Surgery for superior hypophyseal artery aneurysms: A new classification and surgical considerations
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/neuroindia.NI_229_17
Source of Support: None, Conflict of Interest: None
Background: Superior hypophyseal artery (SHA) aneurysms form a unique subgroup of paraclinoid aneurysms having a propensity to grow to a large size in the suprasellar region resulting in compression of the optic nerve, chiasma, and/or tract.
Keywords: Aneurysms, internal carotid artery, paraclinoid, subarachnoid hemorrhage, superior hypophyseal artery, surgery
The superior hypophyseal artery (SHA) forms the largest branching artery arising from the medial wall of the paraclinoid segment of the internal carotid artery (ICA; encompassing the entire length of the ICA, distal to the proximal dural ring, and including the clinoidal and supraclinoidal segment of ICA, up to the origin of the posterior communicating artery). Paraclinoid aneurysms, arising from either the clinoidal or supraclinoidal segment of the internal carotid artery, constitute 5–14% of all intracranial aneurysms., SHA aneurysms form a unique subgroup of paraclinoid aneurysms that are rare and have features that sets them apart from all other intracranial aneurysms. Apart from their characteristic presentation of subarachnoid hemorrhage (SAH), the medial direction of the fundus of SHAs and their propensity to grow to a large size in the suprasellar region below the optic nerve, chiasma, and tract, may often lead to their detection prior to their rupture, by causing mass effect on the optic apparatus and compromising perforator vessels to the optic pathway, hypothalamus, and pituitary gland. The tortuosity of the ICA in this region, the difficulty in effectively cannulating the SHA, and the failure to relieve mass effect of the aneurysm on the surrounding structures often precludes an effective endovascular obliteration of these aneurysms. Even when successfully coiled, their large size often causes coil compaction leading to a residual or recurrent aneurysm. The complexity of the surgical anatomy of this region; the proximity of the SHA to the complex neurovascular structures in the suprasellar area; and the need to often perform intricate procedures of an anterior clinoidectomy, optic nerve decompression, and mobilization, and opening of the carotid collar to gain proximal control of the ICA and to define the proximal aspect of the aneurysmal neck also adds to the degree of difficulty during surgical approach to these aneurysms. Surgery is, however, effective in, nearly almost always, excluding the entire aneurysm from circulation with a low recurrence rate. The SHA is not a single vessel but constitutes an arcade of 4–5 medially directed blood vessels arising from the ICA, and SHA aneurysms may occur from any of the segments of ICA. Due to their rarity and the failure to address the varying nature of the complex anatomy encountered in different locations on the medial surface of the ICA from which these aneurysms may arise, the surgical approach to SHA aneurysms has not been systematically addressed. In this study, a new classification of SHA is proposed that helps in identifying the surgical issues encountered in dealing with SHA aneurysms arising from different segments of the medial surface of ICA. The intricacies in the surgical approach that help in the successful exclusion of these aneurysms from the circulation and in significantly reducing the operative morbidity and mortality are also discussed.
From January 2009 to June 2016, 14 patients (mean age 49.43 ± 11.28 years; 4 male patients and 10 female patients) with an SHA aneurysm who underwent surgery were included in our study. Data regarding their demographic profile, mode of presentation [either as SAH, causing mass effect (assessed as visual acuity compromise or field defect), and/or presenting with altered sensorium/unilateral limb weakness], modified Hunt and Hess grade at presentation, Fisher grade on computed tomographic (CT) scan study, and the presence of hydrocephalus/vasospasm/infarct, were assessed [Table 1]. The patients also underwent a computed tomographic angiography (CTA)/digital subtraction angiography (DSA) to assess the size of the aneurysm; its site of origin and relationship to the optic pathway and other neurovascular structures in the vicinity; the dominance of the circulation; the cross flow across the anterior communicating artery from the opposite internal carotid artery (ICA) or across the posterior communicating artery from the vertebro-basilar system; the side of ICA affected; the relationship of proximal neck of the aneurysm with the anterior clinoid process, the distal dural ring, the sella and the dorsum sellae/posterior clinoid process; and, multiplicity of aneurysms [Table 2].
Classification of superior hypophyseal artery aneurysms
The classification of these SHAs was done based on the 4 parameters that have an important bearing on surgery: a. size: (small: <1 cm, large: 1–2.5 cm and giant: >2.5 cm); b. origin of the neck of the SHA aneurysm and direction of its fundus; c. relationship of the aneurysm to important structures in its vicinity such as the anterior clinoid process/distal dural ring, the sella, and the dorsum sellae/posterior clinoid process; and, d. whether the aneurysms were saccular or fusiform [Table 3]; [Figure 1]. Thus, 7 categories of SHAs were identified in our study: a. antero-supero-medial: This aneurysm arose from the proximal supraclinoidal ICA, just distal to the origin of the ophthalmic artery, with its fundus pointing anteriorly, located directly below the ipsilateral optic nerve and chiasma and elevating them [Figure 2]a and [Figure 2]b; b. Antero-infero-medial: This aneurysm also arose from either the clinoidal or proximal supraclinoidal ICA, just distal to the origin of the ophthalmic artery, with its fundus directed antero-infero-medially towards the tuberculum sellae. The fundus of the aneurysm was located below the optic nerve. No elevation of optic nerve was seen as the aneurysm was directed antero-infero-medially [Figure 3]a and [Figure 3]b; c. Supero-medial: The neck of the aneurysm arose from the medial aspect of supraclinoid ICA with the fundus of the aneurysm directed towards the suprasellar region and located directly below the optic nerve and chiasma, elevating them. In the case of large aneurysms, the proximal part of the neck also occasionally arose from the clinoidal segment of the ICA [Figure 4]a and [Figure 4]b; d. Infero-medial: The location of the neck of this aneurysm was similar to that seen in the supero-medial aneurysms. However, its fundus lay much below the optic nerve and chiasm, growing into the sella and compressing the pituitary stalk and gland. There was no optic pathway compression [Figure 5]a,[Figure 5]b,[Figure 5]c; e. Postero-medial: The location of the neck of this aneurysm was similar to that seen in the supero-medial aneurysms. However, its fundus was directed towards the dorsum sellae below the chiasm or proximal optic tract, deeply situated in the vicinity of membrane of Liliequest, diaphragma sellae, and pituitary stalk. The fundus of the aneurysm may often be in close proximity to the A1 segment of anterior cerebral artery or its medial lenticulostriate perforators, and on occasions, may be closely adherent to the dorsum sellae [Figure 6]a and [Figure 6]b; f. Fusiform: This was a wide-necked aneurysm growing on both sides of ICA but predominantly on the medial side [Figure 7]a,[Figure 7]b,[Figure 7]c; and, g. Giant: More than 2.5-cm sized with a wide neck, directed medially, encompassing the medial surface of the clinoidal and/or supraclinoid ICA [Figure 8]a,[Figure 8]b,[Figure 8]c; [Table 3].
Patients with paraclinoid segment aneurysms, not predominantly arising from the medial aspect of the clinoidal-supraclinoidal ICA and not categorized as SHA aneurysms; transitional aneurysms that extended from the cavernous to supraclinoidal segment of the ICA; and, ICA aneurysms that underwent coiling were excluded from the study. In our department, surgery is not performed for any patient in Hunt and Hess grade 5 (decerebrating or with no motor response), with or without doll's eye movements, as well as with hemodynamic and respiratory instability.
After a plain and contrast CT scan, performed immediately on arrival of the patient in the emergency room, had ascertained the presence of either SAH or the presence of a large/giant aneurysm, a CTA with three-dimensional reconstructed images was also performed. The aneurysmal characteristics as well as the site of origin of its neck on the ICA was determined, and the dominance of ICA and its relationship to the side of the ICA harboring the aneurysm were noted. In case the aneurysmal characteristics were not obvious on CTA, or the dominance of the ICA and its cross flow could not be unequivocally established on CTA, a DSA with cross flow and balloon test occlusion was also performed. A well-established cross flow was determined when there was rapid filling of distal branches of both ICA and MCA branches from the contralateral ICA and/or the vertebrobasilar system after balloon test occlusion of the ICA on the side harboring the aneurysm. If the flow to both sides right up to the distal branches of the anterior and middle cerebral arteries occurred spontaneously on an unilateral ICA angiogram, it established the adequate patency of the anterior communicating artery and further balloon test occlusion of the ipsilateral ICA was not done. The clinical response of the patient and his/her ability to tolerate ipsilateral balloon test occlusion for 20 min was also assessed in cases where flow across the anterior communicating artery was present; however, the adequacy of flow and a rapid filling of the contralateral distal circulation could not be unequivocally established. A balloon test occlusion of the ipsilateral ICA was not done if there was hypoplasia of the A1 segment of anterior cerebral artery (ACA) on either side.
In case an adequate control of the supraclinoid ICA was not available proximal to the SHA aneurysm, neck control of the ICA was undertaken prior to the craniotomy. Patients underwent a standard pterional craniotomy on the side of the aneurysm with extensive drilling of the sphenoid ridge. A wide dissection of the Sylvain fissure was performed and both the transylvian and subfrontal corridors were utilized to expose the ipsilateral ICA harboring the SHA aneurysm and its ACA and MCA branches. If the proximal neck of the aneurysm was obscured by the anterior clinoid process, an intradural anterior clinoidal drilling was performed. This consisted of exposing the planum sphenoidale and anterior clinoidal process; the reflection of the dura overlying the optic foramen and anterior clinoid process, thus protecting the falciform ligament overlying the optic nerve; deroofing the optic foramen to expose the proximal 8 mm to 1 cm of the optic nerve covered by its fascial sheath (this manoeuvre also disconnected the anterior root of the anterior clinoid process); coring out of the anterior clinoid process to disconnect it from its posterior root (the optic strut); dissecting off the mobilized anterior clinoid process from the carotid collar to expose the distal dural ring and the carotid collar wrapped around the clinoidal segment of the ICA that was seen curving laterally to enter the cavernous sinus; dissecting off and removing the distal dural ring and the carotid collar from the clinoidal ICA to expose the ICA proximal to the neck of the SHA aneurysm; defining the proximal and distal aspects of the neck of the aneurysm (the optic foramen deroofing facilitated mobilization of the optic nerve medially to expose the proximal part of fundus of the aneurysm);clipping of the aneurysmal neck after ensuring that there were no major neurovascular structures in close vicinity, especially the optic nerve, chiasma and tract, ophthalmic artery, the hypothalamic and medial lenticulostriate perforators, and the pituitary stalk; and, in the case of giant aneurysms, opening of the aneurysm to remove the thrombosis/blood to relieve the mass effect. In the cases presenting with an acute bleeding and presence of hydrocephalus evident of the preoperative CT scan, the lamina terminalis was also opened to drain third ventricular cerebrospinal fluid. In case acute SAH was associated with cerebral edema and brain swelling, a duroplasty was also carried out.
The points noted during surgery included whether or not the anterior clinoidal process was drilled and the distal dural ring was opened; the size and location of the neck as well as the size and direction of the fundus of the aneurysm based upon the current classification; intraoperative rupture of the aneurysm; temporary clipping of the ICA; and whether or not fenestrated or multiple clips were applied to the neck of the aneurysm [Table 4].
During surgery, although proximal (either in the neck or of the clinoidal/supracliniodal ICA proximal to the aneurysm) and distal control of ICA was taken in every case, temporary clipping of the ICA prior to the actual clipping of the aneurysm was only undertaken when there was an intraoperative rupture of the aneurysm or to render a turgid large-to-giant aneurysm relatively soft prior to its clipping. In the case of giant aneurysms, whenever non-fenestrated clips were used, a two-clip technique was utilized. The first clip was applied slightly distal to the neck of the aneurysm, and subsequently, a more proximal clip was applied to the neck of the aneurysm parallel to the previously applied clip. This ensured adequacy of neck closure and the ready availability of proximal portion of the neck of the aneurysm in case the fundus ruptured during the placement of the initial clip. Following surgery, an adequate fluid resuscitation and hypertensive therapy was administered in patients with clinical vasospasm (neurological deterioration in the presence of radiological vasospasm but without any evidence of cerebral infarction, edema, or hydrocephalus).
Outcome and follow-up
The Modified Rankin Scale (MRS) score was assigned to patients at admission and the outcome of patients was assessed utilizing the same score at discharge and at least 8 months. The outcome was divided into a favorable (mRS 0–2) and an unfavorable one (mRS 3–6) based upon the MRS scores.
The 14 patients with SHA aneurysms presented either with SAH (n = 11; 78.57%) or mass effect due to the large sized aneurysm (n = 3; 21.42%). Visual acuity and field deficits were present in 2 and absent in 7 patients. In 5 patients, visual assessment could not be done due to their altered sensorium. The assessment of the modified Hunt and Hess grade revealed that 3 patients did not have a SAH but presented with a large/giant unruptured aneurysm that was identifiable on the preoperative CT/MRI scan (grade 0); 3 each, respectively, presented with mild (grade 1) and moderate-to-severe headache (grade 2), respectively; 2 were in a state of confusion (grade 3); and, 3 were in a state of stupor (grade 4) [Table 1]. In the 11 patients who presented with SAH, the Fisher grade was 2 (less than 1 mm thick blood in the subarachnoid spaces) in 6 patients, the grade was 3 (greater than 1 mm thick blood in the subarachnoid spaces) in 2 patients, and the grade was 4 (intracerebral/intraventricular hemorrhage) in 3 patients, respectively. Associated hydrocephalus was present in 2 patients that, however, resolved over time and did not require a cerebrospinal fluid diversion in any of them. The SHA aneurysm arose from the right ICA in 6 and from the left in 8 patients. The size of the aneurysm was less than 1 cm in 5, 1–2.4 cm in 7, and greater than 2.5 cm in 2 patients. Multiple aneurysms were seen in 3 patients with the additional aneurysm/s being an anterior communicating artery aneurysm; posterior communicating and middle cerebral artery aneurysms; and a carotico-opthalmic aneurysm of the contralateral side, in one patient each, respectively. In the former 2 patients, the concurrent aneurysms could be approached from the ipsilateral side during the primary surgery for the ruptured SHA aneurysm; the latter patient harboring the additional contralateral carotico-opthalmic aneurysm underwent a subsequent second stage surgery to clip the unruptured aneurysm on the side contralateral to the SHA aneurysm. Among the 11 patients, who also underwent a preoperative DSA and in whom the cross flow and balloon occlusion test of the ipsilateral ICA harboring the SHA aneurysm was performed, an adequate cross flow was present in 4 and was absent in 7 patients. The status of cross flow could not be determined in 3 patients in whom the surgery was directly performed after the CTA study as ICAs on both the sides were found to be co-dominant with bilaterally patent A1 segments of the anterior cerebral artery [Table 2].
The classification of SHA aneurysms, their frequency of distribution according to the present classification, and the challenges faced during their surgical clipping are summarized in [Table 3] and [Table 4], and their outcome is given in [Table 5]. There were 2 small antero-supero-medial directed SHA aneurysms that required anterior clinoidal drilling, opening of the distal dura ring and carotid collar, and mobilization of the ipsilateral optic nerve. An intraoperative rupture in one of them required temporary clipping for less than 3 min. Both of them could be successfully clipped with a single clip, thus relieving pressure on the ipsilateral optic nerve that was stretched over the aneurysm. All were admitted with a favorable mRS and all had a favorable mRS at follow-up.
The single antero-infero-medially directed, small SHA aneurysm was deeply embedded in the anterior wall of the sella inferior to the tuberculum sellae and was covered by the optic nerve. Despite optic nerve mobilization, it could not be clearly visualized as it was arising from the medial aspect of the genu of the ICA, below the origin of the ipsilateral ophthalmic artery, just distal to the lateral turn of the ICA towards the cavernous sinus. Thus, it was wrapped with muscle, bemsheet, and fibrin glue. The aneurysm, despite having presented with a SAH, was too small to be successfully embolized. The patient, who presented with an mRS 2 was discharged with mRS 1, and after 14 months of follow-up, had no complaints with no rebleed.
Of the 3 patients with supero-medially directed SHA aneurysm, 2 small aneurysms required only anterior clinoid drilling for gaining proximal control of the ICA; whereas, the third aneurysm that was large in size with its proximal neck reaching proximal to the distal dural ring also required opening of the carotid collar. There was intraoperative rupture of the aneurysm in two of them with one requiring a short (less than 3 min) temporary clipping time. A fenestrated clip (with the ICA trunk in the fenestration), and two straight nonfenestrated clips, were required in 1 patient each, respectively; whereas in the third patient, the aneurysm could be clipped with a single nonfenestrated clip. Two patients were admitted with a favorable mRS and were discharged with a favorable mRS whereas 1 was admitted with an unfavorable mRS who died. This patient constituted the only perioperative mortality in the series (1/14; 7.14%). He presented with modified Hunt and Hess grade 4 with hydrocephalus and intraventricular hemorrhage that required an external ventricular drainage, which was continued in the postoperative period until the resolution of the hemorrhage. He developed severe postoperative meningitis.
All 3 of the infero-medially directed aneurysms with the fundus reaching the sellar region were large. They all required anterior clinoidal drilling, opening of the carotid collar, and mobilization of the optic nerve for visualization of the neck of the infero-medially directed aneurysm in the carotico-optic space. Two of them could be successfully clipped with a straight clip; whereas the third patient required a fenestrated (taking the parent ICA in fenestration) clip. Two were admitted with an unfavorable mRS and 1 with a favorable mRS. All 3 had an unfavorable mRS at the follow-up visit.
There were two large postero-medially directed aneurysms in whom clinoidal drilling was not required. They were successfully clipped with one requiring a straight and the other a fenestrated clip, respectively. All were admitted with a favorable mRS and all had a favorable mRS at follow-up.
The single large fusiform aneurysm with its predominant component arising from the medial wall of the ICA required vessel reconstruction using 3 clips; one right-angled fenestrated clip parallel to the ICA trunk and two straight fenestrated clips addressing the residual fundus of the aneurysm (taking both the ICA and the right angled fenestrated clip placed parallel to its trunk, in its fenestration). This patient had a favorable mRS at a 6-month follow-up visit.
One of the two giant SHA aneurysms required a fenestrated clip whereas the other was secured using a two-clip technique. No temporary clips on the ICA were applied in both these patients. Following their successful clipping, clot was evacuated from the lumen of these two giant aneurysms. One each was admitted with a favorable and an unfavorable mRS, respectively, which continued in the period of follow-up.
Thus, at admission 10 (71.42%) patients presented in a good grade mRS score and 4 (28.57%) in a poor grade mRS score. At follow-up ranging from 8–84 months (median: 17.5 ± 26.78 months, 9 (64.28%) patients presented in a good grade mRS score, and 5 (35.71%) in a poor grade mRS score. No overt endocrinal deficits were noted in the postoperative period in these patients.
Need for a classification of superior hypophyseal artery aneurysms
The SHA, in fact, does not represent one vessel but includes several branches forming an arterial arcade, varying in number from 1 to 5. SHA aneurysms may arise from the origin of any of these branches along the entire length of the medial wall of the clinoidal and supraclinoidal segment of ICA up to the posterior communicating artery. The variability in the sites of origin and size of the neck and fundus of these aneurysms is responsible for the difference in the operative difficulty and varied risk factors encountered at different segments of ICA.
Day  divided the SHA aneurysms into a. the paraclinoid variant that includes aneurysms burrowed ventrally into the clinoid process (Nutik's ventral paraclinoid aneurysms), or burrowed inferomedial to the ICA (Kobayashi's carotid cave aneurysms); and  b. the suprasellar variant that includes aneurysms located below the optic chiasma. Although this classification is useful, it still does not specifically focus on several variants of the SHA aneurysms arising from the medial wall of the ICA distal to the ophthalmic artery and proximal to the posterior communicating artery that have significant differences in the nuances required to successfully clip them and in the surgical difficulties encountered. These variants of the SHA aneurysms include the aneurysms burrowed into the anterior wall of the sella below the dorsum sellae; those that point supero-medially lifting up the optic nerve, those that are located deep within the sella in close proximity to the pituitary salk, and those that point dorso-medially toward the traversing A1 segment of the anterior communicating artery and in proximity to the medial lenticulostriate and hypothalamic perforators.
Considering the length of the medial segment of the ICA involved, the number of branching vessels that may be implicated, the varying frames of neurovascular structures encountered around the SHA aneurysm based on the site of the origin of its neck (in the proximal, middle, or distal segments of the medial surface of ICA); the direction of the aneurysmal fundus (whether pointing superomedially or inferomedially, anteriorly, or posteriorly); and the consequent differences in the surgical nuances involved, there was an urgent need to formulate a cogent classification that encompasses all the subtle variations of SHA aneurysms and helps in deciding an effective and a safe operative plan.
Anatomical and surgical considerations in each subgroup
In the supero-medial group (antero-supero-medial, supero-medial and postero-supero-medial), the aneurysmal fundus was found directly below the ipsilateral optic nerve, chiasm, or the tract, thus elevating it. In addition, in the antero-supero-medial SHA aneurysms, the neck and fundus were in close proximity to the laterally curving ICA just distal to the ophthalmic artery and the distal dural ring; and in the postero-supero-medial SHA aneurysm, the fundus often lay in close proximity to the A1 segment of ACA, recurrent artery of Heubner running parallel to the A1 segment, and the ICA bifurcation. The optic pathway, hypothalamic, and medial lenticulostriate perforators also required a careful separation from its neck and fundus. Thus, these blood vessels needed to be carefully dissected and preserved. Occasionally, this aneurysm may be closely adherent to the dorsum sellae, and therefore, the underlying bone may prevent the proper application of the clip at its neck to adequately exclude the aneurysm. The antero-infero-medial SHA aneurysm in our patient was hidden from the view as one approached it utilizing the ipsilateral pterional craniotomy as it was embedded in the anterior wall of the sella below the tuberculum sellae. The genu of ICA and the overlying optic nerve also prevented its visualization. This aneurysm, therefore, had to be wrapped rather than clipped. Perhaps a contralateral fronto-temporal approach would have led to a better visualization of this aneurysm. The infero-medial aneurysm was in close vicinity to the sellar structures such as the pituitary stalk and gland. The ICA trunk prevented an adequate visualization of its neck. Thus, a right-angled fenestrated clip was applied taking the ICA trunk in the fenestration. A similar problem was encountered in the case of fusiform aneurysm predominantly arising from the medial wall of the SHA where, in addition to a right-angled fenestrated clip, two further straight fenestrated clips had to be applied perpendicular to the blades of the right angled one, taking both the ICA trunk as well as the blades of the right angled fenestrated clips in its fenestration to completely obliterate the fundus of the aneurysm. Giant SHA aneurysms may arise from any segment of the medial wall of the ICA and cause optic pathway compression. Anterior clinoidal drilling and optic nerve mobilization by drilling the roof of the optic canal helps in mobilizing the optic nerve medially as well as in the visualization of the neck of these aneurysms. A particular precaution that was especially taken in the case of these medially directed giant SHA aneurysms was not to mobilize the fundus of the aneurysm after successful securing of its neck and evacuation of the thrombosed blood clot within. This should be the usual practice to prevent an inadvertent perforator vessel damage. A deviated pituitary stalk may also need to be carefully separated whenever this type of aneurysm is encountered. We found the two-clip technique to be particularly helpful in the successful clipping of these giant aneurysms. The first clip is placed a little distally on the aneurysm neck; the second clip is then placed more proximally. Thus, in case the aneurysm ruptures during the clipping, the proximal part of the aneurysm neck is available to apply a second clip proximal to the first one. We have always preferred an intradural anterior clinoidal drilling in these cases to directly visualize both the aneurysm and the distal ICA. However, an extradural drilling may also be feasible in case the proximal part of aneurysm is not in close proximity to the clinoidal segment of ICA. The optic nerve decompression has always preceded the anterior clinoidal drilling in our cases. The optic nerve decompression was required to mobilize the optic nerve medially, to gain control of the ICA proximal to the aneurysm or to access the proximal part of neck of the aneurysm, and to create space for the application of the proximal blade of the clip in case the neck of the aneurysm was arising close to the clinoidal segment of ICA. During the intradural clinoidal drilling, the optic nerve decompression divided the anterior root of the clinoid process; and, the coring of the anterior clinoid process further detached the optic strut/posterior root of the anterior clinoid process, following which the mobile anterior clinoid process could be dissected off the carotid collar (the fascial layer between the proximal and distal dural rings, covering the clinoidal segment of ICA).
Why is superior hypophyseal artery sparing not attempted during aneurysmal clipping?
Why did we not attempt to dissect and spare the SHA during aneurysmal clipping, as is the general practice for other aneurysms (despite the fact that the SHA supplies the pituitary stalk, optic nerve, and chiasm and that sacrifice of the SHA may lead to visual impairment or pituitary hypofunctioning)? Horiuchi et al., in an elegant study among 190 paraclinoid aneurysms stated that with unilateral SHA sacrifice, no significant difference was observed between patients in groups with or without visual impairment. However, the simultaneous clipping of bilateral SHA aneurysms was a major factor associated with postoperative visual deterioration. The multiple branches and the arterial arcade that represents the SHA often forms a collateral circulation, some of the branches of which may continue to supply the optic apparatus despite clipping of the SHA branch harboring the aneurysm. Bilateral SHA may also supply the optic nerve and chiasma. This may be an important factor in preventing significant visual impairment despite compromise of the branch at the origin of which the SHA aneurysm is situated., It is also possible that the patient has visual impairment due to the unilateral aneurysmal compression but due to the preserved vision on the opposite site, the binocularity of vision prevented him/her from being aware of the visual deficits; however, when surgery was performed on the contralateral side also, which also added to the compromise of vision, the patient became acutely aware of his/her deficits due to bilateral visual acuity/field or binocularity compromise.
Why not endovascular management?
SHA aneurysms represent sidewall aneurysms projecting medially. This prevents the direct jet flow of blood from entering the aneurysmal sac. This facilitates adequate stasis and thrombosis of the aneurysm following coil embolization. The relatively minimally invasive nature of the endovascular procedure and the technically challenging aspects involved in performing surgery in this region (anterior clinoidal drilling, opening the dural ring, preserving the patency of the small perforating branches supplying the optic chiasm and the pituitary gland, especially in the case of larger aneurysms, and immense variability in size and site of aneurysms in this location) have prompted the utilization of endovascular coil embolization or stent-assisted techniques in addressing these aneurysms, especially because higher complication rates after surgery have been reported. However, the SHA aneurysms are unique in that they often present as large-to-giant sized aneurysms with a wide neck and often cause mass effect on the optic pathway. Coil embolization does not relieve the mass effect on the optic pathway; coils may cause a retrograde parent vessel luminal obstruction; and, in large-to-giant aneurysms, many studies have reported a high coil compaction rate leading to regrowth/recurrence of the aneurysm., Many complex SHA aneurysms have a wide neck and, thereby, require stent-assisted techniques. The rationale for the latter therapy is that, besides preventing coil herniation into the parent vessel, stent deployment diverts blood flow away from the lumen and enhances thrombosis in the aneurysm. However, in a prominent study with stent-associated embolization of SHA, primary aneurysm occlusion was achieved in 68.3% of patients with 7% having a significant residual neck (2 mm) or incomplete treatment. The endovascular clinical trial JR NET reported that the rate of complete aneurysm occlusion was only 57.7% and that of residual aneurysm was 10%. Wang et al., reported a complete occlusion rate of only 43.7%, the rate of residual aneurysm being 23.2%, and the recurrence rate being 12.5%. Thus, surgery, despite its complexities, is often preferable to ensure a near-total aneurysmal occlusion and to relieve the mass effect of large-to-giant SHA aneurysms on the optic pathway. Surgical clipping may also be required to further exclude the residual/recurrent SHA aneurysms, after a failed embolization procedure, from the circulation.
Limitations and strengths
The retrospective nature of the analysis, the small cohort of patients recruited, too few cases included in each of the subgroups for any meaningful statistical analysis to be carried out, and the relatively short follow-up are some of the limitations of the study. The persons evaluating outcome were primarily involved in the care of the patient. Despite the fact that during surgery, the clipped aneurysms were punctured to ensure the adequacy of clipping, the follow-up angiograms to assess for the presence of residual/recurrent aneurysms were not carried out in every patient. The inclusion of representative cases in each category despite the relative rarity of these aneurysms, the detailed observations related to the distinct surgical issues that arise in each subgroup of the classification, the surgical nuances highlighted that may have a distinct role to play in improving the outcome, and the fact that focused surgical studies on SHA have been rarely reported are, however, the relative strengths of the study.
A new classification of the SHA aneurysms is proposed that takes into account the intricate surgical issues involved in their management. A detailed knowledge regarding their anatomical variations and surgical considerations helps in achieving gratifying surgical results.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]