| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Neuroradiology |
1 From the Department of Neuroradiology and Therapeutic Angiography, Lariboisière Hospital, Paris, France (J.N.V., A.A., M.R., J.J.M.), and Department of Biophysics and Medical Informatics, Fernand Vidal Hospital (E.V.), Lariboisière Saint Louis Medical University of Paris 7, France. Received November 30, 2001; revision requested February 6, 2002; revision received May 6; accepted June 27. Address correspondence to J.N.V., Department of Diagnostic and Interventional Neuroradiology, Groupe Hospitalo-Universitaire Pitié-Salpétrière, 47-83 Boulevard de l’Hôpital, 75651 Paris Cedex 13, France (e-mail: jean-noel.vallee@psl.ap-hop-paris.fr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
MATERIALS AND METHODS: During 6 years, 55 unselected consecutive saccular aneurysms in 53 patients (mean age, 47 years) were treated with GDC occlusion. Forty-one (75%) aneurysms were ruptured (Hunt-Hess and Fisher grades were assigned in patients); 14 (25%), unruptured. Clinical and angiographic evaluations were performed 6 months after treatment and during follow-up (mean follow-up, 2 years). Multivariate analysis was used to determine factors predictive of outcome.
RESULTS: GDC occlusion was a success in 52 (95%) aneurysms, a failure in two (4%), and not attempted in one (2%). Occlusion at final follow-up, evaluated in 44 aneurysms, was complete in 34 (77%), near complete in four (9%), and incomplete in six (14%). At 6–12 months, mean aneurysmal occlusion rate significantly worsened because of revascularization (P < .001) but improved at final follow-up because of reembolization in 10 aneurysms (P = .009); it remained stable (P = .351) between initial and final follow-up. Multivariate binary logistic regression indicated that before treatment started, aneurysmal neck size was the only independent predictor of initial occlusion rate (P = .002) and revascularization (P = .004). After the initial procedure, sac size and initial occlusion rate were independent predictors of revascularization (P = .004 and .008, respectively), irrespective of neck size. Occlusion rate at 6–12-month follow-up was the only independent predictor of that at final follow-up (P = .021), regardless of shape of aneurysm. Overall morbidity was 2% (one of 51); mortality, 6% (three of 51). Mortality correlated significantly with Hunt-Hess grade 4 at admission (P = .003) and incidence of vasospasm (P = .058).
CONCLUSION: GDC occlusion proved to be a safe effective therapeutic alternative to surgery in patients with ruptured or unruptured basilar tip aneurysms. Morphologic and clinical factors were respectively identified as predictors of the optimal anatomic and clinical outcomes.
© RSNA, 2003
Index terms: Aneurysm, basilar, 1753.1264, 1753.73 • Aneurysm, therapy, 1753.1264 • Interventional procedures, utilization, 1753.1264
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
|
Aneurysmal Features
Location of the aneurysmal neck.—Of the 55 aneurysms, 22 (40%) were located at the tip of the basilar artery, 17 (31%) were located at the junction of the basilar artery and the posterior cerebral artery, eight (15%) were located at the junction of the basilar artery and the superior cerebellar artery, and eight (15%) were located at the junction of the basilar artery, the posterior cerebral artery, and the superior cerebellar artery.
Size.—Angiographic measurements were obtained in all 55 aneurysms. Aneurysmal sac size ranged from 2 to 30 mm. It was less than 5 mm in 12 (22%) aneurysms, 5–9 mm in 16 (29%) aneurysms, 10–14 mm in 14 (25%) aneurysms, 15–24 mm in six (11%) aneurysms, and 25 mm or larger in seven (13%) aneurysms. Neck size ranged from 2 to 10 mm. It was 4 mm or smaller in 29 (53%) aneurysms and larger than 4 mm in 26 (47%) aneurysms.
Ratio of sac to neck size.—The ratio of sac to neck size (RSN) was 3.0 or greater in 11 (20%) aneurysms, 3.0–1.5 in 29 (53%) aneurysms, and less than 1.5 in 15 (27%) aneurysms.
Wall margin.—The wall margin was regular in 39 (71%) aneurysms and irregular in three (5%), and a false sac was present in 13 (24%) aneurysms.
Procedures
All procedures were performed in the angiography room, and the patients received general anesthesia. Systemic heparin sodium (Heparin Choay; Sanofi-Synthelabo, Le Plessis-Robinson, France) was administered as an intravenous bolus of 50 IU per kilogram of body weight, and then it was infused continuously to maintain an activated partial thromboplastin time of at least three times the normal level throughout the procedure. The femoral approach was used, and dominant vertebral artery angiography with the nonionic contrast medium iohexol (Omnipaque 300; Nycomed, Oslo, Norway) was performed to define the optimal projection for the best definition of the aneurysmal neck and of its relationships with the parent vessel and adjacent arterial branches.
A microcatheter (Tracker 18; Boston Scientific/Target, Natick, Mass) was generally used; however, for aneurysms smaller than 5 mm in diameter, a microcatheter (Tracker 10; Boston Scientific/Target) was used. The microcatheter was placed coaxially through a 6-F guide catheter (Envoy, Cordis Neurovascular, Miami, Fla; or Platform, Boston Scientific/Target). By using magnified fluoroscopy and a digital road map, the microcatheter (with its tip appropriately shaped with steam to accommodate the vascular anatomic curves close to the aneurysm) was directed into the aneurysm with the aid of a 0.012- or 0.014-inch guide wire (Transend or Fas-Dasher; Boston Scientific/Target). The aneurysm was embolized with a GDC coil system with a coil diameter of 0.010 inch (GDC-10s; Target Therapeutics) or, if the aneurysm was a large or giant nonruptured type, with a GDC coil system with a coil diameter of 0.018 inch (GDC-18s; Target Therapeutics).
For wide-neck aneurysms and/or for those with an RSN less than 1.5, the remodeling technique (25) was used to prevent the coils from bulging into the parent vessel. The procedure was performed with the protection of a balloon occlusion system (Magic B1 occlusion balloon, Balt Extrusion, Montmorency, France; or Cirrus Balloon Occlusion System, Medtronic MIS, Sunnyvale, Calif). With the balloon occlusion system (Magic B1; Balt Extrusion), the balloon had an expansion size of 6 mm in diameter and 9 mm in length. The balloon was temporarily inflated in front of the aneurysmal neck during the delivery of each GDC. After embolization, the patient was transferred to the neurologic intensive care unit. Effective intravenous administration of systemic heparin was continued for at least 24 hours to prevent clotting, according to a protocol initiated in September 1993 for dealing with wide-neck aneurysms, slowing of flow, or bulging of the coil into the parent vessel.
Methods and Measurements
The measurements, estimation of occlusion rate, and timing of EVT determination were performed by three of us (J.N.V., A.A., and M.R.) in consensus. The follow-up angiograms were evaluated by the same three physicians in consensus. Clinical evaluation was performed by the clinicians of our institution and was reviewed by the same three physicians in consensus.
Angiographic aneurysmal sac and neck measurements were based on the largest sac and neck sizes that were rigorously compared with the mean 3-mm diameter of the contrast material–filled lumen of the basilar artery distal to the vertebrobasilar junction on the same angiographic image (26) and the largest diameter of the first coil introduced. The sac was classified as small, large, or giant when it was less than 15 mm, 15–24 mm, or 25 mm or larger, respectively. The neck was classified as narrow or wide when it was 4 mm or smaller or larger than 4 mm, respectively.
The aneurysmal occlusion rate was estimated angiographically from the ratio of the embolized aneurysmal area to the initial aneurysmal area. According to an occlusion rate of 100%, 90% or greater, and less than 90%, respectively, occlusion was complete when the sac and neck were densely packed with no contrast material visible, near complete when a tiny neck remnant was opacified, and incomplete when the sac and/or neck remnant was opacified. Revascularization was defined as reopacification of the initially embolized aneurysm at angiographic follow-up, which led to a reduction in the occlusion rate.
The timing of EVT was early when treatment started within 3 days after the onset of SAH, intermediate when it started between days 4 and 14 after the onset of SAH, and late when it started at 15 days or later after the onset of SAH.
Clinical and angiographic evaluations were performed at the end of the procedure and then during follow-up, at 6 months (for ruptured aneurysms only), 12 months, 3 years, and 5 years. Clinical evaluation was graded according to the Glasgow Outcome Scale Score (GOSS) (27).
Statistical Analysis
The qualitative and quantitative variables tested included baseline characteristics of the patients and therapeutic and posttherapeutic variables. Baseline characteristics of the patients were age, sex, clinical findings at presentation (ie, ruptured or unruptured aneurysm), aneurysmal features (eg, sac and neck sizes, RSN, wall margin characteristics, and number of aneurysms). Those for ruptured aneurysms were Hunt-Hess and Fisher grades on the day of EVT and timing of EVT. Therapeutic variables were treatment feasibility, technical complications, and initial aneurysmal occlusion. Posttherapeutic variables included presence of procedure-related neurologic deficits, occurrence of vasospasm, aneurysmal occlusion at 6–12-month and final follow-up, revascularization, reembolization, GOSS, and morbidity-mortality.
Independent predictors of treatment feasibility, technical complications, and immediate and long-term anatomic and clinical results were identified by means of a multivariate logistic model by using either the maximum likelihood estimate or the exact method (LOGXACT; Cytel Software, Cambridge, Mass), with a P value of .05 considered to indicate a statistically significant difference. These predictors were identified among the factors previously screened (P < .2) with univariate analysis by using the Fisher exact test for qualitative variables and the Student t test for quantitative variables.
During the periods of treatment and follow-up, the mean evolutionary trend of the aneurysmal occlusion rate (ie, continuous variables) was tested by using the Wilcoxon signed rank test for paired values, with Bonferroni correction and with differences with a P value of .05 considered statistically significant, to take account of the multiple comparisons.
The correlations between the changes in the occlusion rate due to revascularization at the 6–12-month follow-up and those due to reembolization at the final follow-up were evaluated with the Spearman rank correlation coefficient.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
Treatment Effectiveness
At the end of the initial procedure, the aneurysmal occlusion was assessed in 52 aneurysms and was classified as complete in 39 (75%) aneurysms (Fig 2), near complete in 12 (23%) aneurysms, and incomplete in one (2%) aneurysm. Tables 1–4 and Figures 3 and 4 show the effects of the aneurysmal angioarchitecture on the occlusion.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Four patients experienced thromboembolic events; these occurred in 6% (four) of 64 procedures and in 8% (four) of 52 treated aneurysms. They occurred in one patient during the procedure, due to displacement of the first coil while the second coil was introduced, and in three patients who had a neurologic deficit within 72 hours of the procedure. Clots were angiographically documented in two of these four patients and were recanalized after superselective infusion of urokinase (Actosolv; Hoechst Houdé, Paris, France). Systemic heparin was administered to the other two patients. One of the four had right cerebellar infarction sequelae but died a few days later of pulmonary complications. The other three recovered, although a residual left thalamic infarction was noted on the final CT scan in one patient.
One case of coil migration occurred; this manifested in 2% (one of 64) of the procedures and in 2% (one of 52) of the treated aneurysms that were outside a small aneurysm with a wide neck after detachment of the last coil. The migrating coil occluded the temporal branch of the right posterior cerebral artery; the coil could not be retrieved and caused lateral homonymous hemianopsia. However, the latter proved to be transient because of the expansion of leptomeningeal arteries in the involved territory. In this case, no impairment in the visual field, as determined by a full visual field evaluation, occurred. During reembolization of this aneurysm, unintentional occlusion of the origin of the P1 segment of the posterior cerebral artery occurred because of coil loop bulging into the parent vessel through the neck; coil loop bulging resulted from a marked reduction of blood flow at this level.
The posterior cerebral artery was revascularized by the posterior communicating artery. Administration of systemic heparin was temporarily continued, and the occlusion was well tolerated with no perforating artery infarction. During one procedure, performed at an early stage, a rupture occurred in the aneurysmal false sac during delivery of the third coil. The bleeding was controlled by means of the same coil that caused the fissure, within seconds of the event, and by means of reversal of anticoagulation with administration of protamine (Protamine Choay; Sanofi-Synthelabo, Le Plessis-Robinson, France). Despite this adverse event, GDC delivery was continued until the aneurysm was completely occluded. This event caused transient deterioration of consciousness, for which the patient needed maintenance of intubation and assisted ventilation. Six days later, vasospasm occurred, for which the patient required angioplasty and maintenance of assisted ventilation. The patient was extubated 11 days after coil delivery, without clinical sequelae.
Complications due to SAH
No hemorrhage recurred in any patient with a ruptured aneurysm between admission and EVT. In the 26 patients with a ruptured aneurysm embolized within 14 days of the onset of SAH, symptomatic vasospasm that was unresponsive to aggressive medical treatment occurred in eight (31%). This vasospasm required endovascular chemical angioplasty with intraarterial infusion of 2 mg of nimodipine in one patient and of 360 mg papaverine chlorhydrate (Papaverine Aguettant; Aguettant, Lyons, France) in five patients. Endovascular mechanical angioplasty by using a balloon occlusion system was performed in two patients. Six (75%) of these eight patients recovered without clinical sequelae, but the other two (25%) died of severe vasospasm. Acute hydrocephalus was noted in four (15%) of 26 patients with 26 aneurysms treated within 14 days of the onset of SAH, but the condition resolved after ventricular drainage.
Clinical Outcome
All 51 patients with 52 treated aneurysms were clinically evaluated at 6–12 months, and 44 (86%) of these patients were evaluated at the final mean follow-up of 2 years. At 6–12 months, 48 (94%) patients had recovered their previous neurologic status and/or remained asymptomatic. Three (6%) of 51 patients died, two were classified with Hunt-Hess grade 4 and died of complications of SAH, and one was classified with Hunt-Hess grade 0 and died of secondary pulmonary complications. At 6–12 months, according to the GOSS, 46 (90%) of 51 patients had a score of 1 (good recovery without neurologic deficits), and three (6%) had a score of 5 (death). Two (4%) of 51 patients who had SAH sequelae at admission remained neurologically unchanged at 6 months, with GOSS scores of 2 (moderate disability) and 3 (severe disability). Table 5 indicates clinical outcomes in relation to findings at the initial clinical presentation. No bleeding or rebleeding was observed in any patient throughout the entire follow-up period.
|
|
In 39 patients with ruptured aneurysms, transient morbidity was 10% (four patients); permanent morbidity, 3% (one patient with a residual left thalamic infarction); and mortality due to perioperative conditions, 3% (one patient with pulmonary complications). In this group, SAH was responsible for transient morbidity of 26% (10 patients) and mortality of 5% (two patients) (Table 6). In regard to the timing of EVT (Table 7), 11 (28%) patients were treated at the initial procedure within 3 days after the onset of SAH; 15 (38%) patients, between days 4 and 14; and 13 (33%) patients, on day 15 or later. Early, intermediate, and late treatments respectively resulted in initial complete occlusion in 73% (eight of 11), 73% (11 of 15), and 77% (10 of 13) of patients and in morbidity-mortality rates of 9% (one of 11 patients), 13% (two of 15 patients), and 8% (one of 13 patients).
|
|
|
2 test). At initial and 6–12-month follow-up, complete occlusion correlated significantly with occlusion rates during follow-up and inversely with revascularization (Table 10). After the initial procedure, by using maximum likelihood exact multivariate binary logistic regression, the initial occlusion rate was an independent predictor of the 6–12-month occlusion rate and of revascularization (P = .008 for both). At the 6–12-month follow-up, the occlusion rate was an independent predictor of the occlusion rate at final follow-up (P = .021).
|
For reembolized aneurysms, the occlusion rate improved significantly after the last procedure and at the final follow-up (P = .004 for both, Wilcoxon signed rank test). The change of score in the occlusion rate because of revascularization at 6–12 months correlated inversely with the change because of reembolization at final follow-up (P < .001, Spearman rank correlation coefficient).
At admission, Hunt-Hess grades correlated significantly with Fisher CT scan grades (P = .029, r = 0.720, Pearson 2 test), and Hunt-Hess grade 4 correlated significantly with mortality (P = .003, r = 0.807, Fisher exact test). There was a relationship of borderline significance between the incidence of vasospasm and mortality (P = .054, r = 0.380, Fisher exact test).
Age and sex, timing of the procedure, and incidence of technical complications did not correlate significantly with treatment feasibility or with immediate or long-term anatomic and clinical results.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
The proximity of posterior circulation aneurysms to the brainstem and the common occurrence of intraventricular hemorrhage, with expansion of the four ventricles, may explain the severity of the effect of blood on the brainstem and the devastating effects of the rupture of the initial aneurysm in the posterior fossa. In the current treatment of cerebral aneurysms reported by Le Roux and Winn (1), long-term follow-up study findings indicated that if patients are untreated, nearly all those harboring ruptured basilar bifurcation aneurysms die or are severely disabled within 3–6 years. Most deaths were caused by rebleeding, which was between two and three times more likely to occur in basilar bifurcation aneurysms than in anterior circulation aneurysms (1). Risk factors for rebleeding included a poor clinical grade, a large aneurysmal size, and a high aneurysm dome pressing against the third ventricle (1).
Since the late 1980s, endovascular techniques have provided an attractive minimally invasive approach for treating basilar bifurcation aneurysms and constitute a therapeutic alternative to surgical clipping (30–44).
Referred Patients
In our series, 53 patients with 55 unselected consecutive ruptured or unruptured saccular basilar tip aneurysms were referred to our department from neurosurgical centers, and in our department, EVT is the treatment of choice for basilar tip aneurysms. The possibilities for the comparison of findings in our series with those in other published series, given endovascular or surgical treatment, are limited, because both our series and most of the previously reported series were retrospective studies, with different referral patterns and with different aneurysmal morphologic characteristics.
Our policy was to treat ruptured aneurysms as soon as possible, generally while we performed diagnostic arteriography, to prevent aneurysmal rebleeding and to treat delayed cerebral ischemia caused by symptomatic arterial vasospasm aggressively with medication or endovascular means. For unruptured aneurysms, our aim was to prevent bleeding and to improve neurologic symptoms caused by a mass effect.
The goal of endovascular therapy with GDC coils was to exclude the aneurysmal sac and neck from the intracranial circulation by means of complete and lasting occlusion and at the same time to preserve the parent artery, without bulging of the coil or slowing of the flow. To prevent delayed compaction and subsequent neck recanalization, coils were packed as densely as possible up to the last one, which could not be introduced into the sac.
Treatment Feasibility
In our series, treatment of one (2%) of the 55 aneurysms was not attempted because it was small and unruptured (sac size, 2 mm) and had a lower expected risk of bleeding (45); furthermore, very small GDCs (2 mm) were not available at the time. Surgical clipping was not attempted with these conditions. Access to the aneurysms was possible in all instances. Occlusion with the GDC was achieved in 52 (95%) of 55 aneurysms but failed in two (4%). The risk of failure was weakly correlated with an RSN less than 1.5. The ability to catheterize and embolize the aneurysm was not affected by age, sex, findings at initial clinical presentation, or, for ruptured aneurysms, the timing of the procedure.
The arterial geometry of the posterior circulation compared with that of the anterior circulation accounted for the relatively easy endovascular access and GDC occlusion of the basilar tip aneurysms with no more difficulties after SAH. This was reported in other studies (30–36,38) of endovascular treatment with success rates of 94.3%–100%. In certain studies (3,6,17,18) of surgical treatment, clipping was not possible in 6.2%–34% of basilar bifurcation aneurysms because of size, a rigid atherosclerotic neck, or severe injury to major vessels caused by the clip.
Anatomic Results
Initial occlusion.—In our series, the initial near complete or incomplete occlusion rate correlated strongly with wide-neck aneurysms and weakly with giant-sac aneurysms. However, neck size was the only factor predictive of the initial occlusion rate, irrespective of sac size, which correlated strongly with neck size. In other studies (30–44) of endovascular treatment, complete angiographic occlusion of posterior circulation aneurysms ranged from 42% to 85% of cases after initial GDC embolization and depended less on their location and more on their anatomic configuration in regard to neck and sac size. Mainly because of a wide neck and the risk of occluding perforating arteries of the brainstem, postsurgical aneurysmal rest has been evaluated in only a few patients. Postsurgical aneurysmal rest was reported (11,18,22,46) in 1.5%–24.7% of basilar bifurcation aneurysms, with a frequency of regrowth of 12.7% (46,47), which probably was underestimated.
Aneurysmal occlusion and revascularization at 6–12 months.—Complete occlusion at 6–12 months correlated strongly with narrow-neck aneurysms and the initial complete occlusion rate and weakly with small- or large-sac aneurysms. Revascularization correlated strongly with wide-neck aneurysms and the initial near complete or incomplete occlusion rate and weakly with giant-sac aneurysms. However, before initial embolization, neck size was the only independent predictor of the occlusion rate at 6–12 months and of revascularization. Afterinitial embolization, sac size and the initial occlusion rate were independent predictors of the occlusion rate at 6–12 months and of revascularization, irrespective of neck size that correlated strongly with the initial occlusion rate.
Subsequently, regular angiographic follow-up is mandatory, particularly for giant or near completely or incompletely occluded aneurysms, owing to their tendency to recanalize. In other studies (30–44) of endovascular treatment, revascularization and aneurysm growth ranged from 22.4% to 52.6% and were noted in some wide-neck and giant aneurysms with poor initial occlusion. Revascularization may be caused by arterial flow that produces coil compaction or coil impaction in the mural thrombus, particularly in giant aneurysms. Revascularization also may be caused by loose packing or the absence of dense packing of the neck, with concomitant physiologic fibrinolysis, which results in recurrence or regrowth of the remnant without any change in the morphology of the original aneurysm. In rare cases, the sac may grow. Short-term recurrences are clearly linked to coil compaction rather than to aneurysmal growth (30–44).
Final occlusion rate.—In our series, complete occlusion at final follow-up correlated strongly with complete occlusion at the initial procedure and at the 6–12-month follow-up. The occlusion rate at final follow-up did not correlate with baseline characteristics of patients, and it especially did not correlate with aneurysmal features. After initial embolization, the initial occlusion rate was the only independent predictor of the occlusion rate at final follow-up. At 6–12 months after treatment, the occlusion rate was the only independent predictor of the occlusion rate at final follow-up.
Mean trend of the aneurysmal occlusion rate.—Overall, the mean aneurysmal occlusion rate worsened significantly at the 6–12-month follow-up because of revascularization but improved significantly at the final follow-up because of reembolization. Therefore, there was no significant change in the mean aneurysmal occlusion rate between the initial and final follow-up after a mean period of 2 years. This meant that reembolization offset revascularization. In nine of 10 cases, reembolization concerned aneurysms that were incompletely occluded at the 6–12-month follow-up, that had a wide neck or a sac size larger than 10 mm, and that had an improvement in the occlusion rate after reembolization at final follow-up. In most of the near completely occluded aneurysms, residual filling was too small to accommodate the smallest coil. Initial and long-term anatomic results were not related to age, sex, findings at initial clinical presentation, occurrence of technical complications, or, for ruptured aneurysms, timing of EVT.
Technical Complications
Perioperative permanent morbidity and mortality were both 2% in our series. Thromboembolic events, the most frequent technical complication, occurred in four (8%) of 51 patients and were weakly correlated with wide-neck aneurysms and giant-sac aneurysms. Clotting occurred early in our experience, was revealed a few days after the procedure in three (6%) patients, and resulted in a transient neurologic deficit due to probable thrombus extension or clot migration from the neck. Therefore, from September 1993 onward, we attempted to reduce clotting by the temporary continuation of administration of systemic heparin in cases of a wide neck, sluggish flow, or coil bulging into the parent vessel. From that time, no clotting was observed, but this did not enable us to determine any conclusions in this respect.
In other studies (30–44) of endovascular treatment, the permanent morbidity and mortality rates due to the procedure ranged from 0% to 5% and from 0% to 6.4%, respectively. Surgery-related permanent morbidity-mortality was reported (3,11,18,20,22) in 4.7%–50% of cases, mainly because of perforating artery injury, and mostly concerned large, complex, or posteriorly oriented aneurysms (1,5,9,10,21,22,46).
Rebleeding
In our series, no bleeding or rebleeding occurred during the mean follow-up of 2 years, despite near complete and incomplete aneurysmal occlusion, which was respectively observed in four (9%) and six (14%) of the 44 aneurysms treated and followed up. In other studies of endovascular treatment (30–44), recurrent SAH rates ranged from 0% to 3.3%, which led to an annual rebleeding rate of 0%–3%. Aneurysms that rebled were known to have been incompletely treated from the outset (38). Partial coiling seems to prevent temporary subsequent bleeding, probably in part because the change in the intraaneurysmal flow induces subsequent thrombosis, which may strengthen the wall (36,48,49).
However, incompletely occluded aneurysms are at risk of rebleeding and must be considered as cases of treatment failure for as long as they are not retreated. The risk of rupture, according to the Laplace formula, increases with sac size and thinning of the wall. Rebleeding after clipping has been reported in 0.5%, 2.2%, and 5.5% of aneurysms at 5, 10, and 15 years, respectively (50), and in 0.79%–1.5% of aneurysmal rests per year (47,51) because of enlargement of the rests that concerned mostly posterior circulation aneurysms. However, there is no consensus regarding surgery for aneurysmal rests (46,47).
Clinical Outcome
In our series, 48 (94%) of 51 patients who were treated recovered their previous neurologic status and/or remained asymptomatic. Overall morbidity was 2% (one of 51 patients), and mortality was 6% (three of 51 patients). Mortality correlated strongly with Hunt-Hess grade 4 and weakly with the incidence of vasospasm. Clinical outcome was not related to age, sex, Fisher grade, aneurysmal features, technical complications, aneurysmal occlusion, or, for ruptured aneurysms, timing of EVT.
In the patients with ruptured basilar bifurcation aneurysms, overall morbidity was 3% (one of 39 patients), and mortality was 8% (three of 39 patients). At the 6-month follow-up, 92% (36 of 39) of patients recovered their premorbidity status, and 90% (35 of 39) of patients had a GOSS of 1. These results were partly caused by the good neurologic status of 77% (30 of 39) of the patients at admission (Hunt-Hess grades 0–2). These findings were consistent with those in other studies (30–44) of endovascular treatment, for which the final clinical outcome was good (GOSS of 1) in 79.5% of patients on average and ranged from 62.5% to 91.2% concerning 50% and 90.6% of the patients with Hunt-Hess grades 1 or 2, respectively, at admission (31,36).
Investigators of the International Cooperative Study on the Timing of Aneurysm Surgery (12,13) reported a morbidity-mortality rate of 47.4% for the treatment of 266 vertebrobasilar aneurysms; in this study, basilar aneurysms were predictors of mortality. The postoperative risk following early surgery was equivalent to the risk of rebleeding and vasospasm in patients waiting for delayed surgery (12,13). In recent studies (14,18–20) about patients undergoing early surgery for basilar aneurysms, the reported morbidity-mortality rates of 28.0%– 32.7% correlated with age, clinical grade, and aneurysm size.
Treatment of unruptured intracranial aneurysms depends on the natural history of these lesions and on the morbidity-mortality rate associated with repair. In an international study (45), cumulative annual rupture rates of about 0.05% and 0.5% were reported for aneurysms smaller than 10 mm in diameter in patients without and with a history of SAH, respectively. For aneurysms 10 mm or larger in diameter, annual rupture rates were less than 1% in both groups; however, in patients without SAH, the rate for the 1st year was 6% for giant aneurysms (sac diameter, 
25 mm). The rupture rate reported in this study (45) was lower than the rates reported in previous studies (52,53).
Although basilar tip aneurysms were predictors of rupture in both groups, the surgical treatment of the aneurysms was controversial (45). Khanna et al (54) reported that the surgical risk of unruptured aneurysms, based on a grading system from 0 to 6, ranged from 0% in grade 0, which included young patients with small anterior circulation aneurysms, to 66.6% in grade 6, which included elderly patients with complex giant posterior circulation aneurysms. In our series, treatment with the GDC proved to be safe in 13 unruptured basilar tip aneurysms, with no morbidity-mortality and a final complete occlusion in 11 (84.6%). However, for wide-neck or giant unruptured aneurysms, the indications for treatment with GDC should be carefully assessed. In reports (31,34–36) about EVT for unruptured basilar aneurysms, 71.4%–100% of patients remained clinically unchanged. The evolution of GCD technology has proved to provide safe treatment of incidental aneurysms (55).
In conclusion, GDC embolization resulted in good anatomic and clinical outcomes in the present series and proved to be a safe effective therapeutic alternative to surgical treatment of patients with ruptured or unruptured basilar tip aneurysms. The findings in this study resulted from the relatively easy endovascular access to the aneurysms, the success of the initial treatment predicted by neck size, and the reembolization of recanalized aneurysms facilitated by particularly careful regular follow-up for giant or near completely or incompletely occluded aneurysms, factors that were predictors of recanalization. Nevertheless, a longer follow-up is required for further evaluation of these results.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, J.N.V., A.A., J.J.M.; study concepts and design, J.N.V., J.J.M.; literature research, J.N.V., M.R., A.A.; clinical studies, J.N.V., A.A., M.R.; data acquisition, J.N.V., M.R., A.A.; data analysis/interpretation, J.N.V., E.V.; statistical analysis, J.N.V., E.V.; manuscript preparation, definition of intellectual content, and editing, J.N.V., A.A., M.R.; manuscript revision/review, J.N.V., A.A., E.V., J.J.M.; manuscript final version approval, J.N.V., J.J.M.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
