Building on controversial data published in New England Journal of Medicine last year, UC San Francisco researchers report that surgically treating unruptured cerebral aneurysms is unnecessary in a large subset of patients.
The UCSF analysis, combining results from numerous investigations, most prominently the NEJM report, showed that treating patients who have small, asymptomatic, unruptured cerebral aneurysms and no history of brain hemorrhage may increase the likelihood that these patients will experience deteriorated health or death.
"Most aneurysms do require surgical treatment, and every patient's situation is unique, so this complex issue needs to be discussed by a patient and his or her physician," said the lead author of the UCSF study, S. Claiborne Johnston, MD, an assistant professor of neurology. "However, our study, based on information that wasn't available before, suggests that in some cases aneurysms should not be treated."
An aneurysm is a widened, or ballooned, area of a blood vessel that results from a weakening in the wall of that vessel. Some four percent to six percent of adults harbor asymptomatic, unruptured cerebral aneurysms, which have an overall rupture rate of between 0.05 percent to 3 percent annually, depending on size. When they rupture, they cause bleeding in the brain. Thirty to 60 percent of patients die as a result, and 15 to 30 percent have permanent disability.
The findings build on results of the International Study of Unruptured Intracranial Aneurysms (ISUIA) reported in NEJM, which had indicated that small aneurysms do not rupture frequently and that the surgical and post-surgical risks associated with treating them were higher than had been thought. But the researchers presenting this study had left the interpretation of the findings up to other research groups, said Johnston. The UCSF study was reported in June in Neurology.
Physicians have debated the value of treating all unruptured cerebral aneurysms for years. Because assessing the outcome of patients depends on a variety of factors and requires balancing early and delayed risks, the issue has been examined through several different decision and cost-utility analyses. Until recently, these analyses have shown that aneurysm treatment is both advisable and cost-effective.
However, said Johnston, these analyses were based on estimates of rupture and complication rates that have been supplanted by newer, more precise estimates, and aneurysm characteristics were not considered. The UCSF study was a complex cost-utility analysis incorporating new data, most notably results of the ISUIA study.
The UCSF researchers developed a model in which they projected the health outcomes and financial costs associated with hypothetical patients with unruptured aneurysms who either received or did not receive treatment. Those receiving treatment either underwent the standard procedure in which a clip is placed on the neck of the aneurysm, or endovascular coil embolization, in which platinum coils are used to fill the aneurysm through a catheter inserted at the femoral artery. (The latter procedure is performed at only a few medical centers, including UCSF.)
The three options - the two treatments and no treatment -- were examined within eight hypothetical patient scenarios, distinguished by the size of a patient's aneurysm, the patient's history of ruptured aneurysms, and symptoms. The most recent health-outcomes data associated with each of these factors was then applied to each scenario. (Rupture rates were taken from the International Study of Unruptured Intracranial Aneurysms.)
While the study showed that aneurysm treatment was the best option for people with unruptured aneurysms that were at least 10 mm, it also showed that treatment increased the risk of a poorer outcome in people with aneurysms smaller than 10mm who had not had bleeding from a different aneurysm. The finding even accounted for the toll that anxiety might take on a patient who does not receive treatment for a small aneurysm. (The risk of rupture in these cases is similar to the risk of dying in an automobile accident, 1 in 5,000.)
The UCSF analysis relied to a large extent on the rupture-rate and risk data compiled through the ISUIA. Some physicians have questioned the ISUIA findings because they disagreed with prior studies. However, said Johnston, the UCSF researchers determined that the true numbers would need to be seven times larger for the coiling procedure and 15 times larger for the clipping procedure to reverse their conclusions. And that, said Johnston, is "highly unlikely."
Aneurysms can result from a congenital defect or develop later in life. One to six percent of adults harbor asymptomatic, unruptured cerebral aneurysms, which may be discovered incidentally when another aneurysm ruptures, or if they produce symptoms, such as headache or cranial nerve palsy, due to compression of neighboring structures.
Co-authors of the UCSF study were Daryl R. Gress, MD, UCSF associate professor of neurology and James G. Kahn, MD, MPH, UCSF associate adjunct professor at the UCSF Institute for Health Policy Studies.
The study was funded by the National Stroke Association and the National Institutes of Health.
Journal
Neurology