Abstract
Background
Although the abnormal venous drainage into the superior ophthalmic vein (SOV) is a well-known entity responsible for ocular symptoms, it remains unclear to what degree it affects the visual function. The purpose of this study was to evaluate the incidence, characteristics and outcomes of the visual disorders in patients with intracranial arteriovenous fistula (AVF) with venous drainage into the SOV.
Methods
This retrospective study involved eight patients diagnosed with intracranial AVFs with abnormal venous drainage into the SOV between January 2014 and December 2016.
Results
The most common location of AVF was the cavernous sinus (CS) in five patients, followed by the intraorbit in two patients and superior sagittal sinus (SSS) in one patient. Visual disorders were detected in three patients (two intraorbit and one CS). The visual field contraction was observed in a patient with intraorbital AVF, and the reduction of visual acuity was confirmed in another patient with intraorbital AVF and a patient with CS dural AVF. All patients underwent an interventional treatment consisting of endovascular embolisation, stereotactic radiosurgery or both, which was selected based on their angioarchitecture. Although angiographic cure of AVF was confirmed in all patients, visual function did not fully recover in two patients with intraorbital AVF.
Conclusions
In cases of intraorbital AVF, visual disorders are more frequent and can result in poorer outcomes compared with other dural AVFs with drainage into the SOV. Early diagnosis and treatment are crucial to preserve the visual function of patients with intraorbital AVF.
Keywords: Visual disorders, superior ophthalmic vein, arteriovenous fistula, intraorbit
Introduction
The superior ophthalmic vein (SOV) is one of the dominant draining pathways of various intracranial arteriovenous fistulas (AVFs). Intracranial AVFs with venous drainage into the SOV can be classified into the following two subtypes: the AVF involving the SOV as a sheer draining pathway (indirect AVF involving the SOV) and the AVF involving the SOV as the fistulous site (direct AVF involving SOV). In the former subtype, the SOV acts as part of the draining system. The cavernous sinus (CS) dural arteriovenous fistula (DAVF) is the most representative example,1 and other DAVFs were also reported involving venous reflux into the SOV.2,3 In the latter subtype, the SOV is both the fistulous site and draining pathway, and the intraorbital AVF is a typical and sole example.4–16 The venous drainage into the SOV is a well-known entity responsible for ocular symptoms (e.g., chemosis and exophthalmos);1–3 however, it remains unclear to what degree it affects the outcomes of visual function.
The purpose of the present study was to evaluate the incidence, characteristics and outcomes of the visual disorders in the cases of intracranial AVF with venous drainage into the SOV.
Material and methods
Patient selection
This study is a retrospective review based on the prospectively collected clinical database. We have maintained an ongoing prospective database on patients who underwent cerebral digital subtraction angiography (DSA) from January 2014. The collected data include demographic, clinical and angiographic information. Between January 2014 and December 2016, 12 patients with cranial AVFs were diagnosed using DSA. Of these patients, eight patients who had been angiographically confirmed to have venous drainage into the SOV were retrospectively investigated. All patients included in this study provided written informed consent for receiving an interventional treatment and oral consent to participate in the study.
Pretherapeutic clinical evaluation
All patients underwent neurological examination by a neurologist. According to the ocular and visual symptoms of each patient, the ophthalmological inspections including visual acuity measurement, visual field test, intraocular pressure measurement and fundus examination were performed.
Pretherapeutic angiographic evaluation
The individual angioarchitecture including feeding arteries, site of fistulas and draining veins were evaluated using a six-vessel DSA. We classified the eight eligible AVFs into two subtypes as follows based on the location of the fistula: the direct AVF involving the SOV and the indirect AVF involving the SOV. ‘The direct AVF involving SOV’ was defined as the cranial AVF having fistula on the SOV itself. ‘The indirect AVF involving the SOV’ was defined as the cranial AVF having fistula on the sites other than the SOV and using the SOV as a sheer draining pathway.
Treatment
All of the eight patients were considered to be not eligible for conservative therapy because of their aggressive symptoms or dangerous angioarchitecture; hence, they underwent an interventional treatment consisting of endovascular embolisation, stereotactic radiosurgery or both, which was selected based on their angioarchitecture.
Post-therapeutic follow-up
All patients were followed up clinically at least 3 and 6 months after treatment. Detailed ophthalmological examinations were performed as needed. Angiographic follow-up was performed between 6 and 12 months after treatment. The clinical follow-up was continued until the symptoms clinically stabilised and AVF was angiographically cured. Each patient was examined by the same neurologist during follow-up.
Outcome measurements
Visual disorders were defined as visual acuity reduction or visual field contraction proven by both subjective symptoms and objective inspections. Based on the follow-up ophthalmological examinations, outcomes of visual functions were divided into the following three categories: improved, unchanged and deteriorated.
Results
Demographic data and angiographic findings
The demographic data and location of AVF of the eight eligible patients are summarised in Table 1. The mean age was 69.9 years (range, 56−81 years) with a female preponderance (87.5%). The most common location of AVF was the CS in five patients, followed by the intraorbit in two patients and superior sagittal sinus (SSS) in one patient. According to the previously mentioned definition, two cases with intraorbital AVF are classified as direct AVF involving the SOV, and six cases with CS or SSS DAVF are classified as indirect AVF involving the SOV.
Table 1.
Patient demographics (N = 8).
| Age, years, mean (range) | 69.9 (56–81) |
| Sex, No. (%) | |
| Female | 7 (87.5) |
| Male | 1 (12.5) |
| Site of fistulas, No. (%) | |
| Cavernous sinus (indirect) | 5 (62.5) |
| Intraorbit (direct) | 2 (25.0) |
| Superior sagittal sinus (indirect) | 1 (12.5) |
direct: direct AVF involving the SOV. indirect: indirect AVF involving the SOV.
Visual disorders
The details of visual disorders and ocular symptoms are described in Table 2. The incidence of visual disorders and ocular symptoms for each fistulous site is summarised in Table 3. Visual disorders were detected in only three of the eight patients, whereas the ocular symptoms were exhibited by all of the eight patients (Tables 2 and 3). Visual field contraction was observed in a patient with intraorbital AVF (Case 1 in Table 2), and the reduction of visual acuity was confirmed in another patient with intraorbital AVF (Case 2 in Table 2) and a patient with CS dural AVF (Case 3 in Table 2).
Table 2.
Demographic data, angiographic findings and initial symptoms of eight patients with SOV drainage.
| No. | Age (y) | Sex | Site of fistula | Classification | Initial symptoms |
Initial ophthalmological inspections |
||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Visual disorders (details) | Ocular symptoms (details) | Intraocular pressure (mmHg) |
Visual acuity |
|||||||||
| Right | Left | Right | Left | |||||||||
| 1 | 69 | Female | Left IO | direct | + | (Visual field contraction) | + | (Diplopia/oculomotor and abducens nerve palsy) | Uninspecteda | Uninspecteda | ||
| 2 | 56 | Male | Left IO | direct | + | (Visual acuity reduction) | + | (Chemosis/proptosis) | 11.0 | 17.7 | 20/133 | 20/3b |
| 3 | 75 | Female | Left CS | indirect | + | (Visual acuity reduction) | + | (Chemosis/proptosis/diplopia/ oculomotor nerve palsy) | 13.0 | 18.3 | 20/17 | 20/222 |
| 4 | 67 | Female | Left CS | indirect | − | + | (Chemosis/proptosis/diplopia/ oculomotor nerve palsy) | Uninspectedc | Uninspectedc | |||
| 5 | 69 | Female | Bilateral CS | indirect | − | + | (Chemosis/proptosis/diplopia/ oculomotor nerve palsy) | Uninspectedc | Uninspectedc | |||
| 6 | 76 | Female | Left CS | indirect | − | + | (Chemosis/proptosis/diplopia/ oculomotor nerve palsy) | 13.0 | 19.7 | 20/22 | 20/25 | |
| 7 | 81 | Female | Right CS | indirect | − | + | (Chemosis/proptosis) | 21.0 | 15.7 | 20/40 | 20/29 | |
| 8 | 66 | Female | SSS | indirect | − | + | (chemosis/proptosis) | 17.7 | 17.4 | 20/40 | 20/40 | |
IO: intraorbit; CS: cavernous sinus; SSS: superior sagittal sinus.
direct: direct AVF involving the SOV; indirect: indirect AVF involving the SOV
No pretherapeutic measurement of visual acuity and intraocular pressure due to the urgent need of the endovascular embolisation.
Subjective blurred vision of left eye for 2 months in Case 2.
No subjective visual acuity reduction or visual field contraction in Case 4 and 5.
Table 3.
Ocular symptoms and visual disorders on each fistulous site.
| Site of fistulas |
||||
|---|---|---|---|---|
|
indirect
|
direct
|
Total (n = 8) | ||
| CS (n = 5) | SSS (n = 1) | IO (n = 2) | ||
| Ocular symptoms, No. (%) | 5 (100) | 1 (100) | 2 (100) | 8 (100) |
| Visual disorders, No. (%) | 1 (20) | 0 | 2 (100) | 3 (37.5) |
IO: intraorbit; CS: cavernous sinus; SSS: superior sagittal sinus.
direct: direct AVF involving the SOV; indirect: indirect AVF involving the SOV.
In terms of fistulous site, both patients with the direct AVF involving the SOV (intraorbital AVF) had visual disorders, whereas only one of the six patients with the indirect AVF involving the SOV (five CS DAVF and one SSS DAVF) experienced visual disorders.
Ocular symptoms
All patients in the study had ocular symptoms (Tables 2 and 3). Seven patients had both chemosis and proptosis resulting from reflux of the ipsilateral SOV (Table 2), whereas the remaining patient with intraorbital AVF, whose affected SOV was antegradely draining, exhibited only external ophthalmoplegia without chemosis and proptosis (Case 1 in Table 2).
Treatments and outcomes
Treatments and outcomes of the three patients exhibiting visual disorders are summarised in Table 4. The treatment modalities were selected on the basis of the angioarchitecture of individual AVFs. The patient with intraorbital AVF presenting with visual field contraction underwent transarterial embolisation (TAE) (Case 1 in Tables 2 and 4); another patient with intraorbital AVF presenting with visual acuity reduction underwent stereotactic radiosurgery (SRS) (Case 2 in Tables 2 and 4); and a patient with CS dural AVF had transvenous embolisation (TVE) (Case 3 in Tables 2 and 4). In all three patients, complete obliteration of AVF was angiographically confirmed, within 14 months after the onset of initial symptoms. However, only the patient with CS DAVF had improved visual function (Case 3). In two patients with intraorbital AVF, visual function did not fully recover.
Table 4.
Treatments and outcomes of three patients who experienced visual disorders.
| No. | Age (y)/sex | Site of fistula | Initial visual disorders | Treatments |
Outcomes of visual function |
|||
|---|---|---|---|---|---|---|---|---|
| Modality | Result | Durationa (months) | Follow-upb (months) | Visual disorders | ||||
| 1 | 69/female | IO | Visual field contraction | TAE | CO | 14 | 28 | Unchanged |
| 2 | 56/male | IO | Visual acuity reduction | SRS | CO | 12 | 24 | Deteriorated |
| 3 | 76/female | CS | Visual acuity reduction | TVE | CO | 10 | 6 | Improved |
Each case number corresponds to that in Table 1.
CS: cavernous sinus; CO: complete obliteration angiographically confirmed; IO: intraorbit; SRS: stereotactic radiosurgery; TAE: transarterial embolization; TVE: transvenous embolization.
Duration between onset of symptoms and angiographic complete obliteration of AVF.
Period between end of treatment and last clinical follow-up.
Representative cases
Case 1 (Tables 2 and 4)
A 69-year-old woman had a 4-month history of diplopia and blurred vision of the left eye prior to visiting our department. The initial ophthalmological examinations showed left oculomotor nerve palsy and severe visual field contraction of the left eye, although fundus examination (inspection only, no photographic evidence) revealed no abnormal findings. The initial visual acuity and intraocular pressure were not described in her medical record. Magnetic resonance imaging and angiography showed multiple AVFs including the left SOV and left CS. The intraorbital AVF consisted of an extremely high-flow shunt supplied by several branches mainly arising from the third segment of the left ophthalmic artery, and antegradely drained into the left CS through the left SOV (Figure 1(a) and (b)). The left CS was another shunting point of multiple AVFs. It was also the drainage pathway of both intraorbital AVF and normal venous return from the left superficial middle cerebral veins. The visual field contraction of the left eye had been rapidly deteriorating during the investigation period and resulted in temporary blindness. The urgent TAE was performed to decrease the rate of blood flow shunted to the left SOV. N-butyl cyanoacrylate (NBCA) with a concentration of 20% was injected through a Marathon microcatheter (Medtronic, Irvine, CA, USA) that was advanced into the end of the third segment of the left ophthalmic artery (Figure 1(c)). Two main feeding arteries were occluded, and shunting blood flow drastically decreased, although the complete obliteration of AVF could not be achieved. We decided to discontinue the embolisation, because the residual feeders arose adjacent to the origin of the left central retinal artery. The visual field contraction partially improved immediately after TAE. We selected gamma-knife radiosurgery (GKRS), rather than TVE, as the treatment for left CS DAVF, because left CS acted as drainage of the residual left intraorbital AVF and part of the normal cerebral venous return. Two months after the TAE, GKRS was performed for the left CS DAVF. The follow-up angiography obtained 10 months after TAE (14 months after the onset of the symptoms) showed complete obliteration of both the left intraorbital AVF and the left CS dural AVF (Figure 1(d)). However, left optic disc pallor was observed during fundus examination at 12 months after TAE (Figure (e) and (f)), and the visual field contraction of her left eye lasted until the last follow-up at 28 months after TAE (Figure 1(g) and (h)).
Figure 1.
Case 1: Serial cerebral angiography (CAG) of the left carotid artery showing treatment progress (a–d) and findings of ophthalmological examination (e–h). (a) Initial CAG. The intraorbital arteriovenous fistula (AVF) consisting of extremely high-flow shunt is shown. The shunt is supplied by several branches (arrowheads) arising from the third segment of the enlarged left ophthalmic artery, and antegradely drains into the left cavernous sinus (asterisk). (b) Superselective angiography of the left ophthalmic artery. A Marathon microcatheter is advanced to the distal end in the third segment of the left ophthalmic artery (arrow). A selective injection of a contrast medium via the Marathon microcatheter reveals the upward component of the fistula fed by tiny branches arising from the ophthalmic artery (arrowhead) and draining into the superior ophthalmic vein (SOV) (asterisk). (c)The last phase of N-butyl cyanoacrylate (NBCA) injection via the Marathon microcatheter. NBCA with concentration of 20% is injected and occludes both upward and downward components of the feeding arteries (arrowheads). (d) CAG obtained 10 months after transarterial embolization. The cure of intraorbital AVF is angiographically confirmed. The choroidal brush of the left eye is clearly observed. (e) Fundus examination of right eye obtained at 12 months after TAE or at 2 months after the angiographic cure of AVF. No abnormal findings. (f) Fundus examination of left eye obtained at 12 months after TAE or at 2 months after the angiographic cure of AVF. Pallor and excavation of optic disc are observed, indicating prolonged ocular hypertension. (g) Visual field examination of the left eye obtained at 6 months after TAE. Visual field of the left eye remains contracted. (h) Visual field examination of the left eye obtained at 28 months after TAE or 18 months after angiographic cure of AVF. Visual field contraction did not improve at all.
Case 2 (Tables 1 and 4)
A 56-year-old man presented with exophthalmos and chemosis of the left eye. He had a 2-month history of left oculomotor palsy prior to his consultation in our department. He complained of blurred vision of the left eye. In fact, his best corrected visual acuity was 20/33 and 20/133 in the left and right eyes, respectively. Ophthalmological inspections revealed relative ocular hypertension of the left eye (left 17.7 mmHg vs right 11.0 mmHg) and severe venous congestion in the left ocular fundus resulting in haemorrhage (Figure 2(c)). Angiography showed a left intraorbital AVF fed by a single short branch arising from the first segment of the left ophthalmic artery (Figure 2(a)). The fistula was located on the confluence of the left SOV and inferior ophthalmic vein (IOV) (Figure 2(a)). The SOV that acted as a sole draining vein was notably stagnated possibly because of thrombosis (Figure 2(a)). We judged that any endovascular embolisation was not appropriate for the patient because of the following reasons, and we considered stereotactic radiosurgery. Specifically, TAE including the first portion of the ophthalmic artery might cause occlusion of the central retinal artery resulting in visual loss, and TVE was likely to be impossible due to the isolated SOV. The patient underwent GKRS at a marginal dose of 18.0 Gy on the prescribed 70% isodose line and a maximal dose of 25.7 Gy with a volume of 52.3 mm3. The follow-up angiography obtained 9 months after GKRS (12 months after the onset of the symptoms) revealed complete obliteration of the AVF (Figure 2(b)). The resolution of his ocular hypertension was also confirmed during the angiographic follow-up. The intraocular pressure of the left eye decreased from 17.7 mmHg at baseline to 10.7 mmHg at 9 months after GKRS. However, the venous congestion and haemorrhage in the left ocular fundus gradually progressed (Figure 2(d)), and visual acuity has declined from 20/33 to 20/200 at 9 months follow-up. His visual acuity did not recover at 24 months follow-up.
Figure 2.
Case 2: Serial cerebral angiography (CAG) of the left carotid artery and fundus examination of the left eye. (a) Initial CAG. Left carotid angiogram shows low-flow shunt supplied by a tiny short artery arising from the first segment of the left ophthalmic artery (arrowhead). The shunt is located on a confluence of the left SOV and inferior ophthalmic vein (IOV) (asterisk). The SOV as a sole draining vein is stagnated possibly because of thrombosis (arrows). (b) CAG obtained at 9 months after gamma-knife radiosurgery (GKRS). The cure of intraorbital AVF is angiographically confirmed. (c) Initial fundus inspection. Severe venous congestion and multiple bleeding are demonstrated. (d) Fundus examination obtained at 3 months after GKRS. Haemorrhagic regions increased compared with those in the initial examination.
Case 3
A 75-year-old woman diagnosed with left CS DAVF had left facial pain mimicking trigeminal neuralgia, diplopia, and exophthalmos and chemosis of the left eye for 9 months. She was not aware of her eyesight deterioration. Ophthalmological examinations revealed visual acuity reduction of the left eye (left vision of 20/222 vs right vision of 20/16) and relative ocular hypertension in the left eye (left 18.3 mmHg vs right 13.0 mmHg). A diagnostic left external carotid angiogram revealed left CS DAVF with various draining pathways including the left SOV (Figure 3(a)). TVE achieved complete obliteration of the CS DAVF (Figure 3(b)). Her ocular symptoms and visual function rapidly improved after the successful treatment. Visual acuity of the left eye markedly improved from 20/222 at baseline to 20/100 at 6 months after TVE, along with a decrease of intraocular pressure (from 18.3 mmHg to 11.3 mmHg).
Figure 3.
Case 3: Serial cerebral angiography (CAG) of the left external carotid artery. (a) Initial CAG. The CS DAVF is shown. The draining pathway consists of left SOV (arrowhead), left IOV (double arrowheads), left superficial middle cerebral vein (arrow), left uncal vein connecting to left basal vein of Rosenthal (dashed arrow), and left superior petrosal sinus connecting to veins around the brain stem (double arrows). (b) CAG at the end of transvenous embolization. The complete obliteration of CS DAVF is angiographically confirmed.
Discussion
Frequency of visual disorders
The results suggest a correlation between the frequency of visual disorders and the fistulous sites. Specifically, both patients with the direct AVF including the SOV (two intraorbital AVFs) had visual disorders, whereas, among the other six patients with the indirect AVF including the SOV (five CSs and one SSS DAVF), only one patient had visual acuity reduction that fully recovered after TVE.
Regarding previous reported cases, intraorbital AVFs that are classified as direct AVFs including the SOV in this research, are an extremely rare subtype of cranial AVFs. As far as we have identified, 13 previous reports gave a detailed description of 14 cases of intraorbital AVFs after 1999.4–16 The symptoms and outcomes of these 14 reported cases are summarised in Table 5. Nine of the 14 patients (64.3%) had visual disorders mainly consisting of visual acuity reduction. In addition, all 14 patients had the ocular symptoms, with chemosis and proptosis being the initial symptoms.
Table 5.
Reported cases of intraorbital AVFs.
| No. | Age | Sex | Initial symptoms |
Treatment |
Outcomes of visual function | Author | Publication year | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Visual disorders | Ocular symptoms | Modality | Result | Durationa | |||||||
| 1 | 73 | F | − | + | Not described | Not described | Not described | Not described | Ohtsuka12 | 1999 | |
| 2 | 51 | M | + | + | TVE (coil) | Cured | 30 months | Improved | Deguchi4 | 2005 | |
| 3 | 33 | M | − | + | TVE (coil) | Cured | Not described | No change | Subramanian5 | 2005 | |
| 4 | 55 | M | − | + | Surgery | Cured | 6 months | No change | Hamada6 | 2006 | |
| 5 | 51 | F | − | + | TVE (coil) | Cured | Not described | No change | Caragine7 | 2006 | |
| 6 | 63 | F | − | + | TVE (coil) | Cured | Not described | No change | Caragine7 | 2006 | |
| 7 | 50 | M | + | + | Observation | Spontaneously resolved | 6 months | Improved | Cheng9 | 2009 | |
| 8 | 61 | M | + | + | TVE (onyx) | Cured | 3 months | Improved | Lin11 | 2010 | |
| 9 | 61 | M | + | + | Surgery | Cured | 6 months | Improved | Wington10 | 2012 | |
| 10 | 81 | F | + | + | TVE (coil) | Cured | 7 days | Improved | Williamson8 | 2012 | |
| 11 | 72 | M | + | + | TVE (coil) | Cured | 6 months | No change | Naqvi13 | 2013 | |
| 12 | 50 | F | + | + | Surgery | Cured | 18 months | Not described | Mishra14 | 2013 | |
| 13 | 68 | M | + | + | TVE (onyx) | Cured | Not described | Not described | Lv15 | 2015 | |
| 14 | 74 | F | + | + | TAE (NBCA) | Cured | 7 months | Improved | Konstas16 | 2017 | |
Duration between onset of symptoms and angiographic complete obliteration of AVF.
In comparison with intraorbital AVFs, CS DAVFs, which are classified as indirect AVFs including the SOV in this research, less frequently present visual disorders, even though the SOV commonly acts as the draining route. According to a prospective study involving 46 cases of CS DAVFs, visual disorders were observed in only 3 cases (6.5%), whereas ocular symptoms were observed in all cases.1
In other cranial AVFs, the venous drainage into the SOV is uncommon except anterior condylar confluence DAVF, and visual disorders due to the reflux into the SOV are extremely rare.2,3 In these cases, intracranial hypertension due to severe cortical venous reflux caused the visual disorders.3
According to the above-mentioned data, it can be said that intraorbital AVFs more often cause visual disorders compared with other cranial DAVFs.
Possible pathophysiology of visual disorders
In patients with abnormal venous drainage into the SOV, the pathophysiology of visual disorders has not been thoroughly explored, possibly because of the low numbers of existing cases and the disregard for its clinical impact. Theoretically, ocular hypertension, retinal venous congestion or both resulting from the drainage failure of the SOV can be considered as the main cause of visual disorders.
As regards our cases, ocular hypertension most likely caused the visual disorder in Cases 1 and 3. Likewise, it is almost certain that retinal venous congestion affected the visual acuity in Case 2.
In Case 1, the intraocular pressure was not measured before endovascular treatment because urgent treatment was necessary to interrupt the progressive visual field defect. Although pretherapeutic ocular hypertension is unverifiable, there are some issues leading to our supposition that ocular hypertension caused her visual field contraction. These are as follows: the glaucoma-like visual field defect of the left eye, the optic disc pallor on the left fundus indicating optic nerve atrophy and the absence of any intraocular structure compressing the left optic nerve. In Case 3, the visual acuity recovered in response to the lowering of intraocular pressure in the early stage after successful TVE. Therefore, given the above-mentioned findings, the ocular hypertension can be considered as the cause of visual disorders in Cases 1 and 3.
In Case 2, the retinal venous congestion reflecting severe stagnation of the SOV was demonstrated. Therefore, it is certain that the retinal venous congestion was the cause of the patient's visual acuity reduction.
Outcomes of visual disorders
Previous articles reported that the visual disorders in the cases of intraorbital AVFs could be well recovered after radical endovascular or surgical treatment (Table 5),8–11,16 and their clinical significance has been almost ignored. However, in our case series, the clinical outcomes of the two patients with intraorbital AVF (Cases 1 and 2) were far from satisfactory. Their visual disorders rapidly deteriorated and did not recover even after treatment, resulting only in angiographic cure. On the other hand, the visual acuity of a patient with CS DAVF (Case 3) was improved after successful TVE. It should be noted that the outcomes of visual function were affected only in the patients with intraorbital AVFs. Although our study included a limited number of cases, the results may suggest that the direct fistulas on the SOV or IOV (the direct AVF involving the SOV) can more aggressively injure the retina or optic nerve, compared with other cranial DAVFs using only the SOV as the draining pathway (the indirect AVF involving the SOV).
Regarding the length of time the fistula has been present, in our cases with intraorbital AVFs, the duration between the onset of the ocular symptoms and agiographic obliteration of AVF is relatively longer than that in previously reported cases (Tables 4 and 5). The longer disease duration might affect the outcome of the cases with intraorbital AVFs.
Limitation of this study
First, the present study is retrospective and had a small sample size. Second, the ophthalmological inspection items were not necessarily standardised on every patient. In some cases, the inspections were insufficient. Further prospective studies are needed to clarify how the abnormal venous drainage into the SOV affects the visual function.
Conclusions
In cases of intraorbital AVF, visual disorders are more frequent and can result in poorer outcomes compared with other dural AVFs with drainage into the SOV. Early diagnosis and treatment are crucial to preserve the visual function of patients with intraorbital AVF.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
References
- 1.Pashapour A, Mohammadian R, Salehpour F, et al. Long-term endovascular treatment outcome of 46 patients with cavernous sinus dural arteriovenous fistulas presenting with ophthalmic symptoms. A non-controlled trial with clinical and angiographic follow-up. Neuroradiol J 2014; 27: 461–470. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Spitze A, Gersztenkorn D, Al-Zubidi N, et al. Transverse and sigmoid sinus dural arteriovenous fistula mimicking idiopathic intracranial hypertension and carotid cavernous fistula. Neuroophthalmology 2014; 38: 29–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Robert T, Botta D, Blanc R, et al. Ocular signs caused by dural arteriovenous fistula without involvement of the cavernous sinus: A case series with review of the literature. AJNR Am J Neuroradiol. Epub ahead of print 26 May 2016. [DOI] [PMC free article] [PubMed]
- 4.Deguchi J, Yamada M, Ogawa R, et al. Transvenous embolization for a purely intraorbital arteriovenous fistula. J Neurosurg 2005; 103: 756–759. [DOI] [PubMed] [Google Scholar]
- 5.Subramanian PS, Gailloud PH, Heck DV, et al. Cook detachable coil embolization of a symptomatic, isolated orbital arteriovenous fistula via a superior ophthalmic vein approach. Neuroradiology 2005; 47: 62–65. [DOI] [PubMed] [Google Scholar]
- 6.Hamada J, Morioka M, Kai Y, et al. Spontaneous arteriovenous fistula of the orbit: case report. Surg Neurol 2006; 65: 55–57. [DOI] [PubMed] [Google Scholar]
- 7.Caragine LP, Halbach VV, Dowd CF, et al. Intraorbital anteriovenous fistulae of the ophthalmic veins treated by transvenous endovascular occlusion: technical case report. Neurosurgery 2006; 58(1 Suppl): ONS–E170. [DOI] [PubMed] [Google Scholar]
- 8.Williamson RW, Decruet AF, Crowley RW, et al. Transvenous coil embolization of an intraorbital arteriovenous fistula: Case report and review of the literature. Neurosurgery 2013; 72: E130–E134. [DOI] [PubMed] [Google Scholar]
- 9.Cheng KC, Chang CH, Lin WC, et al. Spontaneous resolution of intraorbital arteriovenous fistulas. Ophthal Plast Reconstr Surg 2009; 25: 245–247. [DOI] [PubMed] [Google Scholar]
- 10.Wigton EH, Wells JR, Harrigan MR, et al. Diagnosis and management of an intraorbital vascular fistula. Ophthal Plast Reconstr Surg 2012; 28: E39–E41. [DOI] [PubMed] [Google Scholar]
- 11.Lin CJ, Blanc R, Clarencon F, et al. Transvenous embolization of an intraorbital arteriovenous fistula using Onyx. J Clin Neurosci 2010; 17: 783–785. [DOI] [PubMed] [Google Scholar]
- 12.Ohtsuka K, Hashimoto M. Clinical findings in a patient with spontaneous arteriovenous fistulas of the orbit. Am J Ophthalmol 1999; 127: 736–737. [DOI] [PubMed] [Google Scholar]
- 13.Naqvi J, Laitt R, Leatherbarrow B, et al. A case of a spontaneous intraorbital arteriovenous fistula: clinicoradiological findings and treatment by transvenous embolisation via the superior ophthalmic vein. Orbit 2013; 32: 124–126. [DOI] [PubMed] [Google Scholar]
- 14.Mishra SS, Panigrahi S, Satpathy PC, et al. Intraorbital arteriovenous fistula with thrombosed varix: Diagnosis and treatment without catheter angiography in a developing country. Surg Neurol Int 2013; 4: 107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lv M, Jiang C, Liu D, et al. Direct percutaneous transorbital puncture under fluoroscopic guidance with a 3D skull reconstruction overlay for embolisation of intraorbital and cavernous sinus dural arteriovenous fistulas. Interv Neuroradiol 2015; 21: 357–361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Konstas AA, Rootman DB, Quiros PA, et al. Transarterial embolization of a spontaneous intraorbital arteriovenous fistula with n-BCA Glue. Ophthal Plast Reconstr Surg 2017; 33: e63–e64. [DOI] [PubMed] [Google Scholar]