risk factor for the development of branch retinal vein occlusion. Branch retinal vein occlusion is usually unilateral, with only 9% of patients having bilateral involvement.
Patients with branch retinal vein occlusion usually complain of sudden onset of blurred vision or a visual field defect. Retinal hemorrhages confined to the distribution of a retinal vein are characteristic for branch retinal vein obstruction ( Fig. 115-5 ). As a result of the distribution, the hemorrhages usually assume a triangular configuration with the apex at the site of blockage. Flame-shaped hemorrhages predominate. Mild obstructions are associated with a relatively small amount of hemorrhage. Complete obstructions result in extensive intraretinal hemorrhages, cotton-wool spot formation, and widespread capillary nonperfusion. If the macular region is involved, macular edema or hemorrhage occurs, which causes decreased visual acuity. Visual acuity may range from 20/20 (6/6) to counting fingers. If the macula is spared, a branch retinal vein obstruction may be asymptomatic, found only on routine examination of the fundus. Occasionally a partial branch retinal vein occlusion with little hemorrhage and edema may progress to a completely occluded vein, with an increase in hemorrhage and edema and a corresponding decrease in visual acuity. Retinal neovascularization occurs in approximately 20% of cases. The incidence of retinal neovascularization rises with increasing area of retinal nonperfusion. Retinal neovascularization typically develops within the first 6–12 months but may occur years later. Vitreous hemorrhage can ensue and may require vitrectomy. Anterior segment neovascularization rarely is seen in patients with branch retinal vein obstruction, unless other ischemic conditions co-exist (e.g., diabetes). With time, the dramatic picture of an acute branch retinal vein obstruction can become much more subtle. Hemorrhages fade with time so that the fundus can look almost normal. Collateral vessels and microvascular abnormalities develop to help drain the affected area. The collateral vessels often cross the horizontal raphe. Proximal to the site of blockage, the retinal vein may become sclerotic. The retinal
Figure 115-6 Branch retinal vein obstruction. Fluorescein angiography of the patient shown in Figure 115-4 . Marked hypofluorescence is present secondary to extensive hemorrhage in the distribution of a superotemporal branch vein. Vessel dilation, tortuosity, and staining is seen in the same distribution.
artery that feeds the affected zone may become narrowed and sheathed, as well. Microaneurysm formation occurs and lipid exudation may be present. Capillary nonperfusion is seen best in the later stages, after the hemorrhages have cleared. Epiretinal membrane and macular retinal pigment epithelial changes as a result of chronic cystoid macular edema sometimes are seen in the late phase of a branch retinal vein obstruction. Retinal detachment, either rhegmatogenous or tractional, is uncommon but may be seen. Exudative localized retinal detachment in the distribution of the branch retinal vein occlusion also is seen if there is severe ischemia.
DIAGNOSIS AND ANCILLARY TESTING
The diagnosis of an acute branch retinal vein obstruction is made by finding retinal hemorrhages in the distribution of an obstructed retinal vein. Usually the retinal vein is dilated and tortuous ( Fig. 115-6 ). The obstruction almost always occurs at an arteriovenous crossing site, with the artery anterior to the vein.
Fluorescein angiography is a helpful adjunct for both establishment of the diagnosis and guidance for the treatment of branch retinal vein obstruction. Arteriolar filling is usually normal, but venous filling in the affected vessel usually is delayed in the acute phase. Hypofluorescence caused by hemorrhage and capillary nonperfusion are common findings, and dilated, tortuous capillaries are seen. Collateral vessels may cross the horizontal raphe. The retinal vessels, particularly the vein walls, may stain with fluorescein. Neovascular fronds may leak fluorescein profusely. In contrast, collateral vessels do not leak fluorescein. Retinal vessels, particularly the vein walls, may stain with fluorescein, especially at the site of the occlusion. Macular edema, which is noted clinically but not angiographically, may indicate ischemia. Classic petaloid cystoid macular edema may involve the entire fovea or just several clock hours, depending on the distribution of the obstruction.
The differential diagnosis of branch retinal vein obstructions is shown in Box 115-4 . Hypertensive retinopathy with marked arteriovenous crossing changes and retinal hemorrhages may look like a branch retinal vein occlusion. A chronic branch retinal vein obstruction with telangiectatic capillaries may be confused with juxtafoveal retinal telangiectasia. Asymmetrical diabetic retinopathy can have a picture similar to a branch vein obstruction or, conversely, obscure the diagnosis of a branch vein obstruction.
Hypertension is the condition most commonly associated with branch retinal vein obstruction. The Eye Disease Case Control Study clearly demonstrated the important association of hypertension with vein obstructions. In that study, more than 50% of branch retinal vein obstructions were associated with hypertension. The study also found an association between vein obstructions and a history of cardiovascular disease, increased body mass index at 20 years of age, glaucoma, and higher serum levels of a2 -globulin. A reduced risk of branch retinal vein obstruction was found with alcohol consumption and increasing levels of high-density lipoprotein cholesterol levels.
A histopathological study of nine branch vein occlusions showed a fresh or recanalized thrombus at the site of the vein occlusion in all eyes. Ischemic atrophy of the retina was found in the distribution of the occlusion in most of the eyes. All eyes showed varied degrees of arteriosclerosis. No thrombus was noted in any of the arteries. Neovascularization of the disc and retina was noted in four eyes and cystoid macular edema was present in five.
The Branch Vein Occlusion Study represented a major advance in the understanding of the treatments for two of the most significant complications of branch vein occlusions, namely macular edema and neovascularization.   The study found that a grid pattern laser treatment helped to reduce macular edema and improved visual acuity. In patients who have 20/40 (6/12) or worse vision and macular edema on fluorescein angiography,
Differential Diagnosis of Branch Retinal Vein Obstruction
Ocular ischemic syndrome
Juxtafoveal retinal telangiectasia
Combined branch retinal artery and branch retinal vein occlusion
Figure 115-7 Branch retinal vein obstruction. Immediate posttreatment view of grid laser treatment for macular edema secondary to a branch retinal vein obstruction.
laser treatment improved the chances of a two-or-more–line improvement in vision on the Snellen chart when compared with untreated controls  (65% versus 37%). Because visual acuity and macular edema may improve spontaneously, patients were not treated with laser for at least 3 months after the development of the vein obstruction, to allow for spontaneous improvement. Also, treatment was delayed if the intraretinal hemorrhage was too dense to allow either photocoagulation or adequate evaluation with fluorescein angiography. Patients who had hemorrhage directly in the fovea were excluded. A fluorescein angiogram less than 1 month old was used to guide treatment. A grid pattern of laser was applied to the area of capillary leakage ( Fig. 115-7 ).
Photocoagulation did not extend closer than the edge of the foveal avascular zone, nor did it extend peripherally beyond the major vascular arcades. The eyes were reevaluated with fluorescein angiography 4 months after treatment, and additional photocoagulation was applied if the vision remained poor and macular edema persisted. Most patients required only one treatment. Typically, a 100?µm spot size is used and medium-white burns, each of 0.1-second duration, are applied to the area of edema. In both treated and controlled groups, patients who had hypertension tended to respond less favorably to laser treatment.
The Branch Vein Occlusion Study Group also evaluated the efficacy and timing of sectorial PRP for retinal neovascularization and vitreous hemorrhage.  In patients with neovascularization treated with laser, only 29% developed vitreous hemorrhage, versus 61% of those untreated. The data showed no advantage with treatment before neovascularization occurred, even if extensive capillary nonperfusion existed. If laser is applied to all nonperfused branch retinal vein obstructions, a large percentage of patients will be treated unnecessarily ( Boxes 115-5 and 115-6 ). Fluorescein angiography can be helpful in guiding laser treatment, because it will help define areas of capillary nonperfusion. A scatter pattern of laser is performed in the affected sector. Typically, 500?µm–sized medium-white burns are applied, extending from the arcade out to the periphery. Fill-in PRP may be applied if neovascularization progresses or if vitreous hemorrhage
Treatment Guidelines for Branch Retinal Vein Occlusion and Macular Edema
FOR MACULAR EDEMA, VISUAL ACUITY OF 20/40 (6/12) OR WORSE
Wait for clearance of retinal hemorrhage to allow adequate fluorescein angiography
Determine if decreased visual acuity is caused by macular edema (versus macular nonperfusion)
If macular edema explains visual loss, and no spontaneous improvement has occurred by 3 months, grid macular photocoagulation is recommended
If capillary nonperfusion explains decreased visual acuity, laser treatment is not advised
Treatment Guidelines for Branch Retinal Vein Occlusion and Neovascularization
Good quality fluorescein angiography is obtained after retinal hemorrhages have cleared sufficiently.
If more than five disc diameters of nonperfusion are present, the patient should be followed at 4-month intervals to seek the development of neovascularization.
If neovascularization develops, panretinal photocoagulation to the involved retinal sector should be applied using argon laser to achieve “medium” white burns, 200–500?mm in diameter—one burn width apart to cover the entire involved segment.
occurs. Vitreous surgery is employed occasionally for nonclearing vitreous hemorrhages, epiretinal membrane, or tractional retinal detachment with macular involvement. The outcomes are generally favorable, although preexisting pathology frequently limits recovery of good vision.
COURSE AND OUTCOME
Without treatment, one third of patients who have branch retinal vein occlusion end up with visual acuity better than 20/40 (6/12). However, two thirds have decreased visual acuity secondary to macular edema, macular ischemia, macular hemorrhage, or vitreous hemorrhage. As noted above, laser treatment for macular edema significantly enhances the chance that the patient’s baseline visual acuity will improve by two lines (65% versus 37%). The mean number of lines of improvement in visual acuity averages 1.33 in treated patients versus 0.23 in the control group. Poor visual prognostic factors include advancing age, male sex, worse baseline visual acuity, and an increased number of risk factors. Good prognosis is associated with a younger age, female sex, and fewer risk factors. Patients should be followed up every 3–4 months.
Approximately 20% of patients with branch retinal vein occlusion will develop neovascularization. Of these patients, about 60% will have episodic vitreous hemorrhages. Fortunately, laser treatment (sector PRP) can reduce this by one half to 30%.
NEW TREATMENTS FOR BRANCH RETINAL VEIN OCCLUSION
New treatments for branch retinal vein occlusion that are being evaluated include sheathotomy    and intravitreal steroid injection G triamcinolone for treatment of cystoid macular edema resulting from branch retinal vein occlusion.
1. Hayreh SS, Zimmerman MB, Podhajsky P. Incidence of various types of retinal vein occlusion and their recurrence and demographic characteristics. Am J Ophthalmol. 1994;117:429–41.
2. The Eye Disease Case-Control Study Group. Risk factors for branch retinal vein occlusion. Am J Ophthalmol. 1993;116:286–96.
3. Zegarra H, Gutman FA, Conforto J. The natural course of central retinal vein occlusion. Ophthalmology. 1979;86:1931–8.
4. Rath EZ, Frank RN, Shin DH, Kim C. Risk factors for retinal vein occlusions. A case-control study. Ophthalmology. 1992;99:509–14.
5. Hayreh SS, Zimmerman B, McCarthy MJ, Podhajsky P. Systemic diseases associated with various types of retinal vein occlusion. Am J Ophthalmol. 2001;131:61–77.
6. Quinlan PM, Elman MJ, Bhatt AK, et al. The natural course of central retinal vein occlusion. Am J Ophthalmol. 1990;110:118–23.
7. Dhote R, Bachmeyer C, Orellou MH, et al. Central retinal vein thrombosis associated with resistance to activated protein C. Am J Ophthalmol. 1995;120:388–9.
8. Glacet-Bernard A, Chabanel A, Lelong F, et al. Elevated erythrocyte aggregation in patients with central retinal vein occlusion and without conventional risk factors. Ophthalmology. 1994;101:1483–7.
9. Vine, AK. Hyperhomocystinemia: a new risk factor for central retinal vein occlusion. Trans Am Ophthalmol Soc. 2000;98:453–503.
10. Hayreh SS. Retinal vein occlusion. Indian J Ophthalmol. 1994;42:109–32.
11. Hayreh SS. Classification of central retinal vein occlusion. Ophthalmology. 1983;90:458–74.
12. The Central Vein Occlusion Study Group. Natural history and clinical management of central retinal vein occlusion. Arch Ophthalmol. 1997;115:486–91.
13. Green WR, Chan CC, Hutchins GM, et al. Central retinal vein occlusion: a prospective histopathologic study of 29 eyes in 28 cases. Retina. 1981;1:27–55.
14. Fong ACO, Schatz H, McDonald HR, et al. Central retinal vein occlusion in young adults (papillophlebitis). Retina. 1991;11:3–11.
15. May DR, Klein ML, Peyman GA, et al. Xenon arc panretinal photocoagulation for central retinal vein occlusion: a randomized prospective study. Br J Ophthalmol. 1979;63:735–43.
16. Magargal LE, Brown GC, Augsburger JJ, et al. Neovascular glaucoma following central retinal vein obstruction. Ophthalmology. 1981;88:1095–101.
17. Sabates R, Hirose T, McMeel JW. Electroretinography in the prognosis and classification of central retinal vein occlusion. Arch Ophthalmol. 1983;101:232–5.
18. Barber C, Galloway NR, Reacher M, et al. The role of the electroretinogram in the management of central retinal vein occlusion. Doc Ophthalmol Proc Ser. 1984;40:149–59.
19. Central Vein Occlusion Study Group. Baseline and early natural history report: the Central Vein Occlusion Study. Arch Ophthalmol. 1993;11:1087–95.
20. Kearns TP. Differential diagnosis of central retinal vein obstruction. Ophthalmology. 1983;90:475–80.
21. Furlan AJ, Whisnant JP, Kearns TP. Unilateral visual loss in bright light; an unusual symptom of carotid artery occlusive disease. Arch Neurol. 1979;36:675–6.
22. Schwab PJ, Okun E, Fahey FL. Reversal of retinopathy in Waldenström’s macroglobulinemia by plasmapheresis. Arch Ophthalmol. 1960;64:515–21.
23. Gutman FA. Evaluation of a patient with central retinal vein occlusion. Ophthalmology. 1983;90:481–3.
24. Stowe GC, Zakov ZN, Albert DM. Central retinal vascular occlusion associated with oral contraceptives. Am J Ophthalmol. 1978;86:798–801.
25. 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;102:1434–44.
26. The Central Vein Occlusion Study Group. Evaluation of grid pattern photocoagulation for macular edema in central vein occlusion: the Central Vein Occlusion Study Group M Report. Ophthalmology. 1995;102:1425–33.
27. McAllister IL, Constable IJ. Laser-induced chorioretinal venous anastomosis for treatment of non-ischemic central retinal vein occlusion. Arch Ophthalmol. 1995;113:456–62.
28. Mitchell P, Smith W, Chang A. Prevalence and associations of retinal vein occlusion in Australia. Arch Ophthalmol. 1996;114:1243–7.
29. Hayreh SS, Zimmerman MMB, Podhajsky P. Incidence of various types of retinal vein occlusion and their recurrence in demographic characteristics. Am J Ophthalmol. 1994;117:429–41.
30. Duker JS, Brown GL. Anterior location of the crossing artery in branch retinal vein occlusion. Arch Ophthalmol. 1989;107:998–1000.
31. Frangieh GT, Green WR, Barraquer-Somers E, Finkelstein D. Histopathologic study of nine branch retinal vein occlusions. Arch Ophthalmol. 1982;100:1132–40.
32. Branch Vein Occlusion Study Group. Argon laser scatter photocoagulation for prevention of neovascularization and vitreous hemorrhage in branch vein occlusion. Arch Ophthalmol. 1986;104:34–41.
33. Finkelstein D. Argon laser photocoagulation for macular edema in branch vein occlusion. Ophthalmology. 1986;93:975–7.
34. Amirikia A, Sioh IV, Murray TG, et al. Outcomes of vitreoretinal surgery for complications of branch retinal vein occlusion. Ophthalmology. 2001;108:372–6.
35. Glacet-Bernard A, Coscas G, Chabanel A, et al. Prognostic factors for retinal vein occlusion. A prospective study of 175 cases. Ophthalmology. 1996;103:551–60.
36. Opremcak EM, Bruce RA. Surgical decompression of branch retinal vein occlusions via arteriovenous crossing sheathotomy: a prospective review of 15 cases. Retina. 1999;19:1–5.
37. Ostelah MD, Charles S. Surgical decompression of branch retinal vein occlusions. Arch Ophthalmol. 1998;106:1469–71.
38. Shah GK, Sharma S, Fineman MS, et al. Arteriovenous adventitial sheathotomy for the treatment of macular edema associated with branch retinal vein occlusion. Am J Ophthalmol. 2000;129:104–6.