Clinoidal segment carotid artery aneurysms are surgically challenging lesions. The aneurysm neck originates proximal to the distal dural ring, and the aneurysms typically are larger. Therefore, endovascular techniques are often considered to be the primary treatment option. Treatment techniques and results for 40 clinoidal segment carotid artery aneurysms that were considered unsuitable for contemporary endovascular intervention are analyzed in this report.
Forty aneurysms in 33 female and 3 male patients were treated surgically. Fifteen patients had bilateral aneurysms; of these patients, four underwent bilateral craniotomies. Twenty-seven aneurysms were 10 to 14 mm in size, eight were 15 to 24 mm, and five were more than 25 mm. The most common presentation was visual loss, which occurred in 13 patients. Seven patients presented with subarachnoid hemorrhage.
Thirty-seven aneurysms were directly repaired with clipping, two were trapped with bypass, and one was trapped without bypass. The complication rate was 10%, with one major stroke, two minor strokes, and one successfully treated brain abscess.
Surgical treatment of clinoidal segment carotid artery aneurysms can produce acceptable outcomes. Specific preoperative and intraoperative techniques facilitate improved surgical results for aneurysms that are not treatable with contemporary endovascular techniques.
Aneurysms of the internal carotid artery (ICA) that arise in the cavernous sinus region have historically been difficult to treat. In the past 15 years, advances in microneurosurgical techniques and anatomic knowledge have facilitated a direct surgical approach to ICA aneurysms in the cavernous sinus and paraclinoid region 2–4,8–10,13,14,16,18,27,28,30,41,45). Currently, a multidisciplinary approach that combines endovascular and direct surgical techniques is often used in the management of these lesions (3,37).
Aneurysms of the ICA that originate from the clinoidal segment (the C3 segment in the Fischer classification or the C5 segment in the Bouthillier classification) may expand through the distal dural ring to cause either optic nerve compression or subarachnoid hemorrhage (1,2,9,10,33,34). The extradural origin of the proximal neck, with extension of the dome into the subarachnoid space, requires precise surgical exposure, with extradural or intradural removal of the anterior clinoid process, opening of the distal and sometimes proximal dural rings, identification of the often-atypical origin of the ophthalmic artery, and frequently treatment of a broad thin aneurysm neck (3,29,42,43,55,56). The outcomes reported for surgical series of ICA aneurysms in this region confirm that these are surgically treacherous lesions (2,10,14,25,45).
The optimal treatment modality for patients with clinoidal segment aneurysms remains unclear. Because of the surgical challenges associated with these lesions, at our institution patients with clinoidal segment aneurysms frequently undergo an initial evaluation by the interventional neuroradiologist, for consideration for endovascular treatment (24,35). If it is estimated that interventional techniques can be used with a risk of less than 10%, then serious consideration is given to using these techniques. However, clinoidal segment aneurysms are often poorly suited for endovascular treatment. Forty clinoidal segment aneurysms for which endovascular treatment was unsuitable are analyzed here. The contraindications to endovascular treatment were as follows: neck width of more than 5 mm, incorporation of the posterior communicating artery into the aneurysm neck, giant or partially thrombosed aneurysm, or failure of a trial balloon occlusion test as a prelude to proximal carotid artery occlusion (19). On the basis of this cumulative experience, a planned surgical approach has been developed to treat complicated clinoidal segment aneurysms that have been excluded from endovascular treatment.
PATIENTS AND METHODS
The terminology for the classification of aneurysms of the carotid artery in the cavernous sinus and paraclinoid region has historically been confusing, in part because of conflicting views regarding the anatomic boundaries of the cavernous sinus. Readers are referred to the recent publication by Kim et al. (32), which provides anatomic dissections and a detailed review of this topic. The aneurysms chosen for this analysis were those in which the aneurysm neck originated within the clinoidal segment but the dome extended intradurally. Determination of the origin was based on intraoperative evaluation, after dissection of the aneurysm neck, and not exclusively on the preoperative angiograms. Although the anatomic features become obscured with large or giant aneurysms, the distal dural ring can be identified at its lateral and medial attachments and traced over the aneurysm to confirm the location of the aneurysm neck proximal to the distal dural ring and hence from the clinoidal segment of the ICA. This analysis excludes rare transitional aneurysms in which the neck originates within the cavernous sinus but the dome projects intradurally (36). The clinoidal segment is the extracavernous extradural segment of the carotid artery between the proximal and distal dural rings; it was termed the C3 segment by Fischer (20). A more recent classification by Bouthillier et al. (7) termed this the C5 segment. Not included in this study were paraclinoidal and ophthalmic segment aneurysms, for which the distal dural ring was opened to gain proximal control but not to expose the aneurysm neck.
Forty aneurysms were treated in 33 women and 3 men, with an average age of 55 years. Fifteen patients had bilateral aneurysms; of these patients, four underwent bilateral craniotomies. Twenty-seven aneurysms were 10 to 14 mm, eight were 15 to 24 mm, and five were more than 25 mm in size. The most common presentation was visual loss resulting from mass effect, which occurred in 13 patients. Seven patients presented with subarachnoid hemorrhage. For the remainder of the patients, the aneurysms were discovered during evaluations of other complaints (primarily headaches). Clinoidal segment aneurysms less than 1 cm in size are usually asymptomatic and, because they remain extradural, there is an extremely low risk of subarachnoid hemorrhage. This is why no aneurysms less than 10 mm in size were surgically treated.
During the preoperative evaluation, computed tomography without contrast agent was performed to identify possible aneurysm neck calcification. Neck calcification significantly decreases the likelihood of successful direct aneurysm clipping; surgeons would thus need to be prepared to perform a bypass procedure during surgery.
Angiographic studies were performed in the interventional neuroradiology suite. With the use of neuroleptic analgesia, arterial sheaths were placed in both femoral arteries. A complete cerebral angiogram was obtained, and additional oblique views of the aneurysm to be treated were obtained to adequately define the anatomic features. In some cases, the aneurysm was selectively catheterized with a microcatheter and an “aneurysmogram” was obtained to better define the anatomic features of the aneurysm and the incorporation of vessels into the aneurysm neck. If it seemed that the aneurysm would not be amenable to selective coil occlusion, trial balloon occlusion and cerebral blood flow (CBF) studies were performed according to the following protocol. A baseline activated clotting time is determined, and the patient receives 6000 to 8000 units of intravenously administered heparin. Serial activated clotting times are obtained throughout the procedure, and additional heparin is administered to maintain the activated clotting time at approximately 2.5 times the baseline value. A 5.0-French, double-lumen, balloon catheter is placed in the cervical segment of the ICA and slowly inflated until the artery is occluded. Occlusion is maintained for 30 minutes, during which time the patient is closely monitored for any change in neurological function. During the period of trial occlusion, the patient's blood pressure is lowered pharmacologically to decrease the mean arterial pressure 10 to 20 mm Hg, compared with baseline values. The collateral circulation through the circle of Willis to the vascular territory to be occluded is assessed during the period of balloon occlusion by obtaining digital subtraction angiograms of the contralateral carotid artery, the ipsilateral external carotid artery, and the dominant vertebral artery. Pre- and post-balloon occlusion radioactive xenon CBF studies are also performed. In addition, 99Tc-ethylene cysteine diethlyester is injected intravenously during the period of trial occlusion, and cerebral-perfusion single-photon emission computed tomographic scans are obtained after the patient leaves the angiography suite. Development of a neurological deficit, a CBF decrease of approximately 30%, or a CBF measurement of less than 25 ml/100 g/min is considered to indicate a failed trial balloon occlusion test.
If the patient did not tolerate the test balloon occlusion, then the ipsilateral leg was prepared in case the aneurysm could not be repaired directly, necessitating a bypass procedure. The cranium was fixed in a Mayfield radiolucent pinion, to facilitate intraoperative angiography. All of these aneurysms were approached through a low frontotemporal craniotomy, in which the frontal osteotomy was located just above the orbital roof and extended to the inner canthus of the ipsilateral eye. After exposure of the cervical carotid artery, the medial one-third of the sphenoid wing, including the anterior clinoid process, was removed extradurally (39). In only a few cases was the final portion of the clinoid process removed intradurally. After dural opening and division of the sylvian fissure, the aneurysm was approached subfrontally. If there was concern regarding parent carotid artery patency after repair of the aneurysm, then either intraoperative angiography via direct common carotid artery puncture or quantitative measurement of CBF through the middle cerebral artery (MCA), using an electromagnetic flowmeter, was performed. The surgical approach is illustrated in the following case examples.
This 67-year-old woman presented with subarachnoid hemorrhage (Hunt and Hess Grade I). A preoperative angiogram demonstrated an oblong clinoidal segment aneurysm (Fig. 1). The aneurysm neck was wide, measuring 5 mm. In a trial balloon occlusion test, 133Xe-CBF measurements decreased from 46 ml/100 g/min to 22 ml/100 g/min. During dissection of the aneurysm proximal to the distal dural ring, the cervical ICA was occluded for 20 minutes. Cerebral protection was provided with thiopental (3 mg/kg, administered intravenously), which induced hypertension to 150 mm Hg. Postoperative angiograms confirmed repair of the aneurysm and preservation of the parent carotid artery (Fig. 2). The origin of the ophthalmic artery in the cavernous sinus can be observed. This case illustrates the value of knowing the potential for collateral blood flow if proximal vessel occlusion is required during aneurysm repair. For occlusion of the cervical carotid artery, an atraumatic clip should be used on the ICA. There is thus continuous flow from the common carotid artery into the external carotid artery, preventing the formation of thrombus at the carotid bifurcation that could possibly embolize.
This 56-year-old woman presented with loss of vision in the right eye and retro-orbital pain. A preoperative angiogram demonstrated that the aneurysm neck was broad and fusiform (Fig. 3). Immediate left-sided paresis was observed during a trial balloon occlusion test of the cervical right ICA. The aneurysm was explored through a right frontotemporal craniotomy, with extradural removal of the medial half of the sphenoid wing and clinoid process (Fig. 4). There was some calcification in the aneurysm neck. Multiple clips and positions were used in attempts to obliterate the aneurysm. However, intraoperative measurements using micro-Doppler ultrasonography revealed that blood flow through the M1 segment decreased from a baseline value of 74 ml/min to 10 ml/min with apparently optimal clip placement, indicating that even in these circumstances there was compromise of the parent carotid artery because of the aneurysm neck calcification. A long vein graft from the cervical external carotid artery to the MCA was performed. CBF was preserved during construction of the bypass, because there was no temporary occlusion of either the cervical or intracranial ICA. This is important, given the results of the preoperative trial balloon occlusion test. After bypass construction, the aneurysm was trapped by occluding the cervical ICA above the bifurcation and intracranially proximal to the posterior communicating artery. This case illustrates the value of knowing the potential for collateral blood flow, identifying aneurysm neck calcification, and preoperatively planning for the high probability of an extracranial-intracranial bypass.
This 67-year-old woman presented with subarachnoid hemorrhage (Hunt and Hess Grade I). The aneurysm originated from the clinoidal segment (Fig. 5). The neck was 6 mm in length and seemed to represent a blowout of the carotid artery. The aneurysm neck was quite thin and started to hemorrhage during the initial dissection (Fig. 6). The cervical carotid artery was therefore occluded. The aneurysm was then repaired with an encircling clip graft and a tandem cutout angled clip. After flow was restored, the intracranial ICA seemed to be narrowed. Intraoperative angiograms confirmed that there was severe stenosis of the carotid artery and that the anterior cerebral artery did not opacify because of low flow through the ICA. The cervical ICA was occluded, and the aneurysm clips were readjusted (Fig. 7). Subsequent intraoperative angiography confirmed better flow through the carotid artery underneath the aneurysm clips, with perfusion of the anterior cerebral artery. Postoperative angiograms demonstrated approximately 50% stenosis of the carotid artery (Fig. 8). The cavernous sinus origin of the ophthalmic artery can be observed. The patient made an uneventful neurological recovery. This case illustrates the advantages of exposing the cervical carotid artery to obtain proximal vascular control and to perform intraoperative angiography for assessment of aneurysm clip placement.
This 35-year-old woman presented with subarachnoid hemorrhage (Hunt and Hess Grade II). On neurological examination, she was noted to have a severe junctional scotoma. Cerebral angiography demonstrated giant left clinoidal segment ICA, anterior communicating artery, and right MCA aneurysms (Fig. 9). The patient underwent surgical repair of the left ICA and anterior communicating artery aneurysms, followed by endovascular treatment of the right MCA aneurysm (Fig. 10). During surgery, it was apparent that the source of hemorrhage was the left ICA aneurysm. Exposure and repair of the left ICA aneurysm was facilitated by temporary occlusion of the cervical ICA, followed by decompression of the aneurysm dome and clipping. During these maneuvers, the patient's blood pressure was maintained at approximately 150 mm Hg, and propofol was used as a cerebral protective agent. Deep hypothermia was not required. Although the patient recovered well from the multiple interventions, she exhibited only modest visual improvement.
Thirty-seven patients experienced successful aneurysm repair via direct clipping. In two cases, a bypass was performed, using a saphenous vein graft from the cervical external carotid artery to an M2 branch of the MCA, followed by aneurysm trapping. In one case, the aneurysm was trapped and no bypass was performed.
There were four complications, for an overall surgical morbidity of 10%. One patient, with a 25-mm aneurysm, experienced a major stroke secondary to ICA thrombosis. Two patients, with aneurysms of 20 and 15 mm, experienced minor strokes (Rankin Scale score of 1). One of these patients exhibited right-hand monoparesis, whereas the other patient experienced subtle word-finding difficulties. One patient developed a brain abscess, which was successfully treated with surgical evacuation and antibiotic administration, without sequelae. This patient was intraoperatively noted to have a pneumatized clinoid process, which was the likely source of infection. Of the 13 patients with preoperative visual losses, follow-up postoperative visual testing demonstrated that vision was improved for 2 patients, vision was worsened for 1 patient, and vision was functionally unchanged for the remaining 10 patients.
Role of endovascular therapy
Endovascular techniques have matured to the point that they can now be considered legitimate primary treatments for transitional and clinoidal segment aneurysms. Table 1 summarizes the published endovascular experience with aneurysms of the cavernous and clinoidal ICA. In many of these reports, it is impossible to determine whether the treated aneurysms were completely intracavernous or clinoidal and whether there was any intradural extension. Initial experience with the use of detachable balloons to occlude the ICA in patients with cavernous carotid artery aneurysms was favorable (6,11,21,26). Higashida et al. (26) reported complete thrombosis and complete or partial symptom resolution for 68 patients with intracavernous aneurysms that were treated via detachable balloon occlusion of the carotid artery proximal to the aneurysm. For patients who experience trial balloon occlusion test failure, extracranial-intracranial bypass has been successfully used before balloon occlusion (22).
More recently, electrolytically detachable coils have been used to selectively occlude the aneurysm while preserving the parent carotid artery (5,24,38). As reported by Halbach et al. (24), among a series of 35 patients with cavernous region aneurysms that were treated with selective endovascular coil occlusion, 58% of cranial nerve palsies were completely resolved and 38% exhibited significant improvement, whereas only one patient experienced worsening of symptoms. Seventy-four percent of patients exhibited more than 90% angiographic occlusion of their aneurysms, and all patients exhibited more than 75% angiographic occlusion. It is important to recognize that the long-term prognosis and significance of subtotal aneurysm obliteration remain to be determined. However, among the seven patients who initially presented with hemorrhage, there was no incidence of rebleeding (mean follow-up period, 29 mo) (24).
Surgical treatment of cavernous sinus and clinoidal segment aneurysms by experienced surgeons may be associated with greater morbidity than is endovascular intervention 15–17,37,44,47). Dolenc (17) advocated consideration of endovascular treatment for patients with true intracavernous aneurysms for whom the aneurysm can be completely excluded from the circulation with preservation of the parent carotid artery. However, if the carotid artery cannot be preserved, he favors attempted surgical repair (18), as do other surgeons. We routinely consider endovascular therapy for our patients with cavernous and clinoidal segment aneurysms as an initial treatment option; a paradigm for therapeutic decision-making is outlined in Figure 11. It is important to note that, although the procedural risk associated with endovascular treatment of clinoidal segment aneurysms may be less than that associated with direct surgical repair, no randomized trial comparing endovascular treatment and surgical treatment of cerebral aneurysms exists, and the long-term efficacy of endovascular treatment has not been established. Clinoidal segment aneurysms are often poorly suited for contemporary endovascular therapy. During the same time period as that for this surgical series, 25 patients with clinoidal segment and paraclinoidal aneurysms were treated via an endovascular approach, with a major complication rate of 4%. In the surgical cases reported here, endovascular treatment was excluded because of a wide aneurysm neck, incorporation of the origin of the posterior communicating artery, a giant or partially thrombosed aneurysm, or failure of trial balloon occlusion of the ICA as a prelude to proximal artery occlusion. In our estimation, the same factors that precluded safe endovascular treatment also made a direct surgical approach significantly more challenging for this group of patients. With recent advances in balloon remodeling, stent-assisted coiling techniques, and three-dimensional coils, more broad-necked aneurysms will probably be successfully treated via endovascular techniques in the future (35,40).
The rationale and methodology for the surgical approach used here warrant review, as follows. 1) Preoperative computed tomography without contrast agent administration. Plain computed tomographic scans are quite valuable for determination of whether the aneurysm neck is partially calcified. The presence of calcification significantly decreases the probability of aneurysm obliteration by direct clip placement. Preoperative consideration must then be given to alternative approaches for treatment of the aneurysm if direct clipping does prove technically impossible. 2) Detailed cerebral angiograms, including superselective intra-aneurysmal catheterization (“aneurysmogram”). In our experience, some aneurysms that were considered to be fusiform on the basis of routine angiograms were found to have surgically repairable necks in detailed multiplanar angiograms subsequently obtained by an interventional neuroradiologist. Review of the anatomic features of the aneurysm, as detailed in the angiograms, with the neuroradiologist who performed the study is very valuable for surgical planning. 3) Trial balloon occlusion with CBF studies. During angiography, trial balloon occlusion of the ICA, combined with some type of CBF assessment, should be performed. Although there is a small risk of cerebral ischemic complications associated with the trial balloon occlusion procedure (<0.5% at our institution), we have found that determination of the collateral blood flow before surgery is pro-foundly advantageous. For patients with insufficient collateral flow, a bypass procedure is indicated if the aneurysm cannot be directly excluded from the circulation and trapping is required (22,48,52,53). Preoperative preparation is imperative. This may involve sterile preparation of a leg for saphenous vein collection, preservation of the ipsilateral superficial temporal artery when the scalp incision is made, or exposure of the petrous portion of the ipsilateral ICA. Saphenous vein grafts from the cervical carotid artery to the distal ICA or from the petrous carotid artery to the distal vessel have been used in this setting, with good results (1,51). Surgeons must also consider the hemispheric tolerance of ipsilateral carotid artery occlusion when temporarily occluding the carotid artery during surgery. An alternative to preoperative trial balloon occlusion is the use of intraoperative electroencephalographic monitoring. It is important to recognize that some surgeons routinely use deep hypothermia and circulatory bypass to provide both proximal control and cerebral protection (50). In this series, we did not find it necessary to use profound hypothermia techniques. 4) Use of a radiolucent head-holder and exposure of the cervical carotid artery. It is important to have the option of performing intraoperative angiography when the surgeon is concerned regarding the patency and caliber of the lumen of the ICA. Therefore, in all cases the head should be fixed in a radiolucent head-holder. It is important to position the patient high on the operating room table, to the extent that the shoulders are hanging over the edge. This facilitates adequate positioning of the image intensifier. Furthermore, it is advantageous to limit the use of self-retaining retractors, for several reasons. First, almost all of these operations can be performed with a single retractor blade placed subfrontal. It is rare that a second temporal retractor is necessary. Less brain retraction increases the likelihood of good neurological results. Second, unless they are removed, additional retractor bars obscure good observation in intraoperative angiograms. Exposure of the cervical carotid artery has several advantages. First, it offers a measure of reassurance (proximal control) when the medial sphenoid wing is removed and the aneurysm is dissected. Second, it provides an easy route for performance of intraoperative angiography through a direct carotid artery puncture. Third, it increases the options in cases in which an extracranial-intracranial bypass procedure proves necessary. 5) Extradural removal of the medial sphenoid wing, including the anterior clinoid process. Extradural removal of the sphenoid wing, as first developed by Dolenc (16), is a safe way to remove the anterior clinoid process and expose the clinoidal segment of the carotid artery. There are many variations on the basic technique for this maneuver (39). For this group of patients, removal was commenced by drilling off the outer sphenoid wing with a diamond burr and punching through into the orbit. The medial sphenoid wing was then removed with a small bone punch, as opposed to a drill. Use of the bone rongeurs decreases the time required to remove the anterior clinoid process. Although many surgeons advocate intradural removal of the anterior clinoid process to avoid premature rupture of the aneurysm dome (10), in this series the clinoid process was removed extradurally without complications. It is important to expose the medial aspect of the superior orbital fissure. It is also necessary to remove the optic strut medial to the planum sphenoidale. Removal of the medial component of the optic strut provides more maneuvering room during clip placement across aneurysm necks that originate from the medial or rostral aspect of the carotid artery. When the anterior clinoid process is being removed, it is important to check for possible pneumatization, as evidenced in this series by one patient with postoperative rhinorrhea. 6) Availability of intraoperative CBF quantification. Intraoperative use of the electromagnetic flowmeter is valuable in establishing sufficient flow through the parent carotid artery. In this series, the flow probe was usually placed on the M1 segment of the MCA, because of the ease of maneuverability. Typical blood flow through the MCA at this level measured approximately 60 to 70 ml/min. When diminished flow was recorded, intraoperative angiography was performed for qualitative assessment of vessel patency. In some cases, the angiographic findings led to clip repositioning.
In our series, there was one major stroke, two minor strokes, and one brain abscess, yielding an overall morbidity rate of 10%. The brain abscess was likely the result of a pneumatized clinoid process but responded well to surgical drainage and antibiotic therapy, without neurological sequelae. The outcomes reported here are similar to the complication rate of 13%, including one death, reported by Al-Rodhan et al. (2) for their series of 23 surgically treated transitional aneurysms. Our results also closely mirror those of Day (10) for his series of 54 patients with ophthalmic segment aneurysms, in which 13% of patients experienced poor outcomes or death.
Diaz et al. (14) treated 32 patients with cavernous region aneurysms. Sixteen aneurysms were clipped directly, whereas the other 16 were trapped and bypassed. Although the authors did not differentiate outcomes with respect to location, they observed two deaths, two minor strokes, and two cases of worsened vision, yielding an overall complication rate of 19%. In the series of para- and infraclinoidal aneurysms reported by Perneczky et al. (45), 22 patients were treated via direct aneurysm occlusion, with either clipping or suture ligation. Those authors reported only deaths; two deaths resulted from neurological causes related to surgery and two deaths resulted from medical complications, yielding an overall mortality rate of 18%.
Thirteen patients in this study had aneurysms greater than 15 mm in size. Kattner et al. (30) reported a 20% morbidity rate for their operative series of 29 patients with large or giant paraclinoidal aneurysms. Those findings are comparable to those for the series reported by Heros et al. (25), in which 34 patients with large or giant paraclinoidal aneurysms were treated by clipping, carotid artery ligation (with or without bypass), or trapping, with a combined morbidity rate of 26% (including visual deterioration). Those two series, as well as this report, support the feasibility of a direct surgical approach to larger aneurysms in the para- and infraclinoid regions of the carotid artery, albeit with an associated increase in morbidity rates, compared with those for smaller aneurysms.
Forty clinoidal segment carotid artery aneurysms that had been excluded from contemporary endovascular treatment for anatomic reasons were treated surgically. The complication rate was 10%. On the basis of this experience, the following recommendations are made. First, the preoperative evaluation should include computed tomographic scans without contrast agent administration, trial balloon occlusion of the ICA with CBF studies, and detailed multiplanar angiography with selective intra-aneurysmal injections. Second, during the operation, consideration should be given to exposure of the cervical carotid artery, extradural removal of the anterior clinoid process, the potential need for intraoperative angiography, and the ability to quantify post-repair CBF.
The authors report a series of “clinoidal segment” carotid artery aneurysms that were surgically treated using standard techniques. These lesions were all either large or giant aneurysms, using traditional size parameters. I challenge a number of tenets of this report, as follows. 1) The definition of the clinoidal segment. The authors refer to previous anatomic studies to clarify the anatomic characteristics of the clinoidal segment. With 20 years of neurosurgical experience, I am frankly unable to determine, on the basis of preoperative imaging studies, whether aneurysms are clinoidal or “paraclinoidal.” I think all of us have fairly reasonable criteria for establishing whether an aneurysm is cavernous. However, lesions in this particular area (which is defined on the basis of a number of variable soft-tissue criteria) are difficult to assess using currently available preoperative diagnostic techniques. Additionally, the bony and soft-tissue dissection necessary to clarify the origin of clinoidal aneurysms converts them into paraclinoidal lesions. I specifically challenge the angiograms represented by Figures 1, 5, and 9, with respect to whether those are actually typical paraclinoidal carotid ophthalmic or superior hypophyseal lesions. 2) Procedural risk. The authors mention that the treatment risks of endovascular techniques “may be” less than those of surgical techniques. This hypothesis is not proven. In fact, numerous publications, including those cited by the authors, have demonstrated exquisitely low treatment risks for even highly complex, large lesions in this area. 3) Durability. Although the treatment risk referred to by the authors for endovascular techniques is quite low, the durability of this form of therapy (particularly for large or giant lesions, as described in this article) is at best uncertain. In my opinion, patients with these lesions are best treated by highly skilled and experienced teams of extravascular and endovascular therapists, in concert with highly experienced diagnostic neuroradiologists using state-of-the-art neuroimaging.
H. Hunt Batjer
The authors present a well-studied large series of patients with aneurysms that originated from the clinoidal segment of the internal carotid artery (ICA). It should be noted that the authors specifically excluded purely intracavernous aneurysms and ophthalmic and other paraclinoidal aneurysms in which the neck was distal to the external ring. Their excellent results (10% morbidity rate) are particularly commendable when it is considered that this is a preselected series of difficult aneurysms that were excluded from endovascular surgery because of factors such as a broad neck or giant size.
The authors discuss in great detail the very careful protocol they follow in the evaluation of these patients. We use a very similar protocol, with only minor modifications. We perform an almost identical preoperative trial balloon occlusion test; the only difference is that we cannot conduct direct blood flow studies in the angiographic suite and must rely on postangiographic single-photon emission computed tomographic findings. We perform “aneurysmography” with the microcatheter only for patients we are considering for endovascular treatment. We have only recently begun to appreciate the importance of examining preoperative computed tomographic scans to check for calcification of the neck. Even more recently, we have begun to use computed tomographic angiography to study the relationship of the neck to the clinoid process and surrounding bony structures. In terms of surgical technique, we used to remove the anterior clinoid process extradurally, as taught to us by Dolenc. In the past several years, we have preferred to complete the removal of the anterior clinoid process intradurally in patients with aneurysms, for fear of rupturing the aneurysm without intradural control when the process is removed extradurally; however, we realize that this fear is not well founded, because I personally have not experienced such an event, nor have these authors, nor has such premature rupture been reported to occur with any frequency in the literature. In the past few years, we have performed an orbitozygomatic osteotomy for more complicated aneurysms; lately, however, we have limited the procedure to removal of the rim and roof of the orbit and the orbitozygomatic process, without full removal of the zygomatic arch. Like the authors, we perform a preliminary bypass graft only for patients who fail the balloon occlusion test, when there is no better option than to sacrifice the carotid artery. In those cases, we always prefer trapping, rather than proximal occlusion (as the authors apparently do), to minimize the risk of embolic complications.
One notable advance in endovascular therapy that may change our future approach to many of the more-difficult aneurysms is the development of flexible intracranial stents. Our endovascular colleagues have begun to use these stents, and they were recently able to place a stent and occlude the aneurysm through the stent, with preservation of the carotid artery, in a giant aneurysm case in which calcification of the neck had prevented me from satisfactorily clipping the aneurysm.
Roberto C. Heros
In this article, Meyer et al. report techniques and results of surgical treatment for patients with aneurysms that were not considered for endovascular treatment and originated in the clinoidal segment but exhibited a dome extending into the intradural subarachnoid space. This article provides a good treatment paradigm reflected by the experience of the authors.
Since the original angiographic description of the clinoidal segment of the ICA by Fischer (1), aneurysms of that region have had many semantic labels, including paraclinoidal aneurysms, ophthalmic artery aneurysms, carotid cave aneurysms, and transitional aneurysms. Regardless of nomenclature, the technical approach to these aneurysms is based on the “battle of the clinoid.” Removal of the anterior clinoid process is critical and is well stressed in this article. However, we strongly advocate intradural drilling of the clinoid process; although this series included no disastrous consequences with extradural drilling, there have been reports of such disastrous effects. We also advocate exposure of the ICA in the neck and think that it provides a certain level of comfort and facilitates intraoperative angiography, proximal control, and possible selection of the donor site for a bypass graft.
In our experience, calcification is a serious obstacle to successful clinical clipping but not an absolute contraindication for an attempt. We strongly agree with the use of intraoperative angiography as an adjunct to treatment of these lesions. It provides a level of comfort for proximal control in the unlikely event of a catastrophic rupture. It makes planning much easier and provides immediate proof of patency of the carotid artery and obliteration of the aneurysm. Overall, this article represents an important contribution to the vascular literature and presents an excellent treatment paradigm for these technically difficult aneurysms.
Meyer et al. reported on a significant series of 40 large clinoidal segment aneurysms that were surgically treated. The complication rate of 10% for this difficult group of lesions is a testament to the current sophistication of microsurgical techniques and to the authors’ skill and experience in performing these operations. The discussion of the preoperative preparations and the intraoperative considerations is a very useful summary of the current surgical approach to paraclinoidal aneurysms.
Trial occlusion of the ICA before surgery to treat highly complex, giant paraclinoidal aneurysms provides valuable information. Intraoperative occlusion of the ICA for prolonged periods is often required, for reconstruction of the carotid artery and elimination of giant aneurysms from the circulation. If the patient cannot tolerate a short period of trial carotid artery occlusion, then temporary carotid artery occlusion during surgery is probably not a wise strategy. In these cases, either extracranial-intracranial bypass with proximal occlusion or deep hypothermic circulatory arrest should be considered as alternative treatments. However, for most paraclinoidal aneurysms, preoperative trial balloon occlusion is probably not necessary. Small aneurysms require only brief periods of temporary carotid artery occlusion for clipping of the aneurysm. Less than 3 minutes of carotid artery occlusion can almost always be tolerated with hypertension, hypothermia, and barbiturate cerebroprotection. Problems arise when these aneurysms cannot be clipped at the neck because of calcifications or other complicating factors that require parent artery sacrifice. In such cases, I prefer to perform an extracranial-intracranial bypass before trial occlusion. Twenty-four to 48 hours after a surgical bypass has been performed, the ICA can be endovascularly occluded, with complete anticoagulation, in conscious patients (1). With this strategy, most preoperative trial balloon occlusions can be avoided.
I do not agree with the contention of the authors that “endovascular techniques have matured to the point that they can now be considered legitimate primary treatments for transitional and clinoidal segment aneurysms.” Endovascular techniques are certainly excellent alternatives when direct clipping cannot be performed. The endovascular approach is the preferred method for performing proximal ICA occlusion, because this technique can be used for conscious patients with full anticoagulation. Coil embolization may be slightly safer than surgical clipping for this subcategory of patients, but the long-term efficacy has never been demonstrated. I think that coiling should be reserved for cases in which clipping is not practical or presents unnecessary risks.
Robert A. Solomon
New York, New York