Introduction

The middle cranial fossa is a common location of meningiomas of the cranial base. A large fraction of these meningiomas arise from the dural surfaces bordering the middle fossa, namely the cavernous sinus, sphenoid wing, tentorium, or convexity, and extend into the middle fossa, filling the concavity of the middle fossa floor secondarily. Meningiomas can, however, arise directly from the floor of the middle fossa, with minimal or no connection to the aforementioned border sites.

These tumors have been less well studied, in large part due to a lack of a firm radiographic definition of “middle fossa” meningioma, and lack of awareness of the fact that tumors do occasionally arise from the dura of the middle fossa floor. These tumors likely have been previously classified as sphenoid wing, cavernous sinus, or other meningiomas. We think that a lack of recognition of the middle fossa floor as a potential site of origin predisposes the surgeon to missing an opportunity to remove the tumor at its origin. To date, there are only two small case series specifically addressing this entity, including our own report, which is summarized here, and a 1994 report from Graziani et al.1

Definition

We define a middle fossa floor meningioma as a histo-logically confirmed meningioma with greater than 75% of its radiographic attachment on the floor of the middle fossa, with less than 25% attachment on either the sphenoid wing, cavernous sinus, petrous ridge/tentorium, or lateral convexity dura, which form the four anatomical boundaries of the middle fossa concavity as determined by magnetic resonance imaging (MRI) ( Fig. 34.1 ). We further subclassify tumors that radiographically had no attachments to boundaries of the middle fossa (class 1), and those that had between 0 and 25% attachment to the sphenoid wing (class 2), cavernous sinus (class 3), dura over the petrous ridge and tentorium (class 4), or convexity dura (class 5) as shown in Table 34.1 .

Note that it is possible in this definition for a tumor to be a middle fossa floor meningioma and still have some cavernous sinus invasion if the principal site of origin is the middle fossa floor. Radiographic examples of these meningiomas can be found in Fig. 34.2 .

Incidence

The exact incidence of middle fossa floor meningiomas is not known, in large part due to a paucity of literature on the topic. Using our own experience as an estimate, between 1991 and 2006, 1213 patients were seen by neurosurgeons at the University of California–San Francisco (UCSF) for meningiomas, of which 1034 patients underwent treatment of their lesion with either open surgery or radiosurgery. A total of 17 patients in this series met our criteria for having a middle fossa floor meningioma. Two of these patients had had previous surgery and were excluded because it was unclear where the initial site of their tumor was located, and it is possible that the middle fossa component could have been missed during the initial resection of a tumor that predominantly arose from the sphenoid wing. Thus we estimate that these lesions represent 1.4% of all known or presumed intracranial meningiomas and 6% of all meningiomas in the middle fossa.

Clinical Presentation

These tumors are often rather large at diagnosis and can present with a wide variety of nonspecific or confusing symptoms. The median patient age at time of surgery was 57 years, and the male to female ratio was 6:9. The median volume of these tumors was 21 cc; however, we have seen volumes in excess of 70 cc, with a maximum tumor diameter of 5.5 cm.

Headache was the most common complaint in this series, being a presenting complaint in 60% of these patients. Six of 15 (40%) patients presented with seizures. Not unexpectedly, trigeminal nerve dysfunction (numbness, palsy, or neuralgia) was also common, present in 33% of these patients. Also common were gait disturbance (three patients) and cognitive decline (four patients). Only five patients had no objective neurological deficit at presentation. These symptoms are presented in Table 34.2 .

Interestingly, hearing loss was a common symptom; five patients demonstrated audiographic evidence of hearing loss on presentation. Other objective symptoms are presented in Table 34.2 .

Nearly identical to our experience, Graziani et al1 noted headache in 55% of their patients and cognitive changes in 36%. Their incidence of auditory complaints was also similar to that in our series. They did not report any pre-operative trigeminal nerve dysfunction in their series, which differs from our experience, but taken together, trigeminal dysfunction appears to be less common than the anatomical location of these tumors would predict.