Silk flow-diverter stent for the treatment of complex intracranial aneurysms: A one-year follow-up multicenter study

Abstract

Background

Flow-diverter stents have been successfully used in the treatment of complex aneurysms with limited therapeutic alternatives. We report our experience using the Silk flow diverter (SFD; Balt Extrusion, Montmorency, France) for the treatment of complex aneurysms in four Argentine centers.

Methods

We conducted a retrospective review of 246 consecutive patients who were treated with the SFD at four Argentine centers between January 2009 and January 2017. The patient and aneurysm characteristics, as well as the details of the procedure, were analyzed. The angiographic and clinical findings were recorded during and immediately after the procedure and at 12-month follow-up.

Results

Angiography follow-up at 12 months was possible in 235 patients (95.5%) with 282 aneurysms. A total of 265 aneurysms (93.9%) presented with complete occlusion of the aneurysmal sac (class 1) and 17 aneurysms (6.1%) presented with partial occlusion (class 2). The 12-month clinical follow-up showed 11 patients with major events (seven, scale 2; five, scale 3; and two, scale 4). The morbidity and mortality rates were 4.2% (11/289) and 2.1% (5/289), respectively.

Conclusions

The treatment of aneurysms with the SFD was associated with a low rate of complications and a high percentage of aneurysmal occlusion. These findings suggest that SFD is an effective and safe alternative in the endovascular treatment of complex aneurysms.

Introduction

Treatment of intracranial aneurysms remains a great challenge for modern medicine. Surgical advances have been implemented in wide subarachnoid dissections and the use of multiple clips for reconstruction of arteries in cases of dysplastic and giant aneurysms.¹ On the other hand, advances in endovascular treatment have evolved with the innovation of flow diverters (FDs), which have emerged as a promising therapeutic option in the treatment of complex cerebral aneurysms.^2–6

A large number of publications have reported the use of FDs,^7–12 but most of the large series use the Pipeline device. In this study, we aimed to report our experience with the Silk flow diverter (SFD) through a retrospective study. To our knowledge, this is the longest reported series to evaluate the safety and effectiveness of the system.

Methods

We conducted a retrospective review of a consecutive series of 246 patients with 293 aneurysms treated with the SFD at four Argentine centers between January 2009 and January 2017. This research was conducted according to the ethics board of each institution. The clinical and angiographic image data collection and analysis were performed by six operators (G.F.T., F.R., S.P., J.G., R.M., and I.A.). A signed informed consent for the SFD procedure was obtained from all patients or their family members in case of altered level of consciousness. Patient demographic data and relevant comorbidities were registered.

The decision to treat with the SFD was made on a case-by-case basis by a multidisciplinary team that included vascular neurosurgeons and interventional neuroradiologists in the following situations: (1) dissecting, fusiform, serpiginous, giant, or wide neck, with a body-neck relationship of 1 or 2/1 (dome ratio to neck less than 2) aneurysms, (2) patients with recanalization of previously treated aneurysms via transcranial or endovascular procedures, (3) incidental dysplastic aneurysms with high risk of impending rupture, and (4) symptomatic small aneurysms not susceptible to being treated with other techniques, or multiple aneurysms. This study included 246 patients with 293 aneurysms, 273 non-ruptured (93.17%) and 20 ruptured previously treated aneurysms, with a recanalization or partial occlusion of the aneurysmal sac in the six-month angiographic follow-up (6.83%). Of the patients who had bled, 11 were previously treated with coils, six of them with stent plus coils, and three had been clipped.

Therapeutic strategy and endovascular procedure

In general, all patients were pretreated with double-antiplatelet therapy. This therapy consisted of 150 mg of acetylsalicylic acid and 75 mg of clopidogrel per day for five days prior to the procedure and was continued after the procedure with the same schedule for six months. Platelet aggregation tests were not available in all hospitals. During the procedure, heparinization was used to achieve a coagulation activation time of 250–300 seconds. Double-antiplatelet therapy was continued longer than six months when an intra-stent stenosis existed in the digital subtraction angiography follow-up. Patients with giant aneurysms were also treated with corticosteroids for 10 days to avoid the inflammatory reaction of the aneurysm wall generated by the mural thrombus.

All procedures were performed by two or three interventional neuroradiologists or neurosurgeons with experience in intracranial stent placement. Stent size selection was based on the measurements obtained in previous angiographies and during the procedure. When a difference existed between the caliber of the arteries proximally and distally to the aneurysm, we selected the size corresponding to the proximal segment. The decision to overlap systems or to place coils in addition to the SFD was left to the operator of each center. The positioning and deployment of the SFD were carried out strictly following the manufacturer’s recommendations (Balt Extrusion, Montmorency, France). All patients were transferred to a monitored neurosurgical intensive care unit for early detection of vasospasm, hemorrhage, and other complications related or not to the procedure.

Occlusion rate

For the evaluation of the degree of aneurysmal occlusion, an angiography was obtained immediately after the procedure and at six- and 12-month image follow-up. Owing to the multicentric nature of the study and the heterogeneity of the samples, we included only angiograms obtained at 12 months in our analysis. The angiographic findings were classified using the Szikora classification¹¹ (class 1, complete stasis; class 2, decreased flow; class 3, no flow changes). In patients with angiographic follow-up, the presence or absence of in-stent stenosis and the parental artery occlusion (PAO) were evaluated. The evaluation of the images was made in the reference hospital by two expert neuroradiologists.

Clinical complications

Intra- and post-procedure complications were classified according to a protocol similar to the one used by Berge et al.⁶ as acute-subacute (those that occur in the first week post-procedure) and delayed complications (those that occur between two weeks and 12 months post-procedure). Clinical results were evaluated at hospital discharge and 12-month follow-up using the modified Rankin scale (mRS).¹³ The permanent morbidity rate was calculated as the percentage of patients with an adverse event that resulted in an mRS > 2 at the time of clinical evaluation (12 months).

Statistics

Statistical analysis was performed using IBM SPSS for Windows, version 21.0 (Armonk, NY, USA: IBM Corp).

Quantitative variables were summarized as mean ± standard deviation and median (minimum–maximum), and qualitative variables were presented as number and percentage. The analyses of odds ratios (ORs) and p values were performed using a fixed-effects approach.

Results

Patients and aneurysms

A total of 246 patients (212 women, 86.2%; 34 men, 13.8%) with 293 aneurysms were treated during the study period. The range and mean age were 16–72 and 51 years, respectively. A total of 128 patients (52%) were asymptomatic, and 118 (48%) were symptomatic (headache, 57 patients (23.2%); history of subarachnoid hemorrhage, 20 (8.1%); paralysis of cranial nerve, 29 (11.8%); cerebral ischemia, seven (2.8%); and five, plegic symptomatology (2.0%)). There were 249 (85%) saccular aneurysms and 44 fusiform aneurysms (15%). With regard to their size, 59 (20.1%) were giant, 53 (18.1%) were large, and 181 (61.8%) were smaller than 10 mm. The majority of the aneurysms (286, 97.6%) were located in the anterior circulation (267, internal carotid artery; four, middle cerebral artery; and four, anterior cerebral artery), and seven (2.4%) were located in the posterior circulation (six, basilar artery; one, cerebral posterior artery). Of the 246 patients with 293 aneurysms, 273 (93.2%) were unruptured and 20 (6.8%) were previously treated ruptured aneurysms that presented with partial occlusion or recanalization of the aneurysmal sac during angiographic follow-up. A total of 209 patients had one single aneurysm (85%); 30 patients, two aneurysms (12%); four patients, three aneurysms (1.6%); and three patients, more than four aneurysms (1.2%).

Procedural results

In four patients, with an aneurysm each, the implantation of Silk was not possible. A total of 246 SFDs were deployed in 242 patients with 289 aneurysms. Of the implanted systems, 4.1% (10/246) showed poor apposition to the arterial wall, characterized by folding/kinking, inadequate opening, or shortening of the stent. In seven patients, obstruction of the parent artery was observed. In one case, with a positive test of carotid occlusion, no additional therapeutic activity was performed, and the patient presented with a distal ischemic event of the middle cerebral artery. In six cases, additional procedures, such as balloon angioplasty in three patients and chemical thrombolysis in three patients (one with tirofiban and two with eptifibatide), were necessary. In three cases, a new overlapped Silk was implanted to cover the whole aneurysm and dysplastic parent artery.

Additional coils were located inside the aneurysm in 21 cases (8.5%). In four cases, a Leo stent (Balt Extrusion, Montmorency, France) was previously implanted as a support to allow anchoring of the SFD. In other patients with a recurrent aneurysm treated with coils and a Neuroform stent (Boston Scientific, Natick, MA, USA), the SFD was implanted through the pre-existing stent. During the positioning of the stent, one patient developed a parenchymal cerebral hemorrhage due to dissection of the middle cerebral artery.

Clinical complications

The 12-month clinical follow-up showed 11 patients with major events (seven, scale 2; five, scale 3; and two, scale 4). The morbidity and mortality rates were 4.2% (11/289) and 2.1% (5/289), respectively. There were 10 ischemic and one hemorrhagic complication. Two deaths were due to acute ischemic infarctions following the treatment of basilar aneurysms (two cases). Of the three remaining patients, one died of contralateral parenchymal hemorrhage and another of a subarachnoid hemorrhage. The cause of death of the third patient was unknown as autopsy was not performed.

There were no significant differences in morbidity and mortality between the group of aneurysms treated with FD and those treated with adjunctive coiling.

Occlusion rate

Immediate angiography after the procedure showed complete occlusion (class 1) in four aneurysms (1.4%), subtotal occlusion (class 2) in 278 aneurysms (96.2%), and incomplete occlusion (class 3) in seven aneurysms (2.4%). Angiographic follow-up at 12 months was possible in 235 patients (95.5%) with 282 aneurysms. We observed 17 in-stent stenoses, four cases of which had parent artery obstruction that was asymptomatic due to satisfactory collateral circulation. A total of 265 aneurysms (93.9%) presented with complete occlusion of the aneurysmal sac (class 1), including those with PAO, and 17 aneurysms (6.1%) presented with partial occlusion (class 2). All aneurysms (21) of the subgroup treated with FD and coiling showed complete occlusion.

The multivariate logistic regression analysis in terms of the degree of occlusion obtained showed no significant differences among the morphology, localization, and size of the aneurysms (Table 1).

Table 1.

Logistic regression analysis of morphology/total occlusion.

	OR	p	CI	χ2
Small	2	<0.75	0.174–23.05	0.316
Large	0	<0.01	0.00	6.84
Giant	1.36	<0.9	0.116–15.94	0.06
Fusiform	0.96	<0.97	0.28–3.32	0.002
Saccular	1.037	<0.97	0.3–3.57	0.002

CI: confidence interval; OR: odds ratio.

In the subgroup of aneurysms of the posterior territory, mortality was 1.4% and morbidity 1.1% with a complete occlusion rate of 100%.

Discussion

In our analysis, technical complications, such as poor stent opening, stent migration, shortening, and obstruction of the parent artery, occurred in 10 cases (4.1%). Most of the difficulties were observed mainly in vessel segments with tight or acute curves and with implantation of an over- or undersized device. Our results agree with those of the longest series published by Briganti et al.⁹ and Pumar et al.⁷ However, the percentage of technical difficulties described in other series was between 25% and 4%, with an average of 11.4%.^14,15 A direct comparison between these findings and our results is not possible because of the great variation in key criteria in the selection of patients and population size. In our series, deployment difficulties occurred in earlier treatments. Thus, we believe that the percentage of difficulties decreased significantly as the learning curve progressed, which allowed us to acquire skills for the optimal development of this technique.

Seven cases (2.8%) of parental artery obstruction were observed. Six of them were resolved without clinical repercussion, three by balloon angioplasty and three by chemical thrombolysis, and the other case, with a positive test of carotid occlusion, had no additional therapeutic activity. These findings are below the mean on average compared with those in other studies in which FDs were used (Berge et al.,⁶ 12.3%; Byrne et al.,¹² 10%), and similar to those in other series (Strauss and Maimon¹⁰, 5%; Szikora et al.,¹¹ 2.4%)., Lubicz et al., 1.9%.¹⁴ In our study, we believe that PAO was due to nonmodifiable factors that increased tension. This decreased the penetrability and navigability of the Vasco microcatheter beyond the aneurysmal neck and ensured sufficient safety margins to navigate the SFD and modifiable factors characterized by technical failure in the preparation and navigability of the system, as well as in the selection of the appropriate size of the stent. All these elements may contribute to inadequate placement, defects in delivery, or poor opening of the SDF.

Few studies reported the patency of side branches that are covered with flow-diverting stents secondary to the treatment of aneurysm.^16–18 We encountered a case of perforator or side vessel occlusion at the time of treatment. A patient with a giant carotid cavernous aneurysm treated with two SFDs developed a peripheral facial paralysis and hemiparesis associated with the occlusion of the right anterior choroidal artery (AChoA). Similarly, Brinjikji et al.³ reported a single case of obstruction of the AChoA without clinical repercussion in 14 AChoA covered with FDs. However, Bhogal et al.¹⁶ reported a 20% occlusion in 140 patients with 147 aneurysms. None of their patients developed either permanent or transient symptoms that could be related to AChoA occlusion. In our case, we believe that the developed AChoA territory infarction in one patient was due to the two overlapping SFDs. These overlapping SFDs reduced the area of the AChoA origin with consequent inadequate flow that the medial and lateral posterior choroidal arteries were unable to collaterally supply.

Angiographic follow-up at 12 months was possible in 235 patients (95.5%) with 282 aneurysms showing 93.9% complete occlusion (265/282). Our results are similar to the 93% to 94% rates demonstrated in the early series by Lylyk et al.,¹⁹ Szikora et al.,¹¹ and the Intracranial Treatment of Aneurysms trial, contrasting the minor degree of occlusion of 78.1% to 86.8% reported by Pumar et al.⁷ and Berge et al.⁶ We believe that this difference in occlusion rates could be related to the degree of intra-aneurysmal thrombosis prior to treatment or the influence of dual-antiplatelet therapy. We could not perform computational measurements on the hemodynamic changes in terms of the degree of aneurysmal flow reduction and its relationship with the thrombotic process and retraction of the aneurysm sac. Despite the poor immediate efficacy in showing a complete stasis after the SFD procedure, our results confirm good midterm efficacy of the SFD to obtain a progressive aneurysm occlusion.

Although in the present study adjunctive coiling was performed in part of the patient cohort, there are several questionable aspects regarding its efficacy and the possibility of increasing the complications of this strategy compared with FD embolization alone. There are authors who have advocated adjunctive coil embolization as a method that improves the degree of occlusion and minimizes the risk of aneurysmatic rupture.^20–22 Others have argued that this technique does not add any advantage in terms of treatment efficacy. Lin et al.²⁰ and Szikora et al.¹¹did not find statistically significant differences in the complication rates between the FD-alone group and the FD-plus-coiling group. In our study we did not observe any potential differences in complications between these two treatment strategies.

FDs are a good tool to reduce the degree of aneurysmal recurrence compared with other vascular techniques. In our series, as well as in the series by Pumar et al.,⁷ Kulcsár et al.,²³ Lubicz et al.,¹⁴ Szikora et al.¹¹ and Berge et al.,⁶ we did not observe any recurrence or recanalization of the aneurysmal occlusion at the 12-month follow-up. However, Byrne et al.¹² described two cases of recanalization in two aneurysms previously excluded by SFD. Based on these findings, we plan to continue our long-term follow-up for five years to determine the real efficacy of SFD in aneurysmal occlusion.

Previous studies and meta-analyses^24–27demonstrated that the morbidity and mortality rates ranged from 0% to 12% and 0% to 7%, respectively. In our study, the morbidity and mortality rates were 4.1% and 2%, respectively. These results are below the proportional mean found in the meta-analysis of Ye et al.,²⁵ who showed an overall morbidity of 9.8% and a global mortality of 3.8% in 2508 patients. The meta-analysis of Arrese et al.²⁶ showed a global morbidity of 9.9% and an overall mortality of 4.1% in 897 patients. Ischemic stroke due to thromboembolism and perforator infarctions are well-described complications of FD use. Our results were similar to those of a multicenter study involving 273 patients with 295 aneurysms that were treated with the SFD or Pipeline embolization device,²⁴ which showed an overall mortality rate of 5.9% and a morbidity rate of 3.7%. In our experience, most of the ischemic complications, which were similar to those described by Berge et al.,⁶ were observed in patients with incorrect SFD deployment and subsequent in-stent thrombosis.

In our analysis, no patient presented with delayed bleeding in contrast to 2.4% (19 hemorrhages/793 patients) and 3% (75 hemorrhages/2508 patients) described in the meta-analyses performed by Kallmes et al.⁸ and Ye et al.,²⁵ respectively.

The recent literature reports preliminary clinical experiences with new devices such as the Flow Re-Direction Endoluminal Device (FRED; Microvention, Tustin, CA, USA),^18,19 Pipeline Flex (Covidien),²⁰ Surpass (Stryker, Freemont, CA, USA),²¹ and p64 (Phenox, Bochum, Germany),^22,23 and for newer devices such as Derivo (Acandis, Pforzheim, Germany) the experience is still preclinical.²⁴ Understanding the safety and efficacy profiles associated with new FDs in the treatment of intracranial aneurysms should therefore be considered. A direct literature review comparison of the effectiveness among these new devices and the SFD used in our study is shown in Table 2.

Table 2.

Literature review of new FDs.

Authors	Complete occlusion rate at F/A	Morbidity	Mortality	Number of patients	FD	Number of aneurysms
Kocer et al.28	78.1%	6%	NA	35	FRED (35)	37
Möhlenbruch et al.29	73%	10.3%	0%	29	FRED (35)	34
Briganti et al.30	83%	0%	0%	20	FRED (24)	24
Fischer et al.31	85.7%	1.7%	0.8	121	p64 (127)	130
Briganti et al.32	88%	2.5%	0%	40	p64 (50)	60
Akgul et al.33	77.8%	8.4	4.3	24	Derivo (26)	34
Wakhloo et al.34	75%	4.2%	2.4	165	Surpass (165)	186
Our results	88.5%	2.9%	0.9%	246	SFD (246)	293

F/A: follow-up angiography; FD: flow diverter; FRED: Flow Re-Direction Endoluminal Device; SFD: Silk flow diverter.

Limitations

The main limitation of this study is its nonrandomized retrospective design, with four participating centers following their standardized practice for treating aneurysms with SFD. Although a broad range of aneurysm types and sizes was included, it was an observational study without preplanned subgroup analysis.

A limitation from the point of view of statistical analysis has been the loss to follow-up of 4.5% of patients.

Despite these inherent limitations, we believe that the conclusions may be relevant.

Conclusions

The SFD is safe and effective with high complete occlusion grades associated with an acceptable rate of neurologic complications.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.