Endovascular Treatment of Aneurysmal Subarachnoid Hemorrhage

Aneurysmal subarachnoid hemorrhage is a deadly disease associated with high morbidity and mortality. Surgical clipping has been the gold standard treatment for more than 70 years. Endovascular therapy is now accepted as a valid alternative therapeutic modality. The authors’ approach emphasizes collaboration between endovascular and surgical specialists. The array of new endovascular techniques has extended beyond the Guglielmi Detachable Coil to include new stents and flow-diverting devices. The future promises expansion of the number of types of aneurysms that are treatable with endovascular techniques.

Intracranial aneurysms are common entities whose natural history and definitive management remain controversial. Their prevalence varies according to study design but is estimated at approximately 2.3% in the general population. The most dreaded complication related to intracranial aneurysms is rupture leading to subarachnoid hemorrhage (SAH), a devastating condition still associated with a 30-day mortality rate of 30% to 40%, despite a consistent decline over the past 3 decades. The annual risk of rupture is 1.3%. Only one-third of patients surviving an aneurysmal SAH remain functionally independent. The primary goal in managing patients presenting with aneurysmal SAH is to prevent a new rupture of the aneurysm, which is associated with an even higher mortality rate. Surgical and endovascular methods are available to achieve this goal. This article reviews endovascular management of ruptured intracranial aneurysms.

Surgical clipping has been the gold standard of treatment for more than 70 years, since Walter Dandy first applied a silver clip to the neck of an unruptured internal carotid artery aneurysm at the The Johns Hopkins Hospital in 1937. The surgical approach to intracranial aneurysms was then refined by the adaptation of microsurgical techniques to the neurosurgical field by Yasargil. The concept of endovascular treatment of intracranial aneurysms, drawing on work performed by Serbinenko in the 1970s, was initially based on the use of balloons inflated within the aneurysmal cavity. In 1991, Guglielmi and coworkers published the first description of the endovascular application of detachable platinum coils (the Guglielmi Detachable Coil) to induce thrombosis and obliteration of intracranial aneurysms in humans. After Food and Drug Administration approval was granted in 1995, the use of endovascular coiling has steadily increased and has been adopted as an alternative technique for the treatment of ruptured and unruptured intracranial aneurysms. Only two randomized, prospective studies comparing endovascular coiling and surgical clipping have been reported. The first one is a single-center study published in 1999, which found no significant difference in the obliteration rates at 12 months in 109 patients with SAH randomly assigned to surgical clipping or endovascular coiling. The second, the International Subarachnoid Aneurysm Trial (ISAT), compared clipping versus coiling in 2143 patients with ruptured intracranial aneurysms. Patients treated endovascularly were at a lower risk of death or dependence at 1 year compared with the surgical group, with an absolute risk reduction of 7.4%, which was maintained for up to 7 years. The risk of rehemorrhage was low but more common after coiling than clipping. The neurosurgical community found these results controversial, pointing to possible selection biases detrimental to the surgical group and to the level of expertise of the neurosurgeons performing the surgical treatments (general vs vascular neurosurgeons). Questions concerning the durability and long-term efficacy of coil embolization and protection against rerupture were also raised. Despite these drawbacks, the ISAT did support the notion that endovascular therapy is a valid alternative to surgical clipping.

Endovascular techniques

Endovascular techniques for the treatment of intracranial aneurysms with conservation of the parent artery, also known as constructive therapies, include standard coil embolization, coil embolization with balloon remodeling or stent assistance, and balloon-assisted liquid polymer embolization ( Fig. 1 A, B, and C). The use of covered stents (or stent grafts) has been proposed as an option for large, fusiform, or wide-necked aneurysms, primarily located in the carotid and vertebral arteries, where the risk of occluding functionally important side branches is relatively low (see Fig. 1 D). The long-term patency of stent grafts placed in relatively small vessels, such as the internal carotid artery, is another potential drawback of this approach and remains currently unknown. Stents with a tight mesh or stents covered with semipermeable membranes (collectively known as flow diverters) may represent an improvement over conventional stent grafts in terms of parent artery and side branches patency. These stents may expand the indications of stent grafting to intracranial lesions, although their use in ruptured aneurysms needs to be carefully evaluated. Parent artery occlusion, also referred to as deconstructive therapy, remains a valid alternative option for nonsurgical candidates whose aneurysms are not amenable to constructive treatment methods.

The clinical condition of patients, the aneurysm location and morphology (in particular the diameter of the neck and its relation to the parent artery), and the presence of branches arising from the sac or the neck are important considerations when choosing the most appropriate treatment plan. The aneurysm neck, in particular its size and relation to the parent artery and potential side branches, is the key feature in determining if coil embolization is an appropriate treatment option. Standard coil embolization is considered feasible for aneurysms with a small neck (<4 mm), a dome-to-neck ratio equal or greater than two, and in the absence of important branches arising from the sac or the neck. Coil embolization is achieved primarily with platinum coils. Although a careful analysis of the aneurysm morphology is essential to planning efficient therapy, aneurysms with a seemingly unfavorable configuration can occasionally respond well to simple coiling ( Fig. 2 ). Advances made in platinum coil technology have tried to address incomplete aneurysm occlusion, which increases the risk of coil compaction and aneurysm recanalization. Reported rates of recanalization are approximately 21% to 28.6% but can be as high as 60% for giant aneurysms. Recently developed hybrid, or biologically active, coils are chemically pretreated to enhance their thrombogenicity in an effort to try decreasing the recanalization rate. Currently available modified coils include polyglycolic acid/lactide copolymer–coated coils (Matrix, Boston Scientific, Natick, MA, USA; Cerecyte, Micrus, Sunnyvale, CA, USA; and Nexus, Micro Therapeutics, Irvine, CA, USA) and hydrogel-coated coils (HydroCoil, MicroVention, Aliso Viejo, CA, USA). Other types of coated or active devices, such as coils with radioactive components or coils coated with biologic material, such as collagen or cells, are in experimental phases. Matrix coils comprise an inner core of platinum covered with a biodegradable polymer (polyglycolide/polylactide) designed to accelerate aneurysm fibrosis, neointima formation, and inflammation. The safety of these coils for aneurysm treatment is similar to that of bare platinum coils. Higher rates of recanalization (32%, from 26.1% for small aneurysms with small necks to 75% for large aneurysms) and thromboembolic events (up to 20% vs 2.5%–11%), however, are reported with Matrix coils versus bare platinum coils. Progressive resorption of the polymer coat leading to loss of volume and instability is a possible explanation for the high recurrence rates associated with these coils. A recently published study of 152 patients with ruptured and unruptured aneurysms treated exclusively with Matrix coils showed similar results with no better recanalization rates than those previously reported for bare platinum coils (recanalization rates: 31.1% for aneurysms <10 mm and 56% for aneurysms >10 mm, with more frequent recanalization in ruptured aneurysms). HydroCoils are standard platinum coils coated with an expandable hydrogel material that results in delayed progressive coil expansion on contact with blood. These coils are supposed to provide superior aneurysm volume filling to bare platinum coils and to promote healing and endothelialization at the aneurysm neck. Despite these features, recanalization rates remain high, up to 27% for large aneurysms. More concerning, however, are reported cases of aseptic meningitis and delayed hydrocephalus, which seem specific to the HydroCoil as no cases have so far been described with polyglycolic acid/lactide copolymer–coated coils alone, although some have occurred when Matrix coils were used in combination with HydroCoils. Delayed perianeurysmal inflammation with dramatic neurologic dysfunction (bilateral visual loss) has been reported after embolization with HydroCoils. Perianeurysmal inflammation leading to visual loss has also been described in cases of paraclinoid aneurysms treated with standard platinum or coated coils. The factors leading to these various coil-related events have not yet been elucidated, and more information is needed regarding the role of coils, clot burden, aneurysm size, and inflammatory mediators in the development of these complications.

Recent advances in stent technology have led to the development of flexible self-expanding nitinol stents (or reconstruction devices) (Neuroform, Boston Scientific Neurovascular, Natick, MA, USA; Enterprise, Cordis Neurovascular, Miami Lakes, FL, USA; and LEO, Balt, Montmorency, France) dedicated to intracranial aneurysm therapy, specifically for the treatment of complex and wide-necked aneurysms. Advantages of these self-expanding intracranial stents over the balloon expandable stents previously used for assisted coiling include improved trackability, which helps navigate tortuous intracranial vasculature (although this is true only for the latest generation of self-expanding stents); improved deliverability; and decreased vessel injury during deployment. Variations in stent design include open-cell (Neuroform) versus closed-cell design (Enterprise and LEO), low radial force (Neuroform) versus high radial force (LEO) versus low radial force/high compression resistance (Enterprise), and stent recoverability after partial deployment (a characteristic inherent to the closed-cell design, with up to 70% of stent length for the Enterprise and up to 90% of the stent length for the LEO). Despite several technical differences in stent design, these devices have been safe and effective in assisting the treatment of cerebral aneurysms. The major argument against the use of these devices for the management of patients with aneurysmal SAH is linked to the need for concurrent antiplatelet therapy. A recent series of patients treated with stent-assisted coiling had a higher rate of hemorrhagic complications in the group presenting with SAH: two patients had a fatal outcome believed related to antiplatelet therapy, one from a massive intraventricular hemorrhage (IVH) secondary to an external ventricular drain change, the other from a parenchymal hemorrhage of unknown origin. The risk of parenchymal hemorrhage in patients taking antiplatelet therapy, however, may be independent of their having suffered a SAH. In the authors’ experience, one patient with a nonruptured clinoid segment aneurysm was successfully treated with stent-assisted coiling, having suffered a nonfatal contralateral occipital lobe hemorrhage 2 days after the procedure. The goal of antiplatelet therapy is to reduce the risk of thromboembolic complications related to the stent placement itself and to the subsequent presence of an intraluminal foreign body, at least until the stent structure is covered by a layer of endothelial cells. Although protocols may slightly vary according to institutional and operator preferences, patients scheduled for elective therapy are typically placed under a combination of oral antiplatelet agents several days prior to the procedure. In the authors’ practice, patients are asked to take clopidogrel (75 mg) and aspirin (325 mg) daily starting 5 days before stent placement. Patients already taking these medications for other purposes are given an additional loading dose of clopidogrel (300 mg) the day preceding treatment. Such a drug regimen is not possible in patients presenting with aneurysmal SAH. There is no clear consensus about antiplatelet therapy in patients with SAH. As an example, the approach adopted at the authors’ institution for ruptured wide-necked aneurysms that benefit from stent-assisted coiling is described. These aneurysms are divided into lesions that likely can be secured initially with partial coiling only but require a follow-up procedure for complete treatment and lesions unlikely to be secured with coiling only. This evaluation is based on the morphology of the aneurysm as depicted with 3-D digital subtraction angiography (DSA). In aneurysms that can be secured initially with partial coiling, the immediate goal is to ensure that the risk of rerupture is eliminated (ie, that no residual flow is left within the aneurysmal cavity, in particular at its dome). This initial treatment may be helped by the use of the balloon remodeling technique and may even achieve definitive therapy ( Fig. 3 ). The residual component, if any, is then addressed at a later date with the assistance of a stent, using standard antiplatelet preparation. Even when this first approach has been elected, a stent may be deployed if it becomes obvious that adequate treatment will not be achieved with coiling alone ( Fig. 4 ). In situations where it seems likely that adequate treatment will not be achieved by coiling alone, a stent may be used as first intention therapy. In such instances, a microcatheter is placed within the aneurysmal cavity first, in order to secure access for subsequent coiling (jailing technique). The use of the jailing technique in this instance principally prevents the unlikely but potentially catastrophic situation in which a stent is deployed but the aneurysm cannot be subsequently catheterized. The stent is then advanced and deployed using a standard technique ( Fig. 5 ). In the authors’ practice, antiplatelet therapy is administered only when the stent has been successfully deployed. In the absence of injectable aspirin (eg, as in the United States), the authors’ regimen consists aspirin (600 mg) administered rectally and clopidogrel (450 mg) delivered via a nasogastric tube. Intravenous heparin (initial dose 5000 IU intravenous bolus, monitored and adjusted using activated clotting time) is added after detachment of the first microcoil within the aneurysmal cavity. A concern with this approach lies in the potential need for subsequent placement of a ventricular shunt under antiplatelet therapy. In order to avoid, as much as possible, such a situation, a CT scan is obtained immediately prior to the procedure and a shunt placed if ventricular enlargement is observed.