Temporal lobe encephalocele appearing as a lytic lesion of the skull base and pterygoid process. (Original Article).

Abstract

Meningoencephalocele is an uncommon condition in which brain tissue, meninges, or both protrude through a defect in the anterior cranialfossa and into the ethmoid sinus or nasal cavity. Much less often, brain tissue, meninges, or both protrude through a defect in the middle cranialfossa and into the sphenoid sinus. We report an unusual case of a middle fossa encephalocele that appeared as a lytic lesion of the skull base. The patient was treated successfully via a unique endoscopic transpterygoid app roach--that is, an endoscopic approach through the maxillary sinus and pterygopalatine fossa and into the pterygoid process.

Introduction

Meningoceles and encephaloceles are uncommon clinical entities, occurring in fewer than 1 in 35,000 persons. (1) In most cases, brain tissue, meninges, or both protrude through a defect in the anterior cranial fossa and into the ethmoid sinus or nasal cavity. Much less often, brain tissue, meninges, or both protrude through a defect in the middle cranial fossa and into the sphenoid sinus. In this article, we report an unusual case of a temporal lobe meningoencephalocele in a cranial bone (rather than a paranasal sinus) that appeared as a lytic lesion of the pterygoid process.

Case report

A 51-year-old woman complained to her primary care physician of a 1-month history of headache, memory loss, and fatigue. Her headaches were located in the occipital region. They occurred daily in the morning, continued over several hours, and abated fully later in the day. The patient's medical history was significant for systemic lupus erythematosus and hypertension.

On physical examination, tendemess was noted over the mastoid area. Her complete blood count and mastoid x-ray findings were normal. However, subsequent computed tomography (CT) of orbit, ear, and sella demonstrated a lytic, expansile, 1.5 x 2.3-cm lesion of the left skull base and the pterygoid process of the sphenoid bone, anterior to the foramen ovale (figure 1). CT also detected associated obliteration of the left vidian canal with extension into the left lateral clivus. Magnetic resonance imaging was attempted but quickly terminated because the patient experienced ear discomfort; she had previously undergone a bilateral stapedectomy, and it was suspected that her stapes prosthesis was made of a ferromagnetic material. A diagnosis of multiple myeloma was considered, but findings on serum protein electrophoresis, urine protein electrophoresis, and urine immunoelectrophoresis were normal. Her kappa chains were also normal, and her urine total protein level was slightly elevated (212 mg/ 24 hr). Findings on bone scanning were negative.

The patient was referred to a hematologic oncologist for investigation of possible multiple myeloma or a metastatic or primary neoplasm. However, the results of Mprotein testing were normal. CTs of the brain, chest, abdomen, and pelvis were obtained to rule out malignancy, and all were negative. Findings on a radiographic skeletal survey of the skull, ribs, pelvis, spine, and extremities were normal, as was a recent mammogram. Ultrasonography of the thyroid was performed to rule out malignancy, and it detected a small area of low echo texture in the lower half of the right lobe. A subsequent single-uptake thyroid scan showed that the patient's 24hour radioactive iodine uptake was low at 9.1% (normal: 15 to 30%); no cold nodules were identified. Because the skull base lesion was considered to be relatively inaccessible, a bone marrow biopsy was obtained to further rule out multiple myeloma, and it was negative.

Because of a concem that the lytic lesion might represent a metastasis, the patient was referred to our otolaryngology unit for biopsy of the skull base lesion. We obtamed a biopsy specimen with a 0[degrees] endoscope by incising the mucosa of the lateral nasal wall over the area of the ascending process of the palatine bone. A 10-mm vertical incision was made just posterior to the inferior turbinate insertion site on the lateral nasal wall. A second vertical incision was made parallel to and 5 mm posterior to the initial incision, just anterior to the eustachian tube orifice. A superior incision was then made between the two, creating an inferiorly based pedicled flap. The flap was elevated from the underlying palatine bone and pterygoid process of the sphenoid bone. Inspection identified a grayish mass situated just below a very thin, translucent bone (figure 2). The bone was elevated and found to have focal areas of dehiscence. The mass was exposed over an area of approximately 10 x 5 mm for biopsy. A smal l biopsy forceps was used to grasp the mass and remove a small piece of tissue for analysis. Clear fluid spewed from the biopsy site momentarily. Inspection revealed that the lesion was not a solid tissue mass but rather a cyst-like structure. The endoscope was placed through the biopsy site, allowing us to see the structure's whitish internal lining as well as strands of irregular tissue across its lumen. These characteristics were consistent with a meningoencephalocele (figure 3). Frozen-section histopathologic analysis revealed that the biopsy sample represented benign neural tissue.

The decision was made to resect the suspected meningoencephalocele, identify the skull base defect, and repair it to reduce and perhaps eliminate the risk of ascending meningitis. On intraoperative review of the earlier CT image, we observed that the sphenoid sinus did not pneumatize the lateral pterygoid recess. We also noted a small area of irregularity near the left foramen rotundum that we originally did not suspect was a skull base defect. Based on our intraoperative findings, we believed that this area represented the skull base defect through which the meningoencephalocele arose. We exposed the skull base near the foramen rotundum via an endoscopic transpterygoid approach. The exposure that had been achieved at the transnasal biopsy site allowed for access to the lower portion of the lesion. However, repair of the skull base defect required a more superior exposure.

We initiated the procedure by resecting the uncinate process and opening the ethmoid cells and sphenoid sinus in the usual manner for endoscopic ethmoidectomy. The maxillary sinus ostium was opened, and the posterior fontanelle mucosa was removed to create a large antrostomy opening. This allowed for maximum visualization of the posterior maxillary sinus wall, the infraorbital nerve, and the orbital floor. The mucosa of the posterior maxillary sinus wall was elevated between the antrostomy site and the infraorbital nerve, exposing the posterior wall of the maxillary sinus. The remaining mucosa of the maxillary sinus was preserved to allow for functional mucociliary clearance through the antrostomy opening.

Next, we directed our attention to the lateral nasal wall just anterior to the insertion of the middle turbinate. The neurovascular bundle was delineated in close proximity to the crista ethmoidalis ossis palatini. (2) Taking care not to traumatize the neurovascular bundle, we placed a 2-mm Kerrison rongeur into the sphenopalatine foramen and removed the thick palatine bone that made up the anterior aspect of the foramen and the posteromedial aspect of the maxillary sinus. Dissection was performed in a medial-to-lateral direction. The thinner bone of the posterior wall of the maxillary sinus was removed with a curved J curette, which exposed the fascia-enveloped contents of the pterygopalatine fossa. The fascial covering was incised, and the underlying adipose tissue was dissected bluntly.

Electrocautery was used to control bleeding from the small vessels in the fatty tissue of the pterygopalatine fossa. A hemaclip was used to ligate the sphenopalatine artery. Care was taken to identify and preserve the sphenopalatine ganglia, vidian nerve, infraorbital nerve, and the lacrimal contributions from the sphenopalatine ganglia. The fatty tissue of the pterygopalatine fossa was ablated with a bipolar forceps. While traversing the pterygopalatine fossa, we encountered a second fascial layer, which we incised to reveal the anterior aspect of the pterygoid process. A drill was used to remove a portion of this bone, which allowed us to enter the pterygoid bone and expose the anterior component of the meningoencephalocele. The opening was enlarged in all directions to widely expose the meningoencephalocele and skull base. Then the bipolar forceps were applied to the meningoencephalocele to widely open the anterior face of the lesion.

Endoscopic examination revealed that the lesion was a cyst-like structure with a large central component that had previously contained the clear fluid. Examination through a 30[degrees] endoscope revealed a 2-mm opening in the superomedial aspect of the structure through which clear fluid pulsed with each heartbeat (figure 4). Examination through the sphenoidotomy site revealed that the sphenoid sinus did not communicate with the lesion and that it was clearly separated by bone. We therefore ascertained that the lesion was a meningoencephalocele that had eroded the inner cortex of the pterygoid process of the sphenoid bone, filled the central portion of this bone, and significantly thinned the outer cortices.

To reduce the lesion, we applied bipolar forceps to the outer aspects of the meningoencephalocele wall. This served to shrink the tissue and allow the cauterized tissue to be resected with through-cutting forceps. After many applications with the bipolar and monopolar forceps, the meningoencephalocele remnant was transected with the bipolar forceps, and the remaining small stalk was reduced through the skull base defect. The defect measured 3 x 2 mm and was slightly ovoid. A left postauricular incision was made, and graft material for the skull base repair was obtained. A 3 x 3-cm area of the loose areolar fascia overlying the temporal muscle and a 3 x 3-cm area of temporal fascia were harvested. Thereafter, a 2 x 1-cm area of the outer cortex of the mastoid bone was harvested. A 4 x 3-mm piece of mastoid cranial bone was sectioned from the graft and prepared for use in repairing the skull base defect. The bone was thinned with a diamond drill under 2.6-power magnification. The middle fossa dura was elevated from the area around the defect, and the bone graft was placed extradurally but intracranially to close the defect. The piece of temporal fascia was placed firmly against the defect and the surrounding bone. The space that the meningoencephalocele had occupied was obliterated, as was the remaining fascia and cranial bone. One piece of cranial bone was used to reconstruct the medial aspect of the lateral nasal wall in the region of the ascending process of the palatine bone; the other was placed slightly lateral to the first just under the fascial graft that covered the skull base defect. The inferiorly pedicled flap was returned to its native position.

Gelfoam was packed around the entire surgical site for hemostasis. A Telfa pack was placed in the nose to support the Gelfoam. The patient was kept at bedrest for 3 days with the head of the bed elevated. Her physical activities were then allowed to progressively increase. No lumbar drain was needed. Postoperatively, the posterior wall of the maxillary sinus healed quickly. Fat in the pterygomaxillary space filled the lateral pterygoid process opening, and the posterior maxillary wall remucosalized in a few weeks (figure 5). The patient was followed closely during the first postoperative year, and she had done well, with no signs of recurrence.

Discussion

With the exception of the common temporal bone encephalocele in the ear, it is highly unusual to find a temporal lobe encephalocele in a cranial bone rather than in a paranasal sinus. Garcia (3) reported one case and Leblanc et al (4) reported three in which a bony defect in the base of the greater wing of the sphenoid bone near the foramen rotundum allowed a meningoencephalocele to protrude into the pterygopalatine fossa.

The exact cause of spontaneous encephaloceles is unknown. It is hypothesized that long-standing pressure swings and pulsations in the cerebrospinal fluid might be a factor. Low-grade or benign intracranial hypertension has also been implicated. (5) Normally occurring arachnoid villi and draining veins that penetrate the skull base have also been identified as possible sites from which a skull base defect could arise. Excessive paranasal sinus pneumatization can thin the skull base and predispose a patient to the condition. Leblanc et al hypothesized that a failure in ossification occurs at the base of the greater wing of the sphenoid bone in the region of the foramen rotundum and pterygoid process. (4)

In our case, we hypothesize that the inner cortical portion of the pterygoid process was less ossified than normal. The meningoencephalocele might have originated at a small dehiscence just lateral to the foramen rotundum, which is typical of temporal lobe encephaloceles. The sparse inner cortex offered limited resistance to the meningoencephalocele as the inner cortical bony elements were pushed aside by the expanding meningoencephalocele. Thereafter, the thicker outer cortex did offer resistance and slowed the expansion. But ultimately, this bone was thinned significantly by the meningoencephalocele. During surgery, we found that the ascending process of the pterygoid bone was thin and had focal areas of dehiscence. This appearance suggested that the dehiscence would have enlarged overtime, thereby predisposing our patient to ascending meningitis.

The surgical approach to this area warrants comment. A neurosurgical approach offers direct access to and wide exposure of the meningoencephalocele, albeit with the morbidity of craniotomy. Standard rhinologic approaches--such as the transseptal and endoscopic transethmoid approach to the sphenoid sinus--would not have permitted access to this lesion because it was not located in the sphenoid sinus. An infratemporal fossa approach could also have been used to access the area, but this procedure also carries a considerable degree of morbidity for such a benign disease. We therefore employed a unique endoscopic surgical approach to address this unusual lesion. The endoscopic transpterygoid approach--that is, an endoscopic approach through the maxillary sinus and pterygopalatine fossa and on to the pterygoid process--can provide direct access to the floor of the middle cranial fossa in the region of the foramen rotundum. The advantages of this technique include the lack of a need for an external incision (except for the bone and fascia graft site), minimal morbidity, and excellent magnified visualization of the surgical site with the endoscope. However, the procedure can be technically difficult because of limited exposure.

We recommend that the endoscopic transpterygoid approach be considered when a biopsy is needed to establish a tissue diagnosis of a disconcerting lesion in this region. We also recommend that this approach be considered in cases of middle fossa encephalocele in this region. Middle fossa encephaloceles usually occur in a lateral recess of the sphenoid sinus. Bolger and Osenbach previously described the successful use of this surgical approach in this clinical setting. (6)

This case provided us with a challenging diagnostic dilemma, and it led us to attempt a unique endoscopic approach to surgical treatment. Although we were pleased with our results, we hesitate to recommend the endoscopic transpterygoid approach for the routine treatment of encephaloceles of the middle fossa that occur in the pterygoid region. More experience is needed to better ascertain the exact role and limitations of this new and promising technique.

References

(1.) Van Nouhuys JM, Bruyn GW. Nasopharyngeal transsphenoidal encephalocele, crater-like hole in the optic disk, and agenesis of the corpus callosum. Pneumoencephalographic visualization. J Psychiat Neurol Neurochir 1964;67:243-58.

(2.) Bolger WE, Borgie RC, Melder P. The role of the crista ethmoidalis in endoscopic sphenopalatine artery ligation. Am J Rhinol 1999; 13:81-6.

(3.) Garcia FR. Un caso de epilepsia del lobulo temporal producida por un encefalocele. Rev Esp Otoneurooftalmol 1971;29:216-20.

(4.) Leblanc R, Tampieri D, Robitaille Y, et al. Developmental anterobasal temporal encephalocele and temporal lobe epilepsy. J Neurosurg 1991;74:933-9.

(5.) Kaufman B, Yonas H, White RJ, Miller CF II. Acquired middle cranial fossa fistulas: Normal pressure and nontraumatic in origin. Neurosurgery 1979;5:466-72.

(6.) Bolger WE, Osenbach R. Endoscopic transpterygoid approach to the lateral sphenoid recess. Ear Nose Throat J 1999;78:36-46.

From the Department of Otolaryngology-Head and Neck Surgery, University of Pennsylvania Health System, Philadelphia.

Reprint requests: William E. Bolger, MD, Department of Surgery, USUHS, 4301 Jones Bridge Rd., Bethesda, MD 20814. Phone: [phone]; fax: [phone]; e-mail: [e-mail]