Article Text
PDF
Abstract
Background Flow diversion with the Pipeline embolization device is a well-established method of intracranial aneurysm treatment. However, deployment of the first-generation device (Pipeline Classic) can be technically challenging. The Pipeline Flex contains the same flow-diverting stent with a modified delivery system.
Objective To compare procedural outcomes between the first-generation device (Pipeline Classic) and the Pipeline Flex.
Methods Thirty-eight of the first 40 consecutive patients who underwent intracranial aneurysm treatment with the Pipeline Flex and 58 of the most recent 60 consecutive patients who underwent treatment with the Pipeline Classic at our institution were evaluated. Patient demographics, aneurysm characteristics, technical procedural details, and early outcomes were analyzed.
Results The two groups were comparable for age, gender, and location of target aneurysms. Use of Pipeline Flex decreased procedure time by 44.2 min (p≤0.001) and fluoroscopy time by 22.0 min (p=0.001) compared with the Pipeline Classic. Similarly, radiation exposure was less in the Flex group with a mean difference of 3473.5 Gy cm2 (p=0.002), while contrast usage was decreased with a mean difference of 22.3 mL (p=0.007). These differences remained significant in multivariate regression analysis. Finally, the rate of device deployment failure was lower in the Flex group (7.1%) than in the Classic group (23.9%) (p=0.034).
Conclusions Use of Pipeline Flex significantly reduces the total procedure and fluoroscopy time, contrast usage, patient radiation exposure, and proportion of recaptured devices in comparison with the Pipeline Classic, probably owing to an enhanced delivery system that allows for more reliable and controlled deployment.
- Aneurysm
- Flow Diverter
Statistics from Altmetric.com
Introduction
Flow diversion with the Pipeline embolization device (PED; ev3/Covidien Neurovascular, Irvine, California, USA) has become an important addition to the armamentarium of endovascular techniques to treat intracranial aneurysms. Although initially approved for treatment of large or giant, wide-neck internal carotid artery aneurysms arising from the petrous to superior hypophyseal segments, the PED is increasingly used to treat smaller aneurysms, and lesions arising more distally in the cerebral vasculature.1 PED flow diversion provides a high rate of complete aneurysm occlusion with a low rate of major adverse events. However, deployment of the first-generation PED, Pipeline Classic, can be technically challenging, particularly in tortuous parent vessel anatomy. Some of the problems often encountered during Pipeline Classic placement include difficulty in freeing the distal end of the device from a constraining capture coil, limited pushability of the delivery wire, and inability to resheath the Pipeline after partial deployment.2 In rare instances, one or more of these factors may result in malpositioning or incomplete expansion of the flow diverter.
Consequently, the second-generation PED, or Pipeline Flex, contains a completely redesigned delivery system, while leaving the implantable device unchanged. The proximal portion of the flow diverter was placed on a resheathing pad to allow for recapture and repositioning of the device after partial deployment. Furthermore, the distal capture coil was replaced by two constraining Teflon sleeves, which allow for increased readiness of device opening and facilitation of stent resheathing by 180° rotation upon device recapture. Finally, the pusher wire was exchanged for a larger, more robust laser-cut hypotube to enhance pushability of the device during delivery.3
In theory, the changes in the Pipeline delivery system should facilitate device deployment, potentially improving procedural outcomes such as fluoroscopy usage and procedure times. Although the Pipeline Flex received the European CE mark of approval in March 2014 and FDA approval in February 2015, to date, only a few single cohort studies describing use of this device have been published.2 ,4 Therefore, we elected to retrospectively compare procedural outcomes of our most recent Pipeline Classic and Flex cases.
Methods
The protocol of this retrospective study was approved by the University of Maryland, Baltimore institutional review board (HP-00064678). From a prospectively maintained database, we retrospectively identified 39 cases from the first 40 consecutive intracranial aneurysm treatments performed with Pipeline Flex at our institution from April 2015 through December 2015 where flow diversion was the primary treatment intention. We subsequently identified the 60 most recent Pipeline Classic treatments performed from December 2014 through March 2015 using the same criteria. Information on patient demographics, aneurysm characteristics, procedural details, and immediate clinical outcomes was obtained for all qualifying patients.
Procedural information was obtained from multiple sources. Total fluoroscopy time and patient radiation exposure (estimated by dose–area product) were originally obtained from data generated by one of two biplane flat panel angiographic systems (Artis zee, Siemens, Erlangen, Germany). Patient demographics, clinical outcomes, and procedural details, such as contrast use, procedure time (interval between the start of patient sedation and sheath removal), number of attempted Pipeline deployments, use of a microwire or compliant balloon to facilitate device opening, and intraprocedural abciximab treatment, were acquired from procedure reports and electronic medical records. Details of aneurysm size as well as adequacy and location of PED deployment w obtained from procedure reports and review of source two-dimensional and rotational three-dimensional angiographic data (TM, EJL).
A preliminary analysis of 39 Pipeline Flex and 60 Pipeline Classic cases was performed, which identified one Pipeline Flex and two Pipeline Classic cases where the fluoroscopy time was greater than 3 SDs above the respective mean. In a subsequent detailed review of the electronic medical record and fluoroscopy images from these cases, the vast majority of the fluoroscopy and procedure time was spent catheterizing the parent artery vasculature, either owing to tortuous parent vessel anatomy and/or complex fusiform aneurysm morphology. As the primary factor leading to the longer fluoroscopy and procedure times was not Pipeline deployment, these cases were excluded from further analysis (Pipeline Flex n=38, Pipeline Classic n=58).
Endovascular procedure
Flow diversion with the PED is increasingly used in our practice for elective treatment of intracranial aneurysms involving the internal carotid artery and distal intracranial vertebral arteries. Flow diversion is also considered for aneurysms at or beyond the Circle of Willis when traditional coiling or stent-assisted coiling measures are ineffective or fail. Adjunctive coiling is generally reserved for aneurysms measuring ≥13 mm in diameter owing to prior published reports of delayed rupture of large and giant aneurysms following flow diversion.5 At our institution, flow diversion is generally not performed in the case of acute aneurysm rupture owing to the need for antiplatelet therapy.
All patients were treated by one of three neurointerventional radiologists (DG, GJ, TM) at a single medical center. One to 4 weeks before their procedure, patients began a dual antiplatelet therapy regimen consisting of 81–325 mg aspirin daily and clopidogrel. Clopidogrel dosing was titrated based on platelet inhibition testing using the P2Y12 assay (VerifyNow; Accumetrics, San Diego, California, USA), to achieve a goal P2Y12 reaction unit value between 60 and 200.
Flow diversion was performed under general endotracheal anesthesia. Access was obtained using a triaxial system, which was initiated by placement of a 6F shuttle sheath (Cook Medical, Bloomington, Indiana, USA) within the distal common carotid artery for treatment of anterior circulation aneurysms, or within the proximal subclavian artery for posterior circulation aneurysms. An intermediate catheter, typically a 6F 058 or 072 Navien distal intracranial support catheter (ev3/Covidien, Mansfield, Massachusetts, USA), was then navigated into the target vessel just proximal to the aneurysm. In rare cases with markedly tortuous parent vessel anatomy, a distal access catheter was used in lieu of an intermediate catheter. Through the intermediate catheter, either a Marksman (Stryker Neurovascular, Bayside Pkwy, Fremont, California, USA) or Excelsior XT-27 (ev3/Covidien) microcatheter was navigated across the neck of the target aneurysm(s) over a microwire. PED deployment then proceeded through the microcatheter using the standard technique.
Generally, placement of one PED was considered sufficient for aneurysm treatment. Placement of more than one PED was reserved for cases of incomplete aneurysm coverage following deployment of the first device or during treatment of wide-neck, large or giant aneurysms. In cases of inadequate device apposition with the parent vessel wall after deployment, the operator typically attempted to facilitate opening of the flow diverter by navigating a Synchro 2 or Transcend microwire (Stryker Neurovascular, Bayside Pkwy, Fremont, California, USA) with a ‘J’ tip through the device. If the PED failed to open adequately with this technique, angioplasty of the stent was then attempted using a compliant balloon, such as a Transform compliant balloon (Stryker Neurovascular). Delayed angiography in the parent vessel was generally performed 15–20 min after PED placement to monitor for early platelet aggregation. If evidence of platelet aggregation was present on delayed angiography (ie, filling defect in the PED or occlusion or stagnation of covered branch vessels or aneurysm(s)), half a loading dose of abciximab was administered. Serial angiograms were then obtained every 10–20 min until documented resolution of platelet aggregation occurred.
After the procedure, patients were closely monitored overnight in an intensive care unit and discharged on postprocedure day 1 or 2. All patients were evaluated in our neurointerventional clinic 2–3 weeks after treatment. Adjustment of clopidogrel dosing was continued based on platelet inhibition testing during the first few weeks after PED placement. Dual antiplatelet therapy was continued for at least 6 months after the procedure.
Statistical analysis
Data are presented as mean and range for continuous variables and as frequency for categorical variables. Univariate analysis was carried out using unpaired t tests, χ2 and Fisher's exact tests. Univariate linear regression analysis was used to test covariates predictive of dependent variables. Factors predictive in univariate analysis were then evaluated by multivariate linear regression analysis. p Values of ≤0.05 were considered statistically significant. Statistical analysis was performed using SPSS V.19.0 (IBM Corp, Armonk, New York, USA).
Results
There was no significant difference in patient age or gender between the two groups (table 1). A total of 70 aneurysms were treated in 58 procedures with the Pipeline Classic, while 46 aneurysms were treated in 38 procedures with the Pipeline Flex (mean of 1.2 aneurysms treated/procedure in each group). All aneurysms were treated electively, with the exception of a ruptured P2 aneurysm treated with the Pipeline Classic after coil embolization failed. Target aneurysm location was similar between the two groups (table 2), with 95.7% in the anterior circulation and 92.9% below the Circle of Willis in the Pipeline Classic group, compared with 93.5% and 95.7%, respectively, in the Pipeline Flex group. There was a small, but significant difference in the average largest aneurysm diameter (table 1); 5.6 mm in the Pipeline Classic group versus 4.3 mm in the Pipeline Flex group (p=0.043). There was no significant difference between the two groups in the rates of adjunctive coiling, intra-procedure abciximab administration, use of a J-tip microwire, or balloon angioplasty (table 4).
Baseline patient and aneurysm characteristics
Aneurysm locations
Use of Pipeline Flex was associated with significant reductions in total procedure time, fluoroscopy time, patient radiation exposure, contrast usage, and rate of deployment failure compared with Pipeline Classic (table 3). Average procedure time in the Classic group was 219.7 min vs 175.5 min in the Flex group, with a mean difference of 44.2 min (95% CI 20.5 to 67.9, p<0.001). The average fluoroscopy time of the Classic group was 69.8 min compared with 47.8 min in the Flex group, with a mean difference of 22.0 min (95% CI 9.3 to 34.7, p=0.001). Radiation exposure was also decreased with use of the newer device, with an average dose–area product of 13 252 Gy cm2 in the Classic group versus 9779 Gy cm2 in the Flex group, resulting in a mean difference of 3473.5 (95% CI 1310.6 to 5636.3, p=0.002). Furthermore, average contrast usage in the Classic group was 140.4 mL compared with 118.1 mL in the Flex group, with a mean difference of 22.3 mL (95% CI 6.4 to 38.3, p=0.007). Finally, there was a significantly higher rate of device deployment failure with the Pipeline classic (23.9%) compared with the Pipeline Flex (7.1%), with p= 0.034. There were two treatment failures (inadequate coverage of the target aneurysm) in the Pipeline Classic group (table 3). In one case, the operator was unable to safely deploy a Pipeline Classic during treatment of a paraclinoid internal carotid artery aneurysm. Treatment was successfully performed 6 months later with a Pipeline Classic device. In another instance, a Pipeline Classic device migrated into the distal middle cerebral artery M1 segment during attempted treatment of two paraclinoid internal carotid artery aneurysms. The patient was successfully treated by placement of a second Pipeline Classic device 6 months afterwards.
Primary procedural outcomes
There was no significant difference between the two groups in postprocedure neurologic deficits or permanent morbidity (table 4). In the Classic group, neurologic deficits were seen after 3/58 procedures (5.2%). One was transient, while the other two cases resulted in permanent morbidity (3.4%). In the Flex group, one transient neurological deficit was seen (2.7%), with no permanent morbidity. There were no deaths in either group.
Secondary and postprocedural outcomes
In univariate linear regression analysis, both fluoroscopy and procedure time were found to have multiple predictors in addition to PED type. Additional independent predictors of fluoroscopy time were presence of adjunctive coiling and aneurysm size (greatest diameter and large/giant status), whereas other independent predictors of procedure time were presence of adjunctive coiling, aneurysm size, and intraprocedure use of abciximab. Adjustment for these variables in multivariate analysis demonstrated that use of Pipeline Flex continued to be associated with significantly lower procedure and fluoroscopy time.
Discussion
Flow diversion, an important development in the management of intracranial aneurysms, has been established as a safe and effective treatment.6–8 However, published reports, and our own anecdotal experience, have highlighted the technical difficulties associated with the first-generation Pipeline delivery system.3 ,9 Frequently, the operator is required to perform aggressive maneuvers to achieve satisfactory deployment, including repeated torqueing and pushing of the delivery wire to free the distal end of the device from the constraining capture coil, ‘wagging’ the PED during deployment to ensure adequate wall apposition, as well as bumping the stent with a microcatheter if complete expansion does not occur after deployment. Although these techniques can be effective, they carry a risk of injury to the parent vessel or device and increase the overall complexity of the procedure. Not surprisingly, the complication rate associated with Pipeline Classic has been shown to be correlated with such maneuvers.10 ,11 Other procedural outcomes, such as fluoroscopy and total procedure times, might also be reasonably expected to be affected.
Procedural outcomes with Pipeline Classic may also be influenced by its inability to be resheathed after partial deployment. Proper placement of a PED can be challenging owing to foreshortening of the device during deployment, as well as proximal migration of the flow diverter in the setting of tortuous parent vessel anatomy. If a Pipeline Classic migrates too far proximally while being delivered, the operator is forced either to remove the device completely and reattempt placement with a new device or place a second PED over the first to ensure adequate aneurysm neck coverage. Either option is likely to increase the overall procedure length and complexity.
As previously described, the delivery system of Pipeline Flex has been designed to deal with the challenges posed by the first-generation device, allowing for increased readiness and reliability of device deployment, as well as the ability to reposition the device as needed. Our study is the first to confirm that the new delivery system does improve procedural outcomes compared with Pipeline Classic, probably owing to a combination of these factors. Specifically, we found significant reductions in fluoroscopy and iodinated contrast use, overall procedure length, and patient radiation exposure with Pipeline Flex. Though there were no significant differences in the incidence of immediate neurologic complications, long-term morbidity, or mortality between the two groups, this is probably due to the rarity of such events and the relatively small samples sizes of the study. Such occurrences are often due to thromboembolic events that occur during or immediately after neurovascular procedures and have been shown to correlate with the procedure length, technique used, and patient age.12 The ability to perform PED deployment with greater ease in a shorter amount of time with the Pipeline Flex may result in fewer complications with larger series of patients.
Although there was a significant difference in mean target aneurysm diameter between the two groups, Pipeline Flex use remained a significant, independent predictor of fluoroscopy time in multivariate analysis when aneurysm size (greatest diameter) and adjunctive coiling were taken into account. Pipeline Flex also remained an independent predictor of procedure time when these same factors as well as abciximab use were controlled for. Additionally, it should be noted that the control group consisted of our most recent Pipeline Classic cases, after we had gained considerable experience with the device. The difference in procedural outcomes would probably have been greater if we had compared our initial Pipeline Classic cases with these initial Pipeline Flex cases.
Furthermore, we demonstrated that use of Pipeline Flex results in a significantly reduced rate of deployment failure. This is probably primarily due to the ability to reposition the Pipeline Flex after partial deployment, which is not possible with the Pipeline Classic. In addition, it is possible that decreased manipulation of Pipeline Flex during deployment results in less chance of twisting or stretching of the device, which, in our experience, often damages the flow diverter, necessitating removal. There was also a difference in the number of treatment failures between the two groups, with two in the Classic group and none in the Flex group. However, owing to the rarity of treatment failures with the PED, it is not possible to draw any definitive conclusions given the size of the study.
Secondary procedures outcomes, including rate of J-microwire or balloon angioplasty to fully open a deployed flow diverter, as well as abciximab use, were also not significantly different between the two groups. However, this may be due to the relative rarity of these events and the smaller sample size. Abciximab usage was associated with significantly longer procedure times, which is not surprising given the time it takes for the drug to take full effect. Again, we continue to image the parent vessel until there is complete resolution of platelet aggregation (often over 20–40 min).
Previous reports demonstrated that flow diversion with Pipeline Classic resulted in lower fluoroscopy and procedure times than more traditional endovascular treatment methods such as stent-assisted coil embolization.1 ,13 Our study illustrates that the second-generation Pipeline device further differentiates flow diversion from these earlier techniques, providing support for more routine use of flow diversion in elective aneurysm repair. As the implantable flow diverter is identical between the two versions of Pipeline, long-term outcomes such as aneurysm closure rate, device and branch vessel patency, should be similar.
Limitations of this study include its retrospective design and relatively small sample size. Despite these limitations, Pipeline Flex still demonstrated a significant effect on procedural outcomes, even when confounding variables such as aneurysm size and adjunctive coiling were taken into account. Larger follow-up studies are necessary to examine the effect of Pipeline Flex on the rate of rarer procedure events, such as treatment failure or neurovascular complications.
Conclusion
Compared with the first-generation PED, Pipeline Classic, the second-generation Pipeline Flex is associated with shorter procedure and fluoroscopy times, reduced contrast use and patient radiation exposure, and a lower device deployment failure rate. These differences are probably attributable to an improved delivery system design that allows for enhanced device deployment and repositioning.
References
Footnotes
Contributors EJL contributed to the project design, monitored data collection and maintained the database, contributed to the statistical analysis plan, organized and analyzed the data, and drafted and revised the paper. TM initiated the project, contributed to the project design, monitored data collection and maintained the database, contributed to the statistical analysis plan, organized and analyzed the data, and drafted and revised the paper. YS contributed to the statistical analysis plan, organized and analyzed the data, and revised the paper. RS contributed to the collection of data and revised the paper. GJ contributed clinical data and revised the paper. DG contributed to the project design, provided clinical data, and revised the paper.
Competing interests None declared.
Ethics approval University of Maryland, Baltimore institutional review board.
Provenance and peer review Not commissioned; externally peer reviewed.