Introduction The most widely used intracranial stents for stent-assisted coiling are Neuroform (NEU) and Enterprise stents (ENT). This study compares published outcomes between the ENT and NEU coil-assist systems and comments on the published safety and efficacy of stent-assisted coiling in general.
Methods A literature search was performed through PubMed for all published series of ENT or NEU stent-assisted coiling of cerebral aneurysms from 2004 to 2014. All studies including 10 or more published cases of stent-assisted coiling with ENT or NEU were included.
Results 47 studies met the inclusion criteria, containing 4238 aneurysms in 4039 patients. 2111 aneurysms were treated with NEU and 2127 were treated with ENT. Mean follow-up was 14.1 months. Overall, thromboembolic events occurred in 6.4% of aneurysms, intracranial hemorrhage (ICH) in 2.6%, permanent morbidity in 3.9%, and mortality in 2.3%. Initial and final 100% angiographic occlusion was seen in 53% and 69% of patients, respectively. Deployment failures (p<0.001), ICH (p=0.001), mortality among all patients (p=0.03), and recanalization (p=0.02) were more commonly reported in NEU treated aneurysms. The ENT system was also associated with higher reported complete occlusion at follow-up (p<0.001).
Conclusions This literature review represents the largest and most robust comparison of stent-assisted coiling devices to date, containing over 4200 aneurysms in more than 4000 patients. Comparative analyses demonstrate that both devices are reported to be safe and effective with comparable permanent morbidity.
Statistics from Altmetric.com
Stent-assisted coiling has emerged as a popular treatment strategy for complex cerebral aneurysms with wide necks or low dome-to-neck ratios. Intracranial stents are hypothesized to improve angiographic aneurysmal occlusion by serving as a scaffold to prevent coil prolapse, enhancing coil packing density, providing greater stability to the coil mass and also promoting re-endothelialization.1–10
The two most commonly used intracranial stents in the USA are Neuroform (NEU; Stryker, Kalamazoo, Michigan, USA) and Enterprise (ENT; Codman & Shurtleff, Raynham Massachusetts, USA) stents. NEU is a self-expanding stent with an open-cell design that received Food and Drug Administration (FDA) approval in 2002. Since that time, the NEU system has undergone multiple iterations including NEU 2, 2 Treo, 3 and EZ. ENT is a self-expanding stent with a closed-cell design that received FDA approval in 2007.
To date, several publications have reviewed the published literature comparing stent-assisted coiling between NEU and ENT. Since the publication of the most recent large review,11 the number of case series published reporting on stent-assisted coiling has reached over 4000 patients. The purposes of this study are to: (1) evaluate the literature in its entirety, including the most recent published series, (2) compare published clinical and angiographic outcomes between the ENT and NEU coil-assist systems; and (3) comment on the safety and efficacy of stent-assisted coiling in general.
A literature search was performed through PubMed for all published series of ENT or NEU stent-assisted coiling of ruptured and unruptured cerebral aneurysms from January 2004 to March 2014. Search terms included ‘Enterprise stent’, ‘Enterprise and aneurysm’, ‘Neuroform stent’, ‘Neuroform and aneurysm’, and ‘cerebral aneurysm and stent’. All abstracts of manuscripts published in English were reviewed. Case series, prospective studies or clinical trials with 10 or more patients containing reported clinical and/or radiological data following intracranial stenting with ENT or NEU stents for the treatment of cerebral aneurysms were included. In cases of doubt, the entire article was reviewed. For studies including data from multiple stent systems, only those patients with clearly listed data for each individual system were included. Furthermore, all studies included in the previous review by Gross and Frerichs11 were included, regardless of the number of cases published. Patients receiving both ENT and NEU stents simultaneously were excluded.
All studies that met our inclusion criteria were reviewed in full, and from each study we extracted the sample size for each stent system used, the mean duration of follow-up and the following outcome measures: deployment failure, thromboembolic events (TE; defined as transient ischemic attack, stroke, or development of asymptomatic thrombus during procedure), delayed in-stent stenosis (defined by the presence of the term ‘moderate’, ‘severe’, or ‘symptomatic’, as ≥50% if quantified, or if not clearly defined by the authors), new peri-procedural intracranial hemorrhage (ICH), permanent morbidity (present at last follow-up), mortality, initial complete (100%) angiographic occlusion of the aneurysm after treatment, complete occlusion of the aneurysm at last follow-up, and recanalization or recurrence of the aneurysm at follow-up. Studies that used terms such as ‘recanalization’, ‘recurrence’, those that listed progression of aneurysm remnant, or those that listed progression to a higher Raymond grade on follow-up were included in the last outcome measure. All recorded data were verified by a second reviewer and collected on a standardized spreadsheet. Deployment failure, initial complete occlusion, complete occlusion at follow-up, and recanalization were assessed per aneurysm while the remaining outcome measures were assessed per patient. If data on a particular outcome measure were not available for a given study, the patients from that study were excluded from the denominator during analysis of that outcome measure.
For each outcome measure a two-tailed Fisher exact test was used to determine significance. p Values <0.05 were considered statistically significant.
A total of 47 studies met the inclusion criteria,1–3 ,6–10 ,12–50 containing a total of 4238 aneurysms treated in 4039 patients. Of these, 2111 aneurysms were treated with NEU and 2127 were treated with ENT. The majority of included studies were retrospective in nature (81%), although a total of nine prospective studies were included. Among these studies, 263 aneurysms (14.5%) were treated after clearly-defined acute aneurysmal subarachnoid hemorrhage; 1271 aneurysms (30.4%) were treated after any rupture (acute, subacute, or not defined). More patients were treated with NEU than ENT after acute subarachnoid hemorrhage (16.2% vs 12.7%, respectively; p=0.04).
For all studies, the mean (SD) duration of follow-up was 14.1 (10.0) months. For studies using the ENT system the mean (SD) duration of follow-up was 10.3 (4.7) months and for studies using the NEU system the mean (SD) duration of follow-up was 17.2 (12.0) months.
Table 1 shows the overall rates of clinical complications and complete angiographic occlusion as well as the comparative analysis between stent types for all patients. Note that the denominator changes based upon the number of patients where each potential complication was reported.
Of the 263 patients treated after clearly-defined acute subarachnoid hemorrhage, 243 patients had morbidity and mortality data reported (145 patients treated with NEU and 98 patients treated with ENT). Due to limited reporting of individual complications for these patients (such as in-stent stenosis or TE complications), only permanent morbidity and mortality could be compared (table 2). Permanent morbidity and mortality in patients with subarachnoid hemorrhage were not significantly different between groups (p=0.63 and 0.15, respectively). In those patients treated electively or in a subacute fashion after subarachnoid hemorrhage, no statistically significant difference was identified in either permanent morbidity (p=0.31) or mortality (p=0.32).
The present study represents the most robust review of the current literature on stent-assisted coiling using NEU and ENT intracranial stents, containing over 4200 aneurysms in more than 4000 patients. Overall, this analysis demonstrates that stent-assisted coiling with these devices produces low permanent morbidity (4%) and mortality (2.3%), low rates of TE complications (6.4%) and ICH (2.6%), and satisfactory angiographic occlusion at last follow-up with nearly 70% of patients having Raymond grade 1 (100%) occlusion. When considering only those patients treated electively or in a subacute fashion after subarachnoid hemorrhage, permanent morbidity and mortality occurred in only 3.6% and 1.7% of patients, respectively. The low clinical complication rate and high degree of occlusion are particularly encouraging, given that most aneurysms treated with stent-assisted coiling are done so for aneurysms with wide necks, poor dome-to-neck ratios, or other complex features that may incur a higher risk of coil prolapse or TE complication if coiling primarily, as well as a higher likelihood of recanalization. Furthermore, the low ICH rate is also encouraging given the requirement for dual antiplatelet therapy after stent deployment. Interestingly, there is a highly statistically significant increase in the rate of ICH for NEU stents. It is likely that this difference is due to earlier iterations of the technology that required an exchange procedure. One may hypothesize that, with newer catheter-based delivery iterations of the NEU, the ICH rate may approach the very low 1.6% rate of the ENT system. Overall, this large literature review suggests that stent-assisted coiling is a safe and effective treatment strategy for intracranial aneurysms.
Stent-assisted aneurysmal coil embolization has become a widely accepted endovascular treatment modality for wide-necked or complex aneurysms. The use of a stent provides a scaffold while the coils are placed into the wide-necked aneurysms and also allows for improved packing density.1–10 ,51 Furthermore, stents may promote endothelialization at the aneurysm neck and may encourage progressive aneurysm occlusion through a flow remodeling phenomenon.14 ,52 These are potential explanations as to why some studies have suggested that aneurysms embolized with stent-assisted coiling have lower recanalization rates than those treated without stent assistance.14 ,53 ,54 NEU, an open-cell design, may potentially improve wall apposition in curving arterial segments. In contrast, the closed-cell design of the ENT may better protect against coil prolapse and allows for stent recapture after partial deployment.
Flow diversion is becoming an increasingly popular alternative means of treating wide-necked or complex cerebral aneurysms on the proximal internal carotid artery (ICA) as it does not require direct aneurysm catheterization. In a large literature review encompassing 905 patients with over 1000 aneurysms treated with the Pipeline Embolization Device (PED; ev3-Covidien, Plymouth, Minnesota, USA),55 both ICH and mortality occurred in 2.3% of patients. At 6 months, 80% of aneurysms showed complete occlusion. Furthermore, recent cost-effectiveness analyses have favored flow diversion over alternative endovascular treatments, an effect that becomes more pronounced as aneurysm size increases.56–58 However, PED is currently only FDA-approved for wide-necked aneurysms from the petrous to superior hypophyseal segment of the ICA, a restriction not shared by ENT or NEU. Furthermore, PED may be particularly unsuited to bifurcating aneurysms. The present study suggests that stent-assisted coiling has comparable safety and efficacy to the PED but appears to have more versatility given that it may be used efficaciously in smaller parent arteries or at bifurcation sites without the need for off-label use. Furthermore, many practitioners have grown comfortable with stent-assisted coiling techniques while newer technology like the PED requires additional training to master procedural nuances. For these reasons, stent-assisted coiling will likely remain a strong consideration for aneurysms with wide necks or poor dome-to-neck ratios, even in light of the increasing popularity of flow diversion.
A comparative analysis between the published results of the NEU and ENT systems demonstrates several important findings. Statistically significant differences favoring ENT were seen in the rates of deployment failure, ICH, mortality, complete aneurysm occlusion at follow-up, and recanalization when all patients were considered. However, when divided into acutely ruptured versus unruptured/subacute rupture, no significant differences between stent technology were found in terms of permanent morbidity or mortality. These results should be interpreted with caution. The representative data include studies across multiple device iterations over the course of a decade and the data do not necessarily represent the procedural device failures, clinical sequelae, or occlusion of the most modern device iterations or procedural techniques used currently. For instance, the higher ICH rate and deployment failure rate with NEU has been suggested to be due to the high rate of stent migration and potential guidewire perforation from earlier generation NEU devices.11 In addition, more patients were treated with NEU after acute subarachnoid hemorrhage. Although we attempted to control for rupture status with regard to morbidity and mortality by segregating those with acutely ruptured aneurysms from those with chronic or unruptured aneurysms, we were unable to control for rupture status in the analysis of angiographic outcomes or other clinical endpoints. Therefore, differences in rupture status may potentially explain the higher recanalization rate and lower rate of 100% angiographic occlusion at last follow-up seen in the NEU group.
There are significant limitations to the present study. Biases of the individual studies, the large number of retrospective series, and heterogeneity between studies limits the validity and generalizability of the results. This study represents a collection of published yet heterogeneous case series without rigorous inclusion criteria or grading of included studies. Additionally, demographic information was not analyzed between groups. This may impart significant bias to the conclusions that are generated. Furthermore, variable reporting using different definitions makes direct comparisons difficult. The short- and long-term clinical significance of these findings is not clear, and the majority of long-term follow-up data in the literature do not extend beyond 18 months. In addition, the number of stents used was not included in the analysis. Given that a large portion of the data is self-reported, the reported recanalization rate of 12% probably significantly underestimates the true rate as numerous studies have demonstrated that self-reported data consistently overestimate effect size compared with core laboratory adjudicated data.59–61 As suggested by Gross and Frerichs,11 different stents may be more or less applicable depending on factors such as the baseline parameters of the patient, the clinical characteristics of the aneurysm, and operator training. Additionally, the differences in cost or availability of these devices may drive the selection of devices. These factors are not accounted for in this evaluation, and may limit the ability to directly compare outcomes found in the literature. Lastly, as previously stated, the NEU system has undergone substantial revisions to its delivery system since its introduction and therefore the results presented here may not apply to the most current iteration.
This study represents the largest and most robust literature review of the two most common stent-assisted coiling devices to date, containing 4238 total aneurysms in 4039 patients. Overall, stent-assisted coiling is safe and effective, with permanent morbidity occurring in 4% of patients and complete angiographic occlusion occurring in nearly 70% at last angiographic follow-up. Comparative analyses show that both devices are reported to be safe and effective with comparable permanent morbidity. However, these data demonstrate that, over the history of the two technologies, ENT is associated with statistically significant lower reports of hardware failure at deployment, ICH and recanalization, as well as higher reported rates of complete occlusion at follow-up.
Contributors All authors contributed to the manuscript through data acquisition, data analysis, manuscript composition, and/or through critical review of the manuscript. All authors provided final approval for publication.
Competing interests JM serves as a consultant to Reverse Medical, Lazarus Effect, Pulsar, Medina Medical, and Edge Therapeutics; has investor interests in Blockade Medical and Medina Medical; and is on the Advisory Board for Codman Neurovascular.
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.