Endovascular coil embolization of intracranial aneurysms has become a widely accepted treatment alternative to surgical clip occlusion, especially after the results of the ISAT were published in 2002.17 In that trial, the coil-treated cohort had fewer poor clinical outcomes after 1 year of follow-up compared with patients who had undergone surgical clipping.
However, concerns have persisted regarding the applicability of ISAT, which excluded almost 80% of eligible aneurysms from the study population, thereby calling into question whether the study population accurately reflected the general population of patients with ruptured aneurysms.17,25 To minimize the problem of selection bias, the prospective, randomized, controlled BRAT was designed as an intent-to-treat study to include all eligible patients presenting with SAH. Regardless of treatment, primary outcome was based on the assigned treatment so that one treatment modality could not benefit comparatively from the crossover of poor-grade patients to the other group. The BRAT was also an attempt to determine whether patients with acute SAH could be randomized and treated effectively in a timely fashion in a multicenter trial.
It was hoped that this intent-to-treat analysis, which included all patients with aneurysmal SAH, would improve understanding of the applicability of the ISAT data and of the roles of surgical and endovascular treatment. The BRAT is an ongoing trial with a 10-year follow-up planned after completion of enrollment. This report presents the outcomes 3 years after treatment.
Methods
The study protocol and all facets of the study were approved and overseen by the Institutional Review Board of St. Joseph's Hospital and Medical Center, Phoenix, Arizona. The initial approval was given on November 12, 2002. The trial is registered at ClinicalTrials.gov (NCT01593267). The study protocol, which was reported in detail in the report of the 1st-year follow-up,15 is summarized below.
Patient Population
As previously reported,15 all patients between the ages of 18 and 80 years who were admitted to the intensive care unit with acute nontraumatic SAH (confirmed by CT or lumbar puncture) between March 2003 and January 2007 were eligible for participation in this study and were included if they or their health care decision surrogate consented. Of the 725 patients who were screened, 500 met the criteria to participate in the study. Consent was obtained from 28 of these patients in error, leaving 472 cases eligible for analysis. One patient withdrew consent, leaving 471 patients who were randomly assigned to treatment and whose data were available at discharge: 238 (mean age 53.1 ± 12.8 years, 70% female) were assigned to clipping and 233 (mean age 54.3 ± 12.0 years, 71% female) were assigned to coil embolization (Fig. 1).
Flowchart showing patient assignments and actual treatments. Clip/Clip = assigned to and treated by surgical clipping. Clip/Coil = assigned to surgical clipping and crossed over to coil embolization; Coil/Coil = assigned to coil embolization and treated by coil embolization; Coil/Clip = assigned to coil embolization and treated by surgical clipping; Consented = informed consent was given; Died before tx = moribund patients assigned to treatment group, but not treated due to their status; tx = treatment.
No anatomical inclusion or exclusion criteria were incorporated into the study design; consequently, patients with nonaneurysmal SAH were also enrolled. As previously reported,15 63 patients received no treatment. Six died before initiation of treatment (3 in each assigned group), and 57 patients had a nonaneurysmal SAH.15 Although our prior analysis of outcomes at Year 1 included these untreated patients, they have been excluded from the analysis of outcomes at Year 3, given that these outcomes cannot be associated with a treatment that did not occur. Of 408 patients that actually underwent treatment, 209 were in the clip group and 199 in the coil group (Fig. 1).
Patients were randomly assigned to undergo coil embolization or clipping as described previously.15 After a patient was assigned to treatment, the micro- or endovascular neurosurgeon reviewed the imaging studies to determine the appropriateness of treatment. Patients were crossed over to the alternate group at the discretion of the assigned treating physician. Of the patients initially randomized to clipping who were actually treated (n = 209), 4 were crossed over to coiling (crossover rate 1.9%). Of those randomized to coil embolization and actually treated (n = 199), 75 were crossed over to clipping (crossover rate 38%, Fig. 1).
The reasons for crossover from one treatment modality to the other varied. Of the 75 coil-to-clip crossover patients, 14 (19%) had a hematoma that required evacuation. The remaining patients were crossed over to clipping because they had anatomical features such as dissections judged to render endovascular coiling less appropriate or because the endovascular surgeon considered coiling to be disproportionately difficult compared with clipping. Patients were not crossed over from coiling to clipping based on patient grade or clinical condition.
Outcome Analysis
A research nurse practitioner acts as coordinator, oversees patient accrual and randomization, collects the follow-up data, and performs the mRS assessments. Outcome data have been collected at 6, 12, and 36 months. Further follow-up is scheduled for 6 and 10 years.
As reported in the 1st-year follow-up of BRAT,15 age; sex; ethnicity; comorbid conditions (diabetes, hypertension, smoking, and use of cocaine or methamphetamine); status at presentation; scores on the Glasgow Coma Scale, Hunt and Hess Scale, and Fisher Scale; and aneurysm size and location were also evaluated. With the exception of location, no significant differences were observed in these variables between the treatment groups.15
The primary outcome, the proportion of patients with a mRS score of 3 to 6 (indicating an outcome of dependency or death) at 3 years, was analyzed on an intent-to-treat basis. The dichotomized mRS outcome was chosen so that results could be compared with those from the ISAT.
Secondary analyses included outcome based on actual treatment as opposed to the primary intent-to-treat analysis and outcomes of patients who crossed over from their assigned group to the alternative treatment group. Sentinel events during a patient's postoperative course included aneurysm-related rebleeding, retreatment, and death. Additional secondary analyses included aneurysm size and degree of aneurysm obliteration. The degree of aneurysm obliteration was assessed from imaging data by an independent nontreating neuroradiologist (R.C.W.).
Statistical Analysis
Our primary analysis, which examined the risk of a poor outcome (defined as mRS score > 2, signifying death or dependency at 1 year), consisted of an intent-to-treat analysis in those cases in which mRS data were reported at 3 years postprocedure (349 individuals); this analysis was performed using logistic regression methods, with assigned treatment as a predictor of outcome. Although our analysis of outcomes at 1 year included patients who were randomized but not treated (due to nonaneurysmal SAH or death before treatment), these patients were excluded from the current analysis for 2 reasons. First, there was no difference in the outcomes between them at 3 years. Second, they were not treated and therefore were not subject to the decision to treat as assigned or cross over to the alternate treatment.
We tested the null hypothesis that no difference in outcome would be detected between the endovascular treatment arm and the surgical treatment arm of the study, based on the dichotomized mRS score. Because our test was 2-sided, a statistically significant difference would be interpreted as evidence in favor of the alternative hypothesis that the 2 treatments were not equivalent.
We then examined multivariable models by adjusting our primary predictor (assigned treatment) by age > 50 years, baseline Hunt and Hess score > II (indicating a poorer neurological status), and location (anterior vs posterior). We assessed for interactions between treatment modality and both age > 50 years and Hunt and Hess score > II by including these terms in the model. Because these interaction terms were not significant, they were not included in the final model.
Secondary analyses of outcome based on actual treatment were conducted using logistic regression models as described above for assigned treatment. Rates of rebleeding and retreatment were compared using odds ratios calculated from 2 × 2 tables using Stata v12.
Results
At the 3-year follow-up, 397 patients were available for independent evaluation by the study coordinator. Of these, 200 had been initially assigned to clipping and 197 to coil embolization, and 349 had actually undergone treatment.
Primary Outcome
At the 3-year follow-up, 64 (35.8%) of the patients assigned to clipping and 51 (30%) of the patients assigned to coiling had a poor outcome (mRS > 2), a difference that failed to reach significance in unadjusted analysis (OR 1.30, 95% CI 0.83–2.04, p = 0.25) or adjusted analysis (OR 1.31, 95% CI 0.82–2.10, p = 0.25). Therefore, the effect of clipping on risk of poor outcome observed at 1 year (OR 1.68, 95% CI 1.06–2.67, p = 0.03) had decreased and was no longer significant (Table 1).15
Patients with mRS scores > 2 across BRAT follow-up by assigned treatment
| Time Point | Available for Analysis | Coil-Assigned (mRS >2) | Clip-Assigned (mRS >2) | OR | 95% CI | p Value* | ||
|---|---|---|---|---|---|---|---|---|
| n | % | n | % | |||||
| randomization | 408 | 199 | 209 | |||||
| discharge | 406 | 127/198 | 64.1 | 147/208 | 70.7 | 1.35 | 0.89–2.05 | 0.16 |
| 6 mos | 341 | 40/171 | 23.4 | 62/170 | 36.5 | 1.88 | 1.18–3.03 | 0.009 |
| 1 yr | 358 | 42/174 | 24.1 | 64/184 | 34.8 | 1.68 | 1.06–2.67 | 0.03 |
| 3 yrs | 349 | 51/170 | 30.0 | 64/179 | 35.8 | 1.30 | 0.83–2.04 | 0.25 |
| 3 yrs–CF† | 366 | 51/178 | 28.7 | 64/188 | 34.0 | 1.29 | 0.83–2.00 | 0.27 |
Odds ratios, confidence intervals, and p values obtained from unadjusted logistic regression.
CF = carry forward; includes patients seen at 1 year but not at 3 years.
Crossover Analysis
Of the 64 patients who crossed to clipping from the intent-to-coil group and had available data at Year 3, 42% had a poor outcome at the 3-year follow-up compared with 23% of the uncrossed coil group (p = 0.007). When the coil-to-clip crossover patients were compared with those assigned to the clip group, there was no significant difference in outcome (poor outcome in 42% vs 34%, respectively, p = 0.26). All 4 patients who crossed over to coiling from clipping had a poor outcome at 3 years (Table 2). At 3 years, 77% of the coil-to-clip crossover patients with a hematoma had an mRS score > 2. The intent-to-treat analysis counted these poor outcomes as attributable to the coiling treatment arm.
Patients with mRS scores > 2 based on actual treatment
| Time Point | Available for Analysis | Coil-Coil (124/199) mRS >2 | Coil-Clip (75/199) mRS >2 | Clip-Clip (205/209) mRS >2 | Clip-Coil (4/209) mRS >2 | ||||
|---|---|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | n | % | ||
| time of treatment | 408 | ||||||||
| discharge | 406 | 72/124 | 58.1 | 55/74 | 74.3 | 144/204 | 70.6 | 3/4 | 75.0 |
| 6 mos | 341 | 18/108 | 16.7 | 22/63 | 34.9 | 59/167 | 35.3 | 3/3 | 100.0 |
| 1 yr | 358 | 20/109 | 18.4 | 22/65 | 33.9 | 61/180 | 33.9 | 3/4 | 75.0 |
| 3 yrs* | 349 | 24/106 | 22.6 | 27/64 | 42.2 | 60/175 | 34.3 | 4/4 | 100.0 |
| 3 yrs–CF† | 366 | 24/111 | 21.6 | 27/64 | 42.2 | 60/184 | 32.6 | 4/4 | 100.0 |
Coil-coil compared to coil-clip at 3 years: p = 0.007. Coil-clip compared to clip-clip at 3 years: p = 0.26. Coil-coil compared to clip-clip at 3 years: p = 0.04.
Includes patients seen at 1 year but not at 3 years.
Aneurysm Size
The number of aneurysms assigned to treatment group by location and size is detailed in Table 3. Overall, the median size of all aneurysms was 6 mm, and aneurysm size was evenly distributed between the coil-assigned (median 6, IQR 4–8, range 1–30) and clip-assigned groups (median 6, IQR 4–8, range 1.5–23). The median size of the aneurysms in the coil-to-clip crossover group (median 5 mm, IQR 3–8 mm, range 1–20 mm) was significantly smaller (p = 0.009) than in the actual coil-treated group (median 6 mm, IQR 5–8 mm, range 2–30 mm). This difference was largely accounted for by the fact that aneurysms 4 mm or less in diameter comprised 35% (97/280) of all aneurysms clipped compared with 23% (30/128) of coiled aneurysms (p = 0.02). Aneurysms 12 mm or greater in diameter accounted for 10.3% (29/280) of aneurysms actually clipped, while these larger aneurysms made up 7.8% (10/128) of all coiled aneurysms, a difference that was not significant (p = 0.42, Table 4).
Number of aneurysms assigned to treatment group by location and size*
| Location | n (%) | Clip Group (n = 209) | Coil Group (n = 199) | Median Aneurysm Size (mm) | IQR | Range | ||
|---|---|---|---|---|---|---|---|---|
| Assigned | Actual | Assigned | Actual | |||||
| anterior circulation | ||||||||
| ICA region | ||||||||
| proximal ICA | 24 (5.9) | 15 | 22 | 9 | 2 | 5 | 3–9 | 2–20 |
| AChA | 9 (2.2) | 4 | 4 | 5 | 5 | 7 | 4–9 | 4–12 |
| PCoA | 84 (20.6) | 39 | 49 | 45 | 35 | 6 | 4.4–8 | 2–16 |
| ICA terminus | 12 (2.9) | 5 | 10 | 7 | 2 | 3.5 | 2–5 | 1–20 |
| MCA region | 60 (14.7) | 29 | 50 | 31 | 10 | 7 | 5–11 | 2–20 |
| ACA region | ||||||||
| ACoA region | 137 (33.6) | 75 | 97 | 62 | 40 | 5 | 4–7 | 1–18 |
| pericallosal | 13 (3.2) | 4 | 8 | 9 | 5 | 5 | 4–9 | 3–12 |
| posterior circulation | ||||||||
| BA tip | 19 (4.7) | 8 | 10 | 11 | 9 | 8 | 6–14 | 1.5–17 |
| P1 | 1 (0.2) | 0 | 0 | 1 | 1 | 13 | NA | NA |
| P2 | 3 (0.7) | 0 | 0 | 3 | 3 | 8 | 8–11 | 8–11 |
| BA trunk | 3 (0.7) | 1 | 2 | 2 | 1 | 8 | 4–9 | 4–9 |
| SCA | 6 (1.5) | 5 | 5 | 1 | 1 | 4 | 3–6 | 2–7 |
| AICA | 1 (0.2) | 1 | 1 | 0 | 0 | 1.5 | NA | NA |
| PICA | 21 (5.1) | 18 | 19 | 3 | 2 | 5 | 4–6 | 3–15 |
| VA-BA junction | 5 (1.2) | 2 | 1 | 3 | 4 | 5 | 4–7 | 4–23 |
| VA | 10 (2.5) | 3 | 2 | 7 | 8 | 4 | 3–8 | 2–30 |
| total | 408 | 209 | 280 | 199 | 128 | |||
ACA = anterior cerebral artery; AChA = anterior choroidal artery; ACoA = anterior communicating artery; AICA = anterior inferior cerebellar artery; BA = basilar artery; ICA = internal carotid artery; MCA = middle cerebral artery; NA = not applicable; PCoA = posterior communicating artery; PICA = posterior inferior cerebellar artery; P1 = first segment of the posterior cerebral artery; P2 = second segment of the posterior cerebral artery; SCA = superior cerebellar artery; VA = vertebral artery.
Assigned and actual aneurysm size distribution*
| Treatment | <4 mm | 4–6 mm | 6–8 mm | 8–10 mm | 10–12 mm | >12 mm | Total of >10 mm |
|---|---|---|---|---|---|---|---|
| assigned | |||||||
| clip | 8.3% | 15.2% | 12.0% | 7.4% | 2.7% | 5.6% | 8.3% |
| coil | 9.8% | 13.7% | 11.3% | 5.6% | 4.2% | 4.2% | 8.3% |
| actual | |||||||
| clip | 14.2% | 19.4% | 14.2% | 9.6% | 3.9% | 7.4% | 11.3% |
| coil | 3.9% | 9.6% | 9.1% | 3.4% | 2.9% | 2.5% | 5.4% |
Aneurysm size ranges include the left end point and not the right.
Distribution and crossovers of 408 anterior and posterior circulation aneurysms
| Anterior Circulation n = 339 | Posterior Circulation n = 69 | ||
|---|---|---|---|
| assigned clip n = 171 | assigned coil n = 168 | assigned clip n = 38 | assigned coil n = 31 |
| actual clip n = 240 | actual coil n = 99 | actual clip n = 40 | actual coil n = 29 |
| clip-to-coil crossover = 0.6% | clip-to-coil crossover = 8.0% | ||
| coil-to-clip crossover = 42% | coil-to-clip crossover = 16% | ||
Aneurysm Location
In the ISAT trial, 97.3% of aneurysms were in the anterior circulation.17 When the entire BRAT cohort of treated aneurysms was analyzed by aneurysm location there was a marked difference in the percentage of crossovers (Table 5) and outcomes. Among patients with anterior circulation aneurysms, there were no significant differences in mRS scores at any time point regardless of treatment (Table 6). Among those with posterior circulation aneurysms, however, mRS scores were significantly better after endovascular management than after surgical treatment at every time point (Table 7).
Patients with anterior circulation aneurysms with an mRS score > 2
| Time Point | Available for Analysis (n) | Assigned Treatment | p Value* | |||
|---|---|---|---|---|---|---|
| Coil (n = 168) | Clip (n = 171) | |||||
| n | % | n | % | |||
| treatment | 339 | |||||
| discharge | 337 | 107/167 | 64.1 | 114/170 | 67.1 | 0.56 |
| 6 mos | 285 | 36/145 | 24.8 | 42/140 | 30.0 | 0.33 |
| 1 yr | 297 | 37/146 | 25.3 | 43/151 | 28.5 | 0.57 |
| 3 yrs | 287 | 44/142 | 31.0 | 43/145 | 29.7 | 0.81 |
| 3 yrs–CF† | 304 | 44/150 | 29.3 | 43/154 | 27.9 | 0.79 |
| Actual Treatment | ||||||
| 3 yrs–CF† | 304 | 21/89 | 23.6 | 66/215 | 30.7 | 0.21 |
Values calculated using χ2 tests.
Includes patients seen at 1 year but not at 3 years.
Patients with posterior circulation aneurysms with an mRS score > 2
| Time Point | Available for Analysis (n) | Assigned Treatment | p Value* | |||
|---|---|---|---|---|---|---|
| Coil (n = 31) | Clip (n = 38) | |||||
| n | % | n | % | |||
| treatment | 69 | |||||
| discharge | 69 | 20/31 | 64.5 | 33/38 | 86.8 | 0.03 |
| 6 mos | 56 | 4/26 | 15.4 | 20/30 | 66.7 | <0.0001 |
| 1 yr | 61 | 5/28 | 17.9 | 21/33 | 63.6 | <0.0001 |
| 3 yrs | 62 | 7/28 | 25.0 | 21/34 | 61.8 | 0.004 |
| 3 yrs–CF† | 62 | 7/28 | 25.0 | 21/34 | 61.8 | 0.004 |
| Actual Treatment | ||||||
| 3 yrs–CF† | 62 | 7/26 | 26.9 | 21/36 | 58.3 | 0.01 |
Values calculated using χ2 tests.
Includes patients seen at 1 year but not at 3 years.
The majority of aneurysms that were crossed over from coiling involved the anterior circulation (n = 70), and only 5 were in the posterior circulation. Of the 408 treated patients, 339 (83%) had anterior circulation aneurysms and 69 (17%) had posterior circulation aneurysms (Tables 3 and 5).
Of the anterior circulation aneurysms, 171 were assigned to clipping and 168 were assigned to coil embolization with a fairly equal distribution of anatomical locations between the treatment groups (Fig. 2A). Of these, 240 were actually treated with clip occlusion and 99 with coil embolization. One patient (1/171, 0.6%) in the clip group crossed over to the coil group while 70 patients (70/168, 42%) in the coil group crossed over to the clip group.
The number of aneurysms at each location in the (A) anterior and (B) posterior circulation randomly assigned to surgical clipping or coil embolization. Aneurysms in the anterior circulation were distributed fairly equally between the two treatment groups. Unfortunately, the distribution of aneurysms involving the posterior circulation aneurysms was poorly matched. AICA = anterior inferior communicating artery; Ant Choroidal = anterior choroidal artery; Ant Com. = anterior communicating artery; BA = basilar artery; ICA = internal carotid artery; MCA = middle cerebral artery; PICA = posterior inferior communicating artery; Post Com. = posterior communicating artery; Prox = proximal; P1 = first segment of the posterior cerebral artery; P2 = second segment of the posterior cerebral artery; SCA = superior cerebellar artery; VA = vertebral artery.
Of the 69 posterior circulation aneurysms, 38 were randomized to the clip group and 31 were randomized to the coil group. Of these, 40 were actually clipped and 29 were treated with coil embolization. Three patients (3/38, 8%) in the clip group crossed over to the coil group while 5 patients (5/31, 16%) in the coil group crossed over to the clip group (Tables 5 and 8). With the exception of the basilar tip aneurysms, the randomization of the posterior circulation aneurysms was unexpectedly skewed. Five of the 6 superior cerebellar artery aneurysms and 18 of the 21 posterior inferior cerebellar artery aneurysms were randomized to clipping (Fig. 2B). In contrast, the majority of posterior cerebral, vertebral, and basilar artery aneurysms were assigned to coil embolization (15 vs 5).
Aneurysm location by assigned and actual treatment
| Aneurysm Location | Assigned Number | Actual Number | Clip-to-Coil Crossover n (%) | Coil-to-Clip Crossover n (%) |
|---|---|---|---|---|
| anterior circulation | ||||
| clip | 171 | 240 | 1/171 (0.6) | |
| coil | 168 | 99 | 70/168 (42.0) | |
| posterior circulation | ||||
| clip | 38 | 40 | 3/38 (8.0) | |
| coil | 31 | 29 | 5/31 (16.0) |
As expected, Hunt and Hess grade was strongly associated with outcomes for the assigned clip and coil groups as a whole (OR 2.28 for patients entering with a Hunt and Hess score greater than II, 95% CI 1.42–3.68, p = 0.0007). Overall, patients with a posterior location aneurysm had worse outcomes at 3 years, although this relationship was not significant in adjusted analysis at alpha = 0.05 level (OR 1.72, 95% CI 0.96–3.09, p = 0.07). Part of the explanation may be the Hunt and Hess score greater than II, which in the posterior circulation was 71.2% and 58% for the clip and coil cohorts, respectively, compared with 46.2% and 49.5% for the clip and coil cohorts, respectively, in the anterior circulation. Location, however, appeared to confound the relationship between treatment and outcome; an interaction term included in the model was highly significant (p = 0.009). Patients with a posterior circulation aneurysm who were assigned to clipping were 5 times as likely to experience a poor outcome as patients with a posterior circulation aneurysm who were assigned to coil embolization (OR 5.09, 95% CI 1.63–15.9, p = 0.005, Table 9). Unfortunately, there was a lack of anatomical parity between the 2 assigned cohorts for posterior circulation aneurysms (Fig. 2B).
Multivariable analysis of assigned patients with poor outcome (mRS score > 2) at 3 years
| Characteristic | OR | 95% CI | p Value |
|---|---|---|---|
| clipping* | 1.31 | 0.82–2.10 | 0.25 |
| age >50 yrs | 1.76 | 1.07–2.94 | 0.03 |
| posterior location | 1.72 | 0.96–3.09 | 0.07 |
| Hunt & Hess grade >II | 2.28 | 1.42–3.68 | 0.0007 |
Patients assigned to surgical clipping (intent to treat).
Rebleeding After Year 1
No occurrences of SAH were documented in the 2nd or 3rd year of BRAT in either treatment group. At the time of this writing, 1 patient had experienced a hemorrhage from an incidentally coiled basilar artery aneurysm 8 years after the initial treatment. A coiled anterior communicating artery aneurysm was believed to have been the source of the original hemorrhage. This patient had not returned for requested follow-up evaluations. Imaging studies at the time of presentation with new SAH showed recanalization of the basilar artery aneurysm. The patient died of the hemorrhage.
Retreatment of Aneurysms
By the end of the 1st year, 10.6% of patients treated with coil embolization required retreatment compared with 4.5% of those treated with clipping (p = 0.03).15 During the 2nd and 3rd years of follow-up, 2 more patients in the coil group required retreatment: one underwent recoiling and the other underwent clipping. In contrast, none of the patients in the clip group have required retreatment.
Of 110 coil-treated patients with a 3-year follow-up, 14 (13%) required retreatment compared with 11 (5%) of the 226 clip-treated patients (p = 0.01).
Aneurysm Obliteration
After initial treatment, complete aneurysm obliteration was achieved in 58% of the coil-treated group. At 3 years, this percentage had decreased to 52%. After initial treatment, complete obliteration was achieved in 85% of the clip-treated group. At 3 years, this percentage was 87% (p < 0.0001, Table 10).
Obliteration of aneurysms based on diagnostic studies after actual treatment
| Time Point | n (%) | p Value* | |
|---|---|---|---|
| Coil | Clip | ||
| postop | 73/126 (57.9) | 229/269 (85.1) | <0.0001 |
| 3 yrs | 36/69 (52.2) | 122/140 (87.1) | <0.0001 |
Values calculated using χ2 tests.
Discussion
The primary goal of BRAT was to determine whether clipping or coiling was the superior treatment for patients with SAH based on the clinical and angiographic outcomes of all patients with a ruptured aneurysm. Including all eligible patients in a trial allows the potential role of each treatment modality to be clarified and selection bias to be minimized. Randomizing all SAH patients prior to initiating treatment has the disadvantage of randomizing nonaneurysmal SAH. However, the all-inclusiveness of this process eliminates any possibility of filtering patients that may not represent the whole cohort of aneurysmal SAH patients. The frequency of crossing over patients from coil embolization to clip occlusion, 38% for the whole group and 42% for patients with anterior circulation aneurysms, indicates a clear-cut need for clipping expertise to be available when treating aneurysms for SAH.
In contrast, designing a study that includes only patients for whom either technique is considered appropriate allows only a small proportion of eligible patients to be randomized, as was the case with ISAT, where only 22% of eligible patients were enrolled in the study.17 Despite the limitations of this study design, the findings of ISAT have been generalized to all patients with aneurysms, resulting in a marked change in management. At a major center in the United Kingdom, the rate of clipping declined from 51% to 31% and the rate of coiling increased from 35 to 68%.10 In the United States, the trend was equally dramatic. The rate of coiling for ruptured intracranial aneurysms increased from 17% in 2002 to 58% in 2008, and for unruptured intracranial aneurysms, it increased from 30% to 63%.27
To Clip or To Coil: Is That the Question?
Three contemporary prospective studies have compared coiling to clipping: the Finnish study,13 ISAT,17 and BRAT.15 Based on their most recent follow-up, none of the 3 studies found a significant difference in rate of independent survival between the 2 treatments.
In the small Finnish study, the 10.2% absolute difference favoring coiling was not statistically significant at 1 year.13 Their enrollment continued, and altogether 138 patients were randomly assigned to microsurgical clipping (n = 71) or endovascular coiling (n = 67) by the end of 1999. No further follow-up of this trial has been published. However, in the Kuopio Intracranial Aneurysm Database (www.uef.fi/ns) as of the end of 2008 (Timo Koivisto and Juha Hernesniemi, personal communication, 2011), there were 4 rehemorrhages in the endovascular group—2 from the originally treated aneurysms and 2 from untreated aneurysms. In the microsurgical group, there were 2 hemorrhages from de novo aneurysms. The mean time to the rehemorrhage was 8 years from the initial bleed.
In contrast, both ISAT17 and BRAT15 demonstrated a benefit associated with coiling 1 year after treatment. Neither study, however, found a significant difference in clinical outcomes between the 2 treatments—at 5 years of follow-up for ISAT18 and at 3 years for BRAT. Furthermore, when anterior circulation aneurysms alone were analyzed in BRAT, there was no significant difference between the 2 treatments at any time point (Table 6). The significant benefit in clinical outcomes noted for coil-treated patients with posterior fossa aneurysms (Table 7) is confounded by an imbalance in the location of aneurysms within the posterior fossa between the 2 treatment groups (Fig. 2B).
Although the ISAT found a significantly lower mortality rate in patients treated via coiling compared with those treated by clipping, the difference may well have been confounded by pretreatment deaths related to the delay of treatment, which was more than 14 hours longer in the clip group compared with the group undergoing endovascular coil placement. In fact, when pretreatment deaths were excluded from the analysis, there was no difference in the mortality or morbidity rates of the 2 treatment arms.1 Furthermore, follow-up data from the ISAT study suggest that younger patients may have better long-term outcomes when treated with clipping compared with coiling.16
The entry criteria for ISAT and BRAT differed markedly. Only 22% of eligible patients were entered into ISAT. Therefore, most patients were excluded from the study.17 Of those entered, 97% had aneurysms located in the anterior circulation. In contrast, BRAT randomized all eligible patients with SAH, irrespective of their presentation: 17% had aneurysms located in the posterior circulation.15
The intent-to-treat design of BRAT led to a 38% rate of crossover from the coil group to the clip group.15 The marked difference in crossover rates between the clipping and coiling arms of this study was not unexpected. The low crossover rate for the patients assigned to microsurgery reflects the fact that traditionally all aneurysms have been treated with clipping. In the hands of an experienced neurosurgeon, the majority of aneurysms can be successfully clipped while avoiding compromise of the parent vessels and branches. Microsurgical exposure also permits simultaneous management of associated hematomas. The large crossover rate for the coil-assigned cohort was predominantly due to the presence of associated hematomas (19%) and/or to an unfavorable aneurysm configuration that the endovascular surgeon felt would require complex stent-coil constructs and would therefore be more effectively treated with clipping.
Because in intent-to-treat analyses the outcomes remain in the assigned cohort independent of the actual treatment, crossovers represent clinical decisions that the treating physicians believe provide the best outcomes for their patients. Although a large crossover cohort makes it difficult to compare the efficacy of clipping and coiling directly, it does provide a solid comparison of the decision-making process for the 2 treatment modalities.
The ISAT excluded 78% of patients who would have been considered eligible in BRAT; 88% of the included patients had a good clinical grade, and more than 97% were located in the anterior circulation.15,17 Thus the outcomes of only anterior circulation aneurysms in BRAT provide a more relevant comparison with those in ISAT. Furthermore, in the coil-assigned cohort, the patients crossed over to clipping would never have been entered into ISAT in the first place, making that uncrossed cohort even more suitable for comparison (Table 11). Even when only patients who were both assigned to and treated with coiling are included, more than twice as many eligible patients were randomized in BRAT (48%) compared with ISAT (22%). More importantly, the clinical grade mix in ISAT was more favorable than that in BRAT. In ISAT, 88% of the patients were WFNS Grade I or II compared with only 60% of the patients with a Hunt and Hess grade of I or II in the coil-treated anterior circulation aneurysm cohort of BRAT.
Comparison of BRAT and ISAT at 1-year follow-up*
| Trial | mRS >2 at 1 yr | HH/WFNS Grade at Presentation | Aneurysm Location | Eligible Randomized (%) | |||
|---|---|---|---|---|---|---|---|
| Coil (%) | Clip (%) | 1–2 (%) | 4–5 (%) | Ant (%) | Post (%) | ||
| ISAT | 24 | 31 | 88 | <5 | 97 | 3 | 22 |
| BRAT | 21 | 31 | 49 | 21 | 83 | 17 | 100 |
| BRAT (ant circ) | 22 | 25 | 52 | 19 | 100 | 0 | 83† |
| BRAT (ant circ-coil) | 20 | 60 | 16 | 100 | 0 | 48‡ | |
BRAT and ISAT follow-up time points coincide only at 1 year. Abbreviations: ant = anterior; ant circ = anterior circulation, assigned cohorts; ant circ-coil = anterior circulation actual coil-treated cohort only; HH = Hunt and Hess; post = posterior.
All anterior circulation aneurysms, which represent 83% of the whole cohort (17% posterior circulation).
Includes all anterior circulation aneurysms assigned to the intent-to-coil cohort that were actually treated with coil embolization.
We compared the 1-year results of the ISAT and BRAT because only data from this time point are available from both studies (the most recent data for ISAT were from 5 years compared with 3 years for BRAT), including the mRS scores, Hunt and Hess/WFNS presentations, aneurysm location, and percentage randomized.15,17 The BRAT data are presented for the entire cohort, restricted to the anterior circulation, and restricted to the anterior circulation coil-only group (Table 11). Despite the significantly higher percentage of poor-grade patients in the BRAT, its outcomes still compare favorably to those of ISAT. Patients with anterior circulation aneurysms in BRAT demonstrated similar outcomes whether assigned to clipping or coiling. Therefore, one can assume that the theoretical outcome of the clip group restricted to the same selection process as the uncrossed coil group would be similar. This is a critical issue given that the degree of aneurysm obliteration is higher and the rates of rebleeding are lower in patients who are treated with clip occlusion, and the primary advantage of coil embolization over clipping is the perceived lower risk associated with initial treatment. Following the ISAT publication this point was debated at length.2–4,11 Based on the BRAT results, a strong case can be made that there is equipoise in treatment-related morbidity and mortality rates for anterior circulation aneurysms.
For the 69 posterior circulation aneurysms in BRAT, the results are less clear. Of these, 40 were actually clipped and 29 were treated with coils. Three patients (8%) in the clip group crossed over to the coil group while 5 patients (16%) in the coil group crossed over to the clip group (Tables 5 and 8). With the exception of the basilar tip aneurysms, the randomization of the posterior circulation aneurysms was unexpectedly skewed. Five of the 6 superior cerebellar artery aneurysms and 18 of the 21 posterior inferior cerebellar artery aneurysms were randomized to clipping (Fig. 2B). In contrast, the majority of posterior cerebral, vertebral, and basilar artery aneurysms were assigned to coiling (15 vs 5). Although there is a highly significant difference favoring coiling for posterior circulation aneurysms in BRAT, the lack of anatomical parity makes it difficult to draw strong conclusions.
A study that evaluated the actual results of change in practice patterns after the results of the ISAT trial was recently published. O'Kelly et al.21 retrospectively analyzed a cohort of adult patients with aneurysmal SAH who underwent aneurysm treatment in Ontario between 1995 and 2004. Their goal was to determine whether endovascular occlusion of the ruptured aneurysm reduced the likelihood of readmission for SAH and the rate of mortality compared with surgical occlusion. Of 3120 patients in their series, 778 underwent coiling and 2342 underwent clipping. In general, the rate of coiling tended to increase as the study years progressed. Rates of mortality (24.8% clipping vs 27.1% coiling, p < 0.0001) and readmission (1.8% vs 3.1%, p = 0.03) were significantly lower in patients who were treated with clips than in those who were treated with coils. Coil embolization was associated with a 25% increase in relative risk of death or recurrent SAH compared with clipping.21
In the ISAT trial,17–19 coil embolization was performed significantly earlier than clipping, creating a marked difference in the number of pretreatment deaths, favoring coiling. In contrast, in the study by O'Kelly et al.,21 morbidity and mortality were higher for coiling and the delay of treatment was reversed, with a time from admission to treatment of 2.68 days for coiling compared with 1.99 days for clipping.
Indeed, after the ISAT publication, a major institution reported a significant increase in the time between presentation and treatment (an increase of more than 1 day) from the time of the study to the last recorded period (1.1 vs 2.3 days, p < 0.05).10
The fact that pretreatment deaths can potentially account for the differences in outcome stresses how even in well-planned studies, factors other than the treatment itself can lead to unexpected effects on outcome. It is clear that patients with an SAH should have intervention as promptly as possible.
Obliteration and Rebleeding of Aneurysms
The goal of aneurysm treatment is to limit the risk of further hemorrhage. Residuals of both clipped and coiled aneurysms represent further risk of aneurysmal growth and rupture.5,6,8 The significant difference between clipping and coiling in terms of complete aneurysm obliteration creates additional concerns in coiled aneurysms: the need for frequent follow-up, the continued risk of SAH, the need for retreatment, and the associated cost. Although the risk of recurrent SAH associated with coiled aneurysms is small, it is significantly greater than in patients undergoing clipping. As the residual aneurysm remnant grows over time, the risk would be expected to increase.5,13 Indeed, in the Finnish study the average time for rebleeding was 8 years (Timo Koivisto and Juha Hernesniemi, personal communication, 2011).
The question of treatment efficacy remains a concern with respect to coil embolization. Torner et al.29 presented the long-term outcome in a series of 4060 patients with unruptured aneurysms. Over an average follow-up period of 8.5 years, 81 patients went on to die of SAH: 3.4% of the deaths occurred in the conservatively treated cohort, 2.8% occurred in the coil-treated group, and 0.5% occurred in the clip-treated group.
A recent study by Dorfer et al.7 reported the results of an analysis of 1597 patients with aneurysms treated at their institution over an 18-year period.7 Follow-up angiography at 6 and 18 months was available in 576 of the 739 patients treated by endovascular coiling. In the group with repeat imaging, 145 patients (24.3%) had minor postcoiling residuals and 127 (22.1%) had significant aneurysm residual or growth. In the latter group, initial treatment with coiling resulted in complete occlusion in 68.5% (87), subtotal occlusion in 22.0% (28), and incomplete occlusion in 9.4% (12). The fact that the 68.5% of the patients with significant aneurysm residual or growth on follow-up were documented to have had complete aneurysm obliteration after initial treatment is troublesome and emphasizes the risk of recurrence of coil-treated aneurysms and the need for continued follow-up. Of the 127 patients with significant aneurysm regrowth, 89 presented originally with an SAH, 13 (15%) of whom suffered a rehemorrhage after their initial coiling. Two of the 13 reruptures were treated endovascularly, 4 were treated with clipping, and 7 patients died of their hemorrhage without further treatment (personal communication, Engelbert Knosp, 2012). These results again emphasize the risk of delayed hemorrhage in those patients with significant aneurysm remnants or regrowth.
In the ISAT trial there was a small but significant difference of a greater risk of rehemorrhage in the coil group as compared with the clip group. During the first year of the ISAT, rehemorrhage occurred in 20 (4.2%) of the patients assigned to the coil group. The ISAT authors stated, “Although, on the basis of these data, the annual risk of the treated aneurysm rebleeding is higher in the patients treated by coiling than in the patients treated by clipping, the risk remains low and is at a similar level to the risk of further SAH from another source, either a pre-existing aneurysm or a newly formed aneurysm.”18
Evolution of Treatment
New innovations in endovascular technology are continually introduced and well reported, while the advances associated with microsurgical clipping are less well recognized. Nonetheless, the microsurgical treatment of aneurysms has changed significantly from the inception of the BRAT trial in 2003. Patients with an SAH now seldom undergo arterial angiography—whether preoperatively, intraoperatively, or postoperatively—unless endovascular treatment is also being considered. At our institution we rely primarily on CT angiography and, to a lesser extent, on MR angiography to determine whether a patient with an aneurysm is an appropriate candidate for clip occlusion and for long-term follow-up.
A major improvement in the surgical armamentarium has been the introduction of intraoperative indocyanine green angiography, which has almost entirely replaced the need for standard intraoperative angiography.24 This convenient, noninvasive method for visualizing the vasculature through the operating microscope allows intraoperative assessment of aneurysm obliteration and patency of the major vessel. Furthermore, its resolution far surpasses that of angiography for demonstrating the patency of small perforators that can be at significant risk during surgery. Finally, avoiding the use of rigid retraction even for the most complex aneurysms, as was done routinely during BRAT, minimizes surgical trauma to adjacent brain tissue.28
Although there is continued and dramatic progress in endovascular technology, there is also a price associated with the use of complex aneurysmal constructs. Early promising results with stents provide tantalizing promise of the ability to treat more complex aneurysms.9 However, the addition of stents, even in very experienced hands, increases the rate of morbidity and mortality. Over the 7 years from January 2002 to January 2009, Piotin et al.23 from Moret's group treated 1325 aneurysms in 1137 patients via endovascular techniques. None of the patients had been treated previously. Of the 1325 aneurysms, half of which presented with an SAH, 1109 (83.5%), were treated with coil embolization alone. The remaining 216 (16.5%), only 16% of which had presented with SAH, were treated with stent-assisted coiling. The recurrence rate in the non–stent-treated group was 33.5% compared with 15% in the stent-treated group. The recurrence rate for aneurysms larger than 10 mm in diameter was significantly higher for both the non–stent-treated and stent-treated groups (55% and 33%, respectively).
Piotin et al.23 concluded that stent-assisted coiling dramatically reduced the recurrence rate, but the price of that reduction was a high rate of morbidity and mortality. The overall rate of permanent, procedure-related morbidity (7.4%) and mortality (4.6%) was 12% for the stent-treated cohort compared with an overall 5% rate of permanent procedure-related morbidity (3.8%) and mortality (1.2%) in the non–stent-treated group. Because these percentages were procedure related and because there were 14% fewer patients than procedures, the morbidity and mortality rate per patient increases accordingly. Commenting on these results, van Rooij et al.30 noted that no cumulative risk of an unruptured aneurysm can outweigh such a high complication rate. Raymond and Darsaut26 concluded that no credible data support the use of stents.
In another recent trial, aneurysms with the same angioarchitecture were treated by coiling or stent-assisted coiling and all were followed up for 2 years. There was no difference in the rate of recanalization between the stent-assisted (17.5%) and non–stent-assisted coil groups (21%, p = 0.895).12 In a commentary on the report by Piotin et al.23 addressing general concerns about stenting aneurysms, Raymond and Darsaut26 concluded that no credible data support the use of stents and that randomized trials are sorely needed.
With the results of ISAT and the initial BRAT study demonstrating improved results at 1 year, generalization of these results to the entire aneurysm population has become widespread. The facts that ISAT was a study limited to just over 20% of eligible patients, leaving the majority of patients considered not to have reached equipoise, and that a large percentage of patients crossed over in BRAT tend to be ignored. Although the editorial that accompanied the ISAT study emphasized that “it is difficult to generalize the results of this study to the entire population of patients with aneurysmal SAH, and application to practice must be limited to those whose characteristics match those randomized in ISAT,”20 precisely the contrary occurred, and many publications state the advantages of endovascular treatment without acknowledging the above limitations. Unfortunately, such statements are common: for example, “After the international subarachnoid aneurysm trial (ISAT), which showed better outcomes with endovascular coiling than with neurosurgical clipping for the treatment of intracranial aneurysms, endovascular coiling is the preferred treatment for many patients”31 or “In the European practice, endovascular treatment with coils is now the first treatment technique of choice for both unruptured and ruptured aneurysms.”22
However, the 3-year follow-up data from BRAT and the 5-year follow-up data from ISAT now show that the benefit of coiling is less clear-cut. Even more critical is that when anterior circulation aneurysms are considered alone in BRAT, there was no benefit to coiling at any time point despite a crossover rate of 42%. It is equally clear that there have been significant advances in endovascular technology and improvements in surgical techniques. A prospective randomized trial is urgently needed to clarify the roles of each treatment.
Conclusions
Including all patients with SAH in a trial helps clarify the potential role of endovascular coiling. In BRAT, for instance, 38% of patients with aneurysms in the intent-to-coil cohort were considered inappropriate candidates for that treatment and crossed over to clipping. If only the anterior circulation aneurysms are considered, that percentage increased to 42% based on the judgment of experienced endovascular surgeons. Furthermore, the risks of incomplete aneurysm obliteration and a subsequent recurrence and rehemorrhage associated with coiling remain worrisome.
In the editorial accompanying the publication of the 1st-year results from BRAT, Dr. Lanzino14 noted that the results favored endovascular management as a reasonable strategy for right of first refusal.15 Using the same logic, the 3-year results of BRAT suggest that clipping can be considered the preferable management strategy for anterior circulation aneurysms, as there is no difference in treatment risk between coiling and clipping and because clipping can treat all aneurysms (crossover rate less than 1%), results in better aneurysm occlusion, provides superior protection from rebleeding, and is associated with less need for frequent follow-up and retreatment.
The devastating outcomes of patients with aneurysmal SAH, irrespective of treatment, leave considerable room to improve the management of this disease. Prevention remains a cornerstone of efforts to limit the ravages of SAH. Because specific benefits and risks are inherent to both coiling and clipping, patients with an aneurysm need access to expertise in both treatment modalities. With the results of both ISAT and BRAT having reached equipoise and with BRAT convincingly demonstrating that patients with an SAH can be randomized and treated promptly, a new trial for the entire aneurysm population that incorporates recent advances in both treatment modalities is justified.
Disclosure
Dr. McDougall reports being a consultant for Covidien and Gore. Funding for this manuscript came from the Barrow Neurological Foundation and the Hanley Aneurysm Fund.
Author contributions to the study and manuscript preparation include the following. Conception and design: Spetzler, McDougall. Acquisition of data: Partovi. Analysis and interpretation of data: Spetzler, Albuquerque. Drafting the article: all authors. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Spetzler. Statistical analysis: Hills.
Acknowledgments
The authors gratefully acknowledge the invaluable contributions of Donnie Straus, M.S., for data management; of Madelon Petersen, R.N., B.S.N., for clinical evaluations and data collection; and of the Neuroscience Publications Office of Barrow Neurological Institute, Phoenix, Arizona, for their assistance with editing and preparing the manuscript. The authors also acknowledge Dr. Richard Clatterbuck's contribution to the original design of the study.
References
- 1↑
Bakker NA, , Metzemaekers JDM, , Groen RJM, , Mooij JJA, & Van Dijk JMC: International subarachnoid aneurysm trial 2009: endovascular coiling of ruptured intracranial aneurysms has no significant advantage over neurosurgical clipping. Neurosurgery 66:961–962, 2010
- 2
Batjer HH: ISAT study: is coiling better than clipping?. Surg Neurol 59:162–165, 2003
- 3
Britz GW: ISAT trial: coiling or clipping for intracranial aneurysms?. Lancet 366:783–785, 2005
- 4
Britz GW, , Newell DW, , West GA, & Lam A: The ISAT trial. Lancet 361:431–432, 2003
- 5↑
Byrne JV, , Sohn MJ, , Molyneux AJ, & Chir B: Five-year experience in using coil embolization for ruptured intracranial aneurysms: outcomes and incidence of late rebleeding. J Neurosurg 90:656–663, 1999
- 6↑
Campi A, , Ramzi N, , Molyneux AJ, , Summers PE, , Kerr RS, & Sneade M, et al.: Retreatment of ruptured cerebral aneurysms in patients randomized by coiling or clipping in the International Subarachnoid Aneurysm Trial (ISAT). Stroke 38:1538–1544, 2007
- 7↑
Dorfer C, , Gruber A, , Standhardt H, , Bavinzski G, & Knosp E: Management of residual and recurrent aneurysms after initial endovascular treatment. Neurosurgery 70:537–554, 2012
- 8↑
Feuerberg I, , Lindquist C, , Lindqvist M, & Steiner L: Natural history of postoperative aneurysm rests. J Neurosurg 66:30–34, 1987
- 9↑
Fiorella D, , Albuquerque FC, , Deshmukh VR, & McDougall CG: Usefulness of the Neuroform stent for the treatment of cerebral aneurysms: results at initial (3-6-mo) follow-up. Neurosurgery 56:1191–1202, 2005
- 10↑
Gnanalingham KK, , Apostolopoulos V, , Barazi S, & O'Neill K: The impact of the international subarachnoid aneurysm trial (ISAT) on the management of aneurysmal subarachnoid haemorrhage in a neurosurgical unit in the UK. Clin Neurol Neurosurg 108:117–123, 2006
- 12↑
Hwang G, , Park H, , Bang JS, , Jin SC, , Kim BC, & Oh CW, et al.: Comparison of 2-year angiographic outcomes of stent- and nonstent-assisted coil embolization in unruptured aneurysms with an unfavorable configuration for coiling. AJNR Am J Neuroradiol 32:1707–1710, 2011
- 13↑
Koivisto T, , Vanninen R, , Hurskainen H, , Saari T, , Hernesniemi J, & Vapalahti M: Outcomes of early endovascular versus surgical treatment of ruptured cerebral aneurysms. A prospective randomized study. Stroke 31:2369–2377, 2000
- 15↑
McDougall CG, , Spetzler RF, , Zabramski JM, , Partovi S, , Hills NK, & Nakaji P, et al.: The Barrow Ruptured Aneurysm Trial. Clinical article. J Neurosurg 116:135–144, 2012
- 16↑
Mitchell P, , Kerr R, , Mendelow AD, & Molyneux AJ: Could late rebleeding overturn the superiority of cranial aneurysm coil embolization over clip ligation seen in the International Subarachnoid Aneurysm Trial?. J Neurosurg 108:437–442, 2008
- 17↑
Molyneux AJ, , Kerr R, , Stratton I, , Sandercock P, , Clarke M, & Shrimpton J, et al.: International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 360:1267–1274, 2002
- 18↑
Molyneux AJ, , Kerr RSC, , Birks J, , Ramzi N, , Yarnold J, & Sneade M, et al.: Risk of recurrent subarachnoid haemorrhage, death, or dependence and standardised mortality ratios after clipping or coiling of an intracranial aneurysm in the International Subarachnoid Aneurysm Trial (ISAT): long-term follow-up. Lancet Neurol 8:427–433, 2009
- 19
Molyneux AJ, , Kerr RSC, , Yu LM, , Clarke M, , Sneade M, & Yarnold JA, et al.: 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 366:809–817, 2005
- 20↑
Nichols DA, , Brown RD Jr, & Meyer FB: Coils or clips in subarachnoid haemorrhage?. Lancet 360:1262–1263, 2002
- 21↑
O'Kelly CJ, , Kulkarni AV, , Austin PC, , Wallace MC, & Urbach D: The impact of therapeutic modality on outcomes following repair of ruptured intracranial aneurysms: an administrative data analysis. Clinical article. J Neurosurg 113:795–801, 2010
- 22↑
Pierot L, , Cognard C, , Ricolfi F, & Anxionnat R: Mid-term anatomic results after endovascular treatment of ruptured intracranial aneurysms with Guglielmi detachable coils and Matrix coils: analysis of the CLARITY series. AJNR Am J Neuroradiol 33:469–473, 2012
- 23↑
Piotin M, , Blanc R, , Spelle L, , Mounayer C, , Piantino R, & Schmidt PJ, et al.: Stent-assisted coiling of intracranial aneurysms: clinical and angiographic results in 216 consecutive aneurysms. Stroke 41:110–115, 2010
- 24↑
Raabe A, , Nakaji P, , Beck J, , Kim LJ, , Hsu FP, & Kamerman JD, et al.: Prospective evaluation of surgical microscope-integrated intraoperative near-infrared indocyanine green videoangiography during aneurysm surgery. J Neurosurg 103:982–989, 2005
- 25↑
Raper DM, & Allan R: International subarachnoid trial in the long run: critical evaluation of the long-term follow-up data from the ISAT trial of clipping vs coiling for ruptured intracranial aneurysms. Neurosurgery 66:1166–1169, 2010
- 26↑
Raymond J, & Darsaut TE: Stenting for intracranial aneurysms: how to paint oneself into the proverbial corner. AJNR Am J Neuroradiol 32:1711–1713, 2011
- 27↑
Smith GA, , Dagostino P, , Maltenfort MG, , Dumont AS, & Ratliff JK: Geographic variation and regional trends in adoption of endovascular techniques for cerebral aneurysms. Clinical article. J Neurosurg 114:1768–1777, 2011
- 28↑
Spetzler RF, & Sanai N: The quiet revolution: retractorless surgery for complex vascular and skull base lesions. Clinical article. J Neurosurg 116:291–300, 2012
- 29↑
Torner J, , Piepgras D, , Meissner I, , Huston J, & Brown R: Long-term follow-up in patients surgically treated for an unruptured intracranial aneurysm. J Neurosurg 108:A866, 2008. (Abstract)
- 30↑
van Rooij WJ, , Sluzewski M, & Peluso JP: Alarmingly high serious complication rate of stent-assisted coiling in unruptured intracranial aneurysms: the need for reflection and reconsideration. Stroke 41:e191–e192, 2010
- 31↑
White PM, , Lewis SC, , Gholkar A, , Sellar RJ, , Nahser H, & Cognard C, et al.: Hydrogel-coated coils versus bare platinum coils for the endovascular treatment of intracranial aneurysms (HELPS): a randomised controlled trial. Lancet 377:1655–1662, 2011
