Neurosurg Rev DOI 10.1007/s10143-014-0517-y
ORIGINAL ARTICLE
Surgical management of ocular symptoms in spheno-orbital meningiomas. Is orbital reconstruction really necessary? Andrea Talacchi & Antonella De Carlo & Antonio D’Agostino & Pierfrancesco Nocini
Received: 12 March 2013 / Revised: 22 August 2013 / Accepted: 27 October 2013 # Springer-Verlag Berlin Heidelberg 2014
Abstract Spheno-orbital meningioma (SOM) is an intriguing tumor because of the many different factors that can influence clinical and oncological outcome after treatment. Reasoning that outcome indicator measurement is key to improving therapy, we retrospectively evaluated the management of proptosis and other ocular symptoms in 47 patients surgically treated for SOM at our department in the last 10 years. This patient series was characterized by a high rate of tumor infiltration of the extradural cranial base. Clinical outcome was assessed by comparing preoperative and postoperative ophthalmological and neurological signs. Acute postoperative complications were reported, and clinical and radiological outcome was assessed at 4–6 months, 12 months, and the last follow-up. Proptosis (measured by Hertel exophthalmometry), visual acuity, visual field defect (measured by Goldmann perimetry), diplopia (measured by the Hess-Lancaster test), and other disturbances were rated as normalized, improved, or unchanged/ worsened. The most common presenting symptoms were proptosis (95.7 %), visual impairment (51 %), and cranial nerve deficit (38.2 %). Surgery via the frontotemporal approach was performed in all 47 cases, with the primary aim to relieve symptoms/signs and maximize tumor resection. Bony orbital reconstruction was never performed. Complete resection was A. Talacchi : A. De Carlo Section of Neurosurgery, Department of Neurological, Neuropsychological, Morphological and Movement Sciences, University of Verona, University Hospital, Verona, Italy A. D’Agostino : P. Nocini Section of Dentistry and Maxillo-Facial Surgery, Department of Surgical Science, University of Verona, University Hospital, Verona, Italy A. Talacchi (*) Institute of Neurosurgery, Department of Neurosciences, University of Verona, University Hospital, P. Stefani 1, 37126 Verona, Italy e-mail: [email protected]
achieved in 51 % of cases (Simpson grades I and II) with minimal morbidity. At a mean follow-up of 52 months (range, 12–112), proptosis normalized in 90.9 % and improved in the remaining patients, visual acuity normalized in 20.8 % and improved in 45.8 % patients, cranial nerve deficit subsided in all but two cases. The recurrence rate was 29.7 %. One of the gold standards of surgical treatment, normalization of proptosis, can be achieved by accurate resection of the superior and lateral orbital walls. In this setting, careful reconstruction of the frontobasal dura is far superior to bony reconstruction. Complete tumor resection should not be pursued at the expense of increased morbidity. Keywords Surgical management . Meningioma . Skull base . Orbital tumors . Proptosis
Introduction Spheno-orbital meningioma (SOM), the modern term for Cushing group’s “hyperostosing meningioma of ala magna,” was introduced in the computed tomography (CT) era [2, 4, 7] and then better defined in the magnetic resonance imaging (MRI) era by the majority of studies that described tumor extension and modality of treatment [9, 14]. The surgical technique involves many steps, each of which is tailored case-by-case in a parsimonious approach to maximizing tumor removal. To this end, outcome measurement can provide feedback information for surgical planning. In the treatment of SOM, patient-oriented outcome may best be centered on disturbances in visual and ocular function, by far the most common clinical symptoms. However, reported outcome, ophthalmologic and neurological evaluation, and performance status may underestimate the real outcome, as referred by patients concerned about posttreatment function and cosmetic appearance [10]. Accordingly, the clinical priority is the
Neurosurg Rev
surgical management of proptosis, the main presenting symptom reported in all surgical series, caused by orbital bone thickening and orbital tumor infiltration. Improving and normalizing proptosis are the main clinical aims of surgical treatment, yet it persists in many, improves in most, and remains stable in roughly 20 % of cases [22]. Because preoperative and postoperative objective assessment has been rarely reported in the literature, the evaluation of treatment is limited, particularly with regard to restoration of facial symmetry and eye bulb position [20, 22]. This is one of the chief reasons for the continuing dilemma of the extent of bone removal and of whether to reconstruct the orbital walls. The aim of this study was to objectively assess eye bulb position before and after treatment, together with other ophthalmometric measures, in a surgical patient series in which orbital reconstruction was not performed.
Proptosis was evaluated by Hertel exophthalmometry, assuming an abnormal value to be ≥2 mm as compared with the healthy eye. The extent of surgical removal was assessed using postoperative MRI images and operative notes. The Simpson classification was adopted: grades I and II, with excision or coagulation of the dura, respectively, were defined as complete removal; grade III with persistent dural enhancement and grade IV with residual mass were defined as incomplete removal.
Material and methods
–
Meningiomas of the sphenoid bone with intraosseus tumor growth involving the orbit were defined as SOM. Out of the 118 SOMs operated on in the last 20 years (1992–2012), accounting for 7 % of all intracranial meningiomas (n= 1759) and 33 % of all sphenoid wing meningiomas (n=360) operated on at the Department of Neurosurgery, University Hospital of Verona, we reviewed the cases treated with a similar strategy (n=61) in the last 10 years and selected only those cases with complete data. We compared the symptoms, signs, radiographic features, microsurgical techniques, and outcome in 47 patients with primary tumor. Hospital charts, operative notes, follow-up clinical notes, radiographic study reports, and neuro-ophthalmological reports were retrospectively analyzed for each patient. Patients or relatives were contacted to obtain further follow-up data when necessary. All patients underwent CT, with thin slices (2 mm) and bone window scans for cranial base assessment, and MRI. Gadolinium-enhanced MRI images and fat suppression sequences were obtained to delineate the intradural, intraorbital, or extracranial tumor portion and to determine the extent of dural infiltration. The maximum tumor diameter was not reported because it is inconsistent in such irregular tumors.
–
Outcome measures
This long and tedious part of the operation starts with drilling the lesser sphenoid wing all the way down to the clinoid under microscope magnification. As a rule, the extent of orbital access is tailored to the individual patient and based on the evidence of tumor and/or on the clinical rationale for bony decompression. Great care is taken to avoid laceration of the periorbita and injury to the oculomotor muscles.
Clinical outcome was assessed by comparing preoperative and postoperative ophthalmological and neurological signs. Acute postoperative complications were reported, and clinical and radiological outcome was assessed at 4–6 months, 12 months, and the last follow-up. Proptosis, visual acuity, visual field defect (measured by Goldmann perimetry), diplopia (measured by the Hess-Lancaster test), and other disturbances were rated as normalized, improved, or unchanged/worsened.
Surgical management In light of our experience and a review of the literature, surgical management rests on the following assumptions: –
–
A surgical approach needs to be wide enough to follow the tumor. A circumferential view is needed to control the medial and lateral parts that are usually involved. The orbital rim and zygomatic bone are usually spared by tumor infiltration, except in very advanced stages. Not all the thickened bone is tumor. Infiltrated pockmarked-like bone should be differentiated from reactive ivory-like bone. This can be done with the help of the microscope. In doubtful cases, only bone decompression is performed (see the following item). Priority is given to the relief of disturbances over radical tumor removal.
Surgical approach The patient is positioned supine with the head turned not more than 20° to the opposite side and slightly tilted. A standard pterional approach is established with variable frontal or temporal extension according to the direction of tumor growth. After interfascial dissection, the temporal muscle is reflected downward and laterally to expose the pterional region, which is normally involved by hyperostosis. After craniotomy, the bone is lifted from the dura. Bone work
Orbital roof The anterior limit is given by the need to preserve the orbital rim. The medial limit is the ethmoidal sinus, but
Neurosurg Rev
usually, only the superolateral and the posterior parts of the roof are removed. Lateral orbital wall To avoid enophthalmos or to improve exophthalmos, this is the most important part of orbitotomy. The lateral orbital wall, because it is the site of extraconal tumor, is usually removed, up to the deepest part, anteriorly to the orbital profile, and inferiorly at the level of the inferior orbital fissure (at the bottom of the wall before it becomes the floor). Extending the drilling anteriorly will favor globe reduction (reentry into the orbital cone); limiting inferior drilling will avoid globe retraction (escape from the orbital cone). Optic canal-superior orbital fissure-and foramen rotundum triangle This is the most important part for visual and oculomotor management, since disturbances arise from stenosis of the optic canal and the superior orbital fissure, respectively. Lateral to the superior orbital fissure, drilling continues to the skull base to expose the foramen rotundum.
glue are positioned where watertight sutures cannot be placed. Suspension of the dura is a key step to prevent brain herniation into the orbit. It is usually tightened on the free margins of the orbital roof. Bone Infiltrated bone is drilled out from the internal theca or removed depending on the depth of tumor infiltration. In this series, the sphenoid ridge was not reconstructed; however, the temporal bone defect could be reconstructed unless it is limited to the lateral part and to the floor of the middle fossa. Polymethylmethacrylate cranioplasty was used for cranial reconstruction. Muscle Scrupulous attention is paid to muscle repositioning. The muscle is carefully suspended along the squamous crest and sutured posteriorly, creating drum-like tension to cover any residual lateral bone defect.
Results Extradural work Orbit The tumor is usually extraconal and dissectable from the preserved periorbita. Occasionally, it may be intraconal. The periorbita is not normally opened unless removal of the “en plaque” extraconal mass is inconsistent with the radiological findings or when it is already disrupted or the tumor mass is globular. Cavernous sinus Only when the tumor extends to the lateral wall of the cavernous sinus is the inner dura peeled back starting from the formen rotundum along the superior orbital fissure to the sinus and then removed. The bulk of the tumor is removed as it emerges from the sinus, taking care not to uncover the underlining cranial nerves.
In all, 47 patients undergoing microsurgical removal for SOM were identified and studied; 21 (44.6 %) were men and 26 (55.4 %) were women; the mean age was 57 years (range, 21– 77 years); the tumor was left-sided in 30 patients (63.8 %) and right-sided in 27 (36.2 %). Signs and symptoms The average medical history was 31 months (range, 6 months–8 years). The most common presenting disturbances were proptosis (n=45, 95.7 %), visual acuity deficit (n=24, 51 %), periorbital and temporal swelling (n=21, 44.6 %), and oculomotor deficits (n=18, 38.2 %) (Table 1). Radiographic findings
Devascularization Before opening the dura, external coagulation is advisable, especially near the entrance of the middle meningeal artery at the foramen spinosus. Intradural removal The dura is opened in a semicircular fashion, and extraarachnoid tumor is dissected peripherally while trying to find the best corridor for further dissection so as to later proceed from the center to the periphery. The tumor is removed using an ultrasonic aspirator. Reconstruction Dural Before applying hemostasis, the dura is removed under microscope control along the anterior and middle skull base and replaced with lyophilized dura or galea. Sponge and fibrin
A typically en plaque tumor was seen in 36 cases (76.6 %) and a globular tumor in 11 (23.4 %) (Figs. 1, 2). Tumor extension was used as a surrogate of tumor size, including other useful information for surgical planning (Table 2). The typical CT features were significant hyperostosis of the external sphenoid wing, to some extent involving the remaining part up to the clinoid process, the adjacent bone, including the middle cranial fossa, the lateral orbital wall, and the orbital roof, the wall of the sphenoidal sinus and the ethmoidal cells. Orbital rim involvement was observed in only one case. Intraconal extension was seen in 14 cases (29 %), sometimes associated with extraconal extension. The main cause of proptosis, defined as the prevalent contributing finding (>50 %), was hyperostosis (n=37, 78.8 %) and tumor mass in the remaining 10 cases (21.2 %). This observation calls attention to the importance of the criteria adopted for orbital drilling.
Neurosurg Rev Table 1 Symptoms and signs before and after treatment
Preoperative (N=47) no. of patients
Postoperative (4–6 months) no. of patients Improved
Normal
Stable/Worse
Proptosis Normal >2 mm >4 mm >6 mm >8 mm Total no. of patients (%) Visual acuity Normal 3–9/10 1–2/10 Blindness Total no. of patients (%) Visual field Normal Concentric deficits Hemianopsia Total no. of patients (%) Cranial nerve deficit Normal
2 15 19 8 3 47
0 4 7 4 1 16 (34)
2 10 11 3 1 27 (57.4)
0 1 1 1 1 4 (8.5)
23 13 5 6 47
0 7 3 0 10 (21.2)
23 5 0 0 28 (59.5)
0 1 2 6 9 (19.1)
41 4 2 47
0 2 1 3 (6.3)
41 1 0 42 (89.3)
0 1 1 2 (4.2)
29
0
28
1
Cranial nerve III, IV, VI Cranial nerve V Total no. of patients (%) Pain Normal Retro-ocular pain Trigeminal pain Total no. of patients (%)
12 6 47
6 4 10 (21.2)
1 2 31 (65.9)
5 0 6 (12.7)
36 6 5 47
0 5 2 7 (14.8)
36 0 3 39(82.9)
0 1 0 1 (2.1)
Surgical treatment In all patients, treatment was via a standardized surgical approach according to the guidelines mentioned above, basically following a clinical rationale. Proptosis was the most frequent presenting symptom and required great care during orbital wall resection. Complete tumor removal was achieved in 24 cases (51 %) and incomplete removal in 23 (49 %) (Table 3). Complete resection was limited by medial and lateral basal bone involvement and by two extradural areas, the cavernous sinus and the superior orbital fissure, the rate depending on the case population and the surgical policy rather than on the surgical technique. The periorbita was left in situ unless the tumor invaded the intraconal space. Maximal tumor removal was always attempted. En plaque tumor was found to be extraconal in 29 cases, intraconal and extraconal in 6, and only intraconal in 1 case. Globular tumors were never found to be only extraconal: seven were strictly intraconal and four were intraconal and extraconal. In about half
of the cases, the muscle was thinned if infiltrated; in two cases, it was completely removed and replaced with fat harvested from the abdomen. The orbital rim was regularly preserved, and the orbital roof and lateral orbital wall were not reconstructed. Dural, calvarial bone and muscle reconstruction was meticulously performed. The theca of the cranial vault was drilled out or removed depending on the depth of tumor infiltration. In five patients, the lateral bone up to the squamous crest was removed and substituted with methylmethacrylate. On histopathology, all tumors were classified as typical meningioma, except one which was atypical. Scheduled gamma knife radiosurgery was performed as complementary treatment in nine cases (seven with Simpson resection grade III, two with Simpson resection grade IV). Complications None of the patients died during the acute postoperative period, and minimal morbidity was described as neurological, with
Neurosurg Rev
Fig. 2 Typical orbital patterns: “en plaque” (a) and globular (b). The surgical approach to proptosis varies accordingly. In the first case, it is a matter of hyperostosis: standard two-wall resection is performed and the tumor removed (extraconal meningioma); in the second, the target of surgery is the intraconal meningioma till the orbital apex, where the tumor infiltrates the anulus of Zinn
Fig. 1 A typical spheno-orbital meningioma. Preoperative CT scan (a) showing hyperostotic bone; preoperative MRI image showing a carpetlike tumor infiltrating the anterior middle fossa dura and continuing into the orbit (b); and postoperative MRI image showing tumor removal and normal eye bulb position (c)
local and systemic complications. New signs were observed only in oculomotor function: five patients developed diplopia (cranial nerve III in four, cranial nerve IV in one), four of which completely recovered within 6 months, and one still presented with diplopia at 12 months (permanent deficit). Worsening of preoperative neurological deficit was observed in three patients, which was transient in two (one with cranial nerve III deficit, one with impaired visual acuity) and permanent in one patient
(blindness) who had presented with a severe preoperative visual deficit. Local complications included extradural hematoma (two patients) and cerebrospinal fluid leak (one patient), all of which required reoperation. Wound infection developed in one patient and systemic endocarditis in another. Globally, 14 (22.9 %) patients were not included or traced for the follow-up due to incomplete ophthalmologic data. No difference was found between the traced and the untraced patients with regard to sex, age, complications, or tumor morphology; however, the rate of total removal differed significantly (55 vs 40 %, respectively; P
ORIGINAL ARTICLE
Surgical management of ocular symptoms in spheno-orbital meningiomas. Is orbital reconstruction really necessary? Andrea Talacchi & Antonella De Carlo & Antonio D’Agostino & Pierfrancesco Nocini
Received: 12 March 2013 / Revised: 22 August 2013 / Accepted: 27 October 2013 # Springer-Verlag Berlin Heidelberg 2014
Abstract Spheno-orbital meningioma (SOM) is an intriguing tumor because of the many different factors that can influence clinical and oncological outcome after treatment. Reasoning that outcome indicator measurement is key to improving therapy, we retrospectively evaluated the management of proptosis and other ocular symptoms in 47 patients surgically treated for SOM at our department in the last 10 years. This patient series was characterized by a high rate of tumor infiltration of the extradural cranial base. Clinical outcome was assessed by comparing preoperative and postoperative ophthalmological and neurological signs. Acute postoperative complications were reported, and clinical and radiological outcome was assessed at 4–6 months, 12 months, and the last follow-up. Proptosis (measured by Hertel exophthalmometry), visual acuity, visual field defect (measured by Goldmann perimetry), diplopia (measured by the Hess-Lancaster test), and other disturbances were rated as normalized, improved, or unchanged/ worsened. The most common presenting symptoms were proptosis (95.7 %), visual impairment (51 %), and cranial nerve deficit (38.2 %). Surgery via the frontotemporal approach was performed in all 47 cases, with the primary aim to relieve symptoms/signs and maximize tumor resection. Bony orbital reconstruction was never performed. Complete resection was A. Talacchi : A. De Carlo Section of Neurosurgery, Department of Neurological, Neuropsychological, Morphological and Movement Sciences, University of Verona, University Hospital, Verona, Italy A. D’Agostino : P. Nocini Section of Dentistry and Maxillo-Facial Surgery, Department of Surgical Science, University of Verona, University Hospital, Verona, Italy A. Talacchi (*) Institute of Neurosurgery, Department of Neurosciences, University of Verona, University Hospital, P. Stefani 1, 37126 Verona, Italy e-mail: [email protected]
achieved in 51 % of cases (Simpson grades I and II) with minimal morbidity. At a mean follow-up of 52 months (range, 12–112), proptosis normalized in 90.9 % and improved in the remaining patients, visual acuity normalized in 20.8 % and improved in 45.8 % patients, cranial nerve deficit subsided in all but two cases. The recurrence rate was 29.7 %. One of the gold standards of surgical treatment, normalization of proptosis, can be achieved by accurate resection of the superior and lateral orbital walls. In this setting, careful reconstruction of the frontobasal dura is far superior to bony reconstruction. Complete tumor resection should not be pursued at the expense of increased morbidity. Keywords Surgical management . Meningioma . Skull base . Orbital tumors . Proptosis
Introduction Spheno-orbital meningioma (SOM), the modern term for Cushing group’s “hyperostosing meningioma of ala magna,” was introduced in the computed tomography (CT) era [2, 4, 7] and then better defined in the magnetic resonance imaging (MRI) era by the majority of studies that described tumor extension and modality of treatment [9, 14]. The surgical technique involves many steps, each of which is tailored case-by-case in a parsimonious approach to maximizing tumor removal. To this end, outcome measurement can provide feedback information for surgical planning. In the treatment of SOM, patient-oriented outcome may best be centered on disturbances in visual and ocular function, by far the most common clinical symptoms. However, reported outcome, ophthalmologic and neurological evaluation, and performance status may underestimate the real outcome, as referred by patients concerned about posttreatment function and cosmetic appearance [10]. Accordingly, the clinical priority is the
Neurosurg Rev
surgical management of proptosis, the main presenting symptom reported in all surgical series, caused by orbital bone thickening and orbital tumor infiltration. Improving and normalizing proptosis are the main clinical aims of surgical treatment, yet it persists in many, improves in most, and remains stable in roughly 20 % of cases [22]. Because preoperative and postoperative objective assessment has been rarely reported in the literature, the evaluation of treatment is limited, particularly with regard to restoration of facial symmetry and eye bulb position [20, 22]. This is one of the chief reasons for the continuing dilemma of the extent of bone removal and of whether to reconstruct the orbital walls. The aim of this study was to objectively assess eye bulb position before and after treatment, together with other ophthalmometric measures, in a surgical patient series in which orbital reconstruction was not performed.
Proptosis was evaluated by Hertel exophthalmometry, assuming an abnormal value to be ≥2 mm as compared with the healthy eye. The extent of surgical removal was assessed using postoperative MRI images and operative notes. The Simpson classification was adopted: grades I and II, with excision or coagulation of the dura, respectively, were defined as complete removal; grade III with persistent dural enhancement and grade IV with residual mass were defined as incomplete removal.
Material and methods
–
Meningiomas of the sphenoid bone with intraosseus tumor growth involving the orbit were defined as SOM. Out of the 118 SOMs operated on in the last 20 years (1992–2012), accounting for 7 % of all intracranial meningiomas (n= 1759) and 33 % of all sphenoid wing meningiomas (n=360) operated on at the Department of Neurosurgery, University Hospital of Verona, we reviewed the cases treated with a similar strategy (n=61) in the last 10 years and selected only those cases with complete data. We compared the symptoms, signs, radiographic features, microsurgical techniques, and outcome in 47 patients with primary tumor. Hospital charts, operative notes, follow-up clinical notes, radiographic study reports, and neuro-ophthalmological reports were retrospectively analyzed for each patient. Patients or relatives were contacted to obtain further follow-up data when necessary. All patients underwent CT, with thin slices (2 mm) and bone window scans for cranial base assessment, and MRI. Gadolinium-enhanced MRI images and fat suppression sequences were obtained to delineate the intradural, intraorbital, or extracranial tumor portion and to determine the extent of dural infiltration. The maximum tumor diameter was not reported because it is inconsistent in such irregular tumors.
–
Outcome measures
This long and tedious part of the operation starts with drilling the lesser sphenoid wing all the way down to the clinoid under microscope magnification. As a rule, the extent of orbital access is tailored to the individual patient and based on the evidence of tumor and/or on the clinical rationale for bony decompression. Great care is taken to avoid laceration of the periorbita and injury to the oculomotor muscles.
Clinical outcome was assessed by comparing preoperative and postoperative ophthalmological and neurological signs. Acute postoperative complications were reported, and clinical and radiological outcome was assessed at 4–6 months, 12 months, and the last follow-up. Proptosis, visual acuity, visual field defect (measured by Goldmann perimetry), diplopia (measured by the Hess-Lancaster test), and other disturbances were rated as normalized, improved, or unchanged/worsened.
Surgical management In light of our experience and a review of the literature, surgical management rests on the following assumptions: –
–
A surgical approach needs to be wide enough to follow the tumor. A circumferential view is needed to control the medial and lateral parts that are usually involved. The orbital rim and zygomatic bone are usually spared by tumor infiltration, except in very advanced stages. Not all the thickened bone is tumor. Infiltrated pockmarked-like bone should be differentiated from reactive ivory-like bone. This can be done with the help of the microscope. In doubtful cases, only bone decompression is performed (see the following item). Priority is given to the relief of disturbances over radical tumor removal.
Surgical approach The patient is positioned supine with the head turned not more than 20° to the opposite side and slightly tilted. A standard pterional approach is established with variable frontal or temporal extension according to the direction of tumor growth. After interfascial dissection, the temporal muscle is reflected downward and laterally to expose the pterional region, which is normally involved by hyperostosis. After craniotomy, the bone is lifted from the dura. Bone work
Orbital roof The anterior limit is given by the need to preserve the orbital rim. The medial limit is the ethmoidal sinus, but
Neurosurg Rev
usually, only the superolateral and the posterior parts of the roof are removed. Lateral orbital wall To avoid enophthalmos or to improve exophthalmos, this is the most important part of orbitotomy. The lateral orbital wall, because it is the site of extraconal tumor, is usually removed, up to the deepest part, anteriorly to the orbital profile, and inferiorly at the level of the inferior orbital fissure (at the bottom of the wall before it becomes the floor). Extending the drilling anteriorly will favor globe reduction (reentry into the orbital cone); limiting inferior drilling will avoid globe retraction (escape from the orbital cone). Optic canal-superior orbital fissure-and foramen rotundum triangle This is the most important part for visual and oculomotor management, since disturbances arise from stenosis of the optic canal and the superior orbital fissure, respectively. Lateral to the superior orbital fissure, drilling continues to the skull base to expose the foramen rotundum.
glue are positioned where watertight sutures cannot be placed. Suspension of the dura is a key step to prevent brain herniation into the orbit. It is usually tightened on the free margins of the orbital roof. Bone Infiltrated bone is drilled out from the internal theca or removed depending on the depth of tumor infiltration. In this series, the sphenoid ridge was not reconstructed; however, the temporal bone defect could be reconstructed unless it is limited to the lateral part and to the floor of the middle fossa. Polymethylmethacrylate cranioplasty was used for cranial reconstruction. Muscle Scrupulous attention is paid to muscle repositioning. The muscle is carefully suspended along the squamous crest and sutured posteriorly, creating drum-like tension to cover any residual lateral bone defect.
Results Extradural work Orbit The tumor is usually extraconal and dissectable from the preserved periorbita. Occasionally, it may be intraconal. The periorbita is not normally opened unless removal of the “en plaque” extraconal mass is inconsistent with the radiological findings or when it is already disrupted or the tumor mass is globular. Cavernous sinus Only when the tumor extends to the lateral wall of the cavernous sinus is the inner dura peeled back starting from the formen rotundum along the superior orbital fissure to the sinus and then removed. The bulk of the tumor is removed as it emerges from the sinus, taking care not to uncover the underlining cranial nerves.
In all, 47 patients undergoing microsurgical removal for SOM were identified and studied; 21 (44.6 %) were men and 26 (55.4 %) were women; the mean age was 57 years (range, 21– 77 years); the tumor was left-sided in 30 patients (63.8 %) and right-sided in 27 (36.2 %). Signs and symptoms The average medical history was 31 months (range, 6 months–8 years). The most common presenting disturbances were proptosis (n=45, 95.7 %), visual acuity deficit (n=24, 51 %), periorbital and temporal swelling (n=21, 44.6 %), and oculomotor deficits (n=18, 38.2 %) (Table 1). Radiographic findings
Devascularization Before opening the dura, external coagulation is advisable, especially near the entrance of the middle meningeal artery at the foramen spinosus. Intradural removal The dura is opened in a semicircular fashion, and extraarachnoid tumor is dissected peripherally while trying to find the best corridor for further dissection so as to later proceed from the center to the periphery. The tumor is removed using an ultrasonic aspirator. Reconstruction Dural Before applying hemostasis, the dura is removed under microscope control along the anterior and middle skull base and replaced with lyophilized dura or galea. Sponge and fibrin
A typically en plaque tumor was seen in 36 cases (76.6 %) and a globular tumor in 11 (23.4 %) (Figs. 1, 2). Tumor extension was used as a surrogate of tumor size, including other useful information for surgical planning (Table 2). The typical CT features were significant hyperostosis of the external sphenoid wing, to some extent involving the remaining part up to the clinoid process, the adjacent bone, including the middle cranial fossa, the lateral orbital wall, and the orbital roof, the wall of the sphenoidal sinus and the ethmoidal cells. Orbital rim involvement was observed in only one case. Intraconal extension was seen in 14 cases (29 %), sometimes associated with extraconal extension. The main cause of proptosis, defined as the prevalent contributing finding (>50 %), was hyperostosis (n=37, 78.8 %) and tumor mass in the remaining 10 cases (21.2 %). This observation calls attention to the importance of the criteria adopted for orbital drilling.
Neurosurg Rev Table 1 Symptoms and signs before and after treatment
Preoperative (N=47) no. of patients
Postoperative (4–6 months) no. of patients Improved
Normal
Stable/Worse
Proptosis Normal >2 mm >4 mm >6 mm >8 mm Total no. of patients (%) Visual acuity Normal 3–9/10 1–2/10 Blindness Total no. of patients (%) Visual field Normal Concentric deficits Hemianopsia Total no. of patients (%) Cranial nerve deficit Normal
2 15 19 8 3 47
0 4 7 4 1 16 (34)
2 10 11 3 1 27 (57.4)
0 1 1 1 1 4 (8.5)
23 13 5 6 47
0 7 3 0 10 (21.2)
23 5 0 0 28 (59.5)
0 1 2 6 9 (19.1)
41 4 2 47
0 2 1 3 (6.3)
41 1 0 42 (89.3)
0 1 1 2 (4.2)
29
0
28
1
Cranial nerve III, IV, VI Cranial nerve V Total no. of patients (%) Pain Normal Retro-ocular pain Trigeminal pain Total no. of patients (%)
12 6 47
6 4 10 (21.2)
1 2 31 (65.9)
5 0 6 (12.7)
36 6 5 47
0 5 2 7 (14.8)
36 0 3 39(82.9)
0 1 0 1 (2.1)
Surgical treatment In all patients, treatment was via a standardized surgical approach according to the guidelines mentioned above, basically following a clinical rationale. Proptosis was the most frequent presenting symptom and required great care during orbital wall resection. Complete tumor removal was achieved in 24 cases (51 %) and incomplete removal in 23 (49 %) (Table 3). Complete resection was limited by medial and lateral basal bone involvement and by two extradural areas, the cavernous sinus and the superior orbital fissure, the rate depending on the case population and the surgical policy rather than on the surgical technique. The periorbita was left in situ unless the tumor invaded the intraconal space. Maximal tumor removal was always attempted. En plaque tumor was found to be extraconal in 29 cases, intraconal and extraconal in 6, and only intraconal in 1 case. Globular tumors were never found to be only extraconal: seven were strictly intraconal and four were intraconal and extraconal. In about half
of the cases, the muscle was thinned if infiltrated; in two cases, it was completely removed and replaced with fat harvested from the abdomen. The orbital rim was regularly preserved, and the orbital roof and lateral orbital wall were not reconstructed. Dural, calvarial bone and muscle reconstruction was meticulously performed. The theca of the cranial vault was drilled out or removed depending on the depth of tumor infiltration. In five patients, the lateral bone up to the squamous crest was removed and substituted with methylmethacrylate. On histopathology, all tumors were classified as typical meningioma, except one which was atypical. Scheduled gamma knife radiosurgery was performed as complementary treatment in nine cases (seven with Simpson resection grade III, two with Simpson resection grade IV). Complications None of the patients died during the acute postoperative period, and minimal morbidity was described as neurological, with
Neurosurg Rev
Fig. 2 Typical orbital patterns: “en plaque” (a) and globular (b). The surgical approach to proptosis varies accordingly. In the first case, it is a matter of hyperostosis: standard two-wall resection is performed and the tumor removed (extraconal meningioma); in the second, the target of surgery is the intraconal meningioma till the orbital apex, where the tumor infiltrates the anulus of Zinn
Fig. 1 A typical spheno-orbital meningioma. Preoperative CT scan (a) showing hyperostotic bone; preoperative MRI image showing a carpetlike tumor infiltrating the anterior middle fossa dura and continuing into the orbit (b); and postoperative MRI image showing tumor removal and normal eye bulb position (c)
local and systemic complications. New signs were observed only in oculomotor function: five patients developed diplopia (cranial nerve III in four, cranial nerve IV in one), four of which completely recovered within 6 months, and one still presented with diplopia at 12 months (permanent deficit). Worsening of preoperative neurological deficit was observed in three patients, which was transient in two (one with cranial nerve III deficit, one with impaired visual acuity) and permanent in one patient
(blindness) who had presented with a severe preoperative visual deficit. Local complications included extradural hematoma (two patients) and cerebrospinal fluid leak (one patient), all of which required reoperation. Wound infection developed in one patient and systemic endocarditis in another. Globally, 14 (22.9 %) patients were not included or traced for the follow-up due to incomplete ophthalmologic data. No difference was found between the traced and the untraced patients with regard to sex, age, complications, or tumor morphology; however, the rate of total removal differed significantly (55 vs 40 %, respectively; P
