Modern surgical outcomes following surgery for ...

See the corresponding editorial in this issue, pp 82–85.

J Neurosurg 119:86–93, 2013 ©AANS, 2013

Modern surgical outcomes following surgery for sphenoid wing meningiomas Clinical article MICHAEL E. SUGHRUE, M.D., MARTIN J. RUTKOWSKI, M.D., C. JARED CHEN, B.S., GOPAL SHANGARI, M.D., ARI J. KANE, M.D., ANDREW T. PARSA, M.D., PH.D., MITCHEL S. BERGER, M.D., AND MICHAEL W. MCDERMOTT, M.D. Brain Tumor Research Center, Department of Neurological Surgery, University of California, San Francisco, California Object. globoid tumors into 3 groups: 1) medial; 2) middle; and 3) lateral. The authors review their experience with resection and postsurgical outcome. Methods. All patients undergoing resection of sphenoid wing meningioma at the authors’ institution over a the sphenoid wing to determine if these tumors behaved differently in terms of symptoms, radiographic characteristics, and postsurgical outcome. Results. A total of 56 patients underwent microsurgical resection for sphenoid wing meningioma during this period. The rates of optic canal invasion (medial 50% vs middle 5% vs lateral 0%; p 0.0001, chi-square test), supraclinoid internal carotid artery encasement (medial 32% vs middle 5% vs lateral 0%; p 0.01, chi-square test), and middle cerebral artery encasement (medial 45% vs middle 24% vs lateral 0%; p 0.01, chi-square test) were all high0.05). Conclusions. The authors report outcomes in a large series of sphenoid wing meningiomas that were treated using modern surgical techniques. (http://thejns.org/doi/abs/10.3171/2012.12.JNS11539)

KEY WORDS

M

constitute approximately 13%–19% of all primary intracranial tumors,2,15 and are the most common tumors of the sphenoid wing in the anterior skull base, accounting for approximately 20% of supratentorial meningiomas.14,15 Cushing and Eigiomas in detail, distinguishing between globoid tumors with a nodular shape and en plaque tumors, which are 9,14 The globoid tumors were further categorized into 3 groups: 1) medial; 2) middle; and 3) lateral.14

the wing roughly demarcating the boundaries of these segments—a medial third, which represents the medial posterior-to-anterior projecting segment most adjacent to the anterior clinoid process (Fig. 1A); a middle third, which runs medial to lateral (Fig. 1B); and a lateral third, which runs anterior to posterior (Fig. 1C), eventually joining with the temporal squamosa. Intuitively, meningiomas that arise in the medial region of the sphenoid wing would seem to present more challenges for neurosurgeons, given their proximity to the optic nerve, the

is divided into thirds, with changes in the orientation of

the CS (Fig. 1D–F).5,7,9,14 By contrast, lateral sphenoid ally, and the optic nerve is less at risk (Fig. 1F and G).

ENINGIOMAS

Abbreviations used in this paper: CS = cavernous sinus; ICA = internal carotid artery; MCA = middle cerebral artery; NS = not California, San Francisco.

86

This article contains some figures that are displayed in color online but in black-and-white in the print edition.

J Neurosurg / Volume 119 / July 2013

Sphenoid wing meningioma outcomes

Fig. 1. Anatomy of various sphenoid wing meningiomas. Representative axial T1-weighted postcontrast MR images depicting medial-third (A), middle-third (B), and lateral-third (C) sphenoid wing meningiomas. The red arrow in panel B depicts the point of origin for this meningioma, which invades the orbit. Artist’s rendition of a small medial sphenoid wing meningioma (D) and a depiction of a larger medial sphenoid wing tumor with canal invasion and involvement of the ICA and MCA (E). Depiction of a large lateral-third sphenoid wing meningioma in axial view (F)—the sylvian fissure is pushed medially and a cleft of brain typically separates the tumor from the vessels—and depiction of the same tumor in coronal view (G).

To date, there are rather limited data analyzing and comparing the surgical outcomes in patients undergoing surgery for different sphenoid wing meningiomas at a single center with modern techniques in the present microsurgical era. We review the experience of resection of meningiomas of the sphenoid wing at a single center, in the radiosurgery era, using modern neurosurgical technology. Patient Population

All patients undergoing neurosurgical intervention resection for meningioma. From this cohort we selected patients undergoing resection for a sphenoid wing meningioma between 1998 and 2007. This study was approved tients provided informed consent. Microsurgical Technique and Perioperative Management

Preoperative evaluation of all patients included T1extent of tumor while T2 images may display the arachnoid layer around the tumor, and also adjacent brain edema, the latter giving some indication of adhesion to the surrounding parenchyma. Preoperative embolization was performed for larger tumors according to the discretion of the treating surgeon, and when the supplying vessels were accessible for intravascular occlusion performed using polyvinyl alcohol particles. The goal of preoperaJ Neurosurg / Volume 119 / July 2013

tive embolization was to both reduce tumor vascularity and intraoperative blood loss, consistent with our prior published experience.8 Surgery was only performed under general anesthesia and after endotracheal intubation. cal navigation systems of a variety of types and manufacturers. Standard microsurgical technique was used. pial plane was usually obliterated throughout the tumorparenchymal interface. These tumors tend to be softer in composition than benign meningioma types, and therefore are usually debulked with use of the Cavitron ultrasonic surgical aspirator and/or the Sonopet (Miwatec, Japan). Tumor-involved bone was removed and cranioplasty done with titanium mesh and/or methyl methacrylate. Intraoperatively, all patients received decadron (10 mg), mannitol (1 g/kg), and ceftriaxone (1 or 2 g) at the time of incision. Postoperatively, all patients were cared for in a neurointensive care unit for 1 day before returning of enoxaparin (40 mg subcutaneously each day) was initiuse of venous thrombosis prophylaxis was not started until after 2001. The incidence of postoperative intracranial hemorrhage was no different in the patient groups before or after routine prophylaxis was begun.6 Data Collection

Clinical information was retrospectively reconstructed using patient medical records, radiological data, and 87

M. E. Sughrue et al. cal facilities. All clinical assessments were performed by a neurosurgeon, and whenever necessary the attending surgeon was consulted to clarify any discrepancies. The fy these tumors based on which portion of the sphenoid wing the base of the tumor arose from. Intraoperative related well with the presumed area of attachment suggested by radiological studies. Although the majority of large tumors made contact with MCA branches at some tumor surrounding at least 270 of the circumference of the supraclinoid ICA or the M1 segment of the MCA, as attached to more than one part of the sphenoid wing (4 tumors), extensive en plaque meningiomas (6 tumors), predominantly hyperostosing meningiomas (7 tumors), and tumors predominantly located in the orbit (8 cases) were excluded from this analysis because they could not

excluded from the surgical outcomes analysis. The demographic characteristics of individual patients can be seen in Table 1. The median age at presentation was 58 years (range 28–84 years). The mean age and sex distribution of these patients did not differ between those with medialthird, middle-third, and lateral-third tumors. The median length of postoperative follow-up was 4.1 years, and did Presenting Symptoms of Patients With Sphenoid Wing Meningiomas

The presenting symptoms of the patients in this series are summarized by site of tumor attachment in Table 2. This complex table can be summarized as follows: visual compromise is predominantly a feature of medialthird sphenoid wing meningiomas, and not unexpectedly declines in frequency as tumors move more laterally on the sphenoid wing. As the tumor origin moves laterally, symptoms become less visual, and more the result of corcommon with the middle-third tumors than with either medial-third or lateral-third tumors.

above, and in our experience the surgical issues these tumors raise are quite different from those in the globoid tumors, and thus these tumors were excluded to ensure a relatively homogeneous cohort. Central pathology review

Outcome After Resection of Sphenoid Wing Meningiomas

In each case, extent of resection was analyzed based on both gross surgical impression of subtotal versus gross-total resection and by using the Simpson grading scale. This information was obtained from the surgeon’s assessment found in the operative notes. In all cases the

Table 3 summarizes the relative patient, histopathological, and imaging characteristics of medial-third, middle-third, and lateral-third sphenoid wing meningiomas. ly between the 3 groups (medial 32 6.2 cm3 vs middle 53 11 cm3 vs lateral 28 8.1 cm3

the surgeon’s intraoperative impression. Mortality data were determined from medical records and were crosschecked against the Social Security Death Index. All data were compiled in an electronic database and crosschecked for accuracy before being subject to any statistical analysis. Statistical Analysis

Binary variables were compared using the Pearson chi-square test, and odds ratios were calculated by comparing the risk of new or worsened neurological dysfunction to the reference variable. Continuous variables were compared using an independent samples t-test or 0.05 after correcting for multiple comparisons by using having new or worse neurological dysfunction after surgery.

similar between the groups (medial 41% vs middle 52% vs lateral 25%; p = NS, chi-square test). The rate of optic canal invasion (medial 50% vs middle 5% vs lateral 0%; p 0.0001, chi-square test), supraclinoid ICA encasement (medial 32% vs middle 5% vs lateral 0%; p 0.01, chi-square test), and proximal MCA encasement (medial 45% vs middle 24% vs lateral 0%; p 0.01, chi-square volvement was generally a feature of middle-third–based tumors (medial 9% vs middle 33% vs lateral 6%; p 0.05, chi-square test). We calculated odds ratios of surgical complications patients with these imaging, pathological, and clinical characteristics. Neurosurgical complications other than of 56 patients, and included 4 wound infections, 1 wound dehiscence, and 1 CSF leak. We noted no morbidity related to the orbital or orbitozygomatic osteotomy. The rate of neurosurgical complications other than new or worsened

Patient Demographic Data

note, location along the sphenoid wing did not differ sig-

A total of 59 patients underwent treatment for sphenoid wing meningioma during this period, of whom 56

10 (19%) of 56 patients. These are summarized by loca-

The 3 patients treated with radiosurgery were included in the analysis of presenting symptoms; however, they were 88

only location of the tumor along the medial third of the J Neurosurg / Volume 119 / July 2013

Sphenoid wing meningioma outcomes TABLE 1: Demographic characteristics in 59 patients with sphenoid wing meningiomas* Case No.

Age (yrs), Sex

Vol (cm3)

Simpson Grade

WHO Grade

1.9 3.5 2.2 2.1 1.5 3.0 2.8 4.5 1.3 3.8 3.0 2.5 2.5 3.5 3.2 5.3 3.0 3.1 1.8 3.0 4.0 2.8

11 37 11 32 9 61 11 77 3 46 67 19 28 27 22 127 36 14 5 17 26 21

3 2 3 4 3 2 4 4 4 3 4 4 4 4 4 3 4 3 2 3 2 3

I I I I I II II II II II I II II I I I II I I II I I

2.5 2.0 2.1 4.7 3.0 2.5 3.3 4.2 4.4 2.2 6.0 5.0 2.2 3.8 2.0 4.0 2.8 4.0 5.0 4.0 2.0

2.0 2.2 2.5 5.4 2.5 2.8 3.2 4.4 3.8 2.5 6.0 4.2 1.5 4.5 2.5 4.1 3.2 2.8 5.4 4.2 2.0

11 9 9 114 20 15 32 74 70 14 198 95 5 72 10 62 27 47 149 76 6

1 3 1 1 3 1 4 1 4 2 4 2 2 1 3 3 4 4 3 4 1

I I I II I I I II II II II I I II I II II II I II II

3.4

3.2

33

3

II

Embo

OZ

Side

Dimensions (cm)

yes yes no no yes yes no yes no yes no yes no no yes no no no no no yes no

yes yes no yes yes yes yes yes no yes no no yes yes no yes yes yes yes no yes yes

rt rt lt lt lt rt lt lt rt lt lt rt lt rt rt rt rt rt rt rt lt lt

2.5 3.5 2.2 4.3 2.5 4.4 2.1 4.2 1.5 3.3 4.4 3.0 3.8 2.8 2.5 5.0 3.8 1.9 1.5 2.5 2.2 3.0

2.3 3.0 2.3 3.5 2.5 4.6 1.8 4.1 1.5 3.7 5.1 2.5 2.9 2.8 2.7 4.8 3.2 2.3 2.0 2.2 3.0 2.5

no yes no no yes yes no yes yes no yes yes no yes yes yes yes yes yes yes no

yes no no no yes yes no yes yes no yes no no yes yes yes yes no yes no no

rt lt lt rt rt rt lt lt rt lt lt rt lt rt lt rt lt lt rt rt rt

2.2 2.0 1.8 4.5 2.6 2.2 3.0 4.0 4.2 2.5 5.5 4.5 1.4 4.2 2.0 3.8 3.0 4.2 5.5 4.5 1.5

yes

yes

lt

3.0

pts w/ medial-third tumors 1 69, F 2 66, F 3 46, F 4 55, F 5 71, F 6 40, M 7 58, F 8 77, F 9 72, F 10 57, F 11 33, M 12 78, F 13 36, F 14 50, M 15 41, F 16 28, M 17 61, F 18 49, F 19 39, F 20 57, F 21 71, F 22 38, M pts w/ middle-third tumors 1 41, F 2 33, F 3 84, F 4 41, M 5 50, M 6 60, F 7 79, F 8 58, M 9 74, M 10 72, M 11 51, M 12 53, M 13 43, F 14 72, F 15 55, F 16 50, M 17 68, M 18 51, F 19 60, M 20 78, M 21 76, F pts w/ lateral-third tumors 1 66, F

(continued)


89

M. E. Sughrue et al. TABLE 1: Demographic characteristics in 59 patients with sphenoid wing meningiomas* (continued) Case No.

Age (yrs), Sex

Embo

OZ

Side

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

60, F 53, F 79, F 46, F 60, F 68, M 46, F 72, F 80, F 59, F 36, F 51, F 75, F 72, M 78, F

yes no no no yes no no yes yes yes yes yes no no no

no yes no yes yes yes yes yes yes no no yes SRS SRS SRS

lt rt lt lt lt lt lt rt lt rt lt lt rt lt rt

Dimensions (cm) 5.0 1.8 1.7 1.5 2.8 3.0 2.0 4.0 3.8 2.0 3.5 3.0 1.0 1.0 1.8

5.2 2.0 2.0 2.5 2.5 2.8 2.5 2.8 4.2 2.0 3.5 2.8 1.2 1.2 1.8

4.8 2.5 2.0 2.2 2.6 3.5 2.8 3.8 4.5 2.2 4.0 3.2 1.3 0.9 1.5

Vol (cm3)

Simpson Grade

WHO Grade

125 9 7 8 18 29 14 43 72 9 49 27 2 1 5

3 4 2 3 2 4 2 2 3 2 1 3 NA NA NA

I I I I II I II I I I I II NA NA NA

* embo = embolization; NA = not applicable; OZ = orbitozygomatic; pts = patients.

can range from a straightforward procedure obtaining gross-total resection to a complicated and risky procedure in which total tumor removal is unwise due to tumor adherence to critical neurovascular structures. Although cranial base meningiomas in this region can assume a va-

(Table 5).

of the tumor, resection of sphenoid wing meningiomas

to the skull base and its contents, with thoughtful system-

TABLE 2: Presenting symptoms summarized by location along the sphenoid wing in 56 patients with meningiomas who underwent microsurgical resection No. Symptom vertigo seizures proptosis headaches language disturbance visual disturbance behavioral change tinnitus dizziness fatigue gait disturbance syncope facial numbness ptosis amenorrhea shortness of breath intracranial hemorrhage palpebral paresis no. of pts

90

%

Medial

Middle

Lateral

Medial

Middle

Lateral

2 2 2 6 0 1 16 1 0 0 0 0 1 1 1 1 0 0 0 23

1 5 4 5 1 1 5 3 0 0 0 0 2 1 0 0 1 1 1 18

3 3 2 7 4 2 1 1 1 1 1 1 0 0 0 0 0 0 0 15

9 9 9 26 0 4 70 4 0 0 0 0 4 4 4 4 0 0 0

6 28 22 28 6 6 28 17 0 0 0 0 11 6 0 0 6 6 6

20 20 13 47 27 13 7 7 7 7 7 7 0 0 0 0 0 0 0


Sphenoid wing meningioma outcomes TABLE 3: Patient and tumor characteristics summarized by lesion location along the sphenoid wing in 59 patients with meningiomas* Value Characteristic mean age in yrs age 65 yrs mean tumor vol (cm3) tumor vol 20 cm3 optic canal invasion orbital involvement CS invasion ICA encasement MCA encasement tumor grade WHO I WHO II WHO III GTR no. of pts

Medial 54

Middle

3.2 8 32 6.2 13 11 2 6 7 10

59

3.2 10 53 11 12 1 7 3 1 5

13 9 0 12 22

10 11 0 15 21

% Lateral 3.3 10 28 8.1 7 0 1 0 0 0

Medial

Middle

Lateral

36

48

63

59 50 9 27 32 45

57 5 33 14 5 24

44 0 6 0 0 0

59 41 0 55

48 52 0 71

56 25 0 69

63

9 4 0 11 16

p Value† NS NS NS NS 0.0001 0.05 0.07 0.01 0.01 NS

0.17

* The mean values are expressed SE. Abbreviation: GTR = gross-total resection. † Calculated according to the chi-square test.

atic analysis of preoperative imaging data, the anatomy of parts, facilitating operative planning. In this study, we aimed to describe outcomes using contemporary surgical techniques, to better clarify the expected rates and outand a growing propensity for observation of smaller meundergoing resection at our center and others, because they are likely to be bigger and more aggressive that those operated on in the past.16 Importantly, medial-third sphenoid wing meningiomas are most notable for their proximity to the ipsilateral optic nerve and optic apparatus, which is the symptom prompting the diagnostic imaging study in more than involvement are frequent features of these tumors, and are more common than with lateral parasylvian tumors. This medial and posterior location of these tumors usually requires an orbital osteotomy to obtain the anteriorand complications by location in 10 patients with sphenoid wing meningiomas* Medial

Middle

Lateral

ptosis CN III palsy (3 pts) global aphasia expressive aphasia (2 pts) visual decline

CN III palsy aphasia

none

* CN = cranial nerve.


to-posterior trajectory necessary for effective visualization of the tumor base and its relationship with the optic nerve. In many cases of larger lateral sphenoid wing tumors, an orbital or orbitozygomatic osteotomy is unnecessary. These lateral tumors frequently create their own unique challenges, namely bone reconstruction, orbital repair, and soft-tissue repair and cosmesis, which require consideration in the preoperative planning. Additionally, when there is hyperostosis of the greater sphenoid wing the middle fossa, an orbitozygomatic approach drops the temporalis muscle bulk away from the line of sight for the surgical approach and facilitates removal of involved bone in the posterolateral orbit above and below the sudum and foramen ovale.4,12,13 Thus, approach selection for lateral sphenoid wing meningiomas should be performed on a case-by-case basis, with a bias toward avoiding the use of orbital osteotomy in these cases unless necessary. Furthermore, the medial tumors are in proximity to postoperative cranial nerve palsies is consequently higher with medially located tumors. In fact, the rates of neurothese lesions, even controlling for other tumor factors. pensity of these tumors to invade the optic canal, encase major vessels, and invade the CS along the path of the cranial nerves. This occurred in our series despite a conscious bias on our part toward limited resections in cases of dense tumor adherence to critical neural and vascular structures. 91

M. E. Sughrue et al. TABLE 5: Calculated odds ratios for the risk of new or worsened characteristics in patients with sphenoid wing meningiomas Characteristic age 65 yrs tumor vol 20 cm3 optic canal invasion orbital involvement CS invasion supraclinoid ICA encasement MCA encasement WHO Grade II GTR medial third of sphenoid wing

p Value* 0.7 1.3 1.4 1.7 1.2 1.3 1.6 0.5 0.8 2.7

0.39 0.45 0.43 0.31 0.57 0.49 0.33 0.22 0.49 0.05

* Calculated according to the chi-square test.

As with many cranial base tumors, there still exists for the management of sphenoid wing meningiomas. In general, our impression is that the pool of surgically treated sphenoid wing meningiomas has generally become increasingly technically complex, because simpler tumors are more frequently undergoing observation or radiosurgery. As experience with these lesions increases, certain ideas are increasingly gaining acceptance, and we in this fashion. First, although vascular encasement is common, and sometimes the arachnoidal plane remains intact enough to dissect the vessel free from the tumor, in many cases the tumor is too adherent to do this, and should be truncated and left as a small pad on the vessel. Previous efforts to remove adherent tumor from encased vessels have been met with rates of vascular injury exceeding 20%.10 In cases of vessel encasement, we conservatively explore to see if the arachnoidal plane has been preserved and can easily be removed from the vessel. If this is not possible, then we shave down the residual tumor to make it as small as possible and leave it, to be folThis conservative approach seems to have reduced the rates of vascular injury, despite the arterial system being encased in nearly half of our medial-third sphenoid wing tumors. We hypothesize that many of these tumor remnants undergo growth arrest and remain dormant, based on data that we have previously published showing that aggressive tumors with near-total resections often do not recur, even with no additional therapy.17 Additionally, we and others have largely ceased chasing these tumors into the CS given the excellent rates of control with adjuvant radiosurgery or radiotherapy in this region.18 The rates of successful tumor control with this approach have generally been good, with control rates approaching 80% over 7–8 years postsurgery without any additional treatment.1,3,14 Presently the major remaining morbidity in modern management of these tumors is cranial neuropathy, typically of the third cranial nerve, and typically with medi92

al-third tumors. We try to minimize the use of cautery near the tentorial junction, near where the third cranial nerve would be expected to enter the oculomotor triangle. Despite this, and despite not entering the CS, we saw partial third nerve palsies in approximately 20% of our medial-third tumor cases. Similar numbers have been reported by other groups. For example, Abdel-Aziz and colleagues1 reported cranial neuropathies in 16% of their 11 noted diplopia in 15.8% of patients. These similar rates suggest that a better understanding of the relationship of tumor to tentorium and clinoid process is needed to reduce this rate further cranial neuropathies.

We present our institutional experience with surgical results support the idea of classifying the sphenoid wing meningiomas into 2 distinct types of tumors, rather than 3, and these differences should guide discussions with patients during the informed consent process as well as in surgical planning of tumor resection.

tumor surgery. Dr Sughrue is supported by the AANS Neurosurgery

that they are not involved in any relationships with companies that make products related to this study. Author contributions to the study and manuscript preparation include the following. Conception and design: McDermott, Sughrue.

uscript: all authors. Approved the final version of the manuscript on behalf of all authors: McDermott. Statistical analysis: Sughrue.

The authors thank Meningioma Mommas for their support in facilitating long-term outcomes research.

involving the cavernous sinus: conservative surgical strategies for better functional outcomes. 54:1375–1384, 2004 illary meningioma: a rare but distinct variant of malignant meningioma. 2:3, 2007 strategies for giant medial sphenoid wing meningiomas: a new scoring system for predicting extent of resection. 150:865–877, 2008 -


Sphenoid wing meningioma outcomes sive, hyperostotic sphenoid wing meningiomas. J Neurosurg 107:905–912, 2007 5. Bonnal J, Thibaut A, Brotchi J, Born J: Invading meningiomas of the sphenoid ridge. J Neurosurg 53:587–599, 1980 Berger MS, et al: Adjuvant enoxaparin therapy may decrease the incidence of postoperative thrombotic events though does not increase the incidence of postoperative intracranial hemorrhage in patients with meningiomas. 93:151– 156, 2009 7. Carrizo A, Basso A: Current surgical treatment for sphenoorbital meningiomas. 574–578, 1998 traoperative blood loss for embolized meningiomas. Neuro50:1231–1237, 2002

during radical resection of medial sphenoid wing meningiomas. 11:99–104, 2001 mas.

67

377–384, 2010

tion after resection of sphenoid wing meningiomas. 7:99–103, 2005 Combined frontotemporal-orbitozygomatic approach for tu-


mors of the sphenoid wing and orbit. 116, 1990

26:107–

sphenoid wing meningiomas: clinical outcome and recurrence rate. 626–639, 2006 noid wing meningiomas with ossesous involvement. Surg 64:37–43, 2005 ningioma resection in the modern neurosurgical era. Clinical article. J Neurosurg 112:913–919, 2010 mott MW, Berger MS, et al: The relevance of Simpson Grade I and II resection in modern neurosurgical treatment of World J Neurosurg 113:1029–1035, 2010 MW, Parsa AT: Factors affecting outcome following treatment of patients with cavernous sinus meningiomas. Clinical ar ticle. J Neurosurg 113:1087–1092, 2010 Manuscript submitted March 28, 2011. Accepted December 4, 2012. Please include this information when citing this paper: published Address correspondence to: Michael W. McDermott, M.D., DeFrancisco, 505 Parnassus Avenue, Box 0112, San Francisco, California 94143. email: [email protected].

93