Treatment of direct carotid-cavernous fistulas with flow diversion – does it work?

Abstract

Direct carotid-cavernous fistulas (CCFs) are high flow lesions that can be challenging to treat. A number of recent reports suggest that flow diversion may be a viable treatment option. We present a case of a post-traumatic CCF successfully treated with flow diversion and provide a review of the literature. Our results suggest that flow diversion is a potentially effective treatment option for CCFs and is most successful when used as an adjunctive therapy.

Introduction

Carotid cavernous fistulas (CCFs) frequently present with chemosis, pulsatile proptosis, ocular bruit, vision loss, and occasionally intracerebral hemorrhage or seizure. Direct fistulas, also known as Barrow Type A fistulas, are high flow lesions consisting of a direct connection between the cavernous internal carotid artery (ICA) and the cavernous sinus. They are usually post-traumatic but may also be the result of aneurysmal rupture. The high flow nature of these lesions makes treatment challenging and the standard treatment with transvenous coil embolization frequently fails to curatively obliterate the fistula. Transarterial embolization with flow diversion represents a therapeutic alternative that may be used in combination with coiling or possibly even as standalone treatment.

Methods

A PubMed literature search was performed from 2008 to 2018 to identify studies investigating the potential safety and efficacy of flow diversion for the treatment of direct CCFs. In addition, an illustrative case is reported in which a post-traumatic direct CCF was incompletely treated with transvenous coil embolization and subsequently cured with placement of a flow diverter.

Results

A total of 27 direct CCF cases treated with flow diversion (including the current illustrative case) were identified (Table 1). Nine patients were primarily treated with flow diversion and 18 patients were treated secondarily with flow diversion. Nineteen patients were women and nine were men. The median age was 54 years (range 17–86 years). Fourteen of the cases were traumatic, 10 iatrogenic and three were spontaneous. Multiple flow diversion devices were used including the pipeline embolization device (PED, Medtronic, Ireland), SILK (Balt Extrusion, Montmorency, France), p64 (Phenox, Germany) and Surpass (Stryker, Kalamazoo, USA). The median total number of devices used was three (range 1–10). Coils were deployed in 12 patients with a median number of coils used of 24 (range 1–85). Follow-up angiograms were obtained at intervals between two weeks and 39 months (median 12 months). All fistulas were cured at the end of the follow-up period. In three patients, the carotid was noted to be occluded at final follow-up without major adverse outcome (modified ranking score (mRS) 0–2). No other complications have been reported. Median mRS was 0 (range 0–3). There was one patient with a mRS of 3 who did not have any complications related to treatment.

Table 1.

Reported cases of carotid-cavernous fistulas treated with flow diversion.

Published	Author	Age	Sex	Etiology	Devices	# Devices	Primary treatment
2010	E. Hauck	64	F	Iatrogenic	PED	5	FD
2011	M. Nadarajah	19	F	Traumatic	PED	4	FD
2015	D. Iancu	47	F	Iatrogenic	Silk	1	Other
2015	J. Chan	33	M	Traumatic	PED	1	Other
2015	N. Pradeep	28	M	Traumatic	PED	2	Other
2015	E. Nossek	59	F	Spontaneous	PED	3	Other
2015	N. Pradeep	23	M	Traumatic	PED	3	Other
2016	C. Wendl	58	F	Iatrogenic	PED	1	FD
2016	C. Wendl	72	M	Iatrogenic	p64	1	FD
2016	C. Wendl	60	M	Iatrogenic	p64	2	FD
2016	C. Wendl	76	F	Iatrogenic	PED, p64	2	FD
2016	K. Amuluru	69	F	Iatrogenic	PED	4	FD
2016	C. Wendl	46	F	Iatrogenic	p64	5	FD
2016	C. Wendl	50	F	Traumatic	p64	6	FD
2016	C. Wendl	64	F	Traumatic	p64	1	Other
2016	C. Wendl	44	F	Iatrogenic	PED	2	Other
2016	C. Wendl	66	F	Traumatic	PED	3	Other
2016	C. Wendl	74	M	Iatrogenic	PED	4	Other
2016	C. Wendl	54	F	Traumatic	p64	6	Other
2016	C. Wendl	74	F	Spontaneous	PED	8	Other
2016	C. Wendl	86	F	Spontaneous	p64, PED	8	Other
2016	C. Wendl	17	F	Traumatic	p64, PED	10	Other
2017	C. Ogilvy	78	F	Traumatic	PED	2	Other
2017	C. Ogilvy	35	M	Traumatic	Surpass	1	Other
2017	C. Ogilvy	22	M	Traumatic	Surpass	2	Other
2018	N. Yoon	42	F	Traumatic	PED	1	Other
2018	Current	22	M	Traumatic	PED	7	Other

F: female; M: male; FD: flow diversion.

Illustrative case

A 22-year-old man presented to the emergency department with a Glasgow Coma Scale of 5 after a fall from a truck. A computed tomography (CT) of the head demonstrated multiple skull base fractures and a large subdural hematoma resulting in 7 mm of midline shift. He underwent emergency surgical hematoma evacuation and was admitted to the neurological intensive care unit. On postoperative day 5, the patient developed a right afferent pupillary defect. A CT angiogram and catheter angiogram were obtained, which revealed a direct right carotid-cavernous fistula (Figure 1). Transvenous coil embolization was performed with a total of 24 detachable coils (Axium™ Detachable Coil System, Covidien) with an overall length of 494 cm. Significant reduction of flow was achieved; however, angiographic follow-up three days later revealed persistence of the fistula.

Figure 1.

(a) Anteroposterior (AP) digital subtraction angiogram of right internal carotid artery demonstrating a direct carotid-cavernous fistula (CCF); (b and c) AP and lateral views of right internal carotid arteriogram demonstrating persistent CCF after transvenous coil embolization; (d) Lateral view of right internal carotid arteriogram at three-month follow up after pipeline placement demonstrates complete resolution of the CCF.

The decision was made to treat the residual CCF with flow diversion and the patient was subsequently loaded with aspirin and Plavix. Under general anesthesia, three overlapping PEDs were deployed into the right cavernous ICA. After deployment of the fourth PED, flow through the CCF was minimal. The deployment of the fifth PED did not further reduce the flow and it was thought that placing additional devices would not be beneficial. The patient remained in the hospital for an additional two weeks and his visual symptoms resolved. He was discharged on hospital day 20, walking with assistance and able to perform modified activities of daily living. Follow-up angiography after three months confirmed complete resolution of the CCF. His vision was intact and he regained the full ability to walk.

Discussion

This case and review of the literature suggests flow diversion may represent a potentially successful treatment option for direct CCF patients. This is somewhat surprising as flow diverters are generally designed to preserve patency wherever there is a flow gradient, that is, branches jailed by a flow diverter will be preserved unless there is abundance of collateral flow. It could be expected that the flow gradient from carotid artery to cavernous sinus would be sufficient to preserve flow through a direct CCF despite the placement of flow diverters. This explains why flow diversion is more commonly used as secondary treatment (67%) and less commonly as primary or stand-alone treatment (33%). Our case example and literature review demonstrate that flow diversion can indeed cure a CCF.

Previously, CCFs have been managed with a number of different techniques including vessel sacrifice, balloon embolization, coil occlusion, stent graft, and n-butyl-2-cyanoacrylate (NBCA) embolization.^1–5 However, transvenous coil embolization has emerged as the generally preferred treatment modality.⁶ In the case of larger CCFs, coiling may be insufficient to close the fistula and adjunctive treatment modalities with liquid embolics may be required.^7–9 These are not without risks, as NBCA can result in retained catheter fragments or leakage of glue into the ICA and Onyx can cause cranial neuropathies.^10,11 Covered stents have also been used as a secondary treatment option; however, their use has been limited due to the poor trackability and high thrombogenicity.^12–14

A better device for this task would be a flexible or self-expanding stent graft.

Flow diverters began to emerge as an alternative treatment option to covered stents for CCFs as early as 2010.¹⁵ Flow diversion was originally developed for the treatment of large or giant wide-necked brain aneurysms but are now being used off-label for a variety of conditions including dissections, pseudoaneurysms and carotid-cavernous fistulas.¹⁶ The only flow diverters approved in the USA by the Food and Drug Administration are the PED and Surpass. Internationally used flow diverters include the Silk device and p64. All of these flow diversion devices have been used to cure CCFs (Table 1).^17–25 While treatment with flow diversion necessitates at least three months of dual antiplatelet therapy and is associated with increased costs, it is a technically easy method of treatment and can prevent occlusion of the ICA when used in combination with coils. Combining flow diversion with non-matching patterns of the braided structures may increase their hemodynamic effect (e.g. PED with Surpass, or p64 with PED). Of the 27 cases of CCFs treated with flow diversion, there have been no reports of failures or serious complications. Therefore, the true complication or success rate of flow diversion for the treatment of CCFs remains unknown. Nevertheless, the current data are promising and future prospective data collection is encouraged.

In summary, our literature review and illustrative case suggest that flow diversion is a potentially safe and effective treatment option for CCFs. Flow diversion is typically used as an adjunctive treatment and should be used cautiously as standalone therapy. Multiple overlapping PEDs are typically placed to increase wall coverage and complete occlusion of the CCF is often achieved over time due to decreased flow, similar to the mechanism of flow diversion for treatment of brain aneurysms.

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.