Article Text
Abstract
Direct caroticocavernous fistula (CCF) has traditionally been treated by detachable balloon placement within the affected cavernous sinus. We describe a case of a direct CCF treated solely with flow-diverting stents. These novel devices may offer a simpler and potentially safer vessel-sparing option in this rare condition.
Statistics from Altmetric.com
Introduction
Direct carotid–cavernous fistula (CCF) is an uncommon but well known complication of head trauma.1 It may occur after either blunt or penetrating head trauma and represents an abnormal arteriovenous communication within the cavernous sinus, usually as a result of laceration or transection of the cavernous internal carotid artery (ICA).2 The high flow within the fistula means that patients usually present with ocular or orbital venous congestive features and cephalic bruit.3 4 The goal of treatment is to preserve the patency of the ipsilateral internal carotid while obliterating the fistula. Thus direct CCFs are inevitably treated with endovascular methods.5
Transarterial and transvenous delivery of balloons, detachable coils and/or liquid embolic agents have been the mainstay of treatment, with occasional use of porous or covered stents.6–17 In the hands of experienced interventionalists, with the use of single or combinations of the above techniques, the rates of successful fistula closure have been reported at between 85% and 99%.18–21
Recently, intracranial flow diverting stents have become available, and this has led to interest in the interventional neuroradiology community as to whether these devices may be a suitable alternative means of repairing direct CCFs.
We report the case of a 19 year old woman who was found to have a post-traumatic, high flow CCF which was treated solely by the use of flow diverting stents.
Case report
The patient was admitted to hospital complaining of a sudden, overnight onset of severe right orbital swelling and excruciating retro-orbital pain.
A year previously she had been involved in a high speed (∼80 kph) motor vehicle accident. She suffered multiple long bone and craniofacial fractures, splenic laceration requiring splenectomy and laceration of the liver.
Initial non-contrast enhanced CT immediately post-trauma showed bilateral hemorrhagic contusions involving the frontotemporal region on the left and the parietal region on the right. An undisplaced fracture through the greater and lesser wings of the right sphenoid was also noted. She was discharged 2 months later with a persistent VIth nerve palsy and associated diplopia on the right attributed to injury to the lateral rectus muscle.
She presented 2 months following her discharge with headache, vomiting, worsening diplopia and mild proptosis. Imaging was reportedly normal, and she settled on a small dose of opiates and was discharged home. She reported multiple similar episodes in the time preceding her present admission.
On examination at her recent presentation, she was found to have marked proptosis, chemosis and an orbital bruit. While her visual acuity was normal, she had a persistent right VIth nerve palsy.
Contrast enhanced CT of the orbits demonstrated a CCF, with a partially thrombosed and grossly distended superior ophthalmic vein. She was noted to have a rudimentary right A1 segment, and the right ICA appeared transected at the midpoint of the inferior limb of the carotid siphon. Due to the rudimentary A1 segment, a vessel preserving procedure was preferred over parent artery occlusion.
Procedure
The patient was given a loading dose of 150 mg clopidogrel and 300 mg aspirin followed by 75 mg clopidogrel and 100 mg aspirin daily for 4 days.
The procedure was carried out under general anesthesia.
A 6 F sheath was inserted into the right common femoral artery, and a 6 F guide catheter was placed within the right internal carotid artery (RICA).
Initial angiographic runs confirmed transection of the RICA at the midpoint of the inferior limb of the carotid siphon with a resultant high flow fistula (figure 1).
The transected segment was traversed with a 0.014 inch wire (Transend EX Platinum, Boston Scientific, Fairmont, California, USA), and a Marksman 0.027 inch microcatheter (ev3 Endovascular Inc, Plymouth, Minnesota, USA) was advanced to the distal paraclinoid segment of the RICA.
A bolus of heparin was administered intravenously (50 IU/kg body weight).
A (4.25 mm × 20 mm) flow diverting stent (Pipeline Embolisation Device, ev3 Endovascular Inc) was deployed from just proximal to the ophthalmic artery across the transected segment. Further flow diverting stents (5 mm × 12 mm, 5 mm × 10 mm, 5 mm × 14 mm; Pipeline Embolisation Device, ev3 Endovascular Inc) were then telescoped within the first to reconstruct the cavernous segment of the RICA. Care was taken to overlap the stents in such a way as to provide for greatest mesh coverage at the site of the transection.
Post-procedure angiograms demonstrated no further shunting into the orbit and substantially reduced posterior drainage of the fistula into the inferior petrosal sinus (figure 2). Increased flow was demonstrated along the reconstituted RICA. Intravenous heparin infusion was continued for 48 h post-procedure, maintaining an activated partial thromboplastin time of 50–90 s.
Two days following the procedure, the CT cerebral angiogram demonstrated no arterialization of the cavernous sinus, superior ophthalmic vein or internal jugular vein, and patency of the stent construct and the intracranial branches.
The patient's proptosis and chemosis had vastly improved and there was no evidence of a cephalic bruit. She was discharged home on clopidogrel 75 mg daily and aspirin 100 mg daily.
A 1 month follow-up angiogram showed a fully reconstituted RICA with no evidence of arteriovenous shunting (figure 3 and 4).
Discussion
Historically, direct CCFs were treated by various surgical approaches. In 1935, Dandy et al described ‘trapping’ of the CCF by the ligation of the cervical and intracranial ICA. This was then followed by direct open embolization of the cavernous sinus by a variety of agents.22 The first report in the literature of a procedure that preserved the ICA was that by Parkinson et al in 1974.23 They described a surgical approach in nine of 11 patients whereby the cavernous sinus was packed and the ICA preserved. In the same year, Serbinenko et al reported the first case of successful embolization of a direct CCF from an endovascular approach using a detachable balloon.24 In 1978, Debrun et al reported the first series in 12 patients of successful treatment of direct CCFs with detachable balloons.7
Throughout the 1980s, detachable balloons were the preferred and widely used method of treatment of direct CCFs.5 8
The balloons available at the time, however, were plagued by a variety of problems that eventually led to the US Food and Drug Administration withdrawing approval in 1991.9 25 The Food and Drug Administration later approved detachable balloons for intracranial use (DSB; Boston Scientific-Target) in 1998 but they were withdrawn from use in the USA in 2003 due to balloon valve leaks.26
Presently, a number of different endovascular treatment options for CCFs are available. The method chosen in a given patient depends on the anatomy of the fistula and operator or institutional preference.
The goal of treatment in direct CCFs is to occlude the site of communication between the ICA and the cavernous sinus while preserving the patency of the ICA. This can be accomplished with transarterial obliteration of the fistula with a detachable balloon, deployment of a covered stent across the area of the fistula or obliteration of the ipsilateral cavernous sinus with coils or other embolic material.6–17 The above methods are sometimes employed in combination with or without a transvenous approach.15–17 27 28 If the defect is large and cannot be repaired, the ICA may have to be sacrificed or trapped.29 30
While the literature states that the success rate for a single or combined approach to closing a direct fistula is between 85% and 99%,18–21 in our experience we have found the techniques to be challenging and time consuming. In some of our previous cases we have found the need to reintervene due to incomplete closure of the CCF. The radiation exposure required for serial angiographic follow-up and very possibly the subsequent repeat intervention continues to add to the overall accumulated radiation dose.31 32
The above procedures were considered as options in our case. It was noted, however, that there was no opacification of the right A1 demonstrated on initial CT angiography. Follow-up angiography confirmed no opacification of the right A1. High flow retrograde shunting was seen within the posterior communicating artery on initial angiography, thus proximal sacrifice of the RICA was unlikely to have been either ideal or satisfactory. Balloon test occlusion and neurological testing was considered unreliable due to the retrograde sump effect of the shunt and the difficulty of achieving an accurate balloon position in the ICA without excluding the ophthalmic and/or the posterior communicating arteries. Thus the risk of an unreliable/difficult balloon test occlusion with or without sacrifice of the vessel was weighed against the risk of a relatively simple stent reconstruction. The latter course was judged to be a safer option given that transvenous packing or sacrifice, with or without surgical bypass, remained possibilities if the flow diversion technique failed.
Flow diverting stents have only recently been introduced onto the market.33 34 In this case, we employed the pipeline embolization device (PED). The PED is a self-expanding, cylindrically shaped, endovascular scaffolding composed of 48 braided strands of cobalt chromium and platinum. A single PED has 30–35% metal surface area coverage when fully deployed within an appropriately size matched vessel.35 36 At the time of writing, the longest PED construct available on the market is 20 mm.
The main advantage of a PED is that it offers the ability to both reconstruct and preserve the vessel with closure of the defect. The low porosity and high pore density values of the PED maintain patency of the parent vessel and side branches.33–35 37
At present, the limited availability of lengths of PEDs means that multiple devices need to be telescoped in order to achieve long segment reconstructions. Judicious use of telescoping can be employed, as in our case, to augment the degree of metal surface area coverage over particular segments.34–36 It is also feasible, if crossing major branch vessels, to oversize the device relative to the parent artery such that the elongation of the braided mesh pattern will lead to a more porous cell structure and less overall metal surface area coverage.32
In our patient, there was complete transection of the RICA within the cavernous sinus. The use of one device did not substantially alter flow through the fistula, slow the anterior (orbital) drainage of the shunt or divert flow such that there was predominant flow into the supraclinoid ICA rather than into the shunt. Multiple devices were thus placed until there was flow into the distal ICA and the shunt drainage was predominantly posterior.
At present the literature suggests that the patient needs to be maintained on clopidogrel for 6 months and aspirin for life.33 35 36 38 39 This long term therapy has obvious potential morbidity in younger patients. Other than general considerations in regard to increased risk of bleeding, patients would need to be advised regarding the risks of contact sports, pregnancy and high impact activities. Recently, some operators (PK Nelson, personal communication, 2010) have suggested that aspirin may be safely ceased at 2 years but we intend to maintain our patients with flow diverters in situ on lifelong aspirin until robust data supporting its cessation are published.
There have been recent reports in the literature of delayed aneurysmal rupture following insertion of flow diverters. The cases described were with the use of SILK flow diverter (Balt Extrusion, Montmorency, France)40 41 in relation to intradural aneurysms and not with the use of PED or SILK in CCF.
To our knowledge, the sole use of flow diverting stents as a treatment option for direct CCF has not been described in the literature previously.
Compared with the use of detachable balloons, coils and vessel sacrifice, we believe that the use of flow diverting stents in treating direct CCF may prove to be safe, simple, efficacious and cost effective.
References
Footnotes
Correction notice This article has been corrected since it was published Online First. The section head has been amended to Head and neck.
Competing interests None.
Patient Consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.