Background and purpose Assisted coiling with stents or balloons enables a higher percentage of complete occlusions of saccular unruptured intracranial aneurysms to be achieved with a reasonable complication rate. The aim of this study was to compare stent-assisted coiling and the balloon remodeling technique in terms of efficacy, stability, and safety for the treatment of comparable unruptured saccular intracranial aneurysms.
Materials and methods 268 patients with 286 saccular unruptured wide-necked intracranial aneurysms were treated at our institution with stent- or balloon-assisted coiling and retrospectively reviewed. Statistical analysis was performed to assess significant differences between the two groups.
Results The rate of complete occlusion at the end of the procedure was higher with stent-assisted coiling than with balloon-assisted coiling (86.8% vs 78%) and the same results were also observed after 6 months (92.1% vs 77.6%; p=0.05). About 50% of major recurrences occurred in large to giant aneurysms (p<0.001). The overall complication rate was similar in the stent-assisted and balloon-assisted groups (10.3% vs 9.3%). Independently of the technique, a higher complication rate was observed with bifurcational aneurysms, particularly in the middle cerebral artery (p=0.016).
Conclusions Stent-assisted coiling achieved better results in terms of complete occlusion and stability than balloon-assisted coiling with a lower rate of recurrence without being associated with a higher risk of intraprocedural complications. Bifurcational and large to giant aneurysms were associated with higher complication rates and higher recurrence rates, respectively, and still represent a challenge for both techniques.
Statistics from Altmetric.com
Endovascular treatment of unruptured intracranial aneurysms provides very good results in terms of angiography and clinical outcome. Since the introduction of detachable coils, the endovascular approach has undergone several technical improvements and the materials used have become more sophisticated so that interventionalists are able to treat even difficult aneurysms. Although the concept of difficulty remains reasonably subjective and is related to the experience of the operator, some characteristics of intracranial aneurysms may increase the risk of intraprocedural complications1 ,2 such as a wide neck (or dome-to-neck ratio), a larger size, the presence of blebs, and vascular tortuosity.3 Among these, the presence of a wide neck is the most relevant factor that may influence the planning of the endovascular approach. The need to contain the coils within the aneurysmal sac and to maintain the patency of the parent vessel(s) has led to the introduction of devices to assist the coiling procedure.4 The aim of this paper was to compare the effectiveness, safety, and stability of two conventional techniques used in the endovascular treatment of unruptured intracranial aneurysms.
Materials and methods
Two hundred and sixty-eight patients (188 women, 80 men; mean age 56.4±14.4 years, range 16–78) with 286 saccular wide-necked unruptured intracranial aneurysms were treated at our institution with stent-assisted coiling (122 aneurysms in 117 patients) or the balloon remodeling technique (RT, 164 aneurysms in 151 patients) from January 2004 to December 2012. In all cases the diagnosis was incidental and secondary to symptoms not necessarily related to the presence of the aneurysm(s) (headache or migraine, dizziness) or other reasons (familial history of subarachnoid hemorrhage). Patients with ruptured, dissecting, blister-like and/or fusiform intracranial aneurysms and those treated with other endovascular techniques (RT and stent combined, intra-arterial or intrasaccular flow diverters, parent vessel occlusion, simple coiling, telescopic stenting, Onyx or Glubran) were excluded from the analysis. All the cases included in the analysis were considered treatable with both techniques and those cases in which one of the two techniques could not have been used were excluded from the analysis. Very small aneurysms (<3 mm) were not included in the analysis since, from our experience, we considered this kind of aneurysm not to be eligible for endovascular treatment. Both single and kissing balloon techniques were used as well as single-stent or Y-stenting techniques. Wide neck was defined either as a neck >4 mm or a dome-to-neck aspect ratio <1.5, or both.
All the patients underwent complete angiographic assessment with a manual occlusion test where necessary (particularly for giant aneurysms to evaluate the presence of contralateral cross-filling, and for anterior communicating artery (ACoA) aneurysms to have clear visualization of the ACoA complex).5 The aneurysms were classified as small (<5 mm), medium (6–15 mm), large (16–25 mm), and giant (>25 mm) and, with regard to the localization, as lateral or bifurcational. The procedure used depended on the therapeutic strategy. For stent-assisted coiling, patients were premedicated 10 days before the procedure with aspirin 300 mg and clopidogrel 75 mg/day (or ticlopidine 250 mg twice daily) and in vitro aggregation tests were performed the day before the procedure to evaluate the residual platelet activity. Dual antiplatelet therapy was prescribed for 3 months (for single stent implantation) or 6 months (if more than one stent was used), after which clopidogrel (or ticlopidine) was stopped while aspirin was prescribed for a further year after the procedure (and not stopped in patients aged >50 years). For RT, patients received no premedication and no additional treatment was prescribed after discharge. The choice of the technique was not randomized and it was taken based on the evolution of the devices, mainly intracranial stents. In about 90% of the procedures we used closed-cell stents—mostly the Enterprise stent (Codman Neurovascular, Raynham, Massachusetts, USA) and in a few cases the Solitaire (Covidien, Irvine, California, USA). Open-cell stents (Neuroform stent; Stryker, Kalamazoo, Michigan, USA) were used in only a few cases at the beginning of our experience. All balloon-assisted coiling procedures were performed with Hyperglide or Hyperform balloons (Covidien). All the procedures were performed under general anesthesia and systemic heparinization. Activated clotting time (ACT) was assessed every 30 min and stents were deployed when ACT values were >250 s.
The angiographic result was assessed by an interventional neuroradiologist with 30 years of experience using the Raymond scale (grade I: complete occlusion; grade II: neck remnant; grade III: sac remnant). Intraprocedural, periprocedural, and late adverse events were registered and classified as asymptomatic or clinically symptomatic (with clinical sequelae), which were subdivided into ischemic (thromboembolism or dissection of the parent vessel or any branch visualized during angiographic controls and successively confirmed by CT scan or MRI) and hemorrhagic events (sac or arterial perforation or rupture, confirmed by CT scan or MRI). All patients underwent a CT scan at the end of the procedure in order to evaluate eventual intraprocedural bleeding. After waking, all the patients were transferred to the neurosurgical ward or to the neurosurgical ICU if necessary. The clinical outcome was independently evaluated by a neurologist at discharge and after 6 months. The modified Rankin Scale (mRS) was used to assess clinical outcome, with mRS 0 considered excellent and mRS 1 as a good clinical outcome. Of the 286 aneurysms, 245 (85%) were examined by digital subtraction angiographic evaluation after 6 months to assess eventual recurrences or delayed vascular morphological modifications (late intrastent stenosis). Major recurrences were defined as a change in the angiographic result from complete occlusion (grade I) or neck remnant (grade II) to sac recurrence, while a minor recurrence was defined as the progression from complete occlusion to a limited regrowth of the neck. Patients with a poor clinical outcome (n=3), those who refused further angiographic controls or were lost during follow-up (n=31), patients who died (n=5), and those in whom endovascular treatment failed and were subjected to clipping (n=2) were not included in the results of the angiographic follow-up. Patient age and gender, aneurysm localization (parent vessel), size and location (bifurcational or wall side), endovascular technique used (RT or stent-assisted coiling), intraprocedural complications, occlusion grade, and clinical outcome were included as variables. Size, location, intraprocedural complications, and occlusion grade were categorized as described previously. The occlusion grade was also categorized as complete or adequate occlusion; both grades I and II of the Raymond scale were included in the adequate occlusions.
A Student t test was used to evaluate differences between the two groups and a χ2 test was used to perform univariate analysis. Binary logistic regression with the backward stepwise method was used to perform multivariate analysis, where the clinical outcome and occlusion grade were set as dichotomous variables as described previously. Variables with a p value <0.05 were considered statistically significant. SPSS V.19 software was used to perform the statistical analysis.
The characteristics of the patients are shown in table 1, and the angiographic results at the end of the procedure and after 6 months and the recurrence rates are reported in table 2. In 3/286 (1%) cases the endovascular treatment was suspended because of difficulty with vascular access and the patients were referred for surgery without intraprocedural complications and were excluded from further analyses. Similar data about the feasibility of endovascular treatment have been reported in the literature.6 ,7 Three of the patients treated with stent-assisted coiling developed transient ischemic attacks after discontinuation of scheduled clopidogrel (or ticlopidine) after 3 or 6 months and dual antiplatelet therapy was restored for another 3 months and no clinical sequelae were registered during the follow-up.
Clinical outcomes at discharge and after 6 months and periprocedural complications are shown in table 3. We observed 26 adverse events (9.7%): 7 ischemic (2.6%) and 19 hemorrhagic (7%), independently of their clinical relevance. In 13/26 cases (5 ischemic, 8 hemorrhagic) we observed no or minimal clinical sequelae (mRS 0–1 at discharge), and 13 adverse events (13/268; 4.8%) were considered symptomatic (mRS 2–6). Two ischemic symptomatic complications occurred (2/268; 0.7%). In one patient with a medium-sized aneurysm of the right carotid siphon treated with RT an acute carotid occlusion was observed after coiling and immediately treated and resolved with urokinase and intravenous tirofiban; the patient was discharged with mRS 4 which remained stable after 6 months. The other patient had a medium-sized aneurysm of the left middle cerebral artery (MCA) which was treated with stent-assisted coiling; a distal thromboembolism was observed and treated with intravenous aspirin 500 mg and the patient was discharged with dysarthria (mRS 2) which improved during follow-up to mRS 1 after 6 months. Among the 11 hemorrhagic symptomatic complications (11/268; 4.1%), five occurred in MCA aneurysms, three in ACoA aneurysms, two in aneurysms in the carotid siphon (one at the internal carotid artery (ICA) bifurcation and one of the supraclinoid ICA) and one in a basilar artery aneurysm; eight aneurysms were medium-sized, two were large and one was small. Seven of the 11 aneurysms were treated with balloon RT and four with stent-assisted coiling. Six patients died (2.2%): three because of distal arterial perforation due to the guidewire, one from spontaneous aneurysmal rupture after the first angiographic run, and two because of sac perforation (one due to ventriculitis which occurred 2 months later after uneventful intraoperative rupture of the aneurysm and successive immediate external ventricular drainage placement). In the other five cases the intraprocedural hemorrhagic complications resulted in permanent moderate to severe neurological deficits (one mRS 2 who improved after 6 months to mRS 1, one mRS 3, two mRS 4, and one mRS 5). In 4/5 cases the hematoma was secondary to distal perforation due to the guidewire; in one case the hematoma was detected during the post-procedural CT scan and in the other three cases the hemorrhage occurred at least 6 h after the procedure. In 1/5 cases brainstem bleeding due to dual antiplatelet therapy occurred 5 days after treatment without intraprocedural complications.
Recurrences and retreatments
Twenty-two recurrences were observed during the follow-up period (22/245; 8.9%). Of these, 16 (16/143; 11.1%) occurred in the RT group and 6 (6/102; 5.8%) occurred in the stent group. However, in 12/22 cases (54%), recurrence (any grade) was observed in large to giant aneurysms (9 treated with RT and 3 with stent-assisted coiling). All recurrences were retreated and, in one case, the patient underwent two retreatments without additional intraprocedural complications.
The introduction of intracranial stents and balloons has contributed to the development of endovascular treatment of intracranial aneurysms. Several aneurysms that were not considered treatable with an endovascular approach are currently being coiled with stent- or balloon-assisted techniques. Although the use of flow diverter stents or intra-aneurysmal flow disruptors is increasing rapidly, there is not yet sufficient evidence of the superiority of these devices compared with conventional techniques, particularly in terms of safety.
Evolution of the treatment
Although no significant differences were observed between the two groups of patients, some interesting data are noteworthy. Small aneurysms were more frequently treated with balloon-assisted coiling, as were bifurcational aneurysms and, in general, all the categories in terms of aneurysmal size and localization. These data are illustrated in figure 1, which shows the evolution of the use of the two techniques at our institutions; since 2003, when we started to use intracranial stents, their use has grown linearly during the last decade. Although there are no commonly accepted indications for the treatment of wide-necked unruptured intracranial aneurysms, until 2003 the endovascular treatment of these aneurysms according to their size, localization, or age was growing rapidly and the only available technique was RT.
The angiographic results at the end of the procedure showed a higher rate of complete occlusion and a lower number of neck remnants with stent-assisted coiling than with RT; this difference was statistically significant (86.8% vs 78%, p<0.001, table 4). The rates of incomplete occlusion (grade III) were similar in the two groups (6% in the RT group vs 6.5% in the stent-assisted coiling group) and show the difficulties of both endovascular techniques in relation to the aneurysmal morphology such as the size (in 6/18 cases the size was large to giant) and localization (in 13/18 cases the sac was situated on a bifurcation). Our data are comparable to those reported in the literature for both techniques, particularly concerning complete occlusion and neck remnants.4 ,8–16
A comparison of the stability of the endovascular treatment after 6 months showed analogous results. A higher rate of persistent occlusions (92.1% vs 77.6%) and a lower number of neck remnants (3.9% vs 12.5%) was observed in the stent-assisted coiling group. However, when considering the recanalization rate, stent-assisted coiling showed significantly better results than the RT group: major recurrences were observed in 9.7% of patients treated with RT compared with 3.9% in the stent group (p=0.034), and this is in line with previously reported data.9 ,10 ,17–20 Furthermore, when considering both major and minor recurrences, the overall recurrence rate was higher in the RT group than in the stent group (11.1% vs 5.8%). Finally, among the 38 large to giant aneurysms, any grade of recurrence was observed in 9/24 cases (37.5%) treated with RT and in 3/14 (21.4%) treated with stent-assisted coiling. Therefore, even in aneurysms with high recurrence rates, stent-assisted coiling seems to improve the stability of the coil mesh. A significant correlation between the size of the aneurysmal sac and the recurrence rate was confirmed by multivariate analysis (p<0.001, table 4).
No significant differences regarding safety were observed between the two groups. The hemorrhagic risk (considering any clinical outcome secondary to bleeding) was the same (6%) in both groups, and the risk of ischemic complications was similar in patients treated with RT and stent-assisted coiling. The results were also comparable when only those complications (ischemic or hemorrhagic) with a neurological deficit (mRS ≥2) were considered. Likewise, independently of the technique used, we observed that almost all the symptomatic complications, ischemic or hemorrhagic, occurred in bifurcational aneurysms (14/15 cases). However, when considering only hemorrhagic complications, the localization of the aneurysmal sac at the level of a bifurcation represented a significant risk factor (OR 4.3, p=0.037), as did older age (OR 5, p=0.024). Statistical analysis showed that the MCA is most significantly correlated with a higher risk of intraprocedural complications (p=0.016). Finally, multivariate analysis showed a strong statistical correlation between mortality and intraprocedural hemorrhagic complications (OR 26.9, p<0.001). Furthermore, although dual antiplatelet therapy was administered in all stent-assisted procedures, the hemorrhagic complication rates were similar between the two groups and were more frequently secondary to distal perforations due to the microguide than to the use of the stent or balloon. Previously reported data about the safety of assisted coiling are comparable to those shown in our series.7 ,11 ,16–18 ,20 ,21
The limitations of the study include its retrospective and single-center design, the absence of an analysis of recanalization rates according to packing density (or number of coils used), and the relatively limited number of patients, even if considerable.
Stent-assisted coiling is associated with higher stability than RT, reducing the number of retreatments without being associated with a higher risk of intraprocedural complications. Large to giant aneurysms and those localized at bifurcations with particular characteristics (arteries originating from the sac, wide neck) still represent a challenge for assisted coiling. The application of flow diverter stents for wall side aneurysms or intrasaccular devices for bifurcational aneurysms has not yet been shown to be superior. Currently, particularly for MCA aneurysms, detailed morphologic analysis represents the basis for considering a surgical option or novel endovascular techniques, independently of the technical expertise.
Contributors All the authors fulfilled the criteria for authorship.
Competing interests None.
Ethics approval The study was approved by the local review board.
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.