Clinoid and paraclinoid aneurysms: surgical anatomy, operative techniques, and outcome

Vascular: Aneurysm

Clinoid and Paraclinoid Aneurysms: Surgical Anatomy, Operative Techniques, and Outcome Orlando De Jesu´s, M.D., Laligam N. Sekhar, M.D., F.A.C.S., and Charles J. Riedel, M.D. Department of Neurological Surgery, The George Washington University Medical Center, Washington, DC

BACKGROUND

Paraclinoid or ophthalmic segment aneurysms arise from the internal carotid artery (ICA) between the roof of the cavernous sinus and the origin of the posterior communicating artery. Clinoid aneurysms arise between the proximal and distal carotid dural rings. The complex anatomy of clinoid and paraclinoid ICA aneurysms often makes them difficult to treat by microsurgery. The natural history of these aneurysms varies, based on their location and anatomic relationships. Accurate preoperative assessment of the origin of these aneurysms is therefore a critical aspect of their management.

two were in good condition (independent, but not working), and one died postoperatively of vasospasm. CONCLUSION

Our increased knowledge of anatomy and refinements in operative techniques have greatly improved the surgical treatment of clinoid and paraclinoid aneurysms. © 1999 by Elsevier Science Inc. KEY WORDS

Clinoid aneurysms, contralateral approach, paraclinoid aneurysms.

METHODS

The authors reviewed 35 clinoid and paraclinoid ICA aneurysms operated in 28 patients and classify them according to their anatomic location and angiographic pattern. The operative techniques, surgical outcomes, and indications for surgery are reviewed. RESULTS

Based on surgical anatomy and angiographic patterns, the aneurysms were classified into two categories: clinoid segment and paraclinoid (ophthalmic) segment. The clinoid segment aneurysms consisted of medial, lateral and anterior varieties. The paraclinoid aneurysms could be classified topographically into medial, posterior and anterior varieties, or based on the artery of origin into ophthalmic, superior, hypophyseal, and posterior paraclinoid aneurysms. Ophthalmic aneurysms were most common (40%), followed by posterior ICA wall aneurysms (29%), superior hypophyseal aneurysms (14%), and clinoid aneurysms (17%). Twenty patients (71%) had single aneurysms. Of the remaining eight, six had bilateral aneurysms and two had unilateral multiple aneurysms. Of the 35 aneurysms, 32 were clipped satisfactorily, as confirmed by intraoperative or postoperative angiography. One small broad-based aneurysm was wrapped, and two others were treated by trapping and bypass techniques. Three patients who had bilateral aneurysms underwent successful clipping of four contralateral, left-sided aneurysms via a right frontotemporal, transorbital approach. On follow-up (mean, 39 months), 25 patients were in excellent condition (returned to their prior occupation), Address reprint requests to: Dr. Laligam N. Sekhar, Department of Neurological Surgery, The George Washington University Medical Center, 2150 Pennsylvania Avenue NW, Washington, DC 20037. Received January 13, 1997; accepted June 16, 1998. © 1999 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

he complex extradural and intradural anatomy of clinoid and paraclinoid internal carotid artery (ICA) aneurysms has made surgical treatment of these lesions difficult. The natural history of these aneurysms varies based on their location and anatomic relationships. Accurate preoperative assessment of the origin of these aneurysms is therefore a critical aspect of their management. Even in recently published series, classification of these lesions has been variable [3,5,11–14,18,31]. Many can be classified based on their site of origin from the ICA and their projection. In this paper, we review our experience with a series of 35 such aneurysms, including the indications for surgery, surgical technique, and outcome.

T

Materials and Methods We reviewed our operative experience with 35 clinoid and paraclinoid ICA aneurysms treated between 1988 and June 1996. The majority of the aneurysms were treated at The George Washington University Medical Center; some were treated at the University of Pittsburgh Medical Center by LNS. We obtained patients’ clinical information from their charts and reviewed their arteriograms and intraoperative photographs. Each patient underwent ce0090-3019/99/$–see front matter PII S0090 –3019(98)00137–2

478 Surg Neurol 1999;51:477–88

De Jesus et al

rebral angiography to determine the size, shape, and exact configuration of the aneurysm. Patients with intraluminal thrombosis had computed tomography (CT) and magnetic resonance imaging (MRI) to assess the true size of the aneurysm, to evaluate its relationship to normal brain structures, and to evaluate the extent of calcification of the aneurysm wall. A carotid occlusion test was performed in patients with giant aneurysms, in patients who had calcified aneurysms, and patients who had large or giant aneurysms with ill-defined necks. A bypass was planned if the patient failed the test, if the aneurysm was found to be unclippable, or in any patient if the ICA had to be sacrificed at surgery. Computerized illustrations were made of all the cases, and the lesions were classified according to their microsurgical anatomy. Follow-up ranged from 6 months to 7 years, with a mean of 39 months. SURGICAL TECHNIQUE A balanced anesthetic technique and intraoperative monitoring of the electroencephalogram (EEG) and the somatosensory-evoked potentials are used. A femoral artery catheter is inserted by the radiologist for intraoperative angiography. If temporary occlusion of the ICA may be required for more than 10 minutes, the anesthesiologist prepares for barbiturate- or etomidate-induced metabolic depression to achieve burst suppression on EEG. The patient’s body temperature is lowered to 34°C to reduce brain metabolism. If temporary vascular occlusion is performed, the mean arterial pressure is raised approximately 20% to improve the distal collateral circulation. For ipsilateral aneurysms, a frontotemporal craniotomy is performed up to the supra-orbital notch, followed by an orbital osteotomy (Figure 1). The superior orbital fissure is unroofed extradurally by the removal of the lesser and greater wings of the sphenoid bone, but the anterior clinoid process (ACP) and the optic nerve bony canal (ONC) are left intact (Figure 2). After dural opening, the sylvian fissure is opened and the aneurysm is inspected. If the aneurysm neck is mostly visible, it can be clipped with minimal mobilization of the optic nerve by sectioning the falciform ligament (Figure 3). However, in most cases, the ONC and the ACP will have to be resected. A small flap of dura is created around the ONC and ACP, and with sutures it can be used to protect the aneurysm during drilling (Figure 4A). The ACP and ONC are removed with a high-speed drill and fine rongeurs. If the aneurysm is very thin walled and adherent to the bone, proximal temporary clipping or temporary trapping of

(A & B)Frontotemporal craniotomy and orbital osteotomy for unilateral aneurysms (A), and bilateral aneurysms (B). For bilateral aneurysms, the bone removal extends more medially.

1

the ICA may be necessary during the bone removal. The optic nerve dural sheath is opened widely to mobilize the nerve and to prevent injury during the clipping of the aneurysm (Figure 4B). The distal dural ring will have to be opened in many patients, and if the aneurysm extends into the cavernous sinus (CS), the proximal dural ring will need to be opened as well, with control of CS bleeding with gel foam (Upjohn Pharmaceuticals, Kalamazoo, MI). For large and giant aneurysms, temporary trap-

Clinoid and Paraclinoid Aneurysms

2

Surg Neurol 479 1999;51:477–88

Exposure achieved after bone removal and orbital decompression.

ping of the aneurysm with suction of the blood through the catheter in the superior thyroid artery is necessary to aid in aneurysm decompression and dissection (Figure 5). If a fair amount of thrombus is present inside the aneurysm, a thrombectomy is required. If the aneurysm neck is very atherosclerotic and calcified, it is often safer to trap it permanently after an EC-IC bypass procedure. To aid in the dissection of smaller aneurysms, a balloon catheter is placed by the interventional neuroradiolo-

(A & B) A dural flap has been created around the optic nerve canal and the anterior clinoid process in A. After bone removal, the optic nerve sheath and the distal dural ring have been opened to expose the aneurysm completely (B).

4

3

Sectioning of the falciform ligament to allow the clipping of the aneurysm.

gist with aspiration of blood through the catheter after distal vessel clipping and temporary balloon occlusion (Figure 6) [17]. In the majority of the aneurysms, clip blades are applied parallel to the artery, although a clip placed transverse to the axis of the artery may occasionally work better. Giant aneurysms frequently require multiple clips placed in tandem or in sequence (Figure 7). Large or giant aneurysms on the medial or posterior wall of the ICA frequently require a fenestrated clip. When an aneurysm is unclippable, because the neck is calcified and sclerotic or the ICA is mostly incorporated by the aneurysm with an ill-defined neck, or when the ICA has to be sacrificed because of a tear in the neck, a saphenous vein graft bypass to the M2 segment of

480 Surg Neurol 1999;51:477–88

5

De Jesus et al

Suction decompression of the aneurysm by cannulation of the superior thyroid artery.

the middle cerebral artery (MCA) from the extracranial ICA or the ECA is necessary. The EEG and SSEP are used to guide and warn the surgeon of ischemia. When it occurs, it may be dealt with by raising the blood pressure further, by opening the temporary clips if possible, or by the placement of an IC bypass expeditiously. Completing the distal end of a vein graft bypass before the manipulation of a complex aneurysm is a step that can shorten the duration of the bypass. In general, both paraclinoid and clinoid aneurysms will need to be dissected completely before clipping, to allow the neck to collapse during the clipping process. Posterior paraclinoid aneurysms are often adherent to the superior wall of the CS. Medial paraclinoid aneurysms may be adherent to the optic nerve or the sellar dura mater. Medial clinoid aneurysms are usually adherent to the dura mater lining the sphenoid bone (medial wall of CS) and inferior to the optic nerve. The ophthalmic artery will usually need to be divided to mobilize the carotid artery and the medial clinoid aneurysm before clipping. This can be done safely without visual loss because of collateral flow from the ECA, which can be checked preoperatively by ECA injection during a test ICA occlusion. An intraoperative angiogram (IOA) is performed in all patients, to confirm the occlusion of the aneurysm and the patency of the ICA. Even if the aneurysm is small, the IOA serves to document the aneurysmal occlusion, if regrowth of an aneurysm occurs later. Even in patients with small aneurysms, we have been surprised to find a residual neck or ICA stenosis during IOA.

Suction decompression of the aneurysm by a balloon catheter placed inside the ICA by a transfemoral technique.

6

At the end of the procedure, the dural defect around the ACP is closed by suturing the basal dura circumferentially to the optic nerve sheath and periorbita. Any opening into the sphenoid sinus through the ONC or ACP must be closed with a fat graft, and a small pericranial graft sutured in place. CONTRALATERAL APPROACH Many paraclinoid and some clinoid aneurysms can be clipped from the other side, particularly in patients with bilateral aneurysms, when the brain is slack. The aneurysm must be less than 1.5 cm, have a well-defined neck, and be projecting medially, anteromedially, or anterosuperiorly to be treated by

Clinoid and Paraclinoid Aneurysms

7

Surg Neurol 481 1999;51:477–88

The placement of fenestrated clips to occlude the aneurysm is shown.

the contralateral approach. Posteriorly projecting aneurysms may be difficult to clip, and posterolaterally or laterally projecting aneurysms are dangerous to clip from this approach. Because proximal control of the ICA cannot be easily obtained for contralateral aneurysms, before clipping a balloon can be temporarily inflated in the cervical ICA and a distal clip placed on the supraclinoid ICA with suction decompression, or induced hypotension can be used. Exposure of the ICA superior to the aneurysm can be obtained by working superior to the contralateral CN II. The craniotomy and orbital osteotomy are extended more medially than for ipsilateral aneurysms. Viewing contralateral aneurysms is easier if the optic chasm is normal or post-fixed in position. If necessary, the bony optic canal and dural sheath are opened to decompress the optic nerve, and to prevent its kinking by the clip. If the chiasm is pre-fixed, a flap of dura is dissected over the planum sphenoidale, and the bone is removed with a highspeed drill, to allow aneurysm visualization and dissection (Figure 8, A & B). Longer aneurysm clips are needed than for ipsilateral aneurysms, because of the narrow space. At the end of the procedure, dural reconstruction must be done carefully by primary suture or by using a pericranial graft.

(A & B) The contralateral approach to a medial paraclinoid aneurysm is shown (A) The initial view with the optic nerve stretched over the aneurysm. (B) A portion of the contralateral ONC and planum sphenoidale have been resected, and the opposite optic nerve has been mobilized to expose the aneurysm neck completely.

8

RESULTS DEMOGRAPHICS Our series included 35 clinoid and paraclinoid aneurysms in 28 patients. There were 6 clinoid (5 medial, 1 lateral) aneurysms, 14 ophthalmic aneurysms, 5 superior hypophyseal aneurysms, and 10 posterior paraclinoid aneurysms. Twenty patients (71%) had single aneurysms. Of the remaining eight, six had bilateral aneurysms and two had unilateral multiple aneurysms (Table 1). One patient had a giant intracavernous aneurysm, a large paraclinoid aneurysm, and two middle cerebral artery bifurcation aneurysms on the same side. Aneurysms were classified as small (,1 cm), large (1–2.5 cm), and giant (.2.5 cm). In our series, 54% of the aneurysms were small, 40% were large, and 6% were giant.

482 Surg Neurol 1999;51:477–88

1

De Jesus et al

Description of Aneurysms

NUMBER DESCRIPTION Single aneurysm Bilateral aneurysms L SHA; R SHA L Posterior, PCOM, A chor; R Posterior L Posterior, SHA; R Posterior L Clin, Oph, CS; R Oph, PCOM, A chor L CS; R Clin L PCOM; R Oph Unilateral, multiple L Oph, CS R CS, Oph, MCA 3 2 Totals 5

NUMBER OF ANEURYSMS

OF CLIN/ PATIENTS PARACLIN OTHER

20 6

20 2 2

2

3 3

3

1 1

1 1

1 1

1 3

35

11

2

28

SHA, superior hypophyseal artery; Posterior, ICA-posterior paraclinoid; A chor, anterior choroidal; PCOM, internal carotid-posterior communicating; Oph, ophthalmic; Clin, clinoid segment; CS, intracavernous; MCA, middle cerebral artery.

CLINICAL PRESENTATION The clinical presentation varied with the location of the aneurysm along the ICA. Symptoms included subarachnoid hemorrhage (24%), optic nerve compression with mild-to-severe visual changes (15%), severe headaches (20%), stroke or bleeding not related to the aneurysm (13%), and unrelated symptoms like dizziness or vertigo (13%). One patient with six aneurysms (including one clinoid and two paraclinoid aneurysms) presented with a complete oculomotor palsy secondary to compression from a posterior communicating artery aneurysm. No patient presented with endocrine deficits. Presenting

2

symptoms were classified according to the aneurysm type (Table 2). Optic nerve compression was more common in ophthalmic and superior hypophyseal aneurysms, while headache was the main symptom in patients with posterior ICA aneurysms. Subarachnoid hemorrhage was common to all types of aneurysms. CLASSIFICATION The aneurysms in this series can be classified as belonging to the clinoid or ophthalmic (paraclinoid segment) of the ICA. The clinoid segment lies between the two dural rings, and is considered by many to be an extracavernous, but extra-arachnoid segment of the ICA. The ophthalmic segment extends from the superior dural ring to the posterior communicating artery. Aneurysms of the clinoid segment arise without an apparent branch site of the ICA. They may be medial, lateral, or anterior clinoid, depending on the surface of the ICA from which they originate. Posterior clinoid segment aneurysms have not been observed. Medial clinoid aneurysms project medially inferior to the optic nerve, and on angiography, they are located medial to the anterior bend of the ICA, especially on anteroposterior and oblique views (Figure 9, A & B). In our series, two of the aneurysms projected into an empty sella, exposing them to the subarachnoid space, and two others had small extensions into the subarachnoid space inferior to the optic nerve. Three of these aneurysms presented with subarachnoid hemorrhage. This group of aneurysms has been classified as “carotid cave aneurysms” by Kobayashi [12]. Lateral clinoid aneurysms arise from the lateral aspect of the clinoid ICA (Figure 9, C & D). These aneurysms can project directly into the subarachnoid space or grow anterolaterally and erode the anterior clinoid process. Care has to be taken while removing the ACP to avoid aneurysmal rupture, and

Summary of Presenting Symptoms According to Aneurysm Type SYMPTOM

Headache Subarachnoid hemorrhage SAH—other aneurysm Intracerebral hemorrhage* Stroke CN II CN III or VI palsy—Another aneurysm Vertigo, dizziness Asymptomatic

CLINOID N 5 6

OPHTHALMIC N 5 14

— 4 (67%) — — 1 (20%) — — 1 (20%)

SHA 55

POSTERIOR N 5 10

TOTAL ANEURYSMS N 5 35

2 (9%) 2 (9%) 1 (9%) 2 (18%)

— 1 (20%) — 1 (20%)

4 (40%) 1 (10%) —

2 (18%) 2 (9%) — 2 (18%)

2 (40%) — — 1 (20%)

7 (20%) 8 (24%) 1 (3%) 3 (10%) 1 (3%) 4 (13%) 2 (6%) 1 (3%) 7 (23%)

N

— 1 (10%) 3 (30%)

*One was attributable to hypertension; two were attributable to an associated arteriovenous malformation.

Clinoid and Paraclinoid Aneurysms

Surg Neurol 483 1999;51:477–88

(A–J) Computerized drawings made from patients’ angiograms showing different types of aneurysms. (A & B) A medial clinoid aneurysm in a patient who presented with subarachnoid hemorrhage. The patient shown in C and D had bilateral ICA aneurysms, on the ipsilateral side ICA-ophthalmic, ICA-PCOM, and ICA anterior choroidal, and on the contralateral side, an ophthalmic, lateral clinoid (arrow), and a small intracavernous (not shown). The contralateral aneurysms shown here were clipped by an ipsilateral approach. The posterior paraclinoid aneurysm shown in E and F were successfully clipped (a small ophthalmic aneurysm is also present). (G & H) A patient with ophthalmic and intracavernous aneurysms, which were clipped successfully. The superior hypophyseal aneurysm shown in I and J was treated by trapping with an ICA to ICA vein graft. ACP, anterior clinoid process.

9

temporary trapping of the ICA may be necessary during the bone resection. On angiographic studies, lateral clinoid aneurysms are located lateral to the anterior bend of the ICA on AP view, but may extend superiorly. On the lateral view, they may be partially hidden, depending on their size. Anterior clinoid aneurysms are rare, only one example being observed by the senior author after this series. These aneurysms may perforate through the superior dural ring to present in the subarachnoid space. In such cases, careful removal of the clinoid process and opening of the superior dural ring are necessary [2]. Paraclinoid (opthalmic segment) aneurysms may be divided into medial, anterior, and posterior para-

clinoid aneurysms (lateral paraclinoid aneurysms have not been observed) based on location of the aneurysm’s neck on the ophthalmic segment of the ICA, or into ophthalmic, superior hypophyseal artery, and posterior paraclinoid aneurysms, based on the arterial bifurcation point at which the aneurysms arise. Posterior paraclinoid aneurysms do not seem to have a demonstrable arterial branch of origin. These are seen to arise from the ICA diametrically opposite to the ACP and proximal to the PCOM on lateral views of the angiogram. They may be mistaken for ICA-PCOM aneurysms, especially if the PCOM is not visualized on the angiogram. The fundus of these aneurysms is usually fused to the superior wall of the CS, and the aneurysm may

484 Surg Neurol 1999;51:477–88

occasionally extend into it. Ophthalmic artery aneurysms arise at the origin of the ophthalmic artery from the ICA, on its anteromedial or medial surface, and project anterosuperiorly. On lateral view angiographic studies, they are seen to arise from the anterior bend of the ICA, just superior to the ACP, and the ophthalmic artery can usually be identified just proximal to the aneurysm (Figure 9, G & H). On lateral view angiograms, the ICA is displaced inferiorly (closing the loop). At surgery, it is usually possible to preserve the ophthalmic artery, although in some patients, the OA may originate from the neck of the aneurysm, and therefore be difficult to preserve. Superior hypophyseal artery (SHA) aneurysms arise from the medial or the posteromedial surface of the ophthalmic artery, and on anterioposterior angiograms can been seen to be located medial to or parallel to the supraclinoid ICA. On lateral view angiograms, the ICA is displaced superiorly (opening the loop). Of course when a supraophthalmic aneurysm reaches giant proportions and incorporates much of the ICA, it is hard to be certain whether it started as a SHA aneurysm or a posterior paraclinoid aneurysm (Figure 9, I & J). The treatment of a giant paraclinoid aneurysm is very similar regardless of how it started. Maximal exposure of the aneurysm’s neck and sac, and of the ICA and CN II allow the surgeon to clip the aneurysm by clip-reconstruction, or to revascularize with aneurysm trapping if clipping is not feasible. No adverse consequences seem to have occurred as a result of the occlusion of a single superior hypophyseal artery, although the potential for loss of vision and hypothalamic-pituitary disturbance exists, especially when a large branch or multiple branches are occluded.

Treatment Of the 35 aneurysms, 31 were clipped satisfactorily as confirmed by intraoperative angiography. One ophthalmic aneurysm was not clipped but was wrapped because of its small size and the absence of a neck. One 3.0-cm superior hypophyseal aneurysm was clipped and a superficial temporal artery (STA)-to-MCA bypass was done prophylactically. On postoperative angiography the aneurysm and ICA were thrombosed. The patient developed a small capsular infarct and a mild hemiparesis, but recovered completely. Another SHA aneurysm was treated by trapping and a cervical ICA-supraclinoid ICA vein graft bypass. This patient developed a small capsular infarct, presumably because of ante-

De Jesus et al

(A–C) This 48-year-old woman presented with abducens palsy from a giant left intracavernous ICA aneurysm. She also had a broad-necked carotid ophthalmic aneurysm that could not be clipped, and two MCA aneurysms at bifurcations of the M2 segment of the MCA. This is shown in the anteroposterior angiograms (A). She was treated by clipping of the MCA aneurysms, cervical ICA to MCA (M2 segment) vein graft, and trapping of the ICA above the level of the ophthalmic aneurysm (B & C).

10

rior choroidal artery thrombosis. She exhibited a mild hemiparesis postoperatively, but recovered to an independent status. A third patient with giant intracavernous, large ophthalmic, and two MCA aneurysms was treated by an ICA-to-MCA vein bypass graft, and trapping of the ICA aneurysms (Figure 10 A–C). She made a normal recovery, despite two postoperative epidural hematomas and brain swelling caused by heparin-induced thrombocytopenia. Four aneurysms in three patients were clipped through the contralateral approach after clipping of the ipsilateral aneurysm (Table 3). These included one patient with an ophthalmic aneurysm, one with an ophthalmic and lateral clinoid aneurysm, and one with an ICA-PCOM aneurysm.

Clinoid and Paraclinoid Aneurysms

Complications The postoperative complications are detailed in Table 2. One patient who had suffered SAH from a posterior paraclinoid aneurysm developed a postoperative epidural hematoma that was removed successfully. She subsequently suffered severe vasospasm unresponsive to hypertensive, hypervolemic therapy and angioplasty, and died of massive brain infarction. Interestingly, she had not undergone an orbital osteomy for aneurysm exposure. Another patient developed recurrent epidural hematomas and postoperative brain swelling secondary to heparin-induced thrombocytopenia; the case is detailed above. Two patients suffered ischemic strokes. One occurred because of the inadvertent occlusion of the ICA during the clipping of a partially thrombosed SHA aneurysm. A prophylactic STA-MCA bypass prevented a major stroke. This was in 1988, before the use of intraoperative angiography. This patient suffered a capsular infarct and hemiparesis, but recovered completely. The ICA occlusion would have been recognized by the use of intraoperative angiography, although whether clip repositioning would have led to the prevention of the stroke is speculative. In another patient, a calcified 2.0-cm SHA aneurysm with no demonstrable neck was trapped, with a vein graft placed from the cervical to the supraclinoid ICA. She suffered an anterior choroidal territory stroke, but recovered to an independent status, with resolution of a mild hemiparesis. An ICA-to-M2 segment vein graft would have definitely reduced the temporary occlusion time of the ICA, and perhaps prevented the cerebral infarction. Three patients suffered postoperative visual loss. In two earlier patients, this was thought to be attributable to inadequate mobilization of the optic nerve, as is currently done. One of these patients suffered blindness, and the other recovered 20/30 visual acuity, with a superior medial field defect. The third patient, with severe preoperative chiasmal and contralateral optic nerve damage (finger counting only in the contralateral eye, and bitemporal hemianopsia) from a 2.0-cm SHA aneurysm, suffered ipsilateral visual loss despite adequate optic nerve mobilization. His ipsilateral vision recovered to 20/25, but his preoperative visual deficits did not improve. In this series, the ophthalmic artery was occluded in three patients during clipping, but none had postoperative visual decline. Six patients exhibited postoperative extraocular muscle deficits, but five recovered completely. The sixth patient, who recovered only partially, had a

Surg Neurol 485 1999;51:477–88

3

Operative Treatment for Clinoid and Paraclinoid Aneurysms

SURGICAL TREATMENT

NUMBER OF CASES

Aneurysm clipped Ipsilateral Contralateral* Aneurysm wrapped Trapping with STA-MCA bypass with ICA-ICA vein graft with ICA-MCA vein graft

28 3 1 1 1 1

*An additional patient had the clipping of an ipsilateral ophthalmic and contralateral ICA-PCOM aneurysm.

very tight brain after the intraoperative rupture of a large paraclinoid aneurysm, during or immediately before craniotomy. She suffered a permanent oculomotor paresis. In addition to this patient, one other suffered intraoperative aneurysmal rupture during dissection, which was easily handled. One patient had a postoperative CSF rhinorrhea through a sphenoidal sinus extension into the ACP. She had significant preoperative communicating hydrocephalus, and her problem was solved by a lumboperitoneal shunt and direct repair of the site leakage.

Outcome Patients’ outcome was assessed at 3 months and 1 year postoperatively, with annual follow-up thereafter, and graded according to the Glasgow Outcomes Scale. Twenty-five patients were in excellent condition (GOS 5, returned to previous occupation), two were in good condition (GOS 4, independent with some disability), and one died postoperatively. When patients with unruptured and

4

Management Complications of Patients with Clinoid and Paraclinoid Aneurysms

COMPLICATION Death (epidural hematoma, and severe vasospasm) Stroke, with good recovery Epidural hematoma, recovered Visual loss complete partial CN III paresis, permanent CSF rhinorrhea* *VP shunt, direct repair.

NUMBER (TOTAL 5 28) PERCENTAGE 1

4%

2 1

8% 4%

1 2 1 14%

4% 8% 4%

486 Surg Neurol 1999;51:477–88

De Jesus et al

ruptured aneurysms are considered separately, of 20 patients with unruptured aneurysms, 18 had an excellent outcome, and 2 had a good outcome; of eight patients with ruptured aneurysms, seven had excellent outcomes, and one died postoperatively.

eral aneurysms [4,5,7,19,21,22,28,30,31]. The difficulties with a prefixed chiasm noted by other authors can be solved by the drilling of the planum sphenoidale, although this increases the risk of CSF leakage. We do not recommend attempting to clip a contralateral aneurysm if the brain is tight, or if the aneurysm is larger than 1.5 cm.

DISCUSSION

INDICATIONS FOR TREATMENT All patients who present with SAH should be treated if the clinoid or paraclinoid aneurysm is the source of hemorrhage. If another aneurysm caused the hemorrhage, and a clinoid or paraclinoid aneurysm is present on the side ipsilateral to the operation for the ruptured aneurysm, concurrent surgical clipping can be performed if conditions are favorable. For unruptured aneurysms, we base our decision to operate on the patient’s physiological age and life expectancy, the size of the aneurysm, and its location. Small clinoid aneurysms (,1 cm) may be observed, unless they project into the subarachnoid space. All other aneurysms are treated if they are $4 mm in size, extend into the subarachnoid space, and if the patient is in good physiological condition and may be reasonably expected to live at least 10 more years.

CLASSIFICATION Multiple systems of classification of these aneurysms have been used by various authors in the literature [5,8,12–14,18,31]. We have used a system that is based on the ICA segment of origin of the aneurysm (clinoid or ophthalmic ICA segment), based on the branch-point of origin of the aneurysm from the ICA, and on the location of the neck in relation to the ICA. When the aneurysm reaches a large or giant proportion, one may be able to only identify the broad segment of origin from the ICA, and it may be difficult to classify it more precisely. Both clinoid and posterior-paraclinoid (ophthalmic segment) aneurysms arise without an apparent arterial branch of origin. Whether they arise at bifurcations of vestigial arteries, or because of stresses at the curvature of the carotid siphon is speculative. SURGICAL TECHNIQUE Some aspects of our technique may be controversial. We recommend the routine use of orbital osteotomy, although in patients with small ophthalmic or superior hypophyseal aneurysms, this may be considered by some surgeons to be unnecessary. In our hands, the addition of the orbital osteotomy adds an hour to the procedure, with no increase of postoperative morbidity in regards to deficits or the length of hospital stay. We do perform a preoperative carotid occlusion test in patients with complex ICA aneurysms to assess the safety of temporary ICA occlusion. However, if the ICA has to be sacrificed at surgery, a bypass procedure is always performed to reduce the risk of a stroke [24]. However, the situation is different if ICA sacrifice is performed by endovascular means in such patients. If the patient has an excellent collateral circulation demonstrated by the absence of neurologic deficit or deficits on SPECT scanning, endovascular proximal occlusion can be safely performed, with a short period of anticoagulation after the occlusion. A number of authors have reported the clipping of contralateral ICA aneurysms, particularly during an operation to clip ipsilateral aneurysms. A contralateral approach has also been used for unilat-

ENDOVASCULAR TECHNIQUE Endovascular techniques may be used to occlude the proximal ICA in case of giant aneurysms, or to obliterate the aneurysm with Gulgielmi detachablecoils (GDC). A number of clinoid and paraclinoid aneurysms have been treated in our institution and others with GDC coils [15,27]. However, the longterm follow-up of aneurysms treated by this technique is not yet available. In a series of 79 patients treated with coiling, with a mean angiographic follow-up of 5– 6 months, incomplete obliteration was observed in 15% of small-necked aneurysms and 84% of large-necked aneurysms [6]. By creating a mass, compression of CNs II, III, and VI may be worsened by GDC coils, and future treatment with surgery would become much more complicated if the aneurysm were to regrow [9]. Because rupture of aneurysms after incomplete endovascular obliteration is well documented, patients treated with GDC coils require regular angiographic follow-up [9,25]. Despite these reservations, endovascular treatment should be considered as an alternative to microsurgical treatment in all patients with clinoid or paraclinoid aneurysms.

We thank Jennifer Pryll (Pogorzala) for the illustrations, and Joseph Reister for preparation of the manuscript.

Clinoid and Paraclinoid Aneurysms

Surg Neurol 487 1999;51:477–88

REFERENCES 1. Adson, AW. Surgical treatment of vascular diseases altering function of eyes. Amer J Ophthal 1942;25: 824 –38. 2. Aldrich F. Anterior (dorsal) paraclinoid aneurysm: case report. Surg Neurol 1991;35:374 – 6. 3. Batjer HH, Kopitnik TA, Giller CA, et al. Surgery for paraclinoid carotid artery aneurysms. J Neurosurg 1994;80:650 – 8. 4. Bergland RM, Ray BS, Torack RM. Anatomical variations in the pituitary gland and adjacent structures in 225 human autopsy cases. J Neurosurg 1968;28:93–9. 5. Day AL. Aneurysms of the ophthalmic segment: a clinical and anatomical analysis. J Neurosurg 1990;72: 677–91. 6. 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;15:815–20. 7. Fries G, Perneczky A, van Lindert E, BahadoriMortasawi F. Contralateral and ipsilateral microsurgical approaches to carotid-ophthalmic aneurysms. Neurosurgery 1997;41:333– 43. 8. Gibo H, Lenkey C, Rhoton AL. Microsurgical anatomy of the supraclinoid portion of the internal carotid artery. J Neurosurg 1981;55:560 –74. 9. Gurian JH, Martin NA, King WA, Duckwiler GR, Guglielmi G, Vinuela F. Neurosurgical management of cerebral aneurysms following unsuccessful or incomplete endovascular embolization. J Neurosur 1995;83: 843– 853. 10. Hassler WE, Meyer B, Rohde V, et al. Pterional approach to the contralateral orbit. Neurosurg 1994;34: 552– 4. 11. Kato Y, Sano H, Hayakawa M, et al. Surgical treatment of internal carotid siphon aneurysms. Neurol Res 1996;18:409 – 415. 12. Kobayashi S, Kyoshima K, Gibo H, et al. Carotid cave aneurysms of the internal carotid artery. J Neurosurg 1989;70:216 –21. 13. Korosue K, Heros RC. “Subclinoid” carotid aneurysm with erosion of the anterior clinoid process and fatal intraoperative rupture. Neurosurgery 1992;31:356–60. 14. Kothandaram P, Dawson BH, Kruyt RC. Carotid ophthalmic aneurysms. A study of 19 patients. J Neurosurg 1971;34:544 – 8. 15. Malisch TW, Guglielmi G, Vinuela F, Duckwiler G, Gobin YP, Martin NA, Frazee JG. Intracranial aneurysms treated with Guglielmi detachable coil: midterm clinical results of a consecutive series of 100 patients. J Neurosurg 1997;87:176 – 83. 16. Milenkovic Z, Gopic H, Antovic P, et al. Contralateral pterional approach to a carotid-ophthalmic aneurysm rupture at surgery. J Neurosurg 1982;57:823–5. 17. Mizoi K, Takahasi A, Yoshimoto T, Fujiwara S, Koshu K. Combined endovasuclar and neurosurgical approach for paraclinoid internal carotid artery aneurysms. Neurosurgery 1993;33:986 –92. 18. Nagasawas, S, Ohta T, Tsuda, E. Surgical results and the related topographic anatomy in paraclinoid internal carotid artery aneurysms. Neurol Res 1996;18: 401– 8. 19. Nakagawa F, Kobayashi S, Takemae T, et al. Aneurysm

20.

21.

22. 23.

24.

25.

26.

27.

28. 29.

30.

31.

protruding from the dorsal wall of the internal carotid. J Neurosurg 1986;65:303– 8. Nakao S, Kikuchi H, Takahashi N. Successful clipping of carotid-ophthalmic aneurysm through a contralateral pterional approach: report of two cases. J Neurosurg 1981;54:532– 6. Nishio S, Matsushima T, Fukui M, et al. Microsurgical anatomy around the origin of the ophthalmic artery with reference to contralateral pterional surgical approach to the carotid-ophthalmic aneurysm. Acta Neurochirurgica 1985;76:82–9. Oshiro EM, Rini DA, Tamargo RJ. Contralateral approaches to bilateral cerebral aneurysms: a microsurgical anatomical study. J Neurosurg 1997;87:163–9. Sachs, E Jr. Arteographic demonstration of collateral circulation through ophthamlmic artery in internal carotid artery thrombosis; report of two cases. J Neurosurg 11 1954:405–9. Sekhar LN, Patel SJ. Permanent occlusion of the internal carotid artery during skull base and vascular surgery: Is it really safe? An editorial Am J Otol 1993; 14:421–2. Strother CM, Lunde S, Graves V, Toutant S, Hieshima GB. Late paraophthalmic aneurysm rupture following endovascular treatment: case report. J Neurosurg 1989;71:777– 80. Tamaki N, Shigekuni K, Kazumasa E, et al. Giant carotid-ophthalmic artery aneruysms: direct clipping utilizing the “trapping-evacuation” technique. J Neurosurg 1991;74:567–72. Thielen KR, Nichols DA, Fulgham JR, Piepgras DG. Endovascular treatment of cerebral aneurysms following incomplete clipping. J Neurosurg 1997;87: 184 –9. Vajda J, Juha´sz J, Pa´sztor E, et al. Contralateral approach to bilateral and ophthalmic aneurysms. Neurosurg 1988;22:662– 8. Walsh FB, King AB. Ocular signs of intracranial saccular aneurysms. Experimental work on collateral circulation through ophthalmic artery. Arch Ophthal 1942;27:1–13. Yamada K, Hayakawa T, Oku Y, et al. Contralateral pterional approach for carotid-ophthalmic aneurysm: usefulness of high resolution metrizamide or blood computer tomographic cisternography. Neurosurg 1984;15:5– 8. Yasargil MG, Gasser JC, Hodosh RM, et al. Carotidophthalmic aneurysm: Direct microsurgical approach. Surg Neurol 1977;8:155– 65.

COMMENTARY

This is a very good article by De Jesu´s et al. The authors retrospectively analyze a series of 28 operated patients with 35 clinoid and paraclinoid aneurysms. Surgery for aneurysms in this location presents special challenges because of the complex anatomy of the region and the need to resect bone with a high-speed drill near the aneurysm sac and the optic nerve. Also, the cervical or petrosal internal carotid artery must be exposed to gain proximal control. Finally, unlike other anterior circulation

488 Surg Neurol 1999;51:477–88

De Jesus et al

aneurysms, there is a greater occurrence of large and giant aneurysms (between 25% and 50% in the literature) in this location. I agree with the authors that it is safe to remove the anterior clinoid process and the optic nerve intradurally and also to have proximal control of the ICA in the neck. I also believe that most paraclinoid and some clinoid aneurysms should be clipped from the contralateral side in case of bilateral aneurysms, and if they are not larger than 1.5 cm in diameter, with a good neck. This contralateral approach is sometimes even easier than the ipsilateral one. Although the results of this study present about 20% morbidity and mortality—similar to that of other surgeons— I doubt that an interventional radiologist could do any better, because 46% of the authors’ series were large and giant aneurysms. Atos Alves de Sousa, M.D. Neurosurgeon Belo Horizonte, Brazil In this paper, De Jesu´s et al review their surgical experience of 35 clinoid or paraclinoid internal carotid artery aneurysms. A number of previous publications have reported the use of skull base surgical techniques, intraoperative angiography, and retrograde suction decompression techniques for these aneurysms. The surgical techniques described here are now popular and their surgical results seem to be similar to those of other institutions. I believe that the retrograde suction decompression technique is now the principle surgical adjunct for successful treatment of large or giant aneurysms in this location. However, the duration of temporary occlusion of the ICA is an important issue with the use of this technique. We routinely perform balloon test occlusion to determine the optimal treatment. If a patient develops neurologic deficits soon after test occlusion (usually within a few minutes), the use of retrograde suction decompression can be very dangerous. In such patients, we prefer to use a combination of high-flow bypass graft and proximal occlusion or trapping of the ICA. Accordingly, in our series of 22 patients with large or giant paraclinoid aneurysms, we have decided preoperatively whether or not to use the revascularization technique based on the results of preoperative hemodynamic studies. Kazuo Mizoi, M.D. Department of Neurosurgery Tohoku University School of Medicine Sendai, Japan

The authors report their experience over approximately 8 years with aneurysms involving the clincoid and paraclinoid segments. The classification scheme offered by the authors is slightly different from previously reported schema, but comparisons between prior reports and this report should be straightforward. I would offer brief comments on several points. The routine use of orbital osteotomy is certainly a controversial addition to the standard pterional approach to this area. Each of the lesions reported by the authors could have been adequately exposed, in my opinion, without the addition of the osteotomy. We have recently begun using this technique in selected cases, and as of this time I remain unconvinced that it offers substantial advantage. The routine use of bypass grafting whenever the carotid artery is sacrificed is also somewhat controversial. Based on a few anecdotal problems, however, I am becoming much more liberal in the use of bypass procedures in this setting. The use of trial balloon occlusion is extremely helpful in that it gives the surgeons a more quantitative measure of the volume of blood needed for revascularization. For example, a patient who clinically tolerates internal carotid occlusion but develops minor blood flow disturbances is probably adequately treated with a superficial temporal graft. On the other hand, a patient who becomes immediately hemiplegic might well be expected to require additional blood volume, mandating a long vein graft. I would mention a minor technical point: I strongly favor the use of the external carotid artery as the donor for the saphenous vein graft, as opposed to the ICA. When using the external carotid, the middle cerebral ischemia time is limited to the minutes required for the distal anastomosis. I commend the authors on a modification of the suction decompression technique, in that they report the use of the superior thyroid artery that is cannulated retrograde. I look forward to trying this procedure in my own practice. This is a comprehensive and informative discussion of this important topic.

H. Hunt Batjer, M.D. Division of Neurological Surgery Northwestern University Medical School Chicago, Illinois