A Recurred Carotid Siphon Pseudoaneurysm after Detachable Coil Embolization

Summary

Massive posterior epistaxis is one of the peculiar symptoms of pseudoaneurysms of the carotid siphon. We experienced a case of trauma-related pseudoaneurysm of the carotid siphon. The lesion was initially silent except for the mass effect. We initially treated the lesion with platinum detachable coil embolization of the pseudoaneurysm sac with preservation of the parent artery. However, the patient had delayed massive epistaxis with recurrence of the pseudoaneurysm. The patient was subsequently managed with endovascular occlusion of the affected internal carotid artery using detachable balloons. Complete internal carotid artery trapping is recommended as an initial treatment modality if the patient can tolerate to the occlusion test. Careful observation and follow-up of the patient is required if the lesion is inevitably managed with coil embolization of the pseudoaneurysm sac alone.

Introduction

Traumatic pseudoaneurysms of the carotid siphon associated with skull base fracture secondary to head trauma have a picturesque clinical triad - history of head trauma, ipsilateral visual loss, and massive epistaxis, which was first described by Maurer et Al in 1961¹There have been some case reports on the lesion mainly featured by massive posterior epistaxis^2-4.

The lesion has been managed by surgical ligation of the involved internal carotid artery (ICA) or endovascular permanent ICA occlusion (ICA trapping) with detachable balloons^5-7. With the advent of platinum detachable coils, coiling of the pseudoaneurysm was tried to maintain the patency of the parent ICA^7-9. In contrast to the results of platinum coil embolization for the treatment of usual saccular intracranial aneurysms, however, early recurrence was reported in the case of traumatic pseudoaneurysm although the procedure was initially successful⁸.

We experienced a case of traumatic pseudoaneurysm of the carotid siphon, which was initially silent except for the mass effect. For the preservation of the parent artery, we initially treated the lesion with detachable coil embolization of the pseudoaneurysm sac only. Despite the angiographically acceptable embolization of the pseudoaneurysm, the lesion recurred presenting delayed massive posterior epistaxis. The patient was subsequently managed with endovascular occlusion of the ICA using detachable balloons. The purposes of this case report are to review the morphologic and clinical significance of traumatic cavernous ICA pseudoaneurysm and to emphasize the value of ICA trapping in the management of the pseudoaneurysm.

Case Report

A 25-year old woman, a victim of motor vehicle accident, presented at the emergency department in a stuporous state with a closed head injury. Her right pupil was slightly dilated with sluggish response to light. No paresis of the extremities was present. Initial CT scan showed diffuse brain swelling and a skull base fracture; a linear fracture line ran across the right side of the sphenoid sinus. Her mental status gradually improved through the early admission period. The follow-up brain MRI, performed on the admission day 7, showed focal high signal intensity at the posterior temporal lobe, which was presumed to be a brain contusion. An oval-shaped signal void lesion was noted in the right carotid siphon area (figure 1A). Cerebral angiography was subsequently performed under the impression of carotid artery pseudoaneurysm.

Figure 1.

A) MR FLAIR image shows an oval signal void in the right carotid siphon area. The sphenoid and right ethmoid sinuses are filled with heterogeneous signal intensity lesions suggesting hemorrhage within the sinuses. An aneurysm was suspected, and angiography was performed subsequently. B) Right common carotid arteriogram shows an oval pseudoaneurysm sac at the horizontal portion of the carotid siphon. The aneurysm sac has a well-defined neck.

At that time, the decreased visual acuity of her right eye and sluggish light reflex of the pupil persisted. Ipsilateral sixth cranial nerve (CN) palsy was also noted. At angiography, there was a pseudoaneurysm directed inferiorly at the horizontal portion of the carotid siphon. Focal ectasia and stenosis of the ICA was also noted at the posterior genu of the carotid siphon, suggesting severe arterial injury. There was no clinical evidence of aneurysm rupture at that time (figure 1B).

Initially Guglielmi Detachable Coil (GDC; Target Therapeutics, Fremont, CA, U.S.A.) embolization was planned because the pseudoaneurysm had a well-defined neck. At that time we were not aware of the serious clinical behavior of such a lesion. The embolization procedure was performed three days later. The size of the aneurysm was significantly increased compared with that at the initial angiography, suggesting progression of the lesion (figure 2A). A total of 132 cm of coils were inserted to obliterate the sac. The procedure was uneventful.

Figure 2.

A) Follow-up angiography performed three days later just before the detachable coil embolization reveals enlargement of the sac. Coil embolization of the sac was performed successfully. B) Right common carotid arteriogram was performed again one week later. The sac is obliterated with tangled coils.

The patient was stable after the procedure. She had no significant neurological abnormality except for right visual loss and the sixth CN palsy. Follow-up angiography was performed one week after the embolization, and the aneurysm was completely obliterated (figure 2B).

However, on the 27^th postembolization day, she suddenly showed massive right-sided epistaxis. The bleeding was so massive that blood transfusion was required to stabilize her vital signs. Nasal packs were placed immediately after the bleeding, and the bleeding ceased. Angiography was performed to evaluate the recurrence of the aneurysm. On scout images, the meshes of the previously embolized coils were spread out suggesting loosening of the coils and enlargement of the aneurysm sac. There was no direct evidence of contrast extravasation (figure 3A). Permanent occlusion of the ipsilateral ICA was planned. Initially we used gold valve balloon #17 (GVB; Cathnet, Paris, France) to occlude the ICA and aneurysm together. It was a straightforward procedure. Occlusion tolerance was tested with repeated neurologic examinations for 30 minutes under systemic heparinization achieved with an intravenous bolus injection of 5000 units of heparin. During the test occlusion she was neurologically stable. The first balloon was detached at the horizontal portion of the carotid siphon followed by GVB #16 and GVB #11 (figure 3B). Intracranial branches of right ICA were filled via intact anterior and posterior communicating arteries. After the ICA occlusion she did not show any additional neurological abnormality. She did not complain of recurrence of epistaxis on three-month follow-up after the occlusion procedure.

Figure 3.

On 27^th postembolization day the patient showed sudden onset of epistaxis. A) Right common carotid arteriogram was performed in emergency, and only to show recurrence of the aneurysm with loosening of the previously embolized coils and enlargement of the sac. B) Detachable balloons were inserted to occlude the parent ICA after negative occlusion test. The first balloon is deployed at the horizontal portion of the carotid siphon. Part of the balloon is herniated through the large opening of the sac. A total of three balloons were detached. The most proximal balloon deployed near the carotid bulb is not shown.

Discussion

Traumatic injury of the cavernous carotid artery usually manifests as cavernous-carotid fistula. Traumatic pseudoaneurysm of this region may also occur although it is a rare condition ². As Maurer et Al ¹ already described, the pseudoaneurysm should be considered when there are visual disturbances and sudden onset of ipsilateral posterior epistaxis in a patient with a history of head trauma. The bleeding can be delayed and is mostly fatal.

The mechanism of epistaxis can be easily explained when we understand the anatomic relationship between the cavernous carotid artery and the sphenoid sinus⁴. The carotid siphon has a close relationship with the lateral wall of the sphenoid sinus. There is a groove formed by the carotid artery impression on the lateral wall of the sphenoid sinus, known as the 'carotidgroove'. The bony wall of the area is so thin that the mean thickness is usually less than 1 mm. In four percent of normal population the bony septum is absent. So the lateral wall of the sphenoid sinus is vulnerable to injury. Once a pseudoaneurysm is formed in the area, the sac of the aneurysm mainly consists of sinus mucosa except for some very fortunate cases in which the sac consists of fibrous tissue such as periosteum. This is why the pseudoaneurysm of the carotid siphon usually presents with posterior epistaxis. The bleeding is usually massive in most cases because it is a large arterial bleeding.

The mortality rate of the bleeding has ranged as high as 50%⁷. Fortunately we were able to perform angiography because the bleeding ceased after vigorous nasal packing.

The lesions have been managed mainly with trapping of the affected ICA either surgically or with endovascular measures ^5-7. Detachable balloons play a vital role in the management of those lesions considering the poor surgical accessibility of cavernous ICA⁹. Before the ICA trapping, an occlusion tolerance test is mandatory. If the patient does not tolerate the occlusion test, we can sacrifice the ICA after external carotid-internal carotid artery bypass surgery. In those cases with occlusion intolerance, detachable coil embolization of the pseudoaneurysm sac can be a treatment alternative ^8,9. As in our case, however, the trial of sparing affected ICA in the treatment of pseudoaneurysm might fail although the initial result of aneurysm isolation using coils seemed successful ⁸. It may recur. The clinical behavior of pseudoaneurysm is different from true aneurysm even if they show similarities on the angiogram. There is no complete limiting wall in pseudoaneurysms. So it is impossible to obliterate the space completely with coils. Teitelbaum et Al ⁹ successfully treated a patient with cavernous ICA aneurysm using GDC. However, they could not follow-up the patient because of the patient's early death due to another unrelated cause.

We cannot tell what would have happened in their case on long-term follow-up. We recommend a management plan of sacrificing the involved carotid artery if the occlusion test is tolerable. When not tolerated, the affected ICA can be trapped following surgical bypass. If we have no other choice but to embolize the sac alone, we have to be alert for the recurrence ofthe epistaxis. It is mandatory to take simple skull radiography periodically to see the morphological change of the embolized platinum coils.

Another ideal treatment method might be covered or graft stents though they are not available for this kind of purpose now. Recently a covered stent has become available in the cardiology field. The stent is not suitable for the neurovascular purposes due to its stiffness. The geometry of the carotid siphon requires a more flexible stent.

Conclusions

As a result of our experience and that of others, intrasaccular embolization of the pseudoaneurysm of carotid siphon is not recommended even though the lesion looks morphologically suitable for platinum coil embolization at angiography. Complete ICA trapping is a more reliable method. Careful observation and follow-up of the patient is required if the lesion is inevitably managed with coil embolization of the pseudoaneurysm sac alone.