ABSTRACT

BACKGROUND:

The Pipeline embolization device (PED) is the latest technology available for intracranial aneurysm treatment.

OBJECTIVE:

To report early postmarket results with the PED.

METHODS:

This study was a prospective registry of patients treated with PEDs at 7 American neurosurgical centers subsequent to Food and Drug Administration approval of this device. Data collected included clinical presentation, aneurysm characteristics, treatment details, and periprocedural events. Follow-up data included degree of aneurysm occlusion and delayed (> 30 days after the procedure) complications.

RESULTS:

Sixty-two PED procedures were performed to treat 58 aneurysms in 56 patients. Thirty-seven of the aneurysms (64%) treated were located from the cavernous to the superior hypophyseal artery segment of the internal carotid artery; 22% were distal to that segment, and 14% were in the vertebrobasilar system. A total of 123 PEDs were deployed with an average of 2 implanted per aneurysm treated. Six devices were incompletely deployed; in these cases, rescue balloon angioplasty was required. Six periprocedural (during the procedure/within 30 days after the procedure) thromboembolic events occurred, of which 5 were in patients with vertebrobasilar aneurysms. There were 4 fatal postprocedural hemorrhages (from 2 giant basilar trunk and 2 large ophthalmic artery aneurysms). The major complication rate (permanent disability/death resulting from perioperative/delayed complication) was 8.5%. Among 19 patients with 3-month follow-up angiography, 68% (13 patients) had complete aneurysm occlusion. Two patients presented with delayed flow-limiting in-stent stenosis that was successfully treated with angioplasty.

CONCLUSION:

Unlike conventional coil embolization, aneurysm occlusion with PED is not immediate. Early complications include both thromboembolic and hemorrhagic events and appear to be significantly more frequent in association with treatment of vertebrobasilar aneurysms.

The Pipeline embolization device (PED; ev3/Covidien, Irvine, California), a flow diverter, is the latest technology available in the treatment of intracranial aneurysms. Flow diversion is based on the principle of endoluminal reconstruction by redirecting blood flow away from the aneurysm through the parent vessel, leading to stagnation of blood in the aneurysm and progressive thrombosis. At the same time, the device allows continual blood flow through sidewall perforators (given their demand for blood supply) and serves as a scaffold for neointimal growth across the aneurysm neck for vessel healing. Initial results with the PED were extremely promising,15 and, in April 2011, on the basis of the results of the Pipeline for Uncoilable or Failed Aneurysms (PUFS) trial, the Food and Drug Administration approved the PED for the treatment of large or giant wide-necked intracranial aneurysms from the petrous to the superior hypophyseal segments of the internal carotid artery (ICA).4 Clinical experience with the PED has since grown rapidly across the country. The purpose of this study was to analyze the preliminary postmarket clinical results after the PED was released to the American neurointerventional community and to assess whether the initial trial results can be generalized to real-world experience.

METHODS

Study Design

The study was a prospective registry of all patients with intracranial aneurysms treated with PEDs at 7 American neurosurgical centers subsequent to Food and Drug Administration approval of the PED. These patients were not enrolled in any other studies (eg, the international Complete Occlusion of Coilable Aneurysms using Pipeline Embolization Device trial).5 In general, the aneurysms treated were lesions considered to be unsuitable for or to carry a high risk of recurrence with standard coil embolization techniques (eg, wide-necked, large or giant, fusiform, and/or dissecting aneurysms). Data collected for each patient included clinical presentation, aneurysm characteristics (such as morphology), parent vessel measurements, treatment details, antiplatelet regimens (and responses if available), and periprocedural events (ie, events occurring during the PED procedure and up to 30 days after the procedure), including rescue procedures for intraprocedural technical events (such as balloon angioplasty for incomplete device deployment). Technical success was defined as both successful delivery of the device to the target vessel and successful deployment of the device. Follow-up data included degree of aneurysm occlusion (documented on a digital subtraction angiogram), delayed (> 30 days after the procedure) events (device migration, in-stent stenosis, and perforator occlusion), and clinical status. Local Institutional Review Board approvals were obtained for the collection and review of patient data for this study.

Pipeline Embolization Device

The PED is a braided mesh cylinder composed of 25% platinum and 75% nickel-cobalt chromium alloy that provides 30% to 35% target vessel coverage.3 It has low porosity (0.02-0.05 mm2) and radial force (approximately 2.0 mN/mm) compared with conventional intracranial stents. The nominal diameter of the device ranges from 2.5 to 5 mm (in 0.25-mm increments), and the nominal length available varies from 10 to 35 mm (in 2-mm increments for the 10- to 20-mm devices and 5-mm increments for the 20- to 35-mm devices). The diameter of the device can expand up to 0.25 mm beyond nominal diameter; and the device has a tendency to foreshorten during deployment, which makes optimal positioning of the device more difficult than with traditional intracranial stents. The PED comes preloaded on a 0.16-in insertion wire and is attached distally to a capture, which holds the distal stent. A radiopaque platinum lead wire (length, 15 mm) extends beyond the end of the device.

Patient Preparation, Procedure, and Follow-up

Patients were pretreated with a dual antiplatelet regimen. Typically, those scheduled for PED embolization on an elective basis were started on aspirin (325 mg/d orally) and clopidogrel (Plavix, 75 mg/d orally; Bristol-Myers Squibb/Sanofi Pharmaceuticals Partnership, Bridgewater, New Jersey, and Princeton, New Jersey), and those who underwent PED embolization on an emergent basis received a loading dose of aspirin (650 mg) and clopidogrel (600 mg). Platelet inhibition tests were optional. In general, response units of < 550 for aspirin and < 235 for Plavix are considered therapeutic levels (VerifyNow Assay System, Accumetrics, Inc, San Diego, California).

If the aspirin response is subtherapeutic, the patient is typically given a loading dose of 650 mg aspirin, and the response is checked in 4 to 6 hours. If the patient is still subtherapeutic, the daily dose is increased until a therapeutic response is obtained. If the Plavix response is subtherapeutic, a loading dose of 600 mg Plavix is administered, and the response is retested 4 to 6 hours later. If it is at goal, the daily regimen of Plavix (75 mg by mouth) is continued. If the Plavix response is not at goal after the loading dose has been given, an additional 600-mg loading dose of Plavix is given, and the response is checked in 4 to 6 hours. If it is at goal after the second loading dose, the patient is then placed on 75 mg Plavix, taken orally twice daily, and started on 1 g Lovaza (GlaxoSmith Kline, Research Triangle Park, North Carolina), which is taken orally on a daily basis, to enhance the Plavix response.6,7 If the Plavix response is not at goal after two 600-mg loading doses have been administered, ticlopidine or prasurgel is substituted for the Plavix.

Typically, patients remain on aspirin for life and clopidogrel (or ticlopidine or prasurgel) for at least 6 months. The duration of clopidogrel depends on the results of follow-up angiograms.

The PED embolization procedures were performed under either conscious sedation or general anesthesia with neuromonitoring at the discretion of the operator. With respect to intraprocedural anticoagulation, an activated coagulation time of > 250 seconds was typical. Intravenous heparin, or eptifibatide in rare cases, may be used in the periprocedural period to prevent or treat thromboembolic events (TEEs).

For each patient, an initial routine clinical follow-up evaluation was performed at 1 month. Subsequently, 3-, 6-, and 12-month clinical and catheter-based angiographic follow-up was planned. Aneurysm treatment was classified as either complete occlusion or residual, and residual was further divided into residual aneurysm neck vs residual aneurysm dome. In cases of fusiform aneurysms, aneurysm treatment was classified as either complete occlusion or residual only, with no further subdivision. The radiographic assessment was reported by the operator from each individual site. Follow-up computed tomographic and magnetic resonance (MR) imaging studies were obtained on an as-needed basis (eg, to assess the involution of mass effect) at the discretion of the treating physician.

Deployment Technique

In general, a Marksman microcatheter (ev3/Covidien) is navigated distal to the landing zone of the device (ie, optimal device position). After the PED is loaded in the Marksman catheter, deployment starts by a combination of unsheathing and advancement of the insertion wire. During deployment, close attention must be paid to the lead wire to ensure that it remains in the main vessel. Once the distal end of the device is deployed, it needs to be detached from the capture by torqueing the insertion wire in a clockwise fashion (with no more than 3 turns). Once the PED is detached from the capture, with the distal end opened, the whole system (ie, PED and microcatheter) can be pulled back together as a unit to reach the landing zone. At this point, further deployment is accomplished mainly by advancing the insertion wire with passive backward migration of the microcatheter. Throughout deployment, the PED can be “packed” (to increase vessel coverage) by advancing the microcatheter forward. This maneuver also helps to open the device. After full deployment, the Marksman microcatheter is brought back through the PED to resheath the distal capture and to ensure that the device is fully opened.

Number of PEDs Deployed and Adjunctive Coiling

The number of PEDs deployed was left to the discretion of the operator. In general, insufficient aneurysm neck coverage evident by an ongoing jet into the aneurysm or rapid clearance of contrast material from the aneurysm without stasis was the main indication for the deployment of additional PEDs (in a telescoping fashion) to achieve further neck coverage. In cases of large fusiform aneurysms, additional PEDs placed in a series were often necessary to achieve complete reconstruction of the parent vessel. Adjunctive coiling and any rescue procedure were also performed at the discretion of the operator.

RESULTS

At 7 participating US neurosurgery centers, 62 PED procedures were performed to treat 58 aneurysms in 56 patients between May 6, 2011, and January 26, 2012. The mean age of our cohort was 59 ± 12 years, and the majority of patients were women (47 of 56, 84%). Six of the patients in this report were included in previous articles.8,9

Most aneurysms treated (n = 37, 63%) were located on the ICA from the cavernous segment to the superior hypophyseal artery segment, in accordance with the PUFS trial entry criteria.4 The remaining 21 aneurysms were treated with off-label use of the PED: 13 aneurysms in the anterior circulation distal to the superior hypophyseal artery segment (23%) and 8 aneurysms in the vertebrobasilar system (14%).

Of the 58 aneurysms, 25 (43%) were discovered incidentally. The remaining lesions manifested with headaches only (3, 5.2%), mass effect (21, 36%), TEEs (5, 8.6%), and subarachnoid hemorrhage (4, 6.9%). Of the 4 patients with a history of subarachnoid hemorrhage from the target aneurysm, 2 were previously treated with coil embolization and the remaining 2 were deemed nontreatable by conventional surgical or endovascular methods. Regarding morphology and sizes, most aneurysms were saccular (52, 90%), although 4 of the 8 vertebrobasilar aneurysms treated were fusiform; 40 aneurysms (69%) were between 7 and 25 mm, 8 (14%) were giant (> 25 mm), and 10 (17%) were small (< 7 mm).

A total of 123 PEDs were deployed. All PEDs were successfully delivered to the target vessel. An average of 2 PEDs were used to treat each aneurysm (2.12 ± 1.3; range, 1-9). Fourteen aneurysms (25%) were treated with adjunctive detachable coils in addition to PED (3 previously coiled; 11 coiled during the PED procedure). Intraprocedural technical events requiring rescue procedures were uncommon, and none led to permanent sequelae: 6 of 123 (4.9%) PEDs placed were incompletely deployed, requiring balloon angioplasty to complete the deployment. Intraprocedural thrombus was noted in 1 device and was successfully treated with intra-arterial thrombolysis with eptifibatide. In 1 case, 2 overlapping PEDs became separated, necessitating the deployment of a third device for rescue.

In our cohort, the major complication rate (permanent disability or death resulting from perioperative or delayed complication) was 8.5%. Six periprocedural TEEs (transient ischemic attack or stroke) occurred, but only 1 resulted in permanent neurologic deficit (1.6%) from a brainstem infarct after PED treatment of a fusiform vertebrobasilar aneurysm. Five of the TEEs occurred in patients with vertebrobasilar aneurysms (including the aforementioned case), and 1 occurred in a patient with a giant dissecting middle cerebral artery aneurysm. There were 4 postoperative hemorrhages (6.9%) in the series, all of which were fatal. Two fatal hemorrhages occurred in patients with ruptured giant basilar trunk aneurysms after the procedure (1 occurring on postprocedural day 1 and 1 occurring 2 months after the procedure). The patient who suffered a delayed fatal hemorrhage underwent adjunctive coiling in addition to PED placement. The 2 other fatal hemorrhages occurred in patients with giant and large ophthalmic aneurysms (the giant aneurysm had recently ruptured and had been treated with coil embolization during a separate procedure before the PED procedure; the large aneurysm was unruptured). In both cases, the hemorrhage occurred soon after PED placement (on postprocedural days 5 and 6, respectively). However, in the case of the giant ophthalmic aneurysm that was previously treated with coiling, the postprocedural hemorrhage was intraparenchymal and was located in the distal vasculature in the same hemisphere as the aneurysm. Two patients, 1 with a giant cavernous aneurysm and 1 with a giant basilar trunk aneurysm, suffered new and worsened cranial neuropathies, respectively, in the early postprocedural period. The patient with the basilar trunk aneurysm had significant improvement of her lower cranial neuropathy after a course of high-dose steroid therapy. Retroperitoneal hemorrhage occurred in the patient with the new postoperative cranial neuropathy, necessitating a transfusion only.

Of the 19 patients with a 3-month follow-up angiogram, 13 patients (68%) showed complete aneurysm occlusion, whereas 5 had a neck remnant only and 2 had residual filling of the aneurysm. The average follow-up of these 19 patients was 3.4 months (range, 3-5 months). In the series, 2 patients (of 56) presented on a delayed basis with flow-limiting distal in-stent stenosis identified on follow-up angiograms; 1 was symptomatic at 2 months after the PED procedure, and 1 was asymptomatic at 3 months after the procedure. Both patients eventually underwent an uneventful balloon angioplasty with excellent radiographic results and complete resolution of symptoms in the symptomatic patient.

Illustrative Cases

Case 1, Postoperative Hemorrhage

This 50-year-old woman presented with vision loss in her eye and a large unruptured left ophthalmic segment aneurysm. She underwent uneventful treatment with 3 PEDs (4 × 18, 4.25 × 16, and 4.25 × 14 mm), with excellent parent vessel reconstruction (Figure 1A and 1B). On postoperative day 6, she presented with a fatal Hunt and Hess grade V, Fisher grade 4 subarachnoid hemorrhage (Figure 1C). The 3 other cases with postoperative hemorrhage were described in detail in previous publications.8,9

FIGURE 1.

Case 1. A, angiogram after deployment of the first Pipeline embolization device (PED; lateral view). B, angiogram after deployment of the third PED showing excellent parent vessel reconstruction (lateral view). C, postoperative noncontrast cranial computed tomographic scan 6 days later showing a Fisher grade 4 subarachnoid hemorrhage.

FIGURE 1.

Case 1. A, angiogram after deployment of the first Pipeline embolization device (PED; lateral view). B, angiogram after deployment of the third PED showing excellent parent vessel reconstruction (lateral view). C, postoperative noncontrast cranial computed tomographic scan 6 days later showing a Fisher grade 4 subarachnoid hemorrhage.

Case 2, TEE8

This 51-year-old man presented to an out-of-town institution with dizziness, right facial droop, slurred speech, right gaze difficulty, diplopia, and difficulty ambulating. He was found to have right pontine and thalamic strokes. He received intravenous thrombolytics and recovered, except for residual dysarthria. A large holobasilar aneurysm was found on angiography during that admission (Figure 2A). He was referred for treatment of the aneurysm within 2 weeks of his stroke. He underwent placement of 9 PEDs of various sizes (Figure 2B).

FIGURE 2.

Case 2. A, anteroposterior (AP) view of left vertebral artery injection demonstrating large vertebrobasilar fusiform aneurysm. B, posttreatment AP angiogram. Clockwise from top left: (C) magnetic resonance (MR) imaging demonstrating numerous areas of positive diffusion-weighted imaging changes in the bilateral occipital lobes, (D) bilateral cerebellum, (E) large thrombosed aneurysm compressing the brainstem, and (F) medial left occipital mild diffusion positivity. G, quantitative MR angiography demonstrating nearly symmetric cerebral blood flow in both posterior cerebral arteries despite the absence of flow detection in the basilar artery, presumably caused by artifact. H, final AP angiographic view (early filling) of the left vertebral artery injection. I, final AP angiographic view (later arterial filling) of the left vertebral artery injection. Reproduced from Siddiqui et al8 with permission from the publisher. © 2012 American Association of Neurological Surgeons.

FIGURE 2.

Case 2. A, anteroposterior (AP) view of left vertebral artery injection demonstrating large vertebrobasilar fusiform aneurysm. B, posttreatment AP angiogram. Clockwise from top left: (C) magnetic resonance (MR) imaging demonstrating numerous areas of positive diffusion-weighted imaging changes in the bilateral occipital lobes, (D) bilateral cerebellum, (E) large thrombosed aneurysm compressing the brainstem, and (F) medial left occipital mild diffusion positivity. G, quantitative MR angiography demonstrating nearly symmetric cerebral blood flow in both posterior cerebral arteries despite the absence of flow detection in the basilar artery, presumably caused by artifact. H, final AP angiographic view (early filling) of the left vertebral artery injection. I, final AP angiographic view (later arterial filling) of the left vertebral artery injection. Reproduced from Siddiqui et al8 with permission from the publisher. © 2012 American Association of Neurological Surgeons.

On the night of the procedure, the patient became nonverbal and developed right-sided hemiplegia and disconjugate gaze. MR imaging revealed multiple new occipital, cerebellar, and left midbrain infarcts (Figure 2C-2F). These strokes were likely a combination of emboli from the aneurysm, emboli from catheter or wire manipulation within a diseased vessel, and perforator infarct related to the coverage of the perforator-rich zone by the construct. He was dependent on mechanical ventilation for 2 weeks. Extubation was performed successfully, but the patient required percutaneous gastrostomy for feeding. At the last follow-up evaluation, which was 4 weeks after the procedure while the patient was still hospitalized, he was following commands with his left side and nodding to questions but remained nonverbal. His discharge angiogram showed a patent device construct, and a quantitative MR angiogram showed patent posterior cerebral arteries with artifact at the construct site (Figure 2G-2I).

Case 3, In-Stent Stenosis

This 54-year-old man presented with a 3-week history of headaches and gradual vision loss in his right eye. On examination, his visual acuity was normal, but a temporal field cut was found in his right eye. Cerebral angiogram revealed a fusiform aneurysm (12.8 mm in maximum diameter) on the right supraclinoid carotid artery (Figure 3A). The supraclinoid carotid artery was reconstructed with a 4.5 × 35-mm PED, along with adjunctive coiling (Figure 3B). The patient did well and was discharged on a dual antiplatelet regimen. He returned for his 3-month angiogram, which revealed excellent reconstruction of the supraclinoid carotid artery but also showed an 80% distal in-stent stenosis (Figure 3C). His vision was stable, but MR perfusion imaging showed delayed mean transit time, delayed time to peak, and increased cerebral blood volume, suggestive of a flow-limiting stenosis. The patient subsequently underwent an uneventful balloon angioplasty with an excellent radiographic result (Figure 3D).

FIGURE 3.

Case 3. A, initial angiogram injection showing a large fusiform aneurysm of the right supraclinoid carotid artery (left, anteroposterior [AP] view; right, lateral view). B, final angiographic runs showing significant stasis in the aneurysm (left, AP view; right, lateral view). C, 3-month follow-up angiogram showing significant distal in-stent stenosis (AP view). D, excellent results with balloon angioplasty (left, AP view; right, lateral view).

FIGURE 3.

Case 3. A, initial angiogram injection showing a large fusiform aneurysm of the right supraclinoid carotid artery (left, anteroposterior [AP] view; right, lateral view). B, final angiographic runs showing significant stasis in the aneurysm (left, AP view; right, lateral view). C, 3-month follow-up angiogram showing significant distal in-stent stenosis (AP view). D, excellent results with balloon angioplasty (left, AP view; right, lateral view).

DISCUSSION

The PED represents the latest technological innovation in the treatment of intracranial aneurysms. Initial results with this device were excellent. In the Pipeline for the Intracranial Treatment of Aneurysms (PITA) trial with 31 patients, follow-up angiography demonstrated a complete occlusion rate of 93.3% at 6 months.2 Only 2 major periprocedural strokes (6.5%) were reported. Szikora et al3 reported the Budapest experience with the PED in the treatment of wide-necked large or giant aneurysms. In that series of 18 patients, the complete occlusion rate was 94.4% at 6 months. One patient experienced device thrombosis, whereas another patient died of rupture of a separate aneurysm. Lylyk et al,1 in the Buenos Aires series of 63 aneurysms, reported a 95% complete occlusion rate at 12 months with no major complications. Most recently, in the PUFS trial, the complete occlusion rate at 180 days was 81.8% with a major complication rate of 5.6% among 107 patients treated. This result is certainly superior to traditional surgical and endovascular treatments for large and giant aneurysms.10,11 On the basis of the PUFS data, the PED gained Food and Drug Administration approval for the treatment of large or giant wide-necked intracranial aneurysms in the ICA from the petrous to the superior hypophyseal segments in April 2011.

Regarding aneurysm occlusion, our postmarket data thus far support the results from earlier studies. Lylyk et al1 reported 3-, 6-, and 12-month complete occlusion rates of 56%, 93%, and 95%, respectively. Among the patients with a 3-month follow-up angiogram in our series, the complete aneurysm occlusion rate was 68%. In this group of patients, 2 aneurysms had residual aneurysm filling, whereas the remaining aneurysms revealed only small neck remnants. Follow-up studies will tell whether these small remnants will go on to complete thrombosis and the complete occlusion rate of the remaining lesions.

As reported by other studies, delivery and deployment of the PED are feasible.14,12 Of the 123 PEDs deployed in our study, all were successfully delivered to the target vessel, and only 4.9% were incompletely deployed, requiring rescue balloon angioplasty with good results. The PUFS trial4 reported 99% successful delivery of the device, and Lylyk et al1 reported 97% success of initial deployment. More recently, Fischer et al,12 in a series of 101 intracranial aneurysms and dissections treated with PED, reported successful deployment of PED in 99% of cases with only 13 cases requiring rescue balloon angioplasty for complete device deployment.

In our cohort, 6 perioperative TEEs occurred, but only 1 (1.6%) resulted in permanent neurologic deficit. This is in keeping with the literature. In the PITA trial, the major perioperative stroke rate was 6.5%.2 In a recent series of 20 patients treated with PED, Lubicz et al13 reported 1 fatal infarct (5%) resulting from in-stent thrombosis. Five of the 6 patients with TEEs in our series harbored vertebrobasilar aneurysms, including the patient who suffered a permanent neurologic deficit from a periprocedural brainstem infarct. We believe perforator-rich zones in the posterior circulation likely account for differences in results of PED treatment in that circulation vs the anterior circulation. Limiting the number of devices used and adding periprocedural anticoagulation may reduce thromboembolic complications after treatment with PED in the posterior circulation. Nevertheless, until we better understand the effect of PED on the hemodynamics and perforators in the posterior circulation, we caution the off-label use of this technology to treat aneurysms of the vertebrobasilar system.

The most surprising finding in the study was the 4 fatal postprocedural hemorrhages in 2 patients with giant basilar trunk aneurysms and 2 patients with large and giant ophthalmic aneurysms. The true incidence of this dreaded complication is unknown but has been reported in isolated cases. Hampton et al14 reported a case of fatal giant ICA terminus aneurysm rupture 5 days after uneventful treatment with PED, and Fischer et al12 reported a case of fatal paraclinoid aneurysm rupture 3 days after treatment with PED and adjunctive coiling. In the same series, 3 additional patients were reported with postprocedural hemorrhage in the same hemisphere as the treated aneurysm.12 Of our 4 patients with postprocedural hemorrhage, 2 had adjunctive coils in the aneurysm in addition to the PED. On the basis of our experience and the literature presented above, adjunctive coiling clearly does not eliminate and does not even appear to reduce the risk of postprocedural aneurysmal hemorrhage.

With regard to the occurrence of hemorrhage in vasculature distal to the treated aneurysm, hemorrhagic transformation of a perioperative ischemic stroke seems the most likely explanation, especially because all patients were on a dual antiplatelet regimen. On the other hand, the mechanism of early and delayed aneurysm rupture after PED implantation remains more speculative. We believe that the postimplantation period is best characterized by a duel between 2 opposing yet concurrent events, namely thrombotic forces that induce inflammation in the surrounding aneurysm wall, thinning and eventually disintegrating it, and continued flow into the aneurysm (albeit at a much slower rate) that, if continued to the point when the aneurysm wall is no longer competent, will result in delayed aneurysm rupture. Early hemorrhage, on the other hand, may simply reflect the hemodynamic forces created because of an imbalance between inflow velocities and volumes as opposed to aneurysmal outflow velocities and volumes. We suspect there may be a difference between these 2 forces, resulting in acute expansion and early rupture of the aneurysm even before thrombosis is established. With respect to the SILK flow diverter (Balt Extrusion, Montmorency, France), Kulcsár et al15 speculated that the presence of a postimplantation inflow jet appears to suggest a high inflow velocity and volume and represents a risk factor for postprocedural hemorrhage.

CONCLUSION

Flow diversion with the PED is a new treatment for intracranial aneurysms. Compared with conventional techniques, aneurysm occlusion with the PED takes months to complete. Early complications include both thromboembolic and hemorrhagic events and appear to be significantly more frequent in conjunction with treatment of vertebrobasilar aneurysms in the perforator-rich posterior circulation. Long-term data are needed to establish long-term efficacy and to understand the delayed complications of this new technology.

REFERENCES

1.
Lylyk
P
,
Miranda
C
,
Ceratto
R
, et al
Curative endovascular reconstruction of cerebral aneurysms with the pipeline embolization device: the Buenos Aires experience
.
Neurosurgery
 .
2009
;
64
(
4
):
632
642
.
2.
Nelson
PK
,
Lylyk
P
,
Szikora
I
,
Wetzel
SG
,
Wanke
I
,
Fiorella
D
The pipeline embolization device for the intracranial treatment of aneurysms trial
.
AJNR Am J Neuroradiol
 .
2011
;
32
(
1
):
34
40
.
3.
Szikora
I
,
Berentei
Z
,
Kulcsar
Z
, et al
Treatment of intracranial aneurysms by functional reconstruction of the parent artery: the Budapest experience with the pipeline embolization device
.
AJNR Am J Neuroradiol
 .
2010
;
31
(
6
):
1139
1147
.
4.
U.S. Food and Drug Administration. Pipeline Embolization Device—P100018
:
Summary of Safety and Effectiveness Data (SSED)
 . http://www.accessdata.fda.gov/cdrh_docs/pdf10/P100018b.pdf. Accessed June 29, 2012.
5.
Wong
GK
,
Kwan
MC
,
Ng
RY
,
Yu
SC
,
Poon
WS
Flow diverters for treatment of intracranial aneurysms: current status and ongoing clinical trials
.
J Clin Neurosci
 .
2011
;
18
(
6
):
737
740
.
6.
Gajos
G
,
Rostoff
P
,
Undas
A
,
Piwowarska
W
Effects of polyunsaturated omega-3 fatty acids on responsiveness to dual antiplatelet therapy in patients undergoing percutaneous coronary intervention: the OMEGA-PCI (OMEGA-3 Fatty Acids After PCI to Modify Responsiveness to Dual Antiplatelet Therapy) study
.
J Am Coll Cardiol
 .
2010
;
55
(
16
):
1671
1678
.
7.
Gajos
G
,
Zalewski
J
,
Nessler
J
,
Zmudka
K
,
Undas
A
,
Piwowarska
W
Polyunsaturated omega-3 fatty acids improve responsiveness to clopidogrel after percutaneous coronary intervention in patients with cytochrome P450 2C19 loss-of-function polymorphism
.
Kardiol Pol
 .
2012
;
70
(
5
):
439
445
.
8.
Siddiqui
AH
,
Abla
AA
,
Kan
P
, et al
Panacea or problem: flow diverters in the treatment of symptomatic large or giant fusiformvertebrobasilar aneurysms
.
J Neurosurg
 .
2012
;
116
(
6
):
1258
1266
.
9.
Velat
GJ
,
Fargen
KM
,
Lawson
MF
,
Hoh
BL
,
Fiorella
D
,
Mocco
J
Delayed intraparenchymal hemorrhage following pipeline embolization device treatment for a giant recanalized ophthalmic aneurysm
.
J Neurointerv Surg
 .
2012
;
4
(
5
):
e24
.
10.
Gao
X
,
Liang
G
,
Li
Z
,
Wei
X
,
Cao
P
A single-centre experience and follow-up of patients with endovascular coiling of large and giant intracranial aneurysms with parent artery preservation
.
J Clin Neurosci
 .
2012
;
19
(
3
):
364
369
.
11.
Sughrue
ME
,
Saloner
D
,
Rayz
VL
,
Lawton
MT
Giant intracranial aneurysms: evolution of management in a contemporary surgical series
.
Neurosurgery
 .
2011
;
69
(
6
):
1261
1270
.
12.
Fischer
S
,
Vajda
Z
,
Aguilar Perez
M
, et al
Pipeline embolization device (PED) for neurovascular reconstruction: initial experience in the treatment of 101 intracranial aneurysms and dissections
.
Neuroradiology
 .
2012
;
54
(
4
):
369
382
.
13.
Lubicz
B
,
Collignon
L
,
Raphaeli
G
,
De Witte
O
Pipeline flow-diverter stent for endovascular treatment of intracranial aneurysms: preliminary experience in 20 patients with 27 aneurysms
.
World Neurosurg
 .
2011
;
76
(
1–2
):
114
119
.
14.
Hampton
T
,
Walsh
D
,
Tolias
C
,
Fiorella
D
Mural destabilization after aneurysm treatment with a flow-diverting device: a report of two cases
.
J Neurointerv Surg
 .
2011
;
3
(
2
):
167
171
.
15.
Kulcsár
Z
,
Houdart
E
,
Bonafé
A
, et al
Intra-aneurysmal thrombosis as a possible cause of delayed aneurysm rupture after flow-diversion treatment
.
AJNR Am J Neuroradiol
 .
2011
;
32
(
1
):
20
25
.

Acknowledgments

Patients participating in this study were from the following centers: Department of Neurosurgery, University at Buffalo; Department of Neurosciences, Stroke and Cerebrovascular Center of New Jersey, Capital Health; Departments of Neurosurgery and Radiology, University of South Florida; Department of Neurological Surgery, University of Florida; Mayfield Clinic, Department of Neurosurgery, University of Cincinnati; Departments of Neurosurgery and Neuroradiology, University of Texas Southwestern; Department of Neurosurgery, and University of Pittsburgh Medical Center. Thirty-eight procedures (61%) were performed at the former Millard Fillmore Gates Circle Hospital (a University at Buffalo affiliate), and 24 procedures were performed at the other participating sites (Capital Health, 4; University of South Florida, 10; University of Florida, 6; Mayfield Clinic, 2; University of Texas Southwestern, 1; University of Pittsburgh, 1). We thank Paul H. Dressel, BFA, for assistance with preparation of the illustrations and Debra J. Zimmer, AAS, CMA-A, for editorial assistance.

COMMENTS

This is an excellent summary of clinical and angiographic follow-up of a large number of patients using a Pipeline embolization device (PED) for the treatment of large aneurysms. In a recent review of complications associated with PED use, Fargen et al1 report an overall cumulative complication rate of 5.3% with a hemorrhagic complication rate of 1.6% in a case series of PED use in 374 patients. Delayed ipsilateral hemisphere hemorrhage or aneurismal rupture continues to be of significant concern with the use of PEDs. Potential mechanisms of such complications include inflammatory degradation of the aneurysmal wall while there continues to be intra-aneurysmal flow, decreased compliance within the stented segment of the vessel and thus higher pressure within the ipsilateral distal vessels, and formation of endoleak secondary to stent apposition to vascular wall. Longer follow-up in larger cohorts will lead to a better understanding of the safety and efficacy of this device.

Aditya S. Pandey

B. Gregory Thompson

Ann Arbor, Michigan

1.
Fargen K.M., et al. A review of reported complications associated with the Pipeline embolization device. World Neurosurg, 2012;77(3-4):403–404.

The Pipeline embolization device (PED) is a flow diverter approved by the Food and Drug Administration for the endovascular treatment of adults with large or giant wide-necked intracranial aneurysms in the internal carotid artery from the petrous to the superior hypophyseal segments. These authors analyzed the preliminary postmarket clinical results after the release of the PED to the American neurointerventional community. The study was a prospective registry of all patients with intracranial aneurysms treated with the PED at 7 neurosurgical centers in the United States. In general, these authors were reviewing aneurysms considered unsuitable for standard coil embolization techniques. Technical success was defined as both successful delivery of the PED and successful deployment of this device. Follow-up data included degree of aneurysm occlusion, delayed events, and clinical status.

There were 7 participating neurosurgery centers in the United States with 62 PED procedures performed to treat 58 aneurysms in 56 patients between May 6, 2011, and January 26, 2012. Eighty-four percent of the patients were women with a mean age of 59 ± 12 years. Sixty-three percent of the aneurysms were located on the internal carotid artery from the cavernous segment to the superior hypophyseal artery segment. Off-label use of the device resulted in the treatment of 13 aneurysms in the anterior circulation distal to the superior hypophyseal artery segment, which accounted for 23% of cases, and 8 aneurysms in the vertebral basilar system, which accounted for 14% of aneurysms; 6.9% of the patients presented with subarachnoid hemorrhage.

Ninety percent of the aneurysms were secular; 4 of the 8 vertebral basilar aneurysms treated were fusiform. Sixty-nine percent of the aneurysms were between 7 and 25 mm; 14% were > 25 mm; and 17% were < 7 mm. One hundred twenty-three PEDs were deployed for the treatment of 58 aneurysms. Finally, 25% of aneurysms were treated with adjunct detachable coils in addition to the PED.

These authors report a major complication rate of 8.5%. Five of the 6 procedural transient ischemic attack or strokes occurred in the treatment of vertebral basilar aneurysms, and one of these resulted in a permanent neurologic deficit resulting from a brainstem infarct. All 4 postoperative hemorrhages were fatal. Two of these fatal hemorrhages occurred in patients with ruptured giant basilar trunk aneurysms: 1 early after the procedure at day 1 and a second 2 months after the procedure. The other 2 were in the management of ophthalmic aneurysms. Two patients suffered worsening cranial neuropathies: 1 from a giant basilar trunk aneurysm and the other from a giant middle cerebral artery aneurysm.

Nineteen of the 56 patients had 3-month follow-up angiograms. Thirteen patients (68%) showed complete aneurysm occlusion; 5 patients showed neck remnant only; and 2 had residual filling of the aneurysm. Two of the 56 patients presented in a delayed fashion with flow-limiting distal in-stent stenosis identified on follow-up angiograms. One patient was symptomatic at 2 months, and 1 patient was symptomatic at 3 months after the procedure. Both patients underwent successful balloon angioplasty.

These authors present valuable additional information to the understanding and use of flow diversion devices in the treatment of aneurysms. These authors discuss a predominantly female cohort of patients with difficult-to-treat aneurysms. Fewer than half of their patients had 3-month follow-up angiograms; however, their occlusion rates seem to be similar to others in the literature. Their data also support the technical feasibility of deploying the PED with a high success rate. This article also further demonstrates the unique challenges of dealing with vertebral basilar aneurysms. The authors report more frequent thromboembolic and hemorrhagic events related to the treatment of fusiform vertebral basilar aneurysms; thus, caution must be used when these aneurysms are managed. I agree with the authors’ conclusion that long-term efficacy and understanding of the late complications require further long-term data. The data presented help further our understanding of the use of the PED.

As technology advances and practitioners have more tools at their disposal for the treatment of aneurysms, the decision-making paradigms become more complex. Additionally, considerations are taken when deciding what devices to use in the ruptured aneurysm setting vs the unruptured aneurysm setting. Treating ruptured aneurysms with devices that require dual antiplatelets makes their periprocedural course more complicated. Because it takes months to see complete occlusion in patients treated with the PED, there is a shift in the way practitioners treating aneurysms need to view a successful treatment. This also raises the question, How many stents need to be placed across the aneurysm to have a successful treatment? Finally, the timing and duration of the dual antiplatelet therapy need to be studied further as more information about long-term outcomes is gained. This device is yet another tool to consider when an aneurysm is encountered in clinical practice.

Dan Surdell

Omaha, Nebraska

ABBREVIATIONS

    ABBREVIATIONS
  • ICA

    internal carotid artery

  • PED

    Pipeline embolization device

  • PITA

    Pipeline for the Intracranial Treatment of Aneurysms

  • PUFS

    Pipeline for Uncoilable or Failed Aneurysms

  • TEE

    thromboembolic event