Background Y-stent assisted coiling has been proposed for the treatment of wide-neck bifurcation aneurysms, but there are many technical variations. We report our single-center experience of Y-stent assisted coiling of bifurcation aneurysms with the closed cell Enterprise stent in order to evaluate the safety and long-term results of this technique. The literature on Y-stenting and its hemodynamic effects are reviewed.
Methods Fifty-two consecutive patients with wide-neck bifurcation aneurysms underwent Y-stent assisted coiling with two Enterprise stents. The procedure was completed in 48 cases (92.3%) and technical failure occurred in 4 cases (7.3%). Cases performed with other stents were excluded. All procedures were performed under double antiplatelet therapy. Periodic clinical and neuroradiological follow-up was performed.
Results Mean neuroradiological follow-up time was 26 months. Complete immediate occlusion was obtained in 87.5% of patients. Two remnants had regrown at follow-up and were recoiled, achieving complete occlusion. The late neuroradiological occlusion rate was: complete occlusion 93.6%, neck remnant 4.3%, sac remnant 2.1%. No in-stent stenosis was detected at follow-up. Among the 48 procedures, two complications occurred (4.2%). Mortality was 2.1%. No delayed ischemic stroke occurred.
Conclusions Y-stent assisted coiling has a high immediate occlusion rate and very good long-term stability. The procedure is relatively safe, although the complication and mortality rates are not negligible. Two Enterprise stents can be safely used for Y-stenting and, indeed, offer the advantage of easier catheterization, delivery and deployment into distal and tortuous vessels than open cell stents.
Statistics from Altmetric.com
Stent-assisted coiling has gained broad acceptance as a safe and effective approach for wide-neck aneurysms. This technique is commonly used also for bifurcation aneurysms, although it has some limitations in cases of unfavorable anatomical configuration or very wide neck.1 Balloon remodeling can help in the management of wide-neck aneurysms with an acceptable complication rate, but it may not be suitable in cases of very complex morphology. Flow diverter stents are occasionally used to treat bifurcation aneurysms, but the long-term results are suboptimal because the aneurysm may remain patent and the fate of the artery covered by the stent is unpredictable.2 Intrasaccular flow disruption has been described, but more extensive clinical experience and follow-up are needed to assess the efficacy and safety of this approach.3
A useful technique is Y-stent assisted coiling, also known as Y-stenting, first proposed by Chow et al in 2004.4 Y-stenting is performed by placing two stents in the parent artery with each stent in one of the bifurcation branches, creating a new bifurcation point across the neck of the aneurysm. Other reports and small series have confirmed the feasibility of the procedure, usually reporting the use of open cell stents or a combination of open and closed cell stents,5–14 but variations to the deployment technique have been proposed.15 The use of two closed cell stents for Y-stenting has been reported16 and has been shown to be safe.17–20 The main concern regarding the use of two closed cell stents for Y-stenting is that the closed cell design may limit the expansion of the second stent at the crossing point, with possible ischemic events. However, closed cell stents may have some technical advantages over open cell stents for Y-stenting, and we believe that the closed cell Enterprise stent (Codman Neurovascular, Raynham, Massachusetts, USA) is well suited for this technique because of its easy deployment and its technical characteristics.
We report our single-center experience of Y-stent assisted coiling of bifurcation aneurysms performed with the Enterprise stent with clinical and neuroradiological follow-up, in order to evaluate the safety and long-term results of this technique.
Materials and methods
We retrospectively reviewed all consecutive patients scheduled for stent-assisted coiling with placement of closed cell stents in a ‘Y’ configuration from April 2010 to March 2013 at our institution. Among 52 patients who underwent attempted Y-stent assisted coiling, the procedure was completed in 48. In the remaining four cases, technical failure occurred and the aneurysm was treated by a different technique. All aneurysms were wide-necked (≥4 mm). The asymmetric origin of one bifurcation branch over the neck plane or from the basal aspect of the aneurysm was not considered as a contraindication. We excluded all cases performed with open cell stents, as well as ‘X-stenting’ or ‘parallel-stenting’ procedures.
The demographic data of the 48 patients who underwent Y-stent assisted coiling are summarized in table 1. The aneurysms were unruptured in all except five patients who underwent elective treatment for recurrence of previously coiled ruptured aneurysms.
Treated aneurysms were located at the middle cerebral artery (MCA) (20 cases), anterior communicating artery (AComA) (14 cases), basilar tip (11 cases), carotid tip (2 cases), and vertebrobasilar junction (1 case). Twelve aneurysms were <7 mm, 33 were 7–15 mm, 3 were >15 mm. The mean aneurysm diameter was 10 mm and the mean neck diameter was 5.2 mm. In all cases the neck was wide with a dome to neck ratio below 1.5. In 68.7% of cases the base of the aneurysm was asymmetric. In 33.3% of cases one bifurcation branch originated from the basal aspect of the aneurysm.
All interventions were performed under double antiplatelet therapy with acetylsalicylic acid 300 mg per day and clopidogrel 75 mg per day or ticlopidine 250 mg twice daily, starting 7–10 days before the intervention. Response to antiplatelet therapy was assessed by light transmittance aggregometry. Double antiplatelet therapy was continued for 6 months, then single therapy with acetylsalicylic acid 100/150 mg per day was continued life-long. Patients were also systemically heparinized to achieve an activated clotting time of 250–300 s at the time of stenting. All procedures were performed under general anesthesia using a biplane angiography unit with three-dimensional rotational capability (Allura; Philips, Best, The Netherlands). From the femoral access, a 6 Fr 90 cm introducer sheet (Arrow; Teleflex, Limerick, Pennsylvania, USA) was navigated into the internal or common carotid artery and then a 6 Fr guiding catheter was coaxially positioned into the internal carotid artery. In the case of posterior circulation aneurysms, the long sheet was advanced into the subclavian artery and the guide catheter was navigated into the vertebral artery. We planned to place the first stent in the more challenging bifurcation branch in case the technique had to be changed due to catheterization failure. The chosen bifurcation branch was catheterized with a Prowler Select Plus microcatheter (Codman Neurovascular) over various guidewires, usually the hydrophilic 0.016 inch glidewire (Terumo, Tokyo, Japan). The aneurysm was catheterized with a microcatheter (Echelon 10; Covidien, Irvine, California, USA) over a 0.012 inch hydrophilic glidewire (Terumo) before stenting. The stent was deployed, partially covering the neck of the aneurysm, and the other bifurcation branch was catheterized crossing the interstices of the stent. In most cases the second stent was deliberately shorter than the first one because the presence of the first stent made its deployment more stable and easier. Various lengths of the Enterprise stent were used (ie, 22, 28, or 37 mm). After stenting the aneurysm was coiled with a variety of bare platinum coils, usually using the jailing technique (figure 1). The patients were discharged on double antiplatelet therapy for 6 months and life-long acetylsalicylic acid.
Neurologic examination and modified Rankin Scale (mRS) measurement were performed in all patients at baseline, immediately after the procedure, at discharge, and at 3 months follow-up. The clinical follow-up was continued on a yearly basis. Primary adverse events included death and stroke and secondary adverse events recorded were transient ischemic attack and the need for re-intervention. Morbidity was considered as mRS >0 at 3 months.
Imaging follow-up and evaluation
The patients who completed Y-stent assisted coiling were analyzed to assess the immediate and long-term clinical and angiographic results of this technique. The imaging follow-up protocol was angiography after 6 and 24 months and MR angiography (MRA) after 1 and 3 years. Aneurysm occlusion was estimated using the Raymond–Roy (RR) scale (RR1, complete obliteration; RR2, neck remnant; RR3, contrast filling of the aneurysm) immediately after the intervention and at follow-up. The presence of hemodynamically significant intracranial stenosis (>50%) was reported. The RR scale was retrospectively assessed by a neuroradiologist not involved in the procedures.
Among the 52 patients, the planned Y-stent assisted coiling was completed in 48 cases (92.3%). Due to operative difficulties, in four patients (7.7%) the Y-stenting was not completed as planned and coiling was performed with a different technique. In two cases, failure was due to the fact that catheterization of the most angulated branch of the bifurcation was not possible while the other two failures were due to difficult catheterization of the second bifurcation branch through the first stent. In all cases embolization was performed with an alternative technique: in one case with single stent-assisted coiling, in one case with balloon remodeling, and in two cases with ‘hybrid Y-stenting with waffle-cone’ technique.21
Clinical outcome and complications
Two procedural complications occurred during Y-stent assisted coiling (4.2%), both due to aneurysm perforation. One of these patients had perforation during catheterization and the procedure was completed; an external ventricular drain was positioned, with complete recovery after 1 month. However, 1 month later she died of an intraventricular hemorrhage secondary to removal of the drain which was infected (overall mortality 2.1%). The other patient had subarachnoid hemorrhage and an intraparenchymal hemorrhage due to aneurysm perforation during placement of the first coil. In this case the first stent bulged inside the aneurysm during catheterization of the second branch and it tightly jailed the microcatheter against the aneurysmal wall so the coil had insufficient space to deploy and perforated the sac. However, the patient had good clinical recovery with residual lateral hemianopsia at follow-up (mRS 1; overall morbidity 2.1%).
Mean clinical follow-up was 28 months. No delayed ischemic strokes occurred; one patient experienced transitory ischemic attacks after discontinuation of ticlopidine so dipyridamole 200 mg plus aspirin 100 mg therapy was started and no other ischemic events were reported over an 18-month follow-up period. No aneurysm ruptures and no intracranial or extracranial hemorrhages occurred during follow-up.
The neuroradiological results are summarized in table 2. The immediate angiographic occlusion rate after Y-stent assisted coiling was RR1 in 42 cases (87.5%), RR2 in 3 cases (6.25%; 2 MCA aneurysms, 1 AComA aneurysm), and RR3 in 3 cases (6.25%; 2 AComA and 1 basilar tip aneurysms). All the patients who survived (n=47) had at least one angiographic follow-up and 33 patients underwent a second angiographic examination after ≥2 years. Thirty-seven patients underwent MRA at 1-year follow-up and 12 underwent MRA at the 3-year follow-up. The mean neuroradiological follow-up time (ie, including angiography and MRA) was 26 months.
At the 6-month follow-up the angiographic occlusion rate was RR1 in 42 cases (89.4%), RR2 in 2 cases (4.2%), and RR3 in 3 cases (6.4%). One initial neck remnant was found occluded and two others were unchanged. One sac remnant of an AComA aneurysm was stable while the other two remnants (one mid-sized AComA and one large basilar tip aneurysm) showed an increase in size. These two remnants were coiled, obtaining complete occlusion, although one of them recurred again and required coiling.
Angiographic follow-up at 2 years was available for 33 patients (including the two retreated patients); the occlusion rate was RR1 in 31 cases (93.9%), RR2 in 1 case (3.05%), and RR3 in 1 case (3.05%).
Overall, the late neuroradiological occlusion rate (ie, including the last angiography or MRA examination) was RR1 in 44 cases (93.6%), RR2 in 2 cases (4.3%), and RR3 in 1 case (2.1%). No recurrences or modifications from previous angiographic follow-up were shown by intermediate or late MRA. No in-stent stenosis was detected during follow-up.
Immediate occlusion rates
The immediate occlusion rate of aneurysms treated by Y-stent assisted coiling reported in the literature is quite high but very variable, ranging from 25% to 90%.9 ,13 ,14 ,18–20 It should be considered that most of these aneurysms are of a complex shape and are challenging for both endovascular and open surgery. In our series the immediate occlusion rate was 87.5%, which is as high as other endovascular techniques and higher than the results of most Y-stenting series.
In our series the stability of occlusion during follow-up was very high, with a late complete occlusion rate of 93.6%, rising to 97.9% if both RR1 and RR2 class aneurysms were included. Our neuroradiological follow-up time is longer than all other published series and our long-term results confirm the effectiveness of Y-stent assisted coiling with closed cell stents. In the literature, taking into account only series with more than 10 patients, the long-term occlusion rate after Y-stent assisted coiling ranges from 63.2% to 97.8%.9 ,13 ,14 ,18–20 Interestingly, the immediate occlusion rate reported by the largest series (193 aneurysms of which 183 were treated by Y-stenting, mostly with closed cell devices) is considerably lower than ours with 25.4% RR1 occlusions, but at 6 months the occlusion rate had risen to 97.8%.18 We believe that this difference is mainly due to the fact that, in that series, the interventionalists did not try to fill all the aneurysms tightly, relying on the flow-diverting effect of stents. In a multicenter study the late occlusion rate was 92% (mean follow-up 9.8 months),19 while in another large series (including few ‘X-stenting’ procedures) the late occlusion rate was 85.8%, which was much higher than the immediate occlusion rate (47.6%).14 A very recent series of Y-stent assisted coiling with open cell devices reported 80% RR1 occlusion and 20% RR2 occlusion with no sac remnant at late follow-up.13 Overall, the long-term occlusion rate appears good in all series, regardless of the kind of stent used.
The Y-shaped stent construct is supposed to have a flow-diverting effect, as shown in a small series of aneurysms occluded after Y-stenting with the Enterprise stent without additional coiling.22 The hemodynamic effect of intracranial stents depends on the disruption and turbulence of intra-aneurysmal flow, with an increase in blood stasis inside the sac. The high hemodynamic effect of Y-stenting has been confirmed by a few studies. In an in vitro model with open cell stents, the vorticity and shear stress inside the sac were decreased by more than 40% at the end of the cardiac cycle.23 Computational fluid dynamic analysis after open cell Y-stenting showed that the angular configuration remodeling immediately redirects wall shear stress from the neck transition zone towards the inert coil mass.24 A comparison between single, Y and crossing-bar stenting, based on particle image velocimetry, showed that the Y-stent configuration achieved flow redirection away from the center of the aneurysmal neck, with a more significant reduction in the cross-neck flow.25 Kono et al performed computational fluid dynamic simulations of different configurations of stent placement, including kissing and crossing Y-stenting with Enterprise. Kissing and crossing Y-stents achieve a greater reduction in intra-aneurysmal flow velocity than single stents because the linear profile of the two stents leads to better flow separation and redirection at the neck level. Moreover, the cell area of the Enterprise stent lowers at the crossing point, perhaps contributing to the hemodynamic effect. Another unexpected conclusion is that the narrowing effect of closed cell Y-stenting could be a hemodynamic advantage since it increases flow modifications, with a potential increase in the long-term stability of the occlusion.26
In our opinion the Enterprise stent is the first choice for this kind of procedure because of technical differences between Y-stenting performed with open or closed cell stents. First, laser-cut closed cell devices have a lower profile, are more navigable, and require smaller delivery catheters than open cell devices. Moreover, closed cell stents are usually easily deployed since they can be partially or fully recapturable. In curved vessels, laser-cut closed cell stents tend to keep their linear profile, even if kinking and flattening at the mid-section can occur in very acute bends. On the other hand, open cell stents in curved vessels undergo continuous outward prolapse of the struts at the convexity towards the aneurysm neck and inward prolapse of the struts into the lumen of the stent at the concavity. The linear profile of two closed cell stents may give a better scaffold to protect the parent artery during coiling, leading to a more anatomical reconstruction of the vessels. Sometimes, kinking of the Enterprise in a very angulated bifurcation can make catheterization difficult or impossible. An in vitro study showed that the passage of a large catheter (0.025 inch lumen) through a Y-stent construct is possible only if open cell stents were used because the cell area is smaller in the case of two crossing Enterprise stents. However, passage of a standard coiling microcatheter was possible with all cell configurations.27
A disadvantage of using two closed cell stents is that the first stent may limit the expansion of the second one, resulting in stent stenosis at the intersection point.28 This may be an important concern, however the long-term results of Y-stenting with closed cell devices confirm that this mechanical aspect does not have a significant clinical impact. Interestingly, the use of new low profile braided stents has been reported but no late data are available.29
The main concern of Y-stenting is the risk of thromboembolic events. The rate of acute and late ischemic complications is low in our series and others, with intraprocedural thromboembolic events ranging from 0% to 16%.9 ,13 ,14 ,18–20 We did not experience any acute in-stent thrombosis and the only transitory ischemia occurred after early interruption of antiplatelet treatment.
In the largest series of Y-stenting with closed cell stents, a global complication rate of 2.7% and a mortality rate of 0.5% were reported.14 In some papers of Y-stenting with two open cell stents, the ischemic complication rate ranged from 0% to 31.6%.7 ,9 ,11 ,13 However, no direct comparison of the safety of open and closed cell Y-stenting has been reported. The development of intracranial stenosis was also unlikely in our series and others. In our experience the most complex step of the technique was the crossing of the interstices of the first stent, which was the most common reason for technical failure and, in one case, it led to a hemorrhagic complication.
Limitations of the study
The main limitations of this study are the relatively low number of patients and the retrospective design. The prevalence of anterior circulation aneurysms is higher in our series than in others,9 ,13 ,19 although it is similar to that reported by the largest studies.14 ,18 It could be argued that this is a bias due to different selection criteria, however we do not believe this substantially limits the wider application of our findings. A prospective study with well-defined inclusion criteria could better establish the real advantages of this technique compared with surgery.
Y-stent assisted coiling has a high immediate occlusion rate and very good long-term stability. The procedure is relatively safe, especially if we take into account the fact that it is reserved for complex aneurysms. However, the rate of complications and the mortality rate are not negligible. Closed cell stents such as the Enterprise are well suited for this procedure without a significant risk of long-term ischemia. We believe that improving patient selection, excluding those with an unfavorable anatomy, will reduce the complication rate.
Competing interests Authors have no competing interests to disclose.
Patient consent Obtained.
Ethical approval No ethical committee approval was required for retrospective studies in the institution.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.