Introduction

The orbit is a window to systemic disease [1]. A range of vascular, neoplastic and autoimmune diseases may have orbital manifestations. Likewise, a dilated superior ophthalmic vein (SOV) can occur secondary to an array of conditions in the brain or orbit including vascular diseases e.g., carotid-cavernous fistula, venous thrombosis; inflammatory conditions e.g., idiopathic orbital inflammation and thyroid eye disease; and may be an early indicator of raised intracranial pressure [2,3,4].

In the absence of any underlying pathology, the mean diameter of the SOV is approximately two millimetres [5,6,7,8,9]. The authors herein review cases of dilated SOVs in the absence of orbital, cavernous sinus or neurological disease.

Methods

Patient population

We conducted a retrospective review of asymptomatic patients who were noted to have a dilated superior ophthalmic vein with a diameter of \(\ge\)5.0 mm on computed tomography (CT) or magnetic resonance imaging (MRI) of the orbits. Patients with a dilated SOV secondary to orbital or cavernous sinus disease were excluded, as were patients with signs or symptoms of raised intracranial pressure. Patient demographics, past medical history, presenting complaint, relevant transthoracic echocardiogram (TTE) findings, indication for neuroimaging, and the SOV diameters were recorded. The study was approved by the Central Adelaide Local Health Network ethics committee and adhered to the principles of the Declaration of Helsinki.

Imaging

All patients were evaluated using Magnetom 3 T Skyra scanner (Siemens, Germany) or SOMATOM Force CT (Siemens AG, Germany). The maximum diameter of the SOV was taken perpendicular to the long axis of the SOV on T2 coronal sequences or coronal CT scan (Fig. 1) [2]. T2 coronal sequences are used in line with previous reports [2, 5]. In patients who did not have scans following the dilated scan result, the SOV sizes from any previous scans were recorded. All measurements were performed on high resolution picture archiving and communication system (PACS) under the supervision of a consultant neuroradiologist (SP).

Fig. 1
figure 1

The Superior ophthalmic vein measurement was taken perpendicular to the long axis of the SOV on T2-fat suppressed coronal sequences

Full size image

The normal diameter of the SOV is approximately 2 mm [5, 6, 8, 10]. We used a value of 5 mm to define a dilated SOV to ensure we captured cases that would fall outside of the normal two standard deviation variation and would otherwise raise suspicion for a pathological underlying cause [6, 8].

Results

We identified nine patients with dilated SOVs. Six (67%) patients were female, and the mean age was 72 years (range 58–89). Patient demographics are provided in Table 1. The maximum diameters of the SOV on the initial and other scans are provided in Table 2. Five patients had follow-up imaging after the dilated SOV scan, whereas neuroimaging from before the dilated SOV scan was evaluated in three other cases. Small fluctuations within the size of the SOV may be attributed to dynamic changes in venous pressures. Patients who had follow up scans after the dilated SOV scan did not show a significant increase in the size of the SOV and remained asymptomatic without any visual complications.

Table 1 Clinical profile of patients with dilated superior ophthalmic vein
Full size table
Table 2 Superior ophthalmic vein measurements
Full size table

Five patients had a history of significant cardiovascular disease. Four patients had previous or current presentations with myocardial infarction requiring revascularisation (coronary artery bypass graft [CABG] or percutaneous coronary intervention [PCI]). One patient had a history of peripheral vascular disease requiring a stent.

Three cases (1, 6, 7) had evidence of ‘reversible’ SOV dilatation likely secondary to raised venous pressures. Case 1 presented with decompensated right heart failure with ascites, peripheral oedema, and facial oedema. Case 7 presented with bilateral pulmonary emboli, in the context of metastatic breast cancer complicated by a malignant pericardial effusion with impending tamponade. They had bilateral SOV dilatation which resolved to within normal limits on follow-up scans (Fig. 2). Similarly, case 6 presented with a myocardial infarction with the TTE showing evidence of left ventricle dysfunction and had an elevated B-type natriuretic peptide (10,784). A review of this patient’s imaging from five years prior revealed that he had pre-existing, although less severe SOV dilatation (left 4.6 mm, right 5.9 mm) that was likely exacerbated by his current presentation of left ventricular dysfunction.

Two cases (cases 3, 5) had significant venous clotting risk factors. Case 3 had a history of Factor V Leiden, whilst Case 5 had a history of bilateral branch retinal vein occlusion (BRVO).

Discussion

This is the first study looking at dilated SOVs in the absence of orbital, cavernous sinus or neurological disease. All the patients were asymptomatic from an ophthalmic perspective and patients with follow up scans after the dilated scan did not show significant progression in this size of the SOV. Analysis of these cases may offer some insights into the potential pathophysiology of dilated SOVs and may reveal an association between cardiovascular disease and dilated SOVs.

Elevated venous pressures may explain the dilated SOVs in three of our patients. These patients presented with decompensated right heart failure (case 1), acute anterior myocardial infarction (case 6) and pericardial effusion (case 7). All of these patients had bilateral SOV dilatations. All of these patients had symptomatic and severe cardiac failure. Case 1 presented with decompensated right heart failure with peripheral and facial oedema such that imaging was performed to exclude superior vena cava obstruction. Case 6 had an acute anterior myocardial infarction. Case 7 had a pericardial effusion, which two days after the CT brain, was urgently drained via pericardiocentesis as it caused a cardiac tamponade. Cases 1 and 7 had follow up scans which showed resolution of the SOV dilatation, likely due to treatment of the underlying cause (decompensated heart failure, pericardial effusion). The SOV dilatation was likely transient and in response to elevated venous pressures secondary to acute cardiac disease. Following stabilisation of the acute disease, the SOV sizes reduced. The other cases where unilateral dilatation was seen may have a different underlying pathophysiology (Fig. 3).

Fig. 2
figure 2

Initial T1-Weighted coronal MRI (A) from Case 1 showing bilateral SOV dilatation. This patient presented with decompensated right heart failure. A follow up scan (B) two years later revealed normal SOVs. Initial coronal CT scan (C) showing bilateral SOV dilatation in Case 7. This patient had a pericardial effusion secondary to breast cancer. A follow up CT scan (D) showed normal SOVs

Full size image

Five patients had a significant history of ischaemic heart disease or peripheral vascular disease. It has been suggested that both arterial and venous dilating disease may arise from a common vascular wall pathology [11]. Risk factors for arterial aneurysms (e.g. hypertension, hypercholesterolemia, male sex, increasing age, smoking) may also play a role in the vascular wall degeneration and dilatation of venous vessels [12]. Increased nitric oxide stimulation plays a role in the pathogenesis of arterial dilatations such as abdominal aortic aneurysms, and has also been linked to the development of venous dilatations including varicose veins [13]. (Fig. 3).

Fig. 3
figure 3

Initial (A) and follow up (B) coronal T2 fat suppressed scans in case 2 showing a left dilated SOV. Initial CT (C) and follow up coronal MRI (D) showing a left dilated SOV in case 3. Case 4 shows a right dilated SOV on initial (E) and follow up (F) scans

Full size image

Local factors may also play a role in the pathophysiology of dilated SOVs. Valvular incompetence is implicated in the development of venous dilatations in the lower limbs [13]. Although the SOV has traditionally been thought to be valveless [6, 14, 15], a cadaveric study analysing twelve specimens of the SOV has found evidence of valves in four of the twelve specimens [16]. Valvular incompetence may have a role to play in the development of dilated SOVs, similar to other areas of the body. Additionally, valvular incompetence is usually asymmetric, potentially explaining the unilateral dilatation in some of our cases.

Two cases (cases 3, 5) had significant venous clotting risk factors. Case 3 had a history of Factor V Leiden, whilst Case 5 had a history of bilateral BRVO. Patients with procoagulant risk factors such as active cancer or Factor V leiden carriers have been reported to have asymptomatic thromboses in other areas of the body [17, 18]. It may be the case that our two patients may have had asymptomatic thromboses of the superior ophthalmic vein, that may have led to the dilated SOVs. There was however no evidence of an acute thrombus.

This study has some limitations. This is a retrospective study on a small sample size, due to the rarity of this condition. Further cases from additional centres may help to clarify the potential role of cardiovascular disease in the development of a dilated SOV, in patients who are otherwise asymptomatic from an ocular perspective.

Ophthalmologists may be consulted to give an opinion on incidental findings such as SOV dilatation. SOV dilatation may simply reflect elevated venous pressures secondary to cardiac failure and may self-resolve with treatment of the underlying condition. Other cases of dilated SOV may be seen in patients with significant ischaemic heart disease, peripheral vascular disease, or venous clotting history. These patients may be safely observed as long as they remain asymptomatic.