A case of delayed distal coil migration after coil embolization of an unruptured distal azygos anterior cerebral artery aneurysm: A case report and literature review

Abstract

Intraprocedural coil migration during endovascular treatment for an aneurysm that might carry serious ischemic complications is well known. On the other hand, delayed coil migration after endovascular treatment for an aneurysm is very rare. A 77-year-old woman was incidentally diagnosed with unruptured aneurysm associated with distal azygos anterior cerebral artery (ACA). The aneurysm was located at the distal bifurcation of the azygos ACA and was wide necked (approximately 7 mm in diameter). Endovascular coil embolization was selected and the aneurysm was occluded successfully, but 29 days after endovascular therapy, follow-up computed tomography (CT) and magnetic resonance (MR) angiography revealed distal coil migration in the peripheral portion of the ACA. In addition, CT on day 57 after therapy revealed the migrated coil had moved more distally. Fortunately, in the course of these events, the patient remained asymptomatic. To the best of our knowledge, this represents the first case of delayed distal coil migration associated with relatively rare azygos ACA aneurysm, and also the first report confirming more distal coil movement over time. In the future, a large number of patients could develop this complication as more aneurysms are aggressively treated with endovascular treatment. Knowledge regarding the possibility of delayed coil migration is thus important.

Introduction

Recent progress in endovascular treatment has been marked, especially in coil embolization for aneurysm. However, this treatment continues to carry risks of unexpected complications.^1–3 Among these, intraprocedural coil migration is well known, occurring in 2%–6% of cases.² However, delayed coil migration after coil embolization of an aneurysm is very rare, with few cases reported to date (Table 1).^4–13 We describe herein a case of delayed distal coil migration in a patient with unruptured distal azygos anterior cerebral artery (ACA) aneurysm, with subsequent confirmation of more distal coil movement. Direct flow stress from the azygos ACA to the neck appeared to have carried the untangled small helical-type coil more distally. No similar cases of delayed coil migration associated with azygos ACA aneurysm or detection of more distal movement of the migrated coil have been reported. In this case, the patient was asymptomatic after this event, but coil migration could be related to poor functional outcomes. We discuss this rare occurrence with consideration of the literature.

Table 1.

Summary of clinical features in 15 cases of delayed coil migration.

First author (year)ref	Age, sex	Initial presentation	Aneurysm location	Aneurysm shape	Dome/ neck, mm	Adjunctive technique	Duration since coiling	Partial or total migration	Post-migration symptoms	Salvage treatment	Outcome
Phatouros (1999)5	69, female	SAH	MCA	NA	4/NA	–	6 months	Partial	Asymptomatic	No therapy	Good
	55, male	SAH	MCA	NA	4/NA	–	3 months	Partial	Asymptomatic	No therapy	Good
	60, male	SAH	MCA	NA	9.5/NA	–	2 months	Partial	Paresthesia of right arm	NA	Good
	38, female	SAH	SHA	Broad neck	6.5/3.5	–	15 minutes	Partial	Right hemiparesis, dysphasia	Unsuccessful endovascular retrieval, open surgery	MCA territory infarct
Thornton (2000)6	44, female	SAH	PCoA	Broad neck	3/2	–				Open surgery, coil removed	Left hemiparesis
	62, male	Incidental	SHA	Broad neck	4/2	–	3 days	Partial	Right hand paresthesia	Open surgery, coil removed	Good
Gao (2006)7	60, female	Incidental	PCoA	Broad neck	NA	Stent- assisted	5 months	Total	Asymptomatic	No therapy	Good
Haraguchi (2006)	77, female	Incidental	Basilar trunk	Broad neck	8.8/5.8	–	13 days	Partial	Consciousness disturbance	Thrombolysis	Hemianopsia and mild ataxia
Fiorella (2009)9	56, male	SAH	A1–A2 junction	Broad neck	4.5/3.9	Balloon- assisted	8 days	Partial	Right lower extremity paralysis	Clopidogrel added	Right lower extremity paralysis
Motegi (2010)10	59, male	Incidental	ICA bifurcation	Broad neck	7/6	Balloon- assisted	3 months	Partial	Transient numbness of face and hand	Open surgery, coil removed	Good
Banerjee (2011)11	24, male	SAH	ACoA	Broad neck	2/2	–	1 month	Total	Asymptomatic	Open surgery, added neck clipping	Good
Wada (2012)4	64, female	SAH	ACho	Broad neck	5.4/3.9	–	24 days	Partial	Aphasia, right hemiparesis	Open surgery, clipping of aneurysm (migrated coil could not be retrieved)	None
Saguchi (2015)12	68, male	SAH	ACoA	Broad neck	5/3	Double catheter	38 days	Total	Asymptomatic	No therapy	Good
Kamide (2017)13	77, female	Incidental	PCoA	Broad neck	6.1/6.5	Double catheter	3 months	Partial	Asymptomatic	Stent-assisted coil embolization	Good
Present case (2017)	77, female	Incidental	DACA	Broad neck	6/4	Double catheter	29 days	Total	Asymptomatic	Clopidogrel and edoxaban were continued	Good

SAH: subarachnoid hemorrhage; MCA: middle cerebral artery; SHA: superior hypophyseal artery; PCoA: posterior communicating artery; ICA: internal carotid artery; ACho: anterior choroidal artery; ACoA: anterior cerebral artery; DACA: distal anterior cerebral artery; NA: not available.

Case report

A 77-year-old woman with a history of hypertension was examined on admission to our hospital for subcortical hemorrhage in the right occipital lobe. An unruptured aneurysm was found on the azygos ACA. Angiography demonstrated a saccular aneurysm including a bleb measuring approximately 7 mm in height and 6 mm in width, with the neck measuring 4 mm at the bifurcation of the azygos ACA (Figure 1(a)). After half a year, coil embolization was performed. Aspirin (100 mg/day) and clopidgrel (75 mg/day) were administered from seven days before the procedure. A guiding catheter (Britetip® Guide Catheter STR 8-Fr 90-cm; Cordis, Fremont, CA, USA) was induced to the proximal portion of the right cervical internal carotid artery through an intermediate catheter (6-Fr Guiding Catheter Cerulean DD6; MEDIKIT, Tokyo, Japan). Before microcatheterization, we administered 5000 U of heparin, followed by a bolus dose of 1000 units of heparin every hour. Two microcatheters, an Excelsior® SL-10 (Pre-shaped 45 angle; Stryker Neurovascular, Cork, Ireland) and an Echelon-10 straight microcatheter (Covidien, Irvine, CA, USA), were navigated into the aneurysm. Bilateral A3 branches were arising from the neck of the aneurysm, so we thought that somewhat smaller coils were better for the first coiling, and a cage formation was created inside the aneurysm using Target® 360 Soft Detachable Coils (4.5 mm × 12 cm; Stryker Neurovascular, Fremont, CA, USA) from the Echelon-10 and Target® 360 Soft Detachable Coils (5 mm × 10 cm; Stryker Neurovascular) from SL-10 (Table 2). In the early stages of coil embolization, since the Echelon10 microcatheter was pushed out near the neck, coil insertion was mainly performed from the SL-10 microcatheter. In addition, since the SL-10 microcatheter was also pushed out near the neck at the time of insertion of the no. 5 three-dimensional (3D)-type coil, many helical-type coils had to be used thereafter (Table 2). In the final phase, many helical-type coils were used, so space in the neck of the aneurysm still seemed to remain. We therefore selected 3D-type coils for the final stage; the last three coils inserted were two Target® Nano™ 360 detachable coils (1.5 mm × 2 cm; Stryker Neurovascular) and a HyperSoft® 3D coil (1.5 mm × 2 cm; MicroVention/Terumo, Tustin, CA, USA) in that order (Table 2). Microcatheters showed frequent kickback, especially when trying to insert the last three coils, and the position of the microcatheter tip was not sufficiently confirmed, but we managed to insert the coils (Figure 1(b)). Postoperative volume embolization ratio (VER) was 28.1%. No coil deviation was seen on skull radiography just after endovascular therapy, and no neurological abnormalities were evident afterward. Both lower limb pain and swelling started on day 3 after endovascular treatment and deep vein thrombosis was diagnosed. Administration of edoxaban was started at 30 mg/day. Although routine follow-up computed tomography (CT) demonstrated no abnormality on postoperative day 22 (Figure 1(c)), a shadow of a coil was confirmed around the corpus callosum by CT on postoperative day 29, and was considered to represent delayed distal coil migration (Figure 2(a)(a)). No new infarction was seen on magnetic resonance (MR) imaging performed the same day (Figure 2(a)(b) and (c)) and the patient remained asymptomatic. CT on day 57 after coil embolization showed a small migrated coil mass that moved further distally (Figure 2(a)(d)), but the patient remained neurologically normal. As of more than one year postoperatively, the patient remains asymptomatic and is undergoing regular follow-up observation. No new abnormal findings are noted on MRI and skull radiography. Edoxaban (30 mg/day) and aspirin (100 mg/day) continue to be administered.

Figure 1.

(a) Three-dimensional computed tomography (CT) (oblique view (a)) angiography demonstrates a saccular aneurysm arising on the azygos anterior cerebral artery (arrow). The size of the aneurysm was approximately 7 mm in height and 6 mm in width, with a neck measuring 4 mm. (b) Two microcatheters were advanced via the intermediate catheter and two framing coils were inserted via two microcatheters, respectively. Regardless of frequent kickback of the microcatheter, coil embolization was continued while coils were taken in and out, pulling and pushing the microcatheter delicately while inserting the coil in and out. (c) CT on day 22 after coil embolization shows no abnormality except for old brain damage in the right occipital lobe due to previous subcortical hemorrhage.

Table 2.

Coils used in coil embolization.

Microcatheter	Eshelon10 (straight)	SL10 (Pre-shaped 45-degree angle)
	① Target 360 soft 4.5 mm × 12 cm (framing)	② Target 360 soft 5 mm × 10 cm (framing)
	③ Target 360 soft 3 mm × 8 cm	④ Target 360 soft 3 mm × 8 cm
		⑤ Target 360 ultra 3 mm × 4 cm
		⑥ ED 2.5 mm × 4 cma
		⑦ ED 2.5 mm × 4 cma
		⑧ ED 2.5 mm × 4 cma
		⑨ ED 2 mm × 3 cma
		⑩ ED 2 mm × 3 cma
		⑪ ED 1.5 mm × 2 cma
	⑫ ED 1.5 mm × 2 cma	⑬ ED 1.5 mm × 2 cma
		⑭ ED 1.5 mm × 2 cma
		⑮ Target Helical 1.5 mm × 2 cma
		⑯ Target Helical 1.5 mm × 2 cma
		⑰ ED 1.5 mm × 2 cma
		⑱ Target 360 ultra 1.5 mm × 2 cm
		⑲ Target 360 ultra 1.5 mm × 2 cm
		⑳ HyperSoft 3D 1.5 mm × 2 cm

Numbers indicate insertion order.

^aHelical type coil.

Figure 2.

(a) Computed tomography (CT) on day 29 after coil embolization shows distal coil migration on the corpus callosum (a); however, magnetic resonance (MR) imaging reveals no infarct lesion on diffusion-weighted imaging (b) and MR angiography (c) shows obstruction of the aneurysm and an artifact in the A4 segment due to the migrated coil (double arrow). Although CT on day 57 after coil embolization shows a small coil mass had moved more distally, the patient was asymptomatic (d). (b) Skull radiography immediately after coil embolization shows no abnormality (a). Fifty-seven days after endovascular treatment (b), a small coil mass that had separated from the main coil ball is seen (arrow). (c) A helical coil appears to have separated from the main coil ball.

Discussion

Intraprocedural coil migration is a well-known complication of coil embolization for intracranial aneurysms.^1–3 Coil migration usually occurs during the embolization procedure or within a day thereafter (acute phase). After the acute phase, within a week after the procedure, proliferation of vessel endothelial cells generally improve coil stability (subacute phase),⁴ so delayed coil migration after coil embolization of an aneurysm, particularly in the chronic phase of several weeks or months, seems very rare.^4–13

Coil migration represents deviation of the coil, and two patterns are seen. One involves part of the coil protruding into the parent artery,^{4–6,8–10,13,14} which can cause thromboembolic events. The other type is complete displacement of the coil mass itself with the blood flow out to the distal end of the parent artery.^7,11,12,15 This type could result in parent artery occlusion with potentially serious ischemic complications if not removed promptly.¹⁶

Rare complications of delayed coil migration can be detected after several months (Table 1), and the modality by which coil migration is detected varies. In general, MR angiography is often used for the follow-up after coil embolization;¹⁷ on the other hand, some have reported cases in which asymptomatic distal coil migration was confirmed by CT in follow-up after the coil embolization^11,12 or by follow-up angiogram.^5,7 In our case, radiography of the skull (Figure 2(b)(a)) immediately after endovascular coil embolization and CT on day 22 after therapy was performed for routine follow-up (Figure 1(c)) revealed no particular abnormalities at that time. We found delayed distal coil migration for the first time with routine follow-up CT on postoperative day 29 (Figure 2(a)(a)), and CT on postoperative day 57 showed more distal movement of the migrated coil (Figure 2(a)(d)). This is the first report to confirm coil movement over time, and a helical-type coil seemed to be the migrated coil according to skull radiography (Figure 2(c)). One of the small, short helical-types from nos. 11 to 17 inserted (Table 2), mainly inserted near the neck, was thought to have not tangled well, and was thus finally extruded out of the coil ball. As reported by Gao et al.⁷ and Banerjee et al.,¹¹ the reasons for the lack of symptoms in our patient might be that the migrated coil was a helical type (Figure 2(c)), providing a stent-like configuration to the migrated spiral coil, and thus acting as a conduit allowing continued blood supply through the peripheral artery. Saguchi et al.¹² also reported an asymptomatic case in which peripheral blood flow did not stop from the migrated point, and in their case as well, many helical-type coils had been inserted in the latter phase of treatment, so the deviated coil seems likely to have been a helical type as in our case. Once the coil migrates, it may move further distally and frequent observations are necessary over time.

In previous reports, factors considered to contribute to delayed coil migration have been stated as aneurysm shape, aneurysm site, broad neck, low dome:neck ratio, coil size and design, use of soft coils, and unstable catheter position.^3,10,12,13 Although the impact of anticoagulation therapy on coil migration might have to be discussed, we have not been able to retrieve data on the relationship between them in PubMed research so far. Balloon- or stent-assisted techniques are effective and used to reduce coil migration. However, several distal migration cases associated with these techniques have been reported,^2,7,9,10,15 either because placed fine coils might escape from stent struts or soft coils simply compressed by the balloon might migrate. Banerjee et al.¹¹ stated that one reason for coil migration might be the actual aneurysm is larger than apparent and ultrasoft coils may eventually be dragged out by blood flow. Kamide et al.¹³ reported the first case of computational fluid dynamics analysis of coil migration after endovascular therapy. They stated that wall shear stress and vortical flow in a residual aneurysm may be factors inducing coil migration. Three months after treating an aneurysm at the internal carotid bifurcation, delayed partial coil migration was detected in a symptomatic case described by Motegi et al.¹⁰ They considered the two soft GDC 10-US coils newly placed in the neck remnant and partly independent of the main coil ball were exposed to blood flow with a vector into the middle cerebral artery. Our case represents the first report of distal coil migration associated with azygos ACA aneurysm. An azygos ACA is a very rare anatomical anomaly with an incidence of about 1.1%,¹⁸ but the frequency of associated aneurysm is relatively high, at 13%–71%,¹⁸ and coil embolization of azygos ACA aneurysms seems likely to increase in the future. Most aneurysms on the azygos ACA arise at the distal bifurcation of the parent artery,^18–20 and Kaspera et al.²⁰ stated that hemodynamic stress related to the azygos bifurcation geometry with a bent course of the artery around the genu of the corpus callosum predisposes to aneurysm formation. In our case, no balloon or stent was used, and under a situation of direct flow from the azygos ACA to the wide neck, an inserted soft, small, short coil that had not tangled sufficiently seems to have been exposed to blood flow, finally being carried distally. The end stage of coil insertion for aneurysms often involves the use of soft, short, small-diameter coils. In our case, when inserting the last three helical coils, the microcatheter showed frequent kickback. In addition, the tip of the microcatheter was hidden by the coil ball during the final stage of treatment, so the state of coils during insertion could not be sufficiently confirmed. Coil insertion was continued because the coils seemed to be smoothly inserted under fine control of the microcatheter, regardless of the frequent kickback. In our case, problems may also have arisen in the insertion order itself, placing 3D-type coils after helical-type coils. This is because 3D-type coils might interfere with helical-type coils that have not entangled completely with framing or filling coils. From the middle phase of coil embolization, we continued to insert many helical-type coils smoothly (Table 2; coil nos. 6–16), so we thought space in the neck remained and we changed to the use of 3D-type coils for the last three (Table 2; coil nos. 17–19). From the initial stage, we should have repositioned the microcatheter again in the proper position and intended to finish coil embolization as much as possible with longer 3D-type coils only, without extensive use of small, short, helical-type coils. Small coils with a small diameter and short length in the final stage of coil embolization do not raise the VER greatly, and coil migration as a complication could greatly affect the prognosis. Particularly in cases of an unruptured aneurysm, and in cases of a wide-necked aneurysm, incomplete coil embolization could be acceptable to some extent in the final stage of endovascular treatment.

Conclusion

Delayed coil migration after endovascular treatment for an aneurysm is very rare. We have described the first case of delayed distal coil migration associated with azygos ACA, and also the first case in which distal coil movement was confirmed. Delayed distal coil migration carries a risk of severe ischemic complications, so if the microcatheter frequently kicks back in the final stage of embolization, small and short coils should not continue to be inserted, especially for wide-necked aneurysms. Since the migrated coil may move more distally over time, more frequent observation is required.

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.