Abstract OBJECTIVE: To explore the surgical access and surgical techniques for saddle-nodal meningioma in order to improve the results of surgical treatment. Methods: We retrospectively analyzed 47 cases of saddle-nodal meningioma treated surgically from January 1985 to December 2004 in our department, and classified them into large, medium and small types according to CT and MR manifestations, and used four different surgical approaches to compare the total resection rate, surgical outcomes and complications. Results: Among 47 tumors, 44 cases (93.6%) were completely resected and 3 cases were subtotal resected. The total resection rate of tumors was not related to tumor size and surgical access (P>0.05). There were 43 cases of good recovery after surgery, with an excellent rate of 91.5%; 2 cases of moderate disability; 1 case of severe disability; and 1 case of death, with a mortality rate of 2.1%. The surgical outcome was not related to tumor size and surgical access (P>0.05). Conclusion: Most saddle-nodal meningiomas can be completely and safely resected. The supraorbital keyhole approach is recommended for small and medium-sized tumors, the unilateral inferior frontal or lateral fissure approach is recommended for medium and large tumors, and the orbitozygomatic approach is recommended for large tumors or those invading the tumor optic nerve canal. Regardless of the approach used, the technique of tumor removal is the same and is the key to total tumor removal.
The incidence of saddle-nodal meningiomas is not too high, accounting for approximately 5-10% of intracranial meningiomas. These tumors originate in the saddle-nodal region, cross the sulcus, and grow in suboptic cross areas, making surgery difficult. From January 1985 to December 2004, the authors used different surgical approaches to remove 47 cases of saddle-nodal meningiomas with relatively satisfactory results, which are reported below.
Clinical data and methods.
1. General data: 18 cases in men and 29 cases in women, age 37-68 years, average 43.4 years. The shortest duration of the disease was 2 months, the longest was 3 years, and the average was 1 year and 1 month.
Clinical manifestations: Visual disturbance was the most common first symptom, with 39 cases, followed by 7 cases of headache and 1 case of incidental finding. Every patient had visual loss in clinical manifestations, including 26 cases of unilateral visual loss, 21 cases of bilateral visual loss, 2 cases of blindness and 5 cases of light perception only. Visual field defects were seen in 32 cases, 22 of which were unilateral temporal blindness, 6 cases were quadrant blindness, and only 4 cases were typical bilateral temporal hemianopia. 7 cases could not examine the visual field, and 8 cases had normal visual field. Headache was uncommon and was seen in only 10 patients. 27 cases had hormone measurements, 6 cases had hypopituitarism, 3 cases had mild hyperprolactinemia, and the rest had normal endocrine function. Funduscopic examination revealed different degrees of optic nerve papillary atrophy on one or both sides in 37 cases.
Imaging: All patients underwent cranial CT examination, and 38 patients after 1991 underwent MR examination. All tumors were located in the dura of the saddle nodes and septum, with forward extension to the pterygoid plateau or lateral extension to the K-artery pool. The tumors were classified according to the maximum diameter of the tumor measured by CT and MR. Tumors <2 cm were classified as small, with 3 cases; 2-4 cm were classified as medium-sized, with 28 cases; and >4 cm were classified as large, with 16 cases. Among the large and medium-sized tumors, 23 patients underwent selective bilateral carotid angiography, which showed that the A1 and A2 segments of the anterior cerebral artery were elevated and posteriorly displaced, and some of them became thin and stiff, suggesting that they were encapsulated by the tumor.
4. Surgical access: All patients in this group were removed by four different surgical accesses under general anesthesia, using microsurgical techniques (Table 1). The unilateral inferior frontal approach was used in 13 cases, among which 10 cases were medium-sized tumors and 3 cases were large tumors. The supraorbital keyhole approach was used in 13 cases, of which 3 cases were small and 10 cases were medium-sized. The lateral fissure approach via the pterygoid point was used in 18 cases, of which 10 cases were large tumors and 8 cases were medium-sized tumors. The lateral orbitozygomatic fissure approach was used in 3 cases, all of which were large tumors.
5. Statistical methods: Statistical analysis software SPSS 11.0 was applied to analyze the relationship between tumor size and surgical access and total tumor resection and surgical outcome by using X2 analysis method.
Results
1.Tumor resection degree: Among 47 cases, 44 cases (93.6%) were completely resected and 3 cases were subtotal resection (Figure 1 and 2). The dural attachments of the saddle nodes were not resected and treated with bipolar electrocoagulation. Table 1 shows the relationship between the surgical approach, tumor size and the extent of tumor resection. Complete resection was achieved in 14 of 16 cases (87.5%) for large tumors, 27 of 28 cases (96.4%) for medium-sized tumors, and 3 of 3 cases for small tumors, and the statistical results showed that tumor size was not related to the degree of tumor resection (P = 0.174). The unilateral inferior frontal approach resulted in complete resection in 11 out of 13 cases (84.6%); the pterygoid approach in 17 out of 18 cases (94.4%); and the supraorbital keyhole and orbitozygomatic approaches in 16 cases, and the statistical results showed that the surgical approach was not related to the degree of resection (P = 0.342).
2. Surgical results: 43 cases recovered well after surgery, with an excellent rate of 91.5%; 2 cases with moderate disability; 1 case with severe disability; 1 case with death, with a mortality rate of 2.1%. Table 2 shows the relationship between surgical access, tumor size and surgical outcomes. Good recovery after surgery was observed in 14 cases (87.5%) of large tumors, 26 cases (92.9%) of medium-sized tumors, and 3 cases (100%) of small tumors, and the statistical results showed that tumor size and surgical outcome were not related (P = 0.236). The unilateral inferior frontal approach, pterygoid approach, supraorbital keyhole and orbitozygomatic approach resulted in good recovery in 11 (84.6%), 13 (100%), 16 (88.9%) and 3 (100%) cases, respectively, and the statistical results showed that the surgical approach was not related to the surgical outcome (P = 1.000). Of the 47 cases with preoperative visual loss, 34 had varying degrees of visual acuity improvement, 9 had no change in visual acuity, and 4 had deterioration in visual acuity after surgery, including 1 case of postoperative blindness in a photoreceptor. 1 case of a huge recurrent saddle node meningioma was resected by a pterygoid approach and died 3 days after surgery due to hypothalamic injury. 35 of the 46 survivors were followed up for 1-14.5 years (mean 7.3 Among them, 31 cases resumed normal work, 3 cases could only do housework and 1 case could not live independently. Three cases of tumor recurrence were found, and one of them was resected by surgery again.
3. Surgical complications: 1 case of a senior patient operated by unilateral inferior frontal approach, who was not awake after surgery, and was found to have cerebral contusion and intracerebral hematoma at the base of frontal lobe by CT, and was operated again for decompression 6 hours after surgery, but failed to recover completely, and the patient was bedridden for a long time. One case of newly developed psychiatric symptoms and two cases of transient enuresis occurred after surgery, both of which recovered completely within 3 months. Postoperative CT review revealed small infarcts in the head of the caudate nucleus in 2 cases, both of which were asymptomatic.
DISCUSSION
The saddle nodes are slight bony elevations that separate the anterior parietal and crossed anterior sulci of the pituitary fossa. Meningioma growth in the saddle node is constrained by the adjacent anatomical structures [1]: the internal K artery laterally; the optic nerve anteriorly; the pituitary stalk, funiculus, and Liliequist’s membrane posteriorly; and the optic cross, end plate, and anterior communicating artery superiorly. This anatomical setting allows the tumor to expand onto the pterygoid plateau and the optic nerve either laterally or bilaterally, as evidenced by preoperative MRI and intraoperative observations. As the optic nerve is fixed to the optic foramen, it becomes angulated and further compressed at the site as the tumor continues to grow. The optic nerve may also be asymmetrically encircled by the tumor. the internal K artery is displaced laterally, but not to an extent beyond the optic nerve. The tumor itself can squeeze in between the optic nerve and the internal K artery and can encircle the artery, but the vessel remains covered by the displaced arachnoid layer. The anterior cerebral artery, located dorsal to the optic cross, is often straightened by the tumor and can be deeper than 3.5 cm into the tumor. Posteriorly the tumor can extend beyond the pterygoid saddle, into the interpeduncular pool, and displace the pituitary stalk, but the tumor does not adhere to the dura of the superior slope or posterior bed process.
The average age at the time of discovery of saddle node meningioma is 30-39 years old, and it is more common in females, with a slightly higher average age in this group. It is detected earlier because it can cause early visual impairment, and most patients have tumors smaller than 4 cm [2]. As it extends from the saddle node area to the superior border of the posterior bed protuberance, the average length is 8 mm (5-13 mm) and the width is 11 mm (6-15 mm). This explains that a tumor as long as it is larger than 1.5 cm can cause clinical symptoms [3]. Smaller tumors can also cause vision loss if they originate medial to the optic foramen. Visual impairment in saddle-nodal meningiomas is different from pituitary tumors, and bilateral temporal hemianopia is rare.
Saddle-nodule meningiomas can be resected via several approaches, such as bilateral frontal, unifrontal, supraorbital keyhole, pterygoid point, and orbitozygomatic approaches. The bilateral inferior frontal approach has many disadvantages and is not recommended in microsurgical conditions, and we have abandoned it. The pterygopoint approach is preferable for saddle-nodal meningiomas and is in accordance with Grisoli’s opinion [4,5]. The pterygopoint approach allows for minimal cerebral traction by removing the pterygoid crest and widely separating the lateral fissure to reach the base of the tumor in the suprasellar region through the space it forms and under the frontal lobe, thus avoiding the obstruction of the internal carotid artery and the optic nerve.
This approach has the following advantages: (1) the release of cerebrospinal fluid from the lateral fissure can fully reduce the intracranial pressure and reveal the tumor without removing the frontal pole and excessively stretching the brain tissue; (2) the posterior pole of the tumor and the optic nerve, internal carotid artery and anterior cerebral artery can be revealed at an early stage, so that the tumor can be separated from the optic nerve and cerebral vessels at an early stage; (3) the frontal vein can be avoided to be compressed and the frontal lobe swollen when the frontal lobe is lifted; (4) at least one side of the olfactory bundle can be preserved to avoid complete olfaction after surgery. (4) at least one side of the olfactory bundle can be preserved to avoid complete loss of smell after surgery; (5) the frontal sinus is preserved intact. The only disadvantage is that the ipsilateral optic nerve and suboptic cross tumor cannot be seen clearly under direct vision, so resection may be difficult.
The unilateral inferior frontal approach is also often used and has the following features: ① Through the lateral fissure pool and carotid pool, cerebrospinal fluid can be released, which can also fully reduce intracranial pressure. (ii) Better visualization, which facilitates the treatment of the tumor base from the saddle node medially. ③It is relatively easy to deal with the tumor remaining medially in the ipsilateral optic nerve. The disadvantage is that it is not easy to separate the adhesions between the tumor and the anterior communicating artery. For smaller saddle node meningiomas, usually less than 3 cm, the supraorbital keyhole can be used to achieve good results. This approach has the following characteristics: (1) the angle of the approach and the extent of exposure are similar to those of the inferior frontal approach. It has the advantages of microinvasive, low surgical response and fast recovery. However, there are shortcomings: ①it is difficult to release cerebrospinal fluid at the beginning, and lumbar puncture is often required for placement of the tube. ②It is difficult to open the optic nerve canal. ③It is inconvenient to deal with the ipsilateral medial optic nerve tumor and to separate the anterior communicating artery, especially when the tumor is hard. For large saddle-nodular meningiomas, an orbitozygomatic approach is necessary to improve the total resection rate [4,6]. Other approaches, including interhemispheric approach, transnasal approach, and interbrow keyhole approach [7], should also be applied, but we lack experience in this area. Our statistical results demonstrate that the surgical access and the rate of total tumor resection are not related to the surgical outcome, indicating that good results can be achieved regardless of the above-mentioned accesses as long as the cases are selected appropriately.
The technique of removing the tumor is the same after choosing the appropriate surgical access, therefore, it is the surgical technique that is the key to total tumor removal [2]. Access is usually from the non-dominant side, or from the side with large tumor invasion, i.e., the side with severe visual function impact. The lateral fissure and the K-artery pool are opened under the microscope from the distal to the proximal side to release the cerebrospinal fluid. As soon as the K-artery arachnoid is opened, the ipsilateral optic nerve is immediately visible. If the optic nerve is covered by the tumor, the midline crest of the pterygoid plateau can be used as an anatomical landmark for spatial orientation. The tumor is first digested with low-intensity bipolar electrocoagulation in the midline of the saddle node or pterygoid plateau to identify the contralateral optic nerve, and the tumor is removed from under this nerve and from the medial aspect of the contralateral internal K artery. The tumor is then resected from the contralateral to the ipsilateral direction under the crossed pool, which helps identify the optic nerve and the K internal artery covered by the tumor. The tumor is then removed from underneath the ipsilateral optic nerve and K internal artery, and finally the tumor is removed from the pituitary stalk and crossed pools separately. The whole procedure, if performed properly, is mildly damaging to brain tissue and less time-consuming. The split resection can protect most of the arachnoid membrane of the internal K artery and the crossed pools, which increases the safety of the operation. The surgical separation should be between the tumor envelope and the arachnoid, leaving a layer of arachnoid membrane overlying the optic nerve, optic cross and pituitary stalk. With a well identified arachnoid plane, the pituitary stalk can be separated from the tumor and never become encapsulated by the tumor [2]. Despite preoperative pituitary stalk compression and intraoperative electrocoagulation of the superior pituitary artery, postoperative pituitary insufficiency occurs rarely. This may be due to the fact that the superior pituitary artery is not the main blood supply to the pituitary gland and pituitary stalk.
The key to protecting visual function is to reduce direct manipulation and injury to the optic nerve and also to avoid damage to its blood supply. The initial intra-tumoral resection must begin centrally, where no optic structures are present. The inferior surface of the optic nerve and optic cross receive blood supply from branches of the superior pituitary artery. During resection of meningiomas, small vessels seen in the arachnoid layer are not electrocoagulated and have a better chance of improving visual function by protecting these vessels [2]. Applying this technique, visual acuity remains stable or improves in 90% of patients with relatively good preoperative visual acuity and good postoperative visual recovery. If the tumor invades the optic canal from under the optic nerve, a 2-mm drill can be used to open the optic canal and remove the tumor with blunt instruments [8].
The difficulty of surgical resection is the relationship of the tumor to the optic nerve, optic cross, anterior cerebral artery, internal K artery and their branches, which are often encircled or displaced. In our group, the A1 and anterior communicating arteries were encircled by the tumor in 16 cases, 14 cases were resected without damaging these vessels, and 2 cases damaged the penetrating branches causing infarction of the head of the caudate nucleus. The tumor was seen to be lobulated on MRI, encircling the internal K artery, which was relatively simple to separate. the internal K artery was posteriorly external and the anterior cerebral artery was posteriorly superior suggesting that complete resection of the tumor was relatively easy. In contrast, meningioma invades the floor of the third ventricle behind the anterior cerebral artery and complete resection of the tumor is difficult. Postoperative deterioration has been associated with intraoperative injury to important branches of the anterior cerebral artery, so identification of the anterior cerebral artery early in surgery is necessary. The blood supply to the saddle node meningioma is mainly from the septal artery, with rare branches of the anterior cerebral artery also providing blood supply. The vessel wall close to the side of the tumor can become thin and even aneurysmatically dilated. Many authors have considered the size of the tumor as a determinant suggesting surgical difficulty [3], since larger tumors cause more severe strain on the adjacent neurovascular vessels. However, the analysis of the surgical findings of our patients does not support this view, perhaps because of the relative infrequency of large saddle-nodal meningiomas. MRI signal intensity analysis confirms the degree of tumor hardness, the presence of calcification, the presence of osteophytes, and the fact that the surrounding vessels are displaced rather than encapsulated, also suggesting the hard nature of the tumor. The tumor invades the optic canal and nasal pay sinus, suggesting the invasive nature of the tumor.
The extent of surgical resection of the tumor determines whether the tumor recurs and regrows. ciric [1] concluded that the size of the saddle node meningioma has an impact on the outcome of surgery, with a mortality rate of 42% in patients with tumors larger than 3 cm and 0 in those smaller than 3 cm. our data do not support that tumor size is not related to the rate of total resection, nor to surgical mortality. Surgical access was also not associated with the extent of resection, with unilateral subfrontal, pterygoid, supraorbital keyhole, and orbitozygomatic access rates of 84.6%, 94.1%, and 100% of total resections, respectively, but not statistically significant. The success of surgery determines postoperative neurological symptoms, visual acuity and long-term outcome. The mortality rate of surgically resected saddle node meningiomas is limited in the literature, with several sources analyzing 0-67% and 0-25% recurrence rates of these tumors [1,9]. The total surgical resection rate in our group was 92.7%, with postoperative visual improvement in 68.3%, no change in 22.0%, and deterioration in 9.8% of patients, superior to the total surgical resection rate of 87.0%, postoperative visual improvement in 55%, no change in 26%, and deterioration in 19% of patients reported by Jallo [2].
In conclusion, most saddle node meningiomas can be completely and safely resected. An important anatomical concept is the nudging of the arachnoid membrane as the tumor grows, and the presence of the arachnoid barrier protects the neurovascular structures during tumor resection. Microsurgical treatment is effective. A supraorbital keyhole approach is preferred for small and medium-sized tumors, a lateral fissure approach at the pterygoid point for medium and large tumors, and an orbitozygomatic approach for large tumors to remove the tumor. Total resection is independent of tumor size and surgical approach, but the characteristics of the tumor and surgical technique play an important role.