Coil Retrieval Following Embolization of Cerebral Aneurysms

Summary

Failed coil embolization of cerebral aneurysms may be occasionally followed by direct surgical treatment. We had 5 patients who underwent coil retrieval and surgical clipping after coil embolization because of periprocedural complications. The patients, ranging in age from 40 to 71, had wide-neck aneurysms located at the anterior communicating artery (AcomA) in 3 patients, the middle cerebral artery (MCA) in 1, and the internal carotidophthalmic artery (IC-Ophthalmic) in 1. They were embolized with Guglielmi detachable coils (GDCs), which had to be retrieved within 8 days because of coil protrusion and migration in 3 patients, aneurysm rupture in 1, and increased mass effect due to coil compaction in 1. Coils were successfully removed with aneurysmotomy or arteriotomy under temporary trapping, aneurysms were then clipped or trapped. Three patients had a good outcome, but one suffered permanent visual disturbance and the other had a motor deficit. Our study revealed that a small AcomA aneurysm had a high risk of complication in a case of complex anatomy of the AcomA-Al-A2 complex with its difficult access. In addition, insufficient packing of the inflow zone in a large and symptomatic aneurysm may cause coil compaction and regrow with increasing mass effect.

The indication and treatment strategy for these aneurysms should be carefully determined.

Introduction

Coil embolization is becoming a major modality for treating intracranial aneurysms. Technological advances in devices and materials have enabled us to treat complicated aneurysms which formerly had to be surgically treated. Thus, more aneurysms will be treated endovascularly in the future. However, as recourse to coil embolization becomes more widespread, the number of periprocedural complications is bound to increase.

In this report, we reviewed cases of failed coil embolization followed by coil retrieval and surgical clipping because of periprocedural complications, and we discussed the causes of such complications and the measures to overcome them.

Methods

From 1997 to 2001,346 patients with cerebral aneurysms were endovascularly treated with coil embolization in our institute and affiliated hospitals. Among this series of patients, 5 (1.4%) required coil retrieval and surgical clipping as a result of periprocedural complications (table 1).

Table 1.

Patient charactaristics of periprocedural complication cases

Case	Age Sex	Location	Initial presentation	Size dome/neck	Complication	Outcome
1	53 M	AcomA	SAH	S/S	perforation	GR
2	55 M	AcomA	incidental	S/W	protrusion	SD
3	71 F	AcomA	incidental	S/W	migration	GR
4	70 F	MCA	incidental	s/w	migration	GR
5	40 M	ICA	visual disturbance	L/W	coil compaction	MD
Size (dome/neck) S/S: small with small neck, S/W: small with wide neck, L/W: large with wide neck

These 5 were comprised of 3 males and 2 females with an age range of 40-71 years (mean 58 years). The clinical presentation was subarachnoid hemorrhage in 1 patient, visual disturbance in 1, and no symptom in 3. The aneurysms were small (< 12 mm) with a small-neck (< 4 mm) in 1 patient, small with a wide-neck (4 mm <) in 3, and large (12 mm <) with a wide-neck in 1. The aneurysms were located in the anterior communicating artery (AcomA) in 3 patients, the middle cerebral artery (MCA) in 1, and the internal carotid-ophthalmic artery (IC-Ophthalmic) in 1.

Procedures were performed under local anesthesia with neuroleptanalgesia. A guiding catheter was inserted using the transfemoral approach under general heparinization. For a ruptured aneurysm, heparinization was started after placing the first coil in the aneurysm. Patients had to undergo a craniotomy within 0 to 8 days (mean 1.8 days) after embolization to retrieve the coils and to clip the aneurysms because of coil protrusion and migration into the parent artery in 3 patients, aneurysm rupture in 1, and increased mass effect due to coil compaction in 1.

Results

In all 5 cases, coils were successfully retrieved with aneurysmotomy or arteriotomy under temporary trapping, after which the aneurysms were clipped or trapped. The patient (case 1) with a perforation of the AcomA aneurysm underwent neck clipping with a partial removal of coils. The large IC-Ophthalmic aneurysm (case 5) which enlarged after embolization was treated with trapping of ICA. The details of those 2 cases are discussed below. There were 3 other complications including coil protrusion in an AcomA aneurysm and migration in AcomA and MCA aneurysms. All of these were small aneurysms with a wide-neck. The patient (case 2) with coil protrusion into A2 suffered an occlusion of the left A2. As repeated fibrinolysis yielded only temporary recanalization, an emergency craniotomy was performed. However, in spite of a successful removal of the coils, clots in the branch remained and caused an infarction of the ACA region, resulting in a right hemiparesis. Both patients with a distal migration of coils (cases 3, 4) underwent immediate surgery to retrieve the coil from the migrated branch (A2 and M2) followed by neck clipping. Ischemic symptoms disappeared due to the surgical recanalization. Good outcomes were obtained in 3 patients, but 2 suffered permanent visual field defect and motor deficit with mass effect and ischemic complications.

Illustrative Cases

Case 1 (figure 1)

Figure 1.

Case 1: A dumbbell-shaped, ruptured aneurysm at part of the Al-A2 bifurcation. A,B) Right oblique, right ICA angiograms before (A) and after (B) the coil embolization of a distal sac. This working view shows the overlap at this point of a proximal sac on the Al-A2 junction. C) Left oblique angiogram revealing the proximal sac (C). Only the distal part of the aneurysm was packed. D) Angiogram after repositioning the microcatheter and inserting the coil shows the coil placed in the extraaneurysmal space.

This 53-year-old man presented a sudden onset of headache. A CT scan revealed diffuse SAH (Hunt & Kosnik grade 2, Fisher group 4), and a cerebral angiogram showed a small lobular AcomA artery aneurysm, the sac of which was 2.5 mm in diameter. Embolization was selected because of his general medical condition and at the request of his family.

Endovascular procedure

A guiding catheter was placed in the right internal carotid artery (ICA), and a microcatheter was advanced into the aneurysm. The acute angle of ICA-A1 bifurcation interfered with both smooth catheterization and stability of the microcatheter in the aneurysm. A standard 10 size GDC (2 mm by 4 cm) was successfully placed in the distal sac. After confirming the complete occlusion of the sac, the working angle was changed to provide better visibility, which revealed a residual sac. The microcatheter was repositioned, and an attempt was made to insert a 10 soft type GDC (3 mm by 6 cm) into the proximal sac. During this procedure, the microcatheter jumped forward and the coil penetrated the aneurysm wall. We retried to reposition the coil back into the aneurysm, but it proved impossible to keep the microcatheter inserted in the aneurysm sac. At that point, we decided on surgical treatment abandoning all efforts at an endovascular approach.

Surgical treatment

The aneurysm was exposed under the usual pterional approach. The coil was penetrating through the junction of the proximal and distal sac of the aneurysm. After applying temporary clips, the penetrated coil was easily removed, and the aneurysm was successfully clipped without removing the coil which was inserted in the distal sac. The postoperative course was uneventful.

In this case, the aneurysm had a dumbbell-shape, making the approach to the proximal sac extremely difficult because of the sharp angulation. In particular, the position of the tip of the microcatheter was barely visible since the neck of the proximal sac was formed to overlap the Al-A2 junction. These conditions risked putting undue pressure on the attachment of the tip of the microcatheter to the aneurysm wall, possibly resulting in coil penetration.

Case 5 (figure 2)

Figure 2.

Case 5: Unruptured, large ICA aneurysm in part of the ophthalmic artery. A) Right ICA angiogram showing a large, wide-necked IC-Ophthalmic aneurysm projecting superio-medially. B) Angiogram just after the embolization showing small remnants of an unembolized aneurismal sac. C)Follow-up angiogram 1 week after embolization. D)The operative view. E) Intraoperative angiogram showing complete trapping just distal to the ophthalmic artery.

During a brain check-up for headache, this 40-year-old man was informed of a right wide-neck IC-Ophthalmic aneurysm, but he refused to undergo a craniotomy.

Endovascular Procedure

A preliminary balloon occlusion test of the right ICA was performed for 30 minutes without any neurological deficits. In the usual manner, a three-dimensional 18 type GDC (14 mm by 30 cm) was placed initially. Various-sized GDCs were used totalling 294 cm in length, but the neck of the aneurysm was difficult to completely obliterate without risking protrusion of the coils. The procedure was finished by a subtotal occlusion with a dome filling of contrast medium.

Post embolization Course

A left temporal hemianopsia and right enlargement of the central scotoma developed 4 days after the procedure. There was no change in the shape of the coils on plain X-ray film. Steroids were administered to reduce the inflammatory effects of embolic materials, but were not effective. The cerebral angiogram at a 1 week follow-up revealed coil compaction and a regrowth of the aneurysm, which was probably the cause of his visual disturbance. Therefore, we decided to try clipping with a removal of the coils to reduce the mass effect.

Surgical Treatment

Under the fronto-temporal approach, the optic canal and the anterior clinoid process were resected. The dural ring was cut, exposing an aneurysm containing GDC that was severely compressing the chiasm and right optic tract. Under proximal and distal trapping of ICA, an aneurysmectomy was done at the dome of the aneurysm. However, the mass of GDC and clots was so large that the aneurysm neck was lacerated. Thus, neck clipping was abandoned, and ICA was trapped between a point distal to the ophthalmic artery and proximal to the origin of the posterior communicating artery. Intraoperative DSA confirmed a successful trapping and a good collateral flow. The patient’s visual symptoms have gradually improved since the operation.

Discussion

There are several situations which require direct surgery after coil embolization. These include the protrusion or migration of coils into the parent artery, rupture of an aneurysm, and an increase in the mass effect due to coil compaction and regrowth of the aneurysm.

Coil Protrusion and Migration

Coil protrusion or migration tend to occur after incomplete packing of a wide-neck aneurysm. Protruded coils may be the source of a thrombus resulting in distal vessel occlusion. If patients with a minimal protrusion of coils have no ischemic symptoms, they can be managed with persistent systemic anticoagulation followed by oral anticoagulation. However, in cases with symptomatic protrusion or migration of coils, they should be retrieved to prevent a further ischemic event. Thornton et A1 reported 2 cases of coil migration, which occurred in a small size aneurysm where coils stretched out of the aneurysm, loosing their memory. Both patients developed ischemic symptoms within 48 hours of embolization. Macroscopic findings of the retrieved coils showed a thrombus around the stretched lead point of the coil, which presumably triggered the distal embolization¹. In our series, 2 cases of coil migration into the parent arteries resulted in good outcomes. The main source of ischemic symptoms may have been due to the small embolisms formed around the migrated coils, which could be recanalized with fibrinolysis followed by antiplatelet therapies. In contrast, in a case of thromboembolic occlusion of A2 due to coil protrusion, a high thrombogenic tendancy with a rapid extension of clotting to distal branches resulted in serious irreversible ischemia in the region concerned. This may indicate that a coil mass outside of the orifice can impair blood flow and fam clots more easily than a helical single coil in the vessel.

Rupture of Aneurysm

Aneurysm rupture during coil embolization is occasionally due to a perforation of the aneurysm wall by the microguidewire, microcatheter, or coil. According to a systematic review reported by Brilstra et A1, the incidence of intraprocedural aneurysm rupture is 2.4%². To forestall these complications, McDougall et A1 emphasized the importance of preventing uncontrolled advancement of the microcatheter by ensuring that no forward pressure is exerted before the guidewire is removed, and by removing the guidewire very slowly under direct fluoroscopic visualization³.

We witnessed the perforation of a ruptured AcomA aneurysm in case 1. Actually, this aneurysm had two daughter sacs, but it was difficult to differentiate the neck of the proximal sac from the AcomA-Al-A2 complex. A three-dimensional angiography or CTA would have been helpful in clarifying the structure around the AcomA-Al-A2 complex, and a procedure might then have been performed under biplane fluoroscopy to manipulate the microcatheter safely. Further, endovascular approaches to an AcomA aneurysm are sometimes difficult. Moret et A1 reported 36 AcomA aneurysms treated with coil embolization, and mentioned that in 3 cases (8%) it proved impossible to introduce the microcatheter into the aneurysm because of atheromatous lesions and loops within the cervical and intracranial vessels⁴. Shanno et A1 reviewed 42 aneurysms treated with GDC in which the treatment failed. Sixteen aneurysms which could not be catheterized because of vascular tortuousity included 9 AcomA aneurysms⁵.

Such a torturous, curved track may pose a great obstacle to advancing the microcatheter, occasionally causing the operator to lose control, resulting in a catastrophic perforation. When treating an AcomA aneurysm endovascularly, the complex anatomy of the AcomA-Al-A2 complex and difficulty in approaching the aneurysm may pose major obstacles to successful coil embolization.

Coil Compaction and Regrowth of Aneurysm

For a large aneurysm present with mass effect, the goal of treatment is to reduce the mass effect as well as to prevent bleeding. Malisch et A1 described 19 patients with aneurysms producing a mass effect who were treated with GDC, resulting in improved or resolved symptoms in 74%⁶. On the other hand,Tsuura et A1, who performed serial MRIs after the embolization of large or giant aneurysms, found that shrinkage tended to occur 2 to 12 months after treatment. Aneurysms occluded by coils tended to shrink much more slowly than aneurysms treated with parent artery balloon occlusion⁷.

Coil compaction and regrowth of the aneurysm exacerbated the symptoms in a few days after coil embolization in case 5. This may be the result of an incomplete occlusion of the aneurysm neck and an insufficient control of the inflow zone. To prevent coil protrusion into the parent artery and to achieve complete obliteration of the aneurysm, the “Balloon remodeling technique” should prove useful in the embolization of these wide-neck aneurysms^8,9.

From a surgical point of view, coils should be easily removed in the early post-coiling period. However, delay more than a few additional weeks, solid and partially organized clots would have formed among the coils, making it very difficult to remove both coils and clots.

Nowadays, the preferred strategy for a large or giant IC-Ophthalmic aneurysm presenting mass effect is trapping the aneurysm if the patient can tolerate the parent artery balloon occlusion test.

Conclusions

The retrieval of packed coils requires superfluous surgery, causing much trouble for patients who had expected a cure with embolization alone.

To avoid such unfortunate second operations, meticulous pretreatment evaluation of the images and access routes, proper selection and management of endovascular devices and strategies should be made mandatory.