Flow Diversion in Middle Cerebral Artery Aneurysms: Is It Really an All-Purpose
Treatment?
Osman Melih Topcuoglu1, Erol Akgul2, Ergun Daglioglu3, Elif Dilara Topcuoglu5, Ahmet Peker6, Ilkay Akmangit4, Deniz Belen3, Anil Arat4,6
-BACKGROUND AND OBJECTIVE:The outcome of flow diversion for middle cerebral artery (MCA) aneurysms, one of the most common sites for intracranial aneurysms, has not been defined thoroughly. We assessed our outcomes in MCA aneurysms (MCAAs) treated by flow diversion, i.e., with either dedicated flow diverters or telescopic stents.
-METHODS:Patients with MCAAs were treated by flow diversion if surgical or other endovascular treatment mo-dalities had failed or were deemed likely to fail. Angio-graphic and clinical outcome of these patients was assessed retrospectively. Aneurysm location on MCA was defined as M1 segment, “true bifurcation” (classical bifurcation of MCA into superior and inferior trunks), “variant bifurcation” (bifurcation of early frontal or early/ distal temporal branches), or M2 segment. Aneurysm morphology was classified as saccular versus dissecting/ fusiform.
-RESULTS:Treatment was attempted in 29 MCAAs. Technical failure rate was 3.4% (1/29). Thirteen of aneu-rysms were fusiform. Of the bifurcation aneuaneu-rysms, most (10/16) were the variant type. Overall and procedure-related mortality/permanent morbidity rates were 10.3% (3/29) and 3.5% (1/29). Total occlusion rates (mean angio-graphic follow-up 10.3 months) for saccular and fusiform aneurysms were 40% and 75%, respectively. In bifurcation aneurysms, occlusion was strongly associated with side-branch occlusion (P< 0.005).
-CONCLUSIONS:In this series, flow diversion for the treatment of MCAAs was safe, was effective in the treat-ment of fusiform MCAAs, and was not as effective at mid-term for MCA bifurcation aneurysms. Unsatisfactory oc-clusion rate in bifurcation aneurysms likely results from residual filling of the aneurysms in cases in which the jailed side branch remains patent.
INTRODUCTION
F
low diversion using dedicatedflow diverters (FDs) or tele-scopic stenting has transformed the endovascular treat-ment of aneurysms.1 Most commonly, flow diversion involves the intra-arterial placement of dedicated FDs. If deemed technically possible, telescopic placement of multiple intracranial stents has been used to achieve a similar angiographic and clinical outcome.2,3 Multicenter prospective studies on the safety and effectiveness of both methods have focused on the treatment of internal carotid artery (ICA) aneurysms4,5; however, ICA aneu-rysms comprise only about 20% of all intracranial aneuaneu-rysms.6 Based on good outcomes in the ICA, indications for flow diversion (and hence FDs) quickly were extended to include aneurysms in the posterior circulation or distal territories of anterior circulation, also including ruptured aneurysms.1Half of cerebral aneurysms are distally located bifurcation le-sions, such as those on the anterior communicating artery or middle cerebral artery (MCA) bifurcations.6 The success rate of
Key words -Aneurysm
-Endovascular treatment -Flow diversion -Flow diverter -Middle cerebral artery Abbreviations and Acronyms
FD: Flow diverter
ICA: Internal carotid artery
MCA: Middle cerebral artery
MCAA: Middle cerebral artery aneurysm
MCA-B: Middle cerebral artery true bifurcation
MCA-BV: Middle cerebral artery variant bifurcation
From the1
Department of Radiology, Kecioren Education and Research Hospital, Ankara;
2
Department of Radiology, Cukurova University Medical School, Adana; Departments of
3
Neurosurgery and4
Radiology, Numune Education and Research Hospital, Ankara;
5
Department of Radiology, Ufuk University Medical School, Ankara; and6
Department of Radiology, Hacettepe University Medical School, Ankara, Turkey
To whom correspondence should be addressed: Anil Arat, M.D. [E-mail:anilarat@hotmail.com]
Citation: World Neurosurg. (2016) 87:317-327.
http://dx.doi.org/10.1016/j.wneu.2015.11.073
Journal homepage:www.WORLDNEUROSURGERY.org
Available online:www.sciencedirect.com
flow diversion in bifurcation aneurysms is still in question,7-9 and its safety in aneurysms distal to the circle of Willis has not yet been determined. One of the major concerns in the use of FDs in these arterial territories is the potential for occlusion of the covered or “jailed” side branch.10,11
There are currently 3 single-center studies reporting the outcome of flow diversion specifically in middle cerebral artery aneurysms (MCAAs).12-14 The current reports about clinical and angiographic results of flow diversion in bifurcation aneurysms claim surprisingly favorable results, given the lower rate of aneu-rysm occlusion with FDs in the experimental setting.8,15 In this report, we aimed to evaluate the results of endovascular man-agement of MCAAs withflow diversion and emphasized the role of aneurysm morphology and MCA anatomy on the outcome. MATERIALS AND METHODS
A retrospective evaluation of our case records was performed to identify patients with MCAAs who were treated from August 2010 to February 2015 byflow diversion. Imaging findings and patient re-cords were evaluated retrospectively for initial presentation, aneu-rysm size and morphology, variations of arterial anatomy, devices used forflow diversion, and outcome on follow-up. The hospital’s Institutional Review Board approved the study design and waived the requirement for informed consent for use of pertinent data for publication (Institutional Review Board no. 2015-995).
Patient Selection
In our practice,flow diversion is the primary treatment for fusi-form MCAAs. Patients with fusifusi-form/dissecting aneurysms essentially are treated with flow diversion regardless of their rupture status if there are no absolute contraindications such as very recent surgery, gastrointestinal bleeding, unshunted hydro-cephalus, or severe thrombocytopenia. In contrast, most patients with saccular aneurysms, including those located at the MCA bifurcation or at the origin of a large MCA perforator, are offered both surgical and endovascular treatment options.
Patients are referred directly for surgical treatment if coiling (bare, balloon- or stent-assisted) fails or is deemed likely to fail on the basis of preprocedure imagingfindings. In these cases surgery generally is successful. If not, we proceed to flow diversion regardless of the MCA branching pattern. Infrequently, a patient with an MCAA deemed unfavorable for coiling refuses surgical treatment or there is a residual/recurrent aneurysm after clipping or stent-assisted coiling. Then, we proceed toflow diversion. We aim to refrain from placing FDs in MCAAs that are suitable for coiling or clipping.
The preferred technique forflow diversion in our practice is placement of FDs. If this fails or if access across the aneurysm neck appears difficult, we switch to telescopic stenting for flow diversion.
Definition of Aneurysmal Location on MCA
In papers discussing endovascular or neurosurgical treatment of cerebral aneurysms, MCAAs typically are categorized on the basis of their location as proximal (M1 segment), bifurcation, or distal aneurysms.16 In the current study, they were categorized somewhat differently, as M1-segment, middle cerebral artery
variant bifurcation (MCA-BV), middle cerebral artery true bifur-cation (MCA-B), or distal MCAAs.
M1-segment aneurysms, also known as horizontal-segment aneurysms, generally are defined on the basis of their location between the MCA origin and the bifurcation, or alternatively from the origin to the end of the lenticulostriate arteries.17Although the MCA usually branches into the superior and inferior trunks (true bifurcation) after the M1 segment, variant anatomy is observed in some patients.17 In addition, because there are often 2 or more major branches proximal to this bifurcation, classifying an MCAA as either bifurcation or M1 segment may not be straightforward.18,19 Inconsistent definitions and classifications make direct comparison of success rates and outcomes in different patient series or treatment modalities challenging.
Endovascular or surgical compromise of a bifurcation branch will have different clinical consequences depending on whether the particular branch is an isolated cortical branch or a major arterial trunk. This is a critical distinction in arterial territories like the MCA, where the number of bifurcation aneurysms far out-weighs the number of sidewall or fusiform aneurysms, and where leptomeningeal collateral supply is limited. The lack of collaterals have further significance when the treatment—flow diversion in particular—can compromise flow in the side branch (“jailed” branch) or alternatively when the success of the treatment potentially relies on the cessation of flow in that branch. As a result, in this report we subdivided MCA bifurcation aneurysms into MCA-BV and MCA-B groups, The MCA-BV group includes proximal frontal, proximal temporal, and distal temporal branch aneurysms. The MCA-B group includes those aneurysms located at a“classical” bifurcation of the MCA into superior and inferior trunks. This classification is not arbitrary but is rather based on previous classifications of aneurysms at these locations: Elshar-kawy et al.17and Ulm et al.20made similar classifications, noting that misclassification of an aneurysm located at an early cortical branch as a true bifurcation aneurysm may lead to unforeseen injury to the lenticulostriate arteries during surgery. In aneurysms managed with flow-diversion techniques, misclassifi-cation of an MCAA may result in underestimation or over-estimation of success rates depending on whetherflow diversion was performed for MCA-B or MCA-BV aneurysms, respectively. To our knowledge, this concern has not been discussed in the current literature onflow diversion for MCAAs.
Morphology of MCAAs
The lenticulostriate aneurysms, true and false bifurcation aneu-rysms, and distal MCAAs reported in the comprehensive study by Elsharkawy et al.17 were all saccular in nature. Current endovascular techniques, however, allow us to treat both saccular and fusiform aneurysms effectively. We therefore included all types of MCAAs in our series, regardless of their morphology. In fact, the M1 and distal MCAAs tended to be fusiform rather than saccular in the cohort of MCAAs selected forflow diversion in this series.
Definition of Angiographic Outcome
There is currently no classification scheme to assess morphologic outcome after flow diversion in bifurcation aneurysms. Angio-graphic outcome assessed with the Byrne classification does not
allow for the designation of the status of the jailed side-branch21; nevertheless, in the current study, residual aneurysmalfilling was described according to the axis 1 of Byrne classification. Both the parent artery and the side branch were defined as “stenosed” if there was a reduction in arterial diameter 50%, otherwise the artery and affected side branch were described as “patent” or “occluded.”
Interventional Procedure
All patients were started on 75 mg clopidogrel and 300 mg aspirin daily for 5e10 days before the procedure. The 3 patients pre-senting with subarachnoid hemorrhage were treated after the acute phase (day 8, 10, or 14).
Endovascular treatments were performed by the authors (A.A., E.A., E.D., I.A.) via the standard transfemoral approach with the patient under general anesthesia. A bolus of 70e100 IU/kg of heparin was given intravenously immediately after insertion of the femoral sheath. A long guiding sheath was placed within the ICA, and a 6-French distal access catheter (Fargo [Balt Extrusion, Montmorency, France], Neuron [Stryker, Fremont, California, USA], or Navien [Covidien, Irvine, California, USA]) was then placed coaxially within the guiding sheath.
In patients treated with FDs, the aneurysm was bypassed with the delivery catheter (Vasco 21; Balt Extrusion) or the Silk device (Balt Extrusion), or directly by the device-catheter assembly for the Surpass device (Stryker) over a soft-tip 0.014-inch microguidewire. A second microcatheter (SL-10 [Stryker] or Echelon 10 [Covidien]) was jailed in patients with saccular aneurysms if partial coiling appeared to be feasible (3 cases). The device was deployed gently to cover the neck of the aneurysm, and partial coiling was per-formed. In cases in which increased wall coverage at the level of the neck was needed, another intracranial stent (“scaffolding stent,” Enterprise [Codman & Shurtleff, Inc., Raynham, Massa-chusetts, USA] or LEO stent, or LEO Baby [Balt Extrusion]) was deployed with the goal of undersizing the diverter before deployment of the FD (3 cases).
The status of the aneurysm sacs and any cortical branches/ prominent perforators originating from the aneurysm were assessed by angiography immediately after the endovascular pro-cedure and at follow-up. In all patients, serial angiograms were performed immediately after placement of the FD to look forflow stagnation within the aneurysm and to rule out thromboembo-lism. Aspirin (300 mg) and clopidogrel (75 mg) were continued for at least 6 months after the procedure in every patient. Aspirin was continued indefinitely, and if restenosis of the parent artery was noted on follow-up angiography, clopidogrel also was continued indefinitely. In cases in which coils also were used, patients were asked to return for clinical follow-up with computed tomography angiography or magnetic resonance angiography 1 month after discharge and for angiographic assessment at 6 months. Follow-up continued with annual computed tomography angiography or magnetic resonance angiography.
Statistical Analysis
The clinical significance of the association between aneurysm occlusion and side branch patency was assessed with the Fisher exact test. P values<0.05 were considered significant.
RESULTS
The current study included 28 patients with 29 MCAAs. There were 14 men and 14 women, with a mean age of 47.2 years (range, 3e65 years), harboring 17 saccular (58.6%, mean diameter 7.4 mm), and 12 dissecting/fusiform aneurysms (41.3%, mean diam-eter 19.1 mm). A total of 16 (55.2%) aneurysms were located at an MCA bifurcation, including 10 at variant bifurcations. A prominent perforator or a cortical branch originated in 18 aneurysms (62%). Three residual/recurrent aneurysms included in this series had been treated previously with coiling, and 4 had been clipped (Figure 1). Patient presentation and aneurysm characteristics are summarized inTable 1.
Flow diversion was performed either with dedicated FDs deployed with or without scaffolding stents. If attempts in placing these devices proved to be futile, we switched to telescopic stents as shown in Table 1. A single FD was placed in 23 aneurysms; multiple FDs were used in only one case (aneurysm 15). Four aneurysms were treated with telescopic stenting (triple LEO implantation in aneurysms 16 and 20; 2 LEO baby stents in aneurysm 21; triple Enterprise stents in aneurysm 23). Three LEO stents, 1 LEO baby, and 1 Enterprise stent were deployed as scaffolding stents (Figure 2). In 3 patients, partial coiling of the aneurysm sac was possible immediately after the placement of the Silk device, during the same endovascular procedure.
Outcomes after the procedure and at long-term follow-up are summarized inTable 2. The technical success rate was 96.5% (28/ 29). The mean duration of follow-up was 10.3 months (range, 2 days to 33 months). On last follow-up, the rate of complete oc-clusion was 55.6% (15/27, Byrne grade 4). The ococ-clusion rates for saccular and fusiform aneurysms were 40.0% (6/15, excluding the failed case and one follow-up angiogram that is pending) and 75.0% (9/12), respectively. All 17 saccular aneurysms had cortical branches/perforator arteries originating from the lesion (Figures 3 and4). In the 4 cases in which the side branch was occluded, the aneurysm itself also was occluded; however, among 11 of 17 saccular aneurysms where the saccular branch remained patent, only 2 aneurysms showed complete occlusion.
Residualfilling of aneurysms that were not completely occluded was mild (Byrne grade 3) in 5 aneurysms, moderate (Byrne grade 2) in 1, and significant (Byrne grade 0e1) in 6. Two fusiform/ dissecting aneurysms treated by telescopic LEO stents were totally occluded (Byrne grade 4), whereas the aneurysm treated with telescopic LEO Baby stents demonstrated near-totalfilling (Byrne grade 1).
We did not encounter perforator injury in any patient, although we used regular stents together with the FD in a many of our patients, as noted inTable 1.
In a comparison of the 6 saccular aneurysms that were classified as occluded (Byrne grade 4) versus the 9 deemed patent (Byrne grade 0e3), aneurysm patency was directly related to the patency of the side-branch (P< 0.005, Fisher exact test).
FDs were withdrawn in 4 patients. In thefirst case, angiograms obtained during deployment demonstrated fish-mouthing at the distal end of the scaffolding stent with impending thrombotic occlusion. The diverter was withdrawn, and distal wall apposition and brisk outflow was obtained after deployment of 2 more LEO stents (Balt Extrusion). There was no residual opacification of the
aneurysm after implantation of the telescopic stents, and FD placement was no longer needed. In the second case, the whole M1 segment was involved in a fusiform aneurysm. A long scaf-folding stent (LEO, 5.5 75 mm) was deployed to obtain aneu-rysm coverage. Because of stent overexpansion within the aneurysm, the largest diameter FD available (Silk, 5.5 mm) showed extreme foreshortening to approximately 12 mm during placement within the LEO stent, so the Silk device was not released. Because multiple large-diameter FDs would have been needed to cover the scaffolding stent, the Silk device was withdrawn and flow-diversion with significant intra-aneurysmal stagnation was ach-ieved by telescopic deployment of 2 large-size LEO stents.
In a case of a giant dissecting aneurysm, the aneurysm was bypassed with a 0.014-mm guidewire but a Surpassflow diverter device (Stryker Neurovascular, Kalamazoo, Michigan) would not navigate over the guidewire. It was withdrawn and exchanged with a LEO stent, which was to be used as a scaffolding stent for a Silk device. Because of tortuosity at the landing zone of the stent, the distal end of the stent failed to open and the stent was withdrawn. The aneurysm was finally treated by 3 telescopically placed En-terprise stents.
In 1 patient, the aneurysm recurred after stent-assisted coiling, and there was further recurrence after placement of a Silk device in a second session. The recurrence was treated by placement of 3 more FDs in a telescopic fashion in 2 more sessions. Although there was
only grade 1filling of the aneurysm on follow-up digital subtraction angiography, the patient had 2 reruptures over 33 months and remained in poor clinical status. Except for this patient, placement of overlapping FDs was avoided in the M1 segment.
Procedure-Related Complications
During 1 procedure, the inferior trunk of MCA thrombosed after Silk deployment and was recanalized with the intra-arterial administration of tissue plasminogen activator. A cerebral hema-toma developed in the ipsilateral hemisphere 2 days after the endovascular treatment, and the patient died 2 weeks later. Another patient had a postoperative hematoma around the thyroid gland 4 hours after the procedure. Emergency angiogram did not reveal any point of bleeding. The patient recovered fully in 3 days, with no need for blood transfusion or surgical intervention. Adverse Events During Follow-Up
An asymptomatic total M1 occlusion was detected on the 6-month follow-up angiogram in a patient with a giant M1 aneurysm. In 2 patients, asymptomatic parent artery stenosis was noted on 6-month follow-up angiography; spontaneous resolution of the stenosis was demonstrated at 24 months in 1 of these.
There were no other procedure-related adverse events during the postoperative course. In 1 patient, a Silk device was placed to treat a dissecting MCAAs during the subacute phase of subarachnoid
Figure 1. Case 10. Images obtained in a 52-year-old woman with a ruptured and previously clipped right-sided wide-necked middle cerebral artery (MCA) bifurcation aneurysm. (A) Digital subtraction angiogram and (B) 3-dimensional digital subtraction angiogram showing the right-sided postsurgical remnant of a saccular MCA bifurcation aneurysm. (C)
Nonsubtracted angiogram obtained after treatment shows a well-deployed flow diverter. (D) Digital subtraction angiogram and (E) 3-dimensional digital subtraction angiogram obtained at 12- and 18-month follow-ups show significant filling of the aneurysm and a patent cortical branch originating from the aneurysm.
hemorrhage. This patient tolerated the procedure well and had no clinical problems at discharge. The patient died 3 months later as the result of a spontaneous contralateral intracranial bleed. She was on 75 mg/day clopidogrel and 300 mg/day aspirin at the time of the hemorrhage. Another patient, who had been treated in multiple sessions with placement of 4 FDs, 2 stents, and coils (as mentioned previously) for a ruptured aneurysm died as the result of rerupture 2 years after thefirst session of endovascular treatment. Otherwise, there were no delayed neurologic events. The 30-day procedural mortality/permanent morbidity was 3.4% (1/29).
DISCUSSION
Among intracranial aneurysms, 14%e43% arise from the MCA.16 MCAAs are considered prone to rupture by some authors,22,23 whereas others believe they have a lower rate of rupture.16Most MCAAs are saccular and arise from the true MCA bifurcation, a MCA-BV, distal MCA branches, or lenticulostriate arteries.16 Fusiform aneurysms of the MCA are less common. Coil embolization has been performed safely and effectively in MCAAs,24 with outcomes approaching surgical results in
Table 1. Patient Characteristics, Aneurysm Characteristics, and Flow-Diversion Device
Aneurysm Age, years/Sex Presentation
Previous
Treatment Location Type Diameter, mm
Cortical Branch/Perforator Originating from Aneurysm
Device Used for Flow Diversion
1 61/F Headache No M1-P Saccular 5.0 Yes Silk
2 52/M Headache No MCA-BV Saccular 5.0 Yes Silk
3 42/F Headache No MCA-BV Saccular 5.2 Yes Silk (þcoiling)
4 58/M Headache No MCA-BV Saccular 6.2 Yes Silk
5 No MCA-BV Saccular 7.6 Yes Silk
6 53/F Incidental No MCA-BV Saccular 3.0 Yes Silk
7 64/M Stroke SAC MCA-B F/D 15.0 Yes Silk
8 44/M Headache Clipping M2 F/D 3.0 No Silk
9 43/M Headache Clipping M1-P Saccular 3.0 Yes Silk
10 52/F SAH Clipping MCA-B Saccular 13.0 Yes Silk
11 50/F Headache No MCA-B Saccular 6.0 Yes Silk
12 54/F SAH No M1 F/D 7.0 No Silk
13 65/M SAH No M1 F/D 25.0 No Silk
14 33/M SAH No M1 F/D 4.0 No Silk
15 60/M SAH SAC M1 F/D 17.0 No 3 Silk/1 Surpass
16 3/M History of trauma No M2 F/D 15.0 No 3 Leo
17 34/F SAH Clipping MCA-B Saccular 9.0 Yes Silk
18 15/M Headache No M1 F/D 35.0 No Silk
19 55/M Headache No MCA-B Saccular 5.0 Yes Silk
20 43/F Headache No M1 F/D 40.0 No 3 Leo
21 63/F Headache No M3 F/D 7.0 No 2 Leo Baby
22 57/M Incidental No MCA-BV Saccular 12.0 Yes Silk (þcoiling)
23 38/M SAH No M1 F/D 30.0 No 3 Enterprise
24 57/M Incidental No MCA-BV Saccular 4.0 Yes Silk
25 65/F Incidental No MCA-B Saccular 18.0 Yes Unsuccessful
26 48/F Incidental No MCA-BV Saccular 3.0 Yes Surpass
27 47/F Headache No M1 F/D 32.0 No Surpass/Leo
28 55/F Incidental No MCA-BV Saccular 4.0 Yes Silk (þcoiling)
29 52/F Incidental SAC MCA-BV Saccular 18 Yes Silk
F, female; M1-P, M1 perforator; M, male; MCA-BV, middle cerebral artery variant bifurcation; SAC, stent-assisted coiling; MCA-B, middle cerebral artery true bifurcation; SAH, subarachnoid hemorrhage; F/D, fusiform/dissecting.
experienced centers16,25,26; however, endovascular ap-proaches16,25,27-29have not outperformed surgery in the routine management of MCA bifurcation aneurysms.30 According to a recent study published by Gory et al.,31stent-assisted coiling for MCAAs was associated with greater rates of periprocedural com-plications and morbidity/mortality rates compared with surgical management.
The wide necks that are typical in MCAAs, as well as the prominent perforators and cortical branches originating from these lesions and tight angles at the origin of the involved branch, frequently have created the need for some form of assistance to coiling in these lesions.28,30,32 Besides the complex angioarchi-tecture of MCAAs, necessitating balloon or stent assistance for coiling, endovascular treatment is complicated further by the complex branching pattern frequently seen in MCAAs.17,20 Ac-cording to a cadaveric study by Ulm et al.,2039% of MCAAs arise from early frontal branches, and this subtype was more common than true bifurcation aneurysms in the current series. It is possible that in the literature, most of these “variant type”
aneurysms are referred to as “MCA bifurcation aneurysms” and intuitively considered as “true bifurcation aneurysms” if a distinction between the 2 types was not sought in particular by the authors as part of their methodology. In a larger series, Elsharkawy et al.17 found that 18% of MCAAs arose from an early cortical branch, this kind an anatomical variation may alter or even preclude conventional endovascular management. In the current series of aneurysms selected for treatment with flow diversion, 10 of 16 (62.5%) bifurcation aneurysms were of this early branching or“variant” type.
Elsharkawy et al.17 point out to the importance of discriminating between a true and a variant bifurcation aneurysm from a surgical standpoint; however, the impact of this variant anatomy on endovascular treatment may be more pronounced. In case of a variant bifurcation: 1) the side branch is a cortical branch that is expected to have a smaller diameter (generally around 1 mm or less) in comparison with an arterial trunk; 2) if coiling is to be performed, the risk of occlusion of these smaller side branches attributable to coil encroachment or
Figure 2. Case 13. Images obtained in a 65-year-old man with a ruptured right-sided middle cerebral artery (MCA) M1 segment aneurysm. (A) Three-dimensional digital subtraction angiogram showing the right-sided MCA M1 segment fusiform/dissecting aneurysm. (B) Immediate posttreatment nonsubtracted angiogram showing a well-opened flow diverter and also a Leo stent, which is used as a scaffolding stent. (D) Digital subtraction angiogram obtained at the 6-month follow-up and (D) coronal reformatted computed tomography angiography obtained at the 18-month follow up both showing total occlusion of the aneurysm.
stent thrombosis is greater; and 3) the acute take-off these branches has resulted in acute branch occlusions that are associ-ated with undue straightening of the branch with laser-cut self-expandable stents or elongation of braided stents in our practice. These characteristics would favor FD as an endovascular option.
Recently,flow diversion has emerged as an alternative to sur-gical treatment or other endovascular approaches in the man-agement of aneurysms, and FDs have changed the treatment algorithm in ICA aneurysms.1Whether the success of FDs in the ICA can be reproduced in other arterial segments remains to be seen.11A search on PubMed and Science Citation Index shows 3 limited series on the use of FDs (mostly Pipeline device) for the
management of MCAAs12,14,33 constitute the noteworthy litera-ture on flow diversion specifically in MCAAs. Cases of MCAAs have been reported within larger series of intracranial aneurysms treated by FDs11,34-37; however, these latter reports also include other bifurcation aneurysms such as anterior communicating ar-tery aneurysms, which frequently rely on the circle of Willis rather than leptomeningeal collateral flow if flow in the jailed side branch is compromised. As a result, these reports are not able to address the role of FDs specifically in MCAAs.
The MCA is one of the most common sites of bifurcation an-eurysms,18and any alteration of the current treatment strategy will affect a significant number of patients who await treatment. We
Table 2. Outcomes After Flow Diversion
Aneurysm
Immediate Angiographic Results
Latest Follow-Up,
Months Modality Type
Results, Byrne Axis 1/2
Status Cortical Branch/Perforator Originating from the Aneurysm
1 No change 10 Angiography Saccular 4/a Occluded
2 Stagnation 16 Angiography Saccular 4/a Patent
3 Occlusion 24 Angiography Saccular 3/a Patent
4 No change 18 Angiography Saccular 4/a Patent
5 No change 16 Angiography Saccular 3/a Patent
6 Stagnation 6 Angiography Saccular 4/a Occluded
7 Stagnation 10 Angiography F/D 4/a Occluded
8 Stagnation 24 Angiography F/D 4/a None
9 Stagnation 18 Angiography Saccular 2/a Patent
10 Stagnation 6 Angiography Saccular 1/a Patent
11 Stagnation 2 days Angiography Saccular 4/a Occluded
12 Stagnation 3 Angiography F/D 3/a None
13 Stagnation 18 CTA F/D 4/a None
14 Stagnation 1.5 Angiography F/D 4/a None
15 Stagnation 33 Angiography F/D 4/a None
16 Stagnation 6 Angiography F/D 4/a None
17 No change 5 Angiography Saccular 0/a Patent
18 Stagnation 7 Angiography F/D 4/c None
19 Stagnation 18 Angiography Saccular 3/a Patent
20 Stagnation 6 Angiography F/D 4/b None
21 Stagnation 5 days MRA F/D 1/a None
22 Stagnation 7 MRA Saccular 4/a Occluded
23 No change 4.5 Angiography F/D 4/a None
24 No change 10 Angiography Saccular 1/a Patent
25 Unsuccessful NA NA Saccular NA NA
26 No change 3 MRA Saccular 3/a Patent
27 No change 6 Angiography F/D 4/a None
28 Stagnation 7 days MRA Saccular 1/a Patent
29 No change Pending Pending Saccular Pending Pending
believe MCA territory should be evaluated as a separate entity when it comes to aneurysm treatment with flow diversion for the following reasons: The appeal of completing treatment of a MCAA by simply placing an FD in an easier-to-access side branch may at times be tempting enough for the interventionalist to overlook conventional surgical33 and endovascular treatment options. The FD approach may turn a time-consuming endo-saccular or surgical treatment into a far shorter and less-complex intervention, and the cost of this “simplification” is negligible based on the current literature. In our series, this cost corre-sponded to a lower rate of treatment efficacy at midterm. More-over, data in the literature regarding the fate of side branches jailed by FDs are scarce, but there is a suggestion that arterial occlusion or stenosis does occur and the clinical course is not inevitably benign, especially in the chronic phase.11 Our results show that branch occlusion or even total MCA occlusion is possible inflow diversion for MCAAs. Lastly, the distal collateral supply is only leptomeningeal in MCA territory, and clinical tolerance to occlusion of an MCA trunk is not guaranteed. This finding is in contrast to aneurysms around the circle of Willis,
such as ICA and anterior communicating artery aneurysms. In this series, it was shown that obliteration of the MCAA after flow diversion may entail arterial branch obliteration in a collateral-deficient territory. Conversely, branch patency is related to residual aneurysm.
We propose that “variant bifurcation” aneurysms are better fitted for flow diversion. The odds for patency of a jailed side branch should increase with the size of that branch. Because patency parallels the presence of residual aneurysm after treat-ment with flow diversion, the chances for persistence of true bifurcation aneurysms with the larger side-branches (MCA trunks in“true bifurcation aneurysms”) may be greater than “variant bi-furcations.” In our series, only half of the true bifurcation aneu-rysms were totally occluded on follow-up. In contrast, the smaller side branch of a“variant bifurcation” will be advantageous for FD. Besides gradual occlusion, which is known to occur after im-plantation of an FD,38 will be better tolerated in this subtype (because of the less-eloquent supply area). This relation between side-branch patency and residual aneurysm is not evident in the article by Saleme et al.11One possible reason for this is the lower
Figure 3. Case 3. Images obtained in a 42-year-old woman with a nonruptured right-sided middle cerebral artery (MCA) variant bifurcation aneurysm at the origin of the right anterior temporal artery. (A) Digital subtraction angiogram and (B) 3-dimensional digital subtraction angiogram showing the right-sided saccular MCA variant bifurcation aneurysm. (C) Digital subtraction angiogram obtained at the 6-month follow-up shows a minimally filling aneurysm and parent artery stenosis. (D) Digital subtracted angiogram obtained at the 24-month follow-up reveals slightly increased interval filling of the aneurysm whereas, parent artery stenosis decreased. A cortical branch originating from the aneurysm is filling on both follow-up studies.
rate of residual aneurysms in their series. We noted, however, that in the subset of MCAAs in the article by Saleme et al., half of the aneurysms were either coiled during the same session or treated (most likely coiled) previously, possibly leading to relatively greater rates of occluded aneurysms.
Some pitfalls remain concerning flow diversion in MCAAs. Perforator injury has been a concern in the treatment of MCAAs. Currently, there are insufficient data supporting the widespread use of FDs for perforator aneurysms located in perforator-rich vascular territories.39 We tried to avoid the use of overlapping FDs in MCAAs to avoid perforator injury. It should be kept in mind that perforators also may arise from early cortical branches.40 When a side branch is prone to occlusion because of flow competition,11,35 the perforator arising from this branch may not be able to provide enough“sump effect” to remain pat-ent. This may be why frequent, small, asymptomatic infarcts have been reported after flow diversion in the lateral lenticulostriate territory in recent reports.34
Another concern with reliance on FDs for bifurcation aneurysms is the limited options for bailout in case of failure. In addition to one of our patients, enlargement of residual aneurysms in the absence of angiographic remnants has been documented after flow diversion.41
The pathologic process involved in this is not well understood and residual aneurysm after flow diversion (arbitrarily called “remodeling” without a clear morphologic definition) is not necessarily benign.33,41-43Furthermore, there is in vitro and in vivo experimental evidence that FDs are more prone to fail at bifurcations even with the latest generation devices.7,8If flow diversion fails, further treatment will be limited to either bypass surgery and sacrifice of the involved branch42
or placement of additional FDs, because endovascular access to the jailed branch will have been lost indefinitely. Any further surgeries will be more complicated in such failures because of the need to adjust antiplatelet medications. Again, if surgical sacrifice of the jailed artery is needed, it would be less risky to sacrifice a variant branch rather than an MCA trunk.
Figure 4. Case 2. Images obtained in a 52-year-old man with a nonruptured right-sided middle cerebral artery (MCA) variant bifurcation aneurysm. (A) Three-dimensional digital subtracted angiogram showing the right-sided saccular MCA variant bifurcation aneurysm giving rise to a frontal branch. (B) Immediate posttreatment angiogram showing a flow diverter covering the neck of the aneurysm. (C) Digital subtraction angiograms obtained at the 6-month follow-up and (D) at the 18-month follow-up showing gradual total occlusion of the aneurysm. The side-branch is smaller and its origin appears triangular compared with its preprocedure diameter, suggesting interval “remodeling” at the origin of the aneurysm.
Because of these concerns, our strategy is to useflow diversion in treating MCAAs primarily 1) when surgical or other endovas-cular approaches are not favored; 2) for lesions that remain after previous surgery or endosaccular treatment (e.g. coils or WEB Aneurysm Embolization System; Sequent Medical, Inc., Aliso Viejo, California, USA),13,44and 3) for MCA-BV aneurysms where delayed spontaneous or surgical occlusion of the side branch is expected to be well tolerated. In the latter situation, we feel that the shortcomings offlow diversion are mitigated.
Our study has several limitations. First of all, a larger number of patients are needed to reachfirm conclusions about an association between the status of aneurysm and the status of the arterial branch. The relatively small number of patients in this series is attributable to both our strict exclusion criteria for treating bifurcation aneu-rysms with flow diversion and the paucity of dissecting-type MCAAs. It remains to be determined whether ourfindings can be validated in larger prospective series. Second, our follow-up period was relatively short. Occlusion rates withflow diversion are expected to increase with longer follow-up and cessation of antiplatelet agents. It needs to be determined whether total aneurysm occlusion occurs during long-term follow-up, whether side branches are still patent at that time, and whether patients will remain asymptomatic during this process. However, despite these limitations, we argue that the current study has important implications regarding selec-tion of MCAAs that are suitable for flow diversion outside the context of a monitored a clinical study.
SUMMARY/CONCLUSIONS
On the basis of our data, we suggest that flow diversion is an indispensable tool for the management of fusiform and dis-secting MCAAs; however, pending the availability of data from monitored studies, FDs should not be considered as the pri-mary treatment for MCA-B aneurysms. They may be used potentially for aneurysms at the origin of an “anomalous” cortical branch, a subset that comprises a significant proportion of all MCAAs.
Our limited series has demonstrated a significant direct relation between the patency of the aneurysm and patency of the perfo-rator/cortical branches that originate from it. This may be a limiting factor whenflow diversion is planned for the treatment of bifurcation aneurysms distal to the circle of Willis, including MCAAs. Larger series with longer angiographic follow-up are needed to reachfirm conclusions.
ACKNOWLEDGMENTS
We thank Shifra Fraifeld for her meticulous assistance. How-ever, the full content of the manuscript, including all data, procedure details, outcomes, and conclusions, remained the sole responsibility of the authors throughout the writing and editorial process. The authors disclose no other conflicts of interest.
REFERENCES
1. Fang S, Lanzino G. Paraclinoid aneurysms: is there a new endovascular standard? Neurol Res. 2014;36:314-322.
2. Akmangit I, Aydin K, Sencer S, Topcuoglu OM, Topcuoglu ED, Daglioglu E, et al. Dual stenting using low-profile LEO Baby stents for the endo-vascular management of challengingıntracranial aneurysms. Am J Neuroradiol. 2015;36:323-329. 3. Kim YJ, Ko JH. Sole stenting with large cell stents
for very small ruptured ıntracranial aneurysms. Interv Neuroradiol. 2014;20:45-53.
4. Becske T, Kallmes DF, Saatci I, McDougall CG, Szikora I, Lanzino G, et al. Pipeline for uncoilable or failed aneurysms: results from a multicenter clinical trial. Radiology. 2013;267:858-868. 5.Nelson PK, Lylyk P, Szikora I, Wetzel SG,
Wanke I, Fiorella D. The pipeline embolization device for the intracranial treatment of aneurysms trial. AJNR Am J Neuroradiol. 2011;32:34-40. 6. Huttunen T, von und zu Fraunberg M, Frosen J,
Lehecka M, Tromp G, Helin K, et al. Saccular intracranial aneurysm disease: distribution of site, size, and age suggests different etiologies for aneurysm formation and rupture in 316 familial and 1454 sporadic eastern Finnish patients. Neurosurgery. 2010;66:631-638 [discussion: 638]. 7. Darsaut TE, Bing F, Makoyeva A, Gevry G,
Salazkin I, Raymond J. Flow Diversion of giant curved sidewall and bifurcation experimental an-eurysms with very-low-porosity devices. World Neurosurg. 2014;82:1120-1126.
8. Darsaut TE, Bing F, Salazkin I, Gevry G, Raymond J. Flow diverters failing to occlude experimental bifurcation or curved sidewall an-eurysms: an in vivo study in canines. J Neurosurg. 2012;117:37-44.
9. Raymond J, Darsaut TE, Makoyeva A, Bing F, Salazkin I. Endovascular treatment with flow diverters may fail to occlude experimental bifur-cation aneurysms. Neuroradiology. 2013;55: 1355-1363.
10. Puffer RC, Kallmes DF, Cloft HJ, Lanzino G. Patency of the ophthalmic artery afterflow diver-sion treatment of paraclinoid aneurysms. J Neurosurg. 2012;116:892-896.
11. Saleme S, Iosif C, Ponomarjova S, Mendes G, Camilleri Y, Caire F, et al. Flow-diverting stents for intracranial bifurcation aneurysm treatment. Neurosurgery. 2014;75:623-631. quiz 631.
12. Briganti F, Delehaye L, Leone G, Sicignano C, Buono G, Marseglia M, et al. Flow diverter device for the treatment of small middle cerebral artery aneurysms. J Neurointerv Surg., 2015. pii: neu-rintsurg-2014-011460. http://dx.doi.org/10.1136/ neurintsurg-2014-011460.
13. Lubicz B, Pezzullo M, Brisbois D, De Witte O, Mine B. Endovascular treatment of proximal su-perior middle cerebral artery aneurysms. Neurora-diology. 2012;54:1267-1273.
14. Zanaty M, Chalouhi N, Tjoumakaris SI, Gonzalez LF, Rosenwasser R, Jabbour P. Flow diversion for complex middle cerebral artery an-eurysms. Neuroradiology. 2014;56:381-387.
15. Darsaut TE, Bing F, Salazkin I, Gevry G, Raymond J. Testingflow diverters in giant fusi-form aneurysms: a new experimental model can show leaks responsible for failures. Am J Neuro-radiol. 2011;32:2175-2179.
16. Zaidat OO, Castonguay AC, Teleb MS, Asif K, Gheith A, Southwood C, et al. Middle cerebral artery aneurysm endovascular and surgical therapies: comprehensive literature review and local experience. Neurosurg Clin N Am. 2014;25: 455-469.
17. Elsharkawy A, Lehecka M, Niemela M, Billon-Grand R, Lehto H, Kivisaari R, et al. A new, more accurate classification of middle cerebral artery aneurysms: computed tomography angiographic study of 1,009 consecutive cases with 1,309 middle cerebral artery aneurysms. Neurosurgery. 2013;73: 94-102 [discussion: 102].
18. Dashti R, Hernesniemi J, Niemelä M, Rinne J, Porras M, Lehecka M, et al. Micro-neurosurgical management of middle cerebral artery bifurcation aneurysms. Surg Neurol. 2007; 67:441-456.
19. Dashti R, Rinne J, Hernesniemi J, Niemelä M, Kivipelto L, Lehecka M, et al. Micro-neurosurgical management of proximal middle cerebral artery aneurysms. Surg Neurol. 2007;68: 6-14.
20. Ulm AJ, Fautheree GL, Tanriover N, Russo A, Albanese E, Rhoton AL Jr, et al. Microsurgical and angiographic anatomy of middle cerebral artery aneurysms: prevalence and significance of early branch aneurysms. Neurosurgery. 2008;62(5 suppl
2):ONS344-ONS352 [discussion: ONS352-ONS353].
21. Kamran M, Yarnold J, Grunwald IQ, Byrne JV. Assessment of angiographic outcomes afterflow diversion treatment of intracranial aneurysms: a new grading schema. Neuroradiology. 2011;53: 501-508.
22. Chyatte D, Porterfield R. Nuances of middle cerebral artery aneurysm microsurgery. Neurosur-gery. 2001;48:339-346.
23. Heros RC, Fritsch MJ. Surgical management of middle cerebral artery aneurysms. Neurosurgery. 2001;48:780-786.
24. Bracard S, Barbier C, Derelle AL, Anxionnat R. Endovascular treatment of aneurisms: pre, intra and post operative management. Eur J Radiol. 2013; 82:1633-1637.
25. Diaz OM, Rangel-Castilla L, Barber S, Mayo RC, Klucznik R, Zhang YJ. Middle cerebral artery an-eurysms: a single-center series comparing endo-vascular and surgical treatment. World Neurosurg. 2014;81:322-329.
26. Eboli P, Ryan RW, Alexander JE, Alexander MJ. Evolving role of endovascular treatment for MCA bifurcation aneurysms: case series of 184 aneu-rysms and review of the literature. Neurol Res. 2014; 36:332-338.
27. Doerfler A, Wanke I, Goericke SL, Wiedemayer H,
Engelhorn T, Gizewski ER, et al. Endovascular treatment of middle cerebral artery aneurysms with electrolytically detachable coils. AJNR Am J Neuroradiol. 2006;27:513-520.
28. Iijima A, Piotin M, Mounayer C, Spelle L, Weill A, Moret J. Endovascular treatment with coils of 149 middle cerebral artery berry aneurysms. Radiology. 2005;237:611-619.
29. Quadros RS, Gallas S, Noudel R, Rousseaux P, Pierot L. Endovascular treatment of middle cere-bral artery aneurysms as first option: a single center experience of 92 aneurysms. Am J Neuro-radiol. 2007;28:1567-1572.
30. Rodriguez-Hernandez A, Sughrue ME, Akhavan S, Habdank-Kolaczkowski J, Lawton MT. Current management of middle cerebral artery aneurysms: surgical results with a“clip first” policy. Neuro-surgery. 2013;72:415-427.
31. Gory B, Rouchaud A, Saleme S, Dalmay F, Riva R, Caire F, et al. Endovascular treatment of middle cerebral artery aneurysms for 120 nonselected pa-tients: a prospective cohort study. AJNR Am J Neuroradiol. 2014;35:715-720.
32. Johnson AK, Heiferman DM, Lopes DK. Stent-assisted embolization of 100 middle cerebral ar-tery aneurysms. J Neurosurg. 2013;118:950-955.
33. Yavuz K, Geyik S, Saatci I, Cekirge HS. Endovas-cular treatment of middle cerebral artery aneu-rysms withflow modification with the use of the pipeline embolization device. AJNR Am J Neuro-radiol. 2014;35:529-535.
34. Gawlitza M, Januel AC, Tall P, Bonneville F, Cognard C. Flow diversion treatment of complex bifurcation aneurysms beyond the circle of Willis: a single-center series with special emphasis on covered cortical branches and perforating arteries. J Neurointerv Surg. 2015 [Epub ahead of print].
35. Iosif C, Camilleri Y, Saleme S, Caire F, Yardin C, Ponomarjova S, et al. Diffusion-weighted imaging-detected ischemic lesions associated with flow-diverting stents in intracranial aneurysms: safety, potential mechanisms, clinical outcome, and con-cerns. J Neurosurg. 2015;122:627-636.
36. Lubicz B, Collignon L, Raphaeli G, Pruvo JP, Bruneau M, De Witte O, et al. Flow-diverter stent for the endovascular treatment of intracranial an-eurysms: a prospective study in 29 patients with 34 aneurysms. Stroke. 2010;41:2247-2253.
37. Pistocchi S, Blanc R, Bartolini B, Piotin M. Flow diverters at and beyond the level of the circle of willis for the treatment ofıntracranial aneurysms. Stroke. 2012;43:1032-1038.
38. Wajnberg E, Silva TS, Johnson AK, Lopes DK. Progressive deconstruction: a novel aneurysm treatment using the pipeline embolization device
for competitiveflow diversion. Neurosurgery. 2014; 10(suppl 1):E161-E166 [discussion: E166]. 39. Vargas J, Walsh K, Turner R, Chaudry I, Turk A,
Spiotta A. Lenticulostriate aneurysms: a case se-ries and review of the literature. J Neurointerv Surg. 2015;7:194-201.
40. Sen T, Esmer AF, Acar HI, Karahan ST, Tuccar E. Arterial vascularisation of the anterior perforated substance. Singapore Med J. 2011;52:410-414. 41. Hampton T, Walsh D, Tolias C, Fiorella D. Mural
destabilization after aneurysm treatment with a flow-diverting device: a report of two cases. J Neurointerv Surg. 2011;3:167-171.
42. Clarençon F, Nouet A, Redondo A, Di Maria F, Iosif C, Le Jean L, et al. Occlusion of M1 segment after superficial temporal artery-middle cerebral artery bypass in a giant M1 aneurysm with Onyx-34 injected via a double-lumen balloon under balloon inflation. BMJ Case Rep. 2013;2013:1-4. 43.Raymond J, Darsaut TE, Kotowski M,
Makoyeva A, Gevry G, Berthelet F, et al. Throm-bosis heralding aneurysmal rupture: an explora-tion of potential mechanisms in a novel giant swine aneurysm model. Am J Neuroradiol. 2013;34: 346-353.
44. Cho YD, Lee WJ, Kim KM, Kang HS, Kim JE, Han MH. Endovascular coil embolization of middle cerebral artery aneurysms of the proximal (M1) segment. Neuroradiology. 2013;55:1097-1102.
Conflict of interest statement: The author declares that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Received 26 August 2015; accepted 24 November 2015 Citation: World Neurosurg. (2016) 87:317-327.
http://dx.doi.org/10.1016/j.wneu.2015.11.073
Journal homepage:www.WORLDNEUROSURGERY.org
Available online:www.sciencedirect.com
1878-8750/$ - see front matterª 2016 Elsevier Inc. All rights reserved.
3251 Riverport Lane, Maryland Heights, MO 63043 Tel: (314) 447-8871 (International)
Tel: (800) 654-2452 (US)
Fax: (314) 447-8029 (US and International) Email: journalscustomerservice-usa@elsevier.com
ELSEVIER
