**2.1 Pathogenesis**

Retinal venous obstruction occurs as a result of Virchow's triad in the retinal vessels with stasis, hypercoagulability and endothelial change. In the retina, disruption of venous return results in elevated intravascular pressure and culminates in retinal haemorrhage and oedema. Retinal ischaemia then follows, leading to non-perfusion of the capillary beds.

Retinal vein occlusion RVO can be classified as either a branch retinal vein occlusion (BRVO) (Figure 1) or a central retinal vein occlusion (CRVO) (Figure 2) depending upon the location of the obstruction (Hayreh 2005). Each of BRVO and CRVO can be further subclassified into ischaemic and non-ischaemic categories.

Ischaemic RVO is characterised by retinal capillary non-perfusion and results in more severe signs and symptoms including significant decrease in visual acuity, cotton wool spots (indicating retinal ischaemia) and a relative afferent pupil defect. An ischemic RVO is usually associated with ocular complications such as intraocular neovascularisation.

In non-ischaemic RVO stasis of the retinal veins occurs. There is leakage from the capillary bed but capillary perfusion is still present. A non-ischaemic RVO is associated with decreased visual acuity as a result of leaking retinal capillaries that cause macular oedema.

Venous Thrombosis and the Eye 161

Fig. 2. Fundus photograph of a central retinal vein occlusion in the right eye (a) with normal

haemorrhages (H) in all four quadrants of the retina, cotton wool spots (C), and optic nerve

The prevalence of RVO is approximately 1 in 200 people, with an estimated 16.4 million people being affected worldwide (Rogers, McIntosh et al. 2010). Prevalence appears to increase with age (Rogers, McIntosh et al. 2010), whereas gender or ethnicity does not appear to affect prevalence (Rogers, McIntosh et al. 2010). Retinal vein occlusion is usually unilateral and very rarely bilateral. A BRVO is four times more common than a CRVO (Rogers, McIntosh et al. 2010), with its prevalence ranging from 0.3% (Wong, Larsen et al.

left eye (b) of the same patient. In central retinal vein occlusion, there are scattered

swelling (ON), which is characterised by indistinct disc margins (black arrows).

**2.2 Prevalence** 

2005) to 1.1% (Mitchell, Smith et al. 1996).

Fig. 1. Fundus photograph of a superior-temporal branch retinal vein occlusion in the right eye (a) with the corresponding fluorescein angiogram (b). The retina shows a sectoral area of retinal haemorrhages (H) and cotton wool spots (C) in the area distal to the vein occlusion (black arrow).

Fig. 1. Fundus photograph of a superior-temporal branch retinal vein occlusion in the right eye (a) with the corresponding fluorescein angiogram (b). The retina shows a sectoral area of retinal haemorrhages (H) and cotton wool spots (C) in the area distal to the vein occlusion

(black arrow).

Fig. 2. Fundus photograph of a central retinal vein occlusion in the right eye (a) with normal left eye (b) of the same patient. In central retinal vein occlusion, there are scattered haemorrhages (H) in all four quadrants of the retina, cotton wool spots (C), and optic nerve swelling (ON), which is characterised by indistinct disc margins (black arrows).

#### **2.2 Prevalence**

The prevalence of RVO is approximately 1 in 200 people, with an estimated 16.4 million people being affected worldwide (Rogers, McIntosh et al. 2010). Prevalence appears to increase with age (Rogers, McIntosh et al. 2010), whereas gender or ethnicity does not appear to affect prevalence (Rogers, McIntosh et al. 2010). Retinal vein occlusion is usually unilateral and very rarely bilateral. A BRVO is four times more common than a CRVO (Rogers, McIntosh et al. 2010), with its prevalence ranging from 0.3% (Wong, Larsen et al. 2005) to 1.1% (Mitchell, Smith et al. 1996).

Venous Thrombosis and the Eye 163

and non-ischemic types of RVO (Figure 3). Optical coherence tomography (OCT) is another investigation that is becoming increasingly important in assessing the changes in retinal architecture and measuring the thickness of the retinal layers. OCT can assist in detecting

Fig. 3. Fluorescein angiogram of a non-ischaemic central retinal vein occlusion (a) compared to an ischaemic central retinal vein occlusion (b). In non-ischaemic central retinal vein occlusion, there is effective capillary perfusion and this is seen as white due to the

fluorescein filling the vessels. In ischaemic central retinal vein occlusion, there are areas of capillary non-perfusion (I) and seen as black due to the absence of fluorescein filling the

vessels.

and quantifying intraretinal cysts and the associated macular oedema (Figure 4).

### **2.3 Risk factors**

The risk factors for RVO include systemic or local conditions which result in vascular stasis or endothelial damage. Systemic risk factors for RVO are often associated with atherosclerotic vessel changes, and these risk factors include age (Hayreh, Zimmerman et al. 1994), hypertension (Mitchell, Smith et al. 1996; Hayreh, Zimmerman et al. 2001) and hyperlipidaemia (1993; 1996). Diabetes mellitus (Dodson, Kritzinger et al. 1992; Hayreh, Zimmerman et al. 2001), peripheral vascular disease and cerebrovascular disease also contribute to a higher likelihood of developing RVO.

Less common systemic risk factors for RVO include those associated with hypercoagulability. These include myeloproliferative disorders (Fegan 2002) such as polycythaemia, myeloma and Waldenstrom macroglobulinaemia. These account for 1% of RVO cases. Acquired and inherited hypercoagulable states (Fegan 2002) may also predispose a patient to RVO. Acquired diseases that have been shown to contribute to RVO include hyperhomocysteinaemia (Chua, Kifley et al. 2005; Janssen, den Heijer et al. 2005), anti-cardiolipin antibodies (Janssen, den Heijer et al. 2005), lupus anticoagulant and antiphospholipid antibodies. Inherited hypercoagulable diseases that have been associated with RVO include factor V Leiden mutation, protein C or S deficiency (Greiner, Hafner et al. 1999), anti-thrombin deficiency, prothrombin gene mutation (Incorvaia, Lamberti et al. 1999)and factor XII deficiency (Incorvaia, Lamberti et al. 1999). Inflammatory diseases such as Behçet's disease, sarcoidosis, Wegener's granulomatosis and Goodpasture syndrome and chronic renal failure have also been proven to be risk factors for RVO. Oestrogen therapy may be a cause of RVO, although the evidence is inconclusive (Kirwan, Tsaloumas et al. 1997). Hormone replacement therapy has not been shown to be a major risk factor for RVO, whereas the contraceptive pill does increase the risk of RVO and is contraindicated in those with RVO.

A local risk factor for the development of a CRVO is primary open angle glaucoma (1996; Mitchell, Smith et al. 1996). This is due to stasis caused by the elevated intraocular pressure in glaucoma reducing arterial perfusion. The mean arterial perfusion pressure is the difference between the mean arterial pressure and the intraocular pressure.

### **2.4 Investigations**

All investigations should be directed towards examining the systemic risk factors of RVO and assessing the ocular complications of neovascularization.

Systemic investigations include screening for atherosclerotic disease and hypercoagulability. This involves measuring blood pressure, electrocardiogram, a full blood count, measuring erythrocyte sedimentation rate, fasting glucose and lipids, a vasculitic screen and a thrombophilia screen. A vasculitic screen involves measuring antinuclear antibodies, rheumatoid factor, anti-neutrophil cytoplasmic antibody, serum protein electrophoresis, complement 3 and 4. Whereas a thrombophilia screen would include measuring serum homocysteine, protein C and S, Factor V Leiden, antithrombin, anticardiolipin antibodies and lupus anticoagulant.

It is important to remember that in patients with BRVO, cardiovascular risk factors are the predominant cause. In contrast, in patients with CRVO, especially when the patient's age is less than 65, there should be a more extensive search for other risk factors such as oral contraceptive use, thrombophilia or inflammatory diseases.

Ocular investigation with a fundus fluorescein angiography allows for visualisation of large retinal vessels and retinal capillary beds. This helps to differentiate between the ischemic

The risk factors for RVO include systemic or local conditions which result in vascular stasis or endothelial damage. Systemic risk factors for RVO are often associated with atherosclerotic vessel changes, and these risk factors include age (Hayreh, Zimmerman et al. 1994), hypertension (Mitchell, Smith et al. 1996; Hayreh, Zimmerman et al. 2001) and hyperlipidaemia (1993; 1996). Diabetes mellitus (Dodson, Kritzinger et al. 1992; Hayreh, Zimmerman et al. 2001), peripheral vascular disease and cerebrovascular disease also

Less common systemic risk factors for RVO include those associated with hypercoagulability. These include myeloproliferative disorders (Fegan 2002) such as polycythaemia, myeloma and Waldenstrom macroglobulinaemia. These account for 1% of RVO cases. Acquired and inherited hypercoagulable states (Fegan 2002) may also predispose a patient to RVO. Acquired diseases that have been shown to contribute to RVO include hyperhomocysteinaemia (Chua, Kifley et al. 2005; Janssen, den Heijer et al. 2005), anti-cardiolipin antibodies (Janssen, den Heijer et al. 2005), lupus anticoagulant and antiphospholipid antibodies. Inherited hypercoagulable diseases that have been associated with RVO include factor V Leiden mutation, protein C or S deficiency (Greiner, Hafner et al. 1999), anti-thrombin deficiency, prothrombin gene mutation (Incorvaia, Lamberti et al. 1999)and factor XII deficiency (Incorvaia, Lamberti et al. 1999). Inflammatory diseases such as Behçet's disease, sarcoidosis, Wegener's granulomatosis and Goodpasture syndrome and chronic renal failure have also been proven to be risk factors for RVO. Oestrogen therapy may be a cause of RVO, although the evidence is inconclusive (Kirwan, Tsaloumas et al. 1997). Hormone replacement therapy has not been shown to be a major risk factor for RVO, whereas the contraceptive pill does increase the risk of RVO and is contraindicated in those

A local risk factor for the development of a CRVO is primary open angle glaucoma (1996; Mitchell, Smith et al. 1996). This is due to stasis caused by the elevated intraocular pressure in glaucoma reducing arterial perfusion. The mean arterial perfusion pressure is the

All investigations should be directed towards examining the systemic risk factors of RVO

Systemic investigations include screening for atherosclerotic disease and hypercoagulability. This involves measuring blood pressure, electrocardiogram, a full blood count, measuring erythrocyte sedimentation rate, fasting glucose and lipids, a vasculitic screen and a thrombophilia screen. A vasculitic screen involves measuring antinuclear antibodies, rheumatoid factor, anti-neutrophil cytoplasmic antibody, serum protein electrophoresis, complement 3 and 4. Whereas a thrombophilia screen would include measuring serum homocysteine, protein C and S, Factor V Leiden, antithrombin, anticardiolipin antibodies

It is important to remember that in patients with BRVO, cardiovascular risk factors are the predominant cause. In contrast, in patients with CRVO, especially when the patient's age is less than 65, there should be a more extensive search for other risk factors such as oral

Ocular investigation with a fundus fluorescein angiography allows for visualisation of large retinal vessels and retinal capillary beds. This helps to differentiate between the ischemic

difference between the mean arterial pressure and the intraocular pressure.

and assessing the ocular complications of neovascularization.

contraceptive use, thrombophilia or inflammatory diseases.

**2.3 Risk factors** 

with RVO.

**2.4 Investigations** 

and lupus anticoagulant.

contribute to a higher likelihood of developing RVO.

and non-ischemic types of RVO (Figure 3). Optical coherence tomography (OCT) is another investigation that is becoming increasingly important in assessing the changes in retinal architecture and measuring the thickness of the retinal layers. OCT can assist in detecting and quantifying intraretinal cysts and the associated macular oedema (Figure 4).

Fig. 3. Fluorescein angiogram of a non-ischaemic central retinal vein occlusion (a) compared to an ischaemic central retinal vein occlusion (b). In non-ischaemic central retinal vein occlusion, there is effective capillary perfusion and this is seen as white due to the fluorescein filling the vessels. In ischaemic central retinal vein occlusion, there are areas of capillary non-perfusion (I) and seen as black due to the absence of fluorescein filling the vessels.

Venous Thrombosis and the Eye 165

Fig. 5. Fundus photograph of an inferior branch retinal vein occlusion in the left eye showing different types of retinal haemorrhages. Superficial retinal haemorrhages (S) are

Similar to BRVO, the most common presentation in patients with CRVO is vision loss. The vision loss is sudden, unilateral, painless and usually profound, in the range of counting

Examination reveals a relative afferent Pupillary defect. While fundus examination demonstrates tortuosity and dilatation of all the branches of the central retinal vein as well as superficial (flame-shaped) and deep (dot-blot) intraretinal haemorrhages. Cotton wool

Fluorescein angiography is the ocular investigation of choice in patients with suspected CRVO. The general features of CRVO using fluorescein angiography include delayed arteriovenous transit time and blockage by retinal haemorrhages. Fluorescein angiography is vital in distinguishing between non-ischaemic and ischaemic CRVO (Figure 3). There is effective capillary perfusion in non-ischaemic CRVO and capillary non-perfusion in

Treatment involves management of the systemic risk factors and local treatment for

With regard to systemic treatment of RVO, there is no strong evidence in current literature for the use of anticoagulant, fibrinolytic agents or antiplatelet agents. Best practice management involves treating the atherosclerotic risk factors, such as hypertension and

Local treatment for BRVO differs from the treatment of CRVO. For BRVO, the Branch Vein Occlusion Study (BVOS) suggested that retinal haemorrhages and any macular oedema should be observed for 3 months to allow for spontaneous resolution. After 3 months, if the macular oedema is still present, treatment may be warranted. The BVOS recommends grid laser photocoagulation for the treatment of macular oedema (1984). Use of intravitreal anti-VEGF agents, such as ranibizumab, has been shown to be beneficial (Campochiaro, Heier et

flame shaped; deep retinal haemorrhages (D) are dot-blot.

spots and oedema of the optic disc or macular may be present.

complications of RVO, such as macular oedema and neovascularisation.

**2.5.2 Central retinal vein occlusion** 

fingers.

ischaemic CRVO.

**2.6 Treatment** 

hyperlipidaemia.

Fig. 4. Optical coherence tomography scan of the macular in an eye with retinal vein occlusion. Nasal to temporal (N→T) cross-sectional image of the macula demonstrates the presence of significant macula oedema and associated intra-retinal cysts (white arrow).
