Neuroradiology/Head and Neck Imaging
Original Research
Comparison of Enterprise With Neuroform Stent-Assisted Coiling of Intracranial Aneurysms
Audio Available | 
Share
OBJECTIVE. The Enterprise stent is the first closed-cell stent designed to treat wide-necked intracranial aneurysms. Advantages of the design can include improvement in keeping coils within an aneurysm and the ability of the stent to be recaptured. We compared the technical and clinical complications of the Enterprise stent with the open-cell Neuroform stent, its primary alternative.
SUBJECTS AND METHODS. Patients undergoing Enterprise and Neuroform stent-assisted aneurysm coiling were enrolled in prospective registries starting in March 2007 and February 2003, respectively. All consecutive patients through December 2011 were included. Deployment success and difficulty, stent movement and misplacement, and procedural complications were compared.
RESULTS. Enterprise deployment success was high (108 of 115 attempts, 93.9%) with 102 aneurysms receiving a stent compared with Neuroform (173 of 214 attempts, 80.8%, p = 0.001) with 163 aneurysms. Enterprise was easier to deploy (1.7% vs 15.9% difficult deployment, p < 0.0001). There were no significant differences in the rates of stent movement, misplacement, or symptomatic hemorrhage. Symptomatic thromboembolic events, however, were more frequent with the Enterprise stent (8.7% vs 1.4%, p = 0.0021). The Enterprise stent enabled treatment of 10 additional aneurysms that could not be treated with Neuroform and had a higher rate of immediate aneurysm occlusion (87.3% vs 73.0%, p = 0.0058).
CONCLUSION. Enterprise was easier to deploy and enabled treatment of additional aneurysms; however, there were more thromboembolic complications. On the basis of these findings, we prefer to use the Neuroform stent first and rely on the Enterprise stent as an easy-to-deliver backup for stent-assisted coiling.
Keywords: Enterprise stent, intracranial aneurysm, Neuroform stent, stent-assisted coil embolization
Endovascular coiling of intracranial aneurysms has become an established less-invasive alternative to surgical clipping in selected cases [1]. However, wide-necked aneurysms are associated with higher recurrence rates and increased complications when treated or are not amenable to endovascular treatment [2–4]. The treatment of wide-necked intracranial aneurysms has been substantially augmented with the advent of stent-assisted coiling, which not only can prevent coil prolapse in the parent artery and increase packing density but also may promote aneurysm thrombosis [5–9].
The first stent system designed specifically for neurovascular use, the Neuroform (Stryker), uses an open-cell design and has evolved through multiple generations since 2002 [10]. The Enterprise stent (Codman), which has a closed-cell design, represents the primary alternative to the Neuroform for the treatment of wide-necked intracranial aneurysms. Advantages of the closed-cell design include the ability of the stent to be partially deployed, recaptured, and redeployed and it may improve retaining coils within the aneurysm.
Although studies report comparable experience between Enterprise and Neuroform stent-assisted coiling, direct comparative data are limited. We have compared the technical and clinical complications of Enterprise stent deployments over a nearly 5-year period with an almost 9-year experience with the Neuroform using single-center prospective registry data.
| Subjects and Methods |   |
|---|
Institutional review board approval was obtained for the study, and informed consent was obtained for Enterprise and Neuroform stent placement (Food and Drug Administration [FDA] Humanitarian Device Exemption [HDE]). Patients undergoing attempted Enterprise and Neuroform stent-assisted coiling were enrolled in HIPAA-compliant prospective registries at our institution. For the Enterprise, stents were used for treatment of wide-necked, saccular, and fusiform aneurysms arising from a parent vessel with a diameter of ≥ 2.5 mm and ≤ 4 mm that were not amenable to surgical clipping (per HDE). For the Neuroform, stents were used for treatment of wide-necked saccular aneurysms arising from a parent vessel with a diameter of ≥ 2 mm and ≤ 4.5 mm (per HDE). Wide-necked aneurysms were defined as having a neck ≥ 4 mm or a dome-to-neck ratio < 2 mm. Stenting was not used simply to improve packing density.
Enterprise enrollment began in March 2007, and Neuroform enrollment (including all generations) began in February 2003. Data for the current study include all consecutive patients through December 2011. The recorded data include patient age, sex, presentation, previous aneurysm treatment, aneurysm location, stent size, deployment success, deployment difficulty (defined later), stent misplacement, stent movement, whether the coiling microcatheter was inserted into the aneurysm before stent deployment (“jailed”) or whether the procedure was staged, coiling outcome (degree of aneurysm occlusion) at the time of stent placement, and symptomatic and asymptomatic 30-day periprocedural complications. Follow-up angiograms typically obtained at 6 and 18 months were reviewed for data on stent patency and aneurysm recurrence.
A difficult placement was defined as one in which one of the following criteria were met: a coil pusher or some other device was necessary to deploy the stent; a small guidewire was placed alongside the stent delivery system within the guide catheter to provide increased support; the stent was placed in an alternative but workable location (waffle-cone technique [11] or stenting of an alternative branch vessel for treatment of a bifurcation aneurysm [e.g., M1 segment in lieu of A1 segment]); the guide catheter was exchanged; or, specific to the Enterprise, the stent was recaptured before deployment.
Procedures were performed by one of three neurointerventional staff members and a neurointerventional fellow. All patients started dual-antiplatelet therapy, typically consisting of aspirin 325 mg per day and clopidogrel (Plavix, Bristol-Myers Squibb/ Sanofi Pharmaceuticals Partnership) 75 mg per day at least 5 days before the procedure. Patients with acute subarachnoid hemorrhage were given loading doses of aspirin (650 mg) and clopidogrel (300, 450, or 600 mg) a few hours before stenting. This only applied to the small percentage of patients with ruptured aneurysms not routinely stent-coiled. Aspirin and clopidogrel were continued for 8 weeks after the procedure. Clopidogrel was then discontinued (unless also prescribed for another medical reason); aspirin was maintained indefinitely.
After induction of general anesthesia, a 6-French sheath was inserted in the right common femoral artery. Typically, stents were placed through a 6-French guide catheter (Envoy, Cordis Endovascular or Neuron, Penumbra) or a 6-French long sheath (Shuttle, Cook Medical) after routine diagnostic cerebral angiography using a high-resolution biplane angiographic unit (Neurostar or Axiom Artis, Siemens Healthcare). Systemic heparinization was initiated to maintain the activated clotting time between 250–350 seconds once the guide catheter was placed.
Enterprise stents were delivered through a Prowler Select Plus microcatheter (Codman) appropriately positioned over a 0.014-inch microwire under digital road map guidance. Neuroform delivery techniques have been reported separately [10]. Repeat angiography was performed after stent placement to verify appropriate stent positioning and to evaluate the parent vessel.
After the procedure, anticoagulation was typically reversed with IV protamine, and the arteriotomy site closed with a mechanical closure device. If there was concern for clot formation on the coils or stent (i.e., angiographic filling defect), anticoagulation was not reversed, and the femoral sheath was removed 6–12 hours later with hemostasis achieved by manual compression. Postprocedure monitoring was performed in a dedicated neurointensive care unit.
The recorded Enterprise and Neuroform data were compared using a two-tailed Fisher exact test with a p < 0.05 accepted for significance.
| Results | |
|---|
Between March 2007 and December 2011, Enterprise stent-assisted coil embolization was attempted in 98 patients (82 women and 16 men; mean age, 57 years; age range, 21–88 years). Similarly, between February 2003 and December 2011, Neuroform stent-assisted coil embolization was attempted for 160 patients (123 women and 37 men; mean age, 54 years; age range, 25–81 years). The patients presented as shown in Table 1. Because of the required antiplatelet regimen, ruptured aneurysms were not routinely stent-coiled and only comprised 4% of patients in both groups. Stented aneurysms were located as shown in Table 2.
Twenty-four aneurysms treated with Enterprise (23.5%) and 47 aneurysms treated with Neuroform (28.8%) were previously coiled (p > 0.05). Eight aneurysms were previously treated surgically in each cohort, seven clipped and one wrapped in the Enterprise group (one both clipped and coiled) and six clipped and two wrapped in the Neuroform group (three both clipped and coiled).
Enterprise stents, which all have an unconstrained diameter of 4.5 mm, come in four lengths delivered with the following frequencies: 14-mm stents (n = 0), 22-mm stents (n = 62), 28-mm stents (n = 36), and 37-mm stents (n = 9). The various generations of Neuroform stents were deployed as follows: Neuroform 1 (n = 8), Neuroform 2 (n = 53), Neuroform 2 Treo (n = 9), Neuroform 3 (n = 73), and Neuroform 3 EZ (n = 32), with sizes ranging from 2.5 × 15 mm to 4.5 × 30 mm.
Enterprise stent deployment success was high (108 of 115 attempts, 93.9%) with 102 aneurysms receiving a stent. The Prowler Select Plus microcatheter could not be positioned across the aneurysm neck in seven cases. Two stents were deployed with difficulty (one guide catheter change and one recapture before deployment). Six cases required a second stent for misplaced stents (n = 4), stent movement on coiling attempt (n = 1, occurred when selecting the aneurysm through the stent interstices), and downsizing because a longer stent could not be unsheathed (n = 1). The coiling microcatheter was inserted into the aneurysm before stent deployment in 70 of 115 cases (60.9%); no stent movement greater than 1 mm was observed in these cases. The jailing technique, which has been used since the beginning of our stent-coiling experience, was preferentially used for smaller aneurysms, generally less than 3–4 mm in diameter, considered a higher risk to catheterize through stent interstices.
Two procedures were staged to allow endothelialization of the stent in the setting of perceived stent instability, and the patients returned after 8 weeks for coiling in a delayed fashion through the stent. Three Enterprise stents were placed to tack down coil tails; one was placed to cover a stretched coil. Five were deployed after Neuroform placement failed, and five were deployed through a previously placed Neuroform that would not retain coils.
Most attempted Neuroform stents were deployed (173 of 214, 80.8%), but an inability to position the delivery system occurred with greater frequency. However, even with successful positioning, the Neuroform could not be successfully unsheathed in 12 cases, a problem not seen with the Enterprise. One Neuroform stent required complete withdrawal after partial deployment because of incorrect positioning. Four stents were not deployed because of movement of a previously placed stent.
There were a total of 34 difficult stent deployments with the Neuroform, usually because a coil pusher or some other device was necessary to deploy the stent (n = 21, not including those delivered using a planned stent-transfer technique). Nine deployments required a small guidewire placed alongside the stent delivery system to provide additional support. Three stents were placed in alternative but workable locations, two using the waffle-cone technique for basilar tip aneurysms and one stenting from the internal carotid artery into the M1 segment instead of the A1 segment for a carotid terminus aneurysm.
There were 12 misplaced Neuroform stents and nine cases of stent movement on a coiling attempt, none in the 37 of 214 cases (17.3%) when the coiling microcatheter was placed in the aneurysm before stent deployment. Forty-five procedures were staged to allow stent endothelialization, mostly early in the experience with only one staged procedure in the last 2 years of Neuroform use.
There were 12 periprocedural (30-day) symptomatic neurologic events, both thromboembolic and hemorrhagic. The events consisted of seven transient ischemic attacks (TIA), three ischemic strokes, and two intracranial hemorrhages. The thromboembolic events occurred in vascular territories distal to the treated vessel. In the cases in which the thromboembolic event was identified intraoperatively (angio-graphically), anticoagulation was not reversed as would have otherwise been done to enable hemostasis at the femoral arteriotomy site.
Three of the seven TIAs had an identifiable cause. One of the TIA patients presented with postprocedural confusion and dysarthria, which resolved over the next day. Intraoperatively, she had a small in-stent thrombus, which improved with intraarterial abciximab (ReoPro, Centocor/Eli Lilly). Another TIA was seen in a patient with intraoperative migration of a small coil mass from the aneurysm lumen to the space between the parent vessel wall and the Enterprise stent, resulting in 50% stenosis of the internal carotid artery; no infarct was seen on MRI. A third TIA presented a week after this patient's procedure and was attributed to noncompliance with antiplatelet therapy.
One ischemic stroke, which occurred in a patient with an incidentally discovered recurrence of a remotely clipped aneurysm, was the result of acute in-stent thrombosis resulting in left middle cerebral artery territory infarction requiring decompressive hemicraniectomy despite IV abciximab. The patient was discharged to a skilled nursing facility after 10 weeks of hospitalization with little neurologic improvement.
Another ischemic stroke occurred after stent-coiling of an incidentally discovered left internal carotid artery aneurysm. The stent was placed across the anterior choroidal artery origin. The patient developed right-sided weakness and aphasia in the recovery area. Emergent follow-up angiography showed no in-stent thrombosis and no filling defects to suggest a thromboembolic event; however, an infarct in the posterior limb of the left internal capsule was confirmed at MRI. With improved symptoms, the patient was ultimately discharged a week later to an inpatient rehabilitation center. Of note, to our knowledge, no patients with stents across the ophthalmic artery origin developed retinal emboli or changes in vision.
A third patient developed left-sided hemiplegia in the recovery area after treatment of an incidentally discovered aneurysm. As in the previous patient, emergent follow-up angiography showed no in-stent thrombosis and no filling defects; however, MRI showed border zone infarcts in the right cerebral hemisphere as well as an embolic-type infarct in the right caudate nucleus. The patient was noted to be severely hypotensive during the procedure. She was discharged to an in-patient rehabilitation center 4 days later, with resolution of her symptoms over the course of the next several weeks.
One of the hemorrhagic complications occurred during treatment of a giant, mostly thrombosed, anterior communicating artery aneurysm. The Enterprise was placed from the right A1 segment to the left A2 segment across the neck of the aneurysm after a failed attempt to place a Neuroform (also counted against the Neuroform cohort). At the end of the procedure, a focus of active extravasation was noted from the right A2 segment, resulting in subarachnoid and intraventricular hemorrhage. It was unclear whether this represented a wire perforation because the right A2 was never intentionally selected. Baseline left-sided weakness worsened to hemiplegia after the procedure. The patient was discharged to an inpatient rehabilitation center a week later.
The other hemorrhage presented 2 days later as an intraparenchymal frontal lobe hematoma ipsilateral to the stent, thought to be related to antiplatelet therapy. Initially, both antiplatelet agents were discontinued. The patient was discharged a week later with out-patient rehabilitation and aspirin therapy and had a full recovery.
Five intraoperative intracranial complications were asymptomatic. One was a case of acute in-stent thrombus that resulted in temporary occlusion of the ophthalmic artery and improved with IV abciximab. Another was a case of stent-assisted coiling of a basilar tip aneurysm, resulting in a detached coil dislodging from behind the stent and embolizing to a P3 posterior cerebral artery branch. One case of stent-assisted coiling of a left-superior hypophyseal aneurysm was complicated by the final coil being caught on the Enterprise stent and ultimately being detached partially within the internal carotid artery lumen. This slowed flow in the ipsilateral anterior cerebral artery. The patient was continued on a heparin drip until the next day and remained asymptomatic.
There were two small iatrogenic carotid cavernous fistulas that were both asymptomatic and closed at follow-up angiography. There was one retroperitoneal hematoma and one groin hematoma requiring blood trans-fusion. No thromboembolic or hemorrhagic complications were observed in the four patients treated in the face of acute subarachnoid hemorrhage. There were no procedural deaths.
Clinical complications with the Neuroform from February 2003 to December 2009 have been detailed separately [10]. During the 2 years from January 2010 to December 2011, there were three additional hemorrhages and one thromboembolic complication in the form of a TIA. All of these occurred during stent-assisted coiling of basilar tip aneurysms. One of the hemorrhages resulted in death.
The TIA was in a patient who underwent Neuroform 3 EZ deployment from the left P1 segment into the basilar artery. She awoke from anesthesia with transient right-sided weakness. MRI showed multiple punctate foci of restricted diffusion in the left posterior inferior cerebellar artery territory consistent with embolic phenomenon. She was discharged 2 days later, without any neurologic defects.
The first hemorrhage was a midbasilar wire perforation during Neuroform 3 deployment from the left P1 segment into the basilar artery using the stent-transfer technique. This was treated with reversal of systemic anticoagulation with protamine as well as mannitol and dexamethasone IV. CT of the head showed extensive subarachnoid and intraventricular hemorrhage, with hydrocephalus requiring a ventriculostomy. The patient developed a left sixth cranial nerve palsy that resolved over a few days and made a good neurologic recovery. However, she ultimately required a ventriculoperitoneal shunt, which was placed 1 month later when she presented with visual complaints, headaches, and altered mental status.
The second hemorrhage occurred while placing a Neuroform 3 EZ stent via the posterior communicating artery horizontally across both P1 segments, a technique that can enable basilar tip aneurysm coiling when the basilar-to-posterior communicating artery angles are acute [12]. The contralateral P1 was difficult to catheterize, and the superior cerebellar artery was entered with the wire and microcatheter several times. Despite no obvious procedural perforation or contrast extravasation, this patient had asymmetric pupils while emerging from anesthesia. Emergent CT showed diffuse subarachnoid hemorrhage; the patient did not recover and died 4 days later.
The third hemorrhage was a distal right posterior cerebral artery wire perforation during microwire exchange for Neuroform 2 deployment. CT showed a focal right occipital subarachnoid and parenchymal hemorrhage with extension into the right lateral ventricle, briefly requiring a ventriculostomy. Aside from a left-sided visual field defect, which improved over several months, she made a good neurologic recovery. Periprocedural symptomatic neurologic events with the Enterprise are compared with the Neuroform in Table 3.
Stent-assisted coiling was performed in 92 of 102 Enterprise-stented aneurysms. Twelve Enterprise-stented aneurysms required retreatment, and three aneurysms were initially incompletely coiled; however, they were all later coiled to near complete occlusion in a second session. Ten Enterprise-stented aneurysms did not receive any coils, seven for flow diversion because the aneurysm remnant was too small to coil (only one went on to occlusion at 19 months follow-up). As previously described, detached coils migrated out of the aneurysm in two patients. One aneurysm did not receive any coils because it could not be selected through the stent; the procedure was terminated when a small carotid cavernous fistula formed.
Neuroform coiling outcome details from 2003 through 2009 have been previously reported [10]. From January 2010 to December 2011, 46 additional aneurysms received a Neuroform stent. Of these, 41 were coiled to near complete occlusion (89.1%). Overall results are compared with Enterprise-stented aneurysms in Table 4.
The mean length of follow-up was 14 months for the 102 aneurysms treated with an Enterprise stent with at least 6 months of follow-up for 80 aneurysms. No vascular imaging follow-up was available for 18 aneurysms. Of the 84 aneurysms with follow-up, 80 were evaluated by catheter angiography and four were evaluated by MR angiography in lieu of catheter angiography.
There was one case of delayed thrombosis and parent artery occlusion at 7 months, which was asymptomatic. Notably, there was no underlying atherosclerosis in this patient. The in-stent stenosis rate was two of 108 deployed Enterprise stents (1.9%). Interestingly, both cases occurred in the same patient (contralateral aneurysms), one side with 50% stenosis at 14 months that decreased to minimal stenosis at 20 months, the other side with 30% stenosis at 17 months.
The rate of in-stent stenosis with the Neuroform stent was 11 of 173 stents (6.4%, p = 0.1412 compared with Enterprise). The stenosis with Neuroform was mild (< 50%) and completely resolved in five patients by a mean follow-up of 23 months (range, 8–50 months) and was mild (< 50%) with no follow-up in four patients. Two patients had 50% in-stent stenosis at 6 months, with no further follow-up.
| Discussion | |
|---|
Stent-assisted coil embolization of intracranial aneurysms has become a useful tool in the endovascular therapy of previously untreatable or difficult-to-treat wide-necked aneurysms. Not only does the stent serve as a scaffold for the placement of coils, which in turn enables dense aneurysm packing [8, 9], but it also may reduce the rate of aneurysm recanalization by the redirection of blood flow and alteration of flow dynamics within the aneurysm lumen [5–7] and by the promotion of endothelialization at the aneurysm neck [13].
The first self-expanding stent designed for intracranial use was the Neuroform. The Neuroform is a microcatheter-delivered nitinol stent with an open-cell design and low radial force. It has evolved through a series of five iterations, known as the Neuroform 1, 2, 2 Treo, 3, and 3 EZ resulting in improved deliverability [10].
The Enterprise is the first closed-cell stent designed to treat wide-necked intracranial aneurysms. Advantages of the design include the ability of the stent to be partially deployed, recaptured, and redeployed and that it may improve keeping coils within the aneurysm. In 2005, Higashida et al. [14] reported their initial experience with five patients, achieving 100% technical success without morbidity or mortality. Subsequently in 2007, the year FDA approval for the Enterprise as an HDE device was granted, Weber et al. [15] reported a series of 31 aneurysms in 30 patients, with successful deployment in all cases and no procedural morbidity; however, two patients experienced “possible or probable” device-related thromboembolic events during a 6-month follow-up period. A more recent Chinese series reviewed 50 aneurysms in 46 patients and found a 100% deployment success rate and a 2% procedural complication rate, which was a death related to microcatheter perforation during coiling after stent placement [16].
The largest series of Enterprise patients reported to date comes from the Interstate Collaboration of Enterprise Stent/Coiling, a 10-center registry [17, 18]. The study enrolled 141 patients who underwent 143 attempted stent deployments [17]. Procedural data showed 6% temporary morbidity, 2.8% permanent morbidity, and 2% mortality (0.8% mortality with an unruptured aneurysm, 12% with a ruptured aneurysm). After enrolling additional patients, the investigators reported midterm results in 213 patients with 219 aneurysms showing 6% delayed angiographic findings, with 3% having significant (> 50%) in-stent stenosis or occlusion [18]. There were seven (3%) delayed thrombotic events, all related to cessation of dual antiplatelet therapy.
Comparative data between the Enterprise and the Neuroform are limited. Izar et al. [19] found a higher stenting success rate with Enterprise (46 of 48, 96%) compared with Neuro-form (42 of 51, 82%). In five of the nine cases in which a Neuroform stent could not be navigated to the landing zone, an Enterprise stent was successfully deployed instead. Those authors reported no significant difference in the periprocedural complication rate, postprocedural hospital stay, packing density, recurrence rate, or number of in-stent stenoses.
In line with previously reported series, our technical success rate was high, 108 successful deployments of 115 attempts (93.9%). This compared favorably with the Neuro-form (173 of 214 attempts, 80.8%) and was statistically significant. Notably, five Enterprise stents were deployed after Neuroform placement failed, and five were deployed through a previously placed Neuroform that would not retain coils, enabling a total of 10 additional aneurysm coilings.
There was also a significant difference between the Enterprise and the Neuroform in the rate of difficult stent deployment, 1.7% and 15.9%, respectively. Although the deliverability of Neuroform did improve over four generations [10], the rate of difficult deployment per attempted stent with the Neuroform 3 alone (16 of 124, 11%) was still significantly higher than the Enterprise (two of 115, 1.8%, p = 0.0010). The coiling microcatheter was inserted into the aneurysm before stent deployment more frequently with the Enterprise, probably due to perceived difficulty in catheterization through the closed-cell design. Despite this, rates of stent movement and misplacement, although higher with the Neuroform, were not significantly different.
In contradiction to previously reported series, the clinical complication rates between the Enterprise and the Neuroform were dissimilar. Although the periprocedural hemorrhage rates were about the same, the periprocedural TIA rate and periprocedural stroke rate for the Enterprise were higher: 6.1% and 2.6% compared with 1.4% and 0% for the Neuroform, respectively. Taken together, the difference in the overall symptomatic peri-procedural (30-day) thromboembolic complication rate was highly significant, 8.7% for the Enterprise and 1.4% for the Neuroform (p = 0.0021).
The phenomenon may be related to the closed-cell design of the Enterprise. A post-procedural MRI study of 58 patients who had stent-assisted coil embolization of an intracranial aneurysm using Enterprise closed-cell stents or Neuroform open-cell stents was recently reported by Heller et al. [20]. They described a distinctive semilunar signal pattern on time-of-flight MR angiography, the crescent sign, representing flow outside the confines of the stent struts. This flow signal corresponds with incomplete stent wall apposition, as confirmed by high-resolution angiographic flat-panel CT. The crescent sign was seen in 18 of 33 Enterprise-treated cases and zero of 25 Neuroform-treated cases. The crescent sign was strongly predictive of ipsilateral post-procedural lesions on diffusion-weighted imaging (DWI) with an odds ratio of 18 (95% CI, 4.33–74.8; p < 0.0001). Fifteen of the 33 Enterprise patients (45%) had ipsilateral lesions on DWI; of these, 12 (80%) had a crescent sign as opposed to only six of 18 patients without lesions. Only one Neuroform patient had an ipsilateral lesion on DWI; this patient's procedure was preceded by a 30-minute carotid test balloon occlusion.
Although the crescent sign was strongly predictive, in the study Heller et al. [20], 20% of patients in the study by treated with the Enterprise who did not have a crescent sign still had an ischemic lesion on DWI, whereas only 4% of patients treated with the Neuroform, none of whom had a crescent sign, had such a lesion. One of their theories is that a small amount of incomplete stent-wall apposition may not have enough flow signal intensity to be detectable at 3 T. An alternative hypothesis, without supporting evidence, would be that the embolic phenomenon may be related to the Enterprise stent's polymer coating material (Parylene, Parylene Coatings) (the Neuroform is uncoated). Heller and Malek [21] also showed that incomplete stent-wall apposition is correlated with large parent vessel size, small radius of curvature, and large angle subtended by the stent. Delivery technique (i.e., micro-catheter pullback, delivery microwire push, or a combination of both) plays an important role in apposition as well [22]. The phenomenon is not correlated with stent length or microcatheter placement before deployment versus sequential coiling technique [21].
As described earlier, seven Enterprise stents in our series were deployed for flow diversion because the aneurysm remnant was too small to coil. Data on the theoretic effect of stenting on aneurysm hemodynamics are limited. A computational fluid dynamics study of basilar trunk aneurysms showed suppression of complex aneurysmal flow by stenting [23]. Stent placement lowered the flow velocity in the aneurysm and decreased aneurysmal wall shear stress; the effects were potentiated by the placement of additional stents. Furthermore, aneurysmal flow turnover time, a measure of stasis, was increased to 114–117% with one stent, 127–128% with two stents, and 141% with three stents. A recent series of six cases of blister aneurysms of the supraclinoid internal carotid artery treated using a stent-in-stent technique showed no residual or recurrent aneurysm in three patients [24]. The remaining three cases required retreatment with coils due to continued aneurysm growth. Of the seven Enterprise stents placed for flow diversion in our experience, only one showed complete thrombosis and remodeling on follow-up angiography at 19 months.
Immediate near-complete aneurysm occlusion, defined by occlusion of at least 90% of the aneurysm lumen at the time of stent placement, occurred more frequently with the Enterprise than with the Neuroform stents (87.3% vs 73.0%, p = 0.0058). This correlates with the theoretic advantage of improvement in keeping coils within the aneurysm lumen using a closed-cell design and may be related to increased packing density.
During a mean imaging follow-up of 14 months available for 84 of 101 aneurysms, there was one case of asymptomatic delayed parent artery occlusion at 7 months. Aside from this case, no in-stent stenosis greater than 50% was observed in the Enterprise cohort. For Neuroform, the rate of in-stent stenosis (11 of 173, 6.4%) was similar to the 5.8% rate in the largest reported series [25]. However, only six stents showed stenosis at the last available short-term follow-up angiogram, four of which were mild (< 50%) with two only 50%. It is likely that the phenomenon of resolving mild in-stent stenosis represents the normal response during the endothelialization process occurring in low-radial-force stenting of nonatheromatous vessels while the patient is undergoing prolonged therapy with aspirin and clopidogrel.
Limitations of the study include that the two cohorts of patients differ in time, operator experience, and antiplatelet testing (VerifyNOW, Accumetrics), which was used during the last 24 months of the study period and may have had an effect on patient selection or possibly resulted in alterations to the original antiplatelet regimen.
There is also likely selection bias in choosing the Neuroform versus the Enterprise for a given aneurysm. For example, Enterprise was used more frequently in the internal carotid artery than Neuroform (Table 2, p = 0.002), likely due to the single available diameter of the Enterprise (4.5 mm) matching the diameter of a larger caliber vessel. The closed-cell design of the Enterprise is also thought to have superior coil retention in a wide-necked aneurysm arising from a large vessel. On the other hand, anterior communicating arteries were almost exclusively treated with Neuroform (p = 0.0003), which is available is smaller unconstrained diameters.
| Conclusion | |
|---|
The Enterprise was easier to deploy than the Neuroform stent and enabled the treatment of additional aneurysms. However, compared with the Neuroform stents, there were more thromboembolic complications in the form of periprocedural TIA and stroke. Hemorrhagic complication rates were similar. On the basis of these findings, we prefer to use the Neuroform stent first and rely on the Enterprise stent as an easy-to-deliver backup for stent-assisted coil embolization.
C. J. Moran is a consultant for Codman, ev3/Covidien, and Stryker.
This work received the 2012 ARRS Residents/Fellows in Radiology Executive Council award.
| References | |
|---|
