Ruptured Intracranial Aneurysms: Acute Endovascular Treatment with ...

Ritva Vanninen, MD Timo Koivisto, MD Tapani Saari, MD Juha Hernesniemi, MD2 Matti Vapalahti, MD

Index terms: Aneurysm, intracranial, 17.73 Aneurysm, rupture, 17.73 Aneurysm, surgery, 17.1267 Aneurysm, therapy, 17.1264, 17.1267 Brain, hemorrhage, 17.434, 17.73 Interventional procedures, utilization, 17.1264 Radiology 1999; 211:325–336 Abbreviations: GDC ⫽ Guglielmi detachable coil MCA ⫽ middle cerebral artery SAH ⫽ subarachnoid hemorrhage 1

From the Departments of Clinical Radiology (R.V., T.S.) and Neurosurgery (T.K., J.H., M.V.), Kuopio University Hospital, Puijonlaaksontie 2, FIN70211 Kuopio, Finland. Received April 28, 1998; revision requested July 6; revision received August 12; accepted October 26. Supported by EVOFunding from Kuopio University Hospital. Address reprint requests to R.V.

2

Current address: Department of Neurosurgery, Helsinki University Hospital, Finland r RSNA, 1999 See also the editorial by Latchaw (pp 306–308) in this issue.

Ruptured Intracranial Aneurysms: Acute Endovascular Treatment with Electrolytically Detachable Coils—A Prospective Randomized Study1 PURPOSE: To compare the use of electrolytically detachable coils versus surgical ligation for the management of acutely ruptured intracranial aneurysm. MATERIALS AND METHODS: A prospective randomized study included 109 patients with acute (⬍72 hours) subarachnoid hemorrhage caused by a ruptured aneurysm (Hunt and Hess grade I–II [n ⫽ 67], grade III [n ⫽ 26], or grade IV–V [n ⫽ 16]). All patients were suitable candidates for both endovascular and surgical treatment and were randomly assigned to undergo coil embolization (n ⫽ 52) or surgical ligation (n ⫽ 57). RESULTS: Significantly better primary angiographic results were achieved after surgery in patients with anterior cerebral artery aneurysm (n ⫽ 55, P ⫽ .005) and after endovascular treatment in those with posterior circulation aneurysm (n ⫽ 11, P ⫽ .045). No significant differences were seen in middle cerebral artery (n ⫽ 19) or internal carotid artery (n ⫽ 24) aneurysms. Early rebleeding occurred in one patient after incomplete coil embolization. The technique-related mortality rate was 4% in the surgical group and 2% in the endovascular group. Clinical outcome (Glasgow Outcome Scale score) at 3 months was not significantly different between treatment groups in terms of intended treatment modality. No late rebleedings had occurred at the time of this writing. CONCLUSION: In selected patients with a recently ruptured intracranial aneurysm, favorable results were achieved by using endovascular treatment. Subsequent acute or late open surgery was sometimes required. The clinical outcome at 3 months was comparable in the endovascular and surgical treatment groups.

Author contributions: Guarantors of integrity of entire study, R.V., T.K.; study design, M.V., J.H., T.S.; definition of intellectual content, R.V., T.K., J.H., T.S., M.V.; literature research, R.V., T.K.; clinical studies, T.S., R.V., J.H., M.V., T.K.; data analysis, T.K., R.V.; statistical analysis, R.V., T.K.; manuscript preparation, R.V., T.K.; manuscript editing, T.K., J.H., M.V., T.S.

Subarachnoid hemorrhage (SAH) is a devastating disease because about half of the patients die due to primary bleeding or to subsequent complications, and many of those who survive need extended rehabilitation (1,2). Early craniotomy and surgical clipping have been advocated to help isolate the ruptured aneurysm from the intracranial circulation, thus preventing rebleeding and enabling aggressive treatment of potential or symptomatic vasospasm (3–8). Endovascular coil embolization is a technique that offers an alternative treatment option in these patients. Several published studies (5,9–15) included only patients with anticipated surgical difficulties or patients in whom such treatment had failed. More recently, the indications for endovascular treatment have widened considerably and may now differ markedly from one institution to another. Embolization accomplished with mechanical or electrolytically detachable coils may be attempted as the primary treatment option in certain anatomic locations (16–18) or without specified anatomic criteria, which leaves surgery for cases where endovascular treatment has failed (19,20). In addition, many of the 325

published reports concerned patients with a wide range of clinical conditions, including acute to not recent SAH, headaches, mass effect, or no symptoms. On the basis of the literature, therefore, a direct comparison between endovascular and surgical treatment is difficult. To our knowledge, no randomized clinical studies have been published in which early endovascular and surgical treatment were compared. We report the results from a prospective randomized study of the treatment of recently ruptured (⬍72 hours) cerebral aneurysms and describe aspects of endovascular treatment in the acute phase. The specific goals of our study were to (a) analyze the technical difficulties and clinical complications related to the technique, (b) evaluate the primary angiographic occlusion rate, and (c) compare the early clinical outcome in endovascularly treated patients with that in surgically treated patients.

MATERIALS AND METHODS Study Design The study design was approved by the ethical committee at our hospital. During the study period (February 1, 1995, to August 31, 1997), all patients who were admitted to our university hospital because of primary SAH were evaluated as potential candidates for the present study. After informed consent was obtained from the patient or the patient’s closest relative, all patients with a ruptured aneurysm that was considered to be suitable for both surgical clipping and endovascular treatment were consecutively included, provided that the following exclusion criteria were not fulfilled: (a) age older than 75 years, (b) bleeding for more than 3 days before the procedure, (c) presence of a large hematoma necessitating surgery, (d) presence of mass effect causing a neurologic deficit, or (e) previous surgery for the ruptured aneurysm. The aneurysm was not considered to be suitable for endovascular treatment and the patient was not considered for random assignment to a treatment group if the following findings from diagnostic angiography were present: (a) neck of the aneurysm wider than the fundus, (b) fusiform aneurysm, (c) neck and its relationship to the parent vessel and adjacent branches not distinguishable, or (d) diameter of the aneurysm smaller than 2 mm (less than the smallest coil available). The patient’s suitability for random assignment and endovascular treatment was always considered according to the morphology of the aneurysm that had 326 • Radiology • May 1999

most probably ruptured (aneurysm irregularity, size, and findings seen at computed tomography [CT]). To avoid selection bias, random assignment was performed separately for patients with a Hunt and Hess (21) grade of I–II, for those with a grade of III, and for those with a grade of IV–V. If total occlusion was not achieved in the first session, a repeat embolization of the residual lumen was attempted in a second session or, in cases of a small neck remnant, at the time of the 3-month angiographic control procedure. Postembolization surgical clipping was considered in patients with unsuccessful or incomplete embolization, and repeat coil embolization was considered in patients with incomplete surgical clipping. After the procedure, both the patients who underwent surgery and those who underwent endovascular treatment received care in a similar manner in the intensive care unit. Follow-up angiography was scheduled after surgical clipping during the primary hospital stay and, in cases of a minor neck remnant, 12 months after clipping or 3, 12, and 36 months after endovascular treatment. The occlusion rate achieved after embolization or surgical ligation was evaluated by using the angiograms at a consensus reading with a neuroradiologist (R.V.) and a neurosurgeon (T.K.). Neuropsychologic examinations were scheduled 3 and 12 months after the occurrence of SAH. (The neuropsychologic examination results are not reported here; they will be reported after the 1-year examination, when the possible transient vs permanent deterioration in neuropsychologic test performance in the two treatment groups can be evaluated more reliably.)

Patients During the study period, 321 patients were admitted to the hospital because of primary SAH. Twenty-four patients did not undergo angiography and were treated conservatively because they were in a moribund state. Of the remaining 297 patients, 47 had a negative angiogram, and eight had an arteriovenous malformation. Of the 242 consecutive patients with angiographically proved aneurysmal SAH, 131 were not randomly assigned in the study for the following reasons: moribund state (n ⫽ 9), age older than 75 years (n ⫽ 11), time since bleeding more than 3 days (n ⫽ 26), previous surgery for ruptured aneurysm (n ⫽ 3), surgical treatment necessary because of a large hematoma (n ⫽ 35) or because of nerve

compression due to mass effect (n ⫽ 2), endovascular treatment not considered possible in view of the angiographic criteria mentioned earlier (n ⫽ 33) or because an interventional neuroradiologist was not available (n ⫽ 6), surgical treatment not considered possible because of aneurysm location (n ⫽ 2), or the patient refused to participate in the study (n ⫽ 2) or was from abroad (n ⫽ 2). Altogether, 111 patients were randomly assigned. Two patients were excluded from the analyses after assignment. For the analyses, therefore, 52 patients were assigned to the endovascular treatment group, and 57 were assigned to the surgical treatment group. In the two cases of exclusion from the analyses after assignment, the decision to assign the patient had been mistakenly based on the characteristics of an unruptured aneurysm, while the ruptured aneurysm, which was partially or totally thrombosed and misinterpreted at initial diagnostic angiography, was not suitable for treatment with either modality. One patient with two medium-sized aneurysms in the middle cerebral artery (MCA) and a very small aneurysm in the P1 segment underwent surgical ligation of the two MCA aneurysms, one of which seemed irregular at preoperative angiography. At craniotomy, these proved to be regular and nonruptured. At control angiography, the P1 segment aneurysm was substantially larger than it had been at initial angiography and had obviously bled. It was treated endovascularly with complete obliteration but was later treated with surgery. In another patient, an aneurysm in the ophthalmic segment was successfully obliterated with coils. Angiography was repeated 1 week later, after rebleeding, and another aneurysm was detected in the basilar artery trunk. This wide-necked basilar aneurysm, which had been totally thrombosed at initial angiography, was not considered to be suitable for endovascular treatment. Sixty-seven of the study patients had a Hunt and Hess grade of I–II, 26 had a grade of III, and 16 had a grade of IV–V. The basic characteristics of the patients are shown in Table 1. Ninety-eight of the ruptured aneurysms were located in the anterior circulation, and 11 were located in the posterior circulation. The aneurysms had a diameter of 2–15 mm.

Embolization Procedure Diagnostic angiography was performed via the femoral artery through a 6-F introducer. Digital subtraction angiography Vanninen et al

TABLE 1 Patient Characteristics Embolization Group

Surgery Group

Characteristic

No. of Patients

Mean*

No. of Patients

Mean*

All patients Male/female Age (y) Hunt and Hess grade I–II III IV–V Fisher grade 0–2 3–5 Ruptured aneurysm site Anterior circulation MCA Anterior cerebral artery Internal carotid artery Posterior circulation Aneurysm size All aneurysms Small (ⱕ7 mm) Medium (8–14 mm) Large (15–24 mm) Giant (ⱖ25 mm) Ruptured aneurysm neck size Small (⬍4 mm) Wide (ⱖ4 mm)

52 28/24 NA

NA NA 49 (16–73)

57 23/34 NA

NA NA 50 (14–75)

31 12 9

NA NA NA

36 14 7

NA NA NA

20 32

NA NA

22 35

NA NA

46 7 27 12 6

NA NA NA NA NA

52 12 28 12 5

NA NA NA NA NA

NA 39 13 0 0 NA 38 14

6 (2–14)† NA NA NA NA 3 (1–7)† NA NA

NA 41 15 1 0 NA 46 11

7 (2–15)† NA NA NA NA 3 (1–8)† NA NA

Note.—There were no statistically significant differences between the embolization and surgery groups. NA ⫽ not applicable. * Numbers in parentheses are the range. † Data are in millimeters.

equipment (Polytron; Siemens Medical Engineering Group, Erlangen, Germany) with a 1,024 ⫻ 1,024 matrix was used. One patient who had undergone previous femoral bypass surgery underwent direct carotid punctures for diagnostic angiography and the subsequent embolization. In all other patients, the internal or common carotid artery and the vertebral arteries were selectively catheterized. Angiography usually was immediately followed by random assignment and, if indicated, endovascular treatment during the same session. At angiography, the site, orientation, and morphology of the aneurysm were carefully evaluated, and the size of the aneurysm and the width of the neck were measured with a digital caliper by using two coins as a reference. All embolization procedures were performed by one or both of two interventional neuroradiologists (T.S., R.V.). A 6-F guiding catheter was advanced to the internal carotid artery near the level of the skull base or to the upper cervical portion of the vertebral artery. A microcatheter (Tracker-10 or Tracker-18; Target Therapeutics, Fremont, Calif) with two tip markers and a guide wire (Dasher-14, usually followed by a Dasher10; Target Therapeutics) were used for superselective catheterization of the aneurysmal Volume 211 • Number 2

sac. Because the aneurysms tended to be very small or small, the Tracker-10 microcatheter was preferred for catheterization because of the smaller risk of perforation. In cases of tortuous vessels, however, when kinking of the smaller catheter prohibited smooth movement of the guide wire or coil, the Tracker-18 catheter enabled delivery of the coils. The distal end of the microcatheter was shaped with steam to conform to the vascular geometry before catheterization. By using digital road mapping, care was taken not to touch the aneurysmal wall with the guide wire or catheter. Intraaneurysmal angiograms (ie, with contrast material directly injected into the aneurysm) were not obtained. A continuous, pressurized flush of heparinized saline solution was maintained in both the guiding catheter and the microcatheter. Once catheterization had been achieved, the sac was filled with Guglielmi detachable coils (GDCs) (GDC-10, GDC-10 Soft, or GDC-10 2-Diameter; Target Therapeutics), which can be electrolytically detached. Complete occlusion of the aneurysmal sac and neck was always attempted. The largest coil, which was selected according to the measured aneurysm diameter, was positioned first to form a basketlike frame in the aneurysm.

The smaller coils were then sequentially delivered into the aneurysm until the lumen was completely occluded and flow inside the aneurysm, as well as in the secondary pouch, was arrested (Fig 1). In aneurysms with an oblong or irregular shape, filling was started at the rupture site, when identifiable, usually in the dome of the aneurysm. If the size of the selected coil proved to be unsuitable, the GDC system allowed removal of the coil and repositioning of the mesh to an optimal position.

Surgery Surgical procedures were performed under balanced anesthesia by a team of neurosurgeons (including T.K., J.H., M.V.) with a collective experience of approximately 2,000 aneurysm operations. A pterional approach was used in all but six cases: Three pericallosal aneurysms were approached frontally and interhemispherically, one posterior inferior cerebellar artery aneurysm was approached suboccipitally and laterally, and two basilar artery aneurysms were approached subtemporally. A standard microsurgical method was used for clipping of the aneurysm neck with a Sugita (Mizhuo Itatogyo, Tokyo, Japan) or Aesculap (AesculapPlatz, Tuttlingen, Germany) clip. If feasible, the aneurysm was opened, coagulated, or both. Hyperventilation was used routinely. All patients received corticosteroids and mannitol. Control angiograms were obtained in most patients during their first hospital stay.

Patient Care In all grade IV–V patients and in a majority of grade III patients, embolization was performed with use of general anesthesia. In patients with a grade I–II aneurysm and in some with a grade III aneurysm, no method of sedation was used. During the procedure, systemic administration of heparin (initial bolus of 5,000 IU in the first five patients and of 2,500 IU in the remaining patients, followed by 1,000 IU after an hour and according to the measured activated clotting time thereafter), nimodipine, corticosteroids, and aspirin (500 mg orally or via a nasogastric tube) were used. Regardless of the method used to secure the ruptured aneurysm, every patient was treated similarly with hypervolemic therapy, prolonged bed rest (10 days), corticosteroids, and intravenous nimodipine. In the intensive care unit, hypervolemia was ascertained at SwanGanz catheter placement and measure-

Endovascular versus Surgical Treatment for Ruptured Intracranial Aneurysms • 327

a.

c.

e.

b.

d.

f.

Figure 1. Angiograms show aneurysms that were occluded at the first embolization session. (a, b) A small aneurysm (arrow) of the right posterior communicating artery is shown (a) before and (b) after embolization. (c, d) An aneurysm (arrow) of the right MCA is shown (c) before and (d) after embolization. An intracerebral hematoma was diagnosed after the procedure (see Table 2), and this aneurysm was subsequently treated with surgery. (e, f) A very small, atypical aneurysm (arrow) of the distal portion of the left posterior inferior cerebellar artery is shown (e) before and (f) after embolization.

ment of the pulmonary artery wedge pressure. Ventricular drainage was performed immediately after embolization, when indicated. Heparin administration was not routinely continued in the intensive care unit. Oral administration of aspirin (250 mg daily) was continued for 3 months after embolization.

Statistical Analyses All analyses were performed with the SPSS PC⫹ statistical package (version 6.01; SPSS, Chicago, Ill). Differences were considered to be statistically significant if the 328 • Radiology • May 1999

P value was less than .05. The influence of treatment modality and different primary clinical and anatomic factors on the observed angiographic results and the 3-month clinical outcome were analyzed. The ␹2 test for dichotomous discrete variables and the Mann-Whitney U test for continuous or ordinal scale variables were used for comparisons between patients in the embolization group and those in the surgery group. The influence of the primary clinical parameters and anatomic factors on the observed 3-month clinical outcome was analyzed. Potential predictors that showed statistically significant

associations in univariate analyses with the Glasgow Outcome Scale (22) score as determined at 3 months were included in a stepwise multiple logistic regression analysis.

RESULTS Comparability of the Study Groups There were no significant differences between the endovascular and surgical treatment groups in terms of age, sex, Hunt and Hess grade, Fisher grade, or site or size of the ruptured aneurysm (Table 1). Vanninen et al

TABLE 2 Technical Complications Associated with Endovascular Treatment Patient Hunt and Sex/Age (y) Hess Grade

Aneurysm Neck Size (mm) Size (mm)

Aneurysm Site

M/31 M/38 F/73

II II IV

Pericallosal artery Anterior communicating artery MCA

3 3 6

2 2 3

M/41 F/60 M/39 F/48 M/31 F/38

II I III III II I–II

MCA Anterior communicating artery MCA Anterior communicating artery Pericallosal artery MCA

7 5 12 4 8 12

4 4 3 2 3 3

Technical Complications of Endovascular Treatment Complications associated with the embolization procedure are shown in Table 2. Four technical complications had no clinical consequences and necessitated no therapy. Of five patients who underwent additional direct surgery, three had a good recovery, one had a severe disability, and one died because of early rebleeding from the incompletely occluded aneurysm. Thus, the mortality rate due to the technique was one death (2%) in 52 patients. The patient who experienced rebleeding during the 1st day after embolization had an incomplete occlusion of an MCA aneurysm (Table 2). Immediately after embolization, another patient was found to have a previously undiagnosed intracerebral hematoma (Table 2). During the procedure, nothing indicated iatrogenic perforation, and complete angiographic occlusion was achieved (Fig 1c, 1d). The time of the occurrence of the second hemorrhage was not clinically obvious, and it is possible that the rebleeding occurred in a patient under anesthesia before embolization. After these two complications, which occurred early in the study, the heparin bolus was reduced from 5,000 to 2,500 IU. In three patients, the aneurysm wall was perforated during coil delivery (Table 2). In one patient with an MCA aneurysm, slight extravasation of contrast material was noted during embolization. Coil delivery was continued, and complete obliteration of the aneurysm was achieved. In two patients who did not receive anesthesia, perforation was followed by severe restlessness in one and by convulsions and unconsciousness in the other, which prevented further delivery of coils. These two patients subsequently underwent surgery and recovered uneventfully. Volume 211 • Number 2

Complication Perforation Perforation Rebleeding after unsuccessful occlusion Hematoma, possible perforation Parent vessel obstruction Perforation Coil protrusion Transient ischemic attack Coil migration of 2 cm

Parent vessel obstruction caused by coils protruding from the aneurysm neck occurred in three patients (Table 2). The protruding coils caused a slight flow obstruction as seen at angiography in a 60-year-old woman with a wide-necked aneurysm of the anterior communicating artery and an aplastic left A1 segment. Because of the obstruction and previously diagnosed polycythemia vera, heparin treatment was continued in the intensive care unit. This treatment led to obvious thrombolysis in the aneurysm, however, and refilling of the secondary pouch was detected on the control angiogram. As a consequence, surgery was performed. Coil protrusion from an anterior communicating artery aneurysm toward the parent vessel also was detected in a 48year-old woman. This protrusion did not seem to cause any flow obstruction, and no treatment was considered to be necessary. Later, however, the patient developed severe vasospasm with ischemia, which, at brain CT, manifested bilaterally as hypoattenuating areas in both the pericallosal and middle cerebral artery territories. Despite aggressive treatment, including balloon angioplasty, the patient remained in a vegetative state. In a third patient with a medium-sized MCA aneurysm, the first coil was positioned completely within the aneurysmal sac at the time of detachment. After detachment, however, approximately 2 cm of the proximal end of the coil migrated into an MCA branch; there were no clinical consequences, and the aneurysm was successfully occluded. A 31-year-old man with a left-sided pericallosal aneurysm had transient paresthesia of the right leg during coil delivery (Table 2). His symptoms were relieved before any medication was administered, and no hypoattenuating areas were visible on the postembolization CT scan. There were no other obvious thrombo-

Treatment

Outcome

Surgery Surgery Surgery

Good recovery Good recovery Death

Surgery Heparin, surgery Coil delivery None None None

Severe disability Good recovery Good recovery Vegetative state Good recovery Good recovery

embolic complications in the patients. At CT, however, one patient with aortic and mitral regurgitation and hypertrophic cardiomyopathy had multiple small, hypoattenuating lesions in the watershed areas of both hemispheres; these lesions were interpreted as being associated with general hypoperfusion rather than with thromboembolism caused by the procedure.

Surgical Results Fifty of the 57 patients underwent postoperative control angiography; in the remaining seven, the occlusion rate was confirmed only at surgery. A ruptured aneurysm was totally ligated in 42 (74%) patients (Table 3). A small neck remnant was detected at postoperative angiography in nine (16%) patients. One of these neck remnants was embolized; the other eight patients with a neck remnant will undergo further angiographic follow-up. In five (9%) patients, a substantial residual part of the aneurysm fundus remained nonoccluded after surgery. Two of these residual aneurysms (one at the basilar artery bifurcation, one in the anterior communicating artery) were subsequently treated by means of embolization, with the result of total obliteration (Figs 2, 3); one underwent spontaneous thrombosis; and one was being followed up. One residual basilar artery aneurysm was treated again with surgery after 3 months, but the patient’s condition deteriorated to a vegetative state. One aneurysm of the basilar artery bifurcation ruptured during surgery, and the patient died. One MCA aneurysm ruptured during surgery; the aneurysm was successfully ligated, but the patient developed severe edema of the MCA territory and died 2 days later. Another patient, whose MCA aneurysm ruptured during surgery, had hypoattenuating areas in the MCA territory at postoperative CT, which were indicative of ischemia.


TABLE 3 Control Angiographic Results of Early Endovascular or Surgical Treatment of Ruptured Aneurysms Endovascular Treatment Location All aneurysms Anterior circulation MCA ACA‡ ICA§ Posterior circulation

Surgical Treatment*

Total Obliteration

Neck Remnant

Marked Filling of Fundus

No Occlusion

Total Obliteration

Neck Remnant


No Occlusion

P Value†

26

18

3

5

42

9

5

1

.015

4 13 6 3

1 9 5 3

1 1 1 0

1 4 0 0

10 23 8 1

2 4 3 0

0 1 1 3

0 0 0 1

.186 .005 .468 .045

* Control angiography was not performed in seven patients, and results were evaluated at surgery. † Statistical results determined with the Mann-Whitney U test. ‡ ACA ⫽ anterior cerebral artery, anterior communicating artery, and pericallosal artery. § ICA ⫽ internal carotid artery, ophthalmic artery, posterior communicating artery, and anterior choroidal artery.

a.

b.

c.

Figure 2. Subsequent embolization after surgical difficulties. (a) Angiogram in a patient who was randomly assigned to the surgical treatment group shows an aneurysm (arrow) of the basilar artery bifurcation. (b) Control angiogram shows that a substantial part of the fundus (arrows) remains open after surgery. Lateral angiogram (not shown) demonstrated that only a small anterior portion of the aneurysmal sac appeared to be ligated. (c) Angiogram shows that the residual aneurysm was subsequently occluded by endovascular means.

One patient developed an abscess in the frontal lobe 2 months after ligation of an aneurysm of the anterior communicating artery, and surgical evacuation was required. The mortality rate due to the technique was two deaths (4%) in 57 patients.

Primary Angiographic Results The primary angiographic results in the endovascular and surgical treatment groups according to aneurysm location are shown in Table 3. Note that the numbers in Table 3 include the initial angiographic results in patients who experienced technical failure and complications and who were subsequently treated with surgery. According to the immediate postembolization angiographic results, the aneurysm sac was completely obliterated with coils at the first session in 26 (50%) of the 330 • Radiology • May 1999

endovascularly treated patients (Fig 1). Subtotal occlusion, with a small neck remnant, was achieved in 18 (35%) patients (Figs 4, 5), and some filling of the fundus of the aneurysm was present in three (6%). The remaining patients subsequently underwent open surgery after an attempt at endovascular treatment that was unsuccessful because the neck of the aneurysm proved to be too wide to retain coils (three [6%] patients) or because of one of the previously described technical complications (five [9%] patients). Of the small aneurysms with a small neck, 61% were totally obliterated at the first session. In aneurysms located in the posterior circulation (n ⫽ 11), the primary angiographic results were significantly better in the endovascular treatment group (P ⫽ .045), whereas in anterior cerebral artery aneurysms (n ⫽ 55), the angiographic

results were significantly better in the surgical group (P ⫽ .005). In aneurysms of the MCA (n ⫽ 19) or the internal carotid artery (n ⫽ 24), no significant differences in angiographic results were seen between the treatment groups (Table 3).

Crossover between Treatment Groups Eleven patients underwent crossover from one treatment modality to the other. As a result of the aforementioned technical complications in endovascular treatment, five patients underwent crossover from endovascular to surgical treatment. In addition, three patients subsequently underwent open surgery after attempts at endovascular treatment failed because the neck of the aneurysm proved to be too wide to retain coils. Three patients underwent crossover from surgical to endovasVanninen et al

a.

b.

c.

Figure 3. Subsequent embolization after surgical difficulties. (a) Angiogram shows a lobulated aneurysm (arrow) of the anterior communicating artery. (b) Angiogram obtained after surgery shows that the aneurysm (arrow) is only partially ligated. (c) Angiogram obtained after endovascular treatment shows the aneurysm filled with coils. A tiny neck remnant (arrowhead) is still visible.

a.

b.

c.

Figure 4. Filling of the aneurysm neck after the first embolization session. (a) Angiogram shows an aneurysm (arrow) of the basilar artery bifurcation. (b) Angiogram shows that the aneurysm was only partially occluded at the first embolization session and that a residual neck remnant (arrowhead) is present. There was concern about the patency of the posterior cerebral arteries. (c) Control angiogram shows that the neck remnant thrombosed spontaneously, with no obstruction of the parent arteries.

cular treatment because of residual aneurysms after surgical treatment.

Clinical Problems Associated with SAH In the endovascular treatment group, three patients required insertion of a permanent shunt device because of hydrocephalus. A shunt insertion operation was required more frequently (11 patients) in the surgical group (P ⫽ .045). No statistically significant difference between treatment groups was observed in Volume 211 • Number 2

terms of the number of patients with clinically symptomatic vasospasm. In severe cases where aggressive treatment— including induction of hypervolemia, hypertension, and hemodilution—was unsuccessful, intracranial balloon angioplasty (three patients) and superselective papaverine injections (one patient) were attempted.

Three-month Angiographic Results Altogether, 39 endovascularly treated patients underwent control angiography

at 3 months; eight patients had undergone surgery, and five had died. Table 4 summarizes these 3-month angiographic results, which include the subsequently successful results of the second embolization performed in three patients (Fig 5) and the results in four patients with spontaneous thrombosis of the neck remnant (Fig 4). In 26 patients, the aneurysm was seen at angiography to be totally obliterated, and a small but acceptable neck remnant remained in 11 aneurysms. In 95% of the patients, therefore, the angiographic result was acceptable. In two pa-


tients, the coils were pressed against the fundus of the sac and collapsed, so that a substantial part of the aneurysm was open. Because the neck-to-fundus ratio of the residual aneurysm was unfavorable, repeat embolization was not considered to be possible. An MCA aneurysm in a 38-year-old patient and a pericallosal aneurysm in a 40-year-old patient (Fig 6) were therefore treated surgically.

Treatment for Nonruptured Aneurysms Twenty-eight (26%) patients had a total of 43 additional nonruptured aneurysms. These additional aneurysms were treated during the same operation, when possible, in the surgical group, and by means of embolization after the acute phase of SAH in the endovascular treatment group, if the aneurysms were suitable for embolization. At the time of this writing, 17 nonruptured aneurysms had been treated surgically, two were embolized, and the rest were being followed up with angiography.

a.

b.

c.

d.

Three-month Clinical Outcome The 3-month clinical outcome, as determined with the Glasgow Outcome Scale, for patients in both treatment groups (according to the intended treatment modality) was comparable (P ⫽ .36, Table 5). Forty-two (81%) patients initially assigned to the endovascular treatment group and 45 (79%) assigned to the surgical treatment group showed good or moderate recovery. Four (8%) patients in the endovascular treatment group and six (11%) in the surgical treatment group had a severe disability or were in a vegetative state. Six (12%) patients in the endovascular treatment group and six (11%) in the surgical treatment group died. There were no statistically significant differences in outcome between treatment groups in terms of anatomic location of the aneurysm or initial Hunt and Hess or Fisher (23) grades. At univariate analyses of dichotomous variables, the location or size of the ruptured aneurysm, the age or sex of the patient, and the treatment modality (intended treatment) were not statistically significant determinants of the 3-month outcome as determined with the Glasgow Outcome Scale. Because the Hunt and Hess grade (P ⬍ .001), the Fisher grade (P ⫽ .009), and the width of the aneurysm neck (P ⫽ .034) were significantly associated with the Glasgow Outcome Scale result, they were included in a back332 • Radiology • May 1999

Figure 5. Filling of the aneurysm neck after the first embolization session. (a) Angiogram shows a relatively wide-necked aneurysm (arrow) of the anterior communicating artery. (b) Angiogram shows a small filling at the aneurysm neck and rather poor packing of the coils. (c) Control angiogram shows that the coils have further compressed toward the aneurysmal fundus, and the neck remnant (arrowhead) has markedly increased in size. (d) Angiogram obtained after repeat embolization shows the residual aneurysm filled with coils, resulting in occlusion.

ward stepwise multiple logistic regression analysis as independent variables. For the final model, the following coefficients of determination were obtained: Hunt and Hess Grade, 0.35; and width of the neck, 0.13.

DISCUSSION Only a certain proportion of patients with recently ruptured intracranial aneurysms are suitable candidates for endovascular treatment. In the present study, with relatively broad inclusion criteria, fewer than half of the patients with angiographically proved aneurysmal SAH were

suitable candidates for endovascular treatment. One of the most important factors limiting the use of endovascular treatment of intracranial aneurysms is the width of the aneurysm neck (24), and this was obvious in the present study.

Alternative Endovascular Techniques Promising results have recently been reported (20,25,26) about the occlusion of wide-necked aneurysms by using temporary inflation of a nondetachable balloon in front of the aneurysm neck during each coil placement (the so-called Vanninen et al

TABLE 4 Angiographic Results at 3-month Follow-up after Early Endovascular Treatment of Ruptured Aneurysms according to Intended Treatment Aneurysm Site

Total Obliteration

Neck Remnant


No Occlusion

Total

26

11

2

0

39

2 14 7 3

1 4 4 2

1 1 0 0

0 0 0 0

4 19 11 5

All aneurysms Anterior circulation MCA ACA* ICA† Posterior circulation

Note.—Of the 52 patients initially assigned to the endovascular treatment group, five died and eight underwent surgery after failed or complicated endovascular treatment. * ACA ⫽ anterior cerebral artery, anterior communicating artery, and pericallosal artery. † ICA ⫽ internal carotid artery, ophthalmic artery, posterior communicating artery, and anterior choroidal artery.

remodeling technique), a method also presently used at our institution. This technique is most suitable for relatively proximal aneurysms occurring in a fairly straight arterial portion. Autologous vein– covered stents (27) or a combination of stents and coils have been used to treat wide-necked carotid artery aneurysms in an experimental setting in a canine model (28), as well as in a clinical setting (29), but such devices are dependent on the aneurysm being located in a straight portion of a vessel. Another possibility for endovascular treatment of intracranial aneurysms is the use of mechanical detachable coils, which have been available since 1994 (16,19). The advantage of these coils is more rapid detachment and lower cost. Cognard et al (19) reported their experience with mechanical detachable coils in 60 aneurysms. Attempted embolization failed in five aneurysms, and, although a 3-month control angiogram was available for only 33 berry aneurysms, 23 aneurysms (70%) were totally occluded. Tournade et al (16), in their study with 53 patients (48 with symptomatic SAH), achieved total occlusion in 47 of 56 saccular aneurysms by using mechanical detachable coils. Thus far, no comparative studies on the effectiveness of the use of GDCs versus mechanical detachable coils have been published, to our knowledge.

Angiographic Results In our study, the primary angiographic results were significantly better in the surgical group when the aneurysm was located in the anterior cerebral artery territory. Surgical treatment for these aneurysms usually is straightforward, whereas catheterization and endovascular treatment of pericallosal aneurysms, in particular, may be difficult and prone to complications due to catheter kinkVolume 211 • Number 2

ing and the small caliber of the vessels. We encountered two aneurysm perforations in this anatomic area. No coils were delivered in two additional patients, because the size of the aneurysm was too small for the smallest coil available in one patient, and the neck of the aneurysm was wider than the fundus at superselective angiography in the other. In retrospect, we can say that these aneurysms should not have been included in the random assignment to a treatment modality. However, when the 3-month clinical outcome in patients with an aneurysm of the anterior cerebral artery was compared on the basis of intended treatment modality, no differences were observed between the treatment groups. The primary angiographic results were significantly better in the endovascular treatment group when the aneurysm was located in the posterior circulation. It is well known that the surgical treatment of some aneurysms of the posterior circulation may be cumbersome in the acute phase of SAH because of the difficult anatomic circumstances (30–32). Accordingly, only one (20%) posterior circulation aneurysm was totally obliterated at surgery. However, the 3-month clinical outcome did not significantly differ in the two treatment groups in terms of the intended treatment modality. For aneurysms of the internal carotid artery, there were no statistically significant differences between the two groups in terms of the primary angiographic results. No complications were encountered at endovascular treatment, and 92% (11 of 12) of the patients showed acceptable primary angiographic results, whereas the remaining patient (with marked filling of the fundus) was in such poor overall condition that additional treatment was not attempted. Aneurysms of the internal carotid artery usually are well suited for endovascular treatment

because of their proximal location, but the surgical results in the present study were equally good. For MCA aneurysms, no statistically significant differences in angiographic results were observed, probably because of the relatively small number of patients. At first, a substantial proportion of all patients with MCA aneurysms were excluded from the random assignment because the patients had hematomas necessitating surgical treatment or the anatomic condition (wide neck) was considered to be unsuitable for endovascular treatment. Endovascular treatment of MCA aneurysms was especially frequently associated with technical difficulties, with complications occurring in 57% (four of seven) of such patients, two of whom required early surgical treatment, while the third underwent late surgery because of coil compression. In addition, one aneurysm was treated with surgery after selective catheterization revealed that the neck of the aneurysm was actually so wide that coil delivery was not attempted. It is, therefore, our experience that MCA aneurysms are seldom suitable for endovascular treatment. Vin ˜ uela et al (5) recently reported the substantial experience of eight interventional neuroradiology centers in the United States where GDCs were used in the treatment of 403 patients with acute SAH. They reported complete occlusion of the aneurysm in 70.8% of small aneurysms with a small neck, in 35% of large aneurysms, and in 50% of giant aneurysms. The present study differs markedly in its design from the multicenter study reported by Vin ˜ uela et al. First, the patients in our study were randomly assigned to receive either endovascular or surgical treatment, whereas the patients in the multicenter study included patients with intracranial aneurysms that were considered by a vascular neurosurgeon to be unclippable and patients for whom the surgical procedure entailed a high risk of morbidity and/or mortality. As a consequence, 10% of the aneurysms in the present study versus 57% of those in the multicenter study were located in the posterior circulation. In addition, our patients were treated within 72 hours, whereas those in the multicenter study were treated within 15 days of the primary hemorrhage. The total aneurysm occlusion rate in our study was 50% of all aneurysms and 61% of small aneurysms with a small neck.

Surgery after Attempted Endovascular Treatment Five patients randomly assigned to the endovascular treatment group under-


went surgery because of a technical complication. Two perforated aneurysms were surgically ligated without complications. In cases where the aneurysm was already packed with coils, ligation was more complicated. One day after incomplete embolization of an MCA aneurysm, a 73-yearold patient experienced rebleeding, which necessitated urgent surgery. While the hematoma was being evacuated, the aneurysm started to bleed from the ruptured fundus. With the help of temporary clipping, the aneurysm was opened, the coils were removed, and the neck of the aneurysm could then be ligated. The patient, however, remained unconscious and died 11 days later. Another patient, who had a previously undiagnosed hematoma after coil placement, had an aneurysm that proved during surgery to be totally occluded. The coils were nevertheless removed, and the aneurysm was ligated the same way as in the previously mentioned patient. In the patient in whom protruding coils caused obstruction of the pericallosal artery, temporary ligation of the right A1 segment made it possible to remove the coils and finally ligate the aneurysm. Unexpected surgical difficulties may be encountered in the treatment for a recurrent aneurysm after coil embolization (33,34). For example, two patients in the present study underwent surgery after the 3-month control angiogram had been obtained. In one patient, ligation of a pericallosal aneurysm was uneventful: The fundus of the ligated aneurysm was opened, but the coils were left in place. In the other patient, surgery of a residual MCA aneurysm was extremely difficult, because it was impossible to remove the tightly attached coils. Temporary clips were used, and the clip eventually had to be replaced over the coil mesh in the neck of the aneurysm. The patient had a good recovery.

Thromboembolic Complications Cognard et al (19,20) reported angiographic signs of thrombosis in 13 (22%) of 57 cases in which mechanical detachable coils were used (19) and in 27 (11%) of 249 procedures in which the use of GDCs was attempted (20); the complications they reported included those that necessitated administration of urokinase. These results are clearly different from ours: We encountered only one mild thromboembolic complication, which resolved before any treatment. This difference is not explained by the use of anticoagulation therapy, because the heparin 334 • Radiology • May 1999

a.

b.

c.

d.

Figure 6. Filling of the aneurysm neck after the first embolization session. (a) Angiogram shows a very small aneurysm (arrow) of the pericallosal artery. (b) Angiogram obtained after initial embolization with one coil shows that the aneurysm is occluded (arrow). (c) Routine control angiogram obtained 3 months after b, however, shows a recurrence of the aneurysm (arrow). Attempted catheterization of the aneurysm was hindered by severe vasospasm in the petrous part of the internal carotid artery, and the aneurysm was treated with surgery. (d) Postoperative control angiogram shows the aneurysm to be obliterated (arrow).

doses were higher and the duration of treatment was longer in the two studies of Cognard et al (19,20). However, Cognard et al did not mention the use of a pressurized flush of heparinized saline solution in the guiding catheter and microcatheter. One explanation for the difference in the number of thromboembolic complications may be the timing of the procedure: All of the patients in our study were treated during the acute phase, before the critical time for vasospasm.

Clinical Outcome Our results indicate that, according to the intended treatment modality, the primary outcome in patients was comparable in the endovascular and surgical treatment groups. In accordance with the results published in previous studies (1,3,6,7,35) on surgical treatment of aneurysms, the clinical outcome was dependent mostly on the initial Hunt and Hess grade. There were no interval bleedings Vanninen et al

TABLE 5 Clinical Outcome (Glasgow Outcome Scale) at 3 Months after Endovascular or Surgical Treatment for Ruptured Aneurysm according to Intended Treatment Endovascular Treatment Outcome

Parameter All patients Hunt and Hess grade I–II III IV–V Fisher grade 0–2 3–5 Aneurysm site Anterior circulation MCA ACA* ICA† Posterior circulation

Surgical Treatment Outcome

Severe Good or Severe Good or Disability or Moderate Disability or Moderate Recovery Vegetative State Death Recovery Vegetative State Death 42

4

6

45

6

6

28 11 3

2 0 2

1 1 4

32 11 2

1 2 3

3 1 2

19 23

1 3

0 6

20 25

0 6

2 4

5 23 9 5

1 1 2 0

1 3 1 1

9 23 10 3

2 2 1 1

1 3 1 1

Note.—There were no significant differences between treatment groups in any of the parameters (Mann-Whitney U test). * ACA ⫽ anterior cerebral artery, anterior communicating artery, and pericallosal artery. † ICA ⫽ internal carotid artery, ophthalmic artery, posterior communicating artery, and anterior choroidal artery.

after coiling other than that in the patient with an incomplete occlusion of an MCA aneurysm who experienced rebleeding during the 1st day after embolization. When we compare our results from the endovascular treatment and the surgical treatment groups, it is noteworthy that most of the surgical procedures were performed or supervised by a team of neurosurgeons with a collective experience of approximately 2,000 aneurysm operations. Because GDC embolization is a relatively new technique, the experience of interventional neuroradiologists is naturally much more limited. A certain learning curve can be observed in the present study, because most of the complications occurred in patients treated at the beginning of the study. To maintain and develop technical skills in this rapidly developing area, a critical volume of procedures should be maintained for the interventional neuroradiologists and neurosurgeons. Because the GDC technique was not developed until 1989 (24) and was disseminated to selected institutions only some 5 years ago, the long-term effectiveness of GDC embolization is not yet well documented. Malisch et al (9) have recently reported midterm clinical outcomes (range, 2–6 years; average, 3.5 years) in 94 patients treated with GDCs. Their results indicate that the procedure is a safe, effective, and reliable means of preventing aneurysm hemorrhage in patients with small or large intracranial Volume 211 • Number 2

aneurysms. The results of Malisch et al were less satisfactory in patients with a giant lesion, in which a postembolization hemorrhage rate of 33% was found. Ongoing technical developments in the GDC system, variations in patient selection criteria, and the growing experience of interventional neuroradiologists will continue to affect the outcome in these patients.

Hydrocephalus and Vasospasm The need for permanent shunt creation for management of hydrocephalus was more common in the surgical treatment group, in which one-fifth of the patients developed hydrocephalus necessitating shunt creation. This finding was unexpected and clearly differed from the results of our previous study (36), in which only 81 (10%) of 835 surgically treated patients with aneurysmal SAH developed hydrocephalus that necessitated shunt creation. Although the present sample was smaller, the groups were comparable with regard to preoperative hydrocephalus, clinical grade, and severity of cisternal bleeding, and the statistical difference between the treatment groups in the development of hydrocephalus is clear. In theory, rinsing of the basal cisterns should lower the risk of hydrocephalus necessitating shunt creation. We suggest that surgery itself may cause inflammation and aseptic arachnoiditis, which disturb normal cerebrospinal fluid flow, thus leading to hydrocephalus.

Gruber et al (37) recently reported that the infarction rate associated with endovascular treatment was higher than that associated with surgery. Their two treatment groups were not balanced; in patients with a Fisher grade of 3, where the groups contained almost equal numbers of patients, those who underwent endovascular treatment had fewer infarcts. In our randomized study, we did not observe significant differences between treatment groups in the occurrence of vasospasm. It is well known that ischemic infarcts are correlated with a higher Hunt and Hess grade and a higher Fisher grade.

Long-term Effects of Endovascular Treatment In the endovascular treatment group, 18 neck remnants were seen at primary control angiography (Table 3). However, the number of neck remnants had decreased to 11 at 3-month control angiography (Table 4). It must be emphasized that a comparison among the results in Tables 3 and 4 is not straightforward, because, of the 52 patients included in the primary endovascular results, five died and eight underwent surgery after failed or complicated endovascular treatment and are therefore not included in Table 4. In four patients, spontaneous thrombosis of the residual aneurysm occurred (Fig 4). The numbers in Table 4 include the primary results, as well as the subsequent successful results of the second embolization procedure performed in three patients (Fig 5). There also were two patients with clinically important recurrent aneurysms (Table 4, Fig 6). Because the attempted repeat embolization was unsuccessful, these two patients subsequently underwent surgery. It is well known that completely surgically ligated aneurysms seldom recur. However, minor aneurysm neck remnants have been known (38–40) to dilate and form a new aneurysm with a risk of rebleeding. These facts stress the importance of early postoperative angiography and either repeat surgery for the incompletely clipped aneurysm or careful longterm angiographic follow-up (38–41). The long-term effects of endovascular treatment are unknown. Despite dense packing of the coils, the orifice of the aneurysm remains open in the blood circulation. The hemodynamic forces may cause compaction of the coils and consequent recanalization of the aneurysm neck. Further, it is unclear whether angiography can be used to evaluate the aneurysm thrombosis rate or the permanence


of an initially produced thrombosis of the aneurysm. Even totally occluded aneurysms have been reported to recanalize, which stresses the need for late, repeated angiographic follow-up and repeat embolization or surgical ligation of the recanalized aneurysm (34,42–44).

9.

10.

11.

Conclusion Only selected patients with a recently ruptured intracranial aneurysm are candidates for endovascular treatment. Favorable results are achieved in these patients, although subsequent acute or late open surgery is sometimes necessary. In our study, the primary angiographic results were better in the surgical group when the aneurysm was located in the anterior cerebral artery territory and in the endovascular group when the aneurysm was located in the posterior circulation. MCA aneurysms were seldom suitable for endovascular treatment. The 3-month clinical outcome of the patients was comparable in the endovascular and surgical treatment groups according to the intended treatment modality. As expected, the outcome seemed to be mostly dependent on the initial Hunt and Hess grade of the patient. Assessment of the long-term clinical effectiveness of endovascular treatment is currently under study. Both patient groups will undergo careful clinical, neuropsychologic, and radiologic follow-up to assess the long-term outcome and effectiveness of endovascular treatment.

12.

13.

14.

15.

16.

17.

18.

19.

References 1. 2. 3.

4.

5.

6.

7.

8.

Schievink WI. Intracranial aneurysms. N Engl J Med 1997; 336:28–40. Kassell NF, Drake CG. Review of the management of saccular aneurysms. Neurol Clin 1983; 1:73–86. Hernesniemi J, Vapalahti M, Niskanen M, et al. One-year outcome in early aneurysm surgery: a 14 years experience. Acta Neurochir (Wien) 1993; 122:1–10. Fogelholm R, Hernesniemi J, Vapalahti M. Impact of early surgery on outcome after aneurysmal subarachnoid hemorrhage: a population-based study. Stroke 1993; 24: 1649–1654. Vin ˜ uela F, Duckwiler G, Mawad M. Guglielmi detachable coil embolization of acute intracranial aneurysm: perioperative anatomical and clinical outcome in 403 patients. J Neurosurg 1997; 86:475–482. Kassell NF, Torner JC, Haley EC Jr, Jane JA, Adams HP, Kongable GL. The International Cooperative Study on the Timing of Aneurysm Surgery. I. Overall management results. J Neurosurg 1990; 73:18–36. Kassell NF, Torner JC, Jane JA, Haley EC Jr, Adams HP. The International Cooperative Study on the Timing of Aneurysm Surgery. II. Surgical results. J Neurosurg 1990; 73:37–47. ¨ hman J, Heiskanen O. Timing of operation O for ruptured supratentorial aneurysms: a prospective randomized study. J Neurosurg 1989; 70:55–60.

336 • Radiology • May 1999

20.

21. 22. 23.

24.

25.

26. 27.

Malisch TW, Guglielmi G, Vinuela F, et al. Intracranial aneurysms treated with the Guglielmi detachable coil: midterm clinical results in a consecutive series of 100 patients. J Neurosurg 1997; 87:176–183. Guglielmi G, Vinuela F, Duckwiler G, et al. Endovascular treatment of posterior circulation aneurysms by electrothrombosis using electrically detachable coils. J Neurosurg 1992; 77:515–524. Casasco AE, Aymard A, Gobin YP, et al. Selective endovascular treatment of 71 intracranial aneurysms with platinum coils. J Neurosurg 1993; 79:3–10. McDougall CG, Halbach VV, Dowd CF, Higashida RT, Larsen DW, Hieshima GB. Endovascular treatment of basilar tip aneurysms using electrolytically detachable coils. J Neurosurg 1996; 84:393–399. Richling B, Gross C, Gruber A, Killer M. Early clinical outcome of patients with ruptured cerebral aneurysms treated by endovascular (GDC) or microsurgical techniques: a single center experience. Intervent Neuroradiol 1995; 1:19–27. Raymond J, Roy D, Bojanowski M, Moumdjian R, L’Esperance G. Endovascular treatment of acutely ruptured and unruptured aneurysms of the basilar bifurcation. J Neurosurg 1997; 86:211–219. Byrne JV, Molyneux AJ, Brennan RP, Renowden SA. Embolisation of recently ruptured intracranial aneurysms. J Neurol Neurosurg Psychiatry 1995; 59:616–620. Tournade A, Courtheoux P, Sengel C, Ozgulle S, Tajahmady T. Saccular intracranial aneurysms: endovascular treatment with mechanical detachable spiral coils. Radiology 1997; 202:481–486. Moret J, Pierot L, Boulin A, Castaings L, Rey A. Endovascular treatment of anterior communicating artery aneurysms using Guglielmi detachable coils. Neuroradiology 1996; 38: 800–805. Nichols DA, Brown RD Jr, Thielen KR, Meyer FB, Atkinson JL, Piepgras DG. Endovascular treatment of ruptured posterior circulation aneurysms using electrolytically detachable coils. J Neurosurg 1997; 87:374–380. Cognard C, Pierot L, Boulin A, et al. Intracranial aneurysms: endovascular treatment with mechanical detachable spirals in 60 aneurysms. Radiology 1997; 202:783–792. Cognard C, Weill A, Castaings L, Rey A, Moret J. Intracranial berry aneurysms: angiographic and clinical results after endovascular treatment. Radiology 1998; 206:499–510. Hunt WE, Hess RM. Surgical risk as related to time of intervention in the repair of intracranial aneurysms. J Neurosurg 1968; 28:14–19. Jennet B, Bond M. Assessment of outcome after severe brain damage: a practical scale. Lancet 1975; 1:480–484. Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery 1980; 6:1–9. Fernandez-Zubillaga A, Guglielmi G, Vinuela F, Duckwiler GR. Endovascular occlusion of intracranial aneurysms with electrically detachable coils: correlation of aneurysm neck size and treatment results. AJNR 1994; 15: 815–820. Moret J, Cognard C, Weill A, Castaings L, Rey A. The ‘‘remodeling technique’’ in the treatment of wide neck intracranial aneurysms: angiographic results and clinical follow-up in 56 cases. Intervent Neuroradiol 1997; 3:21–35. Levy DI, Ku A. Balloon-assisted coil placement in wide-necked aneurysms: technical note. J Neurosurg 1997; 86:724–727. Marotta TR, Buller C, Taylor D, Morris C, Zwimpfer T. Autologous vein–covered stent repair of a cervical internal carotid artery

28.

29.

30.

31. 32.

33.

34.

35.

36.

37.

38.

39.

40. 41.

42.

43.

44.

pseudoaneurysm: technical case report. Neurosurgery 1998; 42:408–412. Szikora I, Guterman LR, Wells KM, Hopkins LN. Combined use of stents and coils to treat experimental wide-necked carotid aneurysms: preliminary results. AJNR 1994; 15:1091–1102. Perez-Cruet MJ, Patwardhan RV, Mawad ME, Rose JE. Treatment of dissecting pseudoaneurysm of the cervical internal carotid artery using a wall stent and detachable coils: case report. Neurosurgery 1997; 40:622–625. Hernesniemi J, Vapalahti M, Niskanen M, Kari A. Management outcome for vertebrobasilar artery aneurysms by early surgery. Neurosurgery 1992; 31:857–861. Peerless SJ, Hernesniemi JA, Gutman FB, Drake CG. Early surgery for ruptured vertebrobasilar aneurysms. J Neurosurg 1994; 80:643–649. Drake CG, Peerless SJ, Hernesniemi JA. Complications of surgery for vertebrobasilar artery aneurysms and final comments. In: Surgery of vertebrobasilar aneurysms: London, Ontario, experience on 1,767 patients. New York, NY: Springer-Verlag, 1996; 300–311. Gurian JH, Martin NA, King WA, Duckwiler GR, Guglielmi G, Vinuela F. Neurosurgical management of cerebral aneurysms following unsuccessful or incomplete endovascular embolization. J Neurosurg 1995; 83:843–853. Mizoi K, Yoshimoto T, Takahashi A, Nagamine Y. A pitfall in the surgery of a recurrent aneurysm after coil embolization and its histological observation: technical case report. Neurosurgery 1996; 39:165–168. Niskanen MM, Hernesniemi JA, Vapalahti MP, Kari A. One-year outcome in early aneurysm surgery: prediction of outcome. Acta Neurochir (Wien) 1993; 123:25–32. Tapaninaho A, Hernesniemi J, Vapalahti M, et al. Shunt-dependent hydrocephalus after subarachnoid haemorrhage and aneurysm surgery: timing of surgery is not a risk factor. Acta Neurochir (Wien) 1993; 123:118–124. Gruber A, Ungersbo ¨ ck K, Reinprecht A, et al. Evaluation of cerebral vasospasm after early surgical and endovascular treatment of ruptured intracranial aneurysms. Neurosurgery 1998; 42:258–268. Lin T, Fox AJ, Drake CG. Regrowth of aneurysm sacs from residual neck following aneurysm clipping. J Neurosurg 1989; 70:556– 560. Suzuki J, Kwak R, Katakura R. Angiographical consideration of residual aneurysms following direct operations on intracranial aneurysms. In: Suzuki J, ed. Cerebral aneurysms; experience with 1000 directly operated cases. Tokyo, Japan: Neuron, 1979; 619–630. Feuerberg I, Lindquist C, Lindqvist M, Steiner L. Natural history of postoperative aneurysm rests. J Neurosurg 1987; 66:30–34. Ebina K, Suzuki M, Andoh A, Saitoh K, Iwabuchi T. Recurrence of cerebral aneurysm after initial neck clipping. Neurosurgery 1982; 11: 764–768. Reul J, Weis J, Spetzger U, Konert T, Fricke C, Thron A. Long-term angiographic and histopathologic findings in experimental aneurysms of the carotid bifurcation embolized with platinum and tungsten coils. AJNR 1997; 18:35–42. Spetzger U, Reul J, Weis J, Bertalanffy H, Thron A, Gilsbach JM. Microsurgically produced bifurcation aneurysms in a rabbit model for endovascular coil embolization. J Neurosurg 1996; 85:488–495. Reul J, Spetzger U, Weis J, Sure U, Gilsbach JM, Thron A. Endovascular occlusion of experimental aneurysms with detachable coils: influence of packing density and perioperative anticoagulation. Neurosurgery 1997; 41: 1160–1165.

Vanninen et al