This editorial was developed for AAN.com, which is publishing expert opinions on a variety of hot topics in neurology.
Joey English, MD, PhD
Assistant Professor, UCSF Neurology and Neurocritical Care and Fellow, UCSF Interventional Neuroradiology
Author Disclosure Statement
The International Subarachnoid Aneurysm Trial (ISAT) was a multicenter, prospective, randomized controlled trial of endovascular coiling versus neurosurgical clipping in patients with ruptured cerebral aneurysms suitable for either therapy. This landmark study demonstrated superior one-year clinical outcomes associated with coiling. However, many proponents of clipping have criticized these results for the lack of longer-term follow-up, particularly as related to the durability of coiling for preventing rebleeding, and because of the need for and risks of surveillance angiography and possible retreatment. These latter concerns, combined with the expense of detachable coils, have also raised questions about the cost effectiveness of coiling, compared to clipping. Recent follow-up data from ISAT have addressed many of these issues, providing additional strong support for the superiority of endovascular coiling over neurosurgical clipping in the treatment of ruptured cerebral aneurysms having favorable anatomy for either treatment.
The initial ISAT results were published in 2002(1) after a planned interim analysis favoring endovascular coiling led to an early termination in enrollment, with the full data set reported in 2005 after a mean follow-up of four years(2).
Based upon the initial ISAT results, the American Society of Interventional and Therapeutic Neuroradiology, and the American Society of Neuroradiology have issued a joint position statement that "patients with subarachnoid hemorrhage and aneurysm anatomy indicating a high likelihood of success with endovascular therapy should be offered that option" as a primary treatment(3). The statement noted, however, that its recommendation "must be tempered by limited data for long-term durability beyond 1 year."
The position statement of the American Association of Neurological Surgeons and Congress of Neurological Surgeons struck a much more conservative tone, noting that a chronic trend toward greater rebleeding following coiling could cause early differences in outcome to "soon disappear," and that it was simply too soon to extrapolate these early data. The neurosurgical community would "await with interest the long-term follow-up data on the 2143 ISAT patients." Longer-term data now exist.
In ISAT, surveillance angiography was required at 6 months for coiled patients and requested for clipping patients, and obtained in 88% and 46% of these respective populations. Additional angiography was then obtained at variable intervals. (In practice, most coiled aneurysms are restudied 6–12 months after initial treatment, while clipped aneurysms are often reevaluated in the immediate post-operative period.) Patients were also followed annually for clinical events (including rebleeding) over 4.5 to 12 years.
|Total number retreated||8.6% (94/1096)||0.9% (9/1012)|
|Mean interval to retreatment|
|Overall||20.7 months||5.7 months|
|After rebleeding||41.4 months||2.3 months|
|Overall risk of rebleeding||0.6% (7/1096)||0.3% (3/1012|
In summary, the need for delayed retreatment was significantly higher in coiled patients, though this is influenced in small part by the lower frequency of surveillance angiography in clipped patients. Importantly, no long-term adverse events or impact on clinical outcomes were seen in these retreatment cases. Finally, the long-term risk of rebleeding after initial coiling was low, and not significantly different from that seen with clipping (p=0.3). The observation that late rebleeding following coiling occurred in a much more delayed fashion, however, suggests that ongoing surveillance angiography may be required in patients with coiled aneurysms.
Might the procedural costs associated with coiling—particularly the cost of the coils and of surveillance angiograms—offset the financial benefits of shorter intensive care unit/hospital stays during the initial treatment phase? Recent cost analyses at several ISAT centers suggest that, while this offset is in fact seen, the overall costs of initial and subsequent treatments in the 12–24-month follow-up period are not significantly different between coiling and clipping(5).
It is also important to note that these studies evaluated health care costs at ISAT centers, but did not take into account important societal factors such as the impact on employment or the cost of long-term care for survivors. In the UK, ISAT patients in the coiling arm were significantly more likely to return to work in the first year of recovery; the trend continued to favor coiling at 24 months, though this was no longer significant. While no comparative data are available concerning the costs of long-term health care, the percentage of ISAT survivors with an mRS of 3-5 (15.5% coiled, 21% clipped) and the observation that seizures are more common following clipping suggest that the expenditures and overall burden to society are likely greater in the clipping population. These issues are particularly important given that the average age of patients in ISAT was 52 years.
Additional support comes from a retrospective cohort study of the treatment of unruptured cerebral aneurysms (1881 clip patients, 654 coil patients) at 249 United States hospitals between 1998 and 2000(6). Coiling was associated with significantly lower mortality (0.9% versus 2.5%), adverse outcomes (6.6% versus 13.2%), length of stay (4.5 days versus 7.4 days) and lower hospital charges ($42,044 versus $47,567). This study also suggests that one criticism of ISAT by the AAN's position statement—namely that the "degree of sub-specialization of neurovascular surgeons" in the US dictates that the largely European-based ISAT "may not be applicable" to US patients—is not particularly valid.
At a minimum, even when considering the costs of endovascular therapy and need for surveillance angiograms, the overall long-term expense of coiling is comparable to clipping. The bigger picture, including the consideration of lost employment and long-term health care costs, suggests that coiling is more cost-effective than clipping in patients with ruptured cerebral aneurysms. The most important bottom line, however, is not simply financial, but clinical. The most recent data from the ISAT demonstrate that the superior clinical outcomes following coiling are maintained for up to seven years(2).
Factors other than aneurysm anatomy are also important, including age, clinical grade, comorbid medical conditions, and aneurysm location. While largely unstudied, the less invasive nature of endovascular coiling is likely to be favored with older age, poor clinical grade, serious comorbid medical conditions, and certain aneurysm locations (e.g., basilar tip). Despite the continued evolution of endovascular techniques (e.g., stent-assisted coiling of wide-necked aneurysms), however, some ruptured aneurysms may not be favorable for endovascular coiling. Aneurysms of the middle cerebral bifurcation, for example, often have complex anatomy for coiling and may be better suited for clipping (less than 15% of aneurysms in ISAT were at this location). For aneurysms with favorable anatomy for coiling, however, the arguments for primary surgical clipping continue to fall.
Dr. English has nothing to disclose.