Neurosurgery_TechChir_2006.pdf

OPERATIVE TECHNIQUES

OLFACTORY GROOVE MENINGIOMAS FROM

NEUROSURGICAL AND EAR, NOSE, AND

THROAT PERSPECTIVES: APPROACHES,TECHNIQUES, AND OUTCOMES

Sergey Spektor, M.D.,Ph.D.Department of Neurosurgery,Hadassah University Hospital,Jerusalem, Israel

Javier Valarezo, M.D.Department of Neurosurgery,Hadassah University Hospital,Jerusalem, Israel

Dan M. Fliss, M.D.Department of Otolaryngology,Head and Neck Surgery,Tel Aviv Sourasky Medical Center,Tel Aviv, Israel

Ziv Gil, M.D., Ph.D.Department of Otolaryngology,Head and Neck Surgery,Tel Aviv Sourasky Medical Center,Tel Aviv, Israel

Jose Cohen, M.D.Department of Neurosurgery,Hadassah University Hospital,Jerusalem, Israel

Jose Goldman, M.D.Department of Neurosurgery,Hadassah University Hospital,Jerusalem, Israel

Felix Umansky, M.D.Department of Neurosurgery,Hadassah University Hospital,Jerusalem, Israel

Reprint requests:Sergey Spektor, M.D., Ph.D.,Department of Neurosurgery,P.O. Box 12000,Hadassah University Hospital,Jerusalem, Israel 91120.Email: [email protected]

Received, September 1, 2004.

Accepted, March 8, 2005.

OBJECTIVE: To review the surgical approaches, techniques, outcomes, and recurrencerates in a series of 80 olfactory groove meningioma (OGM) patients operated onbetween 1990 and 2003.METHODS: Eighty patients underwent 81 OGM surgeries. Tumor diameter variedfrom 2 to 9 cm (average, 4.6 cm). In 35 surgeries (43.2%), the tumor was removedthrough bifrontal craniotomy; nine operations (11.1%) were performed through aunilateral subfrontal approach; 18 surgeries (22.2%) were performed through a pteri-onal approach; seven surgeries (8.6%) were carried out using a fronto-orbital crani-otomy; and 12 procedures (14.8%) were accomplished via a subcranial approach.Nine patients (11.3%) had undergone surgery previously and had recurrent tumor.RESULTS: Total removal was obtained in 72 patients (90.0%); subtotal removal wasachieved in 8 patients (10.0%). Two patients, one with total and one with subtotalremoval, had atypical (World Health Organization Grade II) meningiomas, whereas 78patients had World Health Organization Grade I tumors. There was no operativemortality and no new permanent focal neurological deficit besides anosmia. Twenty-five patients (31.3%) experienced surgery-related complications. There were no re-currences in 75 patients (93.8%) 6 to 164 months (mean, 70.8 mo) after surgery. Threepatients (3.8%) were lost to follow-up. In two patients (2.5%) with subtotal removal,the residual evidenced growth on computed tomography and/or magnetic resonanceimaging 1 year after surgery. One of them had an atypical meningioma. The second,a multiple meningiomata patient, was operated on twice in this series.CONCLUSION: A variety of surgical approaches are used for OGM resection. Anapproach tailored to the tumor’s size, location, and extension, combined with modernmicrosurgical cranial base techniques, allows full OGM removal with minimal per-manent morbidity, excellent neurological outcome, and very low recurrence rates.

KEY WORDS: Cranial base, Olfactory groove meningioma, Radical resection, Recurrence, Surgicalapproach, Tumor extension

Neurosurgery 57[ONS Suppl 3]:ONS-268–ONS-280, 2005 DOI: 10.1227/01.NEU.0000176409.70668.EB

Olfactory groove meningiomas (OGMs)arise in the midline over the cribriformplate and frontosphenoidal suture (18,

31). It is well known that most of these tumorsoccupy the floor of the anterior cranial fossa,extending all the way from the crista galli to thetuberculum sellae (4). There are some similari-ties between posteriorly extending OGMs andtuberculum sellae meningiomas. The main dis-tinguishing feature is the location of the optic

apparatus in relation to the tumor (18). OGMspush the optic nerves and the chiasm down-ward and posteriorly as they grow (31, 45).Tuberculum sellae meningiomas elevate the chi-asm and displace the optic nerve superolater-ally; thus, the tumor occupies a subchiasmalposition (2, 18, 19). These benign, slow-growingtumors frequently achieve large size before de-tection. Surgical removal of small to mid-sizeOGMs usually is not difficult. However in a

ONS-268 | VOLUME 57 | OPERATIVE NEUROSURGERY 4 | OCTOBER 2005 www.neurosurgery-online.com

significant proportion of patients, the tumor is very large and/orinfiltrates or involves surrounding structures, making its re-moval challenging.

Several surgical approaches can be applied for tumor re-moval. Traditionally, frontal or bifrontal craniotomy has beenused with subfrontal approach to the tumor (4, 6, 10, 29, 31, 35,43–45). More recently, some surgeons have used a pterionalapproach (17, 33, 38, 46, 47). More aggressive approaches havebeen proposed for resection of OGMs expanding into theparanasal sinuses and orbits, including transbasal (9), subcra-nial (13, 36), and fronto-orbital approaches (41), frontal orbifrontal craniotomy combined with orbital or nasal osteoto-mies (3, 18), and craniofacial resection (15).

This article presents results from the surgical treatment of 80patients with OGMs via bifrontal, unilateral subfrontal, pteri-onal, fronto-orbital, and subcranial approaches from 1990through 2003, the largest patient series and the longest post-operative follow-up reported in the recent neurosurgicalliterature.

PATIENTS AND METHODS

Patient Population

From September 1990 through October 2003, our neurosur-gical team operated on 68 patients with OGM tumors inHadassah University Hospital, Jerusalem, and 12 patientsjointly with an ear, nose, and throat (ENT) team at the Ichilov-Sourasky Medical Center, Tel Aviv. One multiple meningiomapatient who underwent surgery at Hadassah was operated ona second time when his meningioma recurred within the firstyear.

Demographic data are presented in Table 1. There was asignificant female predominance (58 patients; 72.5%). Patientage ranged from 16 to 85 years (average, 55 yr). Headache in41 patients (51.3%) and anosmia in 39 patients (48.8%) werethe most common presenting symptoms. At admission, 21patients (26.3%) had both mental and personality changes.Twenty-two patients (27.5%) demonstrated visual impair-ment, including 11 (13.8%) with papilledema. Nine patients(11.3%) had epilepsy at presentation (Table 2). All patientsunderwent preoperative and postoperative neuroimaging

studies, including computed tomography and/or magneticresonance imaging (MRI), and an ophthalmic examination.

Fifty-eight patients (72.5%) had tumors 4 cm or more indiameter. Tumors were centered on the midline in 73 patients(91.3%). Seven tumors extended laterally (six right [7.5%]; oneleft [1.3%]). Thirty-eight patients (47.5%) had purely intracra-nial tumors with no optic nerve involvement; 21 patients(26.3%) had intracranial tumors including optic nerve involve-ment; 15 (18.8%) patients had intracranial tumors with exten-sion to the ethmoidal sinus, frontal sinus, nasal cavity, and/ororbital area; and 6 patients (7.5%) had both optic nerve in-volvement and paranasal sinus extension (Table 3).

Nine patients (11.3%) referred for recurrent tumor had beenoperated on previously via regular craniotomy (using a bifrontal,frontal, or pterional approach) from 1 to 16 years earlier (average,5.8 yr). Among the nine patients, five patients had intracranialtumors with no optic nerve involvement, two patients had intra-cranial tumors with optic nerve involvement, one had both para-nasal extension and optic nerve involvement, and one had para-nasal extension but no optic nerve involvement. MRI studiesdepicting OGM extension patterns characteristic of tumors in ourseries are shown in Figure 1.

Surgical Techniques

The surgical goal was radical tumor resection unless saferemoval was precluded by significant invasion of the anterior

TABLE 1. Demographic data of 80 patients operated forolfactory groove meningioma

Characteristic No. of patients (%)

SexMale 22 (27.5%)Female 58 (72.5%)

Age (yr)Median 55Range 16–85

TABLE 2. Presenting signs and symptoms in olfactory groovemeningioma patients

Signs and symptoms No. of patients (%)

Olfactory impairment 47 (58.8%)

Headache 41 (51.3%)

Visual impairment 22 (27.5%)

Mental change 21 (26.3%)

Papilledema 11 (13.8%)

Epilepsy 9 (11.3%)

Nasal obstruction 8 (10.0%)

Motor deficit 4 (5.0%)

Incontinence 3 (3.8%)

Optic atrophy 3 (3.8%)

Foster Kennedy 3 (3.8%)

Sinusitis 2 (2.5%)

Exophthalmos 1 (1.3%)

Telecanthus 1 (1.3%)

Facial deformity 1 (1.3%)

OLFACTORY GROOVE MENINGIOMA SURGERY

NEUROSURGERY VOLUME 57 | OPERATIVE NEUROSURGERY 4 | OCTOBER 2005 | ONS-269

communicating artery complex or optic nerves. After tumorremoval, the dural attachment was coagulated and the durawas stripped. Underlying hyperostosis was drilled away.

Thirty-five surgeries (43.2%) in patients with large tumorswere performed through a bifrontal craniotomy, and ninesurgeries (11.1%) were performed through a unilateral sub-frontal approach. Twelve patients (14.8%) with tumors ex-tending far into the paranasal sinus, nasal cavity, and/ororbits were operated on via a subcranial approach. We usedthe pterional approach in 18 surgeries (22.2%) and the fronto-orbital approach in 7 surgeries (8.6%). Figure 2 illustrates eachof the approaches, Table 4 summarizes data on tumor exten-sion and surgical approach, and Table 5 summarizes data ontumor size and surgical approach.

Bifrontal, unilateral-subfrontal, and pterional approacheswere performed as described previously (17, 31, 35, 46). Thefronto-orbital approach was made through a frontotemporalcraniotomy with a single-piece bone flap containing the fron-tal bone, frontozygomatic process, supraorbital rim, and ante-rior part of the orbital roof, as proposed by Delashaw et al. (7).

The subcranial approach was performed using the tech-nique of Raveh et al. (36), with modifications as describedelsewhere (12, 13). With the patient in the supine position, avery low bicoronal skin flap beneath the orbital rims andnasion is elevated, preserving the pericranium. The periorbitbeneath both orbital roofs is separated. Orbital and frontalosteotomies are made, and the bone flap, containing bothorbital rims, orbital roofs, nasion, frontal sinuses, and a part ofthe ethmoidal air cells, is elevated. This approach provides an

immediate and direct route to the cribriform plate and olfac-tory grooves, ethmoidal and sphenoid sinuses, nasal cavities,and both orbits. The optic canals may be opened if necessary,exposing the optic nerves.

An autologous pericranial flap, fat, temporal muscle andtemporal fascia, autologous fascia lata, and fibrin glue wereused for reconstruction and prophylaxis of rhinorrhea, regard-less of the surgical approach, whenever the frontal and/orethmoidal sinuses were opened. Naturally, the subcranial ap-proach created the most significant intracranial and nasopha-ryngeal anterior cranial base defect. Meticulous multilayerreconstruction was performed for these patients, as previouslydescribed (11), and they were managed with continuous lum-bar drain for a minimum of 5 days.

RESULTS

The most frequent postoperative complication was cerebro-spinal fluid (CSF) leak, which occurred in 10 patients (12.5%)and resulted in meningitis in 4 patients (5%). CSF leak stoppedafter several days of lumbar draining in eight patients. Twopatients with persistent rhinorrhea underwent reoperationand their sinuses were obliterated. Rhinorrhea recurred in oneof these patients and led to meningitis and death 6 monthsafter surgery. Four patients (5%) experienced intracranial he-matoma with resulting surgical evacuation in two cases. Threehemorrhages occurred in the tumor bed; the fourth was aremote contralateral, frontoparietal, epidural hematoma.Three patients (3.8%) experienced seizures immediately aftersurgery. Five patients (6.3%) had deep vein thrombosis andtwo patients (2.5%) experienced pulmonary embolism. Therewas no new permanent neurological deficit besides anosmia.Table 6 summarizes postoperative complications according tosurgical approach.

Total removal was obtained in 72 patients (90%). Subtotalremoval was achieved in eight patients (10%) with tumors 4cm or larger and capsule adherent to the optic nerve and/orvascular structures, precluding safe, complete removal (Table7). Two patients, one with total and one with subtotal removal,had atypical (World Health Organization [WHO] Grade II)meningiomas, whereas 78 patients had WHO Grade I tumors(meningothelial, transitional, psammomatous, or secretory).Postoperative Karnofsky score was 100 in 37 patients (46.3%),90 in 28 patients (35.0%), 80 in 11 patients (13.8%), 70 in 3patients (3.8%), and 50 in 1 patient who had a Karnofsky scoreof 30 before surgery.

We examined retrospectively the postoperative history ofthe 80 patients in this study, who are now 6 to 164 months(average, 70.8 mo) from surgery, with telephone contactand/or medical examination, as well as file review. Seventy-two patients (90.0%) are alive, 1 patient died 6 months afterOGM surgery of meningitis resulting from persistent rhinor-rhea, 4 patients died of causes unrelated to surgery, and 3patients (3.8%) who had total removal of their WHO I menin-giomas were lost to follow-up.

TABLE 3. Tumor location and invasion patterns

Tumor location and invasion No. of patients (%)

Intracranial tumor, no optic nerveinvolvement

38 (47.5%)

Intracranial tumor, optic nerveinvolvement

21 (26.2%)

Intracranial tumors with paranasalextension

21 (26.3%)

Ethmoidal, optic nerve 4Ethmoidal, nasal, orbital 4Ethmoidal, frontal, nasal, orbital 3Ethmoidal 2Ethmoidal, nasal, optic nerve 1Ethmoidal, nasal 1Ethmoidal, nasal, orbital, sphenoidal 1Ethmoidal, frontal, orbital, sphenoidal 1Ethmoidal, frontal 1Ethmoidal, nasal, orbital, optic nerve 1Frontal 1Orbital, maxillar, infratemporal 1

Total patients 80 (100.0%)

SPEKTOR ET AL.


There has been no clinical or radiographic (MRI) olfactorygroove recurrence documented to date in any of the 72 pa-tients with total removal. Eight patients had subtotal removal.Six of these patients are alive and have had no regrowth from17 to 90 months after surgery (average, 63.5 mo). Residualtumor showed growth on MRI images within 1 year of surgeryin two of these eight patients. One patient had multiple me-ningiomas at presentation, including a 6-cm OGM. Residualtumor evidenced growth within 1 year after subtotal resectionvia a bifrontal approach. He underwent surgery a second timein this series with a subfrontal approach, and residual tumoragain showed growth on MRI within 1 year. He is alive 112months after the second surgery, but his multiple meningio-mata and his overall clinical state preclude further surgery.The second patient underwent surgery at age 85 years via thepterional approach for a 7-cm atypical meningioma. He dem-onstrated residual regrowth 1 year after surgery but did notundergo reoperation. He died of a massive stroke 2 years aftersurgery.

DISCUSSION

Surgical techniques have evolved a great deal since Cush-ing’s pioneering 1938 publication describing OGM removal in22 patients through a unilateral subfrontal approach withpartial frontal lobe resection (6). Although OGM is the subjectof many publications, optimal surgical policy is still to bedefined. Today, the range of approaches varies from a verywide bifrontal craniotomy (10) to minimally invasive endo-scopic techniques (20) and craniofacial approaches (15, 23, 26,34). For many years, unilateral or bifrontal craniotomy fol-lowed by subfrontal access to the tumor have been consideredstandard approaches for OGM resection (4, 27, 29, 31, 35, 43,45).

Mortality rates in the literature vary from 0% (18, 29, 32, 37,46) to 17% (43) and even 22.7% (6) in the older literature.Complications include postoperative epilepsy, postoperativehematoma, hemiparesis, visual and mental deterioration, boneflap infection, and CSF leak. Surgical approaches have contin-

FIGURE 1. MRI scans showing common growth patterns of olfactory groove meningiomas in patients from this study.



ued to evolve over time. The pterional approach (17, 33, 46)and approaches complemented with fronto-orbital osteoto-mies (1, 3, 40, 41) are now well described. There are also a fewreports describing removal of OGM through craniofacial (15,23, 34) and subcranial approaches (13, 14, 21, 26).

In our experience, the procedure evolved from surgerythrough a standard unilateral or bilateral subfrontal approachto resection through a pterional approach, and later to pteri-onal approach extended by fronto-orbital osteotomy and sub-cranial approach. Tables 6 to 10 describe the outcomes, advan-tages, and disadvantages of these surgical approaches.

Bifrontal Craniotomy with Subfrontal Approach

The advantage of the bilateral subfrontal approach, widesymmetrical anterior cranial fossa exposure, is described inmany papers (10, 24, 29, 31, 35). This approach providesexcellent opportunity for radical tumor resection, drilling ofhyperostosis in the cribriform plate area of the planum sphe-

noidale and tuberculum sellae, and unroofing of optic nerveswhen necessary.

Disadvantages are also well known. The frontal sinuses areusually opened; nevertheless, the anterior inferior border ofthe craniotomy is not low enough to prevent significant brainretraction. In patients with large tumors, swollen brain oftenherniates into the craniotomy window, frequently demandingpartial resection of a frontal lobe. The most important struc-tures—the optic apparatus, carotids, and the anterior commu-nicating complex—come into view late, after a major part ofthe tumor has been removed. The superior sagittal sinusshould be divided, compromising venous drainage from thefrontal lobes and thus contributing to brain edema. Preserva-tion of olfaction is problematic.

In our series, 35 surgeries (43.8%) were performed with thisapproach. Total resection was achieved in 31 surgeries, andsubtotal resection was achieved in 4 surgeries. There were norecurrences in 32 patients during the observation period; 2patients with total removal were lost to follow-up. One mul-tiple meningiomata patient evidenced growth of his residualtumor within the first year after surgery. Five patients hadCSF leak through the frontal sinuses, complicated with men-ingitis in three patients. Two patients had frontal contusion asa result of retraction against swollen brain during the earlysteps of surgery. Two patients experienced intraparenchyma-tous frontal hemorrhage; one hemorrhage required surgicalevacuation. Three patients experienced deep vein thrombosis,one had a pulmonary embolism, and one experiencedconvulsions.

Unilateral Frontal Craniotomy withSubfrontal Approach

This approach has the advantage of sparing the contralateralfrontal lobe and the superior sagittal sinus. The disadvantagesinclude those described above. In addition, this approach pro-vides only a small opening with a very narrow view.

Nine surgeries (11.1%) wereperformed through a unilat-eral subfrontal approach. Weachieved total resection inseven patients and subtotal re-section in two patients. Onepatient underwent surgerytwice in our series, first via abifrontal approach and 1 yearlater via a subfrontal approachafter the tumor recurred. Thissame patient underwent reop-eration for a persistent CSFleak after the subfrontal resec-tion, and again experiencedtumor regrowth. No other pa-tient experienced recurrenceafter subfrontal resection. Onepatient experienced pulmo-

FIGURE 2. Diagramsshowing surgical ap-proaches used for OGM re-section in this study. A, bi-frontal craniotomy withsubfrontal approach. B,unilateral-frontal craniot-omy with subfrontal ap-proach. C, pterional approach. D, fronto-orbital approach. E, subcranial approach.

TABLE 4. Tumor extension patterns according to surgical approach

IntracranialIntracranial and

optic nerveinvolvement

Intracranial andparanasal sinus

extension

Intracranial,paranasal sinus,and optic nerve

involvement

Total

Bifrontal 19 8 4 4 35

Unilateral subfrontal 5 3 1 9

Pterional 13 4 1 18

Fronto-orbital 1 6 7

Subcranial 11 1 12

Total surgeries 38 21 15 7 81a

a Eighty-one surgeries were performed on 80 patients.

SPEKTOR ET AL.


nary embolism. One patient with total removal of his WHOGrade I meningioma was lost to follow-up.

Pterional Approach

This relatively new approach to OGM (17, 46) providesseveral significant advantages, clearly described by Bricolo (5),compared with the subfrontal approaches. It is less time con-suming than bifrontal craniotomy and avoids CSF leak, be-cause the frontal sinus is not transected. The basal cisterns maybe opened before the tumor is even touched, releasing the CSFand providing fair brain relaxation. The optic nerve may belocalized and secured before the tumor is manipulated. Theipsilateral internal carotid artery comes into view early, andthe anterior cerebral arteries may be followed into the tumor(Fig. 3). Turazzi et al. (46) proved that this approach provides

shorter distance to the tumorcompared with the subfron-tal route. The pterional ap-proach also spares venousstructures (33) and providesthe possibility of early tumordetachment from the cranialbase and debulking.

The major disadvantage isa narrow working angle. Inpatients with high-riding tu-mors, the upper part of thetumor is in a blind area, andsignificant brain retractionmay be demanded to visual-ize it. The working distanceto the opposite side of menin-gioma is very long. Orbitalroof bulging obscures the ol-factory groove, and so signif-icant frontal lobe retractionmay be necessary to improvethe working angle.

Initially, we chose this ap-proach only for resection ofrelatively small tumors, butwith experience we also usedit for larger tumors, with thelimitations mentioned above.We applied this approach forremoval of OGM in 18 sur-geries (22.2%), with total re-section in 15 surgeries. Resid-ual tumor increased in sizewithin the first year after sur-gery in one patient operatedon at age 85 years who hadsubtotal resection of his atyp-ical 7-cm meningioma. Heexperienced no complica-

tions, but died following a massive stroke 2 years after OGMsurgery. Two other patients who had subtotal resection, a72-year-old man and an 83-year-old woman, both with denovo intracranial WHO Grade I meningiomas, as well as allpatients with total resection, are alive and have experiencedno tumor regrowth. One patient had convulsions, and two haddeep vein thrombosis. No patient operated on with the pteri-onal approach was lost to follow-up.

Fronto-orbital Approach

Adding basal osteotomy to frontal, bifrontal, or pterionalcraniotomy provides a lower angle of attack and eliminatesthe need for brain retraction. Al-Mefty (1) proposed wideningbifrontal craniotomy with unilateral orbital rim osteotomy.Sekhar et al. (40) perform frontotemporal craniotomy and

TABLE 5. Tumor size and surgical approach

Size (cm) BifrontalUnilateralsubfrontal

PterionalFronto-orbital

Subcranial Total (%)

�2 1 2 3 (3.7%)

3–4 17 5 11 1 3 37 (45.7%)

5–6 18 2 5 4 3 32 (39.5%)

�6 2 1 2 4 9 (11.1%)

Total surgeries 35 9 18 7 12 81 (100%)

TABLE 6. Complications according to surgical approacha

ApproachNo. of

patientsCSFleak

Meningitis Bleeding Convulsion DVT PE Reoperation Death

Bifrontal 35 5 3 2b 1 3 1 2c 1d

Unilateralsubfrontal

9 1 1 1e

Pterional 18 1 2

Fronto-orbital 7 1 2f 1g

Subcranial 12 3 1 1

Totalsurgeries

81 10 4 4 3 5 2 4 1

a CSF, cerebrospinal fluid; DVT, deep vein thrombosis; PE, pulmonary embolus.b Two patients experienced hemorrhage in the surgical site.c One patient was reoperated for hemorrhage in the surgical site, one for persistent rhinorrhea.d In one patient, persistent rhinorrhea led to meningitis and death 6 months after surgery.e One patient reoperated for persistent rhinorrhea.f One patient developed a remote, contralateral, fontal parietal, epidural hematoma, and one experienced hemorrhagein the surgical site. The second patient experienced coagulopathy secondary to hepatic cirrhosis.g The contralateral epidermal hematoma was evacuated.



half-frontal craniotomy, and then add an orbital osteotomyjust across the midline. Babu et al. (3) combine frontal crani-otomy with orbital osteotomy.

We perform fronto-orbital craniotomy as described byDelashaw et al. (7), removing a bone flap that includes theorbital rim and frontozygomatic process of the frontal bone ina single piece. We modify this approach by placing the pos-terior border of the craniotomy further back, as in frontotem-poral (pterional) craniotomy.

We found this approach tobe the most suitable for resec-tion of large, high-riding me-ningiomas. It has many ad-vantages, including benefits ofboth the subfrontal and pteri-onal approaches. Orbital os-teotomy brings the operatorimmediately to the tumor in-sertion, and the pterional com-ponent of the approach en-ables early identification of theipsilateral optic nerve and con-trol over the internal carotidartery. Orbital roof removaland frontozygomatic osteot-omy provide a direct angle ofattack and improved tumorexposure, as proven quantita-tively by Schwartz et al. (39).This approach offers the short-est working distance to the tu-

mor, even in comparison with the pterional approach, and elim-inates the need for brain retraction. As with pterionalcraniotomy, the basal cisterns are opened and the brain is relaxedbefore tumor manipulation (Fig. 4). After tumor detachment anddebulking, the upper posterior aspect of the tumor, usually in-volving the anterior communicating artery complex and hypo-thalamus, is easily accessed, as demonstrated by Schwartz et al.The anterior cerebral arteries can be manipulated without brainretraction because of the lower attack angle.

This approach provides the best conditions for brain pres-ervation without retraction and meticulous dissection. If nec-essary, the planum sphenoidale and tuberculum sellae may bedrilled and the optic nerves unroofed. After tumor resection,there is much room for treatment of the anterior cranial basetumor insertion area. Reconstruction and closure are easy andnot time consuming. This is now our approach of choice forsurgery of large OGMs.

The approach has disadvantages. It is more time consum-ing. The frontal sinus is frequently opened wide, necessitatingits obliteration with a pericranial flap. However, when thesesinuses are small, fronto-orbital osteotomy lateral to the su-praorbital notch may obviate frontal sinus penetration.

We performed seven surgeries (8.6%) in patients with largetumors using this approach. Six patients had optic nerve in-volvement and five had visual impairment before surgery. Weachieved total resection in all seven patients. Six of sevenpatients are alive and have had no recurrence. One womanoperated on at age 73 years died later with no evidence ofrecurrence. In one patient, CSF leak developed through theethmoidal sinus, which was opened during extensive cranialbase drilling. A pericranial flap laid over the defect did notprevent this complication. The CSF leak was cured by severaldays of spinal drainage. Two patients experienced hemor-rhage. One of them had a contralateral epidural hematoma

TABLE 7. Outcome according to surgical approach

ApproachTotal

resectionSubtotalresection

RecurrenceResidualregrowth

Medianpostoperative

Karnofsky score

Bifrontal 31 4 1a 90

Unilateralsubfrontal

7 2 1a 90

Pterional 15 3 1 100

Fronto-orbital 7 90

Subcranial 12 100

Total 72 9b 0 3c 90

a Olfactory groove meningioma regrew twice in this multiple meningiomata patient: once after initial bifrontal removal,and again after a second attempt at removal via a subfrontal approach.b Includes two subtotal resections in two surgeries for one multiple meningiomata patient.c Includes residual regrowth twice after two surgeries in one multiple meningiomata patient.

TABLE 8. Surgical mortality in published olfactory groovemeningioma series

Series (ref. no.)No. of

patientsOperativemortality

Cushing and Eisenhardt, 1938 (6) 22 5 (22.7%)

Solero et al., 1983 (43) 98 17 (17.3%)

Ojemann, 1991 (31) 17 1 (5.9%)

Ransohoff and Nockels, 1993 (35) 33 2 (6.1%)

Schaller et al., 1994 (38) 28 1 (3.6%)

Tsikoudas and Martin-Hirsch, 1999 (45) 13 2 (15%)

Turazzi et al., 1999 (46) 37 1 (2.7%)

Hentschel and DeMonte, 2003 (18) 13 0

Obeid and Al-Mefty, 2003 (29) 15 0

Present study 80 0a

a In one patient, persistent rhinorrhea led to meningitis and death 6 monthsafter surgery.

SPEKTOR ET AL.


that was evacuated surgically. The second patient, who hadcirrhosis and coagulopathy, experienced a hemorrhage in thetumor bed that dissolved without surgical intervention. Nopatient operated on with this approach was lost to follow-up.

Subcranial Approach

This approach, described elsewhere (13, 36), was developed forremoval of invasive cranial base tumors, mainly by ENT surgeons.We applied the subcranial approach for resection of 12 OGMs.

The advantage to this approach is wide exposure of the intra-cranial anterior cranial base compartment in continuity withneighboring structures, including the frontal, sphenoid, and eth-moidal sinuses, the orbits, and the nasal cavity. The optic canalmay be drilled, exposing the optic nerves and removing thetumor from the optic canals. This access provides optimum con-ditions for radical tumor removal (Figs. 5 and 6). The disadvan-tage of the approach is its greater technical difficulty, includingcomplex reconstruction. Both the approach and reconstructionare time consuming. A spinal drain is obligatory to avoid per-manent postoperative CSF leak.

This approach is not ourfirst choice. We use it only forpatients with primary or re-current tumors invading theparanasal sinuses and/or theorbits. In this series, 12 oper-ations were performed withthe subcranial approach. Sixpatients had been operatedon elsewhere through sub-frontal, frontal, or pterionalapproaches from 1 to 6 yearsearlier (average, 3.5 yr). All12 patients had meningiomaspenetrating into ethmoidaland/or frontal sinuses. Inone patient, the tumor in-volved orbits and optic ca-nals, necessitating extensive

drilling. Total resection was achieved in all 12 procedures. Allpatients initially were managed with postoperative spinaldrain. Three had significant CSF leak during the first daysafter surgery that demanded some extra period of spinaldrain. One patient experienced meningitis, and one had con-vulsions. All patients operated on with this approach are alivewith no clinical or imaging evidence of OGM recurrence.

Choosing the Optimal Approach andRecurrence Prevention

Technical developments and improved microsurgical tech-niques have significantly reduced mortality rates after OGMresection from 17.3 to 22.7% (6, 43), to nearly 0% (46) (Table 8).Current debate regarding surgical strategy and tactics focuseson the extent of tumor resection and the optimal approach.The ideal, most radical resection is Simpson Grade I, whichdemands bone drilling beneath the tumor insertion (42). His-torically, some authors (22, 31, 32, 35) weighed the high risk ofCSF leak, potentially complicated with meningitis, against thepotential for recurrence after radical anterior cranial base dril-

TABLE 9. Resection and rates of recurrence in published olfactory groove meningioma series

Series (ref. no.)No. of

patients% total

resectionFollow-up, yr (range)

%recurrence

Mirimanoff et al., 1985 (25)a 22 77.3 5 30

10 41

Tsikoudas and Martin-Hirsch, 1999 (45) 13 100 Approximately 10 (1–22) 30.8

Turazzi et al., 1999 (46) 37 100 Average, 4 (1–8) 0

Hentschel and DeMonte, 2003 (18) 13 84.6 Average, 2 (0–5) 0

Obeid and Al-Mefty, 2003 (29) 15 93.3 Average, 3.7 (1–7.3) 0

Present study 80 90 Average, 5.9 (0.5–13.7) 2.5

a Mirimanoff et al. reported recurrance rates separately for patients who were followed for 5 and 10 years after surgery.

TABLE 10. Comparison of surgical approachesa

Paranasalsinus

transectionBrain retraction

SSSligation

CSF leakSurgical field

exposureComplexity

Timeconsuming

Approach

Yes Significant Yes High risk Good Moderate �� Bifrontal

Yes Very significant No Moderate risk Poor Low � Unilateral subfrontal

No Moderate No Low risk Good Low � Pterional

Yes Slight No Moderate risk Very good Moderate �� Fronto-orbital

Yes Slight No High risk Very good High �� Subcranial

a SSS, superior sagittal sinus; CSF, cerebrospinal fluid; �, relative amount of time.



ling and recommended only “prudent” drilling of the cranialbase. Indeed, the rate of OGM recurrence is relatively low insome reports (31, 35, 43). However, other studies demon-strated OGM recurrence rates of up to 30% over 10 years inpatients whose tumor is not radically removed (45). Miri-manoff et al. (25) reported 30% recurrence rates for OGMmeningiomas at 5 years and 41% at 10 years after surgery(Table 9).

In recent publications, there is an obvious trend towardmore aggressive tumor resection. Obeid and Al-Mefty (29)and Hentschel and DeMonte (18) advocate radical tumor re-section, including drilling of hyperostotic bone and pursuingtumor extending into the paranasal sinuses, with consequentmeticulous cranial base reconstruction. We strongly supportthis concept.

DeMonte (8) notes that at least 15% of OGMs grow into theethmoidal sinuses. In our series, 21 patients (26.3%) demon-strated extension to the paranasal sinuses, including 6 patientswith both paranasal sinus and optic nerve involvement. Mor-ris et al. (28) stress that extracranial spread of OGM should notbe underappreciated, because managing regrowth of residualdisease is difficult, and suggest radical resection in the firstinstance. All four of their patients sought treatment at the ENTservice.

Nine patients in our series were operated on elsewhere andhad recurrent tumors at presentation. We observed similarregrowth patterns. Recurrent tumor originating from the pe-riphery of the original tumor typically involved the falx andanterior cranial base—the cribriform plate, planum sphenoi-dal, and tuberculum sellae, expanding downward into theethmoidal sinuses—and sometimes invaded the dura of theoptic nerves (Figs. 5 and 6).

Some patients who develop recurrent tumor expanding intothe paranasal sinuses or orbits may be referred to ENT sur-geons and be lost to neurosurgical follow-up. In our series,ENT surgeons diagnosed six patients with recurrent tumorinvolving the ethmoidal, sphenoidal, and frontal sinuses, na-sal cavity, orbits, and even the optic canals. All had been

operated on elsewhere by neurosurgeons. Several of thesepatients were definitely informed by their surgeons that theirtumors had been totally removed. The cases of these patientsdemanded a subcranial approach with extensive anterior cra-nial base drilling, which was not performed during primaryremoval. Reoperation was definitely more challenging be-cause of scars and adhesions from the earlier procedures, aswell as the need to unroof and manipulate the optic nervesand to perform extensive cranial base reconstruction. We can-not accept that “prudent” drilling of the anterior cranial baseis sufficient to prevent tumor recurrence.

Several approaches with different levels of complexity havebeen proposed for the surgical treatment of meningiomas withextensive anterior cranial base involvement. Persky and Som(34) suggested a combination frontal craniotomy and lateral

FIGURE 3. Illustrations of the pterional approach. A, T1-weightedgadolinium-enhanced axial MRI scan demonstrating a large OGM. B, aretractor (R) has been introduced beneath the frontal lobe and the parachi-asmatic cistern has been opened, providing excellent visualization of thecaudal part of the tumor (T), the right olfactory tract (OT), the opticnerve (ON), and internal carotid artery (ICA).

FIGURE 4. Illustrations of the fronto-orbital approach. A and B,T1-weighted gadolinium-enhanced coronal (A) and sagittal (B) MRI scansdemonstrating a large OGM. C, intraoperative photograph of right-sided,fronto-orbital osteotomy. D, macroscopic view after a right-sided, fronto-orbital approach has been performed and the dura mater opened. Note per-fect brain relaxation. With the naked eye, and without retraction, the rightoptic nerve is visualized perfectly at the posterior end of the tumor(arrow). FL, frontal lobe; T, tumor; BD, basal dura mater. E and F, post-operative T1-weighted gadolinium-enhanced (E) coronal and (F) sagittalMRI scans demonstrating radical tumor removal.

SPEKTOR ET AL.


rhinotomy. Maiuri et al. (23) used combined subfrontal andnasal approaches. Goffin et al. (15) selected craniofacial resec-tion. A telecanthal approach was proposed by Fujitsu et al.(14), and Hallacq et al. (16) operated using a frontal sinusapproach. Ohata et al. (30) suggested a bi-orbitofrontobasalinterhemispheric approach.

We applied the subcranial approach in 12 patients andconsistently achieved excellent tumor exposure. All 12 hadextensive paranasal sinus and/or orbital involvement. Six hadlocal recurrence secondary to the lack of meticulous tumorremoval from all surrounding structures. Tumor has not re-curred in these patients after their second surgery. Thesepatients have been followed for 11 to 105 months (average,62.3 mo) after surgery. Moore et al. (26) also concluded thatthe subcranial approach is preferable to traditional frontalcraniotomy combined with lateral rhinotomy. We did not findmore extensive approaches necessary.

The choice of approach is clearly important and probablycontributes to the near zero mortality and reduced morbidityrates reported in some recent papers describing experiencewith the pterional approach (17, 46). However, we do notbelieve that the pterional approach is entirely responsible forimproved outcomes with these patients compared with earlierreports. Certainly widespread improvements in surgical tech-niques and instrumentation also contributed significantly tothese excellent results.

We present here the largest OGM series with the longestfollow-up published in modern neurosurgical literature basedon an integrated, multidisciplinary approach to OGM patientmanagement and using a variety of surgical techniques. Wehave achieved good clinical results—zero operative mortality,low morbidity, and very low recurrence rates. Based on our

FIGURE 5. Illustrations of the subcranial approach. A–D, T1-weightedgadolinium-enhanced sagittal (A), coronal (B), and axial (C and D) MRI scansdemonstrating recurrent OGM invading the ethmoidal sinus and right orbit. E–G,intraoperative macroscopic pictures of the subcranial approach. H–K, T1-weightedgadolinium-enhanced sagittal (H), coronal (I), and axial (J and K) MRI scansdemonstrating absence of the tumor and anterior cranial base reconstruction with thefascia lata (H, arrow).

FIGURE 6. Images from a patient who hadundergone surgery for OGM via a right-sided pterional craniotomy 5 years earlier. A,intraoperative macroscopic picture at an earlystep of a subcranial approach. B and C, T1-weighted gadolinium-enhancedcoronal MRI scans. Note tumor spread along the falx and ethmoidal sinusinvasion.



experience with multiple surgical approaches, we tried tointroduce an algorithm facilitating the choice of an optimalapproach for OGM removal in each situation (Table 10).

We find that the pterional, fronto-orbital, and subcranialapproaches generally are the most effective. For small tumorsthat are purely intracranial and do not involve the anteriorcranial base, a regular pterional approach is sufficient. Largeand very large high-riding OGMs are better removed via thefronto-orbital approach, because swollen brain pushes suchtumors downward as soon as part of the anterior cranial baseis removed. The fronto-orbital approach also enables sufficientresection of meningiomas with moderate anterior cranial baseinvasion. Tumors that have significantly infiltrated the ante-rior cranial base, invaded ethmoidal and sphenoid sinuses,and extended into the nasal cavity and orbits are best removedthrough the subcranial approach. The goal of OGM surgeryshould be radical resection. Hyperostotic bone and sinus in-vasion are not limitations.

CONCLUSION

OGMs should be approached with the techniques and atti-tudes applied to cranial base tumors. Modern microsurgicaltechniques and careful choice of surgical approach help toachieve excellent results in the surgical treatment of OGMs.

REFERENCES

1. Al-Mefty O: Tuberculum sellae and olfactory groove meningioma, in SekharLN, Janecka IP (eds): Surgery of Cranial Base Tumors. New York, Raven Press,1993, pp 507–519.

2. Al-Mefty O, Smith R: Tuberculum sellae meningiomas, in Al-Mefty O (ed):Meningiomas. New York, Raven Press, 1991, pp xix, 395–411, 630.

3. Babu R, Barton A, Kasoff SS: Resection of olfactory groove meningiomas:Technical note revisited. Surg Neurol 44:567–572, 1995.

4. Bakay L, Cares HL: Olfactory meningiomas: Report on a series of twenty-five cases. Acta Neurochir (Wien) 26:1–12, 1972.

5. Bricolo A: Commentary on Babu R, Barton A, Kasoff SS. Resection ofolfactory groove meningiomas: Technical note revisited. Surg Neurol 44:567–572, 1995.

6. Cushing H, Eisenhardt L: The olfactory meningiomas with primary anos-mia, in Cushing H, Eisenhardt L (eds): Meningiomas: Their Classification,Regional Behavior, Life History, and Surgical Results. Springfield, Charles CThomas, 1938, pp 250–282.

7. Delashaw JB Jr, Jane JA, Kassell NF, Luce C: Supraorbital craniotomy byfracture of the anterior orbital roof: Technical note. J Neurosurg 79:615–618,1993.

8. DeMonte F: Surgical treatment of anterior basal meningiomas. J Neurooncol29:239–248, 1996.

9. Derome PJ, Guiot G: Bone problems in meningiomas invading the base ofthe skull. Clin Neurosurg 25:435–451, 1978.

10. El Gindi S: Olfactory groove meningioma: Surgical techniques and pitfalls.Surg Neurol 54:415–417, 2000.

11. Fliss DM: Skull base reconstruction after anterior subcranial tumor resec-tion. Neurosurg Focus 12(5):Article 10, 2002.

12. Fliss D, Zucker G, Amir A, Gatot A, Cohen M, Spektor S: The subcranialapproach for anterior skull base tumors. Otolaryngol Head Neck Surg11:238–253, 2000.

13. Fliss DM, Zucker G, Cohen A, Amir A, Sagi A, Rosenberg L, Leiberman A,Gatot A, Reichenthal E: Early outcome and complications of the extendedsubcranial approach to the anterior skull base. Laryngoscope 109:153–160,1999.

14. Fujitsu K, Saijoh M, Aoki F, Sakata K, Fujii S, Mochimatsu Y, Kuwabara T:Telecanthal approach for meningiomas in the ethmoid and sphenoid si-nuses. Neurosurgery 28:714–720, 1991.

15. Goffin J, Fossion E, Plets C, Mommaerts M, Vrielinck L: Craniofacial resec-tion for anterior skull base tumours. Acta Neurochir (Wien) 110:33–37, 1991.

16. Hallacq P, Moreau JJ, Fischer G, Beziat JL: Frontal sinus approach to olfac-tory groove meningiomas [in French]. Neurochirurgie 45:329–337, 1999.

17. Hassler W, Zentner J: Pterional approach for surgical treatment of olfactorygroove meningiomas. Neurosurgery 25:942–947, 1989.

18. Hentschel S, DeMonte F: Olfactory groove meningiomas. Neurosurg Focus14(6):Article 4, 2003.

19. Jallo GI, Benjamin V: Tuberculum sellae meningiomas: Microsurgical anat-omy and surgical technique. Neurosurgery 51:1432–1440, 2002.

20. Jho HD, Alfieri A: Endoscopic glabellar approach to the anterior skull base:A technical note. Minim Invasive Neurosurg 45:185–188, 2002.

21. Kawakami K, Yamanouchi Y, Kawamura Y, Matsumura H: Operative ap-proach to the frontal skull base: Extensive transbasal approach. Neuro-surgery 28:720–725, 1991.

22. Lagares A, Lobato RD, Castro S, Alday R, De la Lama A, Alen JF, GonzalezP: Meningioma of the olfactory groove: review of a series of 27 cases [inSpanish]. Neurocirugia (Astur) 12:17–22, 2001.

23. Maiuri F, Salzano FA, Motta S, Colella G, Sardo L: Olfactory groove menin-gioma with paranasal sinus and nasal cavity extension: removal by com-bined subfrontal and nasal approach. J Craniomaxillofac Surg 26:314–317,1998.

24. Mayfrank L, Gilsbach JM: Interhemispheric approach for microsurgicalremoval of olfactory groove meningiomas. Br J Neurosurg 10:541–545, 1996.

25. Mirimanoff RO, Dosoretz DE, Linggood RM, Ojemann RG, Martuza RL:Meningioma: Analysis of recurrence and progression following neurosur-gical resection. J Neurosurg 62:18–24, 1985.

26. Moore CE, Ross DA, Marentette LJ: Subcranial approach to tumors of theanterior cranial base: Analysis of current and traditional surgical techniques.Otolaryngol Head Neck Surg 120:387–390, 1999.

27. Morley T: Tumors of the cranial meninges, in Youmans J (ed): NeurologicalSurgery. Philadelphia, W.B. Saunders, 1973, pp 1388–1411.

28. Morris KM, Campbell D, Stell PM, MacKenzie I, Miles JB: Meningiomaspresenting with paranasal sinus involvement. Br J Neurosurg 4:511–515,1990.

29. Obeid F, Al-Mefty O: Recurrence of olfactory groove meningiomas. Neuro-surgery 53:534–543, 2003.

30. Ohata K, Hakuba A, Nagai K, Morino M, Iwa Y: A biorbitofrontobasalinterhemispheric approach for suprasellar lesions. Mt Sinai J Med 64:217–221, 1997.

31. Ojemann RG: Olfactory groove meningiomas, in Al-Mefty O (ed): Meningi-omas. New York, Raven Press, 1991, pp 383–392.

32. Ojemann RG, Swan K: Surgical management of olfactory groove, suprasellarand medial sphenoid ridge meningiomas, in Schmidek HH, Sweet WH(eds): Operative Neurosurgical Techniques. Orlando, Grune & Stratton, 1988,pp 531–545.

33. Paterniti S, Fiore P, Levita A, La Camera A, Cambria S: Venous saving inolfactory meningioma’s surgery. Clin Neurol Neurosurg 101:235–237, 1999.

34. Persky MS, Som ML: Olfactory groove meningioma with paranasal sinusand nasal cavity extension: A combined approach. Otolaryngology 86:ORL-714–ORL-720, 1978.

35. Ransohoff J, Nockels R: Olfactory groove and planum meningiomas, inApuzzo M (ed): Brain Surgery: Complication Avoidance and Management. NewYork, Churchill Livingstone, 1993, pp 177–185.

36. Raveh J, Laedrach K, Speiser M, Chen J, Vuillemin T, Seiler R, Ebeling U,Leibinger K: The subcranial approach for fronto-orbital and anteroposteriorskull-base tumors. Arch Otolaryngol Head Neck Surg 119:385–393, 1993.

37. Samii M, Ammirati M: Olfactory groove meningiomas, in Samii M (ed):Surgery of the Skull Base: Meningiomas. Berlin, Springer-Verlag, 1992, pp15–25.

38. Schaller C, Veit R, Hassler W: Microsurgical removal of olfactory groovemeningiomas via the pterional approach. Skull Base Surg 4:189–192, 1994.

SPEKTOR ET AL.


39. Schwartz MS, Anderson GJ, Horgan MA, Kellogg JX, McMenomey SO,Delashaw JB Jr: Quantification of increased exposure resulting from orbitalrim and orbitozygomatic osteotomy via the frontotemporal transsylvianapproach. J Neurosurg 91:1020–1026, 1999.

40. Sekhar LN, Nanda A, Sen CN, Snyderman CN, Janecka IP: The extendedfrontal approach to tumors of the anterior, middle, and posterior skull base.J Neurosurg 76:198–206, 1992.

41. Sekhar LN, Tzortzidis F: Resection of tumors by the fronto-orbital approach,in Sekhar LN, de Oliveira EP (eds): Cranial Microsurgery: Approaches andTechniques. New York, Thieme, 1999, pp 61–75.

42. Simpson D: The recurrence of intracranial meningioma after surgical treat-ment. J Neurol Neurosurg 20:22–39, 1957.

43. Solero CL, Giombini S, Morello G: Suprasellar and olfactory meningiomas: Reporton a series of 153 personal cases. Acta Neurochir (Wien) 67:181–194, 1983.

44. Symon L: Olfactory groove and suprasellar meningioma, in Krayenbühl H(ed): Advances and Technical Standards in Neurosurgery. New York, Springer-Verlag, 1983, p 67.

45. Tsikoudas A, Martin-Hirsch DP: Olfactory groove meningiomas. ClinOtolaryngol 24:507–509, 1999.

46. Turazzi S, Cristofori L, Gambin R, Bricolo A: The pterional approach for themicrosurgical removal of olfactory groove meningiomas. Neurosurgery45:821–826, 1999.

47. Yasargil MG: Microneurosurgery. New York, Thieme Stratton, 1984.

AcknowledgmentWe thank Shifra Fraifeld for her help in preparing this manuscript.

COMMENTS

This is a retrospective review of 80 olfactory groove meningiomas(OGM) operated on over a period of approximately 13 years. The

authors describe their surgical technique and outcomes. Close toone-half of the operations were performed using a bifrontal craniot-omy approach. The other surgical approaches included a unilateralfrontal craniotomy, a parietal craniotomy, a fronto-orbital-zygomaticcraniotomy, and a subcranial cranial base approach. The authorsdiscuss advantages and disadvantages of each approach.

Although most of the operations were performed using the bifrontalcraniotomy approach, the authors do not seem to advocate this approach.Instead, they favor the pterional approach, with or without the additionof the fronto-orbital-zygomatic cranial base dissection. They base theirrecommendation on several factors, including avoidance of brain retrac-tion, access to the basal cisterns, the potential for a cerebrospinal fluid(CSF) leak, and the complexity of cranial base reconstruction. The authorsclaim no neurological morbidity other than anosmia in any of their 80cases. On the other hand, they report an approximately 30% operativemorbidity (mostly consisting of CSF leakage, and a few cases of menin-gitis, seizures, deep vein thrombosis, etc.).

I fully agree with the authors that the choice of the operativeapproach should be tailored to the tumor location, size (high riding ornot), relationship to the carotid arteries and its branches, presence orabsence of significant preoperative frontal lobe vasogenic edemaand/or visual deficits, and many other factors.

The principles for removal of OGMs in our practice have been verysimilar to those of other cranial base operations: an exposure thatwould bring the surgeon close to the tumor, early release of CSF so asto allow for brain relaxation, avoidance of brain retraction, meticulouspreservation of all arterial and venous channels not belonging to thetumor capsule, adequate internal decompression of the tumor beforethe tumor capsule is separated from the surrounding neuro-vascularstructures along the tumor capsule-arachnoid interface, meticulousclosure of the open sinus cavities so as to prevent a postoperative CSF

rhinorrhea, and seizure prophylaxis that is usually begunintraoperatively.

Smaller tumors can be removed using the pterional approach. Inlarger tumors, we prefer to have an additional midline access as well.This makes it possible to divide the anterior falx if necessary. Thereason we like to have midline access for larger tumors is that theipsilateral optic nerve may conceal part of the tumor behind it, requir-ing perhaps undue surgical manipulation of the optic nerve to visu-alize and free this tumor portion. In addition, in larger tumors, thecraniotomy is augmented with a fronto-orbital-clinoidal extraduralcranial base dissection. Such a dissection that includes removal of thesupraorbital rim brings the surgeon even closer to the tumor andaffords additional maneuvering space without the need for brainretraction. Our technique of sinus closure includes either exenterationof the frontal sinus mucosa or stripping away and collapsing it towardthe frontonasal canal, packing the sinus with an autologous fat graftand using a vascularized pericranial flap to seal off the frontal sinusesfrom the operative site by tacking the pericranial flap to the underly-ing dura with a surface- to- surface suture technique.

As far as accomplishing a complete tumor removal (Simpson 1) wetend to strike a balance between the degree of cranial base removaland the potential for postoperative complications, especially a CSFrhinorrhea. Our recurrence rate after the removal of OGMs has beenlower compared with meningiomas in other locations.

Ivan S. CiricEvanston, Illinois

The authors present what could be one of the largest series of OGMstreated using different surgical approaches chosen on the basis of

the characteristics of the tumors. The criteria they use in their deci-sions are acceptable and are in line with what is found in the litera-ture. The results achieved are good, and one may have the impressionof an accurate and competent management of this pathology. In thearticle, the authors show their preference toward the bilateral subfron-tal approach in agreement with a consolidated tendency; and themajority of their patients are operated on with this approach. Thesame authors emphasize some limits of the pterional approach, whichin our hands (1) we do not encounter, such as the difficulty in drillingthe osteoma at the olfactory groove and in controlling the controlateraltumor expansion. The pterional approach allows a safer dissection ofthe optic nerves and of the anterior arteries of the Willis Circle as wellas the frontobasal ones, with little or no damage to the basal frontallobe as stated by Yasargil’s comment on our above-mentioned article(1). The problem of recurrences from the ethmoidal plane, which ismanaged in cooperation with ear, nose, and throat surgeons, is com-mented on exhaustively, whereas posterior recurrences in the para-sellar area, which are often difficult to treat, are never mentioned. Noris the possibility of preserving the olfaction mentioned, which maygive great satisfaction using the pterional approach. In conclusion, thearticle presents a large caseload that is well treated and managed,following present trends.

Sergio TurazziAlbino BricoloVerona, Italy

1. Turazzi A, Cristofori L, Gambin R, Bricolo A: The pterional approach for themicrosurgical removal of olfactory groove meningiomas. Neurosurgery 45:821–826, 1999.



The authors report their experience during a 13-year period with 80patients who underwent a surgical treatment (81 OGM surgeries).

This well-written article describes the different techniques for access tomeningiomas in the olfactory grove. The authors review the potentialadvantages and disadvantages associated with the use of determinedsurgical approach to OGMs, as well as outcomes and recurrence rates. Inthis series, the main approaches were through bifrontal craniotomy,unilateral subfrontal approach, pterional approach with fronto-orbitalcraniotomy, and subcranial approach. Total removal was obtained in90% of the patients, subtotal in 10%. No operative mortality and no newpermanent focal neurological deficit other than anosmia were identifiedin this series. However, 31.3% developed surgery-related complications.These patients’ outcomes prove that proper use of modern microsurgicaltechniques leads to excellent results.

Given the increasing number of operations on the anterior fossa inpatients with OGMs, the selection of the most appropriate approach isespecially important. Previous publications have described many surgi-cal managements of these tumors. For patients with large and smallOGMs, we prefer the extended frontopterional approach. This approachmay accomplish complete removal and the preservation of vascularstructures surrounding the tumor. During the craniotomy, care is takenat the frontobasal anterior extension to reach the midline (falx and cris-ta).The dissection of the Sylvian fissure is very important and is followedby opening the carotid and interpeduncular cisterns. These proceduresallow the release of CSF and relaxation of the brain. The pterionalapproach also allows early visualization of the neurovascular structuresand their dissection. Control of the posterior part of the tumor in relationto the most important structures (optic nerves and internal carotid artery)may be obtained by this approach. Dissection of the anterior cerebralarteries from the capsule may prove difficult because of adherence orencasement. The artery recurrent of Heubner should always be pre-served during the dissection. A bifrontal approach is more direct to thetumors. However, in our opinion the late visualization of the optic nervesand the anterior cerebral complex are disadvantageous points to beconsidered in the choice of this approach.

Evandro P. de OliveiraSão Paulo, Brazil

Spector et al. have reported a large series of OGMs operated on overa 13-year period with excellent results. They used a variety of

approaches to these tumors. The patients were not selected for theap-proaches in any prospective fashion. However, their comments re-garding the operative approaches are useful.

I agree that for small tumors, a small pterional craniotomy, with anorbitotomy is adequate. For larger tumors, I use an unilateral fronto-tempororal craniotomy, with unilateral orbitotomy. For giant-sizedtumors, I prefer a frontotemporal craniotomy which extends justacross the midline, with an orbitotomy which also extends across themidline. In smaller tumors, preservation of olfaction, at least on thecontralateral side, is possible, and preferable. For the larger tumors,layered cranial base reconstruction is very important to avoid CSFleakage. The surgeon should be aware of where these tumors tend torecur, especially near the optic canal, in the anterior fossa bones, andin the ethmoid sinus. Particularly in younger patients, these areasshould receive careful attention during the resection.

More recently, some surgeons have been using an endoscopic ap-proach to the resection of these tumors. While it has the advantage ofno brain retraction, the various critical structures are seen at the endof the operation, or not very well, and if there is a vascular injury, itwould be very difficult to manage by the endoscopic approach. CSFleakage has been a problem for the endoscopic approaches. Thesesurgeons seem to be performing mostly partial tumor resection andradiosurgery for the larger tumors, and have not published any sur-gical series of resection and follow-up. For the present, such endo-scopic approaches may remain primarily in the hands of selectedsurgeons, because of safety and efficacy issues.

Laligam N. SekharSeattle, Washington

This is a pretty large series of OGMs (N�80) treated by differentsurgical approaches. The results are excellent with no mortality

after complete removal in 90% of the cases. Certainly, the best treat-ment is radical resection which is better achieved through a subfrontalapproach with deposition of the medial part of the fronto-orbitalridge. This approach is useful in resecting completely the zone ofinsertion including the bone and, when it exists, the ethmoidonasalextension of the meningioma. This is not possible with the pterionalapproach and is difficult with a single standard of subfrontal ap-proach. This article nicely reviews the advantages and disadvantagesof each technique and the complications (mostly CSF leak) related tothem. A better assessment of the benefits of each approach could havebeen done with cognitive and psychometric tests.

Bernard GeorgeParis, France

SPEKTOR ET AL.


Documents

Neurosurgery_TechChir_2006.pdf