Year : 1994 | Volume
: 42 | Issue : 3 | Page : 109--132
Retinal vein occlusion
Sohan Singh Hayreh
Department of Ophthalmology, College of Medicine, University of Iowa, Iowa, USA
Sohan Singh Hayreh
Department of Ophthalmology, College of Medicine, University of Iowa, Iowa City 52242
In this review of the retinal vein occlusion (RVO), I have summarized recent advances on several controversial and clinically important topics: classification of RVO into six distinct clinical entities; pathogeneses and demographic characteristics of various types of RVO; differentiation of non-ischemic from ischemic central retinal vein occlusion (CRVO); differentiation of hemi-CRVO (HCRVO) from major branch RVO (BRVO); and the course, complications and management of various types of RVO
|How to cite this article:|
Hayreh SS. Retinal vein occlusion.Indian J Ophthalmol 1994;42:109-132
|How to cite this URL:|
Hayreh SS. Retinal vein occlusion. Indian J Ophthalmol [serial online] 1994 [cited 2020 Apr 2 ];42:109-132
Available from: http://www.ijo.in/text.asp?1994/42/3/109/25574
Retinal vein occlusion (RVO) is the most common retinal vascular occlusive disorder and is usually associated with a variable amount of loss of vision. Generally, RVO tends to be considered as one disease which is not only incorrect but also cause of most of the confusion on the subject. Our studies  have shown that from the point of view of pathogenesis, clinical picture, prognosis and management, RVO in fact consists of six distinct clinical entities. For over 150 years an immense amount of literature has accumulated on the subject. The objective of this paper is not to review that but to review in detail some recent advances on the subject, particularly controversial but clinically important topics which I have investigated systematically over the years, clinically and experimentally.
CLASSIFICATION OF RETINAL VEIN OCCLUSION
n 1962, when I was investigating experimentally the pathogenesis of optic disc edema in raised intracranial pressure, , in rhesus monkeys, I accidentally discovered that central retinal vein occlusion (CRVO) consisted of two distinct entities - non-ischemic and ischemic types. , This was afterwards further confirmed by us on clinical studies  and more experimental studies,  and subsequently by many other investigators.
When half of the retina is involved by RVO, it was always assumed in the past that this represented a variant of branch retinal vein occlusion (BRVO), and that the major vein from that half of the retina was blocked either at the arteriovenous crossing on the optic disc, or at the sharp edge of a deeply cupped optic disc. During the mid-seventies, we critically and systematically investigated these patients  Our studies revealed a hitherto unknown type of RVO, which we designated, after a good deal of debate, "hemicentral retinal vein occlusion" (HCRVO). These eyes in fact have blockage of one of two trunks of the CRV within the optic nerve [Figure 1] B. On going back over my previous anatomical studies on central retinal artery done during the late fifties I found an illustration (based on serial reconstruction of the central retinal artery and vein within the optic nerve) showing a CRV with two trunks in the anterior part of the optic nerve [Figure 1] in Hayreh et al [l9]. I also then discovered that Ida Mann, describing the development of central retinal vessels, had reported that from the third month of intrauterine life, there are always two trunks of the CRV in the optic nerve, one on either side of the central retinal artery, and one of the two central retinal venous trunks usually disappears before birth.  Chopdar later on reported that 20.5% of eyes in the general population show a dual-trunked CRV as a congenital anomaly  Our studies,  showed that HCRVO, like the CRVO, also consists of ischemic and non-ischemic types.
Based on this information and on that from our long-term prospective studies in over 1300 eyes with RVO during the past 25 years, it has become evident that, for any logical discussion and clinical management of RVO, it is imperative to classify RVO into the following six distinct clinical entities:
A. Central retinal vein occlusion (CRVO): This consists of:
I. Non-ischemic CRVO (or venous stasis retinopathy) [Figure 2][Figure 3][Figure 4]
II. Ischemic CRVO (or hemorrhagic retinopathy) [Figure 5]
B. Hemi-central retinal vein occlusion (HCRVO): This also consists of:
III. Non-ischemic HCRVO (or hemi-venous stasis retinopathy) [Figure 6]
IV. Ischemic HCRVO (or hemi-hemorrhagic retinopathy) [Figure 7]
C. Branch retinal vein occlusion (BRVO): This consists of:
V. Major BRVO [Figure 8] VI. Macular BRVO [Figure 9]
A good understanding of pathogenesis of a disease is absolutely fundamental to a scientific grasp of the clinical features of the disease and its logical management. A very basic and important fact not pointed out hitherto in any consideration of pathogenesis of various types of RVO is that, like all ocular vascular occlusive disorders, all types of RVO are multifactorial in origin and usually no single factor on its own causes the occlusion. A whole host of local and systemic factors acting in different combinations and to different extents may produce the vascular occlusion. In such a multifactorial scenario in various types of RVO, a particular factor or combination of factors may be present in one case and not in another, or a factor may play a major role in one and only a subsidiary one in another. Moreover, the role of the various factors may vary, with some as predisposing factors and others as precipitating ones in one group and vice versa in another. Thus each patient with RVO may have a unique combination of systemic and local factors which finally produce an episode of one or another type of RVO; the common tendency to blame one systemic disease for this is erroneous. It is also essential to comprehend that CRVO and HCRVO pathogenetically are very different from BRVO. In conclusion, it is a mistake to try to explain all types of RVO by one common pathogenetic mechanism.
A. Central retinal vein occlusion
I have discussed the pathogenesis of CRVO elsewhere in detail ,, In the center of the optic nerve the central retinal artery and vein lie side by side enclosed in a common fibrous tissue envelope [Figure 1]A. Klien and Olwin  and Klein  postulated the following three occlusive mechanisms in CRVO: (a) occlusion of the vein by external compression by sclerotic adjacent structures (i.e. central retinal artery and fibrous tissue envelope) and secondary endothelial proliferation; (b) occlusion by primary venous wall disease (degenerative or inflammatory in nature); and (c) hemodynamic disturbances produced by a variety of factors (e.g., subendothelial atheromatous lesions in the central retinal artery, arterial spasm, sudden reduction of blood pressure, blood dyscrasias, etc., and further aggravated by arteriosclerosis or unfavorable anatomic relations). These produce stagnation of the flow in the vein and result in primary thrombus formation in susceptible eyes. In this connection it is relevant to take into account Virchow's triad for thrombus formation: (i) slowing of blood stream, (ii) changes in vessel wall, and (iii) changes in the blood.
Non-ischemic CRVO (venous stasis retinopathy/): In our studies we have found that 81% of patients with CRVO initially belong to this category .  It seems that the site of occlusion in the CRV in this type is neither in the lamina cribrosa nor in the adjacent retrolaminar region but further back. The severity of retinopathy would depend on the site of occlusion - the farther back the occlusion, the milder the retinopathy, because of the availability of more and more collateral channels [Figure 1]A.
In elderly persons sclerotic changes in the structures adjacent to the CRV and secondary endothelial proliferation , in it within the optic nerve cause narrowing of the lumen of the CRV. This produces circulatory stasis and stagnation thrombosis. Hematologic abnormalities can also contribute to this process. Also, hemodynamic disturbances on the arterial side may play an important role in the development of this thrombosis. This is because blood flow in the retinal vessels depends upon the perfusion pressure (Perfusion pressure = mean arterial blood pressure minus the venous pressure). With venous stasis, secondary to narrowing of the lumen of the CRV (from partial thrombosis and/or other reasons described above), the venous pressure proximal to the site of thrombus rises and that results in fall of perfusion pressure and sluggish circulation. A fall in systemic arterial blood pressure would further lower the perfusion pressure. Recent studies with ambulatory blood pressure monitoring have established beyond doubt that during sleep there is a significant fall of blood pressure .  This may convert a partial thrombosis to complete thrombosis because of poor, sluggish circulation during sleep. The fact that many CRVO patients wake up with poor vision strongly suggests that nocturnal arterial hypotension plays an important role as the final insult in precipitating CRVO in persons who are susceptible.  Some of these patients experience transient recurrent thrombosis of the CRV before they finally develop permanent thrombosis. This is suggested by the fact that some of these patients complain of episodes of amaurosis fugax where they either (a) wake up with marked blurring of vision or (b) during the day get transient episodes of marked blurring of vision - in both cases the vision clears after a short interval. My explanation for these episodes of amaurosis fugax is that, as the thrombus progresses to completely occlude the CRV, it causes sudden stoppage of blood flow in the retinal vascular bed - > transient ischemia of the retina and associated visual loss; this sudden stoppage of venous outflow would also result in sudden rise in the blood pressure to the arterial level in retinal vascular bed proximal to the site of CRV thrombus. Since it is a fresh thrombus, it cannot withstand this high sudden arterial blood pressure, and pops out, like a champagne cork, resulting in restoration of retinal circulation and normal visual function. A gradual and progressive increase in size of the thrombus in the CRV and nocturnal arterial hypotension for many hours during sleep then finally, one day, produce permanent irreversible occlusion of CRV.
It is well-established that CRVO is significantly more common in patients with raised intraocular pressure OOP) and glaucoma. In our studies 22% of patients with CRVO had IOP of >22 mm Hg.  The pressure in the CRV at the optic disc depends upon the IOP, the former being always higher than the latter to maintain blood flow. A rise of IOP would produce retinal venous stasis and sluggish venous outflow - one of the factors in Virchow's triad for thrombus formation.
In young persons all the available evidence suggests that phlebitis of the CRV is responsible for thrombosis.  CRVO due to phlebitis has been given different eponyms including "papillophlebitis," "retinal vasculitis," "mild retinal and papillary vasculitis" and "optic disc vasculitis-type Il." 
Non-ischemic CRVO associated with cilio-retinal artery occlusion: In an eye with a cilio-retinal artery [Figure 1]A, CRVO usually results in transient occlusion of the cilio-retinal artery [Figure 10]. The mechanism of cilio-retinal artery occlusion in these eyes is as follows: Normally perfusion pressure in the retinal arterial vascular bed is higher than in the ciliary arterial vascular bed (and consequently in the cilio-retinal artery); additionally the retinal vascular bed has autoregulation while the choroidal vascular bed has no autoregulation. When there is sudden stoppage of blood flow by aa thrombus in the CRV, the blood pressure in the entire retinal capillary bed transiently rises to the level of central retinal artery blood pressure. As pointed out earlier, with the blood pressure in the cilio-retinal arteries being lower than in the central retinal artery, and the blood pressure in the retinal capillary bed being at the level of the central retinal artery, the cilio-retinal arteries cannot pump blood into a high pressure retinal capillary bed, resulting in a physiologic block in the cilio-retinal artery circulation. Within a day or, two, with the development of venous collaterals by the CRV in the optic nerve [Figure 1]A, the blood pressure in the retinal vascular bed falls to below the level of the blood pressure in the cilio-retinal artery, resulting in almost normal filling of the cilio-retinal artery once again. However, in the meantime, the retina supplied by the cilio-retinal artery has usually been damaged irreversibly by ischemia, resulting in visual loss - the severity of retinal ischemia in the area supplied by the cilio-retinal artery and associated visual loss depends upon the length of time elapsed before the circulation was re-established.
Ischemic CRVO (Herrrorrhagic retinopathy): It seems that in this type of CRVO, the site of occlusion is most probably in the region of lamina cribrosa or immediately posterior to that, as shown by most of the histopathological reports (since enucleated eyes invariably belong to the ischemic type of CRVO). These reports also show senile degenerative changes in the wall of the CRV and adjacent central retinal artery, resulting in marked narrowing of the lumen of the vein and the artery. Possibly sclerotic changes in the lamina cribrosa may also contribute to it. These changes would produce circulatory stasis and venous thrombosis. In patients with these predisposing changes, a fall of systemic blood pressure during sleep would finally complete the thrombotic process (as discussed above) sub This is suggested by the fact that these patients commonly discover marked visual loss on waking up in the morning. In these eyes the retina also suffers from focal or more extensive ischemia, usually of a recurrent transient nature. The mechanism of this is as follows: (a) In these eyes there is a marked rise of venous pressure because the site of CRVO is in the region of the lamina cribrosa or immediately behind that, with only a few small collaterals left to drain away the blood [Figure 1]A. (b) Nocturnal arterial hypotension may lower the blood pressure markedly during the night.(c) Retinal blood flow depends upon perfusion pressure (perfusion pressure = mean arterial pressure minus venous pressure). From this it is evident that in these eyes a combination of very low mean blood pressure at night with a very high venous pressure from CRV narrowing would cause a precipitous fall of perfusion pressure (associated with almost total arrest of retinal blood flow) during sleep. The importance of transient retinal ischemia in the production of ischemic CRVO was shown clearly by our experimental studies on the subject." With retinal ischemia there is also ischemic capillaropathy. As the blood pressure returns to normal levels or even hypertensive levels during waking hours, there is restoration of retinal circulation (although sluggish) and associated rise in intraluminal pressure in ischemic retinal capillaries that ruptures the weakened ischemic capillaries and produces extensive retinal hemorrhages.
In some patients non-ischemic CRVO changes to ischemic CRVO either overnight or gradually. I speculate that this may be caused by two mechanisms:
(a) As discussed above, blood flow in the retina depends upon perfusion pressure. In non-ischemic CRVO the perfusion pressure is low due to a rise in venous pressure. A further precipitous fall in perfusion pressure from marked nocturnal arterial hypotension would produce retinal ischemia during sleeping hours and may convert a non-ischemic to ischemic CRVO  in susceptible individuals, particularly elderly persons with cardiovascular and other systemic risk factors. This seems to be the most frequent sequence in eyes with overnight change from non-ischemic to ischemic CRVO. (b) All the available evidence indicates that in non ischemic CRVO the site of thrombosis is farther back in the optic nerve. In these cases one additional factor may be gradual extension of the thrombotic process in CRV towards the optic disc up to or close to the lamina cribrosa, involving and eliminating the available collaterals in the optic nerve which previously protected these eyes from developing ischemic CRVO [Figure 1]A. This would convert non-ischemic CRVO (usually gradually) to ischemic CRVO.
B. Hemi-central retinal vein occlusion
The pathogenesis of this type of RVO is similar to that of CRVO, and, like CRVO, it is also of non-ischemic and ischemic types. Usually only one of the two trunks of the CRV is involved" [Figure 1]B. Occasionally, however, both trunks may be involved - presumably due to site of occlusion being in the main trunk of CRV after the union of the two trunks, and this on routine examination would appear to be ordinary CRVO. I have seen a number of eyes where occlusion of one of the two trunks has a non-ischemic pattern and occlusion of the second trunk an ischemic pattern'; this is presumably determined by the location of the site of occlusion in the two trunks and the number of available collaterals in each trunk.
C. Branch retinal vein occlusion
It seems the first case of BRVO was reported by Leber in 1877.  Koyanagi  in 1928 first reported the association between BRVO and arterio-venous crossing, and now it is well established that site of occlusion in BRVO is invariably at the arterio-venous crossing [Figure 8]. It has been suggested in the literature that arteriosclerosis and arterio-venous crossing of retinal branch vessels play important roles in development of BRVO. During recent years a number of studies [32 ] have reported that at the site of occlusion in BRVO the retinal arteriole lies anterior to the occluded vein in 93%-100% of cases, and that such an arrangement is significantly more common in BRVO than in the corresponding and other retinal arterio-venous crossings in control groups - these findings confirm the original observations of Jensen in 1936.  Since the first report by Ammann in 1899,  it is also well-known that BRVO is seen more commonly in the temporal than nasal part of the retina, and of the temporal retina more frequently in the superior than inferior quadrant. The arterio-venous crossings have been shown to be more frequent in the supero-temporal quadrant than elsewhere , and situated closer to the optic disc in the supero-temporal than infero-temporal quadrant. 
Seitz,  in hypertensive patients on histologic studies, found thickening of walls and narrowing of the lumen of both retinal arteriole and vein. At the arterio-venous crossing, the vessels shared a common vascular wall and a common, thickened, adventitial and glial sheath, irrespective of which vessel was anterior. He found no compression of the underlying vessel, and attributed the indentation and obscuration of the underlying vein, when the arteriole crossed over the vein, to the deeper position of the vein, rather than to any true compression of the vein. In the only histopathologic report on BRVO in the literature that I am aware of, Rabinowicz et a1  studied four BRVOs in three eyes. They found that in 3 of the 4 BRVOs the lumen of the vein at the arteriovenous crossing was "fully patent," with complete occlusion in the fourth one. They found a marked degree of arteriolar changes in all. They postulated that arterial insufficiency was the "primary factor in the pathogenesis of" BRVO and hence felt that the term BRVO was a "misnomer" for this condition.
From all this information, it becomes evident that the arterio-venous crossing plays an important role in the pathogenesis of BRVO, and that the anterior position of the arteriole at the crossing somehow renders the underlying vein vulnerable to occlusion. The clinical picture and fluorescein angiographic studies in BRVO do show the site of obstruction to blood flow in the retinal vein at the arteriovenous crossing. From this clinical information it would seem logical to assume that sclerotic retinal arteriole probably compresses the accompanying vein because of a common thickened, adventitial and glial sheath; however, histopathological studies  have so far failed to confirm this view. Moreover, the very low incidence of BRVO inspite of the very high incidence of anterior location of the arteriole at the arterio-venous crossing in patients with arteriosclerosis and hypertension clearly indicates that, in the multifactorial etiology of BRVO, factors other than simple anatomical arrangement must play important roles. In addition to all that, focal phlebitis is a wellknown cause of BRVO, e.g., in toxoplasmic chorio-retinitis and retinal vasculitis. Similarly BRVO is seen in dysproteinemias, sickle cell disease and other hematologic disorders. Unlike CRVO and HCRVO, glaucoma and raised IOP play no role in the pathogenesis of BRVO.
We have recently investigated this topic in detail in 1,108 patients with RVO seen prospectively and consecutively in my clinic.
Distribution of various types of RVO: The various types of RVO may develop alone or in different permutations and combinations in one or both eyes, seen at the same time or on different occasions. [Table 1] gives details of distribution of various types of RVO in our series. Each episode of RVO in an eye was counted as an independent event. For example, an eye might have had a recurrence of non-ischemic CRVO and the last episode of that might have had in addition major BRVO superimposed on it: that would mean 3 episodes of RVO in that eye.
Laterality of RVO: Of the 6 types of RVO, we found that only ischemic CRVO and major BRVO showed evidence that RVO occurs more often in one (major BRVO in right eye in 57%) or the other (ischemic CRVO in left eye in 57%) eye. Among the other 4 types of RVO there was no significant difference between the involvement of the two eyes.
Age and gender at first onset of RVO: In our series the male/female ratio in non-ischemic CRVO was 279/221, in ischemic CRVO 100/84, in non-ischemic HCRVO 62/58, in ischemic HCRVO 27/14, in major BRVO 121 /145 and in macular BRVO 71/50. Thus all categories of RVO, except major BRVO, were more common in men than in women.
Our study showed that no age is immune to RVO. The onset of the first episode of RVO occurred anywhere between the ages of 14 and 92 years, with 51 % having the first episode at age >_ 65 years. Similar trends of a higher proportion of women being>_65 years at first onset of RVO compared to men were seen for all types of RVO. A comparison of the two types of CRVO showed a significantly higher proportion of patients with ischemic CRVO with first onset at _> 65 years (67%) compared to those with non-ischemic CRVO (44%). A comparison of CRVO, HCRVO and BRVO showed a higher proportion of patients aged ? 65 years at first onset in HCRVO (65%) and BRVO (63%) than in CRVO (56%). CRVO had more patients with age at first onset at Recurrence of RVO in the same and/or fellow eye:
In our study we tried to answer a number of important questions on this topic. These include the following:
1. Risk of same type of RVO developing in the fellow eye: Within 2 years the fellow eye developed non-ischemic CRVO in 6.6% and major BRVO in 3.4%, and within 4 years in 7.7% and 6.6%, respectively. For ischemic CRVO this risk is 5.6% within 2.8 years, for non-ischemic HCRVO 3.5% within 2.2 years, for ischemic HCRVO 7.4% at 0.4 year, and for macular BRVO 4.0% at 3.3 years.
2. Risk of recurrence of same type of RVO in the same eye: For non-ischemic CRVO this risk is 0.9% within 2.5 years and 2.2% within 5 years from the onset of first episode. For non-ischemic HCRVO it is 3.7% within 6.1 years. No recurrence was seen in ischemic
CRVO, ischemic HCRVO, macular BRVO, and in the same branch of major BRVO.
3. Chances of developing another episode of any type of RVO in the same or fellow eye: For the same eye
it is 0.9% within 2 years and 2.5% within 4 years. For the fellow eye it is 7.7% within 2 years and 11.9% within 4 years.
4. Chances of conversion of non-ischemic CRVO to ischemic CRVO: It was 9.4% within 6 months and 12.6% within 18 months from onset of non-ischemic CRVO. In patients > 65 years old there was a significantly higher rate of conversion compared to the young ( > 65 years old it was 13.2% at 6 months and 18.6% at 18 months whereas in middle-aged it was 6.7% and 8.1%, respectively.
5. Presence of more than one type of RVO in an eye at the same time or on different occasions: We found
that various types of RVO can occur either as a single RVO or in association with other types in one or both eyes in all permutations and combinations, and that an eye can have more than one episode of RVO of the same or a different type [Table 1]. Such combinations are not a rare occurrence.
6. Relative incidence of ischemic vs non-ischemic CRVO: In our series of 620 consecutive CRVO cases, 81% were non-ischemic and 19% ischemic CRVO. sub There is strong evidence to suggest that non-ischemic CRVO may be more common than that because a number of non-ischemic CRVOs may remain asymptomatic and resolve spontaneously, without the patient ever consulting an ophthalmologist. Proof of this statement is the detection of retino-ciliary venous collaterals oil the optic disc ( a footprint of this disease) during a routine ophthalmic evaluation in totally asymptomatic patients or in glaucoma patients later on. A ratio of 81 %: 19% for non-ischemic vs ischemic CRVO is highly important clinically because non-ischemic CRVO is a comparatively benign disease while ocular NV and other seriously blinding complications of CRVO are seen in only ischemic CRVO (see below).
7. Relative incidence of ischemic vs non-ischemic HCRVO and their extent and distribution : In our series of 154 consecutive HCRVO cases, there were 78% non-ischemic and 22% ischemic HCRVO.  Like nonischemic CRVO, there is evidence to suggest that nonischemic HCRVO may be even more common because it can remain asymptomatic as well and its footprint once again is the presence of venous collaterals on the optic disc. As discussed above, HCRVO may be due to occlusion of the superior or inferior trunk of the CRV [Figure 1]B. Usually each of the two trunks has venous drainage from one half of the retina (in 70% of non-ischemic HCRVO and 74% of ischemic HCRVO) ; however, one trunk may drain more than the other. In our studies we found that the area drained by one trunk may vary from 90° to 315 ° of the retina.  We found it less than 180° in 14% of non-ischemic HCRVO and 10% of ischemic HCRVO, and more than 180° in 17% and 16%, respectively.  Superior distribution of HCRVO was seen in 55% in non-ischemic type and 35% of the ischemic type and inferiorly in 45% and 65%, respectively. 
Diagnosis of CRVO and BRVO in general is not a problem because of their classical features. In the past when RVO involved half of the retina, it was always considered to be a variant of BRVO but our studies showed that this is generally HCRVO  - pathogenetically and clinically a variant of CRVO.
Central retinal vein occlusion: Patients with nonischemic CRVO may have no symptoms and it may be detected as an incidental finding on a routine ophthalmic examination. Retinal venous stasis with mild retinal hemorrhages per se is asymptomatic. It is almost always the development of macular edema that makes it symptomatic. When visual symptoms are present, it may be vague blurring of vision, frequently involving the central vision with an almost normal peripheral vision. The blurring in these cases is frequently more marked on waking up in the morning, improving to a variable extent after a few hours or in the afternoon. The onset of symptoms is usually gradual but it may be discovered suddenly on waking one morning. It is not rare to find some of these patients complain of episodes of amaurosis fugax before the constant blur. During episodes of amaurosis fugax, the patient may complain of a purple blur, particularly when non-ischemic CRVO is associated with cilioretinal artery occlusion.
In ischemic CRVO, on the other hand, there is always marked deterioration of vision; this is frequently noticed suddenly on waking one morning. As in non-ischemic CRVO some patients may complain of episodes of amaurosis fugax before the constant visual blurring.
While the diagnosis of CRVO is not difficult, the main problem is differentiation of non-ischemic from ischemic CRVO. This differentiation is crucial in the correct management of CRVO because non-ischemic CRVO is a comparatively benign condition but ischemic CRVO is often associated with serious blinding complications (see in the following pages).
Differentiation of non-ischemic from ischemic CRVO: In my systematic studies of CRVO for well over 25 years I have developed a number of diagnostic criteria to make such a differentiation during the early acute phase of CRVO.  The diagnosis becomes much more certain after a follow-up of about a year or longer; however, at that stage it may be too late, since the risk of developing anterior segment neovascularization (NV) in ischemic CRVO is greatest during the first 7 months or so and minimal after that, as discussed later. The criteria to differentiate non-ischemic from ischemic CRVO were further investigated in a protective stitly for their sensitivity and specificity and are discussed in detail elsewhere. various clinical tests used for the differentiation can be divided into morphological and functional tests; the following is a summary of our findings. 
Morphological tests: These consist of ophthalmoscopy and fluorescein fundus angiography. Ophthalmologists almost universally use ophthalmoscopic and fluorescein angiographic appearances to evaluate and manage CRVO and to differentiate ischemic from non-ischemic CRVO. It is absolutely essential to know how reliable and valid is the information provided by these two tests.
1. Ophthalmoscopy: Our long-term, prospective, natural history studies indicate that ophthalmoscopic findings in CRVO change much more with time than any other parameter investigated in our study,  so that there is virtually a continuous evolution of ophthalmoscopic lesions, as is evident from the following examples:
(a). Retinal hemorrhages: These are an invariable finding in CRVO [Figure 2][Figure 4][Figure 5). The most diagnostic feature of hemorrhages in CRVO is their location in the peripheral retina, seen best on indirect ophthalmoscopy [Figure 3],[Figure 4]A. Our study  revealed that the only ophthalmoscopic parameter to show reasonable sensitivity (81-84%) and specificity (72-74%) in differentiation of the two types of CRVO was the presence of more extensive hemorrhages in posterior retina in ischemic CRVO than in non-ischemic CRVO during the first 3 months of the disease only; and there was no statistically significant difference between the two types of CRVO in the amount of retinal hemorrhages after that period. However, the hemorrhages vary markedly not only from eye to eye (in both ischemic and non-ischemic CRVO) but also with time since onset. For example, in eyes with ischemic CRVO, although there are usually extensive hemorrhages during the early stages of disease, there may be minimal hemorrhages during its very early phase (i.e., within a day or so after the onset), increasing in severity over the following days and weeks, and starting to resolve after about 6 months. There is a small subgroup of eyes with ischemic CRVO which never develop extensive retinal hemorrhages in spite of evidence of marked retinal ischemia (in such cases there is no other etiology to account for retinal ischemia). Non-ischemic CRVO associated with extensive retinal hemorrhages is not at all a rarity [Figure 4].
(b). Cotton-wool spots: These represent focal inner retinal ischemic spots and are constantly evolving. Cotton-wool spots, which have always been regarded as diagnostic of ischemic CRVO, did not show high enough sensitivity and specificity in our study to help in differentiating the two types of CRVO, since non-ischemic CRVO can also have some cotton-wool spots [Figure 4]A.
I have often shown a set of color fundus photographs (of moderate to marked severity of non-ischemic CRVO and of ischemic CRVO mixed together) to a large gathering of ophthalmologists at many meetings of various ophthalmic societies and asked them to classify the type of CRVO from the photographs - at least 3/4 of the ophthalmologists in the audience failed to do so correctly in most cases. This and the data in our study  clearly reflect the problems associated with differentiation between the two types of CRVO from ophthalmoscopic examination only.
2, Fluorescein fundus angiography: For differentiation of ischemic from non-ischemic CRVO, retinal capillary obliteration is the fundamental information required from angiography [Figure 2]B,[Figure 5]B. During the early stages of CRVO, angiography has many serious limitations in providing this information, including the following:
(a). In our study,  fluorescein angiography failed to provide the required information on the retinal capillary non-perfusion in at least one third of the patients due to extensive retinal hemorrhages, poor quality angiograms (from hazy media, poor circulation, inadequate pupil dilation, and many other reasons) or inability to perform angiography.
(b). Even when retinal capillaries are satisfactorily seen on fluorescein angiograms, the findings may be misleading because:
(i) Retinal capillary obliteration is a progressive phenomenon. Available evidence indicates that it takes at least 3-4 weeks, often longer, after the development of ischemic CRVO, for retinal capillaries to be obliterated ,,,[39 ] so that if a patient is seen during the first 2-3 weeks after the onset of ischernic CRVO, fluorescein angiography may show perfectly normal filling of the retinal capillaries inspite of retinal ischemia (see [Figure 7] and [Figure 11] in Hayreh ). Some eyes with mild ischemic CRVO may develop slowly progressive retinal capillary obliteration many months or even years after the onset of CRVO. Under these circumstances, the initial observation of normal retinal capillary perfusion may mislead one into believing that this is non-ischemic CRVO when it is in fact ischemic.
(ii) It is well established now that in various retinopathies associated with retinal capillary obliteration, e.g., diabetic retinopathy, ischemic CRVO,BRVO and others, the retinal capillary obliteration starts first in the peripheral retina and then slowly progresses towards the posterior pole. -Therefore, ,standard fluorescein fundus angiography covering usually only the central 30°, and rarely 60°, of the posterior pole (i.e. optic disc and macular region) may provide no information about the peripheral retina. Sampling by a standard 30° (sometimes even with a 60°) angiography may reveal almost normal capillary perfusion at the posterior pole when in fact the retina may have extensive peripheral retinal capillary obliteration [Figure 11]. Magargal et a1 sub calculated the so-called "ischemic index" from fluorescein angiograms covering 30° or 45° area of the posterior pole and used that as the primary criterion to differentiate CRVO into ischemic and non-ischemic type. This "ischemic index" unfortunately gives no information about peripheral retinal capillary nonperfusion. The statement by Magargal et a1  that "eyes exhibiting extensive capillary nonperfusion of the posterior pole also had widespread ischemia in the retinal periphery" does not mean that the reverse is also true [Figure 11], as revealed frequently by our studies. From the absence of retinal capillary obliteration on angiography of the posterior pole, such an eye would be diagnosed as having nonischermic CRVO when in fact there is ischemic CRVO [Figure 11].
(iii) Even if there are perfectly satisfactory angiograms, there can be problems in their evaluation. For example, Welch and Augsburger  studied reproducibility and accuracy of eight retinal specialists in assessing the extent of retinal capillary nonperfusion on fluorescein angiography in CRVO and found that the proportion of agreement with the correct classification was less than 60% for all the specialists. Johnson et a1  in a forced choice analysis of ophthalmologists, to decide which fluorescein angiograms belonged to the NV group in their CRVO patients with and without iris NV, found that only 36% of patients who developed iris NV could be discriminated from those who did not.
Thus; from previous findings, from our study,  and from our much larger natural history studies on about 700 eyes with CRVO, it is evident that fluorescein fundus angiograms fail to provide any information about retinal capillary obliteration in at least about one third of the acute cases; of the remaining two thirds, because of the latent period between the ischemic insult to the retina and the development of capillary obliteration, and because of inadequate sampling of peripheral retina, angiography gives misleading information about retinal capillary obliteration in an unknown proportion of cases. A test which provides reliable information, at best in only 50-60% of cases, is clinically unreliable. Having made that highly important qualification about its limitations, however, one must point out that once the required information is available, fluorescein fundus angiography is an extremely helpful test with a high sensitivity and specificity. It is generally believed that the presence of any retinal capillary obliteration is a sign of ischemic CRVO; but our studies showed that this is not true, as isolated, small, focal retinal capillary obliteration is still compatible with non-ischemic CRVO.
Functional tests: These are visual acuity, visual fields plotted with Goldmann perimeter, relative afferent pupillary defect (RAPD) and electroretinography (ERG).
1, Visual acuity: In our stLdy,  a visual acuity of (a) better than 6/60 was found in 58.4% of the nonischemic CRVO eyes and only 1.7% of the ischemic CRVO; (b) 6/60 or better was seen in 80.9% of nonischemic and 6.7% of the ischemic CRVO, and (c) 6/120 or worse in 19.1% and 93.3% of the non-ischemic and ischemic, respectively. Visual acuity of 6/120 or worse in ischemic CRVO gave the best sensitivity (91%) and specificity (88%) during the first 3 months of CRVO in differentiation of ischemic from non-ischemic CRVO. Thus, if the visual acuity during the acute phase of CRVO is better than 6/120, it is in all likelihood a non-ischermic CRVO (in 88%), but the opposite is not necessarily true. Since visual acuity is influenced by the various types of macular lesion (e.g., edema, hemorrhages, cystic changes and ischemia, localized foveal and macular lesions), it can be markedly reduced even when the rest of the retina may be non-ischemic and functioning normally. That is why perimetry is much more helpful than visual acuity in differentiation of the two types of CRVO. If the media is hazy due to lens or vitreous opacities, or there are optic nerve lesions or macular degeneration or amblyopia, then visual acuity does not give reliable information.
2. Perimetry: We have found that visual field plotting with a Goldmann perimeter is far more helpful in differentiation of the two types of CRVO than a standard 30° automated perimeter. This is because the latter provides information about the central field only and not the peripheral fields extending up to 70-80° (provided only by the Goldmann, perimeter). Almost all eyes with CRVO have central scotoma, and the highly useful differentiating information is provided by the peripheral visual fields only. In the entire sample, 71% of eyes with nonischemic CRVO could see all the 3 targets (I-2e, I-4e, V-4e) of the Goldmann perimeter used by us, and the remaining 29% could see I-4e and V-4e; however, among the eyes with ischemic CRVO the corresponding figures were 8% and 63%, respectively, with 18% able to see only V-4e and 10% unable to see any target." Eyes with ischemic CRVO showed one visual field defect or another in 100% with I-2e, in 96-100% with I-4e and in 71-82% with V-4e during the first year of the disease; by contrast, the corresponding figures for eyes with non-ischemic CRVO were 54-78%, 38-48% and 12-17%, respectively. Sensitivities and specificities of visual field findings are almost as good as those of ERG in differentiating ischemic from non-ischemic CRVO during the first year after the onset of CRVO. However, if an eye ;has an unrelated optic nerve lesion, the visual fields do not give reliable information.
3. Relative afferent pupillary defect (RAPD): Our studies,  showed that this simple test is a highly useful and reliable functional test in differentiation of ischemic from non-ischemic CRVO. The RAPD is recorded in log units of neutral density filters by the method described elsewhere.  In our study,  the sensitivity and specificity of the RAPD for detection of ischemic CRVO were 88% and 90%, respectively, when RAPD of > 0.70 log units of neutral density filters was used as a cutoff; and 80% and 97%, respectively, when RAPD ? 0.90 log units was used as a cutoff. RAPD has many advantages over fluorescein fundus angiography and ophthalmoscopy, including the following:
a. RAPD gives reliable information (with much higher sensitivity and specificity) at the earliest stages of CRVO and throughout the course of the disease. The duration of the CRVO appears to have no statistically significant influence on RAPD.
b. RAPD provides reliable information in spite of hazy media (even when there is only a very hazy view of the fundus with indirect ophthalmoscope, e.g., with dense opacities of cornea, lens and vitreous) and extensive retinal hemorrhages.
c. RAPD (like the other functional tests) can detect conversion of non-ischemic to ischemic CRVO very early (within days), when morphological tests usually give little indication.
d. It is a non-invasive test, easy to perform as a part of routine eye examination, inexpensive, readily available and objective.
RAPD, however, like other tests, has limitations, including the following:
a. To test for RAPD, it is essential to have a normal fellow eye, and normal optic nerves and pupils in both eyes, therefore this test cannot be performed on eyes using miotics or other drops affecting the pupils, nor when there is any optic nerve disorder in either or both eyes, nor when the fellow eye is abnormal.
b. The amount of RAID is influenced by the size of central scotoma because it is modified more by the number of retinal ganglion cells involved than by the area of the retina  Thus, a patient with a large, dense central scotoma associated with marked macular edema due to non-ischemic CRVO would show a larger RAPD than otherwise expected. A study of relationship between the visual field defects and RAPD would help to correct this.
4. Electroretinography (ERG): This is also an objective functional test, very useful in the differentiation of ischemic from non-ischemic CRVO. Our study has shown that, of the various ERG parameters investigated by us,  the parameter with the best sensitivity (80-90%) and specificity (70-80%) in such a differentiation was the amplitude of the b-wave (both photopic and scotopic). In ischemic CRVO b-wave amplitude was: (a) reduced to 60% of the normal mean value, or (b) reduced by one standard deviation or more below the normal mean value, or (c) the ratio between the CRVO-eye and its fellow, normal eye was  ERG has the same advantages as RAPD except that it is not so easy to perform during a routine eye examination, requires expensive equipment, and is not as readily available. However, it has a distinct advantage over RAPD testing in that it does not require a normal fellow eye, normal pupils or normal optic nerve.
Combined information provided by RAPD and/or ERG: If the information provided by ERG (i.e. b-wave amplitude of 60% of the normal mean value in both photopic and scotopic ERG) is combined with that provided by RAPD ( > 0.70 log units), we found in our study sub that both RAPD and ERG provided the correct information in differentiation of ischemic from non-ischemic CRVO in 73-76% of the eyes, RAPD alone supplied the information (with ERG within normal limits) in an additional 16-19% and ERG alone (with RAPD within normal limits) in 5-8% more. Thus, combined information provided by ERG and/or RAPD was helpful in differentiating 97% of the CRVO cases.
Should ocular NV be used as the only feature in differentiating ischemic from non-ischemic CRVO? It has been advocated that ocular NV, particularly iris NV, should be used as the sole criterion. For example, in a number of recent studies  the two types of CRVO have been differentiated using iris NV as the sole or a definite criterion. I believe that this is not a valid way to classify CRVO for the following reasons:
a. Although ocular NV is usually regarded as the most serious complication of ischemic CRVO, it does not occur in all eyes with ischemic CRVO  (see below and [Figure 12]. We know definitely that there is a group of eyes with retinal capillary obliteration and other evidence of retinal ischemia, which never develop any type of ocular NV. 
b. Our experience has shown that ischemic CRVO
eyes, even if they do not develop NV, still usually have a poorer prognosis for visual outcome than nonischemic CRVO.
c. When we compared the results of the above 6 test-parameters in ischemic CRVO with and without ocular NV, we found that there were no statistically significant differences between the various test-parameters in the two types of ischemic CRVO. 
d. To divide CRVO into two types simply on the basis of presence or absence of iris NV only, would not differentiate ischemic from non-iscehmic CRVO, because about one third of the eyes with ischemic CRVO, but without any iris NV, would be grouped with non-ischemic CRVO; thus, eyes categorized as having "CRVO without ocular NV" in fact comprise not only non-ischemic CRVO eyes but also about one third of the ischemic CRVO eyes without ocular NV.
From this discussion, it is evident that to use iris NV as the sole criterion in differentiating CRVO into its two types would be totally unwarranted in any study dealing with evaluation of prognosis and management of CRVO in ischemic and non-ischemic CRVO.
In conclusion, in our study,  we found none of the 6 tests had 100% sensitivity and specificity in differentiating the two types of CRVO during the early acute phase, so that no one test can be considered a "gold standard," however, combined information from all 6 is almost always reliable. The 4 functional tests proved to be overall much superior to the two morphologic tests in differentiating ischemic from nonischemic CRVO: with RAPD most reliable in uniocular CRVO (with a fellow normal eye), followed closely by ERG in all cases, and combined information from RAPD and ERG differentiated 97% of cases; perimetry was the next most reliable, followed by visual acuity. The two morphologic tests performed worst; fluorescein angiography either provided no information at all on retinal capillary non-perfusion (in at least 1 /3rd of the eyes during the early, acute phase) because of multiple limitations, or sometimes provided misleading information. Ophthalmoscopic appearance is the least reliable, most misleading parameter.
Hemi-central retinal vein occlusion
Our clinical studies have shown that hemi-CRVO, pathogenetically a variant of CRVO, is a distinct clinical entity, often erroneously diagnosed as a major BRVO.  When the trunk draining about 90° of the retina is occluded, on a casual examination it mimics major BRVO but a critical evaluation would reveal HCRVO. The symptomatology in non-ischemic HCRVO is usually similar to that described for nonischemic CRVO above. In ischemic HCRVO the visual loss is much worse, involving the visual field in the sector of the retina involved as well as central visual loss from macular involvement.
Differentiation of hemi-CRVO froth major BRVO:
This distinction is important because major BRVO is almost always of ischemic type while HCRVO may be non-ischemic (in 78%) or ischemic (22%) type; with very different prognosis and complications for the two types (see below). In major BRVO the site of occlusion is almost always at the arteriovenous crossing, usually near the optic disc and very rarely over the optic disc [Figure 8]. In contrast to that, in HCRVO the site of occlusion is within the optic nerve and the two venous trunks entering the optic nerve may be seen clearly [Figure 6][Figure 7]. The difference in site of occlusion in the two types of RVO produces the following distinctions between them: (a) In HCRVO the corresponding part of the optic disc shows edema but the disc is normal in major BRVO unless the arteriovenous crossing is situated in the optic disc cup. (b) The location of venous collaterals connecting the occluded vein with the surrounding patent veins (to short-circuit the blood from the former to the latter in an attempt to re-establish circulation) gives important information about the site of occlusion; in major BRVO, these are located on the retina away from the optic disc but in HCRVO they are located either on the optic disc or within the optic nerve.
In HCRVO, as in CRVO, the incidence of raised IOP or glaucoma is much higher (in about one third of the cases) than in the general population, but not so in major BRVO. 
In differentiation of non-ischemic from ischemic HCRVO, the various tests discussed above in CRVO do not help much except that the visual fields with a Goldmann perimeter show, in addition to central scotoma, a dense segmental defect with larger isopters, and fluorescein angiography later on shows retinal capillary non-perfusion in the involved retina (Figure 6]B, [Figure 7]B.
Branch retinal vein occlusion: Major BRVO is invariably symptomatic, with visual blurring involving usually the central vision and also frequently the sector of visual field corresponding to the area of the retina involved. In macular BRVO there is always the central visual disturbances, with normal peripheral vision. Since macular BRVO always involves a sector of the macular retina radiating centrifugally from the fovea [Figure 9], there is no difficulty in differentiating it from major BRVO.
The various types of RVO run a self-limited course, taking from a few weeks to many years for the retinopathy to resolve. In the meantime some of these eyes can develop various complications, including the ones discussed below.
The main complications of the various types of RVO are as follows:
1. Macular edema: This is one of the major complications in all types of RVO. However, it does not involve all eyes with RVO, e.g., mild cases of non-ischemic CRVO/HCRVO, early cases of BRVO or when the BRVO does not involve the macular region. Chronic macular edema later on may produce cystoid macular degeneration, macular pigmentary degeneration (very much resembling age-related macular degeneration) and/or premacular gliosis/fibrosis/ membrane (cellophane maculopathy) - all resulting in central scotoma. I have seen patients with non-ischemic CRVO/HCRVO and BRVO who have had microcystic macular edema lasting for years, finally clearing up without leaving any residual retinal or visual changes; however, once there are secondary macular retinal pigment epithelial degenerative changes and/or development of cellophane maculopathy, prognosis for visual recovery on resolution of macular edema is poor.
2. Ocular NV: This is the dreaded complication. Our studies have shown that it is a complication of ischemic CRVO/HCRVO and major BRVO only.  It is extremely important to point out that ocular NV is NOT seen in non-ischemic CRVO/HCRVO or macular BRVO. If an eye with one of the latter diagnoses has ocular NV, the diagnosis is probably incorrect, or, most probably, there is associated diabetic retinopathy or other type of retinopathy with extensive retinal capillary obliteration or associated carotid artery disease. 
Ischemic CRVO: We studied the incidence of various types of NV in ischemic CRVO and the findings are discussed at length elsewhere. Very briefly, the ocular NV is seen only when the retinopathy is of moderate or marked severity. The cumulative probability of developing various types of ocular NV in this disease is shown graphically in [Figure 12], which provides four very important pieces of information:
(a). Every eye with ischemic CRVO does not develop ocular NV.
(b). When ocular NV develops, the commonest site is the anterior segment, much less frequently the posterior segment.
(c). The greatest risk of developing anterior segment NV is during the first 7 months, after which the risk falls dramatically to minimal.
(d). This completely debunks the commonly held belief that every eye with CRVO that develops iris NV or angle NV is destined to develop the dire complication of neovascular glaucoma (NVG). In fact, about one-third of the eyes with iris NV and about one-quarter of the eyes with both iris and angle NV never develop NVG.
This study also showed that age of the patient did not influence the incidence of NV. We also found that none of the 6 clinical tests used to differentiate the two types of CRVO (see above) seem to be able to predict definitely the development of various types of ocular NV in ischemic CRVO. 
In order to place ocular NV in CRVO in true perspective, it is important to point out two important facts: (i) Ischemic CRVO constitute only one fifth of all CRVO cases. (ii) NVG, the most dreaded complication of CRVO, is seen at the maximum in about 50% of ischemic CRVO cases only. This would indicate that the overall incidence of NVG in entire CRVO group is only 10% at the maximum - a highly important fact in any management considerations for CRVO.
Ischemic HCRVO: Ocular NV in this type is very different from that seen with ischemic CRVO. In our studies, 58.1% of the eyes developed ocular NV overall, with retinal and/or optic disc NV being the commonest (retinal NV in 41.9% and disc NV in 29%) and anterior segment NV uncommon (iris NV in 12.9%, angle NV in 6.5%).  I have seen NV glaucoma develop in only one eye during my entire experience. The severer and more extensive the retinopathy, the more marked the NV.
Major BRVO: NV is seen much less commonly in this than in CRVO/HCRVO. In our studies, 28.8% of eyes developed ocular NV overall, with retinal and/or disc NV being the commonest (retinal NV in 24.1% and optic disc NV in 28.8%) and anterior segment NV extremely rare (iris NV in 1.6% and angle NV in 0.5%)  - the latter is seen essentially in eyes with BRVO involving half or more of the retina. Like ischemic HCRVO, the severer and more extensive the retinopathy, the more marked the NV.
3. Vitreous hemorrhage: In RVO this may be either secondary to retinal/optic disc NV or due to rupture of the blood through the internal limiting membrane, particularly in eyes with many sub-internal limiting membrane hemorrhages. In my experience, the latter source of vitreous hemorrhage is much more common than the former, and can occur in all types of RVO, usually during the early stages of the disease. Vitreous hemorrhage from retinal/disc NV usually occurs during the late stages of the disease. Sometimes the vitreous hemorrhage can occur from intraretinal microvascular abnormalities secondary to RVO or development of posterior vitreous detachment. Therefore, it is important to be aware of the fact that presence of vitreous hemorrhages in RVO does not always mean the presence of retinal/disc NV.
4. Cilio-retinal artery occlusion: The major cause of serious visual loss in non-ischemic CRVO is the development of associated occlusion of a cilio-retinal artery, particularly when the artery involves a large sector of the retina (resulting in a large sectoral visual field defect) or involving the entire maculopapillar bundle (resulting in a large absolute centrocecal defect - [Figure 10]. The mechanism of development of cilia-retinal artery occlusion in these eyes is discussed above.
MANAGEMENT OF RVO
This still remains uncertain, with treatment options varying from logical to totally absurd being championed from time to time. The therapeutic regimens advocated and tried include anticoagulants, fibrinolytic agents, low molecular weight dextran infusion, carbon dioxide inhalation, vasodilators, hyperbaric oxygen, ocular hypotensive therapy, surgical decompression of CRVO, hemodilution, photocoagulation, systemic corticosteroids and a host of others. The literature on many of these was well summarized by Sedney  and I have also reviewed it in the past.  In almost none of these studies RVO was differentiated into its various types. Successes and beneficial effects claimed for many therapies in most cases simply represent the natural history of the disease, and that basic fact has been ignored in almost all the studies. This is well demonstrated by [Table 2] which is based on analysis (done many years ago) of visual acuity outcome on resolution of various types of RVO, in our prospective natural history study of RVO.
In the management of non-ischemic CRVO 4 fundamental facts have to be borne in mind. (i) The only visually disabling problem that requires management in this disease is macular edema. (ii) Nonischemic CRVO does not develop ocular NV. (iii) It is a self-limiting disorder. (iv) As shown by our natural history visual acuity data analysis in [Table 2], two third of the patients finally have 6/12 or better visual acuity without any treatment, and 6/12 visual acuity is legally considered normal.
1. Anticoagulants: This therapy has been popular for well over half a century, with some claiming varying degrees of success while others believe that it makes no significant difference in outcome as compared to untreated cases.  A critical review reveals that the cases where success has been claimed appear to have been those that would have shown almost equally good results if left alone. In my experience of over 1200 cases of RVO, I have seen many patients who were started on anticoagulants by ophthalmologists, and developed excessive amount of retinal hemorrhages, with devastating results - often converting a benign form of RVO into blinding RVO. Not only that, but I also have many patients who developed RVO while on anticoagulants for various systemic diseases. I have found that patients on antiplatelet agents, such as aspirin; are also liable to develop excessive retinal hemorrhages which adversely affect the outcome. Thus, I have come to the conclusion that anticoagulants and antiplatelet agents are not only of no therapeutic value but are definitely harmful in RVO and should never be used.
2. Fibrinolytic or thrombolytic agents: The object of this therapy is to dissolve the preformed thrombus. From the studies reported in the literature there is little evidence that these have any beneficial effects.  Moreover, administration of these drugs can cause a significant hazard of bleeding, e.g., vitreous and cerebral hemorrhage. There is only one controlled randomized study on the subject and the authors of that study do not advise this treatment for CRVO. ,
3. Surgical decompression of CRVO: This was first advocated by Vasco-Posada  who cut the scleral ring and adjacent part of the dural sheath of the optic nerve. More recently Spoor, at the various meetings in the United States, has advocated optic nerve sheath decompression. Having performed extensive scientific research in various aspects of the field I find that both these procedures are of NO scientific merit at all because these are based on totally erroneous concepts and inadequate basic knowledge of the subject, and could be dangerous.
4. Hemodilution: A few studies have suggested presence of abnormal blood viscosity in RVO patients. , Based on that assumption, two German groups have recently advocated its use in CRVO, claiming some improvement in visual acuity in non
ischemic CRVO  however, in ischemic CRVO they make contradictory claims - one group claiming hemodilution is more effective in ischemic than nonischemic CRVO  while the other claiming no beneficial effect in ischemic CRVO.  I personally do not advocate this therapy because of: (a) the conflicting claims based on very small samples, (b) hazards of hemodilution therapy, (c) chronic nature of the disease, (d) the absence of definite proof that most patients with CRVO have hyperviscosity (as revealed by our studies), and (e) above all I have found similar spontaneous visual improvement in patients with nonischemic CRVO as a part of natural history study (see above).
5. Systemic corticosteroids: In non-ischemic CRVO, as discussed above, the primary cause of visual disability is the presence of macular edema. Therefore in these cases the most important management consideration is controlling macular edema. There is no definite therapy available to treat macular edema. In our prospective studies on the subject for about 25 years, I have found that there is a definite group of patients with non-ischemic CRVO who respond to systemic corticosteroids, with resolution of macular edema and visual improvement while they are on therapy. With systemic corticosteroids I have seen a similar reduction of macular edema of diverse noninflammatory etiology in many patients. The most Critical thing to remember in systemic corticosteroid therapy for macular edema is that it does not work in every patient, suggesting that persons can be divided into responders and non-responders (as in glaucoma), depending probably upon the presence or absence of steroid receptors. Some workers have argued that steroid therapy has no role in non-ischemic CRVO because it is not an inflammatory disease and as such there is no justification for this therapy  ;however, at the same time they do admit that they have never tried this treatment!  We have ample evidence that systemic corticosteroids do work in many non-inflammatory diseases, e.g., cerebral edema of traumatic or ischemic origin, so that the concept that steroids work only in inflammatory disorders is totally erroneous.
In the management of non-ischemic CRVO with macular edema and a visual acuity of worse than 6/12, 1 first explain fully the side-effects and limitations of steroid therapy, stressing strongly that not every patient responds. I offer corticosteroid therapy as an option. If they opt for this therapy, I use a starting dose of 80 mg prednisone daily orally. Those who are going to respond to this therapy usually report a dramatic visual improvement. I find that if a patient shows no worthwhile improvement in visual acuity and/or fields within 2 weeks on 80 mg prednisone, there is little likelihood of that patient responding to the therapy; in that case I stop the treatment. If a patient does respond favorably, then, after 2 weeks of 80 mg prednisone, I very slowly start to taper down the dosage, monitoring the visual acuity and visual field every 2-3 weeks. In the vast majority of these patients when the dosage is reduced to about 40 mg Prednisone, the macular edema starts to develop again and the vision starts to deteriorate. There is a marked inter-individual variation in the minimum dose that is required to keep macular retina free of edema. By this time many of the patients start to develop systemic side-effects of corticosteroids. It is important to stress to the patient that the treatment is simply helping to reduce or eliminate macular edema to prevent long-term permanent macular changes, and that it is NOT a treatment of the CRVO which has to take its own natural course. The exception to this may be the young persons, in many of whom available evidence indicates that the CRVO is due to phlebitis" which may be influenced by the corticosteroid therapy. It is not at all uncommon for those who respond to treatment to be on a maintenance dose of about 40 mg Prednisone or so for many months or even over a year. For example, in one young man with non-ischemic CRVO in his only eye and a visual acuity of 6/60 initially, visual acuity immediately improved to 6/9-6/12 with 80 mg Prednisone. To maintain that visual acuity he required a maintenance dose of 30-40 mg Prednisone for almost 3 years, and every attempt to go any lower immediately produced worsening of macular edema and deterioration of visual acuity. He finally ended up with a visual acuity of 6/6 and no visual or systemic disability.
I have found steroid therapy of no help in ischemic RVO inspite of having macular edema, most probably because of irreversible ischemic damage to the macular retina.
(6). Systemic acetazolamide (Diamox): I have found that some patients (but not all) with non-ischemic CRVO and macular edema respond to this therapy; but once again the macular edema is under control only so long the patient is taking the drug. I usually give sustained release acetazolamide (Diamox Sequels) 500 mg twice daily. Unfortunately this drug can also produce severe systemic side-effects in some patients. If a patient does not respond within 2 weeks, I discontinue the therapy. I have not found this helpful in ischemic RVO.
(7). Ocular hypotensive therapy: Ophthalmologists often start patients with RVO on ocular hypotensive therapy, e.g., topical beta blockers, pilocarpine, etc., or even systemic carbonic anhydrases. This is done under the totally erroneous impression that lowering the IOP improves the retinal blood flow. With venous outflow obstruction, lowering the IOP does not influence the retinal blood flow. Moreover, our clinical and experimental studies have clearly shown that development of CRVO, HCRVO and major BRVO, by some unknown mechanism, lowers the IOP  usually more than any of the ocular hypotensive agents do. Thus the eyes with RVO usually have normal IOP unless the eye has NVG. In eyes with normal IOP, ocular hypotensive therapies seldom lower the pressure appreciably. Furthermore, lowering the IOP to very low levels would enhance macular edema.
Since there is a high incidence of ocular hypertension or glaucoma in eyes with CRVO and HCRVO, development of CRVO or HCRVO mandates treatment of ocular hypertension in the reduce the risk of development of RVO in that eye.
(8). Photocoagulation: Panretinal photocoagulation (PRP) is almost universally considered the treatment of choice in RVO to prevent development of ocular NV, and macular grid photocoagulation in management of macular edema. This practice is based on many reports of the beneficial effects of photocoagulation. Recently , we critically reviewed the literature on the subject and that revealed serious flaws in most of the studies.
(i). Central retinal vein occlusion: Since there is no retinal ischemia in non-ischemic CRVO, there is absolutely NO indication for PRP in this disease. The role of grid photocoagulation in the management of macular edema in non-ischemic CRVO is still experimental, with no definite proof of its efficacy and safety. I personally do not use this.
In ischemic CRVO, the theoretical fundamental reason advocated for PRP is to prevent development of ocular NV and associated blinding complications of NVG and/or vitreous hemorrhage. We conducted the first long-term (10-year) prospective, planned study of argon laser PRP in ischemic CRVO.  In our study, on comparing the lasered eyes versus the non-lasered eyes, there was no statistically significant difference between the two groups in the .incidence of development of angle NV, NVG, retinal and/or optic disc NV, or vitreous hemorrhage, or in visual acuity. Our study, however, did show a statistically significant (p=0.04) difference in the incidence of iris NV between the two groups, with iris NV less prevalent in the lasered group than in the non-lasered group, but only when the PRP was performed within 90 days after the onset of CRVO; however, iris NV per se is of little importance (see below). The other parameter which showed a statistically significant difference between the two groups was the peripheral visual fields - the lasered group suffered a significantly (p 0.03) greater loss than the non-laser group [Figure 13]. The results of our study turned out totally opposite to our original expectation since PRP has been universally regarded as the well-established treatment for ischemic CRVO.
Magargal and co-workers  claimed that none of the eyes treated by them with PRP developed NVG attributable to ischemic CRVO and "iris neovascularization subsequently regressed in each case."  They concluded that "Prophylactic PRP in high-risk ischemic CRVO eyes appears to eliminate virtually the devastating complications of NVG."  This is a truly amazing claim-I00% effectiveness of PRP in 100 eyes! In contrast to their claim of 100% success, Laatikainen et al  in their prospective randomized study, found no statistically significant difference in NV cony plications between treated and untreated groups. We, likewise, found no statistically significant difference in the incidence of NVG between the treated and untreated control population. If PRP were 100% effective in preventing NVG, as claimed by Magargal et a1 , surely in our study, conducted prospectively and meticulously over a 10-year period, we would have seen at least some statistically significant difference in the incidence of NVG in the lasered group as compared to the non-lasered group.
The original rationale for advocating PRP in ischemic CRVO was the proven beneficial effect of PRP on ocular NV in proliferative diabetic retinopathy; however, this did not take into consideration the marked disparities in the disease processes between ischemic CRVO and proliferative diabetic retinopathy and in their responses to PRP. This is because proliferative diabetic retinopathy is a chronic, slowly progressive disease while ischemic CRVO is an acute catastrophe associated with severe and extensive retinal ischemia but with a self-limiting course. One can compare ischemic CRVO to a hurricane which develops suddenly and inflicts extensive, devastating damage to a house, and disappears. By contrast, diabetic retinopathy is like a slow leak which would undermine the house gradually over a period of years, very slowly and insidiously. Measures which would successfully control the damage caused by a slow leak are totally useless against a hurricane! The extent of retinal ischemia, and hence the quantity of the presumed vasoproliferative factor(s), in ischemic CRVO is many, many times that in average proliferative diabetic retinopathy. PRP may be able to cope with the mild amount of retinal ischemia seen in the usual proliferative diabetic retinopathy but is totally inadequate and ineffective when there is the severe, extensive, sudden retinal ischemia of ischemic CRVO. This basic fact has been totally ignored by the advocates of PRP.
If all these facts are put together, a very different perspective on ischemic CRVO emerges - and, consequently, a different perspective on its management. In ischemic CRVO, if the eyes either do not develop NVG (and about 50% will not - [Figure 12] or their high IOP due to NVG is controlled satisfactorily by the means discussed elsewhere,  once the retinopathy burns itself out, they are frequently left with a large, dense central scotoma (from the macular lesions invariably seen in this disease), but with a normal or a reasonably good peripheral visual field. The latter is functionally very useful. PRP, by producing marked constriction and loss of the peripheral visual field, in eyes with large, absolute, central scotoma, thus actually does more harm than good to the majority of eyes [Figure 13], especially when it confers no statistically significant protection against NVG or any other benefit. In the group which would never have developed NVG in the first place (about 50% - [Figure 12] and thus did not need PRP, the high risk of marked peripheral visual field loss from PRP combined with the naturally developing central scotoma from the disease itself [Figure 13], is a serious disabling complication of PRP without any countervailing benefit.
Currently there is a multicenter study going on in the United States to evaluate the role of PRP in ischemic CRVO.  Unfortunately, in my opinion, the design of that study has serious flaws which could vitiate its results, e.g.: (i) In differentiating ischemic from non-ischemic CRVO, the authors have used 10 disc area or more of retinal capillary nonperfusion on fluorescein angiography as almost the sole criterion to make such a differentiation. As discussed above, our studies showed this to be a highly deceptive and unreliable guide for making such a differentiation during the acute phase.  (ii) The study assumes that any ischemic CRVO eye with iris NV is certain to go on to develop NVG and therefore deserves prompt mandatory PRP. Our study has shown this to be a false assumption, because [Figure 12] shows that about 30%, 40% and 50% of the eyes with ischemic CRVO never develop iris NV, angle NV or NVG, respectively, and every eye with iris NV does not develop NVG.  Iris NV may be worrisome and an indication for closer observation but I have followed very closely for years some ischemic CRVO eyes with iris NV or both iris and angle NV which never developed NVG, and the iris and angle NV resolved spontaneously. The advocated objective of PRP is to prevent development of NVG. Much more importantly, by treating all patients with iris NV, and not randomizing them to "treatment" or "no treatment," the Study puts a cloud over its use of iris NV as an outcome measure. (iii) Most importantly, no consideration is given in the Study design to obtaining information on the effect of PRP on peripheral visual fields in ischemic CRVO which was found to be the most significant and disabling finding in our study.
(ii). Hemi-central retinal vein occlusion: In non-ischemic HCRVO, once again there is no justification to do scatter photocoagulation to the involved region.
We conducted a prospective study to ascertain whether scatter argon laser photocoagulation to the involved sector in ischemic HCRVO was beneficial.  The results are discussed below with those in major BRVO.
(iii). Branch retinal vein occlusion: Since 1968, a large number of publications concerning the role of photocoagulation in the management of BRVO have appeared - only 3 of them prospective studies ,,[59 ]sub In our study we included both major BRVO and ischemic HCRVO to investigate the effect of argon laser scatter photocoagulation on retinal and/or disc NV and vitreous hemorrhages and also on visual acuity, visual fields and macular retinal lesions.  Our study showed that peripheral scatter photocoagulation significantly reduced the risk of development of retinal NV and vitreous hemorrhage, did not affect the visual acuity and macular retinal lesions, but produced a significant worsening in the peripheral visual fields compared to the untreated eyes. The multicenter tria1  also found a significant beneficial effect of the peripheral scatter laser treatment on NV and vitreous hemorrhage; however, the multicenter study did not evaluate the effects on peripheral visual fields. Since loss in the lower part of the visual field can produce marked disability and BRVO involving the superior retina is common, a significant worsening of visual fields with laser treatment becomes a very important, clinically relevant finding. The prevalent practice of treating every BRVO patient with peripheral scatter laser photocoagulation is not justified in view of the high probability of visual field loss after treatment, when the risk of developing vitreous hemorrhage is only 14% (19% develop NV without treatment, and 73% of these eyes with NV are at risk of developing vitreous hemorrhages).  This confirms the observations from the previous multicenter study  that "there may be no advantage to treating prior to the development of NV." Thus, our study  showed about 85% of patients with BRVO would be treated unnecessarily and exposed to the unwarranted risk of visual field loss. ; Peripheral scatter laser therapy should therefore be done only if there is NV, to balance the beneficial effect of therapy for preventing vitreous hemorrhage against its detrimental effect on the visual fields.
The role of grid photocoagulation in macular edema secondary to BRVO (over the area of capillary leakage seen on fluorescein angiography) was investigated by a multicenter study.  In that study, a comparison of treated to control patients showed that a gain of at least 2 lines of visual acuity from baseline was significantly greater in treated eyes, and the authors recommended "laser photocoagulation for patients with macular edema associated with branch vein occlusion who meet the eligibility criteria" of their study.
TERMINOLOGY: Finally, the use of a correct and universal name for a disease is absolutely essential for proper communication. Our clinical ,], and experimental ,,, ] studies during the past 30 years on pathogenesis, clinical features and other aspects of CRVO help to clarify the current profusion - and confusion - of eponyms used by different authors for the two types of CRVO. The terms "partial," "incomplete," "imminent," "threatened," "incipient" or "impending" CRVO describing non-ischemic CRVO are invalid , because our experimenta ,, and clinicall ,, studies have demonstrated that the CRVO is almost always complete in these cases. For non-ischemic CRVO the terms "hyperpermeabilityresponse-macular-edema-type " or "hyperpermeable type" are incorrect, because macular edema: (a) is seen in both the non-ischemic and ischemic CRVO and significantly (psub (b) is always secondary to hyperpermeability of retinal capillaries in CRVO, and (c) is absent in a number of non-ischemic CRVOs.used the terms "perfused" and "non-perfused" for nonischemic CRVO, respectively sub - these terms again are incorrect because non-ischemic CRVO can also have patches of non-perfused retinal capillaries and ischemic CRVO does have variable amount of retinal capillary perfusion. A third category of CRVO, called "mixed,"  "indeterminant" or "indeterminant perfusion ["44] type, has been postulated by some authors. We feel this third category simply represents a testing artifact because of the poor differentiating power of the two tests (ophthahnoscopy and fluorescein fundus angiography) used by those authors. When combined information from the 6 clinical tests discussed above is used, this third category disappears completely,
I have used the currently popular terms "ischemic" and "non-ischemic" CRVO to denote the two types. These terms are reasonable, however, they do present some problems from the strictly scientific point of view. For example, our studies , have shown that the presence of up to 0.6 log units of RAPD and a reduction in ERG mean b-wave amplitude by up to 40% are compatiable with non-ischemic CRVO; however, these RAPD and ERG parameters indicate a certain amount of retinal ischemia in those eyes. I feel that "Venous Stasis Retinopathy" may be the most suitable term for these eyes, because venous stasis may or may not be associated with a certain amount of retinal hypoxia, depending upon the severity of the stasis. In all eyes with "ischemic CRVO" there is definite evidence of retinal ischemia, so such a term is scientifically valid; however, for this group the term "Hemorrhagic Retinopathy" has been used since 1855 ,[64 ] and our study  revealed that during the first 3 months retinal hemorrhages were statistically (pI am grateful to Mrs. Georgiane Parkes-ferret and Mrs. Jill House for their secretarial help, and to my wife Shelagh for her help in the preparation of this manuscript. These studies were supported by research grants numbers EY-1151, EY-1576 and RR-59 from the National Institutes of Health, and in part by unrestricted grants from Research to Prevent Blindness, Inc., New York, and from Alcon Research Institute, Fort Worth, Texas. Dr. S.S. Hayreh is a Research to Prevent Blindness Senior Scientific Investigator.
|1||Hayreh SS, Rojas P, Podhajsky P, Montague P, Woolson RF. Ocular neovascularization with retinal vascular oc|
|2||Hayreh SS, Zimmerman MB, Podhajsky P. Seasonal variations in the onset of retinal vein occlusion. Br J Ophthalmol. 76:706-710, 1992.|
|3||Hayreh SS, Zimmerman MB, Podhajsky P. Incidence of various types of retinal vein occlusion and their recurrence and demographic characteristics. Am J Ophthalmol. 117:429-441, 1994.|
|4||Hayreh SS, Pathogenesis of oedema of the optic disc (papilloedema). Thesis for Ph.D., University of London, 1965.|
|5||Hayreh SS, Pathogenesis of oedema of the optic disc. Doc Ophalmol. 24:289-411, 1968.|
|6||Hayreh SS. An experimental study of the central retinal vein occlusion. Trans Ophthalmol Soc UK. 84:586-593, 1964.|
|7||Hayreh SS. Occlusion of the central retinal vessels. Br J Ophthalmol. 49:626-645, 1965.|
|8||Hayreh SS. Pathogeneis of occlusion of the central retinal vessels. Am J Ophthalmol. 72:998-1011, 1971.|
|9||Hayreh SS. So-called 'central retinal vein occlusion.' I. Pathogenesis, terminology, clinical features. Ophthalmologica. 172:1-13, 1976.|
|10||Hayreh SS. Central retinal vein occlusion: Differential diagnosis and management. Trans Am Acad Ophthalmol Otolaryngol. 83:OP-379-391, 1977.|
|11||Hayreh SS. Central retinal vein occlusion. In: Mausolf FA, editor. The eye and systemic disease, ed. 2. St.Louis: CV Mosby, 1980, 223-275.|
|12||Hayreh SS. Classification of central retinal vein occlusion. Ophthalmology. 90:458-474, 1983.|
|13||Servais GE, Thompson HS, Hayreh SS. Relative afferent pupillary defect in central retinal vein occlusion. Ophthalmology. 93:301-303, 1986.|
|14||Hayreh SS, Klugman MR, Podhajsky P, Kolder HE. Electroretinography in central retinal vein occlusion: Correlation of electroretinographic changes with pupillary abnormalities. Graefes Arch Clin Exp Ophthalmol. 227:549-561, 1989.|
|15||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. 228:201-217, 1990.|
|16||Hayreh SS, van Heuven WAJ, Hayreh MS. Experimental retinal vascular occlusion. I. Pathogenesis of central retinal vein occlusion. Arch Ophthalmol. 96:311-323, 1978.|
|17||Juarez CP,Tso MOM, van Heuven WAJ, Hayreh MS, 'Hayreh SS. Experimental retinal vascular occlusion II: A clinico-pathologic correlative study of simultaneous occlusion of central retinal vein and artery. hit Ophthalmol. 9:77-87, 1986.|
|18||Juarez CP, Tso MOM, van Heuven WAJ, Hayreh MS, Hayreh SS. Experimental retinal vascular occlusion III. An ultrastructural study of simultaneous occlusion of central retinal vein and artery. hit Ophthalmol. 9:89101, 1986.|
|19||Hayreh SS, Hayreh MS. Hemi-central retinal vein occlusion. Pathogenesis, clinical features, and natural history. Arch Ophthalmol. 98:1600-1609, 1980.|
|20||Hayreh SS. A study of the central artery of retina in human being in its intra-orbital and intra-neural course. Thesis for Master of Surgery, Panjab University, 1958.|
|21||Mann I. The development of the human eye. New York: Crone & Stratton Inc., 1969:228.|
|22||Chopdar A. Dual trunk central retinal vein: Incidence in clin:=al practice. Arch Ophthalmol. 102:85-87, 1984.|
|23||Klien BA, Olwin JH. A survey of the pathogenesis of retinal venous occlusion. Arch Ophthalmol. 56:207-247, 1956.|
|24||Klien BA. Sidelights on retinal venous occlusion. Am J Ophthalmol. 61:25-35, 1966.|
|25||Hayreh SS, Zimmerman MB, Podhajsky P, Alward WLM. Nocturnal arterial hypotension and its role in optic nerve head and ocular ischemic disorders. Am J Ophthalmol. 117:603-624, 1994.|
|26||Hayreh SS, March W, Phelps CD. Ocular hypotony following retinal vein occlusion. Arch Ophthalmol. 96:827833, 1978.|
|27||Leber T. Die Krankheite der Netzhaut U. des Schnerven. In: Graefe-Saemisch Handbuch der Cesamten Augenheikunde. Leipzig, 1st ed., vol.5, pp. 531, 1877.|
|28||Koyanagi Y. Die Bedeutung der Gefasskreuzung fur die Entstehung der Astthrombose der retinalen Zentralvene. Klin Mbl Augenheilk. 81:219-231, 1928.|
|29||Duker JS, Brown GC. Anterior location of the crossing artery in branch retinal vein obstruction. Arch Ophthalmol. 107:998-1000, 1989.|
|30||Weinberg D, Dodwell DG, Fern SA. Anatomy of arteriovenous crossings in branch retinal vein occlusion. Am J Ophthalmol. 109:298-302, 1990.|
|31||Feist RM, Richo BH, Shapiro MJ, Farber M. Branch retinal vgin occlusion and quadratic variation in arteriovenous crossings. Am J Ophthalmol. 113:664-668, 1992.|
|32||Zhao J, Sastry SM, Sperduto RD, Chew EY, Remaley NA. Arteriovenous crossing patterns in branch retinal vein occlusion. Ophthalmology. 100:423-428, 1993.|
|33||Jensen VA. Clinical studies of tributary thrombosis in the central retinal veins. II. The ophthalmoscopic distribution of the retinal vessels in the normal eyeground. (Studies of the ramifications of the vessels, their course, and the arterio-venous intersections.) Acta Ophthalmol. 10 (Suppl): 1-193, 1936.|
|34||Ammann E: Die Netzhautblutungen bei Blut - and Gefafserkrankungen. Beit zur Augenheilk. 4:211-280, 1899.|
|35||Weinberg DV, Egan KM. Seddon JM. Asymmetric distribution of arteriovenous crossings in the normal retina. Ophthalmology. 100:31-36, 1993.|
|36||Seitz R. The retinal vessels. St. Louis, CV Mosby, 1964, 20-74.|
|37||Rabinowicz IM, Litman S, Michaelson IC. Branch venous thrombosis - A pathological report. Trans Ophthalmol Soc UK. 88:191-210, 1968.|
|38||Hockley DJ, Tripathi RC, Ashton N. Experimental retinal branch vein occlusion in monkey: Histopathological and ultrastructural studies. Trans Ophthalmol Soc UK. 96:202-209, 1976.|
|39||Magargal LE, Brown CC, Augsburger JJ, Parrish RK II. Neovascular glaucoma following central retinal vein obstruction. Ophthalmology. 88:1095-1101, 1981.|
|40||Magargal LE, Donoso LA, Sanborn GE. Retinal ischemia and risk of neovascularization following central retinal vein obstruction. Ophthalmology. 89:1241-1245, 1982.|
|41||Welch JC, Augsburger JJ. Assessment of angiographic retinal capillary nonperfusion in central retinal vein occlusion. Am J Ophthalmol. 103:761-766, 1987.|
|42||Johnson MA, Marcus S, Elman MJ, McPhee TJ. Neovascularization in central retinal vein occlusion: Electroretinographic findings. Arch Ophthalmol. 106:348352, 1988.|
|43||Kaye SB, Harding SP. Early electroretinography in unilateral central retinal vein occlusion as a predictor of rubeosis iridis. Arch Ophthalmol. 106:353-356, 1988.|
|44||Central Vein Occlusion Study Group. Baseline and early natural history report - The central vein occlusion study. Arch Ophthalmol. 111:1087-1095, 1993.|
|45||Sedney SC. Photocoagulation in retinal vein occlusion. The Hague: W Junk, 1976.|
|46||Kohner EM, Pettilt JE, Hamilton AM, et al. Streptokinase in central retinal vein occlusion: a controlled clinical trial. Br Med J.1:550-553, 1976.|
|47||Kohner EM, Laatikainen L, Oughton J. The management of central vein occlusion. Ophthalmology. 90:484487, 1983.|
|48||Vasco-Posada J. Modification of the circulation in the posterior pole of the eye. Ann Ophthalmol. 4:48-59, 1972.|
|49||Ring CP, Pearson TC, Sanders MD, Wetherley-Mein G. Viscosity and retinal vein thrombosis. Br J Ophthalmol. 60:397-410, 1976.|
|50||Wolf S, Arend 0, Bertram B, Remky A, Schulte K, Wald KJ, et al. Hemodilution therapy in central retinal vein occlusion: One-year results of a prospective randomized study. Graefes Arch Clin Exp Ophthalmol. 232:33-39, 1994.|
|51||Hansen LL, Wiek J, Wiederholt M: A randomised prospective study of treatment of non-ischemic central vein occlusion by isovolaemic haemodilution. Br J Ophthalmol. 73:895-899, 1989.|
|52||Hansen LL, Danisevskis P. Arntz H-R, Hoevener G, Wiederholt M. A randomised prospective study on treatment of central retinal vein occlusion by isovolaemic haemodilution and photocoagulation. Br J Ophthalmol. 69:108-116, 1985.|
|53||Hayreh SS, Klugman MR, Podhajsky P, Servais GE, Perkins ES. Argon laser panretinal photocoagulation in ischemic central retinal vein occlusion - A 10-year prospective study. Graefes Arch Clin Exp Ophthalmol. 228:281-296, 1990.|
|54||Hayreh SS, Rubenstein L, Podhjsky P. Argon laser scatter photocoagulation in treatment of branch retinal vein occlusion - A prospective clinical trial. Ophthalmologica. 206:1-14, 1993.|
|55||Magargal LE, Brown GC, Augsburger JJ, Donoso LA. Efficacy of panretinal photocoagulation in preventing neovascular glaucoma following ischemic central retinal vein obstruction. Ophthalmology. 89:780-784, 1982.|
|56||Laatikainen L, Kohner EM, Khoury D, Blach RK. Panretinal photocoagulation in central retinal venous occlusion: a randomised controlled clinical study. Br J Ophthalmol. 61:741-753, 1977.|
|57||Branch Vein Occlusion Study Group. Argon laser scatter photocoagulation for prevention of neovascularization and vitreous hemorrhage in branch vein occlusion. Arch Ophthalmol. 104:34-41, 1986.|
|58||Branch Vein Occlusion Study Group. Argon laser photocoagulation for macular edema in branch vein occlusion. Am J OphthalmoL 98:271-282, 1984.|
|59||Shilling JS, Jones CA. Retinal branch vein occlusion: A study of argon laser photocoagulation in the treatment of macular edema. Br J Ophthalmol. 68:196-198, 1984.|
|60||Hayreh SS. So-called 'central retinal vein occlusion' 11. Venous stasis retinopathy. Ophthalmologica. 172:14-37, 1976.|
|61||van Heuven WAJ, Hayreh MS, Hayreh SS. Experimental central retinal vascular occlusion: Blood-retinal barrier alterations and retinal lesions. Trans Ophthalmol Soc UK. 97:588-618, 1977.|
|62||Coscas C, Dhermy P. Occlusions veineuses retineinnes. Rapport de la Societe Francaise d'Ophtalmologie. Masson:Paris, 280-281, 1978.|
|63||Leber T. Die Thrombose der Zentralvene and die hamorrhagische Retinitis. In: Graefe-Saemisch-Hess Handbuch der Gesamten Augenheilkunde. Leipzig:Englemann, 2nd ed., vol. 7, 1915, 355-459.|
|64||Liebreich R. Ophthalmoskopische Notizen.3. Apoplexia retinae. Albrecht von Graefes Arch Ophthalmol. 1(2):346-351, 1855.|