Article Text
PDF
Abstract
Flow-diverting stents, including the Pipeline embolization device (PED) and Silk, have been beneficial in the treatment of aneurysms previously unable to be approached via endovascular techniques. Recurrent aneurysms for which stent-assisted embolization has failed are a therapeutic challenge, given the existing intraluminal construct with continued blood flow into the aneurysm. We report our experience using flow-diverting stents in the repair of 25 aneurysms for which stent-assisted embolization had failed. Nineteen (76%) of these aneurysms at the 12-month follow-up showed improved Raymond class occlusion, with 38% being completely occluded, and all aneurysms demonstrated decreased filling. One patient developed a moderate permanent neurologic deficit. Appropriate stent sizing, proximal and distal construct coverage, and preventing flow diverter deployment between the previously deployed stent struts are important considerations to ensure wall apposition and prevention of endoleak. Flow diverters are shown to be a reasonable option for treating previously stented recurrent cerebral aneurysms.
- Aneurysm
- Angiography
- Flow Diverter
- Intervention
Statistics from Altmetric.com
Introduction
Endovascular management of cerebral aneurysms has seen an unprecedented advancement since the publications of the International Subarachnoid Aneurysm Trial study in 2005, which documented the effectiveness of such treatment for cerebral aneurysms.1 Stent-assisted coiling (SAC) has enabled the treatment of a large number of wide-neck and fusiform cerebral aneurysms, which were previously considered unsuitable for endovascular treatment. The initial experience with SAC was with flexible, first-generation nitinol self-expanding stents (NSES), which improve the durability of endovascular therapy for wide-neck aneurysms. NSES provided safe and durable treatment for complex aneurysms equivalent to that obtained with coiling alone for simple aneurysms.2 For more challenging aneurysms that recur after initial SAC occlusion, options for re-treatment include clipping, vessel sacrifice, use of liquid embolic agent, further re-coiling, and/or overlapping stent placement.3 Each of these approaches can be technically difficult and of uncertain efficacy.
In 2011, the Pipeline embolization device (PED) (Covidien/ev3, USA) was approved in the USA as a primary treatment for large and giant wide-necked aneurysms involving the internal carotid artery from the petrous to the superior hypophyseal segments. Flow-diverting stents, such as PED and Silk (Balt, France), disrupt blood flow into the aneurysm at the neck, inducing thrombosis in the aneurysm sac, while preserving physiological flow in the parent vessel and adjacent branches. The introduction of flow-diverting stents has significantly increased the occlusion rates for treated wide-neck and large aneurysms. Use of flow-diverting stents as a rescue strategy for treatment failure after SAC with NSES has not been well documented. Although there have been reports of its success under such circumstances, there are various technical nuances associated with the use of flow diverters in these cases, including appropriate apposition over previous stents and anatomical results. We report our experience using flow-diverting stents in the repair of aneurysms for which stent-assisted embolization had failed.
Methods
After obtaining institutional review board approval, a retrospective review was conducted of all patients for whom initial aneurysm treatment with SAC or stent and liquid embolic agent (Onyx 500; Covidien, USA) had failed and secondarily, underwent repair with flow diversion, using either PED or Silk stents. The use of flow diverters for this indication is off-label. This treatment was judged by the senior author to be the best alternative as measured by the risk and potential long-term durability. Twenty-three of these aneurysms had previously undergone SAC, one had been stented after the use of Onyx and one stented after both Onyx and coiling. All stents were NSES with a metal-to-artery surface ratio of 6%. The stent most commonly used alone was Neuroform (Stryker, Fremont, California, USA) (n=10) followed by Enterprise (Codman, Raynham, Massachusetts, USA) (n=2), and Leo+ (Balt, France) (n=1). The other 12 cases had more than one of these devices overlapping (multi-stent construct). All interventions included in the study were the first aneurysm re-treatment, with the exception of one second re-treatment, one third re-treatment, and one fourth re-treatment. Patient demographics are presented in table 1.
Patient demographics
Twenty-one of the aneurysms were saccular with a wide neck (>4 mm) and four were fusiform. Of the saccular aneurysms, four were giant (>25 mm), 14 large (11–25 mm), two moderate (5–10 mm), and one small (<5 mm). Nine of these patients had aneurysms that had previously ruptured. The mean time between NSES placement and flow-diverting stent placement was 19 months (range 2–55 months).
Patients received pretreatment with 325 mg aspirin and 75 mg clopidogrel to achieve P2Y12 reaction units <250 and aspirin reaction units >350. Additionally, intraoperative anticoagulation was achieved with heparin to achieve an activated clotting time >250 s. Platelet reactivity was analyzed using the VerifyNow System (Accumetrics, San Diego, California, USA). Each procedure was performed with either a 6F or 8F Neuron guide catheter (Penumbra, Alameda, California, USA). Intracranial access was accomplished with a Marksman microcatheter (Covidien, USA) and Synchro 0.014″ microwire (Stryker). PED constructs were placed to completely cover the previous NSES constructs proximally and distally, with attention also given to obtaining appropriate vessel wall apposition.
Results
This series included 25 patients (18 women, 7 men) with a mean age of 51 years (range 29–68 years). Aneurysms were successfully accessed and a PED or Silk stent was successfully deployed in all cases except for one PED case. A large fusiform aneurysm of the basilar artery could not be treated, because during stent deployment, the PED became caught in the struts of the existing stent, which prevented proper deployment.
At the 12-month follow-up, angiography showed Raymond class I, II, and III occlusion in nine, seven, and three patients, respectively, with asymptomatic ipsilateral carotid occlusion in an additional two patients (table 2 and figure 1). Three patients were recently treated and have not yet undergone follow-up imaging. Of the three cases who had Raymond class III aneurysm occlusion, the giant right cavernous fusiform aneurysm demonstrated that the PED was not completely apposed proximally against the vessel wall, resulting in an endoleak (figure 2). This patient had required balloon angioplasty of a narrowed PED segment at the time of initial re-treatment. The other two were found to have decreased, but persistent aneurysm filling.
Results of aneurysm re-treatment with flow diversion at the 12-month follow-up
Digital subtracted cerebral angiograms of (A, B) a giant, right cavernous internal carotid artery (ICA) fusiform aneurysm with two previously deployed overlapping Neuroform stents and coils; (C, D) a left ophthalmic artery giant saccular aneurysm with previously deployed Enterprise stent and Onyx; (E, F) a right cavernous ICA large fusiform aneurysm with two previously deployed overlapping Neuroform stents and coils. (A, C, E) Preoperative angiograms; (B, D, F) 6-month follow-up angiograms.
Native view cerebral angiograms and DynaCT of a deployed Pipeline embolization device with focal narrowed segment (arrows) in a residual giant, right cavernous internal carotid artery fusiform aneurysm with two previously deployed overlapping Neuroform stents and coils.
Clinical complications
One patient with a right ophthalmic aneurysm developed a carotid-cavernous fistula and an oculomotor nerve palsy within 24 h of PED deployment. This was immediately embolized with liquid embolic agent and the patient's symptoms resolved.
Over 2 months after PED placement, a patient who was non-compliant with clopidogrel use, with a large left posterior communicating artery aneurysm (12 mm×19 mm), developed mild right-sided weakness and numbness, which resolved after restarting an appropriate antiplatelet agent regimen.
A patient with a giant right posterior communicating artery aneurysm (30 mm×33 mm) developed a right temporal intracerebral hemorrhage 2 days after his procedure. The hematoma did not require evacuation and angiography revealed no aneurysm recurrence with only a very small neck remnant. The patient initially had a dense left-sided hemiparesis, which improved significantly after rehabilitation, and the patient was thereafter independent in all activities of daily living.
Technical complications
Three technical complications occurred during the procedure. The first was failure to deploy the flow diverter owing to the stent catching on the struts of the previously deployed stents, which led to an abortion of the procedure. The other two complications were difficulties with navigating the microcatheter around previously deployed stents, but both these were eventually successfully stented; one of these cases was aborted, but successfully completed the following day. None of these technical complications resulted in any transient or permanent clinical neurologic deficits.
Procedure length
For those patients who underwent PED placement, procedures took, on average, 104 min, ranging from 47 to 266 min. Two endovascular interventions that were aborted took 22 and 59 min. Of the three procedures with technical complications described previously, the procedure times were 154, 266, and 59 min. The average procedure time for procedures that were without complications or abortions was 78 min, ranging from 47 to 116 min. Data were not collected for those patients who underwent Silk placement.
Discussion
Since the use of endovascular treatment of aneurysms became widespread, technological progress in this field has increased considerably, enabling treatment of highly complex intracranial aneurysms with minimal morbidity and mortality.4–8 However, endovascular treatment may not be successful in all cases, with a small percentage of patients experiencing failure of treatment, recurrences, or complications.3 ,9–13 Patients for whom previous coiling, SAC or NSES alone has failed are technically challenging owing to difficulty in accessing the aneurysm through previously deployed hardware. Depending on the length of time from the initial treatment, there may be partial endothelialization of the coil mass or stent, making microcatheter access challenging. One of the largest series published to date on the risks of aneurysm re-treatment showed that the overall complication rate was lower than for the initial treatment.14 However, of 311 patients from eight institutions, the 14 reported complications included three deaths and five permanent disabilities from subarachnoid hemorrhage and ischemic stroke. Re-treatment without having to access the previously treated aneurysm may be a safer alternative. This may be more significant in aneurysms with wide necks or large size owing to their known higher recurrence rate with endovascular treatment. As a result, PED may be an effective tool in the management of complicated aneurysms for which stent-assisted embolization or NSES alone has failed. In this series, flow diversion would have been our first choice for giant and large aneurysms had this technology been available at the time of the initial treatment, and therefore, in large and giant aneurysms for which SAC has failed, we offered the best endovascular treatment available. The smaller aneurysms for which SAC initial treatment failed were deemed to have a higher risk from recoiling than from placing a flow diverter.
There have been previous studies that included aneurysms for which an initial endovascular treatment failed and which were re-treated with PED.15–17 Although previous stenting was an exclusion criterion of the PUFS trial, the durability of flow diversion for treatment of complex aneurysms was demonstrated in that study by the 74% effective treatment rate and 5.6% major stroke or mortality rate.15 Although this was not the study's primary focus, Nelson et al18 reported 12 aneurysms re-treated with PED resulting in occlusion rates of ∼90% at 6-month follow-up. Fischer et al19 published a series that included 30 lesions for which treatment with NSES had failed and which were re-treated with PED. Occlusion rates in re-treatment cases at 6 months when PED had been deployed within another previously placed stent were similar to those obtained when a PED was deployed in an unstented vessel: 65% and 69%, respectively. However, the incidence of adverse events was higher (13%) with a previous stent than without (2%). The authors felt that neo-endothelialization of the new stent might be delayed, or in some cases prevented, when this process has already taken place over a previous stent. These cases may therefore require longer periods of antiplatelet therapy.19 We demonstrate in this series a complete aneurysm occlusion rate of 38% and significant decrease in aneurysmal flow in all cases; with an acceptable complication rate, including only one major stroke (4%) and no deaths, in these patients with complex, recurrent pathologies with limited salvage options.
Lylyk et al5 reported seven aneurysms in which a PED was placed through a previously deployed stent. Re-treatment failed in only one of these lesions—a giant aneurysm—at the 12-month follow-up. This demonstrates the efficacy with which PED can occlude aneurysms for which initial treatment has failed, but also highlights a potential obstacle. Deploying PED within a previous stent may impair stent-to-vessel wall apposition. This may lead to a leak between the PED and vessel wall, resulting in persistent aneurysm filling, also known as an endoleak. We feel this underscores the important technical considerations of creating a flow-diverting construct that is long enough to adequately cover the previous stent construct at least proximally and also distally, when possible, allowing the PED to seat properly against native vessel wall. Appropriate flow-diverting stent sizing to the vessel is crucial to avoid endoleak. Post-flow diverter angioplasty with a compliant balloon may be necessary to obtain the best apposition of the implant to the vessel wall. In one of our cases, we could not achieve complete apposition despite correct PED sizing and subsequent angioplasty (figure 2). After further analysis, it was found that in this case the flow diverter was not well apposed on a focal point, probably because during access of the previously deployed NSES, the PED microcatheter and wire did not navigate smoothly. The distal access with the flow diverter delivery microcatheter was probably through the cells of the previously deployed NSES. Care must be taken to prevent microcatheter advancement and flow-diverting stent deployment through the cell of the previously deployed stent, as this will inhibit stent wall apposition. Use of a J-tip wire can be an advantageous technique, as it will prevent wire entry through the cells of previous stents. Also, every attempt must be made to make a smooth, single pass of the wire and the microcatheter, as the presence of any resistance, especially on a curvature, is a sign of probable malpositioning. When this is recognized, the wire and/or catheter should be retracted and another trajectory for the wire should be attempted. Other methods, such as the use of a balloon catheter to facilitate a passage through the true lumen of the stent may be a plausible technique if the balloon navigates through this often challenging anatomy, but this was not attempted in our series. Previously deployed closed-cell stents implanted around tight curves present the biggest challenge to re-access, and should be approached with additional caution. Preprocedure or intraprocedural CT did not prove to be helpful for assessing stent apposition in our experience.
Limitations to this study include, but are not limited to, selection bias, lack of an adequate control group, and non-blinded review of angiograms. Referral patterns might have led to a preselected group of patients with aneurysms more favorable to endovascular treatment. Because aneurysms for which previous endovascular treatment has failed may be treated by open clipping in some cases, aneurysms referred to the senior author might be particularly amenable to flow diversion. A control group with matching characteristics is not feasible, and historical control group comparison is limited by differences in patient selection and study protocol.
We feel the use of flow-diverting stents to treat aneurysms for which initial treatment with SAC or NSES has failed is safe and effective, provided certain nuances are kept in mind. Most of the aneurysms treated here were large or giant and have high complication and recurrence rates after initial treatment with endovascular therapy. Flow diverters provided a reasonable option for re-treating these lesions. We share our experience with use of a flow diverter after SAC failure and some technical tips acquired during our experience.
References
Footnotes
Contributors The authors, as noted by their initials, made the following contributions to the paper as outlined by the ICMJE guidelines: conception or design: DKL, RM, MEF; data acquisition, revision of the work, final approval, agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved: DKL, RM, MEF, DMH, JTB, MKK, AKJ, KMK; analysis, or interpretation of data: DKL, RM, MEF, DMH, JTB, MKK, AKJ; drafting the work: DKL, DMH, JTB, MKK, AKJ.
Competing interests DKL has financial and research relationships with Penumbra, ev3, and Stryker. AKJ is cofounder of Gaudi Vascular, Inc. DMH, JTB, MKK, KMK, MEF, and RM have no financial disclosures.
Ethics approval Rush University institutional review board.
Provenance and peer review Not commissioned; externally peer reviewed.