Article Text
PDF
Abstract
Background The ‘blister-type’ aneurysm is one of the most devastating cerebrovascular lesions. Flow diversion with stent reconstruction is an emerging treatment and has shown promising initial results.
Objective To evaluate the experience of one institution using stent reconstruction for pseudoaneurysms of the supraclinoid internal carotid artery and to compare with a review of the literature.
Methods A retrospective review from one institution identified eight patients with ‘blister’ aneurysms over a 47-month period. The Raymond scale was used to classify the aneurysms. Clinical data were obtained using the modified Rankin Scale (mRS) and the National Institute of Health Stroke Scale. A literature review was performed and compared with our results. Clinical and angiographic data were obtained.
Results After treatment, two aneurysms were Raymond class 1 (25%) and six were class 3 (75%). Of the class 3 aneurysms, two required retreatment, three (50%) progressed to complete occlusion and three (50%) had persistent aneurysm filling. Clinical data revealed two patients with mRS score of 0 (25%), five with mRS score of 1 (62.5%) and one with mRS score of 2 (12.5%). From the literature review, residual filling was evident in nine patients (64.3%) and complete occlusion in four (28.6%). On follow-up angiography, nine (64.3%) were occluded, two (14.3%) had residual neck filling and one (7.1%) had persistent aneurysm filling. Thirteen patients (92.9%) had an mRS score of 2 or better. Combining the available experience, patients demonstrated either improvement (n=9, 41%) or stability (n=11, 50%). Only two (9%) had progression requiring retreatment.
Conclusions Endovascular stent remodeling of ‘blister-type’ aneurysms is a safe and effective strategy.
- Stent
Statistics from Altmetric.com
Introduction
The term ‘blister-type’ aneurysm, or pseudoaneurysm, refers to a subset of aneurysms that are fundamentally different from the typical intracranial ‘berry’ aneurysm. These pseudoaneurysms are often found arising from the supraclinoid internal carotid artery (ICA) and are frequently associated with subarachnoid hemorrhage (SAH). These aneurysms have a malignant natural history and are difficult to treat both surgically and via endovascular means when the parent vessel must be preserved. Open microsurgery has been associated with high rates of intraoperative rupture, and clip-wrapping techniques have been associated with high postoperative growth and re-hemorrhage rates.1–5 Often the parent vessel must be sacrificed to achieve durable aneurysmal exclusion.
When treated via endovascular means, several techniques have been described including parent vessel sacrifice, unassisted and stent-supported coil embolization,6–21 with varying degrees of success. In many cases these lesions involve a long segment of parent vessel and either lack a saccular component or the saccular component is too small to accommodate coils. Alternatively, treatment of these aneurysms using a reconstructive rather than a deconstructive strategy by stenting alone has been shown to have some early success.10 ,11 ,18 ,22 ,23 Stent placement across the involved segment of the vessel alters the hemodynamic stress applied to the aneurysm by causing flow diversion and acts as a scaffold to promote endothelialization across the neck. However, concerns surrounding hemorrhagic complications secondary to long-term antiplatelet use in patients treated with this strategy have tempered enthusiasm for this technique, particularly since these aneurysms most commonly present in the setting of SAH.
The current series summarizes the available experience in treating pseudoaneurysms arising from the supraclinoid ICA using stenting as monotherapy. We evaluate the mid- to long-term angiographic and clinical outcomes by combining the experience at a single high-volume tertiary care center with the available literature. The aim is to determine if this strategy is a feasible option for the treatment of these difficult lesions.
Methods
Retrospective review
Prospective databases are maintained at our institution for all patients undergoing intracranial stent implantation. These databases were searched retrospectively to identify patients with supraclinoid ICA aneurysms of the ‘blister-type’ who underwent treatment with one or more stents, without the introduction of coils. The medical records, conventional angiographic and non-invasive imaging data, and procedural reports of the patients were reviewed.
Patients
Of all the intracranial aneurysms treated during a 47-month period, eight were identified as being ‘blister-type’ aneurysms in the supraclinoid ICA that were treated with the previously described stent monotherapy technique10 using either a Neuroform (Stryker, Natick, Massachusetts, USA) or Enterprise (Cordis Neurovascular, Miami Lakes, Florida, USA) intracranial stent. Six of the aneurysms were treated in the acute rupture setting, one was treated in the subacute rupture setting (4 weeks post-hemorrhage) and one was treated in the unruptured setting.
Technique
The placement of the intracranial stents was performed using previously described techniques.10 ,24 ,25 All procedures were performed with the patient under general anesthesia with paralytic agents. The patients received pharmacologic platelet inhibition, although specific strategies, dosing and route of administration varied between patients (table 1). The adequacy of platelet inhibition was assessed after the procedure with platelet aggregrometry. The initial treatment angiograms included high magnification views which demonstrated the aneurysms to the best advantage (diagnostic working angle). This view sometimes differed from that used for the positioning and deployment of the stent (stenting working angle).
Summary of the demographic, aneurysmal, procedure characteristics and follow-up of patients undergoing stent monotherapy for blister-type aneurysms of the internal carotid artery (ICA)
All endovascular procedures described in the current series were performed via transfemoral arterial access. Full heparinization (activated clotting time >250 s) was instituted before the introduction of a guiding catheter (6 or 7 Fr Envoy; Cordis Neurovascular) or sheath (6 Fr KSAW Shuttle Select; Cook) into the targeted cervical vessel. Access across the lesion was achieved with a microcatheter (SL-10, Stryker; Echelon-14, Covidien) over a 0.014 inch microwire (Synchro-14, Stryker; Transcend Soft Tip, Stryker). The microcatheter was then exchanged over a 300 cm, 0.014 inch Transcend Floppy or Choice Floppy (Stryker) microwire for a Neuroform (Stryker) or Enterprise (Cordis Neurovascular) stent delivery system. After deployment of the stent, the delivery system was removed. Additional stent delivery system(s) were introduced over the in situ exchange wire and additional stent(s) were deployed, overlapping previously placed stent(s). After stent deployment, control angiography was performed to assess the integrity of the parent and daughter branch vessels as well as the stagnation of flow within the target lesion.
The most commonly employed strategy was to deploy two overlapping stents (n=7). Three stents were used in another. In no patient was a single stent deemed to be satisfactory based on serial angiography. Four Neuroform and 12 Enterprise stents were deployed in total.
Follow-up angiography
All follow-up cerebral angiograms included high magnification views that were aligned using the orientation of the stent markers with respect to osseous landmarks to reproduce the initial diagnostic working angles. When feasible, a ‘down the barrel’ view was obtained along the long axis of the stent. Aneurysms were assessed based on the Raymond classification scheme. Given that these aneurysms are typically broad-based with no significant saccular component, the aneurysms were classified as either Raymond class 1 (complete resolution) or class 3 (residual aneurysm).26
Clinical follow-up
Patients were evaluated and classified using the modified Rankin Scale (mRS) and the National Institute of Health Stroke Scale (NIHSS) with information available from the most recent clinical follow-up.
Literature review
An extensive online literature search was performed to identify other studies reporting results of using similar treatment strategies for these lesions. Online databases available through our institution including Pubmed and Medline were searched using keywords ‘blister aneurysm’, ‘stent’, ‘pseudoaneurysm’ and ‘flow diversion’.
Results
Patients
The blister aneurysms identified at our institution occurred in five women and three men aged 62.1±12.2 years (range 50–84). The majority were treated in the acute rupture setting. A delay in diagnosis due to inadequate angiography at another facility led to one patient being treated in the subacute rupture setting (4 weeks post-rupture). Four patients required ventriculostomy placement at the time of presentation (prior to endovascular treatment).
Initial procedural complications
There was one intraprocedural complication encountered in the treatment of the patients in this series. One patient required emergency thrombolysis with 23 mg intra-arterial abciximab followed by loading with aspirin and clopidogrel for in-stent thrombosis and occlusion. Of note, this patient had had a total of three stents placed across his aneurysm and was on therapeutic dual antiplatelet therapy when the thrombosis occurred. The thrombosis resolved and the patient did not suffer any permanent sequelae.
There were no aneurysmal re-bleeds or strokes during the initial treatment and there were no deaths associated with the treatment.
Angiographic results and durability
Immediately after treatment, two aneurysms were Raymond class 1 (25%) and six were class 3 (75%).26
Follow-up angiography was performed in seven of the eight patients (87.5%) in this series at 10.9±11.1 months after the initial treatment (range 1–28 months). Due to advanced age and evidence of stent patency on MR, one patient declined follow-up angiography.
Of the two patients with complete occlusion on initial treatment, one underwent follow-up angiography and was found to have persistent occlusion of his aneurysm (figure 1). Among the six patients with residual aneurysm filling, two (33%) required retreatment (figure 2).
Cerebral angiography in a 61–65-year-old woman with a ruptured left internal carotid artery dorsal wall aneurysm. (A,B) Initial angiography showing a 3 mm blister aneurysm arising just distal to the posterior communicating artery. Two Neuroform stents were placed across the aneurysm. (C) Initial post-stenting angiography showing complete occlusion of the aneurysm. (D,E) Follow-up angiography performed 10 months after stenting showing continued complete occlusion of the aneurysm.
Cerebral angiography from initial angiogram (A) and 1 month after stent placement (B) in a 51–55-year-old woman with a ruptured right internal carotid artery dorsal wall blister aneurysm. Follow-up angiography showing progression of the lesion despite stent therapy. This patient ultimately required parent vessel occlusion after STA-MCA bypass.
Of the six patients with an initial Raymond class 3 aneurysm, three (50%) were found to have complete occlusion on follow-up imaging at 18.7±13.7 months post-stenting (figure 3). Residual aneurysm filling was seen in the other 50% at follow-up at 3.3±2.5 months. Ultimately, four of the eight aneurysms (50%) were completely occluded on follow-up angiography.
Initial cerebral angiography in a 61–65-year-old woman showing a right internal carotid artery dorsal wall blister aneurysm (A) before and (B) after placement of one Neuroform and one Enterprise stent showing initial residual aneurysm filling. (C) Follow-up angiography performed 28 months after stent placement showing no residual aneurysm filling and widely patent stent complex.
There were no re-ruptures, andnone of the patients have had symptoms or imaging consistent with post-treatment aneurysmal rupture or SAH.
Complications related to platelet inhibition
There were no ventriculostomy-related hemorrhages in any patient requiring an external ventricular drain (EVD) for hydrocephalus (n=4). None of these patients required permanent cerebrospinal fluid diversion. There were no groin hematomas.
Mid- and long-term functional outcomes
Clinical follow-up data were available for all patients at 12.4±10.5 months (range 1–34). There were two patients with an mRS score of 0 (25%), five with an mRS score of 1 (62.5%) and one patient with an mRS score of 2 (12.5%) at the latest clinical follow-up. None of the patients have died. Of the six patients who presented in the acutely ruptured setting, two (33.3%) had an mRS score of 0, three (50%) had an mRS score of 1 and one patient (16.7%) had an mRS score of 2. The patients who presented in the unruptured and subacute ruptured settings both had an mRS score of 1 at their most recent follow-up.
Of the six patients with acutely ruptured aneurysms, four (66.7%) had a NIHSS score of 0, one (16.7%) had a NIHSS score of 1 and one (16.7%) had a NIHSS score of 2. The patient who presented with subacute rupture had a NIHSS score of 0 and the patient with the unruptured aneurysm had a NIHSS score of 1 at the most recent clinical follow-up appointment.
Literature review
An extensive literature search was performed and multiple series were identified that had included patients treated with similar stent monotherapy techniques.10 ,11 ,18 ,19 ,22 ,23 Only studies reporting supraclinoid ICA aneurysms were included. All the studies were published between 2006 and 2012. Individual patient data were collected from publicly available sources and combined to facilitate comparison with the results of this present series. A total of 14 additional patients were identified from six reports (table 2).
Summary of the demographic, aneurysmal, procedural characteristics and follow-up data of all patients included in the meta-analysis
All patients were treated in either the acute (n=11, 78.6%) or subacute (2–6 weeks post-rupture, n=3, 21.4%) rupture setting. Hunt and Hess and Fisher grades were available for 11 of the 14 patients. There was no information regarding the need for EVD placement or EVD-related hemorrhages for any patient in the literature review. Various antiplatelet and anticoagulant regimens were used during treatment of these patients, which are summarized in table 2.
All patients were followed clinically and angiographically for varying lengths of time. The angiographic outcomes were analyzed using the Raymond classification scale and clinical outcomes were reported using the mRS. Of the additional 14 patients, one was treated with stenting after re-rupture during surgical clipping in the setting of SAH.10 A single Neuroform stent was deposited across the diseased segment of the vessel to reinforce the muslin gauze wrap placed during surgery. The aneurysm remained secured after this treatment and angiography at 4 months demonstrated complete resolution.
On initial angiography, residual filling was evident in nine patients (64.3%) and complete occlusion was found in four (28.6%). One patient had no documented initial classification but, due to the description of an enlarging pseudoaneurysm after surgical wrapping, a classification of residual aneurysm (class 3) can be assumed. On follow-up angiography (7.25±3.7 months), nine of 14 (64.3%) were completely occluded, two (14.3%) had residual aneurysm neck filling and one (7.1%) had persistent aneurysm filling. Two patients required retreatment with parent vessel occlusion due to persistent aneurysm growth. Of the nine patients who were initially class 3, six (66.7%) of them spontaneously improved to class 1, two (22.2%) required retreatment with parent vessel occlusion and one (11.1%) remained stable.
Clinical follow-up data were recorded for all the patients in the literature review. The recording of the mRS score differed between the studies and is summarized in table 2. Of the 14 patients, only one (7.1%) had an mRS score of >2; the remaining 13 patients (92.9%) had an mRS score of 2 or better.
Combining all the available experience, the majority of patients demonstrated either angiographic improvement (as defined by Raymond classification; n=9, 41%) or stability (n=11, 50%). Only two (9%) demonstrated progression (as defined by aneurysm growth) requiring retreatment (figure 4).
Graphic representation of angiographic results at initial presentation and then at follow-up showing interval progression to aneurysm occlusion in a significant portion of the aneurysms that initially had residual filling after stent placement. Data combined from retrospective analysis and literature revision.
Comparing our data with that of the literature review, it is noteworthy that our results were similar to those found in other studies. With regard to angiographic results, our dataset demonstrated angiographic improvement in 50% of patients compared with 66.7% in the literature review. In addition, clinical data from our dataset also accurately reflected those of the other studies in terms of ‘good’ functional outcome (mRS ≤2; 100% in our dataset, 92.9% in the literature review).
Discussion
‘Blister-type’ pseudoaneurysms have a poor prognosis if not diagnosed correctly and promptly treated. These aneurysms have high rates of re-bleeding before treatment as well as high rates of rupture during surgical clipping.1–5 27–29 The increasing awareness of the potential benefits provided by intracranial stents as flow diverters has led to their use as a reconstructive strategy to address these challenging aneurysms.
Traditional treatment options for intracranial ‘blister-type’ aneurysms
Historically, the most definitive treatment for these lesions was parent vessel occlusion, when possible. The endovascular technique for vessel occlusion was initially described in patients presenting with dissecting fusiform aneurysms of the vertebral artery and it was shown to provide immediate protection from re-hemorrhage.30–35 In many instances, however, parent vessel sacrifice is not possible and more reconstructive treatment strategies must be employed. Surgical clipping of these aneurysms may not be an option because they commonly lack a definitive saccular component and the parent vessel wall is often friable and involves a long vessel segment. These qualities tend to result in high rates of intraoperative rupture.1 ,36
Endovascular treatment using a traditional approach of coil embolization (with or without adjunctive stenting) is difficult because a saccular component often does not exist or is too small to accommodate the introduction of coils safely.4 ,10 ,15 ,18 ,19 ,33 ,37 ,38 In fact, the placement of coils into the saccular component of one of these lesions is a potentially dangerous maneuver and may cause perforation and re-hemorrhage. It has been theorized that the small saccular component may represent a contained rupture as opposed to a true aneurysmal sac segment of a diseased vessel wall.
Stenting for aneurysm treatment
Initially, stents were incorporated in the setting of stent-assisted coil introduction into saccular aneurysms13 ,39 and became more common after the development of the Neuroform stent.14 ,25 More recently it has been shown that stent deposition across the aneurysm neck leads to significant flow diversion away from the aneurysm while redirecting it along the parent vessel. It has been shown that the flow dynamics of the aneurysms are altered by the placement of these stents, and there is a significantly decreased amount of intra-aneurysmal flow and shear stress on the aneurysm wall after placement of the stents.9 Although the immediate effects of flow diversion theoretically provide some immediate protection from recurrent hemorrhage, stent endothelialization with intravascular remodeling accounts for the long-term durability of the endovascular remodeling strategy.40 As such, endovascular stent reconstruction as monotherapy has been shown to be a practical and effective treatment option for cerebral aneurysms at various locations.10–12 ,18 ,19 ,22 ,23 ,41 ,42
Since 1997, stents have been successfully used to treat fusiform aneurysms of the basilar artery.15 ,38 ,39 ,43 They were subsequently used as monotherapy for ruptured intracranial aneurysms with varying degrees of success.6 ,7 ,17 ,43 ,44 In many cases the aneurysms that were stented were shown to progress to complete occlusion on follow-up angiography. Poor results and complications that have arisen have included lack of thrombosis of the aneurysm, progressive aneurysm expansion and in-stent thrombosis, all requiring further treatment.45–47 Additionally, due to the requirement of platelet inhibition in patients undergoing stent placement, its utility as a viable treatment tool in the setting of SAH has been questioned.
Implications for stenting as monotherapy for the treatment of ICA ‘blister-type’ aneurysms
The safety and efficacy of this treatment strategy has previously been reported for pseudoaneurysms located throughout the cerebrovascular system.10 We chose to focus on pseudoaneurysms occurring along the supraclinoid ICA to critically evaluate a relatively homogeneous patient population. The current series and subsequent literature review involves a small number of patients (n=22); however, it is larger than any existing series of ICA ‘blister-type’ aneurysms treated with any treatment strategy and is significant, considering their relative rarity.
This evaluation has highlighted a few important considerations. First, stenting as stand-alone therapy is effective in acutely securing a ruptured aneurysm at this location. Most of the aneurysms treated were ruptured and there were no re-ruptures following treatment. This compares very favorably with the surgical experience of these lesions. Secondly, the results are durable. Although most aneurysms demonstrated residual filling at initial treatment, many went on to complete occlusion or showed no growth on follow-up studies. Third, no reports of EVD-related hemorrhagic complications are described among this group of patients. Lastly, the clinical outcomes are favorable, with most of the patients achieving an mRS score of 0 or 1 and all achieving an NIHSS score of ≤2 in our series. Taken together, these findings strongly suggest that endovascular stent reconstruction is an effective strategy for the treatment of these lesions,10 ,11 ,18 ,19 ,22 ,23 ,48–,50 even in the setting of rupture.
Limitations
Some important limitations exist in this study including the small overall sample size, the importance and variability of the experience of the operator, the lack of standardization on technical procedural details and the different time intervals in clinical and angiographic follow-up among the patients. These factors could all potentially lead to some bias in the results of this study. Further follow-up of these cases will be beneficial to our understanding of their pathology and response to flow diversion.
Conclusions
Endovascular stent remodeling of ‘blister-type’ aneurysms as a monotherapy has shown some promise as a treatment strategy when a reconstructive therapy is the only feasible option. Given the acceptable clinical outcomes seen in our experience and the literature review, it may be reasonable to consider this as a treatment strategy for these aneurysms when vessel sacrifice is not an option.
References
Footnotes
-
Contributors KMW: data collection, data interpretation, drafting manuscript, critically reviewing manuscript. FKH: conception and design, drafting manuscript, critically reviewing manuscript. SIM: drafting manuscript, critically reviewing manuscript. AMS: conception and design, data collection, data interpretation, drafting manuscript, critically reviewing manuscript.
-
Competing interests None.
-
Ethics approval Ethics approval was obtained from the Cleveland Clinic.
-
Provenance and peer review Not commissioned; externally peer reviewed.