To determine the effectiveness and cost-effectiveness of coil embolization compared with surgical clipping to treat intracranial aneurysms.
Endovascular coil embolization is a percutaneous approach to treat an intracranial aneurysm from within the blood vessel without the need of a craniotomy. In this procedure, a microcatheter is inserted into the femoral artery near the groin and navigated to the site of the aneurysm. Small helical platinum coils are deployed through the microcatheter to fill the aneurysm, and prevent it from further expansion and rupture. Health Canada has approved numerous types of coils and coil delivery systems to treat intracranial aneurysms. The most favoured are controlled detachable coils. Coil embolization may be used with other adjunct endovascular devices such as stents and balloons.
Intracranial aneurysms are the dilation or ballooning of part of a blood vessel in the brain. Intracranial aneurysms range in size from small (<12 mm in diameter) to large (12–25 mm), and to giant (>25 mm). There are 3 main types of aneurysms. Fusiform aneurysms involve the entire circumference of the artery; saccular aneurysms have outpouchings; and dissecting aneurysms have tears in the arterial wall. Berry aneurysms are saccular aneurysms with well-defined necks.
Intracranial aneurysms may occur in any blood vessel of the brain; however, they are most commonly found at the branch points of large arteries that form the circle of Willis at the base of the brain. In 85% to 95% of patients, they are found in the anterior circulation. Aneurysms in the posterior circulation are less frequent, and are more difficult to treat surgically due to inaccessibility.
Most intracranial aneurysms are small and asymptomatic. Large aneurysms may have a mass effect, causing compression on the brain and cranial nerves and neurological deficits. When an intracranial aneurysm ruptures and bleeds, resulting in a subarachnoid hemorrhage (SAH), the mortality rate can be 40% to 50%, with severe morbidity of 10% to 20%. The reported overall risk of rupture is 1.9% per year and is higher for women, cigarette smokers, and cocaine users, and in aneurysms that are symptomatic, greater than 10 mm in diameter, or located in the posterior circulation. If left untreated, there is a considerable risk of repeat hemorrhage in a ruptured aneurysm that results in increased mortality.
In Ontario, intracranial aneurysms occur in about 1% to 4% of the population, and the annual incidence of SAH is about 10 cases per 100,000 people. In 2004-2005, about 660 intracranial aneurysm repairs were performed in Ontario.
Treatment of Intracranial Aneurysms
Treatment of an unruptured aneurysm attempts to prevent the aneurysm from rupturing. The treatment of a ruptured intracranial aneurysm aims to prevent further hemorrhage. There are 3 approaches to treating an intracranial aneurysm.
Small, asymptomatic aneurysms less than 10 mm in diameter may be monitored without any intervention other than treatment for underlying risk factors such as hypertension.
Open surgical clipping, involves craniotomy, brain retraction, and placement of a silver clip across the neck of the aneurysm while a patient is under general anesthesia. This procedure is associated with surgical risks and neurological deficits.
Endovascular coil embolization, introduced in the 1990s, is the health technology under review.
The Medical Advisory Secretariat searched the International Health Technology Assessment (INAHTA) Database and the Cochrane Database of Systematic Reviews to identify relevant systematic reviews. OVID Medline, Medline In-Process and Other Non-Indexed Citations, and Embase were searched for English-language journal articles that reported primary data on the effectiveness or cost-effectiveness of treatments for intracranial aneurysms, obtained in a clinical setting or analyses of primary data maintained in registers or institutional databases. Internet searches of Medscape and manufacturers’ databases were conducted to identify product information and recent reports on trials that were unpublished but that were presented at international conferences. Four systematic reviews, 3 reports on 2 randomized controlled trials comparing coil embolization with surgical clipping of ruptured aneurysms, 30 observational studies, and 3 economic analysis reports were included in this review.
Safety and Effectiveness
Coil embolization appears to be a safe procedure. Complications associated with coil embolization ranged from 8.6% to 18.6% with a median of about 10.6%. Observational studies showed that coil embolization is associated with lower complication rates than surgical clipping (permanent complication 3-7% versus 10.9%; overall 23% versus 46% respectively, p=0.009). Common complications of coil embolization are thrombo-embolic events (2.5%–14.5%), perforation of aneurysm (2.3%–4.7%), parent artery obstruction (2%–3%), collapsed coils (8%), coil malposition (14.6%), and coil migration (0.5%–3%).
Randomized controlled trials showed that for ruptured intracranial aneurysms with SAH, suitable for both coil embolization and surgical clipping (mostly saccular aneurysms <10 mm in diameter located in the anterior circulation) in people with good clinical condition:Coil embolization resulted in a statistically significant 23.9% relative risk reduction and 7% absolute risk reduction in the composite rate of death and dependency compared to surgical clipping (modified Rankin score 3–6) at 1-year.
The advantage of coil embolization over surgical clipping varies widely with aneurysm location, but endovascular treatment seems beneficial for all sites.
There were less deaths in the first 7 years following coil embolization compared to surgical clipping (10.8% vs 13.7%). This survival benefit seemed to be consistent over time, and was statistically significant (log-rank p= 0.03).
Coil embolization is associated with less frequent MRI-detected superficial brain deficits and ischemic lesions at 1-year.
The 1- year rebleeding rate was 2.4% after coil embolization and 1% for surgical clipping. Confirmed rebleeding from the repaired aneurysm after the first year and up to year eight was low and not significantly different between coil embolization and surgical clipping (7 patients for coil embolization vs 2 patients for surgical clipping, log-rank p=0.22).
Observational studies showed that patients with SAH and good clinical grade had better 6-month outcomes and lower risk of symptomatic cerebral vasospasm after coil embolization compared to surgical clipping.
For unruptured intracranial aneurysms, there were no randomized controlled trials that compared coil embolization to surgical clipping. Large observational studies showed that:
The risk of rupture in unruptured aneurysms less than 10 mm in diameter is about 0.05% per year for patients with no pervious history of SAH from another aneurysm. The risk of rupture increases with history of SAH and as the diameter of the aneurysm reaches 10 mm or more.
Coil embolization reduced the composite rate of in hospital deaths and discharge to long-term or short-term care facilities compared to surgical clipping (Odds Ratio 2.2, 95% CI 1.6–3.1, p<0.001). The improvement in discharge disposition was highest in people older than 65 years.
In-hospital mortality rate following treatment of intracranial aneurysm ranged from 0.5% to 1.7% for coil embolization and from 2.1% to 3.5% for surgical clipping. The overall 1-year mortality rate was 3.1% for coil embolization and 2.3% for surgical clipping. One-year morbidity rate was 6.4% for coil embolization and 9.8% for surgical clipping. It is not clear whether these differences were statistically significant.
Coil embolization is associated with shorter hospital stay compared to surgical clipping.
For both ruptured and unruptured aneurysms, the outcome of coil embolization does not appear to be dependent on age, whereas surgical clipping has been shown to yield worse outcome for patients older than 64 years.
Angiographic Efficiency and Recurrences
The main drawback of coil embolization is its low angiographic efficiency. The percentage of complete aneurysm occlusion after coil embolization (27%–79%, median 55%) remains lower than that achieved with surgical clipping (82%–100%). However, about 90% of coiled aneurysms achieve near total occlusion or better. Incompletely coiled aneurysms have been shown to have higher aneurysm recurrence rates ranging from 7% to 39% for coil embolization compared to 2.9% for surgical clipping. Recurrence is defined as refilling of the neck, sac, or dome of a successfully treated aneurysm as shown on an angiogram. The long-term clinical significance of incomplete occlusion following coil embolization is unknown, but in one case series, 20% of patients had major recurrences, and 50% of these required further treatment.
A large international randomized trial reported that the survival benefit from coil embolization was sustained for at least 7 years. The rebleeding rate between year 2 and year 8 following coil embolization was low and not significantly different from that of surgical clipping. However, high quality long-term angiographic evidence is lacking. Accordingly, there is uncertainty about long-term occlusion status, coil durability, and recurrence rates. While surgical clipping is associated with higher immediate procedural risks, its long-term effectiveness has been established.
Indications and Contraindications
Coil embolization offers treatment for people at increased risk for craniotomy, such as those over 65 years of age, with poor clinical status, or with comorbid conditions. The technology also makes it possible to treat surgical high-risk aneurysms.
Not all aneurysms are suitable for coil embolization. Suitability depends on the size, anatomy, and location of the aneurysm. Aneurysms more than 10 mm in diameter or with an aneurysm neck greater than or equal to 4 mm are less likely to achieve total occlusion. They are also more prone to aneurysm recurrences and to complications such as coil compaction or parent vessel occlusion. Aneurysms with a dome to neck ratio of less than 1 have been shown to have lower obliteration rates and poorer outcome following coil embolization. Furthermore, aneurysms in the middle cerebral artery bifurcation are less suitable for coil embolization. For some aneurysms, treatment may require the use of both coil embolization and surgical clipping or adjunctive technologies, such as stents and balloons, to obtain optimal results.
Information from 3 countries indicates that coil embolization is a rapidly diffusing technology. For example, it accounted for about 40% of aneurysm treatments in the United Kingdom.
In Ontario, coil embolization is an insured health service, with the same fee code and fee schedule as open surgical repair requiring craniotomy. Other costs associated with coil embolization are covered under hospitals’ global budgets. Utilization data showed that in 2004-2005, coil embolization accounted for about 38% (251 cases) of all intracranial aneurysm repairs in the province. With the 2005 publication of the positive long-term survival data from the International Subarachnoid Aneursym Trial, the pressure for diffusion will likely increase.
Recent economic studies show that treatment of unruptured intracranial aneurysms smaller than 10 mm in diameter in people with no previous history of SAH, either by coil embolization or surgical clipping, would not be effective or cost-effective. However, in patients with aneurysms that are greater than or equal to 10 mm or symptomatic, or in patients with a history of SAH, treatment appears to be cost-effective.
In Ontario, the average device cost of coil embolization per case was estimated to be about $7,500 higher than surgical clipping. Assuming that the total number of intracranial aneurysm repairs in Ontario increases to 750 in the fiscal year of 2007, and assuming that up to 60% (450 cases) of these will be repaired by coil embolization, the difference in device costs for the 450 cases (including a 15% recurrence rate) would be approximately $3.8 million. This figure does not include capital costs (e.g. $3 million for an angiosuite), additional human resources required, or costs of follow-up. The increase in expenditures associated with coil embolization may be offset partially, by shorter operating room times and hospitalization stays for endovascular repair of unruptured aneurysms; however, the impact of these cost savings is probably not likely to be greater than 25% of the total outlay since the majority of cases involve ruptured aneurysms. Furthermore, the recent growth in aneurysm repair has predominantly been in the area of coil embolization presumably for patients for whom surgical clipping would not be advised; therefore, no offset of surgical clipping costs could be applied in such cases. For ruptured aneurysms, downstream cost savings from endovascular repair are likely to be minimal even though the savings for individual cases may be substantial due to lower perioperative complications for endovascular aneurysm repair.
The two Guidance documents issued by the National Institute of Clinical Excellence (UK) in 2005 support the use of coil embolization for both unruptured and ruptured (SAH) intracranial aneurysms, provided that procedures are in place for informed consent, audit, and clinical governance, and that the procedure is performed in specialist units with expertise in the endovascular treatment of intracranial aneurysms.
For people in good clinical condition following subarachnoid hemorrhage from an acute ruptured intracranial aneurysm suitable for either surgical clipping or endovascular repair, coil embolization results in improved independent survival in the first year and improved survival for up to seven years compared to surgical clipping. The rebleeding rate is low and not significantly different between the two procedures after the first year. However, there is uncertainty regarding the long-term occlusion status, durability of the stent graft, and long-term complications.
For people with unruptured aneurysms, level 4 evidence suggests that coil embolization may be associated with comparable or less mortality and morbidity, shorter hospital stay, and less need for discharge to short-term rehabilitation facilities. The greatest benefit was observed in people over 65 years of age. In these patients, the decision regarding treatment needs to be based on the assessment of the risk of rupture against the risk of the procedure, as well as the morphology of the aneurysm.
In people who require treatment for intracranial aneurysm, but for whom surgical clipping is too risky or not feasible, coil embolization provides survival benefits over surgical clipping, even though the outcomes may not be as favourable as in people in good clinical condition and with small aneurysms. The procedure may be considered under the following circumstances provided that the aneurysm is suitable for coil embolization:
Patients in poor/unstable clinical or neurological state
Patients at high risk for surgical repair (e.g. people>age 65 or with comorbidity), or
Aneurysm(s) with poor accessibility or visibility for surgical treatment due to their location (e.g. ophthalmic or basilar tip aneurysms)
Compared to small aneurysms with a narrow neck in the anterior circulation, large aneurysms (> 10 mm in diameter), aneurysms with a wide neck (>4mm in diameter), and aneurysms in the posterior circulation have lower occlusion rates and higher rate of hemorrhage when treated with coil embolization.
The extent of aneurysm obliteration after coil embolization remains lower than that achieved with surgical clipping. Aneurysm recurrences after successful coiling may require repeat treatment with endovascular or surgical procedures. Experts caution that long-term angiographic outcomes of coil embolization are unknown at this time. Informed consent for and long-term follow-up after coil embolization are recommended.
The decision to treat an intracranial aneurysm with surgical clipping or coil embolization needs to be made jointly by the neurosurgeon and neuro-intervention specialist, based on the clinical status of the patient, the size and morphology of the aneurysm, and the preference of the patient.
The performance of endovascular coil embolization should take place in centres with expertise in both neurosurgery and endovascular neuro-interventions, with adequate treatment volumes to maintain good outcomes. Distribution of the technology should also take into account that patients with SAH should be treated as soon as possible with minimal disruption.