Rates of Delayed Rebleeding From Intracranial Aneurysms Are Low After Surgical and Endovascular Treatment
Background and Purpose— Although results of the randomized International Subarachnoid Aneurysm Trial suggested that coil embolization was superior to surgical clipping 1 year after treatment, a paucity of data on long-term outcomes has been a major concern.
Methods— In an ambidirectional cohort study, 9 institutions with expertise in intracranial aneurysm treatment identified all ruptured saccular aneurysms treated 1996 to 1998. After an initial medical record review, all patients meeting entry criteria were contacted by postal questionnaire or telephone. Possible reruptures were adjudicated independently by a neurologist, a neurosurgeon, and a neurointerventional radiologist. Rates of delayed (>1 year) and early rerupture and retreatment were evaluated using Kaplan-Meier survival analysis and the log-rank test.
Results— A total of 1010 patients (711 surgically clipped, 299 treated with coil embolization) were included. Patients treated with coil embolization were older, more likely to have smaller aneurysms arising from the posterior circulation, and less likely to have middle cerebral artery aneurysms. Rerupture of the index aneurysm after 1 year occurred in 1 patient treated with coil embolization during 904 person-years of follow-up (annual rate 0.11%) and in no patients treated with surgical clipping during 2666 person-years (P=0.11). Aneurysm retreatment after 1 year was more frequent in patients treated with coil embolization (P<0.0001), but major complications were rare during retreatment.
Conclusions— Rerupture of aneurysms treated by either coil embolization or surgical clipping is rare after the first year. Late retreatment is more common after coil embolization than after clipping but complication rates are low. Thus, late events are unlikely to overwhelm differences between procedures at 1-year follow-up.
After subarachnoid hemorrhage attributable to rupture of an intracranial aneurysm, the risk of rerupture is high unless the aneurysm is treated. The most common methods for treating intracranial aneurysms are surgical clipping and endovascular embolization with detachable coils. The International Subarachnoid Aneurysm Trial (ISAT) compared these techniques in randomized patients with recent intracranial aneurysm rupture.1,2 At 1-year follow-up, death or disability was less frequent among those treated with coil embolization. However, there were more reruptures during the first year of follow-up among patients treated with coil embolization. Though the outcomes of reruptures during the first year were captured in the 1-year primary outcome, there were also more reruptures after the first year in those randomized to coil embolization (seven reruptures during 3258 person-years of follow-up; rate 0.21% per year, exact binomial 95% CI, 0.09% to 0.44%) compared with those treated surgically (2, one of whom had crossed over to coil embolization, during 3107 person-years; rate 0.03%, 95% CI, 0% to 0.18%, among those actually treated with clipping). No other studies have directly compared the effectiveness of surgical clipping and endovascular coil embolization in terms of reducing subsequent rebleeding rates, the major goal of treating aneurysms.
Although the ISAT findings have led to increased use of coil embolization, many have been concerned about the limited information on the long-term efficacy of coils in reducing risk of rebleeding.3–7 As a consequence of this and other concerns, many institutions continue to prefer surgical clipping to coil embolization in patients with aneurysms that can be treated by either method.
The Cerebral Aneurysm Rerupture After Treatment (CARAT) study was designed to directly compare rerupture rates after subarachnoid hemorrhage in patients treated initially with coil embolization and surgical clipping. Using an ambidirectional cohort design, all patients treated at 9 US hospitals from 1996 to 1998 were followed for up to 9 years to determine whether delayed rerupture of treated aneurysms could overwhelm the initial short-term benefits of coil embolization. We hypothesized that rates of rerupture >1 year after treatment would be very low for patients treated with both coil embolization and surgical clipping.
Investigators at 9 hospitals in the western US that used both coil embolization and surgical clipping to treat >30 patients with subarachnoid hemorrhage each year were asked to participate; all agreed. All patients discharged between January 1, 1996 and December 31, 1998 with a primary diagnosis of subarachnoid hemorrhage were identified by a medical record search through hospital administrative databases. Detailed medical records were reviewed. Patients were included if subarachnoid hemorrhage was attributable to rupture of an intracranial aneurysm and a treatment attempt of this index aneurysm was made with surgery or endovascular coiling, at the discretion of the treating physicians, but not both. Patients were excluded if any of the following was present: (1) age younger than 18 years at follow-up; (2) no clip or coil was placed; (3) the aneurysm was not saccular (ie, it was fusiform, dissecting, or a pseudoaneurysm); (4) an intracranial arteriovenous malformation or fistula was present; (5) vessel occlusion was used to treat the aneurysm; (6) an endovascular balloon was used for embolization; or (7) the patient did not have a known US Social Security number, thus preventing follow-up for mortality based on national records.
All study procedures were approved by human subject review boards at each participating institution. Patient consent was not required for inclusion in the cohort because risk to the patient was considered minimal, but consent was required, either by return postcard or by telephone, for follow-up contact.
Information about the patient, ruptured and unruptured aneurysms, and details of the procedure were abstracted from medical records. Any aneurysm considered a possible source of the initial subarachnoid hemorrhage and treated during the first procedure was defined as an index aneurysm, so some patients had >1 index aneurysm. Quality was monitored by comparing data with centralized abstraction of the first 5 medical records and of a random sample of 5% of subsequent records.
A variety of public resources were used to locate patients. The US Social Security Death Index is a publicly available resource for identifying whether someone registered with Social Security, which is the vast majority of US residents, is deceased.8 It captures an estimated 93% of all deaths of US citizens, and an even larger percentage of those registered with Social Security.9 The Social Security Death Index was searched by Social Security number, name, and date-of-birth. When at least 2 of these 3 search criteria were met for a patient, death certificates were obtained for confirmation and to establish cause of death. For those still alive, current contact information was extracted from institutional databases and, when this was found to be incorrect, from other public sources, including national search services that rely on billing and license information.
A postal questionnaire was mailed to all living patients along with a letter from the investigators at the institution of treatment. Questions were designed to detect any suggestion of recurrent subarachnoid hemorrhage, including medical history and symptoms. If there was no response, a second questionnaire was mailed 2 to 4 weeks later. Patients were subsequently telephoned if there was no response to mailings or if clarification of questionnaire responses was required.
For all instances of possible subarachnoid hemorrhage and all deaths, associated medical records for the patient were gathered. After masking of information that could reveal the treatment modality or the identity of the patient, records were reviewed independently by members of an adjudication panel composed of 1 neurologist, 1 neurosurgeon, and 1 neurointerventionalist. Adjudicators were asked to determine whether, more likely than not, the treated index aneurysm reruptured. Agreement of 2 of 3 reviewers was required to classify an event as a rerupture.
The prespecified primary outcome of the study was the annual risk of nonprocedural rebleeding from the index treated aneurysm after 1 year of follow-up based on Kaplan-Meier life-table analysis of all included patients, stratified by the modality first used to treat the index aneurysm. We were primarily interested in events after 1 year follow-up because these were not captured in the primary outcome of ISAT and could influence treatment decisions. Patients were classified based on the first treatment modality used in order to parallel an intention-to-treat analysis and account for potential problems related to failed initial treatment. All those surviving to 1-year follow-up were included, and subsequent death, first rerupture after 1 year, and last patient contact were defined as censoring events. Exact binomial confidence intervals of annual rates were calculated, and rates were compared by treatment modality using the log-rank test. Rates of rerupture during the first year of follow-up and rates of index aneurysm retreatment were determined using similar methods.
Baseline characteristics of patients were compared using the Wilcoxon rank-sum test for continuous and ordinal data and Fisher exact test for dichotomous variables. Kaplan-Meier survival analysis with log-rank tests was used to compare rates of rerupture and retreatment. All statistical analyses were performed with SAS (version 8e, SAS Institute).
Role of the Sponsor
The study was supported by an unrestricted grant from Boston Scientific to the University of California, San Francisco. The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report, and had no access to data before public presentation.
Overall, 1010 patients met inclusion criteria: 711 treated with surgical clipping and 299 with coil embolization. Patients treated with coil embolization were older and were more likely to have smaller and posterior-circulation aneurysms (Table 1).
Maximum duration of follow-up was 9.6 years (mean 4.4 years) for clipped patients (3127 person-years) and 8.9 years (mean 3.7 years) for coiled patients (1089 person-years). Overall, 241 patients (24%) died. The primary outcome of the study, rerupture of the index aneurysm >1 year after treatment, occurred in no patient treated with clipping and 1 treated with coiling (Table 2 and Figure 1). Annual rates of rerupture after 1 year were 0.11% (95% CI, 0% to 0.63%) for coiling and 0% (95% CI, 0% to 0.14%) for clipping (P=0.11). Nonprocedural hemorrhage from the index aneurysm in the first year was more frequent after coil embolization than after clipping (Table 2).
The single late rerupture occurred 14 months after coil embolization in a 41-year-old woman. Five aneurysms were initially identified and all were treated with coil embolization, but a 4-mm anterior-communicating–artery aneurysm was considered the likely source of bleeding and it was defined as the index. A follow-up angiogram showed interval growth of the index aneurysm, which was not considered amenable to further treatment; 2 new aneurysms were also identified. The patient returned with a subarachnoid hemorrhage, Hunt and Hess Grade II, 2 months later and the index aneurysm was established as the source based on inspection at the time of surgical clipping. The patient survived the event with no residual disability and all other aneurysms have been subsequently clipped.
Intraprocedural hemorrhage was reported more frequently in patients treated with surgical clipping (17.6% versus 4.5% for coiling; P<0.001). However, the difference between postoperative and preoperative Rankin scores were smaller in those with rupture during surgery (median change, 1 point worsening) compared with those with rupture during coil embolization (2 points worsening; P<0.001).
The index aneurysm was retreated during the first year in 23 patients (7.7%) treated with coil embolization and 12 (1.7%) treated with surgical clipping (Table 3 and Figure 2). After accounting for censoring, the annual rate of retreatment was 13.3% for coiling and 2.6% for clipping in the first year (P<0.0001). No clipped aneurysm was retreated after the first year, whereas for coiling the rate was 4.5% in the second year and 1.1% annually thereafter (Table 3). Overall, retreatment was complicated by a life-threatening or disabling event in 4/35 (11%) patients initially treated with coil embolization (groin pseudoaneurysm, third nerve palsy, gastrointestinal hemorrhage, and chronic headaches), and in 2/12 (17%) of those initially treated with clipping (both deaths).
In a sample of centers that treat a large number of intracranial aneurysms, we found that both surgical clipping and endovascular coil embolization were highly effective at reducing the risk of late rerupture, with annual hemorrhage rates of 0.11% per year for coil embolization and with no late reruptures among those treated with surgical clipping. For subarachnoid hemorrhage in patients in whom the ruptured aneurysm is untreated, annual rates of rerupture after the first year have been estimated to be 3.5%.10 Thus, both surgical clipping and coil embolization dramatically reduce risk of rebleeding.
We included all aneurysms treated during a specific time period rather than just those amenable to both procedures. Although this observational design can introduce important differences in pretreatment prognosis between the treatment groups, the sample is likely to be more representative than that of a randomized trial, such as ISAT, which tended to include smaller aneurysms in the anterior circulation.1 In fact, there were differences in aneurysm characteristics between those treated with surgical clipping and coil embolization, such as location and size, that may have influenced treatment durability. Nonetheless, our results for rebleeding were similar to those of ISAT, in which annual rates of rerupture after 1 year were 0.21% for those treated with coiling and 0.03% for those treated with clipping. This similarity suggests that ISAT results on rerupture are likely generalizable to patients treated at large US centers.
To gain a broader understanding of the implications of late rerupture on treatment efficacy, we combined our data on rerupture after 1 year with those from ISAT. In the 2 groups together, there was 1 rerupture during 5771 person-years of follow-up after surgical clipping (annual rate 0.02%, 95% CI, 0% to 0.10%) and 8 reruptures during 4162 person-years after coiling (annual rate 0.19%, 95% CI, 0.08% to 0.38%). Projecting these rates forward, assuming that 60% of reruptures are associated with new moderate disability or worse,11 we modeled whether the absolute 7.4% benefit in terms of reduced disability and death with coil embolization at 1 year demonstrated in ISAT (which captured complications from reruptures and retreatments during the first year)2 would be eliminated during long-term follow-up because of greater risk for rerupture after coil embolization. Based on these assumptions, equivalent disability and death would occur 73 years after treatment. Even at a rerupture rate of 0.38%, at the upper limit of the 95% CI, it would take 34 years for the early benefit of coil embolization to be eliminated by late rerupture. Thus, the 1-year results of ISAT are unlikely to be overwhelmed by difference in late rerupture risk during the normal lifespan of patients.
Patients initially treated with coil embolization were more likely to undergo retreatment of the index aneurysm. Most occurred during the first 2 years, but the rate of subsequent retreatment remained elevated at 1.1% per year afterward. No late retreatment led to death or disability, but the number of events was small. If we project rates of retreatment forward and assume rates of new moderate disability or death with retreatment are similar to those reported for coil embolization of unruptured aneurysms (6% at 1 year from the International Study of Unruptured Intracranial Aneurysms),2 retreatment would not be expected to overwhelm the early benefit of coiling even after 100 years of follow-up. Even with a rate of death or disability from retreatment of 20%, >30 years would need to elapse before the early benefit of coil embolization was eliminated. Thus, it is unlikely that a greater need for late retreatment with coil embolization will overwhelm its early benefit for most patients with subarachnoid hemorrhage.
Nonetheless, coil embolization is associated with greater risk of rerupture and late retreatment. These risks, as well as the necessity for follow-up imaging, should be discussed with patients and carefully integrated into the consent process. Furthermore, delayed retreatment justifies attempts to produce more durable endovascular treatments, such as with coated coils or adjunctive use of stents.12–14 Safer and more efficient methods of evaluating aneurysm occlusion during follow-up, such as by MR angiography or skull radiographs, could also reduce the impact of monitoring on patients.15–18
Our study has several limitations. First, there were important differences in patients treated with coil embolization and surgical clipping, particularly with respect to aneurysm location. These imbalances, inherent in observational studies, may have impacted relative rates of retreatment; however, with extremely low rates of late rerupture in both treatment groups, they did not contribute significantly to rerupture rate estimates. Second, the small number of reruptures prevented us from evaluating predictors of rerupture and from attempting to adjust for differences in patient characteristics. Third, radiographs were not available for review. We had access to imaging reports but methods for evaluating aneurysm occlusion and for assessing the need for retreatment almost certainly varied between institutions. Nonetheless, the importance of imaging results in assessing the success of treatment is questionable. Imaging results after treatment have not been associated with rates of rerupture, and only clinical events, such as rerupture and complications from retreatment, have clinical relevance.19 Finally, in order to obtain long-term follow-up, we included cases treated in the late 1990s, and techniques have evolved since then that may have altered rates of rerupture and retreatment.
University of California, San Francisco: S. Claiborne Johnston, MD, PhD, Principal Investigator (PI); Christopher F. Dowd, MD (Co-PI); Michael T. Lawton, MD (Co-PI); Daryl R. Gress, MD; Randall T. Higashida, MD; Van V. Halbach, MD; Shoujun Zhao, MD, PhD; Katherine H. Katsura, BS; Kristin J. Fong, BS; Vanja C. Douglas, MD; Rosalyn Ventura, MD; Jacob S. Elkins, MD; Mai N. Nguyen-Huynh, MD.
Barrow Neurological Institute of St Joseph’s Hospital and Medical Center: Cameron G. McDougall, MD (Site PI); Robert F. Spetzler, MD; Joseph M. Zabramski, MD; Heidi K. Jahnke, RN, BSN. Mayo Clinic: David G. Piepgras, MD (Site PI); Douglas A. Nichols, MD; Denise R. Gravenhof; Debra Herzig, RN. Houston Methodist Hospital: Michel E. Mawad, MD (Site PI); Denise Meyer, RN. Stanford University Medical Center: Gary K. Steinberg, MD, PhD (Site PI); Michael P. Marks, MD; Desiree Luu, RN; Hanna Yi, RN. University of California, Los Angeles: Gary R. Duckwiler, MD (Site PI); Neil A. Martin, MD; Henry Adapon, MD. University of Southern California: Steven L. Giannotta, MD (Site PI); Donald W. Larsen, MD; George P. Teitelbaum, MD; Dawn Fishback, PA-C; Evangeline Thomson, RN. University of Texas, Southwestern: Duke S. Samson, MD (Site PI); Phillip D. Purdy, MD; Robert E. Replogle, MD; Jerri Thomas, BS.
This work was funded by an unrestricted grant from Boston Scientific, Inc. Dr Johnston is supported by the National Institutes of Health (NS 02254).
- Received January 7, 2006.
- Revision received March 2, 2006.
- Accepted March 17, 2006.
Molyneux AJ, Kerr RS, Yu LM, Clarke M, Sneade M, Yarnold JA, Sandercock P. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised comparison of effects on survival, dependency, seizures, rebleeding, subgroups, and aneurysm occlusion. Lancet. 2005; 366: 809–817.
Sesso HD, Paffenbarger RS, Lee IM. Comparison of National Death Index and World Wide Web death searches. Am J Epidemiol. 2000; 152: 107–111.
Hill ME. Response to “Comparison of National Death Index and World Wide Web Death Searches.” Am J Epidemiol. 2001; 153: 719.
Hop JW, Rinkel GJ, Algra A, van Gijn J. Case-fatality rates and functional outcome after subarachnoid hemorrhage: a systematic review. Stroke. 1997; 28: 660–664.
Fiorella D, Albuquerque FC, McDougall CG. Durability of aneurysm embolization with matrix detachable coils. Neurosurgery. 2006; 58: 51–58;discussion 58–59.
Cloft HJ, Kallmes DF. Aneurysm packing with HydroCoil Embolic System versus platinum coils: initial clinical experience. Am J Neuroradiol. 2004; 25: 60–62.
Cottier JP, Bleuzen-Couthon A, Gallas S, Vinikoff-Sonier CB, Bertrand P, Domengie F, Barantin L, Herbreteau D. Follow-up of intracranial aneurysms treated with detachable coils: comparison of plain radiographs, 3D time-of-flight MRA and digital subtraction angiography. Neuroradiology. 2003; 45: 818–824.
Yamada N, Hayashi K, Murao K, Higashi M, Iihara K. Time-of-flight MR angiography targeted to coiled intracranial aneurysms is more sensitive to residual flow than is digital subtraction angiography. Am J Neuroradiol. 2004; 25: 1154–1157.
Hwang GB, Niimi Y, Setton A, Pryor J, Baltsavias G, Albert R, Hartman J. The accuracy of plain skull x-ray examination as a predictor of recanalization following Guglielmi Detachable Coil embolization in the treatment of cerebral aneurysms. Intervent Neuroradiol. 2000; 6: 195–202.
Johnston SC, Higashida RT, Barrow DL, Caplan LR, Dion JE, Hademenos G, Hopkins LN, Molyneux A, Rosenwasser RH, Vinuela F, Wilson CB. Recommendations for the endovascular treatment of intracranial aneurysms: a statement for healthcare professionals from the Committee on Cerebrovascular Imaging of the American Heart Association Council on Cardiovascular Radiology. Stroke. 2002; 33: 2536– 2544.