Article Text
PDF
Abstract
Background Anterior choroidal artery (AChA) aneurysms represent a small subset of cerebral aneurysms. The Pipeline Embolization Device (PED) has been successfully applied to various aneurysms of the supraclinoid internal carotid artery (ICA). The treatment of these aneurysms requires special attention due to the eloquent territory supplied by the AChA. We report the largest and first dedicated series of flow diversion treatment of AChA aneurysms.
Methods Four institutional neurointerventional databases were reviewed for cases of intracranial aneurysms treated with PED. Patient and aneurysm data as well as angiographic imaging were reviewed for all cases of AChA aneurysms treated with PED. AChA aneurysms were defined as aneurysms distal to the AChA and proximal to the ICA terminus, with or without the incorporation of the AChA.
Results Eighteen AChA aneurysms were treated during the study period. All aneurysms were successfully treated with a mean follow-up of 19.1 months. The large majority of aneurysms (15/18, 83.3%) were completely obliterated. No patients suffered from intra- or post-procedural complications. A1 stenosis was a common occurrence, seen in 10 of 16 (62.5%) covered anterior cerebral arteries (ACAs), although all were asymptomatic. All AChAs remained patent at last follow-up.
Conclusions The PED can be used successfully in AChA aneurysms with a good safety and efficacy profile. All AChAs remained patent. Collateral flow networks, especially for the ACA, affect long-term branch vessel patency. Treatment with PED for AChA aneurysms appears to be a reasonable option to consider and should be evaluated in a larger cohort.
- artery
- hemorrhage
- aneurysm
- angiography
- flow diverter
Statistics from Altmetric.com
Introduction
Since its initial approval for the treatment of large and giant intracranial aneurysms from the petrous to the superior hypophyseal segment of the internal carotid artery (ICA), the Pipeline Embolization Device (PED; Medtronic, Minneapolis, Minnesota, USA) has been used for other anterior circulation aneurysms, especially distal sidewall aneurysms of the ICA.
Anterior choroidal artery (AChA) aneurysms are relatively uncommon, accounting for 2–5% of all cerebral aneurysms.1 As a result, even large series include only small numbers of AChA aneurysms. These lesions are typically saccular and arise at or near the origin of the AChA, with distal lesions on the AChA rather uncommon.2 Intraventricular hemorrhage occurs in approximately one quarter and acute hydrocephalus in just over half of AChA aneurysm-related subarachnoid hemorrhage (SAH). In the Kuopio Cerebral Aneurysm Database approximately two-fifths of AChA aneurysms were ruptured; these represented 8% of all SAH cases but only 2% of all aneurysms overall.2
Some of the particular concerns while considering treatment of these aneurysms are related to the proximity of the AChA and the eloquence of the territory supplied by the AChA.3 Occlusion of the AChA and infarction of its territory classically results in contralateral hemiparesis, hemianesthesia, and homonymous hemianopsia.4 AChA aneurysms can be treated by the gamut of neurovascular techniques including microsurgical clipping, coiling, assisted-coiling, or flow diversion (FD). Each of the first two has its own strengths and caveats; FD is the subject of this evaluation. AChA territory infarction has been reported as a consequence of both clipping and endovascular therapy.5–11
Traditional endovascular coil embolization requires direct microcatheterization of the often small and multilobed AChA aneurysms which may prove technically challenging. In addition, it is often necessary to leave a neck remnant during coil embolization to preserve the patency of the AChA, given their frequent shared origin. Thus, FD offers an attractive alternative: it has worked well for other aneurysms of the carotid siphon, avoids direct aneurysmal catheterization, and may offer better occlusion rates for these aneurysms that are challenging for coiling.3 To date, the results of FD on AChA aneurysms has not been reported in detail as a separate entity.
Methods
Four institutional endovascular databases were searched for patients treated with FD. Aneurysms of the AChA were identified. AChA aneurysms were defined as aneurysms that were distal to the AChA but proximal to the ICA terminus, with or without the incorporation of the AChA. Aneurysms that involved the posterior communicating artery (PCoA) or whole communicating segment were excluded. From these, we selected patients who had at least 3 months of clinical and angiographic follow-up. Clinical data and all applicable imaging were reviewed. The study was approved by the local institutional review boards. Due to the small number of cases, statistical analyses were not performed.
Procedures
All procedures were performed under general anesthesia after obtaining appropriate consents. Patients were systemically heparinized with an activated clotting time of ≥200 s throughout the procedure. PEDs were delivered via a Marksman (Medtronic) or compatible microcatheter in standard fashion. Satisfactory wall apposition was confirmed with cone-beam CT.
Antiplatelet regimen
All patients were treated with a dual antiplatelet regimen (aspirin 81 or 325 mg and clopidogrel 75 mg) 5–7 days prior to elective placement of PED. While platelet inhibition was checked for some procedures, it was not yet available at our institution at the time of some of the procedures in the series. When available, it was checked immediately preceding the procedure with the P2Y12 assay (VerifyNow; Accumetrics, San Diego, California, USA). Clopidogrel non-responders were changed to an alternative antiplatelet agent. An aspirin response unit value of ≤550 and a Plavix Response Unit of ≤220 were considered to indicate an appropriate level of platelet inhibition for treatment. Patients were maintained on aspirin and clopidogrel for at least 6 months following PED placement and aspirin indefinitely thereafter.
Results
Retrospective review of the four institutional databases yielded a total of 18 AChA aneurysms from 2010 to 2017. All aneurysms either involved the AChA at their proximal neck (n=16) or arose from the ICA between the AChA take-off and the ICA terminus (n=2). None of these cases presented in ruptured status, although this reflects a selection bias due to the antiplatelet requirements for FD treatment. Mean follow-up was 19.1 months (range 3–54 months). All lesions were successfully treated, with partial remnant in three cases and complete obliteration in the remainder. Full information for all aneurysms treated is shown in online supplementary table 1.
Supplementary file 1
Jailed branches
The outcome of the various branches jailed was assessed on the most recent angiographic follow-up (online supplementary table 1). The A1 was jailed in 16/18 cases (88.9%), the PCoA in all 18 cases (100%), the AChA in all cases, and the ophthalmic artery in 14/18 cases (77.7%). Of these, the ophthalmic artery was fully patent in 11 cases and showed a slight reduction in 3 (21.4%). The jailed A1 was occluded in 1 case, stenotic in 9 cases (62.5%), and remained fully patent in 6. The PCoA was occluded in 2, downregulated in 5 (38.9% occlusion/downregulation), and was patent in 11. All AChA were patent with no flow reduction, although there was some reduction in caliber or stenosis in 3 cases. All stenoses and occlusions were asymptomatic.
Case example
A middle-aged patient was found to have a right AChA aneurysm during routine imaging for a meningioma. Diagnostic angiography demonstrated a saccular aneurysm measuring 3.3 mm with a 2.8 mm neck, incorporating the AChA at the base. Due to a family history and personal risk factors, treatment was recommended. He was treated with a single 3.75×14 mm PED placed from the M1 to the supraclinoid ICA, covering the aneurysm and the AChA. At 6-month digital subtraction angiography (DSA) the aneurysm had decreased in size, with moderate stenosis of the A1 and reduction of flow. At 18-month, 2.5-year, and 4.5-year DSA there was progressive reduction in the aneurysm size until complete obliteration with some eventual improvement of A1 stenosis (figure 1). The AChA remained patent throughout and the patient was asymptomatic without any complications.
Treatment of anterior choroidal artery (AChA) aneurysm in the example case. (A) CBCT-A showing the aneurysm (IA) and relevant local anatomy. The size of the A1 can be appreciated to be normal. (B) At 6-month follow-up there is focal severe stenosis at the A1 origin (white arrow) with reduction in size of the aneurysm. (C) At last follow-up at 4.5 years post-embolization the aneurysm is completely obliterated, with now only moderate stenosis of the A1 origin (white arrow) and a patent AChA. (D) Both anterior cerebral artery territories are supplied well by the contralateral side via a large anterior communicating artery.
Discussion
AChA aneurysms are relatively uncommon2 and, as such, have not been given direct consideration with regard to FD treatment. Even at centers with a large endovascular experience, there have been only limited reports or mention of the use of FDs to treat AChA aneurysms,12 despite their frequent use in other supraclinoid ICA aneurysms. We review our group’s results with FD specifically for AChA aneurysms along with some of the technical nuances.
Surgical treatment of AChA aneurysms was described initially by Drake et al 5 and then by Yasargil et al.8 The largest series by Friedman et al focused on ischemic complications, which is the most concerning.13 The mechanisms of ischemia include iatrogenic clip occlusion, clip torsion, and vasospasm. Due to the eloquence of the AChA territory, these ischemic complications can be quite devastating; as such, treatment methodologies that avoid this are intriguing.
Coil-based treatment of AChA aneurysms has been described in several series,7 14 the largest of which included 90 aneurysms by Kang et al.6 In their series, they found a favorable clinical outcome in 90% of patients. However, due to the importance of preserving the AChA origin during coiling, sometimes the therapeutic goal may be limited to subtotal treatment. As such, the rate of complete occlusion was only 17%, with most (77%) being only near-complete.
Classification of AChA aneurysms
Several classification schemata for AChA aneurysms have been used.2 6 13 Some are most applicable for microsurgical treatment, focusing on the angle of take-off of the vessel and the viewability in the surgical window.2 13 That described by Kang et al used two main types based on the location of the AChA relative to the aneurysm neck, which is quite applicable for coil embolization.6 They classified two general types, one in which the AChA arises from the junction of the ICA and the aneurysm (Type A), and the other (Type B) in which there is a clear gap between the ICA and the AChA. Separately, they considered a few observed variations of the AChA. By our assessment, this classification may be overly nuanced for applicability in FD. Rather, we simply look at aneurysms that incorporate the AChA directly and those that do not.
Using this schema, the former is by far the most common, representing 16 of 18 aneurysms in our series. The latter accounted for the other two cases. This is in line with the much larger series by Lehecka et al, in which aneurysms arising from the AChA itself were extremely rare, as were aneurysms occurring between the AChA and another vessel such as an anomalous anterior temporal artery (ATA) (or between a duplicated AChA).2
AChA, ICA, and MCA anatomy: therapeutic implications of collaterals and perforators
A variety of anatomical considerations impact AChA aneurysm FD treatment selection and planning, as well as the expected outcome and potential deficits. These are complex, multifactorial, and include (1) length of communicating ICA and pre-bifurcation; (2) AChA origin; (3) ipsilateral A1 dominance, robustness of contralateral A1, and collateral flow across ACoA; (4) collateralization of AChA-irrigated territory; and (5) extent of AChA anastomoses with other vessels.
A1 jailing outcome and collateral flow
There is great variability in the anatomy of the anterior cerebral artery (ACA) and anterior communicating artery (ACoA), and the collateralization of flow between the hemispheres.15 This variation affects the outcome of a jailed A1 in a flow-dependent manner, as has been discussed previously.16 In our study the rate of jailed A1 stenosis/occlusion was high (62.5%), although all were asymptomatic due to cross-flow. This type of chronic occlusion has been described in parent vessels to be likely asymptomatic and dependent on the presence of collaterals.16 Gascou et al reported 5/5 patients who had reduced ACA flow and stenosis when jailed,17 with three of these patients having transient symptoms without definite infarct. The high rate of jailed A1 stenosis/occlusion requires evaluation in a larger cohort and is an important question to answer in the age of FD for treatment of distal ICA and proximal MCA aneurysms.
AChA origin and relevant supraclinoid ICA and M1 anatomy
The AChA (along with its aneurysms) arises off the inferolateral posterior wall of the ICA approximately 3 mm distal to PCoA take-off,5 18 and the distance between AChA and ICA bifurcation averages 4 mm (range 1.5–9 mm).19 This essentially guarantees that FD treatment of any AChA aneurysm also jails the PCoA, a potential concern in cases of a fetal posterior cerebral artery (PCA). It is notable that a larger PCoA (and especially fetal PCA) generally annexes some of the AChA territory, and these two arteries have inversely proportional sizes.20
Variant origins of AChA include a mono-original vessel with early bifurcation into two separate vessels or a double origin vessel arising off the ICA, or origin from ICA bifurcation, PCoA, or M1.21–25 In addition, the variant of an ATA originating from the ICA distal to the AChA was seen in two of our cases. Awareness of the course of the AChA is critical in any treatment modality.
The proximity of the AChA to the ICA terminus often necessitates placement of the FD in the M1 (89% of our cases). Thus, the pre-bifurcation M1 length is also a critical consideration, as to avoid intrusion of the FD into the MCA bifurcation in a short M1 segment, jailing one of the M2s. Lastly, the M1 segment contains critical lenticulostriate vessels26; while these have been shown to be safely jailed,27 it should ideally be minimized.
AChA collateral flow and outcome
While the AChA supplies many critical structures including the visual apparatus and cerebral peduncle, intentional occlusion when used to treat Parkinson’s tremor only resulted in a 6% rate of hemiplegia.28 29 This study and others showed that AChA-irrigated parenchyma can be salvaged by other collateral supply such as direct perforators from the C7 segment of the ICA, PCoA, and PCA. Although all AChAs remained patent in our study, the potential for compensatory blood supply in the case of AChA occlusion is present, and probably more easily activated in the case of gradual occlusion (as in FD) than acute occlusion (as with a clip).
The fate of the jailed AChA deserves special consideration. Bhogal et al found that, among 91 cases of AChA jailing (for a variety of aneurysm locations), there were no cases of reduced flow or occlusion, as demonstrated in an earlier smaller series of 21 jailings.30 Raz et al 31 described a single case of AChA coverage leading to asymptomatic occlusion for a PCoA aneurysm. FD for PCoA aneurysms resulted in occlusion of the vessel in approximately half of cases,32 with the fate of the remaining PCoAs evenly divided between diminished flow and complete patency (similar to our findings).
The extremely low occlusion rate of the AChA can be explained by its eloquent territory (maintaining demand for this vessel) and is in keeping with our findings. However, there have been notable cases of symptomatic occlusion following jailing of the AChA with PED; even if the rate is low, due to the severity of the symptomatology, it should be given special attention.33 34 This adverse outcome can occur even in very delayed fashion, reported to be as long as 624 days post-procedure.35
In our experience and that of others, the incorporation of vessels at the dome or neck of an aneurysm is a predictor of treatment failure with FD.16 36 37 This same concept should apply to AChA aneurysms, especially in light of the fact that 16/18 aneurysms in our series incorporated the index vessel at the proximal neck. According to our previous reports, this should prevent successful occlusion of the aneurysm, but we suspect that the small caliber of the AChA does not prevent aneurysm occlusion with FD. We postulate that there may be a threshold of flow through the vessel incorporated, under which aneurysm obliteration and endothelialization across the stent may still occur. In other words, flow through a fetal PCA may be too high to allow occlusion of a fetal PCA aneurysm37 whereas flow through an AChA may be low enough to allow aneurysm occlusion. Interestingly, this may be why the time to occlusion may be longer in some of these aneurysms that incorporate the AChA, such as the example case. For the rare AChA aneurysm that arises separately from the AChA, its obliteration rate should mirror that of other carotid sidewall aneurysms.
Based on anatomy and experience, we note that FD treatment for AChA aneurysms should be attempted with more caution in the case of a large PCoA/fetal PCA, short C7 ICA, short M1, absent ACoA (especially with an azygos ACA), and lack of a good collateral network. These factors may only minimally affect the choice to treat by FD and need to be appraised in a larger cohort.
Distal AChA aneurysms
Distal AChA aneurysms are exceedingly rare but, when they occur, they are commonly fusiform.2 13 38–42 They are not ideally treated by FD due to retrograde flow to fill the aneurysm by extensive anastomoses distally. Instead, they may be treated by coil embolization,43 clip ligation,21 or vessel sacrifice.38
Limitations and future directions
As a retrospective report, this study has several limitations. First, the cases described have a clear selection bias; these are aneurysms or patients who were determined not to be strong candidates for either surgical or traditional endovascular management. Next, they represent ‘off-label’ cases and that should be considered in management decisions. As such, we present an initial experience for AChA aneurysms, with the expectation of prospective studies to assess the outcomes more rigorously in a larger cohort. This report demonstrates the feasibility and satisfactory results of FD treatment of AChA aneurysms across multiple centers. Based on our findings, it seems reasonable to consider FD in the treatment armamentarium for these lesions, at least in cases that are unfavorable for traditional options. The indications and contraindications for using FD should remain flexible until further experience can be garnered and other groups corroborate our experience. Decisions should be made on an individual basis, and coiling or clipping may still be safer until stronger data supporting FD in these aneurysms are available.
Conclusions
The PED is safe and provides good outcomes in the treatment of select AChA aneurysms. In our series it was largely successful, with obliteration rates reflecting those seen in larger series of ICA sidewall aneurysms and the preservation of the AChA in all cases. Further study is required to understand the unique function of FD for this uncommon subset of aneurysms.
References
Footnotes
Contributors All authors met the criteria for authorship.
Competing interests None declared.
Ethics approval Institutional Review Boards at each of the involved institutions.
Provenance and peer review Not commissioned; externally peer reviewed.