BRAT
The Barrow Ruptured Aneurysm Trial
Year of Publication: 2011
Authors: Cameron G. McDougall, Robert F. Spetzler, Joseph M. Zabramski, ..., Felipe C. Albuquerque
Journal: Journal of Neurosurgery
Citation: J Neurosurg. November 4, 2011
PDF: https://www.researchgate.net/profile/Rob...saWNhdGlvbiJ9fQ
PDF: Clinical Question
Is one treatment modality (surgical clipping or endovascular coil embolization) superior to the other for treatment of acutely ruptured cerebral aneurysms, as measured by clinical and angiographic outcomes?
Bottom Line
At 1 year after treatment, a policy of intent to treat favoring coil embolization resulted in significantly fewer poor outcomes (10.5% absolute difference) compared to surgical clipping (23.2% vs 33.7%, OR 1.68, p = 0.02). Although most aneurysms assigned to the coil group were treated by coil embolization (62.3%), a substantial number crossed over to surgical clipping. No patient treated by coil embolization suffered recurrent hemorrhage during the first year. High-quality surgical clipping remains important as an alternative treatment modality.
Major Points
- Single-center prospective randomized controlled trial comparing surgical clipping vs endovascular coil embolization for ruptured intracranial aneurysms
- 725 patients screened, 500 enrolled with informed consent, 471 eligible for analysis (238 assigned to clip, 233 assigned to coil)
- No anatomical exclusions - all patients with aneurysmal SAH were eligible regardless of aneurysm location or characteristics
- Intent-to-treat design with crossover allowed based on treating surgeon's clinical judgment
- Primary outcome: poor outcome defined as mRS score > 2 (death or dependency) at 1 year
- At 1 year, 33.7% of patients assigned to clipping had poor outcomes vs 23.2% assigned to coiling (OR 1.68, 95% CI 1.08-2.61, p = 0.02)
- When only patients receiving assigned treatment analyzed: 33.9% clip vs 18.4% coil had poor outcomes (OR 2.28, p = 0.005)
- 75 patients crossed over from coil to clip (most common reasons: anatomical features unfavorable for coiling, hematoma requiring evacuation, or multiple aneurysms with uncertain rupture source)
- Patients who crossed over from coil to clip had same outcomes as those assigned to and treated with clipping
- No rebleeding occurred in any patient treated by coil embolization during first year
- Retreatment more common in coil group: 10.62% vs 4.49% in clip group at 1 year (OR 2.57, p = 0.03)
- Age > 50 years and Hunt & Hess grade > II were independent predictors of poor outcome
- Results consistent with ISAT findings despite different trial design and broader inclusion criteria
Design
Study Type: Prospective randomized controlled trial
Randomization: 1
Blinding: Outcome assessments performed by independent nurse practitioners not involved in treatment, but assessors were not blinded to treatment modality
Enrollment Period: March 2003 to January 2007
Follow-up Duration: 1 year (with planned follow-up to 6 years)
Centers: 1
Countries: United States
Sample Size: 471
Analysis: Intent-to-treat analysis using logistic regression methods. Multivariable models adjusted for age > 50 years and Hunt & Hess score > II. Secondary analyses examined outcomes based on actual treatment received. Wilcoxon rank-sum test for continuous variables and Mantel-Haenszel chi-square test for categorical variables.
Inclusion Criteria
- Age 18-80 years
- Admitted to ICU with acute nontraumatic subarachnoid hemorrhage
- SAH confirmed by CT scan or lumbar puncture
- Presentation within 14 days of hemorrhage
- Patient or health care decision surrogate provided informed consent
- No anatomical exclusions - all aneurysm locations and sizes included
Exclusion Criteria
- Traumatic subarachnoid hemorrhage
- Presentation more than 14 days after hemorrhage
- Age < 18 years or > 80 years
Baseline Characteristics
| Characteristic | Control | Active |
|---|---|---|
| Total patients | 238 | 233 |
| Mean age (years) | 53.1 ± 12.8 | 54.3 ± 12.0 |
| Female | 69.7% | 71.2% |
| Race - Caucasian | 63.9% | 67.8% |
| Race - Hispanic | 26.5% | 21.5% |
| Race - African American | 5.5% | 3.9% |
| Race - Asian | 1.3% | 3.4% |
| Race - Other | 2.9% | 3.4% |
| Diabetes | 8.4% | 7.3% |
| Hypertension | 43.3% | 44.6% |
| Smoking history | 61.8% | 62.2% |
| Cocaine use | 8.8% | 9.0% |
| Methamphetamine use | 7.1% | 8.6% |
| Mean GCS score | 12.3 ± 3.6 | 12.5 ± 3.6 |
| Mean Hunt & Hess grade | 2.6 ± 1.1 | 2.6 ± 1.1 |
| Mean Fisher grade | 2.7 ± 0.7 | 2.7 ± 0.6 |
| Aneurysm mean size (mm) | 6.8 ± 4.1 | 6.6 ± 4.0 |
| Aneurysm median size (mm) | 6.0 (IQR 4-8) | 6.0 (IQR 4-8) |
| Posterior circulation aneurysm | 16.0% | 13.7% |
| Anterior circulation aneurysm | 73.1% | 72.5% |
| Angiography negative | 10.9% | 13.3% |
| Hunt & Hess Grade I | 13.4% | 13.3% |
| Hunt & Hess Grade II | 38.7% | 39.9% |
| Hunt & Hess Grade III | 29.8% | 26.2% |
| Hunt & Hess Grade IV | 12.2% | 14.6% |
| Hunt & Hess Grade V | 5.9% | 6.0% |
Arms
| Field | Control | Endovascular Coil Embolization |
|---|---|---|
| Intervention | Microsurgical clipping of ruptured intracranial aneurysm. Patients assigned to surgical clipping service in alternating fashion. Crossover to endovascular coiling allowed if treating surgeon determined it would provide better outcome. When possible, aneurysm treated within 24 hours of admission. All patients received standardized pre- and postoperative care including external ventricular drainage as needed and vasospasm prophylaxis and treatment. | Endovascular coil embolization of ruptured intracranial aneurysm. Patients assigned to endovascular service in alternating fashion. Crossover to surgical clipping allowed if treating surgeon determined aneurysm more appropriate for surgical treatment (e.g., anatomical features unfavorable for coiling, need for hematoma evacuation, multiple aneurysms with uncertain rupture source). When possible, aneurysm treated within 24 hours of admission. First-generation Matrix coils used in majority of procedures. All patients received standardized pre- and postoperative care including external ventricular drainage as needed and vasospasm prophylaxis and treatment. |
| Duration | Treatment during acute hospitalization, with 1-year follow-up | Treatment during acute hospitalization, with 1-year follow-up |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| Poor outcome defined as modified Rankin Scale (mRS) score > 2 at 1 year, indicating death or dependency | Primary | 69 of 205 patients (33.7%) | 46 of 198 patients (23.2%) | 10.50% | 0.02 |
| Poor outcome (mRS > 2) at 1 year - patients who received assigned treatment only | Secondary | 61 of 180 patients (33.9%) | 20 of 109 patients (18.4%) | OR 2.28 | 0.005 |
| Poor outcome (mRS > 2) at 1 year - actual treatment received (all treated patients) | Secondary | 83 of 245 patients treated with clip (33.9%) | 23 of 113 patients treated with coil (20.4%) | OR 2.01 | 0.01 |
| Rebleeding before treatment | Secondary | 2 patients | 0 patients | ||
| Rebleeding after treatment during initial hospitalization | Secondary | 1 patient assigned to clip | 1 patient assigned to coil but crossed over to clip | ||
| Rebleeding during first year after treatment | Secondary | No rebleeding in clip-treated patients after initial hospitalization | 0 patients treated with coil | ||
| Retreatment by 1 year - intent-to-treat | Secondary | 7 of 238 (2.94%) | 16 of 232 (6.9%) | OR 2.44 | 0.05 |
| Retreatment by 1 year - actual treatment | Secondary | 11 of 245 (4.49%) | 12 of 113 (10.62%) | OR 2.57 | 0.03 |
| Crossover from assigned coil to clip treatment | Secondary | 75 of 199 treated patients (37.7%) | |||
| Crossover from assigned clip to coil treatment | Secondary | 4 patients | |||
| Poor outcome in patients who crossed from coil to clip | Secondary | 22 of 65 (33.9%) - same as clip/clip group | |||
| Death before treatment | Adverse | 3 patients | 3 patients | ||
| Retreatment during initial hospitalization | Adverse | 5 patients assigned to clip (2.10%) | 7 patients assigned to coil (3.02%) | OR 1.45 | 0.53 |
Subgroup Analysis
Multivariable analysis showed that age > 50 years (OR 2.03, 95% CI 1.23-3.42, p = 0.007) and Hunt & Hess grade > II (OR 3.51, 95% CI 2.21-5.68, p < 0.0001) were each independently associated with poor outcome at 1 year, regardless of treatment modality. No significant interactions between treatment modality and either age or baseline Hunt & Hess score were found. For good grade patients (Hunt & Hess I-II), poor outcomes were 9.4% for coil and 19.8% for clip groups.
Criticisms
- Single-center study limits generalizability
- Not powered sufficiently to demonstrate definitive differences - designed as pilot study
- Outcome assessors not blinded to treatment modality
- Substantial crossover (37.7%) from coil to clip group, though this was by design
- Decisions about crossing over somewhat subjective and may vary across institutions
- First-generation Matrix coils used in majority of coil procedures have since been withdrawn from market due to high recurrence rates
- No patient lost to 1-year follow-up data collection for 68 patients (14.4%)
- Relatively small sample size compared to ISAT
- Higher proportion of poor grade patients (19.3% Hunt & Hess IV-V) compared to ISAT (12% WFNS IV-V)
- Cannot determine if incomplete coil occlusion of complex aneurysms would increase complication or recurrence rates
- Angiographic outcomes not yet reported at 1 year
Funding
Not specified in manuscript
Based on: BRAT (Journal of Neurosurgery, 2011)
Authors: Cameron G. McDougall, Robert F. Spetzler, Joseph M. Zabramski, ..., Felipe C. Albuquerque
Citation: J Neurosurg. November 4, 2011
Content summarized and formatted by NeuroTrials.ai.