Background and purpose In patients with aneurysmal subarachnoid hemorrhage, deciding the optimal treatment strategy is challenging in the presence of severe and symptomatic vasospasm and the existing literature is limited. The purpose of this study was to evaluate the safety and effectiveness of endovascular treatment of intracranial aneurysms and the feasibility of stent-assisted coiling during severe and symptomatic vasospasm.
Materials and methods Fifteen consecutive patients with 18 aneurysms who underwent endovascular treatment during severe and symptomatic vasospasm were included in the study. Patient and aneurysm characteristics, treatment technique, and clinical and angiographic outcomes were retrospectively evaluated.
Results Aneurysms were treated by the following techniques: single catheter coiling in 9, stent-assisted coiling in 8 (3 aneurysms with Y-configuration double stents), and balloon-assisted coiling in 1. All patients showed angiographic improvement of vasospasm after treatment including a noticeable dilation of the spastic parent arteries following deployment of the stents. According to the immediate post-treatment angiography results, 14 aneurysms (77.8%) had class I occlusion and 4 (22.2%) had class II occlusion. Three patients died during the course of subarachnoid hemorrhage. Mortality was related to the poor grade (Hunt and Hess grade V) and cardiovascular complications in two and refractory vasospasm one patient. The remaining patients had favorable clinical outcomes at follow-up (modified Rankin Scale 0–2).
Conclusions Endovascular treatment of intracranial aneurysms during severe and symptomatic vasospasm is safe and effective. Stent-assisted coiling in the presence of severe vasospasm is also feasible in wide-necked aneurysms with the additional benefits of mechanical vasodilation.
Statistics from Altmetric.com
Symptomatic vasospasm is one of the major causes of increased morbidity and mortality after aneurysmal subarachnoid hemorrhage (SAH),1 and angiographic evidence of vasospasm is detected in approximately 70% of patients with the maximum peak between 5 and 14 days.2 ,3 The optimal approach in the management of aneurysmal SAH is to treat the ruptured aneurysm at an early stage to avoid the risk of rebleeding and then to focus on treatment of the vasospasm.4–6 However, early stage treatment may not be possible in all cases for various reasons. For example, the period between aneurysm rupture and treatment may be prolonged as a result of an initial misdiagnosis and delay in seeking medical attention, delayed transfer to the referral center, poor clinical grade of the patient, or accompanying medical problems that preclude early treatment. Depending upon the extended period, encountering patients with symptomatic and severe vasospasm at the time of admission is not uncommon. There is no definite consensus on the optimal treatment strategy in this subgroup of patients, but surgery is generally avoided during the period of vasospasm.7 ,8 On the other hand, endovascular treatment has been reported to be an effective modality in a small number of studies,9–12 with the opportunity to treat both the vasospasm and aneurysm in a single session. In spite of the varied treatment protocols and techniques described in existing endovascular studies, there are no data regarding the feasibility of stent-assisted coiling in the presence of severe vasospasm or evaluation of the effect of stenting on the spastic vessels in the current literature.
The purpose of this study was to evaluate the safety and effectiveness of endovascular treatment of intracranial aneurysms and also the feasibility of stent-assisted coiling in the presence of severe and symptomatic vasospasm after aneurysmal SAH.
Materials and methods
This study was approved by the institutional review board and consent forms were obtained from all patients or their relatives before the procedure.
Fifteen consecutive patients with aneurysmal SAH who underwent endovascular aneurysm treatment during severe and symptomatic vasospasm from June 2011 to May 2016 were included in the study. Patient and aneurysm characteristics, treatment technique, and clinical and angiographic outcomes were retrospectively evaluated.
The aneurysm size, morphology, and degree of vasospasm were determined via angiographic evaluation. Aneurysm sizes were classified as small (<10 mm) or large (≥10 and ≤25 mm). Wide-necked aneurysms are defined as those with a dome-to-neck ratio of <2:1 and/or a neck length of >4 mm. Complex aneurysms are defined as multilobulated aneurysms or aneurysms with vascular branches originating from the aneurysm sac. As in previous studies,9 ,10 the degree of vasospasm was classified as mild (<25%), moderate (25–50%), and severe (>50%) on the basis of the estimated original arterial diameter. Follow-up or post-treatment angiograms were used as a reference for estimating original vessel diameters and were retrospectively compared side by side with pretreatment angiograms by two neuroradiologists; grading of vasospasm was done using these comparative measurements. During the comparisons the mean of two separate measurements from the point of maximum vasospasm or from the narrowest segment of the vessel when the entire vessel was spasmodic was used for calculations. All measurements were done with an electronic cursor on the digital images to minimize observer bias.
Vasospasm was defined as focal if it was limited to the vascular territory of the ruptured aneurysm or diffuse if more than one vascular territory was affected. Symptomatic vasospasm was defined as deterioration in the level of consciousness and/or the development of new focal neurologic signs not related to other conditions such as hydrocephalus, seizures, electrolyte disturbance, or infection.
Technique and protocol
The decision as to the treatment modality and technique to be used was made by consensus of the neurosurgical and neurointerventional physicians after careful evaluation of each patient as part of a multidisciplinary decision-making process. Our institutional policy is to treat both vasospasm and aneurysm in the same setting if symptomatic vasospasm is present at the time of admission in patients with aneurysmal SAH. The endovascular approach is therefore preferred as the primary treatment modality in such patients. Furthermore, endovascular treatment is preferred over surgery in the following circumstances: patients with multiple aneurysms requiring multiple craniotomies; patients with diffuse vasospasm where the surgical space is restricted due to brain swelling; patients with complex aneurysms requiring a complex surgical approach with temporary and multiple clipping; patients with accompanying medical comorbidities that preclude open surgery; and patients with poor grades.
In all patients both the aneurysm and vasospasm were treated in the same session. All procedures were performed under general anesthesia by a standard transfemoral approach with a 6 F introducer sheath and 6 F guiding catheter combination. After catheterization of the internal carotid artery (ICA) or vertebral artery, 1 mg nimodipine (diluted in a 20 mL syringe) was administered through a guiding catheter within 10–15 min by maintaining the main arterial blood pressure above 110 mm Hg. The aneurysm sac was then selectively catheterized with the guidance of road-mapping and treated by one of the following techniques: single catheter coiling, balloon-assisted coiling, or stent-assisted coiling with bare platinum coils. The decision to use stent-assisted coiling was based on the aneurysm morphology. Wide-necked and/or complex aneurysms which were considered to be difficult to treat with single catheter coiling or balloon-assisted coiling were treated with stent-assisted coiling. Stents used in this study were Solitaire AB (Covidien/ev3, Irvine, California, USA), LVIS Jr (MicroVention, Tustin, California, USA), and ACCLINO flex (Acandis, Pforzheim, Germany). In case of stent-assisted coiling, the aneurysms were embolized using the jailing technique. Any change in the diameter of the spastic vessels was assessed after administration of nimodipine, selective catheterization of the aneurysm, deployment of the stent, during coiling, and also at the end of the procedure. The degree of vasodilation was graded as mild (improved by less than one angiographic grade), moderate (improved by at least one angiographic grade), and complete (improved by two grades or if normalized). In patients with diffuse vasospasm and/or patients with residual vasospasm at the end of the procedure, additional doses of nimodipine (1–2 mg) were administered for each vascular territory (related or not related to the aneurysm) until angiographic vasodilation was achieved. In order to avoid thrombus formation around the guiding catheter during the prolonged infusion of nimodipine and also to minimize the risk of thromboembolism induced by the blockage of blood flow by the microcatheters and guidewires in the spastic vessels, all patients were heparinized shortly after catheterization of the parent artery with a guiding catheter and an activated clotting time of 250–300 s was maintained during the whole procedure. In patients undergoing stent-assisted coiling, 600 mg clopidogrel and 100 mg aspirin or 60 mg prasugrel alone were loaded via a nasogastric tube just before treatment and maintained on daily doses of 75 mg clopidogrel and 100 mg aspirin or 10 mg prasugrel. Standard medical management, including induced hypertension and calcium channel blockers (peroral nimodipine at a dose of 60 mg every 4 hours for 21 days), was continued after endovascular treatment until reversal of neurologic deficits. In cases of refractory vasospasm, chemical angioplasty with selective IA administration of nimodipine and/or milrinone was performed in separate sessions.
The preoperative clinical condition of the patients was evaluated using the Hunt and Hess classification.13 Clinical outcome was measured immediately after the procedure, at discharge, and at the time of follow-up by the modified Rankin Scale (mRS).14 Aneurysm occlusion was evaluated according to the Raymond scale (I, complete occlusion; II, neck remnant; and III, residual aneurysm).15
Patient and aneurysm characteristics, treatment technique, angiographic and clinical outcomes are summarized in table 1.
The study group consisted of five men and 10 women of mean age 49.3±13.1 years (range 25–70) with a total of 18 aneurysms. All patients were referred from other hospitals to our institution 2–10 days after initial bleeding and were treated within 48 hours of admission. Two patients had multiple aneurysms and detected aneurysms were treated in a single stage in one patient and in two stages in the other patient. The aneurysms were treated with the following techniques: single catheter coiling in nine aneurysms, stent-assisted coiling in eight aneurysms (three aneurysms with Y-configuration double stents and five with a single stent), and balloon-assisted coiling in one aneurysm. In one patient balloon angioplasty was performed to the spastic proximal parent artery (middle cerebral artery, MCA) before embolization of the right MCA aneurysm in order to achieve passage of the microcatheter. No periprocedural complications were encountered during treatment.
All patients had symptomatic vasospasm at the time of treatment with the exception of one in which it was undetermined because of poor grade (Hunt and Hess grade V). Despite the poor clinical grade, the patient was treated because of the presence of adequate brainstem function. Three patients had cerebral ischemia and infarction on the pretreatment CT scan which was related to spastic vascular territories and attributable to the vasospasm, as confirmed by angiography. Vasospasm was focal in nine patients and diffuse in six. The distribution and degree of vasospasm and also the degree of vasodilation in the course of treatment is summarized in table 2.
All patients had a variable degree of angiographic dilation in the spastic vessels after the endovascular procedure. The most prominent vasodilation was seen after obliteration of the aneurysm whereas the degree of dilation was mild to moderate after administration of nimodipine and catheterization of the aneurysm in most of the patients (figure 1). Furthermore, moderate to complete dilation was observed in the spastic parent arteries after deployment of the stents in patients treated with stent-assisted coiling (figure 2). In six patients with diffuse vasospasm, additional doses of nimodipine (1–2 mg) were administered in the same session to the vascular territories not related to the aneurysm for the treatment of symptomatic vasospasm and, in two patients with residual vasospasm, additional nimodipine (1–2 mg) was administered at the end of the procedure. The neurologic status was improved immediately after the procedure in 13 patients while no obvious clinical improvement was observed in two patients. Additional chemical angioplasty was performed in separate sessions with selective IA administration of nimodipine and milrinone in three patients. Two of these patients recovered without serious neurologic deficits while refractory vasospasm and delayed cerebral ischemia led to death in one patient. Hydrocephalus requiring ventriculostomy or ventriculoperitoneal shunt was seen in two patients. One of these patients was treated with stent-assisted coiling and ventriculostomy was performed under dual antiplatelet therapy; no hemorrhagic complications related to the procedure were encountered.
Three of the 15 patients died in the course of SAH. Mortality was related to the gravity of SAH (Hunt and Hess grade V) and cardiovascular complications in two patients and refractory vasospasm and delayed cerebral ischemia in one patient. The remaining patients had favorable clinical outcomes (mRS 0–2) at discharge and also at follow-up. Angiographic follow-up imaging was available in nine patients for 11 aneurysms for a mean duration of 23.8±16.2 months (range 3–52). Immediate post-treatment angiography showed that 14 aneurysms (77.8%) had class I occlusion and four (22.2%) had class II occlusion while the last follow-up angiography showed that eight aneurysms (72.7%) had class I occlusion and three (27.3%) had class II occlusion.
Deciding the optimal timing of treatment is challenging after aneurysmal SAH, but early treatment before the period of vasospasm has gained favor in recent years with the advantage of preventing rebleeding.6 ,16 However, early treatment is not always feasible for various reasons and it is possible to encounter patients with severe and symptomatic vasospasm during treatment. Furthermore, severe and acute angiographic vasospasm may also occur in the early days following SAH in some patients, especially those with poor neurologic grade.17 The optimal treatment strategy in the presence of severe and symptomatic vasospasm is still uncertain and the current literature on the issue is limited. Surgery is mostly avoided in the course of vasospasm.7 ,8 During the period of vasospasm the surgical space may be restricted, due particularly to swollen and retracted ischemic brain tissue, and difficult intraoperative surgical conditions may be encountered. Surgical manipulation and temporary clipping of spastic cerebral blood vessels may worsen arterial narrowing, decrease distal blood flow, and possibly aggravate the latent cerebral ischemia.18 However, poorer clinical results observed with surgery may not be directly influenced by vasospasm itself but may result from other factors such as hypotension, dehydration, or hypoxia encountered during surgery, and a few studies have shown the feasibility of aneurysmal clipping in the setting of vasospasm.19 ,20 Compared with surgery, endovascular treatment offers advantages such as less invasiveness, avoidance of brain manipulation, and the opportunity to treat both the aneurysm and vasospasm in the same setting. However, it is difficult to make comparisons and to reach a definitive conclusion as existing endovascular studies are limited and consist of inhomogeneous small patient cohorts.9–12 Moreover, treatment protocols and techniques differ among them. Only one study has compared surgery with endovascular treatment in patients with symptomatic vasospasm on presentation and no statistically significant difference in terms of clinical outcome was found between the two treatment modalities.21
Among the previously reported endovascular studies, Murayama et al 9 used papaverine as the pharmacologic vasodilator agent and balloon angioplasty was performed from distal to proximal to the spastic segments after obliteration of the aneurysm in nine of the 12 patients. However, in three patients mechanical and/or chemical angioplasty was performed before coiling because of severe vasospasm in the proximal portion of the parent artery. In two other studies the usefulness of continuous IA nimodipine infusion through the guiding catheter during the whole procedure was evaluated.10 ,12 All patients had various degrees of vasodilation with continuous nimodipine infusion and no apparent side effects were observed in either study. On the other hand, Kurata et al 11 reported a series without the use of pharmacologic vasodilation and evaluated the effect of catheterization alone on spastic vessels. The endovascular procedure was successful in 17 patients and all cases had mild dilation of the spastic segments. They also mentioned that vasospasm of the M2 and A2 segments could be treated with microcatheterization only. In the current study, 1 mg (initial dose) of nimodipine (diluted in a 20 mL syringe) was administered through a guiding catheter after catheterization of the ICA or vertebral artery. Continuous infusion was not performed, unlike in previous studies.10 ,12 However, in two patients with residual vasospasm, additional doses of nimodipine were administered at the end of the procedure. In patients with diffuse vasospasm, nimodipine was administered to the vascular territories not related to the aneurysm for the treatment of symptomatic vasospasm in the same session. Following the initial administration of nimodipine, dilation of the spastic vessels was mostly mild but still sufficient to allow passage of the microcatheter and catheterization of the aneurysmal sac. However, in one patient balloon angioplasty was performed from proximal to the distal spastic parent artery in order to achieve passage of the microcatheter. In spite of the previous study results reported by Kurata et al,11 catheterization of spastic vessels did not show mechanical dilation or aggravation of spasm in the current series. On the other hand, noticeable dilation was observed in the spastic arteries after deployment of the stents in patients treated with stent-assisted coiling.
The feasibility of stent-assisted coiling in the treatment of intracranial aneurysms during severe vasospasm has not previously been reported in the literature despite the fact that the use of stents in the treatment of wide-necked aneurysms in the acute stage of SAH has been shown to be an acceptable choice.22 There is only one recently published case series evaluating the effect of temporary stenting with self-expandable retrievable stents in the treatment of delayed vasospasm.23 The study concluded that temporary stenting with stent retrievers can provide long-lasting cerebral vasodilation in patients with delayed cerebral vasospasm, but the radial force was not sufficient to cause vasodilation in some instances. To our knowledge, this is the first study to evaluate the feasibility of stent-assisted coiling and the effect of stenting on the spastic vessels in the endovascular treatment of intracranial aneurysms in the presence of severe vasospasm following aneurysmal SAH. In the current study, eight wide-necked and complex aneurysms were treated with the assistance of low profile stents (except one with Solitaire AB) which can be delivered through a 0.017 inch microcatheter, which has the advantage of easy navigation through spastic vessels. No serious technical difficulties were encountered during treatment and deployment of the stents even in the Y-configuration, and a noticeable vasodilation was observed in stented spastic segments where the effect was considered to be mechanical. Despite the relatively low radial force of the stents compared with balloon angioplasty, the underlying physiological mechanism may be similar to the effects of balloon angioplasty such as compression of the connective tissue, stretching of the internal elastic lamina, and a combination of compression and stretching of the smooth muscles.24 In addition to the initial mechanical vasodilation effect, the stented vascular segments seem to be resistant to refractory vasospasm as observed in patient 14 (figure 2). In that patient, no prominent spasm was seen in the stented segments in the course of refractory vasospasm while the other vascular territories were significantly affected.
In patients with severe and symptomatic vasospasm on admission, the advantage of endovascular treatment is the ability to treat both the vasospasm and aneurysm in the same session. In the current study the patients had various degrees of angiographic improvement in the spastic vessels and the neurologic status was improved immediately after the procedure in most of the patients. Angiographically, the most obvious vasodilation was observed after obliteration of the aneurysm sac in most of the patients. In addition to the delayed vasodilator effect of nimodipine, the securing of the aneurysm and exclusion of the aneurysmal sac from the circulation may possibly induce some degree of vasodilation effect, especially on the pre- or post-aneurysmal portion of the spastic parent arteries. Even so, angiographic improvement is not always correlated with good clinical results. In the study by Murayama et al 9 the clinical outcomes were summarized as good recovery in six patients, moderately disabled in two, severely disabled in three, and death in one patient. They concluded that the patients with diffuse angiographic vasospasm and high Hunt and Hess grades (IV and V) continue to have a poor prognosis despite treatment and angiographic improvement. Similar clinical results have been demonstrated by Kurata et al.11 On the other hand, better clinical outcomes have been achieved in two other studies,10 ,12 which differ from the previous studies in using continuous IA nimodipine infusion during the whole procedure. However, it is not possible to determine the true effect of the treatment technique and protocol on the clinical outcomes in these small and inhomogeneous cohorts. In the current study, 12 of 15 patients (80%) had favorable clinical outcomes (mRS 0–2). The factors affecting the clinical outcomes were considered to be the degree and distribution of vasospasm and the Hunt and Hess grades. The use of stents did not directly influence poor clinical outcomes since mortality was related to the high Hunt and Hess grade (V), cardiovascular complications, and refractory vasospasm.
Our study is limited by its retrospective nature and the patient population which is too small to allow between-study comparisons and to reach a definitive decision. Our definition of the underlying physiological mechanism for the effect of stenting on the spastic vessels is hypothetical since no reports on this issue have yet been published. Because of the combined use of a pharmacologic agent (nimodipine) with both early and delayed vasodilator effect, it was not possible to determine exactly the stage which shows the real impact of angiographic vasodilation. Further studies with extended series should be undertaken to confirm and establish the physiologic and histologic basis for our outcomes. Further work is also needed to confirm our hypothesis that obliteration of the aneurysm sac and exclusion of the aneurysm from the circulation may induce a vasodilatory effect on the pre- and post-aneurysmal portion of spastic parent arteries. Even so, the results of this study may help in the choice of treatment technique and protocol in patients with severe and symptomatic vasospasm at the time of presentation.
Endovascular treatment of intracranial aneurysms during severe and symptomatic vasospasm is safe and effective and provides the opportunity to treat both vasospasm and the aneurysm in the same session. Stent-assisted coiling in the presence of severe vasospasm is also feasible in wide-necked and complex aneurysms with some additional benefits of mechanical vasodilation. However, further studies with extended series are needed to confirm our conclusions.
Contributors CA designed the study. CA, OK and ED collected the data. CA and MG performed the analysis. CA wrote the manuscript. CA and SA edited and reviewed the manuscript. CA is guarantor.
Competing interests None declared.
Patient consent Obtained.
Ethics approval This study was approved by Başkent University Institutional Review Board (project no KA16/245).
Provenance and peer review Not commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.