Characteristics and endovascular treatment of intracranial vertebral artery aneurysms
Objective: The clinical and angiographic characteristics of vertebral artery (VA) aneurysm were evaluated to demonstrate the safety and efficacy of endovascular techniques of VA aneurysms. Materials and Methods: Case records of 38 consecutive patients with 40 VA aneurysms admitted during a 2-year period were reviewed. The data analyzed included age, sex, size of aneurysm, ruptured or unruptured, endovascular techniques, angiographic results after embolization, duration of follow-up, angiographic follow-up results and Glasgow Outcome Score at follow-up. Results: Of the 38 patients, 33 patients had 35 dissecting aneurysms and five patients had five saccular aneurysms. Seventeen (42.5%) aneurysms were ruptured. Of the 34 patients treated with endovascular techniques, immediate post-procedural angiograms showed complete and subtotal occlusion (>90%) of 27 (67.5%) aneurysms and incomplete and no occlusion of 13 (32.5%) aneurysms, including four conservatively treated aneurysms. A clinical improvement or stable outcome was achieved in all the patients (100%) during a mean 12.1-month follow-up. There was no complication related to endovascular treatment and no rebleeding during the follow-up period. Angiographic follow-up (mean of 7.2 months, range 1-18 months) was available in all the patients. Complete and subtotal occlusion was observed in 31 (81.6%) patients, including one spontaneous thrombosis of a conservatively treated VA dissecting aneurysm. Recanalization in two patients (5.9%) at 6 and 9 months did not require retreatment. Conclusion: This series demonstrates the safety and efficacy of multimodality of endovascular techniques for VA aneurysms.
Keywords: Aneurysm, endovascular management, vertebral artery
Vertebral artery (VA) aneurysms are of two types: saccular and dissecting aneurysms. Strategies of VA aneurysm management remain unestablished and are individualized for each patient on the basis of clinical aspects and angiographic finding. ,, Conventional surgical repair is technically demanding with high morbidity and mortality. ,,, The continuous advances in endovascular techniques have provided multimodalities in the management of VA aneurysms and established the safety and feasibility of endovascular treatment. ,,,,,,, But, the number of subjects studied in these studies is small, and most do not include characteristics of VA aneurysms and unruptured VA aneurysms.
This study was a retrospective review of 38 consecutive patients referred for evaluation and treatment of VA aneurysms between 2008 and 2010. Only aneurysms involving VA above C2 and without involving the basilar trunk were included. Patients were divided into ruptured (17) and unruptured (21), and patients with unruptured VA aneurysm presented with cerebral ischemia, mass effect or discovered incidentally after imaging for unrelated pathology. Diagnosis of subarachnoid hemorrhage (SAH) was based on computed tomography (CT) findings. The clinical status of the patients was graded according to the Hunt and Hess grading (HandH) system:  seven patients were in grade I, eight in grade II and two in grade III. All patients received triple-H therapy (after embolization) and selective calcium channel blockers (nicardipine). External ventricular drainage was used for the management of acute hydrocephalus with gradual weaning over a period in one patient. Diagnostic cerebral digital subtraction angiography was performed in all patients. Diagnosis of dissecting aneurysm is classically made by the angiographic appearance, with a double lumen, a focal vessel wall irregularity, a pre-aneurysmal narrowing and an aneurysmal dilatation. On magnetic resonance imaging (MRI), intramural thrombus, an irregular vessel wall and a vessel wall flab could be seen.
Angiograms were analyzed for size, type, location of the aneurysm and the relationship of the aneurysm and the posterior inferior cerebella artery (PICA). Aneurysms were classified as follows: saccular and dissecting; lesion inferior to the origin of PICA (infra-PICA), lesion of VA involving the PICA (PICA) and lesion of the supra-PICA VA without extension into the vertebrobasilar junction.
In patients with unruptured aneurysms, endovascular therapy was limited to cases with progressive enlargement of aneurysmal dilatation or recurrent ischemic symptoms. Endovascular treatment included stent placement with or without coiling in 11 patients and parent VA occlusion in six patients [Table 1]. Stent placement was performed using commercially available stents in China. Balloon occlusion test was performed for patients with unruptured VA dissecting aneurysms selected for VA occlusion. Patients with ruptured VA aneurysms and PICA involved type, our technique included flow diversion by stent alone, stent coiling technique or coil embolization to preserve PICA take off [Figure 1]. Post-operative follow-up angiograms were performed at 1 month and 3-6 months post-embolization. Occlusion grades subsequent to coiling were measured as complete occlusion (100% occlusion), subtotal occlusion (>90% occlusion) and incomplete occlusion (<90% occluded). These grades of occlusion were determined by a consensus between two interventional neuroradiologists.
Systemic anticoagulation was achieved by administering an intravenous bolus injection of 3000-4000 units of heparin at early embolization with an additional 1000 units/h to prevent embolic events. After the procedure, patients were kept on heparin for at least 24 h. Patients with unruptured VA aneurysms were put on 200 mg aspirin and 75 mg clopidogrel per day for 3 days before the procedure. If acute stent placement was indicated, the patient was preloaded with 300 mg clopidogrel 2 h before the procedure. Post-procedure, patients were kept on a dual anti-platelet regimen for 1 month and then aspirin for 6 months.
Follow-up and statistical analysis
For ease of analysis and reporting, the five-point Glasgow Outcome Scale (GOS) score was categorized as either favorable (moderate disability or good recovery) or unfavorable (dead, vegetative or severe disability).  Univariate analysis was performed on each variable using the chi-square test, Fisher' exact test or Student's test, as appropriate. A P-value <0.05 was considered significant. All statistical analyses were performed using SPSS software version 16.0 (SPSS Inc., Chicago, IL, USA).
Clinical and angiographic characteristics in the 38 patients (mean age: 47.2 ± 11.3 years, range 18-75 years, male gender 73.7%) are presented in [Table 1]. Hypertension (28.9%) was the most common risk factor, while 11 (28.9%) patients had no associated medical comorbidity. The mean aneurismal size was 11.1 ± 6.5 mm (range 3-33 mm). Right VA was the site of aneurysm in 19 (50%) patients and left VA in 17 (44.7%) patients, while two (5.3%) patients had bilateral VA dissecting aneurysms. Thirty-three patients had 35 dissecting aneurysms and five patients had five saccular aneurysms. Supra-PICA-type aneurysm was observed in 15 (39.5%) patients, PICA involving type in 11 (28.9%) patients and infra-PICA type in 12 (31.6%) patients. Pre-aneurysmal narrowing and an aneurysmal dilatation was the most common angiographic sign in 23 (65.7%) aneurysms, while intramural thrombus, an irregular vessel wall and a vessel wall flab were seen in the other 12 (34.3%) aneurysms.
Ruptured VA aneurysms
Among the 17 patients (11 males and six females, mean age of 43.7 ± 11.5 years, hypertension in six) with SAH [Table 1], five patients demonstrated five saccular aneurysms and 12 patients demonstrated 13 dissecting aneurysms on serial angiography. The mean size of the aneurysm was 9.8 ± 5.7 mm. At presentation, five (29.4%) patients had supra-PICA aneurysm, seven (41.2%) had PICA involving aneurysm and five (29.4%) had supra-PICA aneurysm. Of the five treated supra-PICA aneurysms, two were treated with a stent coiling technique, two were treated with coil embolization [Figure 2] and one was treated with a trapping procedure. Of the seven treated PICA involving aneurysms, two were managed with stent alone, four were managed with coil embolization and one was treated with stent-assisted coiling preserving PICA. One of the five treated infra-PICA aneurysms was managed with trapping and four with stent coiling technique.
Unruptured VA aneurysms
Twenty-one patients (17 males and four females, mean age of 50.1 ± 10.5 years) had 22 unruptured VA dissecting aneurysms [Table 1], and four of these patients were managed conservatively and 17 patients were managed with endovascular procedures. Of these patients, 11 (52.4%) presented with either headaches or no symptoms, three (14.3%) presented with progressive brainstem symptoms and five patients were known cases of hypertension. The remaining seven (33.3%) patients had symptoms of brainstem ischemia. The average size of 22 aneurysms was 12.2 ± 7.1 mm. At the time of diagnosis, 10 (47.7%) patients had infra-PICA aneurysm, VA aneurysms involving the PICA were demonstrated in four (19.0%) patients and infra-PICA aneurysms in seven (33.3%) patients. Of the eight treated supra-PICA aneurysms, three were treated with a trapping procedure, four with stent coiling technique and one with stent alone. Of the three treated PICA involving aneurysms, one was managed with proximal occlusion, one with stent alone and one with stent coiling technique. Two of the six treated infra-PICA dissecting aneurysms were managed with trapping and four with the stent coiling technique. Four asymptomatic patients who had two supra-PICA, one PICA and one infra-PICA aneurysms were managed conservatively.
Post-procedural angiographic results
There were no procedure-related complications. Among the 34 treated patients, complete and subtotal occlusion of the aneurysms was achieved in 25 (73.5%) patients [Table 1]. Nine (26.5%) patients had incomplete occlusions on angiography. In nine patients, occlusion was intentional incomplete to preserve PICA or arterial perforators during the procedure.
Clinical and angiographic follow-up
The mean length of follow-up was 12.1 ± 7.3 months (range 1-30 months). Clinical outcome was excellent in 37 and good in one. During follow-up, there were no episodes recurrent hemorrhage and no deaths. Complete and subtotal occlusion was observed in 31 (81.6%) patients. Recanalization seen in two patients (5.9%) at 6 and 9 months did not require retreatment; one of the recanalization was caused by primary incomplete occlusion. One patient had angiographic-confirmed spontaneous thrombosis of the dissecting aneurysm. Angiography studies performed at 3-6 months displayed stable in the other three patients treated conservatively, and had good outcomes without recurrent hemorrhage.
Neither age (P = 0.078) nor sex (P = 0.293) had any significant association with rupture. VA aneurysm-related SAH occurred in 54.5% (five of 11) of the patients with pre-existing hypertension, whereas this was seen in 40.7% (11 of 27) of the patients with normal blood pressure. In the univariate analysis, hypertension was not a significant risk factor for rupture of VA aneurysm (P = 0.491). Although VA aneurysms rupture rates did not differe greatly according to aneurysm location (P = 0.314), more than half of the supra-PICA and infra-PICA aneurysms were in the unruptured group and more than half of the PICA involving aneurysms were in the ruptured group [Figure 3]. The correlation between aneurysm size and rupture was not statistically significant (P = 0.499), but [Figure 4] presents the relation between the number of patients with ruptured aneurysms and aneurysmal size. Aneurysm type (P = 0.012) and treatment modalities (P = 0.008) differed greatly between ruptured and unruptured VA aneurysms. All saccular VA aneurysms presented with SAH and all unruptured aneurysms were dissecting. Post-procedural angiographic results (P = 0.307) and clinical (P = 1.00) and angiographic outcomes (P = 0.104) were also not statistically significant.
Patients with VA aneurysms commonly present with symptoms related to mass effect and less commonly with SAH. ,,,, In our patients, in this study, 87.5% of VA aneurysms were dissecting and 12.5% were saccular type. The most common finding in our series was a pseudoaneurysm with irregular vessel lumen narrowing proximally. The etiology of VA aneurysms remains unclear. Both underlying atherosclerotic disease and recurrent arterial dissections secondary to hypertension have been implicated, ,, causing pseudoaneurysms of a fusiform architecture.  However, only 30% VA aneurysm cases had a hypertension history in our patients.
Endovascular treatment of VA aneurysm has emerged as an alternative to surgical treatment and consists of proximal occlusion, internal coil trapping, stent-assisted coiling or stent placement. ,,, When the aneurysm site is located in the infra-PICA, proximal occlusion may be effective and safe. However, if the aneurysm site is located in the supra-PICA, it is highly likely that many of the perforators could be occluded by internal trapping at the time of endovascular procedures. Endovascular internal trapping is only indicated for large and partially thrombosed dissecting aneurysms at this site. , Introduction of stents into the neurovascular system has enabled the preservation of the affected parent artery. This "reconstructive" technique is suggested to alter the intra-aneurysmal flow dynamics and lead to thrombus formation within the aneurysm. ,, Multiple stents can also be overlapped in a telescopic fashion to form a larger structure.  Despite early promise with these devices, , no long-term angiographic and clinical follow-up are available.
Kurata et al. reported thromboembolic events in one of their 18 patients with ruptured VA dissection with infarction in PICA territory.  No perioperative complications in 14 patients treated either with stent placement or stent-assisted coiling by Ahn and colleagues. In this study, in six of the 12 patients, follow-up angiograms at 6-12 months showed incomplete occlusion.  In the study by Iihara and colleagues, the treatment-related morbidity was 16.7%, and in this study of the 18 patients, 12 patients were without involvement of PICA and six patients were with PICA involved type.  Rabinov et al. reported a favorable outcome in 68% of the 25 treated patients with 20% mortality at 3.5 years mean clinical follow-up.  Good recovery was reported in 64% of the patients who presented with SAH and 100% good recovery in patients with VA distal aneurysms by Iihara and colleagues.  The mortality reported in the total treated cohort was 5.5%. Our analysis suggests that in VA aneurysms, stenting and stent-assisted coiling are feasible with an acceptably low periprocedural major morbidity and mortality. Our results are consistent with the earlier observations. Approximately 8-80% of the coiled aneurysms showed recanalization at angiographic follow-up, depending primarily on the size and location of the aneurysm. ,,, In our study, recanalization was observed in two patients, and our study also emphasizes the need for long-term follow-up angiographic studies. Retreatment may be required in case recanalization or regrowth of the pseudoaneurysm should occur. , Newer stent-like devices are being developed for intracranial aneurysms to serve as flow diverters within the parent artery. 
The major limitation of our study is that all the patients treated were in good SAH grades and clinical stable. Most of the patients were transferred from other facilities. The mean time from ictus to admission to our institution was 8.6 months. In our study, multiple endovascular techniques were used for treatment. Despite the aforementioned limitation, a favorable clinical and angiographic outcome in our study adds to the safety and feasibility of endovascular techniques as standard modality for the treatment of VA aneurysms.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]